Universidad ORT Uruguay

Facultad de Ingeniería

Adquisidor de actividad eléctrica del cerebro,

señales de electroencefalograma (EEG).

Entregado como requisito para la obtención del título deIngeniero en Electrónica

Daniel Merlinski - 146455

Diego Alonso - 153259

Gonzalo de León - 158545

Tutores: André Fonseca

Ismael Garrido

2014

Declaración de autoría

Nosotros, Daniel Merlinski, Diego Alonso y Gonzalo de León, declaramos queel trabajo que se presenta en esa obra es de nuestra propia mano. Podemosasegurar que:

La obra fue producida en su totalidad mientras realizábamos el trabajonal de carrera;

Cuando hemos consultado el trabajo publicado por otros, lo hemos atri-buido con claridad;

Cuando hemos citado obras de otros, hemos indicado las fuentes. Conexcepción de estas citas, la obra es enteramente nuestra;

En la obra, hemos acusado recibo de las ayudas recibidas;

Cuando la obra se basa en trabajo realizado conjuntamente con otros,hemos explicado claramente qué fue contribuido por otros, y qué fuecontribuido por nosotros;

Ninguna parte de este trabajo ha sido publicada previamente a su entre-ga, excepto donde se han realizado las aclaraciones correspondientes.

Daniel Merlinski Diego Alonso Gonzalo de León

10 de Abril de 2014

2

Agradecimientos

En primer lugar queremos agradecer a nuestras respectivas familias por elapoyo y conanza brindados durante todos estos años, sin ellos nada de estohubiera sido posible.

Agradecer a nuestros amigos y compañeros a lo largo de la carrera, MartínLjubicic, Sebastián Barbat, Christian Kuster, Paolo Principi, Sebastián Barda-costa, Joaquin Oldan, Tabaré Camacho, Andrés Burel, Gerónimo Peradotto,Florencia Pla, Agustín García y Gonzalo Guerrero, por acompañarnos duranteeste largo emprendimiento, contagiándonos su alegría, enseñándonos diferentesmaneras de ver y disfrutar la vida. Gracias por hacer divertida esta trayectoriade aprendizaje y conocimientos.

A la Universidad ORT por brindarnos la posibilidad de desarrollarnos en elámbito académico, personal y laboral.

Al Ing. Andrés Azar, a la Dra. Neuropediatra Elisa Sirio y a la Dra. Anes-tesista Carmen Estela por su asesoramiento técnico sobre la temática elegida.

Por último queremos agradecer a nuestros tutores André Fonseca e IsmaelGarrido por la ayuda brindada a lo largo de este proyecto.

3

Abstract

En el presente documento se expone la investigación llevada a cabo paradesarrollar un equipo adquisidor de señales de electroencefalograma, así comoel desarrollo propiamente dicho, acompañado de un software que demuestre lasfuncionalidades y el potencial del equipo.

El sistema utiliza electrodos del tipo activos que se sitúan sobre el cuerocabelludo, para obtener la señal eléctrica generada por las neuronas. Esta se-ñal se transmite a una placa para su amplicación, la que a su vez realiza unltrado grueso de la misma. Luego es capturada por un microcontrolador quese encarga de la digitalización y acondicionamiento. Finalmente, se transmi-ten los datos a un PC donde se realiza un ltrado más no, para su posteriorvisualización y se ejecutan algoritmos que detectan patrones en las señales.

Las principales características que se destacan sobre el trabajo realizado, esque se logró un equipo compacto, móvil, exible y de bajo costo.

El emprendimiento de este proyecto fue motivado por las expectativas quegenera un equipo de estas características en un sinnúmero de aplicaciones quepodría tener el mismo. Dichas aplicaciones podrían contemplar una ampliavariedad de campos, desde las consolas de videojuegos, la robótica y el controlde procesos hasta la medicina. Como claro ejemplo de aplicación y del potencialde este prototipo, cabe destacar la posibilidad de que personas que presentanincapacidad motriz puedan valerse por sí mismas.

4

Índice

Declaración de autoría 2

Agradecimientos 3

Abstract 4

Palabras Clave 10

Glosario 12

1. Introducción 15

1.1. Objetivos del proyecto . . . . . . . . . . . . . . . . . . . . . . . 15

2. Descripción del sistema 17

3. Estudio previo en neurología 20

3.1. Introducción a la anatomía del cerebro humano . . . . . . . . . 203.2. Introducción a la electroencefalografía . . . . . . . . . . . . . . . 22

3.2.1. Posicionamiento de los electrodos, Sistema 10 - 20 [1] . . 223.2.2. Conguración y montaje de los electrodos . . . . . . . . 23

3.3. Parámetros de la señal de electroencefalografía . . . . . . . . . . 243.4. Ritmos cerebrales . . . . . . . . . . . . . . . . . . . . . . . . . . 27

3.4.1. Ritmo Alfa . . . . . . . . . . . . . . . . . . . . . . . . . 273.4.2. Ritmo Beta . . . . . . . . . . . . . . . . . . . . . . . . . 30

3.5. Artefactos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 313.6. Potenciales cerebrales . . . . . . . . . . . . . . . . . . . . . . . . 34

4. Electrodos: sensores de potencial bioeléctrico 35

4.1. Responsabilidades . . . . . . . . . . . . . . . . . . . . . . . . . . 354.2. Desarrollo . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 354.3. Implementación nal . . . . . . . . . . . . . . . . . . . . . . . . 38

4.3.1. Material utilizado para la construcción . . . . . . . . . . 384.3.2. Diseño . . . . . . . . . . . . . . . . . . . . . . . . . . . . 39

5

4.3.3. Técnica de medida . . . . . . . . . . . . . . . . . . . . . 414.3.4. Electrodo de referencia . . . . . . . . . . . . . . . . . . . 414.3.5. Montaje y posicionamiento . . . . . . . . . . . . . . . . . 41

5. Interferencia 44

5.1. Introducción . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 445.2. Fuentes de interferencia . . . . . . . . . . . . . . . . . . . . . . . 445.3. Soluciones para minimizar los efectos . . . . . . . . . . . . . . . 46

6. Proyecto OpenEEG 49

7. Placa Amplicadora 50

7.1. Descripción . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507.2. Diseño . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 517.3. Alimentación del circuito . . . . . . . . . . . . . . . . . . . . . . 51

7.3.1. Tierra virtual y plano de tierra . . . . . . . . . . . . . . 527.4. Etapas del circuito . . . . . . . . . . . . . . . . . . . . . . . . . 54

7.4.1. Filtro supresor de ondas de radio . . . . . . . . . . . . . 547.4.2. Circuito de protección . . . . . . . . . . . . . . . . . . . 577.4.3. Etapa de amplicación diferencial . . . . . . . . . . . . . 587.4.4. Segunda etapa de amplicación . . . . . . . . . . . . . . 617.4.5. Filtrado antialiasing . . . . . . . . . . . . . . . . . . . . 647.4.6. Circuito de pierna derecha . . . . . . . . . . . . . . . . . 677.4.7. Elección de los amplicadores operacionales . . . . . . . 69

7.4.7.1. Características de los integrados . . . . . . . . . 707.5. Ruido provocado por los Amplicadores [15]

. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 717.6. Respuesta en frecuencia de la placa amplicadora . . . . . . . . 75

8. Etapa Digital 78

8.1. Conversor analógico-digital . . . . . . . . . . . . . . . . . . . . . 788.2. Comunicación microcontrolador-computadora . . . . . . . . . . 798.3. Elección de módulos para comunicación BlueTooth . . . . . . . 808.4. Elección del microcontrolador . . . . . . . . . . . . . . . . . . . 818.5. Algoritmo del microcontrolador para conversión A/D y envío de

datos . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 82

9. Software Monitor de señales cerebrales 85

9.1. Implementación del software . . . . . . . . . . . . . . . . . . . . 859.2. Serial . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

6

9.3. Buer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 869.4. Proceso . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 86

9.4.1. Acondicionador . . . . . . . . . . . . . . . . . . . . . . . 869.4.1.1. Filtro 40 Hz . . . . . . . . . . . . . . . . . . . . 87

9.4.2. Diezmado . . . . . . . . . . . . . . . . . . . . . . . . . . 889.4.3. Detección de ritmos y generación de señales de control . 89

9.4.3.1. Interfaz cerebro computadora - BCI . . . . . . . 899.4.3.2. Reconocimiento de ritmos cerebrales . . . . . . 91

9.4.4. BCI basada en ritmo alfa occipital . . . . . . . . . . . . 959.4.4.1. Estimador potencia Alfa . . . . . . . . . . . . . 969.4.4.2. Filtro pasa banda . . . . . . . . . . . . . . . . . 969.4.4.3. Filtro de suavizado . . . . . . . . . . . . . . . . 989.4.4.4. Umbral . . . . . . . . . . . . . . . . . . . . . . 1019.4.4.5. Umbral adaptativo . . . . . . . . . . . . . . . . 1049.4.4.6. Performance de la BCI basada en ritmo alfa . . 105

9.4.5. BCI basada en ritmo alfa frontal . . . . . . . . . . . . . 1159.4.6. BCI basada en potenciales miográcos . . . . . . . . . . 115

9.4.6.1. Detección del parpadeo . . . . . . . . . . . . . 1169.4.6.2. Señales de control a partir de los parpadeos . . 116

9.4.7. Visualización . . . . . . . . . . . . . . . . . . . . . . . . 1209.4.7.1. Velocidad de barrido . . . . . . . . . . . . . . . 120

9.5. Servidor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1209.5.1. Comandos . . . . . . . . . . . . . . . . . . . . . . . . . . 120

10.Aplicación de prueba de la BCI 122

11.Pruebas realizadas sobre el funcionamiento del sistema 123

11.1. Pruebas con cable sin mallar y mallado . . . . . . . . . . . . . . 12311.2. Medición del ruido . . . . . . . . . . . . . . . . . . . . . . . . . 12511.3. Frecuencia de corte y atenuación en la banda de paso y a la

frecuencia de 512 Hz . . . . . . . . . . . . . . . . . . . . . . . . 12711.4. Vericación de la frecuencia de muestreo y envío de datos del

microcontrolador . . . . . . . . . . . . . . . . . . . . . . . . . . 12711.5. Pruebas realizadas al software de monitoreo . . . . . . . . . . . 128

12.Conclusiones 129

12.1. Planicación . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12912.2. Posibles mejoras a futuro . . . . . . . . . . . . . . . . . . . . . . 132

Referencias Bibliográcas 134

7

A. Apéndice 136

A.1. Ecuaciones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137A.1.1. Filtro supresor de ondas de radio . . . . . . . . . . . . . 137A.1.2. Etapa de amplicación variable . . . . . . . . . . . . . . 140A.1.3. Filtro antialiasing . . . . . . . . . . . . . . . . . . . . . . 142A.1.4. Exigencia de CMRR . . . . . . . . . . . . . . . . . . . . 147A.1.5. Cálculo del ruido producido por los amplicadores ope-

racionales . . . . . . . . . . . . . . . . . . . . . . . . . . 147A.2. Manual de usuario . . . . . . . . . . . . . . . . . . . . . . . . . 152

A.2.1. Instrucciones de puesta en marcha . . . . . . . . . . . . . 152A.2.2. Alimentación del electroencefalógrafo . . . . . . . . . . . 153A.2.3. Software de monitoreo . . . . . . . . . . . . . . . . . . . 154

A.2.3.1. Descripción . . . . . . . . . . . . . . . . . . . . 154A.2.3.2. Requerimientos . . . . . . . . . . . . . . . . . . 155A.2.3.3. Funcionamiento . . . . . . . . . . . . . . . . . . 155

A.2.4. Programa de prueba para la BCI [17] . . . . . . . . . . . 160A.2.4.1. Requerimientos . . . . . . . . . . . . . . . . . . 160A.2.4.2. Instrucciones de puesta en marcha del robot . . 161A.2.4.3. Instrucciones de uso . . . . . . . . . . . . . . . 162

A.3. Código del microcontrolador ATmega 2560 . . . . . . . . . . . . 163A.4. Códigos de Scicoslab para ltros digitales . . . . . . . . . . . . . 168

A.4.1. Código de ltro Gauss . . . . . . . . . . . . . . . . . . . 169A.4.2. Código Filtro IIR pasa bajo . . . . . . . . . . . . . . . . 172A.4.3. Código ltro IIR pasa banda . . . . . . . . . . . . . . . . 175A.4.4. Código para generar ltro pasa bajo IIR . . . . . . . . . 179A.4.5. Código para generar ltro pasa banda IIR . . . . . . . . 183A.4.6. Código para modelado de la BCI . . . . . . . . . . . . . 188

A.5. Códigos de JAVA para el software de monitoreo . . . . . . . . . 192A.5.1. Clases del dominio . . . . . . . . . . . . . . . . . . . . . 193

A.5.1.1. Clase buer . . . . . . . . . . . . . . . . . . . . 193A.5.1.2. Clase escribir archivo . . . . . . . . . . . . . . . 198A.5.1.3. Clase FFT . . . . . . . . . . . . . . . . . . . . 201A.5.1.4. Clase gráca . . . . . . . . . . . . . . . . . . . 204A.5.1.5. Clase leer archivo . . . . . . . . . . . . . . . . . 212A.5.1.6. Clase procesamiento de señal . . . . . . . . . . 215A.5.1.7. Clase proceso . . . . . . . . . . . . . . . . . . . 227A.5.1.8. Clase serial . . . . . . . . . . . . . . . . . . . . 237A.5.1.9. Clase sistema . . . . . . . . . . . . . . . . . . . 243A.5.1.10. Clase servidor . . . . . . . . . . . . . . . . . . . 246

8

A.5.2. Clases de Interfaz . . . . . . . . . . . . . . . . . . . . . . 249A.5.2.1. Clase ventana . . . . . . . . . . . . . . . . . . . 249A.5.2.2. Clase main . . . . . . . . . . . . . . . . . . . . 285

A.6. Código de programa de comunicación PC - NXT . . . . . . . . . 287A.7. Código robot NXT . . . . . . . . . . . . . . . . . . . . . . . . . 291A.8. Diagrama esquemático de la placa amplicadora . . . . . . . . . 295A.9. Layout de la placa amplicadora . . . . . . . . . . . . . . . . . . 298

B. Anexo 300

B.1. Hoja de datos ATmega 2560 . . . . . . . . . . . . . . . . . . . . 301B.2. Hoja de datos LM78xx . . . . . . . . . . . . . . . . . . . . . . . 710B.3. Hoja de datos INA114BP . . . . . . . . . . . . . . . . . . . . . 739B.4. Hoja de datos TLC277 . . . . . . . . . . . . . . . . . . . . . . . 757B.5. Hoja de datos HC05 . . . . . . . . . . . . . . . . . . . . . . . . 795B.6. Hoja de datos HC06 . . . . . . . . . . . . . . . . . . . . . . . . 809

9

Palabras Clave

EEG

Electrodo

Filtrado

Microcontrolador

BCI

BT

ATMEGA 2560

Fusiformes

Ritmo cerebral

ERS

ERD

Artefacto

PR

PE

BRD

DRL

EMI

CMRR

Vcm

Cable mallado

VGND

OpenEEG

Placa Arduino

INA114BP

TLC277P

Filtro antialiasing

10

Conversor A/D

Diagrama de Bode

AAMI

IEC

RMS

Relación señal-ruido

LM7805

FTDI

puerto COM

Filtros IIR

Filtros FIR

Enventanado

Tiempo de establecimiento

Nivel de background

11

Glosario

EEG: Electroencefalograma, registro de los potenciales eléctricos que se orig-

inan en la corteza cerebral.

Electrodo: Conductor eléctrico utilizado para hacer contacto con una parte

no metálica de un circuito.

Filtrado: Discriminación de una determinada frecuencia o gama de frecuen-

cias de una señal eléctrica.

Microcontrolador: Circuito integrado programable, capaz de ejecutar las

órdenes grabadas en su memoria.

BCI: Interfaz Cerebro-Computadora (por sus siglas en inglés, Brain Com-

puter Interface).

BT: BlueTooth, protocolo de comunicaciones diseñado especialmente para

dispositivos de bajo consumo, que requieren corto alcance de emisión y basados

en transceptores de bajo costo.

ATMEGA 2560: Microcontrolador utilizado para este proyecto.

Fusiformes: Que tengan una forma redondeada.

Ritmo cerebral: Oscilaciones eléctricas que genera el cerebro.

ERS: Evento relativo a la sincronización.

ERD: Evento relativo a la desincronización.

Artefacto: Toda señal eléctrica que aparece en el registro de electroence-

falografía y no pertenece a la actividad bioeléctrica del cerebro.

PR: Potenciales Relacionados, generan una disminución en la amplitud base

del ritmo cerebral.

PE: Potenciales Evocados, cambios en la señal bioeléctrica producto de es-

tímulos externos.

BRD: Archivo del programa EAGLE Circuit Board File.

DRL: Circuito derivador de pierna derecha (por sus siglas en inglés, Driven

12

Right Leg), es el circuito encargado de generar la tensión para el electrodo de

referencia, además de aumentar la relación de rechazo al modo común.

EMI: Interferencias electromagnéticas (por sus siglas en inglés, ElectroMag-

netic Interference ), son señales de origen externo que pueden introducir un

error en la señal a medir.

CMRR: Relación de rechazo al modo común (por sus siglas en inglés, Com-

mon Mode Rejection Ratio), representa la habilidad del amplicador para rec-

hazar las tensiones de modo común presentes en las entradas.

Vcm: Tensión de modo común, diferencia de potencial entre el usuario y la

masa del circuito.

Cable mallado: Tipo de cable recubierto por una malla o un tubo metálico

que actúa de jaula de Faraday para evitar el acople de ruidos y otras interfer-

encias.

VGND: Es la referencia de la señal medida.

OpenEEG: Proyecto open source creado con el n de construir electroence-

falógrafos de bajo costo.

Placa Arduino: Placa sobre la cual viene montado el microcontrolador

ATmega 2560.

INA114BP: Amplicador de instrumentación utilizado para la primera etapa

de la placa.

TLC277P: Amplicador operacional utilizado para la segunda etapa la

placa.

Filtro antialiasing: Filtro pasa bajo analógico que precede a la etapa de

conversión analógica-digital, cuya función es limitar el ancho de banda de señal

a la mitad de la frecuencia de muestreo.

Conversor A/D: Conversor analógico digital.

Diagrama de Bode: Representación gráca que sirve para caracterizar la

respuesta en frecuencia de un sistema.

AAMI: Association for the Advancement of Medical Instrumentation.

IEC: International Electrotechnical Commission.

RMS: Valor ecaz (por sus siglas en inglés, Root Mean Square), es una

medida estadística de magnitud.

13

Relación señal-ruido: Valor medido en decibeles, que se utiliza para com-

parar el nivel de la señal de interés con el nivel de ruido. Considerando valores

medidos en voltios, se calcula como 20× log( SenalRuido

).

LM7805: Regulador de tensión utilizado para la alimentación de la placa.

FTDI: Conversor Serie-USB (TTL) que permite conectar dispositivos TTL

por USB.

puerto COM: Interfaz de comunicación serial de datos digitales.

Filtros IIR: Filtros de respuesta impulsiva infnita.

Filtros FIR: Filtros de respuesta impulsiva nita.

Enventanado: Proceso de multiplicar una señal por una ventana.

Tiempo de establecimiento: Tiempo necesario para que la señal alcanze

en 95% del valor nal.

Nivel de background: Nivel base de un ritmo cerebral.

14

1. Introducción

El proyecto consiste en la creación de un prototipo compacto, móvil y exiblepara la adquisición de señales similares a las obtenidas mediante un electro-encefalografo. Para lograr el objetivo se construyeron tres placas electrónicas,todas ellas realizadas por el grupo. La primera fue realizada con componentesdisponibles en plaza (excepto los amplicadores operacionales) mientras que lasegunda y tercera son placas doble faz y se montaron con componentes traídosdesde EEUU.

El emprendimiento de este proyecto fue motivado por las expectativas quegenera un equipo de estas características en un sinnúmero de aplicaciones quepodría tener el mismo. Dichas aplicaciones podrían contemplar una ampliavariedad de campos, desde el de las consolas de video juegos, la robótica y elcontrol de procesos hasta la medicina.

1.1. Objetivos del proyecto

El objetivo del proyecto fue realizar un prototipo que cumpla con las siguien-tes características:

Compacto y móvil: se desea obtener un prototipo con estas característi-cas para que sea fácilmente trasladable, cómodo de usar y que no requieraprácticamente ningún mantenimiento.

Bajo costo: como parte de los objetivos, se priorizó que el dispositivo lo-grado sea de bajo costo.

Componentes accesibles: los componentes utilizados deberán ser consegui-dos sin mayores dicultades.

Independencia en la implementación: el prototipo logrado debe ser lomás universal posible, es decir, debe funcionar para cualquier persona y sobre

15

cualquier sistema operativo.

A lo largo del desarrollo del proyecto dichos conceptos fueron establecidoscomo premisas a la hora de tomar decisiones.

16

2. Descripción del sistema

Este capítulo tiene como objetivo dar una descripción global del proyecto,con el objetivo de facilitar la lectura de los siguientes capítulos.

El prototipo logrado se puede dividir en 2 grupos, por un lado está la etapade adquisición de señal y por otro lado el sistema de monitoreo. Para cumplirla primera etapa se diseñó un equipo compacto que permita adquirir la señaleléctrica presente en el cerebro. El adquisidor a su vez se compone de 3 ele-mentos fundamentales, que son los electrodos, la placa de amplicación y elmicrocontrolador.

Como segunda parte, se realizó un software de monitoreo que graque enun eje de coordenadas y en tiempo real, las señales eléctricas generadas, y quepermita la aplicación en un caso de uso con la construcción de una InterfazCerebro-Computadora (BCI, por sus siglas en inglés, Brain Computer Inter-face).

En la gura 2.1 se muestra un diagrama de bloques general del sistema, se-gún lo mencionado recientemente, seguido de una breve introducción de cadaelemento.

Figura 2.1.: Diagrama global del sistema.

Electrodos: los electrodos se sitúan sobre la supercie del cráneo, son utili-zados para la obtención de las señales bioeléctricas generadas por el cerebro.

17

Se compone únicamente de un pequeño circuito y del cable mallado a travésdel cual se transmite la señal.

Placa de amplicación: esta parte del sistema se encarga de amplicarla señal de µVobtenida con los electrodos y de realizar un ltrado primario(ltrado grueso). La señal amplicada luego se transmite a la siguiente etapamediante un cable convencional.

Placa Arduino con microcontrolador ATMEGA: este microcontrola-dor recibe las señales de la placa amplicadora. Es el encargado de digitalizary enviar los datos a un PC mediante BlueTooth (BT). Se incorpora un móduloadicional que permite la comunicación utilizando dicho protocolo.

Sistema de monitoreo: El software de monitoreo diseñado realiza el trata-miento de los datos para gracar la señal obtenida y características particularesde la misma. Además funciona como un servidor local, enviando informaciónsobre la señal a posibles clientes. Consta de software y un pequeño hardwareadicional para hacer posible la comunicación por BT.

Los electrodos utilizados son del tipo activos y particularmente en este pro-yecto, se trabajó con los electrodos de Joerg Hansmann. El hardware o cascosobre el cual fueron montados los electrodos es un diseño propio. Se estable-cieron determinadas características a la hora de realizar el diseño del casco; elmismo debería ser ajustable a distintos usuarios para que pueda ser utilizadopor varias personas, además era de vital importancia lograr que los puntos decontacto de los electrodos puedan ejercer una presión adecuada sobre el cuerocabelludo. También era deseable poder colocar los electrodos en ciertas zonasde interés, y nalmente se debería lograr todo lo anterior con un diseño robustoy simple, para poder ser transportado fácilmente. En el capítulo 4 se explicarála evolución del prototipo y los distintos elementos que lo componen.

El desarrollo del hardware de la placa de amplicación se basa en un modeloopen source obtenido de la página openEEG.org, con algunas modicacio-nes realizadas para lograr un diseño más compacto, económico, con un mayorrechazo a la interferencia de la red eléctrica y con una mayor excursión de laseñal de salida. En el capítulo 7 se explica en detalle este elemento del proyecto.

Los elementos de software de este proyecto son tres, el software del micro-

18

controlador ATMEGA, el software de monitoreo y la aplicación de prueba parala BCI. Todos son independientes entre sí y se conectan en el orden que estánmencionados. Los primeros dos se comunican utilizando BT, mientras que losúltimos dos se conectan localmente mediante sockets. La conexiones entre mi-crocontrolador y el software de monitoreo es automática una vez que ambosestán en funcionamiento, como se explicará en la sección 8.3, esto evita que elusuario requiera conocimiento en este ámbito. La conexión entre el software demonitoreo (servidor) y la aplicación de prueba (cliente) se completa al aceptarla petición del cliente desde el servidor.

El software desarrollado para el microcontrolador tiene como objetivo la di-gitalización de la señal analógica a una tasa de muestreo adecuada, para suposterior transmisión a un PC. El mismo puede ser implementado sobre cual-quier microcontrolador, cuyos registros utilizados, frecuencia de reloj y tensiónde operación, sean iguales a las del ATMEGA 2560. El funcionamiento en de-talle de este sistema se explica en el capítulo 8.

El software diseñado para el monitoreo puede ser ejecutado sobre cualquiersistema operativo. Este elemento permite la visualización de la señal de EEGobtenida en el tiempo, además de algunas señales que se consideran de interés,el funcionamiento de este software se explicará en el capítulo 9.

19

3. Estudio previo en neurología

Para el comienzo del proyecto, fue necesario realizar un estudio introductoriosobre todos los temas relacionados a la electroencefalografía. El estudio quese realizó se considera de carácter fundamental para el éxito en la posteriorimplementación, teniendo como objetivo profundizar el conocimiento en lassiguientes áreas:

En dónde y porqué se generan las señales bioeléctricas.

Tipo de señales que se pueden obtener en amplitud y frecuencia.

Posición en la que se deben colocar los electrodos para obtener las señales.

Ritmos cerebrales detectables.

Ruidos e interferencias en señales de electroencefalograma.

3.1. Introducción a la anatomía del cerebro

humano

El cerebro humano se divide en dos hemisferios, derecho e izquierdo, a su vezlos hemisferios se subdividen en regiones denominadas lóbulos dependiendo desu funcionalidad. Estas subregiones son, lóbulo frontal, occipital, temporal yparietal.

La corteza cerebral es un capa de tejido gris de aproximadamente 10.000 mi-llones de neuronas que cubre la supercie de los hemisferios cerebrales. Es laencargada de interpretar los estímulos provenientes de los órganos sensoriales,del control de los movimientos, de producir el razonamiento y el juicio.

Las neuronas que constituyen el tejido nervioso del cerebro son células exci-tables, es decir posen la capacidad de recibir, interpretar y trasmitir impulsoseléctricos los cuales pueden ser captados por electrodos situados sobre el cuerocabelludo.

20

Lóbulo frontal:situado en la parte anterior de la cabeza, alberga la corteza motora pri-maria, se encarga del movimiento ocular voluntario e interviene en elrazonamiento y en el habla.

Lóbulo occipital:se encuentra en la parte posterior de la cabeza, contiene a la cortezavisual.

Lóbulo parietal:esta ubicado entre el lóbulo frontal y occipital debajo del hueso parie-tal, interviene en la interpretación del dolor, reconocimiento espacial ysentido del tacto.

Lóbulo temporal:se sitúa debajo del lóbulo parietal, contiene a la corteza primaria auditiva,interviene en la memoria y en el sentido del olfato.

Figura 3.1.: Encéfalo.

21

3.2. Introducción a la electroencefalografía

El electroencefalograma (EEG) es el registro de los potenciales eléctricos quese originan en la corteza cerebral. Dicha actividad es adquirida por electrodossituados sobre la supercie del cuero cabelludo. Los pequeños potenciales re-cogidos por los electrodos son trasmitidos al electroencefalógrafo que amplicae imprime en una pantalla los grafos de las señales obtenidas.

3.2.1. Posicionamiento de los electrodos, Sistema 10 -20 [1]

Los electroencefalógrafos en la actualidad poseen entre 20 y 40 canales lo queles permiten capturar la actividad bioeléctrica de varias regiones del cerebrosimultáneamente. La posición los electrodos fue estandarizada internacional-mente según el denominado sistema 10 − 20 (propuesto por Jasper en 1958),cuyo nombre deriva de la distancia que hay entre dos electrodos subyacentes,esta puede ser 10 o 20 por ciento de la distancia que existe entre la región pos-terior y anterior de la cabeza (gura 3.2). Dicha estandarización tiene comoobjetivo comparar las señales eléctricas de un sujeto en el tiempo o con las deotros individuos.

Este sistema se basa en la relación entre la ubicación de los electrodos y lazona subyacente de la corteza cerebral. De esta manera cada región de la super-cie del cráneo se identica con una letra que representa al lóbulo subyacentey un número que determina su posición relativa. Los números pares correspon-den al hemisferio derecho y los impares al hemisferio izquierdo. Las letras 'F','O', 'T' y 'P' corresponden a los lóbulos frontales, occipitales, temporales yparietales respectivamente. También se prevé en la nomenclatura del sistemala letra 'C' que identica a los electrodos situados en el centro del cráneo, laletra 'A' para los puntos utilizados como referencia y 'Z' (cero) como subíndicepara indicar las posición media entre dos electrodos.

22

Figura 3.2.: Sistema 10/20. [1]

3.2.2. Conguración y montaje de los electrodos

Los electrodos se pueden congurar de manera monopolar, bipolar o lapla-ciana según se observa en la gura 3.3. [2]

Monopolar:se registra el potencial eléctrico de cada electrodo en comparación al elec-trodo de referencia.

Bipolar (Diferencial):se registra la diferencia de potencial entre dos electrodos.

Laplaciana:se registra la diferencia de potencial entre un electrodo primario y el pro-medio de varios electrodos secundarios que rodean al primario.

Figura 3.3.: Conguraciones de los electrodos. [2]

23

El montaje de los electrodos determina como se conectan entre sí. Los mon-tajes expuestos a continuación son utilizados para sistemas de canales diferen-ciales (gura 3.4). [1]

Montaje circular (halo):describe una circunferencia.

Montaje longitudinal (para sagital):sigue una linea anteroposterior en ambos hemisferios cerebrales.

Montaje transversal (tiara media):sigue una linea transversal que recorre ambos hemisferios.

Figura 3.4.: Montaje circular, montaje longitudinal y montaje transversal. [1]

3.3. Parámetros de la señal de

electroencefalografía

A continuación se detallarán los parámetros básicos en el análisis electroen-cefalográco.

Frecuencia

24

Es la característica más importante, las señales EEG tienen frecuenciasque van desde pocos Hz hasta 100 Hz. El espectro se divide principal-mente en 5 bandas que se nombran con las siguientes letras griegas, δ ,θ, α, β, γ, (ver cuadro 3.1).

delta (δ) < 4 Hztheta (θ) 4− 8 Hzalfa (α) 8− 12 Hzbeta (β) 12− 32 Hz

gamma (γ) > 32 Hz

Cuadro 3.1.: Bandas de la señal EEG.[1]

Es importante tener en cuenta que los límites de estos rangos no sonprecisos y que pueden presentar leves diferencias, según la biografía.

Amplitud

Normalmente existe menor amplitud en la región anterior respecto ala zona posterior. Si bien las amplitudes absolutas de las señales carecende importancia, este parámetro cumple un rol importante cuando se pre-senta asimetría en las señales que provienen de regiones homólogas deambos hemisferios.

Simetría

En condiciones normales se espera obtener señales similares tanto enfrecuencia como en amplitudes que provengan de áreas homologas deambos hemisferios. En consecuencia existen anomalías de asimetría enfrecuencia, asimetría en amplitud o asimetría de amplitud y frecuencia,este último es el caso más común y se observa en la gura 3.5.

25

Figura 3.5.: Se observa en áreas izquierdas una asimetría de mayor amplitudy menor frecuencia constituida por ondas delta a 2 Hz. [1]

Morfología

Debido a que no se puede denir la morfología de una señal ya que laforma de las señales puede variar con cada persona, identicaremos a lasseñales normales como aquellas que presenten características fusiformes,es decir que tengan una forma redondeada, como se muestra en la gura3.6 A. La gura 3.6 B presenta lo inverso a lo anterior, ondas irregulares,con cambios bruscos, no fusiforme.

Figura 3.6.: Morfología de la señal EEG. [1]

26

3.4. Ritmos cerebrales

En ciertas condiciones, el cerebro es capaz de generar determinadas oscilacio-nes, llamadas ritmos cerebrales. Estos ritmos, pueden ser detectados utilizandocualquier método de electroencefalografía, desde los invasivos a los no invasivosrealizados en el cuero cabelludo.

Los ritmos cerebrales aparecen en distintas frecuencias, los más importan-tes son el ritmo alfa (α) y el beta (β). El primero se encuentra en el rango de8−12 Hz (mismo rango que la banda alfa), mientras que el segundo en el rangode 12−24 Hz. Por lo general, se cumple que la potencia de los ritmos disminuyeal aumentar la frecuencia, debido a que los ritmos de más alta frecuencia, songenerados por grupos de neuronas más pequeños.

Existen distintos tipos de ritmos cerebrales, estos aparecen o son bloqueadoscon distintas actividades cerebrales (por ejemplo tareas visuales, motoras, deconcentración, etc.). Cada ritmo tiene sus propias características, pero en lamayoría de ellos se observa que los ritmos son generados por grupos de neuro-nas que se encuentran en reposo, es decir que no están realizando su actividadtípica. Dentro de la zona occipital, en la región encargada del procesamientovisual puede observarse un ejemplo de lo mencionado, si estamos con los ojosabiertos la potencia del ritmo alfa es muy pequeña, pero al cerrarlos, esta au-menta signicativamente. La relación entre la potencia alfa con ojos cerradosy con ojos abiertos varía según la persona, sexo y edad e incluso puede variarun poco para una misma persona en cuestión de minutos.

A este fenómeno de intensicación del ritmo se le llama sincronización asocia-da al evento o ERS (por sus siglas en inglés Event Related Synchronization, loque en español se traduce como evento relacionado a la sincronización), mien-tras que al fenómeno de la atenuación del ritmo se le llama desincronizaciónasociada al evento o ERD (por sus siglas en inglés Event Related Desynchro-nization).

3.4.1. Ritmo Alfa

El ritmo alfa fue estudiado desde los comienzos de la electroencefalografíaa principios del siglo XX, y su estudio y análisis continúa hasta hoy en día.Siempre es el punto de comienzo para el análisis visual de los estudios de EEG,investigando si el ritmo alfa está presente y si sus características son apropia-

27

das para la edad. Además de proveer pistas acerca el estado de ansiedad oexcitación del paciente, la presencia y características del ritmo alfa occipitalson determinantes críticos en la evaluación del signicado de otras actividadespresentes.

Frecuencia

En los adultos, el rango normal de la frecuencia del ritmo occipital alfa,es usualmente de 8−12 Hz. A continuación se muestra la distribución dela frecuencia central del ritmo alfa en un estudio realizado a 200 adultoscon edades entre 24 y 35 años (gura 3.7), y la evolución de la mismadesde temprana edad, (gura 3.8). [3]

Figura 3.7.: Distribución de la frecuencia del ritmo Alfa.

28

Figura 3.8.: Evolución de la frecuencia del ritmo Alfa.

Amplitud

El ritmo alfa se encuentra siempre que las grabaciones son hechas direc-tamente sobre las regiones apropiadas y en individuos no anestesiados.Ciertamente, una gran cantidad de generadores del ritmo alfa existen enla corteza y las profundidades del cerebro, y producen ritmos de altovoltaje. Por otra parte, el voltaje del ritmo alfa obtenido en el cuerocabelludo, generalmente, no es mucho mayor que el nivel de ruido de losamplicadores (ver 7.5). Los estudios han demostrado que entre el 6 y7 por ciento de los adultos sanos, tienen ritmos alfa por debajo de los15µV, el 75% entre 15− 45µV y el resto entre 45− 100µV.

Regulación

Una buena regulación de la frecuencia y el voltaje del ritmo alfa occipi-tal es característica de los electroencefalogramas de aproximadamente el80% de los adultos jóvenes. Al aumentar la edad, hay una tendencia de

29

la actividad alfa de volverse menos regulada (alteración en amplitud yfrecuencia).

Distribución

La región occipital es el sitio de mayor voltaje del ritmo alfa en el 65% delos adultos y el 95% de los niños. Sin embargo, es posible que en algunosindividuos normales la amplitud de la actividad alfa en la región centraly temporal, sea igual o más alta que en la occipital.

En el 32% de los jóvenes adultos normales, la actividad alfa está am-pliamente distribuida. Los estudios con electrodos implantados muestranque existen múltiples generadores del ritmo alfa en el cerebro humano,no sólo en la región occipital, sino que también en la región central ytemporal. Estos generadores se superponen e inuencian entre sí, por lotanto lo que es medido en el cuero cabelludo reeja un promedio de laactividad alfa.

Comportamiento

El ritmo Alfa aparece como un ERS al cerrar los ojos y como un ERDluego abrirlos. El comportamiento del ritmo alfa en relación a otros even-tos distintos al cierre-apertura de los ojos, es variable. Hans Berger en1924, descubrió que la amplitud del ritmo alfa de un individuo disminuyeo es bloqueada durante períodos de concentración mental, como la rea-lización de cálculos. Se ha sugerido, pero no completamente demostrado,que en algunos sujetos el ritmo alfa puede ser bloqueado o disminuido enamplitud por la ansiedad.

3.4.2. Ritmo Beta

Este ritmo está asociado a la actividad motora y por esta razón se localizaprincipalmente en la corteza sensomotora (gura 3.1). La frecuencia de esteritmo se extiende desde los 12 Hz hasta 24 Hzen la mayoría de la población.

Amplitud

30

Las señales de gran amplitud correspondientes al ritmo beta, están pre-sentes en la corteza cerebral en pacientes no anestesiados. Esta actividades mayormente atenuada en el cuero cabelludo. En el 98 % de los niñosy adultos normales, la amplitud está por debajo de 20µV, en el 70 % esde 10µV o menos. La actividad beta con tensiones de 25µV o más, esgeneralmente considerada anormal en los electroencefalogramas, aunquela presencia de estos altos voltajes tiene poca utilidad en los diagnósticos,excepto cuando se sospecha la ingestión de drogas por parte del paciente.

Distribución

A pesar de que el ritmo Beta se encuentra principalmente en la cor-teza sensomotora, también es posible encontrarlo con mayor intensidaden la región frontocentral u occipital, aunque este aumento no parecetener importancia alguna.

Comportamiento

El ritmo Beta aparece como un ERS al realizar los movimientos y comoun ERD luego de estos. Cuando se mueve un extremo izquierdo, se ge-nera el ritmo en el lóbulo parietal derecho y viceversa.

3.5. Artefactos

Se dene como artefacto o articio a toda señal eléctrica que aparece en elregistro de electroencefalografía y no pertenece a la actividad bioeléctrica delcerebro.

Se pueden distinguir dos grupos de articios, por un lado se encuentran losque proviene por culpa de deciencias técnicas como electrodos mal jados,mala calibración del equipo de electroencefalografía, interferencia de corrientealterna (gura 3.9), etc. Este tipo de articios son evitables tomando las pre-cauciones necesarias, la degradación de la señal de EEG sufrida por culpa deeste tipo de artefactos vuelve inutilizable el registro.

31

Figura 3.9.: Interferencia por corriente alterna. [1]

El segundo grupo de articios, son los articios siológicos. Estos dependendel sujeto y no pueden ser evitados, tienen como ventaja que al ser de cortaduración solo invalidan parte del registro. Es necesario identicarlos ya que deno ser así las conclusiones obtenidas a partir de dicho registro serian incorrec-tas.

Estas señales espurias en su mayoría son producto de la contracción de losmúsculos frontales, temporales y maseteros. Los electrodos captan las seña-les eléctricas provenientes de los músculos, que son de orden mayor que lospotenciales generados por las neuronas, sumado a que el movimiento de loselectrodos, producto de la contracción, modica las impedancias de los mis-mos registrando señales anómalas.

El parpadeo de los ojos también es captado al igual que el movimiento ocular,debido a que el ojo está polarizado constituyendo un dipolo córnea-retina quegenera diferencias de potencial que simula señales de baja frecuencia (gura3.10).

32

Figura 3.10.: Articio por parpadeo. [1]

En ocasiones los electrodos captan los potenciales cardíacos, estos últimosson fácilmente distinguibles por su periodicidad (gura 3.11).

Figura 3.11.: Interferencia por electrocardiograma. [1]

33

3.6. Potenciales cerebrales

Los potenciales cerebrales se pueden clasicar en dos grupos, potenciales es-pontáneos y evocados. En el primer grupo está comprendido por los ritmoscerebrales (ver sección 3.4) y los potenciales relacionados (PR), estos últimosson producto de eventos puntuales como la actividad motora o concentración.Los PR generan una disminución en la amplitud base del ritmo cerebral (nivelde background), a este fenómeno se conoce como ERD, los ERS suceden cuan-do los PR se extinguen. Un ejemplo de este fenómeno sucede en la ERD y ERSdel ritmo alfa occipital al abrir y cerrar los parpados. Este tipo de potencialesson de gran interés para las BCI (ver 9.4.3.1) ya que son controlados por lavoluntad del sujeto.

Los potenciales evocados (PE) son cambios en la señal bioeléctrica productode estímulos externos aplicados sobre los órganos sensoriales o nervios perifé-ricos. Debido a que las variaciones, que son producto de estos estímulos, sondemasiado pequeñas con respecto al EEG de fondo se utiliza la técnica depromediación. Dicha técnica consiste en aplicar una serie de estímulos conse-cutivos para posteriormente promediar las señales de EEG obtenidas. Dadoque los estímulos mantienen una relación temporal, al aumentar el número demuestras, aumentara la amplitud del PE, por otro lado como el ritmo de fondodel EEG no mantiene ningún relación temporal, tenderá a desaparecer de laseñal promedio.

Los potenciales evocados son muy utilizados por los neurólogos ya que brin-dan información patológica del cerebro. No son utilizados para el uso de BCI(ver 9.4.3.1) dado que dichas señales no son controladas por el sujeto y elanálisis de estos eventos no se pueden realizar en tiempo real.

34

4. Electrodos: sensores depotencial bioeléctrico

4.1. Responsabilidades

Los electrodos son el primer eslabón de la cadena de medida. Tienen la mi-sión de integrar en el circuito a una persona, con el n de poder medir lospotenciales bioeléctricos generados por las neuronas. Siempre serán posiciona-dos sobre el cuero cabelludo, de forma de obtener un buen contacto con la piel.

En base a estos requisitos, se realizó una investigación inicial de los tipos deelectrodos disponibles. En este capítulo se describen los electrodos utilizadospara el proyecto y las distintas modicaciones realizadas sobre este componentehasta la actualidad.

4.2. Desarrollo

Uno de los puntos más importantes en la medición de señales de EEG, esque la señal de entrada es del orden de los microvoltios.

Para distinguir las señales EEG de otras señales, normalmente se utilizanamplicadores de instrumentación, que permiten tomar 2 señales y rechazanla parte de la señal que es común en ambas entradas como se muestra en lagura 4.1.

35

Figura 4.1.: Esquema de un electrodo pasivo o convencional.

El caso ideal sería cuando el amplicador rechaza todo el ruido, dejando limpiala señal de EEG.

El punto que se analizó en este caso, fue la inuencia de la impedancia piel-electrodo para la medición de la señal. Esta impedancia se encuentra en seriecon la impedancia de entrada del amplicador de instrumentación, es decir quecuanto mayor sea dicha impedancia, menor es la señal que llega al amplica-dor, esto signica que como primer premisa se debe obtener una impedanciapiel-electrodo pequeña. Por otra parte, las impedancias vistas por ambas seña-les deben ser iguales, para que el ruido que llegue al amplicador sea de igualmagnitud, de lo contrario, las señales ruidosas de entrada del amplicador ten-drán diferentes amplitudes y no se rechazará todo el ruido común. La segundapremisa entonces, será obtener impedancias iguales, para lo cual generalmentese utiliza un gel conductor.

Con el objetivo de lograr mejorar el proyecto en estos dos aspectos se comenzóa investigar y trabajar sobre electrodos del tipo activos, que fundamentalmenteincorporan un amplicador operacional, en la modalidad de seguidor de ten-sión, montado físicamente sobre el electrodo, como se muestra en la gura 4.2.

36

Figura 4.2.: Esquema de un electrodo activo.

Cuando se coloca el amplicador operacional cercano a la zona de medición,la alta impedancia de entrada y baja impedancia de salida del mismo, actúadirectamente sobre las premisas planteadas. Cuanto mayor es la impedanciade entrada del amplicador operacional, menor peso tiene la impedancia piel-electrodo, es decir, sigue siendo un factor importante, pero no como en el casode los electrodos pasivos. Por otra parte las impedancias de salida son bajas eiguales, lo cual contribuye a que llegue el mismo nivel de señal a la etapa deamplicación. [4]

Para el proyecto en particular, es válido decir que el tipo de electrodos uti-lizados fue evolucionando a medida que se deseaba mejorar la calidad de laseñal obtenida y la versatilidad del dispositivo.

Al comenzar el proyecto, se decidió utilizar electrodos pasivos de plata clo-rada (AgCl). Una ventaja de este tipo de electrodos es que son de uso comúnen los hospitales, para estudios de electrocardiograma (señales del orden de1000 veces mayores) y por lo tanto no es difícil conseguirlos. Su precio es ac-cesible para público en general, su diseño hace que se adapte perfectamenteen cualquier parte del cráneo y fundamentalmente no era necesario ningún ti-po de acondicionamiento para poder incluirlo en el prototipo. Como grandesdesventajas se encontró que estos electrodos no son re-utilizables, es decir, senecesitaba una gran cantidad de ellos para poder realizar todas las pruebasnecesarias durante el proyecto. No se pueden utilizar en supercies con pelo,lo cual limitaba mucho la zonas de estudio disponibles y los sujetos de prueba.

37

Finalmente, la relación señal-ruido que ofrecen es la más baja, con respecto aotro tipo de modelos.

El siguiente paso fue la implementación de electrodos activos, básicamente unseguidor de tensión, que aumenta la impedancia del sistema de amplicacióncon respecto a la impedancia piel-electrodo, como ya se explicó anteriormen-te. La principal desventaja que presentaron al inicio, era que requerían de unestudio previo para la construcción y tiempo de implementación en cuanto aldiseño de las placas, compra de componentes y construcción. Luego, este tipode electrodos resultó ser muy superior a los anteriores por varias razones. El ti-po de construcción permitía utilizarlo aún en zonas con pelo, son re-utilizablesy no requieren de ningún gel conductor para su utilización.

4.3. Implementación nal

4.3.1. Material utilizado para la construcción

Debido a que el material estará en contacto directo con la piel se debentomar algunas consideraciones. La piel es un tejido conductivo cuyo materialintracelular y extracelular está compuesto de soluciones electrolíticas, en lacual la corriente es transportada por iones, por otro lado, los electrodos estáncompuestos principalmente por una supercie de metal, altamente conductivo,en el cual la corriente es transportada por electrones, es decir, el electrododebe convertir las corrientes iónicas, que son el mecanismo de conducción delas señales bioeléctricas en los tejidos, en corrientes eléctricas, en consecuencia,esto genera que la interfaz piel-electrodo, sea una interfaz ruidosa. [5]

Cuando se habla de materiales más utilizados para la construcción de electro-dos de supercie, se trata de oro, plata, acero inoxidable, platino entre otros. Eldesafío consiste en lograr que la impedancia de contacto piel-electrodo se man-tenga constante en el tiempo o lo más constante posible. Sobre este aspecto,basados en un estudio realizado por A.Searle y L.Kirkup sobre la impedanciade contacto piel-electrodo, para distintos materiales de electrodos, el materialseleccionado para los electrodos del proyecto, tanto para los activos como parael electrodo de referencia, fue plata (Ag).

La gráca de la gura 4.3 muestra los resultados obtenidos por A.Searle yL.Kirkup [6] de la impedancia con respecto al tiempo (La gráca muestra un

38

promedio de datos tomados sobre 5 sujetos de prueba). [6]

Figura 4.3.: Impedancia del contacto piel-electrodo en el tiempo para distintosmateriales de construcción.

Como se observa, la respuesta más constante se obtiene para el caso de elec-trodos Ag/AgCl.

Otro de los aspectos importantes al seleccionar el material, es la caracterís-tica de polarización de los mismos. Esto signica que la carga eléctrica puedeacumularse en la supercie del material generando un voltaje de continua quepuede rondar los 200 mV, lo cual es de magnitud mucho mayor a lo que sedesea medir. Es necesario entonces un material cuyo potencial no varíe con-siderablemente, cuando la corriente pasa a través de él. En este sentido, loselectrodos de Ag/AgCl cumplen las características deseadas. [4]

Para la construcción de los electrodos de este proyecto se utilizó como ma-terial de contacto con la piel, alambre de plata para los electrodos del lóbulooccipital y una placa de plata para el electrodo frontal y para el electrodo dereferencia.

4.3.2. Diseño

El diseño del circuito de los electrodos se baso en el prototipo Jarek Foltynsky,con las modicaciones sugeridas por Joerg Hansmann. [7]

39

La elección de este diseño de electrodos activos en vez de otros disponibles[[8]], se debió a que el mismo era respaldado en otros proyectos, a su vez laguía de construcción era la más completa, con detalle de las pruebas realizadas,sumado a un circuito preciso, prolijo, y con la posibilidad de adquirir todossus materiales sin mayores dicultades.

En las guras 4.4 y 4.5 se muestra el circuito de los electrodos activos y eldiseño BRD1 de los mismos, respectivamente.

Figura 4.4.: Circuito de electrodo activo.

1BRD, archivo del programa EAGLE Circuit Board File.

40

Figura 4.5.: Diseño del circuito de los electrodos activos.

4.3.3. Técnica de medida

La técnica más común para la adquisición de biopotenciales utiliza tres elec-trodos. Mediante dos de ellos se captura la señal bioeléctrica en modo dife-rencial y el electrodo restante opera como electrodo de masa o electrodo dereferencia. Dicha técnica es la utilizada en este proyecto para la medición.

4.3.4. Electrodo de referencia

Estos electrodos miden con respecto a un electrodo de referencia, tambiénconocido como electrodo DRL (driven right leg). El funcionamiento en detalledel mismo, se explicará más adelante (ver subsección 7.4.6). En lo que reerea la construcción, el electrodo de referencia de este proyecto, consiste de unaplaca de plata.

4.3.5. Montaje y posicionamiento

Como ya se mencionó anteriormente, para obtener una buena señal es fun-damental, tener un buen contacto entre el electrodo y la piel. Es necesario quela presión que ejerce el electrodo sea uniforme, que tenga la menor movilidadposible, evitando así las interferencias por movimiento. También se consideróimportante a la hora de diseñar el prototipo, que la posición pueda ser ajus-table previendo de este modo, que sea adaptable al sujeto de prueba y podermedir potenciales en distintos puntos de interés.

41

Es válido mencionar que para los electrodos pasivos, no fue necesario ningúnhardware, ya que los mismos vienen con un autoadhesivo muy práctico.

Una vez que se comenzó a trabajar con electrodos activos, los mismos fuerondispuestos sobre una banda elástica, como se muestra en la gura 4.6. Estonos permitió comenzar a trabar y realizar las primeras pruebas del prototipo.

Figura 4.6.: Electrodos montados sobre banda elástica.

Este diseño tenía como ventaja que los electrodos eran móviles, deslizándosea través del elástico, pero introdujo algunos problemas ya que la presión queotorgaba la vincha no se traducía en un buen contacto electrodo-piel, lo quegeneraba una importante inestabilidad en la medida. Esto trajo como conse-cuencia, una pequeña modicación en la disposición de los componentes deestos electrodos, se decidió pasar la parte de contacto hacia el centro, de modode poder ejercer una presión más uniforme. En la gura 4.7, se muestra elBRD, de dichos electrodos. Otra de las desventajas de este modelo, se encuen-tra en el electrodo DRL, el mismo debía ser sostenido por el usuario, haciendopresión con el dedo pulgar, lo que afectaba la concentración y generaba arte-factos provocados por los músculos.

42

Figura 4.7.: Electrodos activos con la parte de contacto en el centro.

Finalmente, en el diseño denitivo, los electrodos construidos (ambos mo-delos) fueron montados sobre el arnés de un casco de protección (gura 4.8)para poder ajustarse alrededor de la cabeza. Los electrodos que van sobre ellóbulo occipital, son los diseñados en primera instancia, mientras que el elec-trodo frontal es el diseñado en segunda instancia. En este último, la zona decontacto, se cambio por una placa de plata. Además se agrega en el casco, elelectrodo DRL, próximo al electrodo frontal, de forma tal que el usuario sedespreocupa del mismo.

Figura 4.8.: Foto del diseño nal de la vincha.

43

5. Interferencia

5.1. Introducción

Una de las dicultades que se presentan en la adquisición de señales biopo-tenciales son las interferencias electromagnéticas (EMI)1 generadas por fuentesexternas y en particular por la red de distribución eléctrica. Esto se debe a quelas señales de EEG son muy susceptibles a ser afectadas por diferentes tiposde señales eléctricas tanto internas al circuito, como externas.

Se denomina interferencia a las señales de origen externo que podrán generarun error en la señal adquirida. Las interferencias son un fenómeno físico queocurre cuando dos o más ondas se superponen para formar una onda resultantede mayor o menor amplitud que las anteriores [9]. En este caso dicho fenómenoocurre con ondas de tensión.

Estas ondas ingresan al sistema de medida de distintas formas y se utilizanvarios métodos y técnicas que permiten la reducción o eliminación de las mis-mas. Para eliminar el efecto de las fuentes de error es necesario entender dedonde provienen y de que manera afectan la señal que se desea medir.

5.2. Fuentes de interferencia

En esta sección se describen algunas fuentes de interferencia, luego, en lagura 5.1 se muestra un modelo reducido de como se ven afectadas las señalespor las interferencias. [10]

1. Capacidad parásita que se forma entre la red de distribución y los cablesde medida y electrodos.

2. Capacidad parásita que se forma entre la red de distribución y el paciente.

1EMI por su sigla en inglés, ElectroMagnetic Interference

44

3. Interferencia generada por campos magnéticos, como los cables de ad-quisición y el usuario forman un circuito de lazo cerrado, si la superciede dicho lazo es atravesada por un campo magnético variable se generancorrientes parásitas que distorsionan la señal medida.

4. Potencial en el contacto: diferencia de potencial entre los electrodos y elpaciente que varía con el movimiento del electrodo.

5. Potencial bioeléctrico: interferencia generada por los músculos, como con-secuencia del movimiento del paciente.

Figura 5.1.: Capacidades parásitas que producen interferencia. [13]

En esta imagen, Z1, Z2 y Zg corresponden a las impedancias de los elec-trodos. La capacidad parásita que se forma entre la red de distribución y loscables que se utilizan para la medición está representada por los capacitoresC1 y C2, mientras que la capacidad que se forma entre la red de distribucióny la etapa de amplicación se representa mediante C3. La capacidad parásitaque se forma entre la red de distribución y el paciente se representa como el ca-pacitor Cb. Zi representa la impedancia de entrada a la etapa de amplicación.

Los valores estimados por distintos autores para estos elementos, son lossiguientes:

45

Cb: Capacidad Línea-Paciente. (0,2 pF− 5 pF) (Pallas Areny, 1991 [11]).

C1, C2: Capacidad Línea-Cable. (≈ 0,1 pF/m) (Grimbergen, 1991 [11]).

Z1, Z2: Impedancia de contacto del electrodo i. (1 kΩ − 1 MΩ) (Rosell,1988 [11]).

Zi: Impedancia de entrada de modo común.

Vcm: Tensión de Modo Común. Es un parámetro de gran importancia,representa la diferencia de potencial entre el usuario y la masa del cir-cuito. Si bien el usuario no tiene un único potencial y se denen variosnodos sobre él, lo que se hace usualmente es considerar al mismo comoisopotencial, basados en que las impedancias de internas son pequeñasen comparación con las demás impedancias, y dicho potencial se denela sobre la impedancia del tercer electrodo.

5.3. Soluciones para minimizar los efectos

Es importante tener en cuenta la importancia de minimizar las interferenciasdesde la captura de la señal, para evitar la propagación de la misma por lasdistintas etapas, ya que si bien existen varias opciones de procesamiento digitalde señales, para eliminar la interferencia de la red luego de la captura, resultadifícil hacerlo sin deteriorar la señal, en particular la fase, y más difícil aúnefectuar estos procesamientos en tiempo real.

La principal causa de interferencia es generada por la red de distribucióneléctrica. Esta interferencia afecta la señal a través de la tensión Vcm, funda-mentalmente a través de 2 mecanismos. En primer lugar porque la Relaciónde Rechazo al Modo Común (CMRR)2 del amplicador es un parámetro nitoy en segundo lugar esta interferencia puede pasar de modo común a mododiferencial debido a las diferencias en las impedancias.

El primer modo es fácilmente reducible mediante los amplicadores de ins-trumentación como se verá más adelante en la subsección 7.4.3. El circuito deadquisición y amplicación debe tener un CMRR elevado, que permita elimi-nar el efecto de dichas señales, lo que lo vuelve un parámetro de diseño crítico

2CMRR representa la habilidad del amplicador para rechazar las tensiones de modo comúnpresentes en las entradas.

46

para el circuito y determina entre otros elementos la elección de los operacio-nales a utilizar.

El segundo mecanismo requiere que la impedancia de entrada de modo co-mún tenga un valor muy elevado. Una solución muy utilizada cuando se utilizansistemas de medición de 3 electrodos es la utilización de circuitos activos, quea partir de una realimentación negativa de Vcm, permiten una disminuciónimportante de la misma. En la subsección 7.4.6 se explicará en detalle el fun-cionamiento de este circuito.

Otros métodos que fueron utilizados en este proyecto para minimizar el efectode la interferencia se enumeran a continuación.

1. Utilización de cables mallado para la transmisión de la señal analógica,conectando la malla del cable a VGND. Esto permite evitar la formaciónde capacidades parásitas entre la pantalla y el cable y además, de estaforma, se provee un camino de baja impedancia a las corrientes que circu-lan por C1 y C2, evitando que circulen a través de las impedancias de loselectrodos, lo cual de otro modo, produciría una tensión de modo diferen-cial considerable, dependiendo de las diferencias entre las impedanciasde los electrodos. También se redujo, dentro de lo posible, la longitud delos cables de los electrodos, de forma de disminuir las capacidades C1 yC2.

2. La etapa de amplicación fue diseñada de forma tal que tuviese bajaganancia a señales en modo común, respecto de la ganancia a señalesdiferenciales.

3. En esta aplicación, no se tomó en cuenta la disminución del área de ujodel campo magnético ya que los electrodos se posicionan muy próximosunos de otros, lo que hace que las áreas expuestas sean reducidas y lainterferencia producida por fem inducida en el lazo de conexión no generedistorsiones importantes. Cuando se trata de este tipo de interferenciaresulta efectivo un blindaje magnético de toda el área involucrada, perono es una solución aplicable para equipos portátiles.

4. Mantener los electrodos lo más limpios posibles.

5. El usuario se debe de mantener lo más quieto posible.

47

6. El equipo se alimenta mediante baterías (o pilas), esto hace que la impe-dancia de aislación sea grande, disminuyendo las interferencias electro-magnéticas.

48

6. Proyecto OpenEEG

En un principio se pretendía realizar un diseño propio del hardware necesa-rio para la obtención de datos, luego de asesoramiento sobre el tema con unaempresa que trabajó con equipos similares, se repensó esta idea debido a quela construcción del hardware tomaría un tiempo cercano a un año, que era eltiempo que se contaba para el desarrollo de este proyecto. En particular serecomendó la utilización del proyecto openEEG [12].

Gran parte del hardware fue diseñado en base al proyecto openEEG , el cualfue creado con el objetivo de poder construir electroencefalógrafos de bajocosto para la adquisición de señales cerebrales, con nes de aplicarlo en en-trenamiento con EEG biofeedback, o en interfaces cerebro-computadora (ver9.4.3.1) o simplemente para observar la actividad cerebral.

Este proyecto consta de todas las partes necesarias en el proceso de la adqui-sición de señales cerebrales, desde el hardware utilizado para la medición delas señales hasta el software para la visualización de las mismas en la compu-tadora. Se realizaron diversos cambios respecto al diseño original con distintosnes:

para reducir los costos

aumentar la seguridad y comodidad para el usuario

facilitar la construcción y aumentar la calidad de la señal medida.

Los cambios realizados, se explicaran en detalle en los capitulos 7 y 8.

49

7. Placa Amplicadora

7.1. Descripción

La placa amplicadora cumple la función de recibir la señal capturada porlos electrodos en el cuero cabelludo (señal del orden de los µV), amplicarlaen una excursión que va desde 0 a 5 V (la referencia de la señal es VGND= 2,5 VVGND = 2,5V) y posteriormente ltrarla para luego ser enviada almicrocontrolador con el objetivo de ser digitalizada.

También cuenta con una etapa llamada circuito de pierna derecha (DRL)1,la cual cumple la función de disminuir el ruido de modo común.

En la gura 7.1 se muestra el diagrama de bloques simplicado para un solocanal.

Figura 7.1.: Esquema de la placa de amplicación.

1DRL por sus siglas en ingles (driven right leg)

50

7.2. Diseño

El diseño fue realizado en base al proyecto OpenEEG como se explicó ante-riormente.

Previo a la construcción, se estudió cada etapa para vericar su correctofuncionamiento teórico. Con este modelo de placa, cada una de ellas es capazde capturar dos canales, por lo tanto, se construyeron dos placas para lograrla adquisición de cuatro canales. La primera placa fue realizada íntegramentepor el grupo de trabajo, con componentes disponibles en plaza (excepto losamplicadores operacionales). Las dos placas nales, son de doble faz y fueronmontadas con componentes traídos desde EEUU.

Es posible también variar la ganancia de cada placa por separado en funciónde la resistencia de un potenciómetro variable.

7.3. Alimentación del circuito

La alimentación del circuito es un tema de vital importancia, esta debe sermuy estable para obtener una señal constante y de buena calidad. Se utilizauna alimentación con fuente simple o única, es decir que se tiene sólo tensiónpositiva, en este caso de 0 a 5 V .

La placa digital diseñada por el proyecto openEEG contaba con toda laetapa de alimentación, además el diseño incluye el microcontrolador ATMELAT90S4433P utilizado. Esta placa fue sustituida por una placa Arduino, conun microcontrolador ATMEL ATMEGA2560, con el objetivo de simplicar suconstrucción, reducir los costos y mejorar la calidad de la señal.

La placa original utilizaba alimentación de la red eléctrica, la que luego setransformaba en corriente continua. Este aspecto fue modicado para el proyec-to y se utiliza alimentación mediante baterías. Este cambio trae considerablesventajas al proyecto. En primer lugar, esta mejora aumenta signicativamen-te la seguridad del usuario evitando el contacto con la red eléctrica (mediantecomponentes). Cabe señalar que openEEG aísla eléctricamente al usuario, peroes aún más seguro utilizar baterías. Además de brindar una mayor protección,

51

al aislar el prototipo se protege al equipo de posibles irregularidades en la redy, como ya se mencionó anteriormente, se aumenta el rechazo a las EMI`s.

Otra gran ventaja de utilizar baterías se debe a que no hay necesidad deutilizar transformadores o fuentes conmutadas, se eliminan los condensadoresy bobinas necesarios para el pasaje de alterna a continua, de manera que seevita el ruido producido por los mismos y se reduce en gran medida el hardwarenecesario, lo que conlleva también a una reducción en el costo. En aplicacionestan delicadas como lo es la obtención de señales cerebrales es muy importanteintentar de eliminar todas las posibles fuentes de interferencia. Por último, eluso de baterías permite la movilidad del prototipo, cumpliendo así con uno delos objetivos propuestos en el proyecto.

La desventaja que se presenta en el uso de baterías, radica en que las mismasse agotan, aunque considerando el poco consumo del prototipo, las bateríaspueden mantenerse cargadas durante días. Además, para mitigar el gasto ge-nerado por el recambio de las mismas, se decidió utilizar baterías recargables.Se midio el consumo del sistema total, resultando en un valor promedio de71 mA. A partir de este valor y conciderando que las baterías utilizadas tienenun poder de 900 mAh, se logra un tiempo de uso continuo teórico de 12,7 horasaproximadamente.

Para obtener los 5 V estables para alimentar la placa amplicadora y el mi-crocontrolador, se utilizó un integrado LM7805 (ver datasheet en anexo B.2),ya que el mismo viene implementado en la placa Arduino. Además este inte-grado no agrega ruido por conmutación ya que funciona como una resistenciavariable para mantener su salida estable. Su función es mantener la salidaconstante en 5 V ante variaciones provocadas por las baterías y acepta alimen-taciones entre 6 V y 20 V, pero es recomendable utilizar entre 7 V y 12 V. Comolas baterías recargables otorgan 1,2 V, fue necesario el uso de 6 baterías paralograr 7,2 V.

7.3.1. Tierra virtual y plano de tierra

Como fue explicado en la sección anterior, para alimentar el microcontro-lador y la placa amplicadora se utilizan 5 V, de esta manera, tanto la placaamplicadora como el microcontrolador trabajan en el mismo rango. La utili-zación de una fuente única, como en este caso, trae consigo el problema de nopoder obtener tensiones negativas respecto a la referencia (0 V). La solución

52

para poder obtener biopotenciales tanto positivos como negativos, consiste enla utilización de una tierra virtual (VGND), cuya tensión es el punto mediodel rango de trabajo, en este caso 2,5 V. Esta tensión será la referencia de losbiopotenciales medidos, es decir que los mismos estarán montados sobre estevalor y así se logran tensiones por debajo o por encima de la referencia.

En la gura 7.2 se puede observar el circuito utilizado para lograr la ten-sión VGND, se utiliza un amplicador operacional en el formato de seguidorpara no cargar al divisor resistivo y mantener VGND constante. El operacio-nal se encuentra acompañado de una serie de condensadores de desacople. Loscondensadores de desacople cumplen la función de igualar el ruido de altafrecuencia en sus dos terminales (cada terminal está conectada a tensiones di-ferentes) de manera que la diferencia de tensión entre ambas no sea afectadapor el ruido. Así se eliminan las posibles uctuaciones debidas a ruidos e inter-ferencias, logrando que las tensiones de continua sean lo más estables posibles.

Figura 7.2.: Circuito VGND.

53

Un factor muy importante para lograr la estabilidad de las tensiones de con-tinua y disminuir el ruido en todas las pistas, es la implementación de un planode tierra. El plano de tierra brinda varios benecios al circuito para la reduc-ción del ruido. En primer lugar, como su área es extensa, la impedancia delplano de tierra es muchísimo más pequeña de lo que sería en caso de utilizarpistas de tierra comunes. Esto contribuye a que la tensión sea homogénea entodos los puntos del circuito. Es importante tener en cuenta que como el circui-to es de doble capa, los planos de tierra de ambas caras deben estar conectadosen múltiples puntos del circuito para lograr una pequeña impedancia y que lastensiones también sean constantes en ambas caras del circuito. El área extensadel plano de tierra también provoca un efecto capacitivo entre el mismo y lasdemás pistas, funcionando como capacitores de desacople para interferenciasde alta frecuencia, aunque el valor de las capacidades depende de la distanciaa las pistas además del área, por lo que es necesario la colocación de otroscondensadores de desacople en las zonas del circuito donde la capacitanciaprovocada por el plano de tierra no sea suciente.

7.4. Etapas del circuito

A continuación se explicarán las distintas etapas del amplicador y sus fun-ciones en su orden correspondiente, es decir que las salidas de cada etapa sonconectadas a las entradas de la siguiente.

7.4.1. Filtro supresor de ondas de radio

En la primera etapa de la placa se encuentran tres capacitores acompañadosde algunas resistencias. El objetivo de dichos componentes es ltrar las altasfrecuencias de radio que pudieron haber entrado al sistema a través del cableque va desde los electrodos a la placa. Un diagrama del ltro se muestra enla gura 7.3. A los puntos CH 1−y CH 1+llegan las señales obtenidas porlos electrodos activos, al utilizar amplicadores operacionales las impedanciasvistas son del orden de los 100 Ω. Esta impedancia es despreciable comparadacon las impedancias vistas por OUT 1y OUT 2, las cuales coresponden a lasimpedancias de entrada del amplicador de instrumentación y son del orden1010Ω.

54

Figura 7.3.: Filtro supresor de ondas de radio.

Las transferencias obtenidas para este ltro son las siguientes (ver desarrolloen apéndice A.1.1):

Vo1 =Vi1 × (0, 242× 10−6 × S + 1) + Vi2 × 22× 10−9 × S

5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1

Vo2 =Vi2 × (0, 242× 10−6 × S + 1) + Vi1 × 22× 10−9 × S

5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1

Sin embargo el impacto de Vi1 en Vo2 y el de Vi2 en Vo1, son despreciablescomparadas con los aportes de Vi1 en Vo1 y el de Vi2 en Vo12, por lo que las trans-ferencias pueden aproximarse a las que se muestran en las ecuaciones 7.1 y 7.2.

Es vital que en el rango de interés de la señal suceda lo explicado, ya quede lo contrario las señales de interés se distorsionarían. El diagrama Bode delas transferencias Vo1

Vi2y Vo2

Vi1(tomando Vi1 = 0 y Vi2 = 0 respectivamente) se

muestra en la gura 7.4, este es igual para ambas transferencias son igualesdebido a la simetría del circuito. El diagrama de Bode de las transferenciasde las ecuaciones 7.1 y 7.2 (ver desarrollo en apéndice A.1.1) se muestra en lagura 7.5.

Vo1 =Vi1 × (0, 242× 10−6 × S + 1)

5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1(7.1)

55

Vo2 =Vi2 × (0, 242× 10−6 × S + 1)

5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1(7.2)

Figura 7.4.: Diagrama de Bode Vo1Vi2

y Vo2Vi1

.

56

Figura 7.5.: Diagrama de Bode Vo1Vi1

y Vo2Vi2

.

La frecuencia de corte que se obtiene es aproximadamente 650 kHz. Es válidoaclarar que la única radio que transmite debajo de esa frecuencia en Uruguayes CX 4 Radio Rural, 610 kHz.

7.4.2. Circuito de protección

El circuito de protección, como lo indica su nombre, se encarga de protegeral usuario de altas corrientes o voltajes, también se encarga de proteger a loscomponentes de descargas electrostáticas. El mismo se muestra en la gura7.6.

57

Figura 7.6.: Circuito de protección.

Mientras las tensiones en las entradas no alcancen los ± 0,2 V respecto aVGND, los transistores no conducen, es decir que funcionan como un circuitoabierto. Si la tensión de una de las entradas se eleva por encima del valormencionado, uno de los dos transistores comienza a conducir y con la ayudade las resistencias evitan que el voltaje supere los 0,7 V además de limitar lacorriente.

7.4.3. Etapa de amplicación diferencial

El circuito construido cuenta con tres etapas de amplicación diferentes. Laprimera etapa está formada por el amplicador de instrumentación INA114BP(ver hoja de dato en anexo B.3), el cual fue usado en el proyecto openEEG.

Los amplicadores de instrumentación son utilizados cuando se realizan me-diciones en las que la señal de interés es difícil de registrar, ya sea por sermuy pequeña o por encontrarse en un ambiente ruidoso. Estos amplicado-res poseen un alto rechazo al modo común (CMRR), el cual se incrementa alaumentar la ganancia, una alta impedancia de entrada y bajas variaciones desus parámetros con la temperatura. El último factor no es muy importante en

58

este caso, ya que las corrientes manejadas son muy pequeñas (del orden de losmA), de manera que generan potencias a través de los operacionales del ordende los mW, aumentando la temperatura de los mismos en menos de 1 °C. Sinembargo las otras dos características son de suma importancia para obteneruna buena calidad de la señal.

En el capítulo 4 ya se habló de las importancias de tener una alta impedanciade entrada, a esto se le suma la necesidad de un buen rechazo al modo común.En este tipo de equipamientos, esta característica es vital debido a que la am-plitud de la interferencia provocada por la red eléctrica puede llegar a ser másde 10.000 veces la amplitud de la señal de interés. Qué tan alto sea el rechazoal modo común dene en gran parte que tan buena es la inmunidad al ruido,pero no alcanza simplemente con que el amplicador de instrumentación tengaun alto rechazo al modo común, ya que el modo común se puede transformaren modo diferencial por diferencias de impedancias, antes de llegar a la etapadiferencial. Para mejorar este aspecto se utiliza el circuito de pierna derechaque es detallado más adelante en 7.4.6.

Los amplicadores nunca son ideales, es decir que siempre introducen ruidopropio. En los circuitos amplicadores de varias etapas, el primer amplicadores generalmente un amplicador de instrumentación, de manera de introducirel menor ruido posible a la señal de salida, ya que esta luego será amplicadaen otras etapas.

En la gura 7.7 se observa el esquemático de la primera etapa de amplica-ción.

59

Figura 7.7.: Esquema primera etapa de amplicación.

El amplicador de instrumentación toma las señales provenientes de dos elec-trodos, amplica la diferencia de estas y les suma la tensión VGND (VGND =2,5 V respecto a tierra) colocada en pin número 5. De esta manera se estable-ce VGND como la referencia y se logra obtener señales negativas(respecto aVGND) con la alimentación de 0 a 5 voltios utilizada por el amplicador.

Teniendo en cuenta las excelentes características de los amplicadores deinstrumentación antes mencionadas, sería lógico utilizar la mayor gananciaposible del amplicador de instrumentación (en el caso del INA es de 10.000con un BW = 100 Hz), sin embargo la ganancia utilizada en el circuito es de12,36, determinada por las resistencias R216 y R217.

Hay una razón por la cual separar las etapas de amplicación y asignar-le una baja ganancia a la primera etapa, es debido a que algunos materialesusados para la construcción de electrodos tienen la característica de ser pola-rizables, como ya se explicó anteriormente, generando un voltaje de continua.Este voltaje puede llegar a los 200 mV en los peores casos. Ahora si asignamosla mayor ganancia posible en la primera etapa, digamos que amplicamos laseñal 10.000 veces, este voltaje de continua se amplicaría a 2000 V, claro queesto es imposible en la realidad, ya que la señal se saturaría y a la salida severían 5 V (el voltaje de alimentación) perdiendo toda información de la señalEEG. Por lo tanto se toma una ganancia de 12,36 de manera que en el peor

60

caso la señal de continua se amplique a casi 2,5 V , dejando un rango de apro-ximadamente 0,03 V para la señal EEG.

Como se explicó en el capítulo 3.4, las señales de EEG no superan los 100µV,que amplicados 12,36 veces serían de 1,24 mV,de manera que son casi 25 vecesmás chicas que el rango libre.

La plata (material que fue utilizado para la construcción de los electrodos) esmenos polarizable que otros materiales también usados, por lo tanto, generanuna menor tensión de continua debido a dicha polarización. Además la pola-rización se presenta más al utilizar gel conductor, el cual no fue necesario porhaber construido electrodos activos; por lo tanto en este caso, podría haberseaprovechado una mayor ganancia en la primera etapa. No se hizo esto paraque el dispositivo sea compatible con los distintos metales utilizados en loselectrodos.

El capacitor C221 colocado a la salida del amplicador, se encarga de blo-quear la tensión de continua antes mencionada, y así poder continuar ampli-cando la señal en las próximas etapas. Considerando la impedancia vista porC221, la cual es de aproximadamente 1 MΩ, se obtiene una frecuencia de cortede 0,16 Hz, de manera de afectar lo menos posible la respuesta en las bajasfrecuencias. Se observa el punto PAD202, el cual comparado respecto a VGND,muestra la tensión de continua bloqueada. La función de los capacitores C215y C218 es la de mantener las tensiones (5 V y 0 V) de alimentación constantes,ya que las uctuaciones de estas distorsionarían la señal de salida.

Finalmente, la tensión Vcm es la suma de las tensiones medidas por los elec-trodos dividida entre dos. Esta tensión, junta con la tensión simétrica obtenidaa partir del otro canal de la placa, son las entradas del circuito DRL.

7.4.4. Segunda etapa de amplicación

Para la segunda etapa de amplicación se utilizó el amplicador operacionalTLC277P (ver hoja de datos en anexo B.4). La gura 7.8 muestra el esquemá-tico de esta etapa.

61

Figura 7.8.: Esquema segunda etapa de amplicación.

La ganancia de este amplicador es variable con la frecuencia, es decir queactúa como un ltro pasa bajo con el objetivo de disminuir la ganancia parael ruido de alta frecuencia. Es importante constatar que la ganancia no varíedemasiado para el rango de frecuencia de interés(0,1 Hz a 40 Hz).

La máxima variación que se puede tener en la ganancia a lo largo del rangode interés es de 0,03 %, valor aceptable ya que el porcentaje de error debido alruido generado por los propios operacionales es mayor, además el ltrado queveremos más adelante también provoca una mayor atenuación de ganancia queeste.

P203 es un preset, cuyo valor de resistencia puede variarse de 0 a 20 kΩ, setomara Rx = P203 + R222 = P203 + 1kΩ. La frecuencia de corte varía muypoco al variar P203, aproximadamente de 1,592 kHz a 1,616 kHz. La ecuación7.3 (ver desarrollo en apéndice A.1.2) indica la frecuencia de corte en funciónde Rx y la gura 7.9 muestra la evolución de la misma al variar Rx. Al crecerla frecuencia, la ganancia del amplicador tiende a la unidad, de aquí se des-prende que conviene establecer la ganancia variable en un alto valor para quela relación de ganancias sea mayor.

fc =Rx + 105

2× π ×√

2× 10−3 ×Rx + 102(7.3)

62

Figura 7.9.: Frecuencia de corte en función de Rx.

Ahora considerando la ganancia constante en el ancho de banda útil, el valorde la misma se calcula como Rvar+101KΩ

Rvar+1KΩ, donde Rvar es el valor de resistencia

del preset P203. La variación de resistencia del mismo, permite obtener ganan-cias con valores desde aproximadamente 6 hasta 100. Considerando el sistematotal, la ganancia puede variar de 1187 a 19776.

Se realizaron pruebas para elegir un correcto valor de ganancia para estaetapa en el laboratorio. En caso de elegir una ganancia muy baja, se perde-ría calidad de la señal en la conversión A/D. Si se elige una ganancia muyalta, podrían ocurrir saturaciones de la señal debido a los artefactos, ruidos ointerferencias (ver 3.5 y 5 respectivamente). La ganancia elegida fue de apro-ximadamente 17800, ésta podría variarse en caso de cambiar a un ambientemás o menos ruidoso, o incluso al cambiar de usuario, ya que la intensidadde los artefactos puede cambiar de manera importante según la persona. Laecuación de la transferencia de esta etapa se muestra en 7.4 (ver desarrollo enapéndice A.1.2) y en la gura 7.10 el diagrama de Bode correspondiente parala ganancia elegida.

VoVi

=10−4 × S +

(1 + 105

Rx

)10−4 × S + 1

(7.4)

63

Figura 7.10.: Diagrama de Bode.

A continuación de la etapa de amplicación, se encuentra un capacitor, elcual es del mismo valor, que el utilizado para bloquear la tensión de polari-zación de los electrodos, sin embargo el principal objetivo de este capacitores bloquear la tensión de oset que puede contener la señal, este capacitor setendrá en cuenta en la transferencia de la siguiente etapa.

7.4.5. Filtrado antialiasing

El ltro antialiasing es la última etapa antes de que la señal sea enviada almicrocontrolador para ser digitalizada. Todos los sistemas que cuentan con una

64

etapa de digitalización requieren de un ltro el cual se presenta antes de quela señal ingrese al conversor A/D. La función del ltro es reducir el ancho debanda de la señal, de manera de limitar las frecuencias de la señal a la mitadde la frecuencia de muestreo del microcontrolador. De esta manera, según elteorema de Nyquist, se evita el llamado aliasing, el cual evita que la señalpueda ser recuperada luego de ser digitalizada.

En la realidad no es posible eliminar completamente las señales a partir dela frecuencia deseada pero alcanza con atenuarlas lo suciente para que el alia-sing no afecte a la señal de forma signicativa. Una atenuación de 40 dB(100veces) para las frecuencias superiores a la mitad de la frecuencia de muestreoes considerada sucientemente buena para evitar el aliasing.

Figura 7.11.: Esquema ltro anti aliasing.

La transferencia determinada por el ltro se observa en la ecuación 7.5 (verdesarrollo en apéndice A.1.3).

VoVi

=0,1m× s2 + 13,195× s

0,1089n× s4 + 1,65µ× s3 + 1,7082m× s2 + 0,81931× s+ 1(7.5)

El diagrama de Bode de la transferencia se observa en la gura 7.12.

65

Figura 7.12.: Diagrama de Bode del ltro antialiasing.

La frecuencia de corte fc del ltro es 75 Hz y la ganancia en la banda depaso es de 16. Esta ganancia, multiplicada por la ganancia que otorgan losamplicadores de las etapas anteriores, dene la ganancia total del sistema.

Se estudió la caída en decibelios a lo largo del espectro de frecuencias deter-minando que aproximadamente a la frecuencia de 490 Hz se produce una caídade 40 dB. Considerando lo anterior y el teorema de Nyquist, si se deseara evitarel aliasing se necesitaría una frecuencia de muestreo para el microcontroladormayor o igual a 2 × 490Hz = 980 Hz. Sin embargo, la banda de interés seextiende hasta los 40 Hz y posteriormente se realiza un ltrado digital, el cualprácticamente elimina las señales cuyas frecuencias estan por encima de los50 Hz(la atenuación a los 50 Hz es de 40 dB), de manera que podría aceptarsealiasing por arriba de esta frecuencia, por lo tanto la frecuencia de muestreomínima sería 540 Hz.

66

7.4.6. Circuito de pierna derecha

El circuito de pierna derecha lleva su nombre por razones históricas, ya queen sus inicios, se usaba únicamente en los electrocardiogramas, en los cuales seposicionaba un tercer electrodo (conectado al circuito DRL)en la pierna dere-cha, lo más lejos posible del corazón.

Desde los orígenes del EEG se utilizó otro electrodo además de los electrodosde medida, el cual inyectaba una señal al usuario. Este electrodo se utilizabasimplemente como referencia para la tensión de los otros, ya que en caso deno contar con una referencia, las corrientes de polarización en el amplicador,pueden provocar que la tensión del sensor crezca o disminuya respecto a latierra análoga del amplicador. Esto provocaría una tensión de modo comúnque podría superar los niveles que el amplicador puede rechazar.

Con el paso de los años se desarrolló un nuevo electrodo de referencia, con ladiferencia que este electrodo se conectaba a un circuito (DRL) de manera quela referencia se convierte en una referencia activa. Este circuito tiene la granventaja que aumenta signicativamente el rechazo al modo común, midiendola señal de modo común obtenida por el amplicador de instrumentación, in-virtiéndola y realimentándola hacia el cuerpo del paciente, de esta manera segeneran corrientes de igual magnitud a las generadas por el modo común, perode sentido opuesto. Se podría decir que el procedimiento es muy similar a losprocesos de control, donde se mide la salida y se realimenta negativamente pa-ra variar la entrada controlando así la salida. La mayor componente del modocomún se debe a la red eléctrica, utilizando esta técnica, el rechazo al modocomún para la frecuencia de la red eléctrica (50 Hz) puede ser incrementadodesde 30 dB a 50 dB.

Otro método que puede utilizarse para referenciar las tensiones de los elec-trodos, es conectar una gran resistencia (1 MΩ) entre la línea de los sensores yla tierra. Este método es más sencillo que utilizar el circuito DRL pero tienela desventaja de disminuir la impedancia de entrada, además de que no reducela señal de modo común.

67

Figura 7.13.: Esquema circuito DRL.

En este proyecto (que se trabaja con fuente única), la referencia del circuitoDRL es la tensión VGND, de esta manera, la realimentación negativa del modocomún está referenciada a esta tensión (montada a VGND), logrando que lareferencia de los potenciales medidos por los electrodos también sea VGND. Sepueden obtener tensiones negativas respecto a la nueva referencia, lo que (enel caso de fuente única) no se podría lograr si la referencia fuera la tierra realdel circuito.

Otro aspecto importante del DRL es que el mismo se encarga de limitar lacorriente que circula por el usuario en caso de que este se conecte a la tierra delcircuito por alguna razón, en ese caso la corriente circulara a través de la resis-tencia R213(500 kΩ) y esta resistencia es la encargada de limitar la corriente.Tampoco se puede aumentar demasiado la resistencia para obtener corrientesmás pequeñas en caso de mal funcionamiento debido a que valores de R213

68

demasiados altos (mayores a 1 MΩ) afectarían la correcta realimentación ne-gativa del modo común.

Según la AAMI (Association for the Advancement of Medical Instrumenta-tion), la máxima corriente que puede circular por el paciente es de 10µA, sinembargo los estándares de la IEC (International Electrotechnical Commission)se admite una corriente de hasta 50µA. Para el valor de R213 utilizado en elproyecto openEEG (200 kΩ) esta corriente máxima es de 5V

200 kΩ= 25µA. Se

podría cambiar este valor de resistencia por 500 kΩ para cumplir con el re-querimiento más restrictivo, pero este cambio tiene la desventaja de disminuirel rango dinámico en la salida por la caída de tensión en la resistencia. Ba-jo operación normal circula una corriente de aproximadamente 1µA, lo queprovocaría una caída de 0,5 V (10% de la excursión total) en la resistencia de500 kΩ, por lo tanto se decidió mantener el valor de la resistencia en 200 kΩ yapegarnos a los estándares de la IEC.

Es importante asegurar la estabilidad del DRL y para lograr esto se utilizandistintos circuitos. En este caso se utiliza el método de compensación por polodominante, que fue propuesto por Winter y Webster en el 1983. Este métodoanaliza la ganancia en lazo abierto, obtenida a partir del modelo visto en elcapítulo 5. Como los valores del modelo son muy variables, se toma un casodesfavorable para asegurar la estabilidad y se obtiene la transferencia total enfunción de la del circuito DRL. Luego como la ganancia en continua (sin teneren cuenta el DRL) es de 0 dB se construye el DRL como un integrador, demanera que su ganancia decaiga a 0 dB antes de la frecuencia del primer polode la transferencia, para así lograr la estabilidad.

7.4.7. Elección de los amplicadores operacionales

Los amplicadores operacionales y de instrumentación que fueron utilizadospara este proyecto son el INA114BP y el TCL277CP, los cuales eran recomen-dados en openEEG. Se estudiaron las características de ambos, se vericó quecumplieran con los requisitos necesarios para la aplicación y luego se compara-ron con otros de similares características para comprobar que su elección fueracorrecta.

Las tres características más importantes del amplicador operacional en estecaso son: la impedancia de entrada, la relación de rechazo al modo común(CMRR) y el ruido inducido por el propio amplicador.

69

7.4.7.1. Características de los integrados

Como se comentó en el capítulo 4 la primera etapa del sistema de adquisi-ción de la señal, está compuesta por los electrodos activos, los cuales utilizanel integrado TLC277CP. En esta primera etapa también es importante teneren cuenta la tensión de oset de salida, ya que la misma se podría agregar ala tensión de polarización de los electrodos y saturar el operacional.

El integrado que utilizaba el diseño original de los electrodos activos imple-mentados, era un TLC272, el cambio de integrado se basó en que no había unadiferencia de precio signicativa (20 $UYU) y el TLC277CP otorgaba una ten-sión de oset a la salida 20 veces menor, lo que disminuye considerablementela probabilidad de saturación del amplicador de instrumentación. Tambiénse utiliza este amplicador operacional en las demás partes del circuito querequieren dicho integrado, sin contar la primera etapa de amplicación quecomo ya se explicó, utiliza el amplicador de instrumentación INA114BP.

El CMRR de este amplicador no es de vital importancia debido a que elmismo no es utilizado en la etapa de entrada como amplicador diferencial,sino que se utiliza como seguidor de tensión en los electrodos activos y luego enetapas de amplicación secundarias. De todas maneras posee un buen CMRRde 65 dB mínimo y 80 dB típico. Sin embargo el CMRR del amplicador deinstrumentación es crítico, por razones ya explicadas. La tensión de modo co-mún más signicativa es la provocada por la red de alimentación, los valoresexpresados a continuación serán en RMS2.

Segun los valores de capacitancias parásitas e impedancias tomados por lamayoría de las literaturas leídas [11] suponen que la tensión de modo comúnen caso de utilizar baterías puede llegar a valores de hasta 100 mV (si utili-zan alimentación de la red pueden llegar a 500 mV), este resultado se obtienesuponiendo el peor valor de Cb (5 pF) y considerando también desfavorable-mente, que toda la tensión de la red de alimentación cae en Cb, lo que generauna corriente de aproximadamente 0,35µA. Luego considerando una elevadaimpedancia piel-electrodo DRL (300 kΩ), la tensión Vcm = 0,35µA× 300 kΩ =105 mV.

También tenemos que tener en cuenta el nivel de rechazo generado por elDRL, que como se dijo anteriormente, es de por lo menos 30 dB, haciendo que

2RMS (Root Mean Square) o valor ecaz, es una medida estadística de una magnitud

70

el valor de Vcm sea de aproximadamente 3,32 mV. Ahora a partir de la ecuaciónCMRR ≥ 20×log( Vcm

αV d) (ver desarrollo en apéndice A.1.4), donde α representa

el porcentaje de error admisible respecto a la señal de interés (Vd) provocadopor la tensión de modo común, se obtiene CMRR ≥ 93,4 dB. Se eligió admitirun error del 1% para las señales más pequeñas, que son de 10µV de ampli-tud y valor RMS de 7,07µV (considerando las señales como sinusoidales). Elintegrado INA114BP presenta como valor mínimo, a la ganancia utilizada, unCMRR de 96 dB, que cumple con lo calculado anteriormente.

7.5. Ruido provocado por los Amplicadores

[15]

El ruido que genera el integrado TLC es de vital importancia cuando estees utilizado como seguidor de tensión en los electrodos activos. Es necesarioasegurarse que el ruido que generen sea por lo menos 20 veces menor que lasseñales de interés más pequeñas (10µV de amplitud). Los amplicadores ope-racionales inducen principalmente un ruido que es la mezcla de ruido rosa conruido blanco. El primero predomina a bajas frecuencias y el segundo (que estápresente en todo el espectro) se hace notar cuando el ruido rosa decae. Esteruido se modela mediante una fuente de tensión en serie con una de las entra-das del amplicador y dos fuentes de corriente en paralelo a las entradas comose muestra en la gura 7.14.

71

Figura 7.14.: Modelo de ruido de un amplicador operacional.

Siempre se supone las densidades espectrales de potencia deinn(t) e inp(t) soniguales, considerando que el ancho de banda de la señal va desde fL a fH , elruido provocado por el amplicador en su entrada se calcula según la ecuación7.6 (ver desarrollo en apéndice A.1.5) , donde Vwn e Iwn son las densidadesespectrales de los ruidos de tensión y corriente provocados por el ruido blancodel amplicador. fcvy fci son las frecuencias en las que el ruido blanco comienzaa predominar sobre el ruido rosa. Estos valores pueden verse en la gura 7.15y 7.16, Rp y Rn son las impedancias vistas desde las patas positiva y negativarespectivamente.

√V 2wn × (fcv × Ln(

fHfL

) + fH − fL) + (R2n +R2

p)× I2wn × (fci × Ln(

fHfL

) + fH − fL)

(7.6)

72

Figura 7.15.: Gráco del espectro del ruido producto de la tensión.

Figura 7.16.: Gráco del espectro del ruido producto de la corriente.

Los demás ruidos, como el ruido térmico, el ruido de disparo y el ruido deuctuación son despreciables en nuestro caso por las pequeñas corrientes y elbajo valor de resistencias equivalentes.

A partir de la ecuación 7.6, de la hoja de datos del integrado y teniendoen cuenta el ruido sólo para las frecuencias de interés (ya que la señal nalse encuentra ltrada en ese rango) que van desde 0,1 Hz hasta 40 Hz se puedecalcular el ruido generado.

El ruido de corriente in para el TLC277CP es nulo, mientras que, según losvalores típicos, Vwn = 25nV√

Hzy fcv = 15 Hz, de donde se obtiene VinRMS =

73

0,285µV.

En el aspecto del ruido generado, también se debe tener en cuenta el ruidoprovocado por el integrado INA114BP, ya que como se dijo anteriormente elTLC277CP actúa como seguidor de tensión en el electrodo activo, sin ampli-car la magnitud de la señal. De la hoja de datos del amplicador de instrumen-tación, se obtiene como valores típicos Vwn = 15nV√

Hzy fcv = 10 Hz, Iwn = 25 ρA√

Hzy fci = 10 Hz. Ahora Rp y Rn se obtienen como paralelos de la impedancia deentrada del INA y de salida del TLC, por lo que su valor no supera los 100Ω,utilizando el segundo término de la ecuación 7.6 se obtiene un valor RMS delruido de corriente de 0,28 nV, despreciable frente a los 0,15µV obtenidos porel ruido de tensión.

Finalmente, se desea saber el ruido RMS total generado por ambos integra-dos, éste no se calcula como la suma de cada uno. Al no estar correlacionadosambos ruidos, el valor RMS se calcula como se indica en la ecuación 7.7 (verdesarrollo en apéndice A.1.5), donde Vitn es el valor RMS del ruido total, Vin1

es el valor RMS del ruido generado por el integrado TLC y Vin2 es el valorRMS del ruido generado por el integrado INA. Se desprecia el ruido genera-do por los integrados que siguen al amplicador de instrumentación, ya queuna vez amplicada la señal, el ruido agregado por los amplicadores se puededespreciar.

Vitn =√V 2in1 + V 2

in2 = 0,32µV (7.7)

A partir de los resultados se comprueba que el valor RMS del ruido generadopor los integrados es aproximadamente 30 veces menor que el valor RMS de laseñal de interés más pequeña (10µV). Luego con el valor del ruido hallado, sepuede denir la relación señal-ruido3 (para las señales más pequeñas), la cualtienen un valor de26,9 dB.

La última característica de gran importancia es la impedancia de entrada,para el TLC es de 1012 Ω, ya se ha mencionado porque una alta impedancia deentrada tiene considerables ventajas. Este valor de impedancia de entrada, seencuentra dentro de los valores más altos de los operacionales, y es importante

3La relación señal-ruido se utiliza para comparar el nivel de la señal con el nivel del ruido,y se calcula como 20× log( Senal

Ruido )

74

que la impedancia de entrada del TLC sea más alta que la del INA, ya que elprimero se encuentra en los electrodos y una gran impedancia en primera etapadisminuye en gran parte el impacto del ruido. No signica que la impedanciadel amplicador de instrumentación pueda ser baja, pero se encuentra muyaliviada ya que el mismo ve desde sus entradas las impedancias de salida delos TLC, que son bajas y casi iguales. La impedancia de entrada del INA esde 1010 Ω, considerablemente alta aunque sea 100 veces más chica que la de laprimera etapa.

Los integrados utilizados cumplen con los requerimientos necesarios para laaplicación, pero no de forma demasiado holgada, se estudió la posibilidad deutilizar otros, pero no se encontraron integrados de este costo o menor, quecumplan con las características necesarias.

7.6. Respuesta en frecuencia de la placa

amplicadora

En las guras 7.17 y 7.18 se observa la respuesta en frecuencia teórica enmodulo y fase respectivamente de la placa de amplicación construida.

La fase, que en la gráca se gráca de 1 Hz hasta 50 Hz, es seudolineal en elrango de interés de 0,1 a 40 Hz. El máximo retardo provocado por la fase esde 3,75 ms, aproximadamente un 2% del período, esto se da para la frecuenciade 40 Hz.

Figura 7.17.: Modulo de la transferencia.

75

Figura 7.18.: Fase de la transferencia.

La transferencia mostrada es la transferencia ideal, esta puede verse modi-cada debido a los errores en los valores de los componentes. Tomando encuenta la tolerancia del 5% de los capacitores utilizados y las tolerancias del1% de las resistencias, la frecuencia de corte puede variar entre 55, 2 Hz y111,9 Hz aproximadamente, mientras que la atenuación a 512 Hz puede variarentre 101,4 y 120,4 veces aproximadamente.

Es importante resaltar que estos valores para los extremos (frecuencia decorte y atenuación a 512 Hz) no se dan al mismo tiempo debido a la compleji-dad del sistema y la cantidad de componentes involucrados, es decir que unamenor frecuencia de corte, no signica siempre, una mayor atenuación a los512 Hz. Dichos valores se obtuvieron realizando una simulación con 100 de lospeores casos (el resultado se observa en la gura 7.19 ). Considerando estos 100casos, se mencionan como detalles importantes para los casos más críticos, queen la banda de paso, la mayor atenuación que se puede obtener es de 1,3 dB,por lo que la banda de paso puede considerarse plana, y las atenuaciones a512 Hz nunca son menores a 40 dB, por lo que no se corre riesgo de distorsiónsignicante por aliasing.

76

Figura 7.19.: Análisis Montecarlo de la transferencia de la placa.

77

8. Etapa Digital

La etapa digital cumple la función de digitalizar la señal analógica y enviarlaa la computadora. La comunicación entre el microcontrolador y la computadorase realiza mediante el protocolo BlueTooth.

8.1. Conversor analógico-digital

La mayoría de los microcontroladores poseen conversores A/D de 8 bits o10 bits. En caso de utilizar un conversor de 8 bits, un LSB equivaldría aproxi-madamente a 19,5µV, dividiendo por la ganancia del amplicador (17800), seobtiene una apreciación cercana a 1,1µV, la cual no es aceptable en relacióna la amplitud de las señales EEG.

Al utilizar 10 bits, siguiendo el razonamiento anterior se obtiene una apre-ciación de aproximadamente 0,27µV, por lo tanto se optó por la resolución de10 bits. También hay conversores dedicados que ofrecen 12 bits o incluso 14bits o 16 bits, pero estos son muy costosos y tal precisión no era necesaria,ya que como se vio anteriormente, el ruido generado por los amplicadoresoperacionales es de aproximadamente0,32µVRMS teóricamente.

Otra característica vital del conversor es que el mismo posea una frecuenciade muestreo adecuada, ya que para evitar el ruido de aliasing en la banda deinterés, como se explica en A.1.3, se necesita una frecuencia de por lo me-nos 540 Hz. Sin embargo es más fácil obtener menos error en la frecuencia demuestreo del microcontrolador al utilizar potencias de 2. Además al aumentarla frecuencia de muestreo podría extenderse el rango de frecuencias de interésen el futuro considerando otro tipo de aplicaciones. Por lo tanto se eligió unafrecuencia de muestreo de 1024 Hz. Ahora como el dispositivo incluye cuatrocanales, el conversor debe poder convertir 4096 muestras por segundo o más.

78

8.2. Comunicación

microcontrolador-computadora

Para comenzar, fue necesario seleccionar el tipo de comunicación entre elmicrocontrolador y la computadora.

En un principio se decidió utilizar comunicación serial por cable. Una vezque el sistema estaba funcionando correctamente, se sustituyó el cable porcomunicación BlueTooth, ya que se consideró que era la forma más sencillay económica para establecer una comunicación inalámbrica entre el micro-controlador y la computadora, de manera de lograr comodidad al usuario yportabilidad al sistema. Para lograr esto se utilizaron módulos BlueTooth queemplean el protocolo SPP (Serial Port Prole). Este protocolo, lee datos se-riales y los transmite por BlueTooth, y viceversa.

Se evaluaron dos características para la elección de los módulos, costo y ve-locidad de transmisión, siendo la característica más importante esta últimadebido a que debe poder enviar información a una velocidad mayor de la quela recibe. Para poder denir la velocidad necesaria, hubo que decidir cómo seenviarían las muestras de los canales para saber exactamente cuántos bytespor segundo debían ser enviados, por lo que se diseñó una trama de envío paralos datos. Como se deseaba poder elegir la cantidad de canales, la trama deenvío debía ser variable en función de la cantidad de canales utilizados. Tam-bién debía tenerse en cuenta que al utilizar 10 bits por cada muestra, cada unade ellas constaría de 1,25 bytes por lo menos, para simplicar el tratamientode datos se decidió utilizar 2 bytes por muestra. En caso de que la velocidadde transmisión no fuese suciente, se podría modicar este aspecto. Ademáscada byte enviado, incluye un bit de inicio, uno de parada y un bit opcionalde paridad para control de errores.

Como se utiliza comunicación mediante protocolo BlueTooth, y el mismo yacuenta con un control de errores, no era indispensable el bit de paridad en lacomunicación serial, ya que el cableado del microcontrolador al módulo Blue-Tooth es mínimo, la probabilidad de error por ruidos que ingresen con el cabletambién lo es, por lo que se decidió luego este aspecto.

La trama nal utilizada se muestra en la gura 8.1, en esta gura se observaque a pesar de que el sistema posee tan solo 4 canales, la trama es adaptablepara n canales. Se utilizó FFFF como bytes de inicio ya que las muestras son

79

de 10 bits, al presentarla con 2 bytes, el máximo valor de muestra es 03FF.

Figura 8.1.: Trama de comunicación microcontrolador-PC.

La primera trama diseñada incluía un byte entre los bytes de inicio y losbytes de las muestras. Este byte indicaba la cantidad de canales utilizadospara que la computadora lea la trama en función de este valor, pero como elusuario elije el número de canales desde la computadora, no era necesario estebyte, por lo que se eliminó de la trama. Una vez denida la trama se puededeterminar cuántos bytes presenta. Al utilizar 4 canales, los datos y el inicioocuparían 10 bytes, sumando 2 bits (de inicio y de parada) por cada byte,obtenemos un total de 12,5 bytes por trama enviada.

Tomando en cuenta que la frecuencia de muestreo es de 1024 Hz y conside-rando que el tiempo que demora el microcontrolador en muestrear es de 65µs(tiempo de muestreo observado en la mayoría de los microcontroladores estu-diados), se debe enviar 12,8 kB en 1 − 4 × 65µs × 1024 = 0,73376 s, lo quedetermina que la velocidad de transmisión debe ser mayor a 17,45 kB/s, equiva-lentes a 139,6kb/s.

En caso de incluir el bit de paridad para control de errores sería necesariouna velocidad mayor a 149,9kb/s. En caso de elegir un microcontrolador conun tiempo de muestreo mayor, se debería reajustar las velocidades necesarias.Estas velocidades no son exigentes para las velocidades de comunicación serial,por lo que se decidió implementar el bit de paridad, a pesar de la mínimaprobabilidad de error de transmisión con la que se cuenta.

8.3. Elección de módulos para comunicación

BlueTooth

Se utilizó el módulo HC-06 (ver hoja de datos en anexo B.6) como escla-vo, para ser conectado al microcontrolador, el cual cuenta con velocidades de

80

transmisión de hasta 1382,4 kb/s, además de ser el más económico entre los mó-dulos BlueTooth estudiados.

Las velocidades de transmisión que permite dicho módulo, y que son útilespara el proyecto, eran 115,2 kb/s, 230,4 kb/s, 460,8 kb/s, 921,6 kb/s y 1382,4 kb/s. Seeligió la velocidad de 230,4 kb/s, a partir de lo calculado anteriormente. Con lavelocidad elegida se podrían agregar 1 canal más, en caso de necesitar mayorvelocidad es posible incrementar la velocidad del módulo.

También fue necesario decidir la recepción de BlueTooth por parte de lacomputadora. Para poder utilizar comunicación inalámbrica en cualquier compu-tadora (aún si esta no contara con BlueTooth integrado), se utilizó un móduloBlueTooth también para la computadora. En principio se intentó utilizar unBlueTooth USB, ya que era la opción más económica y más utilizada general-mente, pero esto trajo problemas de compatibilidad con el módulo HC-06, porlo que se optó por utilizar el módulo HC-05, de la misma familia que el anterior,con la diferencia que cuenta con un rmware distinto el cual permite utilizarel módulo tanto como esclavo o como maestro. Este cambio tuvo la desven-taja de tener que agregar un módulo FTDI, el cual funciona como conversorUSB-Serial para conectar el HC-05 a la computadora. Por otra lado tiene elbenecio de que una vez congurado para conectarse con el HC-06, sólo esnecesario prenderlo y la conexión se realiza automáticamente. Otra ventaja esque no fue necesario modicar el código en el programa de la computadora res-pecto a la recepción y envío de datos (la cual era serial), ya que la computadorave al FTDI como un puerto COM, el cual es utilizado para la comunicaciónserial directamente. Dicho de otra forma, la computadora ve la transmisión dedatos de igual forma que una transmisión serial común.

8.4. Elección del microcontrolador

Para la etapa digital se eligió una placa Arduino con un microcontroladorATmega2560 (ver hoja de datos en anexo B.1).

En la sección 7.3, donde fue tratado el tema de la alimentación del circuito,se menciona la ventaja de utilizar una placa Arduino, la cual ya cuenta conun integrado LM7805 para lograr la tensión estable de 5 V. Además de esto,la misma brinda otras ventajas. Se incluyen pineras, desde las cuales se puedeacceder a las terminales del microcontrolador, por lo que no es necesaria laconstrucción de un hardware alrededor del microcontrolador para utilizar sus

81

terminales. Las placas son, en su mayoría, de reducido tamaño, lo que es im-portante para lograr el objetivo de portabilidad. Lograr tamaños tan reducidossería sumamente difícil de lograr sin las herramientas necesarias. Finalmente,una ventaja muy importante que otorga Arduino es que incluye un móduloFTDI (o un ATmega16u2 en algunos casos), el cual se utiliza para progra-mar el microcontrolador, de manera que la placa se conecta directamente ala computadora, sin la necesidad de programadores intermedios, además depoder comunicarse por serial utilizando el FTDI. [16]

A la hora de elegir la placa, se tuvo en cuenta el microcontrolador que incluyecada una. La mayoría utilizan los microcontroladores ATmega8, ATmega128,ATmega168, ATmega328, ATmega32u4 o ATmega2560. También se tuvo encuenta un modelo que utilizaba un microcontrolador ARM Cortex-M3, el cualofrecía una velocidad mucho mayor, pero se descartó porque su precio era máselevado que el resto, y como la idea era realizar el procesamiento en la compu-tadora y utilizar el microcontrolador solamente para digitalizar la señal, seconsideró que no eran necesarias tan altas velocidades de procesamiento porparte de la placa digital. Además el ARM Cortex-M3 funciona con 3,3 V sola-mente, lo que disminuye el rango de operación.

Excepto el microcontrolador ATmega32u4, que no presentaba frecuenciasde muestreo adecuadas, todos los microcontroladores cumplían con los reque-rimientos necesarios, se eligió el ATmega2560 debido a que ya se conocía laarquitectura del mismo y se contaba con uno. Su tiempo de conversión mínimoal realizar conversiones de 10 bits es de 65 µs, tiempo utilizado anteriormentepara el calculo de la velocidad necesaria. Como ya se explicó anteriormente,con la velocidad elegida, podría agregarse un canal más. Para agregar más ca-nales hay dos opciones, aumentar la velocidad de envío, o cambiar el algoritmodel microcontrolador (gura 8.2) de manera de que las conversiones se realicenparalelamente al envío de datos y así sólo se perdería tiempo por conversionesdurante la primera conversión. Esto no se implementó así debido a que no eranecesario en nuestro caso, y el algoritmo se volvería más complejo.

8.5. Algoritmo del microcontrolador para

conversión A/D y envío de datos

En la gura 8.2 se muestra el diagrama de ujo del algoritmo utilizado enel microcontrolador, las guras 8.3 y 8.4 muestran las acciones realizadas por

82

las interrupciones del timer y del ADC respectivamente. La interrupción deltimer se ejecuta cada 1 ms y la interrupción del ADC se ejecuta cada vez quese completa una conversión.

A continuación se explican los signicados de las variables para lograr unentendimiento más sencillo del algoritmo:

Conversión: variable booleana que se convierte en verdadera cuando setermina una conversión.

ADC: resultado de la conversión del microcontrolador.

Canal actual: próximo canal que se muestreará para la conversión

Cantidad canales: cantidad total de canales muestreados.

Envió datos: variable booleana que indica si se debe muestrear y mandardatos.

Datos: datos recibidos por la comunicación serial.

Figura 8.2.: Diagrama de ujo del algoritmo del microcontrolador ATmega.

83

Figura 8.3.: Interrupción del timer.

Figura 8.4.: Interrupción ADC.

84

9. Software Monitor de señalescerebrales

Su labor es la de recibir, procesar, desplegar en pantalla y almacenar la infor-mación que proviene del microcontrolador. El software nos permite visualizarlas señales de EEG que provienen de cada canal, así como también señalesde interés generadas a partir del estudio de dichos datos con el n de tomaracciones de control para una aplicación.

9.1. Implementación del software

El software se implemento en el lenguaje de programación Java. Los motivosde dicha elección se debieron a que es un lenguaje multiplataforma, no soloa nivel de sistemas operativos, sino que también cabe la posibilidad de correrdicho programa en dispositivos portátiles como tablets o smartphones. El mis-mo pose un buen soporte en cuanto a interfaz gráca lo cual es necesario paradesplegar grácos en tiempo real.

El programa se dividió es distintos sub-procesos. Los principales son (Se-rial) se encarga de recibir la información a través de un puerto COM virtual,(Buer), almacena los datos recibidos, (Proceso), encargado del tratamien-to de las señales y (Servidor), responsable de enviar los comándos de controla la aplicación.

Figura 9.1.: Diagrama general del software.

85

9.2. Serial

Al inicio de la conexión, se buscan todo los puertos COM presentes, luegoes el usuario quien indica en cuál de ellos se encuentra el dispositivo de ad-quisición, una vez indicado esto el mismo realiza la conexión de ser posible.Si se logra con éxito, este proceso pasa a quedar a la espera de datos. Unavez que se obtiene uno o un grupo de estos, se envían al proceso Buer. Sibien el microcontrolador envía los datos de las señales capturadas de formaregular, dado que se trabaja sobre un COM virtual, los datos llegan de maneraasincrónica, por lo que el proceso Serial actúa como un evento.

9.3. Buer

Los datos llegan en un formato de trama especico, por lo cual se puedeseparar la información de los diferentes canales (esta versión permite cómomáximo 4 canales), este proceso se encarga de almacenar las señales por sepa-rado y descartar datos incorrectos.

9.4. Proceso

Es el eslabón más importante del software. Toma las señales almacenadaspor Buer, las acondiciona y procesa para así poder tomar decisiones de con-trol.

Se puede sub dividir en etapa de acondicionamiento, visualización , detecciónde ritmos y generación de señales de control (implementación de BCI basadaen ritmos cerebrales).

A continuación se explicará cada sub etapa del bloque proceso.

9.4.1. Acondicionador

Esta etapa tiene dos funciones, ltrar la señal y diezmar. En primer lugarporque el sistema de adquisición tiene un ltro antialiasing con una frecuen-cia de corte de 75 Hz. Las señales de interés se encuentran en el rango de 0,1a 40 Hz. En segundo lugar, como el sistema muestrea a 1024 Hz, y es innece-sario contar con tantas muestras para representar y analizar la señal de interés.

86

9.4.1.1. Filtro 40 Hz

Debido a que el análisis se realiza en tiempo real es necesario minimizar loscálculos para que no se produzcan grandes latencias, por esto, los ltros uti-lizados en el sistema son de respuesta impulsiva innita (IIR), dado que conestos se obtiene resultados similares a los ltros de respuesta impulsiva nita(FIR) de mayor orden.

Los ltros presentes en el software fueron diseñados en el programa Scicos-Lab, en el apéndice A.4 se encuentran los script para su construcción.

Es necesario ltrar la señal por un ltro pasa banda para obtener las señalesde interés, pero debido a que se ltra la continua de forma analógica solo bastaun ltro pasa bajo con frecuencia de corte de 40 Hz para obtener la informa-ción, la única componente de continua que presenta la señal es la tensión dereferencia VGND la cual es constante y conocida.

Consideraciones para el diseñoSe quiere que la respuesta del mismo sea plana en la banda de paso y de

fase lineal para no distorsionar la señal (retardo para diferentes frecuencias),pero se desea la mayor atenuación posible a los 50 Hz con el n de disminuirla interferencia inducida por la red eléctrica.

DiseñoEn función a estas consideraciones se decidió utilizar un ltro Chebyshev del

tipo II.

Estos ltros presentan una respuesta plana en la banda de paso, importan-te para evitar la distorsión en amplitud, un rizado constante en la banda derechazo que no afecta a la señal de interés y una fase seudolineal, que generauna deformación tolerable (2% de error con respecto al defasaje) en la señal,ya sea para su análisis, como para su visualización.

ConstrucciónSe construyó un ltro Chebyshev del tipo II de orden 10, con frecuencia de

corte de 40 Hz que opera a 1024 muestras por segundo. La atenuación obtenidaes de 40 dB a los 50 Hz.

Se probo con diferentes tonos puros la respuesta del ltro, con el n de com-probar su correcto funcionamiento dado que podrían existir errores(deformaciones

87

de la transferencia teórica obtenida) por errores de cálculo debido a la sensi-bilidad que presenta modicar el largo de palabra de los coecientes teóricos.

A modo comparativo se construyó el mimo ltro pero FIR, fue necesario queel orden del mismo fuera cercano a 300 para lograr la misma frecuencia decorte y atenuación a 50 Hz.

Figura 9.2.: Respuesta en frecuencia del ltro de 40 Hz.

El retardo promedio debido a la seudolinealidad1 de su fase es de 11, 5 ms.

La mayor desviación de fase, respecto de la recta de aproximación de fase,ocurre en 40 Hz, generando un retardo de aproximadamente 6 ms, el cual noes apreciable para el ojo humano.

9.4.2. Diezmado

Según el teorema de muestreo de Nyquist-Shannon, se debe muestrear aldoble de la frecuencia más alta que presenta la señal, por ende la mínima can-tidad de muestras necesarias para analizar la señal de interés es 80Hz. Para

1Denimos como fase seudolineal, a toda fase que para cada frecuencia dentro del rangode interés, no presente un desvió respecto al retardo lineal mayor al 20% del periodo dedichas frecuencias.

88

obtener esta frecuencia de muestreo sería necesario diezmar por 12,8, esto noes posible sin recurrir a técnicas mixtas de interpolado y diezmado.

Nos limitaremos a utilizar coecientes enteros para diezmar. No es necesarioutilizar un ltro anti-aliasing antes del diezmado porque la señal ya fue ltraday solo presenta componentes menores a 40 Hz.

9.4.3. Detección de ritmos y generación de señales decontrol

Los sistemas que utilizan los potenciales bioeléctricos cerebrales como seña-les de control son denominados Interfaz Cerebro Computadora (BCI).

9.4.3.1. Interfaz cerebro computadora - BCI

La función de las BCI es establecer una comunicación entre el cerebro y unamáquina, sin la necesidad de acciones motoras. Esto permite llevar a cabo ope-raciones sencillas.

Las BCI generalmente utilizan potenciales espontáneos para relacionar dis-tintos estados mentales de un sujeto. Lo más común es utilizar los ritmoscerebrales, en particular el ritmo alfa occipital dado que es el que presenta lamayor potencia y es más simple de bloquear/desbloquear a voluntad.

Básicamente se puede decir que una BCI se compone de 3 partes, la primerade ellas se encarga de capturar la señal de EEG. La segunda etapa detecta losestados mentales a partir de la información obtenida, genera señales de controla partir de los estados y retroalimenta al usuario a través de los grácos dedichas señales permitiéndole un mejor manejo de la interfaz. La última etapase encarga de ejecutar dichas órdenes, como por ejemplo prender y apagar unaluz.

89

Figura 9.3.: Esquema general BCI.

Del esquema general (g 9.3) se desprende que el sistema total desarrolladoen este proyecto integrado por el sistema de adquisición, el software Monitorde señales cerebrales encargado del tratamiento de señales y la aplicación decontrol, conforman una BCI.

Si bien todo el sistema completo es una BCI, por un tema de comodidady convención, nos referiremos como BCI a todo bloque encargado de detectarpatrones cerebrales y generar señales de control, por lo que un mismo sistemapuede estar constituido por varios de estos bloques.

BCI basadas en ritmos cerebralesSon las más utilizadas debido a su sencillez, ya en 1967 Edmond Dewan lo-

gro enviar información atreves de código Morse utilizando la sincronización ydesincronización del ritmo alfa.

Generalmente son interfaces on/o, utilizadas para controlar interruptoresen modo toggle, es decir, una activación los coloca en posición de encendido yla siguiente en posición de apagado. Debido a esto, la capacidad de transmisiónde datos (bits), depende directamente de la capacidad del usuario para activarla interfaz.

90

Figura 9.4.: BCI On/O.

Dado que se quiere diseñar una aplicación que sea controlada por ritmoscerebrales es necesario reconocer a los mismos a partir de la señal obtenida.

9.4.3.2. Reconocimiento de ritmos cerebrales

Se evaluaron dos tipos de técnicas para reconocer ritmos cerebrales, el aná-lisis en frecuencia y el análisis en potencia.

Análisis en frecuenciaComo ya se mencionó antes, los ritmos cerebrales posen características par-

ticulares, por ejemplo presentan su propia banda de frecuencias.

Una primera idea para poder identicar la presencia de los mismos fue rea-lizar un análisis en frecuencia, se utilizo la herramienta matemática transfor-mada de Fourier, para poder obtener así la distribución espectral de la señalde EEG. Dado que la distribución espectral de una captura entera carece deinterés, debido a que se pierde la resolución temporal, es decir se pierde toda lainformación relacionada a los instantes en que aparece o desaparece una com-ponente de frecuencia. Se optó por calcular la transformada de la señal para unnúmero nito de muestras. Esto equivale a tomar el registro original y dividir-los en tramos. De esta manera se puede observar como varia la distribución alo largo del tiempo para lograr determinar la presencia o ausencia de los ritmos.

El inconveniente que trajo dicha práctica es que truncar la señal se traduceen el enventanado2 de la señal. Al momento de calcular su transformada, seobtendrá la transformada de la señal convolucionada con la transformada de

2Denominamos enventanado al proceso de multiplicar una señal por una ventana.

91

la ventana.

[x(t)× w(t)]⇒ [X(f) ~W (f)]

El caso más sencillo es calcular la distribución espectral para un tiempoτ ,por ejemplo un segundo de captura. Esto es lo mismo que utilizar una ventanarectangular, el resultado obtenido es la transformada de la señal convolucio-nada con un sinc.

x(t)× [u(t)× u(τ − t)]⇒ X(f) ~ τ × sinc(πfτ)× e−j2πfτ2

Este factor es muy importante debido que altera signicativamente la distri-bución espectral. Para aclarar esto último pasaremos a mencionar un ejemplo.

Supongamos x(t) = cos(2πfot), si nuestra ventana temporal fuera innita elresultado de la transformada de la señal seria:

X(f) = δ(f−fo)+δ(f+fo)2

En cambio si tenemosy(t) = cos(2πfot)×[u(t− τ2)×u( τ

2−t)], que es lo mismo

que la señal anterior solo que con un enventanado rectangular de tiempo τ . Elresultado obtenido al realizar la transformada será:

Y (f) = τ×sinc(πτ(f−fo))+τ×sinc(πτ(f+fo))2

92

Figura 9.5.: Distorsión espectral sufrida por el enventanado de la señal.

Debido a que el sinc es una señal que nuca se extingue, ahora ya no tenemostoda la potencia de la señal concentrada en una frecuencia, sino que se dis-perso, mayoritariamente se encuentra alojada en el lóbulo principal del sinc,un 90% aproximadamente, el resto se encuentra distribuido en todo el espectro.

El ancho del lóbulo decrece a medida que τ aumenta, por lo tanto una ven-tana más grande, genera una representación de la distribución espectral másexacta. Es decir la densidad espectral de la señal sufre una menor distorsión.El caso limite se da para τ innito, en este caso el sinc aparece como un deltaque no afecta la transformada.

Existe una relación de compromiso en la elección del tiempo de enventanado.Mayor tiempo genera una representación más real, pero se pierde la resolucióntemporal, mejorar la resolución temporal disminuyendo τ genera perdida en la

93

resolución espectral.

Si bien lo mencionado anteriormente es el análisis para tiempo continuo, elrazonamiento en tiempo discreto es análogo. En la práctica se utilizo el algorit-mo de FFT (Fast Fourier Transform) para obtener la transformada de Fourierdiscreta (TFD) como aproximación de la transformada de Fourier de tiempodiscreto (TFTD).

Se utilizaron diferentes tipos de ventana para lograr un truncamiento mássuave, dado que la suavidad en el dominio del tiempo se traduce en un cam-bio brusco en frecuencia, de esta manera se lograría maximizar la resoluciónespectral sin perder resolución temporal.

El análisis en frecuencia no aporto resultados signicativos, a pesar de latécnica de enventanado no se logro una resolución temporal y en frecuenciaque satisciera las necesidades. Dado que los ritmos no tienen una duraciónconstante en el tiempo, fue imposible determinar un tamaño de ventana op-tima, lo ideal es variar τ de forma adaptativa a la duración relativa de cadaritmo. Dicho esto último, se optó por abandonar el análisis de frecuencia.

Análisis de potenciaUnas de las técnicas más utilizadas para detectar la presencia de componen-

tes frecuenciales en una señal en tiempo real concite en medir la potencia mediade la señal en una banda de frecuencia especíca. Esto se consigue medianteel ltrado de la señal por un ltro pasa banda que contenga la frecuencia deinterés, posteriormente se calcula el valor RMS cuadrado de la señal ltrada,y se compara con un umbral.

94

Figura 9.6.: Análisis de potencia.

El cálculo del valor RMS solo aplica para señales periódicas. Debido a quela señal de EEG no es periódica, no es posible calcular el mismo. Una posibleestimación de la potencia media se logra calculando la potencia en un interva-lo te tiempo Dt, en tiempo discreto esto equivale a tomar las muestras de laseñal ltrada correspondientes a ese intervalo, elevar cada muestra al cuadradoy promediar. Lo obtenido es una aproximación de la potencia media sobre unaresitencia de 1 Ωen cada intervalo de tiempo Dt.

Es posible de esta manera detectar la presencia de ritmos cerebrales, soloes necesario calcular la potencia en la banda correspondiente a cada ritmo ycomparar la misma con un umbral adecuado.

9.4.4. BCI basada en ritmo alfa occipital

Es una interfaz On/O como la descrita en la gura 9.4, se basa en detectarel ritmo alfa occipital. Es el ritmo más fácil para la mayoría de los individuosde bloquear/desbloquear, es natural que al cerrar los ojos la potencia del mis-mo incremente una 10 veces con respecto al nivel de base o background queexiste con los ojos abiertos.

Como método de detección del ritmo alfa se utilizo el análisis en potencia.

Dado que no es posible medir la potencia del ritmo, se estima la misma. Estose logra ltrando la señal de EEG por un ltro pasa banda que seleccione elritmo alfa, posteriormente se elevan las muestras al cuadrado, se ltran con

95

un pasa bajos para suavizar los cambios brusco de energía, nalmente se pro-media un cierto tiempo Dt, obteniéndose así una estimación de potencia. Dt

determina cuantas muestras por minuto genera el estimador.

9.4.4.1. Estimador potencia Alfa

Las variaciones de energía producto de la sincronización/desincronizacióndel ritmo son mucho más lentas que la resolución temporal que se pose de1024 muestras por segundo. Con el n de evitar el exceso de cálculo se decidióoperar con 256 muestras para los ltros, si bien esta cantidad de muestras essuciente para obtener los detalles de la señal, es excesiva para la visualizaciónde la señal de potencia si se quiere utilizar como retroalimentación del usuario,generalmente las BCI utilizan 8 muestra por segundo, valor que se considerórazonable para la visualización.

Figura 9.7.: Estimador potencia Alfa.

9.4.4.2. Filtro pasa banda

Este ltro tiene la función de seleccionar el ritmo alfa de la señal. Recorde-mos que el mismo se extiende de 8 a 12 Hz.

Consideraciones para el diseñoDado que solo se quiere estimar la energía, no importa que la señal se dis-

torsioné, es necesario que el ltro tenga una caída rápida, pero que su ordenno sea demasiado alto para evitar latencias.

Diseño

96

Dado las consideraciones se utilizo un ltro elíptico ya que asegura una caídamuy rápida minimizando la banda de transición, pose un rizado constante en labanda de paso. Como gran desventaja tiene una fase no lineal que distorsionala señal, pero que no afecta el estudio en potencia.

Se obtuvo un ltro de orden 4, centrado en la frecuencia 10 Hz con un anchode banda de 4 Hz que tiene un error de ± 0,05 en la banda de paso y 0,1 en labanda de rechazo. La atenuación a los 50 Hz para este ltro es de 42 dB.

Figura 9.8.: Grafo ltro pasa banda.

97

Observando la gráca se puede llegar a pensar que la fase es lineal, esto no escierto, si bien se asemeja a una recta, la misma esta sumada a una constante,que se transforma en un retardo variable en el tiempo.

Debido a la fase no lineal del ltro no es posible determinar un retardo pro-medio, el mismo depende de la frecuencia. Para una señal de 10 Hz el desfasajees de 2, 75°, el retardo sufrido es 764µs.

9.4.4.3. Filtro de suavizado

Su labor es la de suavizar la señal representativa de energía eliminando lasvariaciones. Esto se logra ltrando la señal con un ltro pasa bajo. La señalobtenida después del ltrado es una aproximación de la estimación de la po-tencia, debido a que ltrar con un pasa bajo es integrar. Al promediar estaseñal se obtiene la estimación de potencia.

Consideraciones para el diseñoEl factor de suavizado está directamente relacionado a la cantidad de com-

ponentes en frecuencia que permite pasar el ltro, es decir, cuanto más chicosea el ancho de banda del ltro, más grande será el factor de suavizado, hacien-do que la señal tenga menos uctuaciones, lo que aumenta la inercia de la señal.

El aumento de la inercia provoca que la estimación sea más robusta, pero senecesita un tiempo mayor para detectar los cambios de potencia. Esto últimoes inconveniente dado que degrada la capacidad de transmisión de informaciónque tiene la BCI.

Debido a que pueden existir variaciones de potencia muy abruptas es ne-cesario que el ltro no tenga un valor de sobretiro muy elevado ya que estoderivaría en estimar valores negativos de potencia cuando la misma presentacaídas importantes, por ejemplo cuando se des-sincroniza el ritmo alfa. Esto sepuede evitar aumentando el ancho de la banda de transición, lo que conllevaa disminuir el factor de suavizado.

DiseñoNo es posible utilizar ltros del tipo Chebyshev o elíptico ya que tienen caí-

das demasiado rápidas que generan valores de sobretiro negativos. Para evitareste mal se diseño un ltro de Gauss, también conocido como ltro de res-

98

puesta impulsiva crítica dado que sus polos se encuentran muy próximos a lacircunferencia unidad.

Este tipo de ltro es muy sencillo, solo posee dos polos y ambos son coinci-dentes.

A partir del ensayo practico se selección la pareja de polos que mejor sa-tisfacía con las necesidades de suavizado e inercia de la señal estimativa depotencia para el ritmo alfa occipital.

Figura 9.9.: Ubicación de los polos.

Transferencia obtenida:

0,000081

z−2 − 1,98199× z−1 + 0,98208

99

Figura 9.10.: Grafos ltro Gauss.

Estudiando la respuesta al escalón se puede vericar que el ltro construidono presenta sobretiros, una de las características deseadas.

El tiempo de establecimiento3 (te) es de 2,05 s. Este valor es importante,debido a que representa el tiempo que le toma al ltro determinar el valorde potencia en régimen a partir de la señal de energía. Es el ltro más lentodel sistema, por ende determinara en gran parte la performance de la BCI.

3Tiempo de establecimiento es el tiempo necesario para que la señal alcance en 95% delvalor nal.

100

Como ya se menciono una mayor latencia en la detección del ritmo degrada lacapacidad de trasmisión de datos.

9.4.4.4. Umbral

Las BCI más básicas utilizan un umbral jo, esto presenta dos inconvenien-te, el primero de ellos implica sintonizar el nivel del umbral para cada sujeto,como ya se menciono en 3.4, este diere para cada persona. El segundo es aúnmás preocupante dado que el nivel alfa de fondo puede varia en cuestión depocos segundos para cada individuo, lo que obliga a ajustar el umbral con lamisma frecuencia de tiempo. Una solución a este problema es comparar la señalde potencia alfa con el umbral pero utilizando histéresis, es decir, si la señalde potencia es superior a C1 × umbral entonces el ritmo alfa está sincroniza-do, si la señal de potencia decae por debajo de C2 × umbral, el ritmo alfa seencuentra des-sincronizado (nivel de background). Esta técnica utilizada paraaumentar la inmunidad frente a variaciones espurias de la señal nos ayuda aevitar los falsos positivos y negativos (detección errónea de sincronización odesincronización).

Umbrales de histéresisPara poder estimar los valores correctos de C1 y C2 fue necesario medir la

potencia de sincronización y de base del ritmo alfa en diferentes sujetos deprueba. Si bien una muestra de 10 personas no es representativa, fue sucientepara obtener conclusiones y determinar los valores de histéresis.

Se realizaron 3 test de pruebas para medir el ritmo alfa occipital. Las mediasse hicieron con electrodos activos situados en O1 − O2.

1. Máxima variación en potencia entre sincronización y ritmo base.En este test se le pidió a los usuarios que mantuvieran abiertos sus par-pados por 20 segundos para medir el nivel base (no es necesario evitarel parpadeo), acto seguido, se les pidió que cerraran los parpados por 5segundos para que entraran en estado de relax visual lo que sincronizael ritmo alfa occipital.

2. Duración efectiva de la sincronización del ritmo.Similar al caso anterior pero con la salvedad de mantener los parpadoscerrados por 30 segundos, el n del test es observar el comportamientode sincronización del ritmo por periodos prolongados de tiempo.

101

3. Sincronización y desincronización periódica.Alternar entre tener los parpados abiertos y cerrados, permaneciendo encada estado por 5 segundos, durante un lapso de 30 segundos.

Recordar que cuando hablamos de potencia nos estamos reriendo a una esti-mación de la potencia media sobre una resistencia de1 Ω.

Figura 9.11.: Máxima variación de potencia.

Figura 9.12.: Duración efectiva de la sincronización.

102

Figura 9.13.: Sincronización y desincronización periódica.

En promedio se observo que durante la sincronización la potencia incrementa10 veces con respecto al nivel base.

Una misma persona puede tener incrementos de potencia de diferente mag-nitud, la tendencia es que la variación de potencia tienda a disminuir a medidaque se repite el ejercicio de sincronización y desincronización.

El menor aumento de potencia registrado fue de 4,5 veces.

En todos los sujetos a excepción de 1, la potencia del ritmo alfa al cerrar losojos aumenta de forma estrictamente monótona.

Según lo observado no es posible mantener el ritmo sincronizado a voluntad,después de cierto tiempo la potencia decrece a valores cercanos al de fondo,aunque permanezcan los parpados cerrados. El periodo que dura el sincronismodiere mucho entre los sujetos, por lo que no se puedo generalizar un tiempode duración efectiva. El menor tiempo registrado es de 4 segundos.

La literatura sugiere que la potencia del ritmo alfa puede variar según ladistancia entre los electrodos, el incremento de la diferencia de potencial esproporcional al cuadrado de la distancia, esto es válido hasta 10 cm [3]. Enlas pruebas realizadas se constato que existen diferencias entre la posición ysimetría en la posición de los electrodos.

Se decidió a partir de los datos obtenidos que C1, coeciente que determina

103

el umbral superior, debe de ser 4 ya que el menor incremento en potencia re-gistrado fue de 4,5 veces el nivel de background. El coeciente C2 responsablede determinar el nivel inferior se jo en 2, dado que en el peor de los casosobservado aún estando en la etapa de sincronización no prolongada la potenciadescendió a valores cercanos a 2 veces el nivel de base.

La solución propuesta para evitar el sintonizado de la BCI es utilizar unumbral adaptativo.

9.4.4.5. Umbral adaptativo

Una forma de lograr que el nivel de base se adapte es mediante un ltradono lineal, si la tendencia de la señal de potencia es incrementar, entonces elnivel de base empieza a aumentar, en caso contrario disminuye.

En el momento del sincronismo del ritmo alfa, la potencia incrementa consi-derablemente, por lo que es razonable que el aumento del nivel de backgroundsea lento ya que este incremento no se debió por un aumento propio del nivel debase, sino que por el sincronismo del ritmo, por otro lado si la potencia decaepor debajo del nivel de background, el mismo debe de adecuarse rápidamentea esta disminución de potencia que es característica del asincronismo del ritmo.

Diagrama de ujo

104

Figura 9.14.: Algoritmo umbral adaptativo.

Siendo Bk las muestras del nivel de background, Pk la potencia actual esti-mada y λ1 λ2 los coecientes de ponderación.

Notar que el nivel de background no solo depende del valor de potencia ac-tual, sino que también se tiene en cuenta el valor pasado del background, estonos permite aumentar la inercia, logrando una mayor robustez.

9.4.4.6. Performance de la BCI basada en ritmo alfa

Las características principales que hacen fuerte la performance de un BCIson la inmunidad frente a los artefactos y la capacidad de trasmisión de datos.

La BCI propuesta no tiene ningún algoritmo especializado en reconocer ar-tefactos en general, por lo que no es posible eliminarlos de la señal de EEG,estos algoritmos tienden a ser muy complejos, generalmente están basados enredes neuronales o en parámetros autoregresivos. La solución encontrada paramitigar esta falencia fue aumentar la robustez en el algoritmo que detecta elritmo alfa, con el n minimizar los falsos positivos debido a los artefactos.

105

Como ya se mencionó el aumento en la robustez provoca que se demore mástiempo en detectar el ritmo, lo que genera que se demore más tiempo en activaro desactivar la interfaz.

A continuación se desarrollará un análisis para estudiar el tiempo que re-quiere la BCI propuesta en detectar el sincronismo del ritmo alfa y activarse.

En el sistema total, despreciando la demora producto de la conversión analó-gico digital, la trasmisión de los datos y los cálculos realizados por el progra-ma, los responsables de determinar el tiempo de detección son los ltros de laplaca amplicadora, el ltro de 40 Hz, el ltro pasa banda centrado en 10 Hzy el ltro Gauss.

Bloques responsables del retardo

Figura 9.15.: Bloques responsables del retardo.

La placa amplicadora tiene una fase lineal en el rango de interés, esto nospermite calcular el retardo sufrido por toda la señal, aproximadamente es de7 ms.

La BCI se activa solo si detecta señales que presenten componentes frecuen-ciales que estén en el entorno de los 10 Hz, por lo que solo estudiaremos elretardo para este tipo de señales en los restantes bloques. Debido a que sonltros digitales, podemos recurrir a la simulación para determinar el retardo ycomparar con los cálculos teóricos.

Estadísticamente el ritmo alfa presenta su mayor componente frecuencial alos 10 Hz, simularemos al mismo como un tono puro de dicha frecuencia.

106

Retardo del ltro 40HzPara medir el retardo que genera el ltro solo basta con medir el desfasaje

entre la señal de entrada y la salida del ltro en el tiempo. La demora es de11 ms, valor semejante al retardo promedio calculado teóricamente.

Retardo del ltro pasa bandaEl retardo medido es de 1 ms, que concuerda con el calculado a partir de la

fase.

Retardo del ltro GaussSe debe recordar que este ltro se encuentra a continuación de un bloque que

potencia la señal al cuadrado, siguiendo bajo la misma hipótesis de modelarel ritmo alfa como un tono puro, la señal de entrada es una continúa más untono puro al doble de frecuencia.

Independientemente de que la señal de entrada sea un tono, siempre la en-trada de este ltro será una continua más componentes del doble de frecuenciaque las que permite pasar el ltro pasa banda centrado en 10 Hz.

Como el ltro Gauss es un pasa bajo muy angosto que solo interactúa conlas señales de continua, no hablamos del retardo, sino que del tiempo de esta-blecimiento. Ya que estas señales presentan variaciones muy bruscas que sacanal ltro del régimen.

La potencia media de un tono puro es A2

2, siendo A la amplitud del tono.

〈P 〉 =1

T

ˆ i+T

i

(Asin(ωt))2dt =A2

2−ˆ i+T

i

A2cos(2ωt)

2dt︸ ︷︷ ︸

0

=A2

2

A continuación se presentará el gráco de la respuesta en el tiempo del ltroGauss a la entrada sin(ωt)2.

107

Figura 9.16.: Respuesta en el tiempo del ltro Gauss a la entrada sin(ωt)2.

La señal demora 2.05 s en llegar al 95% del valor nal de potencia, estetiempo corresponde al tiempo de establecimiento del ltro.

Si suponemos que todos los bloques que interviene en la latencia de la detec-ción del sincronismo del ritmo alfa son ideales, podemos considerar que cadauno retarda la señal de forma ideal.

Como los bloques están concatenados, la latencia total, es la suma de losretardos de cada bloque. Por lo tanto τ = 2,07 s. Este es el tiempo que le llevaal sistema en detectar la potencia en régimen, pero no es el tiempo que letoma detectar la presencia del ritmo alfa, ya que no se está tenido en cuentael umbral de detección y el incremento de potencia. Cuanto mayor sea la dife-rencia de potencia entre el nivel de sincronismo y el nivel de base, más rápidoresponderá la BCI. Para aclarar esta idea se prosigue a un ejemplo.

Supongamos que existe un incremento de 10 veces entre la potencia de sin-cronización y el nivel de background. También sabemos que le toma 2,07 s a la

108

BCI estimar este incremento de potencia, recordar que la misma se activa sieste incremento supera 4 veces el nivel del umbral. Suponiendo que es lineal eltiempo que le lleva estimar la potencia, el estimador demorara 4

10τ en detectar

el incremento de 4 veces, en estas condiciones la BCI se activara en T1 = 1,08 sdespués de que se sincronizo el ritmo alfa.

El menor incremento registrado de potencia fue de 4,5 veces, por lo tanto eltiempo de activación será de T2 = 1,84 s

Figura 9.17.: Tiempo de detección del ritmo alfa. B representa el nivel de back-ground.

La BCI implementada funciona en modo toggle, la primera detección la activay la segunda la desactiva, claro está que para detectar por segunda vez el rit-mo alfa, es necesario que la BCI detecte la desincronización del mismo primero.

El tiempo necesario para detectar la desincronización dependerá de la po-tencia del ritmo alfa, cuanta más alta sea esta, mayor será el tiempo. Si consi-deramos que una vez que se detecta el ritmo, el usuario abre sus párpados paradesincronizarlo e ignoramos la inercia del ltro,es decir, estamos suponiendoque el detector reconoce automáticamente que hay un cambio de potencia. El

109

tiempo de detección de la desincronización del ritmo es constante.

Haciendo un razonamiento análogo al anterior, se puede determinar el tiem-po que demora la BCI en detectar la desincronización.

La desincronización del ritmo no es más que una caída brusca de potencia,por lo tanto al estimador le tomara τ segundos en determinar la potencia enrégimen (en este caso la potencia de background), recordar que la BCI detectael asincronismo del ritmo alfa cuando la potencia del mismo es inferior a 2veces el nivel de base. El tiempo de detección del asincronismo de ritmo es de24τ = Ta = 1,04 s.

Figura 9.18.: Tiempo de detección de asincronismo del ritmo alfa.

Siendo T = T′= Ta = 1,04 s.

El tiempo necesario para mandar un símbolo queda determinado por el tiem-po de detección del ritmo alfa, lo llamaremos T (Pα) debido a que depende dela potencia del mismo, y del tiempo de detección del asincronismo del ritmoTa. En la BCI implementada, cada símbolo es un bit.

110

Tbit = T (Pα) + Ta

Figura 9.19.: Tiempo de bit.

En promedio Tbit = T1 + Ta = 2,12 s, el mayor tiempo previsto se da paralos usuarios que presenten un incremento de potencia de 4,5 veces el nivel debackground, en este caso Tbit = T2 + Ta = 2,88 s.

La velocidad de trasmisión de la BCI dependerá del usuario, en el peor delos casos la velocidad será 20,8 b/min, en promedio se espera 28,3 b/min.

El tiempo que demora la BCI en estimar la potencia no es lineal, esto se

111

observa en la respuesta al impulso del ltro Gauss (gura 9.10), debido a estano linealidad sabemos que el ltro presenta inercia. Es necesario recurrir a lasimulación para obtener velocidad real de trasmisión (ver código en apéndiceA.4.6).

Para simular un incremento de 4.5 veces en la potencia del ritmo, generamosuna señal compuesta de 3 sinusoides, dos de ellas de amplitud unitaria quesimulan el ritmo base, la tercera de amplitud

√4,5 que simula el sincronismo

del ritmo alfa.

x(t) = sin(20πt)× p[0, t1] +√

4,5× sin(20πt)× p[t1, t2] + sin(20πt)× p[t2, t3]

112

Figura 9.20.: Tiempo de bit (incremento 4.5).

El tiempo de bit obtenido prácticamente es de 3 s, valor semejante al calcu-lado teóricamente de 2,88 s, el error cometido en la estimación teórica es de un4%.

Procediendo análogamente al caso anterior simularemos un incremento de10 veces entre la potencia de background y el sincronismo.

113

Figura 9.21.: Tiempo de bit (incremento 10).

El tiempo de bit obtenido es de 2,30 s, el error cometido en el calculo teóricoes de un 13%, es lógico que el error en la estimación de la velocidad aumentea medida que crezca el incremento de potencia, esto se debe a que la inerciadel ltro inuye cada vez más en la estimación.

La velocidad en promedio esperada basada en la simulación es de 26,08 b/min.

En la práctica la máxima velocidad de trasmisión alcanzada fue de 15 b/min,independiente del incremento de potencia de los usuarios, todos lograron ve-

114

locidades próximas a los 12 b/min. Como ya se menciono antes, el ejercicio desincronizar y desincronizar el ritmo alfa consecutivamente conlleva a una dis-minución del incremento de potencia acompañada de una latencia entre elmomento en que se cierran los párpados y se sincroniza el ritmo.

9.4.5. BCI basada en ritmo alfa frontal

Es muy similar a la BCI basada en el ritmo alfa occipital, la diferencia esque utiliza la señal de EEG proveniente de los electrodos Fz −O1.

El ritmo alfa frontal se sincroniza con la relajación y se bloquea con la con-centración (por ejemplo al hacer cálculos matemáticos). Presenta al igual queel alfa occipital un nivel de base que se encuentra presenta cuando los sujetosestán en un estado mental consiente pero sin relajarse o concentrarse.

No fue posible utilizar este ritmo como señal de control, ya que es muy com-plicado controlar la sincronización y desincronización a voluntad, es necesarioque el usuario tenga un entrenamiento previo antes de poder utilizar la BCI.

A pesar de que no se implemento la BCI en sí, el software calcula la señalestimativa de potencia de este ritmo, y la misma se muestra por pantalla.

9.4.6. BCI basada en potenciales miográcos

Los potenciales miográcos son los encargados de producir las contraccionesde los músculos. Si bien estos potenciales no son producto de la actividad ce-rebral, son evocados por potenciales neuronales, por lo cual es común que lasBCI los implementen.

Estos potenciales presentan tensiones generalmente muy superiores a los po-tenciales cerebrales, el rango se extiende entre 50µV a 30 mV.

Se implemento una BCI que utiliza los potenciales generados por la contrac-ción de los párpados.

La señal generada por el movimiento de los párpados es el artefacto máscomún en la señal de EEG, es inevitable ya que no es posible que un sujeto dejede parpadear por periodos prologados. La ventaja visible de este artefacto esque es muy simple de distinguir y se presenta de la misma forma en la mayoríade las personas.

115

9.4.6.1. Detección del parpadeo

La señal eléctrica proveniente de los pardos se asemeja mucho a una señalpulso (g 9.22 ), el ancho de los mismo es relativamente constante y similaren todas las personas. Estos pulsos presentan amplitudes de por lo menos dosveces a la señal de EEG de fondo, dado esto y a su morfología es posible dise-ñar un algoritmo para su detección.

Si la señal de EEG en el tiempo supera un umbralSUP y se mantiene porencima del mismo por un cierto tiempo Tparpadeo y posteriormente la señalcae por debajo de un umbralINF , se considera que se detecto el parpadeo.Los valores de las constantes umbralSUP , umbralINF y Tparpadeo se obtuvieronprácticamente.

Para detectar los parpadeos se utilizo la señal proveniente de los electrodosFz −O1.

Figura 9.22.: Efecto del parpadeo.

9.4.6.2. Señales de control a partir de los parpadeos

No es posible utilizar un único parpadeo como una señal de control ya quees un acto involuntario. Lo que sí es posible controlar la cantidad de parpa-

116

deos seguidos, por ejemplo activar y desactivar un comando dependiendo delnúmero de parpadeos seguidos.

Se implemento una BCI basado en el número de parpadeos seguidos, se de-tectan dos o tres parpadeos. A continuación se presentara un diagrama de ujodel algoritmo utilizado para detectar parpadeos.

117

Figura 9.23.: Algoritmo detección de parpadeos.

118

Figura 9.24.: Bloque detectar parpadeos.

119

9.4.7. Visualización

Este bloque genera todos los grácos del programa, permite visualizar lasseñales capturadas por el adquisidor, así como también señales generadas porel propio sistema como lo son la estimación de potencia y el nivel de back-ground. La visualización de estas señales son utilizadas por el usuario comoretroalimentación, lo que le permite tener un mejor control de la BCI.

9.4.7.1. Velocidad de barrido

El sistema recibe 1024 muestras por segundo, si visualizamos la señal conla misma tasa de muestras con la que se obtiene, estaríamos gracando 1024puntos (pixel) por segundo. Esta velocidad es muy excesiva para el ojo humano.

Si se quiere visualizar una señal sinusoidal de 1Hz por lo menos es necesariocontar con 3 muestras para que una persona la reconozca como tal, la señal deinterés tiene un ancho de banda que se extiende hasta los 40Hz, por lo tantosi se quiere visualizar todo el rango es necesario contar con 120 muestras.

9.5. Servidor

Con el n de no limitarnos a controlar una única aplicación se decidió gene-rar el bloque servidor el cual transmite los datos de control generados por elsoftware de monitoreo a través de un socket. De esta manera hay un sinfín deaplicaciones que se pueden controlar, lo único que se precisa es un traductor,el cual se encarga de vincular el software Monitor con la aplicación.

El traductor puede ser programado en cualquier lenguaje, simplemente tie-ne que ser capaz de conectarse a través de socket e interpretar los comandasque el Monitor envía.

El servidor escucha en el puerto 5000.

9.5.1. Comandos

Existen dos BCI implementadas en el sistema, una de ellas funciona en modotoggle, es la BCI basada en ritmo alfa, se envía un 1 cuando dicho ritmo seactiva y 0 cuando se desactiva.

120

El algoritmo de detección de parpadeo, BCI basada en potenciales miográ-cos, envía un 2 al detectar dos parpadeos y un 3 al detectar tres parpadeos.

Dado que el sistema opera con múltiples canales y ambas BCI utilizan losdatos de todos ellos, el servidor envía un dato más, éste indica el canal quecapturó la señal que generó alguno de los comándos posibles. Dicho esto, elservidor envía un vector que contiene dos bytes, dispuestos de la siguiente for-ma: [canal, comando].

Antes de que se cierre la aplicación Monitor, se envía el comando 4 porel socket para indicarle al cliente el n de la conexión.

121

10. Aplicación de prueba de laBCI

El software de monitoreo funciona como servidor, escuchando al puerto 5000.De esta manera cualquier programa, puede funcionar como cliente conectándo-se a este puerto mediante sockets, y recibir datos sobre las acciones del usuario.

Como aplicación de la BCI se decidió controlar el movimiento de un robotLEGO NXT. La misma consta de dos programas, uno corre en el NXT y elotro en el PC. Este último comunica al software de monitoreo con el NXTmediante comunicación BT.

El software del monitoreo funciona en modo toggle, es decir que cuandodetecta presencia del ritmo Alfa, envía un 1 al cliente si está en estado apagadoy un 0 si está en estado prendido. Al detectar dos parpadeos envía un 2 y aldetectar 3 parpadeos envía un 3. También indica en que canal fue detectadoel evento para poder reconocer la ubicación del canal que registra el mismo,esto también sirve para diferenciar de que usuario proviene el evento, en elcaso de que haya más de uno. Como el movimiento del NXT fue diseñado paraser controlado con una persona sola y con dos canales, el primero detectandoparpadeos y el segundo la presencia del ritmo alfa, el programa que se comunicacon el NXT, sólo se comunica con el mismo para avisarle de estos eventos endichos canales.

122

11. Pruebas realizadas sobre elfuncionamiento del sistema

Una vez construido el sistema, se realizaron diferentes pruebas sobre el mis-mo, se probaron señales con y sin cable mallado, se midió el ruido presente enla entrada, se determinó la frecuencia de corte y la atenuación a la mitad dela frecuencia de muestreo (512 Hz), se verico que el microcontrolador mues-treará a una frecuencia adecuada y transmitiera todos los datos debidos, y queel software de la computadora recolecte los datos, los procese y los muestre auna velocidad mayor de la que llegan al buer. Todas las medidas se realizaronen un laboratorio de la Universidad ORT Uruguay y podrían brindar distintosresultados en ambientes más, o menos ruidos.

11.1. Pruebas con cable sin mallar y mallado

Con el n de vericar si fue correcta la decisión de mallar el cable que conec-ta la salida de los electrodos activos, con la placa amplicadora, se realizaronpruebas con cables mallados y sin mallar.

Para poder realizar las pruebas, era necesario utilizar señales de prueba delorden de las señales EEG. Como no se contaba con un generador de señal quepudiera producir señales de esa magnitud, se utilizó un divisor resistivo, pre-sente en la placa amplicadora, que divide la señal introducida en su entradapor un factor de 20.000. Luego esta señal es inyectada a los electrodos y semide la salida de la placa amplicadora con un osciloscopio digital. Para estaspruebas se utilizó la ganancia de la placa amplicadora casi en su máximo(19600), ya que como los electrodos no estaban en contacto con el cuerpo delusuario (por lo que no se producirían artefactos), y se podía variar la entradacomo se deseara, no ocurrirían saturaciones.

En la gura 11.1 se observa el canal 1 y 2 del osciloscopio, que representanla señal de entrada al divisor resistivo y de salida de la placa amplicadorarespectivamente. En este caso se utilizó un cable de 1 metro sin mallar. Se

123

observa una clara presencia de la señal de la red de distribución superpuestaa la señal de entrada.

Figura 11.1.: Prueba con cable sin mallar

En la gura 11.2 se observa el canal 2 del osciloscopio, que representa laseñal de salida de la placa amplicadora, en este caso con un cable mallado de1 metro de largo. En este caso la señal de salida reproduce la señal de entradasin ninguna distorsión visible.

124

Figura 11.2.: Prueba con cable mallado

De esta manera, queda clara la importancia de utilizar cable mallado paradisminuir en gran medida la interferencia captada por los cables.

11.2. Medición del ruido

Se realizó una medición del ruido, para poder compararlo con la señal deinterés, y comprobar que el mismo sea varias veces menor que la señal.

Para poder medir el ruido, sin que las señales del cuerpo intereran, se reali-zó un modelo de la conexión de los electrodos al cuerpo. Se conectaron loselectrodos mediante dos resistencias de 300 kΩ, caso desfavorable para la re-sistencia entre los electrodos, y fueron conectados al electrodo DRL medianteuna resistencia de 300 kΩ, considerando también una alta impedancia para elelectrodo. El ruido presente en este caso será representativo del ruido genera-do por los amplicadores operacionales, y la interferencia que ingrese por elcableado y la placa, tanto diferencial como de modo común.

En la gura 11.3 se observa el ruido medido por el osciloscopio, para deter-minar un valor del mismo se utilizó un tester Fluke, midiendo de esta manera,

125

un valor de ruido aproximado de 5 mV RMS. Este valor hallado representa elruido presente en la salida, al dividirlo por la ganancia total, encontramos elruido en la entrada, el cual resulta en un valor de aproximadamente de 0,26µVRMS, valor de ruido aún más pequeño que el calculado teóricamente parael ruido producido por los amplicadores (ver subsección 7.5), aún cuando esteruido también está compuesto por el ruido que ingresa a los cables. De esto sedesprende que los valores reales del ruido producido por los integrados estánpor debajo de los valores típicos, y que la interferencia que ingresa por loscables al ser mallados, es del mismo orden o menor, que el ruido antes men-cionado, al igual que la interferencia captada por la placa al utilizar plano detierra.

Figura 11.3.: Medición del ruido

En la gura 11.4 se observa la señal medida por el software propio, la cualdebería contener menos ruido que la señal medida a la salida de la placa am-plicadora debido al ltrado digital posterior que se realiza en el software.Efectivamente, se observan variaciones de ±1 LSB respecta al nivel de con-tinua (VGND), que equivalen a ±4,9 mV, y representan menos que los 5mVRMS.

126

Figura 11.4.: Medición del ruido por el software propio

11.3. Frecuencia de corte y atenuación en la

banda de paso y a la frecuencia de 512

Hz

Se midió la frecuencia de corte, resultando en un valor de 95 Hz, la cualestá dentro del rango previsto según el estudio teórico. La mayor caída en labanda de interés es de aproximadamente 0,61 dB, a la frecuencia de 40 Hz yla atenuación lograda a 512 Hz fue próxima a los 40,3 dB, valores que estántambién dentro del rango previsto.

11.4. Vericación de la frecuencia de

muestreo y envío de datos del

microcontrolador

Se vericó que el microcontrolador tuviese la frecuencia de muestreo desea-da y que esta se mantuviera constante, ya que de lo contrario, los ltros nofuncionarían correctamente, además de provocar distorsión de la señal debidoal aliasing. La frecuencia de muestreo más cercana a la deseada que se pudolograr fue de 1024,6 Hz, una diferencia de 0,059% respecto a los 1024 Hz, ygracias a que el microcontrolador es un sistema de tiempo real y su ADC tieneun tiempo de conversión constante, la frecuencia de muestreo es constante.

127

Para comprobar que cada canal está siendo muestreado y enviado correcta-mente, se colocaron distintas tensiones de continua en los canales del ADC; secomprobó que las conversiones sean correctas y sean enviadas correctamente.Se dejó corriendo el microcontrolador durante 10 horas de corrido sin inte-rrumpirse ni caerse.

11.5. Pruebas realizadas al software de

monitoreo

Se realizaron pruebas sobre el software de monitoreo ya que la PC utilizadano funciona en tiempo real y las señales que se deseaban medir si lo son. Laspruebas fueron realizadas en un PC con Windows 7 Professional de 64 bit, pro-cesador Pentium(R) Dual-Core CPU T4500 @ 2.30 GHz y 2 GB de memoriaRAM. Se midió la cantidad de datos que se almacenaban en el buer y la mis-ma nunca superó 3 datos. Esto no se debe a que el tiempo de procesamiento delprograma demore el tiempo que tardan en llegar 3 datos, sino de cada cuantotiempo, el sistema operativo atienda el evento del puerto serial y le avisa alsoftware que levante datos del buer, tiempo que depende de las tareas que es-te ejecutando el ordenador, las cuales no incluyen solo el software de monitoreo.

Lo importante a destacar es que el software no extrae datos del buer hastaque termine su procesamiento de los datos que fueron leídos, por lo que sededuce que, aún considerando que lo único que este atendiendo el sistemaoperativo sea el software, el tiempo de procesamiento nunca puede ser mayorque el tiempo que demoran en llegar tres datos (3 bytes). El tiempo máximoque puede haber entre 3 datos es despreciable por los humanos, por lo tantola visualización de la señal puede ser considerada como tiempo real. El hechoque el procesamiento sea más veloz que la llegada de datos es muy razonable,ya que la velocidad de envío es 230400 bps, mientras que la frecuencia delprocesador (juntando ambos núcleos) es de 4,60 GHz.

128

12. Conclusiones

Los autores consideramos que se logró cumplir con los principales objetivosdel proyecto. Se obtuvo un equipo compacto, robusto, móvil, de bajo consumoy económico en conjunto con un software de monitoreo multiplataforma, abier-to a la expansión, fácil de utilizar y que permite el desarrollo de un sinnúmerode aplicaciones de control para la BCI, las cuales pueden ser programada encualquier lenguaje de programación que incorpore sockets.

Además se diseñó una aplicación básica para la prueba de la BCI, que de-muestra la potencia del proyecto.

También se tuvo una experiencia muy disfrutable y enriquecedora de trabajode grupo.

12.1. Planicación

En Marzo del 2013, en inicio de este proyecto, se realizó una planicacióndetallada del las etapas por las cuales transitaría el mismo (ver gura 12.1).

129

Figura 12.1.: Diagrama de Gantt.

130

Luego, esta planicación sufrió modicaciones debido a diversos obstáculosque hubo que sortear durante la realización del trabajo.

Como dicultades importantes encontradas durante le ejecución, podemosmencionar la fabricación de la placa amplicadora. La falta de equipos y he-rramientas para construir un circuito doble capa hizo necesario la compra dela misma, lo que generó importantes demoras en la obtención del amplica-dor denitivo. También es válido destacar que, a pesar de su precariedad, laplaca que construimos, de simple faz, sirvió para realizar las primeras medidas.

Otro obstáculo importante que hubo que sortear, fue la necesidad de imple-mentar electrodos del tipo activos, debido a que los pasivos no dieron buenosresultados. Esto trajo aparejado construir el pequeño circuito que incluyen,además de un hardware que permita el montaje de los mismos sobre la cabezalo cual requirió de varias versiones y pruebas, tomando un tiempo importante.

La presión adecuada, uniforme y constante que deben ejercer los electrodosfue un gran problema de este proyecto debido a que cuando esto no ocurre, segeneran saturaciones en la señal. Este problema no fue detectado inmediata-mente, ya que se pensó que el problema podría deberse a la placa amplicadora,a un fallo en el muestreo del microcontrolador o tal vez debido a interferen-cias. Se realizaron pruebas que conrmaron que la fuente del problema era lavincha, lo que llevo a la utilización del casco logrado en el proyecto.

Otra dicultad se encontró al momento de la detección de ritmos cerebrales,la primera idea para atacar este problema fue mediante el análisis en frecuenciautilizando la TFD, dicha técnica no soluciono el problema debido al efecto quegeneraba el enventanado de la señal, lo que se traducía en una distorsión en elespectro de la misma. Fue necesario entonces, recurrir al análisis de potenciapara la detección. El primer algoritmo para estimar la potencia utilizaba l-tros con caídas bruscas que le otorgaban a la BCI una velocidad de respuestaelevada, pero con una tasa de detecciones erróneas, por lo que se buscó utilizarltros más lentos que mejoraban la robustez en cuanto al error, sin perjudicardemasiado la velocidad de respuesta.

Dado que el programa Monitor de EEG tiene que hacer varias tareas ensimultaneo, obtener las señales de electroencefalograma, acondicionarlas, gra-carlas y analizarlas, se pensó en dividir el software en varios hilos, donde cadauno ejecutara una tarea independiente y pudieran correr al unísono. La di-

131

cultad que trajo dicha práctica fue la gran cantidad de recursos que consumíael programa en la computadora que estaba corriendo. La solución a esto fueminimizar la cantidad de hilos utilizando eventos y procesos sincronizados.

12.2. Posibles mejoras a futuro

En esta sección se comentaran posibles mejoras que se podrían realizar en elproyecto.

La utilización de componentes SMD1 en la construcción del hardwaretraerían numerosas ventajas al mismo. La principal ventaja es la reduc-ción signicativa del volumen y tamaño del hardware, proporcionandouna mayor comodidad para su movilidad, además de captar menos EMIdebido a su pequeño tamaño y menor ruido diferencial por el incrementode la densidad de componentes. Incluso se podría migrar la placa am-plicadora y el microcontrolador al casco de medición, de manera de nonecesitar cables que salgan de este. Al hacer esto, además de incrementarla comodidad, se eliminaría el ruido captado por los cables. También sereducirían los costos de los componentes o se aumentaría la precisión,ya que por lo general los componentes SMD son más económicos quelos componentes clásicos equivalentes. Se debería considerar los cuidadosnecesarios a tener en cuenta en dicha implementación. La alimentacióndebería estar presente en el casco mismo, por lo que sería necesario la re-ducción del tamaño de la batería, provocando una menor duración de lamisma. Para solucionar este problema, se podría comprar componentesque consuman una menor corriente, aumentando el precio de los mismos.También se podría alimentar con una batería de 5 V, evitando así el usodel conversor LM7805, el cual al realizar la conversión, consume potenciano utilizada por la placa. Además serían necesarios instrumentos de altocosto para la construcción del hardware.

Implementación de algoritmos de reconocimientos de patrones a la BCI(por ejemplo con redes neuronales, modelos autoregresivos de media mó-vil, etc.), para poder lograr de esta manera la detección de potencialesespontáneos que son prácticamente invisibles frente al análisis de poten-

1SMD (Surface Mount Device) son componentes de montaje supercial

132

cia utilizado, y así aumentar la performance del sistema.

Construcción de un casco de medición especíca para la tarea, con elec-trodos que provean un mejor contacto y móviles para poder obtenerdistintos puntos de medición de señales cerebrales y ajustables según lamorfología del cráneo del usuario.

Implementar algoritmos que realicen un ltrado de artefactos, de manerade obtener señales más limpias y menos propensas a saturaciones.

Detección del ritmo Beta, para poder observar la potencia del mismo yutilizar la misma en las aplicaciones de control.

Sustitución de los electrodos utilizados por electrodos capacitivos, de ma-nera que la medición de los biopotenciales, no dependa tan críticamentedel contacto electrodo-piel. Con los electrodos capacitivos, incluso se po-drían realizar mediciones a través de la ropa, por lo que el aislamiento quegenera el pelo, no degrada la intensidad de la señal. La implementaciónde estos electrodos es considerablemente más compleja que los electro-dos convencionales y de costo más elevado, pero sin duda disminuyen lainvasividad y los problemas de contacto.

Implementación de una fuente de mayor tensión, para poder utilizar unamayor ganancia en la primera etapa.

Migración del software de monitoreo a un smartphone o tablet, lo quepermitiría aumentar la portabilidad del equipo y simplicar su uso.

133

Referencias Bibliográcas

[1] N. Fejerman and C. S. Medina,Convulsiones en la infancia, diagnóstico ytratamiento. 2da ed., Buenos Aires: El Ateneo, 1990.

[2] E. M. Spinelli. Interfaces para control cerebral. Interne:http://hdl.handle.net/10915/1361, 29 Noviembre, 2013 [4 Julio, 2013].

[3] D. D. Daly and T. A. Pedley,Current Practice of Clinical Electroencepha-lography. 2da ed., New York: Raven Press, 1990.

[4] R. Panek. Dry active electrodes theory. Internet:http://radek.superhost.pl/active_electrodes/index.php?action=theory, 6Febrero, 2014 [10 Abril, 2013].

[5] Merletti, Roberto; Parker, Philip,Electromyography - Physiology, Enginee-ring, and Noninvasive Applications. 1ra ed., New Jersey: John Wiley &Sons, 2004.

[6] A. Searle end L. Kirkup, A direct comparison of wet, dry and isolatingbioelectric recordings electrodes. 1ra ed., Sydney. Departament of AppliedPhysics: University of Technology, 2007.

[7] J. Hansmann. Guia para la construcción de electrodos activos. Internet:http:users.dcc.uchile.cl/~peortega/ae/, 21 Febrero, 2012 [10 Abril, 2013].

[8] Proyecto OpenEEG. Internet: http://openeeg.sourceforge.net/doc/hw/ae.html, 6 Febrero, 2014 [10 Abril, 2013].

[9] Wikipedia. Denición de interferencia. Internet:http://es.wikipedia.org/wiki/Interferencia, 9 Mayo, 2014 [16 Mayo,2013].

[10] Elementos de diseño de circuitos de Amplicación del ECG, XII seminariode ingeniería biomédica, facultades medicina de ingeniería, Universidad dela República, Montevideo, Uruguay.

134

[11] E. M. Spinelli. Amplicadores de Instrumen-tación en Aplicaciones Biomédicas. Internet:http://sedici.unlp.edu.ar/bitstream/handle/10915/1362/Documento_completo__.pdf?sequence=49, 17 Octubre, 2013 [3 Mayo, 2013].

[12] Proyecto OpenEEG. Internet: http://openeeg.sourceforge.net/, 2 Junio,2014 [20 Marzo, 2013]

[13] J. G.Webster, Médical Instrumentation Applications and Design. 2da ed,Boston: Houghton Miin, 1978.

[14] B. Winter and J. G. Webster,Driven-Right-Leg Circuit Design. BiomedicalEngineering, IEEE Transactions on, vol. BME-30, Enero 1983.

[15] R. Pindado Rico, Electrónica analógica integrada. 1ra ed., Barcelona: Mar-combo, 1997.

[16] Arduino home page. Internet: http://arduino.cc, 2 Junio, 2014 [23 Marzo,2013].

[17] Proyecto leJOS. Internet: http://lejos.org, 2 Junio, 2014 [18 Enero, 2014].

135

A. Apéndice

136

A.1. Ecuaciones

A.1.1. Filtro supresor de ondas de radio

Todas las tensiones se expresaran tomando como referencia la tensión VGND,la cual es referencia de Vi1 y Vi2. Se calcula la transferencia para el circuito dela gura A.1.

Figura A.1.: Filtro supresor de ondas de radio.

Debido a la simetría del circuito, las tensiones de salida tendrán la mismaecuación en esencia, sólo que permutandoVi1 por Vi2. Para calcular las ten-siones a la salida se utiliza superposición, primero calculando el impacto deVi1 en Vo1, al que llamaremos Vo11. El impacto de Vi2 en Vo1, debería ser casinulo, ya que de lo contrario, las señales Vi1 y Vi2 interferirían una con la otra,provocando un error en la medida del canal, al ser restadas luego. Se llamaraVo21al impacto de Vi1 en Vo2. El circuito que se utiliza para calcular el impactode Vi1, se muestra en la gura A.2.

Figura A.2.: Adaptación del ltro supresor de ondas de radio al utilizar super-posición.

137

Vi1 − Vo11

2, 2× 103= (Vo11 − Vo21)× S × C2 + Vo11 × S × C1

Vi1 − Vo11 =

2, 2×103×Vo11×S×C2−2, 2×103×Vo21×S×C2 +2, 2×103×Vo11×S×C1

Vo11

(1 + 2, 2× 103 × S × C2 + 1 + 2, 2× 103 × S × C1

)= (A.1)

Vi1 + 2, 2× 103 × S × C2 × Vo21

Sustituyendo C2 = 10 pF y C1 = 100 pF en A.1,

Vo11 ×(0, 242× 10−6 × S + 1

)= Vi1 + 22× 10−9 × S × Vo21

Vo11 =Vi1 + 22× 10−9 × S × Vo21

0, 242× 10−6 × S + 1(A.2)

(Vo11 − Vo21)× S × C2 =Vo21 − 0

2, 2× 103+ (Vo21 − 0)× S × C1

Vo11 × 2, 2× 103 × S × C2 − Vo21 × 2, 2× 103 × S × C2 =

Vo21 + 2, 2× 103 × Vo21 × S × C1

Vo11 × 2, 2× 103 × S × C2 = (A.3)

Vo21 ×(1 + 2, 2× 103 × S × C2 + 2, 2× 103 × S × C1

)

138

Sustituyendo los valores de C1 yC2 en A.3,

Vo21 =22× 10−9 × Vo11 × S0, 242× 10−6 × S + 1

(A.4)

Sustituyendo A.4 en A.2,

Vo11 =Vi1 + 22× 10−9 × S × 22×10−9×Vo×S

0,242×10−6×S+1

0, 242× 10−6 × S + 1

Vo11 ×(0, 242× 10−6 × S + 1

)2=

Vi1 ×(0, 242× 10−6 × S + 1

)+ 4, 84× 10−16 × Vo11 × S2

Vo11 ×(5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1

)=

Vi1 ×(0, 242× 10−6 × S + 1

)Vo11 =

Vi1 × (0, 242× 10−6 × S + 1)

5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1(A.5)

Luego se halla el aporte de Vi2 a Vo1, el cual llamaremos Vo12. Como el circuitoes simétrico, el impacto de Vi2 en Vo1, sera igual que el de Vi1en Vo2, por lotanto se calcula Vo21.Sustituyendo A.5 en A.4,

Vo21 =22× 10−9 × Vi1×(0,242×10−6×S+1)

5,808×10−14×S2+0,484×10−6×S+1× S

0, 242× 10−6 × S + 1

Vo21 =Vi1 × 22× 10−9 × S

5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1(A.6)

139

Por lo tanto a partir de A.6 y por la simetría mencionada, se obtiene,

Vo12 =Vi2 × 22× 10−9 × S

5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1(A.7)

Sumando A.5 y A.7, se obtiene nalmente,

Vo1 = Vo11 + Vo12 =Vi1 × (0, 242× 10−6 × S + 1) + Vi2 × 22× 10−9 × S

5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1

Análogamente,

Vo2 = Vo21 + Vo22 =Vi2 × (0, 242× 10−6 × S + 1) + Vi1 × 22× 10−9 × S

5, 808× 10−14 × S2 + 0, 484× 10−6 × S + 1

A.1.2. Etapa de amplicación variable

Se calcula la transferencia del circuito que se observa en la gura A.3.

Figura A.3.: Amplicador de ganancia variable.

140

R ‖ 1

S.C=

R

R× C × S + 1

0− ViRx

=(Vi − Vo)× (R× C × S + 1)

R

Rx = Rvar + 103 Ω

−Vi ×R = Rx × Vi × (R× C × S + 1)−Rx × Vo (R× C × S + 1)

Vo × (Rx ×R× C × S +Rx) = Vi × (Rx ×R× C × S +Rx +R)

VoVi

=Rx ×R× C × S +Rx +R

Rx ×R× C × S +Rx

=R× C × S +

(1 + R

Rx

)R× C × S + 1

Sustituyendo por los valores de R y C,

VoVi

=10−4 × S +

(1 + 105

Rx

)10−4 × S + 1

Ahora para hallar la frecuencia de corte,

‖ VoVi‖=(

1 +105

Rx

)× 1√

2=⇒

Tomando S = j × w,

10−8 × w2 + 1 + 2×105

Rx+ 1010

R2x

10−8 × w2 + 1=

(1 +

2× 105

Rx

+1010

R2x

)× 1

2

141

2× 10−8 × w2 + 2 +4× 105

Rx

+2× 1010

R2x

=

(10−8 × w2 + 1

)×(

1 +2× 105

Rx

+1010

R2x

)

10−8 × w2 + 1 +2× 105

Rx

+1010

R2x

= 10−8 × w2 ×(

2× 105

Rx

+1010

R2x

)

10−8 × w2 ×R2x +R2

x +Rx × 2× 105 + 1010 =

10−8 × w2 ×(2× 105 ×Rx + 1010

)

10−8 × w2 ×(R2x − 2× 105 ×Rx − 1010

)= −R2

x −Rx × 2× 105 − 1010

w2 =R2x +Rx × 2× 105 + 1010

−R2x × 10−8 + 2× 10−3 ×Rx + 102

Como Rxvaría entre 1KΩ y 21KΩ, 2× 10−3 ×Rx + 102 R2x × 10−8=⇒

w2 w(Rx + 105)

2

2× 10−3 ×Rx + 102=⇒

fc =Rx + 105

2× π ×√

2× 10−3 ×Rx + 102

A.1.3. Filtro antialiasing

Se calcula la transferencia para el circuito de la gura .

142

Figura A.4.: Filtro antialiasing.

(Vi − V

′

x

)× S × C4 =

V′x

106+V′x − Vx104

106 × Vi × S × C4 − V′

x × S × C4 × 106 = V′

x + 100× V ′x − 100× Vx

V′

x

(106 × S × C4 + 101

)= 106 × Vi × S × C4 + 100× Vx

V′

x =106 × Vi × S × C4 + 100× Vx

106 × S × C4 + 101(A.8)

Sustituyendo el valor de C4 = 1µF en A.8,

V′

x =Vi × S + 100× Vx

S + 101(A.9)

Vx − V +

15× 103= V + × S × C3

Vx = V + ×(15× 103 × S × C3 + 1

)(A.10)

143

Sustituyendo el valor de C3 = 220 nF en A.10,

Vx = V + ×(3, 3× 10−3 × S + 1

)(A.11)

Sustituyendo A.11 en A.9,

V′

x =Vi × S + 100× V + × (3, 3× 10−3 × S + 1)

S + 101= (A.12)

Vi × S + V + × (0, 33× S + 100)

S + 101

V′x − Vx104

= (Vx − Vo)× S × C1 + V + × S × C3 (A.13)

Sustituyendo los valores de C3y C1 = 33 nF en A.13,

V′

x − Vx =

0, 33× 10−3 × Vx × S − 0, 33× 10−3 × Vo × S + 2, 2× 10−3 × V + × S

2, 2× 10−3 × V + × S = (A.14)

V′

x−V +×(3, 3× 10−3 × S + 1

)×(0, 33× 10−3 × S + 1

)+0, 33×10−3×S×Vo

Sustituyendo A.11 en A.14,

144

2, 2× 10−3 × V + × S + V + ×(1, 089× 10−6 × S2 + 3, 63× 10−3 × S + 1

)=

V′

x + 0, 33× 10−3 × Vo × S

V +×(1, 089× 10−6 × S2 + 5, 83× 10−3 × S + 1

)= V

′

x + 0, 33× 10−3×Vo×S(A.15)

Sustituyendo A.12en A.15,

V + ×(1, 089× 10−6 × S2 + 5, 83× 10−3 × S + 1

)=

Vi × S + V + × (0, 33× S + 100)

S + 101+ 0, 33× 10−3 × Vo × S

V + ×(1, 089× 10−6 × S2 + 5, 83× 10−3 × S + 1

)× (S + 101) =

Vi × S + V + × (0, 33× S + 100) + 0, 33× 10−3 × Vo × S × (S + 101)

V + ×(1, 089× 10−6 × S23 + 5, 94× 10−3 × S2 + 1, 589× S + 101

)=

Vi×S+V +× (0, 33× S + 100) +Vo×(0, 33× 10−3 × S2 + 33, 33× 10−3 × S

)

V + ×(1, 089× 10−6 × S23 + 5, 94× 10−3 × S2 + 1, 259× S + 1

)=

Vi × S + Vo ×(0, 33× 10−3 × S2 + 33, 33× 10−3 × S

)

145

V + =Vi × S + Vo × (0, 33× 10−3 × S2 + 33, 33× 10−3 × S)

1, 089× 10−6 × S23 + 5, 94× 10−3 × S2 + 1, 259× S + 1(A.16)

V + = V − =⇒ 0− V +

8, 2× 103=

V + − Vo100× 103

+(V + − Vo

)× (S × C2)

−V +

8, 2=V + − Vo + 100× 103 × S × C2 (V + − Vo)

100

−12, 195×V + = V +×(100× 103 × S × C2 + 1

)−Vo×

(100× 103 × S × C2 + 1

)

V0×(100× 103 × S × C2 + 1

)= V +×

(100× 103 × S × C2 + 13, 195

)(A.17)

Sustituyendo A.16 en A.17,

V0 ×(100× 103 × S × C2 + 1

)= (A.18)

Vi×(100×103×S2×C2+13,195×S)+Vo×(0,33×10−3×S2+33,33×10−3×S)×(100×103×S×C2+13,195)1,089×10−6×S23+5,94×10−3×S2+1,259×S+1

Sustituyendo C2 = 1 nF en A.18,

Vo(0, 1× 10−3 × S + 1

) (1, 089× 10−6 × S23 + 5, 94× 10−3 × S2 + 1, 259× S + 1

)=

Vi×(0, 1× 10−3 × S2 + 13, 195× S

)+Vo×

(33× 10−9 × S3 + 4, 3577× 10−3 × S2 + 0, 43979× S

)

146

Vo×(0, 1089× 10−9 × S4 + 1, 683× 10−6 × S3 + 6, 0659× 10−3 × S2 + 1, 2591× S + 1

)=

Vi×(0, 1× 10−3 × S2 + 13, 195× S

)+Vo×

(33× 10−9 × S3 + 4, 3577× 10−3 × S2 + 0, 43979× S

)

Vo×(0, 1089× 10−9 × S4 + 1, 65× 10−6 × S3 + 1, 7082× 10−3 × S2 + 0, 81931× S + 1

)=

Vi ×(0, 1× 10−3 × S2 + 13, 195× S

)VoVi

= 0,1×10−3×S2+13,195×S0,1089×10−9×S4+1,65×10−6×S3+1,7082×10−3×S2+0,81931×S+1

A.1.4. Exigencia de CMRR

ACM×VCM ≤ α×V0 = α×Ad.Vd =⇒ 20×log(ACM×VCM) ≤ 20×log(α×Ad.Vd) =⇒

20× log

(1

ACM × VCM

)≥ 20× log

(1

α× Ad.Vd

)(A.19)

CMRR = 20× log

(AdACM

)= 20× log

(Ad × VCMACM × VCM

)(A.20)

De A.19y A.20, se desprende que:

CMRR ≥ 20× log

(Ad × VCMα× Ad × Vd

)= 20× log

(VCMα× Vd

)(A.21)

A.1.5. Cálculo del ruido producido por losamplicadores operacionales

En la sección 7.4.7.1 se deseaba conocer el valor RMS del ruido provocadopor los integrados para compararlo con la señal de interés. El valor RMS de

147

un ruido de voltajeVn(t), se calcula como:

VnRMS =

√√√√√ 1

T×

T

0

vn(t)2.dt

Análogamente,

InRMS =

√√√√√ 1

T×

T

0

in(t)2.dt

Si se tiene más de una fuente de ruido sumadas en un punto, digamos vn1(t)y vn2, el valor RMS equivalente se calcula como:

VnRMS =

√√√√√ 1

T×

T

0

(vn1(t) + vn2(t))2 .dt =

√√√√√ 1

T×

T

0

(vn1(t)2 + vn2(t)2 + 2.vn1(t).vn2(t)) .dt =

√√√√√ 1

T×

T

0

vn1(t)2.dt+1

T×

T

0

vn2(t)2.dt+2

T×

T

0

2.vn1(t).vn2(t).dt = (A.22)

√√√√√V 2nRMS1 + V 2

nRMS2 +2

T×

T

0

2.vn1(t).vn2(t).dt (A.23)

Si los ruidos no son correlacionados, como sucede generalmente entonces,

148

2

T×

T

0

vn1(t).vn2(t).dt = 0 =⇒ VnRMS =√V 2nRMS1 + V 2

nRMS2

Análogamente para ruidos de corriente InRMS1e InRMS, no correlacionados,

InRMS =√InRMS1 + InRMS2

Para calcular los valores RMS de ruido, los fabricantes de integrados incluyenen las hojas de datos de los componentes, las densidades espectrales de potenciade los ruidos de tensión y corriente.Siendo Vn(f) e In(f) las densidades espectrales de potencia de los ruidos de

tensión y de corriente vn(t) e in(t), se puede calcular sus valores RMS parauna banda de frecuencia de fLa fHcomo:

VnRMS =

√√√√√ fHˆ

fL

Vn(f)2.df (A.24)

InRMS =

√√√√√ fHˆ

fL

In(f)2.df

La ecuación de las densidades espectrales de potencia de los ruidos provoca-dos por los integrados se aproxima como:

Vn(f) =

√V 2wn ×

(fcvf

+ 1

)

In(f) =

√I2wn ×

(fcif

+ 1

)

149

En la sección 7.4.7.1 se explica los signicados de las variables, y en la gura7.14 se observa el modelo para el ruido de los amplicadores operacionales.Luego utilizando la ecuación A.24,

VnRMS =

√√√√√ fHˆ

fL

V 2wn ×

(fcvf

+ 1

)df = Vwn ×

√√√√√ fHˆ

fL

fcvf.df +

fHˆ

fL

1.df =

Vwn×√fcv × Ln(f) |fHfL +f |fHfL = Vwn×

√fcv × (Ln(fH)− Ln(fL)) + fH − fL =

VnRMS = Vwn ×

√fcv × Ln

(fHfL

)+ fH − fL (A.25)

Realizando el mismo razonamiento para el ruido de corriente,

InRMS=Iwn ×

√fci × Ln

(fHfL

)+ fH − fL (A.26)

Como nos interesa el ruido de tensión, se calcula la tensión (V InRMS) provo-cada por el ruido de corriente. Observando la gura 7.14, se nota que el ruidode corriente es provocado por dos fuentes de corriente (inn e inp). Estas no soncorrelacionadas por lo que:

InRMS =√I2nnRMS + I2

npRMS =⇒ V InRMS =√V I2

nnRMS + V I2npRMS =

√I2nnRMS ×R2

n + I2npRMS ×R2

p

150

Rn y Rp son las resistencias vistas por las fuentes de corriente. Las densida-des espectrales de los ruidos de corriente no son exactamente iguales, pero sison muy parecidas y siempre se toman como iguales, por lo tanto,

I2nnRMS = I2

npRMS = I2nRMS =⇒

V InRMS =√

(R2p +R2

n)× I2nRMS (A.27)

Sustituyendo A.26 en A.27,

V InRMS =

√(R2

p +R2n)× I2

wn ×(fci × Ln

(fHfL

)+ fH − fL

)

Ahora para hallar el ruido de tensión equivalente (VnRMST ), combinamos losruidos V InRMS y VnRMS, los cuales tampoco son correlacionados,por lo que,

VnRMST =√V I2

nRMS + V 2nRMS (A.28)

Finalmente sustituyendo A.25y A.26 en A.28,

VnRMST =

√V 2wn ×

(fcv × Ln

(fHfL

)+ fH − fL

)+ (R2

p +R2n)× I2

wn ×(fci × Ln

(fHfL

)+ fH − fL

)

151

A.2. Manual de usuario

A.2.1. Instrucciones de puesta en marcha

La primera acción que debe realizar el usuario es ponerse la vincha sobrela cabeza, la presión de los electrodos debe ser ejercida lo más directo posiblesobre el cuero cabelludo, por lo que se recomienda correr el pelo con los dedos.Una vez que se encontró una posición adecuada y cómoda, se debe apretar lavincha con la rosca delantera, de forma tal que el contacto sea bueno, peroteniendo la precaución de no lastimarse.

Los electrodos montados sobre la vincha, poseen cables con conectores deltipo DIN de 5 pines, y un electrodo de referencia con conector AUX machode 3.5 mm. Los conectores DIN y AUX mencionados pueden apreciarse en lagura A.5, de izquierda a derecha respectivamente.

Figura A.5.: Conectores de utilizados para los electrodos.

Los conectores deben insertarse en la placa de amplicación y digitalización(electroencefalógrafo). Esta caja tiene 4 conectores DIN machos y 2 conectoreshembra de 3,5mm (ver gura A.6). Cada conector DIN corresponde a un canaly los conectores de 3,5 mm corresponden al electrodo de referencia (o DRL).Si el electroencefalógrafo está siendo utilizado por una sola persona, solo senecesita conectar un electrodo de referencia, si se utiliza para dos personas, se

152

deben utilizar las dos entradas para los electrodos de referencia de cada usuario.

Figura A.6.: Conectores del electroencefalógrafo.

El electroencefalógrafo fue diseñado para soportar desde 1 hasta 4 canalesde medida. En caso de que se desee utilizar menos, se deben utilizar las cone-xiones del canal 1, hasta la cantidad deseada. Luego, cada uno de los canalesutilizados corresponderá a una gráca en el software de monitoreo.

También se debe conectar el dispositivo BT de la computadora a un puertoUSB. El mismo tiene una luz roja, la cual debe empezar a parpadear a unafrecuencia constante.

A.2.2. Alimentación del electroencefalógrafo

El dispositivo se puede alimentar de 2 formas (ver gura A.7), mediantebaterías, utilizando una pila de entre 7 a 12v o 6 pilas AAA. La segunda alter-nativa es conectar el electroencefalógrafo al puerto USB de una computadora,mediante un cable USB de impresora, en este caso se recomienda no enchufarla computadora a la red eléctrica, para evitar variaciones en la alimentación e

153

inducción de ruido. Se debe tener en cuenta que cuando se alimenta el electro-encefalógrafo, también se alimentan los electrodos de la vincha y el dispositivode comunicación BT.

Figura A.7.: Conectores de alimentación del electroencefalógrafo.

Pasados 10 segundos de la alimentación del electroencefalógrafo, los módulosBT deberían conectarse automáticamente, provocando que el parpadeo del leddel módulo BT del PC cambie a series de dos parpadeos cada cierto tiempo.En caso que esto no ocurra, verique el estado de las baterías.

A.2.3. Software de monitoreo

A.2.3.1. Descripción

El software de monitoreo permite la visualización y almacenamiento de losbiopotenciales y de su potencia en la banda Alfa. La visualización puede rea-lizarse midiendo señales en tiempo real o utilizando una grabación antigua.También funciona como servidor esperando conexiones al puerto 5000 median-te sockets, para poder conectarse a distintos programas controladores y así

154

formar una BCI. Se incluyen algoritmos para la detección de la activación ydesactivación del ritmo alfa así como también la detección de 2 y 3 parpadeos.En caso de tener un cliente conectado, y detectarse alguna de las situacionesdescritas anteriormente, se envían datos dicho cliente para informar del eventoocurrido.

A.2.3.2. Requerimientos

Para utilizar el software de monitoreo se requiere:

Instalar el ambiente de ejecución de java (JRE). El software se probó conlas versiones 7 y 8, pero debería funcionar con versiones anteriores. ElJRE puede descargarse de la siguiente página:http://www.oracle.com/technetwork/java/javase/downloads/

En caso de utilizar el sistema operativo Windows, se debe descargarrxtxSerial.dll y colocarla en la ruta C:\Program Files\Java\jre8\bin(suponiendo que se descargó la versión 8 de 64 bits del JRE y se instalóen la ruta por defecto). Ya sea para 32 o 64 bits, el dll mencionado sepuede descargar de siguiente la página:http://jlog.org/rxtx-win.html.En caso de utilizar Linux, las instrucciones para instalar la librería RXTXpueden encontrarse en la siguiente página:http://www.agaveblue.org/howtos/Comm_How-To.shtml.

Instalar driver para la placa Arduino Atmega2560, se recomienda des-cargar la versión 1.0.5 del software de Arduino que ya incluye los drivers,el cual puede ser descargado en la siguiente página:http://arduino.cc/en/Main/Software#toc2

A.2.3.3. Funcionamiento

EnWindows, como primer paso se debe ejecutar el archivo PoyectoEEG.jarel programa. En Linux para ejecutar el programa, se debe abrir la consola, si-tuarla en la carpeta donde se encuentra los archivos del programa, e introducirla siguiente línea:

java -jar ProyectoEEG.jar

155

Al ejecutar el programa se encuentra la siguiente pantalla (ver gura A.8):

Figura A.8.: Pantalla principal.

Una vez que se ejecuta el programa, el segundo paso consiste en seleccionar elpuerto de comunicación, la cantidad de canales que se utilizarán y la velocidadde comunicación. Para ello se debe ingresar al menú Conguración, opciónConguración de la conexión (Ctrl+C) como se muestra en la gura A.9.

Figura A.9.: Congurar la conexión.

Si no hay puertos conectados, el programa informa mediante una ventana(ver gura A.10). En dicho caso verique la conexión USB de módulo BT.

156

Figura A.10.: Ventana que informa que no hay puertos conectados.

En ambos casos, se abre la ventana de conexión (ver gura A.11), la diferenciaes que si no hay puertos conectados, no aparecerá ninguna elección en Puertos,se puede buscar nuevamente puertos disponible presionando el botón Buscar.

Figura A.11.: Ventana de conexión.

Luego se selecciona el puerto en el cual está conectado el módulo de comuni-cación BT. Se debe indicar la velocidad de la comunicación. La velocidad parael módulo BT elegido en este proyecto es de 230400 bps, si se desea modicar,se debe cambiar la conguración de los módulos de comunicación, además decambiar la velocidad de transmisión en el software del microcontrolador. Tam-bién se elige la cantidad de canales conectados al sistema, seleccionando unaopción en Cantidad de canales. Finalmente se presiona el botón Conectar para

157

establecer la conexión con el microcontrolador y se cierra la ventana. Luegoen la pantalla principal se presiona el botón Capturar para que el programacomience a gracar los datos recibidos, así como la potencia en la banda Alfade las señales. Los botones de Stop y Play proporcionan la funcionalidad parapoder pausar y analizar cualquier sección de los datos obtenidos, y luego reto-mar la captura de datos.

Dentro de Menú, seleccionando la opción Abrir EEG (ver gura A.12), sepuede levantar un registro anterior y analizar el mismo.

Figura A.12.: Opción abrir registro anterior o aceptar cliente.

La grabación se ejecuta a la misma velocidad que se vería si se estuviesemidiendo en tiempo real, y puede pararse utilizando la opción Stop en la pan-talla principal. Para analizar cualquier sección de la grabación el programacuenta con un scroll en la parte superior de la ventana. La reproducción de lagrabación puede retomarse utilizando botón Play, en la pantalla principal.

Las grabaciones no tienen por qué haber sido hechas con el software propio,puede levantarse cualquier grabación que cumpla con el formato utilizado. Eneste sentido, el programa utiliza el formato tipo empleado por Scilab o Scicos-lab, es decir, un archivo de texto con un valor de medida por canal en cadala (separados por tabuladores).

Finalmente, en cualquier caso, ya sea en tiempo real o grabado, el programamuestra en las grácas de la izquierda las señales en el tiempo, correspondientea cada canal. En las ventanas del sector derecho, se muestra la potencia delritmo alfa correspondiente al canal de la izquierda, en color celeste, mientrasque en rojo se muestra el nivel de alfa de background. En la gura A.13 semuestra un ejemplo de dos señales medidas.

158

Figura A.13.: Grácas típicas.

En cualquier momento puede conectarse mediante un socket, un cliente alpuerto 5000 para recibir distintos eventos del usuario y que el cliente reali-ce distintas tareas en función de estos eventos. Para aceptar la petición deconexión del cliente, se debe seleccionar la opción Aceptar cliente (Ctrl+A),dentro de Menú (ver gura A.12).Si no se encuentra ningún cliente conectado,aparecerá el mensaje que se observa en la gura A.14. La opción Testear BCI(Ctrl+R), dentro del menú Conguración (ver gura A.15) puede utilizarsepara comprobar la conexión entre cliente y servidor, y para probar el progra-ma del cliente.

Figura A.14.: Mensaje no hay cliente conectado.

159

Figura A.15.: Opción Envío al cliente.

Una vez nalizado el monitoreo, al cerrar el programa se puede elegir siguardar o no los datos obtenidos, eligiendo el nombre del archivo y su ruta.

A.2.4. Programa de prueba para la BCI [17]

Para probar la BCI se realizó un programa que corre en un controlador LEGONXT y un software de computadora que actuará como cliente conectándose alpuerto 5000, encargado de recibir los eventos identicados por el software demonitoreo y transmitirlo al NXT mediante comunicación BT.

Los programas mencionados anteriormente son programados en el lenguajede programación JAVA. Para programar el robot LEGO en este lenguaje, seutiliza la máquina virtual de JAVA leJOS NXJ, la cual es desarrollada por elproyecto de código abierto leJOS y fue programado en Windows 7.

A.2.4.1. Requerimientos

Para utilizar leJOS NXJ se requiere:

Instalar el kit de desarrollo de Java (JDK) de 32 bits, versión 1.5 osuperior. El mismo puede descargarse de la siguiente página:http://www.oracle.com/technetwork/java/javase/downloads/

Para correr leJOS NXJ en Windows se necesita un controlador USB enla PC. El mismo puede descargarse en la página:http://www.lego.com/en-us/mindstorms/downloads/nxt/nxt-fantom-driver/Si se utiliza un Windows, al ejecutar el instalador del driver, se puedegenerar un problema diciendo que no se encuentra LegoMindstormsNXT-driver64Supp.msi, en este caso se debe abrir el archivo setup.ini y luego dela línea [LegoMindstormsNXTdriver64Supp.msi] se debe insertar el si-guiente código Path=Products\LEGO_NXT_Driver_64\NXT_D02\LegoMindstormsNXTdriver64.msi(sin las comillas). Luego al instalar el driver, si se utiliza Windows de

160

64 bits, se debe desactivar la opción LEGO MINDSTORMS NXT x64Driver Support, la cual no es necesaria.

Para realizar la comunicación BT se necesita un pendrive BT si el PCno lo trae integrado, además se requiere una pila BT, en el proyecto seutilizó la pila de Microsoft.

Instalación de leJOS NXJ, el cual puede descargarse en la página:http://sourceforge.net/projects/lejos/les/lejos-NXJ/Se recomienda descargar el ejecutable de la última versión, y dentro de laúltima versión, el archivo de nombre leJOS_NXJ_0.9.1beta-3_win32_setup.exe.

A.2.4.2. Instrucciones de puesta en marcha del robot

A continuación se explican las instrucciones que deben ejecutarse solamentela primera vez.

Cuando se naliza la instalación de leJOS NXJ, se da la opción LaunchNXJ Flash utility, en caso de que el controlador LEGO no tenga el rmwarede leJOS (o tenga una versión antigua), se debe marcar esta opción y seguirlos pasos que aparecen. Es necesario que el LEGO esté conectado al PC me-diante cable USB y este prendido. Si se desea volver a ashear el NXT o sedesea ashear otro NXT se puede escribir en la consola nxjashg y se abriráel programa de asheo.

Luego se debe emparejar el PC con el NXT mediante BT (ya sea integradoo pendrive BT). Es importante corroborar que el NXT este en modo visible yvericar de conocer el pin. Por defecto este último es 1234.

Finalmente se debe cargar el programa al NXT. Como en el paso anteriorya se emparejaron el PC y el NXT, se puede cargar el programa vía BT. Elnombre del mismo es ProgramaNXT. Para cargarlo, se debe abrir una con-sola y situado en la carpeta donde se encuentre el archivo del programa, paraluego introducir la siguiente línea:

nxjupload -r ProgramaNXT.nxj.

El programa se cargará y ejecutará automáticamente. Si se desea que el pro-grama no se ejecute sólo luego de cargarlo, se debe quitar el -r.

161

A.2.4.3. Instrucciones de uso

Como se mencionó en A.2.4.2 esas instrucciones sólo deben realizarse la pri-mera vez, en las siguientes ocasiones sólo se deben seguir las instrucciones quese detallan aquí.

Se debe ejecutar el software de monitoreo (para esto se deben seguir los pa-sos explicados antes en este manual). Finalmente se debe ejecutar el programaque comunica el PC con el LEGO y que funciona como cliente del software demonitoreo. El nombre del programa es ComunicacionPC_NXT. Para esto sedebe abrir la consola en la carpeta donde se encuentra el programa e introducirla siguiente línea:

nxjpc ComunicacionPC_NXT.

El programa de comunicación realizara la petición de conexión al softwarede monitoreo (servidor) automáticamente y luego para que este acepte la co-nexión se debe seleccionar la opción Aceptar cliente, dentro de Menú.

162

A.3. Código del microcontrolador ATmega

2560

163

volatile byte voltageH = 0; volatile byte voltageL = 0; volatile byte cantidadCanales = 1; volatile byte canalActual = 1; volatile unsigned long int microsegundos = 0; volatile unsigned long int microsTotal = 0; volatile int contador = 0; volatile boolean conversionTerminada = false; volatile byte datosRecibidos = 0; volatile boolean envioDatos = false; // the setup routine runs once when you press reset: void setup() cli(); //Deshabilito las interrupciones //power reduction PRR1 = B00111101; PRR0 = B10101110; //CONFIGURACIÓN ADC ADCSRB = B00000000; //habilito ADC bitSet(ADCSRA,ADEN); //autotrigger off bitClear(ADCSRA,ADATE); //elijo Aref = AVcc, 8 bits de resolución y el como canal ADC0 ADMUX = B01000001; //habilito la interrupciónde conversión completa bitSet(ADCSRA,ADIE); //CONFIGURACIÓN SERIAL UBRR2H = 0; UBRR2L = 8; //cargo 8 para vel = 230,4KBaud con u2x2=1 bitSet(UCSR2A,U2X2); //indico que se agregue el bit de paridad bitSet(UCSR2C,UPM21); //Habilito la transmisión bitSet(UCSR2B,TXEN2); //Habilito la recepción bitSet(UCSR2B,RXEN2);

//Setup Timer2 / TCCR2B = 0x00; //Deshabilito el timer 2 durante la configuración TIFR2 = 0x00; //Limpio la bandera de interrupción de overflow TIMSK2 = 0x01; //Habilito la interrupción de overflow TCCR2A = 0x00; //Modo normal de operación sei(); //Habilito las interrupciones //prendo el pin 13 que se utiliza para generar Vref pinMode(13,OUTPUT); digitalWrite(13,1); ISR(TIMER2_OVF_vect) //Se utiliza el delay para lograr la frecuencia de muestreo lo más cercana posible a 1024 Hz delayMicroseconds(3); //si estoy enviando datos, comienzo la conversión if(envioDatos==true) bitSet(ADCSRA,ADSC); TCNT2 = 135; //cargo 135 como carga inicial ISR(ADC_vect) //limpio la bandera de interrupción bitClear(ADCSRA,ADIF); //leo la conversión voltageL = ADCL; voltageH = ADCH; //indico que se realizo una conversión conversionTerminada = true; // the loop routine runs over and over again forever: void loop() //mientras no se termino la conversión o el envío esta detenido while((conversionTerminada == false) || (envioDatos == false)) //recibí datos if(UCSR2A & (1<<RXC2)) //leo el registro donde se reciben los datos datosRecibidos = UDR2; if(datosRecibidos>0 && datosRecibidos<5) //Preescaler a 128 y comienza a andar el timer 2

TCCR2B = 0x05; //comienzo el muestreo y el envío de datos envioDatos=true; //defino la cantidad de canales cantidadCanales = datosRecibidos; else if(datosRecibidos == 5) //detendo el envío de datos y las conversiones envioDatos=false; //reinicio las variables canalActual=1; ADMUX = B01000001; conversionTerminada = false; //si el muestreo fue del canal uno if(canalActual == 1) //envío bytes de principio de la trama mandarByte2(255); mandarByte2(255); //envío bytes alto y bajo de la muestra mandarByte2(voltageL); mandarByte2(voltageH); //si el muestreo fue de otro canal else //envío bytes alto y bajo de la muestra mandarByte2(voltageL); mandarByte2(voltageH); //luego de enviar los datos conversionTerminada = false; canalActual++; if(canalActual > cantidadCanales) //si ya se muestrearon todos los canales vuelvo a muestrear el primero ADMUX = B01000001; canalActual = 1; else //de lo contrario slecciono el próximo canal de muestreo ADMUX++; //comienzo la conversión bitSet(ADCSRA,ADSC); void mandarSerialln(String unTexto) int i = 0; while(i <= unTexto.length()) while ( !( UCSR0A & (1<<UDRE0)) );//mientras que el registro donde se escribe lo que se desea transmitir esta lleno

if(i<unTexto.length()) UDR0 = unTexto[i]; else UDR0 = '\n'; i++; void mandarSerial(String unTexto) int i = 0; while(i < unTexto.length()) while ( bitRead(UCSR0A,UDRE0)==0 );//mientras que el registro donde se escribe lo que se desea transmitir esta lleno UDR0 = unTexto[i]; i++; void mandarByte(byte unByte) while ( !( UCSR0A & (1<<UDRE0)) );//mientras que el registro donde se escribe lo que se desea transmitir esta lleno UDR0 = unByte; void mandarByte2(byte unByte) while ( !( UCSR2A & (1<<UDRE2)) );//mientras que el registro donde se escribe lo que se desea transmitir esta lleno UDR2 = unByte; void mandarSerialln2(String unTexto) int i = 0; while(i <= unTexto.length()) while ( !( UCSR2A & (1<<UDRE2)) );//mientras que el registro donde se escribe lo que se desea transmitir esta lleno if(i<unTexto.length()) UDR2 = unTexto[i]; else //UDR2 = '\n'; i++;

A.4. Códigos de Scicoslab para ltros digitales

168

A.4.1. Código de ltro Gauss

169

//*************************************************** //*****************Filtro Gauss********************** clear; clf; format('v',25); // Frecuencia de muestreo fm = 256; //Numero de puntos y vector de phi Np = 10000; v_phi = 1/Np*[0:Np-1]; numPBO = poly([0.000081],'z','cooef'); denPBO = poly([0.991 0.991],'z','root'); H_z_GaussO =numPBO/denPBO v_h_PBO = freq(numPBO,denPBO,exp(%i*2*%pi*v_phi)); numPBF = poly([0.000049],'z','cooef'); denPBF = poly([0.993 0.993],'z','root'); H_z_GaussF =numPBF/denPBF v_h_PBF = freq(numPBF,denPBF,exp(%i*2*%pi*v_phi)); scf(0); clf(); xgrid(); plot2d(v_phi(1:Np/256)*fm,abs(v_h_PBO(1:Np/256)),5) plot2d(v_phi(1:Np/256)*fm,abs(v_h_PBF(1:Np/256)),2) xtitle('Transferencia','Hz','Modulo') u = ones(1:Np); n = [1:Np]; v_u_PBO = rtitr(numPBO,denPBO,u); T3_PBO = 1; while(v_u_PBO(T3_PBO)<0.95) T3_PBO =T3_PBO+1; end T3_PBO

scf(1); clf(); xgrid(); plot2d(n/fm,v_u_PBO(1:Np),2) plot2d3(T3_PBO/fm,1) plot2d(n/fm,0.95*u) xtitle('Respuesta al escalón','tiempo(s)','Amplitud') // Respuesta a sin(x)^2 f = 10/fm; n = [1:Np]; u = sin(2*%pi*f*n); U = u**2; g = rtitr(numPBO,denPBO,U); // Metodos para buscar tiempo de asentamiento (95 % valor final) // En regimen la potencia de un sin(x) es 1/2 P = 0.5*0.95; Te=0; for i=1:Np if(g(i)> P) then Te = i; break; end end Te/fm scf(3); clf(); xgrid(); plot2d(n(1:Te*2)/fm,g(1:Te*2)) plot2d(n(1:Te*2)/fm,ones(1:Te*2)*P,5) plot2d3(Te/fm,P,5) xtitle('','tiempo(s)','Modulo(W)')

A.4.2. Código Filtro IIR pasa bajo

172

//**************************************************************** //*************Diseño de un filtro pasa-bajo IIR****************** //***************Parametros del tiempo discreto******************* // Limpiar consola clear; clf; // Cambiar formato format('v',25); // Parametros fm = 1024; fp = 40; fs = 50; delta_p = 0.01; delta_s = 0.01; // Generar filtro pasa bajo exec('GeneraPasaBajoIIR.sci'); [num,den] = pasaBajoIIR(fp,fs,10,delta_p,delta_s,3,fm); // Grafo de la solucion obtenida Np = 1000; v_phi = 1/Np*[0:Np-1]; v_h_phi = freq(num,den,exp(%i*2*%pi*v_phi)); scf(0); clf(); xgrid(); L = 59; plot2d(v_phi(1:L)*fm,abs(v_h_phi(1:L)),2) plot2d(v_phi(1:L)*fm,ones(1:L)*(1+delta_p),5) plot2d(v_phi(1:L)*fm,ones(1:L)*(1-delta_p),5) plot2d(v_phi(1:L)*fm,ones(1:L)*(delta_s),5) plot2d3(fp,1+delta_p,5) plot2d3(fs,1+delta_p,5) xtitle('Transferencia','Hz','Modulo') // Grafico de amplitudes en escala logaritmica

scf(1); clf(); xgrid(); plot2d(v_phi(1:Np/2),abs(v_h_phi(1:Np/2)),2, logflag='nl') plot2d(v_phi(1:Np/2),ones(1:Np/2)*(1+delta_p),5,logflag='nl') plot2d(v_phi(1:Np/2),ones(1:Np/2)*(1-delta_p),5,logflag='nl') plot2d(v_phi(1:Np/2),ones(1:Np/2)*(delta_s),5,logflag='nl') plot2d3(fs/fm,1+delta_p,5,logflag='nl') plot2d3(fp/fm,1+delta_p,5,logflag='nl') xtitle('Modulo de la transferencia en escala logaritmica') // fase del filtro zs = %e**(%i*2*%pi*v_phi); imgIIR = horner(num,zs)./horner(den,zs); [fase,dB] = phasemag(imgIIR,'c'); scf(2); clf(); xgrid(); plot2d(v_phi*fm,fase,5) xtitle('Fase de la transferencia','frecuencia(Hz)','fase(grad)') // Respuesta al sin f = 10/fm; n = [1:Np]; u = sin(2*%pi*f*n); g = rtitr(num,den,u); L = length(g); a = 1; for i=2:L-1 if(abs(g(i-1))<abs(g(i)) & abs(g(i))>abs(g(i+1))) then j(a) = abs(g(i)); N(a) = i; a=a+1; end end scf(4); clf(); xgrid(); plot2d(n/fm,(g)) plot2d(n/fm,u); plot2d(n/fm,0.95*ones(1:Np)) plot2d(N/fm,j,5) xtitle('Respuesta a sin( )','Tiempo(s)','Amplitud')

A.4.3. Código ltro IIR pasa banda

175

//***************************************************************** //*************Diseño de un filtro pasa-banda IIR****************** //***************Parametros del tiempo discreto******************** clear; clf; format('v',25); // Filtro pasa banda // Frecuencia de muestreo fm = 256 // Inicio y fin banda de paso fp1 = 9; fp2 = 11; // Fin y Inicio de banda de rechazo fs1 = 7; fs2 = 13; // Error delta_p = 0.01; delta_s = 0.01; exec('GeneraPasaBandaIIR.sce'); [num, den] = pasaBandaIIR(fs1,fs2,fp1,fp2,2,delta_p,delta_s,4,fm); H_z = num/den // Grafo de la solucion obtenida Np = 1000; v_phi = 1/Np*[0:Np-1]; v_h_phi = freq(num,den,exp(%i*2*%pi*v_phi)); scf(1); clf(); xgrid(); plot2d(v_phi(1:Np/2)*fm,abs(v_h_phi(1:Np/2)),2) plot2d(v_phi(1:Np/2)*fm,ones(1:Np/2)*(1+delta_p),5) plot2d(v_phi(1:Np/2)*fm,ones(1:Np/2)*(1-delta_p),5) plot2d(v_phi(1:Np/2)*fm,ones(1:Np/2)*(delta_s),5) plot2d3(fs1,1+delta_p,5)

plot2d3(fs2,1+delta_p,5) plot2d3(fp1,1+delta_p,1) plot2d3(fp2,1+delta_p,1) xtitle('Transferencia','Hz','Magnitud') f = 9/fm; n = [1:Np]; u = 512*ones(1:Np);//sin(2*%pi*f*n); g = rtitr(num,den,u); L = length(g); a = 1; for i=2:L-1 if(abs(g(i-1))<abs(g(i)) & abs(g(i))>abs(g(i+1))) then j(a) = abs(g(i)); N(a) = i; a=a+1; end end scf(2); clf(); xgrid(); plot2d(n/fm,(g)) plot2d(N/fm,j,5) xtitle('Respuesta a sin( )','Tiempo(s)','Amplitud') // fase del filtro zs = %e**(%i*2*%pi*v_phi); imgIIR = horner(num,zs)./horner(den,zs); [fase,dB] = phasemag(imgIIR,'c'); scf(3); clf(); xgrid(); plot2d(v_phi*fm,fase,5) xtitle('Fase de la transferencia','frecuencia(Hz)','fase(grad)')

A.4.4. Código para generar ltro pasa bajo IIR

179

//*********************************************************************** //*****************Funcion crear filtro pasa_bajos IIR******************* // fs frecuencia fin banda de paso Hz // fp frecuencia comienzo banda de paso Hz // of orden del filtro // delta_p error banda pasante // delta_s error banda de rechazo // tipoFiltro (1 butt) (2 cheb1) (3 cheb2) (4 ellip) // fm frecuencia de muestreo Hz function [num,den] = pasaBajoIIR(fp,fs,of,delta_p,delta_s,tipoFiltro,fm) // defino fin del la banda de rechazo phi_s = fs/fm; // defino comienzo de la banda de paso phi_p = fp/fm; //*********Transformo los parametros al tiempo continuo*********** // defino T parametro para mantener las dimensiones T = 10**-4; f_s = 1/(%pi*T)*tan(%pi*phi_s); f_p = 1/(%pi*T)*tan(%pi*phi_p); delta_1 = (2*delta_p)/(1+delta_p); delta_2 = delta_s/(1+delta_p); // defino delta_g, parametro para centrar el error de la banda de // paso en 1. delta_g = 1 + delta_p; // diseño del filtro pasa-bajos del tiempo continuo utilizando la funcion // predefinida analpf();

// defino vector de error de rizado permitido en las bandas de paso y rechazo ve = [delta_1 delta_2]; // utilizo funcion analpf(); filtro = 'butt'; omega = f_p; if(tipoFiltro == 2) then filtro = 'cheb1'; else if(tipoFiltro == 3) then filtro = 'cheb2'; omega = f_s else if(tipoFiltro == 4) then filtro = 'ellip'; end end end [h_s, p_s, c_s,g_s] = analpf(of,'cheb2',ve,2*%pi*(omega)); // paso al dominio del tiempo discreto // caso no hay ceros en 's' if length(c_s) == 0 then v_c = (-1)*ones(1:length(p_s)); v_p = (1+(T/2)*p_s)./(1-(T/2)*p_s); G = g_s/abs(prod((2/T)-p_s)); end // caso hay ceros en 's' if length(c_s) <> 0 then v_c_aux = (-1)*ones(1:length(p_s)-length(c_s)); v_c_aux2 = (1+(T/2)*c_s)./(1-(T/2)*c_s); v_c = [v_c_aux v_c_aux2]; v_p = (1+(T/2)*p_s)./(1-(T/2)*p_s); G = g_s*abs(prod((2/T)-c_s))/abs(prod((2/T)-p_s)); end

// correcion de ganacia para lograr rizado centrado en 1 G = G*delta_g; // creacion de los polinomios numerador y denominador de // la transferencia del tiempo discreto num = G*real(poly(v_c,'z','roots')); den = real(poly(v_p,'z','roots')); endfunction;

A.4.5. Código para generar ltro pasa banda IIR

183

//*************************************************************************** //**********************Genera pasa banda IIR************************** function [num, den] = pasaBandaIIR(fs1,fs2,fp1,fp2,of,delta_p,delta_s,tipoFiltro,fm) // defino fin de la banda de rechazo phi_s1 = fs1/fm; // defino comienzo de la banda de rechazo phi_s2 = fs2/fm; // defino comienzo de la banda de paso phi_p1 = fp1/fm; // defino fin de la banda de paso phi_p2 = fp2/fm; //*********Transformo los parametros al tiempo continuo*********** // defino T parametro para mantener las dimensiones T = 10**-4; f_s1 = 1/(%pi*T)*tan(%pi*phi_s1); f_s2 = 1/(%pi*T)*tan(%pi*phi_s2); f_p1 = 1/(%pi*T)*tan(%pi*phi_p1); f_p2 = 1/(%pi*T)*tan(%pi*phi_p2); delta_1 = (2*delta_p)/(1+delta_p); delta_2 = delta_s/(1+delta_p); // defino delta_g, parametro para centrar el error de la banda de // paso en 1.

delta_g = 1 + delta_p; // transformo el pasa-banda en un filtro pasa-bajos mediante la inversion // defino fo fo = sqrt(f_p1*f_p2); theta_p1 = abs((f_p1^2 - fo^2)/f_p1); theta_p2 = abs((f_p2^2 - fo^2)/f_p2); theta_s1 = abs((f_s1^2 - fo^2)/f_s1); theta_s2 = abs((f_s2^2 - fo^2)/f_s2); // diseño del filtro pasa-bajos del tiempo continuo utilizando la funcion // predefinida analpf(); filtro = 'butt'; omega = theta_p1; if(tipoFiltro == 2) then filtro = 'cheb1'; else if(tipoFiltro == 3) then filtro = 'cheb2'; omega = theta_s1 else if(tipoFiltro == 4) then filtro = 'ellip'; end end end // defino vector de error de rizado permitido en las bandas de paso y rechazo ve = [delta_1 delta_2]; // utilizo funcion analpf(); [h_theta, p_theta, c_theta,g_theta] = analpf(of,'ellip',ve,2*%pi*omega); h_theta; p_theta;

c_theta; // deshacemos inversion de tiempo continuo g_s = g_theta; // paso al mundo de Laplace // defino cak wo = 2*%pi*fo; c_s = [0.5*(c_theta + sqrt(c_theta.^2 - 4*wo^2)) 0.5*(c_theta - sqrt(c_theta.^2 - 4*wo^2))]; // agrego ceros en phi = 0 si D-N s^D-N if length(p_theta)-length(c_theta) <>0 then c_s = [c_s zeros(1,length(p_theta)-length(c_theta))]; end; p_s = [0.5*(p_theta + sqrt(p_theta.^2 - 4*wo^2)) 0.5*(p_theta - sqrt(p_theta.^2 - 4*wo^2))]; // paso al dominio del tiempo discreto // correcion de ganacia para lograr rizado centrado en 1 G = g_s*abs(prod((2/T)-c_s))/abs(prod((2/T)-p_s)); G = G*delta_g; v_c = (1+(T/2)*c_s)./(1-(T/2)*c_s); v_p = (1+(T/2)*p_s)./(1-(T/2)*p_s); if (length(c_s) <> length(p_s)) then v_c = [v_c (-1)*ones(1,length(p_s) - length(c_s))]; end // creacion de los polinomios numerador y denominador de // la transferencia del tiempo discreto num = G*real(poly(v_c,'z','roots'));

den = real(poly(v_p,'z','roots')); H_z = num/den endfunction

A.4.6. Código para modelado de la BCI

188

//****** Modelado tiempo de respuesta estimador ritmo alfa***** // frcuencia de muestreo fm = 1024; f = 10/fm; Np = 8000; n = [1:Np]; A = (fm*0.75); // 656 muestras v_n = [1:A]; A = length(v_n); u = [sin(2*%pi*f*[1:(Np-A)/2]) sqrt(10)*sin(2*%pi*f*[1:A]) sin(2*%pi*f*[1:(Np-A)/2])] ; u_2 = sin(2*%pi*f*n); //Generar filtro 40 Hz fp = 40; fs = 50; delta_p = 0.01; delta_s = 0.01; exec('GeneraPasaBajoIIR.sci'); [num40,den40] = pasaBajoIIR(fp,fs,12,delta_p,delta_s,3,fm); a_u = rtitr(num40,den40,u); a_u2 = rtitr(num40,den40,u_2); // diezmado a_u=a_u(1:4:Np); a_u2=a_u2(1:4:Np); // filtro pasa banda fm = fm/4; fp1 = 9; fp2 = 11; fs1 = 7; fs2 = 13; delta_p = 0.01; delta_s = 0.01; exec('GeneraPasaBandaIIR.sce');

[num, den] = pasaBandaIIR(fs1,fs2,fp1,fp2,2,delta_p,delta_s,4,fm); v_u = rtitr(num,den,a_u); v_u2 = rtitr(num,den,a_u2); V_u = (v_u)^2; V_u2 = (v_u2)^2; // filtro Gauss numPBO = poly([0.000081],'z','cooef'); denPBO = poly([0.991 0.991],'z','root'); y = rtitr(numPBO,denPBO,V_u); y2 = rtitr(numPBO,denPBO,V_u2); L = length(y) a = 0; p1 = []; p2 = []; while (a <= L/32) s1 = 0; s2 = 0; for i=1:32 if(i+a*32 < L) then s1 = s1 + y(i+a*32); s2 = s2 + y2(i+a*32); end end p1(a+1) = s1/32; p2(a+1) = s2/32; a = a+1; end //Diezmado de la señal scf(0); clf(); subplot(2,1,2) xgrid(); L = length(p1)-1 plot2d(n(1:L)/8,(p1(1:L)),2) plot2d(n(1:L)/8,ones(1:L)*0.5,5)

plot2d(n(1:L)/8,ones(1:L)*2,6) plot2d(n(1:L)/8,ones(1:L),3) xtitle('','Tiempo(s)','Potencia(W)') subplot(2,1,1) xgrid plot2d(n/1024,u,2) xtitle('','','Amplitud(V)')

A.5. Códigos de JAVA para el software de

monitoreo

192

A.5.1. Clases del dominio

A.5.1.1. Clase buer

193

package Dominio; import java.io.File; import java.util.ArrayList; /** * *************************** Clase Buffer******************************* * * Los datos llegan en un formato de trama especifico, * por lo cual se puede separar la información de los diferentes canales * (en esta versión máximo 4 canales), * este proceso se encarga de almacenar las señales por separado y * descartar datos incorrectos. * */ public class Buffer private Sistema miSistema; private ArrayList listaDatos; private ArrayList canales; private int cantidadDeCanales; public Buffer(Sistema unSistema) miSistema = unSistema; listaDatos = new ArrayList(); cantidadDeCanales = 1; canales = new ArrayList(); /** * * @param cant setea la cantidad de canales con la que esta * trabajando el adquisidor, se genera un ArrayLits por cada * canal. */ public void setCantidadDeCanales(int cant) cantidadDeCanales = cant; if (canales.isEmpty()) for (int c = 0; c < cantidadDeCanales; c++) canales.add(new ArrayList());

/** * * @return con cuantos canales opera el sistema. */ public int getCantidadDeCanales() return cantidadDeCanales; /** * * @param unDato es un nuevo dato ingresado a la pila general del * buffer. * Este metodo es invocado por el Objeto Serial cuando obtiene * datos nuevosdel adquisidor de señales */ public void addDato(int unDato) listaDatos.add(unDato); trama(); /** * * @param canal * @param indice * @return el dato guardado en el buffer corespondiente al * parametro canal en la posicion indicada por el parametro indice */ public synchronized double getDato(int canal, int indice) double dato = 0; try dato = (double) ((ArrayList) (canales.get(canal))).get(indice); catch (Exception e) System.out.println(e.getMessage() + " getDato"); return dato; /** * * @return la lista que contiene todos los ArrayList corespondientes * a cada canal */

public ArrayList getCanales() return canales; /** * * @param unPath parametro que contiene la ruta para cargar datos de una * captura previamente guardada */ public void cargarDeArchivo(String unPath) canales = LeerArchivo.leer(canales, unPath); System.out.println(((ArrayList) canales.get(0)).size()); this.setCantidadDeCanales(canales.size()); miSistema.getProceso().start(); /** * * @param unFichero es el nombre del archivo que contendra los datos de * una captura que se desea guardar */ public void guardarArchivo(File unFichero) EscribirArchivo.escribirArchivo(canales, unFichero); /** * * Metodo encargado de extraer los datos de la señales de la trama, * guardarlos en el ArrayList corespondiente, y descartar los datos * erroneos */ public synchronized void trama() if (listaDatos.size() >= 2) if ((int) listaDatos.get(0) == -1 && (int) listaDatos.get(1) == -1) if (listaDatos.size() >= 2 + 2 * cantidadDeCanales) int canal = 0; for (int i = 2; i < 2 + 2 * cantidadDeCanales; i = i + 2) int valorL = (int) listaDatos.get(i); if (valorL < 0) valorL = valorL + 256; int valor = 256 * (int) listaDatos.get(i + 1) + valorL; ((ArrayList) canales.get(canal)).add((double) valor);

canal++; for (int i = 0; i < 2 + 2 * cantidadDeCanales; i++) listaDatos.remove(0); else for (int i = 0; i < 2; i++) String a = listaDatos.remove(0).toString(); System.out.println("Descartando " + a);

A.5.1.2. Clase escribir archivo

198

package Dominio; import java.io.*; import java.util.ArrayList; /** * *************************Clase EscribirArchivo****************************** * * Encargada de guardar los datos de las señales de las capturas realizadas. * Los datos se guardan como una tabla, cada columna coresponde a un canal. * Las columnas de datos estan separadas mediante un espacio en * blanco(tabulador) */ public class EscribirArchivo /** * * @param canales indica la cantidad de canales a guardad * @param unFichero indica el nombre del archivo * * */ public static void escribirArchivo(ArrayList canales,File unFichero) File fichero = null; try fichero = unFichero; if (!fichero.exists()) if (fichero.createNewFile()) System.out.println("El fichero se ha creado correctamente"); else System.out.println("No ha podido ser creado el fichero"); BufferedWriter out = new BufferedWriter(new OutputStreamWriter(new FileOutputStream(fichero, true), "UTF8")); int cantidadDeCanales = canales.size(); for (int i = 0; i < ((ArrayList)canales.get(0)).size(); i++) for(int c = 0;c < cantidadDeCanales;c++) double valor = (double)(((ArrayList)canales.get(c)).get(i)); int valorInt = (int)valor; out.write(String.valueOf(valorInt)+"\t"); out.newLine();

out.close(); catch (Exception e) e.printStackTrace();

A.5.1.3. Clase FFT

201

package Dominio; import edu.emory.mathcs.jtransforms.fft.FloatFFT_1D; /** * ****************************Clase FFT******************************* * Utiliza el algortimo FFT para calcular la TFD */ public class FFT /** * * @param unaLista Array con los datos de la señal * @return un Array con los datos de la TFD de la señal */ public static float[] getFFT(float[] unaLista) float[] miLista = unaLista; FloatFFT_1D joseGervasio = new FloatFFT_1D(miLista.length); joseGervasio.realForward(miLista); return miLista; /** * * @param unaLista Array con los datos de la TFD de la señal * @return Array con los datos del Modulo de la TFD de la señal */ public static float[] getModuloFFT(float[] unaLista) float[] miLista = unaLista; FloatFFT_1D joseGervasio = new FloatFFT_1D(miLista.length); joseGervasio.realForward(miLista); float modulo[] = new float[miLista.length / 2 + 1]; //asigno el primer y ultimo elemento modulo[0] = miLista[0]; modulo[modulo.length - 1] = miLista[1]; for (int i = 2; i < miLista.length; i = i + 1) miLista[i] = miLista[i] * miLista[i]; for (int i = 2; i < miLista.length - 1; i = i + 2) modulo[i / 2] = (float) Math.sqrt((double) (miLista[i]) + (double) (miLista[i + 1])); return modulo;

A.5.1.4. Clase gráca

204

package Dominio; import java.awt.Canvas; import java.awt.Color; import java.awt.Graphics; import java.awt.Image; import java.text.DecimalFormat; import java.util.ArrayList; /** * ******************************Clase Grafica********************************* * * Encargada de generar todos los graficos del programa. */ public class Grafica extends Canvas /** * Metodo para graficar una linea * @param g panel grafico(lienzo) en donde se desea dibujar * @param xi posicion de origen de la linea (x) * @param yi posicion de origen de la linea (y) * @param xf posicion final de la linea (x) * @param yf posicion final de la linea (y) * @param unColor color de la linea */ public static void graficarLinea(Graphics g, int xi, int yi, int xf, int yf, Color unColor) g.setColor(unColor); g.drawLine(xi, yi, xf, yf); /** * Borrar en forma rectangular * @param g panel grafico(lienzo) en donde se desea borrar * @param xi posicion de origen (x) rectangulo * @param yi posicion de origen (y) rectangulo * @param ancho ancho del rectangulo * @param alto alto del rectangulo */ public static void borrar(Graphics g, int xi, int yi, int ancho, int alto) g.clearRect(xi, yi, ancho, alto);

/** * Dibujar rectangulo(Barra) * @param g panel grafico(lienzo) en donde se desea dibujar * @param xi posicion de origen (x) * @param yi posicion de origen (y) * @param ancho grosor de la barra * @param alto alto de la barra * @param unColor color */ public static void graficarBarras(Graphics g, int xi, int yi, int ancho, int alto, Color unColor) g.setColor(unColor); g.fill3DRect(xi, yi, ancho, alto, true); /** * Incertar texto * @param g panel grafico(lienzo) en donde se desea dibujar * @param txi posicion de origen (x) * @param tyi posicion de origen (y) * @param texto texto a incertar * @param unColor color de texto */ public static void texto(Graphics g, int txi, int tyi, String texto, Color unColor) g.setColor(unColor); g.drawString(texto, txi, tyi); /** * Generar cuadricula para graficos * @param g panel grafico(lienzo) en donde se desea dibujar * @param ancho ancho del panel * @param alto alto del panel * @param valorMaxSeñal valor maximo de la señal a graficar * @param valorMinSeñal valor minimo de la señal a graficar * @param porcentajeUtilizableDelPanel tamaño de la grafica en % con respecto al lienzo * @param div cantidad de diviciones de la cadricula * @param valorReal valor real de la señal no discretizada */ public static void panelGrafico(Graphics g, int ancho, int alto, double valorMaxSeñal, double valorMinSeñal, int porcentajeUtilizableDelPanel, int div,double valorReal)

float escalado = (float) (valorReal/(float)(valorMaxSeñal-valorMinSeñal)); for (int i = 0; i <= div; i++) float tamaño = (float) (valorMaxSeñal - valorMinSeñal) / div; int valor = Grafica.escalar((int) (valorMaxSeñal - i * tamaño), alto, valorMaxSeñal, valorMinSeñal, porcentajeUtilizableDelPanel); g.setColor(Color.LIGHT_GRAY); g.drawLine(2, valor, ancho - 2, valor); g.setColor(Color.LIGHT_GRAY); DecimalFormat formatter = new DecimalFormat("#.##"); g.drawString(formatter.format(((int) (valorMaxSeñal - i * tamaño)*escalado)), 2, valor - 2); /* * Adapta el valor absoluto de la variable a graficar a las proporciones del panel grafico. * int valor es la variabla a graficar. * int alto es el alto del panel. * double valorMaxSeñal es el valor maximo de pico de la señal a graficar. * int porcentajeUtilizableDelPanel indca la porcion de panel en la que se graficara. */ public static int escalar(int valor, int alto, double valorMaxSeñal, double valorMinSeñal, int porcentajeUtilizableDelPanel) int tamaño = alto * porcentajeUtilizableDelPanel / 100; int Max = (alto - tamaño) / 2; double tamañoSeñal = valorMaxSeñal - valorMinSeñal; double escala = tamaño / tamañoSeñal; double eje = valorMinSeñal * escala; int retorno = (int) (tamaño - (valor * escala - eje) + Max); return retorno; /** * Graficar señal * @param unaLista ArrayList con los datos de la señal a graficar * @param origen posicion de origen en la lista * @param fin posicion de fin en la lista * @param tiempo parametro de tiempo con el que se visualizara * @param frecMuestreo frecuencia de muestreo de la señal * @param porcentaje % del tamaño de la grafica * @param unPanel panel de dibujo * @param imagen objeto imagen del panel

* @param max maxima valor de la señal * @param min minimo valor de la señal * @param divTiempo divicion de tiempo(separacion en entre las * lineas verticales de la cuadricula * @param valorReal valor de la señal real sin discretizar */ public static void graficarSeñal(ArrayList unaLista,int origen,int fin,int tiempo,int frecMuestreo, int porcentaje, javax.swing.JPanel unPanel, Image imagen, int max, int min, double divTiempo,double valorReal) imagen.getGraphics().setColor(Color.BLACK); imagen.getGraphics().fill3DRect(0, 0, unPanel.getWidth() - 1, unPanel.getHeight() - 1, false); int offset = 40; int valorAnteriorCordenadas = offset; int valorAnteriorOrdenadas = escalar((int) Math.round((double)unaLista.get(origen)), unPanel.getHeight(), max, min, porcentaje); for (int i = origen; i < fin - 1; i = i + 1*tiempo) int ordenada = ((i-origen)/tiempo)+offset; Grafica.graficarLinea(imagen.getGraphics(), valorAnteriorCordenadas, valorAnteriorOrdenadas, ordenada, escalar((int) Math.round((double)unaLista.get(i)), unPanel.getHeight(), max, min, porcentaje), Color.CYAN); valorAnteriorCordenadas = (ordenada); valorAnteriorOrdenadas = escalar((int) Math.round((double)unaLista.get(i)), unPanel.getHeight(), max, min, porcentaje); if(i%(frecMuestreo/divTiempo) == 0) Grafica.graficarLinea(imagen.getGraphics(),ordenada, 0, ordenada, escalar(0, unPanel.getHeight(), max, min, porcentaje), Color.LIGHT_GRAY); Grafica.texto(imagen.getGraphics(), ordenada, unPanel.getHeight()-5,String.valueOf((float)(i)/frecMuestreo), Color.LIGHT_GRAY); Grafica.panelGrafico(imagen.getGraphics(),unPanel.getWidth(), unPanel.getHeight(), max, min, porcentaje, 6,valorReal); // Se "pega" la imagen sobre el componente unPanel.getGraphics().drawImage(imagen, 0, 0, unPanel); /** * * Graficar 2 señal en el mismo panel * @param unaLista ArrayList con los datos de la 1° señal a graficar

* @param otraLista ArrayList con los datos de la 2° señal a graficar * @param origen posicion de origen en la lista * @param fin posicion de fin en la lista * @param tiempo parametro de tiempo con el que se visualizara * @param frecMuestreo frecuencia de muestreo de la señal * @param porcentaje % del tamaño de la grafica * @param unPanel panel de dibujo * @param imagen objeto imagen del panel (tecnica del doble buffer) * para evitar parpadeo en la grafico * @param max maxima valor de la señal * @param min minimo valor de la señal * @param divTiempo divicion de tiempo(separacion en entre las * lineas verticales de la cuadricula * @param valorReal valor de la señal real sin discretizar */ public static void graficarSeñal(ArrayList unaLista,ArrayList otraLista,int origen,int fin,int tiempo,int frecMuestreo, int porcentaje, javax.swing.JPanel unPanel, Image imagen, int max, int min, double divTiempo,double valorReal) imagen.getGraphics().setColor(Color.BLACK); imagen.getGraphics().fill3DRect(0, 0, unPanel.getWidth() - 1, unPanel.getHeight() - 1, false); int offset = 40; int valorAnteriorAbsisa = offset; int valorAnteriorCordenadas1 = escalar((int) Math.round((double)unaLista.get(origen)), unPanel.getHeight(), max, min, porcentaje); int valorAnteriorCordenadas2 = escalar((int) Math.round((double)otraLista.get(origen)), unPanel.getHeight(), max, min, porcentaje); for (int i = origen; i < fin - 1; i = i + 1*tiempo) int absisa = ((i-origen)/tiempo)+offset; Grafica.graficarLinea(imagen.getGraphics(), valorAnteriorAbsisa, valorAnteriorCordenadas1, absisa, escalar((int) Math.round((double)unaLista.get(i)), unPanel.getHeight(), max, min, porcentaje), Color.CYAN); Grafica.graficarLinea(imagen.getGraphics(), valorAnteriorAbsisa, valorAnteriorCordenadas2, absisa, escalar((int) Math.round((double)otraLista.get(i)), unPanel.getHeight(), max, min, porcentaje), Color.RED); valorAnteriorAbsisa = absisa; valorAnteriorCordenadas1 = escalar((int) Math.round((double)unaLista.get(i)), unPanel.getHeight(), max, min, porcentaje);

valorAnteriorCordenadas2 = escalar((int) Math.round((double)otraLista.get(i)), unPanel.getHeight(), max, min, porcentaje); if(i%(frecMuestreo/divTiempo) == 0) Grafica.graficarLinea(imagen.getGraphics(),absisa, 0, absisa, escalar(0, unPanel.getHeight(), max, min, porcentaje), Color.LIGHT_GRAY); Grafica.texto(imagen.getGraphics(), absisa, unPanel.getHeight()-5,String.valueOf((float)(i)/frecMuestreo), Color.LIGHT_GRAY); Grafica.panelGrafico(imagen.getGraphics(),unPanel.getWidth(), unPanel.getHeight(), max, min, porcentaje, 6,valorReal); // Se "pega" la imagen sobre el componente unPanel.getGraphics().drawImage(imagen, 0, 0, unPanel); /** * Grafico de la TFD de la señal * @param unaLista Array de datos de la señal * @param porcentaje tamaño del grafico * @param unPanel panel de dibujo * @param imagen objeto imagen del panel (tecnica del doble buffer) * para evitar parpadeo en la grafico * @param cantidadDatosGraficar indice de datos a graficar * @param max maximo valor */ public static void graficarFFT(float unaLista[], int porcentaje, javax.swing.JPanel unPanel, Image imagen, int cantidadDatosGraficar, int max) for (int i = 50; i < unaLista.length; i++) if (max < (int) unaLista[i]) max = (int) unaLista[i]; //hago el escalado de todos los puntos imagen.getGraphics().setColor(Color.BLACK); imagen.getGraphics().fill3DRect(0, 0, unPanel.getWidth() - 1, unPanel.getHeight() - 1, false); int offset = 35; for (int i = 0; i <=500;i++) Grafica.graficarLinea(imagen.getGraphics(),i+offset, Grafica.escalar((int)(unaLista[i]), unPanel.getHeight(), max, 0, porcentaje),

i+1+offset, Grafica.escalar((int)(unaLista[i+1]), unPanel.getHeight(), max, 0, porcentaje), Color.blue); if (i % 50 == 0) Grafica.graficarLinea(imagen.getGraphics(),i + offset, 0, i+offset, escalar(0, unPanel.getHeight(), max, 0, porcentaje), Color.LIGHT_GRAY); Grafica.texto(imagen.getGraphics(), i+offset, Grafica.escalar(0, unPanel.getHeight(), max, 0, porcentaje) + 12, String.valueOf((float)(i+1)/10), Color.LIGHT_GRAY); Grafica.panelGrafico(imagen.getGraphics(), unPanel.getWidth(), unPanel.getHeight(), max, 0, porcentaje, 6,max); unPanel.getGraphics().drawImage(imagen, 0, 0, unPanel);

A.5.1.5. Clase leer archivo

212

package Dominio; import java.io.*; import java.util.ArrayList; /** * *****************************Clase LeerArchivo**************************** * Encargada de levantar los datos de las señales previamentes guardadas */ public class LeerArchivo /** * * @param listaDatos ArrayList en donde se guardaran los * datos de las señales * @param unPath Ruta al archivo * @return returna una ArrayList con ArrayList que posen los datos, * cada ArrayList representa un canal(columna) */ public static ArrayList leer(ArrayList listaDatos, String unPath) try // Abrimos el archivo FileInputStream fstream = new FileInputStream(unPath); // Creamos el objeto de entrada DataInputStream entrada = new DataInputStream(fstream); // Creamos el Buffer de Lectura BufferedReader buffer = new BufferedReader(new InputStreamReader(entrada)); String strLinea; String aux[]; int cantidadDeCanales; // Leer el archivo linea por linea if ((strLinea = buffer.readLine()) != null) aux = strLinea.split("\t"); cantidadDeCanales = aux.length; listaDatos = new ArrayList(); for (int c = 0; c < cantidadDeCanales; c++) listaDatos.add(new ArrayList()); ((ArrayList) listaDatos.get(c)).add((double) Integer.parseInt(aux[c])); while ((strLinea = buffer.readLine()) != null) //separamos palabras de la línea por espacios

// System.out.println(strLinea); aux = strLinea.split("\t"); for (int c = 0; c < cantidadDeCanales; c++) ((ArrayList) listaDatos.get(c)).add((double) Integer.parseInt(aux[c])); // Cerramos el archivo entrada.close(); catch (Exception e) //Catch de excepciones System.err.println("Ocurrio un error: " + e.getMessage()); return listaDatos;

A.5.1.6. Clase procesamiento de señal

215

package Dominio; import java.awt.*; import java.util.ArrayList; /** * *********************Clase ProcesamientoDeSeñal****************************** * Cuenta con los filtros para el acondicionamiento de la señal, tambien tiene * algoritmos para la detecion de ritmos cerebrales y el artefacto del parpadeo. */ public class ProcesamientoDeSeñal private int idProceso; private Sistema miSistema; private double[] bsBp; private double[] asBp; private double[] asLp; private double b0Lp; private double[] entRet; private double[] salRet; private double[] salRetLp; private double[] entRetAlfaO; private double[] salRetAlfaO; private double[] bsBpAlfaO; private double[] asBpAlfaO; private double[] bsLpO; private double[] asLpO; private double[] entRetLpAlfaO; private double[] salRetLpAlfaO; private double[] entRetAlfaF; private double[] salRetAlfaF; private double[] bsBpAlfaF; private double[] asBpAlfaF; private double[] bsLpF; private double[] asLpF; private double[] entRetLpAlfaF; private double[] salRetLpAlfaF; private double[] alfaFondoO; private double[] alfaFondoF; private double lambdaGrande; private double lambdaChico; private boolean hayAlfa; private boolean BCIon;

private boolean pestaneoOn; private boolean pestaneoDobleAceptado; private int nroMuestra; private double sumaBetas; private double sumaAlfas; private double[] as40HzA1024; private double[] bs40HzA1024; private double[] entRet40HzA1024; private double[] salRet40HzA1024; private int pestaneos; private int datosAltos; private int ultimoIndice; private int ultimoIndicePestaneo; private boolean bajeDe3; public ProcesamientoDeSeñal(Sistema unSistema, int unIdProceso) miSistema = unSistema; idProceso = unIdProceso; bsBp = new double[]0.0531, -0.0888777, 0.0888777, -0.0531; asBp = new double[]1.6996, -0.9409; asLp = new double[]1.98, -0.9801; b0Lp = 0.0001; entRet = new double[]0, 0, 0; salRet = new double[]0, 0; salRetLp = new double[]0, 0; alfaFondoO = new double[]66, 99, 99; alfaFondoF = new double[]99, 99, 99; lambdaGrande = 0.999; lambdaChico = 0.925; hayAlfa = false; BCIon = false; pestaneoOn = false; bajeDe3 = false; pestaneoDobleAceptado = false; nroMuestra = 0; sumaBetas = 0; sumaAlfas = 0; as40HzA1024 = new double[]7.91679715312964, -28.3913039259658, 60.6998666580502, -85.6361396764067, 83.2702965142858, -56.4977510207761, 26.4032860857237, -8.13183465486160, 1.49012431463849, -0.123355819059518;

bs40HzA1024 = new double[]0.008566106102783551, -0.06178385272602577, 0.210020980477029, -0.448931132008092, 0.680318395787897, -0.776366624025321, 0.680318395787897, -0.448931132008092, 0.210020980477029, -0.06178385272602575, 0.008566106102783551; entRet40HzA1024 = new double[]0, 0, 0, 0, 0, 0, 0, 0, 0, 0; salRet40HzA1024 = new double[]0, 0, 0, 0, 0, 0, 0, 0, 0, 0; //con filtro nuevo el alfa ehhh entRetAlfaO = new double[]0, 0, 0, 0; salRetAlfaO = new double[]0, 0, 0, 0; asBpAlfaO = new double[]-3.825152192772064196902, 5.6019011122486714882029, -3.718228889479577858879, 0.9449032700541467599820; bsBpAlfaO = new double[]0.0103736125175646030916, -0.0377169054112898999298, 0.0547208187879623733640, -0.0377169054112898999298, 0.0103736125175646013569; asLpO = new double[]-1.98, 0.9800999999999999712230, 0; bsLpO = new double[]0.0001000000000000000048, 0, 0, 0; entRetLpAlfaO = new double[]0, 0, 0; salRetLpAlfaO = new double[]0, 0, 0; entRetAlfaF = new double[]0, 0, 0, 0; salRetAlfaF = new double[]0, 0, 0, 0; asBpAlfaF = new double[]-3.825152192772064196902, 5.6019011122486714882029, -3.718228889479577858879, 0.9449032700541467599820; bsBpAlfaF = new double[]0.0103736125175646030916, -0.0377169054112898999298, 0.0547208187879623733640, -0.0377169054112898999298, 0.0103736125175646013569; asLpF = new double[]-1.985999999999999987565, 0.9860489999999999533031, 0; bsLpF = new double[]0.0000489999999999999984, 0, 0, 0; entRetLpAlfaF = new double[]0, 0, 0; salRetLpAlfaF = new double[]0, 0, 0; /** * * @return El numero identificador del proceso */ public int getIdProceso() return idProceso;

/** * Detecta el ritmos cerebral Alfa Occipital, envia por socket las señales * de control de la activacion de la BCI. * * @param unaEntrada ArrayList de los datos de la señal * @param unOrigen puntero de inicio en la lista * @param unFin puntero de final de datos a tratar en la lista * @return ArrayList con la potencia del ritmo alfa */ public ArrayList ritmoAlfaO(ArrayList unaEntrada, int unOrigen, int unFin) double[] salidaBp = new double[unFin - unOrigen]; double[] salidaLp = new double[unFin - unOrigen]; ArrayList salida = new ArrayList(); double entrada; for (int i = 0; i < unFin - unOrigen; i++) nroMuestra++; entrada = (double) unaEntrada.get(i + unOrigen) - 512; salidaBp[i] = entrada * bsBpAlfaO[0] + entRetAlfaO[0] * bsBpAlfaO[1] + entRetAlfaO[1] * bsBpAlfaO[2] + entRetAlfaO[2] * bsBpAlfaO[3] + entRetAlfaO[3] * bsBpAlfaO[4] - asBpAlfaO[0] * salRetAlfaO[0] - asBpAlfaO[1] * salRetAlfaO[1] - asBpAlfaO[2] * salRetAlfaO[2] - asBpAlfaO[3] * salRetAlfaO[3]; entRetAlfaO[3] = entRetAlfaO[2]; entRetAlfaO[2] = entRetAlfaO[1]; entRetAlfaO[1] = entRetAlfaO[0]; entRetAlfaO[0] = entrada; salRetAlfaO[3] = salRetAlfaO[2]; salRetAlfaO[2] = salRetAlfaO[1]; salRetAlfaO[1] = salRetAlfaO[0]; salRetAlfaO[0] = salidaBp[i]; salidaBp[i] = Math.pow(salidaBp[i], 2); salidaLp[i] = salidaBp[i] * bsLpO[0] + entRetLpAlfaO[0] * bsLpO[1] + entRetLpAlfaO[1] * bsLpO[2] + entRetLpAlfaO[2] * bsLpO[3] - asLpO[0] * salRetLpAlfaO[0] - asLpO[1] * salRetLpAlfaO[1] - asLpO[2] * salRetLpAlfaO[2]; sumaAlfas = sumaAlfas + salidaLp[i]; entRetLpAlfaO[2] = entRetLpAlfaO[1]; entRetLpAlfaO[1] = entRetLpAlfaO[0]; entRetLpAlfaO[0] = salidaBp[i]; salRetLpAlfaO[2] = salRetLpAlfaO[1]; salRetLpAlfaO[1] = salRetLpAlfaO[0]; salRetLpAlfaO[0] = salidaLp[i]; if (nroMuestra == 32) nroMuestra = 0;

salida.add(sumaAlfas / 32); sumaAlfas = 0; if (alfaFondoO[1] > alfaFondoO[2]) alfaFondoO[0] = lambdaGrande * alfaFondoO[1] + (1 - lambdaGrande) * (double) salida.get(i / 32); else alfaFondoO[0] = lambdaChico * alfaFondoO[1] + (1 - lambdaChico) * (double) salida.get(i / 32); if (hayAlfa && (double) salida.get(i / 32) < (2 * alfaFondoO[0])) hayAlfa = false; Toolkit.getDefaultToolkit().beep(); else if (!hayAlfa && (double) salida.get(i / 32) > (4 * alfaFondoO[0])) hayAlfa = true; Toolkit.getDefaultToolkit().beep(); byte[] vec = (byte) idProceso, 0; if (BCIon) vec[1] = 0; BCIon = false; pestaneoOn = false; else vec[1] = 1; BCIon = true; pestaneoOn = false; this.miSistema.getServidor().enviarPorSocket(vec); alfaFondoO[2] = alfaFondoO[1]; alfaFondoO[1] = alfaFondoO[0]; return salida; /** * Detecta el ritmos cerebral Alfa Frontal. * * @param unaEntrada ArrayList de los datos de la señal * @param unOrigen puntero de inicio en la lista * @param unFin puntero de final de datos a tratar en la lista * @return ArrayList con la potencia del ritmo alfa */

public ArrayList ritmoAlfaF(ArrayList unaEntrada, int unOrigen, int unFin) double[] salidaBp = new double[unFin - unOrigen]; double[] salidaLp = new double[unFin - unOrigen]; ArrayList salida = new ArrayList(); double entrada; for (int i = 0; i < unFin - unOrigen; i++) nroMuestra++; entrada = (double) unaEntrada.get(i + unOrigen) - 512; salidaBp[i] = entrada * bsBpAlfaF[0] + entRetAlfaF[0] * bsBpAlfaF[1] + entRetAlfaF[1] * bsBpAlfaF[2] + entRetAlfaF[2] * bsBpAlfaF[3] + entRetAlfaF[3] * bsBpAlfaF[4] - asBpAlfaF[0] * salRetAlfaF[0] - asBpAlfaF[1] * salRetAlfaF[1] - asBpAlfaF[2] * salRetAlfaF[2] - asBpAlfaF[3] * salRetAlfaF[3]; entRetAlfaF[3] = entRetAlfaF[2]; entRetAlfaF[2] = entRetAlfaF[1]; entRetAlfaF[1] = entRetAlfaF[0]; entRetAlfaF[0] = entrada; salRetAlfaF[3] = salRetAlfaF[2]; salRetAlfaF[2] = salRetAlfaF[1]; salRetAlfaF[1] = salRetAlfaF[0]; salRetAlfaF[0] = salidaBp[i]; salidaBp[i] = Math.pow(salidaBp[i], 2); salidaLp[i] = salidaBp[i] * bsLpF[0] + entRetLpAlfaF[0] * bsLpF[1] + entRetLpAlfaF[1] * bsLpF[2] + entRetLpAlfaF[2] * bsLpF[3] - asLpF[0] * salRetLpAlfaF[0] - asLpF[1] * salRetLpAlfaF[1] - asLpF[2] * salRetLpAlfaF[2]; sumaAlfas = sumaAlfas + salidaLp[i]; entRetLpAlfaF[2] = entRetLpAlfaF[1]; entRetLpAlfaF[1] = entRetLpAlfaF[0]; entRetLpAlfaF[0] = salidaBp[i]; salRetLpAlfaF[2] = salRetLpAlfaF[1]; salRetLpAlfaF[1] = salRetLpAlfaF[0]; salRetLpAlfaF[0] = salidaLp[i]; if (nroMuestra == 32) nroMuestra = 0; salida.add(sumaAlfas / 32); sumaAlfas = 0; if (alfaFondoF[1] > alfaFondoF[2]) alfaFondoF[0] = lambdaGrande * alfaFondoF[1] + (1 - lambdaGrande) * (double) salida.get(i / 32); else alfaFondoF[0] = lambdaChico * alfaFondoF[1] + (1 - lambdaChico) * (double) salida.get(i / 32); if (hayAlfa && (double) salida.get(i / 32) < (2 * alfaFondoF[0]))

hayAlfa = false; else if (!hayAlfa && (double) salida.get(i / 32) > (4 * alfaFondoF[0])) hayAlfa = true; alfaFondoF[2] = alfaFondoF[1]; alfaFondoF[1] = alfaFondoF[0]; return salida; /** * Detecta el ritmos cerebral Beta. * * @param unaEntrada ArrayList de los datos de la señal * @param unOrigen puntero de inicio en la lista * @param unFin puntero de final de datos a tratar en la lista * @return ArrayList con la potencia del ritmo alfa */ public ArrayList ritmoBeta(ArrayList unaEntrada, int unOrigen, int unFin) double[] salidaBp = new double[unFin - unOrigen]; double[] salidaLp = new double[unFin - unOrigen]; ArrayList salida = new ArrayList(); for (int i = 0; i < unFin - unOrigen; i++) nroMuestra++; salidaBp[i] = (double) unaEntrada.get(i + unOrigen) * bsBp[0] + entRet[0] * bsBp[1] + entRet[1] * bsBp[2] + entRet[2] * bsBp[3] + asBp[0] * salRet[0] + asBp[1] * salRet[1]; entRet[2] = entRet[1]; entRet[1] = entRet[0]; entRet[0] = (double) unaEntrada.get(i + unOrigen); salRet[1] = salRet[0]; salRet[0] = salidaBp[i]; salidaBp[i] = Math.pow(salidaBp[i], 2); salidaLp[i] = salidaBp[i] * b0Lp + asLp[0] * salRetLp[0] + asLp[1] * salRetLp[1]; sumaBetas = sumaBetas + salidaLp[i]; salRetLp[1] = salRetLp[0]; salRetLp[0] = salidaLp[i]; if (nroMuestra == 32) nroMuestra = 0; salida.add(sumaBetas / 32); sumaBetas = 0;

return salida; /** * Filtro de 40 Hz * * @param entrada valor de entrada de la señal * @return retorna los valore de la señal filtrada */ public double filtro40HzA1024(double entrada) double salida = 0; salida = entrada * bs40HzA1024[0]; for (int j = 0; j < 10; j++) salida += entRet40HzA1024[j] * bs40HzA1024[j + 1] + as40HzA1024[j] * salRet40HzA1024[j]; entRet40HzA1024[9] = entRet40HzA1024[8]; entRet40HzA1024[8] = entRet40HzA1024[7]; entRet40HzA1024[7] = entRet40HzA1024[6]; entRet40HzA1024[6] = entRet40HzA1024[5]; entRet40HzA1024[5] = entRet40HzA1024[4]; entRet40HzA1024[4] = entRet40HzA1024[3]; entRet40HzA1024[3] = entRet40HzA1024[2]; entRet40HzA1024[2] = entRet40HzA1024[1]; entRet40HzA1024[1] = entRet40HzA1024[0]; entRet40HzA1024[0] = entrada; salRet40HzA1024[9] = salRet40HzA1024[8]; salRet40HzA1024[8] = salRet40HzA1024[7]; salRet40HzA1024[7] = salRet40HzA1024[6]; salRet40HzA1024[6] = salRet40HzA1024[5]; salRet40HzA1024[5] = salRet40HzA1024[4]; salRet40HzA1024[4] = salRet40HzA1024[3]; salRet40HzA1024[3] = salRet40HzA1024[2]; salRet40HzA1024[2] = salRet40HzA1024[1]; salRet40HzA1024[1] = salRet40HzA1024[0]; salRet40HzA1024[0] = salida; return salida; /** *

* @return el valor del nivel de background del ritmo alfa occipital */ public double getAlfaFondoO() return (this.alfaFondoO[0]); /** * * @return el valor del nivel de background del ritmo alfa frontal */ public double getAlfaFondoF() return this.alfaFondoF[0]; /** * * @param entrada Valor de la señal de entrada * @param unIndice indice de tiempo */ public void reconocerParpadeosDobles(double entrada, int unIndice) if (unIndice > ultimoIndice) ultimoIndice = unIndice; if (entrada > 750) datosAltos++; // System.out.println("datos: " + datosAltos); else if (entrada < 600 && pestaneos == 1) bajeDe3 = true; if (datosAltos <= 50 && datosAltos > 2) pestaneos++; //System.out.println("pestaneos: " + pestaneos); if (pestaneos == 1) ultimoIndicePestaneo = unIndice; bajeDe3 = false; else if (pestaneos == 2 && (unIndice - ultimoIndicePestaneo) < 100 && bajeDe3 == true) System.out.println("Activo! " + ultimoIndicePestaneo); if (pestaneoOn) byte[] vec = 0; this.miSistema.getServidor().enviarPorSocket(vec); pestaneoOn = false; BCIon = false;

else if (BCIon) byte[] vec = 0; this.miSistema.getServidor().enviarPorSocket(vec); pestaneoOn = false; BCIon = false; else byte[] vec = 2; this.miSistema.getServidor().enviarPorSocket(vec); pestaneoOn = true; BCIon = false; pestaneos = 0; //bajeDe3 = false; else pestaneos = 0; //bajeDe3 = false; if (unIndice - ultimoIndicePestaneo > 100) pestaneos = 0; //bajeDe3 = false; datosAltos = 0; /** * * @param entrada valor de la entrada de la señal * @param unIndice indice de tiempo */ public void reconocerParpadeosMultiples(double entrada, int unIndice) if (unIndice > ultimoIndice) ultimoIndice = unIndice; if (entrada > 750) datosAltos++; else if (entrada < 600 && (pestaneos == 1 || pestaneos == 2)) bajeDe3 = true; if (datosAltos <= 50 && datosAltos > 2) pestaneos++; if (pestaneos == 1)

ultimoIndicePestaneo = unIndice; bajeDe3 = false; else if (pestaneos == 2 && (unIndice - ultimoIndicePestaneo) < 100 && bajeDe3 == true) ultimoIndicePestaneo = unIndice; bajeDe3 = false; pestaneoDobleAceptado = true; else if (pestaneos == 3 && (unIndice - ultimoIndicePestaneo) < 100 && bajeDe3 == true) byte[] vec = (byte) idProceso, 3; this.miSistema.getServidor().enviarPorSocket(vec); BCIon = false; pestaneoDobleAceptado = false; pestaneos = 0; else pestaneoDobleAceptado = true; pestaneos = 0; if (unIndice - ultimoIndicePestaneo > 100) if (pestaneoDobleAceptado) byte[] vec = (byte) idProceso, 2; this.miSistema.getServidor().enviarPorSocket(vec); BCIon = false; pestaneoDobleAceptado = false; pestaneos = 0; datosAltos = 0;

A.5.1.7. Clase proceso

227

package Dominio; import java.util.ArrayList; /** *****************************Clase Proceso******************************** * * Es el eslabón más importante del software. Toma las señales almacenadas * por Buffer, las acondiciona y procesa para así poder tomar decisiones de * control. */ public class Proceso extends Thread Thread t; private Sistema miSistema; // ********************FFT*********************** private int tamañoBufferFFT = 256; private int tamañoCerosFFT = 256 * 9; // ********************************************** private boolean play; private int escalaTiempo; private ArrayList filtrado; private ArrayList señal; private ArrayList diesmado_8; private ArrayList diesmado_4; private ArrayList backGround; private ArrayList ritmoAlfa; ProcesamientoDeSeñal señal1; ProcesamientoDeSeñal señal2; int indiceRitmos; int indiceGraficaTiempo; int indiceGraficaRitmos; int indiceFiltrado; int indiceDiesmado; int indiceFFT; boolean tire; private int cantidadDeCanales; public Proceso(Sistema unSistema) t = new Thread(this); miSistema = unSistema; filtrado = new ArrayList(); señal = new ArrayList(); diesmado_8 = new ArrayList();

diesmado_4 = new ArrayList(); backGround = new ArrayList(); ritmoAlfa = new ArrayList(); indiceGraficaTiempo = 0; indiceGraficaRitmos = 0; indiceFiltrado = 0; indiceDiesmado = 0; escalaTiempo = 1; play = true; tire = true; señal1 = new ProcesamientoDeSeñal(miSistema,0); señal2 = new ProcesamientoDeSeñal(miSistema,1); indiceRitmos = 0; indiceFFT = 0; //Generar Objetos public void run() cantidadDeCanales = miSistema.getBuffer().getCantidadDeCanales(); for (int c = 0; c < cantidadDeCanales; c++) filtrado.add(new ArrayList()); señal.add(new ProcesamientoDeSeñal(miSistema,c)); diesmado_8.add(new ArrayList()); diesmado_4.add(new ArrayList()); backGround.add(new ArrayList()); ritmoAlfa.add(new ArrayList()); miSistema.getVentana().nuevoPanel(String.valueOf(c),"Canal: "+String.valueOf(c+1), 50, 600/cantidadDeCanales*c, 700, 590/cantidadDeCanales); miSistema.getVentana().nuevoPanel(String.valueOf(c),"RitmoAlfa: "+String.valueOf(c+1), 770, 600/cantidadDeCanales*c, 500, 590/cantidadDeCanales); while (true) try //t.sleep(10); t.yield(); if (play) miSistema.getVentana().setScrollMax(((ArrayList)diesmado_8.get(0)).size()/ escalaTiempo); miSistema.getVentana().setScroll(indiceGraficaTiempo); indiceFiltrado = this.filtrarPro(indiceFiltrado);

indiceDiesmado = this.diesmarPro(indiceDiesmado); indiceRitmos = ritmoAlfaPro(indiceRitmos); int aux = indiceGraficaTiempo; int aux2 = indiceGraficaRitmos; for(int c = 0;c < cantidadDeCanales;c++) aux = this.recorrer(indiceGraficaTiempo, 1, 1, 128,((ArrayList)this.diesmado_8.get(c)), miSistema.getVentana().getPanel(2*c), 1, 1024,5); aux2 = this.recorrer(indiceGraficaRitmos, 1, 1, 8,((ArrayList)this.ritmoAlfa.get(c)),((ArrayList)this.backGround.get(c)), miSistema.getVentana().getPanel((c*2)+1), 0.1, 131072,3.125); if (((ArrayList)diesmado_8.get(0)).size() > indiceGraficaTiempo) ((ProcesamientoDeSeñal)señal.get(c)).reconocerParpadeosMultiples((double) ((ArrayList)diesmado_8.get(c)).get(indiceGraficaTiempo), indiceGraficaTiempo); indiceGraficaRitmos = aux2; indiceGraficaTiempo = aux; else indiceFiltrado = this.filtrarPro(indiceFiltrado); indiceDiesmado = this.diesmarPro(indiceDiesmado); indiceRitmos = ritmoAlfaPro(indiceRitmos); for(int c = 0;c < cantidadDeCanales;c++) this.recorrer(miSistema.getVentana().getScroll(), 1, 1, 128,((ArrayList)this.diesmado_8.get(c)), miSistema.getVentana().getPanel(c*2), 1, 1024,5); this.recorrer(miSistema.getVentana().getScroll()/16, 1, 1, 8,((ArrayList)this.ritmoAlfa.get(c)),((ArrayList)this.backGround.get(c)), miSistema.getVentana().getPanel((c*2)+1), 0.1, 131072,3.125); catch (Exception e) System.out.println(e.toString()); // **************************************************************************************** // // Constantes del metodo: tamañoBufferFFT, tamañoCerrosFFT, coeficienteDeSuavizado

// El coeficiente de suavisado determina que tan suave son los bordes de la ventana // de muestreo de la FFT. (coeficiente = 0 implica ventana de pulso unitario) // ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ /** * Metodo para calcular y graficar FTD, utiliza la clase grafica y FFT * @param unIndice inicio de los datos de la lista a tratar * @param unaLista ArrayList de datos * @param unPanel panel grafico. * @return returna el indice de datos tratados */ public int FFT(int unIndice, ArrayList unaLista, javax.swing.JPanel unPanel) if (unaLista.size() > unIndice + tamañoBufferFFT) float listaAux[] = new float[tamañoBufferFFT + tamañoCerosFFT]; double valorMedio = this.mediaMovil(unaLista, unIndice, unIndice + tamañoBufferFFT); for (int i = 0; i < tamañoBufferFFT; i++) float aux = (float) ((double) unaLista.get(i + unIndice) - valorMedio); listaAux[i] = aux * ventanaGauss(i, tamañoBufferFFT, 0.9); for (int i = tamañoBufferFFT; i < tamañoBufferFFT + tamañoCerosFFT; i++) listaAux[i] = 0; listaAux = FFT.getModuloFFT(listaAux); Grafica.graficarFFT(listaAux, 85, unPanel, miSistema.getVentana().getDobleBuffer(unPanel.getWidth(), unPanel.getHeight()), tamañoBufferFFT, 3000); return unIndice + 32; return unIndice; /** * Ventana de suavizado de Gauss * @param k coeficiente * @param N cantidad de coeficientes * @param sigma varianza de la ventana * @return valor de la ventana muestreada en k */ public float ventanaGauss(int k, int N, double sigma)

double exponente = (double) Math.pow((k - (N - 1) / 2) / (sigma * (N - 1) / 2), 2); double resul = (double) Math.exp(exponente * 0.5); return (float) resul; /** * Ventana de suavizado de Blackman * @param k coeficiente * @param N numero de coeficiente * @return valor de la ventana muestreada en k */ public float ventanaBlackman(int k, int N) double pi = Math.PI; return (float) (0.42 + 0.5 * Math.cos(2 * pi * k / (N - 1)) + 0.08 * Math.cos(4 * pi * k / (N - 1))); /** * Metodo utilizado para generar graficas, utiliza la clase Grafica * @param unIndice inicio de los datos de la lista a tratar * @param unIncremento cantidad de datos tratados por llamado del metodo * @param escalamientoDeTiempo indica como ajustar el panel grafico * @param frecMuestreo frecuencia de muestreo de la señal * @param unaLista ArrayList de datos * @param unPanel panel grafico * @param divTiempo divicion de tiempo * @param max maximo valor de señal * @param valorReal valor real de la señal sin cuantificar * @return indice de datos tratados */ public int recorrer(int unIndice, int unIncremento, int escalamientoDeTiempo, int frecMuestreo, ArrayList unaLista, javax.swing.JPanel unPanel, double divTiempo, int max,double valorReal) int tamañoVentana = unPanel.getWidth() - 30; if (unaLista.size() > unIndice * escalamientoDeTiempo) if (unIndice > tamañoVentana * escalamientoDeTiempo) Grafica.graficarSeñal(unaLista, unIndice - tamañoVentana * escalamientoDeTiempo, unIndice, escalamientoDeTiempo, frecMuestreo, 85, unPanel, miSistema.getVentana().getDobleBuffer(unPanel.getWidth(), unPanel.getHeight()), max, 0, divTiempo,valorReal); else Grafica.graficarSeñal(unaLista, 0, unIndice, escalamientoDeTiempo, frecMuestreo, 85, unPanel,

miSistema.getVentana().getDobleBuffer(unPanel.getWidth(), unPanel.getHeight()), max, 0, divTiempo,valorReal); unIndice = unIndice + unIncremento; return unIndice; /** * Metodo utilizado para generar 2 graficos, utiliza la clase Grafica * @param unIndice inicio de los datos de la lista a tratar * @param unIncremento cantidad de datos tratados por llamado del metodo * @param escalamientoDeTiempo indica como ajustar el panel grafico * @param frecMuestreo frecuencia de muestreo de la señal * @param unaLista ArrayList de datos * @param otraLista ArrayList de datos * @param unPanel panel grafico * @param divTiempo divicion de tiempo * @param max maximo valor de señal * @param valorReal valor real de la señal sin cuantificar * @return indice de datos tratados */ public int recorrer(int unIndice, int unIncremento, int escalamientoDeTiempo, int frecMuestreo, ArrayList unaLista, ArrayList otraLista, javax.swing.JPanel unPanel, double divTiempo, int max, double valorReal) int tamañoVentana = unPanel.getWidth() - 30; if (unaLista.size() > unIndice * escalamientoDeTiempo) if (unIndice > tamañoVentana * escalamientoDeTiempo) Grafica.graficarSeñal(unaLista, otraLista, unIndice - tamañoVentana * escalamientoDeTiempo, unIndice, escalamientoDeTiempo, frecMuestreo, 85, unPanel, miSistema.getVentana().getDobleBuffer(unPanel.getWidth(), unPanel.getHeight()), max, 0, divTiempo,valorReal); else Grafica.graficarSeñal(unaLista, otraLista, 0, unIndice, escalamientoDeTiempo, frecMuestreo, 85, unPanel, miSistema.getVentana().getDobleBuffer(unPanel.getWidth(), unPanel.getHeight()), max, 0, divTiempo,valorReal); unIndice = unIndice + unIncremento; return unIndice; /**

* Algoritmo para filtrar la señal, utiliza la clase ProcesamientoDeSeñal * @param unIndice indice de valor a filtrar * @return indice de valor filtrado */ public int filtrarPro(int unIndice) if (((ArrayList)miSistema.getBuffer().getCanales().get(0)).size() > unIndice) for (int c = 0; c < cantidadDeCanales; c++) ((ArrayList) (filtrado.get(c))).add(((ProcesamientoDeSeñal) señal.get(c)).filtro40HzA1024(miSistema.getBuffer().getDato(c, unIndice))); unIndice++; return unIndice; /** * Diesmado de la señal * @param unIndice indice del valor a diesmar * @return indice del valor filtrado */ public int diesmarPro(int unIndice) if (((ArrayList) filtrado.get(0)).size() >= unIndice + 32) for (int i = unIndice; i < unIndice + 32; i = i + 4) for (int c = 0; c < cantidadDeCanales; c++) if ((i - unIndice) % 8 == 0) ((ArrayList)diesmado_8.get(c)).add(((ArrayList) filtrado.get(c)).get(i)); ((ArrayList)diesmado_4.get(c)).add(((ArrayList) filtrado.get(c)).get(i)); return unIndice + 32; else return unIndice; /** * Algoritmo para detectar el ritmo alfa,utiliza la clase * ProcesamientoDeSeñal * @param unIndice indice de valor a filtrar * @return indice de valor filtrado

*/ public int ritmoAlfaPro(int unIndice) try if (((ArrayList) diesmado_4.get(0)).size() > unIndice + 32) try for(int c = 0;c < cantidadDeCanales;c++) ((ArrayList)ritmoAlfa.get(c)).addAll(((ProcesamientoDeSeñal) señal.get(c)).ritmoAlfaO(((ArrayList) diesmado_4.get(c)), unIndice, unIndice + 32)); ((ArrayList)backGround.get(c)).add(((ProcesamientoDeSeñal)señal.get(c)).getAlfaFondoO()); return unIndice + 32; catch (Exception e) System.out.println("Error en Algoritmos de procesamiento"); catch (Exception e) System.out.println("Error en algoritmo ritmoAlfaPro"); return unIndice; /** * * @param estado indica si el sistema esta graficando */ public void setPlay(boolean estado) this.play = estado; /** * Algoritmo para sacar la media movil de la señal * @param unaLista ArrayList de los datos de la señal * @param inicio puntero de inicio de los datos a tratar * @param fin puntero de fin de los datos a tratar * @return */ public double mediaMovil(ArrayList unaLista, int inicio, int fin) double suma = 0; for (int i = inicio; i < fin; i++) suma = (double) unaLista.get(i) + suma;

return (suma / (fin - inicio)); /** * Algoritmo para calcular el modulo de la FFT * @param indice inidice de datos tratados */ public void FFTmodulo(int indice) float listaAux[] = new float[tamañoBufferFFT]; float listaAux2[] = new float[tamañoBufferFFT]; for (int i = 0; i < tamañoBufferFFT; i++) listaAux2[i] = listaAux[i]; for (int i = tamañoBufferFFT; i < tamañoBufferFFT + tamañoCerosFFT; i++) listaAux[i] = 0; listaAux = FFT.getModuloFFT(listaAux); for (int i = 0; i < 50; i++) System.out.println(listaAux2[i]); System.out.println("Indice " + indice + " " + listaAux[i]);

A.5.1.8. Clase serial

237

package Dominio; /** * ****************************Clase Serial****************************** * * Al inicio de la conexión busca todo los puertos COM presentes, * es el usuario quien indica en cuál de ellos se encuentra el dispositivo * de adquisición, una vez indicado esto el mismo realiza la conexión de ser * posible. Si se logra con éxito, este proceso pasa a quedar a la espera de * datos. Una vez que se obtiene uno o un grupo de estos, se envían al proceso * Buffer. Si bien el microcontrolador envía los datos de las señales capturadas * de forma regular, dado que se trabaja sobre un COM virtual, los datos llegan * de manera asincrónica, por lo que el proceso Serial actúa como un evento. * * */ import java.io.InputStream; import java.io.OutputStream; import gnu.io.CommPortIdentifier; import gnu.io.SerialPort; import gnu.io.SerialPortEvent; import gnu.io.SerialPortEventListener; import java.util.ArrayList; import java.util.Enumeration; public class Serial implements SerialPortEventListener SerialPort serialPort; private boolean conectado; private InputStream input; private OutputStream output; private int DATA_RATE; private static final int TIME_OUT = 2000; //Tiempo de espera para apertura del puerto private int valorF; private String miDato; private Sistema miSistema; public Serial(Sistema unSistema) miSistema = unSistema; conectado = false;

/** * @return boolean que indica si el adquisidor esta conectado */ public boolean getConectado() return conectado; /** * * @return todos los puertos COM */ public ArrayList getPuertos() Enumeration portEnum = CommPortIdentifier.getPortIdentifiers(); ArrayList puertos = new ArrayList(); while (portEnum.hasMoreElements()) CommPortIdentifier currPortId = (CommPortIdentifier) portEnum.nextElement(); puertos.add(currPortId.getName()); System.out.println("Puertos: " + currPortId.getName()); if(puertos.isEmpty()) miSistema.getVentana().mensaje("No se encontraron puertos", "PUERTOS COM", 2); return puertos; /** * Inicializacion del puerto COM * @param unPuerto purtoCOM virtual * @param unDataRate velocidad de trasmicion */ public void initialize(Object unPuerto,int unDataRate) DATA_RATE = unDataRate; valorF = 0; miDato = ""; CommPortIdentifier portId = null; Enumeration portEnum = CommPortIdentifier.getPortIdentifiers(); //Busco todos los puertos COM hasta que encuentro el selecionado while (portEnum.hasMoreElements()) CommPortIdentifier currPortId = (CommPortIdentifier) portEnum.nextElement();

if (currPortId.getName().equals(unPuerto)) portId = currPortId; break; if (portId == null) miSistema.getVentana().mensaje("Imposible conectar", "ERROR",1); return; try serialPort = (SerialPort) portId.open(this.getClass().getName(), TIME_OUT); //Seteo los parametros del puerto serialPort.setSerialPortParams(DATA_RATE, SerialPort.DATABITS_8, SerialPort.STOPBITS_1, SerialPort.PARITY_EVEN); //Abro los stream de entrada y salida input = serialPort.getInputStream(); output = serialPort.getOutputStream(); //Agrego los eventos a escuchar serialPort.addEventListener(this); serialPort.notifyOnDataAvailable(true); //Conectado this.conectado = true; catch (Exception e) System.err.println(e.toString()); close(); System.exit(0); /** * Metodo para cerrar el puerto COM */ public synchronized void close() if (serialPort != null) serialPort.removeEventListener(); serialPort.close();

/** * Leer los datos del puerto serie cada vez que se produce un evento * @param oEvent */ public synchronized void serialEvent(SerialPortEvent oEvent) if (oEvent.getEventType() == SerialPortEvent.DATA_AVAILABLE) //leerDatosAscii(); leerDatosHex(); /** * trato los datos del puerto serie como Hex */ public void leerDatosHex() try int available = input.available(); byte chunk[] = new byte[available]; input.read(chunk, 0, available); for (int i = 0; i < available; i++) miSistema.getBuffer().addDato(chunk[i]); catch (Exception e) System.err.println(e.toString()); /** * trato los datos del puerto serie como Ascii */ public void leerDatosAscii() try int available = input.available(); byte chunk[] = new byte[available]; input.read(chunk, 0, available); String dato = new String(chunk); for (int i = 0; i < dato.length(); i++) if (dato.substring(i, i + 1).equals("@")) valorF = Integer.parseInt(miDato); miSistema.getBuffer().addDato(valorF); miDato = "";

else miDato = miDato + dato.substring(i, i + 1); catch (Exception e) System.err.println(e.toString()); miDato = ""; /** * Envia datos(int) por el puerto serial * @param dato a enviar */ public void enviarPorSerial(int dato) try output.write(dato); catch (Exception e) System.out.println(e.toString());

A.5.1.9. Clase sistema

243

package Dominio; import Ventanas.Ventana; /** * *****************************Clase Sistema********************************* * * Clase global del sistema, utilizada para la gestion de las demas. */ public class Sistema private Buffer miBuffer; private Proceso miProceso; private Serial miSerial; private Ventana miVentana; private Servidor miServidor; public Sistema() miBuffer = null; miProceso = null; miSerial = null; miVentana = null; miServidor = null; public void setBuffer(Buffer unBuffer) miBuffer = unBuffer; public void setProceso(Proceso unProceso) miProceso = unProceso; public void setSerial(Serial unSerial) miSerial = unSerial; public void setVentana(Ventana unaVentana) miVentana = unaVentana; public void setServidor(Servidor unServidor) miServidor = unServidor; public Buffer getBuffer() return miBuffer; public Proceso getProceso()

return miProceso; public Serial getSerial() return miSerial; public Ventana getVentana() return miVentana; public Servidor getServidor() return miServidor;

A.5.1.10. Clase servidor

246

package Dominio; /** * ****************************Clase Servidor********************************** * * Se utiliza para enviar los comando de control BCI por socket * * Hay 4 BCI alfa ON/OFF y 4 BCI por deteccion de parpadeos, cada una * corresponde a un canal * * La trama enviada esta compuesta por dos Byte, el primero de ellos coresponde * al canal y el segundo a la señal de control * * comandos 1 y 0 corresponden a On y Off de ritmo alfa * comandos 2 y 3 corresponden a 2 y 3 parpadeos * * trama = [canal,comando], con canal de 0 al 3. * */ import java.net.*; import java.io.*; import java.util.logging.Level; import java.util.logging.Logger; public class Servidor static final int PUERTO = 5000; // Puerto de escucha del sistema ServerSocket sc; Socket so; DataOutputStream salida; BufferedReader entrada; private Sistema miSistema; /** * Inicia el Socket * @param unSistema */ public void initServer(Sistema unSistema) miSistema = unSistema; try sc = new ServerSocket(PUERTO);

so = new Socket(); catch (Exception e) System.out.println("Error al iniciar"); /** * Acepta la conexion con el cliente */ public void acept() try sc.setSoTimeout(1); so = sc.accept(); entrada = new BufferedReader(new InputStreamReader(so.getInputStream())); salida = new DataOutputStream(so.getOutputStream()); this.miSistema.getVentana().mensaje("Conexión exitosa con cliente", "", -1); catch (IOException ex) this.miSistema.getVentana().mensaje("No se encontró cliente conectado", "ERROR SOCKET", 2); Logger.getLogger(Servidor.class.getName()).log(Level.SEVERE, null, ex); /** * Metodo para enviar por Socket * @param unDato */ public void enviarPorSocket(byte[] unDato) try this.salida.write(unDato); catch (Exception e) Logger.getLogger(Servidor.class.getName()).log(Level.SEVERE, null, e);

A.5.2. Clases de Interfaz

A.5.2.1. Clase ventana

249

package Ventanas; import Dominio.Sistema; import java.awt.Color; import java.awt.Graphics; import java.awt.Image; import java.awt.Toolkit; import java.awt.event.WindowAdapter; import java.awt.event.WindowEvent; import java.io.File; import java.util.ArrayList; import javax.swing.*; import javax.swing.BorderFactory; import javax.swing.Icon; import javax.swing.ImageIcon; import javax.swing.JOptionPane; import javax.swing.JPanel; import org.jvnet.substance.SubstanceLookAndFeel; /** * *************************** Clase Ventana******************************* */ public final class Ventana extends javax.swing.JFrame /** * Creates new form Ventana */ private Image dobleBuffer; private Sistema miSistema; private ArrayList listaPaneles; public Ventana() initComponents(); //this.setExtendedState(this.MAXIMIZED_BOTH); public Ventana(Sistema unSistema) JFrame.setDefaultLookAndFeelDecorated(true); SubstanceLookAndFeel.setSkin("org.jvnet.substance.skin.BusinessBlackSteelSkin"); initComponents();

this.setExtendedState(this.MAXIMIZED_BOTH); miSistema = unSistema; listaPaneles = new ArrayList(); this.botonConectar.setEnabled(false); this.baudRates.setSelectedIndex(1); this.setDefaultCloseOperation(WindowConstants.DO_NOTHING_ON_CLOSE); addWindowListener(new WindowAdapter() @Override public void windowClosing(WindowEvent we) miSistema.getSerial().enviarPorSerial(5); int eleccion; eleccion = JOptionPane.showConfirmDialog(null, "¿Desea gurdar antes de salir?", "Confirme guardar", JOptionPane.YES_NO_OPTION); byte[] vec = 4; if (eleccion == 0) guardarArrchivo(); miSistema.getServidor().enviarPorSocket(vec); miSistema.getSerial().enviarPorSerial(5); System.exit(0); else miSistema.getSerial().enviarPorSerial(5); miSistema.getServidor().enviarPorSocket(vec); System.exit(0); ); /** * Genera un nuevo Panel Grafico * @param unNombre nombre del panel * @param unTitulo titulo del panel * @param x posicion en x del panel * @param y posicion en y del panel * @param largo largo del panel * @param ancho ancho del panel */ public void nuevoPanel(String unNombre, String unTitulo, int x, int y, int largo, int ancho) JInternalFrame j = new JInternalFrame(unTitulo, true);

j.setName(unNombre); j.setBounds(x, y, largo, ancho); j.setResizable(true); j.setVisible(true); j.setLayer(1); // ImageIcon fot = new ImageIcon(ClassLoader.getSystemResource("imagenes/cerebro.png")); // ImageIcon icono = new ImageIcon(fot.getImage().getScaledInstance(30, 30, Image.SCALE_DEFAULT)); // j.setFrameIcon(icono); JPanel p = new JPanel(); p.setBounds(0, 0, 200, 50); p.setPreferredSize(j.getPreferredSize()); p.setBackground(Color.black); p.setDoubleBuffered(true); j.add(p); this.panelEscritorio.add(j); this.panelEscritorio.repaint(); this.listaPaneles.add(p); /** * * @return lista de Panels */ public ArrayList getListaPaneles() return listaPaneles; /** * * @param unIndice indica el panel a selecionar * @return un panel especifico */ public javax.swing.JPanel getPanel(int unIndice) return (javax.swing.JPanel) listaPaneles.get(unIndice); /** * * @return Indica si el tiempo se encuentra operando en tiempo real */ public boolean tiempoReal()

return tiempoReal.isSelected(); /** * * @param max seteo la posicion maxima del scrol */ public void setScrollMax(int max) scroll.setMaximum(max); /** * * @param pos setea la posicion del scrol */ public void setScroll(int pos) this.scroll.setValue(pos); /** * * @return posicion del scrol */ public int getScroll() return this.scroll.getValue(); /** * Tecnica del dobleBuffer (para refrescar la imagen sin producir efecto de * parpadeo) * @param ancho de la imagen * @param largo de la imagen * @return */ public Image getDobleBuffer(int ancho, int largo) dobleBuffer = createImage(ancho, largo); return dobleBuffer; /** * Para setear texto * @param unTexto */

public void setTexto(String unTexto) texto.setText(unTexto); public void setTexto1(String unTexto) texto1.setText(unTexto); public void setColor(int unColor) if (unColor == 0) texto1.setForeground(Color.BLUE); else texto1.setForeground(Color.red); /** * Mensajes por ventana * @param unObject quien lo llama * @param unTitulo Titulo * @param unTipo tipo */ public void mensaje(Object unObject, String unTitulo, int unTipo) JOptionPane.showMessageDialog(this, unObject, unTitulo, unTipo); /** * Ventana Guardar Archivo, utiliza la clase EscrivirArchivo */ public void guardarArrchivo() int respuesta = this.elegirArchivo.showSaveDialog(null); // Si apretamos en aceptar ocurrirá esto if (respuesta == JFileChooser.APPROVE_OPTION) File selectedFile = elegirArchivo.getSelectedFile(); miSistema.getBuffer().guardarArchivo(selectedFile); // Si apretamos en cancelar o cerramos la ventana ocurrirá esto public void ventanaConfiguracionShow() ArrayList misPuertos = miSistema.getSerial().getPuertos(); for (int i = 0; i < misPuertos.size(); i++) this.Puertos.addItem(misPuertos.get(i)); this.botonConectar.setEnabled(true);

this.ventanaConfiguracion.setLocation(100, 100); this.ventanaConfiguracion.setSize(this.ventanaConfiguracion.getPreferredSize()); this.ventanaConfiguracion.setVisible(true); public void ventanaElectrodosShow() this.ventanaPruebaBCI.setLocation(100, 100); this.ventanaPruebaBCI.setSize(this.ventanaPruebaBCI.getPreferredSize()); this.ventanaPruebaBCI.setVisible(true); public void ventanaArrchivoShow() this.ventanaElegirArchivo.setLocation(100, 100); this.ventanaElegirArchivo.setSize(this.ventanaElegirArchivo.getPreferredSize()); this.ventanaElegirArchivo.setVisible(true); /** * This method is called from within the constructor to initialize the form. * WARNING: Do NOT modify this code. The content of this method is always * regenerated by the Form Editor. */ @SuppressWarnings("unchecked") // <editor-fold defaultstate="collapsed" desc="Generated Code"> private void initComponents() ventanaConfiguracion = new javax.swing.JDialog(); jPanel2 = new javax.swing.JPanel(); tiempoReal = new javax.swing.JRadioButton(); tiempoSimulado = new javax.swing.JRadioButton(); panelSerial = new javax.swing.JPanel(); Puertos = new javax.swing.JComboBox(); botonBuscar = new javax.swing.JButton(); botonConectar = new javax.swing.JButton(); jLabel1 = new javax.swing.JLabel(); jLabel2 = new javax.swing.JLabel(); baudRates = new javax.swing.JComboBox(); CantidadCanales = new javax.swing.JComboBox(); jLabel3 = new javax.swing.JLabel();

grupoTiempo = new javax.swing.ButtonGroup(); ventanaElegirArchivo = new javax.swing.JDialog(); elegirArchivo = new javax.swing.JFileChooser(); ventanaPruebaBCI = new javax.swing.JDialog(); jPanel1 = new javax.swing.JPanel(); BCI1 = new javax.swing.JPanel(); botonAlfaON1 = new javax.swing.JButton(); botonAlfaOFF1 = new javax.swing.JToggleButton(); botonDosParpadeos1 = new javax.swing.JButton(); botonTresParpadeos1 = new javax.swing.JToggleButton(); jPanel3 = new javax.swing.JPanel(); botonAlfaON2 = new javax.swing.JButton(); botonAlfaOFF2 = new javax.swing.JToggleButton(); botonDosParpadeos2 = new javax.swing.JButton(); botonTresParpadeos2 = new javax.swing.JToggleButton(); jPanel4 = new javax.swing.JPanel(); botonAlfaON4 = new javax.swing.JButton(); botonAlfaOFF4 = new javax.swing.JToggleButton(); botonDosParpadeos4 = new javax.swing.JButton(); botonTresParpadeos4 = new javax.swing.JToggleButton(); BCI2 = new javax.swing.JPanel(); botonAlfaON3 = new javax.swing.JButton(); botonAlfaOFF3 = new javax.swing.JToggleButton(); botonDosParpadeos3 = new javax.swing.JButton(); botonTresParpadeos3 = new javax.swing.JToggleButton(); panelFondo = new javax.swing.JPanel(); texto = new javax.swing.JLabel(); jToolBar1 = new javax.swing.JToolBar(); botonEnviar = new javax.swing.JButton(); botonIniciar = new javax.swing.JButton(); stop = new javax.swing.JButton(); jSeparator1 = new javax.swing.JToolBar.Separator(); scroll = new javax.swing.JScrollBar(); panelEscritorio = new javax.swing.JDesktopPane(); Fondo = new javax.swing.JLabel(); panel2 = new javax.swing.JPanel(); texto1 = new javax.swing.JLabel(); jMenuBar1 = new javax.swing.JMenuBar(); jMenu1 = new javax.swing.JMenu(); menuAbrrir = new javax.swing.JMenuItem(); menuAceptarCliente = new javax.swing.JMenuItem(); menuConfiguracion = new javax.swing.JMenu(); menuConfiguacion = new javax.swing.JMenuItem(); menuElectrodos = new javax.swing.JMenuItem();

ventanaConfiguracion.setTitle("Configuración"); ventanaConfiguracion.setModal(true); ventanaConfiguracion.setName("ventanaConfiguracion"); // NOI18N ventanaConfiguracion.setResizable(false); jPanel2.setBorder(javax.swing.BorderFactory.createTitledBorder("Tiempo")); grupoTiempo.add(tiempoReal); tiempoReal.setSelected(true); tiempoReal.setText("Real"); grupoTiempo.add(tiempoSimulado); tiempoSimulado.setText("Simulado"); javax.swing.GroupLayout jPanel2Layout = new javax.swing.GroupLayout(jPanel2); jPanel2.setLayout(jPanel2Layout); jPanel2Layout.setHorizontalGroup( jPanel2Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(jPanel2Layout.createSequentialGroup() .addGap(16, 16, 16) .addGroup(jPanel2Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addComponent(tiempoSimulado) .addComponent(tiempoReal, javax.swing.GroupLayout.PREFERRED_SIZE, 56, javax.swing.GroupLayout.PREFERRED_SIZE)) .addContainerGap(37, Short.MAX_VALUE)) ); jPanel2Layout.setVerticalGroup( jPanel2Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(jPanel2Layout.createSequentialGroup() .addContainerGap() .addComponent(tiempoReal) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.UNRELATED)

.addComponent(tiempoSimulado) .addContainerGap(javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE)) ); panelSerial.setBorder(javax.swing.BorderFactory.createTitledBorder("Configuración Serial")); Puertos.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) PuertosActionPerformed(evt); ); botonBuscar.setText("Buscar"); botonBuscar.setBorder(javax.swing.BorderFactory.createEmptyBorder(1, 1, 1, 1)); botonBuscar.setOpaque(false); botonBuscar.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonBuscarActionPerformed(evt); ); botonConectar.setText("Conectar"); botonConectar.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonConectarActionPerformed(evt); ); jLabel1.setText("Puertos"); jLabel2.setText("Baud Rates"); baudRates.setModel(new javax.swing.DefaultComboBoxModel(new String[] "9600", "230400", "500000" )); javax.swing.GroupLayout panelSerialLayout = new javax.swing.GroupLayout(panelSerial); panelSerial.setLayout(panelSerialLayout); panelSerialLayout.setHorizontalGroup(

panelSerialLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(panelSerialLayout.createSequentialGroup() .addGap(18, 18, 18) .addGroup(panelSerialLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(panelSerialLayout.createSequentialGroup() .addGroup(panelSerialLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING, false) .addComponent(jLabel2, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(jLabel1, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE)) .addGap(26, 26, 26) .addGroup(panelSerialLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING, false) .addComponent(Puertos, 0, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(baudRates, 0, 83, Short.MAX_VALUE))) .addGroup(panelSerialLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.TRAILING, false) .addComponent(botonConectar, javax.swing.GroupLayout.Alignment.LEADING, javax.swing.GroupLayout.DEFAULT_SIZE, 91, Short.MAX_VALUE) .addComponent(botonBuscar, javax.swing.GroupLayout.Alignment.LEADING, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE))) .addContainerGap(59, Short.MAX_VALUE)) ); panelSerialLayout.setVerticalGroup( panelSerialLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(panelSerialLayout.createSequentialGroup() .addContainerGap()

.addGroup(panelSerialLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.BASELINE) .addComponent(jLabel1) .addComponent(Puertos, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)) .addGap(11, 11, 11) .addGroup(panelSerialLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.BASELINE) .addComponent(jLabel2) .addComponent(baudRates, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)) .addGap(18, 18, 18) .addComponent(botonBuscar, javax.swing.GroupLayout.PREFERRED_SIZE, 26, javax.swing.GroupLayout.PREFERRED_SIZE) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.UNRELATED) .addComponent(botonConectar) .addContainerGap(javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE)) ); CantidadCanales.setModel(new javax.swing.DefaultComboBoxModel(new String[] "1", "2", "3", "4" )); CantidadCanales.setName("CantidadCanales"); // NOI18N CantidadCanales.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) CantidadCanalesActionPerformed(evt); ); jLabel3.setText("Cantidad de canales"); javax.swing.GroupLayout ventanaConfiguracionLayout = new javax.swing.GroupLayout(ventanaConfiguracion.getContentPane());

ventanaConfiguracion.getContentPane().setLayout(ventanaConfiguracionLayout); ventanaConfiguracionLayout.setHorizontalGroup( ventanaConfiguracionLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(ventanaConfiguracionLayout.createSequentialGroup() .addContainerGap() .addGroup(ventanaConfiguracionLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addComponent(panelSerial, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE) .addGroup(ventanaConfiguracionLayout.createSequentialGroup() .addComponent(jPanel2, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE) .addGap(18, 18, 18) .addComponent(jLabel3) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(CantidadCanales, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE))) .addContainerGap(102, Short.MAX_VALUE)) ); ventanaConfiguracionLayout.setVerticalGroup( ventanaConfiguracionLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(javax.swing.GroupLayout.Alignment.TRAILING, ventanaConfiguracionLayout.createSequentialGroup() .addContainerGap() .addComponent(panelSerial, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)

.addGroup(ventanaConfiguracionLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(ventanaConfiguracionLayout.createSequentialGroup() .addGap(18, 18, 18) .addComponent(jPanel2, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)) .addGroup(ventanaConfiguracionLayout.createSequentialGroup() .addGap(27, 27, 27) .addGroup(ventanaConfiguracionLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.BASELINE) .addComponent(CantidadCanales, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE) .addComponent(jLabel3)))) .addContainerGap(70, Short.MAX_VALUE)) ); elegirArchivo.setMinimumSize(new java.awt.Dimension(500, 500)); elegirArchivo.setPreferredSize(new java.awt.Dimension(680, 500)); elegirArchivo.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) elegirArchivoActionPerformed(evt); ); javax.swing.GroupLayout ventanaElegirArchivoLayout = new javax.swing.GroupLayout(ventanaElegirArchivo.getContentPane()); ventanaElegirArchivo.getContentPane().setLayout(ventanaElegirArchivoLayout); ventanaElegirArchivoLayout.setHorizontalGroup( ventanaElegirArchivoLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(ventanaElegirArchivoLayout.createSequentialGroup() .addComponent(elegirArchivo, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)

.addContainerGap(24, Short.MAX_VALUE)) ); ventanaElegirArchivoLayout.setVerticalGroup( ventanaElegirArchivoLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(ventanaElegirArchivoLayout.createSequentialGroup() .addComponent(elegirArchivo, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE) .addGap(0, 34, Short.MAX_VALUE)) ); ventanaPruebaBCI.setTitle("Ventana manejo Robot"); BCI1.setBorder(javax.swing.BorderFactory.createTitledBorder("BCI 1")); BCI1.setToolTipText("BCI 1"); botonAlfaON1.setText("AlfaON"); botonAlfaON1.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonAlfaON1ActionPerformed(evt); ); botonAlfaOFF1.setText("AlfaOFF"); botonAlfaOFF1.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonAlfaOFF1ActionPerformed(evt); ); botonDosParpadeos1.setText("2 parpadeos"); botonDosParpadeos1.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonDosParpadeos1ActionPerformed(evt); ); botonTresParpadeos1.setText("3 parpadeos"); botonTresParpadeos1.addActionListener(new java.awt.event.ActionListener()

public void actionPerformed(java.awt.event.ActionEvent evt) botonTresParpadeos1ActionPerformed(evt); ); javax.swing.GroupLayout BCI1Layout = new javax.swing.GroupLayout(BCI1); BCI1.setLayout(BCI1Layout); BCI1Layout.setHorizontalGroup( BCI1Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(BCI1Layout.createSequentialGroup() .addContainerGap() .addGroup(BCI1Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addComponent(botonTresParpadeos1, javax.swing.GroupLayout.DEFAULT_SIZE, 110, Short.MAX_VALUE) .addComponent(botonDosParpadeos1, javax.swing.GroupLayout.Alignment.TRAILING, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaON1, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaOFF1, javax.swing.GroupLayout.Alignment.TRAILING, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE)) .addContainerGap()) ); BCI1Layout.setVerticalGroup( BCI1Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(BCI1Layout.createSequentialGroup() .addContainerGap(javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaON1) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonAlfaOFF1)

.addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonDosParpadeos1) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonTresParpadeos1)) ); jPanel3.setBorder(javax.swing.BorderFactory.createTitledBorder("BCI 2")); botonAlfaON2.setText("AlfaON"); botonAlfaON2.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonAlfaON2ActionPerformed(evt); ); botonAlfaOFF2.setText("AlfaOFF"); botonAlfaOFF2.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonAlfaOFF2ActionPerformed(evt); ); botonDosParpadeos2.setText("2 parpadeos"); botonDosParpadeos2.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonDosParpadeos2ActionPerformed(evt); ); botonTresParpadeos2.setText("3 parpadeos"); botonTresParpadeos2.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonTresParpadeos2ActionPerformed(evt); ); javax.swing.GroupLayout jPanel3Layout = new javax.swing.GroupLayout(jPanel3); jPanel3.setLayout(jPanel3Layout); jPanel3Layout.setHorizontalGroup(

jPanel3Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(javax.swing.GroupLayout.Alignment.TRAILING, jPanel3Layout.createSequentialGroup() .addContainerGap() .addGroup(jPanel3Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.TRAILING) .addComponent(botonTresParpadeos2, javax.swing.GroupLayout.DEFAULT_SIZE, 104, Short.MAX_VALUE) .addComponent(botonDosParpadeos2, javax.swing.GroupLayout.Alignment.LEADING, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaOFF2, javax.swing.GroupLayout.Alignment.LEADING, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaON2, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE)) .addContainerGap()) ); jPanel3Layout.setVerticalGroup( jPanel3Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(jPanel3Layout.createSequentialGroup() .addContainerGap() .addComponent(botonAlfaON2) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaOFF2) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonDosParpadeos2) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonTresParpadeos2) .addContainerGap()) );

jPanel4.setBorder(javax.swing.BorderFactory.createTitledBorder("BCI 4")); botonAlfaON4.setText("AlfaON"); botonAlfaON4.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonAlfaON4ActionPerformed(evt); ); botonAlfaOFF4.setText("AlfaOFF"); botonAlfaOFF4.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonAlfaOFF4ActionPerformed(evt); ); botonDosParpadeos4.setText("2 parpadeos"); botonDosParpadeos4.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonDosParpadeos4ActionPerformed(evt); ); botonTresParpadeos4.setText("3 parpadeos"); botonTresParpadeos4.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonTresParpadeos4ActionPerformed(evt); ); javax.swing.GroupLayout jPanel4Layout = new javax.swing.GroupLayout(jPanel4); jPanel4.setLayout(jPanel4Layout); jPanel4Layout.setHorizontalGroup( jPanel4Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(javax.swing.GroupLayout.Alignment.TRAILING, jPanel4Layout.createSequentialGroup() .addContainerGap()

.addGroup(jPanel4Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.TRAILING) .addComponent(botonTresParpadeos4, javax.swing.GroupLayout.DEFAULT_SIZE, 104, Short.MAX_VALUE) .addComponent(botonDosParpadeos4, javax.swing.GroupLayout.Alignment.LEADING, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaOFF4, javax.swing.GroupLayout.Alignment.LEADING, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaON4, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE)) .addContainerGap()) ); jPanel4Layout.setVerticalGroup( jPanel4Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(jPanel4Layout.createSequentialGroup() .addContainerGap(javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaON4) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonAlfaOFF4) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonDosParpadeos4) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonTresParpadeos4)) ); BCI2.setBorder(javax.swing.BorderFactory.createTitledBorder("BCI 3")); BCI2.setToolTipText("BCI 1"); botonAlfaON3.setText("AlfaON"); botonAlfaON3.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonAlfaON3ActionPerformed(evt);

); botonAlfaOFF3.setText("AlfaOFF"); botonAlfaOFF3.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonAlfaOFF3ActionPerformed(evt); ); botonDosParpadeos3.setText("2 parpadeos"); botonDosParpadeos3.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonDosParpadeos3ActionPerformed(evt); ); botonTresParpadeos3.setText("3 parpadeos"); botonTresParpadeos3.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonTresParpadeos3ActionPerformed(evt); ); javax.swing.GroupLayout BCI2Layout = new javax.swing.GroupLayout(BCI2); BCI2.setLayout(BCI2Layout); BCI2Layout.setHorizontalGroup( BCI2Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(BCI2Layout.createSequentialGroup() .addContainerGap() .addGroup(BCI2Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addComponent(botonTresParpadeos3, javax.swing.GroupLayout.DEFAULT_SIZE, 103, Short.MAX_VALUE) .addComponent(botonDosParpadeos3, javax.swing.GroupLayout.Alignment.TRAILING, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE)

.addComponent(botonAlfaON3, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaOFF3, javax.swing.GroupLayout.Alignment.TRAILING, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE)) .addContainerGap()) ); BCI2Layout.setVerticalGroup( BCI2Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(BCI2Layout.createSequentialGroup() .addContainerGap(javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(botonAlfaON3) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonAlfaOFF3) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonDosParpadeos3) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addComponent(botonTresParpadeos3)) ); javax.swing.GroupLayout jPanel1Layout = new javax.swing.GroupLayout(jPanel1); jPanel1.setLayout(jPanel1Layout); jPanel1Layout.setHorizontalGroup( jPanel1Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(jPanel1Layout.createSequentialGroup() .addGap(20, 20, 20) .addGroup(jPanel1Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.TRAILING) .addComponent(BCI2, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)

.addComponent(BCI1, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)) .addGap(18, 18, 18) .addGroup(jPanel1Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addComponent(jPanel3, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE) .addComponent(jPanel4, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)) .addContainerGap(20, Short.MAX_VALUE)) ); jPanel1Layout.setVerticalGroup( jPanel1Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(jPanel1Layout.createSequentialGroup() .addGap(34, 34, 34) .addGroup(jPanel1Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING, false) .addComponent(BCI1, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) .addComponent(jPanel3, javax.swing.GroupLayout.PREFERRED_SIZE, 0, Short.MAX_VALUE)) .addGap(29, 29, 29) .addGroup(jPanel1Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addComponent(BCI2, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE) .addComponent(jPanel4, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE)) .addContainerGap(38, Short.MAX_VALUE))

); BCI1.getAccessibleContext().setAccessibleDescription("B"); javax.swing.GroupLayout ventanaPruebaBCILayout = new javax.swing.GroupLayout(ventanaPruebaBCI.getContentPane()); ventanaPruebaBCI.getContentPane().setLayout(ventanaPruebaBCILayout); ventanaPruebaBCILayout.setHorizontalGroup( ventanaPruebaBCILayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(ventanaPruebaBCILayout.createSequentialGroup() .addContainerGap() .addComponent(jPanel1, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE) .addContainerGap(51, Short.MAX_VALUE)) ); ventanaPruebaBCILayout.setVerticalGroup( ventanaPruebaBCILayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(ventanaPruebaBCILayout.createSequentialGroup() .addComponent(jPanel1, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE) .addGap(0, 16, Short.MAX_VALUE)) ); setDefaultCloseOperation(javax.swing.WindowConstants.EXIT_ON_CLOSE); setTitle("EEG Project"); setIconImage(getIconImage()); setName("ventanaPrincipal"); // NOI18N panelFondo.setBackground(new java.awt.Color(0, 0, 0)); texto.setForeground(new java.awt.Color(0, 204, 204)); jToolBar1.setBackground(new java.awt.Color(0, 0, 0)); jToolBar1.setFloatable(false);

jToolBar1.setForeground(new java.awt.Color(102, 102, 102)); jToolBar1.setRollover(true); botonEnviar.setForeground(new java.awt.Color(255, 255, 255)); botonEnviar.setText("Capturar"); botonEnviar.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonEnviarActionPerformed(evt); ); jToolBar1.add(botonEnviar); botonIniciar.setForeground(new java.awt.Color(255, 255, 255)); botonIniciar.setText("Play"); botonIniciar.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) botonIniciarActionPerformed(evt); ); jToolBar1.add(botonIniciar); stop.setForeground(new java.awt.Color(255, 255, 255)); stop.setText("Stop"); stop.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) stopActionPerformed(evt); ); jToolBar1.add(stop); jToolBar1.add(jSeparator1); scroll.setBackground(new java.awt.Color(102, 102, 102)); scroll.setForeground(new java.awt.Color(0, 204, 204)); scroll.setOrientation(javax.swing.JScrollBar.HORIZONTAL); scroll.setCursor(new java.awt.Cursor(java.awt.Cursor.DEFAULT_CURSOR)); panelEscritorio.setBorder(javax.swing.BorderFactory.createEtchedBorder()); panelEscritorio.setOpaque(false); Fondo.setBackground(new java.awt.Color(51, 51, 255)); Fondo.setForeground(new java.awt.Color(51, 102, 255));

Fondo.setIcon(new javax.swing.ImageIcon(getClass().getResource("/Imagenes/EEG Project.png"))); // NOI18N Fondo.setBounds(0, 0, 1586, 980); panelEscritorio.add(Fondo, javax.swing.JLayeredPane.DEFAULT_LAYER); panel2.setBackground(new java.awt.Color(255, 255, 255)); panel2.setBorder(javax.swing.BorderFactory.createLineBorder(new java.awt.Color(51, 51, 255))); javax.swing.GroupLayout panel2Layout = new javax.swing.GroupLayout(panel2); panel2.setLayout(panel2Layout); panel2Layout.setHorizontalGroup( panel2Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGap(0, 362, Short.MAX_VALUE) ); panel2Layout.setVerticalGroup( panel2Layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGap(0, 189, Short.MAX_VALUE) ); panel2.setBounds(200, 200, 364, 191); panelEscritorio.add(panel2, javax.swing.JLayeredPane.DEFAULT_LAYER); texto1.setForeground(new java.awt.Color(0, 204, 204)); javax.swing.GroupLayout panelFondoLayout = new javax.swing.GroupLayout(panelFondo); panelFondo.setLayout(panelFondoLayout); panelFondoLayout.setHorizontalGroup( panelFondoLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(panelFondoLayout.createSequentialGroup() .addComponent(scroll, javax.swing.GroupLayout.DEFAULT_SIZE, 1125, Short.MAX_VALUE)

.addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.UNRELATED) .addComponent(texto, javax.swing.GroupLayout.PREFERRED_SIZE, 81, javax.swing.GroupLayout.PREFERRED_SIZE) .addGap(18, 18, 18) .addComponent(texto1, javax.swing.GroupLayout.PREFERRED_SIZE, 81, javax.swing.GroupLayout.PREFERRED_SIZE) .addGap(27, 27, 27)) .addComponent(panelEscritorio) .addComponent(jToolBar1, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) ); panelFondoLayout.setVerticalGroup( panelFondoLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addGroup(panelFondoLayout.createSequentialGroup() .addComponent(jToolBar1, javax.swing.GroupLayout.PREFERRED_SIZE, 31, javax.swing.GroupLayout.PREFERRED_SIZE) .addPreferredGap(javax.swing.LayoutStyle.ComponentPlacement.RELATED) .addGroup(panelFondoLayout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addComponent(scroll, javax.swing.GroupLayout.PREFERRED_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.PREFERRED_SIZE) .addComponent(texto, javax.swing.GroupLayout.PREFERRED_SIZE, 23, javax.swing.GroupLayout.PREFERRED_SIZE) .addComponent(texto1, javax.swing.GroupLayout.PREFERRED_SIZE, 23, javax.swing.GroupLayout.PREFERRED_SIZE)) .addGap(11, 11, 11) .addComponent(panelEscritorio, javax.swing.GroupLayout.DEFAULT_SIZE, 951, Short.MAX_VALUE)) ); jMenu1.setText("Menu");

menuAbrrir.setAccelerator(javax.swing.KeyStroke.getKeyStroke(java.awt.event.KeyEvent.VK_O, java.awt.event.InputEvent.CTRL_MASK)); menuAbrrir.setText("Abrrir EEG"); menuAbrrir.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) menuAbrrirActionPerformed(evt); ); jMenu1.add(menuAbrrir); menuAceptarCliente.setAccelerator(javax.swing.KeyStroke.getKeyStroke(java.awt.event.KeyEvent.VK_A, java.awt.event.InputEvent.CTRL_MASK)); menuAceptarCliente.setText("Aceptar cliente"); menuAceptarCliente.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) menuAceptarClienteActionPerformed(evt); ); jMenu1.add(menuAceptarCliente); jMenuBar1.add(jMenu1); menuConfiguracion.setText("Configuración"); menuConfiguracion.addMouseListener(new java.awt.event.MouseAdapter() public void mouseClicked(java.awt.event.MouseEvent evt) menuConfiguracionMouseClicked(evt); ); menuConfiguacion.setAccelerator(javax.swing.KeyStroke.getKeyStroke(java.awt.event.KeyEvent.VK_C, java.awt.event.InputEvent.CTRL_MASK)); menuConfiguacion.setText("Configuración"); menuConfiguacion.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) menuConfiguacionActionPerformed(evt); ); menuConfiguracion.add(menuConfiguacion);

menuElectrodos.setAccelerator(javax.swing.KeyStroke.getKeyStroke(java.awt.event.KeyEvent.VK_T, java.awt.event.InputEvent.CTRL_MASK)); menuElectrodos.setText("Testear BCI"); menuElectrodos.addActionListener(new java.awt.event.ActionListener() public void actionPerformed(java.awt.event.ActionEvent evt) menuElectrodosActionPerformed(evt); ); menuConfiguracion.add(menuElectrodos); jMenuBar1.add(menuConfiguracion); setJMenuBar(jMenuBar1); javax.swing.GroupLayout layout = new javax.swing.GroupLayout(getContentPane()); getContentPane().setLayout(layout); layout.setHorizontalGroup( layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addComponent(panelFondo, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) ); layout.setVerticalGroup( layout.createParallelGroup(javax.swing.GroupLayout.Alignment.LEADING) .addComponent(panelFondo, javax.swing.GroupLayout.DEFAULT_SIZE, javax.swing.GroupLayout.DEFAULT_SIZE, Short.MAX_VALUE) ); pack(); // </editor-fold> /* @Override public Image getIconImage() Image retValue = Toolkit.getDefaultToolkit(). getImage(ClassLoader.getSystemResource("imagenes/cerebro.png")); return retValue; */

private void menuConfiguracionMouseClicked(java.awt.event.MouseEvent evt) // TODO add your handling code here: private void botonIniciarActionPerformed(java.awt.event.ActionEvent evt) miSistema.getProceso().setPlay(true); private void botonEnviarActionPerformed(java.awt.event.ActionEvent evt) if(miSistema.getSerial().getConectado()) miSistema.getSerial().enviarPorSerial(this.CantidadCanales.getSelectedIndex()+1); miSistema.getBuffer().setCantidadDeCanales(this.CantidadCanales.getSelectedIndex()+1); miSistema.getProceso().start(); System.out.println("capturando"); else this.mensaje("Imposible capturar", "ERROR",2 ); private void stopActionPerformed(java.awt.event.ActionEvent evt) miSistema.getProceso().setPlay(false); private void menuAbrrirActionPerformed(java.awt.event.ActionEvent evt) int respuesta = this.elegirArchivo.showOpenDialog(null); // Si apretamos en aceptar ocurrirá esto if (respuesta == JFileChooser.APPROVE_OPTION) File selectedFile = elegirArchivo.getSelectedFile(); System.out.println(selectedFile.toString()); this.miSistema.getBuffer().cargarDeArchivo(selectedFile.toString()); // Si apretamos en cancelar o cerramos la ventana ocurrirá esto //Elegiremos archivos del directorio

private void elegirArchivoActionPerformed(java.awt.event.ActionEvent evt) // TODO add your handling code here: private void botonBuscarActionPerformed(java.awt.event.ActionEvent evt) ArrayList misPuertos = miSistema.getSerial().getPuertos(); Puertos.removeAll(); for (int i = 0; i < misPuertos.size(); i++) this.Puertos.addItem(misPuertos.get(i)); private void botonConectarActionPerformed(java.awt.event.ActionEvent evt) if (Puertos.getSelectedItem() == null) this.mensaje("Seleccione un puerto", "ATENCION", 2); else String baudRate = (String) this.baudRates.getSelectedItem(); System.out.println(baudRate); miSistema.getSerial().initialize(Puertos.getSelectedItem(), Integer.valueOf(baudRate)); private void PuertosActionPerformed(java.awt.event.ActionEvent evt) // TODO add your handling code here: private void menuConfiguacionActionPerformed(java.awt.event.ActionEvent evt) // TODO add your handling code here: this.ventanaConfiguracionShow(); private void menuElectrodosActionPerformed(java.awt.event.ActionEvent evt) // TODO add your handling code here: this.ventanaElectrodosShow();

private void CantidadCanalesActionPerformed(java.awt.event.ActionEvent evt) // TODO add your handling code here: miSistema.getBuffer().setCantidadDeCanales(this.CantidadCanales.getSelectedIndex()+1); private void botonTresParpadeos1ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 0,3; miSistema.getServidor().enviarPorSocket(vec); private void botonAlfaON1ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 0,1; miSistema.getServidor().enviarPorSocket(vec); private void botonDosParpadeos1ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 0,2; miSistema.getServidor().enviarPorSocket(vec); private void botonAlfaOFF1ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 0,0; miSistema.getServidor().enviarPorSocket(vec); private void menuAceptarClienteActionPerformed(java.awt.event.ActionEvent evt) miSistema.getServidor().acept(); private void botonAlfaON2ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 1,1;

miSistema.getServidor().enviarPorSocket(vec); private void botonAlfaOFF2ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 1,0; miSistema.getServidor().enviarPorSocket(vec); private void botonDosParpadeos2ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 1,2; miSistema.getServidor().enviarPorSocket(vec); private void botonTresParpadeos2ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 1,3; miSistema.getServidor().enviarPorSocket(vec); private void botonAlfaON3ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 2,1; miSistema.getServidor().enviarPorSocket(vec); private void botonAlfaOFF3ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 2,0; miSistema.getServidor().enviarPorSocket(vec); private void botonDosParpadeos3ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 2,2; miSistema.getServidor().enviarPorSocket(vec);

private void botonTresParpadeos3ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 2,3; miSistema.getServidor().enviarPorSocket(vec); private void botonAlfaON4ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 3,1; miSistema.getServidor().enviarPorSocket(vec); private void botonAlfaOFF4ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 3,0; miSistema.getServidor().enviarPorSocket(vec); private void botonDosParpadeos4ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 3,2; miSistema.getServidor().enviarPorSocket(vec); private void botonTresParpadeos4ActionPerformed(java.awt.event.ActionEvent evt) byte[] vec = 3,3; miSistema.getServidor().enviarPorSocket(vec); /** * @param args the command line arguments */ // Variables declaration - do not modify private javax.swing.JPanel BCI1; private javax.swing.JPanel BCI2; private javax.swing.JComboBox CantidadCanales; private javax.swing.JLabel Fondo; private javax.swing.JComboBox Puertos; private javax.swing.JComboBox baudRates; private javax.swing.JToggleButton botonAlfaOFF1;

private javax.swing.JToggleButton botonAlfaOFF2; private javax.swing.JToggleButton botonAlfaOFF3; private javax.swing.JToggleButton botonAlfaOFF4; private javax.swing.JButton botonAlfaON1; private javax.swing.JButton botonAlfaON2; private javax.swing.JButton botonAlfaON3; private javax.swing.JButton botonAlfaON4; private javax.swing.JButton botonBuscar; private javax.swing.JButton botonConectar; private javax.swing.JButton botonDosParpadeos1; private javax.swing.JButton botonDosParpadeos2; private javax.swing.JButton botonDosParpadeos3; private javax.swing.JButton botonDosParpadeos4; private javax.swing.JButton botonEnviar; private javax.swing.JButton botonIniciar; private javax.swing.JToggleButton botonTresParpadeos1; private javax.swing.JToggleButton botonTresParpadeos2; private javax.swing.JToggleButton botonTresParpadeos3; private javax.swing.JToggleButton botonTresParpadeos4; private javax.swing.JFileChooser elegirArchivo; private javax.swing.ButtonGroup grupoTiempo; private javax.swing.JLabel jLabel1; private javax.swing.JLabel jLabel2; private javax.swing.JLabel jLabel3; private javax.swing.JMenu jMenu1; private javax.swing.JMenuBar jMenuBar1; private javax.swing.JPanel jPanel1; private javax.swing.JPanel jPanel2; private javax.swing.JPanel jPanel3; private javax.swing.JPanel jPanel4; private javax.swing.JToolBar.Separator jSeparator1; private javax.swing.JToolBar jToolBar1; private javax.swing.JMenuItem menuAbrrir; private javax.swing.JMenuItem menuAceptarCliente; private javax.swing.JMenuItem menuConfiguacion; private javax.swing.JMenu menuConfiguracion; private javax.swing.JMenuItem menuElectrodos; private javax.swing.JPanel panel2; private javax.swing.JDesktopPane panelEscritorio; private javax.swing.JPanel panelFondo; private javax.swing.JPanel panelSerial; private javax.swing.JScrollBar scroll; private javax.swing.JButton stop; private javax.swing.JLabel texto;

private javax.swing.JLabel texto1; private javax.swing.JRadioButton tiempoReal; private javax.swing.JRadioButton tiempoSimulado; private javax.swing.JDialog ventanaConfiguracion; private javax.swing.JDialog ventanaElegirArchivo; private javax.swing.JDialog ventanaPruebaBCI; // End of variables declaration

A.5.2.2. Clase main

285

package Ventanas; import Dominio.Buffer; import Dominio.Proceso; import Dominio.Serial; import Dominio.Sistema; import Dominio.Servidor; /** * ***************************Clase main************************************** * * Se encarga de ejecutar el sistema */ public class EEG public static void main(String[] args) // TODO code application logic here Sistema miSistema = new Sistema(); Buffer unB = new Buffer(miSistema); Serial unS = new Serial(miSistema); Proceso unProceso = new Proceso(miSistema); Ventana unV = new Ventana(miSistema); Servidor unServidor = new Servidor(); unServidor.initServer(miSistema); miSistema.setVentana(unV); miSistema.setBuffer(unB); miSistema.setSerial(unS); miSistema.setProceso(unProceso); miSistema.setServidor(unServidor); unV.setVisible(true);

A.6. Código de programa de comunicación

PC - NXT

287

import java.io.DataInputStream; import java.io.DataOutputStream; import java.io.IOException; import lejos.pc.comm.NXTCommLogListener; import lejos.pc.comm.NXTConnector; import java.io.* ; import java.net.* ; public class ComunicacionPC_NXT //Se crea una variable que indica que la conexión sera local static final String HOST = "localhost"; //Se elige el puerto de conexión static final int PUERTO = 5000; public static void main(String[] args) //Se crea el objeto que se utilizara para la conexión BT con el NXT NXTConnector conn = new NXTConnector(); // conn.addLogListener(new NXTCommLogListener() public void logEvent(String mensaje) System.out.println("Log de envío BT: "+mensaje); public void logEvent(Throwable throwable) System.out.println("Log de envío BT - rastreo de pila: "); throwable.printStackTrace(); ); //me conecto via BT al NXT boolean connected = conn.connectTo("btspp://"); //Si fallo la conexión, lo indico y termino el programa if (!connected) System.err.println("Conexión a NXT fallida"); System.exit(1);

//Se crea el canal de comunicación bidireccional DataOutputStream dos = new DataOutputStream(conn.getOutputStream()); DataInputStream dis = new DataInputStream(conn.getInputStream()); byte data; boolean recibiCanal = false; byte canal = 0; try //Se crea el socket local conectandose al puerto 5000 Socket skCliente = new Socket(HOST,PUERTO); //Se crea comunicación bidireccional InputStream aux = skCliente.getInputStream(); DataInputStream flujo = new DataInputStream( aux ); OutputStream auxOut = skCliente.getOutputStream(); DataOutputStream flujoOut = new DataOutputStream(auxOut); //hasta que el servidor cierre la conexión while(connected) try //se espera el envío de datos data = flujo.readByte(); //Si se recibe el comando de finalizar if(data==4) connected = false; dos.writeByte(data); dos.flush(); else if(recibiCanal) if(canal==0) if((data==2) || (data==3)) dos.writeByte(data); dos.flush();

else if(canal==1) if((data==0) || (data==1)) dos.writeByte(data); dos.flush(); recibiCanal = false; else canal=data; recibiCanal = true; catch (IOException ioe) System.out.println("Error enviando datos:"); System.out.println(ioe.getMessage()); try //se cierra el socket skCliente.close(); //se cierran los canales de comunicación con el NXT dis.close(); dos.close(); //Se cierra la comunicación con el NXT conn.close(); catch (IOException ioe) System.out.println("Error cerrando la conexión:"); System.out.println(ioe.getMessage()); catch( Exception e ) System.out.println( e.getMessage() );

A.7. Código robot NXT

291

import java.io.DataInputStream;

import lejos.nxt.Motor;

import lejos.robotics.navigation.DifferentialPilot;

import java.io.DataOutputStream;

import lejos.nxt.LCD;

import lejos.nxt.comm.BTConnection;

import lejos.nxt.comm.Bluetooth;

public class ProgramaNXT

public static void main(String [] args) throws Exception

//Se crea un objeto pilot que controla los motores del NXT

DifferentialPilot pilot= new DifferentialPilot(5.6, 10.3, Motor.A, Motor.B);

//Se indica la velocidad de movimiento en cm/s

pilot.setTravelSpeed(8);

boolean terminar =false;

//se dibuja el mensaje esperando en la pantalla del NXT

LCD.drawString("Esperando...",0,0);

LCD.refresh();

//Se espera una conexion por BT

BTConnection btc = Bluetooth.waitForConnection();

//Se limpia la pantalla y se imprime conectado

LCD.clear();

LCD.drawString("Conectado",0,0);

LCD.refresh();

//se cren los canales de comunicación

DataInputStream dis = btc.openDataInputStream();

DataOutputStream dos = btc.openDataOutputStream();

byte recibo=0;

//mientras no reciba el comando de terminar

while (!terminar)

//espero el envío de datos

recibo=dis.readByte();

//si recibe 1 avanza

if(recibo==1)

pilot.forward();

LCD.clear();

LCD.drawString("Conectado",0,0);

LCD.drawString("Estoy avanzando", 0, 1);

//si recibe 0 para

else if(recibo==0)

pilot.stop();

LCD.clear();

LCD.drawString("Conectado",0,0);

LCD.drawString("Estoy quieto", 0, 1);

//si recibe 12 gira a la derecha

else if(recibo==2)

pilot.setRotateSpeed(22.5);

pilot.rotateRight();

LCD.clear();

LCD.drawString("Conectado",0,0);

LCD.drawString("Giro a la", 0, 1);

LCD.drawString("derecha", 0, 2);

//si recibe 3 gira a la izquierda

else if(recibo==3)

pilot.setRotateSpeed(22.5);

pilot.rotateLeft();

LCD.clear();

LCD.drawString("Conectado",0,0);

LCD.drawString("Giro a la", 0, 1);

LCD.drawString("izquierda", 0, 2);

else if(recibo==4)

terminar=true;

//se cierra la conexión

dis.close();

dos.close();

Thread.sleep(100);

LCD.clear();

LCD.drawString("Cerrando",0,0);

LCD.refresh();

btc.close();

LCD.clear();

A.8. Diagrama esquemático de la placa

amplicadora

295

02/06/2014 08:34:18 p.m. D:\Lalo\Proyecto\ModularEEG-v1.1.1\modEEGamp_v1.1.sch (Sheet: 1/1)

INA114P

2k2

TLC277P

2k2

10k

200k

U_cal

Cal_GND

1µF

1nF

1nF

100nF

100nF

100p

F10

0pF

2k22k2

10pF

2k2

BC547 BC557

2k22k2 2k2

BC547 BC557

100nF

100nF

TLC277P

TLC277P

INA114P

2k2

2k2

1µF

100nF

100nF

100p

F10

0pF

2k22k2

10pF

2k2

BC547 BC557

2k22k2 2k2

BC547 BC557

10k

20k

COM

AGND

IC2032

16

V+ 7

V- 4

8

35

R216

IC201B

6

57

R217

R218

R213R_LEG

CH1-

CH1+

PAD203

PAD204

C221

C212

C216

C218

C215

C20

7C

206

R211R204

C21

0

R212

2

13

Q206

2

13

Q202

R210R203 R209

2

13Q208

2

13 Q204

R22

6

C219

C226

IC204A

2

31

IC204B

6

57

IC204PWR

GN

DV

CC

48

IC2022

16

V+ 7

V- 4

8

35

R214

R215

CH2-

CH2+

C220C217

C214

C20

5C

204

R207R202

C20

9

R208

2

13

Q205

2

13

Q201

R206R201 R205

2

13Q207

2

13 Q203

R22

4

PA

D20

1P

AD

202

R219

P20

1

13

2

PAD205

PAD206

AGND

AGND

VGND

VGND

VGND

VGND

VGND

VGND

VGND

VGND

R_LEG

right leg driver

Electrode

Electrode

Channel 1 -

Channel 1 +

Right LegElectrode

u ca

l

simulates electrode impedance

HP 1 pole fc=0.16 Hz

DRL design from http://www.biosemi.com/publications/artikel7.htm, fig.3

HF rejectionESD protectionand user current limiter

Electrode

Electrode

HP 1 pole fc=0.16 Hz

Channel 2 -

Channel 2 +

DC checkpoint

DC checkpoint

Electrode

Electrode

G=12.2

G=12.2

Total gain = 7812.5Input Voltage Resolution (1 LSB) = 0.5 uVppInput Voltage Full Scale (10 Bit) = 512uVpp

Input can handle up to +/-100mV DC electrode offset

Authors: Moritz v. Buttlar, Joerg Hansmann, Andreas R

See http://openeeg.sf.net for more information.

ModularEEG - 2-channel EEG amplifier module

AGND

+5V/2

Date: 07/08/2013 05:46:47 p.m. Sheet: 1/1

REV:

TITLE:

Document Number:

modEEGamp_v1.1

AGND

+5V/2

AGND

+5V/2

13

If P201 is needed, adjust potentiometer so DRL=0mV (referred to VGND)when _all_ amplifier inputs are shorted to the DRL output (R_LEG).

If you only want to use one channel, never let the other channel float.

Important usage instructions for the DRL.

Always connect the unused terminals to VGND, or the DRL will not work properly.

P201 is not needed when INA114 instrumentation amplifiers are used.You may replace it with a short wire from pin 2 and pin 1 (VGND).

Right-leg driver (DRL) notes:

TLC277P

100nF

10nF

100nF 47

µF

47µF tan 1ohm

100

U_cal

Cal_GND

10nF

PINHD-2X17

10k

10k

100n

F

100k1k

1M

1nF

20k

100k8.2k

1M

1nF

1µF

100n

F

TLC277P TLC277P

100n

F

100k1k

1M

1nF

20k

100k8.2k

1M1nF

1µF

100n

F

TLC277P TLC277P

1M1M

10k 15k

33nF 5%

220n

F 5

%33nF 5%

220n

F 5

%10k 15k

220n

F 5

%

220n

F 5

%

7.5k

7.5k

2

31

84

IC201AC211

C20

2

C208

C21

3

C201

R240

PAD203

PAD204

C20

3

J2011 23 45 67 89 10

11 1213 1415 1617 1819 2021 2223 2425 2627 2829 3031 3233 34

R238

R239

C23

0R225R222

C224

13

2

P203

R235R234

R23

6

C233

C228

C22

3

IC206A

2

31

IC206B

6

57

IC206PWR

GN

DV

CC

48

C22

7R221R220

C222

13

2

P202

R230R229

R23

1C232

C229

C22

5

IC205A

2

31

IC205B

6

57

GN

DV

CC

48

IC205PWR

212121

212121

R223R227

R232 R228

C234

C23

1C236

C23

5

R237 R233

C1

C2

R1

R2

AGND

VGND

VGND

VGND

VGND

VGND

VGND

VGND

VGND

VGND

ADIN1

+5V/2

2.0V

2.0V

PWM

ADIN0

ADIN0P

ADIN0P

ADIN1P

ADIN1P

ref. voltage+2V, buffered

Output Voltage Full Scale (10 Bit) = 4.000 Vp-p (Range 0..4V)

Vcc

ana

log

2.0V buf

PWM cal (PB1)

250µVpp +/-10%, 0.1 .. 100Hz Square wave Calibration Signal

voltage divider 1:20000

simulates electrode impedance

HP 1 pole fc=0.16 Hz

HP 1 pole fc=0.16 Hz

Ground plane is VGND

G=16

G=16G=6..100G=40

G=40G=6..100

5Vp-p +/-10% => 250µVp-p +/-10%

when using 2 or 3 amplifier boardsOnly insert IC201 on _one_ board

The solder-jumpers (SJ201 - SJ206)

boards to share the same connector.

2 channels must be selected on each

solder.

to the right, are used for channelmapping. This allows multiple amplifier

3rd order "Besselworth" filter, fc = 59 Hz.

=> 3.9mV bitstep at 10 bit resolution.

board by closing two jumper gaps with

Use SJ201 and SJ204 for the first board.

The 3rd pole is located on the digital board.

+

AGND

+

AGND

+5V/2

13

+5V/2

AGND

+5V/2

13

AGND

AGND

A.9. Layout de la placa amplicadora

298

02/06/2014 08:39:37 p.m. f=2.00 D:\Lalo\Proyecto\ModularEEG-v1.1.1\modEEGamp_v1.1.brd

31

31

31

11

11

3

11

11

3

3

ModularEEGamplifier v1.1.0

-

+

-

+

IC20

3

R216

IC20

1

R217

R218

R21

3

C211C202

C208

C213

C20

1

R240

U_calCal_GND

C22

1

C203

C212

C21

6

C21

8

C215

J201

R238

C207

C206

R211

R204

C210

R212

Q20

6Q202

R210

R203R209

Q208Q204

R239

C230R

225

R22

2

R22

6C224

P203

R235R234

R236

C233

C228

C223

C219

C226

IC20

6

IC20

4

IC20

2

R214R215

C22

0

C21

7

C214

C205

C204

R207

R202

C209

R208

Q20

5Q201

R206

R201R205

Q207Q203

C227

R22

1R

220

R22

4

C222P20

2

R230R229

R231

C232

C229

C225

IC20

5

R219

R223R227

P201

R232

R228

C234

C231

C236

C235

R237

R233

COM

AGND

C1

C2

R1 R

2

B. Anexo

300

B.1. Hoja de datos ATmega 2560

301

2549A–AVR–03/05

Features• High Performance, Low Power AVR® 8-Bit Microcontroller• Advanced RISC Architecture

– 135 Powerful Instructions – Most Single Clock Cycle Execution– 32 x 8 General Purpose Working Registers– Fully Static Operation– Up to 16 MIPS Throughput at 16 MHz– On-Chip 2-cycle Multiplier

• Non-volatile Program and Data Memories– 64K/128K/256K Bytes of In-System Self-Programmable Flash

Endurance: 10,000 Write/Erase Cycles– Optional Boot Code Section with Independent Lock Bits

In-System Programming by On-chip Boot ProgramTrue Read-While-Write Operation

– 4K Bytes EEPROMEndurance: 100,000 Write/Erase Cycles

– 8K Bytes Internal SRAM– Up to 64K Bytes Optional External Memory Space– Programming Lock for Software Security

• JTAG (IEEE std. 1149.1 compliant) Interface– Boundary-scan Capabilities According to the JTAG Standard– Extensive On-chip Debug Support– Programming of Flash, EEPROM, Fuses, and Lock Bits through the JTAG Interface

• Peripheral Features– Two 8-bit Timer/Counters with Separate Prescaler and Compare Mode– Four 16-bit Timer/Counter with Separate Prescaler, Compare- and Capture Mode– Real Time Counter with Separate Oscillator– Four 8-bit PWM Channels– Six/Twelve PWM Channels with Programmable Resolution from 2 to 16 Bits

(ATmega1281/2561, ATmega640/1280/2560)– Output Compare Modulator– 8/16-channel, 10-bit ADC– Two/Four Programmable Serial USART (ATmega1281/2561,ATmega640/1280/2560)– Master/Slave SPI Serial Interface– Byte Oriented 2-wire Serial Interface– Programmable Watchdog Timer with Separate On-chip Oscillator– On-chip Analog Comparator– Interrupt and Wake-up on Pin Change

• Special Microcontroller Features– Power-on Reset and Programmable Brown-out Detection– Internal Calibrated Oscillator– External and Internal Interrupt Sources– Six Sleep Modes: Idle, ADC Noise Reduction, Power-save, Power-down, Standby,

and Extended Standby• I/O and Packages

– 51/86 Programmable I/O Lines (ATmega1281/2561, ATmega640/1280/2560)– 64-lead (ATmega1281/2561)– 100-lead (ATmega640/1280/2560)– 100-lead TQFP (64-lead TQFP Option)

• Temperature Range:– -40°C to 85°C Industrial

• Speed Grade:– ATmega1281/2561V/ATmega640/1280/2560V:

0 - 4 MHz @ 1.8 - 5.5V, 0 - 8 MHz @ 2.7 - 5.5V– ATmega640/1280/1281/2560/2561:

0 - 8 MHz @ 2.7 - 5.5V, 0 - 16 MHz @ 4.5 - 5.5V

8-bit Microcontroller with 256K Bytes In-SystemProgrammable Flash

ATmega1281/2561/VATmega640/1280/2560/V

Advance Information

2 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Pin Configurations Figure 1. Pinout ATmega640/1280/2560

GN

D

VC

C

PA0

(AD

0)

PA1

(AD

1)

PA2

(AD

2)

PA3 (AD3)

PA4 (AD4)

PA5 (AD5)

PA6 (AD6)

PA7 (AD7)

PG2 (ALE)

AV

CC

GN

D

AR

EF

PF

0 (A

DC

0)

PF

1 (A

DC

1)

PF

2 (A

DC

2)

PF

3 (A

DC

3)

PF

4 (A

DC

4/T

CK

)

PF

5 (A

DC

5/T

MS

)

PF

6 (A

DC

6/T

DO

)

PF

7 (A

DC

7/T

DI)

ATmega640/1280/2560

100 99 98 97 96 95 94 93 92 91 90 89 88 87 86 85 84 83 82 81 80 79 78 77 76

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

17

18

19

20

21

22

23

24

25

75

74

73

72

71

70

69

68

67

66

65

64

63

62

61

60

59

58

57

56

55

54

53

52

51

26 28 29 3127 3630 32 35 3733 34 38 39 40 41 42 43 44 45 46 47 48 49 50

PK

0 (A

DC

8/P

CIN

T16

)

PK

1 (A

DC

9/P

CIN

T17

)

PK

2 (A

DC

10/P

CIN

T18

)

PK

3 (A

DC

11/P

CIN

T19

)

PK

4 (A

DC

12/P

CIN

T20

)

PK

5 (A

DC

13/P

CIN

T21

)

PK

6 (A

DC

14/P

CI N

T22

)

PK

7 (A

DC

15/P

CI N

T23

)

(OC2B) PH6

(TO

SC

2) P

G3

(TO

SC

1) P

G4

RE

SE

T

(T4)

PH

7

(IC

P4)

PL0

VC

C

GN

D

XTA

L2

XTA

L1

PL6

PL7

GND

VCC

(OC0B) PG5

VCC

GND

(RXD2) PH0

(TXD2) PH1

(XCK2) PH2

(OC4A) PH3

(OC4B) PH4

(OC4C) PH5

(RXD0/PCINT8) PE0

(TXD0) PE1

(XCK0/AIN0) PE2

(OC3A/AIN1) PE3

(OC3B/INT4) PE4

(OC3C/INT5) PE5

(T3/INT6) PE6

(CLKO/ICP3/INT7) PE7

(SS/PCINT0) PB0

(SCK/PCINT1) PB1

(MOSI/PCINT2) PB2

(MISO/PCINT3) PB3

(OC2A/PCINT4) PB4

(OC1A/PCINT5) PB5

(OC1B/PCINT6) PB6

(OC

0A/O

C1C

/PC

INT

7) P

B7

PC7 (A15)

PC6 (A14)

PC5 (A13)

PC4 (A12)

PC3 (A11)

PC2 (A10)

PC1 (A9)

PC0 (A8)

PG1 (RD)

PG0 (WR)

(TX

D1/

INT

3) P

D3

(IC

P1)

PD

4

(XC

K1)

PD

5

(T1)

PD

6

(T0)

PD

7

(SC

L/IN

T0)

PD

0

(SD

A/IN

T1)

PD

1

(RX

D1/

INT

2) P

D2

(IC

P5)

PL1

(T5)

PL2

(OC

5A)

PL3

(OC

5B)

PL4

PJ6 (PCINT15)

PJ5 (PCINT14)

PJ4 (PCINT13)

PJ3 (PCINT12)

PJ2 (XCK3/PCINT11)

PJ1 (TXD3/PCINT10)

PJ0 (RXD3/PCINT9)

PJ7

(OC

5C)

PL5

INDEX CORNER

3

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 2. Pinout ATmega1281/2561

Note: The large center pad underneath the QFN/MLF package is made of metal and internallyconnected to GND. It should be soldered or glued to the board to ensure good mechani-cal stability. If the center pad is left unconnected, the package might loosen from theboard.

Disclaimer Typical values contained in this datasheet are based on simulations and characteriza-tion of other AVR microcontrollers manufactured on the same process technology. Minand Max values will be available after the device is characterized.

ATmega1281/2561

(RXD0/PCINT8/PDI) PE0

(TXD0/PDO) PE1

(XCK0/AIN0) PE2

(OC3A/AIN1) PE3

(OC3B/INT4) PE4

(OC3C/INT5) PE5

(T3/INT6) PE6

(ICP3/CLKO/INT7) PE7

(SS/PCINT0) PB0

(OC0B) PG5

(SCK/PCINT1) PB1

(MOSI/PCINT2) PB2

(MISO/PCINT3) PB3

(OC2A/ PCINT4) PB4

(OC1A/PCINT5) PB5

(OC1B/PCINT6) PB6

(OC

0A/O

C1C

/PC

INT

7) P

B7

(TO

SC

2) P

G3

(TO

SC

1) P

G4

RE

SE

T

VC

C

GN

D

XTA

L2

XTA

L1

(SC

L/IN

T0)

PD

0

(SD

A/IN

T1)

PD

1

(RX

D1/

INT

2) P

D2

(TX

D1/

INT

3) P

D3

(IC

P1)

PD

4

(XC

K1)

PD

5

PA3 (AD3)

PA4 (AD4)

PA5 (AD5)

PA6 (AD6)

PA7 (AD7)

PG2 (ALE)

PC7 (A15)

PC6 (A14)

PC5 (A13)

PC4 (A12)

PC3 (A11)

PC2 (A10)

PC1 (A9)

PC0 (A8)

PG1 (RD)

PG0 (WR)

AV

CC

GN

D

AR

EF

PF

0 (A

DC

0)

PF

1 (A

DC

1)

PF

2 (A

DC

2)

PF

3 (A

DC

3)

PF

4 (A

DC

4/T

CK

)

PF

5 (A

DC

5/T

MS

)

PF

6 (A

DC

6/T

DO

)

PF

7 (A

DC

7/T

DI)

GN

D

VC

C

PA0

(AD

0)

PA1

(AD

1)

PA2

(AD

2)

(T1)

PD

6

(T0)

PD

7

INDEX CORNER

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

16

64 63 62 61 60 59 58 57 56 55 54 53 52 51 50 49

48

47

46

45

44

43

42

41

40

39

38

37

36

35

34

33

17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32

4 ATmega640/1280/1281/2560/25612549A–AVR–03/05

OverviewThe ATmega640/1280/1281/2560/2561 is a low-power CMOS 8-bit microcontroller based on the AVR enhanced RISCarchitecture. By executing powerful instructions in a single clock cycle, the ATmega640/1280/1281/2560/2561 achievesthroughputs approaching 1 MIPS per MHz allowing the system designer to optimize power consumption versus processingspeed.

Block Diagram

Figure 3. Block Diagram

CPU

GND

VCC

RESET

PowerSupervision

POR / BOD &RESET

WatchdogOscillator

WatchdogTimer

OscillatorCircuits /

ClockGeneration

XTAL1

XTAL2

PC7..0 PORT C (8)

PA7..0 PORT A (8)

PORT D (8)

PD7..0

PORT B (8)

PB7..0

PORT E (8)

PE7..0

PORT F (8)

PF7..0

PORT J (8)

PJ7..0

PG5..0 PORT G (6)

PORT H (8)

PH7..0

PORT K (8)

PK7..0

PORT L (8)

PL7..0

XRAM

TWI SPI

EEPROM

JTAG

8bit T/C 0 8bit T/C 2

16bit T/C 1

16bit T/C 3

SRAMFLASH

16bit T/C 4

16bit T/C 5

USART 2

USART 1

USART 0

Internal Bandgap reference

Analog Comparator

A/DConverter

USART 3

NOTE:Shaded parts only appearin the 100 pin version.

The ADC, T/C4 and T/C5 only have full functionality in the 100 pin version.

5

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The AVR core combines a rich instruction set with 32 general purpose working registers.All the 32 registers are directly connected to the Arithmetic Logic Unit (ALU), allowingtwo independent registers to be accessed in one single instruction executed in one clockcycle. The resulting architecture is more code efficient while achieving throughputs up toten times faster than conventional CISC microcontrollers.

The ATmega640/1280/1281/2560/2561 provides the following features: 64K/128K/256Kbytes of In-System Programmable Flash with Read-While-Write capabilities, 4K bytesEEPROM, 8K bytes SRAM, 54/86 general purpose I/O lines, 32 general purpose work-ing registers, Real Time Counter (RTC), six flexible Timer/Counters with comparemodes and PWM, 4 USARTs, a byte oriented 2-wire Serial Interface, a 16-channel, 10-bit ADC with optional differential input stage with programmable gain, programmableWatchdog Timer with Internal Oscillator, an SPI serial port, IEEE std. 1149.1 compliantJTAG test interface, also used for accessing the On-chip Debug system and program-ming and six software selectable power saving modes. The Idle mode stops the CPUwhile allowing the SRAM, Timer/Counters, SPI port, and interrupt system to continuefunctioning. The Power-down mode saves the register contents but freezes the Oscilla-tor, disabling all other chip functions until the next interrupt or Hardware Reset. InPower-save mode, the asynchronous timer continues to run, allowing the user to main-tain a timer base while the rest of the device is sleeping. The ADC Noise Reductionmode stops the CPU and all I/O modules except Asynchronous Timer and ADC, to min-imize switching noise during ADC conversions. In Standby mode, the Crystal/ResonatorOscillator is running while the rest of the device is sleeping. This allows very fast start-upcombined with low power consumption. In Extended Standby mode, both the mainOscillator and the Asynchronous Timer continue to run.

The device is manufactured using Atmel’s high-density nonvolatile memory technology.The On-chip ISP Flash allows the program memory to be reprogrammed in-systemthrough an SPI serial interface, by a conventional nonvolatile memory programmer, orby an On-chip Boot program running on the AVR core. The boot program can use anyinterface to download the application program in the application Flash memory. Soft-ware in the Boot Flash section will continue to run while the Application Flash section isupdated, providing true Read-While-Write operation. By combining an 8-bit RISC CPUwith In-System Self-Programmable Flash on a monol i thic chip, the AtmelATmega640/1280/1281/2560/2561 is a powerful microcontroller that provides a highlyflexible and cost effective solution to many embedded control applications.

The ATmega640/1280/1281/2560/2561 AVR is supported with a full suite of programand system development tools including: C compilers, macro assemblers, programdebugger/simulators, in-circuit emulators, and evaluation kits.

6 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Comparison Between ATmega1281/2561 and ATmega640/1280/2560

Each device in the ATmega640/1280/1281/2560/2561 family differs only in memory sizeand number of pins. Table 1 summarizes the different configurations for the six devices.

Pin Descriptions

VCC Digital supply voltage.

GND Ground.

Port A (PA7..PA0) Port A is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for eachbit). The Port A output buffers have symmetrical drive characteristics with both high sinkand source capability. As inputs, Port A pins that are externally pulled low will sourcecurrent if the pull-up resistors are activated. The Port A pins are tri-stated when a resetcondition becomes active, even if the clock is not running.

Por t A a lso se rves the func t ions o f va r ious spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 88.

Port B (PB7..PB0) Port B is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for eachbit). The Port B output buffers have symmetrical drive characteristics with both high sinkand source capability. As inputs, Port B pins that are externally pulled low will sourcecurrent if the pull-up resistors are activated. The Port B pins are tri-stated when a resetcondition becomes active, even if the clock is not running.

Port B has better driving capabilities than the other ports.

Por t B a lso se rves the func t ions o f va r ious spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 89.

Port C (PC7..PC0) Port C is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for eachbit). The Port C output buffers have symmetrical drive characteristics with both high sinkand source capability. As inputs, Port C pins that are externally pulled low will sourcecurrent if the pull-up resistors are activated. The Port C pins are tri-stated when a resetcondition becomes active, even if the clock is not running.

Por t C a lso se rves the func t ions o f spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 92.

Port D (PD7..PD0) Port D is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for eachbit). The Port D output buffers have symmetrical drive characteristics with both high sinkand source capability. As inputs, Port D pins that are externally pulled low will source

Table 1. Configuration Summary

Device Flash EEPROM RAMGeneral

Purpose I/O pins16 bits resolution

PWM channelsSerial

USARTsADC

Channels

ATmega640 64KB 4KB 8KB 86 12 4 16

ATmega1280 128KB 4KB 8KB 86 12 4 16

ATmega1281 128KB 4KB 8KB 54 6 2 8

ATmega2560 256KB 4KB 8KB 86 12 4 16

ATmega2561 256KB 4KB 8KB 54 6 2 8

7

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

current if the pull-up resistors are activated. The Port D pins are tri-stated when a resetcondition becomes active, even if the clock is not running.

Por t D a lso se rves the func t ions o f va r ious spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 94.

Port E (PE7..PE0) Port E is an 8-bit bi-directional I/O port with internal pull-up resistors (selected for eachbit). The Port E output buffers have symmetrical drive characteristics with both high sinkand source capability. As inputs, Port E pins that are externally pulled low will sourcecurrent if the pull-up resistors are activated. The Port E pins are tri-stated when a resetcondition becomes active, even if the clock is not running.

Por t E a lso se rves the func t ions o f va r ious spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 96.

Port F (PF7..PF0) Port F serves as analog inputs to the A/D Converter.

Port F also serves as an 8-bit bi-directional I/O port, if the A/D Converter is not used.Port pins can provide internal pull-up resistors (selected for each bit). The Port F outputbuffers have symmetrical drive characteristics with both high sink and source capability.As inputs, Port F pins that are externally pulled low will source current if the pull-upresistors are activated. The Port F pins are tri-stated when a reset condition becomesactive, even if the clock is not running. If the JTAG interface is enabled, the pull-up resis-tors on pins PF7(TDI), PF5(TMS), and PF4(TCK) will be activated even if a resetoccurs.

Port F also serves the functions of the JTAG interface.

Port G (PG5..PG0) Port G is a 6-bit I/O port with internal pull-up resistors (selected for each bit). The Port Goutput buffers have symmetrical drive characteristics with both high sink and sourcecapability. As inputs, Port G pins that are externally pulled low will source current if thepull-up resistors are activated. The Port G pins are tri-stated when a reset conditionbecomes active, even if the clock is not running.

Por t G a lso se rves the func t ions o f va r ious spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 102.

Port H (PH7..PH0) Port H is a 8-bit bi-directional I/O port with internal pull-up resistors (selected for eachbit). The Port H output buffers have symmetrical drive characteristics with both high sinkand source capability. As inputs, Port H pins that are externally pulled low will sourcecurrent if the pull-up resistors are activated. The Port H pins are tri-stated when a resetcondition becomes active, even if the clock is not running.

Por t H a lso se rves the func t ions o f va r ious spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 104.

Port J (PJ7..PJ0) Port J is a 8-bit bi-directional I/O port with internal pull-up resistors (selected for eachbit). The Port J output buffers have symmetrical drive characteristics with both high sinkand source capability. As inputs, Port J pins that are externally pulled low will sourcecurrent if the pull-up resistors are activated. The Port J pins are tri-stated when a resetcondition becomes active, even if the clock is not running.

Por t J a l so serves the func t ions o f va r ious spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 106.

Port K (PK7..PK0) Port K serves as analog inputs to the A/D Converter.

8 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Port K is a 8-bit bi-directional I/O port with internal pull-up resistors (selected for eachbit). The Port K output buffers have symmetrical drive characteristics with both high sinkand source capability. As inputs, Port K pins that are externally pulled low will sourcecurrent if the pull-up resistors are activated. The Port K pins are tri-stated when a resetcondition becomes active, even if the clock is not running.

Por t K a lso se rves the func t ions o f va r ious spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 108.

Port L (PL7..PL0) Port L is a 8-bit bi-directional I/O port with internal pull-up resistors (selected for eachbit). The Port L output buffers have symmetrical drive characteristics with both high sinkand source capability. As inputs, Port L pins that are externally pulled low will sourcecurrent if the pull-up resistors are activated. The Port L pins are tri-stated when a resetcondition becomes active, even if the clock is not running.

Por t L a l so se rves the func t ions o f va r ious spec ia l fea tu res o f theATmega640/1280/1281/2560/2561 as listed on page 110.

RESET Reset input. A low level on this pin for longer than the minimum pulse length will gener-ate a reset, even if the clock is not running. The minimum pulse length is given in Table23 on page 58. Shorter pulses are not guaranteed to generate a reset.

XTAL1 Input to the inverting Oscillator amplifier and input to the internal clock operating circuit.

XTAL2 Output from the inverting Oscillator amplifier.

AVCC AVCC is the supply voltage pin for Port F and the A/D Converter. It should be externallyconnected to VCC, even if the ADC is not used. If the ADC is used, it should be con-nected to VCC through a low-pass filter.

AREF This is the analog reference pin for the A/D Converter.

About Code Examples

This documentation contains simple code examples that briefly show how to use variousparts of the device. Be aware that not all C compiler vendors include bit definitions in theheader files and interrupt handling in C is compiler dependent. Please confirm with theC compiler documentation for more details.

These code examples assume that the part specific header file is included before com-pilation. For I/O registers located in extended I/O map, "IN", "OUT", "SBIS", "SBIC","CBI", and "SBI" instructions must be replaced with instructions that allow access toextended I/O. Typically "LDS" and "STS" combined with "SBRS", "SBRC", "SBR", and"CBR".

9

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

AVR CPU Core

Introduction This section discusses the AVR core architecture in general. The main function of theCPU core is to ensure correct program execution. The CPU must therefore be able toaccess memories, perform calculations, control peripherals, and handle interrupts.

Architectural Overview Figure 4. Block Diagram of the AVR Architecture

In order to maximize performance and parallelism, the AVR uses a Harvard architecture– with separate memories and buses for program and data. Instructions in the programmemory are executed with a single level pipelining. While one instruction is being exe-cuted, the next instruction is pre-fetched from the program memory. This conceptenables instructions to be executed in every clock cycle. The program memory is In-System Reprogrammable Flash memory.

The fast-access Register File contains 32 x 8-bit general purpose working registers witha single clock cycle access time. This allows single-cycle Arithmetic Logic Unit (ALU)operation. In a typical ALU operation, two operands are output from the Register File,

FlashProgramMemory

InstructionRegister

InstructionDecoder

ProgramCounter

Control Lines

32 x 8GeneralPurpose

Registrers

ALU

Statusand Control

I/O Lines

EEPROM

Data Bus 8-bit

DataSRAM

Dire

ct A

ddre

ssin

g

Indi

rect

Add

ress

ing

InterruptUnit

SPIUnit

WatchdogTimer

AnalogComparator

I/O Module 2

I/O Module1

I/O Module n

10 ATmega640/1280/1281/2560/25612549A–AVR–03/05

the operation is executed, and the result is stored back in the Register File – in oneclock cycle.

Six of the 32 registers can be used as three 16-bit indirect address register pointers forData Space addressing – enabling efficient address calculations. One of the theseaddress pointers can also be used as an address pointer for look up tables in Flash pro-gram memory. These added function registers are the 16-bit X-, Y-, and Z-register,described later in this section.

The ALU supports arithmetic and logic operations between registers or between a con-stant and a register. Single register operations can also be executed in the ALU. Afteran arithmetic operation, the Status Register is updated to reflect information about theresult of the operation.

Program flow is provided by conditional and unconditional jump and call instructions,able to directly address the whole address space. Most AVR instructions have a single16-bit word format. Every program memory address contains a 16- or 32-bit instruction.

Program Flash memory space is divided in two sections, the Boot Program section andthe Application Program section. Both sections have dedicated Lock bits for write andread/write protection. The SPM instruction that writes into the Application Flash memorysection must reside in the Boot Program section.

During interrupts and subroutine calls, the return address Program Counter (PC) isstored on the Stack. The Stack is effectively allocated in the general data SRAM, andconsequently the Stack size is only limited by the total SRAM size and the usage of theSRAM. All user programs must initialize the SP in the Reset routine (before subroutinesor interrupts are executed). The Stack Pointer (SP) is read/write accessible in the I/Ospace. The data SRAM can easily be accessed through the five different addressingmodes supported in the AVR architecture.

The memory spaces in the AVR architecture are all linear and regular memory maps.

A flexible interrupt module has its control registers in the I/O space with an additionalGlobal Interrupt Enable bit in the Status Register. All interrupts have a separate InterruptVector in the Interrupt Vector table. The interrupts have priority in accordance with theirInterrupt Vector position. The lower the Interrupt Vector address, the higher the priority.

The I/O memory space contains 64 addresses for CPU peripheral functions as ControlRegisters, SPI, and other I/O functions. The I/O Memory can be accessed directly, or asthe Data Space locations following those of the Register File, 0x20 - 0x5F. In addition,the ATmega640/1280/1281/2560/2561 has Extended I/O space from 0x60 - 0x1FF inSRAM where only the ST/STS/STD and LD/LDS/LDD instructions can be used.

ALU – Arithmetic Logic Unit

The high-performance AVR ALU operates in direct connection with all the 32 generalpurpose working registers. Within a single clock cycle, arithmetic operations betweengeneral purpose registers or between a register and an immediate are executed. TheALU operations are divided into three main categories – arithmetic, logical, and bit-func-tions. Some implementations of the architecture also provide a powerful multipliersupporting both signed/unsigned multiplication and fractional format. See the “Instruc-tion Set” section for a detailed description.

Status Register The Status Register contains information about the result of the most recently executedarithmetic instruction. This information can be used for altering program flow in order toperform conditional operations. Note that the Status Register is updated after all ALUoperations, as specified in the Instruction Set Reference. This will in many cases

11

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

remove the need for using the dedicated compare instructions, resulting in faster andmore compact code.

The Status Register is not automatically stored when entering an interrupt routine andrestored when returning from an interrupt. This must be handled by software.

The AVR Status Register – SREG – is defined as:

• Bit 7 – I: Global Interrupt Enable

The Global Interrupt Enable bit must be set for the interrupts to be enabled. The individ-ual interrupt enable control is then performed in separate control registers. If the GlobalInterrupt Enable Register is cleared, none of the interrupts are enabled independent ofthe individual interrupt enable settings. The I-bit is cleared by hardware after an interrupthas occurred, and is set by the RETI instruction to enable subsequent interrupts. The I-bit can also be set and cleared by the application with the SEI and CLI instructions, asdescribed in the instruction set reference.

• Bit 6 – T: Bit Copy Storage

The Bit Copy instructions BLD (Bit LoaD) and BST (Bit STore) use the T-bit as source ordestination for the operated bit. A bit from a register in the Register File can be copiedinto T by the BST instruction, and a bit in T can be copied into a bit in a register in theRegister File by the BLD instruction.

• Bit 5 – H: Half Carry Flag

The Half Carry Flag H indicates a Half Carry in some arithmetic operations. Half Carry Isuseful in BCD arithmetic. See the “Instruction Set Description” for detailed information.

• Bit 4 – S: Sign Bit, S = N ⊕ V

The S-bit is always an exclusive or between the Negative Flag N and the Two’s Comple-ment Overflow Flag V. See the “Instruction Set Description” for detailed information.

• Bit 3 – V: Two’s Complement Overflow Flag

The Two’s Complement Overflow Flag V supports two’s complement arithmetics. Seethe “Instruction Set Description” for detailed information.

• Bit 2 – N: Negative Flag

The Negative Flag N indicates a negative result in an arithmetic or logic operation. Seethe “Instruction Set Description” for detailed information.

• Bit 1 – Z: Zero Flag

The Zero Flag Z indicates a zero result in an arithmetic or logic operation. See the“Instruction Set Description” for detailed information.

• Bit 0 – C: Carry Flag

The Carry Flag C indicates a carry in an arithmetic or logic operation. See the “Instruc-tion Set Description” for detailed information.

Bit 7 6 5 4 3 2 1 0

I T H S V N Z C SREG

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

12 ATmega640/1280/1281/2560/25612549A–AVR–03/05

General Purpose Register File

The Register File is optimized for the AVR Enhanced RISC instruction set. In order toachieve the required performance and flexibility, the following input/output schemes aresupported by the Register File:

• One 8-bit output operand and one 8-bit result input

• Two 8-bit output operands and one 8-bit result input

• Two 8-bit output operands and one 16-bit result input

• One 16-bit output operand and one 16-bit result input

Figure 5 shows the structure of the 32 general purpose working registers in the CPU.

Figure 5. AVR CPU General Purpose Working Registers

Most of the instructions operating on the Register File have direct access to all registers,and most of them are single cycle instructions.

As shown in Figure 5, each register is also assigned a data memory address, mappingthem directly into the first 32 locations of the user Data Space. Although not being phys-ically implemented as SRAM locations, this memory organization provides greatflexibility in access of the registers, as the X-, Y- and Z-pointer registers can be set toindex any register in the file.

7 0 Addr.

R0 0x00

R1 0x01

R2 0x02

…

R13 0x0D

General R14 0x0E

Purpose R15 0x0F

Working R16 0x10

Registers R17 0x11

…

R26 0x1A X-register Low Byte

R27 0x1B X-register High Byte

R28 0x1C Y-register Low Byte

R29 0x1D Y-register High Byte

R30 0x1E Z-register Low Byte

R31 0x1F Z-register High Byte

13

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The X-register, Y-register, and Z-register

The registers R26..R31 have some added functions to their general purpose usage.These registers are 16-bit address pointers for indirect addressing of the data space.The three indirect address registers X, Y, and Z are defined as described in Figure 6.

Figure 6. The X-, Y-, and Z-registers

In the different addressing modes these address registers have functions as fixed dis-placement, automatic increment, and automatic decrement (see the instruction setreference for details).

Stack Pointer The Stack is mainly used for storing temporary data, for storing local variables and forstoring return addresses after interrupts and subroutine calls. The Stack Pointer Regis-ter always points to the top of the Stack. Note that the Stack is implemented as growingfrom higher memory locations to lower memory locations. This implies that a StackPUSH command decreases the Stack Pointer.

The Stack Pointer points to the data SRAM Stack area where the Subroutine and Inter-rupt Stacks are located. This Stack space in the data SRAM must be defined by theprogram before any subroutine calls are executed or interrupts are enabled. The StackPointer must be set to point above 0x0200. The initial value of the stack pointer is thelast address of the internal SRAM. The Stack Pointer is decremented by one when datais pushed onto the Stack with the PUSH instruction, and it is decremented by threewhen the return address is pushed onto the Stack with subroutine call or interrupt. TheStack Pointer is incremented by one when data is popped from the Stack with the POPinstruction, and it is incremented by three when data is popped from the Stack withreturn from subroutine RET or return from interrupt RETI.

The AVR Stack Pointer is implemented as two 8-bit registers in the I/O space. The num-ber of bits actually used is implementation dependent. Note that the data space in someimplementations of the AVR architecture is so small that only SPL is needed. In thiscase, the SPH Register will not be present.

15 XH XL 0

X-register 7 0 7 0

R27 (0x1B) R26 (0x1A)

15 YH YL 0

Y-register 7 0 7 0

R29 (0x1D) R28 (0x1C)

15 ZH ZL 0

Z-register 7 0 7 0

R31 (0x1F) R30 (0x1E)

Bit 15 14 13 12 11 10 9 8

SP15 SP14 SP13 SP12 SP11 SP10 SP9 SP8 SPH

SP7 SP6 SP5 SP4 SP3 SP2 SP1 SP0 SPL

7 6 5 4 3 2 1 0

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 1 0 0 0 0 1

1 1 1 1 1 1 1 1

14 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Extended Z-pointer Register for ELPM/SPM - RAMPZ

For ELPM/SPM instructions, the Z-pointer is a concatenation of RAMPZ, ZH, and ZL, asshown in Figure 7. Note that LPM is not affected by the RAMPZ setting.

Figure 7. The Z-pointer used by ELPM and SPM

The actual number of bits is implementation dependent. Unused bits in an implementa-tion will always read as zero. For compatibility with future devices, be sure to write thesebits to zero.

Extended Indirect Register - EIND

For EICALL/EIJMP instructions, the Indirect-pointer to the subroutine/routine is a con-catenation of EIND, ZH, and ZL, as shown in Figure 8. Note that ICALL and IJMP arenot affected by the EIND setting.

Figure 8. The Indirect-pointer used by EICALL and EIJMP

The actual number of bits is implementation dependent. Unused bits in an implementa-tion will always read as zero. For compatibility with future devices, be sure to write thesebits to zero.

Instruction Execution Timing

This section describes the general access timing concepts for instruction execution. TheAVR CPU is driven by the CPU clock clkCPU, directly generated from the selected clocksource for the chip. No internal clock division is used.

Figure 9 shows the parallel instruction fetches and instruction executions enabled by theHarvard architecture and the fast-access Register File concept. This is the basic pipelin-ing concept to obtain up to 1 MIPS per MHz with the corresponding unique results forfunctions per cost, functions per clocks, and functions per power-unit.

Bit 7 6 5 4 3 2 1 0

RAMPZ7 RAMPZ6 RAMPZ5 RAMPZ4 RAMPZ3 RAMPZ2 RAMPZ1 RAMPZ0 RAMPZ

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit (Individually)

7 0 7 0 7 0

RAMPZ ZH ZL

Bit (Z-pointer) 23 16 15 8 7 0

Bit 7 6 5 4 3 2 1 0

EIND7 EIND6 EIND5 EIND4 EIND3 EIND2 EIND1 EIND0 EIND

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit (Individual-ly)

7 0 7 0 7 0

EIND ZH ZL

Bit (Indirect-pointer)

23 16 15 8 7 0

15

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 9. The Parallel Instruction Fetches and Instruction Executions

Figure 10 shows the internal timing concept for the Register File. In a single clock cyclean ALU operation using two register operands is executed, and the result is stored backto the destination register.

Figure 10. Single Cycle ALU Operation

Reset and Interrupt Handling

The AVR provides several different interrupt sources. These interrupts and the separateReset Vector each have a separate program vector in the program memory space. Allinterrupts are assigned individual enable bits which must be written logic one togetherwith the Global Interrupt Enable bit in the Status Register in order to enable the interrupt.Depending on the Program Counter value, interrupts may be automatically disabledwhen Boot Lock bits BLB02 or BLB12 are programmed. This feature improves softwaresecurity. See the section “Memory Programming” on page 335 for details.

The lowest addresses in the program memory space are by default defined as the Resetand Interrupt Vectors. The complete list of vectors is shown in “Interrupts” on page 69.The list also determines the priority levels of the different interrupts. The lower theaddress the higher is the priority level. RESET has the highest priority, and next is INT0– the External Interrupt Request 0. The Interrupt Vectors can be moved to the start ofthe Boot Flash section by setting the IVSEL bit in the MCU Control Register (MCUCR).Refer to “Interrupts” on page 69 for more information. The Reset Vector can also bemoved to the start of the Boot Flash section by programming the BOOTRST Fuse, see“Memory Programming” on page 335.

When an interrupt occurs, the Global Interrupt Enable I-bit is cleared and all interruptsare disabled. The user software can write logic one to the I-bit to enable nested inter-rupts. All enabled interrupts can then interrupt the current interrupt routine. The I-bit isautomatically set when a Return from Interrupt instruction – RETI – is executed.

clk

1st Instruction Fetch

1st Instruction Execute2nd Instruction Fetch

2nd Instruction Execute3rd Instruction Fetch

3rd Instruction Execute4th Instruction Fetch

T1 T2 T3 T4

CPU

Total Execution Time

Register Operands Fetch

ALU Operation Execute

Result Write Back

T1 T2 T3 T4

clkCPU

16 ATmega640/1280/1281/2560/25612549A–AVR–03/05

There are basically two types of interrupts. The first type is triggered by an event thatsets the Interrupt Flag. For these interrupts, the Program Counter is vectored to theactual Interrupt Vector in order to execute the interrupt handling routine, and hardwareclears the corresponding Interrupt Flag. Interrupt Flags can also be cleared by writing alogic one to the flag bit position(s) to be cleared. If an interrupt condition occurs while thecorresponding interrupt enable bit is cleared, the Interrupt Flag will be set and remem-bered until the interrupt is enabled, or the flag is cleared by software. Similarly, if one ormore interrupt conditions occur while the Global Interrupt Enable bit is cleared, the cor-responding Interrupt Flag(s) will be set and remembered until the Global InterruptEnable bit is set, and will then be executed by order of priority.

The second type of interrupts will trigger as long as the interrupt condition is present.These interrupts do not necessarily have Interrupt Flags. If the interrupt condition disap-pears before the interrupt is enabled, the interrupt will not be triggered.

When the AVR exits from an interrupt, it will always return to the main program and exe-cute one more instruction before any pending interrupt is served.

Note that the Status Register is not automatically stored when entering an interrupt rou-tine, nor restored when returning from an interrupt routine. This must be handled bysoftware.

When using the CLI instruction to disable interrupts, the interrupts will be immediatelydisabled. No interrupt will be executed after the CLI instruction, even if it occurs simulta-neously with the CLI instruction. The following example shows how this can be used toavoid interrupts during the timed EEPROM write sequence..

When using the SEI instruction to enable interrupts, the instruction following SEI will beexecuted before any pending interrupts, as shown in this example.

Assembly Code Example

in r16, SREG ; store SREG value

cli ; disable interrupts during timed sequence

sbi EECR, EEMPE ; start EEPROM write

sbi EECR, EEPE

out SREG, r16 ; restore SREG value (I-bit)

C Code Example

char cSREG;

cSREG = SREG; /* store SREG value */

/* disable interrupts during timed sequence */

__disable_interrupt();

EECR |= (1<<EEMPE); /* start EEPROM write */

EECR |= (1<<EEPE);

SREG = cSREG; /* restore SREG value (I-bit) */

17

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Interrupt Response Time The interrupt execution response for all the enabled AVR interrupts is five clock cyclesminimum. After five clock cycles the program vector address for the actual interrupt han-dling routine is executed. During these five clock cycle period, the Program Counter ispushed onto the Stack. The vector is normally a jump to the interrupt routine, and thisjump takes three clock cycles. If an interrupt occurs during execution of a multi-cycleinstruction, this instruction is completed before the interrupt is served. If an interruptoccurs when the MCU is in sleep mode, the interrupt execution response time isincreased by five clock cycles. This increase comes in addition to the start-up time fromthe selected sleep mode.

A return from an interrupt handling routine takes five clock cycles. During these fiveclock cycles, the Program Counter (three bytes) is popped back from the Stack, theStack Pointer is incremented by three, and the I-bit in SREG is set.

Assembly Code Example

sei ; set Global Interrupt Enable

sleep ; enter sleep, waiting for interrupt

; note: will enter sleep before any pending

; interrupt(s)

C Code Example

__enable_interrupt(); /* set Global Interrupt Enable */

__sleep(); /* enter sleep, waiting for interrupt */

/* note: will enter sleep before any pending interrupt(s) */

18 ATmega640/1280/1281/2560/25612549A–AVR–03/05

19

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

AVR ATmega640/1280/1281/2560/2561 MemoriesThis section describes the different memories in the ATmega640/1280/1281/2560/2561.The AVR architecture has two main memory spaces, the Data Memory and the ProgramMemory space. In addition, the ATmega640/1280/1281/2560/2561 features anEEPROM Memory for data storage. All three memory spaces are linear and regular.

In-System Reprogrammable Flash Program Memory

The ATmega640/1280/1281/2560/2561 contains 64K/128K/256K bytes On-chip In-Sys-tem Reprogrammable Flash memory for program storage. Since all AVR instructionsare 16 or 32 bits wide, the Flash is organized as 32K/64K/128K x 16. For software secu-rity, the Flash Program memory space is divided into two sections, Boot Programsection and Application Program section.

The Flash memory has an endurance of at least 10,000 write/erase cycles. TheATmega640/1280/1281/2560/2561 Program Counter (PC) is 15/16/17 bits wide, thusaddressing the 32K/64K/128K program memory locations. The operation of Boot Pro-gram section and associated Boot Lock bits for software protection are described indetail in “Boot Loader Support – Read-While-Write Self-Programming” on page 317.“Memory Programming” on page 335 contains a detailed description on Flash dataserial downloading using the SPI pins or the JTAG interface.

Constant tables can be allocated within the entire program memory address space (seethe LPM – Load Program Memory instruction description and ELPM - Extended LoadProgram Memory instruction description).

Timing diagrams for instruction fetch and execution are presented in “Instruction Execu-tion Timing” on page 14.

Figure 11. Program Memory Map

Program Memory

0x0000

0x7FFF/0xFFFF/0x1FFFF

Application Flash Section

Boot Flash Section

20 ATmega640/1280/1281/2560/25612549A–AVR–03/05

SRAM Data Memory Figure 12 shows how the ATmega640/1280/1281/2560/2561 SRAM Memory isorganized.

The ATmega640/1280/1281/2560/2561 is a complex microcontroller with more periph-eral units than can be supported within the 64 location reserved in the Opcode for the INand OUT instructions. For the Extended I/O space from $060 - $1FF in SRAM, only theST/STS/STD and LD/LDS/LDD instructions can be used.

The first 4,608/8,704 Data Memory locations address both the Register File, the I/OMemory, Extended I/O Memory, and the internal data SRAM. The first 32 locationsaddress the Register file, the next 64 location the standard I/O Memory, then 416 loca-tions of Extended I/O memory and the next 8,192 locations address the internal dataSRAM.

An op t iona l ex te rna l da ta SRAM can be used w i th theATmega640/1280/1281/2560/2561. This SRAM will occupy an area in the remainingaddress locations in the 64K address space. This area starts at the address followingthe internal SRAM. The Register file, I/O, Extended I/O and Internal SRAM occupies thelowest 4,608/8,704 bytes, so when using 64KB (65,536 bytes) of External Memory,60,478/56,832 Bytes of External Memory are available. See “External Memory Inter-face” on page 29 for details on how to take advantage of the external memory map.

When the addresses accessing the SRAM memory space exceeds the internal datamemory locations, the external data SRAM is accessed using the same instructions asfor the internal data memory access. When the internal data memories are accessed,the read and write strobe pins (PG0 and PG1) are inactive during the whole accesscycle. External SRAM operation is enabled by setting the SRE bit in the XMCRARegister.

Accessing external SRAM takes one additional clock cycle per byte compared to accessof the internal SRAM. This means that the commands LD, ST, LDS, STS, LDD, STD,PUSH, and POP take one additional clock cycle. If the Stack is placed in externalSRAM, interrupts, subroutine calls and returns take three clock cycles extra because thethree-byte program counter is pushed and popped, and external memory access doesnot take advantage of the internal pipe-line memory access. When external SRAM inter-face is used with wait-state, one-byte external access takes two, three, or four additionalclock cycles for one, two, and three wait-states respectively. Interrupts, subroutine callsand returns will need five, seven, or nine clock cycles more than specified in the instruc-tion set manual for one, two, and three wait-states.

The five different addressing modes for the data memory cover: Direct, Indirect with Dis-placement, Indirect, Indirect with Pre-decrement, and Indirect with Post-increment. Inthe Register file, registers R26 to R31 feature the indirect addressing pointer registers.

The direct addressing reaches the entire data space.

The Indirect with Displacement mode reaches 63 address locations from the baseaddress given by the Y- or Z-register.

When using register indirect addressing modes with automatic pre-decrement and post-increment, the address registers X, Y, and Z are decremented or incremented.

The 32 general purpose working registers, 64 I/O registers, and the 4,196/8,192 bytes ofinternal data SRAM in the ATmega640/1280/1281/2560/2561 are all accessible throughall these addressing modes. The Register File is described in “General Purpose Regis-ter File” on page 12.

21

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 12. Data Memory Map

32 Registers64 I/O Registers

Internal SRAM(8192 x 8)

$0000 - $001F$0020 - $005F

$2200$21FF

$FFFF

$0060 - $01FF

Data Memory

External SRAM(0 - 64K x 8)

416 Ext I/O Reg.$0200

22 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Data Memory Access Times This section describes the general access timing concepts for internal memory access.The internal data SRAM access is performed in two clkCPU cycles as described in Figure13.

Figure 13. On-chip Data SRAM Access Cycles

EEPROM Data Memory The ATmega640/1280/1281/2560/2561 contains 4K bytes of data EEPROM memory. Itis organized as a separate data space, in which single bytes can be read and written.The EEPROM has an endurance of at least 100,000 write/erase cycles. The accessbetween the EEPROM and the CPU is described in the following, specifying theEEPROM Address Registers, the EEPROM Data Register, and the EEPROM ControlRegister.

For a detailed description of SPI, JTAG and Parallel data downloading to the EEPROM,see page 349, page 353, and page 338 respectively.

EEPROM Read/Write Access The EEPROM Access Registers are accessible in the I/O space.

The write access time for the EEPROM is given in Table 3. A self-timing function, how-ever, lets the user software detect when the next byte can be written. If the user codecontains instructions that write the EEPROM, some precautions must be taken. Inheavily filtered power supplies, VCC is likely to rise or fall slowly on power-up/down. Thiscauses the device for some period of time to run at a voltage lower than specified asminimum for the clock frequency used. See “Preventing EEPROM Corruption” on page27. for details on how to avoid problems in these situations.

In order to prevent unintentional EEPROM writes, a specific write procedure must be fol-lowed. Refer to the description of the EEPROM Control Register for details on this.

When the EEPROM is read, the CPU is halted for four clock cycles before the nextinstruction is executed. When the EEPROM is written, the CPU is halted for two clockcycles before the next instruction is executed.

clk

WR

RD

Data

Data

Address Address valid

T1 T2 T3

Compute Address

Rea

dW

rite

CPU

Memory Access Instruction Next Instruction

23

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The EEPROM Address Register – EEARH and EEARL

• Bits 15..12 – Res: Reserved Bits

These bits are reserved bits and will always read as zero.

• Bits 11..0 – EEAR8..0: EEPROM Address

The EEPROM Address Registers – EEARH and EEARL specify the EEPROM addressin the 4K bytes EEPROM space. The EEPROM data bytes are addressed linearlybetween 0 and 4096. The initial value of EEAR is undefined. A proper value must bewritten before the EEPROM may be accessed.

The EEPROM Data Register – EEDR

• Bits 7..0 – EEDR7.0: EEPROM Data

For the EEPROM write operation, the EEDR Register contains the data to be written tothe EEPROM in the address given by the EEAR Register. For the EEPROM read oper-ation, the EEDR contains the data read out from the EEPROM at the address given byEEAR.

The EEPROM Control Register – EECR

• Bits 7..6 – Res: Reserved Bits

These bits are reserved bits and will always read as zero.

• Bits 5, 4 – EEPM1 and EEPM0: EEPROM Programming Mode Bits

The EEPROM Programming mode bit setting defines which programming action that willbe triggered when writing EEPE. It is possible to program data in one atomic operation(erase the old value and program the new value) or to split the Erase and Write opera-tions in two different operations. The Programming times for the different modes areshown in Table 2. While EEPE is set, any write to EEPMn will be ignored. During reset,the EEPMn bits will be reset to 0b00 unless the EEPROM is busy programming.

Bit 15 14 13 12 11 10 9 8

– – – – EEAR11 EEAR10 EEAR9 EEAR8 EEARH

EEAR7 EEAR6 EEAR5 EEAR4 EEAR3 EEAR2 EEAR1 EEAR0 EEARL

7 6 5 4 3 2 1 0

Read/Write R R R R R/W R/W R/W R/W

R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 X X X X

X X X X X X X X

Bit 7 6 5 4 3 2 1 0

MSB LSB EEDR

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – EEPM1 EEPM0 EERIE EEMPE EEPE EERE EECR

Read/Write R R R/W R/W R/W R/W R/W R/W

Initial Value 0 0 X X 0 0 X 0

24 ATmega640/1280/1281/2560/25612549A–AVR–03/05

• Bit 3 – EERIE: EEPROM Ready Interrupt Enable

Writing EERIE to one enables the EEPROM Ready Interrupt if the I bit in SREG is set.Writing EERIE to zero disables the interrupt. The EEPROM Ready interrupt generates aconstant interrupt when EEPE is cleared.

• Bit 2 – EEMPE: EEPROM Master Programming Enable

The EEMPE bit determines whether setting EEPE to one causes the EEPROM to bewritten. When EEMPE is set, setting EEPE within four clock cycles will write data to theEEPROM at the selected address If EEMPE is zero, setting EEPE will have no effect.When EEMPE has been written to one by software, hardware clears the bit to zero afterfour clock cycles. See the description of the EEPE bit for an EEPROM write procedure.

• Bit 1 – EEPE: EEPROM Programming Enable

The EEPROM Write Enable Signal EEPE is the write strobe to the EEPROM. Whenaddress and data are correctly set up, the EEPE bit must be written to one to write thevalue into the EEPROM. The EEMPE bit must be written to one before a logical one iswritten to EEPE, otherwise no EEPROM write takes place. The following procedureshould be followed when writing the EEPROM (the order of steps 3 and 4 is notessential):

1. Wait until EEPE becomes zero.

2. Wait until SELFPRGEN in SPMCSR becomes zero.

3. Write new EEPROM address to EEAR (optional).

4. Write new EEPROM data to EEDR (optional).

5. Write a logical one to the EEMPE bit while writing a zero to EEPE in EECR.

6. Within four clock cycles after setting EEMPE, write a logical one to EEPE.

The EEPROM can not be programmed during a CPU write to the Flash memory. Thesoftware must check that the Flash programming is completed before initiating a newEEPROM write. Step 2 is only relevant if the software contains a Boot Loader allowingthe CPU to program the Flash. If the Flash is never being updated by the CPU, step 2can be omitted. See “Memory Programming” on page 335 for details about Bootprogramming.

Caution: An interrupt between step 5 and step 6 will make the write cycle fail, since theEEPROM Master Write Enable will time-out. If an interrupt routine accessing theEEPROM is interrupting another EEPROM access, the EEAR or EEDR Register will bemodified, causing the interrupted EEPROM access to fail. It is recommended to havethe Global Interrupt Flag cleared during all the steps to avoid these problems.

When the write access time has elapsed, the EEPE bit is cleared by hardware. The usersoftware can poll this bit and wait for a zero before writing the next byte. When EEPEhas been set, the CPU is halted for two cycles before the next instruction is executed.

Table 2. EEPROM Mode Bits

EEPM1 EEPM0Programming

Time Operation

0 0 3.4 ms Erase and Write in one operation (Atomic Operation)

0 1 1.8 ms Erase Only

1 0 1.8 ms Write Only

1 1 – Reserved for future use

25

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• Bit 0 – EERE: EEPROM Read Enable

The EEPROM Read Enable Signal EERE is the read strobe to the EEPROM. When thecorrect address is set up in the EEAR Register, the EERE bit must be written to a logicone to trigger the EEPROM read. The EEPROM read access takes one instruction, andthe requested data is available immediately. When the EEPROM is read, the CPU ishalted for four cycles before the next instruction is executed.

The user should poll the EEPE bit before starting the read operation. If a write operationis in progress, it is neither possible to read the EEPROM, nor to change the EEARRegister.

The calibrated Oscillator is used to time the EEPROM accesses. Table 3 lists the typicalprogramming time for EEPROM access from the CPU.

Table 3. EEPROM Programming Time

Symbol Number of Calibrated RC Oscillator Cycles Typ Programming Time

EEPROM write (from CPU)

26,368 3.3 ms

26 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The following code examples show one assembly and one C function for writing to theEEPROM. The examples assume that interrupts are controlled (e.g. by disabling inter-rupts globally) so that no interrupts will occur during execution of these functions. Theexamples also assume that no Flash Boot Loader is present in the software. If suchcode is present, the EEPROM write function must also wait for any ongoing SPM com-mand to finish.

Note: 1. See “About Code Examples” on page 8.

Assembly Code Example()

EEPROM_write:

; Wait for completion of previous write

sbic EECR,EEPE

rjmp EEPROM_write

; Set up address (r18:r17) in address register

out EEARH, r18

out EEARL, r17

; Write data (r16) to Data Register

out EEDR,r16

; Write logical one to EEMPE

sbi EECR,EEMPE

; Start eeprom write by setting EEPE

sbi EECR,EEPE

ret

C Code Example(1)

void EEPROM_write(unsigned int uiAddress, unsigned char ucData)

/* Wait for completion of previous write */

while(EECR & (1<<EEPE))

;

/* Set up address and Data Registers */

EEAR = uiAddress;

EEDR = ucData;

/* Write logical one to EEMPE */

EECR |= (1<<EEMPE);

/* Start eeprom write by setting EEPE */

EECR |= (1<<EEPE);

27

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The next code examples show assembly and C functions for reading the EEPROM. Theexamples assume that interrupts are controlled so that no interrupts will occur duringexecution of these functions.

Note: 1. See “About Code Examples” on page 8.

Preventing EEPROM Corruption

During periods of low VCC, the EEPROM data can be corrupted because the supply volt-age is too low for the CPU and the EEPROM to operate properly. These issues are thesame as for board level systems using EEPROM, and the same design solutions shouldbe applied.

An EEPROM data corruption can be caused by two situations when the voltage is toolow. First, a regular write sequence to the EEPROM requires a minimum voltage tooperate correctly. Secondly, the CPU itself can execute instructions incorrectly, if thesupply voltage is too low.

EEPROM data corrupt ion can easi ly be avoided by fo l lowing th is designrecommendation:

Keep the AVR RESET active (low) during periods of insufficient power supply voltage.This can be done by enabling the internal Brown-out Detector (BOD). If the detectionlevel of the internal BOD does not match the needed detection level, an external lowVCC reset Protection circuit can be used. If a reset occurs while a write operation is inprogress, the write operation will be completed provided that the power supply voltage issufficient.

Assembly Code Example(1)

EEPROM_read:

; Wait for completion of previous write

sbic EECR,EEPE

rjmp EEPROM_read

; Set up address (r18:r17) in address register

out EEARH, r18

out EEARL, r17

; Start eeprom read by writing EERE

sbi EECR,EERE

; Read data from Data Register

in r16,EEDR

ret

C Code Example(1)

unsigned char EEPROM_read(unsigned int uiAddress)

/* Wait for completion of previous write */

while(EECR & (1<<EEPE))

;

/* Set up address register */

EEAR = uiAddress;

/* Start eeprom read by writing EERE */

EECR |= (1<<EERE);

/* Return data from Data Register */

return EEDR;

28 ATmega640/1280/1281/2560/25612549A–AVR–03/05

I/O Memory The I/O space definition of the ATmega640/1280/1281/2560/2561 is shown in “RegisterSummary” on page 385.

All ATmega640/1280/1281/2560/2561 I/Os and peripherals are placed in the I/O space.All I/O locations may be accessed by the LD/LDS/LDD and ST/STS/STD instructions,transferring data between the 32 general purpose working registers and the I/O space.I/O Registers within the address range 0x00 - 0x1F are directly bit-accessible using theSBI and CBI instructions. In these registers, the value of single bits can be checked byusing the SBIS and SBIC instructions. Refer to the instruction set section for moredetails. When using the I/O specific commands IN and OUT, the I/O addresses 0x00 -0x3F must be used. When addressing I/O Registers as data space using LD and STins t ruc t ions , 0x20 mus t be added to these addresses . TheATmega640/1280/1281/2560/2561 is a complex microcontroller with more peripheralunits than can be supported within the 64 location reserved in Opcode for the IN andOUT instructions. For the Extended I/O space from 0x60 - 0x1FF in SRAM, only theST/STS/STD and LD/LDS/LDD instructions can be used.

For compatibility with future devices, reserved bits should be written to zero if accessed.Reserved I/O memory addresses should never be written.

Some of the Status Flags are cleared by writing a logical one to them. Note that, unlikemost other AVRs, the CBI and SBI instructions will only operate on the specified bit, andcan therefore be used on registers containing such Status Flags. The CBI and SBIinstructions work with registers 0x00 to 0x1F only.

The I/O and peripherals control registers are explained in later sections.

General Purpose I/O Registers The ATmega640/1280/1281/2560/2561 contains three General Purpose I/O Registers.These registers can be used for storing any information, and they are particularly usefulfor storing global variables and Status Flags. General Purpose I/O Registers within theaddress range 0x00 - 0x1F are directly bit-accessible using the SBI, CBI, SBIS, andSBIC instructions.

General Purpose I/O Register 2 – GPIOR2

General Purpose I/O Register 1 – GPIOR1

General Purpose I/O Register 0 – GPIOR0

Bit 7 6 5 4 3 2 1 0

MSB LSB GPIOR2

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

MSB LSB GPIOR1

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

MSB LSB GPIOR0

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

29

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

External Memory Interface

With all the features the External Memory Interface provides, it is well suited to operateas an interface to memory devices such as External SRAM and Flash, and peripheralssuch as LCD-display, A/D, and D/A. The main features are:• Four different wait-state settings (including no wait-state).• Independent wait-state setting for different extErnal Memory sectors (configurable sector

size).• The number of bits dedicated to address high byte is selectable.• Bus keepers on data lines to minimize current consumption (optional).

Overview When the eXternal MEMory (XMEM) is enabled, address space outside the internalSRAM becomes available using the dedicated External Memory pins (see Figure 2 onpage 3, Table 36 on page 88, Table 42 on page 92, and Table 54 on page 102). Thememory configuration is shown in Figure 14.

Figure 14. External Memory with Sector Select

Using the External Memory Interface

The interface consists of:

• AD7:0: Multiplexed low-order address bus and data bus.

• A15:8: High-order address bus (configurable number of bits).

• ALE: Address latch enable.

• RD: Read strobe.

• WR: Write strobe.

The control bits for the External Memory Interface are located in two registers, the Exter-nal Memory Control Register A – XMCRA, and the External Memory Control Register B– XMCRB.

When the XMEM interface is enabled, the XMEM interface will override the setting in thedata direction registers that corresponds to the ports dedicated to the XMEM interface.

Memory Configuration A

0x0000

0x21FF

External Memory(0-60K x 8)

0xFFFF

Internal memory

SRL[2..0]

SRW11SRW10

SRW01SRW00

Lower sector

Upper sector

0x2200

30 ATmega640/1280/1281/2560/25612549A–AVR–03/05

For details about the port override, see the alternate functions in section “I/O-Ports” onpage 81. The XMEM interface will auto-detect whether an access is internal or external.If the access is external, the XMEM interface will output address, data, and the controlsignals on the ports according to Figure 16 (this figure shows the wave forms withoutwait-states). When ALE goes from high-to-low, there is a valid address on AD7:0. ALE islow during a data transfer. When the XMEM interface is enabled, also an internal accesswill cause activity on address, data and ALE ports, but the RD and WR strobes will nottoggle during internal access. When the External Memory Interface is disabled, the nor-mal pin and data direction settings are used. Note that when the XMEM interface isdisabled, the address space above the internal SRAM boundary is not mapped into theinternal SRAM. Figure 15 illustrates how to connect an external SRAM to the AVR usingan octal latch (typically “74 x 573” or equivalent) which is transparent when G is high.

Address Latch Requirements Due to the high-speed operation of the XRAM interface, the address latch must beselected with care for system frequencies above 8 MHz @ 4V and 4 MHz @ 2.7V.When operating at conditions above these frequencies, the typical old style 74HC serieslatch becomes inadequate. The External Memory Interface is designed in compliance tothe 74AHC series latch. However, most latches can be used as long they comply withthe main timing parameters. The main parameters for the address latch are:

• D to Q propagation delay (tPD).

• Data setup time before G low (tSU).

• Data (address) hold time after G low (TH).

The External Memory Interface is designed to guaranty minimum address hold timeafter G is asserted low of th = 5 ns. Refer to tLAXX_LD/tLLAXX_ST in “External Data MemoryTiming” Tables 169 through Tables 176 on pages 376 - 378. The D-to-Q propagationdelay (tPD) must be taken into consideration when calculating the access time require-ment of the external component. The data setup time before G low (tSU) must notexceed address valid to ALE low (tAVLLC) minus PCB wiring delay (dependent on thecapacitive load).

Figure 15. External SRAM Connected to the AVR

Pull-up and Bus-keeper The pull-ups on the AD7:0 ports may be activated if the corresponding Port register iswritten to one. To reduce power consumption in sleep mode, it is recommended to dis-able the pull-ups by writing the Port register to zero before entering sleep.

The XMEM interface also provides a bus-keeper on the AD7:0 lines. The bus-keepercan be disabled and enabled in software as described in “External Memory Control Reg-ister B – XMCRB” on page 35. When enabled, the bus-keeper will keep the previousvalue on the AD7:0 bus while these lines are tri-stated by the XMEM interface.

D[7:0]

A[7:0]

A[15:8]

RD

WR

SRAM

D Q

G

AD7:0

ALE

A15:8

RD

WR

AVR

31

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Timing External Memory devices have different timing requirements. To meet these require-ments, the XMEM interface provides four different wait-states as shown in Table 5. It isimportant to consider the timing specification of the External Memory device beforeselecting the wait-state. The most important parameters are the access time for theexternal memory compared to the set-up requirement. The access time for the ExternalMemory is defined to be the time from receiving the chip select/address until the data ofthis address actually is driven on the bus. The access time cannot exceed the time fromthe ALE pulse must be asserted low until data is stable during a read sequence (SeetLLRL+ tRLRH - tDVRH in Tables 169 through Tables 176 on pages 376 - 378). The differentwait-states are set up in software. As an additional feature, it is possible to divide theexternal memory space in two sectors with individual wait-state settings. This makes itpossible to connect two different memory devices with different timing requirements tothe same XMEM interface. For XMEM interface timing details, please refer to Table 169to Table 176 and Figure 161 to Figure 164 in the “External Data Memory Timing” onpage 376.

Note that the XMEM interface is asynchronous and that the waveforms in the followingfigures are related to the internal system clock. The skew between the internal andexternal clock (XTAL1) is not guarantied (varies between devices temperature, and sup-ply voltage). Consequently, the XMEM interface is not suited for synchronous operation.

Figure 16. External Data Memory Cycles without Wait-state (SRWn1=0 and SRWn0=0)

Note: 1. SRWn1 = SRW11 (upper sector) or SRW01 (lower sector), SRWn0 = SRW10 (uppersector) or SRW00 (lower sector). The ALE pulse in period T4 is only present if thenext instruction accesses the RAM (internal or external).

ALE

T1 T2 T3

Writ

eR

ead

WR

T4

A15:8 AddressPrev. addr.

DA7:0 Address DataPrev. data XX

RD

DA7:0 (XMBK = 0) DataPrev. data Address

DataPrev. data AddressDA7:0 (XMBK = 1)

System Clock (CLKCPU)

XXXXX XXXXXXXX

32 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 17. External Data Memory Cycles with SRWn1 = 0 and SRWn0 = 1(1)

Note: 1. SRWn1 = SRW11 (upper sector) or SRW01 (lower sector), SRWn0 = SRW10 (uppersector) or SRW00 (lower sector).The ALE pulse in period T5 is only present if the next instruction accesses the RAM(internal or external).

Figure 18. External Data Memory Cycles with SRWn1 = 1 and SRWn0 = 0(1)

Note: 1. SRWn1 = SRW11 (upper sector) or SRW01 (lower sector), SRWn0 = SRW10 (uppersector) or SRW00 (lower sector).The ALE pulse in period T6 is only present if the next instruction accesses the RAM(internal or external).

ALE

T1 T2 T3

Writ

eR

ead

WR

T5

A15:8 AddressPrev. addr.

DA7:0 Address DataPrev. data XX

RD

DA7:0 (XMBK = 0) DataPrev. data Address

DataPrev. data AddressDA7:0 (XMBK = 1)

System Clock (CLKCPU)

T4

ALE

T1 T2 T3

Writ

eR

ead

WR

T6

A15:8 AddressPrev. addr.

DA7:0 Address DataPrev. data XX

RD

DA7:0 (XMBK = 0) DataPrev. data Address

DataPrev. data AddressDA7:0 (XMBK = 1)

System Clock (CLKCPU)

T4 T5

33

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 19. External Data Memory Cycles with SRWn1 = 1 and SRWn0 = 1(1)

Note: 1. SRWn1 = SRW11 (upper sector) or SRW01 (lower sector), SRWn0 = SRW10 (uppersector) or SRW00 (lower sector).The ALE pulse in period T7 is only present if the next instruction accesses the RAM(internal or external).

External Memory Control Register A – XMCRA

• Bit 7 – SRE: External SRAM/XMEM Enable

Writing SRE to one enables the External Memory Interface.The pin functions AD7:0,A15:8, ALE, WR, and RD are activated as the alternate pin functions. The SRE bit over-rides any pin direction settings in the respective data direction registers. Writing SRE tozero, disables the External Memory Interface and the normal pin and data direction set-tings are used.

• Bit 6..4 – SRL2:0: Wait-state Sector Limit

It is possible to configure different wait-states for different External Memory addresses.The external memory address space can be divided in two sectors that have separatewait-state bits. The SRL2, SRL1, and SRL0 bits select the split of the sectors, see Table4 and Figure 14. By default, the SRL2, SRL1, and SRL0 bits are set to zero and theentire external memory address space is treated as one sector. When the entire SRAMaddress space is configured as one sector, the wait-states are configured by theSRW11 and SRW10 bits.

ALE

T1 T2 T3

Writ

eR

ead

WR

T7

A15:8 AddressPrev. addr.

DA7:0 Address DataPrev. data XX

RD

DA7:0 (XMBK = 0) DataPrev. data Address

DataPrev. data AddressDA7:0 (XMBK = 1)

System Clock (CLKCPU)

T4 T5 T6

Bit 7 6 5 4 3 2 1 0

SRE SRL2 SRL1 SRL0 SRW11 SRW10 SRW01 SRW00 XMCRA

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

34 ATmega640/1280/1281/2560/25612549A–AVR–03/05

• Bit 3..2 – SRW11, SRW10: Wait-state Select Bits for Upper Sector

The SRW11 and SRW10 bits control the number of wait-states for the upper sector ofthe external memory address space, see Table 5.

• Bit 1..0 – SRW01, SRW00: Wait-state Select Bits for Lower Sector

The SRW01 and SRW00 bits control the number of wait-states for the lower sector ofthe external memory address space, see Table 5.

Note: 1. n = 0 or 1 (lower/upper sector).For further details of the timing and wait-states of the External Memory Interface, seeFigures 16 through Figures 19 for how the setting of the SRW bits affects the timing.

Table 4. Sector limits with different settings of SRL2..0

SRL2 SRL1 SRL0 Sector Limits

0 0 xLower sector = N/AUpper sector = 0x2200 - 0xFFFF

0 1 0Lower sector = 0x2200 - 0x3FFFUpper sector = 0x4000 - 0xFFFF

0 1 1Lower sector = 0x2200 - 0x5FFFUpper sector = 0x6000 - 0xFFFF

1 0 0Lower sector = 0x2200 - 0x7FFFUpper sector = 0x8000 - 0xFFFF

1 0 1Lower sector = 0x2200 - 0x9FFFUpper sector = 0xA000 - 0xFFFF

1 1 0Lower sector = 0x2200 - 0xBFFFUpper sector = 0xC000 - 0xFFFF

1 1 1Lower sector = 0x2200 - 0xDFFFUpper sector = 0xE000 - 0xFFFF

Table 5. Wait States(1)

SRWn1 SRWn0 Wait States

0 0 No wait-states

0 1 Wait one cycle during read/write strobe

1 0 Wait two cycles during read/write strobe

1 1 Wait two cycles during read/write and wait one cycle before driving out new address

35

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

External Memory Control Register B – XMCRB

• Bit 7– XMBK: External Memory Bus-keeper Enable

Writing XMBK to one enables the bus keeper on the AD7:0 lines. When the bus keeperis enabled, AD7:0 will keep the last driven value on the lines even if the XMEM interfacehas tri-stated the lines. Writing XMBK to zero disables the bus keeper. XMBK is notqualified with SRE, so even if the XMEM interface is disabled, the bus keepers are stillactivated as long as XMBK is one.

• Bit 6..3 – Res: Reserved Bits

These bits are reserved and will always read as zero. When writing to this address loca-tion, write these bits to zero for compatibility with future devices.

• Bit 2..0 – XMM2, XMM1, XMM0: External Memory High Mask

When the External Memory is enabled, all Port C pins are default used for the highaddress byte. If the full 60KB address space is not required to access the External Mem-ory, some, or all, Port C pins can be released for normal Port Pin function as describedin Table 6. As described in “Using all 64KB Locations of External Memory” on page 36,it is possible to use the XMMn bits to access all 64KB locations of the External Memory.

Using all Locations of External Memory Smaller than 64 KB

Since the external memory is mapped after the internal memory as shown in Figure 14,the external memory is not addressed when addressing the first 8,704 bytes of dataspace. It may appear that the first 8,704 bytes of the external memory are inaccessible(external memory addresses 0x0000 to 0x21FF). However, when connecting an exter-nal memory smaller than 64 KB, for example 32 KB, these locations are easily accessedsimply by addressing from address 0x8000 to 0xA1FF. Since the External MemoryAddress bit A15 is not connected to the external memory, addresses 0x8000 to 0xA1FFwill appear as addresses 0x0000 to 0x21FF for the external memory. Addressing aboveaddress 0xA1FF is not recommended, since this will address an external memory loca-tion that is already accessed by another (lower) address. To the Application software,the external 32 KB memory will appear as one linear 32 KB address space from 0x2200to 0xA1FF. This is illustrated in Figure 20.

Bit 7 6 5 4 3 2 1 0

XMBK – – – – XMM2 XMM1 XMM0 XMCRB

Read/Write R/W R R R R R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 6. Port C Pins Released as Normal Port Pins when the External Memory isEnabled

XMM2 XMM1 XMM0 # Bits for External Memory Address Released Port Pins

0 0 0 8 (Full 56KB space) None

0 0 1 7 PC7

0 1 0 6 PC7 - PC6

0 1 1 5 PC7 - PC5

1 0 0 4 PC7 - PC4

1 0 1 3 PC7 - PC3

1 1 0 2 PC7 - PC2

1 1 1 No Address high bits Full Port C

36 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 20. Address Map with 32 KB External Memory

Using all 64KB Locations of External Memory

Since the External Memory is mapped after the Internal Memory as shown in Figure 14,only 56KB of External Memory is available by default (address space 0x0000 to 0x21FFis reserved for internal memory). However, it is possible to take advantage of the entireExternal Memory by masking the higher address bits to zero. This can be done by usingthe XMMn bits and control by software the most significant bits of the address. By set-ting Port C to output 0x00, and releasing the most significant bits for normal Port Pinoperation, the Memory Interface will address 0x0000 - 0x2FFF. See the following codeexamples.

Care must be exercised using this option as most of the memory is masked away.

0x0000

0x21FF

0xFFFF

0x2200

0x7FFF 0x8000

0x90FF 0x9100

0x0000

0x7FFF

Memory Configuration A

Internal Memory

(Unused)

AVR Memory Map External 32K SRAM

External

Memory

37

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: 1. See “About Code Examples” on page 8.

Assembly Code Example(1)

; OFFSET is defined to 0x4000 to ensure; external memory access; Configure Port C (address high byte) to; output 0x00 when the pins are released; for normal Port Pin operation

ldi r16, 0xFFout DDRC, r16ldi r16, 0x00out PORTC, r16; release PC7:6ldi r16, (1<<XMM1)sts XMCRB, r16; write 0xAA to address 0x0001 of external; memoryldi r16, 0xaasts 0x0001+OFFSET, r16; re-enable PC7:6 for external memoryldi r16, (0<<XMM1)sts XMCRB, r16; store 0x55 to address (OFFSET + 1) of; external memoryldi r16, 0x55sts 0x0001+OFFSET, r16

C Code Example(1)

#define OFFSET 0x4000

void XRAM_example(void)

unsigned char *p = (unsigned char *) (OFFSET + 1);

DDRC = 0xFF;

PORTC = 0x00;

XMCRB = (1<<XMM1);

*p = 0xaa;

XMCRB = 0x00;

*p = 0x55;

38 ATmega640/1280/1281/2560/25612549A–AVR–03/05

39

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

System Clock and Clock Options

Clock Systems and their Distribution

Figure 21 presents the principal clock systems in the AVR and their distribution. All ofthe clocks need not be active at a given time. In order to reduce power consumption, theclocks to modules not being used can be halted by using different sleep modes, asdescribed in “Power Management and Sleep Modes” on page 51. The clock systemsare detailed below.

Figure 21. Clock Distribution

CPU Clock – clkCPU The CPU clock is routed to parts of the system concerned with operation of the AVRcore. Examples of such modules are the General Purpose Register File, the Status Reg-ister and the data memory holding the Stack Pointer. Halting the CPU clock inhibits thecore from performing general operations and calculations.

I/O Clock – clkI/O The I/O clock is used by the majority of the I/O modules, like Timer/Counters, SPI, andUSART. The I/O clock is also used by the External Interrupt module, but note that someexternal interrupts are detected by asynchronous logic, allowing such interrupts to bedetected even if the I/O clock is halted. Also note that start condition detection in the USImodule is carried out asynchronously when clkI/O is halted, TWI address recognition inall sleep modes.

Flash Clock – clkFLASH The Flash clock controls operation of the Flash interface. The Flash clock is usuallyactive simultaneously with the CPU clock.

General I/OModules

AsynchronousTimer/Counter

CPU Core RAM

clkI/O

clkASY

AVR ClockControl Unit

clkCPU

Flash andEEPROM

clkFLASH

Source clock

Watchdog Timer

WatchdogOscillator

Reset Logic

ClockMultiplexer

Watchdog clock

Calibrated RCOscillator

Timer/CounterOscillator

CrystalOscillator

Low-frequencyCrystal Oscillator

External Clock

ADC

clkADC

System ClockPrescaler

40 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Asynchronous Timer Clock – clkASY

The Asynchronous Timer clock allows the Asynchronous Timer/Counter to be clockeddirectly from an external clock or an external 32 kHz clock crystal. The dedicated clockdomain allows using this Timer/Counter as a real-time counter even when the device isin sleep mode.

ADC Clock – clkADC The ADC is provided with a dedicated clock domain. This allows halting the CPU andI/O clocks in order to reduce noise generated by digital circuitry. This gives more accu-rate ADC conversion results.

Clock Sources The device has the following clock source options, selectable by Flash Fuse bits asshown below. The clock from the selected source is input to the AVR clock generator,and routed to the appropriate modules.

Note: 1. For all fuses “1” means unprogrammed while “0” means programmed.

Default Clock Source The device is shipped with internal RC oscillator at 8.0MHz and with the fuse CKDIV8programmed, resulting in 1.0MHz system clock. The startup time is set to maximum andtime-out period enabled. (CKSEL = "0010", SUT = "10", CKDIV8 = "0"). The default set-ting ensures that all users can make their desired clock source setting using anyavailable programming interface.

Clock Startup Sequence Any clock source needs a sufficient VCC to start oscillating and a minimum number ofoscillating cycles before it can be considered stable.

To ensure sufficient VCC, the device issues an internal reset with a time-out delay (tTOUT)after the device reset is released by all other reset sources. “On-chip Debug System” onpage 56 describes the start conditions for the internal reset. The delay (tTOUT) is timedfrom the Watchdog Oscillator and the number of cycles in the delay is set by the SUTxand CKSELx fuse bits. The selectable delays are shown in Table 8. The frequency ofthe Watchdog Osc i l l a to r i s vo l t age dependen t as shown in“ATmega640/1280/1281/2560/2561 Typical Characteristics – Preliminary Data” on page381.

Table 7. Device Clocking Options Select(1)

Device Clocking Option CKSEL3..0

Low Power Crystal Oscillator 1111 - 1000

Full Swing Crystal Oscillator 0111 - 0110

Low Frequency Crystal Oscillator 0101 - 0100

Internal 128 kHz RC Oscillator 0011

Calibrated Internal RC Oscillator 0010

External Clock 0000

Reserved 0001

Table 8. Number of Watchdog Oscillator Cycles

Typ Time-out (VCC = 5.0V) Typ Time-out (VCC = 3.0V) Number of Cycles

0 ms 0 ms 0

4.1 ms 4.3 ms 512

65 ms 69 ms 8K (8,192)

41

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Main purpose of the delay is to keep the AVR in reset until it is supplied with minimumVcc. The delay will not monitor the actual voltage and it will be required to select a delaylonger than the Vcc rise time. If this is not possible, an internal or external Brown-OutDetection circuit should be used. A BOD circuit will ensure sufficient Vcc before itreleases the reset, and the time-out delay can be disabled. Disabling the time-out delaywithout utilizing a Brown-Out Detection circuit is not recommended.

The oscillator is required to oscillate for a minimum number of cycles before the clock isconsidered stable. An internal ripple counter monitors the oscillator output clock, andkeeps the internal reset active for a given number of clock cycles. The reset is thenreleased and the device will start to execute. The recommended oscillator start-up timeis dependent on the clock type, and varies from 6 cycles for an externally applied clockto 32K cycles for a low frequency crystal.

The start-up sequence for the clock includes both the time-out delay and the start-uptime when the device starts up from reset. When starting up from Power-save or Power-down mode, Vcc is assumed to be at a sufficient level and only the start-up time isincluded.

Low Power Crystal Oscillator

Pins XTAL1 and XTAL2 are input and output, respectively, of an inverting amplifierwhich can be configured for use as an On-chip Oscillator, as shown in Figure 22. Eithera quartz crystal or a ceramic resonator may be used.

This Crystal Oscillator is a low power oscillator, with reduced voltage swing on theXTAL2 output. It gives the lowest power consumption, but is not capable of driving otherclock inputs, and may be more susceptible to noise in noisy environments. In thesecases, refer to the “Full Swing Crystal Oscillator” on page 43.

C1 and C2 should always be equal for both crystals and resonators. The optimal valueof the capacitors depends on the crystal or resonator in use, the amount of stray capac-itance, and the electromagnetic noise of the environment. Some initial guidelines forchoosing capacitors for use with crystals are given in Table 9. For ceramic resonators,the capacitor values given by the manufacturer should be used.

Figure 22. Crystal Oscillator Connections

The Low Power Oscillator can operate in three different modes, each optimized for aspecific frequency range. The operating mode is selected by the fuses CKSEL3..1 asshown in Table 9.

XTAL2

XTAL1

GND

C2

C1

42 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Notes: 1. The frequency ranges are preliminary values. Actual values are TBD.2. This option should not be used with crystals, only with ceramic resonators.3. If 8 MHz frequency exceeds the specification of the device (depends on VCC), the

CKDIV8 Fuse can be programmed in order to divide the internal frequency by 8. Itmust be ensured that the resulting divided clock meets the frequency specification ofthe device.

The CKSEL0 Fuse together with the SUT1..0 Fuses select the start-up times as shownin Table 10.

Notes: 1. These options should only be used when not operating close to the maximum fre-quency of the device, and only if frequency stability at start-up is not important for theapplication. These options are not suitable for crystals.

2. These options are intended for use with ceramic resonators and will ensure fre-quency stability at start-up. They can also be used with crystals when not operatingclose to the maximum frequency of the device, and if frequency stability at start-up isnot important for the application.

Table 9. Low Power Crystal Oscillator Operating Modes(3)

Frequency Range(1) (MHz) CKSEL3..1Recommended Range for Capacitors

C1 and C2 (pF)

0.4 - 0.9 100(2) –

0.9 - 3.0 101 12 - 22

3.0 - 8.0 110 12 - 22

8.0 - 16.0 111 12 - 22

Table 10. Start-up Times for the Low Power Crystal Oscillator Clock Selection

Oscillator Source / Power Conditions

Start-up Time from Power-down and

Power-save

Additional Delay from Reset (VCC = 5.0V) CKSEL0 SUT1..0

Ceramic resonator, fast rising power

258 CK 14CK + 4.1 ms(1) 0 00

Ceramic resonator, slowly rising power

258 CK 14CK + 65 ms(1) 0 01

Ceramic resonator, BOD enabled

1K CK 14CK(2) 0 10

Ceramic resonator, fast rising power

1K CK 14CK + 4.1 ms(2) 0 11

Ceramic resonator, slowly rising power

1K CK 14CK + 65 ms(2) 1 00

Crystal Oscillator, BOD enabled

16K CK 14CK1

01

Crystal Oscillator, fast rising power

16K CK 14CK + 4.1 ms1

10

Crystal Oscillator, slowly rising power

16K CK 14CK + 65 ms1

11

43

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Full Swing Crystal Oscillator

Pins XTAL1 and XTAL2 are input and output, respectively, of an inverting amplifierwhich can be configured for use as an On-chip Oscillator, as shown in Figure 22. Eithera quartz crystal or a ceramic resonator may be used.

This Crystal Oscillator is a full swing oscillator, with rail-to-rail swing on the XTAL2 out-put. This is useful for driving other clock inputs and in noisy environments. The currentconsumption is higher than the “Low Power Crystal Oscillator” on page 41. Note that theFull Swing Crystal Oscillator will only operate for Vcc = 2.7 - 5.5 volts.

C1 and C2 should always be equal for both crystals and resonators. The optimal valueof the capacitors depends on the crystal or resonator in use, the amount of stray capac-itance, and the electromagnetic noise of the environment. Some initial guidelines forchoosing capacitors for use with crystals are given in Table 12. For ceramic resonators,the capacitor values given by the manufacturer should be used.

The operating mode is selected by the fuses CKSEL3..1 as shown in Table 11.

Notes: 1. The frequency ranges are preliminary values. Actual values are TBD.2. If 8 MHz frequency exceeds the specification of the device (depends on VCC), the

CKDIV8 Fuse can be programmed in order to divide the internal frequency by 8. Itmust be ensured that the resulting divided clock meets the frequency specification ofthe device.

Figure 23. Crystal Oscillator Connections

Table 11. Full Swing Crystal Oscillator operating modes(2)

Frequency Range(1) (MHz) CKSEL3..1Recommended Range for Capacitors

C1 and C2 (pF)

0.4 - 16 011 12 - 22

XTAL2

XTAL1

GND

C2

C1

44 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Notes: 1. These options should only be used when not operating close to the maximum fre-quency of the device, and only if frequency stability at start-up is not important for theapplication. These options are not suitable for crystals.

2. These options are intended for use with ceramic resonators and will ensure fre-quency stability at start-up. They can also be used with crystals when not operatingclose to the maximum frequency of the device, and if frequency stability at start-up isnot important for the application.

Table 12. Start-up Times for the Full Swing Crystal Oscillator Clock Selection

Oscillator Source / Power Conditions

Start-up Time from Power-down and

Power-save

Additional Delay from Reset (VCC = 5.0V) CKSEL0 SUT1..0

Ceramic resonator, fast rising power

258 CK 14CK + 4.1 ms(1) 0 00

Ceramic resonator, slowly rising power

258 CK 14CK + 65 ms(1) 0 01

Ceramic resonator, BOD enabled

1K CK 14CK(2) 0 10

Ceramic resonator, fast rising power

1K CK 14CK + 4.1 ms(2) 0 11

Ceramic resonator, slowly rising power

1K CK 14CK + 65 ms(2) 1 00

Crystal Oscillator, BOD enabled

16K CK 14CK1

01

Crystal Oscillator, fast rising power

16K CK 14CK + 4.1 ms1

10

Crystal Oscillator, slowly rising power

16K CK 14CK + 65 ms1

11

45

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Low Frequency Crystal Oscillator

The device can utilize a 32.768 kHz watch crystal as clock source by a dedicated LowFrequency Crystal Oscillator. The crystal should be connected as shown in Figure 22.When this Oscillator is selected, start-up times are determined by the SUT Fuses andCKSEL0 as shown in Table 13.

Note: 1. These options should only be used if frequency stability at start-up is not importantfor the application.

Calibrated Internal RC Oscillator

The calibrated internal RC Oscillator by default provides a 8.0 MHz clock. The frequencyis nominal value at 3V and 25°C. The device is shipped with the CKDIV8 Fuse pro-grammed. See “System Clock Prescaler” on page 48 for more details. This clock maybe selected as the system clock by programming the CKSEL Fuses as shown in Table14. If selected, it will operate with no external components. During reset, hardware loadsthe calibration byte into the OSCCAL Register and thereby automatically calibrates theRC Oscillator. At 3V and 25°C, this calibration gives a frequency of 8 MHz ± 1%. Theoscillator can be calibrated to any frequency in the range 7.3 - 8.1 MHz within ±1%accuracy, by changing the OSCCAL register. When this Oscillator is used as the chipclock, the Watchdog Oscillator will still be used for the Watchdog Timer and for theReset Time-out. For more information on the pre-programmed calibration value, see thesection “Calibration Byte” on page 338.

Notes: 1. The device is shipped with this option selected.2. The frequency ranges are preliminary values. Actual values are TBD.3. If 8 MHz frequency exceeds the specification of the device (depends on VCC), the

CKDIV8 Fuse can be programmed in order to divide the internal frequency by 8.

When this Oscillator is selected, start-up times are determined by the SUT Fuses asshown in Table 15 on page 46.

Table 13. Start-up Times for the Low Frequency Crystal Oscillator Clock Selection

Power Conditions

Start-up Time from Power-down and

Power-save

Additional Delay from Reset (VCC = 5.0V) CKSEL0 SUT1..0

BOD enabled 1K CK 14CK(1) 0 00

Fast rising power 1K CK 14CK + 4.1 ms(1) 0 01

Slowly rising power 1K CK 14CK + 65 ms(1) 0 10

Reserved 0 11

BOD enabled 32K CK 14CK 1 00

Fast rising power 32K CK 14CK + 4.1 ms 1 01

Slowly rising power 32K CK 14CK + 65 ms 1 10

Reserved 1 11

Table 14. Internal Calibrated RC Oscillator Operating Modes(1)(3)

Frequency Range(2) (MHz) CKSEL3..0

7.3 - 8.1 0010

46 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. The device is shipped with this option selected.

Oscillator Calibration Register – OSCCAL

• Bits 7..0 – CAL7..0: Oscillator Calibration Value

The Oscillator Calibration Register is used to trim the Calibrated Internal RC Oscillatorto remove process variations from the oscillator frequency. The factory-calibrated valueis automatically written to this register during chip reset, giving an oscillator frequency of8.0 MHz at 25°C. The application software can write this register to change the oscillatorfrequency. The oscillator can be calibrated to any frequency in the range 7.3 - 8.1 MHzwithin ±1% accuracy. Calibration outside that range is not guaranteed.

Note that this oscillator is used to time EEPROM and Flash write accesses, and thesewrite times will be affected accordingly. If the EEPROM or Flash are written, do not cali-brate to more than 8.8 MHz. Otherwise, the EEPROM or Flash write may fail.

The CAL7 bit determines the range of operation for the oscillator. Setting this bit to 0gives the lowest frequency range, setting this bit to 1 gives the highest frequency range.The two frequency ranges are overlapping, in other words a setting of OSCCAL = 0x7Fgives a higher frequency than OSCCAL = 0x80.

The CAL6..0 bits are used to tune the frequency within the selected range. A setting of0x00 gives the lowest frequency in that range, and a setting of 0x7F gives the highestfrequency in the range. Incrementing CAL6..0 by 1 will give a frequency increment ofless than 2% in the frequency range 7.3 - 8.1 MHz.

Table 15. Start-up times for the internal calibrated RC Oscillator clock selection

Power ConditionsStart-up Time from Power-

down and Power-saveAdditional Delay from

Reset (VCC = 5.0V) SUT1..0

BOD enabled 6 CK 14CK 00

Fast rising power 6 CK 14CK + 4.1 ms 01

Slowly rising power 6 CK 14CK + 65 ms(1) 10

Reserved 11

Bit 7 6 5 4 3 2 1 0

CAL7 CAL6 CAL5 CAL4 CAL3 CAL2 CAL1 CAL0 OSCCAL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value Device Specific Calibration Value

47

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

128 kHz Internal Oscillator

The 128 kHz internal Oscillator is a low power Oscillator providing a clock of 128 kHz.The frequency is nominal at 3V and 25°C. This clock may be select as the system clockby programming the CKSEL Fuses to “11” as shown in Table 16.

Note: 1. The frequency is preliminary value. Actual value is TBD.

When this clock source is selected, start-up times are determined by the SUT Fuses asshown in Table 17.

External Clock To drive the device from an external clock source, XTAL1 should be driven as shown inFigure 24. To run the device on an external clock, the CKSEL Fuses must be pro-grammed to “0000”.

Figure 24. External Clock Drive Configuration

When this clock source is selected, start-up times are determined by the SUT Fuses asshown in Table 20..

Table 16. 128 kHz Internal Oscillator Operating Modes

Nominal Frequency CKSEL3..0

128 kHz 0011

Table 17. Start-up Times for the 128 kHz Internal Oscillator

Power ConditionsStart-up Time from Power-

down and Power-saveAdditional Delay from

Reset SUT1..0

BOD enabled 6 CK 14CK 00

Fast rising power 6 CK 14CK + 4 ms 01

Slowly rising power 6 CK 14CK + 64 ms 10

Reserved 11

Table 18. Crystal Oscillator Clock Frequency

Nominal Frequency CKSEL3..0

0 - 16 MHz 0000

NC

EXTERNALCLOCKSIGNAL

XTAL2

XTAL1

GND

48 ATmega640/1280/1281/2560/25612549A–AVR–03/05

When applying an external clock, it is required to avoid sudden changes in the appliedclock frequency to ensure stable operation of the MCU. A variation in frequency of morethan 2% from one clock cycle to the next can lead to unpredictable behavior. If changesof more than 2% is required, ensure that the MCU is kept in Reset during the changes.

Note that the System Clock Prescaler can be used to implement run-time changes ofthe internal clock frequency while still ensuring stable operation. Refer to “System ClockPrescaler” on page 48 for details.

Clock Output Buffer The device can output the system clock on the CLKO pin. To enable the output, theCKOUT Fuse has to be programmed. This mode is suitable when the chip clock is usedto drive other circuits on the system. The clock also will be output during reset, and thenormal operation of I/O pin will be overridden when the fuse is programmed. Any clocksource, including the internal RC Oscillator, can be selected when the clock is output onCLKO. If the System Clock Prescaler is used, it is the divided system clock that isoutput.

Timer/Counter Oscillator The device can operate its Timer/Counter2 from an external 32.768 kHz watch crystal ora external clock source. See Figure 22 on page 41 for crystal connection.

Applying an external clock source to TOSC1 requires EXCLK in the ASSR Register writ-ten to logic one. See “Asynchronous operation of the Timer/Counter” on page 189 forfurther description on selecting external clock as input instead of a 32 kHz crystal.

System Clock Prescaler The ATmega640/1280/1281/2560/2561 has a system clock prescaler, and the systemclock can be divided by setting the “Clock Prescale Register – CLKPR” on page 49. Thisfeature can be used to decrease the system clock frequency and the power consump-tion when the requirement for processing power is low. This can be used with all clocksource options, and it will affect the clock frequency of the CPU and all synchronousperipherals. clkI/O, clkADC, clkCPU, and clkFLASH are divided by a factor as shown in Table20.

When switching between prescaler settings, the System Clock Prescaler ensures thatno glitches occurs in the clock system. It also ensures that no intermediate frequency ishigher than neither the clock frequency corresponding to the previous setting, nor theclock frequency corresponding to the new setting.

The ripple counter that implements the prescaler runs at the frequency of the undividedclock, which may be faster than the CPU's clock frequency. Hence, it is not possible todetermine the state of the prescaler - even if it were readable, and the exact time it takesto switch from one clock division to the other cannot be exactly predicted. From the timethe CLKPS values are written, it takes between T1 + T2 and T1 + 2 * T2 before the newclock frequency is active. In this interval, 2 active clock edges are produced. Here, T1 is

Table 19. Start-up Times for the External Clock Selection

Power ConditionsStart-up Time from Power-

down and Power-saveAdditional Delay from

Reset (VCC = 5.0V) SUT1..0

BOD enabled 6 CK 14CK 00

Fast rising power 6 CK 14CK + 4.1 ms 01

Slowly rising power 6 CK 14CK + 65 ms 10

Reserved 11

49

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

the previous clock period, and T2 is the period corresponding to the new prescalersetting.

To avoid unintentional changes of clock frequency, a special write procedure must befollowed to change the CLKPS bits:

1. Write the Clock Prescaler Change Enable (CLKPCE) bit to one and all other bitsin CLKPR to zero.

2. Within four cycles, write the desired value to CLKPS while writing a zero toCLKPCE.

Interrupts must be disabled when changing prescaler setting to make sure the write pro-cedure is not interrupted.

Clock Prescale Register – CLKPR

• Bit 7 – CLKPCE: Clock Prescaler Change Enable

The CLKPCE bit must be written to logic one to enable change of the CLKPS bits. TheCLKPCE bit is only updated when the other bits in CLKPR are simultaneously written tozero. CLKPCE is cleared by hardware four cycles after it is written or when CLKPS bitsare written. Rewriting the CLKPCE bit within this time-out period does neither extend thetime-out period, nor clear the CLKPCE bit.

• Bits 3..0 – CLKPS3..0: Clock Prescaler Select Bits 3 - 0

These bits define the division factor between the selected clock source and the internalsystem clock. These bits can be written run-time to vary the clock frequency to suit theapplication requirements. As the divider divides the master clock input to the MCU, thespeed of all synchronous peripherals is reduced when a division factor is used. The divi-sion factors are given in Table 20.

The CKDIV8 Fuse determines the initial value of the CLKPS bits. If CKDIV8 is unpro-grammed, the CLKPS bits will be reset to “0000”. If CKDIV8 is programmed, CLKPS bitsare reset to “0011”, giving a division factor of 8 at start up. This feature should be used ifthe selected clock source has a higher frequency than the maximum frequency of thedevice at the present operating conditions. Note that any value can be written to theCLKPS bits regardless of the CKDIV8 Fuse setting. The Application software mustensure that a sufficient division factor is chosen if the selected clock source has a higherfrequency than the maximum frequency of the device at the present operating condi-tions. The device is shipped with the CKDIV8 Fuse programmed.

Bit 7 6 5 4 3 2 1 0

CLKPCE – – – CLKPS3 CLKPS2 CLKPS1 CLKPS0 CLKPR

Read/Write R/W R R R R/W R/W R/W R/W

Initial Value 0 0 0 0 See Bit Description

50 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Table 20. Clock Prescaler Select

CLKPS3 CLKPS2 CLKPS1 CLKPS0 Clock Division Factor

0 0 0 0 1

0 0 0 1 2

0 0 1 0 4

0 0 1 1 8

0 1 0 0 16

0 1 0 1 32

0 1 1 0 64

0 1 1 1 128

1 0 0 0 256

1 0 0 1 Reserved

1 0 1 0 Reserved

1 0 1 1 Reserved

1 1 0 0 Reserved

1 1 0 1 Reserved

1 1 1 0 Reserved

1 1 1 1 Reserved

51

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Power Management and Sleep Modes

Sleep modes enable the application to shut down unused modules in the MCU, therebysaving power. The AVR provides various sleep modes allowing the user to tailor thepower consumption to the application’s requirements.

To enter any of the five sleep modes, the SE bit in SMCR must be written to logic oneand a SLEEP instruction must be executed. The SM2, SM1, and SM0 bits in the SMCRRegister select which sleep mode (Idle, ADC Noise Reduction, Power-down, Power-save, or Standby) will be activated by the SLEEP instruction. See Table 21 for a sum-mary. If an enabled interrupt occurs while the MCU is in a sleep mode, the MCU wakesup. The MCU is then halted for four cycles in addition to the start-up time, executes theinterrupt routine, and resumes execution from the instruction following SLEEP. The con-tents of the Register File and SRAM are unaltered when the device wakes up fromsleep. If a reset occurs during sleep mode, the MCU wakes up and executes from theReset Vector.

F igu re 21 on page 39 p resen ts the d i f fe ren t c lock sys tems in t heATmega640/1280/1281/2560/2561, and their distribution. The figure is helpful in select-ing an appropriate sleep mode.

Sleep Mode Control Register – SMCR

The Sleep Mode Control Register contains control bits for power management.

• Bits 3, 2, 1 – SM2..0: Sleep Mode Select Bits 2, 1, and 0

These bits select between the five available sleep modes as shown in Table 21.

Note: 1. Standby modes are only recommended for use with external crystals or resonators.

• Bit 1 – SE: Sleep Enable

The SE bit must be written to logic one to make the MCU enter the sleep mode when theSLEEP instruction is executed. To avoid the MCU entering the sleep mode unless it isthe programmer’s purpose, it is recommended to write the Sleep Enable (SE) bit to onejust before the execution of the SLEEP instruction and to clear it immediately after wak-ing up.

Bit 7 6 5 4 3 2 1 0

– – – – SM2 SM1 SM0 SE SMCR

Read/Write R R R R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 21. Sleep Mode Select

SM2 SM1 SM0 Sleep Mode

0 0 0 Idle

0 0 1 ADC Noise Reduction

0 1 0 Power-down

0 1 1 Power-save

1 0 0 Reserved

1 0 1 Reserved

1 1 0 Standby(1)

1 1 1 Extended Standby(1)

52 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Idle Mode When the SM2..0 bits are written to 000, the SLEEP instruction makes the MCU enterIdle mode, stopping the CPU but allowing the SPI, USART, Analog Comparator, ADC,2-wire Serial Interface, Timer/Counters, Watchdog, and the interrupt system to continueoperating. This sleep mode basically halts clkCPU and clkFLASH, while allowing the otherclocks to run.

Idle mode enables the MCU to wake up from external triggered interrupts as well asinternal ones like the Timer Overflow and USART Transmit Complete interrupts. Ifwake-up from the Analog Comparator interrupt is not required, the Analog Comparatorcan be powered down by setting the ACD bit in the Analog Comparator Control and Sta-tus Register – ACSR. This will reduce power consumption in Idle mode. If the ADC isenabled, a conversion starts automatically when this mode is entered.

ADC Noise Reduction Mode

When the SM2..0 bits are written to 001, the SLEEP instruction makes the MCU enterADC Noise Reduction mode, stopping the CPU but allowing the ADC, the external inter-rupts, 2-wire Serial Interface address match, Timer/Counter2 and the Watchdog tocontinue operating (if enabled). This sleep mode basically halts clkI/O, clkCPU, and clk-FLASH, while allowing the other clocks to run.

This improves the noise environment for the ADC, enabling higher resolution measure-ments. If the ADC is enabled, a conversion starts automatically when this mode isentered. Apart form the ADC Conversion Complete interrupt, only an External Reset, aWatchdog System Reset, a Watchdog interrupt, a Brown-out Reset, a 2-wire serial inter-face interrupt, a Timer/Counter2 interrupt, an SPM/EEPROM ready interrupt, anexternal level interrupt on INT7:4 or a pin change interrupt can wakeup the MCU fromADC Noise Reduction mode.

Power-down Mode When the SM2..0 bits are written to 010, the SLEEP instruction makes the MCU enterPower-down mode. In this mode, the external Oscillator is stopped, while the externalinterrupts, the 2-wire Serial Interface, and the Watchdog continue operating (if enabled).Only an External Reset, a Watchdog Reset, a Brown-out Reset, 2-wire Serial Interfaceaddress match, an external level interrupt on INT7:4, an external interrupt on INT3:0, ora pin change interrupt can wake up the MCU. This sleep mode basically halts all gener-ated clocks, allowing operation of asynchronous modules only.

Note that if a level triggered interrupt is used for wake-up from Power-down mode, thechanged level must be held for some time to wake up the MCU. Refer to “External Inter-rupts” on page 75 for details.

When waking up from Power-down mode, there is a delay from the wake-up conditionoccurs until the wake-up becomes effective. This allows the clock to restart and becomestable after having been stopped. The wake-up period is defined by the same CKSELFuses that define the Reset Time-out period, as described in “Clock Sources” on page40.

Power-save Mode When the SM2..0 bits are written to 011, the SLEEP instruction makes the MCU enterPower-save mode. This mode is identical to Power-down, with one exception:

If Timer/Counter2 is enabled, it will keep running during sleep. The device can wake upfrom either Timer Overflow or Output Compare event from Timer/Counter2 if the corre-sponding Timer/Counter2 interrupt enable bits are set in TIMSK2, and the GlobalInterrupt Enable bit in SREG is set.

If Timer/Counter2 is not running, Power-down mode is recommended instead of Power-save mode.

53

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The Timer/Counter2 can be clocked both synchronously and asynchronously in Power-save mode. If the Timer/Counter2 is not using the asynchronous clock, theTimer/Counter Oscillator is stopped during sleep. If the Timer/Counter2 is not using thesynchronous clock, the clock source is stopped during sleep. Note that even if the syn-chronous clock is running in Power-save, this clock is only available for theTimer/Counter2.

Standby Mode When the SM2..0 bits are 110 and an external crystal/resonator clock option is selected,the SLEEP instruction makes the MCU enter Standby mode. This mode is identical toPower-down with the exception that the Oscillator is kept running. From Standby mode,the device wakes up in six clock cycles.

Extended Standby Mode When the SM2..0 bits are 111 and an external crystal/resonator clock option is selected,the SLEEP instruction makes the MCU enter Extended Standby mode. This mode isidentical to Power-save mode with the exception that the Oscillator is kept running.From Extended Standby mode, the device wakes up in six clock cycles.

Notes: 1. Only recommended with external crystal or resonator selected as clock source.2. If Timer/Counter2 is running in asynchronous mode.3. For INT7:4, only level interrupt.

Table 22. Active Clock Domains and Wake-up Sources in the Different Sleep Modes.

Active Clock Domains Oscillators Wake-up Sources

Sleep Mode clk C

PU

clk F

LA

SH

clk I

O

clk A

DC

clk A

SY

Mai

n C

lock

So

urc

e E

nab

led

Tim

er O

sc

En

able

d

INT

7:0

and

P

in C

han

ge

TW

I Ad

dre

ss

Mat

ch

Tim

er2

SP

M/

EE

PR

OM

Rea

dy

AD

C

WD

T In

terr

up

t

Oth

er I/

O

Idle X X X X X(2) X X X X X X X

ADCNRM X X X X(2) X(3) X X(2) X X X

Power-down X(3) X X

Power-save X X(2) X(3) X X X

Standby(1) X X(3) X X

Extended Standby

X(2) X X(2) X(3) X X X

54 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Power Reduction Register

The Power Reduction Register, PRR, provides a method to stop the clock to individualperipherals to reduce power consumption. The current state of the peripheral is frozenand the I/O registers can not be read or written. Resources used by the peripheral whenstopping the clock will remain occupied, hence the peripheral should in most cases bedisabled before stopping the clock. Waking up a module, which is done by clearing thebit in PRR, puts the module in the same state as before shutdown.

Module shutdown can be used in Idle mode and Active mode to significantly reduce theoverall power consumption. See “Supply Current of IO modules” on page 381 for exam-ples. In all other sleep modes, the clock is already stopped.

Power Reduction Register 0 - PRR0

• Bit 7 - PRTWI: Power Reduction TWI

Writing a logic one to this bit shuts down the TWI by stopping the clock to the module.When waking up the TWI again, the TWI should be re initialized to ensure properoperation.

• Bit 6 - PRTIM2: Power Reduction Timer/Counter2

Writing a logic one to this bit shuts down the Timer/Counter2 module in synchronousmode (AS2 is 0). When the Timer/Counter2 is enabled, operation will continue likebefore the shutdown.

• Bit 5 - PRTIM0: Power Reduction Timer/Counter0

Writing a logic one to this bit shuts down the Timer/Counter0 module. When theTimer/Counter0 is enabled, operation will continue like before the shutdown.

• Bit 4 - Res: Reserved bit

This bit is reserved bit and will always read as zero.

• Bit 3 - PRTIM1: Power Reduction Timer/Counter1

Writing a logic one to this bit shuts down the Timer/Counter1 module. When theTimer/Counter1 is enabled, operation will continue like before the shutdown.

• Bit 2 - PRSPI: Power Reduction Serial Peripheral Interface

Writing a logic one to this bit shuts down the Serial Peripheral Interface by stopping theclock to the module. When waking up the SPI again, the SPI should be re initialized toensure proper operation.

• Bit 1 - PRUSART0: Power Reduction USART0

Writing a logic one to this bit shuts down the USART0 by stopping the clock to the mod-ule. When waking up the USART0 again, the USART0 should be re initialized to ensureproper operation.

• Bit 0 - PRADC: Power Reduction ADC

Writing a logic one to this bit shuts down the ADC. The ADC must be disabled beforeshut down. The analog comparator cannot use the ADC input MUX when the ADC isshut down.

Bit 7 6 5 4 3 2 1 0

PRTWI PRTIM2 PRTIM0 – PRTIM1 PRSPI PRUSART0 PRADC PRR0

Read/Write R/W R/W R/W R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

55

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Power Reduction Register 1 - PRR1

• Bit 7..6 - Res: Reserved bits

These bits are reserved and will always read as zero.

• Bit 5 - PRTIM5: Power Reduction Timer/Counter5

Writing a logic one to this bit shuts down the Timer/Counter5 module. When theTimer/Counter5 is enabled, operation will continue like before the shutdown.

• Bit 4 - PRTIM4: Power Reduction Timer/Counter4

Writing a logic one to this bit shuts down the Timer/Counter4 module. When theTimer/Counter4 is enabled, operation will continue like before the shutdown.

• Bit 3 - PRTIM3: Power Reduction Timer/Counter3

Writing a logic one to this bit shuts down the Timer/Counter3 module. When theTimer/Counter3 is enabled, operation will continue like before the shutdown.

• Bit 2 - PRUSART3: Power Reduction USART3

Writing a logic one to this bit shuts down the USART3 by stopping the clock to the mod-ule. When waking up the USART3 again, the USART3 should be re initialized to ensureproper operation.

• Bit 1 - PRUSART2: Power Reduction USART2

Writing a logic one to this bit shuts down the USART2 by stopping the clock to the mod-ule. When waking up the USART2 again, the USART2 should be re initialized to ensureproper operation.

• Bit 0 - PRUSART1: Power Reduction USART1

Writing a logic one to this bit shuts down the USART1 by stopping the clock to the mod-ule. When waking up the USART1 again, the USART1 should be re initialized to ensureproper operation.

Minimizing Power Consumption

There are several issues to consider when trying to minimize the power consumption inan AVR controlled system. In general, sleep modes should be used as much as possi-ble, and the sleep mode should be selected so that as few as possible of the device’sfunctions are operating. All functions not needed should be disabled. In particular, thefollowing modules may need special consideration when trying to achieve the lowestpossible power consumption.

Analog to Digital Converter If enabled, the ADC will be enabled in all sleep modes. To save power, the ADC shouldbe disabled before entering any sleep mode. When the ADC is turned off and on again,the next conversion will be an extended conversion. Refer to “Analog to Digital Con-verter” on page 274 for details on ADC operation.

Analog Comparator When entering Idle mode, the Analog Comparator should be disabled if not used. Whenentering ADC Noise Reduction mode, the Analog Comparator should be disabled. Inother sleep modes, the Analog Comparator is automatically disabled. However, if theAnalog Comparator is set up to use the Internal Voltage Reference as input, the AnalogComparator should be disabled in all sleep modes. Otherwise, the Internal Voltage Ref-erence will be enabled, independent of sleep mode. Refer to “Analog Comparator” onpage 271 for details on how to configure the Analog Comparator.

Bit 7 6 5 4 3 2 1 0

– – PRTIM5 PRTIM4 PRTIM3 PRUSART3 PRUSART2 PRUSART1 PRR1

Read/Write R R R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

56 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Brown-out Detector If the Brown-out Detector is not needed by the application, this module should be turnedoff. If the Brown-out Detector is enabled by the BODLEVEL Fuses, it will be enabled inall sleep modes, and hence, always consume power. In the deeper sleep modes, thiswill contribute significantly to the total current consumption. Refer to “Brown-out Detec-tion” on page 59 for details on how to configure the Brown-out Detector.

Internal Voltage Reference The Internal Voltage Reference will be enabled when needed by the Brown-out Detec-tion, the Analog Comparator or the ADC. If these modules are disabled as described inthe sections above, the internal voltage reference will be disabled and it will not be con-suming power. When turned on again, the user must allow the reference to start upbefore the output is used. If the reference is kept on in sleep mode, the output can beused immediately. Refer to “Internal Voltage Reference” on page 62 for details on thestart-up time.

Watchdog Timer If the Watchdog Timer is not needed in the application, the module should be turned off.If the Watchdog Timer is enabled, it will be enabled in all sleep modes, and hence,always consume power. In the deeper sleep modes, this will contribute significantly tothe total current consumption. Refer to “Interrupts” on page 69 for details on how to con-figure the Watchdog Timer.

Port Pins When entering a sleep mode, all port pins should be configured to use minimum power.The most important is then to ensure that no pins drive resistive loads. In sleep modeswhere both the I/O clock (clkI/O) and the ADC clock (clkADC) are stopped, the input buff-ers of the device will be disabled. This ensures that no power is consumed by the inputlogic when not needed. In some cases, the input logic is needed for detecting wake-upconditions, and it will then be enabled. Refer to the section “Digital Input Enable andSleep Modes” on page 85 for details on which pins are enabled. If the input buffer isenabled and the input signal is left floating or have an analog signal level close to VCC/2,the input buffer will use excessive power.

For analog input pins, the digital input buffer should be disabled at all times. An analogsignal level close to VCC/2 on an input pin can cause significant current even in activemode. Digital input buffers can be disabled by writing to the Digital Input Disable Regis-ters (DIDR2, DIDR1 and DIDR0). Refer to “Digital Input Disable Register 2 – DIDR2” onpage 293, “Digital Input Disable Register 1 – DIDR1” on page 273 and “Digital Input Dis-able Register 0 – DIDR0” on page 293 for details.

On-chip Debug System If the On-chip debug system is enabled by the OCDEN Fuse and the chip enters sleepmode, the main clock source is enabled, and hence, always consumes power. In thedeeper sleep modes, this will contribute significantly to the total current consumption.

There are three alternative ways to disable the OCD system:

• Disable the OCDEN Fuse.

• Disable the JTAGEN Fuse.

• Write one to the JTD bit in MCUCR.

57

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

System Control and Reset

Resetting the AVR During reset, all I/O Registers are set to their initial values, and the program starts exe-cution from the Reset Vector. The instruction placed at the Reset Vector must be a JMP– Absolute Jump – instruction to the reset handling routine. If the program neverenables an interrupt source, the Interrupt Vectors are not used, and regular programcode can be placed at these locations. This is also the case if the Reset Vector is in theApplication section while the Interrupt Vectors are in the Boot section or vice versa. Thecircuit diagram in Figure 25 shows the reset logic. Table 23 defines the electrical param-eters of the reset circuitry.

The I/O ports of the AVR are immediately reset to their initial state when a reset sourcegoes active. This does not require any clock source to be running.

After all reset sources have gone inactive, a delay counter is invoked, stretching theinternal reset. This allows the power to reach a stable level before normal operationstarts. The time-out period of the delay counter is defined by the user through the SUTand CKSEL Fuses. The different selections for the delay period are presented in “ClockSources” on page 40.

Reset Sources The ATmega640/1280/1281/2560/2561 has five sources of reset:

• Power-on Reset. The MCU is reset when the supply voltage is below the Power-on Reset threshold (VPOT).

• External Reset. The MCU is reset when a low level is present on the RESET pin for longer than the minimum pulse length.

• Watchdog Reset. The MCU is reset when the Watchdog Timer period expires and the Watchdog is enabled.

• Brown-out Reset. The MCU is reset when the supply voltage VCC is below the Brown-out Reset threshold (VBOT) and the Brown-out Detector is enabled.

• JTAG AVR Reset. The MCU is reset as long as there is a logic one in the Reset Register, one of the scan chains of the JTAG system. Refer to the section “IEEE 1149.1 (JTAG) Boundary-scan” on page 301 for details.

58 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 25. Reset Logic

Notes: 1. Values are guidelines only. Actual values are TBD.2. The Power-on Reset will not work unless the supply voltage has been below VPOT

(falling)

Power-on Reset A Power-on Reset (POR) pulse is generated by an On-chip detection circuit. The detec-tion level is defined in Table 23. The POR is activated whenever VCC is below thedetection level. The POR circuit can be used to trigger the start-up Reset, as well as todetect a failure in supply voltage.

A Power-on Reset (POR) circuit ensures that the device is reset from Power-on. Reach-ing the Power-on Reset threshold voltage invokes the delay counter, which determineshow long the device is kept in RESET after VCC rise. The RESET signal is activatedagain, without any delay, when VCC decreases below the detection level.

Table 23. Reset Characteristics(1)

Symbol Parameter Condition Min Typ Max Units

VPOT

Power-on Reset Threshold Voltage (rising)

TBD TBD TBD V

Power-on Reset Threshold Voltage (falling)(2) TBD TBD TBD V

VRST RESET Pin Threshold Voltage TBD TBD TBD V

tRSTMinimum pulse width on RESET Pin

TBD TBD TBD ns

MCU StatusRegister (MCUSR)

Brown-outReset CircuitBODLEVEL [2..0]

Delay Counters

CKSEL[3:0]

CKTIMEOUT

WD

RF

BO

RF

EX

TR

F

PO

RF

DATA BUS

ClockGenerator

SPIKEFILTER

Pull-up Resistor

JTR

F

JTAG ResetRegister

WatchdogOscillator

SUT[1:0]

Power-on ResetCircuit

59

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 26. MCU Start-up, RESET Tied to VCC

Figure 27. MCU Start-up, RESET Extended Externally

External Reset An External Reset is generated by a low level on the RESET pin. Reset pulses longerthan the minimum pulse width (see Table 23) will generate a reset, even if the clock isnot running. Shorter pulses are not guaranteed to generate a reset. When the appliedsignal reaches the Reset Threshold Voltage – VRST – on its positive edge, the delaycounter starts the MCU after the Time-out period – tTOUT – has expired.

Figure 28. External Reset During Operation

Brown-out Detection ATmega640/1280/1281/2560/2561 has an On-chip Brown-out Detection (BOD) circuitfor monitoring the VCC level during operation by comparing it to a fixed trigger level. Thetrigger level for the BOD can be selected by the BODLEVEL Fuses. The trigger level

V

RESET

TIME-OUT

INTERNALRESET

tTOUT

VPOT

VRST

CC

RESET

TIME-OUT

INTERNALRESET

tTOUT

VPOT

VRST

VCC

CC

60 ATmega640/1280/1281/2560/25612549A–AVR–03/05

has a hysteresis to ensure spike free Brown-out Detection. The hysteresis on the detec-tion level should be interpreted as VBOT+ = VBOT + VHYST/2 and VBOT- = VBOT - VHYST/2.

Note: 1. VBOT may be below nominal minimum operating voltage for some devices. Fordevices where this is the case, the device is tested down to VCC = VBOT during theproduction test. This guarantees that a Brown-Out Reset will occur before VCC dropsto a voltage where correct operation of the microcontroller is no longer guaranteed.The test is performed using BODLEVEL = 110 forATmega640/1280/1281/2560/2561 and BODLEVEL = 101 forATmega640/1280/2560/1L.

When the BOD is enabled, and VCC decreases to a value below the trigger level (VBOT-in Figure 29), the Brown-out Reset is immediately activated. When VCC increases abovethe trigger level (VBOT+ in Figure 29), the delay counter starts the MCU after the Time-out period tTOUT has expired.

The BOD circuit will only detect a drop in VCC if the voltage stays below the trigger levelfor longer than tBOD given in Table 23.

Figure 29. Brown-out Reset During Operation

Table 24. BODLEVEL Fuse Coding(1)

BODLEVEL 2..0 Fuses Min VBOT Typ VBOT Max VBOT Units

111 BOD Disabled

110 1.8

V101 2.7

100 4.3

011

Reserved010

001

000

Table 25. Brown-out Characteristics

Symbol Parameter Min Typ Max Units

VHYST Brown-out Detector Hysteresis 50 mV

tBOD Min Pulse Width on Brown-out Reset ns

VCC

RESET

TIME-OUT

INTERNALRESET

VBOT-VBOT+

tTOUT

61

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Watchdog Reset When the Watchdog times out, it will generate a short reset pulse of one CK cycle dura-tion. On the falling edge of this pulse, the delay timer starts counting the Time-out periodtTOUT. See “Watchdog Timer” on page 56. for details on operation of the WatchdogTimer.

Figure 30. Watchdog Reset During Operation

MCU Status Register – MCUSR

The MCU Status Register provides information on which reset source caused an MCUreset.

• Bit 4 – JTRF: JTAG Reset Flag

This bit is set if a reset is being caused by a logic one in the JTAG Reset Registerselected by the JTAG instruction AVR_RESET. This bit is reset by a Power-on Reset, orby writing a logic zero to the flag.

• Bit 3 – WDRF: Watchdog Reset Flag

This bit is set if a Watchdog Reset occurs. The bit is reset by a Power-on Reset, or bywriting a logic zero to the flag.

• Bit 2 – BORF: Brown-out Reset Flag

This bit is set if a Brown-out Reset occurs. The bit is reset by a Power-on Reset, or bywriting a logic zero to the flag.

• Bit 1 – EXTRF: External Reset Flag

This bit is set if an External Reset occurs. The bit is reset by a Power-on Reset, or bywriting a logic zero to the flag.

• Bit 0 – PORF: Power-on Reset Flag

This bit is set if a Power-on Reset occurs. The bit is reset only by writing a logic zero tothe flag.

To make use of the Reset Flags to identify a reset condition, the user should read andthen Reset the MCUSR as early as possible in the program. If the register is cleared

CK

CC

Bit 7 6 5 4 3 2 1 0

– – – JTRF WDRF BORF EXTRF PORF MCUSR

Read/Write R R R R/W R/W R/W R/W R/W

Initial Value 0 0 0 See Bit Description

62 ATmega640/1280/1281/2560/25612549A–AVR–03/05

before another reset occurs, the source of the reset can be found by examining theReset Flags.

Internal Voltage Reference

ATmega640/1280/1281/2560/2561 features an internal bandgap reference. This refer-ence is used for Brown-out Detection, and it can be used as an input to the AnalogComparator or the ADC.

Voltage Reference Enable Signals and Start-up Time

The voltage reference has a start-up time that may influence the way it should be used.The start-up time is given in Table 26. To save power, the reference is not always turnedon. The reference is on during the following situations:

1. When the BOD is enabled (by programming the BODLEVEL [2..0] Fuse).

2. When the bandgap reference is connected to the Analog Comparator (by setting the ACBG bit in ACSR).

3. When the ADC is enabled.

Thus, when the BOD is not enabled, after setting the ACBG bit or enabling the ADC, theuser must always allow the reference to start up before the output from the Analog Com-parator or ADC is used. To reduce power consumption in Power-down mode, the usercan avoid the three conditions above to ensure that the reference is turned off beforeentering Power-down mode.

Note: 1. Values are guidelines only. Actual values are TBD.

Table 26. Internal Voltage Reference Characteristics(1)

Symbol Parameter Condition Min Typ Max Units

VBG Bandgap reference voltage TBD TBD 1.1 TBD V

tBG Bandgap reference start-up time TBD 40 70 µs

IBGBandgap reference current consumption

TBD 10 TBD µA

63

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Watchdog Timer ATmega640/1280/1281/2560/2561 has an Enhanced Watchdog Timer (WDT). Themain features are:• Clocked from separate On-chip Oscillator• 3 Operating modes

– Interrupt– System Reset– Interrupt and System Reset

• Selectable Time-out period from 16ms to 8s• Possible Hardware fuse Watchdog always on (WDTON) for fail-safe mode

Figure 31. Watchdog Timer

The Watchdog Timer (WDT) is a timer counting cycles of a separate on-chip 128 kHzoscillator. The WDT gives an interrupt or a system reset when the counter reaches agiven time-out value. In normal operation mode, it is required that the system uses theWDR - Watchdog Timer Reset - instruction to restart the counter before the time-outvalue is reached. If the system doesn't restart the counter, an interrupt or system resetwill be issued.

In Interrupt mode, the WDT gives an interrupt when the timer expires. This interrupt canbe used to wake the device from sleep-modes, and also as a general system timer. Oneexample is to limit the maximum time allowed for certain operations, giving an interruptwhen the operation has run longer than expected. In System Reset mode, the WDTgives a reset when the timer expires. This is typically used to prevent system hang-up incase of runaway code. The third mode, Interrupt and System Reset mode, combines theother two modes by first giving an interrupt and then switch to System Reset mode. Thismode will for instance allow a safe shutdown by saving critical parameters before a sys-tem reset.

The Watchdog always on (WDTON) fuse, if programmed, will force the Watchdog Timerto System Reset mode. With the fuse programmed the System Reset mode bit (WDE)and Interrupt mode bit (WDIE) are locked to 1 and 0 respectively. To further ensure pro-gram security, alterations to the Watchdog set-up must follow timed sequences. Thesequence for clearing WDE and changing time-out configuration is as follows:

128kHzOSCILLATOR

OSC

/2K

OSC

/4K

OSC

/8K

OSC

/16K

OSC

/32K

OSC

/64K

OSC

/128

KO

SC/2

56K

OSC

/512

KO

SC/1

024K

WDP0 WDP1WDP2WDP3

WATCHDOGRESET

WDE

WDIF

WDIE

MCU RESET

INTERRUPT

64 ATmega640/1280/1281/2560/25612549A–AVR–03/05

1. In the same operation, write a logic one to the Watchdog change enable bit(WDCE) and WDE. A logic one must be written to WDE regardless of the previ-ous value of the WDE bit.

2. Within the next four clock cycles, write the WDE and Watchdog prescaler bits(WDP) as desired, but with the WDCE bit cleared. This must be done in oneoperation.

The following code example shows one assembly and one C function for turning off theWatchdog Timer. The example assumes that interrupts are controlled (e.g. by disablinginterrupts globally) so that no interrupts will occur during the execution of thesefunctions.

65

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: 1. The example code assumes that the part specific header file is included.

Note: If the Watchdog is accidentally enabled, for example by a runaway pointer orbrown-out condition, the device will be reset and the Watchdog Timer will stay enabled.If the code is not set up to handle the Watchdog, this might lead to an eternal loop oftime-out resets. To avoid this situation, the application software should always clear theWatchdog System Reset Flag (WDRF) and the WDE control bit in the initialisation rou-tine, even if the Watchdog is not in use.

Assembly Code Example(1)

WDT_off:

; Turn off global interrupt

cli

; Reset Watchdog Timer

wdr

; Clear WDRF in MCUSR

in r16, MCUSR

andi r16, (0xff & (0<<WDRF))

out MCUSR, r16

; Write logical one to WDCE and WDE

; Keep old prescaler setting to prevent unintentional time-out

in r16, WDTCSR

ori r16, (1<<WDCE) | (1<<WDE)

out WDTCSR, r16

; Turn off WDT

ldi r16, (0<<WDE)

out WDTCSR, r16

; Turn on global interrupt

sei

ret

C Code Example(1)

void WDT_off(void)

__disable_interrupt();

__watchdog_reset();

/* Clear WDRF in MCUSR */

MCUSR &= ~(1<<WDRF);

/* Write logical one to WDCE and WDE */

/* Keep old prescaler setting to prevent unintentional time-out */

WDTCSR |= (1<<WDCE) | (1<<WDE);

/* Turn off WDT */

WDTCSR = 0x00;

__enable_interrupt();

66 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The following code example shows one assembly and one C function for changing thetime-out value of the Watchdog Timer.

Note: 1. The example code assumes that the part specific header file is included.

Note: The Watchdog Timer should be reset before any change of the WDP bits, since achange in the WDP bits can result in a time-out when switching to a shorter time-outperiod.

Assembly Code Example(1)

WDT_Prescaler_Change:

; Turn off global interrupt

cli

; Reset Watchdog Timer

wdr

; Start timed sequence

in r16, WDTCSR

ori r16, (1<<WDCE) | (1<<WDE)

out WDTCSR, r16

; -- Got four cycles to set the new values from here -

; Set new prescaler(time-out) value = 64K cycles (~0.5 s)

ldi r16, (1<<WDE) | (1<<WDP2) | (1<<WDP0)

out WDTCSR, r16

; -- Finished setting new values, used 2 cycles -

; Turn on global interrupt

sei

ret

C Code Example(1)

void WDT_Prescaler_Change(void)

__disable_interrupt();

__watchdog_reset();

/* Start timed equence */

WDTCSR |= (1<<WDCE) | (1<<WDE);

/* Set new prescaler(time-out) value = 64K cycles (~0.5 s) */

WDTCSR = (1<<WDE) | (1<<WDP2) | (1<<WDP0);

__enable_interrupt();

67

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Watchdog Timer Control Register - WDTCSR

• Bit 7 - WDIF: Watchdog Interrupt Flag

This bit is set when a time-out occurs in the Watchdog Timer and the Watchdog Timer isconfigured for interrupt. WDIF is cleared by hardware when executing the correspondinginterrupt handling vector. Alternatively, WDIF is cleared by writing a logic one to the flag.When the I-bit in SREG and WDIE are set, the Watchdog Time-out Interrupt isexecuted.

• Bit 6 - WDIE: Watchdog Interrupt Enable

When this bit is written to one and the I-bit in the Status Register is set, the WatchdogInterrupt is enabled. If WDE is cleared in combination with this setting, the WatchdogTimer is in Interrupt Mode, and the corresponding interrupt is executed if time-out in theWatchdog Timer occurs.

If WDE is set, the Watchdog Timer is in Interrupt and System Reset Mode. The firsttime-out in the Watchdog Timer will set WDIF. Executing the corresponding interruptvector will clear WDIE and WDIF automatically by hardware (the Watchdog goes to Sys-tem Reset Mode). This is useful for keeping the Watchdog Timer security while usingthe interrupt. To stay in Interrupt and System Reset Mode, WDIE must be set after eachinterrupt. This should however not be done within the interrupt service routine itself, asthis might compromise the safety-function of the Watchdog System Reset mode. If theinterrupt is not executed before the next time-out, a System Reset will be applied.

• Bit 4 - WDCE: Watchdog Change Enable

This bit is used in timed sequences for changing WDE and prescaler bits. To clear theWDE bit, and/or change the prescaler bits, WDCE must be set.

Once written to one, hardware will clear WDCE after four clock cycles.

• Bit 3 - WDE: Watchdog System Reset Enable

WDE is overridden by WDRF in MCUSR. This means that WDE is always set whenWDRF is set. To clear WDE, WDRF must be cleared first. This feature ensures multipleresets during conditions causing failure, and a safe start-up after the failure.

• Bit 5, 2..0 - WDP3..0: Watchdog Timer Prescaler 3, 2, 1 and 0

The WDP3..0 bits determine the Watchdog Timer prescaling when the Watchdog Timeris running. The different prescaling values and their corresponding time-out periods areshown in Table 28 on page 68.

Bit 7 6 5 4 3 2 1 0

WDIF WDIE WDP3 WDCE WDE WDP2 WDP1 WDP0 WDTCSR

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 X 0 0 0

Table 27. Watchdog Timer Configuration

WDTON WDE WDIE Mode Action on Time-out

0 0 0 Stopped None

0 0 1 Interrupt Mode Interrupt

0 1 0 System Reset Mode Reset

0 1 1Interrupt and System Reset Mode

Interrupt, then go to System Reset Mode

1 x x System Reset Mode Reset

68 ATmega640/1280/1281/2560/25612549A–AVR–03/05

.

Table 28. Watchdog Timer Prescale Select

WDP3 WDP2 WDP1 WDP0Number of WDT Oscillator

CyclesTypical Time-out at

VCC = 5.0V

0 0 0 0 2K (2048) cycles 16 ms

0 0 0 1 4K (4096) cycles 32 ms

0 0 1 0 8K (8192) cycles 64 ms

0 0 1 1 16K (16384) cycles 0.125 s

0 1 0 0 32K (32768) cycles 0.25 s

0 1 0 1 64K (65536) cycles 0.5 s

0 1 1 0 128K (131072) cycles 1.0 s

0 1 1 1 256K (262144) cycles 2.0 s

1 0 0 0 512K (524288) cycles 4.0 s

1 0 0 1 1024K (1048576) cycles 8.0 s

1 0 1 0

Reserved

1 0 1 1

1 1 0 0

1 1 0 1

1 1 1 0

1 1 1 1

69

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Interrupts This section describes the specifics of the interrupt handling as performed inATmega640/1280/1281/2560/2561. For a general explanation of the AVR interrupt han-dling, refer to “Reset and Interrupt Handling” on page 15.

Interrupt Vectors in ATmega640/1280/1281/2560/2561

Table 29. Reset and Interrupt Vectors

VectorNo.

ProgramAddress(2) Source Interrupt Definition

1$0000(1) RESET External Pin, Power-on Reset, Brown-out Reset,

Watchdog Reset, and JTAG AVR Reset

2 $0002 INT0 External Interrupt Request 0

3 $0004 INT1 External Interrupt Request 1

4 $0006 INT2 External Interrupt Request 2

5 $0008 INT3 External Interrupt Request 3

6 $000A INT4 External Interrupt Request 4

7 $000C INT5 External Interrupt Request 5

8 $000E INT6 External Interrupt Request 6

9 $0010 INT7 External Interrupt Request 7

10 $0012 PCINT0 Pin Change Interrupt Request 0

11 $0014 PCINT1 Pin Change Interrupt Request 1

12 $0016(3) PCINT2 Pin Change Interrupt Request 2

13 $0018 WDT Watchdog Time-out Interrupt

14 $001A TIMER2 COMPA Timer/Counter2 Compare Match A

15 $001C TIMER2 COMPB Timer/Counter2 Compare Match B

16 $001E TIMER2 OVF Timer/Counter2 Overflow

17 $0020 TIMER1 CAPT Timer/Counter1 Capture Event

18 $0022 TIMER1 COMPA Timer/Counter1 Compare Match A

19 $0024 TIMER1 COMPB Timer/Counter1 Compare Match B

20 $0026 TIMER1 COMPC Timer/Counter1 Compare Match C

21 $0028 TIMER1 OVF Timer/Counter1 Overflow

22 $002A TIMER0 COMPA Timer/Counter0 Compare Match A

23 $002C TIMER0 COMPB Timer/Counter0 Compare match B

24 $002E TIMER0 OVF Timer/Counter0 Overflow

25 $0030 SPI, STC SPI Serial Transfer Complete

26 $0032 USART0 RX USART0 Rx Complete

27 $0034 USART0 UDRE USART0 Data Register Empty

28 $0036 USART0 TX USART0 Tx Complete

29 $0038 ANALOG COMP Analog Comparator

30 $003A ADC ADC Conversion Complete

31 $003C EE READY EEPROM Ready

70 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Notes: 1. When the BOOTRST Fuse is programmed, the device will jump to the Boot Loaderaddress at reset, see “Memory Programming” on page 335.

2. When the IVSEL bit in MCUCR is set, Interrupt Vectors will be moved to the start ofthe Boot Flash Section. The address of each Interrupt Vector will then be the addressin this table added to the start address of the Boot Flash Section.

3. Only available in ATmega640/1280/2560

Table 30 shows reset and Interrupt Vectors placement for the various combinations ofBOOTRST and IVSEL settings. If the program never enables an interrupt source, theInterrupt Vectors are not used, and regular program code can be placed at these loca-tions. This is also the case if the Reset Vector is in the Application section while theInterrupt Vectors are in the Boot section or vice versa.

32 $003E TIMER3 CAPT Timer/Counter3 Capture Event

33 $0040 TIMER3 COMPA Timer/Counter3 Compare Match A

34 $0042 TIMER3 COMPB Timer/Counter3 Compare Match B

35 $0044 TIMER3 COMPC Timer/Counter3 Compare Match C

36 $0046 TIMER3 OVF Timer/Counter3 Overflow

37 $0048 USART1 RX USART1 Rx Complete

38 $004A USART1 UDRE USART1 Data Register Empty

39 $004C USART1 TX USART1 Tx Complete

40 $004E TWI 2-wire Serial Interface

41 $0050 SPM READY Store Program Memory Ready

42 $0052(3) TIMER4 CAPT Timer/Counter4 Capture Event

43 $0054 TIMER4 COMPA Timer/Counter4 Compare Match A

44 $0056 TIMER4 COMPB Timer/Counter4 Compare Match B

45 $0058 TIMER4 COMPC Timer/Counter4 Compare Match C

46 $005A TIMER4 OVF Timer/Counter4 Overflow

47 $005C(3) TIMER5 CAPT Timer/Counter5 Capture Event

48 $005E TIMER5 COMPA Timer/Counter5 Compare Match A

49 $0060 TIMER5 COMPB Timer/Counter5 Compare Match B

50 $0062 TIMER5 COMPC Timer/Counter5 Compare Match C

51 $0064 TIMER5 OVF Timer/Counter5 Overflow

52 $0066(3) USART2 RX USART2 Rx Complete

53 $0068(3) USART2 UDRE USART2 Data Register Empty

54 $006A(3) USART2 TX USART2 Tx Complete

55 $006C(3) USART3 RX USART3 Rx Complete

56 $006E(3)) USART3 UDRE USART3 Data Register Empty

57 $0070(3) USART3 TX USART3 Tx Complete

Table 29. Reset and Interrupt Vectors (Continued)

VectorNo.

ProgramAddress(2) Source Interrupt Definition

71

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: 1. The Boot Reset Address is shown in Table 139 on page 330 through Table 147 onpage 334. For the BOOTRST Fuse “1” means unprogrammed while “0” meansprogrammed.

The most typical and general program setup for the Reset and Interrupt VectorAddresses in ATmega640/1280/1281/2560/2561 is:

Table 30. Reset and Interrupt Vectors Placement(1)

BOOTRST IVSEL Reset Address Interrupt Vectors Start Address

1 0 0x0000 0x0002

1 1 0x0000 Boot Reset Address + 0x0002

0 0 Boot Reset Address 0x0002

0 1 Boot Reset Address Boot Reset Address + 0x0002

Address Labels Code Comments

0x0000 jmp RESET ; Reset Handler

0x0002 jmp INT0 ; IRQ0 Handler

0x0004 jmp INT1 ; IRQ1 Handler

0x0006 jmp INT2 ; IRQ2 Handler

0x0008 jmp INT3 ; IRQ3 Handler

0x000A jmp INT4 ; IRQ4 Handler

0x000C jmp INT5 ; IRQ5 Handler

0x000E jmp INT6 ; IRQ6 Handler

0x0010 jmp INT7 ; IRQ7 Handler

0x0012 jmp PCINT0 ; PCINT0 Handler

0x0014 jmp PCINT1 ; PCINT1 Handler

0x0016 jmp PCINT2 ; PCINT2 Handler

0X0018 jmp WDT ; Watchdog Timeout Handler

0x001A jmp TIM2_COMPA ; Timer2 CompareA Handler

0x001C jmp TIM2_COMPB ; Timer2 CompareB Handler

0x001E jmp TIM2_OVF ; Timer2 Overflow Handler

0x0020 jmp TIM1_CAPT ; Timer1 Capture Handler

0x0022 jmp TIM1_COMPA ; Timer1 CompareA Handler

0x0024 jmp TIM1_COMPB ; Timer1 CompareB Handler

0x0026 jmp TIM1_COMPC ; Timer1 CompareC Handler

0x0028 jmp TIM1_OVF ; Timer1 Overflow Handler

0x002A jmp TIM0_COMPA ; Timer0 CompareA Handler

0x002C jmp TIM0_COMPB ; Timer0 CompareB Handler

0x002E jmp TIM0_OVF ; Timer0 Overflow Handler

0x0030 jmp SPI_STC ; SPI Transfer Complete Handler

0x0032 jmp USART0_RXC ; USART0 RX Complete Handler

0x0034 jmp USART0_UDRE ; USART0,UDR Empty Handler

0x0036 jmp USART0_TXC ; USART0 TX Complete Handler

0x0038 jmp ANA_COMP ; Analog Comparator Handler

0x003A jmp ADC ; ADC Conversion Complete Handler

0x003C jmp EE_RDY ; EEPROM Ready Handler

0x003E jmp TIM3_CAPT ; Timer3 Capture Handler

0x0040 jmp TIM3_COMPA ; Timer3 CompareA Handler

0x0042 jmp TIM3_COMPB ; Timer3 CompareB Handler

0x0044 jmp TIM3_COMPC ; Timer3 CompareC Handler

0x0046 jmp TIM3_OVF ; Timer3 Overflow Handler

0x0048 jmp USART1_RXC ; USART1 RX Complete Handler

72 ATmega640/1280/1281/2560/25612549A–AVR–03/05

When the BOOTRST Fuse is unprogrammed, the Boot section size set to 8K bytes andthe IVSEL bit in the MCUCR Register is set before any interrupts are enabled, the mosttypical and general program setup for the Reset and Interrupt Vector Addresses is:

Address Labels Code Comments

0x00000 RESET: ldi r16,high(RAMEND) ; Main program start

0x00001 out SPH,r16 ; Set Stack Pointer to top of RAM

0x00002 ldi r16,low(RAMEND)

0x00003 out SPL,r160x00004 sei ; Enable interrupts

0x00005 <instr> xxx

;

.org 0x1F002

0x1F002 jmp EXT_INT0 ; IRQ0 Handler

0x1F004 jmp EXT_INT1 ; IRQ1 Handler

... ... ... ;

0x1FO70 jmp USART3_TXC ; USART3 TX Complete Handler

When the BOOTRST Fuse is programmed and the Boot section size set to 8K bytes, themost typical and general program setup for the Reset and Interrupt Vector Addresses is:

0x004A jmp USART1_UDRE ; USART1,UDR Empty Handler

0x004C jmp USART1_TXC ; USART1 TX Complete Handler

0x004E jmp TWI ; 2-wire Serial Handler

0x0050 jmp SPM_RDY ; SPM Ready Handler

0x0052 jmp TIM4_CAPT ; Timer4 Capture Handler

0x0054 jmp TIM4_COMPA ; Timer4 CompareA Handler

0x0056 jmp TIM4_COMPB ; Timer4 CompareB Handler

0x0058 jmp TIM4_COMPC ; Timer4 CompareC Handler

0x005A jmp TIM4_OVF ; Timer4 Overflow Handler

0x005C jmp TIM5_CAPT ; Timer5 Capture Handler

0x005E jmp TIM5_COMPA ; Timer5 CompareA Handler

0x0060 jmp TIM5_COMPB ; Timer5 CompareB Handler

0x0062 jmp TIM5_COMPC ; Timer5 CompareC Handler

0x0064 jmp TIM5_OVF ; Timer5 Overflow Handler

0x0066 jmp USART2_RXC ; USART2 RX Complete Handler

0x0068 jmp USART2_UDRE ; USART2,UDR Empty Handler

0x006A jmp USART2_TXC ; USART2 TX Complete Handler

0x006C jmp USART3_RXC ; USART3 RX Complete Handler

0x006E jmp USART3_UDRE ; USART3,UDR Empty Handler

0x0070 jmp USART3_TXC ; USART3 TX Complete Handler

;

0x0072 RESET: ldi r16, high(RAMEND) ; Main program start

0x0073 out SPH,r16 ; Set Stack Pointer to top of RAM

0x0074 ldi r16, low(RAMEND)

0x0075 out SPL,r16

0x0076 sei ; Enable interrupts

0x0077 <instr> xxx

... ... ... ...

73

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Address Labels Code Comments

.org 0x0002

0x00002 jmp EXT_INT0 ; IRQ0 Handler

0x00004 jmp EXT_INT1 ; IRQ1 Handler

... ... ... ;

0x00070 jmp USART3_TXC ; USART3 TX Complete Handler

;

.org 0x1F0000x1F000 RESET: ldi r16,high(RAMEND) ; Main program start

0x1F001 out SPH,r16 ; Set Stack Pointer to top of RAM

0x1F002 ldi r16,low(RAMEND)

0x1F003 out SPL,r160x1F004 sei ; Enable interrupts

0x1F005 <instr> xxx

When the BOOTRST Fuse is programmed, the Boot section size set to 8K bytes and theIVSEL bit in the MCUCR Register is set before any interrupts are enabled, the most typ-ical and general program setup for the Reset and Interrupt Vector Addresses is:

Address Labels Code Comments

;

.org 0x1F0000x1F000 jmp RESET ; Reset handler0x1F002 jmp EXT_INT0 ; IRQ0 Handler

0x1F004 jmp EXT_INT1 ; IRQ1 Handler

... ... ... ;

0x1F070 jmp USART3_TXC ; USART3 TX Complete Handler

;

0x1F072 RESET: ldi r16,high(RAMEND) ; Main program start

0x1F073 out SPH,r16 ; Set Stack Pointer to top of RAM

0x1F074 ldi r16,low(RAMEND)

0x1F075 out SPL,r160x1F076 sei ; Enable interrupts

0x1FO77 <instr> xxx

Moving Interrupts Between Application and Boot Space

The General Interrupt Control Register controls the placement of the Interrupt Vectortable.

MCU Control Register – MCUCR

• Bit 1 – IVSEL: Interrupt Vector Select

When the IVSEL bit is cleared (zero), the Interrupt Vectors are placed at the start of theFlash memory. When this bit is set (one), the Interrupt Vectors are moved to the begin-ning of the Boot Loader section of the Flash. The actual address of the start of the BootFlash Section is determined by the BOOTSZ Fuses. Refer to the section “Memory Pro-gramming” on page 335 for details. To avoid unintentional changes of Interrupt Vectortables, a special write procedure must be followed to change the IVSEL bit:

Bit 7 6 5 4 3 2 1 0

JTD – – PUD – – IVSEL IVCE MCUCR

Read/Write R/W R R R/W R R R/W R/W

Initial Value 0 0 0 0 0 0 0 0

74 ATmega640/1280/1281/2560/25612549A–AVR–03/05

1. Write the Interrupt Vector Change Enable (IVCE) bit to one.

2. Within four cycles, write the desired value to IVSEL while writing a zero toIVCE.

Interrupts will automatically be disabled while this sequence is executed. Interrupts aredisabled in the cycle IVCE is set, and they remain disabled until after the instruction fol-lowing the write to IVSEL. If IVSEL is not written, interrupts remain disabled for fourcycles. The I-bit in the Status Register is unaffected by the automatic disabling.Note: If Interrupt Vectors are placed in the Boot Loader section and Boot Lock bit BLB02 is pro-

grammed, interrupts are disabled while executing from the Application section. IfInterrupt Vectors are placed in the Application section and Boot Lock bit BLB12 is pro-gramed, interrupts are disabled while executing from the Boot Loader section. Refer tothe section “Memory Programming” on page 335 for details on Boot Lock bits.

• Bit 0 – IVCE: Interrupt Vector Change Enable

The IVCE bit must be written to logic one to enable change of the IVSEL bit. IVCE iscleared by hardware four cycles after it is written or when IVSEL is written. Setting theIVCE bit will disable interrupts, as explained in the IVSEL description above. See CodeExample below.

Assembly Code Example

Move_interrupts:

; Enable change of Interrupt Vectors

ldi r16, (1<<IVCE)

out MCUCR, r16

; Move interrupts to Boot Flash section

ldi r16, (1<<IVSEL)

out MCUCR, r16

ret

C Code Example

void Move_interrupts(void)

/* Enable change of Interrupt Vectors */

MCUCR = (1<<IVCE);

/* Move interrupts to Boot Flash section */

MCUCR = (1<<IVSEL);

75

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

External Interrupts The External Interrupts are triggered by the INT7:0 pin or any of the PCINT23..0 pins.Observe that, if enabled, the interrupts will trigger even if the INT7:0 or PCINT23..0 pinsare configured as outputs. This feature provides a way of generating a softwareinterrupt.

The Pin change interrupt PCI2 will trigger if any enabled PCINT23:16 pin toggles, Pinchange interrupt PCI1 if any enabled PCINT15:8 toggles and Pin change interruptsPCI0 will trigger if any enabled PCINT7..0 pin toggles. PCMSK2, PCMSK1 andPCMSK0 Registers control which pins contribute to the pin change interrupts. Pinchange interrupts on PCINT23 ..0 are detected asynchronously. This implies that theseinterrupts can be used for waking the part also from sleep modes other than Idle mode.

The External Interrupts can be triggered by a falling or rising edge or a low level. This isset up as indicated in the specification for the External Interrupt Control Registers –EICRA (INT3:0) and EICRB (INT7:4). When the external interrupt is enabled and is con-figured as level triggered, the interrupt will trigger as long as the pin is held low. Notethat recognition of falling or rising edge interrupts on INT7:4 requires the presence of anI/O clock, described in “Clock Systems and their Distribution” on page 39. Low levelinterrupts and the edge interrupt on INT3:0 are detected asynchronously. This impliesthat these interrupts can be used for waking the part also from sleep modes other thanIdle mode. The I/O clock is halted in all sleep modes except Idle mode.

Note that if a level triggered interrupt is used for wake-up from Power-down, therequired level must be held long enough for the MCU to complete the wake-up to triggerthe level interrupt. If the level disappears before the end of the Start-up Time, the MCUwill still wake up, but no interrupt will be generated. The start-up time is defined by theSUT and CKSEL Fuses as described in “System Clock and Clock Options” on page 39.

Pin Change Interrupt Timing

An example of timing of a pin change interrupt is shown in Figure 32.

Figure 32.

clk

PCINT(n)

pin_lat

pin_sync

pcint_in_(n)

pcint_syn

pcint_setflag

PCIF

PCINT(0)

pin_syncpcint_syn

pin_latD Q

LE

pcint_setflagPCIF

clk

clkPCINT(0) in PCMSK(x)

pcint_in_(0) 0

x

76 ATmega640/1280/1281/2560/25612549A–AVR–03/05

External Interrupt Control Register A – EICRA

The External Interrupt Control Register A contains control bits for interrupt sensecontrol.

• Bits 7..0 – ISC31, ISC30 – ISC00, ISC00: External Interrupt 3 - 0 Sense Control Bits

The External Interrupts 3 - 0 are activated by the external pins INT3:0 if the SREG I-flagand the corresponding interrupt mask in the EIMSK is set. The level and edges on theexternal pins that activate the interrupts are defined in Table 31. Edges on INT3..INT0are registered asynchronously. Pulses on INT3:0 pins wider than the minimum pulsewidth given in Table 32 will generate an interrupt. Shorter pulses are not guaranteed togenerate an interrupt. If low level interrupt is selected, the low level must be held untilthe completion of the currently executing instruction to generate an interrupt. If enabled,a level triggered interrupt will generate an interrupt request as long as the pin is heldlow. When changing the ISCn bit, an interrupt can occur. Therefore, it is recommendedto first disable INTn by clearing its Interrupt Enable bit in the EIMSK Register. Then, theISCn bit can be changed. Finally, the INTn interrupt flag should be cleared by writing alogical one to its Interrupt Flag bit (INTFn) in the EIFR Register before the interrupt is re-enabled.

Note: 1. n = 3, 2, 1or 0.When changing the ISCn1/ISCn0 bits, the interrupt must be disabled by clearing itsInterrupt Enable bit in the EIMSK Register. Otherwise an interrupt can occur whenthe bits are changed.

External Interrupt Control Register B – EICRB

• Bits 7..0 – ISC71, ISC70 - ISC41, ISC40: External Interrupt 7 - 4 Sense Control Bits

The External Interrupts 7 - 4 are activated by the external pins INT7:4 if the SREG I-flagand the corresponding interrupt mask in the EIMSK is set. The level and edges on the

Bit 7 6 5 4 3 2 1 0

ISC31 ISC30 ISC21 ISC20 ISC11 ISC10 ISC01 ISC00 EICRA

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 31. Interrupt Sense Control(1)

ISCn1 ISCn0 Description

0 0 The low level of INTn generates an interrupt request.

0 1 Any edge of INTn generates asynchronously an interrupt request.

1 0 The falling edge of INTn generates asynchronously an interrupt request.

1 1 The rising edge of INTn generates asynchronously an interrupt request.

Table 32. Asynchronous External Interrupt Characteristics

Symbol Parameter Condition Min Typ Max Units

tINTMinimum pulse width for asynchronous external interrupt

50 ns

Bit 7 6 5 4 3 2 1 0

ISC71 ISC70 ISC61 ISC60 ISC51 ISC50 ISC41 ISC40 EICRB

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

77

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

external pins that activate the interrupts are defined in Table 33. The value on theINT7:4 pins are sampled before detecting edges. If edge or toggle interrupt is selected,pulses that last longer than one clock period will generate an interrupt. Shorter pulsesare not guaranteed to generate an interrupt. Observe that CPU clock frequency can belower than the XTAL frequency if the XTAL divider is enabled. If low level interrupt isselected, the low level must be held until the completion of the currently executinginstruction to generate an interrupt. If enabled, a level triggered interrupt will generate aninterrupt request as long as the pin is held low.

Note: 1. n = 7, 6, 5 or 4.When changing the ISCn1/ISCn0 bits, the interrupt must be disabled by clearing itsInterrupt Enable bit in the EIMSK Register. Otherwise an interrupt can occur whenthe bits are changed.

External Interrupt Mask Register – EIMSK

• Bits 7..0 – INT7 – INT0: External Interrupt Request 7 - 0 Enable

When an INT7 – INT0 bit is written to one and the I-bit in the Status Register (SREG) isset (one), the corresponding external pin interrupt is enabled. The Interrupt Sense Con-trol bits in the External Interrupt Control Registers – EICRA and EICRB – defineswhether the external interrupt is activated on rising or falling edge or level sensed. Activ-ity on any of these pins will trigger an interrupt request even if the pin is enabled as anoutput. This provides a way of generating a software interrupt.

External Interrupt Flag Register – EIFR

• Bits 7..0 – INTF7 - INTF0: External Interrupt Flags 7 - 0

When an edge or logic change on the INT7:0 pin triggers an interrupt request, INTF7:0becomes set (one). If the I-bit in SREG and the corresponding interrupt enable bit,INT7:0 in EIMSK, are set (one), the MCU will jump to the interrupt vector. The flag iscleared when the interrupt routine is executed. Alternatively, the flag can be cleared bywriting a logical one to it. These flags are always cleared when INT7:0 are configured aslevel interrupt. Note that when entering sleep mode with the INT3:0 interrupts disabled,the input buffers on these pins will be disabled. This may cause a logic change in inter-

Table 33. Interrupt Sense Control(1)

ISCn1 ISCn0 Description

0 0 The low level of INTn generates an interrupt request.

0 1 Any logical change on INTn generates an interrupt request

1 0The falling edge between two samples of INTn generates an interrupt request.

1 1The rising edge between two samples of INTn generates an interrupt request.

Bit 7 6 5 4 3 2 1 0

INT7 INT6 INT5 INT4 INT3 INT2 INT1 IINT0 EIMSK

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

INTF7 INTF6 INTF5 INTF4 INTF3 INTF2 INTF1 IINTF0 EIFR

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

78 ATmega640/1280/1281/2560/25612549A–AVR–03/05

nal signals which will set the INTF3:0 flags. See “Digital Input Enable and Sleep Modes”on page 85 for more information.

Pin Change Interrupt Control Register - PCICR

• Bit 2 – PCIE2: Pin Change Interrupt Enable 1

When the PCIE2 bit is set (one) and the I-bit in the Status Register (SREG) is set (one),pin change interrupt 2 is enabled. Any change on any enabled PCINT23..16 pin willcause an interrupt. The corresponding interrupt of Pin Change Interrupt Request is exe-cuted from the PCI2 Interrupt Vector. PCINT23..16 pins are enabled individually by thePCMSK2 Register.

• Bit 1 – PCIE1: Pin Change Interrupt Enable 1

When the PCIE1 bit is set (one) and the I-bit in the Status Register (SREG) is set (one),pin change interrupt 1 is enabled. Any change on any enabled PCINT15..8 pin willcause an interrupt. The corresponding interrupt of Pin Change Interrupt Request is exe-cuted from the PCI1 Interrupt Vector. PCINT15..8 pins are enabled individually by thePCMSK1 Register.

• Bit 0 – PCIE0: Pin Change Interrupt Enable 0

When the PCIE0 bit is set (one) and the I-bit in the Status Register (SREG) is set (one),pin change interrupt 0 is enabled. Any change on any enabled PCINT7..0 pin will causean interrupt. The corresponding interrupt of Pin Change Interrupt Request is executedfrom the PCI0 Interrupt Vector. PCINT7..0 pins are enabled individually by the PCMSK0Register.

Pin Change Interrupt Flag Register – PCIFR

• Bit 2 – PCIF2: Pin Change Interrupt Flag 1

When a logic change on any PCINT23..16 pin triggers an interrupt request, PCIF2becomes set (one). If the I-bit in SREG and the PCIE2 bit in EIMSK are set (one), theMCU will jump to the corresponding Interrupt Vector. The flag is cleared when the inter-rupt routine is executed. Alternatively, the flag can be cleared by writing a logical one toit.

• Bit 1 – PCIF1: Pin Change Interrupt Flag 1

When a logic change on any PCINT15..8 pin triggers an interrupt request, PCIF1becomes set (one). If the I-bit in SREG and the PCIE1 bit in EIMSK are set (one), theMCU will jump to the corresponding Interrupt Vector. The flag is cleared when the inter-rupt routine is executed. Alternatively, the flag can be cleared by writing a logical one toit.

Bit 7 6 5 4 3 2 1 0

– – – PCIE2 PCIE1 PCIE0 PCICR

Read/Write R R R R R R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – – PCIF2 PCIF1 PCIF0 PCIFR

Read/Write R R R R R R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

79

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• Bit 0 – PCIF0: Pin Change Interrupt Flag 0

When a logic change on any PCINT7..0 pin triggers an interrupt request, PCIF0becomes set (one). If the I-bit in SREG and the PCIE0 bit in EIMSK are set (one), theMCU will jump to the corresponding Interrupt Vector. The flag is cleared when the inter-rupt routine is executed. Alternatively, the flag can be cleared by writing a logical one toit.

Pin Change Mask Register 2 – PCMSK2

• Bit 7..0 – PCINT23..16: Pin Change Enable Mask 23..16

Each PCINT23..16-bit selects whether pin change interrupt is enabled on the corre-sponding I/O pin. If PCINT23..16 is set and the PCIE2 bit in PCICR is set, pin changeinterrupt is enabled on the corresponding I/O pin. If PCINT23..16 is cleared, pin changeinterrupt on the corresponding I/O pin is disabled.

Pin Change Mask Register 1 – PCMSK1

• Bit 7..0 – PCINT15..8: Pin Change Enable Mask 15..8

Each PCINT15..8-bit selects whether pin change interrupt is enabled on the correspond-ing I/O pin. If PCINT15..8 is set and the PCIE1 bit in EIMSK is set, pin change interruptis enabled on the corresponding I/O pin. If PCINT15..8 is cleared, pin change interrupton the corresponding I/O pin is disabled.

Pin Change Mask Register 0 – PCMSK0

• Bit 7..0 – PCINT7..0: Pin Change Enable Mask 7..0

Each PCINT7..0 bit selects whether pin change interrupt is enabled on the correspond-ing I/O pin. If PCINT7..0 is set and the PCIE0 bit in PCICR is set, pin change interrupt isenabled on the corresponding I/O pin. If PCINT7..0 is cleared, pin change interrupt onthe corresponding I/O pin is disabled.

Bit 7 6 5 4 3 2 1 0

PCINT23 PCINT22 PCINT21 PCINT20 PCINT19 PCINT18 PCINT17 PCINT16 PCMSK2

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PCINT15 PCINT14 PCINT13 PCINT12 PCINT11 PCINT10 PCINT9 PCINT8 PCMSK1

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PCINT7 PCINT6 PCINT5 PCINT4 PCINT3 PCINT2 PCINT1 PCINT0 PCMSK0

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

80 ATmega640/1280/1281/2560/25612549A–AVR–03/05

81

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

I/O-Ports

Introduction All AVR ports have true Read-Modify-Write functionality when used as general digitalI/O ports. This means that the direction of one port pin can be changed without uninten-tionally changing the direction of any other pin with the SBI and CBI instructions. Thesame applies when changing drive value (if configured as output) or enabling/disablingof pull-up resistors (if configured as input). Each output buffer has symmetrical drivecharacteristics with both high sink and source capability. The pin driver is strong enoughto drive LED displays directly. All port pins have individually selectable pull-up resistorswith a supply-voltage invariant resistance. All I/O pins have protection diodes to bothVCC and Ground as indicated in Figure 33. Refer to “Electrical Characteristics” on page367 for a complete list of parameters.

Figure 33. I/O Pin Equivalent Schematic

All registers and bit references in this section are written in general form. A lower case“x” represents the numbering letter for the port, and a lower case “n” represents the bitnumber. However, when using the register or bit defines in a program, the precise formmust be used. For example, PORTB3 for bit no. 3 in Port B, here documented generallyas PORTxn. The physical I/O Registers and bit locations are listed in “Register Descrip-tion for I/O-Ports” on page 112.

Three I/O memory address locations are allocated for each port, one each for the DataRegister – PORTx, Data Direction Register – DDRx, and the Port Input Pins – PINx. ThePort Input Pins I/O location is read only, while the Data Register and the Data DirectionRegister are read/write. However, writing a logic one to a bit in the PINx Register, willresult in a toggle in the corresponding bit in the Data Register. In addition, the Pull-upDisable – PUD bit in MCUCR disables the pull-up function for all pins in all ports whenset.

Using the I/O port as General Digital I/O is described in “Ports as General Digital I/O” onpage 82. Most port pins are multiplexed with alternate functions for the peripheral fea-tures on the device. How each alternate function interferes with the port pin is describedin “Alternate Port Functions” on page 86. Refer to the individual module sections for afull description of the alternate functions.

Cpin

Logic

Rpu

See Figure"General Digital I/O" for

Details

Pxn

82 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note that enabling the alternate function of some of the port pins does not affect the useof the other pins in the port as general digital I/O.

Ports as General Digital I/O

The ports are bi-directional I/O ports with optional internal pull-ups. Figure 34 shows afunctional description of one I/O-port pin, here generically called Pxn.

Figure 34. General Digital I/O(1)

Note: 1. WRx, WPx, WDx, RRx, RPx, and RDx are common to all pins within the same port.clkI/O, SLEEP, and PUD are common to all ports.

Configuring the Pin Each port pin consists of three register bits: DDxn, PORTxn, and PINxn. As shown in“Register Description for I/O-Ports” on page 112, the DDxn bits are accessed at theDDRx I/O address, the PORTxn bits at the PORTx I/O address, and the PINxn bits atthe PINx I/O address.

The DDxn bit in the DDRx Register selects the direction of this pin. If DDxn is writtenlogic one, Pxn is configured as an output pin. If DDxn is written logic zero, Pxn is config-ured as an input pin.

If PORTxn is written logic one when the pin is configured as an input pin, the pull-upresistor is activated. To switch the pull-up resistor off, PORTxn has to be written logiczero or the pin has to be configured as an output pin. The port pins are tri-stated whenreset condition becomes active, even if no clocks are running.

If PORTxn is written logic one when the pin is configured as an output pin, the port pin isdriven high (one). If PORTxn is written logic zero when the pin is configured as an out-put pin, the port pin is driven low (zero).

clk

RPx

RRx

RDx

WDx

PUD

SYNCHRONIZER

WDx: WRITE DDRx

WRx: WRITE PORTxRRx: READ PORTx REGISTERRPx: READ PORTx PIN

PUD: PULLUP DISABLE

clkI/O: I/O CLOCK

RDx: READ DDRx

D

L

Q

Q

RESET

RESET

Q

QD

Q

Q D

CLR

PORTxn

Q

Q D

CLR

DDxn

PINxn

DAT

A B

US

SLEEP

SLEEP: SLEEP CONTROL

Pxn

I/O

WPx

0

1

WRx

WPx: WRITE PINx REGISTER

83

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Toggling the Pin Writing a logic one to PINxn toggles the value of PORTxn, independent on the value ofDDRxn. Note that the SBI instruction can be used to toggle one single bit in a port.

Switching Between Input and Output

When switching between tri-state (DDxn, PORTxn = 0b00) and output high (DDxn,PORTxn = 0b11), an intermediate state with either pull-up enabled DDxn, PORTxn =0b01) or output low (DDxn, PORTxn = 0b10) must occur. Normally, the pull-upenabled state is fully acceptable, as a high-impedant environment will not notice the dif-ference between a strong high driver and a pull-up. If this is not the case, the PUD bit inthe MCUCR Register can be set to disable all pull-ups in all ports.

Switching between input with pull-up and output low generates the same problem. Theuser must use either the tri-state (DDxn, PORTxn = 0b00) or the output high state(DDxn, PORTxn = 0b11) as an intermediate step.

Table 34 summarizes the control signals for the pin value.

Reading the Pin Value Independent of the setting of Data Direction bit DDxn, the port pin can be read throughthe PINxn Register bit. As shown in Figure 34, the PINxn Register bit and the precedinglatch constitute a synchronizer. This is needed to avoid metastability if the physical pinchanges value near the edge of the internal clock, but it also introduces a delay. Figure35 shows a timing diagram of the synchronization when reading an externally appliedpin value. The maximum and minimum propagation delays are denoted tpd,max and tpd,minrespectively.

Figure 35. Synchronization when Reading an Externally Applied Pin value

Table 34. Port Pin Configurations

DDxn PORTxnPUD

(in MCUCR) I/O Pull-up Comment

0 0 X Input No Tri-state (Hi-Z)

0 1 0 Input YesPxn will source current if ext. pulled low.

0 1 1 Input No Tri-state (Hi-Z)

1 0 X Output No Output Low (Sink)

1 1 X Output No Output High (Source)

XXX in r17, PINx

0x00 0xFF

INSTRUCTIONS

SYNC LATCH

PINxn

r17

XXX

SYSTEM CLK

tpd, max

tpd, min

84 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Consider the clock period starting shortly after the first falling edge of the system clock.The latch is closed when the clock is low, and goes transparent when the clock is high,as indicated by the shaded region of the “SYNC LATCH” signal. The signal value islatched when the system clock goes low. It is clocked into the PINxn Register at the suc-ceeding positive clock edge. As indicated by the two arrows tpd,max and tpd,min, asingle signal transition on the pin will be delayed between ½ and 1½ system clockperiod depending upon the time of assertion.

When reading back a software assigned pin value, a nop instruction must be inserted asindicated in Figure 36. The out instruction sets the “SYNC LATCH” signal at the positiveedge of the clock. In this case, the delay tpd through the synchronizer is 1 system clockperiod.

Figure 36. Synchronization when Reading a Software Assigned Pin Value

The following code example shows how to set port B pins 0 and 1 high, 2 and 3 low, anddefine the port pins from 4 to 7 as input with pull-ups assigned to port pins 6 and 7. Theresulting pin values are read back again, but as previously discussed, a nop instructionis included to be able to read back the value recently assigned to some of the pins.

out PORTx, r16 nop in r17, PINx

0xFF

0x00 0xFF

SYSTEM CLK

r16

INSTRUCTIONS

SYNC LATCH

PINxn

r17tpd

85

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: 1. For the assembly program, two temporary registers are used to minimize the timefrom pull-ups are set on pins 0, 1, 6, and 7, until the direction bits are correctly set,defining bit 2 and 3 as low and redefining bits 0 and 1 as strong high drivers.

Digital Input Enable and Sleep Modes

As shown in Figure 34, the digital input signal can be clamped to ground at the input ofthe schmitt-trigger. The signal denoted SLEEP in the figure, is set by the MCU SleepController in Power-down mode, Power-save mode, and Standby mode to avoid highpower consumption if some input signals are left floating, or have an analog signal levelclose to VCC/2.

SLEEP is overridden for port pins enabled as external interrupt pins. If the external inter-rupt request is not enabled, SLEEP is active also for these pins. SLEEP is alsooverridden by various other alternate functions as described in “Alternate Port Func-tions” on page 86.

If a logic high level (“one”) is present on an asynchronous external interrupt pin config-ured as “Interrupt on Rising Edge, Falling Edge, or Any Logic Change on Pin” while theexternal interrupt is not enabled, the corresponding External Interrupt Flag will be setwhen resuming from the above mentioned Sleep mode, as the clamping in these sleepmode produces the requested logic change.

Unconnected Pins If some pins are unused, it is recommended to ensure that these pins have a definedlevel. Even though most of the digital inputs are disabled in the deep sleep modes asdescribed above, floating inputs should be avoided to reduce current consumption in allother modes where the digital inputs are enabled (Reset, Active mode and Idle mode).

Assembly Code Example(1)

...

; Define pull-ups and set outputs high

; Define directions for port pins

ldi r16,(1<<PB7)|(1<<PB6)|(1<<PB1)|(1<<PB0)

ldi r17,(1<<DDB3)|(1<<DDB2)|(1<<DDB1)|(1<<DDB0)

out PORTB,r16

out DDRB,r17

; Insert nop for synchronization

nop

; Read port pins

in r16,PINB

...

C Code Example

unsigned char i;

...

/* Define pull-ups and set outputs high */

/* Define directions for port pins */

PORTB = (1<<PB7)|(1<<PB6)|(1<<PB1)|(1<<PB0);

DDRB = (1<<DDB3)|(1<<DDB2)|(1<<DDB1)|(1<<DDB0);

/* Insert nop for synchronization*/

__no_operation();

/* Read port pins */

i = PINB;

...

86 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The simplest method to ensure a defined level of an unused pin, is to enable the internalpull-up. In this case, the pull-up will be disabled during reset. If low power consumptionduring reset is important, it is recommended to use an external pull-up or pull-down.Connecting unused pins directly to VCC or GND is not recommended, since this maycause excessive currents if the pin is accidentally configured as an output.

Alternate Port Functions Most port pins have alternate functions in addition to being general digital I/Os. Figure37 shows how the port pin control signals from the simplified Figure 34 can be overrid-den by alternate functions. The overriding signals may not be present in all port pins, butthe figure serves as a generic description applicable to all port pins in the AVR micro-controller family.

Figure 37. Alternate Port Functions(1)

Note: 1. WRx, WPx, WDx, RRx, RPx, and RDx are common to all pins within the same port.clkI/O, SLEEP, and PUD are common to all ports. All other signals are unique for eachpin.

clk

RPx

RRxWRx

RDx

WDx

PUD

SYNCHRONIZER

WDx: WRITE DDRx

WRx: WRITE PORTxRRx: READ PORTx REGISTER

RPx: READ PORTx PIN

PUD: PULLUP DISABLE

clkI/O: I/O CLOCK

RDx: READ DDRx

D

L

Q

Q

SET

CLR

0

1

0

1

0

1

DIxn

AIOxn

DIEOExn

PVOVxn

PVOExn

DDOVxn

DDOExn

PUOExn

PUOVxn

PUOExn: Pxn PULL-UP OVERRIDE ENABLEPUOVxn: Pxn PULL-UP OVERRIDE VALUEDDOExn: Pxn DATA DIRECTION OVERRIDE ENABLEDDOVxn: Pxn DATA DIRECTION OVERRIDE VALUEPVOExn: Pxn PORT VALUE OVERRIDE ENABLEPVOVxn: Pxn PORT VALUE OVERRIDE VALUE

DIxn: DIGITAL INPUT PIN n ON PORTxAIOxn: ANALOG INPUT/OUTPUT PIN n ON PORTx

RESET

RESET

Q

Q D

CLR

Q

Q D

CLR

Q

QD

CLR

PINxn

PORTxn

DDxn

DAT

A B

US

0

1DIEOVxn

SLEEP

DIEOExn: Pxn DIGITAL INPUT-ENABLE OVERRIDE ENABLEDIEOVxn: Pxn DIGITAL INPUT-ENABLE OVERRIDE VALUESLEEP: SLEEP CONTROL

Pxn

I/O

0

1

PTOExn

PTOExn: Pxn, PORT TOGGLE OVERRIDE ENABLE

WPx: WRITE PINx

WPx

87

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 35 summarizes the function of the overriding signals. The pin and port indexesfrom Figure 37 are not shown in the succeeding tables. The overriding signals are gen-erated internally in the modules having the alternate function.

The following subsections shortly describe the alternate functions for each port, andrelate the overriding signals to the alternate function. Refer to the alternate functiondescription for further details.

Table 35. Generic Description of Overriding Signals for Alternate Functions

Signal Name Full Name Description

PUOE Pull-up Override Enable

If this signal is set, the pull-up enable is controlled by the PUOV signal. If this signal is cleared, the pull-up is enabled when DDxn, PORTxn, PUD = 0b010.

PUOV Pull-up Override Value

If PUOE is set, the pull-up is enabled/disabled when PUOV is set/cleared, regardless of the setting of the DDxn, PORTxn, and PUD Register bits.

DDOE Data Direction Override Enable

If this signal is set, the Output Driver Enable is controlled by the DDOV signal. If this signal is cleared, the Output driver is enabled by the DDxn Register bit.

DDOV Data Direction Override Value

If DDOE is set, the Output Driver is enabled/disabled when DDOV is set/cleared, regardless of the setting of the DDxn Register bit.

PVOE Port Value Override Enable

If this signal is set and the Output Driver is enabled, the port value is controlled by the PVOV signal. If PVOE is cleared, and the Output Driver is enabled, the port Value is controlled by the PORTxn Register bit.

PVOV Port Value Override Value

If PVOE is set, the port value is set to PVOV, regardless of the setting of the PORTxn Register bit.

PTOE Port Toggle Override Enable

If PTOE is set, the PORTxn Register bit is inverted.

DIEOE Digital Input Enable Override Enable

If this bit is set, the Digital Input Enable is controlled by the DIEOV signal. If this signal is cleared, the Digital Input Enable is determined by MCU state (Normal mode, sleep mode).

DIEOV Digital Input Enable Override Value

If DIEOE is set, the Digital Input is enabled/disabled when DIEOV is set/cleared, regardless of the MCU state (Normal mode, sleep mode).

DI Digital Input This is the Digital Input to alternate functions. In the figure, the signal is connected to the output of the schmitt trigger but before the synchronizer. Unless the Digital Input is used as a clock source, the module with the alternate function will use its own synchronizer.

AIO Analog Input/Output

This is the Analog Input/output to/from alternate functions. The signal is connected directly to the pad, and can be used bi-directionally.

88 ATmega640/1280/1281/2560/25612549A–AVR–03/05

MCU Control Register – MCUCR

• Bit 4 – PUD: Pull-up Disable

When this bit is written to one, the pull-ups in the I/O ports are disabled even if the DDxnand PORTxn Registers are configured to enable the pull-ups (DDxn, PORTxn = 0b01).See “Configuring the Pin” on page 82 for more details about this feature.

Alternate Functions of Port A The Port A has an alternate function as the address low byte and data lines for theExternal Memory Interface.

Table 37 and Table 38 relates the alternate functions of Port A to the overriding signalsshown in Figure 37 on page 86.

Note: 1. ADA is short for ADdress Active and represents the time when address is output. See“External Memory Interface” on page 29 for details.

Bit 7 6 5 4 3 2 1 0

JTD – – PUD – – IVSEL IVCE MCUCR

Read/Write R/W R R R/W R R R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 36. Port A Pins Alternate Functions

Port Pin Alternate Function

PA7 AD7 (External memory interface address and data bit 7)

PA6 AD6 (External memory interface address and data bit 6)

PA5 AD5 (External memory interface address and data bit 5)

PA4 AD4 (External memory interface address and data bit 4)

PA3 AD3 (External memory interface address and data bit 3)

PA2 AD2 (External memory interface address and data bit 2)

PA1 AD1 (External memory interface address and data bit 1)

PA0 AD0 (External memory interface address and data bit 0)

Table 37. Overriding Signals for Alternate Functions in PA7..PA4

SignalName PA7/AD7 PA6/AD6 PA5/AD5 PA4/AD4

PUOE SRE SRE SRE SRE

PUOV ~(WR | ADA(1)) • PORTA7 • PUD

~(WR | ADA) • PORTA6 • PUD

~(WR | ADA) • PORTA5 • PUD

~(WR | ADA) • PORTA4 • PUD

DDOE SRE SRE SRE SRE

DDOV WR | ADA WR | ADA WR | ADA WR | ADA

PVOE SRE SRE SRE SRE

PVOV A7 • ADA | D7 OUTPUT • WR

A6 • ADA | D6 OUTPUT • WR

A5 • ADA | D5 OUTPUT • WR

A4 • ADA | D4 OUTPUT • WR

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI D7 INPUT D6 INPUT D5 INPUT D4 INPUT

AIO – – – –

89

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Alternate Functions of Port B The Port B pins with alternate functions are shown in Table 39.

The alternate pin configuration is as follows:

• OC0A/OC1C/PCINT7, Bit 7

OC0A, Output Compare Match A output: The PB7 pin can serve as an external outputfor the Timer/Counter0 Output Compare. The pin has to be configured as an output(DDB7 set “one”) to serve this function. The OC0A pin is also the output pin for the PWMmode timer function.

OC1C, Output Compare Match C output: The PB7 pin can serve as an external outputfor the Timer/Counter1 Output Compare C. The pin has to be configured as an output(DDB7 set (one)) to serve this function. The OC1C pin is also the output pin for thePWM mode timer function.

Table 38. Overriding Signals for Alternate Functions in PA3..PA0

Signal Name PA3/AD3 PA2/AD2 PA1/AD1 PA0/AD0

PUOE SRE SRE SRE SRE

PUOV ~(WR | ADA) • PORTA3 • PUD

~(WR | ADA) • PORTA2 • PUD

~(WR | ADA) • PORTA1 • PUD

~(WR | ADA) • PORTA0 • PUD

DDOE SRE SRE SRE SRE

DDOV WR | ADA WR | ADA WR | ADA WR | ADA

PVOE SRE SRE SRE SRE

PVOV A3 • ADA | D3 OUTPUT • WR

A2• ADA | D2 OUTPUT • WR

A1 • ADA | D1 OUTPUT • WR

A0 • ADA | D0 OUTPUT • WR

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI D3 INPUT D2 INPUT D1 INPUT D0 INPUT

AIO – – – –

Table 39. Port B Pins Alternate Functions

Port Pin Alternate Functions

PB7OC0A/OC1C/PCINT7 (Output Compare and PWM Output A for Timer/Counter0, Output Compare and PWM Output C for Timer/Counter1 or Pin Change Interrupt 7)

PB6OC1B/PCINT6 (Output Compare and PWM Output B for Timer/Counter1 or Pin Change Interrupt 6)

PB5OC1A/PCINT5 (Output Compare and PWM Output A for Timer/Counter1 or Pin Change Interrupt 5)

PB4OC2A/PCINT4 (Output Compare and PWM Output A for Timer/Counter2 or Pin Change Interrupt 4)

PB3 MISO/PCINT3 (SPI Bus Master Input/Slave Output or Pin Change Interrupt 3)

PB2 MOSI/PCINT2 (SPI Bus Master Output/Slave Input or Pin Change Interrupt 2)

PB1 SCK/PCINT1 (SPI Bus Serial Clock or Pin Change Interrupt 1)

PB0 SS/PCINT0 (SPI Slave Select input or Pin Change Interrupt 0)

90 ATmega640/1280/1281/2560/25612549A–AVR–03/05

PCINT7, Pin Change Interrupt source 7: The PB7 pin can serve as an external interruptsource.

• OC1B/PCINT6, Bit 6

OC1B, Output Compare Match B output: The PB6 pin can serve as an external outputfor the Timer/Counter1 Output Compare B. The pin has to be configured as an output(DDB6 set (one)) to serve this function. The OC1B pin is also the output pin for the PWMmode timer function.

PCINT6, Pin Change Interrupt source 6: The PB7 pin can serve as an external interruptsource.

• OC1A/PCINT5, Bit 5

OC1A, Output Compare Match A output: The PB5 pin can serve as an external outputfor the Timer/Counter1 Output Compare A. The pin has to be configured as an output(DDB5 set (one)) to serve this function. The OC1A pin is also the output pin for the PWMmode timer function.

PCINT5, Pin Change Interrupt source 5: The PB7 pin can serve as an external interruptsource.

• OC2A/PCINT4, Bit 4

OC2A, Output Compare Match output: The PB4 pin can serve as an external output forthe Timer/Counter2 Output Compare. The pin has to be configured as an output (DDB4set (one)) to serve this function. The OC2A pin is also the output pin for the PWM modetimer function.

PCINT4, Pin Change Interrupt source 4: The PB7 pin can serve as an external interruptsource.

• MISO/PCINT3 – Port B, Bit 3

MISO: Master Data input, Slave Data output pin for SPI channel. When the SPI isenabled as a master, this pin is configured as an input regardless of the setting ofDDB3. When the SPI is enabled as a slave, the data direction of this pin is controlled byDDB3. When the pin is forced to be an input, the pull-up can still be controlled by thePORTB3 bit.

PCINT3, Pin Change Interrupt source 3: The PB7 pin can serve as an external interruptsource.

• MOSI/PCINT2 – Port B, Bit 2

MOSI: SPI Master Data output, Slave Data input for SPI channel. When the SPI isenabled as a slave, this pin is configured as an input regardless of the setting of DDB2.When the SPI is enabled as a master, the data direction of this pin is controlled byDDB2. When the pin is forced to be an input, the pull-up can still be controlled by thePORTB2 bit.

PCINT2, Pin Change Interrupt source 2: The PB7 pin can serve as an external interruptsource.

• SCK/PCINT1 – Port B, Bit 1

SCK: Master Clock output, Slave Clock input pin for SPI channel. When the SPI isenabled as a slave, this pin is configured as an input regardless of the setting of DDB1.

91

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

When the SPI0 is enabled as a master, the data direction of this pin is controlled byDDB1. When the pin is forced to be an input, the pull-up can still be controlled by thePORTB1 bit.

PCINT1, Pin Change Interrupt source 1: The PB7 pin can serve as an external interruptsource.

• SS/PCINT0 – Port B, Bit 0

SS: Slave Port Select input. When the SPI is enabled as a slave, this pin is configuredas an input regardless of the setting of DDB0. As a slave, the SPI is activated when thispin is driven low. When the SPI is enabled as a master, the data direction of this pin iscontrolled by DDB0. When the pin is forced to be an input, the pull-up can still be con-trolled by the PORTB0 bit.

Table 40 and Table 41 relate the alternate functions of Port B to the overriding signalsshown in Figure 37 on page 86. SPI MSTR INPUT and SPI SLAVE OUTPUT constitutethe MISO signal, while MOSI is divided into SPI MSTR OUTPUT and SPI SLAVEINPUT.

PCINT0, Pin Change Interrupt source 0: The PB7 pin can serve as an external interruptsource..

Table 40. Overriding Signals for Alternate Functions in PB7..PB4

Signal Name PB7/OC0A/OC1C PB6/OC1B PB5/OC1A PB4/OC2

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE OC0/OC1C ENABLE OC1B ENABLE OC1A ENABLE OC2 ENABLE

PVOV OC0/OC1C OC1B OC1A OC2B

DIEOE PCINT7 • PCIE0 PCINT6 • PCIE0 PCINT5 • PCIE0 PCINT4 • PCIE0

DIEOV 1 1 1 1

DI PCINT7 INPUT PCINT6 INPUT PCINT5 INPUT PCINT4 INPUT

AIO – – – –

92 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Alternate Functions of Port C The Port C alternate function is as follows:

Table 43 and Table 44 relate the alternate functions of Port C to the overriding signalsshown in Figure 37 on page 86.

Table 41. Overriding Signals for Alternate Functions in PB3..PB0

Signal Name PB3/MISO PB2/MOSI PB1/SCK PB0/SS

PUOE SPE • MSTR SPE • MSTR SPE • MSTR SPE • MSTR

PUOV PORTB3 • PUD PORTB2 • PUD PORTB1 • PUD PORTB0 • PUD

DDOE SPE • MSTR SPE • MSTR SPE • MSTR SPE • MSTR

DDOV 0 0 0 0

PVOE SPE • MSTR SPE • MSTR SPE • MSTR 0

PVOV SPI SLAVE OUTPUT SPI MSTR OUTPUT SCK OUTPUT 0

DIEOE PCINT3 • PCIE0 PCINT2 • PCIE0 PCINT1 • PCIE0

PCINT0 • PCIE0

DIEOV 1 1 1 1

DI SPI MSTR INPUT

PCINT3 INPUT

SPI SLAVE INPUT

PCINT2 INPUT

SCK INPUT

PCINT1 INPUT

SPI SS

PCINT0 INPUT

AIO – – – –

Table 42. Port C Pins Alternate Functions

Port Pin Alternate Function

PC7 A15(External Memory interface address bit 15)

PC6 A14(External Memory interface address bit 14)

PC5 A13(External Memory interface address bit 13)

PC4 A12(External Memory interface address bit 12)

PC3 A11(External Memory interface address bit 11)

PC2 A10(External Memory interface address bit 10)

PC1 A9(External Memory interface address bit 9)

PC0 A8(External Memory interface address bit 8)

93

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 43. Overriding Signals for Alternate Functions in PC7..PC4

Signal Name PC7/A15 PC6/A14 PC5/A13 PC4/A12

PUOE SRE • (XMM<1) SRE • (XMM<2) SRE • (XMM<3) SRE • (XMM<4)

PUOV 0 0 0 0

DDOE SRE • (XMM<1) SRE • (XMM<2) SRE • (XMM<3) SRE • (XMM<4)

DDOV 1 1 1 1

PVOE SRE • (XMM<1) SRE • (XMM<2) SRE • (XMM<3) SRE • (XMM<4)

PVOV A15 A14 A13 A12

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI – – – –

AIO – – – –

Table 44. Overriding Signals for Alternate Functions in PC3..PC0

Signal Name PC3/A11 PC2/A10 PC1/A9 PC0/A8

PUOE SRE • (XMM<5) SRE • (XMM<6) SRE • (XMM<7) SRE • (XMM<7)

PUOV 0 0 0 0

DDOE SRE • (XMM<5) SRE • (XMM<6) SRE • (XMM<7) SRE • (XMM<7)

DDOV 1 1 1 1

PVOE SRE • (XMM<5) SRE • (XMM<6) SRE • (XMM<7) SRE • (XMM<7)

PVOV A11 A10 A9 A8

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI – – – –

AIO – – – –

94 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Alternate Functions of Port D The Port D pins with alternate functions are shown in Table 45.

The alternate pin configuration is as follows:

• T0 – Port D, Bit 7

T0, Timer/Counter0 counter source.

• T1 – Port D, Bit 6

T1, Timer/Counter1 counter source.

• XCK1 – Port D, Bit 5

XCK1, USART1 External clock. The Data Direction Register (DDD5) controls whetherthe clock is output (DDD5 set) or input (DDD5 cleared). The XCK1 pin is active onlywhen the USART1 operates in Synchronous mode.

• ICP1 – Port D, Bit 4

ICP1 – Input Capture Pin 1: The PD4 pin can act as an input capture pin forTimer/Counter1.

• INT3/TXD1 – Port D, Bit 3

INT3, External Interrupt source 3: The PD3 pin can serve as an external interrupt sourceto the MCU.

TXD1, Transmit Data (Data output pin for the USART1). When the USART1 Transmitteris enabled, this pin is configured as an output regardless of the value of DDD3.

• INT2/RXD1 – Port D, Bit 2

INT2, External Interrupt source 2. The PD2 pin can serve as an External Interruptsource to the MCU.

RXD1, Receive Data (Data input pin for the USART1). When the USART1 receiver isenabled this pin is configured as an input regardless of the value of DDD2. When theUSART forces this pin to be an input, the pull-up can still be controlled by the PORTD2bit.

Table 45. Port D Pins Alternate Functions

Port Pin Alternate Function

PD7 T0 (Timer/Counter0 Clock Input)

PD6 T1 (Timer/Counter1 Clock Input)

PD5 XCK1 (USART1 External Clock Input/Output)

PD4 ICP1 (Timer/Counter1 Input Capture Trigger)

PD3 INT3/TXD1 (External Interrupt3 Input or USART1 Transmit Pin)

PD2 INT2/RXD1 (External Interrupt2 Input or USART1 Receive Pin)

PD1 INT1/SDA (External Interrupt1 Input or TWI Serial DAta)

PD0 INT0/SCL (External Interrupt0 Input or TWI Serial CLock)

95

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• INT1/SDA – Port D, Bit 1

INT1, External Interrupt source 1. The PD1 pin can serve as an external interrupt sourceto the MCU.

SDA, 2-wire Serial Interface Data: When the TWEN bit in TWCR is set (one) to enablethe 2-wire Serial Interface, pin PD1 is disconnected from the port and becomes theSerial Data I/O pin for the 2-wire Serial Interface. In this mode, there is a spike filter onthe pin to suppress spikes shorter than 50 ns on the input signal, and the pin is driven byan open drain driver with slew-rate limitation.

• INT0/SCL – Port D, Bit 0

INT0, External Interrupt source 0. The PD0 pin can serve as an external interrupt sourceto the MCU.

SCL, 2-wire Serial Interface Clock: When the TWEN bit in TWCR is set (one) to enablethe 2-wire Serial Interface, pin PD0 is disconnected from the port and becomes theSerial Clock I/O pin for the 2-wire Serial Interface. In this mode, there is a spike filter onthe pin to suppress spikes shorter than 50 ns on the input signal, and the pin is driven byan open drain driver with slew-rate limitation.

Table 46 and Table 47 relates the alternate functions of Port D to the overriding signalsshown in Figure 37 on page 86.

Table 46. Overriding Signals for Alternate Functions PD7..PD4

Signal Name PD7/T0 PD6/T1 PD5/XCK1 PD4/ICP1

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 XCK1 OUTPUT ENABLE 0

DDOV 0 0 1 0

PVOE 0 0 XCK1 OUTPUT ENABLE 0

PVOV 0 0 XCK1 OUTPUT 0

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI T0 INPUT T1 INPUT XCK1 INPUT ICP1 INPUT

AIO – – – –

96 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. When enabled, the 2-wire Serial Interface enables Slew-Rate controls on the outputpins PD0 and PD1. This is not shown in this table. In addition, spike filters are con-nected between the AIO outputs shown in the port figure and the digital logic of theTWI module.

Alternate Functions of Port E The Port E pins with alternate functions are shown in Table 48.

Note: 1. Only for ATmega1281/2561. For ATmega1281/2561 these functions are placedon MISO/MOSI pins.

• INT7/ICP3 – Port E, Bit 7

INT7, External Interrupt source 7: The PE7 pin can serve as an external interruptsource.

ICP3 - Input Capture Pin 3: The PE7 pin can act as an input capture pin forTimer/Counter3.

Table 47. Overriding Signals for Alternate Functions in PD3..PD0(1)

Signal Name PD3/INT3/TXD1 PD2/INT2/RXD1 PD1/INT1/SDA PD0/INT0/SCL

PUOE TXEN1 RXEN1 TWEN TWEN

PUOV 0 PORTD2 • PUD PORTD1 • PUD PORTD0 • PUD

DDOE TXEN1 RXEN1 TWEN TWEN

DDOV 1 0 SDA_OUT SCL_OUT

PVOE TXEN1 0 TWEN TWEN

PVOV TXD1 0 0 0

DIEOE INT3 ENABLE INT2 ENABLE INT1 ENABLE INT0 ENABLE

DIEOV 1 1 1 1

DI INT3 INPUT INT2 INPUT/RXD1 INT1 INPUT INT0 INPUT

AIO – – SDA INPUT SCL INPUT

Table 48. Port E Pins Alternate Functions

Port Pin Alternate Function

PE7INT7/ICP3/CLK0 (External Interrupt 7 Input, Timer/Counter3 Input Capture Trigger or Divided System Clock)

PE6 INT6/ T3 (External Interrupt 6 Input or Timer/Counter3 Clock Input)

PE5INT5/OC3C (External Interrupt 5 Input or Output Compare and PWM Output C for Timer/Counter3)

PE4INT4/OC3B (External Interrupt4 Input or Output Compare and PWM Output B for Timer/Counter3)

PE3AIN1/OC3A (Analog Comparator Negative Input or Output Compare and PWM Output A for Timer/Counter3)

PE2AIN0/XCK0 (Analog Comparator Positive Input or USART0 external clock input/output)

PE1 PDO(1)/TXD0 (Programming Data Output or USART0 Transmit Pin)

PE0PDI(1)/RXD0/PCINT8 (Programming Data Input, USART0 Receive Pin or Pin Change Interrupt 8)

97

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

CLKO - Divided System Clock: The divided system clock can be output on the PE7 pin.The divided system clock will be output if the CKOUT Fuse is programmed, regardlessof the PORTE7 and DDE7 settings. It will also be output during reset.

• INT6/T3 – Port E, Bit 6

INT6, External Interrupt source 6: The PE6 pin can serve as an external interruptsource.

T3, Timer/Counter3 counter source.

• INT5/OC3C – Port E, Bit 5

INT5, External Interrupt source 5: The PE5 pin can serve as an External Interruptsource.

OC3C, Output Compare Match C output: The PE5 pin can serve as an External outputfor the Timer/Counter3 Output Compare C. The pin has to be configured as an output(DDE5 set “one”) to serve this function. The OC3C pin is also the output pin for thePWM mode timer function.

• INT4/OC3B – Port E, Bit 4

INT4, External Interrupt source 4: The PE4 pin can serve as an External Interruptsource.

OC3B, Output Compare Match B output: The PE4 pin can serve as an External outputfor the Timer/Counter3 Output Compare B. The pin has to be configured as an output(DDE4 set (one)) to serve this function. The OC3B pin is also the output pin for the PWMmode timer function.

• AIN1/OC3A – Port E, Bit 3

AIN1 – Analog Comparator Negative input. This pin is directly connected to the negativeinput of the Analog Comparator.

OC3A, Output Compare Match A output: The PE3 pin can serve as an External outputfor the Timer/Counter3 Output Compare A. The pin has to be configured as an output(DDE3 set “one”) to serve this function. The OC3A pin is also the output pin for the PWMmode timer function.

• AIN0/XCK0 – Port E, Bit 2

AIN0 – Analog Comparator Positive input. This pin is directly connected to the positiveinput of the Analog Comparator.

XCK0, USART0 External clock. The Data Direction Register (DDE2) controls whetherthe clock is output (DDE2 set) or input (DDE2 cleared). The XCK0 pin is active onlywhen the USART0 operates in Synchronous mode.

• PDO/TXD0 – Port E, Bit 1

PDO, SPI Serial Programming Data Output. During Serial Program Downloading, thispin is used as data output line for the ATmega1281/2561. For ATmega640/1280/2560this function is placed on MISO.

TXD0, USART0 Transmit pin.

98 ATmega640/1280/1281/2560/25612549A–AVR–03/05

• PDI/RXD0/PCINT8 – Port E, Bit 0

PDI, SPI Serial Programming Data Input. During Serial Program Downloading, this pinis used as data input line for the ATmega1281/2561. For ATmega640/1280/2560 thisfunction is placed on MOSI.

RXD0, USART0 Receive Pin. Receive Data (Data input pin for the USART0). When theUSART0 receiver is enabled this pin is configured as an input regardless of the value ofDDRE0. When the USART0 forces this pin to be an input, a logical one in PORTE0 willturn on the internal pull-up.

PCINT8, Pin Change Interrupt source 8: The PE0 pin can serve as an external interrupt source.

Table 49 and Table 50 relates the alternate functions of Port E to the overriding signalsshown in Figure 37 on page 86.

Table 49. Overriding Signals for Alternate Functions PE7..PE4

Signal Name PE7/INT7/ICP3 PE6/INT6/T3 PE5/INT5/OC3C PE4/INT4/OC3B

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE 0 0 OC3C ENABLE OC3B ENABLE

PVOV 0 0 OC3C OC3B

DIEOE INT7 ENABLE INT6 ENABLE INT5 ENABLE INT4 ENABLE

DIEOV 1 1 1 1

DI INT7 INPUT/ICP3 INPUT

INT7 INPUT/T3 INPUT

INT5 INPUT INT4 INPUT

AIO – – – –

99

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: 1. PDO/PDI only available at PE1/PE0 for ATmega1281/2561.

Table 50. Overriding Signals for Alternate Functions in PE3..PE0

Signal Name PE3/AIN1/OC3A PE2/AIN0/XCK0

PE1/PDO(1)/TXD0

PE0/PDI(1)/RXD0/PCINT8

PUOE 0 0 TXEN0 RXEN0

PUOV 0 0 0 PORTE0 • PUD

DDOE 0 XCK0 OUTPUT ENABLE

TXEN0 RXEN0

DDOV 0 1 1 0

PVOE OC3B ENABLE XCK0 OUTPUT ENABLE

TXEN0 0

PVOV OC3B XCK0 OUTPUT TXD0 0

DIEOE 0 0 0 PCINT8 • PCIE1

DIEOV 0 0 0 1

DI 0 XCK0 INPUT – RXD0

PE0 0 0 0 PCINT8 INPUT

AIO AIN1 INPUT AIN0 INPUT – –

100 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Alternate Functions of Port F The Port F has an alternate function as analog input for the ADC as shown in Table 51.If some Port F pins are configured as outputs, it is essential that these do not switchwhen a conversion is in progress. This might corrupt the result of the conversion. If theJTAG interface is enabled, the pull-up resistors on pins PF7(TDI), PF5(TMS), andPF4(TCK) will be activated even if a Reset occurs.

• TDI, ADC7 – Port F, Bit 7

ADC7, Analog to Digital Converter, Channel 7.

TDI, JTAG Test Data In: Serial input data to be shifted in to the Instruction Register orData Register (scan chains). When the JTAG interface is enabled, this pin can not beused as an I/O pin.

• TDO, ADC6 – Port F, Bit 6

ADC6, Analog to Digital Converter, Channel 6.

TDO, JTAG Test Data Out: Serial output data from Instruction Register or Data Regis-ter. When the JTAG interface is enabled, this pin can not be used as an I/O pin.

The TDO pin is tri-stated unless TAP states that shift out data are entered.

• TMS, ADC5 – Port F, Bit 5

ADC5, Analog to Digital Converter, Channel 5.

TMS, JTAG Test Mode Select: This pin is used for navigating through the TAP-controllerstate machine. When the JTAG interface is enabled, this pin can not be used as an I/Opin.

• TCK, ADC4 – Port F, Bit 4

ADC4, Analog to Digital Converter, Channel 4.

TCK, JTAG Test Clock: JTAG operation is synchronous to TCK. When the JTAG inter-face is enabled, this pin can not be used as an I/O pin.

Table 51. Port F Pins Alternate Functions

Port Pin Alternate Function

PF7 ADC7/TDI (ADC input channel 7 or JTAG Test Data Input)

PF6 ADC6/TDO (ADC input channel 6 or JTAG Test Data Output)

PF5 ADC5/TMS (ADC input channel 5 or JTAG Test Mode Select)

PF4 ADC4/TCK (ADC input channel 4 or JTAG Test ClocK)

PF3 ADC3 (ADC input channel 3)

PF2 ADC2 (ADC input channel 2)

PF1 ADC1 (ADC input channel 1)

PF0 ADC0 (ADC input channel 0)

101

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• ADC3 – ADC0 – Port F, Bit 3..0

Analog to Digital Converter, Channel 3..0.

Table 52. Overriding Signals for Alternate Functions in PF7..PF4

Signal Name PF7/ADC7/TDI PF6/ADC6/TDO PF5/ADC5/TMS PF4/ADC4/TCK

PUOE JTAGEN JTAGEN JTAGEN JTAGEN

PUOV 1 0 1 1

DDOE JTAGEN JTAGEN JTAGEN JTAGEN

DDOV 0 SHIFT_IR + SHIFT_DR

0 0

PVOE 0 JTAGEN 0 0

PVOV 0 TDO 0 0

DIEOE JTAGEN JTAGEN JTAGEN JTAGEN

DIEOV 0 0 0 0

DI – – – –

AIO TDI/ADC7 INPUT ADC6 INPUT TMS/ADC5 INPUT

TCK/ADC4 INPUT

Table 53. Overriding Signals for Alternate Functions in PF3..PF0

Signal Name PF3/ADC3 PF2/ADC2 PF1/ADC1 PF0/ADC0

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE 0 0 0 0

PVOV 0 0 0 0

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI – – – –

AIO ADC3 INPUT ADC2 INPUT ADC1 INPUT ADC0 INPUT

102 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Alternate Functions of Port G The Port G alternate pin configuration is as follows:

• OC0B – Port G, Bit 5

OC0B, Output Compare match B output: The PG5 pin can serve as an external outputfor the TImer/Counter0 Output Compare. The pin has to be configured as an output(DDG5 set) to serve this function. The OC0B pin is also the output pin for the PWMmode timer function.

• TOSC1 – Port G, Bit 4

TOSC2, Timer Oscillator pin 1: When the AS2 bit in ASSR is set (one) to enable asyn-chronous clocking of Timer/Counter2, pin PG4 is disconnected from the port, andbecomes the input of the inverting Oscillator amplifier. In this mode, a Crystal Oscillatoris connected to this pin, and the pin can not be used as an I/O pin.

• TOSC2 – Port G, Bit 3

TOSC2, Timer Oscillator pin 2: When the AS2 bit in ASSR is set (one) to enable asyn-chronous clocking of Timer/Counter2, pin PG3 is disconnected from the port, andbecomes the inverting output of the Oscillator amplifier. In this mode, a Crystal Oscillatoris connected to this pin, and the pin can not be used as an I/O pin.

• ALE – Port G, Bit 2

ALE is the external data memory Address Latch Enable signal.

• RD – Port G, Bit 1

RD is the external data memory read control strobe.

• WR – Port G, Bit 0

WR is the external data memory write control strobe.

Table 55 and Table 56 relates the alternate functions of Port G to the overriding signalsshown in Figure 37 on page 86.

Table 54. Port G Pins Alternate Functions

Port Pin Alternate Function

PG5 OC0B (Output Compare and PWM Output B for Timer/Counter0)

PG4 TOSC1 (RTC Oscillator Timer/Counter2)

PG3 TOSC2 (RTC Oscillator Timer/Counter2)

PG2 ALE (Address Latch Enable to external memory)

PG1 RD (Read strobe to external memory)

PG0 WR (Write strobe to external memory)

103

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 55. Overriding Signals for Alternate Functions in PG5..PG4

Signal Name — — PG5/OC0B PG4/TOSC1

PUOE – – – AS2

PUOV – – – 0

DDOE – – – AS2

DDOV – – – 0

PVOE – – OC0B Enable 0

PVOV – – OC0B 0

PTOE – – – –

DIEOE – – – AS2

DIEOV – – – EXCLK

DI – – – –

AIO – – – T/C2 OSC INPUT

Table 56. Overriding Signals for Alternate Functions in PG3..PG0

Signal Name PG3/TOSC2 PG2/ALE/A7 PG1/RD PG0/WR

PUOE AS2 • EXCLK SRE SRE SRE

PUOV 0 0 0 0

DDOE AS2 • EXCLK SRE SRE SRE

DDOV 0 1 1 1

PVOE 0 SRE SRE SRE

PVOV 0 ALE RD WR

PTOE – – – –

DIEOE AS2 • EXCLK 0 0 0

DIEOV 0 0 0 0

DI – – – –

AIO T/C2 OSC OUTPUT – – –

104 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Alternate Functions of Port H The Port H alternate pin configuration is as follows:

• T4 – Port H, Bit 7

T4, Timer/Counter4 counter source.

• OC2B – Port H, Bit 6

OC2B, Output Compare Match B output: The PH6 pin can serve as an external output for theTimer/Counter2 Output Compare B. The pin has to be configured as an output (DDH6 set) toserve this function. The OC2B pin is also the output pin for the PWM mode timer function.

• OC4C – Port H, Bit 5

OC4C, Output Compare Match C output: The PH5 pin can serve as an external output for theTimer/Counter4 Output Compare C. The pin has to be configured as an output (DDH5 set) toserve this function. The OC4C pin is also the output pin for the PWM mode timer function.

• OC4B – Port H, Bit 4

OC4B, Output Compare Match B output: The PH4 pin can serve as an external output for theTimer/Counter2 Output Compare B. The pin has to be configured as an output (DDH4 set) toserve this function. The OC4B pin is also the output pin for the PWM mode timer function.

• OC4A – Port H, Bit 3

OC4C, Output Compare Match A output: The PH3 pin can serve as an external outputfor the Timer/Counter4 Output Compare A. The pin has to be configured as an output(DDH3 set) to serve this function. The OC4A pin is also the output pin for the PWMmode timer function.

• XCK2 – Port H, Bit 2

XCK2, USART2 External Clock: The Data Direction Register (DDH2) controls whether the clock isoutput (DDH2 set) or input (DDH2 cleared). The XC2K pin is active only when the USART2 oper-ates in synchronous mode.

• TXD2 – Port H, Bit 1

TXD2, USART2 Transmit Pin.

Table 57. Port H Pins Alternate Functions

Port Pin Alternate Function

PH7 T4 (Timer/Counter4 Clock Input)

PH6 OC2B(Output Compare and PWM Output B for Timer/Counter2)

PH5 OC4C(Output Compare and PWM Output C for Timer/Counter4)

PH4 OC4B(Output Compare and PWM Output B for Timer/Counter4)

PH3 OC4A(Output Compare and PWM Output A for Timer/Counter4)

PH2 XCK2 (USART2 External Clock)

PH1 TXD2 (USART2 Transmit Pin)

PH0 RXD2 (USART2 Receive Pin)

105

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• RXD2 – Port H, Bit 0

RXD2, USART2 Receive pin: Receive Data (Data input pin for the USART2). When theUSART2 Receiver is enabled, this pin is configured as an input regardless of the valueof DDH0. When the USART2 forces this pin to be an input, a logical on in PORTH0 willturn on the internal pull-up.

Table 58. Overriding Signals for Alternate Functions in PH7..PH4

Signal Name PH7/T4 PH6/OC2B PH5/OC4C PH4/OC4B

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE 0 OC2B ENABLE OC4C ENABLE OC4B ENABLE

PVOV 0 OC2B OC4C OC4B

PTOE – – – –

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI T4 INPUT 0 0 0

AIO – – – –

Table 59. Overriding Signals for Alternate Functions in PH3..PH0

Signal Name PH3/OC4A PH2/XCK2 PH1/TXD2 PH0/RXD2

PUOE 0 0 TXEN2 RXEN2

PUOV 0 0 0 PORTH0 • PUD

DDOE 0 XCK2 OUTPUT ENABLE

TXEN2 RXEN2

DDOV 0 1 1 0

PVOE OC4A ENABLE XCK2 OUTPUT ENABLE

TXEN2 0

PVOV OC4A XCK2 TXD2 0

PTOE – – – –

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI 0 XC2K INPUT 0 RXD2

AIO – – – –

106 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Alternate Functions of Port J The Port J alternate pin configuration is as follows:

• PCINT15:12 - Port J, Bit 6:3

PCINT15:12, Pin Change Interrupt Source 15:12. The PJ6:3 pins can serve as ExternalInterrupt Sources

• XCK2/PCINT11 - Port J, Bit 2

XCK2, USART 2 External Clock. The Data Direction Register (DDJ2) controls whetherthe clock is output (DDJ2 set) or input (DDJ2 cleared). The XCK2 pin is active only whenthe USART2 operates in synchronous mode.

PCINT11, Pin Change Interrupt Source 11. The PJ2 pins can serve as External InterruptSources

• TXD3/PCINT10 - Port J, Bit 1

TXD3, USART3 Transmit pin

PCINT10, Pin Change Interrupt Source 10. The PJ2 pins can serve as External InterruptSources

• RXD3/PCINT9 - Port J, Bit 0

RXD3, USART3 Receive pin. Receive Data (Data input pin for the USART3). When theUSART3 Receiver is enabled, this pin is configured as an input regardless of the valueof DDJ0. When the USART3 forces this pin to be an input, a logical one in PORTJ0 willturn on the internal pull-up.

PCINT9, Pin Change Interrupt Source 9. The PJ2 pins can serve as External InterruptSources

Table 61 and Table 62 relates the alternate functions of Port J to the overriding signalsshown in Figure 37 on page 86

Table 60. Port J Pins Alternate Functions

Port Pin Alternate Function

PJ7 –

PJ6 PCINT15 (Pin Change Interrupt 15)

PJ5 PCINT14 (Pin Change Interrupt 14)

PJ4 PCINT13 (Pin Change Interrupt 13)

PJ3 PCINT12 (Pin Change Interrupt 12)

PJ2 XCK3/PCINT11 (USART3 External Clock or Pin Change Interrupt 11)

PJ1 TXD3/PCINT10 (USART3 Transmit Pin or Pin Change Interrupt 10)

PJ0 RXD3/PCINT9 (USART3 Receive Pin or Pin Change Interrupt 9)

107

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 61. Overriding Signals for Alternate Functions in PJ7..PJ4

Signal Name PJ7 PJ6/ PCINT15 PJ5/ PCINT14 PJ4/ PCINT13

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE 0 0 0 0

PVOV 0 0 0 0

PTOE - - - -

DIEOE 0 PCINT15·PCIE1 PCINT14·PCIE1 PCINT13·PCIE1

DIEOV 0 1 1 1

DI 0 PCINT15 INPUT PCINT14 INPUT PCINT13 INPUT

AIO - - - -

Table 62. Overriding Signals for Alternate Functions in PJ3..PJ0

Signal Name PJ3/PCINT12PJ2/XCK3/PCINT11

PJ1/TXD3/PCINT10

PJ0/RXD3/PCINT9

PUOE 0 0 TXEN3 RXEN3

PUOV 0 0 0 PORTJ0·PUD

DDOE 0 XCK3 OUTPUT ENABLE

TXEN3 RXEN3

DDOV 0 1 1 0

PVOE 0 XCK3 OUTPUT ENABLE

TXEN3 0

PVOV 0 XCK3 TXD3 0

PTOE - - - -

DIEOE PCINT12·PCIE1 PCINT11·PCIE1 PCINT10·PCIE1 PCINT9·PCIE1

DIEOV 1 1 1 1

DI PCINT12 INPUT PCINT11 INPUTXCK3 INPUT

PCINT10 INPUT PCINT9 INPUT RXD3

AIO - - - -

108 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Alternate Functions of Port K The Port K alternate pin configuration is as follows:

• ADC15:8/PCINT23:16 – Port K, Bit 7:0

ADC15:8, Analog to Digital Converter, Channel 15 - 8.

PCINT23:16, Pin Change Interrupt Source 23:16. The PK7:0 pins can serve as ExternalInterrupt Sources.

Table 63. Port K Pins Alternate Functions

Port Pin Alternate Function

PK7 ADC15/PCINT23 (ADC Input Channel 15 or Pin Change Interrupt 23)

PK6 ADC14/PCINT22 (ADC Input Channel 14 or Pin Change Interrupt 22)

PK5 ADC13/PCINT21 (ADC Input Channel 13 or Pin Change Interrupt 21)

PK4 ADC12/PCINT20 (ADC Input Channel 12 or Pin Change Interrupt 20)

PK3 ADC11/PCINT19 (ADC Input Channel 11 or Pin Change Interrupt 19)

PK2 ADC10/PCINT18 (ADC Input Channel 10 or Pin Change Interrupt 18)

PK1 ADC9/PCINT17 (ADC Input Channel 9 or Pin Change Interrupt 17)

PK0 ADC8 /PCINT16 (ADC Input Channel 8 or Pin Change Interrupt 16)

Table 64. Overriding Signals for Alternate Functions in PK7..PK4

Signal NamePK7/ADC15/PCINT23

PK6/ADC14/PCINT22

PK5/ADC13/PCINT21

PK4/ADC12/PCINT20

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE 0 0 0 0

PVOV 0 0 0 0

PTOE – – – –

DIEOE PCINT23 • PCIE2

PCINT22 • PCIE2

PCINT21 • PCIE2

PCINT20 • PCIE2

DIEOV 1 1 1 1

DI PCINT23 INPUT

PCINT22 INPUT

PCINT21 INPUT

PCINT20 INPUT

AIO ADC15 INPUT ADC14 INPUT ADC13 INPUT ADC12 INPUT

109

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 65. Overriding Signals for Alternate Functions in PK3..PK0

Signal NamePK3/ADC11/PCINT19

PK2/ADC10/PCINT18

PK1/ADC9/PCINT17

PK0/ADC8/PCINT16

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE 0 0 0 0

PVOV 0 0 0 0

PTOE – – – –

DIEOE PCINT19 • PCIE2

PCINT18 • PCIE2

PCINT17 • PCIE2

PCINT16 • PCIE2

DIEOV 1 1 1 1

DI PCINT19 INPUT PCINT18 INPUT PCINT17 INPUT

PCINT16 INPUT

AIO ADC11 INPUT ADC10INPUT ADC9 INPUT ADC8 INPUT

110 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Alternate Functions of Port L The Port L alternate pin configuration is as follows:

• OC5C – Port L, Bit 5

OC5C, Output Compare Match C output: The PL5 pin can serve as an external outputfor the Timer/Counter5 Output Compare C. The pin has to be configured as an output(DDL5 set) to serve this function. The OC5C pin is also the output pin for the PWMmode timer function.

• OC5B – Port L, Bit 4

OC5B, Output Compare Match B output: The PL4 pin can serve as an external outputfor the Timer/Counter 5 Output Compare B. The pin has to be configured as an output(DDL4 set) to serve this function. The OC5B pin is also the output pin for the PWMmode timer function.

• OC5A – Port L, Bit 3

OC5A, Output Compare Match A output: The PL3 pin can serve as an external outputfor the Timer/Counter 5 Output Compare A. The pin has to be configured as an output(DDL3 set) to serve this function. The OC5A pin is also the output pin for the PWMmode timer function.

• T5 – Port L, Bit 2

T5, Timer/Counter5 counter source.

• ICP5 – Port L, Bit 1

ICP5, Input Capture Pin 5: The PL1 pin can serve as an Input Capture pin forTimer/Counter5.

• ICP4 – Port L, Bit 0

ICP4, Input Capture Pin 4: The PL0 pin can serve as an Input Capture pin forTimer/Counter4.

Table 66. Port L Pins Alternate Functions

Port Pin Alternate Function

PL7 –

PL6 –

PL5 OC5C (Output Compare and PWM Output C for Timer/Counter5)

PL4 OC5B (Output Compare and PWM Output B for Timer/Counter5)

PL3 OC5A (Output Compare and PWM Output A for Timer/Counter5)

PL2 T5 (Timer/Counter5 Clock Input)

PL1 ICP5 (Timer/Counter5 Input Capture Trigger)

PL0 ICP4 (Timer/Counter4 Input Capture Trigger)

111

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 67 and Table 68 relates the alternate functions of Port L to the overriding signalsshown in Figure 37 on page 86.

Table 67. Overriding Signals for Alternate Functions in PL7..PL4

Signal Name PL7 PL6 PL5/OC5C PL4/OC5B

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE – – 0 0

DDOV – – 0 0

PVOE – – OC5C ENABLE OC5B ENABLE

PVOV – – OC5C OC5B

PTOE – – – –

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI 0 0 0 0

AIO – – – –

Table 68. Overriding Signals for Alternate Functions in PL3..PL0

Signal Name PL3/OC5A PL2/T5 PL1/ICP5 PL0/ICP4

PUOE 0 0 0 0

PUOV 0 0 0 0

DDOE 0 0 0 0

DDOV 0 0 0 0

PVOE OC5A ENABLE 0 0 0

PVOV OC5A 0 0 0

PTOE – – – –

DIEOE 0 0 0 0

DIEOV 0 0 0 0

DI 0 T5 INPUT ICP5 INPUT ICP4 INPUT

AIO – – – –

112 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Register Description for I/O-Ports

Port A Data Register – PORTA

Port A Data Direction Register – DDRA

Port A Input Pins Address – PINA

Port B Data Register – PORTB

Port B Data Direction Register – DDRB

Port B Input Pins Address – PINB

Port C Data Register – PORTC

Port C Data Direction Register – DDRC

Port C Input Pins Address – PINC

Bit 7 6 5 4 3 2 1 0

PORTA7 PORTA6 PORTA5 PORTA4 PORTA3 PORTA2 PORTA1 PORTA0 PORTA

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDA7 DDA6 DDA5 DDA4 DDA3 DDA2 DDA1 DDA0 DDRA

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PINA7 PINA6 PINA5 PINA4 PINA3 PINA2 PINA1 PINA0 PINA

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

Bit 7 6 5 4 3 2 1 0

PORTB7 PORTB6 PORTB5 PORTB4 PORTB3 PORTB2 PORTB1 PORTB0 PORTB

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDB7 DDB6 DDB5 DDB4 DDB3 DDB2 DDB1 DDB0 DDRB

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PINB7 PINB6 PINB5 PINB4 PINB3 PINB2 PINB1 PINB0 PINB

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

Bit 7 6 5 4 3 2 1 0

PORTC7 PORTC6 PORTC5 PORTC4 PORTC3 PORTC2 PORTC1 PORTC0 PORTC

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDC7 DDC6 DDC5 DDC4 DDC3 DDC2 DDC1 DDC0 DDRC

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PINC7 PINC6 PINC5 PINC4 PINC3 PINC2 PINC1 PINC0 PINC

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

113

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Port D Data Register – PORTD

Port D Data Direction Register – DDRD

Port D Input Pins Address – PIND

Port E Data Register – PORTE

Port E Data Direction Register – DDRE

Port E Input Pins Address – PINE

Port F Data Register – PORTF

Port F Data Direction Register – DDRF

Port F Input Pins Address – PINF

Bit 7 6 5 4 3 2 1 0

PORTD7 PORTD6 PORTD5 PORTD4 PORTD3 PORTD2 PORTD1 PORTD0 PORTD

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDD7 DDD6 DDD5 DDD4 DDD3 DDD2 DDD1 DDD0 DDRD

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PIND7 PIND6 PIND5 PIND4 PIND3 PIND2 PIND1 PIND0 PIND

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

Bit 7 6 5 4 3 2 1 0

PORTE7 PORTE6 PORTE5 PORTE4 PORTE3 PORTE2 PORTE1 PORTE0 PORTE

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDE7 DDE6 DDE5 DDE4 DDE3 DDE2 DDE1 DDE0 DDRE

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PINE7 PINE6 PINE5 PINE4 PINE3 PINE2 PINE1 PINE0 PINE

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

Bit 7 6 5 4 3 2 1 0

PORTF7 PORTF6 PORTF5 PORTF4 PORTF3 PORTF2 PORTF1 PORTF0 PORTF

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDF7 DDF6 DDF5 DDF4 DDF3 DDF2 DDF1 DDF0 DDRF

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PINF7 PINF6 PINF5 PINF4 PINF3 PINF2 PINF1 PINF0 PINF

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

114 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Port G Data Register – PORTG

Port G Data Direction Register – DDRG

Port G Input Pins Address – PING

Port H Data Register – PORTH

Port H Data Direction Register – DDRH

Port H Input Pins Address – PINH

Port J Data Register – PORTJ

Port J Data Direction Register – DDRJ

Port J Input Pins Address – PINJ

Bit 7 6 5 4 3 2 1 0

– – PORTG5 PORTG4 PORTG3 PORTG2 PORTG1 PORTG0 PORTG

Read/Write R R R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – DDG5 DDG4 DDG3 DDG2 DDG1 DDG0 DDRG

Read/Write R R R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – PING5 PING4 PING3 PING2 PING1 PING0 PING

Read/Write R R R/W R/W R/W R/W R/W R/W

Initial Value 0 0 N/A N/A N/A N/A N/A N/A

Bit 7 6 5 4 3 2 1 0

PORTH7 PORTH6 PORTH5 PORTH4 PORTH3 PORTH2 PORTH1 PORTH0 PORTH

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDH7 DDH6 DDH5 DDH4 DDH3 DDH2 DDH1 DDH0 DDRH

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PINH5 PINH5 PINH5 PINH4 PINH3 PINGH PINH1 PINH0 PINH

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

Bit 7 6 5 4 3 2 1 0

PORTJ7 PORTJ6 PORTJ5 PORTJ4 PORTJ3 PORTJ2 PORTJ1 PORTJ0 PORTJ

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDJ7 DDJ6 DDJ5 DDJ4 DDJ3 DDJ2 DDJ1 DDJ0 DDRJ

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PINJ5 PINJ5 PINJ5 PINJ4 PINJ3 PINGJ PINJ1 PINJ0 PINJ

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

115

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Port K Data Register – PORTK

Port K Data Direction Register – DDRK

Port K Input Pins Address – PINK

Port L Data Register – PORTL

Port L Data Direction Register – DDRL

Port L Input Pins Address – PINL

Bit 7 6 5 4 3 2 1 0

PORTK7 PORTK6 PORTK5 PORTK4 PORTK3 PORTK2 PORTK1 PORTK0 PORTK

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDK7 DDK6 DDK5 DDK4 DDK3 DDK2 DDK1 DDK0 DDRK

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PINK5 PINK5 PINK5 PINK4 PINK3 PINGK PINK1 PINK0 PINK

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

Bit 7 6 5 4 3 2 1 0

PORTL7 PORTL6 PORTL5 PORTL4 PORTL3 PORTL2 PORTL1 PORTL0 PORTL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

DDL7 DDL6 DDL5 DDL4 DDL3 DDL2 DDL1 DDL0 DDRL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

PINL5 PINL5 PINL5 PINL4 PINL3 PINGL PINL1 PINL0 PINL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value N/A N/A N/A N/A N/A N/A N/A N/A

116 ATmega640/1280/1281/2560/25612549A–AVR–03/05

117

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

8-bit Timer/Counter0 with PWM

Timer/Counter0 is a general purpose 8-bit Timer/Counter module, with two independentOutput Compare Units, and with PWM support. It allows accurate program executiontiming (event management) and wave generation. The main features are:• Two Independent Output Compare Units• Double Buffered Output Compare Registers• Clear Timer on Compare Match (Auto Reload)• Glitch Free, Phase Correct Pulse Width Modulator (PWM)• Variable PWM Period• Frequency Generator• Three Independent Interrupt Sources (TOV0, OCF0A, and OCF0B)

Overview A simplified block diagram of the 8-bit Timer/Counter is shown in Figure 38. For theactual placement of I/O pins, refer to “Pinout ATmega640/1280/2560” on page 2. CPUaccessible I/O Registers, including I/O bits and I/O pins, are shown in bold. The device-specific I/O Register and bit locations are listed in the “8-bit Timer/Counter RegisterDescription” on page 128.

Figure 38. 8-bit Timer/Counter Block Diagram

Registers The Timer/Counter (TCNT0) and Output Compare Registers (OCR0A and OCR0B) are8-bit registers. Interrupt request (abbreviated to Int.Req. in the figure) signals are all vis-ible in the Timer Interrupt Flag Register (TIFR0). All interrupts are individually maskedwith the Timer Interrupt Mask Register (TIMSK0). TIFR0 and TIMSK0 are not shown inthe figure.

The Timer/Counter can be clocked internally, via the prescaler, or by an external clocksource on the T0 pin. The Clock Select logic block controls which clock source and edgethe Timer/Counter uses to increment (or decrement) its value. The Timer/Counter isinactive when no clock source is selected. The output from the Clock Select logic isreferred to as the timer clock (clkT0).

The double buffered Output Compare Registers (OCR0A and OCR0B) are comparedwith the Timer/Counter value at all times. The result of the compare can be used by theWaveform Generator to generate a PWM or variable frequency output on the Output

Clock Select

Timer/Counter

DAT

A B

US

OCRnA

OCRnB

=

=

TCNTn

WaveformGeneration

WaveformGeneration

OCnA

OCnB

=

FixedTOP

Value

Control Logic

= 0

TOP BOTTOM

Count

Clear

Direction

TOVn(Int.Req.)

OCnA(Int.Req.)

OCnB(Int.Req.)

TCCRnA TCCRnB

TnEdge

Detector

( From Prescaler )

clkTn

118 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Compare pins (OC0A and OC0B). See “Output Compare Unit” on page 119. for details.The Compare Match event will also set the Compare Flag (OCF0A or OCF0B) whichcan be used to generate an Output Compare interrupt request.

Definitions Many register and bit references in this section are written in general form. A lower case“n” replaces the Timer/Counter number, in this case 0. A lower case “x” replaces theOutput Compare Unit, in this case Compare Unit A or Compare Unit B. However, whenusing the register or bit defines in a program, the precise form must be used, i.e.,TCNT0 for accessing Timer/Counter0 counter value and so on.

The definitions in Table 69 are also used extensively throughout the document.

Timer/Counter Clock Sources

The Timer/Counter can be clocked by an internal or an external clock source. The clocksource is selected by the Clock Select logic which is controlled by the Clock Select(CS02:0) bits located in the Timer/Counter Control Register (TCCR0B). For details onclock sources and prescaler, see “Timer/Counter0, Timer/Counter1, Timer/Counter3,Timer/Counter4, and Timer/Counter5 Prescalers” on page 169.

Counter Unit The main part of the 8-bit Timer/Counter is the programmable bi-directional counter unit.Figure 39 shows a block diagram of the counter and its surroundings.

Figure 39. Counter Unit Block Diagram

Signal description (internal signals):

count Increment or decrement TCNT0 by 1.

direction Select between increment and decrement.

clear Clear TCNT0 (set all bits to zero).

clkTn Timer/Counter clock, referred to as clkT0 in the following.

top Signalize that TCNT0 has reached maximum value.

bottom Signalize that TCNT0 has reached minimum value (zero).

Table 69. Definitions

BOTTOM The counter reaches the BOTTOM when it becomes 0x00.

MAX The counter reaches its MAXimum when it becomes 0xFF (decimal 255).

TOP The counter reaches the TOP when it becomes equal to the highestvalue in the count sequence. The TOP value can be assigned to be thefixed value 0xFF (MAX) or the value stored in the OCR0A Register. Theassignment is dependent on the mode of operation.

DATA BUS

TCNTn Control Logic

count

TOVn(Int.Req.)

Clock Select

top

TnEdge

Detector

( From Prescaler )

clkTn

bottom

direction

clear

119

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Depending of the mode of operation used, the counter is cleared, incremented, or dec-remented at each timer clock (clkT0). clkT0 can be generated from an external or internalclock source, selected by the Clock Select bits (CS02:0). When no clock source isselected (CS02:0 = 0) the timer is stopped. However, the TCNT0 value can be accessedby the CPU, regardless of whether clkT0 is present or not. A CPU write overrides (haspriority over) all counter clear or count operations.

The counting sequence is determined by the setting of the WGM01 and WGM00 bitslocated in the Timer/Counter Control Register (TCCR0A) and the WGM02 bit located inthe Timer/Counter Control Register B (TCCR0B). There are close connections betweenhow the counter behaves (counts) and how waveforms are generated on the OutputCompare outputs OC0A and OC0B. For more details about advanced countingsequences and waveform generation, see “Modes of Operation” on page 121.

The Timer/Counter Overflow Flag (TOV0) is set according to the mode of operationselected by the WGM02:0 bits. TOV0 can be used for generating a CPU interrupt.

Output Compare Unit The 8-bit comparator continuously compares TCNT0 with the Output Compare Regis-ters (OCR0A and OCR0B). Whenever TCNT0 equals OCR0A or OCR0B, thecomparator signals a match. A match will set the Output Compare Flag (OCF0A orOCF0B) at the next timer clock cycle. If the corresponding interrupt is enabled, the Out-put Compare Flag generates an Output Compare interrupt. The Output Compare Flag isautomatically cleared when the interrupt is executed. Alternatively, the flag can becleared by software by writing a logical one to its I/O bit location. The Waveform Gener-ator uses the match signal to generate an output according to operating mode set by theWGM02:0 bits and Compare Output mode (COM0x1:0) bits. The max and bottom sig-nals are used by the Waveform Generator for handling the special cases of the extremevalues in some modes of operation (“Modes of Operation” on page 121).

Figure 40 shows a block diagram of the Output Compare unit.

Figure 40. Output Compare Unit, Block Diagram

OCFnx (Int.Req.)

= (8-bit Comparator )

OCRnx

OCnx

DATA BUS

TCNTn

WGMn1:0

Waveform Generator

top

FOCn

COMnX1:0

bottom

120 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The OCR0x Registers are double buffered when using any of the Pulse Width Modula-tion (PWM) modes. For the normal and Clear Timer on Compare (CTC) modes ofoperation, the double buffering is disabled. The double buffering synchronizes theupdate of the OCR0x Compare Registers to either top or bottom of the countingsequence. The synchronization prevents the occurrence of odd-length, non-symmetricalPWM pulses, thereby making the output glitch-free.

The OCR0x Register access may seem complex, but this is not case. When the doublebuffering is enabled, the CPU has access to the OCR0x Buffer Register, and if doublebuffering is disabled the CPU will access the OCR0x directly.

Force Output Compare In non-PWM waveform generation modes, the match output of the comparator can beforced by writing a one to the Force Output Compare (FOC0x) bit. Forcing CompareMatch will not set the OCF0x Flag or reload/clear the timer, but the OC0x pin will beupdated as if a real Compare Match had occurred (the COM0x1:0 bits settings definewhether the OC0x pin is set, cleared or toggled).

Compare Match Blocking by TCNT0 Write

All CPU write operations to the TCNT0 Register will block any Compare Match thatoccur in the next timer clock cycle, even when the timer is stopped. This feature allowsOCR0x to be initialized to the same value as TCNT0 without triggering an interrupt whenthe Timer/Counter clock is enabled.

Using the Output Compare Unit

Since writing TCNT0 in any mode of operation will block all Compare Matches for onetimer clock cycle, there are risks involved when changing TCNT0 when using the OutputCompare Unit, independently of whether the Timer/Counter is running or not. If thevalue written to TCNT0 equals the OCR0x value, the Compare Match will be missed,resulting in incorrect waveform generation. Similarly, do not write the TCNT0 valueequal to BOTTOM when the counter is down-counting.

The setup of the OC0x should be performed before setting the Data Direction Registerfor the port pin to output. The easiest way of setting the OC0x value is to use the ForceOutput Compare (FOC0x) strobe bits in Normal mode. The OC0x Registers keep theirvalues even when changing between Waveform Generation modes.

Be aware that the COM0x1:0 bits are not double buffered together with the comparevalue. Changing the COM0x1:0 bits will take effect immediately.

Compare Match Output Unit

The Compare Output mode (COM0x1:0) bits have two functions. The Waveform Gener-ator uses the COM0x1:0 bits for defining the Output Compare (OC0x) state at the nextCompare Match. Also, the COM0x1:0 bits control the OC0x pin output source. Figure 41shows a simplified schematic of the logic affected by the COM0x1:0 bit setting. The I/ORegisters, I/O bits, and I/O pins in the figure are shown in bold. Only the parts of thegeneral I/O Port Control Registers (DDR and PORT) that are affected by the COM0x1:0bits are shown. When referring to the OC0x state, the reference is for the internal OC0xRegister, not the OC0x pin. If a system reset occur, the OC0x Register is reset to “0”.

121

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 41. Compare Match Output Unit, Schematic

The general I/O port function is overridden by the Output Compare (OC0x) from theWaveform Generator if either of the COM0x1:0 bits are set. However, the OC0x pindirection (input or output) is still controlled by the Data Direction Register (DDR) for theport pin. The Data Direction Register bit for the OC0x pin (DDR_OC0x) must be set asoutput before the OC0x value is visible on the pin. The port override function is indepen-dent of the Waveform Generation mode.

The design of the Output Compare pin logic allows initialization of the OC0x state beforethe output is enabled. Note that some COM0x1:0 bit settings are reserved for certainmodes of operation. See “8-bit Timer/Counter Register Description” on page 128.

Compare Output Mode and Waveform Generation

The Waveform Generator uses the COM0x1:0 bits differently in Normal, CTC, and PWMmodes. For all modes, setting the COM0x1:0 = 0 tells the Waveform Generator that noaction on the OC0x Register is to be performed on the next Compare Match. For com-pare output actions in the non-PWM modes refer to Table 70 on page 128. For fastPWM mode, refer to Table 71 on page 128, and for phase correct PWM refer to Table72 on page 129.

A change of the COM0x1:0 bits state will have effect at the first Compare Match afterthe bits are written. For non-PWM modes, the action can be forced to have immediateeffect by using the FOC0x strobe bits.

Modes of Operation The mode of operation, i.e., the behavior of the Timer/Counter and the Output Comparepins, is defined by the combination of the Waveform Generation mode (WGM02:0) andCompare Output mode (COM0x1:0) bits. The Compare Output mode bits do not affectthe counting sequence, while the Waveform Generation mode bits do. The COM0x1:0bits control whether the PWM output generated should be inverted or not (inverted ornon-inverted PWM). For non-PWM modes the COM0x1:0 bits control whether the out-put should be set, cleared, or toggled at a Compare Match (See “Compare MatchOutput Unit” on page 146.).

For detailed timing information see “Timer/Counter Timing Diagrams” on page 126.

PORT

DDR

D Q

D Q

OCnxPinOCnx

D QWaveformGenerator

COMnx1

COMnx0

0

1

DAT

A B

US

FOCn

clkI/O

122 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Normal Mode The simplest mode of operation is the Normal mode (WGM02:0 = 0). In this mode thecounting direction is always up (incrementing), and no counter clear is performed. Thecounter simply overruns when it passes its maximum 8-bit value (TOP = 0xFF) and thenrestarts from the bottom (0x00). In normal operation the Timer/Counter Overflow Flag(TOV0) will be set in the same timer clock cycle as the TCNT0 becomes zero. TheTOV0 Flag in this case behaves like a ninth bit, except that it is only set, not cleared.However, combined with the timer overflow interrupt that automatically clears the TOV0Flag, the timer resolution can be increased by software. There are no special cases toconsider in the Normal mode, a new counter value can be written anytime.

The Output Compare Unit can be used to generate interrupts at some given time. Usingthe Output Compare to generate waveforms in Normal mode is not recommended,since this will occupy too much of the CPU time.

Clear Timer on Compare Match (CTC) Mode

In Clear Timer on Compare or CTC mode (WGM02:0 = 2), the OCR0A Register is usedto manipulate the counter resolution. In CTC mode the counter is cleared to zero whenthe counter value (TCNT0) matches the OCR0A. The OCR0A defines the top value forthe counter, hence also its resolution. This mode allows greater control of the CompareMatch output frequency. It also simplifies the operation of counting external events.

The timing diagram for the CTC mode is shown in Figure 42. The counter value(TCNT0) increases until a Compare Match occurs between TCNT0 and OCR0A, andthen counter (TCNT0) is cleared.

Figure 42. CTC Mode, Timing Diagram

An interrupt can be generated each time the counter value reaches the TOP value byusing the OCF0A Flag. If the interrupt is enabled, the interrupt handler routine can beused for updating the TOP value. However, changing TOP to a value close to BOTTOMwhen the counter is running with none or a low prescaler value must be done with caresince the CTC mode does not have the double buffering feature. If the new value writtento OCR0A is lower than the current value of TCNT0, the counter will miss the CompareMatch. The counter will then have to count to its maximum value (0xFF) and wraparound starting at 0x00 before the Compare Match can occur.

For generating a waveform output in CTC mode, the OC0A output can be set to toggleits logical level on each Compare Match by setting the Compare Output mode bits totoggle mode (COM0A1:0 = 1). The OC0A value will not be visible on the port pin unlessthe data direction for the pin is set to output. The waveform generated will have a maxi-

TCNTn

OCn(Toggle)

OCnx Interrupt Flag Set

1 4Period 2 3

(COMnx1:0 = 1)

123

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

mum frequency of fOC0 = fclk_I/O/2 when OCR0A is set to zero (0x00). The waveformfrequency is defined by the following equation:

The N variable represents the prescale factor (1, 8, 64, 256, or 1024).

As for the Normal mode of operation, the TOV0 Flag is set in the same timer clock cyclethat the counter counts from MAX to 0x00.

Fast PWM Mode The fast Pulse Width Modulation or fast PWM mode (WGM02:0 = 3 or 7) provides a highfrequency PWM waveform generation option. The fast PWM differs from the other PWMoption by its single-slope operation. The counter counts from BOTTOM to TOP thenrestarts from BOTTOM. TOP is defined as 0xFF when WGM2:0 = 3, and OCR0A whenWGM2:0 = 7. In non-inverting Compare Output mode, the Output Compare (OC0x) iscleared on the Compare Match between TCNT0 and OCR0x, and set at BOTTOM. Ininverting Compare Output mode, the output is set on Compare Match and cleared atBOTTOM. Due to the single-slope operation, the operating frequency of the fast PWMmode can be twice as high as the phase correct PWM mode that use dual-slope opera-tion. This high frequency makes the fast PWM mode well suited for power regulation,rectification, and DAC applications. High frequency allows physically small sized exter-nal components (coils, capacitors), and therefore reduces total system cost.

In fast PWM mode, the counter is incremented until the counter value matches the TOPvalue. The counter is then cleared at the following timer clock cycle. The timing diagramfor the fast PWM mode is shown in Figure 43. The TCNT0 value is in the timing diagramshown as a histogram for illustrating the single-slope operation. The diagram includesnon-inverted and inverted PWM outputs. The small horizontal line marks on the TCNT0slopes represent Compare Matches between OCR0x and TCNT0.

Figure 43. Fast PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOV0) is set each time the counter reaches TOP. Ifthe interrupt is enabled, the interrupt handler routine can be used for updating the com-pare value.

In fast PWM mode, the compare unit allows generation of PWM waveforms on theOC0x pins. Setting the COM0x1:0 bits to two will produce a non-inverted PWM and an

fOCnxfclk_I/O

2 N 1 OCRnx+( )⋅ ⋅--------------------------------------------------=

TCNTn

OCRnx Update andTOVn Interrupt Flag Set

1Period 2 3

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)

OCRnx Interrupt Flag Set

4 5 6 7

124 ATmega640/1280/1281/2560/25612549A–AVR–03/05

inverted PWM output can be generated by setting the COM0x1:0 to three: Setting theCOM0A1:0 bits to one allows the OC0A pin to toggle on Compare Matches if theWGM02 bit is set. This option is not available for the OC0B pin (See Table 71 on page128). The actual OC0x value will only be visible on the port pin if the data direction forthe port pin is set as output. The PWM waveform is generated by setting (or clearing)the OC0x Register at the Compare Match between OCR0x and TCNT0, and clearing (orsetting) the OC0x Register at the timer clock cycle the counter is cleared (changes fromTOP to BOTTOM).

The PWM frequency for the output can be calculated by the following equation:

The N variable represents the prescale factor (1, 8, 64, 256, or 1024).

The extreme values for the OCR0A Register represents special cases when generatinga PWM waveform output in the fast PWM mode. If the OCR0A is set equal to BOTTOM,the output will be a narrow spike for each MAX+1 timer clock cycle. Setting the OCR0Aequal to MAX will result in a constantly high or low output (depending on the polarity ofthe output set by the COM0A1:0 bits.)

A frequency (with 50% duty cycle) waveform output in fast PWM mode can be achievedby setting OC0x to toggle its logical level on each Compare Match (COM0x1:0 = 1). Thewaveform generated will have a maximum frequency of fOC0 = fclk_I/O/2 when OCR0A isset to zero. This feature is similar to the OC0A toggle in CTC mode, except the doublebuffer feature of the Output Compare unit is enabled in the fast PWM mode.

Phase Correct PWM Mode The phase correct PWM mode (WGM02:0 = 1 or 5) provides a high resolution phasecorrect PWM waveform generation option. The phase correct PWM mode is based on adual-slope operation. The counter counts repeatedly from BOTTOM to TOP and thenfrom TOP to BOTTOM. TOP is defined as 0xFF when WGM2:0 = 1, and OCR0A whenWGM2:0 = 5. In non-inverting Compare Output mode, the Output Compare (OC0x) iscleared on the Compare Match between TCNT0 and OCR0x while upcounting, and seton the Compare Match while down-counting. In inverting Output Compare mode, theoperation is inverted. The dual-slope operation has lower maximum operation frequencythan single slope operation. However, due to the symmetric feature of the dual-slopePWM modes, these modes are preferred for motor control applications.

In phase correct PWM mode the counter is incremented until the counter value matchesTOP. When the counter reaches TOP, it changes the count direction. The TCNT0 valuewill be equal to TOP for one timer clock cycle. The timing diagram for the phase correctPWM mode is shown on Figure 44. The TCNT0 value is in the timing diagram shown asa histogram for illustrating the dual-slope operation. The diagram includes non-invertedand inverted PWM outputs. The small horizontal line marks on the TCNT0 slopes repre-sent Compare Matches between OCR0x and TCNT0.

fOCnxPWMfclk_I/O

N 256⋅------------------=

125

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 44. Phase Correct PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOV0) is set each time the counter reaches BOT-TOM. The Interrupt Flag can be used to generate an interrupt each time the counterreaches the BOTTOM value.

In phase correct PWM mode, the compare unit allows generation of PWM waveforms onthe OC0x pins. Setting the COM0x1:0 bits to two will produce a non-inverted PWM. Aninverted PWM output can be generated by setting the COM0x1:0 to three: Setting theCOM0A0 bits to one allows the OC0A pin to toggle on Compare Matches if the WGM02bit is set. This option is not available for the OC0B pin (See Table 72 on page 129). Theactual OC0x value will only be visible on the port pin if the data direction for the port pinis set as output. The PWM waveform is generated by clearing (or setting) the OC0xRegister at the Compare Match between OCR0x and TCNT0 when the counter incre-ments, and setting (or clearing) the OC0x Register at Compare Match between OCR0xand TCNT0 when the counter decrements. The PWM frequency for the output whenusing phase correct PWM can be calculated by the following equation:

The N variable represents the prescale factor (1, 8, 64, 256, or 1024).

The extreme values for the OCR0A Register represent special cases when generating aPWM waveform output in the phase correct PWM mode. If the OCR0A is set equal toBOTTOM, the output will be continuously low and if set equal to MAX the output will becontinuously high for non-inverted PWM mode. For inverted PWM the output will havethe opposite logic values.

At the very start of period 2 in Figure 44 OCnx has a transition from high to low eventhough there is no Compare Match. The point of this transition is to guarantee symmetryaround BOTTOM. There are two cases that give a transition without Compare Match.

• OCR0A changes its value from MAX, like in Figure 44. When the OCR0A value is MAX the OCn pin value is the same as the result of a down-counting Compare

TOVn Interrupt Flag Set

OCnx Interrupt Flag Set

1 2 3

TCNTn

Period

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)

OCRnx Update

fOCnxPCPWMfclk_I/O

N 510⋅------------------=

126 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Match. To ensure symmetry around BOTTOM the OCn value at MAX must correspond to the result of an up-counting Compare Match.

• The timer starts counting from a value higher than the one in OCR0A, and for that reason misses the Compare Match and hence the OCn change that would have happened on the way up.

Timer/Counter Timing Diagrams

The Timer/Counter is a synchronous design and the timer clock (clkT0) is thereforeshown as a clock enable signal in the following figures. The figures include informationon when Interrupt Flags are set. Figure 45 contains timing data for basic Timer/Counteroperation. The figure shows the count sequence close to the MAX value in all modesother than phase correct PWM mode.

Figure 45. Timer/Counter Timing Diagram, no Prescaling

Figure 46 shows the same timing data, but with the prescaler enabled.

Figure 46. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)

Figure 47 shows the setting of OCF0B in all modes and OCF0A in all modes exceptCTC mode and PWM mode, where OCR0A is TOP.

Figure 47. Timer/Counter Timing Diagram, Setting of OCF0x, with Prescaler (fclk_I/O/8)

clkTn(clkI/O/1)

TOVn

clkI/O

TCNTn MAX - 1 MAX BOTTOM BOTTOM + 1

TOVn

TCNTn MAX - 1 MAX BOTTOM BOTTOM + 1

clkI/O

clkTn(clkI/O/8)

OCFnx

OCRnx

TCNTn

OCRnx Value

OCRnx - 1 OCRnx OCRnx + 1 OCRnx + 2

clkI/O

clkTn(clkI/O/8)

127

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 48 shows the setting of OCF0A and the clearing of TCNT0 in CTC mode and fastPWM mode where OCR0A is TOP.

Figure 48. Timer/Counter Timing Diagram, Clear Timer on Compare Match mode, withPrescaler (fclk_I/O/8)

OCFnx

OCRnx

TCNTn(CTC)

TOP

TOP - 1 TOP BOTTOM BOTTOM + 1

clkI/O

clkTn(clkI/O/8)

128 ATmega640/1280/1281/2560/25612549A–AVR–03/05

8-bit Timer/Counter Register Description

Timer/Counter Control Register A – TCCR0A

• Bits 7:6 – COM01A:0: Compare Match Output A Mode

These bits control the Output Compare pin (OC0A) behavior. If one or both of theCOM0A1:0 bits are set, the OC0A output overrides the normal port functionality of theI/O pin it is connected to. However, note that the Data Direction Register (DDR) bit cor-responding to the OC0A pin must be set in order to enable the output driver.

When OC0A is connected to the pin, the function of the COM0A1:0 bits depends on theWGM02:0 bit setting. Table 70 shows the COM0A1:0 bit functionality when theWGM02:0 bits are set to a normal or CTC mode (non-PWM).

Table 71 shows the COM0A1:0 bit functionality when the WGM01:0 bits are set to fastPWM mode.

Note: 1. A special case occurs when OCR0A equals TOP and COM0A1 is set. In this case,the Compare Match is ignored, but the set or clear is done at TOP. See “Fast PWMMode” on page 123 for more details.

Bit 7 6 5 4 3 2 1 0

COM0A1 COM0A0 COM0B1 COM0B0 – – WGM01 WGM00 TCCR0A

Read/Write R/W R/W R/W R/W R R R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 70. Compare Output Mode, non-PWM Mode

COM0A1 COM0A0 Description

0 0 Normal port operation, OC0A disconnected.

0 1 Toggle OC0A on Compare Match

1 0 Clear OC0A on Compare Match

1 1 Set OC0A on Compare Match

Table 71. Compare Output Mode, Fast PWM Mode(1)

COM0A1 COM0A0 Description

0 0 Normal port operation, OC0A disconnected.

0 1 WGM02 = 0: Normal Port Operation, OC0A Disconnected.WGM02 = 1: Toggle OC0A on Compare Match.

1 0 Clear OC0A on Compare Match, set OC0A at TOP

1 1 Set OC0A on Compare Match, clear OC0A at TOP

129

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 72 shows the COM0A1:0 bit functionality when the WGM02:0 bits are set tophase correct PWM mode.

Note: 1. A special case occurs when OCR0A equals TOP and COM0A1 is set. In this case,the Compare Match is ignored, but the set or clear is done at TOP. See “Phase Cor-rect PWM Mode” on page 124 for more details.

• Bits 5:4 – COM0B1:0: Compare Match Output B Mode

These bits control the Output Compare pin (OC0B) behavior. If one or both of theCOM0B1:0 bits are set, the OC0B output overrides the normal port functionality of theI/O pin it is connected to. However, note that the Data Direction Register (DDR) bit cor-responding to the OC0B pin must be set in order to enable the output driver.

When OC0B is connected to the pin, the function of the COM0B1:0 bits depends on theWGM02:0 bit setting. Table 70 shows the COM0A1:0 bit functionality when theWGM02:0 bits are set to a normal or CTC mode (non-PWM).

Table 71 shows the COM0B1:0 bit functionality when the WGM02:0 bits are set to fastPWM mode.

Note: 1. A special case occurs when OCR0B equals TOP and COM0B1 is set. In this case,the Compare Match is ignored, but the set or clear is done at TOP. See “Fast PWMMode” on page 123 for more details.

Table 72. Compare Output Mode, Phase Correct PWM Mode(1)

COM0A1 COM0A0 Description

0 0 Normal port operation, OC0A disconnected.

0 1 WGM02 = 0: Normal Port Operation, OC0A Disconnected.WGM02 = 1: Toggle OC0A on Compare Match.

1 0 Clear OC0A on Compare Match when up-counting. Set OC0A on Compare Match when down-counting.

1 1 Set OC0A on Compare Match when up-counting. Clear OC0A on Compare Match when down-counting.

Table 73. Compare Output Mode, non-PWM Mode

COM01 COM00 Description

0 0 Normal port operation, OC0B disconnected.

0 1 Toggle OC0B on Compare Match

1 0 Clear OC0B on Compare Match

1 1 Set OC0B on Compare Match

Table 74. Compare Output Mode, Fast PWM Mode(1)

COM01 COM00 Description

0 0 Normal port operation, OC0B disconnected.

0 1 Reserved

1 0 Clear OC0B on Compare Match, set OC0B at TOP

1 1 Set OC0B on Compare Match, clear OC0B at TOP

130 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Table 72 shows the COM0B1:0 bit functionality when the WGM02:0 bits are set tophase correct PWM mode.

Note: 1. A special case occurs when OCR0B equals TOP and COM0B1 is set. In this case,the Compare Match is ignored, but the set or clear is done at TOP. See “Phase Cor-rect PWM Mode” on page 124 for more details.

• Bits 3, 2 – Res: Reserved Bits

These bits are reserved bits and will always read as zero.

• Bits 1:0 – WGM01:0: Waveform Generation Mode

Combined with the WGM02 bit found in the TCCR0B Register, these bits control thecounting sequence of the counter, the source for maximum (TOP) counter value, andwhat type of waveform generation to be used, see Table 76. Modes of operation sup-ported by the Timer/Counter unit are: Normal mode (counter), Clear Timer on CompareMatch (CTC) mode, and two types of Pulse Width Modulation (PWM) modes (see“Modes of Operation” on page 147).

Notes: 1. MAX = 0xFF 2. BOTTOM = 0x00

Table 75. Compare Output Mode, Phase Correct PWM Mode(1)

COM0A1 COM0A0 Description

0 0 Normal port operation, OC0B disconnected.

0 1 Reserved

1 0 Clear OC0B on Compare Match when up-counting. Set OC0B on Compare Match when down-counting.

1 1 Set OC0B on Compare Match when up-counting. Clear OC0B on Compare Match when down-counting.

Table 76. Waveform Generation Mode Bit Description

Mode WGM2 WGM1 WGM0

Timer/Counter Mode of Operation TOP

Update ofOCRx at

TOV FlagSet on(1)(2)

0 0 0 0 Normal 0xFF Immediate MAX

1 0 0 1 PWM, Phase Correct

0xFF TOP BOTTOM

2 0 1 0 CTC OCRA Immediate MAX

3 0 1 1 Fast PWM 0xFF TOP MAX

4 1 0 0 Reserved – – –

5 1 0 1 PWM, Phase Correct

OCRA TOP BOTTOM

6 1 1 0 Reserved – – –

7 1 1 1 Fast PWM OCRA TOP TOP

131

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Timer/Counter Control Register B – TCCR0B

• Bit 7 – FOC0A: Force Output Compare A

The FOC0A bit is only active when the WGM bits specify a non-PWM mode.

However, for ensuring compatibility with future devices, this bit must be set to zero whenTCCR0B is written when operating in PWM mode. When writing a logical one to theFOC0A bit, an immediate Compare Match is forced on the Waveform Generation unit.The OC0A output is changed according to its COM0A1:0 bits setting. Note that theFOC0A bit is implemented as a strobe. Therefore it is the value present in theCOM0A1:0 bits that determines the effect of the forced compare.

A FOC0A strobe will not generate any interrupt, nor will it clear the timer in CTC modeusing OCR0A as TOP.

The FOC0A bit is always read as zero.

• Bit 6 – FOC0B: Force Output Compare B

The FOC0B bit is only active when the WGM bits specify a non-PWM mode.

However, for ensuring compatibility with future devices, this bit must be set to zero whenTCCR0B is written when operating in PWM mode. When writing a logical one to theFOC0B bit, an immediate Compare Match is forced on the Waveform Generation unit.The OC0B output is changed according to its COM0B1:0 bits setting. Note that theFOC0B bit is implemented as a strobe. Therefore it is the value present in theCOM0B1:0 bits that determines the effect of the forced compare.

A FOC0B strobe will not generate any interrupt, nor will it clear the timer in CTC modeusing OCR0B as TOP.

The FOC0B bit is always read as zero.

• Bits 5:4 – Res: Reserved Bits

These bits are reserved bits and will always read as zero.

• Bit 3 – WGM02: Waveform Generation Mode

See the description in the “Timer/Counter Control Register A – TCCR0A” on page 128.

• Bits 2:0 – CS02:0: Clock Select

The three Clock Select bits select the clock source to be used by the Timer/Counter.

Bit 7 6 5 4 3 2 1 0

FOC0A FOC0B – – WGM02 CS02 CS01 CS00 TCCR0B

Read/Write W W R R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 77. Clock Select Bit Description

CS02 CS01 CS00 Description

0 0 0 No clock source (Timer/Counter stopped)

0 0 1 clkI/O/(No prescaling)

0 1 0 clkI/O/8 (From prescaler)

0 1 1 clkI/O/64 (From prescaler)

1 0 0 clkI/O/256 (From prescaler)

132 ATmega640/1280/1281/2560/25612549A–AVR–03/05

If external pin modes are used for the Timer/Counter0, transitions on the T0 pin willclock the counter even if the pin is configured as an output. This feature allows softwarecontrol of the counting.

Timer/Counter Register – TCNT0

The Timer/Counter Register gives direct access, both for read and write operations, tothe Timer/Counter unit 8-bit counter. Writing to the TCNT0 Register blocks (removes)the Compare Match on the following timer clock. Modifying the counter (TCNT0) whilethe counter is running, introduces a risk of missing a Compare Match between TCNT0and the OCR0x Registers.

Output Compare Register A – OCR0A

The Output Compare Register A contains an 8-bit value that is continuously comparedwith the counter value (TCNT0). A match can be used to generate an Output Compareinterrupt, or to generate a waveform output on the OC0A pin.

Output Compare Register B – OCR0B

The Output Compare Register B contains an 8-bit value that is continuously comparedwith the counter value (TCNT0). A match can be used to generate an Output Compareinterrupt, or to generate a waveform output on the OC0B pin.

Timer/Counter Interrupt Mask Register – TIMSK0

• Bits 7..3, 0 – Res: Reserved Bits

These bits are reserved bits and will always read as zero.

• Bit 2 – OCIE0B: Timer/Counter Output Compare Match B Interrupt Enable

When the OCIE0B bit is written to one, and the I-bit in the Status Register is set, theTimer/Counter Compare Match B interrupt is enabled. The corresponding interrupt is

1 0 1 clkI/O/1024 (From prescaler)

1 1 0 External clock source on T0 pin. Clock on falling edge.

1 1 1 External clock source on T0 pin. Clock on rising edge.

Table 77. Clock Select Bit Description (Continued)

CS02 CS01 CS00 Description

Bit 7 6 5 4 3 2 1 0

TCNT0[7:0] TCNT0

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR0A[7:0] OCR0A

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR0B[7:0] OCR0B

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – – – – OCIE0B OCIE0A TOIE0 TIMSK0

Read/Write R R R R R R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

133

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

executed if a Compare Match in Timer/Counter occurs, i.e., when the OCF0B bit is set inthe Timer/Counter Interrupt Flag Register – TIFR0.

• Bit 1 – OCIE0A: Timer/Counter0 Output Compare Match A Interrupt Enable

When the OCIE0A bit is written to one, and the I-bit in the Status Register is set, theTimer/Counter0 Compare Match A interrupt is enabled. The corresponding interrupt isexecuted if a Compare Match in Timer/Counter0 occurs, i.e., when the OCF0A bit is setin the Timer/Counter 0 Interrupt Flag Register – TIFR0.

• Bit 0 – TOIE0: Timer/Counter0 Overflow Interrupt Enable

When the TOIE0 bit is written to one, and the I-bit in the Status Register is set, theTimer/Counter0 Overflow interrupt is enabled. The corresponding interrupt is executed ifan overflow in Timer/Counter0 occurs, i.e., when the TOV0 bit is set in theTimer/Counter 0 Interrupt Flag Register – TIFR0.

Timer/Counter 0 Interrupt Flag Register – TIFR0

• Bits 7..3, 0 – Res: Reserved Bits

These bits are reserved bits and will always read as zero.

• Bit 2 – OCF0B: Timer/Counter 0 Output Compare B Match Flag

The OCF0B bit is set when a Compare Match occurs between the Timer/Counter andthe data in OCR0B – Output Compare Register0 B. OCF0B is cleared by hardwarewhen executing the corresponding interrupt handling vector. Alternatively, OCF0B iscleared by writ ing a logic one to the flag. When the I-bit in SREG, OCIE0B(Timer/Counter Compare B Match Interrupt Enable), and OCF0B are set, theTimer/Counter Compare Match Interrupt is executed.

• Bit 1 – OCF0A: Timer/Counter 0 Output Compare A Match Flag

The OCF0A bit is set when a Compare Match occurs between the Timer/Counter0 andthe data in OCR0A – Output Compare Register0. OCF0A is cleared by hardware whenexecuting the corresponding interrupt handling vector. Alternatively, OCF0A is clearedby writing a logic one to the flag. When the I-bit in SREG, OCIE0A (Timer/Counter0Compare Match Interrupt Enable), and OCF0A are set, the Timer/Counter0 CompareMatch Interrupt is executed.

• Bit 0 – TOV0: Timer/Counter0 Overflow Flag

The bit TOV0 is set when an overflow occurs in Timer/Counter0. TOV0 is cleared byhardware when executing the corresponding interrupt handling vector. Alternatively,TOV0 is cleared by writing a logic one to the flag. When the SREG I-bit, TOIE0(Timer/Counter0 Overflow Interrupt Enable), and TOV0 are set, the Timer/Counter0Overflow interrupt is executed.

The setting of this flag is dependent of the WGM02:0 bit setting. Refer to Table 76,“Waveform Generation Mode Bit Description” on page 130.

Bit 7 6 5 4 3 2 1 0

– – – – – OCF0B OCF0A TOV0 TIFR0

Read/Write R R R R R R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

134 ATmega640/1280/1281/2560/25612549A–AVR–03/05

135

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

16-bit Timer/Counter (Timer/Counter1, Timer/Counter3, Timer/Counter4 and Timer/Counter5)

The 16-bit Timer/Counter unit allows accurate program execution timing (event man-agement), wave generation, and signal timing measurement. The main features are:• True 16-bit Design (i.e., Allows 16-bit PWM)• Three independent Output Compare Units• Double Buffered Output Compare Registers• One Input Capture Unit• Input Capture Noise Canceler• Clear Timer on Compare Match (Auto Reload)• Glitch-free, Phase Correct Pulse Width Modulator (PWM)• Variable PWM Period• Frequency Generator• External Event Counter• Twenty independent interrupt sources (TOV1, OCF1A, OCF1B, OCF1C, ICF1, TOV3,

OCF3A, OCF3B, OCF3C, ICF3, TOV4, OCF4A, OCF4B, OCF4C, ICF4, TOV5, OCF5A, OCF5B, OCF5C and ICF5)

Overview Most register and bit references in this section are written in general form. A lower case“n” replaces the Timer/Counter number, and a lower case “x” replaces the Output Com-pare unit channel. However, when using the register or bit defines in a program, theprecise form must be used, i.e., TCNT1 for accessing Timer/Counter1 counter valueand so on.

A simplified block diagram of the 16-bit Timer/Counter is shown in Figure 49. For theactual placement of I/O pins, see “Pinout ATmega640/1280/2560” on page 2 and“Pinout ATmega1281/2561” on page 3. CPU accessible I/O Registers, including I/O bitsand I/O pins, are shown in bold. The device-specific I/O Register and bit locations arelisted in the “16-bit Timer/Counter Register Description” on page 157.

The Power Reduction Timer/Counter1 bit, PRTIM1, in “Power Reduction Register 0 -PRR0” on page 54 must be written to zero to enable Timer/Counter1 module.

The Power Reduction Timer/Counter3 bit, PRTIM3, in “Power Reduction Register 1 -PRR1” on page 55 must be written to zero to enable Timer/Counter3 module.

The Power Reduction Timer/Counter4 bit, PRTIM4, in “Power Reduction Register 1 -PRR1” on page 55 must be written to zero to enable Timer/Counter4 module.

The Power Reduction Timer/Counter5 bit, PRTIM5, in “Power Reduction Register 1 -PRR1” on page 55 must be written to zero to enable Timer/Counter5 module.

T imer /Counter4 and T imer /Counter5 on ly have fu l l func t iona l i t y in theATmega640/1280/2560.

136 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 49. 16-bit Timer/Counter Block Diagram(1)

Note: 1. Refer to Figure 1 on page 2, Table 38 on page 89, and Table 44 on page 93 forTimer/Counter1 and 3 and 3 pin placement and description.

Registers The Timer/Counter (TCNTn), Output Compare Registers (OCRnA/B/C), and Input Cap-ture Register (ICRn) are all 16-bit registers. Special procedures must be followed whenaccessing the 16-bit registers. These procedures are described in the section “Access-ing 16-bit Registers” on page 137. The Timer/Counter Control Registers (TCCRnA/B/C)are 8-bit registers and have no CPU access restrictions. Interrupt requests (shorten asInt.Req.) signals are all visible in the Timer Interrupt Flag Register (TIFRn). All interruptsare individually masked with the Timer Interrupt Mask Register (TIMSKn). TIFRn andTIMSKn are not shown in the figure since these registers are shared by other timerunits.

The Timer/Counter can be clocked internally, via the prescaler, or by an external clocksource on the Tn pin. The Clock Select logic block controls which clock source and edgethe Timer/Counter uses to increment (or decrement) its value. The Timer/Counter isinactive when no clock source is selected. The output from the clock select logic isreferred to as the timer clock (clkTn).

The double buffered Output Compare Registers (OCRnA/B/C) are compared with theTimer/Counter value at all time. The result of the compare can be used by the WaveformGenerator to generate a PWM or variable frequency output on the Output Compare pin(OCnA/B/C). See “Output Compare Units” on page 144.. The compare match event will

ICFn (Int.Req.)

TOVn

(Int.Req.)

Clock Select

Timer/Counter

DA

TA

BU

S

ICRn

=

=

=

TCNTn

Waveform

Generation

Waveform

Generation

Waveform

Generation

OCnA

OCnB

OCnC

Noise

CancelerICPn

=

Fixed

TOP

Values

Edge

Detector

Control Logic

= 0

TOP BOTTOM

Count

Clear

Direction

OCFnA

(Int.Req.)

OCFnB

(Int.Req.)

OCFnC

(Int.Req.)

TCCRnA TCCRnB TCCRnC

( From Analog

Comparator Ouput )

TnEdge

Detector

( From Prescaler )

TCLK

OCRnC

OCRnB

OCRnA

137

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

also set the Compare Match Flag (OCFnA/B/C) which can be used to generate an Out-put Compare interrupt request.

The Input Capture Register can capture the Timer/Counter value at a given external(edge triggered) event on either the Input Capture pin (ICPn) or on the Analog Compar-ator pins (See “Analog Comparator” on page 271.) The Input Capture unit includes adigital filtering unit (Noise Canceler) for reducing the chance of capturing noise spikes.

The TOP value, or maximum Timer/Counter value, can in some modes of operation bedefined by either the OCRnA Register, the ICRn Register, or by a set of fixed values.When using OCRnA as TOP value in a PWM mode, the OCRnA Register can not beused for generating a PWM output. However, the TOP value will in this case be doublebuffered allowing the TOP value to be changed in run time. If a fixed TOP value isrequired, the ICRn Register can be used as an alternative, freeing the OCRnA to beused as PWM output.

Definitions The following definitions are used extensively throughout the document:

Accessing 16-bit Registers

The TCNTn, OCRnA/B/C, and ICRn are 16-bit registers that can be accessed by theAVR CPU via the 8-bit data bus. The 16-bit register must be byte accessed using tworead or write operations. Each 16-bit timer has a single 8-bit register for temporary stor-ing of the high byte of the 16-bit access. The same Temporary Register is sharedbetween all 16-bit registers within each 16-bit timer. Accessing the low byte triggers the16-bit read or write operation. When the low byte of a 16-bit register is written by theCPU, the high byte stored in the Temporary Register, and the low byte written are bothcopied into the 16-bit register in the same clock cycle. When the low byte of a 16-bit reg-ister is read by the CPU, the high byte of the 16-bit register is copied into the TemporaryRegister in the same clock cycle as the low byte is read.

Not all 16-bit accesses uses the Temporary Register for the high byte. Reading theOCRnA/B/C 16-bit registers does not involve using the Temporary Register.

To do a 16-bit write, the high byte must be written before the low byte. For a 16-bit read,the low byte must be read before the high byte.

The following code examples show how to access the 16-bit timer registers assumingthat no interrupts updates the temporary register. The same principle can be useddirectly for accessing the OCRnA/B/C and ICRn Registers. Note that when using “C”,the compiler handles the 16-bit access.

Table 78. Definitions

BOTTOM The counter reaches the BOTTOM when it becomes 0x0000.

MAX The counter reaches its MAXimum when it becomes 0xFFFF (decimal 65535).

TOP The counter reaches the TOP when it becomes equal to the highest value in the count sequence. The TOP value can be assigned to be one of the fixed values: 0x00FF, 0x01FF, or 0x03FF, or to the value stored in the OCRnA or ICRn Register. The assignment is dependent of the mode of operation.

138 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. See “About Code Examples” on page 8.

The assembly code example returns the TCNTn value in the r17:r16 register pair.

It is important to notice that accessing 16-bit registers are atomic operations. If an inter-rupt occurs between the two instructions accessing the 16-bit register, and the interruptcode updates the temporary register by accessing the same or any other of the 16-bitTimer Registers, then the result of the access outside the interrupt will be corrupted.Therefore, when both the main code and the interrupt code update the temporary regis-ter, the main code must disable the interrupts during the 16-bit access.

Assembly Code Examples(1)

...

; Set TCNTn to 0x01FF

ldi r17,0x01

ldi r16,0xFF

out TCNTnH,r17

out TCNTnL,r16

; Read TCNTn into r17:r16

in r16,TCNTnL

in r17,TCNTnH

...

C Code Examples(1)

unsigned int i;

...

/* Set TCNTn to 0x01FF */

TCNTn = 0x1FF;

/* Read TCNTn into i */

i = TCNTn;

...

139

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The following code examples show how to do an atomic read of the TCNTn Registercontents. Reading any of the OCRnA/B/C or ICRn Registers can be done by using thesame principle.

Note: 1. See “About Code Examples” on page 8.

The assembly code example returns the TCNTn value in the r17:r16 register pair.

Assembly Code Example(1)

TIM16_ReadTCNTn:

; Save global interrupt flag

in r18,SREG

; Disable interrupts

cli

; Read TCNTn into r17:r16

in r16,TCNTnL

in r17,TCNTnH

; Restore global interrupt flag

out SREG,r18

ret

C Code Example(1)

unsigned int TIM16_ReadTCNTn( void )

unsigned char sreg;

unsigned int i;

/* Save global interrupt flag */

sreg = SREG;

/* Disable interrupts */

__disable_interrupt();

/* Read TCNTn into i */

i = TCNTn;

/* Restore global interrupt flag */

SREG = sreg;

return i;

140 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The following code examples show how to do an atomic write of the TCNTn Registercontents. Writing any of the OCRnA/B/C or ICRn Registers can be done by using thesame principle.

Note: 1. See “About Code Examples” on page 8.

The assembly code example requires that the r17:r16 register pair contains the value tobe written to TCNTn.

Reusing the Temporary High Byte Register

If writing to more than one 16-bit register where the high byte is the same for all registerswritten, then the high byte only needs to be written once. However, note that the samerule of atomic operation described previously also applies in this case.

Assembly Code Example(1)

TIM16_WriteTCNTn:

; Save global interrupt flag

in r18,SREG

; Disable interrupts

cli

; Set TCNTn to r17:r16

out TCNTnH,r17

out TCNTnL,r16

; Restore global interrupt flag

out SREG,r18

ret

C Code Example(1)

void TIM16_WriteTCNTn( unsigned int i )

unsigned char sreg;

unsigned int i;

/* Save global interrupt flag */

sreg = SREG;

/* Disable interrupts */

__disable_interrupt();

/* Set TCNTn to i */

TCNTn = i;

/* Restore global interrupt flag */

SREG = sreg;

141

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Timer/Counter Clock Sources

The Timer/Counter can be clocked by an internal or an external clock source. The clocksource is selected by the Clock Select logic which is controlled by the Clock Select(CSn2:0) bits located in the Timer/Counter control Register B (TCCRnB). For details onclock sources and prescaler, see “Timer/Counter0, Timer/Counter1, Timer/Counter3,Timer/Counter4, and Timer/Counter5 Prescalers” on page 169.

Counter Unit The main part of the 16-bit Timer/Counter is the programmable 16-bit bi-directionalcounter unit. Figure 50 shows a block diagram of the counter and its surroundings.

Figure 50. Counter Unit Block Diagram

Signal description (internal signals):

Count Increment or decrement TCNTn by 1.

Direction Select between increment and decrement.

Clear Clear TCNTn (set all bits to zero).

clkTn Timer/Counter clock.

TOP Signalize that TCNTn has reached maximum value.

BOTTOM Signalize that TCNTn has reached minimum value (zero).

The 16-bit counter is mapped into two 8-bit I/O memory locations: Counter High(TCNTnH) containing the upper eight bits of the counter, and Counter Low (TCNTnL)containing the lower eight bits. The TCNTnH Register can only be indirectly accessedby the CPU. When the CPU does an access to the TCNTnH I/O location, the CPUaccesses the high byte temporary register (TEMP). The temporary register is updatedwith the TCNTnH value when the TCNTnL is read, and TCNTnH is updated with thetemporary register value when TCNTnL is written. This allows the CPU to read or writethe entire 16-bit counter value within one clock cycle via the 8-bit data bus. It is impor-tant to notice that there are special cases of writing to the TCNTn Register when thecounter is counting that will give unpredictable results. The special cases are describedin the sections where they are of importance.

Depending on the mode of operation used, the counter is cleared, incremented, or dec-remented at each timer clock (clkTn). The clkTn can be generated from an external orinternal clock source, selected by the Clock Select bits (CSn2:0). When no clock sourceis selected (CSn2:0 = 0) the timer is stopped. However, the TCNTn value can beaccessed by the CPU, independent of whether clkTn is present or not. A CPU write over-rides (has priority over) all counter clear or count operations.

The counting sequence is determined by the setting of the Waveform Generation modebits (WGMn3:0) located in the Timer/Counter Control Registers A and B (TCCRnA andTCCRnB). There are close connections between how the counter behaves (counts) and

TEMP (8-bit)

DATA BUS (8-bit)

TCNTn (16-bit Counter)

TCNTnH (8-bit) TCNTnL (8-bit)Control Logic

Count

Clear

Direction

TOVn(Int.Req.)

Clock Select

TOP BOTTOM

TnEdge

Detector

( From Prescaler )

clkTn

142 ATmega640/1280/1281/2560/25612549A–AVR–03/05

how waveforms are generated on the Output Compare outputs OCnx. For more detailsabout advanced counting sequences and waveform generation, see “Modes of Opera-tion” on page 147.

The Timer/Counter Overflow Flag (TOVn) is set according to the mode of operationselected by the WGMn3:0 bits. TOVn can be used for generating a CPU interrupt.

Input Capture Unit The Timer/Counter incorporates an input capture unit that can capture external eventsand give them a time-stamp indicating time of occurrence. The external signal indicatingan event, or multiple events, can be applied via the ICPn pin or alternatively, for theTimer/Counter1 only, via the Analog Comparator unit. The time-stamps can then beused to calculate frequency, duty-cycle, and other features of the signal applied. Alter-natively the time-stamps can be used for creating a log of the events.

The Input Capture unit is illustrated by the block diagram shown in Figure 51. The ele-ments of the block diagram that are not directly a part of the input capture unit are grayshaded. The small “n” in register and bit names indicates the Timer/Counter number.

Figure 51. Input Capture Unit Block Diagram

Note: The Analog Comparator Output (ACO) can only trigger the Timer/Counter1 ICP – notTimer/Counter3, 4 or 5.

When a change of the logic level (an event) occurs on the Input Capture Pin (ICPn),alternatively on the analog Comparator output (ACO), and this change confirms to thesetting of the edge detector, a capture will be triggered. When a capture is triggered, the16-bit value of the counter (TCNTn) is written to the Input Capture Register (ICRn). TheInput Capture Flag (ICFn) is set at the same system clock as the TCNTn value is copiedinto ICRn Register. If enabled (TICIEn = 1), the input capture flag generates an inputcapture interrupt. The ICFn flag is automatically cleared when the interrupt is executed.Alternatively the ICFn flag can be cleared by software by writing a logical one to its I/Obit location.

ICFn (Int.Req.)

AnalogComparator

WRITE ICRn (16-bit Register)

ICRnH (8-bit)

NoiseCanceler

ICPn

EdgeDetector

TEMP (8-bit)

DATA BUS (8-bit)

ICRnL (8-bit)

TCNTn (16-bit Counter)

TCNTnH (8-bit) TCNTnL (8-bit)

ACIC* ICNC ICESACO*

143

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Reading the 16-bit value in the Input Capture Register (ICRn) is done by first reading thelow byte (ICRnL) and then the high byte (ICRnH). When the low byte is read the highbyte is copied into the high byte Temporary Register (TEMP). When the CPU reads theICRnH I/O location it will access the TEMP Register.

The ICRn Register can only be written when using a Waveform Generation mode thatutilizes the ICRn Register for defining the counter’s TOP value. In these cases theWaveform Generation mode (WGMn3:0) bits must be set before the TOP value can bewritten to the ICRn Register. When writing the ICRn Register the high byte must be writ-ten to the ICRnH I/O location before the low byte is written to ICRnL.

For more information on how to access the 16-bit registers refer to “Accessing 16-bitRegisters” on page 137.

Input Capture Trigger Source The main trigger source for the input capture unit is the Input Capture Pin (ICPn).Timer/Counter1 can alternatively use the analog comparator output as trigger source forthe input capture unit. The Analog Comparator is selected as trigger source by settingthe analog Comparator Input Capture (ACIC) bit in the Analog Comparator Control andStatus Register (ACSR). Be aware that changing trigger source can trigger a capture.The input capture flag must therefore be cleared after the change.

Both the Input Capture Pin (ICPn) and the Analog Comparator output (ACO) inputs aresampled using the same technique as for the Tn pin (Figure 62 on page 169). The edgedetector is also identical. However, when the noise canceler is enabled, additional logicis inserted before the edge detector, which increases the delay by four system clockcycles. Note that the input of the noise canceler and edge detector is always enabledunless the Timer/Counter is set in a Waveform Generation mode that uses ICRn todefine TOP.

An input capture can be triggered by software by controlling the port of the ICPn pin.

Noise Canceler The noise canceler improves noise immunity by using a simple digital filtering scheme.The noise canceler input is monitored over four samples, and all four must be equal forchanging the output that in turn is used by the edge detector.

The noise canceler is enabled by setting the Input Capture Noise Canceler (ICNCn) bitin Timer/Counter Control Register B (TCCRnB). When enabled the noise canceler intro-duces additional four system clock cycles of delay from a change applied to the input, tothe update of the ICRn Register. The noise canceler uses the system clock and is there-fore not affected by the prescaler.

Using the Input Capture Unit The main challenge when using the Input Capture unit is to assign enough processorcapacity for handling the incoming events. The time between two events is critical. If theprocessor has not read the captured value in the ICRn Register before the next eventoccurs, the ICRn will be overwritten with a new value. In this case the result of the cap-ture will be incorrect.

When using the Input Capture interrupt, the ICRn Register should be read as early in theinterrupt handler routine as possible. Even though the Input Capture interrupt has rela-tively high priority, the maximum interrupt response time is dependent on the maximumnumber of clock cycles it takes to handle any of the other interrupt requests.

Using the Input Capture unit in any mode of operation when the TOP value (resolution)is actively changed during operation, is not recommended.

Measurement of an external signal’s duty cycle requires that the trigger edge is changedafter each capture. Changing the edge sensing must be done as early as possible afterthe ICRn Register has been read. After a change of the edge, the Input Capture Flag

144 ATmega640/1280/1281/2560/25612549A–AVR–03/05

(ICFn) must be cleared by software (writing a logical one to the I/O bit location). Formeasuring frequency only, the clearing of the ICFn Flag is not required (if an interrupthandler is used).

Output Compare Units The 16-bit comparator continuously compares TCNTn with the Output Compare Regis-ter (OCRnx). If TCNT equals OCRnx the comparator signals a match. A match will setthe Output Compare Flag (OCFnx) at the next timer clock cycle. If enabled (OCIEnx =1), the Output Compare Flag generates an Output Compare interrupt. The OCFnx Flagis automatically cleared when the interrupt is executed. Alternatively the OCFnx Flagcan be cleared by software by writing a logical one to its I/O bit location. The WaveformGenerator uses the match signal to generate an output according to operating mode setby the Waveform Generation mode (WGMn3:0) bits and Compare Output mode(COMnx1:0) bits. The TOP and BOTTOM signals are used by the Waveform Generatorfor handling the special cases of the extreme values in some modes of operation (See“Modes of Operation” on page 147.)

A special feature of Output Compare unit A allows it to define the Timer/Counter TOPvalue (i.e., counter resolution). In addition to the counter resolution, the TOP valuedefines the period time for waveforms generated by the Waveform Generator.

Figure 52 shows a block diagram of the Output Compare unit. The small “n” in the regis-ter and bit names indicates the device number (n = n for Timer/Counter n), and the “x”indicates Output Compare unit (A/B/C). The elements of the block diagram that are notdirectly a part of the Output Compare unit are gray shaded.

Figure 52. Output Compare Unit, Block Diagram

The OCRnx Register is double buffered when using any of the twelve Pulse Width Mod-ulation (PWM) modes. For the Normal and Clear Timer on Compare (CTC) modes ofoperation, the double buffering is disabled. The double buffering synchronizes theupdate of the OCRnx Compare Register to either TOP or BOTTOM of the countingsequence. The synchronization prevents the occurrence of odd-length, non-symmetricalPWM pulses, thereby making the output glitch-free.

OCFnx (Int.Req.)

= (16-bit Comparator )

OCRnx Buffer (16-bit Register)

OCRnxH Buf. (8-bit)

OCnx

TEMP (8-bit)

DATA BUS (8-bit)

OCRnxL Buf. (8-bit)

TCNTn (16-bit Counter)

TCNTnH (8-bit) TCNTnL (8-bit)

COMnx1:0WGMn3:0

OCRnx (16-bit Register)

OCRnxH (8-bit) OCRnxL (8-bit)

Waveform GeneratorTOP

BOTTOM

145

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The OCRnx Register access may seem complex, but this is not case. When the doublebuffering is enabled, the CPU has access to the OCRnx Buffer Register, and if doublebuffering is disabled the CPU will access the OCRnx directly. The content of the OCR1x(Buffer or Compare) Register is only changed by a write operation (the Timer/Counterdoes not update this register automatically as the TCNT1 and ICR1 Register). ThereforeOCR1x is not read via the high byte temporary register (TEMP). However, it is a goodpractice to read the low byte first as when accessing other 16-bit registers. Writing theOCRnx Registers must be done via the TEMP Register since the compare of all 16 bitsis done continuously. The high byte (OCRnxH) has to be written first. When the highbyte I/O location is written by the CPU, the TEMP Register will be updated by the valuewritten. Then when the low byte (OCRnxL) is written to the lower eight bits, the high bytewill be copied into the upper 8-bits of either the OCRnx buffer or OCRnx Compare Reg-ister in the same system clock cycle.

For more information of how to access the 16-bit registers refer to “Accessing 16-bitRegisters” on page 137.

Force Output Compare In non-PWM Waveform Generation modes, the match output of the comparator can beforced by writing a one to the Force Output Compare (FOCnx) bit. Forcing comparematch will not set the OCFnx Flag or reload/clear the timer, but the OCnx pin will beupdated as if a real compare match had occurred (the COMn1:0 bits settings definewhether the OCnx pin is set, cleared or toggled).

Compare Match Blocking by TCNTn Write

All CPU writes to the TCNTn Register will block any compare match that occurs in thenext timer clock cycle, even when the timer is stopped. This feature allows OCRnx to beinitialized to the same value as TCNTn without triggering an interrupt when theTimer/Counter clock is enabled.

Using the Output Compare Unit

Since writing TCNTn in any mode of operation will block all compare matches for onetimer clock cycle, there are risks involved when changing TCNTn when using any of theOutput Compare channels, independent of whether the Timer/Counter is running or not.If the value written to TCNTn equals the OCRnx value, the compare match will bemissed, resulting in incorrect waveform generation. Do not write the TCNTn equal toTOP in PWM modes with variable TOP values. The compare match for the TOP will beignored and the counter will continue to 0xFFFF. Similarly, do not write the TCNTn valueequal to BOTTOM when the counter is downcounting.

The setup of the OCnx should be performed before setting the Data Direction Registerfor the port pin to output. The easiest way of setting the OCnx value is to use the ForceOutput Compare (FOCnx) strobe bits in Normal mode. The OCnx Register keeps itsvalue even when changing between Waveform Generation modes.

Be aware that the COMnx1:0 bits are not double buffered together with the comparevalue. Changing the COMnx1:0 bits will take effect immediately.

146 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Compare Match Output Unit

The Compare Output mode (COMnx1:0) bits have two functions. The Waveform Gener-ator uses the COMnx1:0 bits for defining the Output Compare (OCnx) state at the nextcompare match. Secondly the COMnx1:0 bits control the OCnx pin output source. Fig-ure 53 shows a simplified schematic of the logic affected by the COMnx1:0 bit setting.The I/O Registers, I/O bits, and I/O pins in the figure are shown in bold. Only the parts ofthe general I/O Port Control Registers (DDR and PORT) that are affected by theCOMnx1:0 bits are shown. When referring to the OCnx state, the reference is for theinternal OCnx Register, not the OCnx pin. If a system reset occur, the OCnx Register isreset to “0”.

Figure 53. Compare Match Output Unit, Schematic

The general I/O port function is overridden by the Output Compare (OCnx) from theWaveform Generator if either of the COMnx1:0 bits are set. However, the OCnx pindirection (input or output) is still controlled by the Data Direction Register (DDR) for theport pin. The Data Direction Register bit for the OCnx pin (DDR_OCnx) must be set asoutput before the OCnx value is visible on the pin. The port override function is generallyindependent of the Waveform Generation mode, but there are some exceptions. Referto Table 79, Table 80 and Table 81 for details.

The design of the Output Compare pin logic allows initialization of the OCnx state beforethe output is enabled. Note that some COMnx1:0 bit settings are reserved for certainmodes of operation. See “16-bit Timer/Counter Register Description” on page 157.

The COMnx1:0 bits have no effect on the Input Capture unit.

Compare Output Mode and Waveform Generation

The Waveform Generator uses the COMnx1:0 bits differently in normal, CTC, and PWMmodes. For all modes, setting the COMnx1:0 = 0 tells the Waveform Generator that noaction on the OCnx Register is to be performed on the next compare match. For com-pare output actions in the non-PWM modes refer to Table 79 on page 158. For fastPWM mode refer to Table 80 on page 158, and for phase correct and phase and fre-quency correct PWM refer to Table 81 on page 159.

PORT

DDR

D Q

D Q

OCnxPinOCnx

D QWaveformGenerator

COMnx1

COMnx0

0

1D

ATA

BU

SFOCnx

clkI/O

147

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

A change of the COMnx1:0 bits state will have effect at the first compare match after thebits are written. For non-PWM modes, the action can be forced to have immediate effectby using the FOCnx strobe bits.

Modes of Operation The mode of operation, i.e., the behavior of the Timer/Counter and the Output Comparepins, is defined by the combination of the Waveform Generation mode (WGMn3:0) andCompare Output mode (COMnx1:0) bits. The Compare Output mode bits do not affectthe counting sequence, while the Waveform Generation mode bits do. The COMnx1:0bits control whether the PWM output generated should be inverted or not (inverted ornon-inverted PWM). For non-PWM modes the COMnx1:0 bits control whether the out-put should be set, cleared or toggle at a compare match (See “Compare Match OutputUnit” on page 146.)

For detailed timing information refer to “Timer/Counter Timing Diagrams” on page 154.

Normal Mode The simplest mode of operation is the Normal mode (WGMn3:0 = 0). In this mode thecounting direction is always up (incrementing), and no counter clear is performed. Thecounter simply overruns when it passes its maximum 16-bit value (MAX = 0xFFFF) andthen restarts from the BOTTOM (0x0000). In normal operation the Timer/Counter Over-flow Flag (TOVn) will be set in the same timer clock cycle as the TCNTn becomes zero.The TOVn Flag in this case behaves like a 17th bit, except that it is only set, not cleared.However, combined with the timer overflow interrupt that automatically clears the TOVnFlag, the timer resolution can be increased by software. There are no special cases toconsider in the Normal mode, a new counter value can be written anytime.

The Input Capture unit is easy to use in Normal mode. However, observe that the maxi-mum interval between the external events must not exceed the resolution of the counter.If the interval between events are too long, the timer overflow interrupt or the prescalermust be used to extend the resolution for the capture unit.

The Output Compare units can be used to generate interrupts at some given time. Usingthe Output Compare to generate waveforms in Normal mode is not recommended,since this will occupy too much of the CPU time.

Clear Timer on Compare Match (CTC) Mode

In Clear Timer on Compare or CTC mode (WGMn3:0 = 4 or 12), the OCRnA or ICRnRegister are used to manipulate the counter resolution. In CTC mode the counter iscleared to zero when the counter value (TCNTn) matches either the OCRnA (WGMn3:0= 4) or the ICRn (WGMn3:0 = 12). The OCRnA or ICRn define the top value for thecounter, hence also its resolution. This mode allows greater control of the comparematch output frequency. It also simplifies the operation of counting external events.

The timing diagram for the CTC mode is shown in Figure 54. The counter value(TCNTn) increases until a compare match occurs with either OCRnA or ICRn, and thencounter (TCNTn) is cleared.

148 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 54. CTC Mode, Timing Diagram

An interrupt can be generated at each time the counter value reaches the TOP value byeither using the OCFnA or ICFn Flag according to the register used to define the TOPvalue. If the interrupt is enabled, the interrupt handler routine can be used for updatingthe TOP value. However, changing the TOP to a value close to BOTTOM when thecounter is running with none or a low prescaler value must be done with care since theCTC mode does not have the double buffering feature. If the new value written toOCRnA or ICRn is lower than the current value of TCNTn, the counter will miss the com-pare match. The counter will then have to count to its maximum value (0xFFFF) andwrap around starting at 0x0000 before the compare match can occur. In many casesthis feature is not desirable. An alternative will then be to use the fast PWM mode usingOCRnA for defining TOP (WGMn3:0 = 15) since the OCRnA then will be doublebuffered.

For generating a waveform output in CTC mode, the OCnA output can be set to toggleits logical level on each compare match by setting the Compare Output mode bits to tog-gle mode (COMnA1:0 = 1). The OCnA value will not be visible on the port pin unless thedata direction for the pin is set to output (DDR_OCnA = 1). The waveform generated willhave a maximum frequency of fOCnA = fclk_I/O/2 when OCRnA is set to zero (0x0000). Thewaveform frequency is defined by the following equation:

The N variable represents the prescaler factor (1, 8, 64, 256, or 1024).

As for the Normal mode of operation, the TOVn Flag is set in the same timer clock cyclethat the counter counts from MAX to 0x0000.

Fast PWM Mode The fast Pulse Width Modulation or fast PWM mode (WGMn3:0 = 5, 6, 7, 14, or 15) pro-vides a high frequency PWM waveform generation option. The fast PWM differs fromthe other PWM options by its single-slope operation. The counter counts from BOTTOMto TOP then restarts from BOTTOM. In non-inverting Compare Output mode, the OutputCompare (OCnx) is set on the compare match between TCNTn and OCRnx, andcleared at TOP. In inverting Compare Output mode output is cleared on compare matchand set at TOP. Due to the single-slope operation, the operating frequency of the fastPWM mode can be twice as high as the phase correct and phase and frequency correctPWM modes that use dual-slope operation. This high frequency makes the fast PWMmode well suited for power regulation, rectification, and DAC applications. High fre-quency allows physically small sized external components (coils, capacitors), hencereduces total system cost.

TCNTn

OCnA(Toggle)

OCnA Interrupt Flag Setor ICFn Interrupt Flag Set(Interrupt on TOP)

1 4Period 2 3

(COMnA1:0 = 1)

fOCnAfclk_I/O

2 N 1 OCRnA+( )⋅ ⋅---------------------------------------------------=

149

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The PWM resolution for fast PWM can be fixed to 8-, 9-, or 10-bit, or defined by eitherICRn or OCRnA. The minimum resolution allowed is 2-bit (ICRn or OCRnA set to0x0003), and the maximum resolution is 16-bit (ICRn or OCRnA set to MAX). The PWMresolution in bits can be calculated by using the following equation:

In fast PWM mode the counter is incremented until the counter value matches eitherone of the fixed values 0x00FF, 0x01FF, or 0x03FF (WGMn3:0 = 5, 6, or 7), the value inICRn (WGMn3:0 = 14), or the value in OCRnA (WGMn3:0 = 15). The counter is thencleared at the following timer clock cycle. The timing diagram for the fast PWM mode isshown in Figure 55. The figure shows fast PWM mode when OCRnA or ICRn is used todefine TOP. The TCNTn value is in the timing diagram shown as a histogram for illus-trating the single-slope operation. The diagram includes non-inverted and inverted PWMoutputs. The small horizontal line marks on the TCNTn slopes represent comparematches between OCRnx and TCNTn. The OCnx Interrupt Flag will be set when a com-pare match occurs.

Figure 55. Fast PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOVn) is set each time the counter reaches TOP. Inaddition the OCnA or ICFn Flag is set at the same timer clock cycle as TOVn is setwhen either OCRnA or ICRn is used for defining the TOP value. If one of the interruptsare enabled, the interrupt handler routine can be used for updating the TOP and com-pare values.

When changing the TOP value the program must ensure that the new TOP value ishigher or equal to the value of all of the Compare Registers. If the TOP value is lowerthan any of the Compare Registers, a compare match will never occur between theTCNTn and the OCRnx. Note that when using fixed TOP values the unused bits aremasked to zero when any of the OCRnx Registers are written.

The procedure for updating ICRn differs from updating OCRnA when used for definingthe TOP value. The ICRn Register is not double buffered. This means that if ICRn ischanged to a low value when the counter is running with none or a low prescaler value,there is a risk that the new ICRn value written is lower than the current value of TCNTn.The result will then be that the counter will miss the compare match at the TOP value.The counter will then have to count to the MAX value (0xFFFF) and wrap around start-ing at 0x0000 before the compare match can occur. The OCRnA Register however, is

RFPWMTOP 1+( )log

2( )log-----------------------------------=

TCNTn

OCRnx / TOP Update and TOVn Interrupt Flag Set and OCnA Interrupt Flag Set or ICFn Interrupt Flag Set (Interrupt on TOP)

1 7Period 2 3 4 5 6 8

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)

150 ATmega640/1280/1281/2560/25612549A–AVR–03/05

double buffered. This feature allows the OCRnA I/O location to be written anytime.When the OCRnA I/O location is written the value written will be put into the OCRnABuffer Register. The OCRnA Compare Register will then be updated with the value inthe Buffer Register at the next timer clock cycle the TCNTn matches TOP. The update isdone at the same timer clock cycle as the TCNTn is cleared and the TOVn Flag is set.

Using the ICRn Register for defining TOP works well when using fixed TOP values. Byusing ICRn, the OCRnA Register is free to be used for generating a PWM output onOCnA. However, if the base PWM frequency is actively changed (by changing the TOPvalue), using the OCRnA as TOP is clearly a better choice due to its double bufferfeature.

In fast PWM mode, the compare units allow generation of PWM waveforms on theOCnx pins. Setting the COMnx1:0 bits to two will produce a non-inverted PWM and aninverted PWM output can be generated by setting the COMnx1:0 to three (see Table onpage 158). The actual OCnx value will only be visible on the port pin if the data directionfor the port pin is set as output (DDR_OCnx). The PWM waveform is generated by set-ting (or clearing) the OCnx Register at the compare match between OCRnx and TCNTn,and clearing (or setting) the OCnx Register at the timer clock cycle the counter iscleared (changes from TOP to BOTTOM).

The PWM frequency for the output can be calculated by the following equation:

The N variable represents the prescaler divider (1, 8, 64, 256, or 1024).

The extreme values for the OCRnx Register represents special cases when generatinga PWM waveform output in the fast PWM mode. If the OCRnx is set equal to BOTTOM(0x0000) the output will be a narrow spike for each TOP+1 timer clock cycle. Setting theOCRnx equal to TOP will result in a constant high or low output (depending on the polar-ity of the output set by the COMnx1:0 bits.)

A frequency (with 50% duty cycle) waveform output in fast PWM mode can be achievedby setting OCnA to toggle its logical level on each compare match (COMnA1:0 = 1).This applies only if OCR1A is used to define the TOP value (WGM13:0 = 15). The wave-form generated will have a maximum frequency of fOCnA = fclk_I/O/2 when OCRnA is set tozero (0x0000). This feature is similar to the OCnA toggle in CTC mode, except the dou-ble buffer feature of the Output Compare unit is enabled in the fast PWM mode.

Phase Correct PWM Mode The phase correct Pulse Width Modulation or phase correct PWM mode (WGMn3:0 = 1,2, 3, 10, or 11) provides a high resolution phase correct PWM waveform generationoption. The phase correct PWM mode is, like the phase and frequency correct PWMmode, based on a dual-slope operation. The counter counts repeatedly from BOTTOM(0x0000) to TOP and then from TOP to BOTTOM. In non-inverting Compare Outputmode, the Output Compare (OCnx) is cleared on the compare match between TCNTnand OCRnx while upcounting, and set on the compare match while downcounting. Ininverting Output Compare mode, the operation is inverted. The dual-slope operation haslower maximum operation frequency than single slope operation. However, due to thesymmetric feature of the dual-slope PWM modes, these modes are preferred for motorcontrol applications.

The PWM resolution for the phase correct PWM mode can be fixed to 8-, 9-, or 10-bit, ordefined by either ICRn or OCRnA. The minimum resolution allowed is 2-bit (ICRn orOCRnA set to 0x0003), and the maximum resolution is 16-bit (ICRn or OCRnA set toMAX). The PWM resolution in bits can be calculated by using the following equation:

fOCnxPWMfclk_I/O

N 1 TOP+( )⋅-----------------------------------=

151

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

In phase correct PWM mode the counter is incremented until the counter value matcheseither one of the fixed values 0x00FF, 0x01FF, or 0x03FF (WGMn3:0 = 1, 2, or 3), thevalue in ICRn (WGMn3:0 = 10), or the value in OCRnA (WGMn3:0 = 11). The counterhas then reached the TOP and changes the count direction. The TCNTn value will beequal to TOP for one timer clock cycle. The timing diagram for the phase correct PWMmode is shown on Figure 56. The figure shows phase correct PWM mode when OCRnAor ICRn is used to define TOP. The TCNTn value is in the timing diagram shown as ahistogram for illustrating the dual-slope operation. The diagram includes non-invertedand inverted PWM outputs. The small horizontal line marks on the TCNTn slopes repre-sent compare matches between OCRnx and TCNTn. The OCnx Interrupt Flag will beset when a compare match occurs.

Figure 56. Phase Correct PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOVn) is set each time the counter reaches BOT-TOM. When either OCRnA or ICRn is used for defining the TOP value, the OCnA orICFn Flag is set accordingly at the same timer clock cycle as the OCRnx Registers areupdated with the double buffer value (at TOP). The Interrupt Flags can be used to gen-erate an interrupt each time the counter reaches the TOP or BOTTOM value.

When changing the TOP value the program must ensure that the new TOP value ishigher or equal to the value of all of the Compare Registers. If the TOP value is lowerthan any of the Compare Registers, a compare match will never occur between theTCNTn and the OCRnx. Note that when using fixed TOP values, the unused bits aremasked to zero when any of the OCRnx Registers are written. As the third period shownin Figure 56 illustrates, changing the TOP actively while the Timer/Counter is running inthe phase correct mode can result in an unsymmetrical output. The reason for this canbe found in the time of update of the OCRnx Register. Since the OCRnx update occursat TOP, the PWM period starts and ends at TOP. This implies that the length of the fall-ing slope is determined by the previous TOP value, while the length of the rising slope is

RPCPWMTOP 1+( )log

2( )log-----------------------------------=

OCRnx/TOP Update andOCnA Interrupt Flag Setor ICFn Interrupt Flag Set(Interrupt on TOP)

1 2 3 4

TOVn Interrupt Flag Set(Interrupt on Bottom)

TCNTn

Period

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)

152 ATmega640/1280/1281/2560/25612549A–AVR–03/05

determined by the new TOP value. When these two values differ the two slopes of theperiod will differ in length. The difference in length gives the unsymmetrical result on theoutput.

It is recommended to use the phase and frequency correct mode instead of the phasecorrect mode when changing the TOP value while the Timer/Counter is running. Whenusing a static TOP value there are practically no differences between the two modes ofoperation.

In phase correct PWM mode, the compare units allow generation of PWM waveforms onthe OCnx pins. Setting the COMnx1:0 bits to two will produce a non-inverted PWM andan inverted PWM output can be generated by setting the COMnx1:0 to three (See Table81 on page 159). The actual OCnx value will only be visible on the port pin if the datadirection for the port pin is set as output (DDR_OCnx). The PWM waveform is gener-ated by setting (or clearing) the OCnx Register at the compare match between OCRnxand TCNTn when the counter increments, and clearing (or setting) the OCnx Register atcompare match between OCRnx and TCNTn when the counter decrements. The PWMfrequency for the output when using phase correct PWM can be calculated by the fol-lowing equation:

The N variable represents the prescaler divider (1, 8, 64, 256, or 1024).

The extreme values for the OCRnx Register represent special cases when generating aPWM waveform output in the phase correct PWM mode. If the OCRnx is set equal toBOTTOM the output will be continuously low and if set equal to TOP the output will becontinuously high for non-inverted PWM mode. For inverted PWM the output will havethe opposite logic values. If OCR1A is used to define the TOP value (WGM13:0 = 11)and COM1A1:0 = 1, the OC1A output will toggle with a 50% duty cycle.

Phase and Frequency Correct PWM Mode

The phase and frequency correct Pulse Width Modulation, or phase and frequency cor-rect PWM mode (WGMn3:0 = 8 or 9) provides a high resolution phase and frequencycorrect PWM waveform generation option. The phase and frequency correct PWMmode is, like the phase correct PWM mode, based on a dual-slope operation. Thecounter counts repeatedly from BOTTOM (0x0000) to TOP and then from TOP to BOT-TOM. In non-inverting Compare Output mode, the Output Compare (OCnx) is clearedon the compare match between TCNTn and OCRnx while upcounting, and set on thecompare match while downcounting. In inverting Compare Output mode, the operationis inverted. The dual-slope operation gives a lower maximum operation frequency com-pared to the single-slope operation. However, due to the symmetric feature of the dual-slope PWM modes, these modes are preferred for motor control applications.

The main difference between the phase correct, and the phase and frequency correctPWM mode is the time the OCRnx Register is updated by the OCRnx Buffer Register,(see Figure 56 and Figure 57).

The PWM resolution for the phase and frequency correct PWM mode can be defined byeither ICRn or OCRnA. The minimum resolution allowed is 2-bit (ICRn or OCRnA set to0x0003), and the maximum resolution is 16-bit (ICRn or OCRnA set to MAX). The PWMresolution in bits can be calculated using the following equation:

fOCnxPCPWMfclk_I/O

2 N TOP⋅ ⋅----------------------------=

RPFCPWMTOP 1+( )log

2( )log-----------------------------------=

153

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

In phase and frequency correct PWM mode the counter is incremented until the countervalue matches either the value in ICRn (WGMn3:0 = 8), or the value in OCRnA(WGMn3:0 = 9). The counter has then reached the TOP and changes the count direc-tion. The TCNTn value will be equal to TOP for one timer clock cycle. The timingdiagram for the phase correct and frequency correct PWM mode is shown on Figure 57.The figure shows phase and frequency correct PWM mode when OCRnA or ICRn isused to define TOP. The TCNTn value is in the timing diagram shown as a histogram forillustrating the dual-slope operation. The diagram includes non-inverted and invertedPWM outputs. The small horizontal line marks on the TCNTn slopes represent comparematches between OCRnx and TCNTn. The OCnx Interrupt Flag will be set when a com-pare match occurs.

Figure 57. Phase and Frequency Correct PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOVn) is set at the same timer clock cycle as theOCRnx Registers are updated with the double buffer value (at BOTTOM). When eitherOCRnA or ICRn is used for defining the TOP value, the OCnA or ICFn Flag set whenTCNTn has reached TOP. The Interrupt Flags can then be used to generate an interrupteach time the counter reaches the TOP or BOTTOM value.

When changing the TOP value the program must ensure that the new TOP value ishigher or equal to the value of all of the Compare Registers. If the TOP value is lowerthan any of the Compare Registers, a compare match will never occur between theTCNTn and the OCRnx.

As Figure 57 shows the output generated is, in contrast to the phase correct mode, sym-metrical in all periods. Since the OCRnx Registers are updated at BOTTOM, the lengthof the rising and the falling slopes will always be equal. This gives symmetrical outputpulses and is therefore frequency correct.

Using the ICRn Register for defining TOP works well when using fixed TOP values. Byusing ICRn, the OCRnA Register is free to be used for generating a PWM output onOCnA. However, if the base PWM frequency is actively changed by changing the TOPvalue, using the OCRnA as TOP is clearly a better choice due to its double bufferfeature.

In phase and frequency correct PWM mode, the compare units allow generation ofPWM waveforms on the OCnx pins. Setting the COMnx1:0 bits to two will produce a

OCRnx/TOP UpdateandTOVn Interrupt Flag Set(Interrupt on Bottom)

OCnA Interrupt Flag Setor ICFn Interrupt Flag Set(Interrupt on TOP)

1 2 3 4

TCNTn

Period

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)

154 ATmega640/1280/1281/2560/25612549A–AVR–03/05

non-inverted PWM and an inverted PWM output can be generated by setting theCOMnx1:0 to three (See Table 81 on page 159). The actual OCnx value will only be vis-ible on the port pin if the data direction for the port pin is set as output (DDR_OCnx). ThePWM waveform is generated by setting (or clearing) the OCnx Register at the comparematch between OCRnx and TCNTn when the counter increments, and clearing (or set-ting) the OCnx Register at compare match between OCRnx and TCNTn when thecounter decrements. The PWM frequency for the output when using phase and fre-quency correct PWM can be calculated by the following equation:

The N variable represents the prescaler divider (1, 8, 64, 256, or 1024).

The extreme values for the OCRnx Register represents special cases when generatinga PWM waveform output in the phase correct PWM mode. If the OCRnx is set equal toBOTTOM the output will be continuously low and if set equal to TOP the output will beset to high for non-inverted PWM mode. For inverted PWM the output will have theopposite logic values. If OCR1A is used to define the TOP value (WGM13:0 = 9) andCOM1A1:0 = 1, the OC1A output will toggle with a 50% duty cycle.

Timer/Counter Timing Diagrams

The Timer/Counter is a synchronous design and the timer clock (clkTn) is thereforeshown as a clock enable signal in the following figures. The figures include informationon when Interrupt Flags are set, and when the OCRnx Register is updated with theOCRnx buffer value (only for modes utilizing double buffering). Figure 58 shows a timingdiagram for the setting of OCFnx.

Figure 58. Timer/Counter Timing Diagram, Setting of OCFnx, no Prescaling

Figure 59 shows the same timing data, but with the prescaler enabled.

fOCnxPFCPWMfclk_I/O

2 N TOP⋅ ⋅----------------------------=

clkTn(clkI/O/1)

OCFnx

clkI/O

OCRnx

TCNTn

OCRnx Value

OCRnx - 1 OCRnx OCRnx + 1 OCRnx + 2

155

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 59. Timer/Counter Timing Diagram, Setting of OCFnx, with Prescaler (fclk_I/O/8)

Figure 60 shows the count sequence close to TOP in various modes. When using phaseand frequency correct PWM mode the OCRnx Register is updated at BOTTOM. Thetiming diagrams will be the same, but TOP should be replaced by BOTTOM, TOP-1 byBOTTOM+1 and so on. The same renaming applies for modes that set the TOVn Flagat BOTTOM.

Figure 60. Timer/Counter Timing Diagram, no Prescaling

Figure 61 shows the same timing data, but with the prescaler enabled.

OCFnx

OCRnx

TCNTn

OCRnx Value

OCRnx - 1 OCRnx OCRnx + 1 OCRnx + 2

clkI/O

clkTn(clkI/O/8)

TOVn (FPWM)and ICFn (if used

as TOP)

OCRnx(Update at TOP)

TCNTn(CTC and FPWM)

TCNTn(PC and PFC PWM)

TOP - 1 TOP TOP - 1 TOP - 2

Old OCRnx Value New OCRnx Value

TOP - 1 TOP BOTTOM BOTTOM + 1

clkTn(clkI/O/1)

clkI/O

156 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 61. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)

TOVn (FPWM)and ICFn (if used

as TOP)

OCRnx(Update at TOP)

TCNTn(CTC and FPWM)

TCNTn(PC and PFC PWM)

TOP - 1 TOP TOP - 1 TOP - 2

Old OCRnx Value New OCRnx Value

TOP - 1 TOP BOTTOM BOTTOM + 1

clkI/O

clkTn(clkI/O/8)

157

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

16-bit Timer/Counter Register Description

Timer/Counter1 Control Register A – TCCR1A

Timer/Counter3 Control Register A – TCCR3A

Timer/Counter4 Control Register A – TCCR4A

Timer/Counter5 Control Register A – TCCR5A

• Bit 7:6 – COMnA1:0: Compare Output Mode for Channel A• Bit 5:4 – COMnB1:0: Compare Output Mode for Channel B• Bit 3:2 – COMnC1:0: Compare Output Mode for Channel C

The COMnA1:0, COMnB1:0, and COMnC1:0 control the output compare pins (OCnA,OCnB, and OCnC respectively) behavior. If one or both of the COMnA1:0 bits are writ-ten to one, the OCnA output overrides the normal port functionality of the I/O pin it isconnected to. If one or both of the COMnB1:0 bits are written to one, the OCnB outputoverrides the normal port functionality of the I/O pin it is connected to. If one or both ofthe COMnC1:0 bits are written to one, the OCnC output overrides the normal port func-tionality of the I/O pin it is connected to. However, note that the Data Direction Register(DDR) bit corresponding to the OCnA, OCnB or OCnC pin must be set in order toenable the output driver.

When the OCnA, OCnB or OCnC is connected to the pin, the function of the COMnx1:0bits is dependent of the WGMn3:0 bits setting. Table 79 shows the COMnx1:0 bit func-tionality when the WGMn3:0 bits are set to a normal or a CTC mode (non-PWM).

Bit 7 6 5 4 3 2 1 0

COM1A1 COM1A0 COM1B1 COM1B0 COM1C1 COM1C0 WGM11 WGM10 TCCR1A

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

COM3A1 COM3A0 COM3B1 COM3B0 COM3C1 COM3C0 WGM31 WGM30 TCCR3A

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

COM4A1 COM4A0 COM4B1 COM4B0 COM4C1 COM4C0 WGM41 WGM40 TCCR4A

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

COM5A1 COM5A0 COM5B1 COM5B0 COM5C1 COM5C0 WGM51 WGM50 TCCR5A

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

158 ATmega640/1280/1281/2560/25612549A–AVR–03/05

.

Table 80 shows the COMnx1:0 bit functionality when the WGMn3:0 bits are set to thefast PWM mode.

Note: A special case occurs when OCRnA/OCRnB/OCRnC equals TOP andCOMnA1/COMnB1/COMnC1 is set. In this case the compare match is ignored, but theset or clear is done at TOP. See “Fast PWM Mode” on page 148. for more details.

Table 81 shows the COMnx1:0 bit functionality when the WGMn3:0 bits are set to thephase correct and frequency correct PWM mode.

Table 79. Compare Output Mode, non-PWM

COMnA1/COMnB1/ COMnC1

COMnA0/COMnB0/ COMnC0 Description

0 0 Normal port operation, OCnA/OCnB/OCnC disconnected.

0 1 Toggle OCnA/OCnB/OCnC on compare match.

1 0 Clear OCnA/OCnB/OCnC on compare match (set output to low level).

1 1 Set OCnA/OCnB/OCnC on compare match (set output to high level).

Table 80. Compare Output Mode, Fast PWM

COMnA1/COMnB1/ COMnC0

COMnA0/COMnB0/ COMnC0 Description

0 0 Normal port operation, OCnA/OCnB/OCnC disconnected.

0 1 WGM13:0 = 14 or 15: Toggle OC1A on Compare Match, OC1B and OC1C disconnected (normal port operation). For all other WGM1 settings, normal port operation, OC1A/OC1B/OC1C disconnected.

1 0 Clear OCnA/OCnB/OCnC on compare match, set OCnA/OCnB/OCnC at TOP

1 1 Set OCnA/OCnB/OCnC on compare match, clear OCnA/OCnB/OCnC at TOP

159

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: A special case occurs when OCRnA/OCRnB/OCRnC equals TOP andCOMnA1/COMnB1//COMnC1 is set. See “Phase Correct PWM Mode” on page 150. formore details.

• Bit 1:0 – WGMn1:0: Waveform Generation Mode

Combined with the WGMn3:2 bits found in the TCCRnB Register, these bits control thecounting sequence of the counter, the source for maximum (TOP) counter value, andwhat type of waveform generation to be used, see Table 82. Modes of operation sup-ported by the Timer/Counter unit are: Normal mode (counter), Clear Timer on Comparematch (CTC) mode, and three types of Pulse Width Modulation (PWM) modes. (See“Modes of Operation” on page 147.).

Table 81. Compare Output Mode, Phase Correct and Phase and Frequency CorrectPWM

COMnA1/COMnB/ COMnC1

COMnA0/COMnB0/ COMnC0 Description

0 0 Normal port operation, OCnA/OCnB/OCnC disconnected.

0 1 WGM13:0 = 8, 9 10 or 11: Toggle OC1A on Compare Match, OC1B and OC1C disconnected (normal port operation). For all other WGM1 settings, normal port operation, OC1A/OC1B/OC1C disconnected.

1 0 Clear OCnA/OCnB/OCnC on compare match when up-counting. Set OCnA/OCnB/OCnC on compare match when downcounting.

1 1 Set OCnA/OCnB/OCnC on compare match when up-counting. Clear OCnA/OCnB/OCnC on compare match when downcounting.

160 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. The CTCn and PWMn1:0 bit definition names are obsolete. Use the WGMn2:0 definitions. However, the functionality andlocation of these bits are compatible with previous versions of the timer.

Timer/Counter1 Control Register B – TCCR1B

Timer/Counter3 Control Register B – TCCR3B

Timer/Counter4 Control Register B – TCCR4B

Table 82. Waveform Generation Mode Bit Description(1)

Mode WGMn3WGMn2(CTCn)

WGMn1(PWMn1)

WGMn0(PWMn0)

Timer/Counter Mode of Operation TOP

Update of OCRnx at

TOVn Flag Set on

0 0 0 0 0 Normal 0xFFFF Immediate MAX

1 0 0 0 1 PWM, Phase Correct, 8-bit 0x00FF TOP BOTTOM

2 0 0 1 0 PWM, Phase Correct, 9-bit 0x01FF TOP BOTTOM

3 0 0 1 1 PWM, Phase Correct, 10-bit 0x03FF TOP BOTTOM

4 0 1 0 0 CTC OCRnA Immediate MAX

5 0 1 0 1 Fast PWM, 8-bit 0x00FF TOP TOP

6 0 1 1 0 Fast PWM, 9-bit 0x01FF TOP TOP

7 0 1 1 1 Fast PWM, 10-bit 0x03FF TOP TOP

8 1 0 0 0 PWM, Phase and Frequency Correct

ICRn BOTTOM BOTTOM

9 1 0 0 1 PWM, Phase and Frequency Correct

OCRnA BOTTOM BOTTOM

10 1 0 1 0 PWM, Phase Correct ICRn TOP BOTTOM

11 1 0 1 1 PWM, Phase Correct OCRnA TOP BOTTOM

12 1 1 0 0 CTC ICRn Immediate MAX

13 1 1 0 1 (Reserved) – – –

14 1 1 1 0 Fast PWM ICRn TOP TOP

15 1 1 1 1 Fast PWM OCRnA TOP TOP

Bit 7 6 5 4 3 2 1 0

ICNC1 ICES1 – WGM13 WGM12 CS12 CS11 CS10 TCCR1B

Read/Write R/W R/W R R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

ICNC3 ICES3 – WGM33 WGM32 CS32 CS31 CS30 TCCR3B

Read/Write R/W R/W R R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

ICNC4 ICES4 – WGM43 WGM42 CS42 CS41 CS40 TCCR4B

Read/Write R/W R/W R R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

161

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Timer/Counter5 Control Register B – TCCR5B

• Bit 7 – ICNCn: Input Capture Noise Canceler

Setting this bit (to one) activates the Input Capture Noise Canceler. When the NoiseCanceler is activated, the input from the Input Capture Pin (ICPn) is filtered. The filterfunction requires four successive equal valued samples of the ICPn pin for changing itsoutput. The input capture is therefore delayed by four Oscillator cycles when the noisecanceler is enabled.

• Bit 6 – ICESn: Input Capture Edge Select

This bit selects which edge on the Input Capture Pin (ICPn) that is used to trigger a cap-ture event. When the ICESn bit is written to zero, a falling (negative) edge is used astrigger, and when the ICESn bit is written to one, a rising (positive) edge will trigger thecapture.

When a capture is triggered according to the ICESn setting, the counter value is copiedinto the Input Capture Register (ICRn). The event will also set the Input Capture Flag(ICFn), and this can be used to cause an Input Capture Interrupt, if this interrupt isenabled.

When the ICRn is used as TOP value (see description of the WGMn3:0 bits located inthe TCCRnA and the TCCRnB Register), the ICPn is disconnected and consequentlythe input capture function is disabled.

• Bit 5 – Reserved Bit

This bit is reserved for future use. For ensuring compatibility with future devices, this bitmust be written to zero when TCCRnB is written.

• Bit 4:3 – WGMn3:2: Waveform Generation Mode

See TCCRnA Register description.

• Bit 2:0 – CSn2:0: Clock Select

The three clock select bits select the clock source to be used by the Timer/Counter, seeFigure 58 and Figure 59.

Bit 7 6 5 4 3 2 1 0

ICNC5 ICES5 – WGM53 WGM52 CS52 CS51 CS50 TCCR5B

Read/Write R/W R/W R R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

162 ATmega640/1280/1281/2560/25612549A–AVR–03/05

If external pin modes are used for the Timer/Countern, transitions on the Tn pin willclock the counter even if the pin is configured as an output. This feature allows softwarecontrol of the counting.

Timer/Counter1 Control Register C – TCCR1C

Timer/Counter3 Control Register C – TCCR3C

Timer/Counter4 Control Register C – TCCR4C

Timer/Counter5 Control Register C – TCCR5C

• Bit 7 – FOCnA: Force Output Compare for Channel A• Bit 6 – FOCnB: Force Output Compare for Channel B• Bit 5 – FOCnC: Force Output Compare for Channel C

The FOCnA/FOCnB/FOCnC bits are only active when the WGMn3:0 bits specifies anon-PWM mode. When writing a logical one to the FOCnA/FOCnB/FOCnC bit, animmediate compare match is forced on the waveform generation unit. TheOCnA/OCnB/OCnC output is changed according to its COMnx1:0 bits setting. Note thatthe FOCnA/FOCnB/FOCnC bits are implemented as strobes. Therefore it is the valuepresent in the COMnx1:0 bits that determine the effect of the forced compare.

A FOCnA/FOCnB/FOCnC strobe will not generate any interrupt nor will it clear the timerin Clear Timer on Compare Match (CTC) mode using OCRnA as TOP.

Table 83. Clock Select Bit Description

CSn2 CSn1 CSn0 Description

0 0 0 No clock source. (Timer/Counter stopped)

0 0 1 clkI/O/1 (No prescaling

0 1 0 clkI/O/8 (From prescaler)

0 1 1 clkI/O/64 (From prescaler)

1 0 0 clkI/O/256 (From prescaler)

1 0 1 clkI/O/1024 (From prescaler)

1 1 0 External clock source on Tn pin. Clock on falling edge

1 1 1 External clock source on Tn pin. Clock on rising edge

Bit 7 6 5 4 3 2 1 0

FOC1A FOC1B FOC1C – – – – – TCCR1C

Read/Write W W W R R R R R

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

FOC3A FOC3B FOC3C – – – – – TCCR3C

Read/Write W W W R R R R R

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

FOC4A FOC4B FOC4C – – – – – TCCR4C

Read/Write W W W R R R R R

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

FOC5A FOC5B FOC3C – – – – – TCCR5C

Read/Write W W W R R R R R

Initial Value 0 0 0 0 0 0 0 0

163

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The FOCnA/FOCnB/FOCnB bits are always read as zero.• Bit 4:0 – Reserved Bits

These bits are reserved for future use. For ensuring compatibility with future devices,these bits must be written to zero when TCCRnC is written.

Timer/Counter1 – TCNT1H and TCNT1L

Timer/Counter3 – TCNT3H and TCNT3L

Timer/Counter4 – TCNT4H and TCNT4L

Timer/Counter5 – TCNT5H and TCNT5L

The two Timer/Counter I/O locations (TCNTnH and TCNTnL, combined TCNTn) givedirect access, both for read and for write operations, to the Timer/Counter unit 16-bitcounter. To ensure that both the high and low bytes are read and written simultaneouslywhen the CPU accesses these registers, the access is performed using an 8-bit tempo-rary High Byte Register (TEMP). This temporary register is shared by all the other 16-bitregisters. See “Accessing 16-bit Registers” on page 137.

Modifying the counter (TCNTn) while the counter is running introduces a risk of missinga compare match between TCNTn and one of the OCRnx Registers.

Writing to the TCNTn Register blocks (removes) the compare match on the followingtimer clock for all compare units.

Bit 7 6 5 4 3 2 1 0

TCNT1[15:8] TCNT1H

TCNT1[7:0] TCNT1L

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

TCNT3[15:8] TCNT3H

TCNT3[7:0] TCNT3L

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

TCNT4[15:8] TCNT4H

TCNT4[7:0] TCNT4L

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

TCNT5[15:8] TCNT5H

TCNT5[7:0] TCNT5L

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

164 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Output Compare Register 1 A – OCR1AH and OCR1AL

Output Compare Register 1 B – OCR1BH and OCR1BL

Output Compare Register 1 C – OCR1CH and OCR1CL

Output Compare Register 3 A – OCR3AH and OCR3AL

Output Compare Register 3 B – OCR3BH and OCR3BL

Output Compare Register 3 C – OCR3CH and OCR3CL

Output Compare Register 4 A – OCR4AH and OCR4AL

Output Compare Register 4 B – OCR4BH and OCR4BL

Bit 7 6 5 4 3 2 1 0

OCR1A[15:8] OCR1AH

OCR1A[7:0] OCR1AL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR1B[15:8] OCR1BH

OCR1B[7:0] OCR1BL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR1C[15:8] OCR1CH

OCR1C[7:0] OCR1CL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR3A[15:8] OCR3AH

OCR3A[7:0] OCR3AL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR3B[15:8] OCR3BH

OCR3B[7:0] OCR3BL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR3C[15:8] OCR3CH

OCR3C[7:0] OCR3CL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR4A[15:8] OCR4AH

OCR4A[7:0] OCR4AL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR4B[15:8] OCR4BH

OCR4B[7:0] OCR4BL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

165

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Output Compare Register 4 C – OCR4CH and OCR4CL

Output Compare Register 5 A – OCR5AH and OCR5AL

Output Compare Register 5 B – OCR5BH and OCR5BL

Output Compare Register 5 C – OCR5CH and OCR5CL

The Output Compare Registers contain a 16-bit value that is continuously comparedwith the counter value (TCNTn). A match can be used to generate an Output Compareinterrupt, or to generate a waveform output on the OCnx pin.

The Output Compare Registers are 16-bit in size. To ensure that both the high and lowbytes are written simultaneously when the CPU writes to these registers, the access isperformed using an 8-bit temporary High Byte Register (TEMP). This temporary registeris shared by all the other 16-bit registers. See “Accessing 16-bit Registers” on page 137.

Input Capture Register 1 – ICR1H and ICR1L

Input Capture Register 3 – ICR3H and ICR3L

Bit 7 6 5 4 3 2 1 0

OCR4C[15:8] OCR4CH

OCR4C[7:0] OCR4CL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR5A[15:8] OCR5AH

OCR5A[7:0] OCR5AL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR5B[15:8] OCR5BH

OCR5B[7:0] OCR5BL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR5C[15:8] OCR5CH

OCR5C[7:0] OCR5CL

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

ICR1[15:8] ICR1H

ICR1[7:0] ICR1L

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

ICR3[15:8] ICR3H

ICR3[7:0] ICR3L

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

166 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Input Capture Register 4 – ICR4H and ICR4L

Input Capture Register 5 – ICR5H and ICR5L

The Input Capture is updated with the counter (TCNTn) value each time an event occurson the ICPn pin (or optionally on the Analog Comparator output for Timer/Counter1).The Input Capture can be used for defining the counter TOP value.

The Input Capture Register is 16-bit in size. To ensure that both the high and low bytesare read simultaneously when the CPU accesses these registers, the access is per-formed using an 8-bit temporary High Byte Register (TEMP). This temporary register isshared by all the other 16-bit registers. See “Accessing 16-bit Registers” on page 137.

Timer/Counter1 Interrupt Mask Register – TIMSK1

Timer/Counter3 Interrupt Mask Register – TIMSK3

Timer/Counter4 Interrupt Mask Register – TIMSK4

Timer/Counter5 Interrupt Mask Register – TIMSK5

• Bit 5 – ICIEn: Timer/Countern, Input Capture Interrupt Enable

When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-bally enabled), the Timer/Countern Input Capture interrupt is enabled. Thecorresponding Interrupt Vector (See “Interrupts” on page 69.) is executed when theICFn Flag, located in TIFRn, is set.

Bit 7 6 5 4 3 2 1 0

ICR4[15:8] ICR4H

ICR4[7:0] ICR4L

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

ICR5[15:8] ICR5H

ICR5[7:0] ICR5L

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – ICIE1 – OCIE1C OCIE1B OCIE1A TOIE1 TIMSK1

Read/Write R R R/W R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – ICIE3 – OCIE3C OCIE3B OCIE3A TOIE3 TIMSK3

Read/Write R R R/W R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – ICIE4 – OCIE4C OCIE4B OCIE4A TOIE4 TIMSK4

Read/Write R R R/W R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – ICIE5 – OCIE5C OCIE5B OCIE5A TOIE5 TIMSK5

Read/Write R R R/W R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

167

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• Bit 3 – OCIEnC: Timer/Countern, Output Compare C Match Interrupt Enable

When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-bally enabled), the Timer/Countern Output Compare C Match interrupt is enabled. Thecorresponding Interrupt Vector (See “Interrupts” on page 69.) is executed when theOCFnC Flag, located in TIFRn, is set.

• Bit 2 – OCIEnB: Timer/Countern, Output Compare B Match Interrupt Enable

When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-bally enabled), the Timer/Countern Output Compare B Match interrupt is enabled. Thecorresponding Interrupt Vector (See “Interrupts” on page 69.) is executed when theOCFnB Flag, located in TIFRn, is set.

• Bit 1 – OCIEnA: Timer/Countern, Output Compare A Match Interrupt Enable

When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-bally enabled), the Timer/Countern Output Compare A Match interrupt is enabled. Thecorresponding Interrupt Vector (See “Interrupts” on page 69.) is executed when theOCFnA Flag, located in TIFRn, is set.

• Bit 0 – TOIEn: Timer/Countern, Overflow Interrupt Enable

When this bit is written to one, and the I-flag in the Status Register is set (interrupts glo-bally enabled), the Timer/Countern Overflow interrupt is enabled. The correspondingInterrupt Vector (See “Interrupts” on page 69.) is executed when the TOVn Flag, locatedin TIFRn, is set.

Timer/Counter1 Interrupt Flag Register – TIFR1

Timer/Counter3 Interrupt Flag Register – TIFR3

Timer/Counter4 Interrupt Flag Register – TIFR4

Timer/Counter5 Interrupt Flag Register – TIFR5

• Bit 5 – ICFn: Timer/Countern, Input Capture Flag

This flag is set when a capture event occurs on the ICPn pin. When the Input CaptureRegister (ICRn) is set by the WGMn3:0 to be used as the TOP value, the ICFn Flag isset when the counter reaches the TOP value.

Bit 7 6 5 4 3 2 1 0

– – ICF1 – OCF1C OCF1B OCF1A TOV1 TIFR1

Read/Write R R R/W R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – ICF3 – OCF3C OCF3B OCF3A TOV3 TIFR3

Read/Write R R R/W R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – ICF4 – OCF4C OCF4B OCF4A TOV4 TIFR4

Read/Write R R R/W R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – ICF5 – OCF5C OCF5B OCF5A TOV5 TIFR5

Read/Write R R R/W R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

168 ATmega640/1280/1281/2560/25612549A–AVR–03/05

ICFn is automatically cleared when the Input Capture Interrupt Vector is executed. Alter-natively, ICFn can be cleared by writing a logic one to its bit location.

• Bit 3– OCFnC: Timer/Countern, Output Compare C Match Flag

This flag is set in the timer clock cycle after the counter (TCNTn) value matches the Out-put Compare Register C (OCRnC).

Note that a Forced Output Compare (FOCnC) strobe will not set the OCFnC Flag.

OCFnC is automatically cleared when the Output Compare Match C Interrupt Vector isexecuted. Alternatively, OCFnC can be cleared by writing a logic one to its bit location.

• Bit 2 – OCFnB: Timer/Counter1, Output Compare B Match Flag

This flag is set in the timer clock cycle after the counter (TCNTn) value matches the Out-put Compare Register B (OCRnB).

Note that a Forced Output Compare (FOCnB) strobe will not set the OCFnB Flag.

OCFnB is automatically cleared when the Output Compare Match B Interrupt Vector isexecuted. Alternatively, OCFnB can be cleared by writing a logic one to its bit location.

• Bit 1 – OCF1A: Timer/Counter1, Output Compare A Match Flag

This flag is set in the timer clock cycle after the counter (TCNTn value matches the Out-put Compare Register A (OCRnA).

Note that a Forced Output Compare (FOCnA) strobe will not set the OCFnA Flag.

OCFnA is automatically cleared when the Output Compare Match A Interrupt Vector isexecuted. Alternatively, OCFnA can be cleared by writing a logic one to its bit location.

• Bit 0 – TOVn: Timer/Countern, Overflow Flag

The setting of this flag is dependent of the WGMn3:0 bits setting. In Normal and CTCmodes, the TOVn Flag is set when the timer overflows. Refer to Table 82 on page 160for the TOVn Flag behavior when using another WGMn3:0 bit setting.

TOVn is automatically cleared when the Timer/Countern Overflow Interrupt Vector isexecuted. Alternatively, TOVn can be cleared by writing a logic one to its bit location.

169

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Timer/Counter0, Timer/Counter1, Timer/Counter3, Timer/Counter4, and Timer/Counter5 Prescalers

Timer/Counter0, 1, 3, 4, and 5 share the same prescaler module, but theTimer/Counters can have different prescaler settings. The description below applies toall Timer/Counters. Tn is used as a general name, n = 0, 1, 3, 4, or 5.

Internal Clock Source The Timer/Counter can be clocked directly by the system clock (by setting the CSn2:0 =1). This provides the fastest operation, with a maximum Timer/Counter clock frequencyequal to system clock frequency (fCLK_I/O). Alternatively, one of four taps from the pres-caler can be used as a clock source. The prescaled clock has a frequency of eitherfCLK_I/O/8, fCLK_I/O/64, fCLK_I/O/256, or fCLK_I/O/1024.

Prescaler Reset The prescaler is free running, i.e., operates independently of the Clock Select logic ofthe Timer/Counter, and it is shared by the Timer/Counter Tn. Since the prescaler is notaffected by the Timer/Counter’s clock select, the state of the prescaler will have implica-tions for situations where a prescaled clock is used. One example of prescaling artifactsoccurs when the timer is enabled and clocked by the prescaler (6 > CSn2:0 > 1). Thenumber of system clock cycles from when the timer is enabled to the first count occurscan be from 1 to N+1 system clock cycles, where N equals the prescaler divisor (8, 64,256, or 1024).

It is possible to use the prescaler reset for synchronizing the Timer/Counter to programexecution. However, care must be taken if the other Timer/Counter that shares thesame prescaler also uses prescaling. A prescaler reset will affect the prescaler periodfor all Timer/Counters it is connected to.

External Clock Source An external clock source applied to the Tn pin can be used as Timer/Counter clock(clkTn). The Tn pin is sampled once every system clock cycle by the pin synchronizationlogic. The synchronized (sampled) signal is then passed through the edge detector. Fig-ure 62 shows a functional equivalent block diagram of the Tn synchronization and edgedetector logic. The registers are clocked at the positive edge of the internal system clock(clkI/O). The latch is transparent in the high period of the internal system clock.

The edge detector generates one clkTn pulse for each positive (CSn2:0 = 7) or negative(CSn2:0 = 6) edge it detects.

Figure 62. Tn/T0 Pin Sampling

The synchronization and edge detector logic introduces a delay of 2.5 to 3.5 systemclock cycles from an edge has been applied to the Tn pin to the counter is updated.

Enabling and disabling of the clock input must be done when Tn has been stable for atleast one system clock cycle, otherwise it is a risk that a false Timer/Counter clock pulseis generated.

Tn_sync(To ClockSelect Logic)

Edge DetectorSynchronization

D QD Q

LE

D QTn

clkI/O

170 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Each half period of the external clock applied must be longer than one system clockcycle to ensure correct sampling. The external clock must be guaranteed to have lessthan half the system clock frequency (fExtClk < fclk_I/O/2) given a 50/50% duty cycle. Sincethe edge detector uses sampling, the maximum frequency of an external clock it candetect is half the sampling frequency (Nyquist sampling theorem). However, due to vari-ation of the system clock frequency and duty cycle caused by Oscillator source (crystal,resonator, and capacitors) tolerances, it is recommended that maximum frequency of anexternal clock source is less than fclk_I/O/2.5.

An external clock source can not be prescaled.

Figure 63. Prescaler for synchronous Timer/Counters

General Timer/Counter Control Register – GTCCR

• Bit 7 – TSM: Timer/Counter Synchronization Mode

Writing the TSM bit to one activates the Timer/Counter Synchronization mode. In thismode, the value that is written to the PSRASY and PSRSYNC bits is kept, hence keep-ing the corresponding prescaler reset signals asserted. This ensures that thecorresponding Timer/Counters are halted and can be configured to the same value with-out the risk of one of them advancing during configuration. When the TSM bit is writtento zero, the PSRASY and PSRSYNC bits are cleared by hardware, and theTimer/Counters start counting simultaneously.

• Bit 0 – PSRSYNC: Prescaler Reset for Synchronous Timer/Counters

When this bit is one, Timer/Counter0 and Timer/Counter1, Timer/Counter3,Timer/Counter4 and Timer/Counter5 prescaler will be Reset. This bit is normally clearedimmediately by hardware, except if the TSM bit is set. Note that Timer/Counter0,Timer/Counter1, Timer/Counter3, Timer/Counter4 and Timer/Counter5 share the sameprescaler and a reset of this prescaler will affect all timers.

PSR10

Clear

Tn

Tn

clkI/O

Synchronization

Synchronization

TIMER/COUNTERn CLOCK SOURCEclk

Tn

TIMER/COUNTERn CLOCK SOURCEclk

Tn

CSn0

CSn1

CSn2

CSn0

CSn1

CSn2

Bit 7 6 5 4 3 2 1 0

TSM – – – – – PSRASY PSRSYNC GTCCR

Read/Write R/W R R R R R R/W R/W

Initial Value 0 0 0 0 0 0 0 0

171

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Output Compare Modulator (OCM1C0A)

Overview The Output Compare Modulator (OCM) allows generation of waveforms modulated witha carrier frequency. The modulator uses the outputs from the Output Compare Unit C ofthe 16-bit Timer/Counter1 and the Output Compare Unit of the 8-bit Timer/Counter0. Formore details about these Timer/Counters see “Timer/Counter0, Timer/Counter1,Timer/Counter3, Timer/Counter4, and Timer/Counter5 Prescalers” on page 169 and “8-bit Timer/Counter2 with PWM and Asynchronous Operation” on page 173.

Figure 64. Output Compare Modulator, Block Diagram

When the modulator is enabled, the two output compare channels are modulatedtogether as shown in the block diagram (Figure 64).

Description The Output Compare unit 1C and Output Compare unit 2 shares the PB7 port pin foroutput. The outputs of the Output Compare units (OC1C and OC0A) overrides the nor-mal PORTB7 Register when one of them is enabled (i.e., when COMnx1:0 is not equalto zero). When both OC1C and OC0A are enabled at the same time, the modulator isautomatically enabled.

The functional equivalent schematic of the modulator is shown on Figure 65. The sche-matic includes part of the Timer/Counter units and the port B pin 7 output driver circuit.

Figure 65. Output Compare Modulator, Schematic

OC1C

Pin

OC1C /

OC0A / PB7

Timer/Counter 1

Timer/Counter 0 OC0A

PORTB7 DDRB7

D QD Q

Pin

COMA01

COMA00

DATABUS

OC1C /

OC0A/ PB7

COM1C1

COM1C0

Modulator

1

0

OC1C

D Q

OC0A

D Q

( From Waveform Generator )

( From Waveform Generator )

0

1

Vcc

172 ATmega640/1280/1281/2560/25612549A–AVR–03/05

When the modulator is enabled the type of modulation (logical AND or OR) can beselected by the PORTB7 Register. Note that the DDRB7 controls the direction of theport independent of the COMnx1:0 bit setting.

Timing Example Figure 66 illustrates the modulator in action. In this example the Timer/Counter1 is set tooperate in fast PWM mode (non-inverted) and Timer/Counter0 uses CTC waveformmode with toggle Compare Output mode (COMnx1:0 = 1).

Figure 66. Output Compare Modulator, Timing Diagram

In this example, Timer/Counter2 provides the carrier, while the modulating signal is gen-erated by the Output Compare unit C of the Timer/Counter1.

The resolution of the PWM signal (OC1C) is reduced by the modulation. The reductionfactor is equal to the number of system clock cycles of one period of the carrier (OC0A).In this example the resolution is reduced by a factor of two. The reason for the reductionis illustrated in Figure 66 at the second and third period of the PB7 output whenPORTB7 equals zero. The period 2 high time is one cycle longer than the period 3 hightime, but the result on the PB7 output is equal in both periods.

1 2

OC0A(CTC Mode)

OC1C(FPWM Mode)

PB7(PORTB7 = 0)

PB7(PORTB7 = 1)

(Period)3

clk I/O

173

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

8-bit Timer/Counter2 with PWM and Asynchronous Operation

Timer/Counter2 is a general purpose, single channel, 8-bit Timer/Counter module. Themain features are:• Single Channel Counter• Clear Timer on Compare Match (Auto Reload)• Glitch-free, Phase Correct Pulse Width Modulator (PWM)• Frequency Generator• 10-bit Clock Prescaler• Overflow and Compare Match Interrupt Sources (TOV2, OCF2A and OCF2B)• Allows Clocking from External 32 kHz Watch Crystal Independent of the I/O Clock

Overview A simplified block diagram of the 8-bit Timer/Counter is shown in Figure 60.. For theactual placement of I/O pins, see “Pin Configurations” on page 2. CPU accessible I/ORegisters, including I/O bits and I/O pins, are shown in bold. The device-specific I/ORegister and bit locations are listed in the “8-bit Timer/Counter Register Description” onpage 184.

The Power Reduction Timer/Counter2 bit, PRTIM2, in “Power Reduction Register 0 -PRR0” on page 54 must be written to zero to enable Timer/Counter2 module.

Figure 67. 8-bit Timer/Counter Block Diagram

Registers The Timer/Counter (TCNT2) and Output Compare Register (OCR2A and OCR2B) are8-bit registers. Interrupt request (abbreviated to Int.Req.) signals are all visible in theTimer Interrupt Flag Register (TIFR2). All interrupts are individually masked with theTimer Interrupt Mask Register (TIMSK2). TIFR2 and TIMSK2 are not shown in thefigure.

Timer/Counter

DAT

A B

US

OCRnA

OCRnB

=

=

TCNTn

WaveformGeneration

WaveformGeneration

OCnA

OCnB

=

FixedTOP

Value

Control Logic

= 0

TOP BOTTOM

Count

Clear

Direction

TOVn(Int.Req.)

OCnA(Int.Req.)

OCnB(Int.Req.)

TCCRnA TCCRnB

clkTn

ASSRn

Synchronization Unit

Prescaler

T/COscillator

clkI/O

clkASY

asynchronous mode select (ASn)

Synchronized Status flags

TOSC1

TOSC2

Status flags

clkI/O

174 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The Timer/Counter can be clocked internally, via the prescaler, or asynchronouslyclocked from the TOSC1/2 pins, as detailed later in this section. The asynchronousoperation is controlled by the Asynchronous Status Register (ASSR). The Clock Selectlogic block controls which clock source the Timer/Counter uses to increment (or decre-ment) its value. The Timer/Counter is inactive when no clock source is selected. Theoutput from the Clock Select logic is referred to as the timer clock (clkT2).

The double buffered Output Compare Register (OCR2A and OCR2B) are comparedwith the Timer/Counter value at all times. The result of the compare can be used by theWaveform Generator to generate a PWM or variable frequency output on the OutputCompare pins (OC2A and OC2B). See “Output Compare Unit” on page 175. for details.The compare match event will also set the Compare Flag (OCF2A or OCF2B) which canbe used to generate an Output Compare interrupt request.

Definitions Many register and bit references in this document are written in general form. A lowercase “n” replaces the Timer/Counter number, in this case 2. However, when using theregister or bit defines in a program, the precise form must be used, i.e., TCNT2 foraccessing Timer/Counter2 counter value and so on.

The definitions in Table 84 are also used extensively throughout the section.

Timer/Counter Clock Sources

The Timer/Counter can be clocked by an internal synchronous or an external asynchro-nous clock source. The clock source clkT2 is by default equal to the MCU clock, clkI/O.When the AS2 bit in the ASSR Register is written to logic one, the clock source is takenfrom the Timer/Counter Oscillator connected to TOSC1 and TOSC2. For details onasynchronous operation, see “Asynchronous Status Register – ASSR” on page 189. Fordetails on clock sources and prescaler, see “Timer/Counter Prescaler” on page 193.

Counter Unit The main part of the 8-bit Timer/Counter is the programmable bi-directional counter unit.Figure 68 shows a block diagram of the counter and its surrounding environment.

Figure 68. Counter Unit Block Diagram

Table 84. Definitions

BOTTOM The counter reaches the BOTTOM when it becomes zero (0x00).

MAX The counter reaches its MAXimum when it becomes 0xFF (decimal 255).

TOP The counter reaches the TOP when it becomes equal to the highestvalue in the count sequence. The TOP value can be assigned to be thefixed value 0xFF (MAX) or the value stored in the OCR2A Register. Theassignment is dependent on the mode of operation.

DATA BUS

TCNTn Control Logic

count

TOVn(Int.Req.)

topbottom

direction

clear

TOSC1

T/COscillator

TOSC2

Prescaler

clkI/O

clkTn

175

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Signal description (internal signals):

count Increment or decrement TCNT2 by 1.

direction Selects between increment and decrement.

clear Clear TCNT2 (set all bits to zero).

clkTn Timer/Counter clock, referred to as clkT2 in the following.

top Signalizes that TCNT2 has reached maximum value.

bottom Signalizes that TCNT2 has reached minimum value (zero).

Depending on the mode of operation used, the counter is cleared, incremented, or dec-remented at each timer clock (clkT2). clkT2 can be generated from an external or internalclock source, selected by the Clock Select bits (CS22:0). When no clock source isselected (CS22:0 = 0) the timer is stopped. However, the TCNT2 value can be accessedby the CPU, regardless of whether clkT2 is present or not. A CPU write overrides (haspriority over) all counter clear or count operations.

The counting sequence is determined by the setting of the WGM21 and WGM20 bitslocated in the Timer/Counter Control Register (TCCR2A) and the WGM22 located in theTimer/Counter Control Register B (TCCR2B). There are close connections betweenhow the counter behaves (counts) and how waveforms are generated on the OutputCompare outputs OC2A and OC2B. For more details about advanced countingsequences and waveform generation, see “Modes of Operation” on page 178.

The Timer/Counter Overflow Flag (TOV2) is set according to the mode of operationselected by the WGM22:0 bits. TOV2 can be used for generating a CPU interrupt.

Output Compare Unit The 8-bit comparator continuously compares TCNT2 with the Output Compare Register(OCR2A and OCR2B). Whenever TCNT2 equals OCR2A or OCR2B, the comparatorsignals a match. A match will set the Output Compare Flag (OCF2A or OCF2B) at thenext timer clock cycle. If the corresponding interrupt is enabled, the Output CompareFlag generates an Output Compare interrupt. The Output Compare Flag is automaticallycleared when the interrupt is executed. Alternatively, the Output Compare Flag can becleared by software by writing a logical one to its I/O bit location. The Waveform Gener-ator uses the match signal to generate an output according to operating mode set by theWGM22:0 bits and Compare Output mode (COM2x1:0) bits. The max and bottom sig-nals are used by the Waveform Generator for handling the special cases of the extremevalues in some modes of operation (“Modes of Operation” on page 178).

Figure 58 on page 154 shows a block diagram of the Output Compare unit.

176 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 69. Output Compare Unit, Block Diagram

The OCR2x Register is double buffered when using any of the Pulse Width Modulation(PWM) modes. For the Normal and Clear Timer on Compare (CTC) modes of operation,the double buffering is disabled. The double buffering synchronizes the update of theOCR2x Compare Register to either top or bottom of the counting sequence. The syn-chronization prevents the occurrence of odd-length, non-symmetrical PWM pulses,thereby making the output glitch-free.

The OCR2x Register access may seem complex, but this is not case. When the doublebuffering is enabled, the CPU has access to the OCR2x Buffer Register, and if doublebuffering is disabled the CPU will access the OCR2x directly.

Force Output Compare In non-PWM waveform generation modes, the match output of the comparator can beforced by writing a one to the Force Output Compare (FOC2x) bit. Forcing comparematch will not set the OCF2x Flag or reload/clear the timer, but the OC2x pin will beupdated as if a real compare match had occurred (the COM2x1:0 bits settings definewhether the OC2x pin is set, cleared or toggled).

Compare Match Blocking by TCNT2 Write

All CPU write operations to the TCNT2 Register will block any compare match thatoccurs in the next timer clock cycle, even when the timer is stopped. This feature allowsOCR2x to be initialized to the same value as TCNT2 without triggering an interrupt whenthe Timer/Counter clock is enabled.

Using the Output Compare Unit

Since writing TCNT2 in any mode of operation will block all compare matches for onetimer clock cycle, there are risks involved when changing TCNT2 when using the OutputCompare channel, independently of whether the Timer/Counter is running or not. If thevalue written to TCNT2 equals the OCR2x value, the compare match will be missed,resulting in incorrect waveform generation. Similarly, do not write the TCNT2 valueequal to BOTTOM when the counter is downcounting.

The setup of the OC2x should be performed before setting the Data Direction Registerfor the port pin to output. The easiest way of setting the OC2x value is to use the ForceOutput Compare (FOC2x) strobe bit in Normal mode. The OC2x Register keeps itsvalue even when changing between Waveform Generation modes.

OCFnx (Int.Req.)

= (8-bit Comparator )

OCRnx

OCnx

DATA BUS

TCNTn

WGMn1:0

Waveform Generator

top

FOCn

COMnX1:0

bottom

177

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Be aware that the COM2x1:0 bits are not double buffered together with the comparevalue. Changing the COM2x1:0 bits will take effect immediately.

Compare Match Output Unit

The Compare Output mode (COM2x1:0) bits have two functions. The Waveform Gener-ator uses the COM2x1:0 bits for defining the Output Compare (OC2x) state at the nextcompare match. Also, the COM2x1:0 bits control the OC2x pin output source. Figure 70shows a simplified schematic of the logic affected by the COM2x1:0 bit setting. The I/ORegisters, I/O bits, and I/O pins in the figure are shown in bold. Only the parts of thegeneral I/O Port Control Registers (DDR and PORT) that are affected by the COM2x1:0bits are shown. When referring to the OC2x state, the reference is for the internal OC2xRegister, not the OC2x pin.

Figure 70. Compare Match Output Unit, Schematic

The general I/O port function is overridden by the Output Compare (OC2x) from theWaveform Generator if either of the COM2x1:0 bits are set. However, the OC2x pindirection (input or output) is still controlled by the Data Direction Register (DDR) for theport pin. The Data Direction Register bit for the OC2x pin (DDR_OC2x) must be set asoutput before the OC2x value is visible on the pin. The port override function is indepen-dent of the Waveform Generation mode.

The design of the Output Compare pin logic allows initialization of the OC2x state beforethe output is enabled. Note that some COM2x1:0 bit settings are reserved for certainmodes of operation. See “8-bit Timer/Counter Register Description” on page 184.

Compare Output Mode and Waveform Generation

The Waveform Generator uses the COM2x1:0 bits differently in normal, CTC, and PWMmodes. For all modes, setting the COM2x1:0 = 0 tells the Waveform Generator that noaction on the OC2x Register is to be performed on the next compare match. For com-pare output actions in the non-PWM modes refer to Table 88 on page 185. For fastPWM mode, refer to Table 89 on page 185, and for phase correct PWM refer to Table90 on page 186.

PORT

DDR

D Q

D Q

OCnxPinOCnx

D QWaveformGenerator

COMnx1

COMnx0

0

1

DAT

A B

US

FOCnx

clkI/O

178 ATmega640/1280/1281/2560/25612549A–AVR–03/05

A change of the COM2x1:0 bits state will have effect at the first compare match after thebits are written. For non-PWM modes, the action can be forced to have immediate effectby using the FOC2x strobe bits.

Modes of Operation The mode of operation, i.e., the behavior of the Timer/Counter and the Output Comparepins, is defined by the combination of the Waveform Generation mode (WGM22:0) andCompare Output mode (COM2x1:0) bits. The Compare Output mode bits do not affectthe counting sequence, while the Waveform Generation mode bits do. The COM2x1:0bits control whether the PWM output generated should be inverted or not (inverted ornon-inverted PWM). For non-PWM modes the COM2x1:0 bits control whether the out-put should be set, cleared, or toggled at a compare match (See “Compare Match OutputUnit” on page 177.).

For detailed timing information refer to “Timer/Counter Timing Diagrams” on page 182.

Normal Mode The simplest mode of operation is the Normal mode (WGM22:0 = 0). In this mode thecounting direction is always up (incrementing), and no counter clear is performed. Thecounter simply overruns when it passes its maximum 8-bit value (TOP = 0xFF) and thenrestarts from the bottom (0x00). In normal operation the Timer/Counter Overflow Flag(TOV2) will be set in the same timer clock cycle as the TCNT2 becomes zero. TheTOV2 Flag in this case behaves like a ninth bit, except that it is only set, not cleared.However, combined with the timer overflow interrupt that automatically clears the TOV2Flag, the timer resolution can be increased by software. There are no special cases toconsider in the Normal mode, a new counter value can be written anytime.

The Output Compare unit can be used to generate interrupts at some given time. Usingthe Output Compare to generate waveforms in Normal mode is not recommended,since this will occupy too much of the CPU time.

Clear Timer on Compare Match (CTC) Mode

In Clear Timer on Compare or CTC mode (WGM22:0 = 2), the OCR2A Register is usedto manipulate the counter resolution. In CTC mode the counter is cleared to zero whenthe counter value (TCNT2) matches the OCR2A. The OCR2A defines the top value forthe counter, hence also its resolution. This mode allows greater control of the comparematch output frequency. It also simplifies the operation of counting external events.

The timing diagram for the CTC mode is shown in Table 71. The counter value (TCNT2)increases until a compare match occurs between TCNT2 and OCR2A, and then counter(TCNT2) is cleared.

Figure 71. CTC Mode, Timing Diagram

An interrupt can be generated each time the counter value reaches the TOP value byusing the OCF2A Flag. If the interrupt is enabled, the interrupt handler routine can be

TCNTn

OCnx(Toggle)

OCnx Interrupt Flag Set

1 4Period 2 3

(COMnx1:0 = 1)

179

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

used for updating the TOP value. However, changing TOP to a value close to BOTTOMwhen the counter is running with none or a low prescaler value must be done with caresince the CTC mode does not have the double buffering feature. If the new value writtento OCR2A is lower than the current value of TCNT2, the counter will miss the comparematch. The counter will then have to count to its maximum value (0xFF) and wraparound starting at 0x00 before the compare match can occur.

For generating a waveform output in CTC mode, the OC2A output can be set to toggleits logical level on each compare match by setting the Compare Output mode bits to tog-gle mode (COM2A1:0 = 1). The OC2A value will not be visible on the port pin unless thedata direction for the pin is set to output. The waveform generated will have a maximumfrequency of fOC2A = fclk_I/O/2 when OCR2A is set to zero (0x00). The waveform fre-quency is defined by the following equation:

The N variable represents the prescale factor (1, 8, 32, 64, 128, 256, or 1024).

As for the Normal mode of operation, the TOV2 Flag is set in the same timer clock cyclethat the counter counts from MAX to 0x00.

Fast PWM Mode The fast Pulse Width Modulation or fast PWM mode (WGM22:0 = 3 or 7) provides a highfrequency PWM waveform generation option. The fast PWM differs from the other PWMoption by its single-slope operation. The counter counts from BOTTOM to TOP thenrestarts from BOTTOM. TOP is defined as 0xFF when WGM22:0 = 3, and OCR2A whenMGM22:0 = 7. In non-inverting Compare Output mode, the Output Compare (OC2x) iscleared on the compare match between TCNT2 and OCR2x, and set at BOTTOM. Ininverting Compare Output mode, the output is set on compare match and cleared atBOTTOM. Due to the single-slope operation, the operating frequency of the fast PWMmode can be twice as high as the phase correct PWM mode that uses dual-slope oper-ation. This high frequency makes the fast PWM mode well suited for power regulation,rectification, and DAC applications. High frequency allows physically small sized exter-nal components (coils, capacitors), and therefore reduces total system cost.

In fast PWM mode, the counter is incremented until the counter value matches the TOPvalue. The counter is then cleared at the following timer clock cycle. The timing diagramfor the fast PWM mode is shown in Figure 61. The TCNT2 value is in the timing diagramshown as a histogram for illustrating the single-slope operation. The diagram includesnon-inverted and inverted PWM outputs. The small horizontal line marks on the TCNT2slopes represent compare matches between OCR2x and TCNT2.

fOCnxfclk_I/O

2 N 1 OCRnx+( )⋅ ⋅--------------------------------------------------=

180 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 72. Fast PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOV2) is set each time the counter reaches TOP. Ifthe interrupt is enabled, the interrupt handler routine can be used for updating the com-pare value.

In fast PWM mode, the compare unit allows generation of PWM waveforms on theOC2x pin. Setting the COM2x1:0 bits to two will produce a non-inverted PWM and aninverted PWM output can be generated by setting the COM2x1:0 to three. TOP isdefined as 0xFF when WGM2:0 = 3, and OCR2A when WGM2:0 = 7 (See Table 86 onpage 184). The actual OC2x value will only be visible on the port pin if the data directionfor the port pin is set as output. The PWM waveform is generated by setting (or clearing)the OC2x Register at the compare match between OCR2x and TCNT2, and clearing (orsetting) the OC2x Register at the timer clock cycle the counter is cleared (changes fromTOP to BOTTOM).

The PWM frequency for the output can be calculated by the following equation:

The N variable represents the prescale factor (1, 8, 32, 64, 128, 256, or 1024).

The extreme values for the OCR2A Register represent special cases when generating aPWM waveform output in the fast PWM mode. If the OCR2A is set equal to BOTTOM,the output will be a narrow spike for each MAX+1 timer clock cycle. Setting the OCR2Aequal to MAX will result in a constantly high or low output (depending on the polarity ofthe output set by the COM2A1:0 bits.)

A frequency (with 50% duty cycle) waveform output in fast PWM mode can be achievedby setting OC2x to toggle its logical level on each compare match (COM2x1:0 = 1). Thewaveform generated will have a maximum frequency of foc2 = fclk_I/O/2 when OCR2A isset to zero. This feature is similar to the OC2A toggle in CTC mode, except the doublebuffer feature of the Output Compare unit is enabled in the fast PWM mode.

TCNTn

OCRnx Update andTOVn Interrupt Flag Set

1Period 2 3

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)

OCRnx Interrupt Flag Set

4 5 6 7

fOCnxPWMfclk_I/O

N 256⋅------------------=

181

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Phase Correct PWM Mode The phase correct PWM mode (WGM22:0 = 1 or 5) provides a high resolution phasecorrect PWM waveform generation option. The phase correct PWM mode is based on adual-slope operation. The counter counts repeatedly from BOTTOM to TOP and thenfrom TOP to BOTTOM. TOP is defined as 0xFF when WGM22:0 = 1, and OCR2A whenMGM22:0 = 5. In non-inverting Compare Output mode, the Output Compare (OC2x) iscleared on the compare match between TCNT2 and OCR2x while upcounting, and seton the compare match while downcounting. In inverting Output Compare mode, theoperation is inverted. The dual-slope operation has lower maximum operation frequencythan single slope operation. However, due to the symmetric feature of the dual-slopePWM modes, these modes are preferred for motor control applications.

In phase correct PWM mode the counter is incremented until the counter value matchesTOP. When the counter reaches TOP, it changes the count direction. The TCNT2 valuewill be equal to TOP for one timer clock cycle. The timing diagram for the phase correctPWM mode is shown on Figure 73. The TCNT2 value is in the timing diagram shown asa histogram for illustrating the dual-slope operation. The diagram includes non-invertedand inverted PWM outputs. The small horizontal line marks on the TCNT2 slopes repre-sent compare matches between OCR2x and TCNT2.

Figure 73. Phase Correct PWM Mode, Timing Diagram

The Timer/Counter Overflow Flag (TOV2) is set each time the counter reaches BOT-TOM. The Interrupt Flag can be used to generate an interrupt each time the counterreaches the BOTTOM value.

In phase correct PWM mode, the compare unit allows generation of PWM waveforms onthe OC2x pin. Setting the COM2x1:0 bits to two will produce a non-inverted PWM. Aninverted PWM output can be generated by setting the COM2x1:0 to three. TOP isdefined as 0xFF when WGM2:0 = 3, and OCR2A when MGM2:0 = 7 (See Table 87 onpage 185). The actual OC2x value will only be visible on the port pin if the data directionfor the port pin is set as output. The PWM waveform is generated by clearing (or setting)the OC2x Register at the compare match between OCR2x and TCNT2 when thecounter increments, and setting (or clearing) the OC2x Register at compare match

TOVn Interrupt Flag Set

OCnx Interrupt Flag Set

1 2 3

TCNTn

Period

OCnx

OCnx

(COMnx1:0 = 2)

(COMnx1:0 = 3)

OCRnx Update

182 ATmega640/1280/1281/2560/25612549A–AVR–03/05

between OCR2x and TCNT2 when the counter decrements. The PWM frequency for theoutput when using phase correct PWM can be calculated by the following equation:

The N variable represents the prescale factor (1, 8, 32, 64, 128, 256, or 1024).

The extreme values for the OCR2A Register represent special cases when generating aPWM waveform output in the phase correct PWM mode. If the OCR2A is set equal toBOTTOM, the output will be continuously low and if set equal to MAX the output will becontinuously high for non-inverted PWM mode. For inverted PWM the output will havethe opposite logic values.

At the very start of period 2 in Figure 73 OCnx has a transition from high to low eventhough there is no Compare Match. The point of this transition is to guarantee symmetryaround BOTTOM. There are two cases that give a transition without Compare Match.

• OCR2A changes its value from MAX, like in Figure 73. When the OCR2A value is MAX the OCn pin value is the same as the result of a down-counting compare match. To ensure symmetry around BOTTOM the OCn value at MAX must correspond to the result of an up-counting Compare Match.

• The timer starts counting from a value higher than the one in OCR2A, and for that reason misses the Compare Match and hence the OCn change that would have happened on the way up.

Timer/Counter Timing Diagrams

The following figures show the Timer/Counter in synchronous mode, and the timer clock(clkT2) is therefore shown as a clock enable signal. In asynchronous mode, clkI/O shouldbe replaced by the Timer/Counter Oscillator clock. The figures include information onwhen Interrupt Flags are set. Figure 74 contains timing data for basic Timer/Counteroperation. The figure shows the count sequence close to the MAX value in all modesother than phase correct PWM mode.

Figure 74. Timer/Counter Timing Diagram, no Prescaling

Figure 75 shows the same timing data, but with the prescaler enabled.

fOCnxPCPWMfclk_I/O

N 510⋅------------------=

clkTn(clkI/O/1)

TOVn

clkI/O

TCNTn MAX - 1 MAX BOTTOM BOTTOM + 1

183

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 75. Timer/Counter Timing Diagram, with Prescaler (fclk_I/O/8)

Figure 76 shows the setting of OCF2A in all modes except CTC mode.

Figure 76. Timer/Counter Timing Diagram, Setting of OCF2A, with Prescaler (fclk_I/O/8)

Figure 77 shows the setting of OCF2A and the clearing of TCNT2 in CTC mode.

Figure 77. Timer/Counter Timing Diagram, Clear Timer on Compare Match mode, withPrescaler (fclk_I/O/8)

TOVn

TCNTn MAX - 1 MAX BOTTOM BOTTOM + 1

clkI/O

clkTn(clkI/O/8)

OCFnx

OCRnx

TCNTn

OCRnx Value

OCRnx - 1 OCRnx OCRnx + 1 OCRnx + 2

clkI/O

clkTn(clkI/O/8)

OCFnx

OCRnx

TCNTn(CTC)

TOP

TOP - 1 TOP BOTTOM BOTTOM + 1

clkI/O

clkTn(clkI/O/8)

184 ATmega640/1280/1281/2560/25612549A–AVR–03/05

8-bit Timer/Counter Register Description

Timer/Counter Control Register A – TCCR2A

• Bits 7:6 – COM2A1:0: Compare Match Output A Mode

These bits control the Output Compare pin (OC2A) behavior. If one or both of theCOM2A1:0 bits are set, the OC2A output overrides the normal port functionality of theI/O pin it is connected to. However, note that the Data Direction Register (DDR) bit cor-responding to the OC2A pin must be set in order to enable the output driver.

When OC2A is connected to the pin, the function of the COM2A1:0 bits depends on theWGM22:0 bit setting. Table 85 shows the COM2A1:0 bit functionality when theWGM22:0 bits are set to a normal or CTC mode (non-PWM).

Table 86 shows the COM2A1:0 bit functionality when the WGM21:0 bits are set to fastPWM mode.

Note: 1. A special case occurs when OCR2A equals TOP and COM2A1 is set. In this case,the Compare Match is ignored, but the set or clear is done at TOP. See “Fast PWMMode” on page 179 for more details.

Bit 7 6 5 4 3 2 1 0

COM2A1 COM2A0 COM2B1 COM2B0 – – WGM21 WGM20 TCCR2A

Read/Write R/W R/W R/W R/W R R R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 85. Compare Output Mode, non-PWM Mode

COM2A1 COM2A0 Description

0 0 Normal port operation, OC0A disconnected.

0 1 Toggle OC2A on Compare Match

1 0 Clear OC2A on Compare Match

1 1 Set OC2A on Compare Match

Table 86. Compare Output Mode, Fast PWM Mode(1)

COM2A1 COM2A0 Description

0 0 Normal port operation, OC2A disconnected.

0 1 WGM22 = 0: Normal Port Operation, OC0A Disconnected.WGM22 = 1: Toggle OC2A on Compare Match.

1 0 Clear OC2A on Compare Match, set OC2A at TOP

1 1 Set OC2A on Compare Match, clear OC2A at TOP

185

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 87 shows the COM2A1:0 bit functionality when the WGM22:0 bits are set tophase correct PWM mode.

Note: 1. A special case occurs when OCR2A equals TOP and COM2A1 is set. In this case,the Compare Match is ignored, but the set or clear is done at TOP. See “Phase Cor-rect PWM Mode” on page 181 for more details.

• Bits 5:4 – COM2B1:0: Compare Match Output B Mode

These bits control the Output Compare pin (OC2B) behavior. If one or both of theCOM2B1:0 bits are set, the OC2B output overrides the normal port functionality of theI/O pin it is connected to. However, note that the Data Direction Register (DDR) bit cor-responding to the OC2B pin must be set in order to enable the output driver.

When OC2B is connected to the pin, the function of the COM2B1:0 bits depends on theWGM22:0 bit setting. Table 88 shows the COM2B1:0 bit functionality when theWGM22:0 bits are set to a normal or CTC mode (non-PWM).

Table 89 shows the COM2B1:0 bit functionality when the WGM22:0 bits are set to fastPWM mode.

Note: 1. A special case occurs when OCR2B equals TOP and COM2B1 is set. In this case,the Compare Match is ignored, but the set or clear is done at TOP. See “Fast PWMMode” on page 179 for more details.

Table 87. Compare Output Mode, Phase Correct PWM Mode(1)

COM2A1 COM2A0 Description

0 0 Normal port operation, OC2A disconnected.

0 1 WGM22 = 0: Normal Port Operation, OC2A Disconnected.WGM22 = 1: Toggle OC2A on Compare Match.

1 0 Clear OC2A on Compare Match when up-counting. Set OC2A on Compare Match when down-counting.

1 1 Set OC2A on Compare Match when up-counting. Clear OC2A on Compare Match when down-counting.

Table 88. Compare Output Mode, non-PWM Mode

COM2B1 COM2B0 Description

0 0 Normal port operation, OC2B disconnected.

0 1 Toggle OC2B on Compare Match

1 0 Clear OC2B on Compare Match

1 1 Set OC2B on Compare Match

Table 89. Compare Output Mode, Fast PWM Mode(1)

COM2B1 COM2B0 Description

0 0 Normal port operation, OC2B disconnected.

0 1 Reserved

1 0 Clear OC2B on Compare Match, set OC2B at TOP

1 1 Set OC2B on Compare Match, clear OC2B at TOP

186 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Table 90 shows the COM2B1:0 bit functionality when the WGM22:0 bits are set tophase correct PWM mode.

Note: 1. A special case occurs when OCR2B equals TOP and COM2B1 is set. In this case,the Compare Match is ignored, but the set or clear is done at TOP. See “Phase Cor-rect PWM Mode” on page 181 for more details.

• Bits 3, 2 – Res: Reserved Bits

These bits are reserved bits and will always read as zero.

• Bits 1:0 – WGM21:0: Waveform Generation Mode

Combined with the WGM22 bit found in the TCCR2B Register, these bits control thecounting sequence of the counter, the source for maximum (TOP) counter value, andwhat type of waveform generation to be used, see Table 91. Modes of operation sup-ported by the Timer/Counter unit are: Normal mode (counter), Clear Timer on CompareMatch (CTC) mode, and two types of Pulse Width Modulation (PWM) modes (see“Modes of Operation” on page 178).

Notes: 1. MAX= 0xFF2. BOTTOM= 0x00

Table 90. Compare Output Mode, Phase Correct PWM Mode(1)

COM2B1 COM2B0 Description

0 0 Normal port operation, OC2B disconnected.

0 1 Reserved

1 0 Clear OC2B on Compare Match when up-counting. Set OC2B on Compare Match when down-counting.

1 1 Set OC2B on Compare Match when up-counting. Clear OC2B on Compare Match when down-counting.

Table 91. Waveform Generation Mode Bit Description

Mode WGM2 WGM1 WGM0

Timer/Counter Mode of Operation TOP

Update ofOCRx at

TOV FlagSet on(1)(2)

0 0 0 0 Normal 0xFF Immediate MAX

1 0 0 1 PWM, Phase Correct

0xFF TOP BOTTOM

2 0 1 0 CTC OCRA Immediate MAX

3 0 1 1 Fast PWM 0xFF TOP MAX

4 1 0 0 Reserved – – –

5 1 0 1 PWM, Phase Correct

OCRA TOP BOTTOM

6 1 1 0 Reserved – – –

7 1 1 1 Fast PWM OCRA TOP TOP

187

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Timer/Counter Control Register B – TCCR2B

• Bit 7 – FOC2A: Force Output Compare A

The FOC2A bit is only active when the WGM bits specify a non-PWM mode.

However, for ensuring compatibility with future devices, this bit must be set to zero whenTCCR2B is written when operating in PWM mode. When writing a logical one to theFOC2A bit, an immediate Compare Match is forced on the Waveform Generation unit.The OC2A output is changed according to its COM2A1:0 bits setting. Note that theFOC2A bit is implemented as a strobe. Therefore it is the value present in theCOM2A1:0 bits that determines the effect of the forced compare.

A FOC2A strobe will not generate any interrupt, nor will it clear the timer in CTC modeusing OCR2A as TOP.

The FOC2A bit is always read as zero.

• Bit 6 – FOC2B: Force Output Compare B

The FOC2B bit is only active when the WGM bits specify a non-PWM mode.

However, for ensuring compatibility with future devices, this bit must be set to zero whenTCCR2B is written when operating in PWM mode. When writing a logical one to theFOC2B bit, an immediate Compare Match is forced on the Waveform Generation unit.The OC2B output is changed according to its COM2B1:0 bits setting. Note that theFOC2B bit is implemented as a strobe. Therefore it is the value present in theCOM2B1:0 bits that determines the effect of the forced compare.

A FOC2B strobe will not generate any interrupt, nor will it clear the timer in CTC modeusing OCR2B as TOP.

The FOC2B bit is always read as zero.

• Bits 5:4 – Res: Reserved Bits

These bits are reserved bits and will always read as zero.

• Bit 3 – WGM22: Waveform Generation Mode

See the description in the “Timer/Counter Control Register A – TCCR2A” on page 184.

• Bit 2:0 – CS22:0: Clock Select

The three Clock Select bits select the clock source to be used by the Timer/Counter, seeTable 92.

Bit 7 6 5 4 3 2 1 0

FOC2A FOC2B – – WGM22 CS22 CS21 CS20 TCCR2B

Read/Write W W R R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

188 ATmega640/1280/1281/2560/25612549A–AVR–03/05

If external pin modes are used for the Timer/Counter0, transitions on the T0 pin willclock the counter even if the pin is configured as an output. This feature allows softwarecontrol of the counting.

Timer/Counter Register – TCNT2

The Timer/Counter Register gives direct access, both for read and write operations, tothe Timer/Counter unit 8-bit counter. Writing to the TCNT2 Register blocks (removes)the Compare Match on the following timer clock. Modifying the counter (TCNT2) whilethe counter is running, introduces a risk of missing a Compare Match between TCNT2and the OCR2x Registers.

Output Compare Register A – OCR2A

The Output Compare Register A contains an 8-bit value that is continuously comparedwith the counter value (TCNT2). A match can be used to generate an Output Compareinterrupt, or to generate a waveform output on the OC2A pin.

Output Compare Register B – OCR2B

The Output Compare Register B contains an 8-bit value that is continuously comparedwith the counter value (TCNT2). A match can be used to generate an Output Compareinterrupt, or to generate a waveform output on the OC2B pin.

Table 92. Clock Select Bit Description

CS22 CS21 CS20 Description

0 0 0 No clock source (Timer/Counter stopped).

0 0 1 clkT2S/(No prescaling)

0 1 0 clkT2S/8 (From prescaler)

0 1 1 clkT2S/32 (From prescaler)

1 0 0 clkT2S/64 (From prescaler)

1 0 1 clkT2S/128 (From prescaler)

1 1 0 clkT2S/256 (From prescaler)

1 1 1 clkT2S/1024 (From prescaler)

Bit 7 6 5 4 3 2 1 0

TCNT2[7:0] TCNT2

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR2A[7:0] OCR2A

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

OCR2B[7:0] OCR2B

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

189

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Asynchronous operation of the Timer/Counter

Asynchronous Status Register – ASSR

• Bit 6 – EXCLK: Enable External Clock Input

When EXCLK is written to one, and asynchronous clock is selected, the external clockinput buffer is enabled and an external clock can be input on Timer Oscillator 1 (TOSC1)pin instead of a 32 kHz crystal. Writing to EXCLK should be done before asynchronousoperation is selected. Note that the crystal Oscillator will only run when this bit is zero.

• Bit 5 – AS2: Asynchronous Timer/Counter2

When AS2 is written to zero, Timer/Counter2 is clocked from the I/O clock, clkI/O. WhenAS2 is written to one, Timer/Counter2 is clocked from a crystal Oscillator connected tothe Timer Oscillator 1 (TOSC1) pin. When the value of AS2 is changed, the contents ofTCNT2, OCR2A, OCR2B, TCCR2A and TCCR2B might be corrupted.

• Bit 4 – TCN2UB: Timer/Counter2 Update Busy

When Timer/Counter2 operates asynchronously and TCNT2 is written, this bit becomesset. When TCNT2 has been updated from the temporary storage register, this bit iscleared by hardware. A logical zero in this bit indicates that TCNT2 is ready to beupdated with a new value.

• Bit 3 – OCR2AUB: Output Compare Register2 Update Busy

When Timer/Counter2 operates asynchronously and OCR2A is written, this bit becomesset. When OCR2A has been updated from the temporary storage register, this bit iscleared by hardware. A logical zero in this bit indicates that OCR2A is ready to beupdated with a new value.

• Bit 2 – OCR2BUB: Output Compare Register2 Update Busy

When Timer/Counter2 operates asynchronously and OCR2B is written, this bit becomesset. When OCR2B has been updated from the temporary storage register, this bit iscleared by hardware. A logical zero in this bit indicates that OCR2B is ready to beupdated with a new value.

• Bit 1 – TCR2AUB: Timer/Counter Control Register2 Update Busy

When Timer/Counter2 operates asynchronously and TCCR2A is written, this bitbecomes set. When TCCR2A has been updated from the temporary storage register,this bit is cleared by hardware. A logical zero in this bit indicates that TCCR2A is readyto be updated with a new value.

• Bit 0 – TCR2BUB: Timer/Counter Control Register2 Update Busy

When Timer/Counter2 operates asynchronously and TCCR2B is written, this bitbecomes set. When TCCR2B has been updated from the temporary storage register,this bit is cleared by hardware. A logical zero in this bit indicates that TCCR2B is readyto be updated with a new value.

Bit 7 6 5 4 3 2 1 0

– EXCLK AS2 TCN2UB OCR2AUB OCR2BUB TCR2AUB TCR2BUB ASSR

Read/Write R R/W R/W R R R R R

Initial Value 0 0 0 0 0 0 0 0

190 ATmega640/1280/1281/2560/25612549A–AVR–03/05

If a write is performed to any of the five Timer/Counter2 Registers while its update busyflag is set, the updated value might get corrupted and cause an unintentional interrupt tooccur.

The mechanisms for reading TCNT2, OCR2A, OCR2B, TCCR2A and TCCR2B are dif-ferent. When reading TCNT2, the actual timer value is read. When reading OCR2A,OCR2B, TCCR2A and TCCR2B the value in the temporary storage register is read.

Asynchronous Operation of Timer/Counter2

When Timer/Counter2 operates asynchronously, some considerations must be taken.

• Warning: When switching between asynchronous and synchronous clocking of Timer/Counter2, the Timer Registers TCNT2, OCR2x, and TCCR2x might be corrupted. A safe procedure for switching clock source is:

1. Disable the Timer/Counter2 interrupts by clearing OCIE2x and TOIE2.

2. Select clock source by setting AS2 as appropriate.

3. Write new values to TCNT2, OCR2x, and TCCR2x.

4. To switch to asynchronous operation: Wait for TCN2UB, OCR2xUB, and TCR2xUB.

5. Clear the Timer/Counter2 Interrupt Flags.

6. Enable interrupts, if needed.

• The CPU main clock frequency must be more than four times the Oscillator frequency.

• When writing to one of the registers TCNT2, OCR2x, or TCCR2x, the value is transferred to a temporary register, and latched after two positive edges on TOSC1. The user should not write a new value before the contents of the temporary register have been transferred to its destination. Each of the five mentioned registers have their individual temporary register, which means that e.g. writing to TCNT2 does not disturb an OCR2x write in progress. To detect that a transfer to the destination register has taken place, the Asynchronous Status Register – ASSR has been implemented.

• When entering Power-save or ADC Noise Reduction mode after having written to TCNT2, OCR2x, or TCCR2x, the user must wait until the written register has been updated if Timer/Counter2 is used to wake up the device. Otherwise, the MCU will enter sleep mode before the changes are effective. This is particularly important if any of the Output Compare2 interrupt is used to wake up the device, since the Output Compare function is disabled during writing to OCR2x or TCNT2. If the write cycle is not finished, and the MCU enters sleep mode before the corresponding OCR2xUB bit returns to zero, the device will never receive a compare match interrupt, and the MCU will not wake up.

• If Timer/Counter2 is used to wake the device up from Power-save or ADC Noise Reduction mode, precautions must be taken if the user wants to re-enter one of these modes: The interrupt logic needs one TOSC1 cycle to be reset. If the time between wake-up and re-entering sleep mode is less than one TOSC1 cycle, the interrupt will not occur, and the device will fail to wake up. If the user is in doubt whether the time before re-entering Power-save or ADC Noise Reduction mode is sufficient, the following algorithm can be used to ensure that one TOSC1 cycle has elapsed:

1. Write a value to TCCR2x, TCNT2, or OCR2x.

2. Wait until the corresponding Update Busy Flag in ASSR returns to zero.

3. Enter Power-save or ADC Noise Reduction mode.

191

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• When the asynchronous operation is selected, the 32.768 kHz Oscillator for Timer/Counter2 is always running, except in Power-down and Standby modes. After a Power-up Reset or wake-up from Power-down or Standby mode, the user should be aware of the fact that this Oscillator might take as long as one second to stabilize. The user is advised to wait for at least one second before using Timer/Counter2 after power-up or wake-up from Power-down or Standby mode. The contents of all Timer/Counter2 Registers must be considered lost after a wake-up from Power-down or Standby mode due to unstable clock signal upon start-up, no matter whether the Oscillator is in use or a clock signal is applied to the TOSC1 pin.

• Description of wake up from Power-save or ADC Noise Reduction mode when the timer is clocked asynchronously: When the interrupt condition is met, the wake up process is started on the following cycle of the timer clock, that is, the timer is always advanced by at least one before the processor can read the counter value. After wake-up, the MCU is halted for four cycles, it executes the interrupt routine, and resumes execution from the instruction following SLEEP.

• Reading of the TCNT2 Register shortly after wake-up from Power-save may give an incorrect result. Since TCNT2 is clocked on the asynchronous TOSC clock, reading TCNT2 must be done through a register synchronized to the internal I/O clock domain. Synchronization takes place for every rising TOSC1 edge. When waking up from Power-save mode, and the I/O clock (clkI/O) again becomes active, TCNT2 will read as the previous value (before entering sleep) until the next rising TOSC1 edge. The phase of the TOSC clock after waking up from Power-save mode is essentially unpredictable, as it depends on the wake-up time. The recommended procedure for reading TCNT2 is thus as follows:

1. Write any value to either of the registers OCR2x or TCCR2x.

2. Wait for the corresponding Update Busy Flag to be cleared.

3. Read TCNT2.

• During asynchronous operation, the synchronization of the Interrupt Flags for the asynchronous timer takes 3 processor cycles plus one timer cycle. The timer is therefore advanced by at least one before the processor can read the timer value causing the setting of the Interrupt Flag. The Output Compare pin is changed on the timer clock and is not synchronized to the processor clock.

Timer/Counter2 Interrupt Mask Register – TIMSK2

• Bit 2 – OCIE2B: Timer/Counter2 Output Compare Match B Interrupt Enable

When the OCIE2B bit is written to one and the I-bit in the Status Register is set (one),the Timer/Counter2 Compare Match B interrupt is enabled. The corresponding interruptis executed if a compare match in Timer/Counter2 occurs, i.e., when the OCF2B bit isset in the Timer/Counter 2 Interrupt Flag Register – TIFR2.

• Bit 1 – OCIE2A: Timer/Counter2 Output Compare Match A Interrupt Enable

When the OCIE2A bit is written to one and the I-bit in the Status Register is set (one),the Timer/Counter2 Compare Match A interrupt is enabled. The corresponding interruptis executed if a compare match in Timer/Counter2 occurs, i.e., when the OCF2A bit isset in the Timer/Counter 2 Interrupt Flag Register – TIFR2.

Bit 7 6 5 4 3 2 1 0

– – – – – OCIE2B OCIE2A TOIE2 TIMSK2

Read/Write R R R R R R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

192 ATmega640/1280/1281/2560/25612549A–AVR–03/05

• Bit 0 – TOIE2: Timer/Counter2 Overflow Interrupt Enable

When the TOIE2 bit is written to one and the I-bit in the Status Register is set (one), theTimer/Counter2 Overflow interrupt is enabled. The corresponding interrupt is executed ifan overflow in Timer/Counter2 occurs, i.e., when the TOV2 bit is set in theTimer/Counter2 Interrupt Flag Register – TIFR2.

Timer/Counter2 Interrupt Flag Register – TIFR2

• Bit 2 – OCF2B: Output Compare Flag 2 B

The OCF2B bit is set (one) when a compare match occurs between the Timer/Counter2and the data in OCR2B – Output Compare Register2. OCF2B is cleared by hardwarewhen executing the corresponding interrupt handling vector. Alternatively, OCF2B iscleared by writing a logic one to the flag. When the I-bit in SREG, OCIE2B(Timer/Counter2 Compare match Interrupt Enable), and OCF2B are set (one), theTimer/Counter2 Compare match Interrupt is executed.

• Bit 1 – OCF2A: Output Compare Flag 2 A

The OCF2A bit is set (one) when a compare match occurs between the Timer/Counter2and the data in OCR2A – Output Compare Register2. OCF2A is cleared by hardwarewhen executing the corresponding interrupt handling vector. Alternatively, OCF2A iscleared by writing a logic one to the flag. When the I-bit in SREG, OCIE2A(Timer/Counter2 Compare match Interrupt Enable), and OCF2A are set (one), theTimer/Counter2 Compare match Interrupt is executed.

• Bit 0 – TOV2: Timer/Counter2 Overflow Flag

The TOV2 bit is set (one) when an overflow occurs in Timer/Counter2. TOV2 is clearedby hardware when executing the corresponding interrupt handling vector. Alternatively,TOV2 is cleared by writing a logic one to the flag. When the SREG I-bit, TOIE2A(Timer/Counter2 Overf low Interrupt Enable), and TOV2 are set (one), theTimer/Counter2 Overflow interrupt is executed. In PWM mode, this bit is set whenTimer/Counter2 changes counting direction at 0x00.

Bit 7 6 5 4 3 2 1 0

– – – – – OCF2B OCF2A TOV2 TIFR2

Read/Write R R R R R R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

193

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Timer/Counter Prescaler Figure 78. Prescaler for Timer/Counter2

The clock source for Timer/Counter2 is named clkT2S. clkT2S is by default connected tothe main system I/O clock clkIO. By setting the AS2 bit in ASSR, Timer/Counter2 is asyn-chronously clocked from the TOSC1 pin. This enables use of Timer/Counter2 as a RealTime Counter (RTC). When AS2 is set, pins TOSC1 and TOSC2 are disconnected fromPort C. A crystal can then be connected between the TOSC1 and TOSC2 pins to serveas an independent clock source for Timer/Counter2. The Oscillator is optimized for usewith a 32.768 kHz crystal. Applying an external clock source to TOSC1 is notrecommended.

For Timer/Counter2, the possible prescaled selections are: clkT2S/8, clkT2S/32, clkT2S/64,clkT2S/128, clkT2S/256, and clkT2S/1024. Additionally, clkT2S as well as 0 (stop) may beselected. Setting the PSRASY bit in GTCCR resets the prescaler. This allows the userto operate with a predictable prescaler.

General Timer/Counter Control Register – GTCCR

• Bit 1 – PSRASY: Prescaler Reset Timer/Counter2

When this bit is one, the Timer/Counter2 prescaler will be reset. This bit is normallycleared immediately by hardware. If the bit is written when Timer/Counter2 is operatingin asynchronous mode, the bit will remain one until the prescaler has been reset. The bitwill not be cleared by hardware if the TSM bit is set. Refer to the description of the “Bit 7– TSM: Timer/Counter Synchronization Mode” on page 170 for a description of theTimer/Counter Synchronization mode.

10-BIT T/C PRESCALER

TIMER/COUNTER2 CLOCK SOURCE

clkI/O clkT2S

TOSC1

AS2

CS20CS21CS22

clk T

2S/8

clk T

2S/6

4

clk T

2S/1

28

clk T

2S/1

024

clk T

2S/2

56

clk T

2S/3

2

0PSRASY

Clear

clkT2

Bit 7 6 5 4 3 2 1 0

TSM – – – – – PSRASY PSRSYNC GTCCR

Read/Write R/W R R R R R R/W R/W

Initial Value 0 0 0 0 0 0 0 0

194 ATmega640/1280/1281/2560/25612549A–AVR–03/05

195

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Serial Peripheral Interface – SPI

The Serial Peripheral Interface (SPI) allows high-speed synchronous data transferbetween the ATmega640/1280/1281/2560/2561 and peripheral devices or between sev-eral AVR devices. The ATmega640/1280/1281/2560/2561 SPI includes the followingfeatures:• Full-duplex, Three-wire Synchronous Data Transfer• Master or Slave Operation• LSB First or MSB First Data Transfer• Seven Programmable Bit Rates• End of Transmission Interrupt Flag• Write Collision Flag Protection• Wake-up from Idle Mode• Double Speed (CK/2) Master SPI Mode

USART can also be used in Master SPI mode, see “USART in SPI Mode” on page 231.

The Power Reduction SPI bit, PRSPI, in “Power Reduction Register 0 - PRR0” on page54 on page 50 must be written to zero to enable SPI module.

Figure 79. SPI Block Diagram(1)

Note: 1. Refer to Figure 1 on page 2, and Table 39 on page 89 for SPI pin placement.

The interconnection between Master and Slave CPUs with SPI is shown in Figure 80.The system consists of two shift Registers, and a Master clock generator. The SPI Mas-ter initiates the communication cycle when pulling low the Slave Select SS pin of thedesired Slave. Master and Slave prepare the data to be sent in their respective shiftRegisters, and the Master generates the required clock pulses on the SCK line to inter-

SP

I2X

SP

I2X

DIVIDER/2/4/8/16/32/64/128

196 ATmega640/1280/1281/2560/25612549A–AVR–03/05

change data. Data is always shifted from Master to Slave on the Master Out – Slave In,MOSI, line, and from Slave to Master on the Master In – Slave Out, MISO, line. Aftereach data packet, the Master will synchronize the Slave by pulling high the Slave Select,SS, line.

When configured as a Master, the SPI interface has no automatic control of the SS line.This must be handled by user software before communication can start. When this isdone, writing a byte to the SPI Data Register starts the SPI clock generator, and thehardware shifts the eight bits into the Slave. After shifting one byte, the SPI clock gener-ator stops, setting the end of Transmission Flag (SPIF). If the SPI Interrupt Enable bit(SPIE) in the SPCR Register is set, an interrupt is requested. The Master may continueto shift the next byte by writing it into SPDR, or signal the end of packet by pulling highthe Slave Select, SS line. The last incoming byte will be kept in the Buffer Register forlater use.

When configured as a Slave, the SPI interface will remain sleeping with MISO tri-statedas long as the SS pin is driven high. In this state, software may update the contents ofthe SPI Data Register, SPDR, but the data will not be shifted out by incoming clockpulses on the SCK pin until the SS pin is driven low. As one byte has been completelyshifted, the end of Transmission Flag, SPIF is set. If the SPI Interrupt Enable bit, SPIE,in the SPCR Register is set, an interrupt is requested. The Slave may continue to placenew data to be sent into SPDR before reading the incoming data. The last incoming bytewill be kept in the Buffer Register for later use.

Figure 80. SPI Master-slave Interconnection

The system is single buffered in the transmit direction and double buffered in the receivedirection. This means that bytes to be transmitted cannot be written to the SPI DataRegister before the entire shift cycle is completed. When receiving data, however, areceived character must be read from the SPI Data Register before the next characterhas been completely shifted in. Otherwise, the first byte is lost.

In SPI Slave mode, the control logic will sample the incoming signal of the SCK pin. Toensure correct sampling of the clock signal, the frequency of the SPI clock should neverexceed fosc/4.

SHIFTENABLE

197

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

When the SPI is enabled, the data direction of the MOSI, MISO, SCK, and SS pins isoverridden according to Table 93. For more details on automatic port overrides, refer to“Alternate Port Functions” on page 86.

Note: 1. See “Alternate Functions of Port B” on page 89 for a detailed description of how todefine the direction of the user defined SPI pins.

The following code examples show how to initialize the SPI as a Master and how to per-form a simple transmission. DDR_SPI in the examples must be replaced by the actualData Direction Register controlling the SPI pins. DD_MOSI, DD_MISO and DD_SCKmust be replaced by the actual data direction bits for these pins. E.g. if MOSI is placedon pin PB5, replace DD_MOSI with DDB5 and DDR_SPI with DDRB.

Table 93. SPI Pin Overrides(1)

Pin Direction, Master SPI Direction, Slave SPI

MOSI User Defined Input

MISO Input User Defined

SCK User Defined Input

SS User Defined Input

198 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. See “About Code Examples” on page 8.

Assembly Code Example(1)

SPI_MasterInit:

; Set MOSI and SCK output, all others input

ldi r17,(1<<DD_MOSI)|(1<<DD_SCK)

out DDR_SPI,r17

; Enable SPI, Master, set clock rate fck/16

ldi r17,(1<<SPE)|(1<<MSTR)|(1<<SPR0)

out SPCR,r17

ret

SPI_MasterTransmit:

; Start transmission of data (r16)

out SPDR,r16

Wait_Transmit:

; Wait for transmission complete

sbis SPSR,SPIF

rjmp Wait_Transmit

ret

C Code Example(1)

void SPI_MasterInit(void)

/* Set MOSI and SCK output, all others input */

DDR_SPI = (1<<DD_MOSI)|(1<<DD_SCK);

/* Enable SPI, Master, set clock rate fck/16 */

SPCR = (1<<SPE)|(1<<MSTR)|(1<<SPR0);

void SPI_MasterTransmit(char cData)

/* Start transmission */

SPDR = cData;

/* Wait for transmission complete */

while(!(SPSR & (1<<SPIF)))

;

199

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The following code examples show how to initialize the SPI as a Slave and how to per-form a simple reception.

Note: 1. See “About Code Examples” on page 8.

Assembly Code Example(1)

SPI_SlaveInit:

; Set MISO output, all others input

ldi r17,(1<<DD_MISO)

out DDR_SPI,r17

; Enable SPI

ldi r17,(1<<SPE)

out SPCR,r17

ret

SPI_SlaveReceive:

; Wait for reception complete

sbis SPSR,SPIF

rjmp SPI_SlaveReceive

; Read received data and return

in r16,SPDR

ret

C Code Example(1)

void SPI_SlaveInit(void)

/* Set MISO output, all others input */

DDR_SPI = (1<<DD_MISO);

/* Enable SPI */

SPCR = (1<<SPE);

char SPI_SlaveReceive(void)

/* Wait for reception complete */

while(!(SPSR & (1<<SPIF)))

;

/* Return Data Register */

return SPDR;

200 ATmega640/1280/1281/2560/25612549A–AVR–03/05

SS Pin Functionality

Slave Mode When the SPI is configured as a Slave, the Slave Select (SS) pin is always input. WhenSS is held low, the SPI is activated, and MISO becomes an output if configured so bythe user. All other pins are inputs. When SS is driven high, all pins are inputs, and theSPI is passive, which means that it will not receive incoming data. Note that the SPIlogic will be reset once the SS pin is driven high.

The SS pin is useful for packet/byte synchronization to keep the slave bit counter syn-chronous with the master clock generator. When the SS pin is driven high, the SPI slavewill immediately reset the send and receive logic, and drop any partially received data inthe Shift Register.

Master Mode When the SPI is configured as a Master (MSTR in SPCR is set), the user can determinethe direction of the SS pin.

If SS is configured as an output, the pin is a general output pin which does not affect theSPI system. Typically, the pin will be driving the SS pin of the SPI Slave.

If SS is configured as an input, it must be held high to ensure Master SPI operation. Ifthe SS pin is driven low by peripheral circuitry when the SPI is configured as a Masterwith the SS pin defined as an input, the SPI system interprets this as another masterselecting the SPI as a slave and starting to send data to it. To avoid bus contention, theSPI system takes the following actions:

1. The MSTR bit in SPCR is cleared and the SPI system becomes a Slave. As aresult of the SPI becoming a Slave, the MOSI and SCK pins become inputs.

2. The SPIF Flag in SPSR is set, and if the SPI interrupt is enabled, and the I-bit inSREG is set, the interrupt routine will be executed.

Thus, when interrupt-driven SPI transmission is used in Master mode, and there exists apossibility that SS is driven low, the interrupt should always check that the MSTR bit isstill set. If the MSTR bit has been cleared by a slave select, it must be set by the user tore-enable SPI Master mode.

SPI Control Register – SPCR

• Bit 7 – SPIE: SPI Interrupt Enable

This bit causes the SPI interrupt to be executed if SPIF bit in the SPSR Register is setand the if the Global Interrupt Enable bit in SREG is set.

• Bit 6 – SPE: SPI Enable

When the SPE bit is written to one, the SPI is enabled. This bit must be set to enableany SPI operations.

• Bit 5 – DORD: Data Order

When the DORD bit is written to one, the LSB of the data word is transmitted first.

When the DORD bit is written to zero, the MSB of the data word is transmitted first.

Bit 7 6 5 4 3 2 1 0

SPIE SPE DORD MSTR CPOL CPHA SPR1 SPR0 SPCR

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

201

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• Bit 4 – MSTR: Master/Slave Select

This bit selects Master SPI mode when written to one, and Slave SPI mode when writtenlogic zero. If SS is configured as an input and is driven low while MSTR is set, MSTR willbe cleared, and SPIF in SPSR will become set. The user will then have to set MSTR tore-enable SPI Master mode.

• Bit 3 – CPOL: Clock Polarity

When this bit is written to one, SCK is high when idle. When CPOL is written to zero,SCK is low when idle. Refer to Figure 81 and Figure 82 for an example. The CPOL func-tionality is summarized below:

• Bit 2 – CPHA: Clock Phase

The settings of the Clock Phase bit (CPHA) determine if data is sampled on the leading(first) or trailing (last) edge of SCK. Refer to Figure 81 and Figure 82 for an example.The CPOL functionality is summarized below:

• Bits 1, 0 – SPR1, SPR0: SPI Clock Rate Select 1 and 0

These two bits control the SCK rate of the device configured as a Master. SPR1 andSPR0 have no effect on the Slave. The relationship between SCK and the OscillatorClock frequency fosc is shown in the following table:

Table 94. CPOL Functionality

CPOL Leading Edge Trailing Edge

0 Rising Falling

1 Falling Rising

Table 95. CPHA Functionality

CPHA Leading Edge Trailing Edge

0 Sample Setup

1 Setup Sample

Table 96. Relationship Between SCK and the Oscillator Frequency

SPI2X SPR1 SPR0 SCK Frequency

0 0 0 fosc/4

0 0 1 fosc/16

0 1 0 fosc/64

0 1 1 fosc/128

1 0 0 fosc/2

1 0 1 fosc/8

1 1 0 fosc/32

1 1 1 fosc/64

202 ATmega640/1280/1281/2560/25612549A–AVR–03/05

SPI Status Register – SPSR

• Bit 7 – SPIF: SPI Interrupt Flag

When a serial transfer is complete, the SPIF Flag is set. An interrupt is generated ifSPIE in SPCR is set and global interrupts are enabled. If SS is an input and is driven lowwhen the SPI is in Master mode, this will also set the SPIF Flag. SPIF is cleared byhardware when executing the corresponding interrupt handling vector. Alternatively, theSPIF bit is cleared by first reading the SPI Status Register with SPIF set, then accessingthe SPI Data Register (SPDR).

• Bit 6 – WCOL: Write COLlision Flag

The WCOL bit is set if the SPI Data Register (SPDR) is written during a data transfer.The WCOL bit (and the SPIF bit) are cleared by first reading the SPI Status Registerwith WCOL set, and then accessing the SPI Data Register.

• Bit 5..1 – Res: Reserved Bits

These bits are reserved bits and will always read as zero.

• Bit 0 – SPI2X: Double SPI Speed Bit

When this bit is written logic one the SPI speed (SCK Frequency) will be doubled whenthe SPI is in Master mode (see Table 96). This means that the minimum SCK period willbe two CPU clock periods. When the SPI is configured as Slave, the SPI is only guaran-teed to work at fosc/4 or lower.

The SPI interface on the ATmega640/1280/1281/2560/2561 is also used for programmemory and EEPROM downloading or uploading. See “Serial Downloading” on page349 for serial programming and verification.

SPI Data Register – SPDR

The SPI Data Register is a read/write register used for data transfer between the Regis-ter File and the SPI Shift Register. Writing to the register initiates data transmission.Reading the register causes the Shift Register Receive buffer to be read.

Data Modes There are four combinations of SCK phase and polarity with respect to serial data,which are determined by control bits CPHA and CPOL. The SPI data transfer formatsare shown in Figure 81 and Figure 82. Data bits are shifted out and latched in on oppo-site edges of the SCK signal, ensuring sufficient time for data signals to stabilize. This isclearly seen by summarizing Table 94 and Table 95, as done below:

Bit 7 6 5 4 3 2 1 0

SPIF WCOL – – – – – SPI2X SPSR

Read/Write R R R R R R R R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

MSB LSB SPDR

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value X X X X X X X X Undefined

203

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 81. SPI Transfer Format with CPHA = 0

Figure 82. SPI Transfer Format with CPHA = 1

Table 97. CPOL Functionality

Leading Edge Trailing eDge SPI Mode

CPOL=0, CPHA=0 Sample (Rising) Setup (Falling) 0

CPOL=0, CPHA=1 Setup (Rising) Sample (Falling) 1

CPOL=1, CPHA=0 Sample (Falling) Setup (Rising) 2

CPOL=1, CPHA=1 Setup (Falling) Sample (Rising) 3

Bit 1Bit 6

LSBMSB

SCK (CPOL = 0)mode 0

SAMPLE IMOSI/MISO

CHANGE 0MOSI PIN

CHANGE 0MISO PIN

SCK (CPOL = 1)mode 2

SS

MSBLSB

Bit 6Bit 1

Bit 5Bit 2

Bit 4Bit 3

Bit 3Bit 4

Bit 2Bit 5

MSB first (DORD = 0)LSB first (DORD = 1)

SCK (CPOL = 0)mode 1

SAMPLE IMOSI/MISO

CHANGE 0MOSI PIN

CHANGE 0MISO PIN

SCK (CPOL = 1)mode 3

SS

MSBLSB

Bit 6Bit 1

Bit 5Bit 2

Bit 4Bit 3

Bit 3Bit 4

Bit 2Bit 5

Bit 1Bit 6

LSBMSB

MSB first (DORD = 0)LSB first (DORD = 1)

204 ATmega640/1280/1281/2560/25612549A–AVR–03/05

205

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

USART The Universal Synchronous and Asynchronous serial Receiver and Transmitter(USART) is a highly flexible serial communication device. The main features are:• Full Duplex Operation (Independent Serial Receive and Transmit Registers)• Asynchronous or Synchronous Operation• Master or Slave Clocked Synchronous Operation• High Resolution Baud Rate Generator• Supports Serial Frames with 5, 6, 7, 8, or 9 Data Bits and 1 or 2 Stop Bits• Odd or Even Parity Generation and Parity Check Supported by Hardware• Data OverRun Detection• Framing Error Detection• Noise Filtering Includes False Start Bit Detection and Digital Low Pass Filter• Three Separate Interrupts on TX Complete, TX Data Register Empty and RX Complete• Multi-processor Communication Mode• Double Speed Asynchronous Communication Mode

Quad USART The ATmega640/1280/1281/2560/2561 has four USART’s, USART0, USART1,USART2, and USART3. The functionality for all four USART’s is described below.USART0, USART1, USART2, and USART3 have different I/O registers as shown in“Register Summary” on page 385.

Overview A simplified block diagram of the USART Transmitter is shown in Figure 83 on page206. CPU accessible I/O Registers and I/O pins are shown in bold.

The Power Reducion USART0 bit, PRUSART0, in “Power Reduction Register 0 -PRR0” on page 54 must be disabled by writing a logical zero to it.

The Power Reducion USART1 bit, PRUSART1, in “Power Reduction Register 1 -PRR1” on page 55 must be disabled by writing a logical zero to it.

The Power Reducion USART2 bit, PRUSART2, in “Power Reduction Register 1 -PRR1” on page 55 must be disabled by writing a logical zero to it.

The Power Reducion USART3 bit, PRUSART3, in “Power Reduction Register 1 -PRR1” on page 55 must be disabled by writing a logical zero to it.

206 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 83. USART Block Diagram(1)

Note: 1. See Figure 1 on page 2, Figure 2 on page 3, Table 45 on page 94, Table 48 on page96, Table 57 on page 104 and Table 60 on page 106 for USART pin placement.

The dashed boxes in the block diagram separate the three main parts of the USART(listed from the top): Clock Generator, Transmitter and Receiver. Control Registers areshared by all units. The Clock Generation logic consists of synchronization logic forexternal clock input used by synchronous slave operation, and the baud rate generator.The XCKn (Transfer Clock) pin is only used by synchronous transfer mode. The Trans-mitter consists of a single write buffer, a serial Shift Register, Parity Generator andControl logic for handling different serial frame formats. The write buffer allows a contin-uous transfer of data without any delay between frames. The Receiver is the mostcomplex part of the USART module due to its clock and data recovery units. The recov-ery units are used for asynchronous data reception. In addition to the recovery units, theReceiver includes a Parity Checker, Control logic, a Shift Register and a two levelreceive buffer (UDRn). The Receiver supports the same frame formats as the Transmit-ter, and can detect Frame Error, Data OverRun and Parity Errors.

PARITYGENERATOR

UBRR[H:L]

UDR (Transmit)

UCSRA UCSRB UCSRC

BAUD RATE GENERATOR

TRANSMIT SHIFT REGISTER

RECEIVE SHIFT REGISTER RxD

TxDPINCONTROL

UDR (Receive)

PINCONTROL

XCK

DATARECOVERY

CLOCKRECOVERY

PINCONTROL

TXCONTROL

RXCONTROL

PARITYCHECKER

DA

TA B

US

OSC

SYNC LOGIC

Clock Generator

Transmitter

Receiver

207

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Clock Generation The Clock Generation logic generates the base clock for the Transmitter and Receiver.The USARTn supports four modes of clock operation: Normal asynchronous, DoubleSpeed asynchronous, Master synchronous and Slave synchronous mode. The UMSELnbit in USART Control and Status Register C (UCSRnC) selects between asynchronousand synchronous operation. Double Speed (asynchronous mode only) is controlled bythe U2Xn found in the UCSRnA Register. When using synchronous mode (UMSELn =1), the Data Direction Register for the XCKn pin (DDR_XCKn) controls whether theclock source is internal (Master mode) or external (Slave mode). The XCKn pin is onlyactive when using synchronous mode.

Figure 84 shows a block diagram of the clock generation logic.

Figure 84. Clock Generation Logic, Block Diagram

Signal description:

txclk Transmitter clock (Internal Signal).

rxclk Receiver base clock (Internal Signal).

xcki Input from XCK pin (internal Signal). Used for synchronous slave operation.

xcko Clock output to XCK pin (Internal Signal). Used for synchronous masteroperation.

fOSC XTAL pin frequency (System Clock).

PrescalingDown-Counter /2

UBRR

/4 /2

fosc

UBRR+1

SyncRegister

OSC

XCKPin

txclk

U2X

UMSEL

DDR_XCK

0

1

0

1

xcki

xcko

DDR_XCKrxclk

0

1

1

0Edge

Detector

UCPOL

208 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Internal Clock Generation – The Baud Rate Generator

Internal clock generation is used for the asynchronous and the synchronous mastermodes of operation. The description in this section refers to Figure 84.

The USART Baud Rate Register (UBRRn) and the down-counter connected to it func-tion as a programmable prescaler or baud rate generator. The down-counter, running atsystem clock (fosc), is loaded with the UBRRn value each time the counter has counteddown to zero or when the UBRRLn Register is written. A clock is generated each timethe counter reaches zero. This clock is the baud rate generator clock output (=fosc/(UBRRn+1)). The Transmitter divides the baud rate generator clock output by 2, 8 or16 depending on mode. The baud rate generator output is used directly by theReceiver’s clock and data recovery units. However, the recovery units use a statemachine that uses 2, 8 or 16 states depending on mode set by the state of the UMSELn,U2Xn and DDR_XCKn bits.

Table 98 contains equations for calculating the baud rate (in bits per second) and forcalculating the UBRRn value for each mode of operation using an internally generatedclock source.

Note: 1. The baud rate is defined to be the transfer rate in bit per second (bps)

BAUD Baud rate (in bits per second, bps)

fOSC System Oscillator clock frequency

UBRRnContents of the UBRRHn and UBRRLn Registers, (0-4095)

Table 98. Equations for Calculating Baud Rate Register Setting

Operating ModeEquation for Calculating

Baud Rate(1)Equation for Calculating

UBRR Value

Asynchronous Normal mode (U2Xn = 0)

Asynchronous Double Speed mode (U2Xn = 1)

Synchronous Master mode

BAUDfOSC

16 UBRRn 1+( )------------------------------------------=

UBRRnfOSC

16BAUD------------------------ 1–=

BAUDfOSC

8 UBRRn 1+( )---------------------------------------=

UBRRnfOSC

8BAUD-------------------- 1–=

BAUDfOSC

2 UBRRn 1+( )---------------------------------------=

UBRRnfOSC

2BAUD-------------------- 1–=

209

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Some examples of UBRRn values for some system clock frequencies are found in Table106 on page 227.

Double Speed Operation (U2Xn)

The transfer rate can be doubled by setting the U2Xn bit in UCSRnA. Setting this bitonly has effect for the asynchronous operation. Set this bit to zero when using synchro-nous operation.

Setting this bit will reduce the divisor of the baud rate divider from 16 to 8, effectivelydoubling the transfer rate for asynchronous communication. Note however that theReceiver will in this case only use half the number of samples (reduced from 16 to 8) fordata sampling and clock recovery, and therefore a more accurate baud rate setting andsystem clock are required when this mode is used. For the Transmitter, there are nodownsides.

External Clock External clocking is used by the synchronous slave modes of operation. The descriptionin this section refers to Figure 84 for details.

External clock input from the XCKn pin is sampled by a synchronization register to mini-mize the chance of meta-stability. The output from the synchronization register mustthen pass through an edge detector before it can be used by the Transmitter andReceiver. This process introduces a two CPU clock period delay and therefore the max-imum external XCKn clock frequency is limited by the following equation:

Note that fosc depends on the stability of the system clock source. It is therefore recom-mended to add some margin to avoid possible loss of data due to frequency variations.

Synchronous Clock Operation When synchronous mode is used (UMSELn = 1), the XCKn pin will be used as eitherclock input (Slave) or clock output (Master). The dependency between the clock edgesand data sampling or data change is the same. The basic principle is that data input (onRxDn) is sampled at the opposite XCKn clock edge of the edge the data output (TxDn)is changed.

Figure 85. Synchronous Mode XCKn Timing.

The UCPOLn bit UCRSC selects which XCKn clock edge is used for data sampling andwhich is used for data change. As Figure 85 shows, when UCPOLn is zero the data willbe changed at rising XCKn edge and sampled at falling XCKn edge. If UCPOLn is set,the data will be changed at falling XCKn edge and sampled at rising XCKn edge.

fXCKfOSC

4-----------<

RxD / TxD

XCK

RxD / TxD

XCKUCPOL = 0

UCPOL = 1

Sample

Sample

210 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Frame Formats A serial frame is defined to be one character of data bits with synchronization bits (startand stop bits), and optionally a parity bit for error checking. The USART accepts all 30combinations of the following as valid frame formats:

• 1 start bit

• 5, 6, 7, 8, or 9 data bits

• no, even or odd parity bit

• 1 or 2 stop bits

A frame starts with the start bit followed by the least significant data bit. Then the nextdata bits, up to a total of nine, are succeeding, ending with the most significant bit. Ifenabled, the parity bit is inserted after the data bits, before the stop bits. When a com-plete frame is transmitted, it can be directly followed by a new frame, or thecommunication line can be set to an idle (high) state. Figure 86 illustrates the possiblecombinations of the frame formats. Bits inside brackets are optional.

Figure 86. Frame Formats

St Start bit, always low.

(n) Data bits (0 to 8).

P Parity bit. Can be odd or even.

Sp Stop bit, always high.

IDLE No transfers on the communication line (RxDn or TxDn). An IDLE line must behigh.

The frame format used by the USART is set by the UCSZn2:0, UPMn1:0 and USBSnbits in UCSRnB and UCSRnC. The Receiver and Transmitter use the same setting.Note that changing the setting of any of these bits will corrupt all ongoing communica-tion for both the Receiver and Transmitter.

The USART Character SiZe (UCSZn2:0) bits select the number of data bits in the frame.The USART Parity mode (UPMn1:0) bits enable and set the type of parity bit. The selec-tion between one or two stop bits is done by the USART Stop Bit Select (USBSn) bit.The Receiver ignores the second stop bit. An FE (Frame Error) will therefore only bedetected in the cases where the first stop bit is zero.

Parity Bit Calculation The parity bit is calculated by doing an exclusive-or of all the data bits. If odd parity isused, the result of the exclusive or is inverted. The relation between the parity bit anddata bits is as follows::

Peven Parity bit using even parity

Podd Parity bit using odd parity

dn Data bit n of the character

10 2 3 4 [5] [6] [7] [8] [P]St Sp1 [Sp2] (St / IDLE)(IDLE)

FRAME

Peven dn 1– … d3 d2 d1 d0 0Podd

⊕ ⊕ ⊕ ⊕ ⊕ ⊕dn 1– … d3 d2 d1 d0 1⊕ ⊕ ⊕ ⊕ ⊕ ⊕

==

211

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

If used, the parity bit is located between the last data bit and first stop bit of a serialframe.

USART Initialization The USART has to be initialized before any communication can take place. The initial-ization process normally consists of setting the baud rate, setting frame format andenabling the Transmitter or the Receiver depending on the usage. For interrupt drivenUSART operation, the Global Interrupt Flag should be cleared (and interrupts globallydisabled) when doing the initialization.

Before doing a re-initialization with changed baud rate or frame format, be sure thatthere are no ongoing transmissions during the period the registers are changed. TheTXCn Flag can be used to check that the Transmitter has completed all transfers, andthe RXC Flag can be used to check that there are no unread data in the receive buffer.Note that the TXCn Flag must be cleared before each transmission (before UDRn iswritten) if it is used for this purpose.

The following simple USART initialization code examples show one assembly and oneC function that are equal in functionality. The examples assume asynchronous opera-tion using polling (no interrupts enabled) and a fixed frame format. The baud rate isgiven as a function parameter. For the assembly code, the baud rate parameter isassumed to be stored in the r17:r16 Registers.

Note: 1. See “About Code Examples” on page 8.

More advanced initialization routines can be made that include frame format as parame-ters, disable interrupts and so on. However, many applications use a fixed setting of thebaud and control registers, and for these types of applications the initialization code can

Assembly Code Example(1)

USART_Init:

; Set baud rate

out UBRRHn, r17

out UBRRLn, r16

; Enable receiver and transmitter

ldi r16, (1<<RXENn)|(1<<TXENn)

out UCSRnB,r16

; Set frame format: 8data, 2stop bit

ldi r16, (1<<USBSn)|(3<<UCSZn0)

out UCSRnC,r16

ret

C Code Example(1)

void USART_Init( unsigned int baud )

/* Set baud rate */

UBRRHn = (unsigned char)(baud>>8);

UBRRLn = (unsigned char)baud;

/* Enable receiver and transmitter */

UCSRnB = (1<<RXENn)|(1<<TXENn);

/* Set frame format: 8data, 2stop bit */

UCSRnC = (1<<USBSn)|(3<<UCSZn0);

212 ATmega640/1280/1281/2560/25612549A–AVR–03/05

be placed directly in the main routine, or be combined with initialization code for otherI/O modules.

Data Transmission – The USART Transmitter

The USART Transmitter is enabled by setting the Transmit Enable (TXEN) bit in theUCSRnB Register. When the Transmitter is enabled, the normal port operation of theTxDn pin is overridden by the USART and given the function as the Transmitter’s serialoutput. The baud rate, mode of operation and frame format must be set up once beforedoing any transmissions. If synchronous operation is used, the clock on the XCKn pinwill be overridden and used as transmission clock.

Sending Frames with 5 to 8 Data Bit

A data transmission is initiated by loading the transmit buffer with the data to be trans-mitted. The CPU can load the transmit buffer by writing to the UDRn I/O location. Thebuffered data in the transmit buffer will be moved to the Shift Register when the ShiftRegister is ready to send a new frame. The Shift Register is loaded with new data if it isin idle state (no ongoing transmission) or immediately after the last stop bit of the previ-ous frame is transmitted. When the Shift Register is loaded with new data, it will transferone complete frame at the rate given by the Baud Register, U2Xn bit or by XCKndepending on mode of operation.

The following code examples show a simple USART transmit function based on pollingof the Data Register Empty (UDREn) Flag. When using frames with less than eight bits,the most significant bits written to the UDRn are ignored. The USART has to be initial-ized before the function can be used. For the assembly code, the data to be sent isassumed to be stored in Register R16

Note: 1. See “About Code Examples” on page 8.

The function simply waits for the transmit buffer to be empty by checking the UDREnFlag, before loading it with new data to be transmitted. If the Data Register Empty inter-rupt is utilized, the interrupt routine writes the data into the buffer.

Sending Frames with 9 Data Bit

If 9-bit characters are used (UCSZn = 7), the ninth bit must be written to the TXB8 bit inUCSRnB before the low byte of the character is written to UDRn. The following code

Assembly Code Example(1)

USART_Transmit:

; Wait for empty transmit buffer

sbis UCSRnA,UDREn

rjmp USART_Transmit

; Put data (r16) into buffer, sends the data

out UDRn,r16

ret

C Code Example(1)

void USART_Transmit( unsigned char data )

/* Wait for empty transmit buffer */

while ( !( UCSRnA & (1<<UDREn)) )

;

/* Put data into buffer, sends the data */

UDRn = data;

213

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

examples show a transmit function that handles 9-bit characters. For the assemblycode, the data to be sent is assumed to be stored in registers R17:R16.

Notes: 1. These transmit functions are written to be general functions. They can be optimized ifthe contents of the UCSRnB is static. For example, only the TXB8 bit of the UCSRnBRegister is used after initialization.

2. See “About Code Examples” on page 8.

The ninth bit can be used for indicating an address frame when using multi processorcommunication mode or for other protocol handling as for example synchronization.

Assembly Code Example(1)(2)

USART_Transmit:

; Wait for empty transmit buffer

sbis UCSRnA,UDREn

rjmp USART_Transmit

; Copy 9th bit from r17 to TXB8

cbi UCSRnB,TXB8

sbrc r17,0

sbi UCSRnB,TXB8

; Put LSB data (r16) into buffer, sends the data

out UDRn,r16

ret

C Code Example(1)(2)

void USART_Transmit( unsigned int data )

/* Wait for empty transmit buffer */

while ( !( UCSRnA & (1<<UDREn))) )

;

/* Copy 9th bit to TXB8 */

UCSRnB &= ~(1<<TXB8);

if ( data & 0x0100 )

UCSRnB |= (1<<TXB8);

/* Put data into buffer, sends the data */

UDRn = data;

214 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Transmitter Flags and Interrupts

The USART Transmitter has two flags that indicate its state: USART Data RegisterEmpty (UDREn) and Transmit Complete (TXCn). Both flags can be used for generatinginterrupts.

The Data Register Empty (UDREn) Flag indicates whether the transmit buffer is ready toreceive new data. This bit is set when the transmit buffer is empty, and cleared when thetransmit buffer contains data to be transmitted that has not yet been moved into the ShiftRegister. For compatibility with future devices, always write this bit to zero when writingthe UCSRnA Register.

When the Data Register Empty Interrupt Enable (UDRIEn) bit in UCSRnB is written toone, the USART Data Register Empty Interrupt will be executed as long as UDREn isset (provided that global interrupts are enabled). UDREn is cleared by writing UDRn.When interrupt-driven data transmission is used, the Data Register Empty interrupt rou-tine must either write new data to UDRn in order to clear UDREn or disable the DataRegister Empty interrupt, otherwise a new interrupt will occur once the interrupt routineterminates.

The Transmit Complete (TXCn) Flag bit is set one when the entire frame in the TransmitShift Register has been shifted out and there are no new data currently present in thetransmit buffer. The TXCn Flag bit is automatically cleared when a transmit completeinterrupt is executed, or it can be cleared by writing a one to its bit location. The TXCnFlag is useful in half-duplex communication interfaces (like the RS-485 standard), wherea transmitting application must enter receive mode and free the communication busimmediately after completing the transmission.

When the Transmit Compete Interrupt Enable (TXCIEn) bit in UCSRnB is set, theUSART Transmit Complete Interrupt will be executed when the TXCn Flag becomes set(provided that global interrupts are enabled). When the transmit complete interrupt isused, the interrupt handling routine does not have to clear the TXCn Flag, this is doneautomatically when the interrupt is executed.

Parity Generator The Parity Generator calculates the parity bit for the serial frame data. When parity bit isenabled (UPMn1 = 1), the transmitter control logic inserts the parity bit between the lastdata bit and the first stop bit of the frame that is sent.

Disabling the Transmitter The disabling of the Transmitter (setting the TXEN to zero) will not become effectiveuntil ongoing and pending transmissions are completed, i.e., when the Transmit ShiftRegister and Transmit Buffer Register do not contain data to be transmitted. When dis-abled, the Transmitter will no longer override the TxDn pin.

Data Reception – The USART Receiver

The USART Receiver is enabled by writing the Receive Enable (RXENn) bit in theUCSRnB Register to one. When the Receiver is enabled, the normal pin operation ofthe RxDn pin is overridden by the USART and given the function as the Receiver’s serialinput. The baud rate, mode of operation and frame format must be set up once beforeany serial reception can be done. If synchronous operation is used, the clock on theXCKn pin will be used as transfer clock.

Receiving Frames with 5 to 8 Data Bits

The Receiver starts data reception when it detects a valid start bit. Each bit that followsthe start bit will be sampled at the baud rate or XCKn clock, and shifted into the ReceiveShift Register until the first stop bit of a frame is received. A second stop bit will beignored by the Receiver. When the first stop bit is received, i.e., a complete serial frameis present in the Receive Shift Register, the contents of the Shift Register will be movedinto the receive buffer. The receive buffer can then be read by reading the UDRn I/Olocation.

215

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The following code example shows a simple USART receive function based on pollingof the Receive Complete (RXCn) Flag. When using frames with less than eight bits themost significant bits of the data read from the UDRn will be masked to zero. The USARThas to be initialized before the function can be used.

Note: 1. See “About Code Examples” on page 8.

The function simply waits for data to be present in the receive buffer by checking theRXCn Flag, before reading the buffer and returning the value.

Receiving Frames with 9 Data Bits

If 9-bit characters are used (UCSZn=7) the ninth bit must be read from the RXB8n bit inUCSRnB before reading the low bits from the UDRn. This rule applies to the FEn,DORn and UPEn Status Flags as well. Read status from UCSRnA, then data fromUDRn. Reading the UDRn I/O location will change the state of the receive buffer FIFOand consequently the TXB8n, FEn, DORn and UPEn bits, which all are stored in theFIFO, will change.

The following code example shows a simple USART receive function that handles bothnine bit characters and the status bits.

Assembly Code Example(1)

USART_Receive:

; Wait for data to be received

sbis UCSRnA, RXCn

rjmp USART_Receive

; Get and return received data from buffer

in r16, UDRn

ret

C Code Example(1)

unsigned char USART_Receive( void )

/* Wait for data to be received */

while ( !(UCSRnA & (1<<RXCn)) )

;

/* Get and return received data from buffer */

return UDRn;

216 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. See “About Code Examples” on page 8.

The receive function example reads all the I/O Registers into the Register File beforeany computation is done. This gives an optimal receive buffer utilization since the bufferlocation read will be free to accept new data as early as possible.

Receive Compete Flag and Interrupt

The USART Receiver has one flag that indicates the Receiver state.

The Receive Complete (RXCn) Flag indicates if there are unread data present in thereceive buffer. This flag is one when unread data exist in the receive buffer, and zerowhen the receive buffer is empty (i.e., does not contain any unread data). If the Receiver

Assembly Code Example(1)

USART_Receive:

; Wait for data to be received

sbis UCSRnA, RXCn

rjmp USART_Receive

; Get status and 9th bit, then data from buffer

in r18, UCSRnA

in r17, UCSRnB

in r16, UDRn

; If error, return -1

andi r18,(1<<FEn)|(1<<DORn)|(1<<UPEn)

breq USART_ReceiveNoError

ldi r17, HIGH(-1)

ldi r16, LOW(-1)

USART_ReceiveNoError:

; Filter the 9th bit, then return

lsr r17

andi r17, 0x01

ret

C Code Example(1)

unsigned int USART_Receive( void )

unsigned char status, resh, resl;

/* Wait for data to be received */

while ( !(UCSRnA & (1<<RXCn)) )

;

/* Get status and 9th bit, then data */

/* from buffer */

status = UCSRnA;

resh = UCSRnB;

resl = UDRn;

/* If error, return -1 */

if ( status & (1<<FEn)|(1<<DORn)|(1<<UPEn) )

return -1;

/* Filter the 9th bit, then return */

resh = (resh >> 1) & 0x01;

return ((resh << 8) | resl);

217

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

is disabled (RXENn = 0), the receive buffer will be flushed and consequently the RXCnbit will become zero.

When the Receive Complete Interrupt Enable (RXCIEn) in UCSRnB is set, the USARTReceive Complete interrupt will be executed as long as the RXCn Flag is set (providedthat global interrupts are enabled). When interrupt-driven data reception is used, thereceive complete routine must read the received data from UDRn in order to clear theRXCn Flag, otherwise a new interrupt will occur once the interrupt routine terminates.

Receiver Error Flags The USART Receiver has three Error Flags: Frame Error (FEn), Data OverRun (DORn)and Parity Error (UPEn). All can be accessed by reading UCSRnA. Common for theError Flags is that they are located in the receive buffer together with the frame for whichthey indicate the error status. Due to the buffering of the Error Flags, the UCSRnA mustbe read before the receive buffer (UDRn), since reading the UDRn I/O location changesthe buffer read location. Another equality for the Error Flags is that they can not bealtered by software doing a write to the flag location. However, all flags must be set tozero when the UCSRnA is written for upward compatibility of future USART implementa-tions. None of the Error Flags can generate interrupts.

The Frame Error (FEn) Flag indicates the state of the first stop bit of the next readableframe stored in the receive buffer. The FEn Flag is zero when the stop bit was correctlyread (as one), and the FEn Flag will be one when the stop bit was incorrect (zero). Thisflag can be used for detecting out-of-sync conditions, detecting break conditions andprotocol handling. The FEn Flag is not affected by the setting of the USBSn bit in UCS-RnC since the Receiver ignores all, except for the first, stop bits. For compatibility withfuture devices, always set this bit to zero when writing to UCSRnA.

The Data OverRun (DORn) Flag indicates data loss due to a receiver buffer full condi-tion. A Data OverRun occurs when the receive buffer is full (two characters), it is a newcharacter waiting in the Receive Shift Register, and a new start bit is detected. If theDORn Flag is set there was one or more serial frame lost between the frame last readfrom UDRn, and the next frame read from UDRn. For compatibility with future devices,always write this bit to zero when writing to UCSRnA. The DORn Flag is cleared whenthe frame received was successfully moved from the Shift Register to the receive buffer.

The Parity Error (UPEn) Flag indicates that the next frame in the receive buffer had aParity Error when received. If Parity Check is not enabled the UPEn bit will always beread zero. For compatibility with future devices, always set this bit to zero when writingto UCSRnA. For more details see “Parity Bit Calculation” on page 210 and “ParityChecker” on page 217.

Parity Checker The Parity Checker is active when the high USART Parity mode (UPMn1) bit is set.Type of Parity Check to be performed (odd or even) is selected by the UPMn0 bit. Whenenabled, the Parity Checker calculates the parity of the data bits in incoming frames andcompares the result with the parity bit from the serial frame. The result of the check isstored in the receive buffer together with the received data and stop bits. The ParityError (UPEn) Flag can then be read by software to check if the frame had a Parity Error.

The UPEn bit is set if the next character that can be read from the receive buffer had aParity Error when received and the Parity Checking was enabled at that point (UPMn1 =1). This bit is valid until the receive buffer (UDRn) is read.

Disabling the Receiver In contrast to the Transmitter, disabling of the Receiver will be immediate. Data fromongoing receptions will therefore be lost. When disabled (i.e., the RXENn is set to zero)the Receiver will no longer override the normal function of the RxDn port pin. The

218 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Receiver buffer FIFO will be flushed when the Receiver is disabled. Remaining data inthe buffer will be lost

Flushing the Receive Buffer The receiver buffer FIFO will be flushed when the Receiver is disabled, i.e., the bufferwill be emptied of its contents. Unread data will be lost. If the buffer has to be flushedduring normal operation, due to for instance an error condition, read the UDRn I/O loca-tion until the RXCn Flag is cleared. The following code example shows how to flush thereceive buffer.

Note: 1. See “About Code Examples” on page 8.

Asynchronous Data Reception

The USART includes a clock recovery and a data recovery unit for handling asynchro-nous data reception. The clock recovery logic is used for synchronizing the internallygenerated baud rate clock to the incoming asynchronous serial frames at the RxDn pin.The data recovery logic samples and low pass filters each incoming bit, thereby improv-ing the noise immunity of the Receiver. The asynchronous reception operational rangedepends on the accuracy of the internal baud rate clock, the rate of the incomingframes, and the frame size in number of bits.

Asynchronous Clock Recovery

The clock recovery logic synchronizes internal clock to the incoming serial frames. Fig-ure 87 illustrates the sampling process of the start bit of an incoming frame. The samplerate is 16 times the baud rate for Normal mode, and eight times the baud rate for DoubleSpeed mode. The horizontal arrows illustrate the synchronization variation due to thesampling process. Note the larger time variation when using the Double Speed mode(U2Xn = 1) of operation. Samples denoted zero are samples done when the RxDn lineis idle (i.e., no communication activity).

Figure 87. Start Bit Sampling

When the clock recovery logic detects a high (idle) to low (start) transition on the RxDnline, the start bit detection sequence is initiated. Let sample 1 denote the first zero-sam-

Assembly Code Example(1)

USART_Flush:

sbis UCSRnA, RXCn

ret

in r16, UDRn

rjmp USART_Flush

C Code Example(1)

void USART_Flush( void )

unsigned char dummy;

while ( UCSRnA & (1<<RXCn) ) dummy = UDRn;

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1 2

STARTIDLE

00

BIT 0

3

1 2 3 4 5 6 7 8 1 20

RxD

Sample(U2X = 0)

Sample(U2X = 1)

219

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

ple as shown in the figure. The clock recovery logic then uses samples 8, 9, and 10 forNormal mode, and samples 4, 5, and 6 for Double Speed mode (indicated with samplenumbers inside boxes on the figure), to decide if a valid start bit is received. If two ormore of these three samples have logical high levels (the majority wins), the start bit isrejected as a noise spike and the Receiver starts looking for the next high to low-transi-tion. If however, a valid start bit is detected, the clock recovery logic is synchronized andthe data recovery can begin. The synchronization process is repeated for each start bit.

Asynchronous Data Recovery When the receiver clock is synchronized to the start bit, the data recovery can begin.The data recovery unit uses a state machine that has 16 states for each bit in Normalmode and eight states for each bit in Double Speed mode. Figure 88 shows the sam-pling of the data bits and the parity bit. Each of the samples is given a number that isequal to the state of the recovery unit.

Figure 88. Sampling of Data and Parity Bit

The decision of the logic level of the received bit is taken by doing a majority voting ofthe logic value to the three samples in the center of the received bit. The center samplesare emphasized on the figure by having the sample number inside boxes. The majorityvoting process is done as follows: If two or all three samples have high levels, thereceived bit is registered to be a logic 1. If two or all three samples have low levels, thereceived bit is registered to be a logic 0. This majority voting process acts as a low passfilter for the incoming signal on the RxDn pin. The recovery process is then repeateduntil a complete frame is received. Including the first stop bit. Note that the Receiver onlyuses the first stop bit of a frame.

Figure 89 shows the sampling of the stop bit and the earliest possible beginning of thestart bit of the next frame.

Figure 89. Stop Bit Sampling and Next Start Bit Sampling

The same majority voting is done to the stop bit as done for the other bits in the frame. Ifthe stop bit is registered to have a logic 0 value, the Frame Error (FEn) Flag will be set.

A new high to low transition indicating the start bit of a new frame can come right afterthe last of the bits used for majority voting. For Normal Speed mode, the first low levelsample can be at point marked (A) in Figure 89. For Double Speed mode the first lowlevel must be delayed to (B). (C) marks a stop bit of full length. The early start bit detec-tion influences the operational range of the Receiver.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 1

BIT n

1 2 3 4 5 6 7 8 1

RxD

Sample(U2X = 0)

Sample(U2X = 1)

1 2 3 4 5 6 7 8 9 10 0/1 0/1 0/1

STOP 1

1 2 3 4 5 6 0/1

RxD

Sample(U2X = 0)

Sample(U2X = 1)

(A) (B) (C)

220 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Asynchronous Operational Range

The operational range of the Receiver is dependent on the mismatch between thereceived bit rate and the internally generated baud rate. If the Transmitter is sendingframes at too fast or too slow bit rates, or the internally generated baud rate of theReceiver does not have a similar (see Table 99) base frequency, the Receiver will notbe able to synchronize the frames to the start bit.

The following equations can be used to calculate the ratio of the incoming data rate andinternal receiver baud rate.

D Sum of character size and parity size (D = 5 to 10 bit)

S Samples per bit. S = 16 for Normal Speed mode and S = 8 for Double Speedmode.

SF First sample number used for majority voting. SF = 8 for normal speed and SF = 4for Double Speed mode.

SM Middle sample number used for majority voting. SM = 9 for normal speed andSM = 5 for Double Speed mode.

Rslow is the ratio of the slowest incoming data rate that can be accepted in relation to thereceiver baud rate. Rfast is the ratio of the fastest incoming data rate that can beaccepted in relation to the receiver baud rate.

Table 99 and Table 100 list the maximum receiver baud rate error that can be tolerated.Note that Normal Speed mode has higher toleration of baud rate variations.

RslowD 1+( )S

S 1– D S⋅ SF+ +-------------------------------------------= Rfast

D 2+( )SD 1+( )S SM+

-----------------------------------=

221

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The recommendations of the maximum receiver baud rate error was made under theassumption that the Receiver and Transmitter equally divides the maximum total error.

There are two possible sources for the receivers baud rate error. The Receiver’s systemclock (XTAL) will always have some minor instability over the supply voltage range andthe temperature range. When using a crystal to generate the system clock, this is rarelya problem, but for a resonator the system clock may differ more than 2% depending ofthe resonators tolerance. The second source for the error is more controllable. The baudrate generator can not always do an exact division of the system frequency to get thebaud rate wanted. In this case an UBRR value that gives an acceptable low error can beused if possible.

Multi-processor Communication Mode

Setting the Multi-processor Communication mode (MPCMn) bit in UCSRnA enables afiltering function of incoming frames received by the USART Receiver. Frames that donot contain address information will be ignored and not put into the receive buffer. Thiseffectively reduces the number of incoming frames that has to be handled by the CPU,in a system with multiple MCUs that communicate via the same serial bus. The Trans-mitter is unaffected by the MPCMn setting, but has to be used differently when it is apart of a system utilizing the Multi-processor Communication mode.

If the Receiver is set up to receive frames that contain 5 to 8 data bits, then the first stopbit indicates if the frame contains data or address information. If the Receiver is set upfor frames with nine data bits, then the ninth bit (RXB8n) is used for identifying addressand data frames. When the frame type bit (the first stop or the ninth bit) is one, the framecontains an address. When the frame type bit is zero the frame is a data frame.

Table 99. Recommended Maximum Receiver Baud Rate Error for Normal Speed Mode(U2Xn = 0)

D# (Data+Parity Bit) Rslow (%) Rfast (%) Max Total Error (%)

Recommended Max Receiver Error (%)

5 93.20 106.67 +6.67/-6.8 ± 3.0

6 94.12 105.79 +5.79/-5.88 ± 2.5

7 94.81 105.11 +5.11/-5.19 ± 2.0

8 95.36 104.58 +4.58/-4.54 ± 2.0

9 95.81 104.14 +4.14/-4.19 ± 1.5

10 96.17 103.78 +3.78/-3.83 ± 1.5

Table 100. Recommended Maximum Receiver Baud Rate Error for Double SpeedMode (U2Xn = 1)

D# (Data+Parity Bit) Rslow (%) Rfast (%) Max Total Error (%)

Recommended Max Receiver Error (%)

5 94.12 105.66 +5.66/-5.88 ± 2.5

6 94.92 104.92 +4.92/-5.08 ± 2.0

7 95.52 104,35 +4.35/-4.48 ± 1.5

8 96.00 103.90 +3.90/-4.00 ± 1.5

9 96.39 103.53 +3.53/-3.61 ± 1.5

10 96.70 103.23 +3.23/-3.30 ± 1.0

222 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The Multi-processor Communication mode enables several slave MCUs to receive datafrom a master MCU. This is done by first decoding an address frame to find out whichMCU has been addressed. If a particular slave MCU has been addressed, it will receivethe following data frames as normal, while the other slave MCUs will ignore the receivedframes until another address frame is received.

Using MPCMn For an MCU to act as a master MCU, it can use a 9-bit character frame format (UCSZn= 7). The ninth bit (TXB8n) must be set when an address frame (TXB8n = 1) or clearedwhen a data frame (TXB = 0) is being transmitted. The slave MCUs must in this case beset to use a 9-bit character frame format.

The following procedure should be used to exchange data in Multi-processor Communi-cation mode:

1. All Slave MCUs are in Multi-processor Communication mode (MPCMn in UCS-RnA is set).

2. The Master MCU sends an address frame, and all slaves receive and read this frame. In the Slave MCUs, the RXCn Flag in UCSRnA will be set as normal.

3. Each Slave MCU reads the UDRn Register and determines if it has been selected. If so, it clears the MPCMn bit in UCSRnA, otherwise it waits for the next address byte and keeps the MPCMn setting.

4. The addressed MCU will receive all data frames until a new address frame is received. The other Slave MCUs, which still have the MPCMn bit set, will ignore the data frames.

5. When the last data frame is received by the addressed MCU, the addressed MCU sets the MPCMn bit and waits for a new address frame from master. The process then repeats from 2.

Using any of the 5- to 8-bit character frame formats is possible, but impractical since theReceiver must change between using n and n+1 character frame formats. This makesfull-duplex operation difficult since the Transmitter and Receiver uses the same charac-ter size setting. If 5- to 8-bit character frames are used, the Transmitter must be set touse two stop bit (USBSn = 1) since the first stop bit is used for indicating the frame type.

Do not use Read-Modify-Write instructions (SBI and CBI) to set or clear the MPCMn bit.The MPCMn bit shares the same I/O location as the TXCn Flag and this might acciden-tally be cleared when using SBI or CBI instructions.

USART Register Description

USART I/O Data Register n– UDRn

The USART Transmit Data Buffer Register and USART Receive Data Buffer Registersshare the same I/O address referred to as USART Data Register or UDRn. The Trans-mit Data Buffer Register (TXB) will be the destination for data written to the UDRnRegister location. Reading the UDRn Register location will return the contents of theReceive Data Buffer Register (RXB).

For 5-, 6-, or 7-bit characters the upper unused bits will be ignored by the Transmitterand set to zero by the Receiver.

Bit 7 6 5 4 3 2 1 0

RXB[7:0] UDRn (Read)

TXB[7:0] UDRn (Write)

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

223

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The transmit buffer can only be written when the UDREn Flag in the UCSRnA Registeris set. Data written to UDRn when the UDREn Flag is not set, will be ignored by theUSART Transmitter. When data is written to the transmit buffer, and the Transmitter isenabled, the Transmitter will load the data into the Transmit Shift Register when theShift Register is empty. Then the data will be serially transmitted on the TxDn pin.

The receive buffer consists of a two level FIFO. The FIFO will change its state wheneverthe receive buffer is accessed. Due to this behavior of the receive buffer, do not useRead-Modify-Write instructions (SBI and CBI) on this location. Be careful when using bittest instructions (SBIC and SBIS), since these also will change the state of the FIFO.

USART Control and Status Register A – UCSRnA

• Bit 7 – RXCn: USART Receive Complete

This flag bit is set when there are unread data in the receive buffer and cleared when thereceive buffer is empty (i.e., does not contain any unread data). If the Receiver is dis-abled, the receive buffer will be flushed and consequently the RXCn bit will becomezero. The RXCn Flag can be used to generate a Receive Complete interrupt (seedescription of the RXCIEn bit).

• Bit 6 – TXCn: USART Transmit Complete

This flag bit is set when the entire frame in the Transmit Shift Register has been shiftedout and there are no new data currently present in the transmit buffer (UDRn). TheTXCn Flag bit is automatically cleared when a transmit complete interrupt is executed,or it can be cleared by writing a one to its bit location. The TXCn Flag can generate aTransmit Complete interrupt (see description of the TXCIEn bit).

• Bit 5 – UDREn: USART Data Register Empty

The UDREn Flag indicates if the transmit buffer (UDRn) is ready to receive new data. IfUDREn is one, the buffer is empty, and therefore ready to be written. The UDREn Flagcan generate a Data Register Empty interrupt (see description of the UDRIEn bit).

UDREn is set after a reset to indicate that the Transmitter is ready.

• Bit 4 – FEn: Frame Error

This bit is set if the next character in the receive buffer had a Frame Error whenreceived. I.e., when the first stop bit of the next character in the receive buffer is zero.This bit is valid until the receive buffer (UDRn) is read. The FEn bit is zero when the stopbit of received data is one. Always set this bit to zero when writing to UCSRnA.

• Bit 3 – DORn: Data OverRun

This bit is set if a Data OverRun condition is detected. A Data OverRun occurs when thereceive buffer is full (two characters), it is a new character waiting in the Receive ShiftRegister, and a new start bit is detected. This bit is valid until the receive buffer (UDRn)is read. Always set this bit to zero when writing to UCSRnA.

Bit 7 6 5 4 3 2 1 0

RXCn TXCn UDREn FEn DORn UPEn U2Xn MPCMn UCSRnA

Read/Write R R/W R R R R R/W R/W

Initial Value 0 0 1 0 0 0 0 0

224 ATmega640/1280/1281/2560/25612549A–AVR–03/05

• Bit 2 – UPEn: USART Parity Error

This bit is set if the next character in the receive buffer had a Parity Error when receivedand the Parity Checking was enabled at that point (UPMn1 = 1). This bit is valid until thereceive buffer (UDRn) is read. Always set this bit to zero when writing to UCSRnA.

• Bit 1 – U2Xn: Double the USART Transmission Speed

This bit only has effect for the asynchronous operation. Write this bit to zero when usingsynchronous operation.

Writing this bit to one will reduce the divisor of the baud rate divider from 16 to 8 effec-tively doubling the transfer rate for asynchronous communication.

• Bit 0 – MPCMn: Multi-processor Communication Mode

This bit enables the Multi-processor Communication mode. When the MPCMn bit is writ-ten to one, all the incoming frames received by the USART Receiver that do not containaddress information will be ignored. The Transmitter is unaffected by the MPCMn set-ting. For more detailed information see “Multi-processor Communication Mode” on page221.

USART Control and Status Register n B – UCSRnB

• Bit 7 – RXCIEn: RX Complete Interrupt Enable n

Writing this bit to one enables interrupt on the RXCn Flag. A USART Receive Completeinterrupt will be generated only if the RXCIEn bit is written to one, the Global InterruptFlag in SREG is written to one and the RXCn bit in UCSRnA is set.

• Bit 6 – TXCIEn: TX Complete Interrupt Enable n

Writing this bit to one enables interrupt on the TXCn Flag. A USART Transmit Completeinterrupt will be generated only if the TXCIEn bit is written to one, the Global InterruptFlag in SREG is written to one and the TXCn bit in UCSRnA is set.

• Bit 5 – UDRIEn: USART Data Register Empty Interrupt Enable n

Writing this bit to one enables interrupt on the UDREn Flag. A Data Register Emptyinterrupt will be generated only if the UDRIEn bit is written to one, the Global InterruptFlag in SREG is written to one and the UDREn bit in UCSRnA is set.

• Bit 4 – RXENn: Receiver Enable n

Writing this bit to one enables the USART Receiver. The Receiver will override normalport operation for the RxDn pin when enabled. Disabling the Receiver will flush thereceive buffer invalidating the FEn, DORn, and UPEn Flags.

• Bit 3 – TXENn: Transmitter Enable n

Writing this bit to one enables the USART Transmitter. The Transmitter will override nor-mal port operation for the TxDn pin when enabled. The disabling of the Transmitter(writing TXENn to zero) will not become effective until ongoing and pending transmis-sions are completed, i.e., when the Transmit Shift Register and Transmit Buffer Register

Bit 7 6 5 4 3 2 1 0

RXCIEn TXCIEn UDRIEn RXENn TXENn UCSZn2 RXB8n TXB8n UCSRnB

Read/Write R/W R/W R/W R/W R/W R/W R R/W

Initial Value 0 0 0 0 0 0 0 0

225

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

do not contain data to be transmitted. When disabled, the Transmitter will no longeroverride the TxDn port.

• Bit 2 – UCSZn2: Character Size n

The UCSZn2 bits combined with the UCSZn1:0 bit in UCSRnC sets the number of databits (Character SiZe) in a frame the Receiver and Transmitter use.

• Bit 1 – RXB8n: Receive Data Bit 8 n

RXB8n is the ninth data bit of the received character when operating with serial frameswith nine data bits. Must be read before reading the low bits from UDRn.

• Bit 0 – TXB8n: Transmit Data Bit 8 n

TXB8n is the ninth data bit in the character to be transmitted when operating with serialframes with nine data bits. Must be written before writing the low bits to UDRn.

USART Control and Status Register n C – UCSRnC

• Bits 7:6 – UMSELn1:0 USART Mode Select

These bits select the mode of operation of the USARTn as shown in Table 101..

Note: 1. See “USART in SPI Mode” on page 231 for full description of the Master SPI Mode(MSPIM) operation

• Bits 5:4 – UPMn1:0: Parity Mode

These bits enable and set type of parity generation and check. If enabled, the Transmit-ter will automatically generate and send the parity of the transmitted data bits withineach frame. The Receiver will generate a parity value for the incoming data and com-pare it to the UPMn setting. If a mismatch is detected, the UPEn Flag in UCSRnA will beset.

Bit 7 6 5 4 3 2 1 0

UMSELn1 UMSELn0 UPMn1 UPMn0 USBSn UCSZn1 UCSZn0 UCPOLn UCSRnC

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 1 1 0

Table 101. UMSELn Bits Settings

UMSELn1 UMSELn0 Mode

0 0 Asynchronous USART

0 1 Synchronous USART

1 0 (Reserved)

1 1 Master SPI (MSPIM)(1)

Table 102. UPMn Bits Settings

UPMn1 UPMn0 Parity Mode

0 0 Disabled

0 1 Reserved

1 0 Enabled, Even Parity

1 1 Enabled, Odd Parity

226 ATmega640/1280/1281/2560/25612549A–AVR–03/05

• Bit 3 – USBSn: Stop Bit Select

This bit selects the number of stop bits to be inserted by the Transmitter. The Receiverignores this setting.

• Bit 2:1 – UCSZn1:0: Character Size

The UCSZn1:0 bits combined with the UCSZn2 bit in UCSRnB sets the number of databits (Character SiZe) in a frame the Receiver and Transmitter use.

• Bit 0 – UCPOLn: Clock Polarity

This bit is used for synchronous mode only. Write this bit to zero when asynchronousmode is used. The UCPOLn bit sets the relationship between data output change anddata input sample, and the synchronous clock (XCKn).

USART Baud Rate Registers – UBRRLn and UBRRHn

Table 103. USBS Bit Settings

USBSn Stop Bit(s)

0 1-bit

1 2-bit

Table 104. UCSZn Bits Settings

UCSZn2 UCSZn1 UCSZn0 Character Size

0 0 0 5-bit

0 0 1 6-bit

0 1 0 7-bit

0 1 1 8-bit

1 0 0 Reserved

1 0 1 Reserved

1 1 0 Reserved

1 1 1 9-bit

Table 105. UCPOLn Bit Settings

UCPOLnTransmitted Data Changed (Output of TxDn Pin)

Received Data Sampled (Input on RxDn Pin)

0 Rising XCKn Edge Falling XCKn Edge

1 Falling XCKn Edge Rising XCKn Edge

Bit 15 14 13 12 11 10 9 8

– – – – UBRR[11:8] UBRRHn

UBRR[7:0] UBRRLn

7 6 5 4 3 2 1 0

Read/Write R R R R R/W R/W R/W R/W

R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

227

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• Bit 15:12 – Reserved Bits

These bits are reserved for future use. For compatibility with future devices, these bitmust be written to zero when UBRRH is written.

• Bit 11:0 – UBRR11:0: USART Baud Rate Register

This is a 12-bit register which contains the USART baud rate. The UBRRH contains thefour most significant bits, and the UBRRL contains the eight least significant bits of theUSART baud rate. Ongoing transmissions by the Transmitter and Receiver will be cor-rupted if the baud rate is changed. Writing UBRRL will trigger an immediate update ofthe baud rate prescaler.

Examples of Baud Rate Setting

For standard crystal and resonator frequencies, the most commonly used baud rates forasynchronous operation can be generated by using the UBRR settings in Table 106 toTable 109. UBRR values which yield an actual baud rate differing less than 0.5% fromthe target baud rate, are bold in the table. Higher error ratings are acceptable, but theReceiver will have less noise resistance when the error ratings are high, especially forlarge serial frames (see “Asynchronous Operational Range” on page 220). The errorvalues are calculated using the following equation:

Error[%]BaudRateClosest Match

BaudRate-------------------------------------------------------- 1–⎝ ⎠

⎛ ⎞ 100%•=

Table 106. Examples of UBRRn Settings for Commonly Used Oscillator Frequencies

Baud Rate (bps)

fosc = 1.0000 MHz fosc = 1.8432 MHz fosc = 2.0000 MHz

U2Xn = 0 U2Xn = 1 U2Xn = 0 U2Xn = 1 U2Xn = 0 U2Xn = 1

UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error

2400 25 0.2% 51 0.2% 47 0.0% 95 0.0% 51 0.2% 103 0.2%

4800 12 0.2% 25 0.2% 23 0.0% 47 0.0% 25 0.2% 51 0.2%

9600 6 -7.0% 12 0.2% 11 0.0% 23 0.0% 12 0.2% 25 0.2%

14.4k 3 8.5% 8 -3.5% 7 0.0% 15 0.0% 8 -3.5% 16 2.1%

19.2k 2 8.5% 6 -7.0% 5 0.0% 11 0.0% 6 -7.0% 12 0.2%

28.8k 1 8.5% 3 8.5% 3 0.0% 7 0.0% 3 8.5% 8 -3.5%

38.4k 1 -18.6% 2 8.5% 2 0.0% 5 0.0% 2 8.5% 6 -7.0%

57.6k 0 8.5% 1 8.5% 1 0.0% 3 0.0% 1 8.5% 3 8.5%

76.8k – – 1 -18.6% 1 -25.0% 2 0.0% 1 -18.6% 2 8.5%

115.2k – – 0 8.5% 0 0.0% 1 0.0% 0 8.5% 1 8.5%

230.4k – – – – – – 0 0.0% – – – –

250k – – – – – – – – – – 0 0.0%

Max. (1) 62.5 kbps 125 kbps 115.2 kbps 230.4 kbps 125 kbps 250 kbps

1. UBRR = 0, Error = 0.0%

228 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Table 107. Examples of UBRRn Settings for Commonly Used Oscillator Frequencies (Continued)

Baud Rate (bps)

fosc = 3.6864 MHz fosc = 4.0000 MHz fosc = 7.3728 MHz

U2Xn = 0 U2Xn = 1 U2Xn = 0 U2Xn = 1 U2Xn = 0 U2Xn = 1

UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error

2400 95 0.0% 191 0.0% 103 0.2% 207 0.2% 191 0.0% 383 0.0%

4800 47 0.0% 95 0.0% 51 0.2% 103 0.2% 95 0.0% 191 0.0%

9600 23 0.0% 47 0.0% 25 0.2% 51 0.2% 47 0.0% 95 0.0%

14.4k 15 0.0% 31 0.0% 16 2.1% 34 -0.8% 31 0.0% 63 0.0%

19.2k 11 0.0% 23 0.0% 12 0.2% 25 0.2% 23 0.0% 47 0.0%

28.8k 7 0.0% 15 0.0% 8 -3.5% 16 2.1% 15 0.0% 31 0.0%

38.4k 5 0.0% 11 0.0% 6 -7.0% 12 0.2% 11 0.0% 23 0.0%

57.6k 3 0.0% 7 0.0% 3 8.5% 8 -3.5% 7 0.0% 15 0.0%

76.8k 2 0.0% 5 0.0% 2 8.5% 6 -7.0% 5 0.0% 11 0.0%

115.2k 1 0.0% 3 0.0% 1 8.5% 3 8.5% 3 0.0% 7 0.0%

230.4k 0 0.0% 1 0.0% 0 8.5% 1 8.5% 1 0.0% 3 0.0%

250k 0 -7.8% 1 -7.8% 0 0.0% 1 0.0% 1 -7.8% 3 -7.8%

0.5M – – 0 -7.8% – – 0 0.0% 0 -7.8% 1 -7.8%

1M – – – – – – – – – – 0 -7.8%

Max. (1) 230.4 kbps 460.8 kbps 250 kbps 0.5 Mbps 460.8 kbps 921.6 kbps

1. UBRR = 0, Error = 0.0%

229

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 108. Examples of UBRRn Settings for Commonly Used Oscillator Frequencies (Continued)

Baud Rate (bps)

fosc = 8.0000 MHz fosc = 11.0592 MHz fosc = 14.7456 MHz

U2Xn = 0 U2Xn = 1 U2Xn = 0 U2Xn = 1 U2Xn = 0 U2Xn = 1

UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error

2400 207 0.2% 416 -0.1% 287 0.0% 575 0.0% 383 0.0% 767 0.0%

4800 103 0.2% 207 0.2% 143 0.0% 287 0.0% 191 0.0% 383 0.0%

9600 51 0.2% 103 0.2% 71 0.0% 143 0.0% 95 0.0% 191 0.0%

14.4k 34 -0.8% 68 0.6% 47 0.0% 95 0.0% 63 0.0% 127 0.0%

19.2k 25 0.2% 51 0.2% 35 0.0% 71 0.0% 47 0.0% 95 0.0%

28.8k 16 2.1% 34 -0.8% 23 0.0% 47 0.0% 31 0.0% 63 0.0%

38.4k 12 0.2% 25 0.2% 17 0.0% 35 0.0% 23 0.0% 47 0.0%

57.6k 8 -3.5% 16 2.1% 11 0.0% 23 0.0% 15 0.0% 31 0.0%

76.8k 6 -7.0% 12 0.2% 8 0.0% 17 0.0% 11 0.0% 23 0.0%

115.2k 3 8.5% 8 -3.5% 5 0.0% 11 0.0% 7 0.0% 15 0.0%

230.4k 1 8.5% 3 8.5% 2 0.0% 5 0.0% 3 0.0% 7 0.0%

250k 1 0.0% 3 0.0% 2 -7.8% 5 -7.8% 3 -7.8% 6 5.3%

0.5M 0 0.0% 1 0.0% – – 2 -7.8% 1 -7.8% 3 -7.8%

1M – – 0 0.0% – – – – 0 -7.8% 1 -7.8%

Max. (1) 0.5 Mbps 1 Mbps 691.2 kbps 1.3824 Mbps 921.6 kbps 1.8432 Mbps

1. UBRR = 0, Error = 0.0%

230 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Table 109. Examples of UBRRn Settings for Commonly Used Oscillator Frequencies (Continued)

Baud Rate (bps)

fosc = 16.0000 MHz fosc = 18.4320 MHz fosc = 20.0000 MHz

U2Xn = 0 U2Xn = 1 U2Xn = 0 U2Xn = 1 U2Xn = 0 U2Xn = 1

UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error UBRR Error

2400 416 -0.1% 832 0.0% 479 0.0% 959 0.0% 520 0.0% 1041 0.0%

4800 207 0.2% 416 -0.1% 239 0.0% 479 0.0% 259 0.2% 520 0.0%

9600 103 0.2% 207 0.2% 119 0.0% 239 0.0% 129 0.2% 259 0.2%

14.4k 68 0.6% 138 -0.1% 79 0.0% 159 0.0% 86 -0.2% 173 -0.2%

19.2k 51 0.2% 103 0.2% 59 0.0% 119 0.0% 64 0.2% 129 0.2%

28.8k 34 -0.8% 68 0.6% 39 0.0% 79 0.0% 42 0.9% 86 -0.2%

38.4k 25 0.2% 51 0.2% 29 0.0% 59 0.0% 32 -1.4% 64 0.2%

57.6k 16 2.1% 34 -0.8% 19 0.0% 39 0.0% 21 -1.4% 42 0.9%

76.8k 12 0.2% 25 0.2% 14 0.0% 29 0.0% 15 1.7% 32 -1.4%

115.2k 8 -3.5% 16 2.1% 9 0.0% 19 0.0% 10 -1.4% 21 -1.4%

230.4k 3 8.5% 8 -3.5% 4 0.0% 9 0.0% 4 8.5% 10 -1.4%

250k 3 0.0% 7 0.0% 4 -7.8% 8 2.4% 4 0.0% 9 0.0%

0.5M 1 0.0% 3 0.0% – – 4 -7.8% – – 4 0.0%

1M 0 0.0% 1 0.0% – – – – – – – –

Max. (1) 1 Mbps 2 Mbps 1.152 Mbps 2.304 Mbps 1.25 Mbps 2.5 Mbps

1. UBRR = 0, Error = 0.0%

231

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

USART in SPI Mode The Universal Synchronous and Asynchronous serial Receiver and Transmitter(USART) can be set to a master SPI compliant mode of operation. The Master SPIMode (MSPIM) has the following features:• Full Duplex, Three-wire Synchronous Data Transfer• Master Operation• Supports all four SPI Modes of Operation (Mode 0, 1, 2, and 3)• LSB First or MSB First Data Transfer (Configurable Data Order)• Queued Operation (Double Buffered)• High Resolution Baud Rate Generator• High Speed Operation (fXCKmax = fCK/2)• Flexible Interrupt Generation

Overview Setting both UMSELn1:0 bits to one enables the USART in MSPIM logic. In this mode ofoperation the SPI master control logic takes direct control over the USART resources.These resources include the transmitter and receiver shift register and buffers, and thebaud rate generator. The parity generator and checker, the data and clock recoverylogic, and the RX and TX control logic is disabled. The USART RX and TX control logicis replaced by a common SPI transfer control logic. However, the pin control logic andinterrupt generation logic is identical in both modes of operation.

The I/O register locations are the same in both modes. However, some of the functional-ity of the control registers changes when using MSPIM.

Clock Generation The Clock Generation logic generates the base clock for the Transmitter and Receiver.For USART MSPIM mode of operation only internal clock generation (i.e. master opera-tion) is supported. The Data Direction Register for the XCKn pin (DDR_XCKn) musttherefore be set to one (i.e. as output) for the USART in MSPIM to operate correctly.Preferably the DDR_XCKn should be set up before the USART in MSPIM is enabled(i.e. TXENn and RXENn bit set to one).

The internal clock generation used in MSPIM mode is identical to the USART synchro-nous master mode. The baud rate or UBRRn setting can therefore be calculated usingthe same equations, see Table 110:

Note: 1. The baud rate is defined to be the transfer rate in bit per second (bps)

BAUD Baud rate (in bits per second, bps)

fOSC System Oscillator clock frequency

UBRRnContents of the UBRRnH and UBRRnL Registers, (0-4095)

SPI Data Modes and Timing

There are four combinations of XCKn (SCK) phase and polarity with respect to serialdata, which are determined by control bits UCPHAn and UCPOLn. The data transfer

Table 110. Equations for Calculating Baud Rate Register Setting

Operating ModeEquation for Calculating Baud

Rate(1)Equation for Calculating

UBRRn Value

Synchronous Master mode

BAUDfOSC

2 UBRRn 1+( )---------------------------------------= UBRRn

fOSC2BAUD-------------------- 1–=

232 ATmega640/1280/1281/2560/25612549A–AVR–03/05

timing diagrams are shown in Figure 90. Data bits are shifted out and latched in onopposite edges of the XCKn signal, ensuring sufficient time for data signals to stabilize.The UCPOLn and UCPHAn functionality is summarized in Table 111. Note that chang-ing the setting of any of these bits will corrupt all ongoing communication for both theReceiver and Transmitter.

Figure 90. UCPHAn and UCPOLn data transfer timing diagrams.

Frame Formats A serial frame for the MSPIM is defined to be one character of 8 data bits. The USARTin MSPIM mode has two valid frame formats:

• 8-bit data with MSB first

• 8-bit data with LSB first

A frame starts with the least or most significant data bit. Then the next data bits, up to atotal of eight, are succeeding, ending with the most or least significant bit accordingly.When a complete frame is transmitted, a new frame can directly follow it, or the commu-nication line can be set to an idle (high) state.

The UDORDn bit in UCSRnC sets the frame format used by the USART in MSPIMmode. The Receiver and Transmitter use the same setting. Note that changing the set-ting of any of these bits will corrupt all ongoing communication for both the Receiver andTransmitter.

16-bit data transfer can be achieved by writing two data bytes to UDRn. A UART trans-mit complete interrupt will then signal that the 16-bit value has been shifted out.

Table 111. UCPOLn and UCPHAn Functionality-

UCPOLn UCPHAn SPI Mode Leading Edge Trailing Edge

0 0 0 Sample (Rising) Setup (Falling)

0 1 1 Setup (Rising) Sample (Falling)

1 0 2 Sample (Falling) Setup (Rising)

1 1 3 Setup (Falling) Sample (Rising)

XCK

Data setup (TXD)

Data sample (RXD)

XCK

Data setup (TXD)

Data sample (RXD)

XCK

Data setup (TXD)

Data sample (RXD)

XCK

Data setup (TXD)

Data sample (RXD)

UCPOL=0 UCPOL=1

UC

PH

A=0

UC

PH

A=1

233

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

USART MSPIM Initialization The USART in MSPIM mode has to be initialized before any communication can takeplace. The initialization process normally consists of setting the baud rate, setting mas-ter mode of operation (by setting DDR_XCKn to one), setting frame format and enablingthe Transmitter and the Receiver. Only the transmitter can operate independently. Forinterrupt driven USART operation, the Global Interrupt Flag should be cleared (and thusinterrupts globally disabled) when doing the initialization.Note: To ensure immediate initialization of the XCKn output the baud-rate register (UBRRn)

must be zero at the time the transmitter is enabled. Contrary to the normal mode USARToperation the UBRRn must then be written to the desired value after the transmitter isenabled, but before the first transmission is started. Setting UBRRn to zero beforeenabling the transmitter is not necessary if the initialization is done immediately after areset since UBRRn is reset to zero.

Before doing a re-initialization with changed baud rate, data mode, or frame format, besure that there is no ongoing transmissions during the period the registers are changed.The TXCn Flag can be used to check that the Transmitter has completed all transfers,and the RXCn Flag can be used to check that there are no unread data in the receivebuffer. Note that the TXCn Flag must be cleared before each transmission (beforeUDRn is written) if it is used for this purpose.

The following simple USART initialization code examples show one assembly and oneC function that are equal in functionality. The examples assume polling (no interruptsenabled). The baud rate is given as a function parameter. For the assembly code, thebaud rate parameter is assumed to be stored in the r17:r16 registers.

234 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. See “About Code Examples” on page 8.

Data Transfer Using the USART in MSPI mode requires the Transmitter to be enabled, i.e. the TXENnbit in the UCSRnB register is set to one. When the Transmitter is enabled, the normalport operation of the TxDn pin is overridden and given the function as the Transmitter'sserial output. Enabling the receiver is optional and is done by setting the RXENn bit inthe UCSRnB register to one. When the receiver is enabled, the normal pin operation ofthe RxDn pin is overridden and given the function as the Receiver's serial input. TheXCKn will in both cases be used as the transfer clock.

After initialization the USART is ready for doing data transfers. A data transfer is initiatedby writing to the UDRn I/O location. This is the case for both sending and receiving datasince the transmitter controls the transfer clock. The data written to UDRn is moved from

Assembly Code Example(1)

USART_Init:

clr r18

out UBRRnH,r18

out UBRRnL,r18

; Setting the XCKn port pin as output, enables master mode.

sbi XCKn_DDR, XCKn

; Set MSPI mode of operation and SPI data mode 0.

ldi r18, (1<<UMSELn1)|(1<<UMSELn0)|(0<<UCPHAn)|(0<<UCPOLn)

out UCSRnC,r18

; Enable receiver and transmitter.

ldi r18, (1<<RXENn)|(1<<TXENn)

out UCSRnB,r18

; Set baud rate.

; IMPORTANT: The Baud Rate must be set after the transmitter is enabled!

out UBRRnH, r17

out UBRRnL, r18

ret

C Code Example(1)

void USART_Init( unsigned int baud )

UBRRn = 0;

/* Setting the XCKn port pin as output, enables master mode. */

XCKn_DDR |= (1<<XCKn);

/* Set MSPI mode of operation and SPI data mode 0. */

UCSRnC = (1<<UMSELn1)|(1<<UMSELn0)|(0<<UCPHAn)|(0<<UCPOLn);

/* Enable receiver and transmitter. */

UCSRnB = (1<<RXENn)|(1<<TXENn);

/* Set baud rate. */

/* IMPORTANT: The Baud Rate must be set after the transmitter is enabled */

UBRRn = baud;

235

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

the transmit buffer to the shift register when the shift register is ready to send a newframe.Note: To keep the input buffer in sync with the number of data bytes transmitted, the UDRn reg-

ister must be read once for each byte transmitted. The input buffer operation is identicalto normal USART mode, i.e. if an overflow occurs the character last received will be lost,not the first data in the buffer. This means that if four bytes are transferred, byte 1 first,then byte 2, 3, and 4, and the UDRn is not read before all transfers are completed, thenbyte 3 to be received will be lost, and not byte 1.

The following code examples show a simple USART in MSPIM mode transfer functionbased on polling of the Data Register Empty (UDREn) Flag and the Receive Complete(RXCn) Flag. The USART has to be initialized before the function can be used. For theassembly code, the data to be sent is assumed to be stored in Register R16 and thedata received will be available in the same register (R16) after the function returns.

The function simply waits for the transmit buffer to be empty by checking the UDREnFlag, before loading it with new data to be transmitted. The function then waits for datato be present in the receive buffer by checking the RXCn Flag, before reading the bufferand returning the value..

Note: 1. See “About Code Examples” on page 8.

Transmitter and Receiver Flags and Interrupts

The RXCn, TXCn, and UDREn flags and corresponding interrupts in USART in MSPIMmode are identical in function to the normal USART operation. However, the receivererror status flags (FE, DOR, and PE) are not in use and is always read as zero.

Assembly Code Example(1)

USART_MSPIM_Transfer:

; Wait for empty transmit buffer

sbis UCSRnA, UDREn

rjmp USART_MSPIM_Transfer

; Put data (r16) into buffer, sends the data

out UDRn,r16

; Wait for data to be received

USART_MSPIM_Wait_RXCn:

sbis UCSRnA, RXCn

rjmp USART_MSPIM_Wait_RXCn

; Get and return received data from buffer

in r16, UDRn

ret

C Code Example(1)

unsigned char USART_Receive( void )

/* Wait for empty transmit buffer */

while ( !( UCSRnA & (1<<UDREn)) );

/* Put data into buffer, sends the data */

UDRn = data;

/* Wait for data to be received */

while ( !(UCSRnA & (1<<RXCn)) );

/* Get and return received data from buffer */

return UDRn;

236 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Disabling the Transmitter or Receiver

The disabling of the transmitter or receiver in USART in MSPIM mode is identical infunction to the normal USART operation.

USART MSPIM Register Description

The following section describes the registers used for SPI operation using the USART.

USART MSPIM I/O Data Register - UDRn

The function and bit description of the USART data register (UDRn) in MSPI mode isidentical to normal USART operation. See “USART I/O Data Register n– UDRn” onpage 222.

USART MSPIM Control and Status Register n A - UCSRnA

•

• Bit 7 - RXCn: USART Receive Complete

This flag bit is set when there are unread data in the receive buffer and cleared when thereceive buffer is empty (i.e., does not contain any unread data). If the Receiver is dis-abled, the receive buffer will be flushed and consequently the RXCn bit will becomezero. The RXCn Flag can be used to generate a Receive Complete interrupt (seedescription of the RXCIEn bit).

• Bit 6 - TXCn: USART Transmit Complete

This flag bit is set when the entire frame in the Transmit Shift Register has been shiftedout and there are no new data currently present in the transmit buffer (UDRn). TheTXCn Flag bit is automatically cleared when a transmit complete interrupt is executed,or it can be cleared by writing a one to its bit location. The TXCn Flag can generate aTransmit Complete interrupt (see description of the TXCIEn bit).

• Bit 5 - UDREn: USART Data Register Empty

The UDREn Flag indicates if the transmit buffer (UDRn) is ready to receive new data. IfUDREn is one, the buffer is empty, and therefore ready to be written. The UDREn Flagcan generate a Data Register Empty interrupt (see description of the UDRIE bit).UDREn is set after a reset to indicate that the Transmitter is ready.

• Bit 4:0 - Reserved Bits in MSPI mode

When in MSPI mode, these bits are reserved for future use. For compatibility with futuredevices, these bits must be written to zero when UCSRnA is written.

USART MSPIM Control and Status Register n B - UCSRnB

• Bit 7 - RXCIEn: RX Complete Interrupt Enable

Writing this bit to one enables interrupt on the RXCn Flag. A USART Receive Completeinterrupt will be generated only if the RXCIEn bit is written to one, the Global InterruptFlag in SREG is written to one and the RXCn bit in UCSRnA is set.

Bit 7 6 5 4 3 2 1 0

RXCn TXCn UDREn - - - - - UCSRnA

Read/Write R/W R/W R/W R R R R R

Initial Value 0 0 0 0 0 1 1 0

Bit 7 6 5 4 3 2 1 0

RXCIEn TXCIEn UDRIE RXENn TXENn - - - UCSRnB

Read/Write R/W R/W R/W R/W R/W R R R

Initial Value 0 0 0 0 0 1 1 0

237

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• Bit 6 - TXCIEn: TX Complete Interrupt Enable

Writing this bit to one enables interrupt on the TXCn Flag. A USART Transmit Completeinterrupt will be generated only if the TXCIEn bit is written to one, the Global InterruptFlag in SREG is written to one and the TXCn bit in UCSRnA is set.

• Bit 5 - UDRIE: USART Data Register Empty Interrupt Enable

Writing this bit to one enables interrupt on the UDREn Flag. A Data Register Emptyinterrupt will be generated only if the UDRIE bit is written to one, the Global InterruptFlag in SREG is written to one and the UDREn bit in UCSRnA is set.

• Bit 4 - RXENn: Receiver Enable

Writing this bit to one enables the USART Receiver in MSPIM mode. The Receiver willoverride normal port operation for the RxDn pin when enabled. Disabling the Receiverwill flush the receive buffer. Only enabling the receiver in MSPI mode (i.e. settingRXENn=1 and TXENn=0) has no meaning since it is the transmitter that controls thetransfer clock and since only master mode is supported.

• Bit 3 - TXENn: Transmitter Enable

Writing this bit to one enables the USART Transmitter. The Transmitter will override nor-mal port operation for the TxDn pin when enabled. The disabling of the Transmitter(writing TXENn to zero) will not become effective until ongoing and pending transmis-sions are completed, i.e., when the Transmit Shift Register and Transmit Buffer Registerdo not contain data to be transmitted. When disabled, the Transmitter will no longeroverride the TxDn port.

• Bit 2:0 - Reserved Bits in MSPI mode

When in MSPI mode, these bits are reserved for future use. For compatibility with futuredevices, these bits must be written to zero when UCSRnB is written.

USART MSPIM Control and Status Register n C - UCSRnC

• Bit 7:6 - UMSELn1:0: USART Mode Select

These bits select the mode of operation of the USART as shown in Table 112. See“USART Control and Status Register n C – UCSRnC” on page 225 for full description ofthe normal USART operation. The MSPIM is enabled when both UMSELn bits are set toone. The UDORDn, UCPHAn, and UCPOLn can be set in the same write operationwhere the MSPIM is enabled.

Bit 7 6 5 4 3 2 1 0

UMSELn1 UMSELn0 - - - UDORDn UCPHAn UCPOLn UCSRnC

Read/Write R/W R/W R R R R/W R/W R/W

Initial Value 0 0 0 0 0 1 1 0

Table 112. UMSELn Bits Settings

UMSELn1 UMSELn0 Mode

0 0 Asynchronous USART

0 1 Synchronous USART

1 0 (Reserved)

1 1 Master SPI (MSPIM)

238 ATmega640/1280/1281/2560/25612549A–AVR–03/05

• Bit 5:3 - Reserved Bits in MSPI mode

When in MSPI mode, these bits are reserved for future use. For compatibility with futuredevices, these bits must be written to zero when UCSRnC is written.

• Bit 2 - UDORDn: Data Order

When set to one the LSB of the data word is transmitted first. When set to zero the MSBof the data word is transmitted first. Refer to the Frame Formats section page 4 fordetails.

• Bit 1 - UCPHAn: Clock Phase

The UCPHAn bit setting determine if data is sampled on the leasing edge (first) or tailing(last) edge of XCKn. Refer to the SPI Data Modes and Timing section page 4 for details.

• Bit 0 - UCPOLn: Clock Polarity

The UCPOLn bit sets the polarity of the XCKn clock. The combination of the UCPOLnand UCPHAn bit settings determine the timing of the data transfer. Refer to the SPI DataModes and Timing section page 4 for details.

USART MSPIM Baud Rate Registers - UBRRnL and UBRRnH

The function and bit description of the baud rate registers in MSPI mode is identical tonormal USART operation. See “USART Baud Rate Registers – UBRRLn and UBRRHn”on page 226.

AVR USART MSPIM vs.AVR SPI

The USART in MSPIM mode is fully compatible with the AVR SPI regarding:

• Master mode timing diagram.

• The UCPOLn bit functionality is identical to the SPI CPOL bit.

• The UCPHAn bit functionality is identical to the SPI CPHA bit.

• The UDORDn bit functionality is identical to the SPI DORD bit.

However, since the USART in MSPIM mode reuses the USART resources, the use ofthe USART in MSPIM mode is somewhat different compared to the SPI. In addition todifferences of the control register bits, and that only master operation is supported bythe USART in MSPIM mode, the following features differ between the two modules:

• The USART in MSPIM mode includes (double) buffering of the transmitter. The SPI has no buffer.

• The USART in MSPIM mode receiver includes an additional buffer level.

• The SPI WCOL (Write Collision) bit is not included in USART in MSPIM mode.

• The SPI double speed mode (SPI2X) bit is not included. However, the same effect is achieved by setting UBRRn accordingly.

• Interrupt timing is not compatible.

• Pin control differs due to the master only operation of the USART in MSPIM mode.

A comparison of the USART in MSPIM mode and the SPI pins is shown in Table 113 onpage 239.

239

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 113. Comparison of USART in MSPIM mode and SPI pins.

USART_MSPIM SPI Comment

TxDn MOSI Master Out only

RxDn MISO Master In only

XCKn SCK (Functionally identical)

(N/A) SS Not supported by USART in MSPIM

240 ATmega640/1280/1281/2560/25612549A–AVR–03/05

241

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

2-wire Serial Interface

Features • Simple yet Powerful and Flexible Communication Interface, only two Bus Lines needed• Both Master and Slave Operation Supported• Device can Operate as Transmitter or Receiver• 7-bit Address Space Allows up to 128 Different Slave Addresses• Multi-master Arbitration Support• Up to 400 kHz Data Transfer Speed• Slew-rate Limited Output Drivers• Noise Suppression Circuitry Rejects Spikes on Bus Lines• Fully Programmable Slave Address with General Call Support• Address Recognition Causes Wake-up When AVR is in Sleep Mode

2-wire Serial Interface Bus Definition

The 2-wire Serial Interface (TWI) is ideally suited for typical microcontroller applications.The TWI protocol allows the systems designer to interconnect up to 128 differentdevices using only two bi-directional bus lines, one for clock (SCL) and one for data(SDA). The only external hardware needed to implement the bus is a single pull-upresistor for each of the TWI bus lines. All devices connected to the bus have individualaddresses, and mechanisms for resolving bus contention are inherent in the TWIprotocol.

Figure 91. TWI Bus Interconnection

TWI Terminology The following definitions are frequently encountered in this section.

The Power Reduction TWI bit, PRTWI bit in “Power Reduction Register 0 - PRR0” onpage 54 must be written to zero to enable the 2-wire Serial Interface.

Device 1 Device 2 Device 3 Device n

SDA

SCL

........ R1 R2

VCC

Table 114. TWI Terminology

Term Description

Master The device that initiates and terminates a transmission. The Master also generates the SCL clock.

Slave The device addressed by a Master.

Transmitter The device placing data on the bus.

Receiver The device reading data from the bus.

242 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Electrical Interconnection As depicted in Figure 91, both bus lines are connected to the positive supply voltagethrough pull-up resistors. The bus drivers of all TWI-compliant devices are open-drain oropen-collector. This implements a wired-AND function which is essential to the opera-tion of the interface. A low level on a TWI bus line is generated when one or more TWIdevices output a zero. A high level is output when all TWI devices trim-state their out-puts, allowing the pull-up resistors to pull the line high. Note that all AVR devicesconnected to the TWI bus must be powered in order to allow any bus operation.

The number of devices that can be connected to the bus is only limited by the buscapacitance limit of 400 pF and the 7-bit slave address space. A detailed specification ofthe electrical characteristics of the TWI is given in “SPI Timing Characteristics” on page372. Two different sets of specifications are presented there, one relevant for busspeeds below 100 kHz, and one valid for bus speeds up to 400 kHz.

Data Transfer and Frame Format

Transferring Bits Each data bit transferred on the TWI bus is accompanied by a pulse on the clock line.The level of the data line must be stable when the clock line is high. The only exceptionto this rule is for generating start and stop conditions.

Figure 92. Data Validity

START and STOP Conditions The Master initiates and terminates a data transmission. The transmission is initiatedwhen the Master issues a START condition on the bus, and it is terminated when theMaster issues a STOP condition. Between a START and a STOP condition, the bus isconsidered busy, and no other master should try to seize control of the bus. A specialcase occurs when a new START condition is issued between a START and STOP con-dition. This is referred to as a REPEATED START condition, and is used when theMaster wishes to initiate a new transfer without relinquishing control of the bus. After aREPEATED START, the bus is considered busy until the next STOP. This is identical tothe START behavior, and therefore START is used to describe both START andREPEATED START for the remainder of this datasheet, unless otherwise noted. Asdepicted below, START and STOP conditions are signalled by changing the level of theSDA line when the SCL line is high.

SDA

SCL

Data Stable Data Stable

Data Change

243

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 93. START, REPEATED START and STOP conditions

Address Packet Format All address packets transmitted on the TWI bus are 9 bits long, consisting of 7 addressbits, one READ/WRITE control bit and an acknowledge bit. If the READ/WRITE bit isset, a read operation is to be performed, otherwise a write operation should be per-formed. When a Slave recognizes that it is being addressed, it should acknowledge bypulling SDA low in the ninth SCL (ACK) cycle. If the addressed Slave is busy, or forsome other reason can not service the Master’s request, the SDA line should be lefthigh in the ACK clock cycle. The Master can then transmit a STOP condition, or aREPEATED START condition to initiate a new transmission. An address packet consist-ing of a slave address and a READ or a WRITE bit is called SLA+R or SLA+W,respectively.

The MSB of the address byte is transmitted first. Slave addresses can freely be allo-cated by the designer, but the address 0000 000 is reserved for a general call.

When a general call is issued, all slaves should respond by pulling the SDA line low inthe ACK cycle. A general call is used when a Master wishes to transmit the same mes-sage to several slaves in the system. When the general call address followed by a Writebit is transmitted on the bus, all slaves set up to acknowledge the general call will pullthe SDA line low in the ack cycle. The following data packets will then be received by allthe slaves that acknowledged the general call. Note that transmitting the general calladdress followed by a Read bit is meaningless, as this would cause contention if severalslaves started transmitting different data.

All addresses of the format 1111 xxx should be reserved for future purposes.

Figure 94. Address Packet Format

SDA

SCL

START STOPREPEATED STARTSTOP START

SDA

SCL

START

1 2 7 8 9

Addr MSB Addr LSB R/W ACK

244 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Data Packet Format All data packets transmitted on the TWI bus are nine bits long, consisting of one databyte and an acknowledge bit. During a data transfer, the Master generates the clock andthe START and STOP conditions, while the Receiver is responsible for acknowledgingthe reception. An Acknowledge (ACK) is signalled by the Receiver pulling the SDA linelow during the ninth SCL cycle. If the Receiver leaves the SDA line high, a NACK is sig-nalled. When the Receiver has received the last byte, or for some reason cannot receiveany more bytes, it should inform the Transmitter by sending a NACK after the final byte.The MSB of the data byte is transmitted first.

Figure 95. Data Packet Format

Combining Address and Data Packets into a Transmission

A transmission basically consists of a START condition, a SLA+R/W, one or more datapackets and a STOP condition. An empty message, consisting of a START followed bya STOP condition, is illegal. Note that the Wired-ANDing of the SCL line can be used toimplement handshaking between the Master and the Slave. The Slave can extend theSCL low period by pulling the SCL line low. This is useful if the clock speed set up by theMaster is too fast for the Slave, or the Slave needs extra time for processing betweenthe data transmissions. The Slave extending the SCL low period will not affect the SCLhigh period, which is determined by the Master. As a consequence, the Slave canreduce the TWI data transfer speed by prolonging the SCL duty cycle.

Figure 96 shows a typical data transmission. Note that several data bytes can be trans-mitted between the SLA+R/W and the STOP condition, depending on the softwareprotocol implemented by the application software.

Figure 96. Typical Data Transmission

1 2 7 8 9

Data MSB Data LSB ACK

AggregateSDA

SDA fromTransmitter

SDA fromReceiver

SCL fromMaster

SLA+R/W Data ByteSTOP, REPEATED

START or NextData Byte

1 2 7 8 9

Data Byte

Data MSB Data LSB ACK

SDA

SCL

START

1 2 7 8 9

Addr MSB Addr LSB R/W ACK

SLA+R/W STOP

245

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Multi-master Bus Systems, Arbitration and Synchronization

The TWI protocol allows bus systems with several masters. Special concerns havebeen taken in order to ensure that transmissions will proceed as normal, even if two ormore masters initiate a transmission at the same time. Two problems arise in multi-mas-ter systems:

• An algorithm must be implemented allowing only one of the masters to complete the transmission. All other masters should cease transmission when they discover that they have lost the selection process. This selection process is called arbitration. When a contending master discovers that it has lost the arbitration process, it should immediately switch to Slave mode to check whether it is being addressed by the winning master. The fact that multiple masters have started transmission at the same time should not be detectable to the slaves, i.e. the data being transferred on the bus must not be corrupted.

• Different masters may use different SCL frequencies. A scheme must be devised to synchronize the serial clocks from all masters, in order to let the transmission proceed in a lockstep fashion. This will facilitate the arbitration process.

The wired-ANDing of the bus lines is used to solve both these problems. The serialclocks from all masters will be wired-ANDed, yielding a combined clock with a highperiod equal to the one from the Master with the shortest high period. The low period ofthe combined clock is equal to the low period of the Master with the longest low period.Note that all masters listen to the SCL line, effectively starting to count their SCL highand low time-out periods when the combined SCL line goes high or low, respectively.

Figure 97. SCL Synchronization Between Multiple Masters

Arbitration is carried out by all masters continuously monitoring the SDA line after out-putting data. If the value read from the SDA line does not match the value the Masterhad output, it has lost the arbitration. Note that a Master can only lose arbitration when itoutputs a high SDA value while another Master outputs a low value. The losing Mastershould immediately go to Slave mode, checking if it is being addressed by the winningMaster. The SDA line should be left high, but losing masters are allowed to generate aclock signal until the end of the current data or address packet. Arbitration will continueuntil only one Master remains, and this may take many bits. If several masters are tryingto address the same Slave, arbitration will continue into the data packet.

TA low TA high

SCL fromMaster A

SCL fromMaster B

SCL BusLine

TBlow TBhigh

Masters StartCounting Low Period

Masters StartCounting High Period

246 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 98. Arbitration Between Two Masters

Note that arbitration is not allowed between:

• A REPEATED START condition and a data bit.

• A STOP condition and a data bit.

• A REPEATED START and a STOP condition.

It is the user software’s responsibility to ensure that these illegal arbitration conditionsnever occur. This implies that in multi-master systems, all data transfers must use thesame composition of SLA+R/W and data packets. In other words: All transmissionsmust contain the same number of data packets, otherwise the result of the arbitration isundefined.

SDA fromMaster A

SDA fromMaster B

SDA Line

SynchronizedSCL Line

START Master A LosesArbitration, SDAA SDA

247

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Overview of the TWI Module

The TWI module is comprised of several submodules, as shown in Figure 99. All regis-ters drawn in a thick line are accessible through the AVR data bus.

Figure 99. Overview of the TWI Module

SCL and SDA Pins These pins interface the AVR TWI with the rest of the MCU system. The output driverscontain a slew-rate limiter in order to conform to the TWI specification. The input stagescontain a spike suppression unit removing spikes shorter than 50 ns. Note that the inter-nal pull-ups in the AVR pads can be enabled by setting the PORT bits corresponding tothe SCL and SDA pins, as explained in the I/O Port section. The internal pull-ups can insome systems eliminate the need for external ones.

Bit Rate Generator Unit This unit controls the period of SCL when operating in a Master mode. The SCL periodis controlled by settings in the TWI Bit Rate Register (TWBR) and the Prescaler bits inthe TWI Status Register (TWSR). Slave operation does not depend on Bit Rate or Pres-caler settings, but the CPU clock frequency in the Slave must be at least 16 times higherthan the SCL frequency. Note that slaves may prolong the SCL low period, therebyreducing the average TWI bus clock period. The SCL frequency is generated accordingto the following equation:

TW

I Uni

t

Address Register(TWAR)

Address Match Unit

Address Comparator

Control Unit

Control Register(TWCR)

Status Register(TWSR)

State Machine andStatus control

SCL

Slew-rateControl

SpikeFilter

SDA

Slew-rateControl

SpikeFilter

Bit Rate Generator

Bit Rate Register(TWBR)

Prescaler

Bus Interface Unit

START / STOPControl

Arbitration detection Ack

Spike Suppression

Address/Data ShiftRegister (TWDR)

248 ATmega640/1280/1281/2560/25612549A–AVR–03/05

• TWBR = Value of the TWI Bit Rate Register.

• TWPS = Value of the prescaler bits in the TWI Status Register.Note: TWBR should be 10 or higher if the TWI operates in Master mode. If TWBR is lower than

10, the Master may produce an incorrect output on SDA and SCL for the reminder of thebyte. The problem occurs when operating the TWI in Master mode, sending Start + SLA+ R/W to a Slave (a Slave does not need to be connected to the bus for the condition tohappen).

Bus Interface Unit This unit contains the Data and Address Shift Register (TWDR), a START/STOP Con-troller and Arbitration detection hardware. The TWDR contains the address or databytes to be transmitted, or the address or data bytes received. In addition to the 8-bitTWDR, the Bus Interface Unit also contains a register containing the (N)ACK bit to betransmitted or received. This (N)ACK Register is not directly accessible by the applica-tion software. However, when receiving, it can be set or cleared by manipulating theTWI Control Register (TWCR). When in Transmitter mode, the value of the received(N)ACK bit can be determined by the value in the TWSR.

The START/STOP Controller is responsible for generation and detection of START,REPEATED START, and STOP conditions. The START/STOP controller is able todetect START and STOP conditions even when the AVR MCU is in one of the sleepmodes, enabling the MCU to wake up if addressed by a Master.

If the TWI has initiated a transmission as Master, the Arbitration Detection hardwarecontinuously monitors the transmission trying to determine if arbitration is in process. Ifthe TWI has lost an arbitration, the Control Unit is informed. Correct action can then betaken and appropriate status codes generated.

Address Match Unit The Address Match unit checks if received address bytes match the seven-bit addressin the TWI Address Register (TWAR). If the TWI General Call Recognition Enable(TWGCE) bit in the TWAR is written to one, all incoming address bits will also be com-pared against the General Call address. Upon an address match, the Control Unit isinformed, allowing correct action to be taken. The TWI may or may not acknowledge itsaddress, depending on settings in the TWCR. The Address Match unit is able to com-pare addresses even when the AVR MCU is in sleep mode, enabling the MCU to wakeup if addressed by a Master. If another interrupt (e.g., INT0) occurs during TWI Power-down address match and wakes up the CPU, the TWI aborts operation and return to it’sidle state. If this cause any problems, ensure that TWI Address Match is the onlyenabled interrupt when entering Power-down.

Control Unit The Control unit monitors the TWI bus and generates responses corresponding to set-tings in the TWI Control Register (TWCR). When an event requiring the attention of theapplication occurs on the TWI bus, the TWI Interrupt Flag (TWINT) is asserted. In thenext clock cycle, the TWI Status Register (TWSR) is updated with a status code identify-ing the event. The TWSR only contains relevant status information when the TWIInterrupt Flag is asserted. At all other times, the TWSR contains a special status codeindicating that no relevant status information is available. As long as the TWINT Flag isset, the SCL line is held low. This allows the application software to complete its tasksbefore allowing the TWI transmission to continue.

The TWINT Flag is set in the following situations:

• After the TWI has transmitted a START/REPEATED START condition.

SCL frequency CPU Clock frequency

16 2(TWBR) 4TWPS⋅+-----------------------------------------------------------=

249

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• After the TWI has transmitted SLA+R/W.

• After the TWI has transmitted an address byte.

• After the TWI has lost arbitration.

• After the TWI has been addressed by own slave address or general call.

• After the TWI has received a data byte.

• After a STOP or REPEATED START has been received while still addressed as a Slave.

• When a bus error has occurred due to an illegal START or STOP condition.

TWI Register Description

TWI Bit Rate Register – TWBR

• Bits 7..0 – TWI Bit Rate Register

TWBR selects the division factor for the bit rate generator. The bit rate generator is afrequency divider which generates the SCL clock frequency in the Master modes. See“Bit Rate Generator Unit” on page 247 for calculating bit rates.

TWI Control Register – TWCR

The TWCR is used to control the operation of the TWI. It is used to enable the TWI, toinitiate a Master access by applying a START condition to the bus, to generate aReceiver acknowledge, to generate a stop condition, and to control halting of the buswhile the data to be written to the bus are written to the TWDR. It also indicates a writecollision if data is attempted written to TWDR while the register is inaccessible.

• Bit 7 – TWINT: TWI Interrupt Flag

This bit is set by hardware when the TWI has finished its current job and expects appli-cation software response. If the I-bit in SREG and TWIE in TWCR are set, the MCU willjump to the TWI Interrupt Vector. While the TWINT Flag is set, the SCL low period isstretched. The TWINT Flag must be cleared by software by writing a logic one to it. Notethat this flag is not automatically cleared by hardware when executing the interrupt rou-tine. Also note that clearing this flag starts the operation of the TWI, so all accesses tothe TWI Address Register (TWAR), TWI Status Register (TWSR), and TWI Data Regis-ter (TWDR) must be complete before clearing this flag.

• Bit 6 – TWEA: TWI Enable Acknowledge Bit

The TWEA bit controls the generation of the acknowledge pulse. If the TWEA bit is writ-ten to one, the ACK pulse is generated on the TWI bus if the following conditions aremet:

1. The device’s own slave address has been received.

2. A general call has been received, while the TWGCE bit in the TWAR is set.

Bit 7 6 5 4 3 2 1 0

TWBR7 TWBR6 TWBR5 TWBR4 TWBR3 TWBR2 TWBR1 TWBR0 TWBRRead/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIE TWCRRead/Write R/W R/W R/W R/W R R/W R R/W

Initial Value 0 0 0 0 0 0 0 0

250 ATmega640/1280/1281/2560/25612549A–AVR–03/05

3. A data byte has been received in Master Receiver or Slave Receiver mode.

By writing the TWEA bit to zero, the device can be virtually disconnected from the 2-wireSerial Bus temporarily. Address recognition can then be resumed by writing the TWEAbit to one again.

• Bit 5 – TWSTA: TWI START Condition Bit

The application writes the TWSTA bit to one when it desires to become a Master on the2-wire Serial Bus. The TWI hardware checks if the bus is available, and generates aSTART condition on the bus if it is free. However, if the bus is not free, the TWI waitsuntil a STOP condition is detected, and then generates a new START condition to claimthe bus Master status. TWSTA must be cleared by software when the START conditionhas been transmitted.

• Bit 4 – TWSTO: TWI STOP Condition Bit

Writing the TWSTO bit to one in Master mode will generate a STOP condition on the 2-wire Serial Bus. When the STOP condition is executed on the bus, the TWSTO bit iscleared automatically. In Slave mode, setting the TWSTO bit can be used to recoverfrom an error condition. This will not generate a STOP condition, but the TWI returns toa well-defined unaddressed Slave mode and releases the SCL and SDA lines to a highimpedance state.

• Bit 3 – TWWC: TWI Write Collision Flag

The TWWC bit is set when attempting to write to the TWI Data Register – TWDR whenTWINT is low. This flag is cleared by writing the TWDR Register when TWINT is high.

• Bit 2 – TWEN: TWI Enable Bit

The TWEN bit enables TWI operation and activates the TWI interface. When TWEN iswritten to one, the TWI takes control over the I/O pins connected to the SCL and SDApins, enabling the slew-rate limiters and spike filters. If this bit is written to zero, the TWIis switched off and all TWI transmissions are terminated, regardless of any ongoingoperation.

• Bit 1 – Res: Reserved Bit

This bit is a reserved bit and will always read as zero.

• Bit 0 – TWIE: TWI Interrupt Enable

When this bit is written to one, and the I-bit in SREG is set, the TWI interrupt request willbe activated for as long as the TWINT Flag is high.

TWI Status Register – TWSR

• Bits 7..3 – TWS: TWI Status

These 5 bits reflect the status of the TWI logic and the 2-wire Serial Bus. The differentstatus codes are described later in this section. Note that the value read from TWSRcontains both the 5-bit status value and the 2-bit prescaler value. The application

Bit 7 6 5 4 3 2 1 0

TWS7 TWS6 TWS5 TWS4 TWS3 – TWPS1 TWPS0 TWSRRead/Write R R R R R R R/W R/W

Initial Value 1 1 1 1 1 0 0 0

251

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

designer should mask the prescaler bits to zero when checking the Status bits. Thismakes status checking independent of prescaler setting. This approach is used in thisdatasheet, unless otherwise noted.

• Bit 2 – Res: Reserved Bit

This bit is reserved and will always read as zero.

• Bits 1..0 – TWPS: TWI Prescaler Bits

These bits can be read and written, and control the bit rate prescaler.

To calculate bit rates, see “Bit Rate Generator Unit” on page 247. The value ofTWPS1..0 is used in the equation.

TWI Data Register – TWDR

In Transmit mode, TWDR contains the next byte to be transmitted. In Receive mode, theTWDR contains the last byte received. It is writable while the TWI is not in the process ofshifting a byte. This occurs when the TWI Interrupt Flag (TWINT) is set by hardware.Note that the Data Register cannot be initialized by the user before the first interruptoccurs. The data in TWDR remains stable as long as TWINT is set. While data is shiftedout, data on the bus is simultaneously shifted in. TWDR always contains the last bytepresent on the bus, except after a wake up from a sleep mode by the TWI interrupt. Inthis case, the contents of TWDR is undefined. In the case of a lost bus arbitration, nodata is lost in the transition from Master to Slave. Handling of the ACK bit is controlledautomatically by the TWI logic, the CPU cannot access the ACK bit directly.

• Bits 7..0 – TWD: TWI Data Register

These eight bits constitute the next data byte to be transmitted, or the latest data bytereceived on the 2-wire Serial Bus.

TWI (Slave) Address Register – TWAR

The TWAR should be loaded with the 7-bit Slave address (in the seven most significantbits of TWAR) to which the TWI will respond when programmed as a Slave Transmitteror Receiver, and not needed in the Master modes. In multimaster systems, TWAR mustbe set in masters which can be addressed as Slaves by other Masters.

Table 115. TWI Bit Rate Prescaler

TWPS1 TWPS0 Prescaler Value

0 0 1

0 1 4

1 0 16

1 1 64

Bit 7 6 5 4 3 2 1 0

TWD7 TWD6 TWD5 TWD4 TWD3 TWD2 TWD1 TWD0 TWDRRead/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 1 1 1 1 1 1 1 1

Bit 7 6 5 4 3 2 1 0

TWA6 TWA5 TWA4 TWA3 TWA2 TWA1 TWA0 TWGCE TWAR

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 1 1 1 1 1 1 1 0

252 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The LSB of TWAR is used to enable recognition of the general call address (0x00).There is an associated address comparator that looks for the slave address (or generalcall address if enabled) in the received serial address. If a match is found, an interruptrequest is generated.

• Bits 7..1 – TWA: TWI (Slave) Address Register

These seven bits constitute the slave address of the TWI unit.

• Bit 0 – TWGCE: TWI General Call Recognition Enable Bit

If set, this bit enables the recognition of a General Call given over the 2-wire Serial Bus.

TWI (Slave) Address Mask Register – TWAMR

• Bits 7..1 – TWAM: TWI Address Mask

The TWAMR can be loaded with a 7-bit Slave Address mask. Each of the bits inTWAMR can mask (disable) the corresponding address bit in the TWI Address Register(TWAR). If the mask bit is set to one then the address match logic ignores the comparebetween the incoming address bit and the corresponding bit in TWAR. Figure 100shows the address match logic in detail.

Figure 100. TWI Address Match Logic, Block Diagram

• Bit 0 – Res: Reserved Bit

This bit is reserved and will always read as zero.

Using the TWI The AVR TWI is byte-oriented and interrupt based. Interrupts are issued after all busevents, like reception of a byte or transmission of a START condition. Because the TWIis interrupt-based, the application software is free to carry on other operations during aTWI byte transfer. Note that the TWI Interrupt Enable (TWIE) bit in TWCR together withthe Global Interrupt Enable bit in SREG allow the application to decide whether or notassertion of the TWINT Flag should generate an interrupt request. If the TWIE bit iscleared, the application must poll the TWINT Flag in order to detect actions on the TWIbus.

When the TWINT Flag is asserted, the TWI has finished an operation and awaits appli-cation response. In this case, the TWI Status Register (TWSR) contains a value

Bit 7 6 5 4 3 2 1 0

TWAM[6:0] – TWAMR

Read/Write R/W R/W R/W R/W R/W R/W R/W R

Initial Value 0 0 0 0 0 0 0 0

AddressMatch

Address Bit Comparator 0

Address Bit Comparator 6..1

TWAR0

TWAMR0

AddressBit 0

253

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

indicating the current state of the TWI bus. The application software can then decidehow the TWI should behave in the next TWI bus cycle by manipulating the TWCR andTWDR Registers.

Figure 101 is a simple example of how the application can interface to the TWI hard-ware. In this example, a Master wishes to transmit a single data byte to a Slave. Thisdescription is quite abstract, a more detailed explanation follows later in this section. Asimple code example implementing the desired behavior is also presented.

Figure 101. Interfacing the Application to the TWI in a Typical Transmission

1. The first step in a TWI transmission is to transmit a START condition. This is done by writing a specific value into TWCR, instructing the TWI hardware to transmit a START condition. Which value to write is described later on. However, it is important that the TWINT bit is set in the value written. Writing a one to TWINT clears the flag. The TWI will not start any operation as long as the TWINT bit in TWCR is set. Immediately after the application has cleared TWINT, the TWI will initiate transmission of the START condition.

2. When the START condition has been transmitted, the TWINT Flag in TWCR is set, and TWSR is updated with a status code indicating that the START condition has successfully been sent.

3. The application software should now examine the value of TWSR, to make sure that the START condition was successfully transmitted. If TWSR indicates other-wise, the application software might take some special action, like calling an error routine. Assuming that the status code is as expected, the application must load SLA+W into TWDR. Remember that TWDR is used both for address and data. After TWDR has been loaded with the desired SLA+W, a specific value must be written to TWCR, instructing the TWI hardware to transmit the SLA+W present in TWDR. Which value to write is described later on. However, it is important that the TWINT bit is set in the value written. Writing a one to TWINT clears the flag. The TWI will not start any operation as long as the TWINT bit in TWCR is set. Immediately after the application has cleared TWINT, the TWI will initiate transmission of the address packet.

4. When the address packet has been transmitted, the TWINT Flag in TWCR is set, and TWSR is updated with a status code indicating that the address packet has

START SLA+W A Data A STOP

1. Applicationwrites to TWCR to

initiatetransmission of

START

2. TWINT set.Status code indicatesSTART condition sent

4. TWINT set.Status code indicates

SLA+W sent, ACKreceived

6. TWINT set.Status code indicates

data sent, ACK received

3. Check TWSR to see if START was sent. Application loads SLA+W into

TWDR, and loads appropriate control signals into TWCR, makin sure that

TWINT is written to one, and TWSTA is written to zero.

5. Check TWSR to see if SLA+W wassent and ACK received.

Application loads data into TWDR, andloads appropriate control signals intoTWCR, making sure that TWINT is

written to one

7. Check TWSR to see if data was sentand ACK received.

Application loads appropriate controlsignals to send STOP into TWCR,

making sure that TWINT is written to one

TWI bus

IndicatesTWINT set

App

licat

ion

Act

ion

TW

IH

ardw

are

Act

ion

254 ATmega640/1280/1281/2560/25612549A–AVR–03/05

successfully been sent. The status code will also reflect whether a Slave acknowledged the packet or not.

5. The application software should now examine the value of TWSR, to make sure that the address packet was successfully transmitted, and that the value of the ACK bit was as expected. If TWSR indicates otherwise, the application software might take some special action, like calling an error routine. Assuming that the status code is as expected, the application must load a data packet into TWDR. Subsequently, a specific value must be written to TWCR, instructing the TWI hardware to transmit the data packet present in TWDR. Which value to write is described later on. However, it is important that the TWINT bit is set in the value written. Writing a one to TWINT clears the flag. The TWI will not start any opera-tion as long as the TWINT bit in TWCR is set. Immediately after the application has cleared TWINT, the TWI will initiate transmission of the data packet.

6. When the data packet has been transmitted, the TWINT Flag in TWCR is set, and TWSR is updated with a status code indicating that the data packet has suc-cessfully been sent. The status code will also reflect whether a Slave acknowledged the packet or not.

7. The application software should now examine the value of TWSR, to make sure that the data packet was successfully transmitted, and that the value of the ACK bit was as expected. If TWSR indicates otherwise, the application software might take some special action, like calling an error routine. Assuming that the status code is as expected, the application must write a specific value to TWCR, instructing the TWI hardware to transmit a STOP condition. Which value to write is described later on. However, it is important that the TWINT bit is set in the value written. Writing a one to TWINT clears the flag. The TWI will not start any operation as long as the TWINT bit in TWCR is set. Immediately after the appli-cation has cleared TWINT, the TWI will initiate transmission of the STOP condition. Note that TWINT is NOT set after a STOP condition has been sent.

Even though this example is simple, it shows the principles involved in all TWI transmis-sions. These can be summarized as follows:

• When the TWI has finished an operation and expects application response, the TWINT Flag is set. The SCL line is pulled low until TWINT is cleared.

• When the TWINT Flag is set, the user must update all TWI Registers with the value relevant for the next TWI bus cycle. As an example, TWDR must be loaded with the value to be transmitted in the next bus cycle.

• After all TWI Register updates and other pending application software tasks have been completed, TWCR is written. When writing TWCR, the TWINT bit should be set. Writing a one to TWINT clears the flag. The TWI will then commence executing whatever operation was specified by the TWCR setting.

In the following an assembly and C implementation of the example is given. Note thatthe code below assumes that several definitions have been made, for example by usinginclude-files.

255

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Assembly Code Example C Example Comments

1 ldi r16, (1<<TWINT)|(1<<TWSTA)|

(1<<TWEN)

out TWCR, r16

TWCR = (1<<TWINT)|(1<<TWSTA)|

(1<<TWEN)Send START condition

2 wait1:

in r16,TWCR

sbrs r16,TWINT

rjmp wait1

while (!(TWCR & (1<<TWINT)))

;Wait for TWINT Flag set. This indicates that the START condition has been transmitted

3 in r16,TWSR

andi r16, 0xF8

cpi r16, START

brne ERROR

if ((TWSR & 0xF8) != START)

ERROR();Check value of TWI Status Register. Mask prescaler bits. If status different from START go to ERROR

ldi r16, SLA_W

out TWDR, r16

ldi r16, (1<<TWINT) | (1<<TWEN)

out TWCR, r16

TWDR = SLA_W;

TWCR = (1<<TWINT) | (1<<TWEN);Load SLA_W into TWDR Register. Clear TWINT bit in TWCR to start transmission of address

4 wait2:

in r16,TWCR

sbrs r16,TWINT

rjmp wait2

while (!(TWCR & (1<<TWINT)))

;Wait for TWINT Flag set. This indicates that the SLA+W has been transmitted, and ACK/NACK has been received.

5 in r16,TWSR

andi r16, 0xF8

cpi r16, MT_SLA_ACK

brne ERROR

if ((TWSR & 0xF8) != MT_SLA_ACK)

ERROR();Check value of TWI Status Register. Mask prescaler bits. If status different from MT_SLA_ACK go to ERROR

ldi r16, DATA

out TWDR, r16

ldi r16, (1<<TWINT) | (1<<TWEN)

out TWCR, r16

TWDR = DATA;

TWCR = (1<<TWINT) | (1<<TWEN);Load DATA into TWDR Register. Clear TWINT bit in TWCR to start transmission of data

6 wait3:

in r16,TWCR

sbrs r16,TWINT

rjmp wait3

while (!(TWCR & (1<<TWINT)))

;Wait for TWINT Flag set. This indicates that the DATA has been transmitted, and ACK/NACK has been received.

7 in r16,TWSR

andi r16, 0xF8

cpi r16, MT_DATA_ACK

brne ERROR

if ((TWSR & 0xF8) != MT_DATA_ACK)

ERROR();Check value of TWI Status Register. Mask prescaler bits. If status different from MT_DATA_ACK go to ERROR

ldi r16, (1<<TWINT)|(1<<TWEN)|

(1<<TWSTO)

out TWCR, r16

TWCR = (1<<TWINT)|(1<<TWEN)|

(1<<TWSTO);Transmit STOP condition

256 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Transmission Modes The TWI can operate in one of four major modes. These are named Master Transmitter(MT), Master Receiver (MR), Slave Transmitter (ST) and Slave Receiver (SR). Severalof these modes can be used in the same application. As an example, the TWI can useMT mode to write data into a TWI EEPROM, MR mode to read the data back from theEEPROM. If other masters are present in the system, some of these might transmit datato the TWI, and then SR mode would be used. It is the application software that decideswhich modes are legal.

The following sections describe each of these modes. Possible status codes aredescribed along with figures detailing data transmission in each of the modes. These fig-ures contain the following abbreviations:

S: START condition

Rs: REPEATED START condition

R: Read bit (high level at SDA)

W: Write bit (low level at SDA)

A: Acknowledge bit (low level at SDA)

A: Not acknowledge bit (high level at SDA)

Data: 8-bit data byte

P: STOP condition

SLA: Slave Address

In Figure 103 to Figure 109, circles are used to indicate that the TWINT Flag is set. Thenumbers in the circles show the status code held in TWSR, with the prescaler bitsmasked to zero. At these points, actions must be taken by the application to continue orcomplete the TWI transfer. The TWI transfer is suspended until the TWINT Flag iscleared by software.

When the TWINT Flag is set, the status code in TWSR is used to determine the appro-priate software action. For each status code, the required software action and details ofthe following serial transfer are given in Table 116 to Table 119. Note that the prescalerbits are masked to zero in these tables.

Master Transmitter Mode In the Master Transmitter mode, a number of data bytes are transmitted to a SlaveReceiver (see Figure 102). In order to enter a Master mode, a START condition must betransmitted. The format of the following address packet determines whether MasterTransmitter or Master Receiver mode is to be entered. If SLA+W is transmitted, MTmode is entered, if SLA+R is transmitted, MR mode is entered. All the status codesmentioned in this section assume that the prescaler bits are zero or are masked to zero.

257

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 102. Data Transfer in Master Transmitter Mode

A START condition is sent by writing the following value to TWCR:

TWEN must be set to enable the 2-wire Serial Interface, TWSTA must be written to oneto transmit a START condition and TWINT must be written to one to clear the TWINTFlag. The TWI will then test the 2-wire Serial Bus and generate a START condition assoon as the bus becomes free. After a START condition has been transmitted, theTWINT Flag is set by hardware, and the status code in TWSR will be 0x08 (see Table116). In order to enter MT mode, SLA+W must be transmitted. This is done by writingSLA+W to TWDR. Thereafter the TWINT bit should be cleared (by writing it to one) tocontinue the transfer. This is accomplished by writing the following value to TWCR:

When SLA+W have been transmitted and an acknowledgement bit has been received,TWINT is set again and a number of status codes in TWSR are possible. Possible sta-tus codes in Master mode are 0x18, 0x20, or 0x38. The appropriate action to be takenfor each of these status codes is detailed in Table 116.

When SLA+W has been successfully transmitted, a data packet should be transmitted.This is done by writing the data byte to TWDR. TWDR must only be written whenTWINT is high. If not, the access will be discarded, and the Write Collision bit (TWWC)will be set in the TWCR Register. After updating TWDR, the TWINT bit should becleared (by writing it to one) to continue the transfer. This is accomplished by writing thefollowing value to TWCR:

This scheme is repeated until the last byte has been sent and the transfer is ended bygenerating a STOP condition or a repeated START condition. A STOP condition is gen-erated by writing the following value to TWCR:

A REPEATED START condition is generated by writing the following value to TWCR:

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 1 X 1 0 X 1 0 X

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 1 X 0 0 X 1 0 X

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 1 X 0 0 X 1 0 X

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 1 X 0 1 X 1 0 X

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 1 X 1 0 X 1 0 X

Device 1MASTER

TRANSMITTER

Device 2SLAVE

RECEIVERDevice 3 Device n

SDA

SCL

........ R1 R2

VCC

258 ATmega640/1280/1281/2560/25612549A–AVR–03/05

After a repeated START condition (state 0x10) the 2-wire Serial Interface can accessthe same Slave again, or a new Slave without transmitting a STOP condition. RepeatedSTART enables the Master to switch between Slaves, Master Transmitter mode andMaster Receiver mode without losing control of the bus.

Table 116. Status codes for Master Transmitter ModeStatus Code(TWSR)Prescaler Bitsare 0

Status of the 2-wire Serial Busand 2-wire Serial Interface Hard-ware

Application Software Response

Next Action Taken by TWI Hardware

To/from TWDR To TWCRSTA STO TWINT TWEA

0x08 A START condition has beentransmitted

Load SLA+W 0 0 1 X SLA+W will be transmitted;ACK or NOT ACK will be received

0x10 A repeated START conditionhas been transmitted

Load SLA+W or

Load SLA+R

0

0

0

0

1

1

X

X

SLA+W will be transmitted;ACK or NOT ACK will be receivedSLA+R will be transmitted;Logic will switch to Master Receiver mode

0x18 SLA+W has been transmitted;ACK has been received

Load data byte or

No TWDR action orNo TWDR action or

No TWDR action

0

10

1

0

01

1

1

11

1

X

XX

X

Data byte will be transmitted and ACK or NOT ACK will be receivedRepeated START will be transmittedSTOP condition will be transmitted andTWSTO Flag will be resetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be reset

0x20 SLA+W has been transmitted;NOT ACK has been received

Load data byte or

No TWDR action orNo TWDR action or

No TWDR action

0

10

1

0

01

1

1

11

1

X

XX

X

Data byte will be transmitted and ACK or NOT ACK will be receivedRepeated START will be transmittedSTOP condition will be transmitted andTWSTO Flag will be resetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be reset

0x28 Data byte has been transmitted;ACK has been received

Load data byte or

No TWDR action orNo TWDR action or

No TWDR action

0

10

1

0

01

1

1

11

1

X

XX

X

Data byte will be transmitted and ACK or NOT ACK will be receivedRepeated START will be transmittedSTOP condition will be transmitted andTWSTO Flag will be resetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be reset

0x30 Data byte has been transmitted;NOT ACK has been received

Load data byte or

No TWDR action orNo TWDR action or

No TWDR action

0

10

1

0

01

1

1

11

1

X

XX

X

Data byte will be transmitted and ACK or NOT ACK will be receivedRepeated START will be transmittedSTOP condition will be transmitted andTWSTO Flag will be resetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be reset

0x38 Arbitration lost in SLA+W or databytes

No TWDR action or

No TWDR action

0

1

0

0

1

1

X

X

2-wire Serial Bus will be released and not addressed Slave mode enteredA START condition will be transmitted when the bus be-comes free

259

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 103. Formats and States in the Master Transmitter Mode

Master Receiver Mode In the Master Receiver mode, a number of data bytes are received from a Slave Trans-mitter (Slave see Figure 104). In order to enter a Master mode, a START condition mustbe transmitted. The format of the following address packet determines whether MasterTransmitter or Master Receiver mode is to be entered. If SLA+W is transmitted, MTmode is entered, if SLA+R is transmitted, MR mode is entered. All the status codesmentioned in this section assume that the prescaler bits are zero or are masked to zero.

S SLA W A DATA A P

$08 $18 $28

R SLA W

$10

A P

$20

P

$30

A or A

$38

A

Other mastercontinues A or A

$38

Other mastercontinues

R

A

$68

Other mastercontinues

$78 $B0To correspondingstates in slave mode

MT

MR

Successfulltransmissionto a slavereceiver

Next transferstarted with arepeated startcondition

Not acknowledgereceived after theslave address

Not acknowledgereceived after a databyte

Arbitration lost in slaveaddress or data byte

Arbitration lost andaddressed as slave

DATA A

n

From master to slave

From slave to master

Any number of data bytesand their associated acknowledge bits

This number (contained in TWSR) correspondsto a defined state of the Two-Wire Serial Bus. Theprescaler bits are zero or masked to zero

S

260 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 104. Data Transfer in Master Receiver Mode

A START condition is sent by writing the following value to TWCR:

TWEN must be written to one to enable the 2-wire Serial Interface, TWSTA must bewritten to one to transmit a START condition and TWINT must be set to clear the TWINTFlag. The TWI will then test the 2-wire Serial Bus and generate a START condition assoon as the bus becomes free. After a START condition has been transmitted, theTWINT Flag is set by hardware, and the status code in TWSR will be 0x08 (See Table116). In order to enter MR mode, SLA+R must be transmitted. This is done by writingSLA+R to TWDR. Thereafter the TWINT bit should be cleared (by writing it to one) tocontinue the transfer. This is accomplished by writing the following value to TWCR:

When SLA+R have been transmitted and an acknowledgement bit has been received,TWINT is set again and a number of status codes in TWSR are possible. Possible sta-tus codes in Master mode are 0x38, 0x40, or 0x48. The appropriate action to be takenfor each of these status codes is detailed in Table 117. Received data can be read fromthe TWDR Register when the TWINT Flag is set high by hardware. This scheme isrepeated until the last byte has been received. After the last byte has been received, theMR should inform the ST by sending a NACK after the last received data byte. Thetransfer is ended by generating a STOP condition or a repeated START condition. ASTOP condition is generated by writing the following value to TWCR:

A REPEATED START condition is generated by writing the following value to TWCR:

After a repeated START condition (state 0x10) the 2-wire Serial Interface can accessthe same Slave again, or a new Slave without transmitting a STOP condition. Repeated

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 1 X 1 0 X 1 0 X

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 1 X 0 0 X 1 0 X

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 1 X 0 1 X 1 0 X

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 1 X 1 0 X 1 0 X

Device 1MASTER

RECEIVER

Device 2SLAVE

TRANSMITTERDevice 3 Device n

SDA

SCL

........ R1 R2

VCC

261

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

START enables the Master to switch between Slaves, Master Transmitter mode andMaster Receiver mode without losing control over the bus.

Table 117. Status codes for Master Receiver ModeStatus Code(TWSR)Prescaler Bitsare 0

Status of the 2-wire Serial Busand 2-wire Serial Interface Hard-ware

Application Software Response

Next Action Taken by TWI HardwareTo/from TWDR

To TWCRSTA STO TWINT TWEA

0x08 A START condition has beentransmitted

Load SLA+R 0 0 1 X SLA+R will be transmittedACK or NOT ACK will be received

0x10 A repeated START conditionhas been transmitted

Load SLA+R or

Load SLA+W

0

0

0

0

1

1

X

X

SLA+R will be transmittedACK or NOT ACK will be receivedSLA+W will be transmittedLogic will switch to Master Transmitter mode

0x38 Arbitration lost in SLA+R or NOTACK bit

No TWDR action or

No TWDR action

0

1

0

0

1

1

X

X

2-wire Serial Bus will be released and not addressed Slave mode will be enteredA START condition will be transmitted when the busbecomes free

0x40 SLA+R has been transmitted;ACK has been received

No TWDR action or

No TWDR action

0

0

0

0

1

1

0

1

Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

0x48 SLA+R has been transmitted;NOT ACK has been received

No TWDR action orNo TWDR action or

No TWDR action

10

1

01

1

11

1

XX

X

Repeated START will be transmittedSTOP condition will be transmitted and TWSTO Flag will be resetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be reset

0x50 Data byte has been received;ACK has been returned

Read data byte or

Read data byte

0

0

0

0

1

1

0

1

Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

0x58 Data byte has been received;NOT ACK has been returned

Read data byte orRead data byte or

Read data byte

10

1

01

1

11

1

XX

X

Repeated START will be transmittedSTOP condition will be transmitted and TWSTO Flag will be resetSTOP condition followed by a START condition will be transmitted and TWSTO Flag will be reset

262 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 105. Formats and States in the Master Receiver Mode

Slave Receiver Mode In the Slave Receiver mode, a number of data bytes are received from a Master Trans-mitter (see Figure 106). All the status codes mentioned in this section assume that theprescaler bits are zero or are masked to zero.

Figure 106. Data transfer in Slave Receiver mode

To initiate the Slave Receiver mode, TWAR and TWCR must be initialized as follows:

S SLA R A DATA A

$08 $40 $50

SLA R

$10

A P

$48

A or A

$38

Other mastercontinues

$38

Other mastercontinues

W

A

$68

Other mastercontinues

$78 $B0To correspondingstates in slave mode

MR

MT

Successfullreceptionfrom a slavereceiver

Next transferstarted with arepeated startcondition

Not acknowledgereceived after theslave address

Arbitration lost in slaveaddress or data byte

Arbitration lost andaddressed as slave

DATA A

n

From master to slave

From slave to master

Any number of data bytesand their associated acknowledge bits

This number (contained in TWSR) correspondsto a defined state of the Two-Wire Serial Bus. The prescaler bits are zero or masked to zero

PDATA A

$58

A

RS

TWAR TWA6 TWA5 TWA4 TWA3 TWA2 TWA1 TWA0 TWGCEvalue Device’s Own Slave Address

Device 3 Device n

SDA

SCL

........ R1 R2

VCC

Device 2MASTER

TRANSMITTER

Device 1SLAVE

RECEIVER

263

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

The upper 7 bits are the address to which the 2-wire Serial Interface will respond whenaddressed by a Master. If the LSB is set, the TWI will respond to the general calladdress (0x00), otherwise it will ignore the general call address.

TWEN must be written to one to enable the TWI. The TWEA bit must be written to oneto enable the acknowledgement of the device’s own slave address or the general calladdress. TWSTA and TWSTO must be written to zero.

When TWAR and TWCR have been initialized, the TWI waits until it is addressed by itsown slave address (or the general call address if enabled) followed by the data directionbit. If the direction bit is “0” (write), the TWI will operate in SR mode, otherwise ST modeis entered. After its own slave address and the write bit have been received, the TWINTFlag is set and a valid status code can be read from TWSR. The status code is used todetermine the appropriate software action. The appropriate action to be taken for eachstatus code is detailed in Table 118. The Slave Receiver mode may also be entered ifarbitration is lost while the TWI is in the Master mode (see states 0x68 and 0x78).

If the TWEA bit is reset during a transfer, the TWI will return a “Not Acknowledge” (“1”)to SDA after the next received data byte. This can be used to indicate that the Slave isnot able to receive any more bytes. While TWEA is zero, the TWI does not acknowledgeits own slave address. However, the 2-wire Serial Bus is still monitored and address rec-ognition may resume at any time by setting TWEA. This implies that the TWEA bit maybe used to temporarily isolate the TWI from the 2-wire Serial Bus.

In all sleep modes other than Idle mode, the clock system to the TWI is turned off. If theTWEA bit is set, the interface can still acknowledge its own slave address or the generalcall address by using the 2-wire Serial Bus clock as a clock source. The part will thenwake up from sleep and the TWI will hold the SCL clock low during the wake up anduntil the TWINT Flag is cleared (by writing it to one). Further data reception will be car-ried out as normal, with the AVR clocks running as normal. Observe that if the AVR isset up with a long start-up time, the SCL line may be held low for a long time, blockingother data transmissions.

Note that the 2-wire Serial Interface Data Register – TWDR does not reflect the last bytepresent on the bus when waking up from these Sleep modes.

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 0 1 0 0 0 1 0 X

264 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Table 118. Status Codes for Slave Receiver ModeStatus Code(TWSR)Prescaler Bitsare 0

Status of the 2-wire Serial Bus and2-wire Serial Interface Hardware

Application Software Response

Next Action Taken by TWI HardwareTo/from TWDR

To TWCRSTA STO TWINT TWEA

0x60 Own SLA+W has been received;ACK has been returned

No TWDR action or

No TWDR action

X

X

0

0

1

1

0

1

Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

0x68 Arbitration lost in SLA+R/W asMaster; own SLA+W has been received; ACK has been returned

No TWDR action or

No TWDR action

X

X

0

0

1

1

0

1

Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

0x70 General call address has been received; ACK has been returned

No TWDR action or

No TWDR action

X

X

0

0

1

1

0

1

Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

0x78 Arbitration lost in SLA+R/W asMaster; General call address hasbeen received; ACK has been returned

No TWDR action or

No TWDR action

X

X

0

0

1

1

0

1

Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

0x80 Previously addressed with ownSLA+W; data has been received;ACK has been returned

Read data byte or

Read data byte

X

X

0

0

1

1

0

1

Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

0x88 Previously addressed with ownSLA+W; data has been received;NOT ACK has been returned

Read data byte or

Read data byte or

Read data byte or

Read data byte

0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1

Switched to the not addressed Slave mode;no recognition of own SLA or GCASwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”Switched to the not addressed Slave mode;no recognition of own SLA or GCA;a START condition will be transmitted when the bus becomes freeSwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”;a START condition will be transmitted when the bus becomes free

0x90 Previously addressed with general call; data has been re-ceived; ACK has been returned

Read data byte or

Read data byte

X

X

0

0

1

1

0

1

Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

0x98 Previously addressed with general call; data has been received; NOT ACK has been returned

Read data byte or

Read data byte or

Read data byte or

Read data byte

0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1

Switched to the not addressed Slave mode;no recognition of own SLA or GCASwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”Switched to the not addressed Slave mode;no recognition of own SLA or GCA;a START condition will be transmitted when the bus becomes freeSwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”;a START condition will be transmitted when the bus becomes free

0xA0 A STOP condition or repeatedSTART condition has been received while still addressed asSlave

No action 0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1

Switched to the not addressed Slave mode;no recognition of own SLA or GCASwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”Switched to the not addressed Slave mode;no recognition of own SLA or GCA;a START condition will be transmitted when the bus becomes freeSwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”;a START condition will be transmitted when the bus becomes free

265

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 107. Formats and States in the Slave Receiver Mode

S SLA W A DATA A

$60 $80

$88

A

$68

Reception of the ownslave address and one ormore data bytes. All areacknowledged

Last data byte receivedis not acknowledged

Arbitration lost as masterand addressed as slave

Reception of the general calladdress and one or more databytes

Last data byte received isnot acknowledged

n

From master to slave

From slave to master

Any number of data bytesand their associated acknowledge bits

This number (contained in TWSR) correspondsto a defined state of the Two-Wire Serial Bus. The prescaler bits are zero or masked to zero

P or SDATA A

$80 $A0

P or SA

A DATA A

$70 $90

$98

A

$78

P or SDATA A

$90 $A0

P or SA

General Call

Arbitration lost as master andaddressed as slave by general call

DATA A

266 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Slave Transmitter Mode In the Slave Transmitter mode, a number of data bytes are transmitted to a MasterReceiver (see Figure 108). All the status codes mentioned in this section assume thatthe prescaler bits are zero or are masked to zero.

Figure 108. Data Transfer in Slave Transmitter Mode

To initiate the Slave Transmitter mode, TWAR and TWCR must be initialized as follows:

The upper seven bits are the address to which the 2-wire Serial Interface will respondwhen addressed by a Master. If the LSB is set, the TWI will respond to the general calladdress (0x00), otherwise it will ignore the general call address.

TWEN must be written to one to enable the TWI. The TWEA bit must be written to oneto enable the acknowledgement of the device’s own slave address or the general calladdress. TWSTA and TWSTO must be written to zero.

When TWAR and TWCR have been initialized, the TWI waits until it is addressed by itsown slave address (or the general call address if enabled) followed by the data directionbit. If the direction bit is “1” (read), the TWI will operate in ST mode, otherwise SR modeis entered. After its own slave address and the write bit have been received, the TWINTFlag is set and a valid status code can be read from TWSR. The status code is used todetermine the appropriate software action. The appropriate action to be taken for eachstatus code is detailed in Table 119. The Slave Transmitter mode may also be entered ifarbitration is lost while the TWI is in the Master mode (see state 0xB0).

If the TWEA bit is written to zero during a transfer, the TWI will transmit the last byte ofthe transfer. State 0xC0 or state 0xC8 will be entered, depending on whether the MasterReceiver transmits a NACK or ACK after the final byte. The TWI is switched to the notaddressed Slave mode, and will ignore the Master if it continues the transfer. Thus theMaster Receiver receives all “1” as serial data. State 0xC8 is entered if the Masterdemands additional data bytes (by transmitting ACK), even though the Slave has trans-mitted the last byte (TWEA zero and expecting NACK from the Master).

While TWEA is zero, the TWI does not respond to its own slave address. However, the2-wire Serial Bus is still monitored and address recognition may resume at any time bysetting TWEA. This implies that the TWEA bit may be used to temporarily isolate theTWI from the 2-wire Serial Bus.

TWAR TWA6 TWA5 TWA4 TWA3 TWA2 TWA1 TWA0 TWGCEvalue Device’s Own Slave Address

TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN – TWIEvalue 0 1 0 0 0 1 0 X

Device 3 Device n

SDA

SCL

........ R1 R2

VCC

Device 2MASTER

RECEIVER

Device 1SLAVE

TRANSMITTER

267

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

In all sleep modes other than Idle mode, the clock system to the TWI is turned off. If theTWEA bit is set, the interface can still acknowledge its own slave address or the generalcall address by using the 2-wire Serial Bus clock as a clock source. The part will thenwake up from sleep and the TWI will hold the SCL clock will low during the wake up anduntil the TWINT Flag is cleared (by writing it to one). Further data transmission will becarried out as normal, with the AVR clocks running as normal. Observe that if the AVR isset up with a long start-up time, the SCL line may be held low for a long time, blockingother data transmissions.

Note that the 2-wire Serial Interface Data Register – TWDR does not reflect the last bytepresent on the bus when waking up from these sleep modes.

Table 119. Status Codes for Slave Transmitter ModeStatus Code(TWSR)Prescaler Bitsare 0

Status of the 2-wire Serial Bus and2-wire Serial Interface Hardware

Application Software Response

Next Action Taken by TWI HardwareTo/from TWDR

To TWCRSTA STO TWINT TWEA

0xA8 Own SLA+R has been received;ACK has been returned

Load data byte or

Load data byte

X

X

0

0

1

1

0

1

Last data byte will be transmitted and NOT ACK should be receivedData byte will be transmitted and ACK should be re-ceived

0xB0 Arbitration lost in SLA+R/W asMaster; own SLA+R has been received; ACK has been returned

Load data byte or

Load data byte

X

X

0

0

1

1

0

1

Last data byte will be transmitted and NOT ACK should be receivedData byte will be transmitted and ACK should be re-ceived

0xB8 Data byte in TWDR has been transmitted; ACK has been received

Load data byte or

Load data byte

X

X

0

0

1

1

0

1

Last data byte will be transmitted and NOT ACK should be receivedData byte will be transmitted and ACK should be re-ceived

0xC0 Data byte in TWDR has been transmitted; NOT ACK has been received

No TWDR action or

No TWDR action or

No TWDR action or

No TWDR action

0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1

Switched to the not addressed Slave mode;no recognition of own SLA or GCASwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”Switched to the not addressed Slave mode;no recognition of own SLA or GCA;a START condition will be transmitted when the bus becomes freeSwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”;a START condition will be transmitted when the bus becomes free

0xC8 Last data byte in TWDR has beentransmitted (TWEA = “0”); ACKhas been received

No TWDR action or

No TWDR action or

No TWDR action or

No TWDR action

0

0

1

1

0

0

0

0

1

1

1

1

0

1

0

1

Switched to the not addressed Slave mode;no recognition of own SLA or GCASwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”Switched to the not addressed Slave mode;no recognition of own SLA or GCA;a START condition will be transmitted when the bus becomes freeSwitched to the not addressed Slave mode;own SLA will be recognized;GCA will be recognized if TWGCE = “1”;a START condition will be transmitted when the bus becomes free

268 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 109. Formats and States in the Slave Transmitter Mode

Miscellaneous States There are two status codes that do not correspond to a defined TWI state, see Table120.

Status 0xF8 indicates that no relevant information is available because the TWINT Flagis not set. This occurs between other states, and when the TWI is not involved in a serialtransfer.

Status 0x00 indicates that a bus error has occurred during a 2-wire Serial Bus transfer.A bus error occurs when a START or STOP condition occurs at an illegal position in theformat frame. Examples of such illegal positions are during the serial transfer of anaddress byte, a data byte, or an acknowledge bit. When a bus error occurs, TWINT isset. To recover from a bus error, the TWSTO Flag must set and TWINT must be clearedby writing a logic one to it. This causes the TWI to enter the not addressed Slave modeand to clear the TWSTO Flag (no other bits in TWCR are affected). The SDA and SCLlines are released, and no STOP condition is transmitted.

S SLA R A DATA A

$A8 $B8

A

$B0

Reception of the ownslave address and one ormore data bytes

Last data byte transmitted.Switched to not addressedslave (TWEA = '0')

Arbitration lost as masterand addressed as slave

n

From master to slave

From slave to master

Any number of data bytesand their associated acknowledge bits

This number (contained in TWSR) correspondsto a defined state of the Two-Wire Serial Bus. The prescaler bits are zero or masked to zero

P or SDATA

$C0

DATA A

A

$C8

P or SAll 1's

A

Table 120. Miscellaneous StatesStatus Code(TWSR)Prescaler Bitsare 0

Status of the 2-wire Serial Busand 2-wire Serial Interface Hard-ware

Application Software Response

Next Action Taken by TWI HardwareTo/from TWDR

To TWCRSTA STO TWINT TWEA

0xF8 No relevant state informationavailable; TWINT = “0”

No TWDR action No TWCR action Wait or proceed current transfer

0x00 Bus error due to an illegalSTART or STOP condition

No TWDR action 0 1 1 X Only the internal hardware is affected, no STOP condi-tion is sent on the bus. In all cases, the bus is released and TWSTO is cleared.

269

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Combining Several TWI Modes

In some cases, several TWI modes must be combined in order to complete the desiredaction. Consider for example reading data from a serial EEPROM. Typically, such atransfer involves the following steps:

1. The transfer must be initiated.

2. The EEPROM must be instructed what location should be read.

3. The reading must be performed.

4. The transfer must be finished.

Note that data is transmitted both from Master to Slave and vice versa. The Master mustinstruct the Slave what location it wants to read, requiring the use of the MT mode. Sub-sequently, data must be read from the Slave, implying the use of the MR mode. Thus,the transfer direction must be changed. The Master must keep control of the bus duringall these steps, and the steps should be carried out as an atomical operation. If this prin-ciple is violated in a multimaster system, another Master can alter the data pointer in theEEPROM between steps 2 and 3, and the Master will read the wrong data location.Such a change in transfer direction is accomplished by transmitting a REPEATEDSTART between the transmission of the address byte and reception of the data. After aREPEATED START, the Master keeps ownership of the bus. The following figure showsthe flow in this transfer.

Figure 110. Combining Several TWI Modes to Access a Serial EEPROM

Multi-master Systems and Arbitration

If multiple masters are connected to the same bus, transmissions may be initiated simul-taneously by one or more of them. The TWI standard ensures that such situations arehandled in such a way that one of the masters will be allowed to proceed with the trans-fer, and that no data will be lost in the process. An example of an arbitration situation isdepicted below, where two masters are trying to transmit data to a Slave Receiver.

Figure 111. An Arbitration Example

Several different scenarios may arise during arbitration, as described below:

Master Transmitter Master Receiver

S = START Rs = REPEATED START P = STOP

Transmitted from master to slave Transmitted from slave to master

S SLA+W A ADDRESS A Rs SLA+R A DATA A P

Device 1MASTER

TRANSMITTER

Device 2MASTER

TRANSMITTER

Device 3SLAVE

RECEIVERDevice n

SDA

SCL

........ R1 R2

VCC

270 ATmega640/1280/1281/2560/25612549A–AVR–03/05

• Two or more masters are performing identical communication with the same Slave. In this case, neither the Slave nor any of the masters will know about the bus contention.

• Two or more masters are accessing the same Slave with different data or direction bit. In this case, arbitration will occur, either in the READ/WRITE bit or in the data bits. The masters trying to output a one on SDA while another Master outputs a zero will lose the arbitration. Losing masters will switch to not addressed Slave mode or wait until the bus is free and transmit a new START condition, depending on application software action.

• Two or more masters are accessing different slaves. In this case, arbitration will occur in the SLA bits. Masters trying to output a one on SDA while another Master outputs a zero will lose the arbitration. Masters losing arbitration in SLA will switch to Slave mode to check if they are being addressed by the winning Master. If addressed, they will switch to SR or ST mode, depending on the value of the READ/WRITE bit. If they are not being addressed, they will switch to not addressed Slave mode or wait until the bus is free and transmit a new START condition, depending on application software action.

This is summarized in Figure 112. Possible status values are given in circles.

Figure 112. Possible Status Codes Caused by Arbitration

OwnAddress / General Call

received

Arbitration lost in SLA

TWI bus will be released and not addressed slave mode will be enteredA START condition will be transmitted when the bus becomes free

No

Arbitration lost in Data

Direction

Yes

Write Data byte will be received and NOT ACK will be returnedData byte will be received and ACK will be returned

Last data byte will be transmitted and NOT ACK should be receivedData byte will be transmitted and ACK should be received

ReadB0

68/78

38

SLASTART Data STOP

271

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Analog Comparator The Analog Comparator compares the input values on the positive pin AIN0 and nega-tive pin AIN1. When the voltage on the positive pin AIN0 is higher than the voltage onthe negative pin AIN1, the Analog Comparator output, ACO, is set. The comparator’soutput can be set to trigger the Timer/Counter1 Input Capture function. In addition, thecomparator can trigger a separate interrupt, exclusive to the Analog Comparator. Theuser can select Interrupt triggering on comparator output rise, fall or toggle. A block dia-gram of the comparator and its surrounding logic is shown in Figure 113.

The Power Reduction ADC bit, PRADC, in “Power Reduction Register 0 - PRR0” onpage 54 must be disabled by writing a logical zero to be able to use the ADC input MUX.

Figure 113. Analog Comparator Block Diagram(2)

Notes: 1. See Table 122 on page 273.2. Refer to Figure 1 on page 2 and Table 38 on page 89 for Analog Comparator pin

placement.

ADC Control and Status Register B – ADCSRB

• Bit 6 – ACME: Analog Comparator Multiplexer Enable

When this bit is written logic one and the ADC is switched off (ADEN in ADCSRA iszero), the ADC multiplexer selects the negative input to the Analog Comparator. Whenthis bit is written logic zero, AIN1 is applied to the negative input of the Analog Compar-ator. For a detailed description of this bit, see “Analog Comparator Multiplexed Input” onpage 273.

Analog Comparator Control and Status Register – ACSR

• Bit 7 – ACD: Analog Comparator Disable

When this bit is written logic one, the power to the Analog Comparator is switched off.This bit can be set at any time to turn off the Analog Comparator. This will reduce power

ACBG

BANDGAPREFERENCE

ADC MULTIPLEXEROUTPUT

ACMEADEN

(1)

Bit 7 6 5 4 3 2 1 0

– ACME – – MUX5 ADTS2 ADTS1 ADTS0 ADCSRB

Read/Write R R/W R R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

ACD ACBG ACO ACI ACIE ACIC ACIS1 ACIS0 ACSR

Read/Write R/W R/W R R/W R/W R/W R/W R/W

Initial Value 0 0 N/A 0 0 0 0 0

272 ATmega640/1280/1281/2560/25612549A–AVR–03/05

consumption in Active and Idle mode. When changing the ACD bit, the Analog Compar-ator Interrupt must be disabled by clearing the ACIE bit in ACSR. Otherwise an interruptcan occur when the bit is changed.

• Bit 6 – ACBG: Analog Comparator Bandgap Select

When this bit is set, a fixed bandgap reference voltage replaces the positive input to theAnalog Comparator. When this bit is cleared, AIN0 is applied to the positive input of theAnalog Comparator. See “Internal Voltage Reference” on page 62.

• Bit 5 – ACO: Analog Comparator Output

The output of the Analog Comparator is synchronized and then directly connected toACO. The synchronization introduces a delay of 1 - 2 clock cycles.

• Bit 4 – ACI: Analog Comparator Interrupt Flag

This bit is set by hardware when a comparator output event triggers the interrupt modedefined by ACIS1 and ACIS0. The Analog Comparator interrupt routine is executed ifthe ACIE bit is set and the I-bit in SREG is set. ACI is cleared by hardware when execut-ing the corresponding interrupt handling vector. Alternatively, ACI is cleared by writing alogic one to the flag.

• Bit 3 – ACIE: Analog Comparator Interrupt Enable

When the ACIE bit is written logic one and the I-bit in the Status Register is set, the Ana-log Comparator interrupt is activated. When written logic zero, the interrupt is disabled.

• Bit 2 – ACIC: Analog Comparator Input Capture Enable

When written logic one, this bit enables the input capture function in Timer/Counter1 tobe triggered by the Analog Comparator. The comparator output is in this case directlyconnected to the input capture front-end logic, making the comparator utilize the noisecanceler and edge select features of the Timer/Counter1 Input Capture interrupt. Whenwritten logic zero, no connection between the Analog Comparator and the input capturefunction exists. To make the comparator trigger the Timer/Counter1 Input Capture inter-rupt, the ICIE1 bit in the Timer Interrupt Mask Register (TIMSK1) must be set.

• Bits 1, 0 – ACIS1, ACIS0: Analog Comparator Interrupt Mode Select

These bits determine which comparator events that trigger the Analog Comparator inter-rupt. The different settings are shown in Table 121.

When changing the ACIS1/ACIS0 bits, the Analog Comparator Interrupt must be dis-abled by clearing its Interrupt Enable bit in the ACSR Register. Otherwise an interruptcan occur when the bits are changed.

Table 121. ACIS1/ACIS0 Settings

ACIS1 ACIS0 Interrupt Mode

0 0 Comparator Interrupt on Output Toggle.

0 1 Reserved

1 0 Comparator Interrupt on Falling Output Edge.

1 1 Comparator Interrupt on Rising Output Edge.

273

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Analog Comparator Multiplexed Input

It is possible to select any of the ADC15..0 pins to replace the negative input to the Ana-log Comparator. The ADC multiplexer is used to select this input, and consequently, theADC must be switched off to utilize this feature. If the Analog Comparator MultiplexerEnable bit (ACME in ADCSRB) is set and the ADC is switched off (ADEN in ADCSRA iszero), MUX5 and MUX2..0 in ADMUX select the input pin to replace the negative inputto the Analog Comparator, as shown in Table 122. If ACME is cleared or ADEN is set,AIN1 is applied to the negative input to the Analog Comparator.

Digital Input Disable Register 1 – DIDR1

• Bit 1, 0 – AIN1D, AIN0D: AIN1, AIN0 Digital Input Disable

When this bit is written logic one, the digital input buffer on the AIN1/0 pin is disabled.The corresponding PIN Register bit will always read as zero when this bit is set. Whenan analog signal is applied to the AIN1/0 pin and the digital input from this pin is notneeded, this bit should be written logic one to reduce power consumption in the digitalinput buffer.

Table 122. Analog Comparator Mulitiplexed Input

ACME ADEN MUX5 MUX2..0 Analog Comparator Negative Input

0 x x xxx AIN1

1 1 x xxx AIN1

1 0 0 000 ADC0

1 0 0 001 ADC1

1 0 0 010 ADC2

1 0 0 011 ADC3

1 0 0 100 ADC4

1 0 0 101 ADC5

1 0 0 110 ADC6

1 0 0 111 ADC7

1 0 1 000 ADC8

1 0 1 001 ADC9

1 0 1 010 ADC10

1 0 1 011 ADC11

1 0 1 100 ADC12

1 0 1 101 ADC13

1 0 1 110 ADC14

1 0 1 111 ADC15

Bit 7 6 5 4 3 2 1 0

– – – – – – AIN1D AIN0D DIDR1

Read/Write R R R R R R R/W R/W

Initial Value 0 0 0 0 0 0 0 0

274 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Analog to Digital Converter

Features • 10-bit Resolution• 0.5 LSB Integral Non-linearity• ± 2 LSB Absolute Accuracy• 65 - 260 µs Conversion Time• Up to 15 kSPS at Maximum Resolution• 16 Multiplexed Single Ended Input Channels• 14 Differential input channels• 4 Differential Input Channels with Optional Gain of 10x and 200x• Optional Left Adjustment for ADC Result Readout• 0 - VCC ADC Input Voltage Range• Selectable 2.56V or 1.1V ADC Reference Voltage• Free Running or Single Conversion Mode• Interrupt on ADC Conversion Complete• Sleep Mode Noise Canceler

The ATmega640/1280/1281/2560/2561 features a 10-bit successive approximationADC. The ADC is connected to an 8-channel Analog Multiplexer which allows eight sin-gle-ended voltage inputs constructed from the pins of Port A. The single-ended voltageinputs refer to 0V (GND).

The device also supports 16 differential voltage input combinations. Four of the differen-tial inputs (ADC1 & ADC0, ADC & ADC2, ADC9 & ADC8 and ADC11 & ADC10) areequipped with a programmable gain stage, providing amplification steps of 0dB (1x),20dB (10x) or 46dB (200x) on the differential input voltage before the ADC conversion.The 16 channels are split in two sections of 8 channels where in each section seven dif-ferential analog input channels share a common negative terminal (ADC1/ADC9), whileany other ADC input in that section can be selected as the positive input terminal. If 1xor 10x gain is used, 8 bit resolution can be expected. If 200x gain is used, 7 bit resolu-tion can be expected.

The ADC contains a Sample and Hold circuit which ensures that the input voltage to theADC is held at a constant level during conversion. A block diagram of the ADC is shownin Figure 114.

The ADC has a separate analog supply voltage pin, AVCC. AVCC must not differ morethan ± 0.3V from VCC. See the paragraph “ADC Noise Canceler” on page 281 on how toconnect this pin.

Internal reference voltages of nominally 1.1V, 2.56V or AVCC are provided On-chip.The voltage reference may be externally decoupled at the AREF pin by a capacitor forbetter noise performance.

The Power Reduction ADC bit, PRADC, in “Power Reduction Register 0 - PRR0” onpage 54 must be disabled by writing a logical zero to enable the ADC.

275

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 114. Analog to Digital Converter Block Schematic

Operation The ADC converts an analog input voltage to a 10-bit digital value through successiveapproximation. The minimum value represents GND and the maximum value representsthe voltage on the AREF pin minus 1 LSB. Optionally, AVCC or an internal 1.1V refer-ence voltage may be connected to the AREF pin by writing to the REFSn bits in theADMUX Register. The internal voltage reference may thus be decoupled by an externalcapacitor at the AREF pin to improve noise immunity.

The analog input channel is selected by writing to the MUX bits in ADMUX. Any of theADC input pins, as well as GND and a fixed bandgap voltage reference, can be selectedas single ended inputs to the ADC. A selection of ADC input pins can be selected aspositive and negative inputs to the differential amplifier.

If differential channels are selected, the voltage difference between the selected inputchannel pair then becomes the analog input to the ADC. If single ended channels areused, the amplifier is bypassed altogether.

The ADC is enabled by setting the ADC Enable bit, ADEN in ADCSRA. Voltage refer-ence and input channel selections will not go into effect until ADEN is set. The ADCdoes not consume power when ADEN is cleared, so it is recommended to switch off theADC before entering power saving sleep modes.

ADC CONVERSIONCOMPLETE IRQ

8-BIT DATABUS

15 0

MU

X2

AD

IE

AD

FR

AD

SC

AD

EN

AD

IFA

DIF

MU

X1

MU

X0

AD

PS

0

AD

PS

1

AD

PS

2

MU

X3

POSINPUTMUX

10-bit DAC

+

-

SAMPLE & HOLDCOMPARATOR

INTERNALREFERENCE(1.1V/2.56V)

MU

X4

AVCC

ADC7

ADC6

ADC5

ADC4

ADC3

ADC2

ADC1

ADC0

RE

FS

[1:0

]

AD

LAR

+

-

CH

AN

NE

L S

ELE

CT

ION

AD

C[9

:0]

ADCMULTIPLEXEROUTPUT

GAINAMPLIFIER

AREF

BANDGAP (1.1V)REFERENCE

SINGLE ENDED / DIFFERENTIAL SELECTION

GND

MU

X5

CONVERSION LOGIC

ADC CTRL & STATUSREGISTER B (ADCSRB)

ADC CTRL & STATUSREGISTER A (ADCSRA)

PRESCALER

ADC MULTIPLEXER SELECT(ADMUX)

MUX DECODER

GA

IN S

ELE

CT

ION

ADC DATA REGISTER(ADCH/ADCL)

NEG INPUTMUX

ADC15

ADC14

ADC13

ADC12

ADC11

ADC10

ADC9

ADC8

TRIGGERSELECT

START

INTERRUPTFLAGS

ADTS[2:0]

276 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The ADC generates a 10-bit result which is presented in the ADC Data Registers,ADCH and ADCL. By default, the result is presented right adjusted, but can optionallybe presented left adjusted by setting the ADLAR bit in ADMUX.

If the result is left adjusted and no more than 8-bit precision is required, it is sufficient toread ADCH. Otherwise, ADCL must be read first, then ADCH, to ensure that the contentof the Data Registers belongs to the same conversion. Once ADCL is read, ADC accessto Data Registers is blocked. This means that if ADCL has been read, and a conversioncompletes before ADCH is read, neither register is updated and the result from the con-version is lost. When ADCH is read, ADC access to the ADCH and ADCL Registers isre-enabled.

The ADC has its own interrupt which can be triggered when a conversion completes.When ADC access to the Data Registers is prohibited between reading of ADCH andADCL, the interrupt will trigger even if the result is lost.

Starting a Conversion A single conversion is started by writing a logical one to the ADC Start Conversion bit,ADSC. This bit stays high as long as the conversion is in progress and will be cleared byhardware when the conversion is completed. If a different data channel is selected whilea conversion is in progress, the ADC will finish the current conversion before performingthe channel change.

Alternatively, a conversion can be triggered automatically by various sources. Auto Trig-gering is enabled by setting the ADC Auto Trigger Enable bit, ADATE in ADCSRA. Thetrigger source is selected by setting the ADC Trigger Select bits, ADTS in ADCSRB(See description of the ADTS bits for a list of the trigger sources). When a positive edgeoccurs on the selected trigger signal, the ADC prescaler is reset and a conversion isstarted. This provides a method of starting conversions at fixed intervals. If the triggersignal still is set when the conversion completes, a new conversion will not be started. Ifanother positive edge occurs on the trigger signal during conversion, the edge will beignored. Note that an Interrupt Flag will be set even if the specific interrupt is disabled orthe Global Interrupt Enable bit in SREG is cleared. A conversion can thus be triggeredwithout causing an interrupt. However, the Interrupt Flag must be cleared in order to trig-ger a new conversion at the next interrupt event.

Figure 115. ADC Auto Trigger Logic

Using the ADC Interrupt Flag as a trigger source makes the ADC start a new conversionas soon as the ongoing conversion has finished. The ADC then operates in Free Run-ning mode, constantly sampling and updating the ADC Data Register. The first

ADSC

ADIF

SOURCE 1

SOURCE n

ADTS[2:0]

CONVERSIONLOGIC

PRESCALER

START CLKADC

.

.

.

. EDGEDETECTOR

ADATE

277

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

conversion must be started by writing a logical one to the ADSC bit in ADCSRA. In thismode the ADC will perform successive conversions independently of whether the ADCInterrupt Flag, ADIF is cleared or not.

If Auto Triggering is enabled, single conversions can be started by writing ADSC inADCSRA to one. ADSC can also be used to determine if a conversion is in progress.The ADSC bit will be read as one during a conversion, independently of how the conver-sion was started.

Prescaling and Conversion Timing

Figure 116. ADC Prescaler

By default, the successive approximation circuitry requires an input clock frequencybetween 50 kHz and 200 kHz to get maximum resolution. If a lower resolution than 10bits is needed, the input clock frequency to the ADC can be higher than 200 kHz to get ahigher sample rate.

The ADC module contains a prescaler, which generates an acceptable ADC clock fre-quency from any CPU frequency above 100 kHz. The prescaling is set by the ADPS bitsin ADCSRA. The prescaler starts counting from the moment the ADC is switched on bysetting the ADEN bit in ADCSRA. The prescaler keeps running for as long as the ADENbit is set, and is continuously reset when ADEN is low.

When initiating a single ended conversion by setting the ADSC bit in ADCSRA, the con-version starts at the following rising edge of the ADC clock cycle.

A normal conversion takes 13 ADC clock cycles. The first conversion after the ADC isswitched on (ADEN in ADCSRA is set) takes 25 ADC clock cycles in order to initializethe analog circuitry.

The actual sample-and-hold takes place 1.5 ADC clock cycles after the start of a normalconversion and 13.5 ADC clock cycles after the start of an first conversion. When a con-version is complete, the result is written to the ADC Data Registers, and ADIF is set. InSingle Conversion mode, ADSC is cleared simultaneously. The software may then setADSC again, and a new conversion will be initiated on the first rising ADC clock edge.

When Auto Triggering is used, the prescaler is reset when the trigger event occurs. Thisassures a fixed delay from the trigger event to the start of conversion. In this mode, thesample-and-hold takes place two ADC clock cycles after the rising edge on the triggersource signal. Three additional CPU clock cycles are used for synchronization logic.

7-BIT ADC PRESCALER

ADC CLOCK SOURCE

CK

ADPS0ADPS1ADPS2

CK

/128

CK

/2

CK

/4

CK

/8

CK

/16

CK

/32

CK

/64

ResetADENSTART

278 ATmega640/1280/1281/2560/25612549A–AVR–03/05

In Free Running mode, a new conversion will be started immediately after the conver-sion completes, while ADSC remains high. For a summary of conversion times, seeTable 123.

Figure 117. ADC Timing Diagram, First Conversion (Single Conversion Mode)

Figure 118. ADC Timing Diagram, Single Conversion

Figure 119. ADC Timing Diagram, Auto Triggered Conversion

Sign and MSB of Result

LSB of Result

ADC Clock

ADSC

Sample & Hold

ADIF

ADCH

ADCL

Cycle Number

ADEN

1 2 12 13 14 15 16 17 18 19 20 21 22 23 24 25 1 2

First ConversionNextConversion

3

MUX and REFSUpdate

MUX and REFSUpdate

ConversionComplete

1 2 3 4 5 6 7 8 9 10 11 12 13

Sign and MSB of Result

LSB of Result

ADC Clock

ADSC

ADIF

ADCH

ADCL

Cycle Number 1 2

One Conversion Next Conversion

3

Sample & Hold

MUX and REFSUpdate

ConversionComplete

MUX and REFSUpdate

1 2 3 4 5 6 7 8 9 10 11 12 13

Sign and MSB of Result

LSB of Result

ADC Clock

TriggerSource

ADIF

ADCH

ADCL

Cycle Number 1 2

One Conversion Next Conversion

ConversionCompletePrescaler

Reset

ADATE

PrescalerReset

Sample &Hold

MUX and REFS Update

279

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 120. ADC Timing Diagram, Free Running Conversion

Differential Channels When using differential channels, certain aspects of the conversion need to be takeninto consideration.

Differential conversions are synchronized to the internal clock CKADC2 equal to half theADC clock. This synchronization is done automatically by the ADC interface in such away that the sample-and-hold occurs at a specific phase of CKADC2. A conversion initi-ated by the user (i.e., all single conversions, and the first free running conversion) whenCKADC2 is low will take the same amount of time as a single ended conversion (13 ADCclock cycles from the next prescaled clock cycle). A conversion initiated by the userwhen CKADC2 is high will take 14 ADC clock cycles due to the synchronization mecha-nism. In Free Running mode, a new conversion is initiated immediately after theprevious conversion completes, and since CKADC2 is high at this time, all automaticallystarted (i.e., all but the first) Free Running conversions will take 14 ADC clock cycles.

If differential channels are used and conversions are started by Auto Triggering, theADC must be switched off between conversions. When Auto Triggering is used, theADC prescaler is reset before the conversion is started. Since the stage is dependent ofa stable ADC clock prior to the conversion, this conversion will not be valid. By disablingand then re-enabling the ADC between each conversion (writing ADEN in ADCSRA to“0” then to “1”), only extended conversions are performed. The result from the extendedconversions will be valid. See “Prescaling and Conversion Timing” on page 277 for tim-ing details.

Table 123. ADC Conversion Time

ConditionSample & Hold (Cycles from Start of Conversion)

Conversion Time (Cycles)

First conversion 13.5 25

Normal conversions, single ended 1.5 13

Auto Triggered conversions 2 13.5

Normal conversions, differential 1.5/2.5 13/14

11 12 13

Sign and MSB of Result

LSB of Result

ADC Clock

ADSC

ADIF

ADCH

ADCL

Cycle Number1 2

One Conversion Next Conversion

3 4

ConversionComplete

Sample & Hold

MUX and REFSUpdate

280 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Changing Channel or Reference Selection

The MUXn and REFS1:0 bits in the ADMUX Register are single buffered through a tem-porary register to which the CPU has random access. This ensures that the channelsand reference selection only takes place at a safe point during the conversion. Thechannel and reference selection is continuously updated until a conversion is started.Once the conversion starts, the channel and reference selection is locked to ensure asufficient sampling time for the ADC. Continuous updating resumes in the last ADCclock cycle before the conversion completes (ADIF in ADCSRA is set). Note that theconversion starts on the following rising ADC clock edge after ADSC is written. The useris thus advised not to write new channel or reference selection values to ADMUX untilone ADC clock cycle after ADSC is written.

If Auto Triggering is used, the exact time of the triggering event can be indeterministic.Special care must be taken when updating the ADMUX Register, in order to controlwhich conversion will be affected by the new settings.

If both ADATE and ADEN is written to one, an interrupt event can occur at any time. Ifthe ADMUX Register is changed in this period, the user cannot tell if the next conversionis based on the old or the new settings. ADMUX can be safely updated in the followingways:

1. When ADATE or ADEN is cleared.

2. During conversion, minimum one ADC clock cycle after the trigger event.

3. After a conversion, before the Interrupt Flag used as trigger source is cleared.

When updating ADMUX in one of these conditions, the new settings will affect the nextADC conversion.

Special care should be taken when changing differential channels. Once a differentialchannel has been selected, the stage may take as much as TBD µs to stabilize to thenew value. Thus conversions should not be started within the first TBD µs after selectinga new differential channel. Alternatively, conversion results obtained within this periodshould be discarded.

The same settling time should be observed for the first differential conversion afterchanging ADC reference (by changing the REFS1:0 bits in ADMUX).

ADC Input Channels When changing channel selections, the user should observe the following guidelines toensure that the correct channel is selected:

In Single Conversion mode, always select the channel before starting the conversion.The channel selection may be changed one ADC clock cycle after writing one to ADSC.However, the simplest method is to wait for the conversion to complete before changingthe channel selection.

In Free Running mode, always select the channel before starting the first conversion.The channel selection may be changed one ADC clock cycle after writing one to ADSC.However, the simplest method is to wait for the first conversion to complete, and thenchange the channel selection. Since the next conversion has already started automati-cally, the next result will reflect the previous channel selection. Subsequent conversionswill reflect the new channel selection.

When switching to a differential gain channel, the first conversion result may have apoor accuracy due to the required settling time for the automatic offset cancellation cir-cuitry. The user should preferably disregard the first conversion result.

281

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

ADC Voltage Reference The reference voltage for the ADC (VREF) indicates the conversion range for the ADC.Single ended channels that exceed VREF will result in codes close to 0x3FF. VREF can beselected as either AVCC, internal 1.1V reference, internal 2.56V reference or externalAREF pin.

AVCC is connected to the ADC through a passive switch. The internal 1.1V reference isgenerated from the internal bandgap reference (VBG) through an internal amplifier. Ineither case, the external AREF pin is directly connected to the ADC, and the referencevoltage can be made more immune to noise by connecting a capacitor between theAREF pin and ground. VREF can also be measured at the AREF pin with a high impedantvoltmeter. Note that VREF is a high impedant source, and only a capacitive load shouldbe connected in a system. The Internal 2.56V reference is generated from the 1.1Vreference.

If the user has a fixed voltage source connected to the AREF pin, the user may not usethe other reference voltage options in the application, as they will be shorted to theexternal voltage. If no external voltage is applied to the AREF pin, the user may switchbetween AVCC, 1.1V and 2.56V as reference selection. The first ADC conversion resultafter switching reference voltage source may be inaccurate, and the user is advised todiscard this result.

If differential channels are used, the selected reference should not be closer to AVCCthan indicated in “ADC Characteristics – Preliminary Data” on page 374.

ADC Noise Canceler The ADC features a noise canceler that enables conversion during sleep mode toreduce noise induced from the CPU core and other I/O peripherals. The noise cancelercan be used with ADC Noise Reduction and Idle mode. To make use of this feature, thefollowing procedure should be used:

1. Make sure that the ADC is enabled and is not busy converting. Single Con-version mode must be selected and the ADC conversion complete interrupt must be enabled.

2. Enter ADC Noise Reduction mode (or Idle mode). The ADC will start a con-version once the CPU has been halted.

3. If no other interrupts occur before the ADC conversion completes, the ADC interrupt will wake up the CPU and execute the ADC Conversion Complete interrupt routine. If another interrupt wakes up the CPU before the ADC con-version is complete, that interrupt will be executed, and an ADC Conversion Complete interrupt request will be generated when the ADC conversion completes. The CPU will remain in active mode until a new sleep command is executed.

Note that the ADC will not be automatically turned off when entering other sleep modesthan Idle mode and ADC Noise Reduction mode. The user is advised to write zero toADEN before entering such sleep modes to avoid excessive power consumption.

If the ADC is enabled in such sleep modes and the user wants to perform differentialconversions, the user is advised to switch the ADC off and on after waking up fromsleep to prompt an extended conversion to get a valid result.

Analog Input Circuitry The analog input circuitry for single ended channels is illustrated in Figure 121. An ana-log source applied to ADCn is subjected to the pin capacitance and input leakage of thatpin, regardless of whether that channel is selected as input for the ADC. When the chan-nel is selected, the source must drive the S/H capacitor through the series resistance(combined resistance in the input path).

282 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The ADC is optimized for analog signals with an output impedance of approximately10 kΩ or less. If such a source is used, the sampling time will be negligible. If a sourcewith higher impedance is used, the sampling time will depend on how long time thesource needs to charge the S/H capacitor, with can vary widely. The user is recom-mended to only use low impedant sources with slowly varying signals, since thisminimizes the required charge transfer to the S/H capacitor.

Signal components higher than the Nyquist frequency (fADC/2) should not be present foreither kind of channels, to avoid distortion from unpredictable signal convolution. Theuser is advised to remove high frequency components with a low-pass filter beforeapplying the signals as inputs to the ADC.

Figure 121. Analog Input Circuitry

Analog Noise Canceling Techniques

Digital circuitry inside and outside the device generates EMI which might affect theaccuracy of analog measurements. If conversion accuracy is critical, the noise level canbe reduced by applying the following techniques:

1. Keep analog signal paths as short as possible. Make sure analog tracks run over the analog ground plane, and keep them well away from high-speed switching digital tracks.

2. The AVCC pin on the device should be connected to the digital VCC supply voltage via an LC network as shown in Figure 122.

3. Use the ADC noise canceler function to reduce induced noise from the CPU.

4. If any ADC port pins are used as digital outputs, it is essential that these do not switch while a conversion is in progress.

ADCn

IIH

1..100 kΩCS/H= 14 pF

VCC/2

IIL

283

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 122. ADC Power Connections, ATmega1281/2561.

Figure 123. ADC Power Connections, ATmega640/1280/2560

VCC

GND

100nF

Analog Ground Plane

(ADC0) PF0

(ADC7) PF7

(ADC1) PF1

(ADC2) PF2

(ADC3) PF3

(ADC4) PF4

(ADC5) PF5

(ADC6) PF6

AREF

GND

AVCC

52

53

54

55

56

57

58

59

60

6161

6262

6363

6464

1

51

PG

5

PA0

10µΗ

100nF

Analog Ground Plane

100

(OC

0B)

PG

5

10µΗ

79

80

81

82

83

84

85

86

87

88

89

90

91

92

93

94

95

96

97

98

99

PJ7

VCC

GND

(ADC15/PCINT23) PK7

(ADC14/PCINT22) PK6

(ADC13/PCINT21) PK5

(ADC12/PCINT20) PK4

(ADC11/PCINT19) PK3

(ADC10/PCINT18) PK2

(ADC9/PCINT17) PK1

(ADC8/PCINT16) PK0

(ADC7/TDI) PF7

(ADC6/TDO) PF6

(ADC5/TMS) PF5

(ADC4/TCK) PF4

(ADC3) PF3

(ADC2) PF2

(ADC1) PF1

(ADC0) PF0

AREF

GND

AVCC

284 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Offset Compensation Schemes

The stage has a built-in offset cancellation circuitry that nulls the offset of differentialmeasurements as much as possible. The remaining offset in the analog path can bemeasured directly by selecting the same channel for both differential inputs. This offsetresidue can be then subtracted in software from the measurement results. Using thiskind of software based offset correction, offset on any channel can be reduced belowone LSB.

ADC Accuracy Definitions An n-bit single-ended ADC converts a voltage linearly between GND and VREF in 2n

steps (LSBs). The lowest code is read as 0, and the highest code is read as 2n-1.

Several parameters describe the deviation from the ideal behavior:

• Offset: The deviation of the first transition (0x000 to 0x001) compared to the ideal transition (at 0.5 LSB). Ideal value: 0 LSB.

Figure 124. Offset Error

• Gain Error: After adjusting for offset, the Gain Error is found as the deviation of the last transition (0x3FE to 0x3FF) compared to the ideal transition (at 1.5 LSB below maximum). Ideal value: 0 LSB

Figure 125. Gain Error

Output Code

VREF Input Voltage

Ideal ADC

Actual ADC

OffsetError

Output Code

VREF Input Voltage

Ideal ADC

Actual ADC

GainError

285

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• Integral Non-linearity (INL): After adjusting for offset and gain error, the INL is the maximum deviation of an actual transition compared to an ideal transition for any code. Ideal value: 0 LSB.

Figure 126. Integral Non-linearity (INL)

• Differential Non-linearity (DNL): The maximum deviation of the actual code width (the interval between two adjacent transitions) from the ideal code width (1 LSB). Ideal value: 0 LSB.

Figure 127. Differential Non-linearity (DNL)

• Quantization Error: Due to the quantization of the input voltage into a finite number of codes, a range of input voltages (1 LSB wide) will code to the same value. Always ± 0.5 LSB.

• Absolute Accuracy: The maximum deviation of an actual (unadjusted) transition compared to an ideal transition for any code. This is the compound effect of offset, gain error, differential error, non-linearity, and quantization error. Ideal value: ± 0.5 LSB.

Output Code

VREF Input Voltage

Ideal ADC

Actual ADC

INL

Output Code

0x3FF

0x000

0 VREF Input Voltage

DNL

1 LSB

286 ATmega640/1280/1281/2560/25612549A–AVR–03/05

ADC Conversion Result

After the conversion is complete (ADIF is high), the conversion result can be found inthe ADC Result Registers (ADCL, ADCH).

For single ended conversion, the result is

where VIN is the voltage on the selected input pin and VREF the selected voltage refer-ence (see Table 125 on page 287 and Table 126 on page 288). 0x000 representsanalog ground, and 0x3FF represents the selected reference voltage minus one LSB.

If differential channels are used, the result is

where VPOS is the voltage on the positive input pin, VNEG the voltage on the negativeinput pin, and VREF the selected voltage reference. The result is presented in two’s com-plement form, from 0x200 (-512d) through 0x1FF (+511d). Note that if the user wants toperform a quick polarity check of the result, it is sufficient to read the MSB of the result(ADC9 in ADCH). If the bit is one, the result is negative, and if this bit is zero, the resultis positive. Figure 128 shows the decoding of the differential input range.

Table 124 shows the resulting output codes if the differential input channel pair (ADCn -ADCm) is selected with a gain of GAIN and a reference voltage of VREF.

Figure 128. Differential Measurement Range

ADCVIN 1024⋅

VREF--------------------------=

ADCVPOS VNEG–( ) 512⋅

VREF-----------------------------------------------------=

0

Output Code

0x1FF

0x000

VREFDifferential InputVoltage (Volts)

0x3FF

0x200

- VREF

287

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Example:

ADMUX = 0xFB (ADC3 - ADC2, 10x gain, 2.56V reference, left adjusted result)

Voltage on ADC3 is 300 mV, voltage on ADC2 is 500 mV.

ADCR = 512 * 10 * (300 - 500) / 2560 = -400 = 0x270.

ADCL will thus read 0x00, and ADCH will read 0x9C. Writing zero to ADLAR rightadjusts the result: ADCL = 0x70, ADCH = 0x02.

ADC Multiplexer Selection Register – ADMUX

• Bit 7:6 – REFS1:0: Reference Selection Bits

These bits select the voltage reference for the ADC, as shown in Table 125. If these bitsare changed during a conversion, the change will not go in effect until this conversion iscomplete (ADIF in ADCSRA is set). The internal voltage reference options may not beused if an external reference voltage is being applied to the AREF pin.

Note: 1. If 10x or 200x gain is selected, only 2.56 V should be used as Internal VoltageReference.

• Bit 5 – ADLAR: ADC Left Adjust Result

The ADLAR bit affects the presentation of the ADC conversion result in the ADC DataRegister. Write one to ADLAR to left adjust the result. Otherwise, the result is right

Table 124. Correlation Between Input Voltage and Output Codes

VADCn Read Code Corresponding Decimal Value

VADCm + VREF / GAIN 0x1FF 511

VADCm + 0.999 VREF / GAIN 0x1FF 511

VADCm + 0.998 VREF / GAIN 0x1FE 510

... ... ...

VADCm + 0.001 VREF / GAIN 0x001 1

VADCm 0x000 0

VADCm - 0.001 VREF / GAIN 0x3FF -1

... ... ...

VADCm - 0.999 VREF / GAIN 0x201 -511

VADCm - VREF / GAIN 0x200 -512

Bit 7 6 5 4 3 2 1 0

REFS1 REFS0 ADLAR MUX4 MUX3 MUX2 MUX1 MUX0 ADMUX

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 125. Voltage Reference Selections for ADC

REFS1 REFS0 Voltage Reference Selection(1)

0 0 AREF, Internal VREF turned off

0 1 AVCC with external capacitor at AREF pin

1 0 Internal 1.1V Voltage Reference with external capacitor at AREF pin

1 1 Internal 2.56V Voltage Reference with external capacitor at AREF pin

288 ATmega640/1280/1281/2560/25612549A–AVR–03/05

adjusted. Changing the ADLAR bit will affect the ADC Data Register immediately,regardless of any ongoing conversions. For a complete description of this bit, see “TheADC Data Register – ADCL and ADCH” on page 292.

• Bits 4:0 – MUX4:0: Analog Channel and Gain Selection Bits

The value of these bits selects which combination of analog inputs are connected to theADC. See Table 126 for details. If these bits are changed during a conversion, thechange will not go in effect until this conversion is complete (ADIF in ADCSRA is set)

ADC Control and Status Register B – ADCSRB

• Bit 3 – MUX5: Analog Channel and Gain Selection Bit

This bit is used together with MUX4:0 in ADMUX to select which combination in of ana-log inputs are connected to the ADC. See Table 126 for details. If this bit is changedduring a conversion, the change will not go in effect until this conversion is complete(ADIF in ADCSRA is set).

This bit can only be used in ATmega640/1280/2560.

Bit 7 6 5 4 3 2 1 0

– ACME – – MUX5 ADTS2 ADTS1 ADTS0 ADCSRB

Read/Write R R/W R R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 126. Input Channel Selections

MUX5..0Single Ended Input

Positive Differential Input

Negative Differential Input Gain

000000 ADC0

N/A

000001 ADC1

000010 ADC2

000011 ADC3

000100 ADC4

000101 ADC5

000110 ADC6

000111 ADC7

001000(1)

N/A

ADC0 ADC0 10x

001001(1) ADC1 ADC0 10x

001010(1) ADC0 ADC0 200x

001011(1) ADC1 ADC0 200x

001100(1) ADC2 ADC2 10x

001101(1) ADC3 ADC2 10x

001110(1) ADC2 ADC2 200x

001111(1) ADC3 ADC2 200x

010000 ADC0 ADC1 1x

010001 ADC1 ADC1 1x

010010 ADC2 ADC1 1x

289

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

010011

N/A

ADC3 ADC1 1x

010100 ADC4 ADC1 1x

010101 ADC5 ADC1 1x

010110 ADC6 ADC1 1x

010111 ADC7 ADC1 1x

011000 ADC0 ADC2 1x

011001 ADC1 ADC2 1x

011010 ADC2 ADC2 1x

011011 ADC3 ADC2 1x

011100 ADC4 ADC2 1x

011101 ADC5 ADC2 1x

011110 1.1V (VBG)N/A

011111 0V (GND)

100000 ADC8

N/A

100001 ADC9

100010 ADC10

100011 ADC11

100100 ADC12

100101 ADC13

100110 ADC14

100111 ADC15

101000(1)

N/A

ADC8 ADC8 10x

101001(1) ADC9 ADC8 10x

101010(1) ADC8 ADC8 200x

101011(1) ADC9 ADC8 200x

101100(1) ADC10 ADC10 10x

101101(1) ADC11 ADC10 10x

101110(1) ADC10 ADC10 200x

101111(1) ADC11 ADC10 200x

110000 ADC8 ADC9 1x

110001 ADC9 ADC9 1x

110010 ADC10 ADC9 1x

110011 ADC11 ADC9 1x

110100 ADC12 ADC9 1x

110101 ADC13 ADC9 1x

Table 126. Input Channel Selections (Continued)

MUX5..0Single Ended Input

Positive Differential Input

Negative Differential Input Gain

290 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. To reach the given accuracy, 10x or 200x Gain should not be used for operating volt-age below 2.7V

110110

N/A

ADC14 ADC9 1x

110111 ADC15 ADC9 1x

111000 ADC8 ADC10 1x

111001 ADC9 ADC10 1x

111010 ADC10 ADC10 1x

111011 ADC11 ADC10 1x

111100 ADC12 ADC10 1x

111101 ADC13 ADC10 1x

111110 Reserved N/A

111111 Reserved N/A

Table 126. Input Channel Selections (Continued)

MUX5..0Single Ended Input

Positive Differential Input

Negative Differential Input Gain

291

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

ADC Control and Status Register A – ADCSRA

• Bit 7 – ADEN: ADC Enable

Writing this bit to one enables the ADC. By writing it to zero, the ADC is turned off. Turn-ing the ADC off while a conversion is in progress, will terminate this conversion.

• Bit 6 – ADSC: ADC Start Conversion

In Single Conversion mode, write this bit to one to start each conversion. In Free Run-ning mode, write this bit to one to start the first conversion. The first conversion afterADSC has been written after the ADC has been enabled, or if ADSC is written at thesame time as the ADC is enabled, will take 25 ADC clock cycles instead of the normal13. This first conversion performs initialization of the ADC.

ADSC will read as one as long as a conversion is in progress. When the conversion iscomplete, it returns to zero. Writing zero to this bit has no effect.

• Bit 5 – ADATE: ADC Auto Trigger Enable

When this bit is written to one, Auto Triggering of the ADC is enabled. The ADC will starta conversion on a positive edge of the selected trigger signal. The trigger source isselected by setting the ADC Trigger Select bits, ADTS in ADCSRB.

• Bit 4 – ADIF: ADC Interrupt Flag

This bit is set when an ADC conversion completes and the Data Registers are updated.The ADC Conversion Complete Interrupt is executed if the ADIE bit and the I-bit inSREG are set. ADIF is cleared by hardware when executing the corresponding interrupthandling vector. Alternatively, ADIF is cleared by writing a logical one to the flag.Beware that if doing a Read-Modify-Write on ADCSRA, a pending interrupt can be dis-abled. This also applies if the SBI and CBI instructions are used.

• Bit 3 – ADIE: ADC Interrupt Enable

When this bit is written to one and the I-bit in SREG is set, the ADC Conversion Com-plete Interrupt is activated.

• Bits 2:0 – ADPS2:0: ADC Prescaler Select Bits

These bits determine the division factor between the XTAL frequency and the inputclock to the ADC.

Bit 7 6 5 4 3 2 1 0

ADEN ADSC ADATE ADIF ADIE ADPS2 ADPS1 ADPS0 ADCSRA

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Table 127. ADC Prescaler Selections

ADPS2 ADPS1 ADPS0 Division Factor

0 0 0 2

0 0 1 2

0 1 0 4

0 1 1 8

1 0 0 16

292 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The ADC Data Register – ADCL and ADCH

ADLAR = 0

ADLAR = 1

When an ADC conversion is complete, the result is found in these two registers. If differ-ential channels are used, the result is presented in two’s complement form.

When ADCL is read, the ADC Data Register is not updated until ADCH is read. Conse-quently, if the result is left adjusted and no more than 8-bit precision (7 bit + sign bit fordifferential input channels) is required, it is sufficient to read ADCH. Otherwise, ADCLmust be read first, then ADCH.

The ADLAR bit in ADMUX, and the MUXn bits in ADMUX affect the way the result isread from the registers. If ADLAR is set, the result is left adjusted. If ADLAR is cleared(default), the result is right adjusted.

• ADC9:0: ADC Conversion Result

These bits represent the result from the conversion, as detailed in “ADC ConversionResult” on page 286.

ADC Control and Status Register B – ADCSRB

1 0 1 32

1 1 0 64

1 1 1 128

Table 127. ADC Prescaler Selections

ADPS2 ADPS1 ADPS0 Division Factor

Bit 15 14 13 12 11 10 9 8

– – – – – – ADC9 ADC8 ADCH

ADC7 ADC6 ADC5 ADC4 ADC3 ADC2 ADC1 ADC0 ADCL

7 6 5 4 3 2 1 0

Read/Write R R R R R R R R

R R R R R R R R

Initial Value 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

Bit 15 14 13 12 11 10 9 8

ADC9 ADC8 ADC7 ADC6 ADC5 ADC4 ADC3 ADC2 ADCH

ADC1 ADC0 – – – – – – ADCL

7 6 5 4 3 2 1 0

Read/Write R R R R R R R R

R R R R R R R R

Initial Value 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– ACME – – MUX5 ADTS2 ADTS1 ADTS0 ADCSRB

Read/Write R R/W R R R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

293

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• Bit 7 – Res: Reserved Bit

This bit is reserved for future use. To ensure compatibility with future devices, this bitmust be written to zero when ADCSRB is written.

• Bit 2:0 – ADTS2:0: ADC Auto Trigger Source

If ADATE in ADCSRA is written to one, the value of these bits selects which source willtrigger an ADC conversion. If ADATE is cleared, the ADTS2:0 settings will have noeffect. A conversion will be triggered by the rising edge of the selected Interrupt Flag.Note that switching from a trigger source that is cleared to a trigger source that is set,will generate a positive edge on the trigger signal. If ADEN in ADCSRA is set, this willstart a conversion. Switching to Free Running mode (ADTS[2:0]=0) will not cause a trig-ger event, even if the ADC Interrupt Flag is set.

Digital Input Disable Register 0 – DIDR0

• Bit 7..0 – ADC7D..ADC0D: ADC7..0 Digital Input Disable

When this bit is written logic one, the digital input buffer on the corresponding ADC pin isdisabled. The corresponding PIN Register bit will always read as zero when this bit isset. When an analog signal is applied to the ADC7..0 pin and the digital input from thispin is not needed, this bit should be written logic one to reduce power consumption inthe digital input buffer.

Digital Input Disable Register 2 – DIDR2

• Bit 7..0 – ADC15D..ADC8D: ADC15..8 Digital Input Disable

When this bit is written logic one, the digital input buffer on the corresponding ADC pin isdisabled. The corresponding PIN Register bit will always read as zero when this bit isset. When an analog signal is applied to the ADC15..8 pin and the digital input from thispin is not needed, this bit should be written logic one to reduce power consumption inthe digital input buffer.

Table 128. ADC Auto Trigger Source Selections

ADTS2 ADTS1 ADTS0 Trigger Source

0 0 0 Free Running mode

0 0 1 Analog Comparator

0 1 0 External Interrupt Request 0

0 1 1 Timer/Counter0 Compare Match A

1 0 0 Timer/Counter0 Overflow

1 0 1 Timer/Counter1 Compare Match B

1 1 0 Timer/Counter1 Overflow

1 1 1 Timer/Counter1 Capture Event

Bit 7 6 5 4 3 2 1 0

ADC7D ADC6D ADC5D ADC4D ADC3D ADC2D ADC1D ADC0D DIDR0

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

ADC15D ADC14D ADC13D ADC12D ADC11D ADC10D ADC9D ADC8D DIDR2

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

294 ATmega640/1280/1281/2560/25612549A–AVR–03/05

295

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

JTAG Interface and On-chip Debug System

Features • JTAG (IEEE std. 1149.1 Compliant) Interface• Boundary-scan Capabilities According to the IEEE std. 1149.1 (JTAG) Standard• Debugger Access to:

– All Internal Peripheral Units– Internal and External RAM– The Internal Register File– Program Counter– EEPROM and Flash Memories

• Extensive On-chip Debug Support for Break Conditions, Including– AVR Break Instruction– Break on Change of Program Memory Flow– Single Step Break– Program Memory Break Points on Single Address or Address Range– Data Memory Break Points on Single Address or Address Range

• Programming of Flash, EEPROM, Fuses, and Lock Bits through the JTAG Interface• On-chip Debugging Supported by AVR Studio®

Overview The AVR IEEE std. 1149.1 compliant JTAG interface can be used for

• Testing PCBs by using the JTAG Boundary-scan capability

• Programming the non-volatile memories, Fuses and Lock bits

• On-chip debugging

A brief description is given in the following sections. Detailed descriptions for Program-ming via the JTAG interface, and using the Boundary-scan Chain can be found in thesections “Programming via the JTAG Interface” on page 353 and “IEEE 1149.1 (JTAG)Boundary-scan” on page 301, respectively. The On-chip Debug support is consideredbeing private JTAG instructions, and distributed within ATMEL and to selected thirdparty vendors only.

Figure 129 shows a block diagram of the JTAG interface and the On-chip Debug sys-tem. The TAP Controller is a state machine controlled by the TCK and TMS signals. TheTAP Controller selects either the JTAG Instruction Register or one of several Data Reg-isters as the scan chain (Shift Register) between the TDI – input and TDO – output. TheInstruction Register holds JTAG instructions controlling the behavior of a Data Register.

The ID-Register, Bypass Register, and the Boundary-scan Chain are the Data Registersused for board-level testing. The JTAG Programming Interface (actually consisting ofseveral physical and virtual Data Registers) is used for serial programming via the JTAGinterface. The Internal Scan Chain and Break Point Scan Chain are used for On-chipdebugging only.

Test Access Port – TAP The JTAG interface is accessed through four of the AVR’s pins. In JTAG terminology,these pins constitute the Test Access Port – TAP. These pins are:

• TMS: Test mode select. This pin is used for navigating through the TAP-controller state machine.

• TCK: Test Clock. JTAG operation is synchronous to TCK.

• TDI: Test Data In. Serial input data to be shifted in to the Instruction Register or Data Register (Scan Chains).

• TDO: Test Data Out. Serial output data from Instruction Register or Data Register.

296 ATmega640/1280/1281/2560/25612549A–AVR–03/05

The IEEE std. 1149.1 also specifies an optional TAP signal; TRST – Test ReSeT –which is not provided.

When the JTAGEN Fuse is unprogrammed, these four TAP pins are normal port pins,and the TAP controller is in reset. When programmed, the input TAP signals are inter-nally pulled high and the JTAG is enabled for Boundary-scan and programming. Thedevice is shipped with this fuse programmed.

For the On-chip Debug system, in addition to the JTAG interface pins, the RESET pin ismonitored by the debugger to be able to detect external reset sources. The debuggercan also pull the RESET pin low to reset the whole system, assuming only open collec-tors on the reset line are used in the application.

Figure 129. Block Diagram

TAPCONTROLLER

TDITDOTCKTMS

FLASHMEMORY

AVR CPU

DIGITALPERIPHERAL

UNITS

JTAG / AVR CORECOMMUNICATION

INTERFACE

BREAKPOINTUNIT

FLOW CONTROLUNIT

OCD STATUSAND CONTROL

INTERNAL SCANCHAIN

MUX

INSTRUCTIONREGISTER

IDREGISTER

BYPASSREGISTER

JTAG PROGRAMMINGINTERFACE

PCInstruction

AddressData

BREAKPOINTSCAN CHAIN

ADDRESSDECODER

ANALOGPERIPHERIAL

UNITS

I/O PORT 0

I/O PORT n

BOUNDARY SCAN CHAIN

Analog inputs

Control & Clock lines

DEVICE BOUNDARY

297

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 130. TAP Controller State Diagram

TAP Controller The TAP controller is a 16-state finite state machine that controls the operation of theBoundary-scan circuitry, JTAG programming circuitry, or On-chip Debug system. Thestate transitions depicted in Figure 130 depend on the signal present on TMS (shownadjacent to each state transition) at the time of the rising edge at TCK. The initial stateafter a Power-on Reset is Test-Logic-Reset.

As a definition in this document, the LSB is shifted in and out first for all Shift Registers.

Assuming Run-Test/Idle is the present state, a typical scenario for using the JTAG inter-face is:

• At the TMS input, apply the sequence 1, 1, 0, 0 at the rising edges of TCK to enter the Shift Instruction Register – Shift-IR state. While in this state, shift the four bits of the JTAG instructions into the JTAG Instruction Register from the TDI input at the rising edge of TCK. The TMS input must be held low during input of the 3 LSBs in order to remain in the Shift-IR state. The MSB of the instruction is shifted in when this state is left by setting TMS high. While the instruction is shifted in from the TDI pin, the captured IR-state 0x01 is shifted out on the TDO pin. The JTAG Instruction selects a particular Data Register as path between TDI and TDO and controls the circuitry surrounding the selected Data Register.

• Apply the TMS sequence 1, 1, 0 to re-enter the Run-Test/Idle state. The instruction is latched onto the parallel output from the Shift Register path in the Update-IR

Test-Logic-Reset

Run-Test/Idle

Shift-DR

Exit1-DR

Pause-DR

Exit2-DR

Update-DR

Select-IR Scan

Capture-IR

Shift-IR

Exit1-IR

Pause-IR

Exit2-IR

Update-IR

Select-DR Scan

Capture-DR

0

1

0 1 1 1

0 0

0 0

1 1

1 0

1

1

0

1

0

0

1 0

1

1

0

1

0

0

00

11

298 ATmega640/1280/1281/2560/25612549A–AVR–03/05

state. The Exit-IR, Pause-IR, and Exit2-IR states are only used for navigating the state machine.

• At the TMS input, apply the sequence 1, 0, 0 at the rising edges of TCK to enter the Shift Data Register – Shift-DR state. While in this state, upload the selected Data Register (selected by the present JTAG instruction in the JTAG Instruction Register) from the TDI input at the rising edge of TCK. In order to remain in the Shift-DR state, the TMS input must be held low during input of all bits except the MSB. The MSB of the data is shifted in when this state is left by setting TMS high. While the Data Register is shifted in from the TDI pin, the parallel inputs to the Data Register captured in the Capture-DR state is shifted out on the TDO pin.

• Apply the TMS sequence 1, 1, 0 to re-enter the Run-Test/Idle state. If the selected Data Register has a latched parallel-output, the latching takes place in the Update-DR state. The Exit-DR, Pause-DR, and Exit2-DR states are only used for navigating the state machine.

As shown in the state diagram, the Run-Test/Idle state need not be entered betweenselecting JTAG instruction and using Data Registers, and some JTAG instructions mayselect certain functions to be performed in the Run-Test/Idle, making it unsuitable as anIdle state.Note: Independent of the initial state of the TAP Controller, the Test-Logic-Reset state can

always be entered by holding TMS high for five TCK clock periods.

For detailed information on the JTAG specification, refer to the literature listed in “Bibli-ography” on page 300.

Using the Boundary-scan Chain

A complete description of the Boundary-scan capabilities are given in the section “IEEE1149.1 (JTAG) Boundary-scan” on page 301.

Using the On-chip Debug System

As shown in Figure 129, the hardware support for On-chip Debugging consists mainly of

• A scan chain on the interface between the internal AVR CPU and the internal peripheral units.

• Break Point unit.

• Communication interface between the CPU and JTAG system.

All read or modify/write operations needed for implementing the Debugger are done byapplying AVR instructions via the internal AVR CPU Scan Chain. The CPU sends theresult to an I/O memory mapped location which is part of the communication interfacebetween the CPU and the JTAG system.

The Break Point Unit implements Break on Change of Program Flow, Single StepBreak, two Program Memory Break Points, and two combined Break Points. Together,the four Break Points can be configured as either:

• 4 single Program Memory Break Points.

• 3 Single Program Memory Break Point + 1 single Data Memory Break Point.

• 2 single Program Memory Break Points + 2 single Data Memory Break Points.

• 2 single Program Memory Break Points + 1 Program Memory Break Point with mask (“range Break Point”).

• 2 single Program Memory Break Points + 1 Data Memory Break Point with mask (“range Break Point”).

A debugger, like the AVR Studio, may however use one or more of these resources forits internal purpose, leaving less flexibility to the end-user.

299

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

A list of the On-chip Debug specific JTAG instructions is given in “On-chip Debug Spe-cific JTAG Instructions” on page 299.

The JTAGEN Fuse must be programmed to enable the JTAG Test Access Port. In addi-tion, the OCDEN Fuse must be programmed and no Lock bits must be set for the On-chip debug system to work. As a security feature, the On-chip debug system is disabledwhen either of the LB1 or LB2 Lock bits are set. Otherwise, the On-chip debug systemwould have provided a back-door into a secured device.

The AVR Studio enables the user to fully control execution of programs on an AVRdevice with On-chip Debug capability, AVR In-Circuit Emulator, or the built-in AVRInstruction Set Simulator. AVR Studio® supports source level execution of Assemblyprograms assembled with Atmel Corporation’s AVR Assembler and C programs com-piled with third party vendors’ compilers.

AVR Studio runs under Microsoft® Windows® 95/98/2000 and Microsoft Windows NT®.

For a full description of the AVR Studio, please refer to the AVR Studio User Guide.Only highlights are presented in this document.

All necessary execution commands are available in AVR Studio, both on source leveland on disassembly level. The user can execute the program, single step through thecode either by tracing into or stepping over functions, step out of functions, place thecursor on a statement and execute until the statement is reached, stop the execution,and reset the execution target. In addition, the user can have an unlimited number ofcode Break Points (using the BREAK instruction) and up to two data memory BreakPoints, alternatively combined as a mask (range) Break Point.

On-chip Debug Specific JTAG Instructions

The On-chip debug support is considered being private JTAG instructions, and distrib-uted within ATMEL and to selected third party vendors only. Instruction opcodes arelisted for reference.

PRIVATE0; 0x8 Private JTAG instruction for accessing On-chip debug system.

PRIVATE1; 0x9 Private JTAG instruction for accessing On-chip debug system.

PRIVATE2; 0xA Private JTAG instruction for accessing On-chip debug system.

PRIVATE3; 0xB Private JTAG instruction for accessing On-chip debug system.

300 ATmega640/1280/1281/2560/25612549A–AVR–03/05

On-chip Debug Related Register in I/O Memory

On-chip Debug Register – OCDR

The OCDR Register provides a communication channel from the running program in themicrocontroller to the debugger. The CPU can transfer a byte to the debugger by writingto this location. At the same time, an internal flag; I/O Debug Register Dirty – IDRD – isset to indicate to the debugger that the register has been written. When the CPU readsthe OCDR Register the 7 LSB will be from the OCDR Register, while the MSB is theIDRD bit. The debugger clears the IDRD bit when it has read the information.

In some AVR devices, this register is shared with a standard I/O location. In this case,the OCDR Register can only be accessed if the OCDEN Fuse is programmed, and thedebugger enables access to the OCDR Register. In all other cases, the standard I/Olocation is accessed.

Refer to the debugger documentation for further information on how to use this register.

Using the JTAG Programming Capabilities

Programming of AVR parts via JTAG is performed via the 4-pin JTAG port, TCK, TMS,TDI, and TDO. These are the only pins that need to be controlled/observed to performJTAG programming (in addition to power pins). It is not required to apply 12V externally.The JTAGEN Fuse must be programmed and the JTD bit in the MCUCR Register mustbe cleared to enable the JTAG Test Access Port.

The JTAG programming capability supports:

• Flash programming and verifying.

• EEPROM programming and verifying.

• Fuse programming and verifying.

• Lock bit programming and verifying.

The Lock bit security is exactly as in parallel programming mode. If the Lock bits LB1 orLB2 are programmed, the OCDEN Fuse cannot be programmed unless first doing achip erase. This is a security feature that ensures no back-door exists for reading out thecontent of a secured device.

The details on programming through the JTAG interface and programming specificJTAG instructions are given in the section “Programming via the JTAG Interface” onpage 353.

Bibliography For more information about general Boundary-scan, the following literature can beconsulted:

• IEEE: IEEE Std. 1149.1-1990. IEEE Standard Test Access Port and Boundary-scan Architecture, IEEE, 1993.

• Colin Maunder: The Board Designers Guide to Testable Logic Circuits, Addison-Wesley, 1992.

Bit 7 6 5 4 3 2 1 0

MSB/IDRD LSB OCDR

Read/Write R/W R/W R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

301

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

IEEE 1149.1 (JTAG) Boundary-scan

Features • JTAG (IEEE std. 1149.1 compliant) Interface• Boundary-scan Capabilities According to the JTAG Standard• Full Scan of all Port Functions as well as Analog Circuitry having Off-chip Connections• Supports the Optional IDCODE Instruction• Additional Public AVR_RESET Instruction to Reset the AVR

System Overview The Boundary-scan chain has the capability of driving and observing the logic levels onthe digital I/O pins, as well as the boundary between digital and analog logic for analogcircuitry having off-chip connections. At system level, all ICs having JTAG capabilitiesare connected serially by the TDI/TDO signals to form a long Shift Register. An externalcontroller sets up the devices to drive values at their output pins, and observe the inputvalues received from other devices. The controller compares the received data with theexpected result. In this way, Boundary-scan provides a mechanism for testing intercon-nections and integrity of components on Printed Circuits Boards by using the four TAPsignals only.

The four IEEE 1149.1 defined mandatory JTAG instructions IDCODE, BYPASS, SAM-PLE/PRELOAD, and EXTEST, as well as the AVR specific public JTAG instructionAVR_RESET can be used for testing the Printed Circuit Board. Initial scanning of theData Register path will show the ID-Code of the device, since IDCODE is the defaultJTAG instruction. It may be desirable to have the AVR device in reset during test mode.If not reset, inputs to the device may be determined by the scan operations, and theinternal software may be in an undetermined state when exiting the test mode. Enteringreset, the outputs of any port pin will instantly enter the high impedance state, makingthe HIGHZ instruction redundant. If needed, the BYPASS instruction can be issued tomake the shortest possible scan chain through the device. The device can be set in thereset state either by pulling the external RESET pin low, or issuing the AVR_RESETinstruction with appropriate setting of the Reset Data Register.

The EXTEST instruction is used for sampling external pins and loading output pins withdata. The data from the output latch will be driven out on the pins as soon as theEXTEST instruction is loaded into the JTAG IR-Register. Therefore, the SAMPLE/PRE-LOAD should also be used for setting initial values to the scan ring, to avoid damagingthe board when issuing the EXTEST instruction for the first time. SAMPLE/PRELOADcan also be used for taking a snapshot of the external pins during normal operation ofthe part.

The JTAGEN Fuse must be programmed and the JTD bit in the I/O Register MCUCRmust be cleared to enable the JTAG Test Access Port.

When using the JTAG interface for Boundary-scan, using a JTAG TCK clock frequencyhigher than the internal chip frequency is possible. The chip clock is not required to run.

Data Registers The Data Registers relevant for Boundary-scan operations are:

• Bypass Register

• Device Identification Register

• Reset Register

• Boundary-scan Chain

302 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Bypass Register The Bypass Register consists of a single Shift Register stage. When the Bypass Regis-ter is selected as path between TDI and TDO, the register is reset to 0 when leaving theCapture-DR controller state. The Bypass Register can be used to shorten the scanchain on a system when the other devices are to be tested.

Device Identification Register Figure 131 shows the structure of the Device Identification Register.

Figure 131. The Format of the Device Identification Register

Version Version is a 4-bit number identifying the revision of the component. The JTAG versionnumber follows the revision of the device. Revision A is 0x0, revision B is 0x1 and so on.

Part Number The part number is a 16-bit code identifying the component. The JTAG Part Number forATmega640/1280/1281/2560/2561 is listed in Table 129.

Manufacturer ID The Manufacturer ID is a 11-bit code identifying the manufacturer. The JTAG manufac-turer ID for ATMEL is listed in Table 130.

Reset Register The Reset Register is a test Data Register used to reset the part. Since the AVR tri-states Port Pins when reset, the Reset Register can also replace the function of theunimplemented optional JTAG instruction HIGHZ.

A high value in the Reset Register corresponds to pulling the external Reset low. Thepart is reset as long as there is a high value present in the Reset Register. Dependingon the fuse settings for the clock options, the part will remain reset for a reset time-outperiod (see “Clock Sources” on page 40) after releasing the Reset Register. The outputfrom this Data Register is not latched, so the reset will take place immediately, as shownin Figure 132.

MSB LSB

Bit 31 28 27 12 11 1 0

Device ID Version Part Number Manufacturer ID 1

4 bits 16 bits 11 bits 1-bit

Table 129. AVR JTAG Part Number

Part Number JTAG Part Number (Hex)

ATmega640 0x9607

ATmega1280 0x9703

ATmega1281 0x9704

ATmega2560 0x9801

ATmega2561 0x9802

Table 130. Manufacturer ID

Manufacturer JTAG Manufactor ID (Hex)

ATMEL 0x01F

303

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 132. Reset Register

Boundary-scan Chain The Boundary-scan Chain has the capability of driving and observing the logic levels onthe digital I/O pins, as well as the boundary between digital and analog logic for analogcircuitry having off-chip connections.

See “Boundary-scan Chain” on page 305 for a complete description.

Boundary-scan Specific JTAG Instructions

The Instruction Register is 4-bit wide, supporting up to 16 instructions. Listed below arethe JTAG instructions useful for Boundary-scan operation. Note that the optional HIGHZinstruction is not implemented, but all outputs with tri-state capability can be set in high-impedant state by using the AVR_RESET instruction, since the initial state for all portpins is tri-state.

As a definition in this datasheet, the LSB is shifted in and out first for all Shift Registers.

The OPCODE for each instruction is shown behind the instruction name in hex format.The text describes which Data Register is selected as path between TDI and TDO foreach instruction.

EXTEST; 0x0 Mandatory JTAG instruction for selecting the Boundary-scan Chain as Data Register fortesting circuitry external to the AVR package. For port-pins, Pull-up Disable, OutputControl, Output Data, and Input Data are all accessible in the scan chain. For Analog cir-cuits having off-chip connections, the interface between the analog and the digital logicis in the scan chain. The contents of the latched outputs of the Boundary-scan chain isdriven out as soon as the JTAG IR-Register is loaded with the EXTEST instruction.

The active states are:

• Capture-DR: Data on the external pins are sampled into the Boundary-scan Chain.

• Shift-DR: The Internal Scan Chain is shifted by the TCK input.

• Update-DR: Data from the scan chain is applied to output pins.

IDCODE; 0x1 Optional JTAG instruction selecting the 32 bit ID-Register as Data Register. The ID-Register consists of a version number, a device number and the manufacturer codechosen by JEDEC. This is the default instruction after power-up.

The active states are:

• Capture-DR: Data in the IDCODE Register is sampled into the Boundary-scan Chain.

• Shift-DR: The IDCODE scan chain is shifted by the TCK input.

D QFromTDI

ClockDR · AVR_RESET

To TDO

From Other Internal andExternal Reset Sources

Internal reset

304 ATmega640/1280/1281/2560/25612549A–AVR–03/05

SAMPLE_PRELOAD; 0x2 Mandatory JTAG instruction for pre-loading the output latches and taking a snap-shot ofthe input/output pins without affecting the system operation. However, the output latchesare not connected to the pins. The Boundary-scan Chain is selected as Data Register.

The active states are:

• Capture-DR: Data on the external pins are sampled into the Boundary-scan Chain.

• Shift-DR: The Boundary-scan Chain is shifted by the TCK input.

• Update-DR: Data from the Boundary-scan chain is applied to the output latches. However, the output latches are not connected to the pins.

AVR_RESET; 0xC The AVR specific public JTAG instruction for forcing the AVR device into the Resetmode or releasing the JTAG reset source. The TAP controller is not reset by this instruc-tion. The one bit Reset Register is selected as Data Register. Note that the reset will beactive as long as there is a logic “one” in the Reset Chain. The output from this chain isnot latched.

The active states are:

• Shift-DR: The Reset Register is shifted by the TCK input.

BYPASS; 0xF Mandatory JTAG instruction selecting the Bypass Register for Data Register.

The active states are:

• Capture-DR: Loads a logic “0” into the Bypass Register.

• Shift-DR: The Bypass Register cell between TDI and TDO is shifted.

Boundary-scan Related Register in I/O Memory

MCU Control Register – MCUCR

The MCU Control Register contains control bits for general MCU functions.

• Bits 7 – JTD: JTAG Interface Disable

When this bit is zero, the JTAG interface is enabled if the JTAGEN Fuse is programmed.If this bit is one, the JTAG interface is disabled. In order to avoid unintentional disablingor enabling of the JTAG interface, a timed sequence must be followed when changingthis bit: The application software must write this bit to the desired value twice within fourcycles to change its value. Note that this bit must not be altered when using the On-chipDebug system.

MCU Status Register – MCUSR

The MCU Status Register provides information on which reset source caused an MCUreset.

Bit 7 6 5 4 3 2 1 0

JTD – – PUD – – IVSEL IVCE MCUCR

Read/Write R/W R R R/W R R R/W R/W

Initial Value 0 0 0 0 0 0 0 0

Bit 7 6 5 4 3 2 1 0

– – – JTRF WDRF BORF EXTRF PORF MCUSR

Read/Write R R R R/W R/W R/W R/W R/W

Initial Value 0 0 0 See Bit Description

305

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

• Bit 4 – JTRF: JTAG Reset Flag

This bit is set if a reset is being caused by a logic one in the JTAG Reset Registerselected by the JTAG instruction AVR_RESET. This bit is reset by a Power-on Reset, orby writing a logic zero to the flag.

Boundary-scan Chain The Boundary-scan chain has the capability of driving and observing the logic levels onthe digital I/O pins, as well as the boundary between digital and analog logic for analogcircuitry having off-chip connection.

Scanning the Digital Port Pins Figure 133 shows the Boundary-scan Cell for a bi-directional port pin. The pull-up func-tion is disabled during Boundary-scan when the JTAG IC contains EXTEST orSAMPLE_PRELOAD. The cell consists of a bi-directional pin cell that combines thethree signals Output Control - OCxn, Output Data - ODxn, and Input Data - IDxn, intoonly a two-stage Shift Register. The port and pin indexes are not used in the followingdescription

The Boundary-scan logic is not included in the figures in the datasheet. Figure 134shows a simple digital port pin as described in the section “I/O-Ports” on page 81. TheBoundary-scan details from Figure 133 replaces the dashed box in Figure 134.

When no alternate port function is present, the Input Data - ID - corresponds to thePINxn Register value (but ID has no synchronizer), Output Data corresponds to thePORT Register, Output Control corresponds to the Data Direction - DD Register, andthe Pull-up Enable - PUExn - corresponds to logic expression PUD · DDxn · PORTxn.

Digital alternate port functions are connected outside the dotted box in Figure 134 tomake the scan chain read the actual pin value. For analog function, there is a direct con-nection from the external pin to the analog circuit. There is no scan chain on theinterface between the digital and the analog circuitry, but some digital control signal toanalog circuitry are turned off to avoid driving contention on the pads.

When JTAG IR contains EXTEST or SAMPLE_PRELOAD the clock is not sent out onthe port pins even if the CKOUT fuse is programmed. Even though the clock is outputwhen the JTAG IR contains SAMPLE_PRELOAD, the clock is not sampled by theboundary scan.

306 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 133. Boundary-scan Cell for Bi-directional Port Pin with Pull-up Function.

D Q D Q

G

0

10

1

D Q D Q

G

0

10

1

0

1

Por

t Pin

(P

Xn)

VccEXTESTTo Next CellShiftDR

Output Control (OC)

Output Data (OD)

Input Data (ID)

From Last Cell UpdateDRClockDR

FF1 LD1

LD0FF0

0

1

Pull-up Enable (PUE)

307

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 134. General Port Pin Schematic Diagram

Scanning the RESET Pin The RESET pin accepts 5V active low logic for standard reset operation, and 12V activehigh logic for High Voltage Parallel programming. An observe-only cell as shown in Fig-ure 135 is inserted for the 5V reset signal.

Figure 135. Observe-only Cell

CLK

RPx

RRx

WRx

RDx

WDx

PUD

SYNCHRONIZER

WDx: WRITE DDRx

WRx: WRITE PORTxRRx: READ PORTx REGISTERRPx: READ PORTx PIN

PUD: PULLUP DISABLE

CLK : I/O CLOCK

RDx: READ DDRx

D

L

Q

Q

RESET

RESET

Q

QD

Q

Q D

CLR

PORTxn

Q

Q D

CLR

DDxn

PINxn

DAT

A B

US

SLEEP

SLEEP: SLEEP CONTROL

Pxn

I/O

I/O

See Boundary-scan Description for Details!

PUExn

OCxn

ODxn

IDxn

PUExn: PULLUP ENABLE for pin PxnOCxn: OUTPUT CONTROL for pin PxnODxn: OUTPUT DATA to pin PxnIDxn: INPUT DATA from pin Pxn

0

1D Q

FromPrevious

Cell

ClockDR

ShiftDR

ToNextCell

From System Pin To System Logic

FF1

308 ATmega640/1280/1281/2560/25612549A–AVR–03/05

ATmega640/1280/1281/2560/2561 Boundary-scan Order

Table 131 shows the Scan order between TDI and TDO when the Boundary-scan chainis selected as data path. Bit 0 is the LSB; the first bit scanned in, and the first bitscanned out. The scan order follows the pin-out order as far as possible. Therefore, thebits of Port A and Port K is scanned in the opposite bit order of the other ports. Excep-tions from the rules are the Scan chains for the analog circuits, which constitute themost significant bits of the scan chain regardless of which physical pin they are con-nected to. In Figure 133, PXn. Data corresponds to FF0, PXn. Control corresponds toFF1, PXn. Bit 4, 5, 6 and 7 of Port F is not in the scan chain, since these pins constitutethe TAP pins when the JTAG is enabled.

Table 131. ATmega640/1280/2560 Boundary-scan Order

Bit Number Signal Name Module

164 PG5.Data Port G

163 PG5.Control

162 PE0.Data Port E

161 PE0.Control

160 PE1.Data

159 PE1.Control

158 PE2.Data

157 PE2.Control

156 PE3.Data

155 PE3.Control

154 PE4.Data

153 PE4.Control

152 PE5.Data

151 PE5.Control

150 PE6.Data

149 PE6.Control

148 PE7.Data

147 PE7.Control

146 PH0.Data Port H

145 PH0.Control

144 PH1.Data

143 PH1.Control

142 PH2.Data

141 PH2.Control

140 PH3.Data

139 PH3.Control

138 PH4.Data

137 PH4.Control

136 PH5.Data

309

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

135 PH5.Control

134 PH6.Data

133 PH6.Control

132 PB0.Data Port B

131 PB0.Control

130 PB1.Data

129 PB1.Control

128 PB2.Data

127 PB2.Control

126 PB3.Data

125 PB3.Control

124 PB4.Data

123 PB4.Control

122 PB5.Data

121 PB5.Control

120 PB6.Data

119 PB6.Control

118 PB7.Data

117 PB7.Control

116 PH7.Data Port H

115 PH7.Control

114 PG3.Data Port G

113 PG3.Control

112 PG4.Data

111 PG4.Control

110 RSTT Reset Logic (Observe Only)

109 PL0.Data Port L

108 PL0.Control

107 PL1.Data

106 PL1.Control

105 PL2.Data

104 PL2.Control

103 PL3.Data

102 PL3.Control

101 PL4.Data

100 PL4.Control

Table 131. ATmega640/1280/2560 Boundary-scan Order (Continued)

Bit Number Signal Name Module

310 ATmega640/1280/1281/2560/25612549A–AVR–03/05

99 PL5.Data

98 PL5.Control

97 PL6.Data

96 PL6.Control

95 PL7.Data

94 PL7.Control

93 PD0.Data Port D

92 PD0.Control

91 PD1.Data

90 PD1.Control

89 PD2.Data

88 PD2.Control

87 PD3.Data

86 PD3.Control

85 PD4.Data

84 PD4.Control

83 PD5.Data

82 PD5.Control

81 PD6.Data

80 PD6.Control

79 PD7.Data

78 PD7.Control

77 PG0.Data Port G

76 PG0.Control

75 PG1.Data

74 PG1.Control

73 PC0.Data Port C

72 PC0.Control

71 PC1.Data

70 PC1.Control

69 PC2.Data

68 PC2.Control

67 PC3.Data

66 PC3.Control

65 PC4.Data

64 PC4.Control

Table 131. ATmega640/1280/2560 Boundary-scan Order (Continued)

Bit Number Signal Name Module

311

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

63 PC5.Data

62 PC5.Control

61 PC6.Data

60 PC6.Control

59 PC7.Data

58 PC7.Control

57 PJ0.Data Port J

56 PJ0.Control

55 PJ1.Data

54 PJ1.Control

53 PJ2.Data

52 PJ2.Control

51 PJ3.Data

50 PJ3.Control

49 PJ4.Data

48 PJ4.Control

47 PJ5.Data

46 PJ5.Control

45 PJ6.Data

44 PJ6.Control

43 PG2.Data Port G

42 PG2.Control

41 PA7.Data Port A

40 PA7.Control

39 PA6.Data

38 PA6.Control

37 PA5.Data

36 PA5.Control

35 PA4.Data

34 PA4.Control

33 PA3.Data

32 PA3.Control

31 PA2.Data

30 PA2.Control

29 PA1.Data

28 PA1.Control

Table 131. ATmega640/1280/2560 Boundary-scan Order (Continued)

Bit Number Signal Name Module

312 ATmega640/1280/1281/2560/25612549A–AVR–03/05

27 PA0.Data

26 PA0.Control

25 PJ7.Data Port J

24 PJ7.Control

23 PK7.Data Port K

22 PK7.Control

21 PK6.Data

20 PK6.Control

19 PK5.Data

18 PK5.Control

17 PK4.Data

16 PK4.Control

15 PK3.Data

14 PK3.Control

13 PK2.Data

12 PK2.Control

11 PK1.Data

10 PK1.Control

9 PK0.Data

8 PK0.Control

7 PF3.Data Port F

6 PF3.Control

5 PF2.Data

4 PF2.Control

3 PF1.Data

2 PF1.Control

1 PF0.Data

0 PF0.Control

Table 131. ATmega640/1280/2560 Boundary-scan Order (Continued)

Bit Number Signal Name Module

313

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 132. ATmega1281/2561 Boundary-scan Order

Bit Number Signal Name Module

100 PG5.Data Port G

99 PG5.Control

98 PE0.Data Port E

97 PE0.Control

96 PE1.Data

95 PE1.Control

94 PE2.Data

93 PE2.Control

92 PE3.Data

91 PE3.Control

90 PE4.Data

89 PE4.Control

88 PE5.Data

87 PE5.Control

86 PE6.Data

85 PE6.Control

84 PE7.Data

83 PE7.Control

82 PB0.Data Port B

81 PB0.Control

80 PB1.Data

79 PB1.Control

78 PB2.Data

77 PB2.Control

76 PB3.Data

75 PB3.Control

74 PB4.Data

73 PB4.Control

72 PB5.Data

71 PB5.Control

70 PB6.Data

69 PB6.Control

68 PB7.Data

67 PB7.Control

66 PG3.Data Port G

314 ATmega640/1280/1281/2560/25612549A–AVR–03/05

65 PG3.Control

64 PG4.Data

63 PG4.Control

62 RSTT Reset Logic (Observe Only)

61 PD0.Data Port D

60 PD0.Control

59 PD1.Data

58 PD1.Control

57 PD2.Data

56 PD2.Control

55 PD3.Data

54 PD3.Control

53 PD4.Data

52 PD4.Control

51 PD5.Data

50 PD5.Control

49 PD6.Data

48 PD6.Control

47 PD7.Data

46 PD7.Control

45 PG0.Data Port G

44 PG0.Control

43 PG1.Data

42 PG1.Control

41 PC0.Data Port C

40 PC0.Control

39 PC1.Data

38 PC1.Control

37 PC2.Data

36 PC2.Control

35 PC3.Data

34 PC3.Control

33 PC4.Data

32 PC4.Control

31 PC5.Data

30 PC5.Control

Table 132. ATmega1281/2561 Boundary-scan Order (Continued)

Bit Number Signal Name Module

315

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Boundary-scan Description Language Files

Boundary-scan Description Language (BSDL) files describe Boundary-scan capabledevices in a standard format used by automated test-generation software. The orderand function of bits in the Boundary-scan Data Register are included in this description.BSDL files are available for ATmega1281/2561 and ATmega640/1280/2560.

29 PC6.Data

28 PC6.Control

27 PC7.Data

26 PC7.Control

25 PG2.Data Port G

24 PG2.Control

23 PA7.Data Port A

22 PA7.Control

21 PA6.Data

20 PA6.Control

19 PA5.Data

18 PA5.Control

17 PA4.Data

16 PA4.Control

15 PA3.Data

14 PA3.Control

13 PA2.Data

12 PA2.Control

11 PA1.Data

10 PA1.Control

9 PA0.Data

8 PA0.Control

7 PF3.Data Port F

6 PF3.Control

5 PF2.Data

4 PF2.Control

3 PF1.Data

2 PF1.Control

1 PF0.Data

0 PF0.Control

Table 132. ATmega1281/2561 Boundary-scan Order (Continued)

Bit Number Signal Name Module

316 ATmega640/1280/1281/2560/25612549A–AVR–03/05

317

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Boot Loader Support – Read-While-Write Self-Programming

The Boot Loader Support provides a real Read-While-Write Self-Programming mecha-nism for downloading and uploading program code by the MCU itself. This featureallows flexible application software updates controlled by the MCU using a Flash-resi-dent Boot Loader program. The Boot Loader program can use any available datainterface and associated protocol to read code and write (program) that code into theFlash memory, or read the code from the program memory. The program code withinthe Boot Loader section has the capability to write into the entire Flash, including theBoot Loader memory. The Boot Loader can thus even modify itself, and it can alsoerase itself from the code if the feature is not needed anymore. The size of the BootLoader memory is configurable with fuses and the Boot Loader has two separate sets ofBoot Lock bits which can be set independently. This gives the user a unique flexibility toselect different levels of protection.

Boot Loader Features • Read-While-Write Self-Programming• Flexible Boot Memory Size• High Security (Separate Boot Lock Bits for a Flexible Protection)• Separate Fuse to Select Reset Vector• Optimized Page(1) Size• Code Efficient Algorithm• Efficient Read-Modify-Write Support

Note: 1. A page is a section in the Flash consisting of several bytes (see Table 158 on page340) used during programming. The page organization does not affect normaloperation.

Application and Boot Loader Flash Sections

The Flash memory is organized in two main sections, the Application section and theBoot Loader section (see Figure 137). The size of the different sections is configured bythe BOOTSZ Fuses as shown in Table 139 on page 330 and Figure 137. These twosections can have different level of protection since they have different sets of Lock bits.

Application Section The Application section is the section of the Flash that is used for storing the applicationcode. The protection level for the Application section can be selected by the applicationBoot Lock bits (Boot Lock bits 0), see Table 134 on page 320. The Application sectioncan never store any Boot Loader code since the SPM instruction is disabled when exe-cuted from the Application section.

BLS – Boot Loader Section While the Application section is used for storing the application code, the The BootLoader software must be located in the BLS since the SPM instruction can initiate a pro-gramming when executing from the BLS only. The SPM instruction can access theentire Flash, including the BLS itself. The protection level for the Boot Loader sectioncan be selected by the Boot Loader Lock bits (Boot Lock bits 1), see Table 135 on page320.

Read-While-Write and No Read-While-Write Flash Sections

Whether the CPU supports Read-While-Write or if the CPU is halted during a BootLoader software update is dependent on which address that is being programmed. Inaddition to the two sections that are configurable by the BOOTSZ Fuses as describedabove, the Flash is also divided into two fixed sections, the Read-While-Write (RWW)section and the No Read-While-Write (NRWW) section. The limit between the RWW-and NRWW sections is given in Table 133 and Figure 136 on page 318. The main differ-ence between the two sections is:

• When erasing or writing a page located inside the RWW section, the NRWW section can be read during the operation.

• When erasing or writing a page located inside the NRWW section, the CPU is halted during the entire operation.

318 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note that the user software can never read any code that is located inside the RWWsection during a Boot Loader software operation. The syntax “Read-While-Write sec-tion” refers to which section that is being programmed (erased or written), not whichsection that actually is being read during a Boot Loader software update.

RWW – Read-While-Write Section

If a Boot Loader software update is programming a page inside the RWW section, it ispossible to read code from the Flash, but only code that is located in the NRWW sec-tion. During an on-going programming, the software must ensure that the RWW sectionnever is being read. If the user software is trying to read code that is located inside theRWW section (i.e., by load program memory, call, or jump instructions or an interrupt)during programming, the software might end up in an unknown state. To avoid this, theinterrupts should either be disabled or moved to the Boot Loader section. The BootLoader section is always located in the NRWW section. The RWW Section Busy bit(RWWSB) in the Store Program Memory Control and Status Register (SPMCSR) will beread as logical one as long as the RWW section is blocked for reading. After a program-ming is completed, the RWWSB must be cleared by software before reading codelocated in the RWW section. See “Store Program Memory Control and Status Register –SPMCSR” on page 321. for details on how to clear RWWSB.

NRWW – No Read-While-Write Section

The code located in the NRWW section can be read when the Boot Loader software isupdating a page in the RWW section. When the Boot Loader code updates the NRWWsection, the CPU is halted during the entire Page Erase or Page Write operation.

Figure 136. Read-While-Write vs. No Read-While-Write

Table 133. Read-While-Write Features

Which Section does the Z-pointerAddress during the Programming?

Which Section can be Read during Programming? CPU Halted?

Read-While-Write Supported?

RWW Section NRWW Section No Yes

NRWW Section None Yes No

Read-While-Write(RWW) Section

No Read-While-Write (NRWW) Section

Z-pointerAddresses RWWSection

Z-pointerAddresses NRWWSection

CPU is HaltedDuring the Operation

Code Located in NRWW SectionCan be Read Duringthe Operation

319

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 137. Memory Sections

Note: 1. The parameters in the figure above are given in Table 139 on page 330.

Boot Loader Lock Bits If no Boot Loader capability is needed, the entire Flash is available for application code.The Boot Loader has two separate sets of Boot Lock bits which can be set indepen-dently. This gives the user a unique flexibility to select different levels of protection.

The user can select:

• To protect the entire Flash from a software update by the MCU.

• To protect only the Boot Loader Flash section from a software update by the MCU.

• To protect only the Application Flash section from a software update by the MCU.

• Allow software update in the entire Flash.

See Table 134 and Table 135 for further details. The Boot Lock bits can be set in soft-ware and in Serial or Parallel Programming mode, but they can be cleared by a ChipErase command only. The general Write Lock (Lock Bit mode 2) does not control theprogramming of the Flash memory by SPM instruction. Similarly, the generalRead/Write Lock (Lock Bit mode 1) does not control reading nor writ ing by(E)LPM/SPM, if it is attempted.

0x0000

Flashend

Program MemoryBOOTSZ = '11'

Application Flash Section

Boot Loader Flash SectionFlashend

Program MemoryBOOTSZ = '10'

0x0000

Program MemoryBOOTSZ = '01'

Program MemoryBOOTSZ = '00'

Application Flash Section

Boot Loader Flash Section

0x0000

Flashend

Application Flash Section

Flashend

End RWW

Start NRWW

Application Flash Section

Boot Loader Flash Section

Boot Loader Flash Section

End RWW

Start NRWW

End RWW

Start NRWW

0x0000

End RWW, End Application

Start NRWW, Start Boot Loader

Application Flash SectionApplication Flash Section

Application Flash Section

Rea

d-W

hile

-Writ

e S

ectio

nN

o R

ead-

Whi

le-W

rite

Sec

tion

Rea

d-W

hile

-Writ

e S

ectio

nN

o R

ead-

Whi

le-W

rite

Sec

tion

Rea

d-W

hile

-Writ

e S

ectio

nN

o R

ead-

Whi

le-W

rite

Sec

tion

Rea

d-W

hile

-Writ

e S

ectio

nN

o R

ead-

Whi

le-W

rite

Sec

tion

End Application

Start Boot Loader

End Application

Start Boot Loader

End Application

Start Boot Loader

320 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. “1” means unprogrammed, “0” means programmed

Note: 1. “1” means unprogrammed, “0” means programmed

Entering the Boot Loader Program

Entering the Boot Loader takes place by a jump or call from the application program.This may be initiated by a trigger such as a command received via USART, or SPI inter-face. Alternatively, the Boot Reset Fuse can be programmed so that the Reset Vector ispointing to the Boot Flash start address after a reset. In this case, the Boot Loader isstarted after a reset. After the application code is loaded, the program can start execut-ing the application code. Note that the fuses cannot be changed by the MCU itself. Thismeans that once the Boot Reset Fuse is programmed, the Reset Vector will alwayspoint to the Boot Loader Reset and the fuse can only be changed through the serial orparallel programming interface.

Note: 1. “1” means unprogrammed, “0” means programmed

Table 134. Boot Lock Bit0 Protection Modes (Application Section)(1)

BLB0 Mode BLB02 BLB01 Protection

1 1 1 No restrictions for SPM or (E)LPM accessing the Application section.

2 1 0 SPM is not allowed to write to the Application section.

3 0 0 SPM is not allowed to write to the Application section, and (E)LPM executing from the Boot Loader section is not allowed to read from the Application section. If Interrupt Vectors are placed in the Boot Loader section, interrupts are disabled while executing from the Application section.

4 0 1 (E)LPM executing from the Boot Loader section is not allowed to read from the Application section. If Interrupt Vectors are placed in the Boot Loader section, interrupts are disabled while executing from the Application section.

Table 135. Boot Lock Bit1 Protection Modes (Boot Loader Section)(1)

BLB1 Mode BLB12 BLB11 Protection

1 1 1 No restrictions for SPM or (E)LPM accessing the Boot Loader section.

2 1 0 SPM is not allowed to write to the Boot Loader section.

3 0 0 SPM is not allowed to write to the Boot Loader section, and (E)LPM executing from the Application section is not allowed to read from the Boot Loader section. If Interrupt Vectors are placed in the Application section, interrupts are disabled while executing from the Boot Loader section.

4 0 1 (E)LPM executing from the Application section is not allowed to read from the Boot Loader section. If Interrupt Vectors are placed in the Application section, interrupts are disabled while executing from the Boot Loader section.

Table 136. Boot Reset Fuse(1)

BOOTRST Reset Address

1 Reset Vector = Application Reset (address 0x0000)

0 Reset Vector = Boot Loader Reset (see Table 139 on page 330)

321

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Store Program Memory Control and Status Register – SPMCSR

The Store Program Memory Control and Status Register contains the control bitsneeded to control the Boot Loader operations.

• Bit 7 – SPMIE: SPM Interrupt Enable

When the SPMIE bit is written to one, and the I-bit in the Status Register is set (one), theSPM ready interrupt will be enabled. The SPM ready Interrupt will be executed as longas the SPMEN bit in the SPMCSR Register is cleared.

• Bit 6 – RWWSB: Read-While-Write Section Busy

When a Self-Programming (Page Erase or Page Write) operation to the RWW section isinitiated, the RWWSB will be set (one) by hardware. When the RWWSB bit is set, theRWW section cannot be accessed. The RWWSB bit will be cleared if the RWWSRE bitis written to one after a Self-Programming operation is completed. Alternatively theRWWSB bit will automatically be cleared if a page load operation is initiated.

• Bit 5 – SIGRD: Signature Row Read

If this bit is written to one at the same time as SPMEN, the next LPM instruction withinthree clock cycles will read a byte from the signature row into the destination register.see “Reading the Signature Row from Software” on page 326 for details. An SPMinstruction within four cycles after SIGRD and SPMEN are set will have no effect. Thisoperation is reserved for future use and should not be used.

• Bit 4 – RWWSRE: Read-While-Write Section Read Enable

When programming (Page Erase or Page Write) to the RWW section, the RWW sectionis blocked for reading (the RWWSB will be set by hardware). To re-enable the RWWsection, the user software must wait until the programming is completed (SPMEN will becleared). Then, if the RWWSRE bit is written to one at the same time as SPMEN, thenext SPM instruction within four clock cycles re-enables the RWW section. The RWWsection cannot be re-enabled while the Flash is busy with a Page Erase or a Page Write(SPMEN is set). If the RWWSRE bit is written while the Flash is being loaded, the Flashload operation will abort and the data loaded will be lost.

• Bit 3 – BLBSET: Boot Lock Bit Set

If this bit is written to one at the same time as SPMEN, the next SPM instruction withinfour clock cycles sets Boot Lock bits, according to the data in R0. The data in R1 andthe address in the Z-pointer are ignored. The BLBSET bit will automatically be clearedupon completion of the Lock bit set, or if no SPM instruction is executed within four clockcycles.

An (E)LPM instruction within three cycles after BLBSET and SPMEN are set in theSPMCSR Register, will read either the Lock bits or the Fuse bits (depending on Z0 inthe Z-pointer) into the destination register. See “Reading the Fuse and Lock Bits fromSoftware” on page 325 for details.

• Bit 2 – PGWRT: Page Write

If this bit is written to one at the same time as SPMEN, the next SPM instruction withinfour clock cycles executes Page Write, with the data stored in the temporary buffer. The

Bit 7 6 5 4 3 2 1 0

SPMIE RWWSB SIGRD RWWSRE BLBSET PGWRT PGERS SPMEN SPMCSR

Read/Write R/W R R/W R/W R/W R/W R/W R/W

Initial Value 0 0 0 0 0 0 0 0

322 ATmega640/1280/1281/2560/25612549A–AVR–03/05

page address is taken from the high part of the Z-pointer. The data in R1 and R0 areignored. The PGWRT bit will auto-clear upon completion of a Page Write, or if no SPMinstruction is executed within four clock cycles. The CPU is halted during the entirePage Write operation if the NRWW section is addressed.

• Bit 1 – PGERS: Page Erase

If this bit is written to one at the same time as SPMEN, the next SPM instruction withinfour clock cycles executes Page Erase. The page address is taken from the high part ofthe Z-pointer. The data in R1 and R0 are ignored. The PGERS bit will auto-clear uponcompletion of a Page Erase, or if no SPM instruction is executed within four clockcycles. The CPU is halted during the entire Page Write operation if the NRWW section isaddressed.

• Bit 0 – SPMEN: Store Program Memory Enable

This bit enables the SPM instruction for the next four clock cycles. If written to onetogether with either RWWSRE, BLBSET, PGWRT’ or PGERS, the following SPMinstruction will have a special meaning, see description above. If only SPMEN is written,the following SPM instruction will store the value in R1:R0 in the temporary page bufferaddressed by the Z-pointer. The LSB of the Z-pointer is ignored. The SPMEN bit willauto-clear upon completion of an SPM instruction, or if no SPM instruction is executedwithin four clock cycles. During Page Erase and Page Write, the SPMEN bit remainshigh until the operation is completed.

Writing any other combination than “10001”, “01001”, “00101”, “00011” or “00001” in thelower five bits will have no effect.Note: Only one SPM instruction should be active at any time.

Addressing the Flash During Self-Programming

The Z-pointer is used to address the SPM commands. The Z pointer consists of the Z-registers ZL and ZH in the register file, and RAMPZ in the I/O space. The number of bitsactually used is implementation dependent. Note that the RAMPZ register is only imple-mented when the program space is larger than 64K bytes.

Since the Flash is organized in pages (see Table 158 on page 340), the ProgramCounter can be treated as having two different sections. One section, consisting of theleast significant bits, is addressing the words within a page, while the most significantbits are addressing the pages. This is shown in Figure 138. Note that the Page Eraseand Page Write operations are addressed independently. Therefore it is of major impor-tance that the Boot Loader software addresses the same page in both the Page Eraseand Page Write operation. Once a programming operation is initiated, the address islatched and the Z-pointer can be used for other operations.

The (E)LPM instruction use the Z-pointer to store the address. Since this instructionaddresses the Flash byte-by-byte, also bit Z0 of the Z-pointer is used.

Bit 23 22 21 20 19 18 17 16

15 14 13 12 11 10 9 8

RAMPZ RAMPZ7 RAMPZ6 RAMPZ5 RAMPZ4 RAMPZ3 RAMPZ2 RAMPZ1 RAMPZ0

ZH (R31) Z15 Z14 Z13 Z12 Z11 Z10 Z9 Z8

ZL (R30) Z7 Z6 Z5 Z4 Z3 Z2 Z1 Z0

7 6 5 4 3 2 1 0

323

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 138. Addressing the Flash During SPM(1)

Note: 1. The different variables used in Figure 138 are listed in Table 141 on page 330.

Self-Programming the Flash

The program memory is updated in a page by page fashion. Before programming apage with the data stored in the temporary page buffer, the page must be erased. Thetemporary page buffer is filled one word at a time using SPM and the buffer can be filledeither before the Page Erase command or between a Page Erase and a Page Writeoperation:

Alternative 1, fill the buffer before a Page Erase

• Fill temporary page buffer

• Perform a Page Erase

• Perform a Page Write

Alternative 2, fill the buffer after Page Erase

• Perform a Page Erase

• Fill temporary page buffer

• Perform a Page Write

If only a part of the page needs to be changed, the rest of the page must be stored (forexample in the temporary page buffer) before the erase, and then be rewritten. Whenusing alternative 1, the Boot Loader provides an effective Read-Modify-Write featurewhich allows the user software to first read the page, do the necessary changes, andthen write back the modified data. If alternative 2 is used, it is not possible to read theold data while loading since the page is already erased. The temporary page buffer canbe accessed in a random sequence. It is essential that the page address used in boththe Page Erase and Page Write operation is addressing the same page. See “SimpleAssembly Code Example for a Boot Loader” on page 328 for an assembly codeexample.

PROGRAM MEMORY

0115

Z - REGISTER

BIT

0

ZPAGEMSB

WORD ADDRESSWITHIN A PAGE

PAGE ADDRESSWITHIN THE FLASH

ZPCMSB

INSTRUCTION WORD

PAGE PCWORD[PAGEMSB:0]:

00

01

02

PAGEEND

PAGE

PCWORDPCPAGE

PCMSB PAGEMSBPROGRAMCOUNTER

324 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Performing Page Erase by SPM

To execute Page Erase, set up the address in the Z-pointer, write “X0000011” toSPMCSR and execute SPM within four clock cycles after writing SPMCSR. The data inR1 and R0 is ignored. The page address must be written to PCPAGE in the Z-register.Other bits in the Z-pointer will be ignored during this operation.

• Page Erase to the RWW section: The NRWW section can be read during the Page Erase.

• Page Erase to the NRWW section: The CPU is halted during the operation.

Filling the Temporary Buffer (Page Loading)

To write an instruction word, set up the address in the Z-pointer and data in R1:R0, write“00000001” to SPMCSR and execute SPM within four clock cycles after writingSPMCSR. The content of PCWORD in the Z-register is used to address the data in thetemporary buffer. The temporary buffer will auto-erase after a Page Write operation orby writing the RWWSRE bit in SPMCSR. It is also erased after a system reset. Note thatit is not possible to write more than one time to each address without erasing the tempo-rary buffer.

If the EEPROM is written in the middle of an SPM Page Load operation, all data loadedwill be lost.

Performing a Page Write To execute Page Write, set up the address in the Z-pointer, write “X0000101” toSPMCSR and execute SPM within four clock cycles after writing SPMCSR. The data inR1 and R0 is ignored. The page address must be written to PCPAGE. Other bits in theZ-pointer must be written to zero during this operation.

• Page Write to the RWW section: The NRWW section can be read during the Page Write.

• Page Write to the NRWW section: The CPU is halted during the operation.

Using the SPM Interrupt If the SPM interrupt is enabled, the SPM interrupt will generate a constant interruptwhen the SPMEN bit in SPMCSR is cleared. This means that the interrupt can be usedinstead of polling the SPMCSR Register in software. When using the SPM interrupt, theInterrupt Vectors should be moved to the BLS section to avoid that an interrupt isaccessing the RWW section when it is blocked for reading. How to move the interruptsis described in “Interrupts” on page 69.

Consideration While Updating BLS

Special care must be taken if the user allows the Boot Loader section to be updated byleaving Boot Lock bit11 unprogrammed. An accidental write to the Boot Loader itself cancorrupt the entire Boot Loader, and further software updates might be impossible. If it isnot necessary to change the Boot Loader software itself, it is recommended to programthe Boot Lock bit11 to protect the Boot Loader software from any internal softwarechanges.

Prevent Reading the RWW Section During Self-Programming

During Self-Programming (either Page Erase or Page Write), the RWW section isalways blocked for reading. The user software itself must prevent that this section isaddressed during the self programming operation. The RWWSB in the SPMCSR will beset as long as the RWW section is busy. During Self-Programming the Interrupt Vectortable should be moved to the BLS as described in “Interrupts” on page 69, or the inter-rupts must be disabled. Before addressing the RWW section after the programming iscompleted, the user software must clear the RWWSB by writing the RWWSRE. See“Simple Assembly Code Example for a Boot Loader” on page 328 for an example.

325

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Setting the Boot Loader Lock Bits by SPM

To set the Boot Loader Lock bits, write the desired data to R0, write “X0001001” toSPMCSR and execute SPM within four clock cycles after writing SPMCSR. The onlyaccessible Lock bits are the Boot Lock bits that may prevent the Application and BootLoader section from any software update by the MCU.

See Table 134 and Table 135 for how the different settings of the Boot Loader bits affectthe Flash access.

If bits 5..2 in R0 are cleared (zero), the corresponding Boot Lock bit will be programmedif an SPM instruction is executed within four cycles after BLBSET and SPMEN are set inSPMCSR. The Z-pointer is don’t care during this operation, but for future compatibility itis recommended to load the Z-pointer with 0x0001 (same as used for reading the lOckbits). For future compatibility it is also recommended to set bits 7, 6, 1, and 0 in R0 to “1”when writing the Lock bits. When programming the Lock bits the entire Flash can beread during the operation.

EEPROM Write Prevents Writing to SPMCSR

Note that an EEPROM write operation will block all software programming to Flash.Reading the Fuses and Lock bits from software will also be prevented during theEEPROM write operation. It is recommended that the user checks the status bit (EEPE)in the EECR Register and verifies that the bit is cleared before writing to the SPMCSRRegister.

Reading the Fuse and Lock Bits from Software

It is possible to read both the Fuse and Lock bits from software. To read the Lock bits,load the Z-pointer with 0x0001 and set the BLBSET and SPMEN bits in SPMCSR.When an (E)LPM instruction is executed within three CPU cycles after the BLBSET andSPMEN bits are set in SPMCSR, the value of the Lock bits will be loaded in the destina-tion register. The BLBSET and SPMEN bits will auto-clear upon completion of readingthe Lock bits or if no (E)LPM instruction is executed within three CPU cycles or no SPMinstruction is executed within four CPU cycles. When BLBSET and SPMEN are cleared,(E)LPM will work as described in the Instruction set Manual.

The algorithm for reading the Fuse Low byte is similar to the one described above forreading the Lock bits. To read the Fuse Low byte, load the Z-pointer with 0x0000 andset the BLBSET and SPMEN bits in SPMCSR. When an (E)LPM instruction is executedwithin three cycles after the BLBSET and SPMEN bits are set in the SPMCSR, the valueof the Fuse Low byte (FLB) will be loaded in the destination register as shown below.Refer to Table 152 on page 337 for a detailed description and mapping of the Fuse Lowbyte.

Similarly, when reading the Fuse High byte, load 0x0003 in the Z-pointer. When an(E)LPM instruction is executed within three cycles after the BLBSET and SPMEN bitsare set in the SPMCSR, the value of the Fuse High byte (FHB) will be loaded in the des-tination register as shown below. Refer to Table 151 on page 337 for detaileddescription and mapping of the Fuse High byte.

Bit 7 6 5 4 3 2 1 0

R0 1 1 BLB12 BLB11 BLB02 BLB01 1 1

Bit 7 6 5 4 3 2 1 0

Rd – – BLB12 BLB11 BLB02 BLB01 LB2 LB1

Bit 7 6 5 4 3 2 1 0

Rd FLB7 FLB6 FLB5 FLB4 FLB3 FLB2 FLB1 FLB0

Bit 7 6 5 4 3 2 1 0

Rd FHB7 FHB6 FHB5 FHB4 FHB3 FHB2 FHB1 FHB0

326 ATmega640/1280/1281/2560/25612549A–AVR–03/05

When reading the Extended Fuse byte, load 0x0002 in the Z-pointer. When an (E)LPMinstruction is executed within three cycles after the BLBSET and SPMEN bits are set inthe SPMCSR, the value of the Extended Fuse byte (EFB) will be loaded in the destina-tion register as shown below. Refer to Table 150 on page 336 for detailed descriptionand mapping of the Extended Fuse byte.

Fuse and Lock bits that are programmed, will be read as zero. Fuse and Lock bits thatare unprogrammed, will be read as one.

Reading the Signature Row from Software

To read the Signature Row from software, load the Z-pointer with the signature byteaddress given in Table 137 on page 326 and set the SIGRD and SPMEN bits inSPMCSR. When an LPM instruction is executed within three CPU cycles after theSIGRD and SPMEN bits are set in SPMCSR, the signature byte value will be loaded inthe destination register. The SIGRD and SPMEN bits will auto-clear upon completion ofreading the Signature Row Lock bits or if no LPM instruction is executed within threeCPU cycles. When SIGRD and SPMEN are cleared, LPM will work as described in theInstruction set Manual.

Note: All other addresses are reserved for future use.

Preventing Flash Corruption During periods of low VCC, the Flash program can be corrupted because the supply volt-age is too low for the CPU and the Flash to operate properly. These issues are the sameas for board level systems using the Flash, and the same design solutions should beapplied.

A Flash program corruption can be caused by two situations when the voltage is too low.First, a regular write sequence to the Flash requires a minimum voltage to operate cor-rectly. Secondly, the CPU itself can execute instructions incorrectly, if the supply voltagefor executing instructions is too low.

Flash corruption can easily be avoided by following these design recommendations (oneis sufficient):

1. If there is no need for a Boot Loader update in the system, program the Boot Loader Lock bits to prevent any Boot Loader software updates.

2. Keep the AVR RESET active (low) during periods of insufficient power supply voltage. This can be done by enabling the internal Brown-out Detector (BOD) if the operating voltage matches the detection level. If not, an external low VCC reset protection circuit can be used. If a reset occurs while a write operation is in progress, the write operation will be completed provided that the power supply voltage is sufficient.

3. Keep the AVR core in Power-down sleep mode during periods of low VCC. This will prevent the CPU from attempting to decode and execute instructions, effec-

Bit 7 6 5 4 3 2 1 0

Rd – – – – – EFB2 EFB1 EFB0

Table 137. Signature Row Addressing

Signature Byte Z-Pointer Address

Device Signature Byte 1 0x0000

Device Signature Byte 2 0x0002

Device Signature Byte 3 0x0004

RC Oscillator Calibration Byte 0x0001

327

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

tively protecting the SPMCSR Register and thus the Flash from unintentional writes.

Programming Time for Flash when Using SPM

The calibrated RC Oscillator is used to time Flash accesses. Table 138 shows the typi-cal programming time for Flash accesses from the CPU.

Table 138. SPM Programming Time

Symbol Min Programming Time Max Programming Time

Flash write (Page Erase, Page Write, and write Lock bits by SPM)

3.7 ms 4.5 ms

328 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Simple Assembly Code Example for a Boot Loader

;-the routine writes one page of data from RAM to Flash; the first data location in RAM is pointed to by the Y pointer; the first data location in Flash is pointed to by the Z-pointer;-error handling is not included;-the routine must be placed inside the Boot space; (at least the Do_spm sub routine). Only code inside NRWW section

can; be read during Self-Programming (Page Erase and Page Write).;-registers used: r0, r1, temp1 (r16), temp2 (r17), looplo (r24), ; loophi (r25), spmcrval (r20); storing and restoring of registers is not included in the routine; register usage can be optimized at the expense of code size;-It is assumed that either the interrupt table is moved to the

Boot; loader section or that the interrupts are disabled.

.equ PAGESIZEB = PAGESIZE*2 ;PAGESIZEB is page size in BYTES, not words.org SMALLBOOTSTARTWrite_page:; Page Eraseldi spmcrval, (1<<PGERS) | (1<<SPMEN)call Do_spm

; re-enable the RWW sectionldi spmcrval, (1<<RWWSRE) | (1<<SPMEN)call Do_spm

; transfer data from RAM to Flash page bufferldi looplo, low(PAGESIZEB) ;init loop variableldi loophi, high(PAGESIZEB) ;not required for PAGESIZEB<=256

Wrloop:ld r0, Y+ld r1, Y+ldi spmcrval, (1<<SPMEN)call Do_spmadiw ZH:ZL, 2sbiw loophi:looplo, 2 ;use subi for PAGESIZEB<=256brne Wrloop

; execute Page Writesubi ZL, low(PAGESIZEB) ;restore pointersbci ZH, high(PAGESIZEB) ;not required for PAGESIZEB<=256ldi spmcrval, (1<<PGWRT) | (1<<SPMEN)call Do_spm

; re-enable the RWW sectionldi spmcrval, (1<<RWWSRE) | (1<<SPMEN)call Do_spm

; read back and check, optionalldi looplo, low(PAGESIZEB) ;init loop variableldi loophi, high(PAGESIZEB) ;not required for PAGESIZEB<=256subi YL, low(PAGESIZEB) ;restore pointersbci YH, high(PAGESIZEB)

Rdloop:elpm r0, Z+ld r1, Y+cpse r0, r1jmp Error

329

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

sbiw loophi:looplo, 1 ;use subi for PAGESIZEB<=256brne Rdloop

; return to RWW section; verify that RWW section is safe to read

Return:in temp1, SPMCSRsbrs temp1, RWWSB ; If RWWSB is set, the RWW section is not

ready yetret; re-enable the RWW sectionldi spmcrval, (1<<RWWSRE) | (1<<SPMEN)call Do_spmrjmp Return

Do_spm:; check for previous SPM complete

Wait_spm:in temp1, SPMCSRsbrc temp1, SPMENrjmp Wait_spm; input: spmcrval determines SPM action; disable interrupts if enabled, store statusin temp2, SREGcli; check that no EEPROM write access is present

Wait_ee:sbic EECR, EEPErjmp Wait_ee; SPM timed sequenceout SPMCSR, spmcrvalspm; restore SREG (to enable interrupts if originally enabled)out SREG, temp2ret

330 ATmega640/1280/1281/2560/25612549A–AVR–03/05

ATmega640 Boot Loader Parameters

In Table 139 through Table 141, the parameters used in the description of the Self-Pro-gramming are given.

Note: 1. The different BOOTSZ Fuse configurations are shown in Figure 137.

Note: 1. For details about these two section, see “NRWW – No Read-While-Write Section” onpage 318 and “RWW – Read-While-Write Section” on page 318.

Table 139. Boot Size Configuration, ATmega640(1)

BO

OT

SZ

1

BO

OT

SZ

0

Bo

ot

Siz

e

Pag

es

Ap

pli-

cati

on

F

lash

Sec

tio

n

Bo

ot

Lo

ader

F

lash

Sec

tio

n

En

d A

pp

licat

ion

S

ecti

on

Bo

ot

Res

et A

dd

ress

(Sta

rt B

oo

t L

oad

erS

ecti

on

)

1 1512

words4

0x0000 - 0x7DFF

0x7E00 - 0x7FFF

0x7DFF 0x7E00

1 01024words

80x0000 - 0x7BFF

0x7C00 - 0x7FFF

0x7BFF 0x7C00

0 12048words

160x0000 - 0x77FF

0x7800 - 0x7FFF

0x77FF 0x7800

0 04096words

320x0000 - 0x6FFF

0x7000 - 0x7FFF

0x6FFF 0x7000

Table 140. Read-While-Write Limit, ATmega640

Section(1) Pages Address

Read-While-Write section (RWW) 224 0x0000 - 0x6FFF

No Read-While-Write section (NRWW) 32 0x7000 - 0x7FFF

Table 141. Explanation of different variables used in Figure 138 and the mapping to theZ-pointer, ATmega640

VariableCorresponding

Z-value(2) Description(1)

PCMSB 14 Most significant bit in the Program Counter. (The Program Counter is 15 bits PC[14:0])

331

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Notes: 1. Z0: should be zero for all SPM commands, byte select for the (E)LPM instruction.2. See “Addressing the Flash During Self-Programming” on page 322 for details about

the use of Z-pointer during Self-Programming.

ATmega1280/1281 Boot Loader Parameters

In Table 142 through Table 143, the parameters used in the description of the Self-Pro-gramming are given.

Note: 1. The different BOOTSZ Fuse configurations are shown in Figure 137.

PAGEMSB 6 Most significant bit which is used to address the words within one page (128 words in a page requires seven bits PC [6:0]).

ZPCMSB Z15 Bit in Z-pointer that is mapped to PCMSB. Because Z0 is not used, the ZPCMSB equals PCMSB + 1.

ZPAGEMSB Z7 Bit in Z-pointer that is mapped to PCMSB. Because Z0 is not used, the ZPAGEMSB equals PAGEMSB + 1.

PCPAGE PC[14:7] Z15:Z8 Program Counter page address: Page select, for Page Erase and Page Write

PCWORD PC[6:0] Z7:Z1 Program Counter word address: Word select, for filling temporary buffer (must be zero during Page Write operation)

Table 141. Explanation of different variables used in Figure 138 and the mapping to theZ-pointer, ATmega640

VariableCorresponding

Z-value(2) Description(1)

Table 142. Boot Size Configuration, ATmega1280/1281(1)

BO

OT

SZ

1

BO

OT

SZ

0

Bo

ot

Siz

e

Pag

es

Ap

pli-

cati

on

F

lash

Sec

tio

n

Bo

ot

Lo

ader

F

lash

Sec

tio

n

En

d A

pp

licat

ion

S

ecti

on

Bo

ot

Res

et A

dd

ress

(Sta

rt B

oo

t L

oad

erS

ecti

on

)

1 1512

words4

0x0000 - 0xFDFF

0xFE00 - 0xFFFF

0xFDFF 0xFE00

1 01024words

80x0000 - 0xFBFF

0xFC00 - 0xFFFF

0xFBFF 0xFC00

0 12048words

160x0000 - 0xF7FF

0xF800 - 0xFFFF

0xF7FF 0xF800

0 04096words

320x0000 - 0xEFFF

0xF000 - 0xFFFF

0xEFFF 0xF000

Table 143. Read-While-Write Limit, ATmega1280/1281

Section(1) Pages Address

Read-While-Write section (RWW) 480 0x0000 - 0xEFFF

No Read-While-Write section (NRWW) 32 0xF000 - 0xFFFF

332 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. For details about these two section, see “NRWW – No Read-While-Write Section” onpage 318 and “RWW – Read-While-Write Section” on page 318.

Notes: 1. Z0: should be zero for all SPM commands, byte select for the (E)LPM instruction.2. See “Addressing the Flash During Self-Programming” on page 322 for details about

the use of Z-pointer during Self-Programming.3. The Z-register is only 16 bits wide. Bit 16 is located in the RAMPZ register in the I/O

map.

ATmega2560/2561 Boot Loader Parameters

In Table 145 through Table 147, the parameters used in the description of the Self-Pro-gramming are given.

Table 144. Explanation of different variables used in Figure 138 and the mapping to theZ-pointer, ATmega1280/1281

VariableCorresponding

Z-value(2) Description(1)

PCMSB 15 Most significant bit in the Program Counter. (The Program Counter is 16 bits PC[15:0])

PAGEMSB 6 Most significant bit which is used to address the words within one page (128 words in a page requires seven bits PC [6:0]).

ZPCMSB Z16(3) Bit in Z-pointer that is mapped to PCMSB. Because Z0 is not used, the ZPCMSB equals PCMSB + 1.

ZPAGEMSB Z7 Bit in Z-pointer that is mapped to PCMSB. Because Z0 is not used, the ZPAGEMSB equals PAGEMSB + 1.

PCPAGE PC[15:7] Z16(3):Z8 Program Counter page address: Page select, for Page Erase and Page Write

PCWORD PC[6:0] Z7:Z1 Program Counter word address: Word select, for filling temporary buffer (must be zero during Page Write operation)

Table 145. Boot Size Configuration, ATmega2560/2561(1)

BO

OT

SZ

1

BO

OT

SZ

0

Bo

ot

Siz

e

Pag

es

Ap

pli-

cati

on

F

lash

Sec

tio

n

Bo

ot

Lo

ader

F

lash

Sec

tio

n

En

d A

pp

licat

ion

S

ecti

on

Bo

ot

Res

et A

dd

ress

(Sta

rt B

oo

t L

oad

erS

ecti

on

)

1 1512

words4

0x00000 - 0x1FDFF

0x1FE00 - 0x1FFFF

0x1FDFF 0x1FE00

1 01024words

80x00000 - 0x1FBFF

0x1FC00 - 0x1FFFF

0x1FBFF 0x1FC00

0 12048words

160x00000 - 0x1F7FF

0x1F800 - 0x1FFFF

0x1F7FF 0x1F800

0 04096words

320x00000 - 0x1EFFF

0x1F000 - 0x1FFFF

0x1EFFF 0x1F000

333

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: 1. The different BOOTSZ Fuse configurations are shown in Figure 137.

Note: 1. For details about these two section, see “NRWW – No Read-While-Write Section” onpage 318 and “RWW – Read-While-Write Section” on page 318.

Table 146. Read-While-Write Limit, ATmega2560/2561

Section(1) Pages Address

Read-While-Write section (RWW) 992 0x00000 - 0x1EFFF

No Read-While-Write section (NRWW) 32 0x1F000 - 0x1FFFF

334 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Notes: 1. Z0: should be zero for all SPM commands, byte select for the (E)LPM instruction.2. See “Addressing the Flash During Self-Programming” on page 322 for details about

the use of Z-pointer during Self-Programming.3. The Z-register is only 16 bits wide. Bit 16 is located in the RAMPZ register in the I/O

map.

Table 147. Explanation of different variables used in Figure 138 and the mapping to theZ-pointer, ATmega2560/2561

VariableCorresponding

Z-value(2) Description(1)

PCMSB 16 Most significant bit in the Program Counter. (The Program Counter is 17 bits PC[16:0])

PAGEMSB 6 Most significant bit which is used to address the words within one page (128 words in a page requires seven bits PC [6:0]).

ZPCMSB Z17:Z16(3) Bit in Z-pointer that is mapped to PCMSB. Because Z0 is not used, the ZPCMSB equals PCMSB + 1.

ZPAGEMSB Z7 Bit in Z-pointer that is mapped to PCMSB. Because Z0 is not used, the ZPAGEMSB equals PAGEMSB + 1.

PCPAGE PC[16:7] Z17(3):Z8 Program Counter page address: Page select, for Page Erase and Page Write

PCWORD PC[6:0] Z7:Z1 Program Counter word address: Word select, for filling temporary buffer (must be zero during Page Write operation)

335

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Memory Programming

Program And Data Memory Lock Bits

The ATmega640/1280/1281/2560/2561 provides six Lock bits which can be left unpro-grammed (“1”) or can be programmed (“0”) to obtain the additional features listed inTable 149. The Lock bits can only be erased to “1” with the Chip Erase command.

Note: 1. “1” means unprogrammed, “0” means programmed

Table 148. Lock Bit Byte(1)

Lock Bit Byte Bit No Description Default Value

7 – 1 (unprogrammed)

6 – 1 (unprogrammed)

BLB12 5 Boot Lock bit 1 (unprogrammed)

BLB11 4 Boot Lock bit 1 (unprogrammed)

BLB02 3 Boot Lock bit 1 (unprogrammed)

BLB01 2 Boot Lock bit 1 (unprogrammed)

LB2 1 Lock bit 1 (unprogrammed)

LB1 0 Lock bit 1 (unprogrammed)

Table 149. Lock Bit Protection Modes(1)(2)

Memory Lock Bits Protection Type

LB Mode LB2 LB1

1 1 1 No memory lock features enabled.

2 1 0

Further programming of the Flash and EEPROM is disabled in Parallel and Serial Programming mode. The Fuse bits are locked in both Serial and Parallel Programming mode.(1)

3 0 0

Further programming and verification of the Flash and EEPROM is disabled in Parallel and Serial Programming mode. The Boot Lock bits and Fuse bits are locked in both Serial and Parallel Programming mode.(1)

BLB0 Mode BLB02 BLB01

1 1 1No restrictions for SPM or (E)LPM accessing the Application section.

2 1 0 SPM is not allowed to write to the Application section.

3 0 0

SPM is not allowed to write to the Application section, and (E)LPM executing from the Boot Loader section is not allowed to read from the Application section. If Interrupt Vectors are placed in the Boot Loader section, interrupts are disabled while executing from the Application section.

4 0 1

(E)LPM executing from the Boot Loader section is not allowed to read from the Application section. If Interrupt Vectors are placed in the Boot Loader section, interrupts are disabled while executing from the Application section.

336 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Notes: 1. Program the Fuse bits and Boot Lock bits before programming the LB1 and LB2.2. “1” means unprogrammed, “0” means programmed

Fuse Bits The ATmega640/1280/1281/2560/2561 has four Fuse bytes. Table 150 - Table 152describe briefly the functionality of all the fuses and how they are mapped into the Fusebytes. Note that the fuses are read as logical zero, “0”, if they are programmed.

Note: 1. See Table 24 on page 60 for BODLEVEL Fuse decoding.

BLB1 Mode BLB12 BLB11

1 1 1No restrictions for SPM or (E)LPM accessing the Boot Loader section.

2 1 0 SPM is not allowed to write to the Boot Loader section.

3 0 0

SPM is not allowed to write to the Boot Loader section, and (E)LPM executing from the Application section is not allowed to read from the Boot Loader section. If Interrupt Vectors are placed in the Application section, interrupts are disabled while executing from the Boot Loader section.

4 0 1

(E)LPM executing from the Application section is not allowed to read from the Boot Loader section. If Interrupt Vectors are placed in the Application section, interrupts are disabled while executing from the Boot Loader section.

Table 149. Lock Bit Protection Modes(1)(2) (Continued)

Memory Lock Bits Protection Type

Table 150. Extended Fuse Byte

Fuse Low Byte Bit No Description Default Value

– 7 – 1

– 6 – 1

– 5 – 1

– 4 – 1

– 3 – 1

BODLEVEL2(1) 2 Brown-out Detector trigger level 1 (unprogrammed)

BODLEVEL1(1) 1 Brown-out Detector trigger level 1 (unprogrammed)

BODLEVEL0(1) 0 Brown-out Detector trigger level 1 (unprogrammed)

337

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: 1. The SPIEN Fuse is not accessible in serial programming mode.

2. The default value of BOOTSZ1..0 results in maximum Boot Size. See Table 139 onpage 330 for details.

3. See “Watchdog Timer Control Register - WDTCSR” on page 67 for details.4. Never ship a product with the OCDEN Fuse programmed regardless of the setting of

Lock bits and JTAGEN Fuse. A programmed OCDEN Fuse enables some parts of theclock system to be running in all sleep modes. This may increase the powerconsumption.

Note: 1. The default value of SUT1..0 results in maximum start-up time for the default clocksource. See Table 23 on page 58 for details.

2. The default setting of CKSEL3..0 results in internal RC Oscillator @ 8 MHz. SeeTable 7 on page 40 for details.

3. The CKOUT Fuse allow the system clock to be output on PORTE7. See “Clock Out-put Buffer” on page 48 for details.

4. See “System Clock Prescaler” on page 48 for details.

The status of the Fuse bits is not affected by Chip Erase. Note that the Fuse bits arelocked if Lock bit1 (LB1) is programmed. Program the Fuse bits before programming theLock bits.

Table 151. Fuse High Byte

Fuse High Byte Bit No Description Default Value

OCDEN(4) 7Enable OCD

1 (unprogrammed, OCD disabled)

JTAGEN 6Enable JTAG

0 (programmed, JTAG enabled)

SPIEN(1) 5Enable Serial Program and Data Downloading

0 (programmed, SPI prog. enabled)

WDTON(3) 4 Watchdog Timer always on 1 (unprogrammed)

EESAVE 3EEPROM memory is preserved through the Chip Erase

1 (unprogrammed, EEPROM not preserved)

BOOTSZ1 2Select Boot Size (see Table 153 for details) 0 (programmed)(2)

BOOTSZ0 1Select Boot Size (see Table 153 for details) 0 (programmed)(2)

BOOTRST 0 Select Reset Vector 1 (unprogrammed)

Table 152. Fuse Low Byte

Fuse Low Byte Bit No Description Default Value

CKDIV8(4) 7 Divide clock by 8 0 (programmed)

CKOUT(3) 6 Clock output 1 (unprogrammed)

SUT1 5 Select start-up time 1 (unprogrammed)(1)

SUT0 4 Select start-up time 0 (programmed)(1)

CKSEL3 3 Select Clock source 0 (programmed)(2)

CKSEL2 2 Select Clock source 0 (programmed)(2)

CKSEL1 1 Select Clock source 1 (unprogrammed)(2)

CKSEL0 0 Select Clock source 0 (programmed)(2)

338 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Latching of Fuses The fuse values are latched when the device enters programming mode and changes ofthe fuse values will have no effect until the part leaves Programming mode. This doesnot apply to the EESAVE Fuse which will take effect once it is programmed. The fusesare also latched on Power-up in Normal mode.

Signature Bytes All Atmel microcontrollers have a three-byte signature code which identifies the device.This code can be read in both serial and parallel mode, also when the device is locked.The three bytes reside in a separate address space.

ATmega640 Signature Bytes:

1. 0x000: 0x1E (indicates manufactured by Atmel).

2. 0x001: 0x96 (indicates 64K bytes Flash memory).

3. 0x002: 0x07 (indicates ATmega640 device when 0x001 is 0x96).

ATmega1280 Signature Bytes:

1. 0x000: 0x1E (indicates manufactured by Atmel).

2. 0x001: 0x97 (indicates 128K bytes Flash memory).

3. 0x002: 0x03 (indicates ATmega1280 device when 0x001 is 0x97).

ATmega1281 Signature Bytes:

1. 0x000: 0x1E (indicates manufactured by Atmel).

2. 0x001: 0x97 (indicates 128K bytes Flash memory).

3. 0x002: 0x04 (indicates ATmega1281 device when 0x001 is 0x97).

ATmega2560 Signature Bytes:

1. 0x000: 0x1E (indicates manufactured by Atmel).

2. 0x001: 0x98 (indicates 256K bytes Flash memory).

3. 0x002: 0x01 (indicates ATmega2560 device when 0x001 is 0x98).

ATmega2561 Signature Bytes:

1. 0x000: 0x1E (indicates manufactured by Atmel).

2. 0x001: 0x98 (indicates 256K bytes Flash memory).

3. 0x002: 0x02 (indicates ATmega2561 device when 0x001 is 0x98).

Calibration Byte The ATmega640/1280/1281/2560/2561 has a byte calibration value for the internal RCOscillator. This byte resides in the high byte of address 0x000 in the signature addressspace. During reset, this byte is automatically written into the OSCCAL Register toensure correct frequency of the calibrated RC Oscillator.

Parallel Programming Parameters, Pin Mapping, and Commands

This section describes how to parallel program and verify Flash Program memory,EEPROM Data memory , Memory Lock b i t s , and Fuse b i t s in theATmega640/1280/1281/2560/2561. Pulses are assumed to be at least 250 ns unlessotherwise noted.

Signal Names In this section, some pins of the ATmega640/1280/1281/2560/2561 are referenced bysignal names describing their functionality during parallel programming, see Figure 139and Table 153. Pins not described in the following table are referenced by pin names.

The XA1/XA0 pins determine the action executed when the XTAL1 pin is given a posi-tive pulse. The bit coding is shown in Table 156.

339

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

When pulsing WR or OE, the command loaded determines the action executed. The dif-ferent commands are shown in Table 157.

Figure 139. Parallel Programming(1)

Note: 1. Unused Pins should be left floating.

Table 153. Pin Name Mapping

Signal Name in Programming Mode Pin Name I/O Function

RDY/BSY PD1 O0: Device is busy programming, 1: Device is ready for new command.

OE PD2 I Output Enable (Active low).

WR PD3 I Write Pulse (Active low).

BS1 PD4 I Byte Select 1.

XA0 PD5 I XTAL Action Bit 0

XA1 PD6 I XTAL Action Bit 1

PAGEL PD7 I Program Memory and EEPROM data Page Load.

BS2 PA0 I Byte Select 2.

DATA PB7-0 I/O Bi-directional Data bus (Output when OE is low).

Table 154. BS2 and BS1 Encoding

BS2 BS1

Flash / EEPROM Address

Flash Data Loading / Reading

Fuse Programming

Reading Fuse and Lock Bits

0 0 Low Byte Low Byte Low Byte Fuse Low Byte

0 1 High Byte High Byte High Byte Lockbits

1 0 Extended High Byte

Reserved Extended Byte Extended Fuse Byte

1 1 Reserved Reserved Reserved Fuse High Byte

VCC

+5V

GND

XTAL1

PD1

PD2

PD3

PD4

PD5

PD6

PB7 - PB0 DATA

RESET

PD7

+12 V

BS1

XA0

XA1

OE

RDY/BSY

PAGEL

PA0

WR

BS2

AVCC

+5V

340 ATmega640/1280/1281/2560/25612549A–AVR–03/05

,Table 155. Pin Values Used to Enter Programming Mode

Pin Symbol Value

PAGEL Prog_enable[3] 0

XA1 Prog_enable[2] 0

XA0 Prog_enable[1] 0

BS1 Prog_enable[0] 0

Table 156. XA1 and XA0 Enoding

XA1 XA0 Action when XTAL1 is Pulsed

0 0 Load Flash or EEPROM Address (High or low address byte determined by BS2 and BS1).

0 1 Load Data (High or Low data byte for Flash determined by BS1).

1 0 Load Command

1 1 No Action, Idle

Table 157. Command Byte Bit Encoding

Command Byte Command Executed

1000 0000 Chip Erase

0100 0000 Write Fuse bits

0010 0000 Write Lock bits

0001 0000 Write Flash

0001 0001 Write EEPROM

0000 1000 Read Signature Bytes and Calibration byte

0000 0100 Read Fuse and Lock bits

0000 0010 Read Flash

0000 0011 Read EEPROM

Table 158. No. of Words in a Page and No. of Pages in the Flash

Flash Size Page Size PCWORDNo. of Pages PCPAGE PCMSB

128K words (256K bytes) 128 words PC[6:0] 1024 PC[16:7] 16

Table 159. No. of Words in a Page and No. of Pages in the EEPROM

EEPROM Size Page Size PCWORDNo. of Pages PCPAGE EEAMSB

4K bytes 8 bytes EEA[2:0] 512 EEA[11:3] 11

341

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Parallel Programming

Enter Programming Mode The following algorithm puts the device in parallel programming mode:

1. Apply 4.5 - 5.5V between VCC and GND.

2. Set RESET to “0” and toggle XTAL1 at least six times.

3. Set the Prog_enable pins listed in Table 155 on page 340 to “0000” and wait at least 100 ns.

4. Apply 11.5 - 12.5V to RESET. Any activity on Prog_enable pins within 100 ns after +12V has been applied to RESET, will cause the device to fail entering pro-gramming mode.

5. Wait at least 50 µs before sending a new command.

Considerations for Efficient Programming

The loaded command and address are retained in the device during programming. Forefficient programming, the following should be considered.

• The command needs only be loaded once when writing or reading multiple memory locations.

• Skip writing the data value 0xFF, that is the contents of the entire EEPROM (unless the EESAVE Fuse is programmed) and Flash after a Chip Erase.

• Address high byte needs only be loaded before programming or reading a new 256 word window in Flash or 256 byte EEPROM. This consideration also applies to Signature bytes reading.

Chip Erase The Chip Erase will erase the Flash and EEPROM(1) memories plus Lock bits. The Lockbits are not reset until the program memory has been completely erased. The Fuse bitsare not changed. A Chip Erase must be performed before the Flash and/or EEPROMare reprogrammed.Note: 1. The EEPRPOM memory is preserved during Chip Erase if the EESAVE Fuse is

programmed.

Load Command “Chip Erase”

1. Set XA1, XA0 to “10”. This enables command loading.

2. Set BS1 to “0”.

3. Set DATA to “1000 0000”. This is the command for Chip Erase.

4. Give XTAL1 a positive pulse. This loads the command.

5. Give WR a negative pulse. This starts the Chip Erase. RDY/BSY goes low.

6. Wait until RDY/BSY goes high before loading a new command.

Programming the Flash The Flash is organized in pages, see Table 158 on page 340. When programming theFlash, the program data is latched into a page buffer. This allows one page of programdata to be programmed simultaneously. The following procedure describes how to pro-gram the entire Flash memory:

A. Load Command “Write Flash”

1. Set XA1, XA0 to “10”. This enables command loading.

2. Set BS1 to “0”.

3. Set DATA to “0001 0000”. This is the command for Write Flash.

4. Give XTAL1 a positive pulse. This loads the command.

B. Load Address Low byte (Address bits 7..0)

342 ATmega640/1280/1281/2560/25612549A–AVR–03/05

1. Set XA1, XA0 to “00”. This enables address loading.

2. Set BS2, BS1 to “00”. This selects the address low byte.

3. Set DATA = Address low byte (0x00 - 0xFF).

4. Give XTAL1 a positive pulse. This loads the address low byte.

C. Load Data Low Byte

1. Set XA1, XA0 to “01”. This enables data loading.

2. Set DATA = Data low byte (0x00 - 0xFF).

3. Give XTAL1 a positive pulse. This loads the data byte.

D. Load Data High Byte

1. Set BS1 to “1”. This selects high data byte.

2. Set XA1, XA0 to “01”. This enables data loading.

3. Set DATA = Data high byte (0x00 - 0xFF).

4. Give XTAL1 a positive pulse. This loads the data byte.

E. Latch Data

1. Set BS1 to “1”. This selects high data byte.

2. Give PAGEL a positive pulse. This latches the data bytes. (See Figure 141 for signal waveforms)

F. Repeat B through E until the entire buffer is filled or until all data within the page isloaded.

While the lower bits in the address are mapped to words within the page, the higher bitsaddress the pages within the FLASH. This is illustrated in Figure 140 on page 343. Notethat if less than eight bits are required to address words in the page (pagesize < 256),the most significant bit(s) in the address low byte are used to address the page whenperforming a Page Write.

G. Load Address High byte (Address bits15..8)

1. Set XA1, XA0 to “00”. This enables address loading.

2. Set BS2, BS1 to “01”. This selects the address high byte.

3. Set DATA = Address high byte (0x00 - 0xFF).

4. Give XTAL1 a positive pulse. This loads the address high byte.

H. Load Address Extended High byte (Address bits 23..16)

1. Set XA1, XA0 to “00”. This enables address loading.

2. Set BS2, BS1 to “10”. This selects the address extended high byte.

3. Set DATA = Address extended high byte (0x00 - 0xFF).

4. Give XTAL1 a positive pulse. This loads the address high byte.

I. Program Page

1. Set BS2, BS1 to “00”

2. Give WR a negative pulse. This starts programming of the entire page of data. RDY/BSY goes low.

3. Wait until RDY/BSY goes high (See Figure 141 for signal waveforms).

J. Repeat B through I until the entire Flash is programmed or until all data has beenprogrammed.

K. End Page Programming

343

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

1. 1. Set XA1, XA0 to “10”. This enables command loading.

2. Set DATA to “0000 0000”. This is the command for No Operation.

3. Give XTAL1 a positive pulse. This loads the command, and the internal write sig-nals are reset.

Figure 140. Addressing the Flash Which is Organized in Pages(1)

Note: 1. PCPAGE and PCWORD are listed in Table 158 on page 340.

Figure 141. Programming the Flash Waveforms(1)

Note: 1. “XX” is don’t care. The letters refer to the programming description above.

Programming the EEPROM The EEPROM is organized in pages, see Table 159 on page 340. When programmingthe EEPROM, the program data is latched into a page buffer. This allows one page ofdata to be programmed simultaneously. The programming algorithm for the EEPROMdata memory is as follows (refer to “Programming the Flash” on page 341 for details onCommand, Address and Data loading):

PROGRAM MEMORY

WORD ADDRESSWITHIN A PAGE

PAGE ADDRESSWITHIN THE FLASH

INSTRUCTION WORD

PAGE PCWORD[PAGEMSB:0]:

00

01

02

PAGEEND

PAGE

PCWORDPCPAGE

PCMSB PAGEMSBPROGRAMCOUNTER

RDY/BSY

WR

OE

RESET +12V

PAGEL

BS2

0x10 ADDR. LOW ADDR. HIGHDATADATA LOW DATA HIGH ADDR. LOW DATA LOW DATA HIGH

XA1

XA0

BS1

XTAL1

XX XX XX

A B C D E B C D E G

F

ADDR. EXT.H

H I

344 ATmega640/1280/1281/2560/25612549A–AVR–03/05

1. A: Load Command “0001 0001”.

2. G: Load Address High Byte (0x00 - 0xFF).

3. B: Load Address Low Byte (0x00 - 0xFF).

4. C: Load Data (0x00 - 0xFF).

5. E: Latch data (give PAGEL a positive pulse).

K: Repeat 3 through 5 until the entire buffer is filled.

L: Program EEPROM page

1. Set BS2, BS1 to “00”.

2. Give WR a negative pulse. This starts programming of the EEPROM page. RDY/BSY goes low.

3. Wait until to RDY/BSY goes high before programming the next page (See Figure 142 for signal waveforms).

Figure 142. Programming the EEPROM Waveforms

Reading the Flash The algorithm for reading the Flash memory is as follows (refer to “Programming theFlash” on page 341 for details on Command and Address loading):

1. A: Load Command “0000 0010”.

2. H: Load Address Extended Byte (0x00- 0xFF).

3. G: Load Address High Byte (0x00 - 0xFF).

4. B: Load Address Low Byte (0x00 - 0xFF).

5. Set OE to “0”, and BS1 to “0”. The Flash word low byte can now be read at DATA.

6. Set BS to “1”. The Flash word high byte can now be read at DATA.

7. Set OE to “1”.

Reading the EEPROM The algorithm for reading the EEPROM memory is as follows (refer to “Programming theFlash” on page 341 for details on Command and Address loading):

1. A: Load Command “0000 0011”.

2. G: Load Address High Byte (0x00 - 0xFF).

3. B: Load Address Low Byte (0x00 - 0xFF).

RDY/BSY

WR

OE

RESET +12V

PAGEL

BS2

0x11 ADDR. HIGHDATA

ADDR. LOW DATA ADDR. LOW DATA XX

XA1

XA0

BS1

XTAL1

XX

A G B C E B C E L

K

345

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

4. Set OE to “0”, and BS1 to “0”. The EEPROM Data byte can now be read at DATA.

5. Set OE to “1”.

Programming the Fuse Low Bits

The algorithm for programming the Fuse Low bits is as follows (refer to “Programmingthe Flash” on page 341 for details on Command and Data loading):

1. A: Load Command “0100 0000”.

2. C: Load Data Low Byte. Bit n = “0” programs and bit n = “1” erases the Fuse bit.

3. Give WR a negative pulse and wait for RDY/BSY to go high.

Programming the Fuse High Bits

The algorithm for programming the Fuse High bits is as follows (refer to “Programmingthe Flash” on page 341 for details on Command and Data loading):

1. A: Load Command “0100 0000”.

2. C: Load Data Low Byte. Bit n = “0” programs and bit n = “1” erases the Fuse bit.

3. Set BS2, BS1 to “01”. This selects high data byte.

4. Give WR a negative pulse and wait for RDY/BSY to go high.

5. Set BS2, BS1 to “00”. This selects low data byte.

Programming the Extended Fuse Bits

The algorithm for programming the Extended Fuse bits is as follows (refer to “Program-ming the Flash” on page 341 for details on Command and Data loading):

1. 1. A: Load Command “0100 0000”.

2. 2. C: Load Data Low Byte. Bit n = “0” programs and bit n = “1” erases the Fuse bit.

3. 3. Set BS2, BS1 to “10”. This selects extended data byte.

4. 4. Give WR a negative pulse and wait for RDY/BSY to go high.

5. 5. Set BS2, BS1 to “00”. This selects low data byte.

Figure 143. Programming the FUSES Waveforms

RDY/BSY

WR

OE

RESET +12V

PAGEL

0x40DATA

DATA XX

XA1

XA0

BS1

XTAL1

A C

0x40 DATA XX

A C

Write Fuse Low byte Write Fuse high byte

0x40 DATA XX

A C

Write Extended Fuse byte

BS2

346 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Programming the Lock Bits The algorithm for programming the Lock bits is as follows (refer to “Programming theFlash” on page 341 for details on Command and Data loading):

1. A: Load Command “0010 0000”.

2. C: Load Data Low Byte. Bit n = “0” programs the Lock bit. If LB mode 3 is pro-grammed (LB1 and LB2 is programmed), it is not possible to program the Boot Lock bits by any External Programming mode.

3. Give WR a negative pulse and wait for RDY/BSY to go high.

The Lock bits can only be cleared by executing Chip Erase.

Reading the Fuse and Lock Bits

The algorithm for reading the Fuse and Lock bits is as follows (refer to “Programmingthe Flash” on page 341 for details on Command loading):

1. A: Load Command “0000 0100”.

2. Set OE to “0”, and BS2, BS1 to “00”. The status of the Fuse Low bits can now be read at DATA (“0” means programmed).

3. Set OE to “0”, and BS2, BS1 to “11”. The status of the Fuse High bits can now be read at DATA (“0” means programmed).

4. Set OE to “0”, and BS2, BS1 to “10”. The status of the Extended Fuse bits can now be read at DATA (“0” means programmed).

5. Set OE to “0”, and BS2, BS1 to “01”. The status of the Lock bits can now be read at DATA (“0” means programmed).

6. Set OE to “1”.

Figure 144. Mapping Between BS1, BS2 and the Fuse and Lock Bits During Read

Reading the Signature Bytes The algorithm for reading the Signature bytes is as follows (refer to “Programming theFlash” on page 341 for details on Command and Address loading):

1. A: Load Command “0000 1000”.

2. B: Load Address Low Byte (0x00 - 0x02).

3. Set OE to “0”, and BS to “0”. The selected Signature byte can now be read at DATA.

4. Set OE to “1”.

Lock Bits 0

1

BS2

Fuse High Byte

0

1

BS1

DATA

Fuse Low Byte 0

1

BS2

Extended Fuse Byte

347

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Reading the Calibration Byte The algorithm for reading the Calibration byte is as follows (refer to “Programming theFlash” on page 341 for details on Command and Address loading):

1. A: Load Command “0000 1000”.

2. B: Load Address Low Byte, 0x00.

3. Set OE to “0”, and BS1 to “1”. The Calibration byte can now be read at DATA.

4. Set OE to “1”.

Parallel Programming Characteristics

Figure 145. Parallel Programming Timing, Including some General TimingRequirements

Figure 146. Parallel Programming Timing, Loading Sequence with TimingRequirements(1)

Note: 1. The timing requirements shown in Figure 145 (i.e., tDVXH, tXHXL, and tXLDX) also applyto loading operation.

Data & Contol(DATA, XA0/1, BS1, BS2)

XTAL1tXHXL

tWLWH

tDVXH tXLDX

tPLWL

tWLRH

WR

RDY/BSY

PAGEL tPHPL

tPLBXtBVPH

tXLWL

tWLBXtBVWL

WLRL

XTAL1

PAGEL

tPLXHXLXHt tXLPH

ADDR0 (Low Byte) DATA (Low Byte) DATA (High Byte) ADDR1 (Low Byte)DATA

BS1

XA0

XA1

LOAD ADDRESS(LOW BYTE)

LOAD DATA (LOW BYTE)

LOAD DATA(HIGH BYTE)

LOAD DATA LOAD ADDRESS(LOW BYTE)

348 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 147. Parallel Programming Timing, Reading Sequence (within the Same Page)with Timing Requirements(1)

Note: 1. The timing requirements shown in Figure 145 (i.e., tDVXH, tXHXL, and tXLDX) also applyto reading operation.

Table 160. Parallel Programming Characteristics, VCC = 5V ± 10%

Symbol Parameter Min Typ Max Units

VPP Programming Enable Voltage 11.5 12.5 V

IPP Programming Enable Current 250 µA

tDVXH Data and Control Valid before XTAL1 High 67 ns

tXLXH XTAL1 Low to XTAL1 High 200 ns

tXHXL XTAL1 Pulse Width High 150 ns

tXLDX Data and Control Hold after XTAL1 Low 67 ns

tXLWL XTAL1 Low to WR Low 0 ns

tXLPH XTAL1 Low to PAGEL high 0 ns

tPLXH PAGEL low to XTAL1 high 150 ns

tBVPH BS1 Valid before PAGEL High 67 ns

tPHPL PAGEL Pulse Width High 150 ns

tPLBX BS1 Hold after PAGEL Low 67 ns

tWLBX BS2/1 Hold after WR Low 67 ns

tPLWL PAGEL Low to WR Low 67 ns

tBVWL BS2/1 Valid to WR Low 67 ns

tWLWH WR Pulse Width Low 150 ns

tWLRL WR Low to RDY/BSY Low 0 1 µs

tWLRH WR Low to RDY/BSY High(1) 3.7 4.5 ms

tWLRH_CE WR Low to RDY/BSY High for Chip Erase(2) 7.5 9 ms

tXLOL XTAL1 Low to OE Low 0 ns

XTAL1

OE

ADDR0 (Low Byte) DATA (Low Byte) DATA (High Byte) ADDR1 (Low Byte)DATA

BS1

XA0

XA1

LOAD ADDRESS(LOW BYTE)

READ DATA (LOW BYTE)

READ DATA(HIGH BYTE)

LOAD ADDRESS(LOW BYTE)

tBVDV

tOLDV

tXLOL

tOHDZ

349

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Notes: 1. tWLRH is valid for the Write Flash, Write EEPROM, Write Fuse bits and Write Lockbits commands.

2. tWLRH_CE is valid for the Chip Erase command.

Serial Downloading Both the Flash and EEPROM memory arrays can be programmed using a serial pro-gramming bus while RESET is pulled to GND. The serial programming interfaceconsists of pins SCK, PDI (input) and PDO (output). After RESET is set low, the Pro-gramming Enable instruction needs to be executed first before program/eraseoperations can be executed. NOTE, in Table 161 on page 349, the pin mapping forserial programming is listed. Not all packages use the SPI pins dedicated for the internalSerial Peripheral Interface - SPI.

Serial Programming Pin Mapping

Figure 148. Serial Programming and Verify(1)

Notes: 1. If the device is clocked by the internal Oscillator, it is no need to connect a clocksource to the XTAL1 pin.

2. VCC - 0.3V < AVCC < VCC + 0.3V, however, AVCC should always be within 1.8 - 5.5V

tBVDV BS1 Valid to DATA valid 0 250 ns

tOLDV OE Low to DATA Valid 250 ns

tOHDZ OE High to DATA Tri-stated 250 ns

Table 160. Parallel Programming Characteristics, VCC = 5V ± 10% (Continued)

Symbol Parameter Min Typ Max Units

Table 161. Pin Mapping Serial Programming

SymbolPins

(TQFP-100)Pins

(TQFP-64) I/O Description

PDI PB2 PE0 I Serial Data in

PDO PB3 PE1 O Serial Data out

SCK PB1 PB1 I Serial Clock

VCC

GND

XTAL1

SCK

PDO

PDI

RESET

+1.8 - 5.5V

AVCC

+1.8 - 5.5V(2)

350 ATmega640/1280/1281/2560/25612549A–AVR–03/05

When programming the EEPROM, an auto-erase cycle is built into the self-timed pro-gramming operation (in the Serial mode ONLY) and there is no need to first execute theChip Erase instruction. The Chip Erase operation turns the content of every memorylocation in both the Program and EEPROM arrays into 0xFF.

Depending on CKSEL Fuses, a valid clock must be present. The minimum low and highperiods for the serial clock (SCK) input are defined as follows:

Low:> 2 CPU clock cycles for fck < 12 MHz, 3 CPU clock cycles for fck >= 12 MHz

High:> 2 CPU clock cycles for fck < 12 MHz, 3 CPU clock cycles for fck >= 12 MHz

Serial Programming Algorithm

When writing serial data to the ATmega640/1280/1281/2560/2561, data is clocked onthe rising edge of SCK.

When reading data from the ATmega640/1280/1281/2560/2561, data is clocked on thefalling edge of SCK. See Figure 149 for timing details.

To program and verify the ATmega640/1280/1281/2560/2561 in the serial programmingmode, the following sequence is recommended (See four byte instruction formats inTable 163):1. Power-up sequence:

Apply power between VCC and GND while RESET and SCK are set to “0”. In some systems, the programmer can not guarantee that SCK is held low during power-up. In this case, RESET must be given a positive pulse of at least two CPU clock cycles duration after SCK has been set to “0”.

2. Wait for at least 20 ms and enable serial programming by sending the Program-ming Enable serial instruction to pin PDI.

3. The serial programming instructions will not work if the communication is out of synchronization. When in sync. the second byte (0x53), will echo back when issuing the third byte of the Programming Enable instruction. Whether the echo is correct or not, all four bytes of the instruction must be transmitted. If the 0x53 did not echo back, give RESET a positive pulse and issue a new Programming Enable command.

4. The Flash is programmed one page at a time. The memory page is loaded one byte at a time by supplying the 7 LSB of the address and data together with the Load Program Memory Page instruction. To ensure correct loading of the page, the data low byte must be loaded before data high byte is applied for a given address. The Program Memory Page is stored by loading the Write Program Memory Page instruction with the address lines 15..8. Before issuing this com-mand, make sure the instruction Load Extended Address Byte has been used to define the MSB of the address. The extended address byte is stored until the command is re-issued, i.e., the command needs only be issued for the first page, and when crossing the 64KWord boundary. If polling (RDY/BSY) is not used, the user must wait at least tWD_FLASH before issuing the next page. (See Table 162.) Accessing the serial programming interface before the Flash write operation completes can result in incorrect programming.

5. The EEPROM array is programmed one byte at a time by supplying the address and data together with the appropriate Write instruction. An EEPROM memory location is first automatically erased before new data is written. If polling is not used, the user must wait at least tWD_EEPROM before issuing the next byte. (See Table 162.) In a chip erased device, no 0xFFs in the data file(s) need to be programmed.

6. Any memory location can be verified by using the Read instruction which returns the content at the selected address at serial output PDO. When reading the

351

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Flash memory, use the instruction Load Extended Address Byte to define the upper address byte, which is not included in the Read Program Memory instruc-tion. The extended address byte is stored until the command is re-issued, i.e., the command needs only be issued for the first page, and when crossing the 64KWord boundary.

7. At the end of the programming session, RESET can be set high to commence normal operation.

8. Power-off sequence (if needed):Set RESET to “1”.Turn VCC power off.

Figure 149. Serial Programming Waveforms

Table 162. Minimum Wait Delay Before Writing the Next Flash or EEPROM Location

Symbol Minimum Wait Delay

tWD_FLASH 4.5 ms

tWD_EEPROM 9.0 ms

tWD_ERASE 9.0 ms

MSB

MSB

LSB

LSB

SERIAL CLOCK INPUT(SCK)

SERIAL DATA INPUT (MOSI)

(MISO)

SAMPLE

SERIAL DATA OUTPUT

352 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Table 163. Serial Programming Instruction Set

Instruction

Instruction Format

OperationByte 1 Byte 2 Byte 3 Byte4

Programming Enable 1010 1100 0101 0011 xxxx xxxx xxxx xxxx Enable Serial Programming after RESET goes low.

Chip Erase 1010 1100 100x xxxx xxxx xxxx xxxx xxxx Chip Erase EEPROM and Flash.

Load Extended Address Byte 0100 1101 0000 0000 cccc cccc xxxx xxxx Defines Extended Address Byte for Read Program Memory and Write Program Memory Page.

Read Program Memory 0010 H000 aaaa aaaa bbbb bbbb oooo oooo Read H (high or low) data o from Program memory at word address c:a:b.

Load Program Memory Page 0100 H000 xxxx xxxx xxbb bbbb iiii iiii Write H (high or low) data i to Program Memory page at word address b. Data low byte must be loaded before Data high byte is applied within the same address.

Write Program Memory Page 0100 1100 aaaa aaaa bbxx xxxx xxxx xxxx Write Program Memory Page at address c:a:b.

Read EEPROM Memory 1010 0000 0000 aaaa bbbb bbbb oooo oooo Read data o from EEPROM memory at address a:b.

Write EEPROM Memory 1100 0000 0000 aaaa bbbb bbbb iiii iiii Write data i to EEPROM memory at address a:b.

Load EEPROM Memory Page (page access)

1100 0001 0000 0000 0000 00bb iiii iiii Load data i to EEPROM memory page buffer. After data is loaded, program EEPROM page.

Write EEPROM Memory Page (page access)

1100 0010 0000 aaaa bbbb bb00 xxxx xxxx Write EEPROM page at address a:b.

Read Lock bits 0101 1000 0000 0000 xxxx xxxx xxoo oooo Read Lock bits. “0” = programmed, “1” = unprogrammed. See Table 148 on page 335 for details.

Write Lock bits 1010 1100 111x xxxx xxxx xxxx 11ii iiii Write Lock bits. Set bits = “0” to program Lock bits. See Table 148 on page 335 for details.

Read Signature Byte 0011 0000 000x xxxx xxxx xxbb oooo oooo Read Signature Byte o at address b.

Write Fuse bits 1010 1100 1010 0000 xxxx xxxx iiii iiii Set bits = “0” to program, “1” to unprogram. See Table 124 on page 287 for details.

Write Fuse High bits 1010 1100 1010 1000 xxxx xxxx iiii iiii Set bits = “0” to program, “1” to unprogram. See Table 123 on page 279 for details.

Write Extended Fuse Bits 1010 1100 1010 0100 xxxx xxxx iiii iiii Set bits = “0” to program, “1” to unprogram. See Table 150 on page 336 for details.

Read Fuse bits 0101 0000 0000 0000 xxxx xxxx oooo oooo Read Fuse bits. “0” = programmed, “1” = unprogrammed. See Table 124 on page 287 for details.

353

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: a = address high bits, b = address low bits, c = address extended bits, H = 0 - Low byte, 1 - High Byte, o = data out, i = data in,x = don’t care

Serial Programming Characteristics

For characteristics of the Serial Programming module see “SPI Timing Characteristics”on page 372.

Programming via the JTAG Interface

Programming through the JTAG interface requires control of the four JTAG specificpins: TCK, TMS, TDI, and TDO. Control of the reset and clock pins is not required.

To be able to use the JTAG interface, the JTAGEN Fuse must be programmed. Thedevice is default shipped with the fuse programmed. In addition, the JTD bit in MCUCSRmust be cleared. Alternatively, if the JTD bit is set, the external reset can be forced low.Then, the JTD bit will be cleared after two chip clocks, and the JTAG pins are availablefor programming. This provides a means of using the JTAG pins as normal port pins inRunning mode while still allowing In-System Programming via the JTAG interface. Notethat this technique can not be used when using the JTAG pins for Boundary-scan or On-chip Debug. In these cases the JTAG pins must be dedicated for this purpose.

During programming the clock frequency of the TCK Input must be less than the maxi-mum frequency of the chip. The System Clock Prescaler can not be used to divide theTCK Clock Input into a sufficiently low frequency.

As a definition in this datasheet, the LSB is shifted in and out first of all Shift Registers.

Programming Specific JTAG Instructions

The Instruction Register is 4-bit wide, supporting up to 16 instructions. The JTAGinstructions useful for programming are listed below.

The OPCODE for each instruction is shown behind the instruction name in hex format.The text describes which Data Register is selected as path between TDI and TDO foreach instruction.

The Run-Test/Idle state of the TAP controller is used to generate internal clocks. It canalso be used as an idle state between JTAG sequences. The state machine sequencefor changing the instruction word is shown in Figure 150.

Read Fuse High bits 0101 1000 0000 1000 xxxx xxxx oooo oooo Read Fuse High bits. “0” = pro-grammed, “1” = unprogrammed. See Table 123 on page 279 for details.

Read Extended Fuse Bits 0101 0000 0000 1000 xxxx xxxx oooo oooo Read Extended Fuse bits. “0” = pro-grammed, “1” = unprogrammed. See Table 150 on page 336 for details.

Read Calibration Byte 0011 1000 000x xxxx 0000 0000 oooo oooo Read Calibration Byte

Poll RDY/BSY 1111 0000 0000 0000 xxxx xxxx xxxx xxxo If o = “1”, a programming operation is still busy. Wait until this bit returns to “0” before applying another command.

Table 163. Serial Programming Instruction Set (Continued)

Instruction

Instruction Format

OperationByte 1 Byte 2 Byte 3 Byte4

354 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 150. State Machine Sequence for Changing the Instruction Word

AVR_RESET (0xC) The AVR specific public JTAG instruction for setting the AVR device in the Reset modeor taking the device out from the Reset mode. The TAP controller is not reset by thisinstruction. The one bit Reset Register is selected as Data Register. Note that the resetwill be active as long as there is a logic “one” in the Reset Chain. The output from thischain is not latched.

The active states are:

• Shift-DR: The Reset Register is shifted by the TCK input.

PROG_ENABLE (0x4) The AVR specific public JTAG instruction for enabling programming via the JTAG port.The 16-bit Programming Enable Register is selected as Data Register. The active statesare the following:

• Shift-DR: The programming enable signature is shifted into the Data Register.

• Update-DR: The programming enable signature is compared to the correct value, and Programming mode is entered if the signature is valid.

Test-Logic-Reset

Run-Test/Idle

Shift-DR

Exit1-DR

Pause-DR

Exit2-DR

Update-DR

Select-IR Scan

Capture-IR

Shift-IR

Exit1-IR

Pause-IR

Exit2-IR

Update-IR

Select-DR Scan

Capture-DR

0

1

0 1 1 1

0 0

0 0

1 1

1 0

1

1

0

1

0

0

1 0

1

1

0

1

0

0

00

11

355

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

PROG_COMMANDS (0x5) The AVR specific public JTAG instruction for entering programming commands via theJTAG port. The 15-bit Programming Command Register is selected as Data Register.The active states are the following:

• Capture-DR: The result of the previous command is loaded into the Data Register.

• Shift-DR: The Data Register is shifted by the TCK input, shifting out the result of the previous command and shifting in the new command.

• Update-DR: The programming command is applied to the Flash inputs

• Run-Test/Idle: One clock cycle is generated, executing the applied command

PROG_PAGELOAD (0x6) The AVR specific public JTAG instruction to directly load the Flash data page via theJTAG port. An 8-bit Flash Data Byte Register is selected as the Data Register. This isphysically the 8 LSBs of the Programming Command Register. The active states are thefollowing:

• Shift-DR: The Flash Data Byte Register is shifted by the TCK input.

• Update-DR: The content of the Flash Data Byte Register is copied into a temporary register. A write sequence is initiated that within 11 TCK cycles loads the content of the temporary register into the Flash page buffer. The AVR automatically alternates between writing the low and the high byte for each new Update-DR state, starting with the low byte for the first Update-DR encountered after entering the PROG_PAGELOAD command. The Program Counter is pre-incremented before writing the low byte, except for the first written byte. This ensures that the first data is written to the address set up by PROG_COMMANDS, and loading the last location in the page buffer does not make the program counter increment into the next page.

PROG_PAGEREAD (0x7) The AVR specific public JTAG instruction to directly capture the Flash content via theJTAG port. An 8-bit Flash Data Byte Register is selected as the Data Register. This isphysically the 8 LSBs of the Programming Command Register. The active states are thefollowing:

• Capture-DR: The content of the selected Flash byte is captured into the Flash Data Byte Register. The AVR automatically alternates between reading the low and the high byte for each new Capture-DR state, starting with the low byte for the first Capture-DR encountered after entering the PROG_PAGEREAD command. The Program Counter is post-incremented after reading each high byte, including the first read byte. This ensures that the first data is captured from the first address set up by PROG_COMMANDS, and reading the last location in the page makes the program counter increment into the next page.

• Shift-DR: The Flash Data Byte Register is shifted by the TCK input.

Data Registers The Data Registers are selected by the JTAG instruction registers described in section“Programming Specific JTAG Instructions” on page 353. The Data Registers relevant forprogramming operations are:

• Reset Register

• Programming Enable Register

• Programming Command Register

• Flash Data Byte Register

356 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Reset Register The Reset Register is a Test Data Register used to reset the part during programming. Itis required to reset the part before entering Programming mode.

A high value in the Reset Register corresponds to pulling the external reset low. Thepart is reset as long as there is a high value present in the Reset Register. Dependingon the Fuse settings for the clock options, the part will remain reset for a Reset Time-outperiod (refer to “Clock Sources” on page 40) after releasing the Reset Register. The out-put from this Data Register is not latched, so the reset will take place immediately, asshown in Figure 132 on page 303.

Programming Enable Register The Programming Enable Register is a 16-bit register. The contents of this register iscompared to the p rog ramming enab le s igna tu re , b ina ry code0b1010_0011_0111_0000. When the contents of the register is equal to the program-ming enable signature, programming via the JTAG port is enabled. The register is resetto 0 on Power-on Reset, and should always be reset when leaving Programming mode.

Figure 151. Programming Enable Register

Programming Command Register

The Programming Command Register is a 15-bit register. This register is used to seri-ally shift in programming commands, and to serially shift out the result of the previouscommand, if any. The JTAG Programming Instruction Set is shown in Table 164. Thestate sequence when shifting in the programming commands is illustrated in Figure 153.

TDI

TDO

DATA

= D Q

ClockDR & PROG_ENABLE

Programming Enable0xA370

357

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 152. Programming Command RegisterTDI

TDO

STROBES

ADDRESS/DATA

FlashEEPROM

FusesLock Bits

358 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Table 164. JTAG Programming Instruction Set a = address high bits, b = address low bits, c = address extended bits, H = 0 - Low byte, 1 - High Byte, o = data out, i = data in, x =don’t care

Instruction TDI Sequence TDO Sequence Notes

1a. Chip Erase 0100011_10000000

0110001_100000000110011_10000000

0110011_10000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

1b. Poll for Chip Erase Complete 0110011_10000000 xxxxxox_xxxxxxxx (2)

2a. Enter Flash Write 0100011_00010000 xxxxxxx_xxxxxxxx

2b. Load Address Extended High Byte 0001011_cccccccc xxxxxxx_xxxxxxxx (10)

2c. Load Address High Byte 0000111_aaaaaaaa xxxxxxx_xxxxxxxx

2d. Load Address Low Byte 0000011_bbbbbbbb xxxxxxx_xxxxxxxx

2e. Load Data Low Byte 0010011_iiiiiiii xxxxxxx_xxxxxxxx

2f. Load Data High Byte 0010111_iiiiiiii xxxxxxx_xxxxxxxx

2g. Latch Data 0110111_000000001110111_00000000

0110111_00000000

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

(1)

2h. Write Flash Page 0110111_00000000

0110101_00000000

0110111_000000000110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

(1)

2i. Poll for Page Write Complete 0110111_00000000 xxxxxox_xxxxxxxx (2)

3a. Enter Flash Read 0100011_00000010 xxxxxxx_xxxxxxxx

3b. Load Address Extended High Byte 0001011_cccccccc xxxxxxx_xxxxxxxx (10)

3c. Load Address High Byte 0000111_aaaaaaaa xxxxxxx_xxxxxxxx

3d. Load Address Low Byte 0000011_bbbbbbbb xxxxxxx_xxxxxxxx

3e. Read Data Low and High Byte 0110010_00000000

0110110_00000000

0110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_ooooooooxxxxxxx_oooooooo

Low byte

High byte

4a. Enter EEPROM Write 0100011_00010001 xxxxxxx_xxxxxxxx

4b. Load Address High Byte 0000111_aaaaaaaa xxxxxxx_xxxxxxxx (10)

4c. Load Address Low Byte 0000011_bbbbbbbb xxxxxxx_xxxxxxxx

4d. Load Data Byte 0010011_iiiiiiii xxxxxxx_xxxxxxxx

4e. Latch Data 0110111_00000000

1110111_00000000

0110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

(1)

4f. Write EEPROM Page 0110011_00000000

0110001_00000000

0110011_000000000110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

(1)

4g. Poll for Page Write Complete 0110011_00000000 xxxxxox_xxxxxxxx (2)

359

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

5a. Enter EEPROM Read 0100011_00000011 xxxxxxx_xxxxxxxx

5b. Load Address High Byte 0000111_aaaaaaaa xxxxxxx_xxxxxxxx (10)

5c. Load Address Low Byte 0000011_bbbbbbbb xxxxxxx_xxxxxxxx

5d. Read Data Byte 0110011_bbbbbbbb0110010_00000000

0110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

xxxxxxx_oooooooo

6a. Enter Fuse Write 0100011_01000000 xxxxxxx_xxxxxxxx

6b. Load Data Low Byte(6) 0010011_iiiiiiii xxxxxxx_xxxxxxxx (3)

6c. Write Fuse Extended Byte 0111011_000000000111001_00000000

0111011_00000000

0111011_00000000

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

(1)

6d. Poll for Fuse Write Complete 0110111_00000000 xxxxxox_xxxxxxxx (2)

6e. Load Data Low Byte(7) 0010011_iiiiiiii xxxxxxx_xxxxxxxx (3)

6f. Write Fuse High Byte 0110111_000000000110101_00000000

0110111_00000000

0110111_00000000

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

(1)

6g. Poll for Fuse Write Complete 0110111_00000000 xxxxxox_xxxxxxxx (2)

6h. Load Data Low Byte(7) 0010011_iiiiiiii xxxxxxx_xxxxxxxx (3)

6i. Write Fuse Low Byte 0110011_000000000110001_00000000

0110011_00000000

0110011_00000000

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

(1)

6j. Poll for Fuse Write Complete 0110011_00000000 xxxxxox_xxxxxxxx (2)

7a. Enter Lock Bit Write 0100011_00100000 xxxxxxx_xxxxxxxx

7b. Load Data Byte(9) 0010011_11iiiiii xxxxxxx_xxxxxxxx (4)

7c. Write Lock Bits 0110011_00000000

0110001_000000000110011_00000000

0110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

xxxxxxx_xxxxxxxx

(1)

7d. Poll for Lock Bit Write complete 0110011_00000000 xxxxxox_xxxxxxxx (2)

8a. Enter Fuse/Lock Bit Read 0100011_00000100 xxxxxxx_xxxxxxxx

8b. Read Extended Fuse Byte(6) 0111010_00000000

0111011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_oooooooo

8c. Read Fuse High Byte(7) 0111110_00000000

0111111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_oooooooo

8d. Read Fuse Low Byte(8) 0110010_00000000

0110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_oooooooo

Table 164. JTAG Programming Instruction (Continued)Set (Continued) a = address high bits, b = address low bits, c = address extended bits, H = 0 - Low byte, 1 - High Byte, o = data out, i= data in, x = don’t care

Instruction TDI Sequence TDO Sequence Notes

360 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Notes: 1. This command sequence is not required if the seven MSB are correctly set by the previous command sequence (which isnormally the case).

2. Repeat until o = “1”.3. Set bits to “0” to program the corresponding Fuse, “1” to unprogram the Fuse.4. Set bits to “0” to program the corresponding Lock bit, “1” to leave the Lock bit unchanged.5. “0” = programmed, “1” = unprogrammed.6. The bit mapping for Fuses Extended byte is listed in Table 150 on page 3367. The bit mapping for Fuses High byte is listed in Table 151 on page 3378. The bit mapping for Fuses Low byte is listed in Table 152 on page 3379. The bit mapping for Lock bits byte is listed in Table 148 on page 33510. Address bits exceeding PCMSB and EEAMSB (Table 158 and Table 159) are don’t care11. All TDI and TDO sequences are represented by binary digits (0b...).

8e. Read Lock Bits(9) 0110110_00000000

0110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_xxoooooo(5)

8f. Read Fuses and Lock Bits 0111010_00000000

0111110_00000000

0110010_000000000110110_00000000

0110111_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_ooooooooxxxxxxx_ooooooooxxxxxxx_ooooooooxxxxxxx_oooooooo

(5)

Fuse Ext. byte

Fuse High byteFuse Low byte

Lock bits

9a. Enter Signature Byte Read 0100011_00001000 xxxxxxx_xxxxxxxx

9b. Load Address Byte 0000011_bbbbbbbb xxxxxxx_xxxxxxxx

9c. Read Signature Byte 0110010_00000000

0110011_00000000

xxxxxxx_xxxxxxxx

xxxxxxx_oooooooo

10a. Enter Calibration Byte Read 0100011_00001000 xxxxxxx_xxxxxxxx

10b. Load Address Byte 0000011_bbbbbbbb xxxxxxx_xxxxxxxx

10c. Read Calibration Byte 0110110_000000000110111_00000000

xxxxxxx_xxxxxxxxxxxxxxx_oooooooo

11a. Load No Operation Command 0100011_000000000110011_00000000

xxxxxxx_xxxxxxxxxxxxxxx_xxxxxxxx

Table 164. JTAG Programming Instruction (Continued)Set (Continued) a = address high bits, b = address low bits, c = address extended bits, H = 0 - Low byte, 1 - High Byte, o = data out, i= data in, x = don’t care

Instruction TDI Sequence TDO Sequence Notes

361

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 153. State Machine Sequence for Changing/Reading the Data Word

Flash Data Byte Register The Flash Data Byte Register provides an efficient way to load the entire Flash pagebuffer before executing Page Write, or to read out/verify the content of the Flash. A statemachine sets up the control signals to the Flash and senses the strobe signals from theFlash, thus only the data words need to be shifted in/out.

The Flash Data Byte Register actually consists of the 8-bit scan chain and a 8-bit tempo-rary register. During page load, the Update-DR state copies the content of the scanchain over to the temporary register and initiates a write sequence that within 11 TCKcycles loads the content of the temporary register into the Flash page buffer. The AVRautomatically alternates between writing the low and the high byte for each new Update-DR state, starting with the low byte for the first Update-DR encountered after enteringthe PROG_PAGELOAD command. The Program Counter is pre-incremented beforewriting the low byte, except for the first written byte. This ensures that the first data iswritten to the address set up by PROG_COMMANDS, and loading the last location inthe page buffer does not make the Program Counter increment into the next page.

During Page Read, the content of the selected Flash byte is captured into the FlashData Byte Register during the Capture-DR state. The AVR automatically alternatesbetween reading the low and the high byte for each new Capture-DR state, starting withthe low byte for the first Capture-DR encountered after entering the PROG_PAGEREADcommand. The Program Counter is post-incremented after reading each high byte,

Test-Logic-Reset

Run-Test/Idle

Shift-DR

Exit1-DR

Pause-DR

Exit2-DR

Update-DR

Select-IR Scan

Capture-IR

Shift-IR

Exit1-IR

Pause-IR

Exit2-IR

Update-IR

Select-DR Scan

Capture-DR

0

1

0 1 1 1

0 0

0 0

1 1

1 0

1

1

0

1

0

0

1 0

1

1

0

1

0

0

00

11

362 ATmega640/1280/1281/2560/25612549A–AVR–03/05

including the first read byte. This ensures that the first data is captured from the firstaddress set up by PROG_COMMANDS, and reading the last location in the page makesthe program counter increment into the next page.

Figure 154. Flash Data Byte Register

The state machine controlling the Flash Data Byte Register is clocked by TCK. Duringnormal operation in which eight bits are shifted for each Flash byte, the clock cyclesneeded to navigate through the TAP controller automatically feeds the state machine forthe Flash Data Byte Register with sufficient number of clock pulses to complete its oper-ation transparently for the user. However, if too few bits are shifted between eachUpdate-DR state during page load, the TAP controller should stay in the Run-Test/Idlestate for some TCK cycles to ensure that there are at least 11 TCK cycles between eachUpdate-DR state.

Programming Algorithm All references below of type “1a”, “1b”, and so on, refer to Table 164.

Entering Programming Mode 1. Enter JTAG instruction AVR_RESET and shift 1 in the Reset Register.

2. Enter instruction PROG_ENABLE and shift 0b1010_0011_0111_0000 in the Programming Enable Register.

Leaving Programming Mode 1. Enter JTAG instruction PROG_COMMANDS.

2. Disable all programming instructions by using no operation instruction 11a.

3. Enter instruction PROG_ENABLE and shift 0b0000_0000_0000_0000 in the programming Enable Register.

4. Enter JTAG instruction AVR_RESET and shift 0 in the Reset Register.

Performing Chip Erase 1. Enter JTAG instruction PROG_COMMANDS.

2. Start Chip Erase using programming instruction 1a.

3. Poll for Chip Erase complete using programming instruction 1b, or wait for tWLRH_CE (refer to Table 160 on page 348).

TDI

TDO

DATA

FlashEEPROM

FusesLock Bits

STROBES

ADDRESS

StateMachine

363

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Programming the Flash Before programming the Flash a Chip Erase must be performed, see “Performing ChipErase” on page 362.

1. Enter JTAG instruction PROG_COMMANDS.

2. Enable Flash write using programming instruction 2a.

3. Load address Extended High byte using programming instruction 2b.

4. Load address High byte using programming instruction 2c.

5. Load address Low byte using programming instruction 2d.

6. Load data using programming instructions 2e, 2f and 2g.

7. Repeat steps 5 and 6 for all instruction words in the page.

8. Write the page using programming instruction 2h.

9. Poll for Flash write complete using programming instruction 2i, or wait for tWLRH (refer to Table 160 on page 348).

10. Repeat steps 3 to 9 until all data have been programmed.

A more efficient data transfer can be achieved using the PROG_PAGELOADinstruction:

1. Enter JTAG instruction PROG_COMMANDS.

2. Enable Flash write using programming instruction 2a.

3. Load the page address using programming instructions 2b, 2c and 2d. PCWORD (refer to Table 158 on page 340) is used to address within one page and must be written as 0.

4. Enter JTAG instruction PROG_PAGELOAD.

5. Load the entire page by shifting in all instruction words in the page byte-by-byte, starting with the LSB of the first instruction in the page and ending with the MSB of the last instruction in the page. Use Update-DR to copy the contents of the Flash Data Byte Register into the Flash page location and to auto-increment the Program Counter before each new word.

6. Enter JTAG instruction PROG_COMMANDS.

7. Write the page using programming instruction 2h.

8. Poll for Flash write complete using programming instruction 2i, or wait for tWLRH (refer to Table 160 on page 348).

9. Repeat steps 3 to 8 until all data have been programmed.

Reading the Flash 1. Enter JTAG instruction PROG_COMMANDS.

2. Enable Flash read using programming instruction 3a.

3. Load address using programming instructions 3b, 3c and 3d.

4. Read data using programming instruction 3e.

5. Repeat steps 3 and 4 until all data have been read.

A more efficient data transfer can be achieved using the PROG_PAGEREADinstruction:

1. Enter JTAG instruction PROG_COMMANDS.

2. Enable Flash read using programming instruction 3a.

3. Load the page address using programming instructions 3b, 3c and 3d. PCWORD (refer to Table 158 on page 340) is used to address within one page and must be written as 0.

4. Enter JTAG instruction PROG_PAGEREAD.

364 ATmega640/1280/1281/2560/25612549A–AVR–03/05

5. Read the entire page (or Flash) by shifting out all instruction words in the page (or Flash), starting with the LSB of the first instruction in the page (Flash) and ending with the MSB of the last instruction in the page (Flash). The Capture-DR state both captures the data from the Flash, and also auto-increments the pro-gram counter after each word is read. Note that Capture-DR comes before the shift-DR state. Hence, the first byte which is shifted out contains valid data.

6. Enter JTAG instruction PROG_COMMANDS.

7. Repeat steps 3 to 6 until all data have been read.

Programming the EEPROM Before programming the EEPROM a Chip Erase must be performed, see “PerformingChip Erase” on page 362.

1. Enter JTAG instruction PROG_COMMANDS.

2. Enable EEPROM write using programming instruction 4a.

3. Load address High byte using programming instruction 4b.

4. Load address Low byte using programming instruction 4c.

5. Load data using programming instructions 4d and 4e.

6. Repeat steps 4 and 5 for all data bytes in the page.

7. Write the data using programming instruction 4f.

8. Poll for EEPROM write complete using programming instruction 4g, or wait for tWLRH (refer to Table 160 on page 348).

9. Repeat steps 3 to 8 until all data have been programmed.

Note that the PROG_PAGELOAD instruction can not be used when programming theEEPROM.

Reading the EEPROM 1. Enter JTAG instruction PROG_COMMANDS.

2. Enable EEPROM read using programming instruction 5a.

3. Load address using programming instructions 5b and 5c.

4. Read data using programming instruction 5d.

5. Repeat steps 3 and 4 until all data have been read.

Note that the PROG_PAGEREAD instruction can not be used when reading theEEPROM.

Programming the Fuses 1. Enter JTAG instruction PROG_COMMANDS.

2. Enable Fuse write using programming instruction 6a.

3. Load data high byte using programming instructions 6b. A bit value of “0” will pro-gram the corresponding fuse, a “1” will unprogram the fuse.

4. Write Fuse High byte using programming instruction 6c.

5. Poll for Fuse write complete using programming instruction 6d, or wait for tWLRH (refer to Table 160 on page 348).

6. Load data low byte using programming instructions 6e. A “0” will program the fuse, a “1” will unprogram the fuse.

7. Write Fuse low byte using programming instruction 6f.

8. Poll for Fuse write complete using programming instruction 6g, or wait for tWLRH (refer to Table 160 on page 348).

365

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Programming the Lock Bits 1. Enter JTAG instruction PROG_COMMANDS.

2. Enable Lock bit write using programming instruction 7a.

3. Load data using programming instructions 7b. A bit value of “0” will program the corresponding lock bit, a “1” will leave the lock bit unchanged.

4. Write Lock bits using programming instruction 7c.

5. Poll for Lock bit write complete using programming instruction 7d, or wait for tWLRH (refer to Table 160 on page 348).

Reading the Fuses and Lock Bits

1. Enter JTAG instruction PROG_COMMANDS.

2. Enable Fuse/Lock bit read using programming instruction 8a.

3. To read all Fuses and Lock bits, use programming instruction 8e.To only read Fuse High byte, use programming instruction 8b.To only read Fuse Low byte, use programming instruction 8c.To only read Lock bits, use programming instruction 8d.

Reading the Signature Bytes 1. Enter JTAG instruction PROG_COMMANDS.

2. Enable Signature byte read using programming instruction 9a.

3. Load address 0x00 using programming instruction 9b.

4. Read first signature byte using programming instruction 9c.

5. Repeat steps 3 and 4 with address 0x01 and address 0x02 to read the second and third signature bytes, respectively.

Reading the Calibration Byte 1. Enter JTAG instruction PROG_COMMANDS.

2. Enable Calibration byte read using programming instruction 10a.

3. Load address 0x00 using programming instruction 10b.

4. Read the calibration byte using programming instruction 10c.

366 ATmega640/1280/1281/2560/25612549A–AVR–03/05

367

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Electrical Characteristics

Absolute Maximum Ratings*

DC Characteristics

Operating Temperature.................................. -55°C to +125°C *NOTICE: Stresses beyond those listed under “Absolute Maximum Ratings” may cause permanent dam-age to the device. This is a stress rating only and functional operation of the device at these or other conditions beyond those indicated in the operational sections of this specification is not implied. Exposure to absolute maximum rating conditions for extended periods may affect device reliability.

Storage Temperature ..................................... -65°C to +150°C

Voltage on any Pin except RESETwith respect to Ground ................................-0.5V to VCC+0.5V

Voltage on RESET with respect to Ground......-0.5V to +13.0V

Maximum Operating Voltage ............................................ 6.0V

DC Current per I/O Pin ............................................... 40.0 mA

DC Current VCC and GND Pins................................ 200.0 mA

TA = -40°C to 85°C, VCC = 1.8V to 5.5V (unless otherwise noted)

Symbol Parameter Condition Min.(5) Typ. Max.(5) Units

VILInput Low Voltage,Except XTAL1 and Reset pin

VCC = 1.8V - 2.4VVCC = 2.4V - 5.5V

-0.5-0.5

0.2VCC(1)

0.3VCC(1) V

VIL1Input Low Voltage,XTAL1 pin

VCC = 1.8V - 5.5V -0.5 0.1VCC(1) V

VIL2Input Low Voltage, RESET pin

VCC = 1.8V - 5.5V -0.5 0.1VCC(1) V

VIH

Input High Voltage, Except XTAL1 and RESET pins

VCC = 1.8V - 2.4VVCC = 2.4V - 5.5V

0.7VCC(2)

0.6VCC(2)

VCC + 0.5VCC + 0.5

V

VIH1Input High Voltage, XTAL1 pin

VCC = 1.8V - 2.4VVCC = 2.4V - 5.5V

0.8VCC(2)

0.7VCC(2)

VCC + 0.5VCC + 0.5

V

VIH2Input High Voltage, RESET pin

VCC = 1.8V - 5.5V 0.9VCC(2) VCC + 0.5 V

VOL Output Low Voltage(3), IOL = 10mA, VCC = 5VIOL = 5mA, VCC = 3V

0.70.5

V

VOH Output High Voltage(4), IOH = -20mA, VCC = 5VIOH = -10mA, VCC = 3V

4.22.3

V

IILInput LeakageCurrent I/O Pin

VCC = 5.5V, pin low(absolute value)

1 µA

IIHInput LeakageCurrent I/O Pin

VCC = 5.5V, pin high(absolute value)

1 µA

RRST Reset Pull-up Resistor 30 60 kΩ

RPU I/O Pin Pull-up Resistor 20 50 kΩ

368 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: 1. "Max" means the highest value where the pin is guaranteed to be read as low

2. "Min" means the lowest value where the pin is guaranteed to be read as high3. Although each I/O port can sink more than the test conditions (20mA at VCC = 5V, 10mA at VCC = 3V) under steady state

conditions (non-transient), the following must be observed:ATmega1281/2561:1.)The sum of all IOL, for ports A0-A7, G2, C4-C7 should not exceed 100 mA.2.)The sum of all IOL, for ports C0-C3, G0-G1, D0-D7 should not exceed 100 mA.3.)The sum of all IOL, for ports G3-G5, B0-B7, E0-E7 should not exceed 100 mA.4.)The sum of all IOL, for ports F0-F7 should not exceed 100 mA.ATmega640/1280/2560:1.)The sum of all IOL, for ports J0-J7, A0-A7, G2 should not exceed 200 mA.2.)The sum of all IOL, for ports C0-C7, G0-G1, D0-D7, L0-L7 should not exceed 200 mA.3.)The sum of all IOL, for ports G3-G4, B0-B7, H0-B7 should not exceed 200 mA.4.)The sum of all IOL, for ports E0-E7, G5 should not exceed 100 mA.5.)The sum of all IOL, for ports F0-F7, K0-K7 should not exceed 100 mA.If IOL exceeds the test condition, VOL may exceed the related specification. Pins are not guaranteed to sink current greaterthan the listed test condition.

4. Although each I/O port can source more than the test conditions (20mA at VCC = 5V, 10mA at VCC = 3V) under steadystate conditions (non-transient), the following must be observed:ATmega1281/2561:1)The sum of all IOH, for ports A0-A7, G2, C4-C7 should not exceed 100 mA.2)The sum of all IOH, for ports C0-C3, G0-G1, D0-D7 should not exceed 100 mA.

ICC

Power Supply Current(6)

Active 1MHz, VCC = 2V

(ATmega640/1280/2560/1V)

0.8 mA

Active 4MHz, VCC = 3V(ATmega640/1280/2560/1L)

5 mA

Active 8MHz, VCC = 5V

(ATmega640/1280/1281/2560/2561)

18 mA

Idle 1MHz, VCC = 2V

(ATmega640/1280/2560/1V)

0.4 0.75 mA

Idle 4MHz, VCC = 3V(ATmega640/1280/2560/1L)

2.2 mA

Idle 8MHz, VCC = 5V

(ATmega640/1280/1281/2560/2561)

8 mA

Power-down modeWDT enabled, VCC = 3V <10 20 µA

WDT disabled, VCC = 3V <1 3 µA

VACIOAnalog Comparator Input Offset Voltage

VCC = 5V

Vin = VCC/2<10 40 mV

IACLKAnalog Comparator Input Leakage Current

VCC = 5VVin = VCC/2

-50 50 nA

tACIDAnalog Comparator Propagation Delay

VCC = 2.7VVCC = 4.0V

750500

ns

TA = -40°C to 85°C, VCC = 1.8V to 5.5V (unless otherwise noted) (Continued)

Symbol Parameter Condition Min.(5) Typ. Max.(5) Units

369

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

3)The sum of all IOH, for ports G3-G5, B0-B7, E0-E7 should not exceed 100 mA.4)The sum of all IOH, for ports F0-F7 should not exceed 100 mA.ATmega640/1280/2560:1)The sum of all IOH, for ports J0-J7, G2, A0-A7 should not exceed 200 mA.2)The sum of all IOH, for ports C0-C7, G0-G1, D0-D7, L0-L7 should not exceed 200 mA.3)The sum of all IOH, for ports G3-G4, B0-B7, H0-H7 should not exceed 200 mA.4)The sum of all IOH, for ports E0-E7, G5 should not exceed 100 mA.5)The sum of all IOH, for ports F0-F7, K0-K7 should not exceed 100 mA.If IOH exceeds the test condition, VOH may exceed the related specification. Pins are not guaranteed to source currentgreater than the listed test condition.

5. All DC Characteristics contained in this datasheet are based on simulation and characterization of other AVR microcontrol-lers manufactured in the same process technology. These values are preliminary values representing design targets, andwill be updated after characterization of actual silicon

6. Values with “Power Reduction Register 1 - PRR1” disabled (0x00).

External Clock Drive Waveforms

Figure 155. External Clock Drive Waveforms

External Clock Drive

Note: All DC Characteristics contained in this datasheet are based on simulation and charac-terization of other AVR microcontrollers manufactured in the same process technology.These values are preliminary values representing design targets, and will be updatedafter characterization of actual silicon.

VIL1

VIH1

Table 165. External Clock Drive

Symbol Parameter

VCC=1.8-5.5V VCC=2.7-5.5V VCC=4.5-5.5V

UnitsMin. Max. Min. Max. Min. Max.

1/tCLCLOscillator Frequency

0 2 0 8 0 16 MHz

tCLCL Clock Period 500 125 62.5 ns

tCHCX High Time 200 50 25 ns

tCLCX Low Time 200 50 25 ns

tCLCH Rise Time 2.0 1.6 0.5 µs

tCHCL Fall Time 2.0 1.6 0.5 µs

∆tCLCL

Change in period from one clock cycle to the next

2 2 2 %

370 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Maximum speed vs. VCC Maximum frequency is depending on VCC. As shown in Figure 156 and Figure 157, theMaximum Frequency vs. VCC curve is linear between 1.8V < VCC < 2.7V and between2.7V < VCC < 4.5V.

Figure 156. Maximum Frequency vs. VCC, ATmegaXXX1V/XXX0V

Figure 157. Maximum Frequency vs. VCC, ATmega640/1280/1281/2560/2561

8 MHz

4 MHz

1.8V 2.7V 5.5V

Safe Operating Area

16 MHz

8 MHz

2.7V 4.5V 5.5V

Safe Operating Area

371

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

2-wire Serial Interface Characteristics

Tab le 166 desc r ibes the requ i rements fo r dev ices connec ted to the 2 -w i re Ser ia l Bus. TheATmega640/1280/1281/2560/2561 2-wire Serial Interface meets or exceeds these requirements under the notedconditions.

Timing symbols refer to Figure 158.

Notes: 1. In ATmega640/1280/1281/2560/2561, this parameter is characterized and not 100% tested.2. Required only for fSCL > 100 kHz.3. Cb = capacitance of one bus line in pF.

Table 166. 2-wire Serial Bus Requirements

Symbol Parameter Condition Min Max Units

VIL Input Low-voltage -0.5 0.3 VCC V

VIH Input High-voltage 0.7 VCC VCC + 0.5 V

Vhys(1) Hysteresis of Schmitt Trigger Inputs 0.05 VCC

(2) – V

VOL(1) Output Low-voltage 3 mA sink current 0 0.4 V

tr(1) Rise Time for both SDA and SCL 20 + 0.1Cb

(3)(2) 300 ns

tof(1) Output Fall Time from VIHmin to VILmax 10 pF < Cb < 400 pF(3) 20 + 0.1Cb

(3)(2) 250 ns

tSP(1) Spikes Suppressed by Input Filter 0 50(2) ns

Ii Input Current each I/O Pin 0.1VCC < Vi < 0.9VCC -10 10 µA

Ci(1) Capacitance for each I/O Pin – 10 pF

fSCL SCL Clock Frequency fCK(4) > max(16fSCL, 250kHz)(5) 0 400 kHz

Rp Value of Pull-up resistor

fSCL ≤ 100 kHz

fSCL > 100 kHz

tHD;STA Hold Time (repeated) START ConditionfSCL ≤ 100 kHz 4.0 – µs

fSCL > 100 kHz 0.6 – µs

tLOW Low Period of the SCL ClockfSCL ≤ 100 kHz(6) 4.7 – µs

fSCL > 100 kHz(7) 1.3 – µs

tHIGH High period of the SCL clockfSCL ≤ 100 kHz 4.0 – µs

fSCL > 100 kHz 0.6 – µs

tSU;STA Set-up time for a repeated START conditionfSCL ≤ 100 kHz 4.7 – µs

fSCL > 100 kHz 0.6 – µs

tHD;DAT Data hold timefSCL ≤ 100 kHz 0 3.45 µs

fSCL > 100 kHz 0 0.9 µs

tSU;DAT Data setup timefSCL ≤ 100 kHz 250 – ns

fSCL > 100 kHz 100 – ns

tSU;STO Setup time for STOP conditionfSCL ≤ 100 kHz 4.0 – µs

fSCL > 100 kHz 0.6 – µs

tBUFBus free time between a STOP and START condition

fSCL ≤ 100 kHz 4.7 – µs

fSCL > 100 kHz 1.3 – µs

VCC 0,4V–

3mA---------------------------- 1000ns

Cb------------------- Ω

VCC 0,4V–

3mA---------------------------- 300ns

Cb---------------- Ω

372 ATmega640/1280/1281/2560/25612549A–AVR–03/05

4. fCK = CPU clock frequency5. This requirement applies to all ATmega640/1280/1281/2560/2561 2-wire Serial Interface operation. Other devices con-

nected to the 2-wire Serial Bus need only obey the general fSCL requirement.6. The actual low period generated by the ATmega640/1280/1281/2560/2561 2-wire Serial Interface is (1/fSCL - 2/fCK), thus

fCK must be greater than 6 MHz for the low time requirement to be strictly met at fSCL = 100 kHz.7. The actual low period generated by the ATmega640/1280/1281/2560/2561 2-wire Serial Interface is (1/fSCL - 2/fCK), thus

the low time requirement will not be strictly met for fSCL > 308 kHz when fCK = 8 MHz. Still,ATmega640/1280/1281/2560/2561 devices connected to the bus may communicate at full speed (400 kHz) with otherATmega640/1280/1281/2560/2561 devices, as well as any other device with a proper tLOW acceptance margin.

Figure 158. 2-wire Serial Bus Timing

SPI Timing Characteristics

See Figure 159 and Figure 160 for details.

Note: 1. In SPI Programming mode the minimum SCK high/low period is:- 2 tCLCL for fCK < 12 MHz- 3 tCLCL for fCK > 12 MHz

tSU;STA

tLOW

tHIGH

tLOW

tof

tHD;STA tHD;DAT tSU;DATtSU;STO

tBUF

SCL

SDA

tr

Table 167. SPI Timing Parameters

Description Mode Min Typ Max

1 SCK period Master See Table 96

ns

2 SCK high/low Master 50% duty cycle

3 Rise/Fall time Master TBD

4 Setup Master 10

5 Hold Master 10

6 Out to SCK Master 0.5 • tsck

7 SCK to out Master 10

8 SCK to out high Master 10

9 SS low to out Slave 15

10 SCK period Slave 4 • tck

11 SCK high/low(1) Slave 2 • tck

12 Rise/Fall time Slave TBD

13 Setup Slave 10

14 Hold Slave tck

15 SCK to out Slave 15

16 SCK to SS high Slave 20

17 SS high to tri-state Slave 10

18 SS low to SCK Slave 20

373

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 159. SPI Interface Timing Requirements (Master Mode)

Figure 160. SPI Interface Timing Requirements (Slave Mode)

MOSI(Data Output)

SCK(CPOL = 1)

MISO(Data Input)

SCK(CPOL = 0)

SS

MSB LSB

LSBMSB

...

...

6 1

2 2

34 5

87

MISO(Data Output)

SCK(CPOL = 1)

MOSI(Data Input)

SCK(CPOL = 0)

SS

MSB LSB

LSBMSB

...

...

10

11 11

1213 14

1715

9

X

16

374 ATmega640/1280/1281/2560/25612549A–AVR–03/05

ADC Characteristics – Preliminary DataTable 168. ADC Characteristics

Symbol Parameter Condition Min(1) Typ(1) Max(1) Units

Resolution

Single Ended Conversion 10 Bits

Differential Conversion

Gain = 1x or 20x8

Bits

Differential ConversionGain = 200x

7Bits

Absolute accuracy (Including INL, DNL, quantization error, gain and offset error)

Single Ended ConversionVREF = 4V, VCC = 4V,ADC clock = 200 kHz

2 2.5 LSB

Single Ended ConversionVREF = 4V, VCC = 4V,ADC clock = 1 MHz

4.5 LSB

Single Ended ConversionVREF = 4V, VCC = 4V,ADC clock = 200 kHz

Noise Reduction Mode

2 LSB

Single Ended ConversionVREF = 4V, VCC = 4V,ADC clock = 1 MHzNoise Reduction Mode

4.5 LSB

Integral Non-Linearity (INL)Single Ended ConversionVREF = 4V, VCC = 4V,ADC clock = 200 kHz

0.5 LSB

Differential Non-Linearity (DNL)Single Ended ConversionVREF = 4V, VCC = 4V,ADC clock = 200 kHz

0.25 LSB

Gain ErrorSingle Ended ConversionVREF = 4V, VCC = 4V,ADC clock = 200 kHz

2 LSB

Offset ErrorSingle Ended ConversionVREF = 4V, VCC = 4V,ADC clock = 200 kHz

2 LSB

Conversion Time Free Running Conversion 13 260 µs

Clock Frequency Single Ended Conversion 50 1000 kHz

AVCC Analog Supply Voltage VCC - 0.3 VCC + 0.3 V

VREF Reference VoltageSingle Ended Conversion 1.0 AVCC V

Differential Conversion 1.0 AVCC - 0.5 V

VIN Input VoltageSingle ended channels GND VREF V

Differential Conversion 0 AVCC V

Input BandwidthSingle Ended Channels 38,5 kHz

Differential Channels 4 kHz

VINT1 Internal Voltage Reference 1.1V 1.0 1.1 1.2 V

375

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Notes: 1. Values are guidelines only. Actual values are TBD

VINT2 Internal Voltage Reference 2.56V 2.4 2.56 2.8 V

RREF Reference Input Resistance 32 kΩ

RAIN Analog Input Resistance 100 MΩ

Table 168. ADC Characteristics (Continued)

Symbol Parameter Condition Min(1) Typ(1) Max(1) Units

376 ATmega640/1280/1281/2560/25612549A–AVR–03/05

External Data Memory Timing

Notes: 1. This assumes 50% clock duty cycle. The half period is actually the high time of the external clock, XTAL1.2. This assumes 50% clock duty cycle. The half period is actually the low time of the external clock, XTAL1.

Table 169. External Data Memory Characteristics, 4.5 - 5.5 Volts, No Wait-state

Symbol Parameter

8 MHz Oscillator Variable Oscillator

UnitMin Max Min Max

0 1/tCLCL Oscillator Frequency 0.0 16 MHz

1 tLHLL ALE Pulse Width 115 1.0tCLCL-10 ns

2 tAVLL Address Valid A to ALE Low 57.5 0.5tCLCL-5(1) ns

3a tLLAX_ST

Address Hold After ALE Low, write access

5 5ns

3b tLLAX_LD

Address Hold after ALE Low, read access

5 5ns

4 tAVLLC Address Valid C to ALE Low 57.5 0.5tCLCL-5(1) ns

5 tAVRL Address Valid to RD Low 115 1.0tCLCL-10 ns

6 tAVWL Address Valid to WR Low 115 1.0tCLCL-10 ns

7 tLLWL ALE Low to WR Low 47.5 67.5 0.5tCLCL-15(2) 0.5tCLCL+5(2) ns

8 tLLRL ALE Low to RD Low 47.5 67.5 0.5tCLCL-15(2) 0.5tCLCL+5(2) ns

9 tDVRH Data Setup to RD High 40 40 ns

10 tRLDV Read Low to Data Valid 75 1.0tCLCL-50 ns

11 tRHDX Data Hold After RD High 0 0 ns

12 tRLRH RD Pulse Width 115 1.0tCLCL-10 ns

13 tDVWL Data Setup to WR Low 42.5 0.5tCLCL-20(1) ns

14 tWHDX Data Hold After WR High 115 1.0tCLCL-10 ns

15 tDVWH Data Valid to WR High 125 1.0tCLCL ns

16 tWLWH WR Pulse Width 115 1.0tCLCL-10 ns

Table 170. External Data Memory Characteristics, 4.5 - 5.5 Volts, 1 Cycle Wait-state

Symbol Parameter

8 MHz Oscillator Variable Oscillator

UnitMin Max Min Max

0 1/tCLCL Oscillator Frequency 0.0 16 MHz

10 tRLDV Read Low to Data Valid 200 2.0tCLCL-50 ns

12 tRLRH RD Pulse Width 240 2.0tCLCL-10 ns

15 tDVWH Data Valid to WR High 240 2.0tCLCL ns

16 tWLWH WR Pulse Width 240 2.0tCLCL-10 ns

377

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table 171. External Data Memory Characteristics, 4.5 - 5.5 Volts, SRWn1 = 1, SRWn0 = 0

Symbol Parameter

4 MHz Oscillator Variable Oscillator

UnitMin Max Min Max

0 1/tCLCL Oscillator Frequency 0.0 16 MHz

10 tRLDV Read Low to Data Valid 325 3.0tCLCL-50 ns

12 tRLRH RD Pulse Width 365 3.0tCLCL-10 ns

15 tDVWH Data Valid to WR High 375 3.0tCLCL ns

16 tWLWH WR Pulse Width 365 3.0tCLCL-10 ns

Table 172. External Data Memory Characteristics, 4.5 - 5.5 Volts, SRWn1 = 1, SRWn0 = 1

Symbol Parameter

4 MHz Oscillator Variable Oscillator

UnitMin Max Min Max

0 1/tCLCL Oscillator Frequency 0.0 16 MHz

10 tRLDV Read Low to Data Valid 325 3.0tCLCL-50 ns

12 tRLRH RD Pulse Width 365 3.0tCLCL-10 ns

14 tWHDX Data Hold After WR High 240 2.0tCLCL-10 ns

15 tDVWH Data Valid to WR High 375 3.0tCLCL ns

16 tWLWH WR Pulse Width 365 3.0tCLCL-10 ns

Table 173. External Data Memory Characteristics, 2.7 - 5.5 Volts, No Wait-state

Symbol Parameter

4 MHz Oscillator Variable Oscillator

UnitMin Max Min Max

0 1/tCLCL Oscillator Frequency 0.0 8 MHz

1 tLHLL ALE Pulse Width 235 tCLCL-15 ns

2 tAVLL Address Valid A to ALE Low 115 0.5tCLCL-10(1) ns

3a tLLAX_ST

Address Hold After ALE Low, write access

5 5ns

3b tLLAX_LD

Address Hold after ALE Low, read access

5 5ns

4 tAVLLC Address Valid C to ALE Low 115 0.5tCLCL-10(1) ns

5 tAVRL Address Valid to RD Low 235 1.0tCLCL-15 ns

6 tAVWL Address Valid to WR Low 235 1.0tCLCL-15 ns

7 tLLWL ALE Low to WR Low 115 130 0.5tCLCL-10(2) 0.5tCLCL+5(2) ns

8 tLLRL ALE Low to RD Low 115 130 0.5tCLCL-10(2) 0.5tCLCL+5(2) ns

9 tDVRH Data Setup to RD High 45 45 ns

10 tRLDV Read Low to Data Valid 190 1.0tCLCL-60 ns

11 tRHDX Data Hold After RD High 0 0 ns

378 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Notes: 1. This assumes 50% clock duty cycle. The half period is actually the high time of the external clock, XTAL1.2. This assumes 50% clock duty cycle. The half period is actually the low time of the external clock, XTAL1.

12 tRLRH RD Pulse Width 235 1.0tCLCL-15 ns

13 tDVWL Data Setup to WR Low 105 0.5tCLCL-20(1) ns

14 tWHDX Data Hold After WR High 235 1.0tCLCL-15 ns

15 tDVWH Data Valid to WR High 250 1.0tCLCL ns

16 tWLWH WR Pulse Width 235 1.0tCLCL-15 ns

Table 173. External Data Memory Characteristics, 2.7 - 5.5 Volts, No Wait-state (Continued)

Symbol Parameter

4 MHz Oscillator Variable Oscillator

UnitMin Max Min Max

Table 174. External Data Memory Characteristics, 2.7 - 5.5 Volts, SRWn1 = 0, SRWn0 = 1

Symbol Parameter

4 MHz Oscillator Variable Oscillator

UnitMin Max Min Max

0 1/tCLCL Oscillator Frequency 0.0 8 MHz

10 tRLDV Read Low to Data Valid 440 2.0tCLCL-60 ns

12 tRLRH RD Pulse Width 485 2.0tCLCL-15 ns

15 tDVWH Data Valid to WR High 500 2.0tCLCL ns

16 tWLWH WR Pulse Width 485 2.0tCLCL-15 ns

Table 175. External Data Memory Characteristics, 2.7 - 5.5 Volts, SRWn1 = 1, SRWn0 = 0

Symbol Parameter

4 MHz Oscillator Variable Oscillator

UnitMin Max Min Max

0 1/tCLCL Oscillator Frequency 0.0 8 MHz

10 tRLDV Read Low to Data Valid 690 3.0tCLCL-60 ns

12 tRLRH RD Pulse Width 735 3.0tCLCL-15 ns

15 tDVWH Data Valid to WR High 750 3.0tCLCL ns

16 tWLWH WR Pulse Width 735 3.0tCLCL-15 ns

Table 176. External Data Memory Characteristics, 2.7 - 5.5 Volts, SRWn1 = 1, SRWn0 = 1

Symbol Parameter

4 MHz Oscillator Variable Oscillator

UnitMin Max Min Max

0 1/tCLCL Oscillator Frequency 0.0 8 MHz

10 tRLDV Read Low to Data Valid 690 3.0tCLCL-60 ns

12 tRLRH RD Pulse Width 735 3.0tCLCL-15 ns

14 tWHDX Data Hold After WR High 485 2.0tCLCL-15 ns

15 tDVWH Data Valid to WR High 750 3.0tCLCL ns

16 tWLWH WR Pulse Width 735 3.0tCLCL-15 ns

379

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Figure 161. External Memory Timing (SRWn1 = 0, SRWn0 = 0

Figure 162. External Memory Timing (SRWn1 = 0, SRWn0 = 1)

ALE

T1 T2 T3

Writ

eR

ead

WR

T4

A15:8 AddressPrev. addr.

DA7:0 Address DataPrev. data XX

RD

DA7:0 (XMBK = 0) DataAddress

System Clock (CLKCPU)

1

4

2

7

6

3a

3b

5

8 12

16

13

10

11

14

15

9

ALE

T1 T2 T3

Writ

eR

ead

WR

T5

A15:8 AddressPrev. addr.

DA7:0 Address DataPrev. data XX

RD

DA7:0 (XMBK = 0) DataAddress

System Clock (CLKCPU)

1

4

2

7

6

3a

3b

5

8 12

16

13

10

11

14

15

9

T4

380 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Figure 163. External Memory Timing (SRWn1 = 1, SRWn0 = 0)

Figure 164. External Memory Timing (SRWn1 = 1, SRWn0 = 1)()

The ALE pulse in the last period (T4-T7) is only present if the next instruction accesses the RAM(internal or external).

ALE

T1 T2 T3

Wri

teR

ea

d

WR

T6

A15:8 AddressPrev. addr.

DA7:0 Address DataPrev. data XX

RD

DA7:0 (XMBK = 0) DataAddress

System Clock (CLKCPU)

1

4

2

7

6

3a

3b

5

8 12

16

13

10

11

14

15

9

T4 T5

ALE

T1 T2 T3

Wri

teR

ea

d

WR

T7

A15:8 AddressPrev. addr.

DA7:0 Address DataPrev. data XX

RD

DA7:0 (XMBK = 0) DataAddress

System Clock (CLKCPU)

1

4

2

7

6

3a

3b

5

8 12

16

13

10

11

14

15

9

T4 T5 T6

381

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

ATmega640/1280/1281/2560/2561 Typical Characteristics – Preliminary Data

TBD

The following charts show typical behavior. These figures are not tested during manu-facturing. All current consumption measurements are performed with all I/O pinsconfigured as inputs and with internal pull-ups enabled. A sine wave generator with rail-to-rail output is used as clock source.

All Active- and Idle current consumption measurements are done with all bits in the PRRregisters set and thus, the corresponding I/O modules are turned off. Also the AnalogComparator is disabled during these measurements. Table 177 on page 381 and Table178 on page 382 show the additional current consumption compared to ICC Active andICC Idle for every I/O module controlled by the Power Reduction Register. See “PowerReduction Register” on page 54 for details.

The power consumption in Power-down mode is independent of clock selection.

The current consumption is a function of several factors such as: operating voltage,operating frequency, loading of I/O pins, switching rate of I/O pins, code executed andambient temperature. The dominating factors are operating voltage and frequency.

The current drawn from capacitive loaded pins may be estimated (for one pin) asCL*VCC*f where CL = load capacitance, VCC = operating voltage and f = average switch-ing frequency of I/O pin.

The parts are characterized at frequencies higher than test limits. Parts are not guaran-teed to function properly at frequencies higher than the ordering code indicates.

The difference between current consumption in Power-down mode with WatchdogTimer enabled and Power-down mode with Watchdog Timer disabled represents the dif-ferential current drawn by the Watchdog Timer.

Supply Current of IO modules

The tables and formulas below can be used to calculate the additional current consump-tion for the different I/O modules in Active and Idle mode. The enabling or disabling ofthe I/O modules are controlled by the Power Reduction Register. See “Power ReductionRegister” on page 54 for details.

Table 177. Additional Current Consumption for the different I/O modules (absolute values)

PRR bit Typical numbers

VCC = 2V, F = 1MHz VCC = 3V, F = 4MHz VCC = 5V, F = 8MHz

PRUSART3 8.0 uA 51 uA 220 uA

PRUSART2 8.0 uA 51 uA 220 uA

PRUSART1 8.0 uA 51 uA 220 uA

PRUSART0 8.0 uA 51 uA 220 uA

PRTWI 12 uA 75 uA 315 uA

PRTIM5 6.0 uA 39 uA 150 uA

PRTIM4 6.0 uA 39 uA 150 uA

PRTIM3 6.0 uA 39 uA 150 uA

PRTIM2 11 uA 72 uA 300 uA

PRTIM1 6.0 uA 39 uA 150 uA

382 ATmega640/1280/1281/2560/25612549A–AVR–03/05

It is possible to calculate the typical current consumption based on the numbers fromTable 177 for other VCC and frequency settings than listed in Table 178.

Example 1 Calculate the expected current consumption in idle mode with USART0, TIMER1, andTWI enabled at VCC = 3.0V and F = 1MHz. From Table 177, third column, we see thatwe need to add 18% for the USART0, 26% for the TWI, and 11% for the TIMER1 mod-ule. Reading from Figure XXXX, we find that the idle current consumption is ~0,075mAat VCC = 3.0V and F = 1MHz. The total current consumption in idle mode with USART0,TIMER1, and TWI enabled, gives:

Example 2 Same conditions as in example 1, but in active mode instead. From Table 178 secondcolumn we see that we need to add 3.3% for the USART0, 4.8% for the TWI, and 2.0%for the TIMER1 module. Reading from Figure XXXX, we find that the active current con-sumption is ~0,42mA at VCC = 3.0V and F = 1MHz. The total current consumption in idlemode with USART0, TIMER1, and TWI enabled, gives:

PRTIM0 4.0 uA 24 uA 100 uA

PRSPI 15 uA 95 uA 400 uA

PRADC 12 uA 75 uA 315 uA

Table 178. Additional Current Consumption (percentage) in Active and Idle mode

PRR bit

Additional Current consumption compared to Active with external clock

Additional Current consumption compared to Idle with external clock

PRUSART3 3.0% 17%

PRUSART2 3.0% 17%

PRUSART1 3.0% 17%

PRUSART0 3.0% 17%

PRTWI 4.4% 24%

PRTIM5 1.8% 10%

PRTIM4 1.8% 10%

PRTIM3 1.8% 10%

PRTIM2 4.3% 23%

PRTIM1 1.8% 10%

PRTIM0 1.5% 8.0%

PRSPI 3.3% 18%

PRADC 4.5% 24%

Table 177. Additional Current Consumption for the different I/O modules (absolute values)(Continued)

PRR bit Typical numbers

VCC = 2V, F = 1MHz VCC = 3V, F = 4MHz VCC = 5V, F = 8MHz

ICCtotal 0.075mA 1 0.18 0.26 0.11+ + +( )• 0.116mA≈ ≈

ICCtotal 0.42mA 1 0.033 0.048 0.02+ + +( )• 0.46mA≈ ≈

383

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Example 3 All I/O modules should be enabled. Calculate the expected current consumption inactive mode at VCC = 3.6V and F = 10MHz. We find the active current consumption with-out the I/O modules to be ~ 4.0mA (from Figure XXXX). Then, by using the numbersfrom Table 178 - second column, we find the total current consumption:

ICCtotal 5.0mA 1 0.03 0.03 0.03 0.03 0.044 0.018 0.018 0.018 0.043 0.018 0.015 0.033 0.045+ + + + + + + + + + + + +( )• 6.9mA≈ ≈

384 ATmega640/1280/1281/2560/25612549A–AVR–03/05

385

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Register SummaryAddress Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Page

(0x1FF) Reserved - - - - - - - -

... Reserved - - - - - - - -

(0x13F) Reserved

(0x13E) Reserved

(0x13D) Reserved

(0x13C) Reserved

(0x13B) Reserved

(0x13A) Reserved

(0x139) Reserved

(0x138) Reserved

(0x137) Reserved

(0x136) UDR3 USART3 I/O Data Register

(0x135) UBRR3H - - - - USART3 Baud Rate Register High Byte

(0x134) UBRR3L USART3 Baud Rate Register Low Byte

(0x133) Reserved - - - - - - - -

(0x132) UCSR3C UMSEL31 UMSEL30 UPM31 UPM30 USBS3 UCSZ31 UCSZ30 UCPOL3

(0x131) UCSR3B RXCIE3 TXCIE3 UDRIE3 RXEN3 TXEN3 UCSZ32 RXB83 TXB83

(0x130) UCSR3A RXC3 TXC3 UDRE3 FE3 DOR3 UPE3 U2X3 MPCM3

(0x12F) Reserved - - - - - - - -

(0x12E) Reserved - - - - - - - -

(0x12D) OCR5CH Timer/Counter5 - Output Compare Register C High Byte

(0x12C) OCR5CL Timer/Counter5 - Output Compare Register C Low Byte

(0x12B) OCR5BH Timer/Counter5 - Output Compare Register B High Byte

(0x12A) OCR5BL Timer/Counter5 - Output Compare Register B Low Byte

(0x129) OCR5AH Timer/Counter5 - Output Compare Register A High Byte

(0x128) OCR5AL Timer/Counter5 - Output Compare Register A Low Byte

(0x127) ICR5H Timer/Counter5 - Input Capture Register High Byte

(0x126) ICR5L Timer/Counter5 - Input Capture Register Low Byte

(0x125) TCNT5H Timer/Counter5 - Counter Register High Byte

(0x124) TCNT5L Timer/Counter5 - Counter Register Low Byte

(0x123) Reserved - - - - - - - -

(0x122) TCCR5C FOC5A FOC5B FOC5C - - - - -

(0x121) TCCR5B ICNC5 ICES5 - WGM53 WGM52 CS52 CS51 CS50

(0x120) TCCR5A COM5A1 COM5A0 COM5B1 COM5B0 COM5C1 COM5C0 WGM51 WGM50

(0x11F) Reserved - - - - - - - -

(0x11E) Reserved - - - - - - - -

(0x11D) Reserved - - - - - - - -

(0x11C) Reserved - - - - - - - -

(0x11B) Reserved - - - - - - - -

(0x11A) Reserved - - - - - - - -

(0x119) Reserved - - - - - - - -

(0x118) Reserved - - - - - - - -

(0x117) Reserved - - - - - - - -

(0x116) Reserved - - - - - - - -

(0x115) Reserved - - - - - - - -

(0x114) Reserved - - - - - - - -

(0x113) Reserved - - - - - - - -

(0x112) Reserved - - - - - - - -

(0x111) Reserved - - - - - - - -

(0x110) Reserved - - - - - - - -

(0x10F) Reserved - - - - - - - -

(0x10E) Reserved - - - - - - - -

(0x10D) Reserved - - - - - - - -

(0x10C) Reserved - - - - - - - -

(0x10B) PORTL PORTL7 PORTL6 PORTL5 PORTL4 PORTL3 PORTL2 PORTL1 PORTL0

(0x10A) DDRL DDL7 DDL6 DDL5 DDL4 DDL3 DDL2 DDL1 DDL0

(0x109) PINL PINL7 PINL6 PINL5 PINL4 PINL3 PINL2 PINL1 PINL0

(0x108) PORTK PORTK7 PORTK6 PORTK5 PORTK4 PORTK3 PORTK2 PORTK1 PORTK0

(0x107) DDRK DDK7 DDK6 DDK5 DDK4 DDK3 DDK2 DDK1 DDK0

(0x106) PINK PINK7 PINK6 PINK5 PINK4 PINK3 PINK2 PINK1 PINK0

(0x105) PORTJ PORTJ7 PORTJ6 PORTJ5 PORTJ4 PORTJ3 PORTJ2 PORTJ1 PORTJ0

(0x104) DDRJ DDJ7 DDJ6 DDJ5 DDJ4 DDJ3 DDJ2 DDJ1 DDJ0

(0x103) PINJ PINJ7 PINJ6 PINJ5 PINJ4 PINJ3 PINJ2 PINJ1 PINJ0

(0x102) PORTH PORTH7 PORTH6 PORTH5 PORTH4 PORTH3 PORTH2 PORTH1 PORTH0

386 ATmega640/1280/1281/2560/25612549A–AVR–03/05

(0x101) DDRH DDH7 DDH6 DDH5 DDH4 DDH3 DDH2 DDH1 DDH0

(0x100) PINH PINH7 PINH6 PINH5 PINH4 PINH3 PINH2 PINH1 PINH0

(0xFF) Reserved - - - - - - - -

(0xFE) Reserved - - - - - - - -

(0xFD) Reserved - - - - - - - -

(0xFC) Reserved - - - - - - - -

(0xFB) Reserved - - - - - - - -

(0xFA) Reserved - - - - - - - -

(0xF9) Reserved - - - - - - - -

(0xF8) Reserved - - - - - - - -

(0xF7) Reserved - - - - - - - -

(0xF6) Reserved - - - - - - - -

(0xF5) Reserved - - - - - - - -

(0xF4) Reserved - - - - - - - -

(0xF3) Reserved - - - - - - - -

(0xF2) Reserved - - - - - - - -

(0xF1) Reserved - - - - - - - -

(0xF0) Reserved - - - - - - - -

(0xEF) Reserved - - - - - - - -

(0xEE) Reserved - - - - - - - -

(0xED) Reserved - - - - - - - -

(0xEC) Reserved - - - - - - - -

(0xEB) Reserved - - - - - - -

(0xEA) Reserved - - - - - - - -

(0xE9) Reserved - - - - - - - -

(0xE8) Reserved - - - - - - - -

(0xE7) Reserved - - - - - - -

(0xE6) Reserved - - - - - - - -

(0xE5) Reserved - - - - - - - -

(0xE4) Reserved - - - - - - - -

(0xE3) Reserved - - - - - - -

(0xE2) Reserved - - - - - - - -

(0xE1) Reserved - - - - - - -

(0xE0) Reserved - - - - - - -

(0xDF) Reserved - - - - - - - -

(0xDE) Reserved - - - - - - - -

(0xDD) Reserved - - - - - - -

(0xDC) Reserved - - - - - - - -

(0xDB) Reserved - - - - - - - -

(0xDA) Reserved - - - - - - - -

(0xD9) Reserved - - - - - - -

(0xD8) Reserved - - - - - - - -

(0xD7) Reserved - - - - - - - -

(0xD6) UDR2 USART2 I/O Data Register

(0xD5) UBRR2H - - - - USART2 Baud Rate Register High Byte

(0xD4) UBRR2L USART2 Baud Rate Register Low Byte

(0xD3) Reserved - - - - - - - -

(0xD2) UCSR2C UMSEL21 UMSEL20 UPM21 UPM20 USBS2 UCSZ21 UCSZ20 UCPOL2

(0xD1) UCSR2B RXCIE2 TXCIE2 UDRIE2 RXEN2 TXEN2 UCSZ22 RXB82 TXB82

(0xD0) UCSR2A RXC2 TXC2 UDRE2 FE2 DOR2 UPE2 U2X2 MPCM2

(0xCF) Reserved - - - - - - - -

(0xCE) UDR1 USART1 I/O Data Register

(0xCD) UBRR1H - - - - USART1 Baud Rate Register High Byte

(0xCC) UBRR1L USART1 Baud Rate Register Low Byte

(0xCB) Reserved - - - - - - - -

(0xCA) UCSR1C UMSEL11 UMSEL10 UPM11 UPM10 USBS1 UCSZ11 UCSZ10 UCPOL1

(0xC9) UCSR1B RXCIE1 TXCIE1 UDRIE1 RXEN1 TXEN1 UCSZ12 RXB81 TXB81

(0xC8) UCSR1A RXC1 TXC1 UDRE1 FE1 DOR1 UPE1 U2X1 MPCM1

(0xC7) Reserved - - - - - - - -

(0xC6) UDR0 USART0 I/O Data Register

(0xC5) UBRR0H - - - - USART0 Baud Rate Register High Byte

(0xC4) UBRR0L USART0 Baud Rate Register Low Byte

(0xC3) Reserved - - - - - - - -

(0xC2) UCSR0C UMSEL01 UMSEL00 UPM01 UPM00 USBS0 UCSZ01 UCSZ00 UCPOL0

(0xC1) UCSR0B RXCIE0 TXCIE0 UDRIE0 RXEN0 TXEN0 UCSZ02 RXB80 TXB80

(0xC0) UCSR0A RXC0 TXC0 UDRE0 FE0 DOR0 UPE0 U2X0 MPCM0

Address Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Page

387

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

(0xBF) Reserved - - - - - - - -

(0xBE) Reserved - - - - - - - -

(0xBD) TWAMR TWAM6 TWAM5 TWAM4 TWAM3 TWAM2 TWAM1 TWAM0 -

(0xBC) TWCR TWINT TWEA TWSTA TWSTO TWWC TWEN - TWIE

(0xBB) TWDR 2-wire Serial Interface Data Register

(0xBA) TWAR TWA6 TWA5 TWA4 TWA3 TWA2 TWA1 TWA0 TWGCE

(0xB9) TWSR TWS7 TWS6 TWS5 TWS4 TWS3 - TWPS1 TWPS0

(0xB8) TWBR 2-wire Serial Interface Bit Rate Register

(0xB7) Reserved - - - - - - - -

(0xB6) ASSR - EXCLK AS2 TCN2UB OCR2AUB OCR2BUB TCR2AUB TCR2BUB

(0xB5) Reserved - - - - - - - -

(0xB4) OCR2B Timer/Counter2 Output Compare Register B

(0xB3) OCR2A Timer/Counter2 Output Compare Register A

(0xB2) TCNT2 Timer/Counter2 (8 Bit)

(0xB1) TCCR2B FOC2A FOC2B - - WGM22 CS22 CS21 CS20

(0xB0) TCCR2A COM2A1 COM2A0 COM2B1 COM2B0 - - WGM21 WGM20

(0xAF) Reserved - - - - - - - -

(0xAE) Reserved - - - - - - - -

(0xAD) OCR4CH Timer/Counter4 - Output Compare Register C High Byte

(0xAC) OCR4CL Timer/Counter4 - Output Compare Register C Low Byte

(0xAB) OCR4BH Timer/Counter4 - Output Compare Register B High Byte

(0xAA) OCR4BL Timer/Counter4 - Output Compare Register B Low Byte

(0xA9) OCR4AH Timer/Counter4 - Output Compare Register A High Byte

(0xA8) OCR4AL Timer/Counter4 - Output Compare Register A Low Byte

(0xA7) ICR4H Timer/Counter4 - Input Capture Register High Byte

(0xA6) ICR4L Timer/Counter4 - Input Capture Register Low Byte

(0xA5) TCNT4H Timer/Counter4 - Counter Register High Byte

(0xA4) TCNT4L Timer/Counter4 - Counter Register Low Byte

(0xA3) Reserved - - - - - - - -

(0xA2) TCCR4C FOC4A FOC4B FOC4C - - - - -

(0xA1) TCCR4B ICNC4 ICES4 - WGM43 WGM42 CS42 CS41 CS40

(0xA0) TCCR4A COM4A1 COM4A0 COM4B1 COM4B0 COM4C1 COM4C0 WGM41 WGM40

(0x9F) Reserved - - - - - - - -

(0x9E) Reserved - - - - - - - -

(0x9D) OCR3CH Timer/Counter3 - Output Compare Register C High Byte

(0x9C) OCR3CL Timer/Counter3 - Output Compare Register C Low Byte

(0x9B) OCR3BH Timer/Counter3 - Output Compare Register B High Byte

(0x9A) OCR3BL Timer/Counter3 - Output Compare Register B Low Byte

(0x99) OCR3AH Timer/Counter3 - Output Compare Register A High Byte

(0x98) OCR3AL Timer/Counter3 - Output Compare Register A Low Byte

(0x97) ICR3H Timer/Counter3 - Input Capture Register High Byte

(0x96) ICR3L Timer/Counter3 - Input Capture Register Low Byte

(0x95) TCNT3H Timer/Counter3 - Counter Register High Byte

(0x94) TCNT3L Timer/Counter3 - Counter Register Low Byte

(0x93) Reserved - - - - - - - -

(0x92) TCCR3C FOC3A FOC3B FOC3C - - - - -

(0x91) TCCR3B ICNC3 ICES3 - WGM33 WGM32 CS32 CS31 CS30

(0x90) TCCR3A COM3A1 COM3A0 COM3B1 COM3B0 COM3C1 COM3C0 WGM31 WGM30

(0x8F) Reserved - - - - - - - -

(0x8E) Reserved - - - - - - - -

(0x8D) OCR1CH Timer/Counter1 - Output Compare Register C High Byte

(0x8C) OCR1CL Timer/Counter1 - Output Compare Register C Low Byte

(0x8B) OCR1BH Timer/Counter1 - Output Compare Register B High Byte

(0x8A) OCR1BL Timer/Counter1 - Output Compare Register B Low Byte

(0x89) OCR1AH Timer/Counter1 - Output Compare Register A High Byte

(0x88) OCR1AL Timer/Counter1 - Output Compare Register A Low Byte

(0x87) ICR1H Timer/Counter1 - Input Capture Register High Byte

(0x86) ICR1L Timer/Counter1 - Input Capture Register Low Byte

(0x85) TCNT1H Timer/Counter1 - Counter Register High Byte

(0x84) TCNT1L Timer/Counter1 - Counter Register Low Byte

(0x83) Reserved - - - - - - - -

(0x82) TCCR1C FOC1A FOC1B FOC1C - - - - -

(0x81) TCCR1B ICNC1 ICES1 - WGM13 WGM12 CS12 CS11 CS10

(0x80) TCCR1A COM1A1 COM1A0 COM1B1 COM1B0 COM1C1 COM1C0 WGM11 WGM10

(0x7F) DIDR1 - - - - - - AIN1D AIN0D

(0x7E) DIDR0 ADC7D ADC6D ADC5D ADC4D ADC3D ADC2D ADC1D ADC0D

Address Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Page

388 ATmega640/1280/1281/2560/25612549A–AVR–03/05

(0x7D) DIDR2 ADC15D ADC14D ADC13D ADC12D ADC11D ADC10D ADC9D ADC8D

(0x7C) ADMUX REFS1 REFS0 ADLAR MUX4 MUX3 MUX2 MUX1 MUX0

(0x7B) ADCSRB - ACME - - MUX5 ADTS2 ADTS1 ADTS0

(0x7A) ADCSRA ADEN ADSC ADATE ADIF ADIE ADPS2 ADPS1 ADPS0

(0x79) ADCH ADC Data Register High byte

(0x78) ADCL ADC Data Register Low byte

(0x77) Reserved - - - - - - - -

(0x76) Reserved - - - - - - - -

(0x75) XMCRB XMBK - - - - XMM2 XMM1 XMM0

(0x74) XMCRA SRE SRL2 SRL1 SRL0 SRW11 SRW10 SRW01 SRW00

(0x73) TIMSK5 - - ICIE5 - OCIE5C OCIE5B OCIE5A TOIE5

(0x72) TIMSK4 - - ICIE4 - OCIE4C OCIE4B OCIE4A TOIE4

(0x71) TIMSK3 - - ICIE3 - OCIE3C OCIE3B OCIE3A TOIE3

(0x70) TIMSK2 - - - - - OCIE2B OCIE2A TOIE2

(0x6F) TIMSK1 - - ICIE1 - OCIE1C OCIE1B OCIE1A TOIE1

(0x6E) TIMSK0 - - - - - OCIE0B OCIE0A TOIE0

(0x6D) PCMSK2 PCINT23 PCINT22 PCINT21 PCINT20 PCINT19 PCINT18 PCINT17 PCINT16

(0x6C) PCMSK1 PCINT15 PCINT14 PCINT13 PCINT12 PCINT11 PCINT10 PCINT9 PCINT8

(0x6B) PCMSK0 PCINT7 PCINT6 PCINT5 PCINT4 PCINT3 PCINT2 PCINT1 PCINT0

(0x6A) EICRB ISC71 ISC70 ISC61 ISC60 ISC51 ISC50 ISC41 ISC40

(0x69) EICRA ISC31 ISC30 ISC21 ISC20 ISC11 ISC10 ISC01 ISC00

(0x68) PCICR - - - - - PCIE2 PCIE1 PCIE0

(0x67) Reserved - - - - - - - -

(0x66) OSCCAL Oscillator Calibration Register

(0x65) PRR1 - - PRTIM5 PRTIM4 PRTIM3 PRUSART3 PRUSART2 PRUSART1

(0x64) PRR0 PRTWI PRTIM2 PRTIM0 - PRTIM1 PRSPI PRUSART0 PRADC

(0x63) Reserved - - - - - - - -

(0x62) Reserved - - - - - - - -

(0x61) CLKPR CLKPCE - - - CLKPS3 CLKPS2 CLKPS1 CLKPS0

(0x60) WDTCSR WDIF WDIE WDP3 WDCE WDE WDP2 WDP1 WDP0

0x3F (0x5F) SREG I T H S V N Z C

0x3E (0x5E) SPH SP15 SP14 SP13 SP12 SP11 SP10 SP9 SP8

0x3D (0x5D) SPL SP7 SP6 SP5 SP4 SP3 SP2 SP1 SP0

0x3C (0x5C) EIND - - - - - - - EIND0

0x3B (0x5B) RAMPZ - - - - - - RAMPZ1 RAMPZ0

0x3A (0x5A) Reserved - - - - - - - -

0x39 (0x59) Reserved - - - - - - - -

0x38 (0x58) Reserved - - - - - - - -

0x37 (0x57) SPMCSR SPMIE RWWSB SIGRD RWWSRE BLBSET PGWRT PGERS SPMEN

0x36 (0x56) Reserved - - - - - - - -

0x35 (0x55) MCUCR JTD - - PUD - - IVSEL IVCE

0x34 (0x54) MCUSR - - - JTRF WDRF BORF EXTRF PORF

0x33 (0x53) SMCR - - - - SM2 SM1 SM0 SE

0x32 (0x52) Reserved - - - - - - - -

0x31 (0x51)OCDR/

MONDROCDR7 OCDR6 OCDR5 OCDR4 OCDR3 OCDR2 OCDR1 OCDR0

Monitor Data Register

0x30 (0x50) ACSR ACD ACBG ACO ACI ACIE ACIC ACIS1 ACIS0

0x2F (0x4F) Reserved - - - - - - - -

0x2E (0x4E) SPDR SPI Data Register

0x2D (0x4D) SPSR SPIF WCOL - - - - - SPI2X

0x2C (0x4C) SPCR SPIE SPE DORD MSTR CPOL CPHA SPR1 SPR0

0x2B (0x4B) GPIOR2 General Purpose I/O Register 2

0x2A (0x4A) GPIOR1 General Purpose I/O Register 1

0x29 (0x49) Reserved - - - - - - - -

0x28 (0x48) OCR0B Timer/Counter0 Output Compare Register B

0x27 (0x47) OCR0A Timer/Counter0 Output Compare Register A

0x26 (0x46) TCNT0 Timer/Counter0 (8 Bit)

0x25 (0x45) TCCR0B FOC0A FOC0B - - WGM02 CS02 CS01 CS00

0x24 (0x44) TCCR0A COM0A1 COM0A0 COM0B1 COM0B0 - - WGM01 WGM00

0x23 (0x43) GTCCR TSM - - - - - PSRASY PSRSYNC

0x22 (0x42) EEARH - - - - EEPROM Address Register High Byte

0x21 (0x41) EEARL EEPROM Address Register Low Byte

0x20 (0x40) EEDR EEPROM Data Register

0x1F (0x3F) EECR - - EEPM1 EEPM0 EERIE EEMPE EEPE EERE

0x1E (0x3E) GPIOR0 General Purpose I/O Register 0

0x1D (0x3D) EIMSK INT7 INT6 INT5 INT4 INT3 INT2 INT1 INT0

Address Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Page

389

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Note: 1. For compatibility with future devices, reserved bits should be written to zero if accessed. Reserved I/O memory addressesshould never be written.

2. I/O registers within the address range $00 - $1F are directly bit-accessible using the SBI and CBI instructions. In these reg-isters, the value of single bits can be checked by using the SBIS and SBIC instructions.

3. Some of the status flags are cleared by writing a logical one to them. Note that the CBI and SBI instructions will operate onall bits in the I/O register, writing a one back into any flag read as set, thus clearing the flag. The CBI and SBI instructionswork with registers 0x00 to 0x1F only.

4. When using the I/O specific commands IN and OUT, the I/O addresses $00 - $3F must be used. When addressing I/O regis-ters as data space using LD and ST instructions, $20 must be added to these addresses. TheATmega640/1280/1281/2560/2561 is a complex microcontroller with more peripheral units than can be supported withinthe 64 location reserved in Opcode for the IN and OUT instructions. For the Extended I/O space from $60 - $1FF in SRAM,only the ST/STS/STD and LD/LDS/LDD instructions can be used.

0x1C (0x3C) EIFR INTF7 INTF6 INTF5 INTF4 INTF3 INTF2 INTF1 INTF0

0x1B (0x3B) PCIFR - - - - - PCIF2 PCIF1 PCIF0

0x1A (0x3A) TIFR5 - - ICF5 - OCF5C OCF5B OCF5A TOV5

0x19 (0x39) TIFR4 - - ICF4 - OCF4C OCF4B OCF4A TOV4

0x18 (0x38) TIFR3 - - ICF3 - OCF3C OCF3B OCF3A TOV3

0x17 (0x37) TIFR2 - - - - - OCF2B OCF2A TOV2

0x16 (0x36) TIFR1 - - ICF1 - OCF1C OCF1B OCF1A TOV1

0x15 (0x35) TIFR0 - - - - - OCF0B OCF0A TOV0

0x14 (0x34) PORTG - - PORTG5 PORTG4 PORTG3 PORTG2 PORTG1 PORTG0

0x13 (0x33) DDRG - - DDG5 DDG4 DDG3 DDG2 DDG1 DDG0

0x12 (0x32) PING - - PING5 PING4 PING3 PING2 PING1 PING0

0x11 (0x31) PORTF PORTF7 PORTF6 PORTF5 PORTF4 PORTF3 PORTF2 PORTF1 PORTF0

0x10 (0x30) DDRF DDF7 DDF6 DDF5 DDF4 DDF3 DDF2 DDF1 DDF0

0x0F (0x2F) PINF PINF7 PINF6 PINF5 PINF4 PINF3 PINF2 PINF1 PINF0

0x0E (0x2E) PORTE PORTE7 PORTE6 PORTE5 PORTE4 PORTE3 PORTE2 PORTE1 PORTE0

0x0D (0x2D) DDRE DDE7 DDE6 DDE5 DDE4 DDE3 DDE2 DDE1 DDE0

0x0C (0x2C) PINE PINE7 PINE6 PINE5 PINE4 PINE3 PINE2 PINE1 PINE0

0x0B (0x2B) PORTD PORTD7 PORTD6 PORTD5 PORTD4 PORTD3 PORTD2 PORTD1 PORTD0

0x0A (0x2A) DDRD DDD7 DDD6 DDD5 DDD4 DDD3 DDD2 DDD1 DDD0

0x09 (0x29) PIND PIND7 PIND6 PIND5 PIND4 PIND3 PIND2 PIND1 PIND0

0x08 (0x28) PORTC PORTC7 PORTC6 PORTC5 PORTC4 PORTC3 PORTC2 PORTC1 PORTC0

0x07 (0x27) DDRC DDC7 DDC6 DDC5 DDC4 DDC3 DDC2 DDC1 DDC0

0x06 (0x26) PINC PINC7 PINC6 PINC5 PINC4 PINC3 PINC2 PINC1 PINC0

0x05 (0x25) PORTB PORTB7 PORTB6 PORTB5 PORTB4 PORTB3 PORTB2 PORTB1 PORTB0

0x04 (0x24) DDRB DDB7 DDB6 DDB5 DDB4 DDB3 DDB2 DDB1 DDB0

0x03 (0x23) PINB PINB7 PINB6 PINB5 PINB4 PINB3 PINB2 PINB1 PINB0

0x02 (0x22) PORTA PORTA7 PORTA6 PORTA5 PORTA4 PORTA3 PORTA2 PORTA1 PORTA0

0x01 (0x21) DDRA DDA7 DDA6 DDA5 DDA4 DDA3 DDA2 DDA1 DDA0

0x00 (0x20) PINA PINA7 PINA6 PINA5 PINA4 PINA3 PINA2 PINA1 PINA0

Address Name Bit 7 Bit 6 Bit 5 Bit 4 Bit 3 Bit 2 Bit 1 Bit 0 Page

390 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Instruction Set SummaryMnemonics Operands Description Operation Flags #Clocks

ARITHMETIC AND LOGIC INSTRUCTIONS

ADD Rd, Rr Add two Registers Rd ← Rd + Rr Z,C,N,V,H 1

ADC Rd, Rr Add with Carry two Registers Rd ← Rd + Rr + C Z,C,N,V,H 1

ADIW Rdl,K Add Immediate to Word Rdh:Rdl ← Rdh:Rdl + K Z,C,N,V,S 2

SUB Rd, Rr Subtract two Registers Rd ← Rd - Rr Z,C,N,V,H 1

SUBI Rd, K Subtract Constant from Register Rd ← Rd - K Z,C,N,V,H 1

SBC Rd, Rr Subtract with Carry two Registers Rd ← Rd - Rr - C Z,C,N,V,H 1

SBCI Rd, K Subtract with Carry Constant from Reg. Rd ← Rd - K - C Z,C,N,V,H 1

SBIW Rdl,K Subtract Immediate from Word Rdh:Rdl ← Rdh:Rdl - K Z,C,N,V,S 2

AND Rd, Rr Logical AND Registers Rd ← Rd • Rr Z,N,V 1

ANDI Rd, K Logical AND Register and Constant Rd ← Rd • K Z,N,V 1

OR Rd, Rr Logical OR Registers Rd ← Rd v Rr Z,N,V 1

ORI Rd, K Logical OR Register and Constant Rd ← Rd v K Z,N,V 1

EOR Rd, Rr Exclusive OR Registers Rd ← Rd ⊕ Rr Z,N,V 1

COM Rd One’s Complement Rd ← 0xFF − Rd Z,C,N,V 1

NEG Rd Two’s Complement Rd ← 0x00 − Rd Z,C,N,V,H 1

SBR Rd,K Set Bit(s) in Register Rd ← Rd v K Z,N,V 1

CBR Rd,K Clear Bit(s) in Register Rd ← Rd • (0xFF - K) Z,N,V 1

INC Rd Increment Rd ← Rd + 1 Z,N,V 1

DEC Rd Decrement Rd ← Rd − 1 Z,N,V 1

TST Rd Test for Zero or Minus Rd ← Rd • Rd Z,N,V 1

CLR Rd Clear Register Rd ← Rd ⊕ Rd Z,N,V 1

SER Rd Set Register Rd ← 0xFF None 1

MUL Rd, Rr Multiply Unsigned R1:R0 ← Rd x Rr Z,C 2

MULS Rd, Rr Multiply Signed R1:R0 ← Rd x Rr Z,C 2

MULSU Rd, Rr Multiply Signed with Unsigned R1:R0 ← Rd x Rr Z,C 2

FMUL Rd, Rr Fractional Multiply Unsigned R1:R0 ← (Rd x Rr) << 1 Z,C 2

FMULS Rd, Rr Fractional Multiply Signed R1:R0 ← (Rd x Rr) << 1 Z,C 2

FMULSU Rd, Rr Fractional Multiply Signed with Unsigned R1:R0 ← (Rd x Rr) << 1 Z,C 2

BRANCH INSTRUCTIONS

RJMP k Relative Jump PC ← PC + k + 1 None 2

IJMP Indirect Jump to (Z) PC ← Z None 2

EIJMP Extended Indirect Jump to (Z) PC ←(EIND:Z) None 2

JMP k Direct Jump PC ← k None 3

RCALL k Relative Subroutine Call PC ← PC + k + 1 None 4

ICALL Indirect Call to (Z) PC ← Z None 4

EICALL Extended Indirect Call to (Z) PC ←(EIND:Z) None 4

CALL k Direct Subroutine Call PC ← k None 5

RET Subroutine Return PC ← STACK None 5

RETI Interrupt Return PC ← STACK I 5

CPSE Rd,Rr Compare, Skip if Equal if (Rd = Rr) PC ← PC + 2 or 3 None 1/2/3

CP Rd,Rr Compare Rd − Rr Z, N,V,C,H 1

CPC Rd,Rr Compare with Carry Rd − Rr − C Z, N,V,C,H 1

CPI Rd,K Compare Register with Immediate Rd − K Z, N,V,C,H 1

SBRC Rr, b Skip if Bit in Register Cleared if (Rr(b)=0) PC ← PC + 2 or 3 None 1/2/3

SBRS Rr, b Skip if Bit in Register is Set if (Rr(b)=1) PC ← PC + 2 or 3 None 1/2/3

SBIC P, b Skip if Bit in I/O Register Cleared if (P(b)=0) PC ← PC + 2 or 3 None 1/2/3

SBIS P, b Skip if Bit in I/O Register is Set if (P(b)=1) PC ← PC + 2 or 3 None 1/2/3

BRBS s, k Branch if Status Flag Set if (SREG(s) = 1) then PC←PC+k + 1 None 1/2

BRBC s, k Branch if Status Flag Cleared if (SREG(s) = 0) then PC←PC+k + 1 None 1/2

BREQ k Branch if Equal if (Z = 1) then PC ← PC + k + 1 None 1/2

BRNE k Branch if Not Equal if (Z = 0) then PC ← PC + k + 1 None 1/2

BRCS k Branch if Carry Set if (C = 1) then PC ← PC + k + 1 None 1/2

BRCC k Branch if Carry Cleared if (C = 0) then PC ← PC + k + 1 None 1/2

BRSH k Branch if Same or Higher if (C = 0) then PC ← PC + k + 1 None 1/2

BRLO k Branch if Lower if (C = 1) then PC ← PC + k + 1 None 1/2

BRMI k Branch if Minus if (N = 1) then PC ← PC + k + 1 None 1/2

BRPL k Branch if Plus if (N = 0) then PC ← PC + k + 1 None 1/2

BRGE k Branch if Greater or Equal, Signed if (N ⊕ V= 0) then PC ← PC + k + 1 None 1/2

BRLT k Branch if Less Than Zero, Signed if (N ⊕ V= 1) then PC ← PC + k + 1 None 1/2

BRHS k Branch if Half Carry Flag Set if (H = 1) then PC ← PC + k + 1 None 1/2

BRHC k Branch if Half Carry Flag Cleared if (H = 0) then PC ← PC + k + 1 None 1/2

BRTS k Branch if T Flag Set if (T = 1) then PC ← PC + k + 1 None 1/2

BRTC k Branch if T Flag Cleared if (T = 0) then PC ← PC + k + 1 None 1/2

391

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

BRVS k Branch if Overflow Flag is Set if (V = 1) then PC ← PC + k + 1 None 1/2

BRVC k Branch if Overflow Flag is Cleared if (V = 0) then PC ← PC + k + 1 None 1/2

BRIE k Branch if Interrupt Enabled if ( I = 1) then PC ← PC + k + 1 None 1/2

BRID k Branch if Interrupt Disabled if ( I = 0) then PC ← PC + k + 1 None 1/2

BIT AND BIT-TEST INSTRUCTIONS

SBI P,b Set Bit in I/O Register I/O(P,b) ← 1 None 2

CBI P,b Clear Bit in I/O Register I/O(P,b) ← 0 None 2

LSL Rd Logical Shift Left Rd(n+1) ← Rd(n), Rd(0) ← 0 Z,C,N,V 1

LSR Rd Logical Shift Right Rd(n) ← Rd(n+1), Rd(7) ← 0 Z,C,N,V 1

ROL Rd Rotate Left Through Carry Rd(0)←C,Rd(n+1)← Rd(n),C←Rd(7) Z,C,N,V 1

ROR Rd Rotate Right Through Carry Rd(7)←C,Rd(n)← Rd(n+1),C←Rd(0) Z,C,N,V 1

ASR Rd Arithmetic Shift Right Rd(n) ← Rd(n+1), n=0..6 Z,C,N,V 1

SWAP Rd Swap Nibbles Rd(3..0)←Rd(7..4),Rd(7..4)←Rd(3..0) None 1

BSET s Flag Set SREG(s) ← 1 SREG(s) 1

BCLR s Flag Clear SREG(s) ← 0 SREG(s) 1

BST Rr, b Bit Store from Register to T T ← Rr(b) T 1

BLD Rd, b Bit load from T to Register Rd(b) ← T None 1

SEC Set Carry C ← 1 C 1

CLC Clear Carry C ← 0 C 1

SEN Set Negative Flag N ← 1 N 1

CLN Clear Negative Flag N ← 0 N 1

SEZ Set Zero Flag Z ← 1 Z 1

CLZ Clear Zero Flag Z ← 0 Z 1

SEI Global Interrupt Enable I ← 1 I 1

CLI Global Interrupt Disable I ← 0 I 1

SES Set Signed Test Flag S ← 1 S 1

CLS Clear Signed Test Flag S ← 0 S 1

SEV Set Twos Complement Overflow. V ← 1 V 1

CLV Clear Twos Complement Overflow V ← 0 V 1

SET Set T in SREG T ← 1 T 1

CLT Clear T in SREG T ← 0 T 1

SEH Set Half Carry Flag in SREG H ← 1 H 1

CLH Clear Half Carry Flag in SREG H ← 0 H 1

DATA TRANSFER INSTRUCTIONS

MOV Rd, Rr Move Between Registers Rd ← Rr None 1

MOVW Rd, Rr Copy Register Word Rd+1:Rd ← Rr+1:Rr None 1

LDI Rd, K Load Immediate Rd ← K None 1

LD Rd, X Load Indirect Rd ← (X) None 2

LD Rd, X+ Load Indirect and Post-Inc. Rd ← (X), X ← X + 1 None 2

LD Rd, - X Load Indirect and Pre-Dec. X ← X - 1, Rd ← (X) None 2

LD Rd, Y Load Indirect Rd ← (Y) None 2

LD Rd, Y+ Load Indirect and Post-Inc. Rd ← (Y), Y ← Y + 1 None 2

LD Rd, - Y Load Indirect and Pre-Dec. Y ← Y - 1, Rd ← (Y) None 2

LDD Rd,Y+q Load Indirect with Displacement Rd ← (Y + q) None 2

LD Rd, Z Load Indirect Rd ← (Z) None 2

LD Rd, Z+ Load Indirect and Post-Inc. Rd ← (Z), Z ← Z+1 None 2

LD Rd, -Z Load Indirect and Pre-Dec. Z ← Z - 1, Rd ← (Z) None 2

LDD Rd, Z+q Load Indirect with Displacement Rd ← (Z + q) None 2

LDS Rd, k Load Direct from SRAM Rd ← (k) None 2

ST X, Rr Store Indirect (X) ← Rr None 2

ST X+, Rr Store Indirect and Post-Inc. (X) ← Rr, X ← X + 1 None 2

ST - X, Rr Store Indirect and Pre-Dec. X ← X - 1, (X) ← Rr None 2

ST Y, Rr Store Indirect (Y) ← Rr None 2

ST Y+, Rr Store Indirect and Post-Inc. (Y) ← Rr, Y ← Y + 1 None 2

ST - Y, Rr Store Indirect and Pre-Dec. Y ← Y - 1, (Y) ← Rr None 2

STD Y+q,Rr Store Indirect with Displacement (Y + q) ← Rr None 2

ST Z, Rr Store Indirect (Z) ← Rr None 2

ST Z+, Rr Store Indirect and Post-Inc. (Z) ← Rr, Z ← Z + 1 None 2

ST -Z, Rr Store Indirect and Pre-Dec. Z ← Z - 1, (Z) ← Rr None 2

STD Z+q,Rr Store Indirect with Displacement (Z + q) ← Rr None 2

STS k, Rr Store Direct to SRAM (k) ← Rr None 2

LPM Load Program Memory R0 ← (Z) None 3

LPM Rd, Z Load Program Memory Rd ← (Z) None 3

LPM Rd, Z+ Load Program Memory and Post-Inc Rd ← (Z), Z ← Z+1 None 3

ELPM Extended Load Program Memory R0 ← (RAMPZ:Z) None 3

ELPM Rd, Z Extended Load Program Memory Rd ← (Z) None 3

Mnemonics Operands Description Operation Flags #Clocks

392 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Note: EICALL and EIJMP do not exist in ATmega640/1280/1281.ELPM does not exist in ATmega640.

ELPM Rd, Z+ Extended Load Program Memory Rd ← (RAMPZ:Z), RAMPZ:Z ←RAMPZ:Z+1 None 3

SPM Store Program Memory (Z) ← R1:R0 None -

IN Rd, P In Port Rd ← P None 1

OUT P, Rr Out Port P ← Rr None 1

PUSH Rr Push Register on Stack STACK ← Rr None 2

POP Rd Pop Register from Stack Rd ← STACK None 2

MCU CONTROL INSTRUCTIONS

NOP No Operation None 1

SLEEP Sleep (see specific descr. for Sleep function) None 1

WDR Watchdog Reset (see specific descr. for WDR/timer) None 1

BREAK Break For On-chip Debug Only None N/A

Mnemonics Operands Description Operation Flags #Clocks

393

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Ordering Information

Notes: 1. This device can also be supplied in wafer form. Please contact your local Atmel sales office for detailed ordering informationand minimum quantities.

2. See “Maximum speed vs. VCC” on page 370.3. Pb-free packaging alternative, complies to the European Directive for Restriction of Hazardous Substances (RoHS direc-

tive). Also Halide free and fully Green.

ATmega1281/2561Speed (MHz)(2) Power Supply Ordering Code Package(1) Operation Range

8 1.8 - 5.5V

ATmega1281/2561V-8AIATmega1281/2561V-8AU(3)

ATmega1281/2561V-8MIATmega1281/2561V-8MU(3)

64A64A64M164M1

Industrial(-40°C to 85°C)

16 4.5 - 5.5V

ATmega1281/2561-16AIATmega1281/2561-16AU(3)

ATmega1281/2561-16MIATmega1281/2561-16MU(3)

64A64M164A64M1

Industrial(-40°C to 85°C)

Package Type

64A 64-Lead, Thin (1.0 mm) Plastic Gull Wing Quad Flat Package (TQFP)

64M1 64-pad, 9 x 9 x 1.0 mm body, lead pitch 0.50 mm, Micro Lead Frame Package (QFN/MLF)

100A 100-Lead, Thin (1.0 mm) Plastic Gull Wing Quad Flat Package (TQFP)

394 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Notes: 1. This device can also be supplied in wafer form. Please contact your local Atmel sales office for detailed ordering informationand minimum quantities.

2. See “Maximum speed vs. VCC” on page 370.3. Pb-free packaging alternative, complies to the European Directive for Restriction of Hazardous Substances (RoHS direc-

tive). Also Halide free and fully Green.

ATmega640/1280/2560Speed (MHz)(2) Power Supply Ordering Code Package(1) Operation Range

81.8 - 5.5V ATmega640/1280/2560V-8AI

ATmega640/1280/2560V-8AU(3)100A100A

Industrial (-40°C to 85°C)

164.5 - 5.5V ATmega640/1280/2560-16AI

ATmega640/1280/2560-16AU(3)100A100A

Industrial (-40°C to 85°C)

Package Type

64A 64-Lead, Thin (1.0 mm) Plastic Gull Wing Quad Flat Package (TQFP)

64M1 64-pad, 9 x 9 x 1.0 mm body, lead pitch 0.50 mm, Micro Lead Frame Package (QFN/MLF)

100A 100-Lead, Thin (1.0 mm) Plastic Gull Wing Quad Flat Package (TQFP)

395

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Packaging Information

100A

2325 Orchard Parkway San Jose, CA 95131

TITLE DRAWING NO.

R

REV.

100A, 100-lead, 14 x 14 mm Body Size, 1.0 mm Body Thickness,0.5 mm Lead Pitch, Thin Profile Plastic Quad Flat Package (TQFP)

C100A

10/5/2001

PIN 1 IDENTIFIER

0˚~7˚

PIN 1

L

C

A1 A2 A

D1

D

e E1 E

B

A – – 1.20

A1 0.05 – 0.15

A2 0.95 1.00 1.05

D 15.75 16.00 16.25

D1 13.90 14.00 14.10 Note 2

E 15.75 16.00 16.25

E1 13.90 14.00 14.10 Note 2

B 0.17 – 0.27

C 0.09 – 0.20

L 0.45 – 0.75

e 0.50 TYP

Notes: 1. This package conforms to JEDEC reference MS-026, Variation AED. 2. Dimensions D1 and E1 do not include mold protrusion. Allowable

protrusion is 0.25 mm per side. Dimensions D1 and E1 are maximum plastic body size dimensions including mold mismatch.

3. Lead coplanarity is 0.08 mm maximum.

COMMON DIMENSIONS(Unit of Measure = mm)

SYMBOL MIN NOM MAX NOTE

396 ATmega640/1280/1281/2560/25612549A–AVR–03/05

64A

2325 Orchard Parkway San Jose, CA 95131

TITLE DRAWING NO.

R

REV.

64A, 64-lead, 14 x 14 mm Body Size, 1.0 mm Body Thickness,0.8 mm Lead Pitch, Thin Profile Plastic Quad Flat Package (TQFP)

B64A

10/5/2001

PIN 1 IDENTIFIER

0˚~7˚

PIN 1

L

C

A1 A2 A

D1

D

e E1 E

B

COMMON DIMENSIONS(Unit of Measure = mm)

SYMBOL MIN NOM MAX NOTE

Notes: 1. This package conforms to JEDEC reference MS-026, Variation AEB. 2. Dimensions D1 and E1 do not include mold protrusion. Allowable

protrusion is 0.25 mm per side. Dimensions D1 and E1 are maximum plastic body size dimensions including mold mismatch.

3. Lead coplanarity is 0.10 mm maximum.

A – – 1.20

A1 0.05 – 0.15

A2 0.95 1.00 1.05

D 15.75 16.00 16.25

D1 13.90 14.00 14.10 Note 2

E 15.75 16.00 16.25

E1 13.90 14.00 14.10 Note 2

B 0.30 – 0.45

C 0.09 – 0.20

L 0.45 – 0.75

e 0.80 TYP

397

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

64M1

2325 Orchard Parkway San Jose, CA 95131

TITLE DRAWING NO.

R

REV. 64M1, 64-pad, 9 x 9 x 1.0 mm Body, Lead Pitch 0.50 mm,

D64M1

8/19/04

COMMON DIMENSIONS(Unit of Measure = mm)

SYMBOL MIN NOM MAX NOTE

A 0.80 0.90 1.00

A1 – 0.02 0.05

b 0.23 0.25 0.28

D 9.00 BSC

D2 5.20 5.40 5.60

E 9.00 BSC

E2 5.20 5.40 5.60

e 0.50 BSC

L 0.35 0.40 0.45

Note: JEDEC Standard MO-220, (SAW Singulation) Fig. 1, VMMD.

TOP VIEW

SIDE VIEW

BOTTOM VIEW

D

E

Marked Pin# 1 ID

SEATING PLANE

A1

C

A

C0.08

123

K 0.20 – –

E2

D2

b e

Pin #1 CornerL

Pin #1 Triangle

Pin #1 Chamfer(C 0.30)

Option A

Option B

Pin #1 Notch(0.20 R)

Option C

K

K

5.40 mm Exposed Pad, Micro Lead Frame Package (MLF)

398 ATmega640/1280/1281/2560/25612549A–AVR–03/05

Errata

ATmega640 rev. A

No errata.

ATmega1280 rev. A

No errata.

ATmega1280 rev. A

No errata.

ATmega2560 rev. A• Non-Read-While-Write area of flash not functional• Part does not work under 2.0 Volts

1. Non-Read-While-Write area of flash not functional

The Non-Read-While-Write area of the flash is not working as expected. The prob-lem is related to the speed of the part when reading the flash of this area.

Problem Fix/Workaround

- Only use the first 248K of the flash.

- If boot functionality is needed, run the code in the Non-Read-While-Write area atmaximum 1/4th of the maximum frequency of the device at any given voltage. Thisis done by writing the CLKPR register before entering the boot section of the code

2. Part does not work under 2.0 Volts

The part does not execute code correctly below 2.0 Volts

Problem Fix/Workaround

Do not use the part at voltages below 2.0 Volts.

ATmega2561 rev. A• Non-Read-While-Write area of flash not functional• Part does not work under 2.0 Volts

1. Non-Read-While-Write area of flash not functional

The Non-Read-While-Write area of the flash is not working as expected. The prob-lem is related to the speed of the part when reading the flash of this area.

Problem Fix/Workaround

- Only use the first 248K of the flash.

- If boot functionality is needed, run the code in the Non-Read-While-Write area atmaximum 1/4th of the maximum frequency of the device at any given voltage. Thisis done by writing the CLKPR register before entering the boot section of the code

2. Part does not work under 2.0 Volts

The part does not execute code correctly below 2.0 Volts

Problem Fix/Workaround

Do not use the part at voltages below 2.0 Volts.

399

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Datasheet Revision History

Please note that the referring page numbers in this section are referring to this docu-ment.The referring revision in this section are referring to the document revision.

Rev. 2549A-12/041. Initial version.

400 ATmega640/1280/1281/2560/25612549A–AVR–03/05

i

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Table of Contents Features................................................................................................ 1

Pin Configurations............................................................................... 2Disclaimer ............................................................................................................. 3

Overview............................................................................................... 4Block Diagram ...................................................................................................... 4Comparison Between ATmega2561 and ATmega2560 ....................................... 5Pin Descriptions.................................................................................................... 6

About Code Examples......................................................................... 8

AVR CPU Core ..................................................................................... 9Introduction........................................................................................................... 9Architectural Overview.......................................................................................... 9ALU – Arithmetic Logic Unit................................................................................ 10Status Register ................................................................................................... 10General Purpose Register File ........................................................................... 12Stack Pointer ...................................................................................................... 13Instruction Execution Timing............................................................................... 14Reset and Interrupt Handling.............................................................................. 15

AVR ATmega256/2560 Memories ..................................................... 18In-System Reprogrammable Flash Program Memory ........................................ 18SRAM Data Memory........................................................................................... 19EEPROM Data Memory...................................................................................... 21I/O Memory......................................................................................................... 27External Memory Interface.................................................................................. 28

System Clock and Clock Options .................................................... 37Clock Systems and their Distribution .................................................................. 37Clock Sources..................................................................................................... 38Low Power Crystal Oscillator.............................................................................. 39Full Swing Crystal Oscillator ............................................................................... 41Low Frequency Crystal Oscillator ....................................................................... 43Calibrated Internal RC Oscillator ........................................................................ 43128 kHz Internal Oscillator.................................................................................. 45External Clock..................................................................................................... 45Clock Output Buffer ............................................................................................ 46Timer/Counter Oscillator..................................................................................... 46System Clock Prescaler...................................................................................... 46

Power Management and Sleep Modes............................................. 49Idle Mode............................................................................................................ 50ADC Noise Reduction Mode............................................................................... 50Power-down Mode.............................................................................................. 50

ii ATmega640/1280/1281/2560/25612549A–AVR–03/05

Power-save Mode............................................................................................... 50Standby Mode..................................................................................................... 51Extended Standby Mode .................................................................................... 51Power Reduction Register .................................................................................. 52Minimizing Power Consumption ......................................................................... 53

System Control and Reset............................................................................................ 55

Internal Voltage Reference................................................................................. 60Watchdog Timer ................................................................................................. 61

Interrupts ............................................................................................ 66Interrupt Vectors in ATmega256/2560................................................................ 66

I/O-Ports.............................................................................................. 72Introduction......................................................................................................... 72Ports as General Digital I/O................................................................................ 73Alternate Port Functions ..................................................................................... 77Register Description for I/O-Ports..................................................................... 103

External Interrupts........................................................................... 107

8-bit Timer/Counter0 with PWM...................................................... 112Overview........................................................................................................... 112Timer/Counter Clock Sources........................................................................... 113Counter Unit...................................................................................................... 113Output Compare Unit........................................................................................ 114Compare Match Output Unit ............................................................................. 115Modes of Operation .......................................................................................... 116Timer/Counter Timing Diagrams....................................................................... 1218-bit Timer/Counter Register Description ......................................................... 123

16-bit Timer/Counter (Timer/Counter1, Timer/Counter3, Tim-er/Counter4 and Timer/Counter5) .................................................. 129

Overview........................................................................................................... 129Accessing 16-bit Registers ............................................................................... 131Timer/Counter Clock Sources........................................................................... 135Counter Unit...................................................................................................... 135Input Capture Unit............................................................................................. 136Output Compare Units ...................................................................................... 138Compare Match Output Unit ............................................................................. 140Modes of Operation .......................................................................................... 141Timer/Counter Timing Diagrams....................................................................... 14816-bit Timer/Counter Register Description ....................................................... 151

Timer/Counter0, Timer/Counter1, Timer/Counter3, Timer/Counter4,

iii

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

and Timer/Counter5 Prescalers...................................................... 163

Output Compare Modulator (OCM1C0A) ....................................... 165Overview........................................................................................................... 165Description........................................................................................................ 165

8-bit Timer/Counter2 with PWM and Asynchronous Operation .. 167Overview........................................................................................................... 167Timer/Counter Clock Sources........................................................................... 168Counter Unit...................................................................................................... 168Output Compare Unit........................................................................................ 169Compare Match Output Unit ............................................................................. 171Modes of Operation .......................................................................................... 172Timer/Counter Timing Diagrams....................................................................... 1768-bit Timer/Counter Register Description ......................................................... 178Asynchronous operation of the Timer/Counter ................................................. 183Timer/Counter Prescaler................................................................................... 187

Serial Peripheral Interface – SPI..................................................... 188SS Pin Functionality.......................................................................................... 193Data Modes ...................................................................................................... 195

USART .............................................................................................. 197Overview........................................................................................................... 197Clock Generation .............................................................................................. 199Frame Formats ................................................................................................. 202USART Initialization.......................................................................................... 203Data Transmission – The USART Transmitter ................................................. 204Data Reception – The USART Receiver .......................................................... 206Asynchronous Data Reception ......................................................................... 210Multi-processor Communication Mode ............................................................. 213USART Register Description ............................................................................ 214Examples of Baud Rate Setting........................................................................ 219

USART in SPI Mode ......................................................................... 223Overview........................................................................................................... 223Clock Generation .............................................................................................. 223SPI Data Modes and Timing............................................................................. 223Frame Formats ................................................................................................. 224Data Transfer.................................................................................................... 226USART MSPIM Register Description ............................................................... 228AVR USART MSPIM vs.AVR SPI............................................................................................................ 230

2-wire Serial Interface...................................................................... 232Features............................................................................................................ 232

iv ATmega640/1280/1281/2560/25612549A–AVR–03/05

2-wire Serial Interface Bus Definition................................................................ 232Data Transfer and Frame Format ..................................................................... 233Multi-master Bus Systems, Arbitration and Synchronization ............................ 236Overview of the TWI Module ............................................................................ 238TWI Register Description.................................................................................. 240Using the TWI................................................................................................... 243Transmission Modes......................................................................................... 247Multi-master Systems and Arbitration............................................................... 260

Analog Comparator ......................................................................... 262Analog Comparator Multiplexed Input .............................................................. 264

Analog to Digital Converter ............................................................ 265Features............................................................................................................ 265Operation.......................................................................................................... 266Starting a Conversion ....................................................................................... 267Prescaling and Conversion Timing................................................................... 268Changing Channel or Reference Selection ...................................................... 271ADC Noise Canceler......................................................................................... 272ADC Conversion Result.................................................................................... 277

JTAG Interface and On-chip Debug System ................................. 285Overview........................................................................................................... 285Test Access Port – TAP.................................................................................... 285TAP Controller .................................................................................................. 287Using the Boundary-scan Chain ....................................................................... 288Using the On-chip Debug System .................................................................... 288On-chip Debug Specific JTAG Instructions ...................................................... 289On-chip Debug Related Register in I/O Memory .............................................. 290Using the JTAG Programming Capabilities ...................................................... 290Bibliography ...................................................................................................... 290

IEEE 1149.1 (JTAG) Boundary-scan .............................................. 291Features............................................................................................................ 291System Overview.............................................................................................. 291Data Registers .................................................................................................. 291Boundary-scan Specific JTAG Instructions ...................................................... 293Boundary-scan Related Register in I/O Memory .............................................. 294Boundary-scan Chain ....................................................................................... 295ATmega256/2560 Boundary-scan Order.......................................................... 298Boundary-scan Description Language Files ..................................................... 305

Boot Loader Support – Read-While-Write Self-Programming..... 306Boot Loader Features....................................................................................... 306Application and Boot Loader Flash Sections .................................................... 306Read-While-Write and No Read-While-Write Flash Sections........................... 306

v

ATmega640/1280/1281/2560/2561

2549A–AVR–03/05

Boot Loader Lock Bits....................................................................................... 308Entering the Boot Loader Program................................................................... 309Addressing the Flash During Self-Programming .............................................. 311Self-Programming the Flash............................................................................. 312

Memory Programming..................................................................... 320Program And Data Memory Lock Bits .............................................................. 320Fuse Bits........................................................................................................... 321Signature Bytes ................................................................................................ 323Calibration Byte ................................................................................................ 323Parallel Programming Parameters, Pin Mapping, and Commands .................. 323Parallel Programming ....................................................................................... 326Serial Downloading........................................................................................... 334Serial Programming Pin Mapping ..................................................................... 334Programming via the JTAG Interface ............................................................... 338

Electrical Characteristics................................................................ 350Absolute Maximum Ratings*............................................................................. 350DC Characteristics............................................................................................ 350External Clock Drive Waveforms...................................................................... 352External Clock Drive ......................................................................................... 352Maximum speed vs. VCC................................................................................... 3532-wire Serial Interface Characteristics .............................................................. 354SPI Timing Characteristics ............................................................................... 355ADC Characteristics – Preliminary Data........................................................... 357External Data Memory Timing .......................................................................... 359

ATmega256/2560 Typical Characteristics – Preliminary Data..... 364Supply Current of IO modules .......................................................................... 364

Register Summary ........................................................................... 367

Instruction Set Summary ................................................................ 372

Ordering Information....................................................................... 375ATmega2561 .................................................................................................... 375ATmega2560 .................................................................................................... 376

Packaging Information .................................................................... 377100A ................................................................................................................. 37764A ................................................................................................................... 37864M1................................................................................................................. 379

Table of Contents ................................................................................. i

vi ATmega640/1280/1281/2560/25612549A–AVR–03/05

Printed on recycled paper.

2549A–AVR–03/05

Disclaimer: The information in this document is provided in connection with Atmel products. No license, express or implied, by estoppel or otherwise, to anyintellectual property right is granted by this document or in connection with the sale of Atmel products. EXCEPT AS SET FORTH IN ATMEL’S TERMS AND CONDI-TIONS OF SALE LOCATED ON ATMEL’S WEB SITE, ATMEL ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY EXPRESS, IMPLIED OR STATUTORYWARRANTY RELATING TO ITS PRODUCTS INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULARPURPOSE, OR NON-INFRINGEMENT. IN NO EVENT SHALL ATMEL BE LIABLE FOR ANY DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDEN-TAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF PROFITS, BUSINESS INTERRUPTION, OR LOSS OF INFORMATION) ARISING OUTOF THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF ATMEL HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Atmel makes norepresentations or warranties with respect to the accuracy or completeness of the contents of this document and reserves the right to make changes to specificationsand product descriptions at any time without notice. Atmel does not make any commitment to update the information contained herein. Atmel’s products are notintended, authorized, or warranted for use as components in applications intended to support or sustain life.

Atmel Corporation Atmel Operations

2325 Orchard ParkwaySan Jose, CA 95131, USATel: 1(408) 441-0311Fax: 1(408) 487-2600

Regional Headquarters

EuropeAtmel SarlRoute des Arsenaux 41Case Postale 80CH-1705 FribourgSwitzerlandTel: (41) 26-426-5555Fax: (41) 26-426-5500

AsiaRoom 1219Chinachem Golden Plaza77 Mody Road TsimshatsuiEast KowloonHong KongTel: (852) 2721-9778Fax: (852) 2722-1369

Japan9F, Tonetsu Shinkawa Bldg.1-24-8 ShinkawaChuo-ku, Tokyo 104-0033JapanTel: (81) 3-3523-3551Fax: (81) 3-3523-7581

Memory2325 Orchard ParkwaySan Jose, CA 95131, USATel: 1(408) 441-0311Fax: 1(408) 436-4314

Microcontrollers2325 Orchard ParkwaySan Jose, CA 95131, USATel: 1(408) 441-0311Fax: 1(408) 436-4314

La ChantrerieBP 7060244306 Nantes Cedex 3, FranceTel: (33) 2-40-18-18-18Fax: (33) 2-40-18-19-60

ASIC/ASSP/Smart CardsZone Industrielle13106 Rousset Cedex, FranceTel: (33) 4-42-53-60-00Fax: (33) 4-42-53-60-01

1150 East Cheyenne Mtn. Blvd.Colorado Springs, CO 80906, USATel: 1(719) 576-3300Fax: 1(719) 540-1759

Scottish Enterprise Technology ParkMaxwell BuildingEast Kilbride G75 0QR, Scotland Tel: (44) 1355-803-000Fax: (44) 1355-242-743

RF/AutomotiveTheresienstrasse 2Postfach 353574025 Heilbronn, GermanyTel: (49) 71-31-67-0Fax: (49) 71-31-67-2340

1150 East Cheyenne Mtn. Blvd.Colorado Springs, CO 80906, USATel: 1(719) 576-3300Fax: 1(719) 540-1759

Biometrics/Imaging/Hi-Rel MPU/High Speed Converters/RF Datacom

Avenue de RochepleineBP 12338521 Saint-Egreve Cedex, FranceTel: (33) 4-76-58-30-00Fax: (33) 4-76-58-34-80

Literature Requestswww.atmel.com/literature

© Atmel Corporation 2005. All rights reserved. Atmel®, logo and combinations thereof, AVR®, and AVR Studio® are registered trademarks,and Everywhere You AreSM are the trademarks of Atmel Corporation or its subsidiaries. Other terms and product names may be trademarks ofothers.

This datasheet has been downloaded from:

www.DatasheetCatalog.com

Datasheets for electronic components.

B.2. Hoja de datos LM78xx

710

©2001 Fairchild Semiconductor Corporation

www.fairchildsemi.com

Rev. 1.0.1

Features• Output Current up to 1A • Output Voltages of 5, 6, 8, 9, 10, 12, 15, 18, 24V • Thermal Overload Protection • Short Circuit Protection• Output Transistor Safe Operating Area Protection

DescriptionThe MC78XX/LM78XX/MC78XXA series of three terminal positive regulators are available in the TO-220/D-PAK package and with several fixed output voltages, making them useful in a wide range of applications. Each type employs internal current limiting,thermal shut down and safe operating area protection, making it essentially indestructible. If adequate heat sinkingis provided, they can deliver over 1A output current.Although designed primarily as fixed voltage regulators,these devices can be used with external components toobtain adjustable voltages and currents.

TO-220

D-PAK

1. Input 2. GND 3. Output

1

1

Internal Block Digram

MC78XX/LM78XX/MC78XXA3-Terminal 1A Positive Voltage Regulator

MC78XX/LM78XX/MC78XXA

2

Absolute Maximum Ratings

Electrical Characteristics (MC7805/LM7805)(Refer to test circuit ,0°C < TJ < 125°C, IO = 500mA, VI = 10V, CI= 0.33µF, CO= 0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Changes in Vo due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Value UnitInput Voltage (for VO = 5V to 18V)(for VO = 24V)

VIVI

3540

VV

Thermal Resistance Junction-Cases (TO-220) RθJC 5 oC/WThermal Resistance Junction-Air (TO-220) RθJA 65 oC/WOperating Temperature Range TOPR 0 ~ +125 oCStorage Temperature Range TSTG -65 ~ +150 oC

Parameter Symbol ConditionsMC7805/LM7805

UnitMin. Typ. Max.

Output Voltage VOTJ =+25 oC 4.8 5.0 5.25.0mA ≤ Io ≤ 1.0A, PO ≤ 15WVI = 7V to 20V 4.75 5.0 5.25 V

Line Regulation (Note1) Regline TJ=+25 oCVO = 7V to 25V - 4.0 100

mVVI = 8V to 12V - 1.6 50

Load Regulation (Note1) Regload TJ=+25 oCIO = 5.0mA to1.5A - 9 100

mVIO =250mA to 750mA - 4 50

Quiescent Current IQ TJ =+25 oC - 5.0 8.0 mA

Quiescent Current Change ∆IQIO = 5mA to 1.0A - 0.03 0.5

mAVI= 7V to 25V - 0.3 1.3

Output Voltage Drift ∆VO/∆T IO= 5mA - -0.8 - mV/ oCOutput Noise Voltage VN f = 10Hz to 100KHz, TA=+25 oC - 42 - µV/Vo

Ripple Rejection RR f = 120HzVO = 8V to 18V 62 73 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25 oC - 2 - VOutput Resistance rO f = 1KHz - 15 - mΩShort Circuit Current ISC VI = 35V, TA =+25 oC - 230 - mAPeak Current IPK TJ =+25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

3

Electrical Characteristics (MC7806)(Refer to test circuit ,0°C < TJ < 125°C, IO = 500mA, VI =11V, CI= 0.33µF, CO= 0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Changes in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol ConditionsMC7806

UnitMin. Typ. Max.

Output Voltage VOTJ =+25 oC 5.75 6.0 6.255.0mA ≤ IO ≤ 1.0A, PO ≤ 15WVI = 8.0V to 21V 5.7 6.0 6.3 V

Line Regulation (Note1) Regline TJ =+25 oCVI = 8V to 25V - 5 120

mVVI = 9V to 13V - 1.5 60

Load Regulation (Note1) Regload TJ =+25 oCIO =5mA to 1.5A - 9 120

mVIO =250mA to750A - 3 60

Quiescent Current IQ TJ =+25 oC - 5.0 8.0 mA

Quiescent Current Change ∆IQIO = 5mA to 1A - - 0.5

mAVI = 8V to 25V - - 1.3

Output Voltage Drift ∆VO/∆T IO = 5mA - -0.8 - mV/ oCOutput Noise Voltage VN f = 10Hz to 100KHz, TA =+25 oC - 45 - µV/Vo

Ripple Rejection RR f = 120HzVI = 9V to 19V 59 75 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25 oC - 2 - VOutput Resistance rO f = 1KHz - 19 - mΩShort Circuit Current ISC VI= 35V, TA=+25 oC - 250 - mAPeak Current IPK TJ =+25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

4

Electrical Characteristics (MC7808)(Refer to test circuit ,0°C < TJ < 125°C, IO = 500mA, VI =14V, CI= 0.33µF, CO= 0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Changes in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol ConditionsMC7808

UnitMin. Typ. Max.

Output Voltage VOTJ =+25 oC 7.7 8.0 8.35.0mA ≤ IO ≤ 1.0A, PO ≤ 15WVI = 10.5V to 23V 7.6 8.0 8.4 V

Line Regulation (Note1) Regline TJ =+25 oCVI = 10.5V to 25V - 5.0 160

mVVI = 11.5V to 17V - 2.0 80

Load Regulation (Note1) Regload TJ =+25 oCIO = 5.0mA to 1.5A - 10 160

mVIO= 250mA to 750mA - 5.0 80

Quiescent Current IQ TJ =+25 oC - 5.0 8.0 mA

Quiescent Current Change ∆IQIO = 5mA to 1.0A - 0.05 0.5

mAVI = 10.5A to 25V - 0.5 1.0

Output Voltage Drift ∆VO/∆T IO = 5mA - -0.8 - mV/ oCOutput Noise Voltage VN f = 10Hz to 100KHz, TA =+25 oC - 52 - µV/VoRipple Rejection RR f = 120Hz, VI= 11.5V to 21.5V 56 73 - dBDropout Voltage VDrop IO = 1A, TJ=+25 oC - 2 - VOutput Resistance rO f = 1KHz - 17 - mΩShort Circuit Current ISC VI= 35V, TA =+25 oC - 230 - mAPeak Current IPK TJ =+25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

5

Electrical Characteristics (MC7809)(Refer to test circuit ,0°C < TJ < 125°C, IO = 500mA, VI =15V, CI= 0.33µF, CO= 0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Changes in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol ConditionsMC7809

UnitMin. Typ. Max.

Output Voltage VOTJ =+25°C 8.65 9 9.355.0mA≤ IO ≤1.0A, PO ≤15WVI= 11.5V to 24V 8.6 9 9.4 V

Line Regulation (Note1) Regline TJ=+25°CVI = 11.5V to 25V - 6 180

mVVI = 12V to 17V - 2 90

Load Regulation (Note1) Regload TJ=+25°CIO = 5mA to 1.5A - 12 180

mVIO = 250mA to 750mA - 4 90

Quiescent Current IQ TJ=+25°C - 5.0 8.0 mA

Quiescent Current Change ∆IQIO = 5mA to 1.0A - - 0.5

mAVI = 11.5V to 26V - - 1.3

Output Voltage Drift ∆VO/∆T IO = 5mA - -1 - mV/ °COutput Noise Voltage VN f = 10Hz to 100KHz, TA =+25 °C - 58 - µV/VoRipple Rejection RR f = 120Hz

VI = 13V to 23V 56 71 - dB

Dropout Voltage VDrop IO = 1A, TJ=+25°C - 2 - VOutput Resistance rO f = 1KHz - 17 - mΩShort Circuit Current ISC VI= 35V, TA =+25°C - 250 - mAPeak Current IPK TJ= +25°C - 2.2 - A

MC78XX/LM78XX/MC78XXA

6

Electrical Characteristics (MC7810)(Refer to test circuit ,0°C< TJ < 125°C, IO = 500mA, VI =16V, CI= 0.33µF, CO=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Changes in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol ConditionsMC7810

UnitMin. Typ. Max.

Output Voltage VOTJ =+25 °C 9.6 10 10.45.0mA ≤ IO≤1.0A, PO ≤15WVI = 12.5V to 25V 9.5 10 10.5 V

Line Regulation (Note1) Regline TJ =+25°CVI = 12.5V to 25V - 10 200

mVVI = 13V to 25V - 3 100

Load Regulation (Note1) Regload TJ =+25°CIO = 5mA to 1.5A - 12 200

mVIO = 250mA to 750mA - 4 400

Quiescent Current IQ TJ =+25°C - 5.1 8.0 mA

Quiescent Current Change ∆IQIO = 5mA to 1.0A - - 0.5

mAVI = 12.5V to 29V - - 1.0

Output Voltage Drift ∆VO/∆T IO = 5mA - -1 - mV/°COutput Noise Voltage VN f = 10Hz to 100KHz, TA =+25 °C - 58 - µV/Vo

Ripple Rejection RR f = 120HzVI = 13V to 23V 56 71 - dB

Dropout Voltage VDrop IO = 1A, TJ=+25 °C - 2 - VOutput Resistance rO f = 1KHz - 17 - mΩShort Circuit Current ISC VI = 35V, TA=+25 °C - 250 - mAPeak Current IPK TJ =+25 °C - 2.2 - A

MC78XX/LM78XX/MC78XXA

7

Electrical Characteristics (MC7812)(Refer to test circuit ,0°C < TJ < 125°C, IO = 500mA, VI =19V, CI= 0.33µF, CO=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Changes in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol ConditionsMC7812

UnitMin. Typ. Max.

Output Voltage VOTJ =+25 oC 11.5 12 12.55.0mA ≤ IO≤1.0A, PO≤15WVI = 14.5V to 27V 11.4 12 12.6 V

Line Regulation (Note1) Regline TJ =+25 oCVI = 14.5V to 30V - 10 240

mVVI = 16V to 22V - 3.0 120

Load Regulation (Note1) Regload TJ =+25 oCIO = 5mA to 1.5A - 11 240

mVIO = 250mA to 750mA - 5.0 120

Quiescent Current IQ TJ =+25 oC - 5.1 8.0 mA

Quiescent Current Change ∆IQIO = 5mA to 1.0A - 0.1 0.5

mAVI = 14.5V to 30V - 0.5 1.0

Output Voltage Drift ∆VO/∆T IO = 5mA - -1 - mV/ oCOutput Noise Voltage VN f = 10Hz to 100KHz, TA =+25 oC - 76 - µV/Vo

Ripple Rejection RR f = 120HzVI = 15V to 25V 55 71 - dB

Dropout Voltage VDrop IO = 1A, TJ=+25 oC - 2 - VOutput Resistance rO f = 1KHz - 18 - mΩShort Circuit Current ISC VI = 35V, TA=+25 oC - 230 - mAPeak Current IPK TJ = +25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

8

Electrical Characteristics (MC7815)(Refer to test circuit ,0°C < TJ < 125°C, IO = 500mA, VI =23V, CI= 0.33µF, CO=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Changes in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol ConditionsMC7815

UnitMin. Typ. Max.

Output Voltage VOTJ =+25 oC 14.4 15 15.65.0mA ≤ IO ≤ 1.0A, PO ≤ 15WVI = 17.5V to 30V 14.25 15 15.75 V

Line Regulation (Note1) Regline TJ =+25 oCVI = 17.5V to 30V - 11 300

mVVI = 20V to 26V - 3 150

Load Regulation (Note1) Regload TJ =+25 oCIO = 5mA to 1.5A - 12 300

mVIO = 250mA to 750mA - 4 150

Quiescent Current IQ TJ =+25 oC - 5.2 8.0 mA

Quiescent Current Change ∆IQIO = 5mA to 1.0A - - 0.5

mAVI = 17.5V to 30V - - 1.0

Output Voltage Drift ∆VO/∆T IO = 5mA - -1 - mV/ oCOutput Noise Voltage VN f = 10Hz to 100KHz, TA =+25 oC - 90 - µV/Vo

Ripple Rejection RR f = 120HzVI = 18.5V to 28.5V 54 70 - dB

Dropout Voltage VDrop IO = 1A, TJ=+25 oC - 2 - VOutput Resistance rO f = 1KHz - 19 - mΩShort Circuit Current ISC VI = 35V, TA=+25 oC - 250 - mAPeak Current IPK TJ =+25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

9

Electrical Characteristics (MC7818)(Refer to test circuit ,0°C < TJ < 125°C, IO = 500mA, VI =27V, CI= 0.33µF, CO=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Changes in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol ConditionsMC7818

UnitMin. Typ. Max.

Output Voltage VOTJ =+25 oC 17.3 18 18.75.0mA ≤ IO ≤1.0A, PO ≤15WVI = 21V to 33V 17.1 18 18.9 V

Line Regulation (Note1) Regline TJ =+25 oCVI = 21V to 33V - 15 360

mVVI = 24V to 30V - 5 180

Load Regulation (Note1) Regload TJ =+25 oCIO = 5mA to 1.5A - 15 360

mVIO = 250mA to 750mA - 5.0 180

Quiescent Current IQ TJ =+25 oC - 5.2 8.0 mA

Quiescent Current Change ∆IQIO = 5mA to 1.0A - - 0.5

mAVI = 21V to 33V - - 1

Output Voltage Drift ∆VO/∆T IO = 5mA - -1 - mV/ oCOutput Noise Voltage VN f = 10Hz to 100KHz, TA =+25 oC - 110 - µV/Vo

Ripple Rejection RR f = 120HzVI = 22V to 32V 53 69 - dB

Dropout Voltage VDrop IO = 1A, TJ=+25 oC - 2 - VOutput Resistance rO f = 1KHz - 22 - mΩShort Circuit Current ISC VI = 35V, TA=+25 oC - 250 - mAPeak Current IPK TJ =+25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

10

Electrical Characteristics (MC7824)(Refer to test circuit ,0°C < TJ < 125°C, IO = 500mA, VI =33V, CI= 0.33µF, CO=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Changes in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol ConditionsMC7824

UnitMin. Typ. Max.

Output Voltage VOTJ =+25 oC 23 24 255.0mA ≤ IO ≤ 1.0A, PO ≤ 15WVI = 27V to 38V 22.8 24 25.25 V

Line Regulation (Note1) Regline TJ =+25 oCVI = 27V to 38V - 17 480

mVVI = 30V to 36V - 6 240

Load Regulation (Note1) Regload TJ =+25 oCIO = 5mA to 1.5A - 15 480

mVIO = 250mA to 750mA - 5.0 240

Quiescent Current IQ TJ =+25 oC - 5.2 8.0 mA

Quiescent Current Change ∆IQIO = 5mA to 1.0A - 0.1 0.5

mAVI = 27V to 38V - 0.5 1

Output Voltage Drift ∆VO/∆T IO = 5mA - -1.5 - mV/ oCOutput Noise Voltage VN f = 10Hz to 100KHz, TA =+25 oC - 60 - µV/Vo

Ripple Rejection RR f = 120HzVI = 28V to 38V 50 67 - dB

Dropout Voltage VDrop IO = 1A, TJ=+25 oC - 2 - VOutput Resistance rO f = 1KHz - 28 - mΩShort Circuit Current ISC VI = 35V, TA=+25 oC - 230 - mAPeak Current IPK TJ =+25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

11

Electrical Characteristics (MC7805A)(Refer to the test circuits. 0°C < TJ < 125°C, Io =1A, V I = 10V, C I=0.33µF, C O=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Change in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Conditions Min. Typ. Max. Unit

Output Voltage VOTJ =+25 oC 4.9 5 5.1

VIO = 5mA to 1A, PO ≤ 15WVI = 7.5V to 20V 4.8 5 5.2

Line Regulation (Note1) Regline

VI = 7.5V to 25VIO = 500mA - 5 50

mVVI = 8V to 12V - 3 50

TJ =+25 oCVI= 7.3V to 20V - 5 50VI= 8V to 12V - 1.5 25

Load Regulation (Note1) Regload

TJ =+25 oCIO = 5mA to 1.5A - 9 100

mVIO = 5mA to 1A - 9 100IO = 250mA to 750mA - 4 50

Quiescent Current IQ TJ =+25 oC - 5.0 6 mA

Quiescent Current Change ∆IQ

IO = 5mA to 1A - - 0.5mAVI = 8 V to 25V, IO = 500mA - - 0.8

VI = 7.5V to 20V, TJ =+25 oC - - 0.8Output Voltage Drift ∆V/∆T Io = 5mA - -0.8 - mV/ oC

Output Noise Voltage VNf = 10Hz to 100KHzTA =+25 oC - 10 - µV/Vo

Ripple Rejection RR f = 120Hz, IO = 500mAVI = 8V to 18V - 68 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25 oC - 2 - VOutput Resistance rO f = 1KHz - 17 - mΩShort Circuit Current ISC VI= 35V, TA =+25 oC - 250 - mAPeak Current IPK TJ= +25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

12

Electrical Characteristics (MC7806A)(Refer to the test circuits. 0°C < TJ < 125°C, Io =1A, V I =11V, C I=0.33µF, C O=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Change in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Conditions Min. Typ. Max. Unit

Output Voltage VOTJ =+25 oC 5.58 6 6.12

VIO = 5mA to 1A, PO ≤ 15WVI = 8.6V to 21V 5.76 6 6.24

Line Regulation (Note1) Regline

VI= 8.6V to 25VIO = 500mA - 5 60

mVVI= 9V to 13V - 3 60

TJ =+25 oCVI= 8.3V to 21V - 5 60VI= 9V to 13V - 1.5 30

Load Regulation (Note1) Regload

TJ =+25 oCIO = 5mA to 1.5A - 9 100

mVIO = 5mA to 1A - 4 100IO = 250mA to 750mA - 5.0 50

Quiescent Current IQ TJ =+25 oC - 4.3 6 mA

Quiescent Current Change ∆IQ

IO = 5mA to 1A - - 0.5mAVI = 9V to 25V, IO = 500mA - - 0.8

VI= 8.5V to 21V, TJ =+25 oC - - 0.8Output Voltage Drift ∆V/∆T IO = 5mA - -0.8 - mV/ oC

Output Noise Voltage VNf = 10Hz to 100KHzTA =+25 oC - 10 - µV/Vo

Ripple Rejection RR f = 120Hz, IO = 500mAVI = 9V to 19V - 65 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25 oC - 2 - VOutput Resistance rO f = 1KHz - 17 - mΩShort Circuit Current ISC VI= 35V, TA =+25 oC - 250 - mAPeak Current IPK TJ=+25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

13

Electrical Characteristics (MC7808A)(Refer to the test circuits. 0°C < TJ < 125°C, Io =1A, V I = 14V, C I=0.33µF, C O=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Change in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Conditions Min. Typ. Max. Unit

Output Voltage VOTJ =+25 oC 7.84 8 8.16

VIO = 5mA to 1A, PO ≤15WVI = 10.6V to 23V 7.7 8 8.3

Line Regulation (Note1) Regline

VI= 10.6V to 25VIO = 500mA - 6 80

mVVI= 11V to 17V - 3 80

TJ =+25 oCVI= 10.4V to 23V - 6 80VI= 11V to 17V - 2 40

Load Regulation (Note1) Regload

TJ =+25 oCIO = 5mA to 1.5A - 12 100

mVIO = 5mA to 1A - 12 100IO = 250mA to 750mA - 5 50

Quiescent Current IQ TJ =+25 oC - 5.0 6 mA

Quiescent Current Change ∆IQIO = 5mA to 1A - - 0.5

mAVI = 11V to 25V, IO = 500mA - - 0.8VI= 10.6V to 23V, TJ =+25 oC - - 0.8

Output Voltage Drift ∆V/∆T IO = 5mA - -0.8 - mV/ oC

Output Noise Voltage VNf = 10Hz to 100KHzTA =+25 oC - 10 - µV/Vo

Ripple Rejection RR f = 120Hz, IO = 500mAVI = 11.5V to 21.5V - 62 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25 oC - 2 - VOutput Resistance rO f = 1KHz - 18 - mΩShort Circuit Current ISC VI= 35V, TA =+25 oC - 250 - mAPeak Current IPK TJ=+25 oC - 2.2 - A

MC78XX/LM78XX/MC78XXA

14

Electrical Characteristics (MC7809A)(Refer to the test circuits. 0°C < TJ < 125°C, Io =1A, V I = 15V, C I=0.33µF, C O=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant, junction temperature. Change in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Conditions Min. Typ. Max. Unit

Output Voltage VOTJ =+25°C 8.82 9.0 9.18

VIO = 5mA to 1A, PO≤15WVI = 11.2V to 24V 8.65 9.0 9.35

Line Regulation (Note1) Regline

VI= 11.7V to 25VIO = 500mA - 6 90

mVVI= 12.5V to 19V - 4 45

TJ =+25°C VI= 11.5V to 24V - 6 90 VI= 12.5V to 19V - 2 45

Load Regulation (Note1) Regload

TJ =+25°CIO = 5mA to 1.0A - 12 100

mVIO = 5mA to 1.0A - 12 100IO = 250mA to 750mA - 5 50

Quiescent Current IQ TJ =+25 °C - 5.0 6.0 mA

Quiescent Current Change ∆IQ

VI = 11.7V to 25V, TJ=+25 °C - - 0.8mAVI = 12V to 25V, IO = 500mA - - 0.8

IO = 5mA to 1.0A - - 0.5Output Voltage Drift ∆V/∆T IO = 5mA - -1.0 - mV/ °C

Output Noise Voltage VNf = 10Hz to 100KHzTA =+25 °C - 10 - µV/Vo

Ripple Rejection RR f = 120Hz, IO = 500mAVI = 12V to 22V - 62 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25 °C - 2.0 - VOutput Resistance rO f = 1KHz - 17 - mΩShort Circuit Current ISC VI= 35V, TA =+25 °C - 250 - mAPeak Current IPK TJ=+25°C - 2.2 - A

MC78XX/LM78XX/MC78XXA

15

Electrical Characteristics (MC7810A)(Refer to the test circuits. 0°C < TJ < 125°C, Io =1A, V I = 16V, C I=0.33µF, C O=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Change in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Conditions Min. Typ. Max. Unit

Output Voltage VO TJ =+25°C 9.8 10 10.2

V IO = 5mA to 1A, PO ≤ 15W VI =12.8V to 25V 9.6 10 10.4

Line Regulation (Note1) Regline

VI= 12.8V to 26V IO = 500mA - 8 100

mV VI= 13V to 20V - 4 50

TJ =+25 °C VI= 12.5V to 25V - 8 100 VI= 13V to 20V - 3 50

Load Regulation (Note1) Regload

TJ =+25 °C IO = 5mA to 1.5A - 12 100

mV IO = 5mA to 1.0A - 12 100 IO = 250mA to 750mA - 5 50

Quiescent Current IQ TJ =+25 °C - 5.0 6.0 mA

Quiescent Current Change ∆IQ

VI = 13V to 26V, TJ=+25 °C - - 0.5mA VI = 12.8V to 25V, IO = 500mA - - 0.8

IO = 5mA to 1.0A - - 0.5Output Voltage Drift ∆V/∆T IO = 5mA - -1.0 - mV/ °C

Output Noise Voltage VN f = 10Hz to 100KHz TA =+25 °C - 10 - µV/Vo

Ripple Rejection RR f = 120Hz, IO = 500mA VI = 14V to 24V - 62 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25°C - 2.0 - VOutput Resistance rO f = 1KHz - 17 - mΩShort Circuit Current ISC VI= 35V, TA =+25 °C - 250 - mAPeak Current IPK TJ=+25 °C - 2.2 - A

MC78XX/LM78XX/MC78XXA

16

Electrical Characteristics (MC7812A)(Refer to the test circuits. 0°C < TJ < 125°C, Io =1A, V I = 19V, C I=0.33µF, C O=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Change in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Conditions Min. Typ. Max. Unit

Output Voltage VO TJ =+25 °C 11.75 12 12.25

V IO = 5mA to 1A, PO ≤15W VI = 14.8V to 27V 11.5 12 12.5

Line Regulation (Note1) Regline

VI= 14.8V to 30V IO = 500mA - 10 120

mV VI= 16V to 22V - 4 120

TJ =+25 °C VI= 14.5V to 27V - 10 120 VI= 16V to 22V - 3 60

Load Regulation (Note1) Regload

TJ =+25 °C IO = 5mA to 1.5A - 12 100

mV IO = 5mA to 1.0A - 12 100 IO = 250mA to 750mA - 5 50

Quiescent Current IQ TJ =+25°C - 5.1 6.0 mA

Quiescent Current Change ∆IQ

VI = 15V to 30V, TJ=+25 °C - 0.8mA VI = 14V to 27V, IO = 500mA - 0.8

IO = 5mA to 1.0A - 0.5Output Voltage Drift ∆V/∆T IO = 5mA - -1.0 - mV/°C

Output Noise Voltage VN f = 10Hz to 100KHz TA =+25°C - 10 - µV/Vo

Ripple Rejection RR f = 120Hz, IO = 500mA VI = 14V to 24V - 60 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25°C - 2.0 - VOutput Resistance rO f = 1KHz - 18 - mΩShort Circuit Current ISC VI= 35V, TA =+25 °C - 250 - mAPeak Current IPK TJ=+25 °C - 2.2 - A

MC78XX/LM78XX/MC78XXA

17

Electrical Characteristics (MC7815A)(Refer to the test circuits. 0°C < TJ < 125°C, Io =1A, V I =23V, C I=0.33µF, C O=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Change in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Conditions Min. Typ. Max. Unit

Output Voltage VO TJ =+25 °C 14.7 15 15.3

V IO = 5mA to 1A, PO ≤15W VI = 17.7V to 30V 14.4 15 15.6

Line Regulation (Note1) Regline

VI= 17.9V to 30V IO = 500mA - 10 150

mV VI= 20V to 26V - 5 150

TJ =+25°C VI= 17.5V to 30V - 11 150 VI= 20V to 26V - 3 75

Load Regulation (Note1) Regload

TJ =+25 °C IO = 5mA to 1.5A - 12 100

mV IO = 5mA to 1.0A - 12 100 IO = 250mA to 750mA - 5 50

Quiescent Current IQ TJ =+25 °C - 5.2 6.0 mA

Quiescent Current Change ∆IQ

VI = 17.5V to 30V, TJ =+25 °C - - 0.8mA VI = 17.5V to 30V, IO = 500mA - - 0.8

IO = 5mA to 1.0A - - 0.5Output Voltage Drift ∆V/∆T IO = 5mA - -1.0 - mV/°C

Output Noise Voltage VN f = 10Hz to 100KHz TA =+25 °C - 10 - µV/Vo

Ripple Rejection RR f = 120Hz, IO = 500mA VI = 18.5V to 28.5V - 58 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25 °C - 2.0 - VOutput Resistance rO f = 1KHz - 19 - mΩShort Circuit Current ISC VI= 35V, TA =+25 °C - 250 - mAPeak Current IPK TJ=+25°C - 2.2 - A

MC78XX/LM78XX/MC78XXA

18

Electrical Characteristics (MC7818A)(Refer to the test circuits. 0°C < TJ < 125°C, Io =1A, V I = 27V, C I=0.33µF, C O=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Change in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Conditions Min. Typ. Max. Unit

Output Voltage VO TJ =+25 °C 17.64 18 18.36

V IO = 5mA to 1A, PO ≤15W VI = 21V to 33V 17.3 18 18.7

Line Regulation (Note1) Regline

VI= 21V to 33V IO = 500mA - 15 180

mV VI= 21V to 33V - 5 180

TJ =+25 °C VI= 20.6V to 33V - 15 180 VI= 24V to 30V - 5 90

Load Regulation (Note1) Regload

TJ =+25°C IO = 5mA to 1.5A - 15 100

mV IO = 5mA to 1.0A - 15 100 IO = 250mA to 750mA - 7 50

Quiescent Current IQ TJ =+25 °C - 5.2 6.0 mA

Quiescent Current Change ∆IQ

VI = 21V to 33V, TJ=+25 °C - - 0.8mA VI = 21V to 33V, IO = 500mA - - 0.8

IO = 5mA to 1.0A - - 0.5Output Voltage Drift ∆V/∆T IO = 5mA - -1.0 - mV/ °C

Output Noise Voltage VN f = 10Hz to 100KHz TA =+25°C - 10 - µV/Vo

Ripple Rejection RR f = 120Hz, IO = 500mA VI = 22V to 32V - 57 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25°C - 2.0 - VOutput Resistance rO f = 1KHz - 19 - mΩShort Circuit Current ISC VI= 35V, TA =+25°C - 250 - mAPeak Current IPK TJ=+25 °C - 2.2 - A

MC78XX/LM78XX/MC78XXA

19

Electrical Characteristics (MC7824A)(Refer to the test circuits. 0°C < TJ < 125°C, Io =1A, V I = 33V, C I=0.33µF, C O=0.1µF, unless otherwise specified)

Note:1. Load and line regulation are specified at constant junction temperature. Change in VO due to heating effects must be taken

into account separately. Pulse testing with low duty is used.

Parameter Symbol Conditions Min. Typ. Max. Unit

Output Voltage VO TJ =+25 °C 23.5 24 24.5

V IO = 5mA to 1A, PO ≤15W VI = 27.3V to 38V 23 24 25

Line Regulation (Note1) Regline

VI= 27V to 38V IO = 500mA - 18 240

mV VI= 21V to 33V - 6 240

TJ =+25 °C VI= 26.7V to 38V - 18 240 VI= 30V to 36V - 6 120

Load Regulation (Note1) Regload

TJ =+25 °C IO = 5mA to 1.5A - 15 100

mV IO = 5mA to 1.0A - 15 100 IO = 250mA to 750mA - 7 50

Quiescent Current IQ TJ =+25 °C - 5.2 6.0 mA

Quiescent Current Change ∆IQ

VI = 27.3V to 38V, TJ =+25 °C - - 0.8mA VI = 27.3V to 38V, IO = 500mA - - 0.8

IO = 5mA to 1.0A - - 0.5Output Voltage Drift ∆V/∆T IO = 5mA - -1.5 - mV/ °C

Output Noise Voltage VN f = 10Hz to 100KHz TA = 25 °C - 10 - µV/Vo

Ripple Rejection RR f = 120Hz, IO = 500mA VI = 28V to 38V - 54 - dB

Dropout Voltage VDrop IO = 1A, TJ =+25 °C - 2.0 - VOutput Resistance rO f = 1KHz - 20 - mΩShort Circuit Current ISC VI= 35V, TA =+25 °C - 250 - mAPeak Current IPK TJ=+25 °C - 2.2 - A

MC78XX/LM78XX/MC78XXA

20

Typical Perfomance Characteristics

Figure 1. Quiescent Current

Figure 3. Output Voltage

Figure 2. Peak Output Current

Figure 4. Quiescent Current

I

MC78XX/LM78XX/MC78XXA

21

Typical Applications

Figure 5. DC Parameters

Figure 6. Load Regulation

Figure 7. Ripple Rejection

Figure 8. Fixed Output Regulator

Input OutputMC78XX/LM78XX

Input OutputMC78XX/LM78XX

Input OutputMC78XX/LM78XX

Input OutputMC78XX/LM78XX

MC78XX/LM78XX/MC78XXA

22

Figure 9. Constant Current Regulator

Notes:(1) To specify an output voltage. substitute voltage value for "XX." A common ground is required between the input and the

Output voltage. The input voltage must remain typically 2.0V above the output voltage even during the low point on the inputripple voltage.

(2) CI is required if regulator is located an appreciable distance from power Supply filter.(3) CO improves stability and transient response.

VO = VXX(1+R2/R1)+IQR2Figure 10. Circuit for Increasing Output Voltage

IRI ≥5 IQVO = VXX(1+R2/R1)+IQR2

Figure 11. Adjustable Output Regulator (7 to 30V)

Input OutputMC78XX/LM78XX

CI

Co

Input OutputMC78XX/LM78XX

CICo

IRI 5IQ≥

Input OutputMC7805LM7805

LM741Co

CI

MC78XX/LM78XX/MC78XXA

23

Figure 12. High Current Voltage Regulator

Figure 13. High Output Current with Short Circuit Protection

Figure 14. Tracking Voltage Regulator

Input

OutputMC78XX/LM78XX

Input

OutputMC78XX/LM78XX

MC78XX/LM78XX

LM741

MC78XX/LM78XX/MC78XXA

24

Figure 15. Split Power Supply ( ±15V-1A)

Figure 16. Negative Output Voltage Circuit

Figure 17. Switching Regulator

MC7815

MC7915

Input

Output

MC78XX/LM78XX

Input Output

MC78XX/LM78XX

MC78XX/LM78XX/MC78XXA

25

Mechanical DimensionsPackage

4.50 ±0.209.90 ±0.20

1.52 ±0.10

0.80 ±0.102.40 ±0.20

10.00 ±0.20

1.27 ±0.10

ø3.60 ±0.10

(8.70)

2.80

±0.

1015

.90

±0.2

0

10.0

8 ±0

.30

18.9

5MA

X.

(1.7

0)

(3.7

0)(3

.00)

(1.4

6)

(1.0

0)

(45°)

9.20

±0.

2013

.08

±0.2

0

1.30

±0.

10

1.30+0.10–0.05

0.50+0.10–0.05

2.54TYP[2.54 ±0.20]

2.54TYP[2.54 ±0.20]

TO-220

MC78XX/LM78XX/MC78XXA

26

Mechancal Dimensions (Continued)

Package

6.60 ±0.20

2.30 ±0.10

0.50 ±0.10

5.34 ±0.30

0.70

±0.

20

0.60

±0.

200.

80 ±

0.20

9.50

±0.

30

6.10

±0.

20

2.70

±0.

209.

50 ±

0.30

6.10

±0.

20

2.70

±0.

20

MIN

0.55

0.76 ±0.10 0.50 ±0.10

1.02 ±0.20

2.30 ±0.20

6.60 ±0.20

0.76 ±0.10

(5.34)

(1.50)

(2XR0.25)

(5.04)

0.89

±0.

10

(0.1

0)(3

.05)

(1.0

0)

(0.9

0)

(0.7

0)

0.91

±0.

10

2.30TYP[2.30±0.20]

2.30TYP[2.30±0.20]

MAX0.96

(4.34)(0.50) (0.50)

D-PAK

MC78XX/LM78XX/MC78XXA

27

Ordering InformationProduct Number Output Voltage Tolerance Package Operating Temperature

LM7805CT ±4% TO-220 0 ~ + 125°C

Product Number Output Voltage Tolerance Package Operating TemperatureMC7805CT

±4%

TO-220

0 ~ + 125°C

MC7806CTMC7808CTMC7809CTMC7810CTMC7812CTMC7815CTMC7818CTMC7824CT

MC7805CDT

D-PAK

MC7806CDTMC7808CDTMC7809CDTMC7810CDTMC7812CDTMC7805ACT

±2% TO-220

MC7806ACTMC7808ACTMC7809ACTMC7810ACTMC7812ACTMC7815ACTMC7818ACTMC7824ACT

MC78XX/LM78XX/MC78XXA

7/2/01 0.0m 001Stock#DSxxxxxxxx

2001 Fairchild Semiconductor Corporation

LIFE SUPPORT POLICY FAIRCHILD’S PRODUCTS ARE NOT AUTHORIZED FOR USE AS CRITICAL COMPONENTS IN LIFE SUPPORT DEVICES OR SYSTEMS WITHOUT THE EXPRESS WRITTEN APPROVAL OF THE PRESIDENT OF FAIRCHILD SEMICONDUCTOR CORPORATION. As used herein:

1. Life support devices or systems are devices or systems which, (a) are intended for surgical implant into the body, or (b) support or sustain life, and (c) whose failure to perform when properly used in accordance with instructions for use provided in the labeling, can be reasonably expected to result in a significant injury of the user.

2. A critical component in any component of a life support device or system whose failure to perform can be reasonably expected to cause the failure of the life support device or system, or to affect its safety or effectiveness.

www.fairchildsemi.com

DISCLAIMER FAIRCHILD SEMICONDUCTOR RESERVES THE RIGHT TO MAKE CHANGES WITHOUT FURTHER NOTICE TO ANY PRODUCTS HEREIN TO IMPROVE RELIABILITY, FUNCTION OR DESIGN. FAIRCHILD DOES NOT ASSUME ANY LIABILITY ARISING OUT OF THE APPLICATION OR USE OF ANY PRODUCT OR CIRCUIT DESCRIBED HEREIN; NEITHER DOES IT CONVEY ANY LICENSE UNDER ITS PATENT RIGHTS, NOR THE RIGHTS OF OTHERS.

B.3. Hoja de datos INA114BP

739

®

INA1141

FEATURES LOW OFFSET VOLTAGE: 50 µV max

LOW DRIFT: 0.25µV/°C max

LOW INPUT BIAS CURRENT: 2nA max

HIGH COMMON-MODE REJECTION:115dB min

INPUT OVER-VOLTAGE PROTECTION:±40V

WIDE SUPPLY RANGE: ±2.25 to ±18V

LOW QUIESCENT CURRENT: 3mA max

8-PIN PLASTIC AND SOL-16

INA114

DESCRIPTIONThe INA114 is a low cost, general purpose instrumen-tation amplifier offering excellent accuracy. Its versa-tile 3-op amp design and small size make it ideal for awide range of applications.

A single external resistor sets any gain from 1 to 10,000.Internal input protection can withstand up to ±40Vwithout damage.

The INA114 is laser trimmed for very low offset voltage(50µV), drift (0.25µV/°C) and high common-moderejection (115dB at G = 1000). It operates with powersupplies as low as ±2.25V, allowing use in batteryoperated and single 5V supply systems. Quiescent cur-rent is 3mA maximum.

The INA114 is available in 8-pin plastic and SOL-16surface-mount packages. Both are specified for the–40°C to +85°C temperature range.

APPLICATIONS BRIDGE AMPLIFIER

THERMOCOUPLE AMPLIFIER

RTD SENSOR AMPLIFIER

MEDICAL INSTRUMENTATION

DATA ACQUISITION

A1

A2

A3

(12)

(11)

6

(10)25kΩ25kΩ

25kΩ25kΩ

(13)7

(7)4

(5)

3

(15)

8

(2)

1

(4)

2VIN

VIN

RG

V+

V–

INA114

DIP (SOIC)

Ref

DIP ConnectedInternally

VO

G = 1 + 50kΩRG

–

+5

Over-VoltageProtection

25kΩ

25kΩ

Over-VoltageProtection

Feedback

PrecisionINSTRUMENTATION AMPLIFIER

®

International Airport Industrial Park • Mailing Address: PO Box 11400, Tucson, AZ 85734 • Street Address: 6730 S. Tucson Bl vd., Tucson, AZ 85706 • Tel: (520) 746-1111 • Twx: 910-952-1111Internet: http://www.burr-brown.com/ • FAXLine: (800) 548-6133 (US/Canada Only) • Cable: BBRCORP • Telex: 066-6491 • FA X: (520) 889-1510 • Immediate Product Info: (800) 548-6132

INA114

INA114

©1992 Burr-Brown Corporation PDS-1142D Printed in U.S.A. March, 1998

SBOS014

®

INA114 2

SPECIFICATIONSELECTRICALAt TA = +25°C, VS = ±15V, RL = 2kΩ, unless otherwise noted.

Specification same as INA114BP/BU.

NOTE: (1) Temperature coefficient of the “50kΩ” term in the gain equation.

INA114BP, BU INA114AP, AU

PARAMETER CONDITIONS MIN TYP MAX MIN TYP MAX UNITS

The information provided herein is believed to be reliable; however, BURR-BROWN assumes no responsibility for inaccuracies or omissions. BURR-BROWN assumesno responsibility for the use of this information, and all use of such information shall be entirely at the user’s own risk. Prices and specifications are subject to changewithout notice. No patent rights or licenses to any of the circuits described herein are implied or granted to any third party. BURR-BROWN does not authorize or warrantany BURR-BROWN product for use in life support devices and/or systems.

INPUTOffset Voltage, RTI

Initial TA = +25°C ±10 + 20/G ±50 + 100/G ±25 + 30/G ±125 + 500/G µVvs Temperature TA = TMIN to TMAX ±0.1 + 0.5/G ±0.25 + 5/G ±0.25 + 5/G ±1 + 10/G µV/°Cvs Power Supply VS = ±2.25V to ±18V 0.5 + 2/G 3 + 10/G µV/V

Long-Term Stability ±0.2 + 0.5/G µV/moImpedance, Differential 1010 || 6 Ω || pF

Common-Mode 1010 || 6 Ω || pFInput Common-Mode Range ±11 ±13.5 VSafe Input Voltage ±40 VCommon-Mode Rejection VCM = ±10V, ∆RS = 1kΩ

G = 1 80 96 75 90 dBG = 10 96 115 90 106 dB

G = 100 110 120 106 110 dBG = 1000 115 120 106 110 dB

BIAS CURRENT ±0.5 ±2 ±5 nAvs Temperature ±8 pA/°C

OFFSET CURRENT ±0.5 ±2 ±5 nAvs Temperature ±8 pA/°C

NOISE VOLTAGE, RTI G = 1000, RS = 0Ωf = 10Hz 15 nV/√Hzf = 100Hz 11 nV/√Hzf = 1kHz 11 nV/√HzfB = 0.1Hz to 10Hz 0.4 µVp-p

Noise Currentf=10Hz 0.4 pA/√Hzf=1kHz 0.2 pA/√HzfB = 0.1Hz to 10Hz 18 pAp-p

GAINGain Equation 1 + (50kΩ/RG) V/VRange of Gain 1 10000 V/VGain Error G = 1 ±0.01 ±0.05 %

G = 10 ±0.02 ±0.4 ±0.5 %G = 100 ±0.05 ±0.5 ±0.7 %G = 1000 ±0.5 ±1 ±2 %

Gain vs Temperature G = 1 ±2 ±10 ±10 ppm/°C50kΩ Resistance(1) ±25 ±100 ppm/°C

Nonlinearity G = 1 ±0.0001 ±0.001 ±0.002 % of FSRG = 10 ±0.0005 ±0.002 ±0.004 % of FSR

G = 100 ±0.0005 ±0.002 ±0.004 % of FSRG = 1000 ±0.002 ±0.01 ±0.02 % of FSR

OUTPUTVoltage IO = 5mA, TMIN to TMAX ±13.5 ±13.7 V

VS = ±11.4V, RL = 2kΩ ±10 ±10.5 VVS = ±2.25V, RL = 2kΩ ±1 ±1.5 V

Load Capacitance Stability 1000 pFShort Circuit Current +20/–15 mA

FREQUENCY RESPONSEBandwidth, –3dB G = 1 1 MHz

G = 10 100 kHzG = 100 10 kHzG = 1000 1 kHz

Slew Rate VO = ±10V, G = 10 0.3 0.6 V/µsSettling Time, 0.01% G = 1 18 µs

G = 10 20 µsG = 100 120 µsG = 1000 1100 µs

Overload Recovery 50% Overdrive 20 µs

POWER SUPPLYVoltage Range ±2.25 ±15 ±18 VCurrent VIN = 0V ±2.2 ±3 mA

TEMPERATURE RANGESpecification –40 85 °COperating –40 125 °CθJA 80 °C/W

®

INA1143

RG

V–IN

V+IN

V–

RG

V+

VO

Ref

1

2

3

4

8

7

6

5

P Package 8-Pin DIPTop View

PIN CONFIGURATIONS ELECTROSTATICDISCHARGE SENSITIVITY

This integrated circuit can be damaged by ESD. Burr-Brownrecommends that all integrated circuits be handled with ap-propriate precautions. Failure to observe proper handling andinstallation procedures can cause damage.

ESD damage can range from subtle performance degradationto complete device failure. Precision integrated circuits maybe more susceptible to damage because very small parametricchanges could cause the device not to meet its publishedspecifications.

NC

RG

NC

V–IN

V+IN

NC

V–

NC

NC

RG

NC

V+

Feedback

VO

Ref

NC

1

2

3

4

5

6

7

8

16

15

14

13

12

11

10

9

U Package SOL-16 Surface-MountTop View

PACKAGEDRAWING TEMPERATURE

PRODUCT PACKAGE NUMBER (1) RANGE

INA114AP 8-Pin Plastic DIP 006 –40°C to +85°CINA114BP 8-Pin Plastic DIP 006 –40°C to +85°CINA114AU SOL-16 Surface-Mount 211 –40°C to +85°CINA114BU SOL-16 Surface-Mount 211 –40°C to +85°C

NOTE: (1) For detailed drawing and dimension table, please see end of datasheet, or Appendix C of Burr-Brown IC Data Book.

PACKAGE/ORDERING INFORMATION

Supply Voltage .................................................................................. ±18VInput Voltage Range .......................................................................... ±40VOutput Short-Circuit (to ground) .............................................. ContinuousOperating Temperature ................................................. –40°C to +125°CStorage Temperature ..................................................... –40°C to +125°CJunction Temperature .................................................................... +150°CLead Temperature (soldering, 10s) ............................................... +300°C

NOTE: (1) Stresses above these ratings may cause permanent damage.

ABSOLUTE MAXIMUM RATINGS (1)

®

INA114 4

INPUT-REFERRED NOISE VOLTAGEvs FREQUENCY

Frequency (Hz)

Inpu

t-R

efer

red

Noi

se V

olta

ge (

nV/√

Hz)

1 10 1k100

1k

100

10

110k

G = 1

G = 10

G = 100, 1000

G = 1000BW Limit

NEGATIVE POWER SUPPLY REJECTIONvs FREQUENCY

Frequency (Hz)

Pow

er S

uppl

y R

ejec

tion

(dB

)

10 100 10k 1M1k

140

120

100

80

60

40

20

0100k

G = 1G = 10

G = 100G = 1000

POSITIVE POWER SUPPLY REJECTIONvs FREQUENCY

Frequency (Hz)

Pow

er S

uppl

y R

ejec

tion

(dB

)

10 100 10k 1M1k

140

120

100

80

60

40

20

0100k

G = 1

G = 10

G = 100

G = 1000

INPUT COMMON-MODE VOLTAGE RANGEvs OUTPUT VOLTAGE

Output Voltage (V)

Com

mon

-Mod

e V

olta

ge (

V)

–15 –10 0 5 15–5

15

10

5

0

–5

–10

–1510

Limited by A1

+ Output Swing

A3 – OutputSwing Limit

A3 + OutputSwing Limit

Limited by A2

– Output SwingLimited by A1

– Output Swing

Limited by A2

+ Output Swing

VD/2–

+–

+

VCM

VO

(Any Gain)

VD/2

COMMON-MODE REJECTION vs FREQUENCY

Frequency (Hz)C

omm

on-M

ode

Rej

ectio

n (d

B)

10 100 10k 100k 1M1k

140

120

100

80

60

40

20

0

G = 1k

G = 100

G = 10

G = 1

G = 100, 1k

G = 10

GAIN vs FREQUENCY

Frequency (Hz)

Gai

n (V

/V)

10 100 10k 100k 1M1k

1k

100

10

1

TYPICAL PERFORMANCE CURVESAt TA = +25°C, VS = ±15V, unless otherwise noted.

®

INA1145

MAXIMUM OUTPUT SWING vs FREQUENCY

Pea

k-to

-Pea

k A

mpl

itude

(V

)

10

32

28

24

20

16

12

8

4

0100 10k 1M

Frequency (Hz)

100k1k

G = 100

G = 1, 10

G = 1000

INPUT BIAS CURRENTvs COMMON-MODE INPUT VOLTAGE

Inpu

t Bia

s C

urre

nt (

mA

)

–45

3

2

1

0

–1

–2

–3–30 –15 0 15 30 45

|Ib1| + |Ib2|

Common-Mode Voltage (V)

NormalOperation

Over-VoltageProtection

Over-VoltageProtection

One Input

Both Inputs

Both Inputs

One Input

INPUT BIAS CURRENTvs DIFFERENTIAL INPUT VOLTAGE

Differential Overload Voltage (V)

Inpu

t Bia

s C

urre

nt (

mA

)

–45

3

2

1

0

–1

–2

–3–30 –15 0 15 30 45

G = 1

G = 10

G = 1000G = 100

INPUT BIAS AND INPUT OFFSET CURRENTvs TEMPERATURE

Temperature (°C)

Inpu

t Bia

s an

d In

put O

ffset

Cur

rent

(nA

)

–40

2

1

0

–1

–2–15 10 35 60 85

±IB

IOS

OFFSET VOLTAGE WARM-UP vs TIME

Time from Power Supply Turn-on (s)

Offs

et V

olta

ge C

hang

e (µ

V)

0

6

4

2

0

–2

–4

–615 30 45 60 75 90 105 120

G ≥ 100

SETTLING TIME vs GAIN

Gain (V/V)

Set

tling

Tim

e (µ

s)

1 100 100010

1200

1000

800

600

400

200

0

0.01%

0.1%

TYPICAL PERFORMANCE CURVES (CONT)At TA = +25°C, VS = ±15V, unless otherwise noted.

®

INA114 6

NEGATIVE SIGNAL SWING vs TEMPERATUE (RL = 2kΩ)

Out

put V

olta

ge (

V)

–75

–16

–14

–12

–10

–8

–6

–4

–2

0125

Temperature (°C)

–50 –25 0 25 50 75 100

VS = ±15V

VS = ±11.4V

VS = ±2.25V

POSITIVE SIGNAL SWING vs TEMPERATUE (RL = 2kΩ)

Out

put V

olta

ge (

V)

–75

16

14

12

10

8

6

4

2

0125

Temperature (°C)

–50 –25 0 25 50 75 100

VS = ±15V

VS = ±11.4V

VS = ±2.25V

QUIESCENT CURRENT AND POWER DISSIPATIONvs POWER SUPPLY VOLTAGE

Qui

esce

nt C

urre

nt (

mA

)

0

2.6

2.5

2.4

2.3

2.2

2.1

2.0

Power Supply Voltage (V)

±3 ±6 ±9 ±12 ±15 ±18

120

100

80

60

40

20

0

Pow

er D

issi

patio

n (m

W)

Power Dissipation

Quiescent Current

QUIESCENT CURRENT vs TEMPERATURE

Qui

esce

nt C

urre

nt (

mA

)

–75

2.8

2.6

2.4

2.2

2.0

1.8125

Temperature (°C)

–50 –25 0 25 50 75 100

OUTPUT CURRENT LIMIT vs TEMPERATURE

Sho

rt C

ircui

t Cur

rent

(m

A)

–40

30

25

20

15

1085

Temperature (°C)

–15 10 35 60

+|ICL|

–|ICL|

SLEW RATE vs TEMPERATURE

Sle

w R

ate

(V/µ

s)

–75

1.0

0.8

0.6

0.4

0.2

0125

Temperature (°C)

–50 –25 0 25 50 75 100

TYPICAL PERFORMANCE CURVES (CONT)At TA = +25°C, VS = ±15V, unless otherwise noted.

®

INA1147

TYPICAL PERFORMANCE CURVES (CONT)At TA = +25°C, VS = ±15V, unless otherwise noted.

LARGE SIGNAL RESPONSE, G = 1 SMALL SIGNAL RESPONSE, G = 1

LARGE SIGNAL RESPONSE, G = 1000 SMALL SIGNAL RESPONSE, G = 1000

+10V

0

–10V

+10V

0

–10V

+200mV

0

–200mV

+100mV

0

–200mV

INPUT-REFERRED NOISE, 0.1 to 10Hz

0.1µV/div

1 s/div

®

INA114 8

APPLICATION INFORMATIONFigure 1 shows the basic connections required for operationof the INA114. Applications with noisy or high impedancepower supplies may require decoupling capacitors close tothe device pins as shown.

The output is referred to the output reference (Ref) terminalwhich is normally grounded. This must be a low-impedanceconnection to assure good common-mode rejection. A resis-tance of 5Ω in series with the Ref pin will cause a typicaldevice to degrade to approximately 80dB CMR (G = 1).

SETTING THE GAIN

Gain of the INA114 is set by connecting a single externalresistor, RG:

Commonly used gains and resistor values are shown inFigure 1.

The 50kΩ term in equation (1) comes from the sum of thetwo internal feedback resistors. These are on-chip metal filmresistors which are laser trimmed to accurate absolute val-

FIGURE 1. Basic Connections.

G = 1 + 50 kΩR

G

(1)

DESIRED RG NEAREST 1% RG

GAIN (Ω) (Ω)

1 No Connection No Connection2 50.00k 49.9k5 12.50k 12.4k10 5.556k 5.62k20 2.632k 2.61k50 1.02k 1.02k100 505.1 511200 251.3 249500 100.2 1001000 50.05 49.92000 25.01 24.95000 10.00 1010000 5.001 4.99

ues. The accuracy and temperature coefficient of theseresistors are included in the gain accuracy and drift specifi-cations of the INA114.

The stability and temperature drift of the external gainsetting resistor, RG, also affects gain. RG’s contribution togain accuracy and drift can be directly inferred from the gainequation (1). Low resistor values required for high gain canmake wiring resistance important. Sockets add to the wiringresistance which will contribute additional gain error (possi-bly an unstable gain error) in gains of approximately 100 orgreater.

NOISE PERFORMANCE

The INA114 provides very low noise in most applications.For differential source impedances less than 1kΩ, the INA103may provide lower noise. For source impedances greaterthan 50kΩ, the INA111 FET-input instrumentation ampli-fier may provide lower noise.

Low frequency noise of the INA114 is approximately0.4µVp-p measured from 0.1 to 10Hz. This is approximatelyone-tenth the noise of “low noise” chopper-stabilized ampli-fiers.

A1

A2

A36

25kΩ25kΩ

25kΩ25kΩ

7

4

3

8

1

2VIN

VIN

RG

V+

V–

INA114

G = 1 + 50kΩRG

–

+5

Over-VoltageProtection

25kΩ

25kΩ

Over-VoltageProtection

Load

VO = G • (VIN – VIN)+ –

0.1µF

0.1µF

Pin numbers arefor DIP packages.

+

–

VO

INA114RG

Also drawn in simplified form:

VO

Ref

VIN–

VIN+

®

INA1149

OFFSET TRIMMING

The INA114 is laser trimmed for very low offset voltage anddrift. Most applications require no external offset adjust-ment. Figure 2 shows an optional circuit for trimming theoutput offset voltage. The voltage applied to Ref terminal issummed at the output. Low impedance must be maintainedat this node to assure good common-mode rejection. This isachieved by buffering trim voltage with an op amp asshown.

FIGURE 2. Optional Trimming of Output Offset Voltage.

INPUT BIAS CURRENT RETURN PATH

The input impedance of the INA114 is extremely high—approximately 1010Ω. However, a path must be provided forthe input bias current of both inputs. This input bias currentis typically less than ±1nA (it can be either polarity due tocancellation circuitry). High input impedance means thatthis input bias current changes very little with varying inputvoltage.

Input circuitry must provide a path for this input bias currentif the INA114 is to operate properly. Figure 3 shows variousprovisions for an input bias current path. Without a biascurrent return path, the inputs will float to a potential whichexceeds the common-mode range of the INA114 and theinput amplifiers will saturate. If the differential source resis-tance is low, bias current return path can be connected to oneinput (see thermocouple example in Figure 3). With highersource impedance, using two resistors provides a balancedinput with possible advantages of lower input offset voltagedue to bias current and better common-mode rejection.

INPUT COMMON-MODE RANGE

The linear common-mode range of the input op amps of theINA114 is approximately ±13.75V (or 1.25V from thepower supplies). As the output voltage increases, however,the linear input range will be limited by the output voltageswing of the input amplifiers, A1 and A2. The common-mode range is related to the output voltage of the completeamplifier—see performance curve “Input Common-ModeRange vs Output Voltage.”

A combination of common-mode and differential inputsignals can cause the output of A1 or A2 to saturate. Figure4 shows the output voltage swing of A1 and A2 expressed interms of a common-mode and differential input voltages.Output swing capability of these internal amplifiers is thesame as the output amplifier, A3. For applications whereinput common-mode range must be maximized, limit theoutput voltage swing by connecting the INA114 in a lowergain (see performance curve “Input Common-Mode VoltageRange vs Output Voltage”). If necessary, add gain after theINA114 to increase the voltage swing.

Input-overload often produces an output voltage that appearsnormal. For example, an input voltage of +20V on one inputand +40V on the other input will obviously exceed the linearcommon-mode range of both input amplifiers. Since bothinput amplifiers are saturated to nearly the same outputvoltage limit, the difference voltage measured by the outputamplifier will be near zero. The output of the INA114 willbe near 0V even though both inputs are overloaded.

INPUT PROTECTION

The inputs of the INA114 are individually protected forvoltages up to ±40V. For example, a condition of –40V onone input and +40V on the other input will not causedamage. Internal circuitry on each input provides low seriesimpedance under normal signal conditions. To provideequivalent protection, series input resistors would contributeexcessive noise. If the input is overloaded, the protectioncircuitry limits the input current to a safe value (approxi-mately 1.5mA). The typical performance curve “Input BiasCurrent vs Common-Mode Input Voltage” shows this input

FIGURE 3. Providing an Input Common-Mode Current Path.

INA114

VIN

VIN

RG

–

+

10kΩ

VO

OPA177

Ref

±10mVAdjustment Range

100Ω

100Ω

100µA1/2 REF200

100µA1/2 REF200

V+

V–

INA114

47kΩ47kΩ

INA114

10kΩ

Microphone,Hydrophone

etc.

Thermocouple

INA114

Center-tap providesbias current return.

®

INA114 10

INA114VIN

–

VIN+

OPA602

511Ω22.1kΩ22.1kΩ

Ref

VO

For G = 100RG = 511Ω // 2(22.1kΩ)effective RG = 505Ω

100ΩShield is driven at thecommon-mode potential.

current limit behavior. The inputs are protected even if nopower supply voltage is present.

OUTPUT VOLTAGE SENSE (SOL-16 package only)

The surface-mount version of the INA114 has a separateoutput sense feedback connection (pin 12). Pin 12 must beconnected to the output terminal (pin 11) for proper opera-tion. (This connection is made internally on the DIP versionof the INA114.)

The output sense connection can be used to sense the outputvoltage directly at the load for best accuracy. Figure 5 showshow to drive a load through series interconnection resis-tance. Remotely located feedback paths may cause instabil-ity. This can be generally be eliminated with a highfrequency feedback path through C1. Heavy loads or longlines can be driven by connecting a buffer inside the feed-back path (Figure 6).

FIGURE 4. Voltage Swing of A1 and A2.

FIGURE 5. Remote Load and Ground Sensing. FIGURE 6. Buffered Output for Heavy Loads.

FIGURE 7. Shield Driver Circuit.

A1

A2

A3

25kΩ25kΩ

25kΩ25kΩ

RG

V+

V–

INA114

VO = G • VD

G = 1 + 50kΩRG25kΩ

25kΩ

VCM – G • VD2

VD 2

VD 2

VCM

VCM + G • VD2

Over-VoltageProtection

Over-VoltageProtection

INA114RG

VIN–

VIN+ Load

Equal resistance here preservesgood common-mode rejection.

C11000pF

OutputSense

Ref

Surface-mount packageversion only.

INA114RG

VIN–

VIN+

IL: ±100mA

OutputSense

Ref

Surface-mount packageversion only.

OPA633

RL

180Ω

®

INA11411

FIGURE 8. RTD Temperature Measurement Circuit.

SEEBECKISA COEFFICIENT R2 R4TYPE MATERIAL ( µV/°C) (R3 = 100Ω) (R5 + R6 = 100Ω)

E Chromel 58.5 3.48kΩ 56.2kΩConstantan

J Iron 50.2 4.12kΩ 64.9kΩConstantan

K Chromel 39.4 5.23kΩ 80.6kΩAlumel

T Copper 38.0 5.49kΩ 84.5kΩConstantan

NOTES: (1) –2.1mV/°C at 200µA. (2) R7 provides down-scale burn-out indication.

FIGURE 9. Thermocouple Amplifier With Cold Junction Compensation.

INA114RG

Ref

VO

100µA

V+ V+

RZ

RTD

1

2

3

Equal line resistance here createsa small common-mode voltagewhich is rejected by INA114.

Resistance in this line causesa small common-mode voltagewhich is rejected by INA114.

VO = 0V at RRTD = RZ

REF200

REF102

R80.6k

4

Ω

R1M

7

ΩR5.23k

2

Ω

R27k

1

Ω

R100

3

Ω

1N4148

Cu

Cu

R50

5

Ω

R100Zero Adj

6

Ω

V+

K

610.0V

4

(1)(2)

2

INA114VO

Ref

®

INA114 12

INA114RG

VO

C10.1µF

OPA602

Ref R11MΩ

f–3dB = 12πR1C1

= 1.59Hz

VIN

+

–

FIGURE 10. ECG Amplifier With Right-Leg Drive.

FIGURE 12. AC-Coupled Instrumentation Amplifier.

INA114RG/2

VOLA

RL

RA

10kΩ

RefG = 102.8kΩ

2.8kΩ

1/2OPA2604

390kΩ

390kΩ

1/2OPA2604

FIGURE 11. Bridge Transducer Amplifier.

INA114RG

100Ω

VO

+10V

BridgeG = 500

Ref

INA114RG

IB

R1

VIN

–

+

A1 IO

Load

IO = • GVINR

Ref

FIGURE 13. Differential Voltage-to-Current Converter.

A1 IB Error

OPA177 ±1.5nAOPA602 1pAOPA128 75fA

PACKAGE OPTION ADDENDUM

www.ti.com 11-Apr-2013

Addendum-Page 1

PACKAGING INFORMATION

Orderable Device Status(1)

Package Type PackageDrawing

Pins PackageQty

Eco Plan(2)

Lead/Ball Finish MSL Peak Temp(3)

Op Temp (°C) Top-Side Markings(4)

Samples

INA114AP ACTIVE PDIP P 8 50 Green (RoHS& no Sb/Br)

CU NIPDAU N / A for Pkg Type INA114AP

INA114APG4 ACTIVE PDIP P 8 50 Green (RoHS& no Sb/Br)

CU NIPDAU N / A for Pkg Type INA114AP

INA114AU ACTIVE SOIC DW 16 40 Green (RoHS& no Sb/Br)

CU NIPDAU Level-3-260C-168 HR -40 to 85 INA114AU

INA114AU/1K ACTIVE SOIC DW 16 1000 Green (RoHS& no Sb/Br)

CU NIPDAU Level-3-260C-168 HR -40 to 85 INA114AU

INA114AU/1KE4 ACTIVE SOIC DW 16 1000 Green (RoHS& no Sb/Br)

CU NIPDAU Level-3-260C-168 HR -40 to 85 INA114AU

INA114AUE4 ACTIVE SOIC DW 16 40 Green (RoHS& no Sb/Br)

CU NIPDAU Level-3-260C-168 HR -40 to 85 INA114AU

INA114AUG4 ACTIVE SOIC DW 16 40 Green (RoHS& no Sb/Br)

CU NIPDAU Level-3-260C-168 HR -40 to 85 INA114AU

INA114BP ACTIVE PDIP P 8 50 Green (RoHS& no Sb/Br)

CU NIPDAU N / A for Pkg Type INA114BP

INA114BPG4 ACTIVE PDIP P 8 50 Green (RoHS& no Sb/Br)

CU NIPDAU N / A for Pkg Type INA114BP

INA114BU ACTIVE SOIC DW 16 40 Green (RoHS& no Sb/Br)

CU NIPDAU Level-3-260C-168 HR INA114BU

INA114BU/1K ACTIVE SOIC DW 16 1000 Green (RoHS& no Sb/Br)

CU NIPDAU Level-3-260C-168 HR INA114BU

INA114BU/1KE4 ACTIVE SOIC DW 16 1000 Green (RoHS& no Sb/Br)

CU NIPDAU Level-3-260C-168 HR INA114BU

INA114BUE4 ACTIVE SOIC DW 16 40 Green (RoHS& no Sb/Br)

CU NIPDAU Level-3-260C-168 HR INA114BU

(1) The marketing status values are defined as follows:ACTIVE: Product device recommended for new designs.LIFEBUY: TI has announced that the device will be discontinued, and a lifetime-buy period is in effect.NRND: Not recommended for new designs. Device is in production to support existing customers, but TI does not recommend using this part in a new design.PREVIEW: Device has been announced but is not in production. Samples may or may not be available.OBSOLETE: TI has discontinued the production of the device.

PACKAGE OPTION ADDENDUM

www.ti.com 11-Apr-2013

Addendum-Page 2

(2) Eco Plan - The planned eco-friendly classification: Pb-Free (RoHS), Pb-Free (RoHS Exempt), or Green (RoHS & no Sb/Br) - please check http://www.ti.com/productcontent for the latest availabilityinformation and additional product content details.TBD: The Pb-Free/Green conversion plan has not been defined.Pb-Free (RoHS): TI's terms "Lead-Free" or "Pb-Free" mean semiconductor products that are compatible with the current RoHS requirements for all 6 substances, including the requirement thatlead not exceed 0.1% by weight in homogeneous materials. Where designed to be soldered at high temperatures, TI Pb-Free products are suitable for use in specified lead-free processes.Pb-Free (RoHS Exempt): This component has a RoHS exemption for either 1) lead-based flip-chip solder bumps used between the die and package, or 2) lead-based die adhesive used betweenthe die and leadframe. The component is otherwise considered Pb-Free (RoHS compatible) as defined above.Green (RoHS & no Sb/Br): TI defines "Green" to mean Pb-Free (RoHS compatible), and free of Bromine (Br) and Antimony (Sb) based flame retardants (Br or Sb do not exceed 0.1% by weightin homogeneous material)

(3) MSL, Peak Temp. -- The Moisture Sensitivity Level rating according to the JEDEC industry standard classifications, and peak solder temperature.

(4) Multiple Top-Side Markings will be inside parentheses. Only one Top-Side Marking contained in parentheses and separated by a "~" will appear on a device. If a line is indented then it is acontinuation of the previous line and the two combined represent the entire Top-Side Marking for that device.

Important Information and Disclaimer:The information provided on this page represents TI's knowledge and belief as of the date that it is provided. TI bases its knowledge and belief on informationprovided by third parties, and makes no representation or warranty as to the accuracy of such information. Efforts are underway to better integrate information from third parties. TI has taken andcontinues to take reasonable steps to provide representative and accurate information but may not have conducted destructive testing or chemical analysis on incoming materials and chemicals.TI and TI suppliers consider certain information to be proprietary, and thus CAS numbers and other limited information may not be available for release.

In no event shall TI's liability arising out of such information exceed the total purchase price of the TI part(s) at issue in this document sold by TI to Customer on an annual basis.

TAPE AND REEL INFORMATION

*All dimensions are nominal

Device PackageType

PackageDrawing

Pins SPQ ReelDiameter

(mm)

ReelWidth

W1 (mm)

A0(mm)

B0(mm)

K0(mm)

P1(mm)

W(mm)

Pin1Quadrant

INA114AU/1K SOIC DW 16 1000 330.0 16.4 10.75 10.7 2.7 12.0 16.0 Q1

INA114BU/1K SOIC DW 16 1000 330.0 16.4 10.75 10.7 2.7 12.0 16.0 Q1

PACKAGE MATERIALS INFORMATION

www.ti.com 26-Jan-2013

Pack Materials-Page 1

*All dimensions are nominal

Device Package Type Package Drawing Pins SPQ Length (mm) Width (mm) Height (mm)

INA114AU/1K SOIC DW 16 1000 367.0 367.0 38.0

INA114BU/1K SOIC DW 16 1000 367.0 367.0 38.0

PACKAGE MATERIALS INFORMATION

www.ti.com 26-Jan-2013

Pack Materials-Page 2

IMPORTANT NOTICE

Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, enhancements, improvements and otherchanges to its semiconductor products and services per JESD46, latest issue, and to discontinue any product or service per JESD48, latestissue. Buyers should obtain the latest relevant information before placing orders and should verify that such information is current andcomplete. All semiconductor products (also referred to herein as “components”) are sold subject to TI’s terms and conditions of salesupplied at the time of order acknowledgment.

TI warrants performance of its components to the specifications applicable at the time of sale, in accordance with the warranty in TI’s termsand conditions of sale of semiconductor products. Testing and other quality control techniques are used to the extent TI deems necessaryto support this warranty. Except where mandated by applicable law, testing of all parameters of each component is not necessarilyperformed.

TI assumes no liability for applications assistance or the design of Buyers’ products. Buyers are responsible for their products andapplications using TI components. To minimize the risks associated with Buyers’ products and applications, Buyers should provideadequate design and operating safeguards.

TI does not warrant or represent that any license, either express or implied, is granted under any patent right, copyright, mask work right, orother intellectual property right relating to any combination, machine, or process in which TI components or services are used. Informationpublished by TI regarding third-party products or services does not constitute a license to use such products or services or a warranty orendorsement thereof. Use of such information may require a license from a third party under the patents or other intellectual property of thethird party, or a license from TI under the patents or other intellectual property of TI.

Reproduction of significant portions of TI information in TI data books or data sheets is permissible only if reproduction is without alterationand is accompanied by all associated warranties, conditions, limitations, and notices. TI is not responsible or liable for such altereddocumentation. Information of third parties may be subject to additional restrictions.

Resale of TI components or services with statements different from or beyond the parameters stated by TI for that component or servicevoids all express and any implied warranties for the associated TI component or service and is an unfair and deceptive business practice.TI is not responsible or liable for any such statements.

Buyer acknowledges and agrees that it is solely responsible for compliance with all legal, regulatory and safety-related requirementsconcerning its products, and any use of TI components in its applications, notwithstanding any applications-related information or supportthat may be provided by TI. Buyer represents and agrees that it has all the necessary expertise to create and implement safeguards whichanticipate dangerous consequences of failures, monitor failures and their consequences, lessen the likelihood of failures that might causeharm and take appropriate remedial actions. Buyer will fully indemnify TI and its representatives against any damages arising out of the useof any TI components in safety-critical applications.

In some cases, TI components may be promoted specifically to facilitate safety-related applications. With such components, TI’s goal is tohelp enable customers to design and create their own end-product solutions that meet applicable functional safety standards andrequirements. Nonetheless, such components are subject to these terms.

No TI components are authorized for use in FDA Class III (or similar life-critical medical equipment) unless authorized officers of the partieshave executed a special agreement specifically governing such use.

Only those TI components which TI has specifically designated as military grade or “enhanced plastic” are designed and intended for use inmilitary/aerospace applications or environments. Buyer acknowledges and agrees that any military or aerospace use of TI componentswhich have not been so designated is solely at the Buyer's risk, and that Buyer is solely responsible for compliance with all legal andregulatory requirements in connection with such use.

TI has specifically designated certain components as meeting ISO/TS16949 requirements, mainly for automotive use. In any case of use ofnon-designated products, TI will not be responsible for any failure to meet ISO/TS16949.

Products Applications

Audio www.ti.com/audio Automotive and Transportation www.ti.com/automotive

Amplifiers amplifier.ti.com Communications and Telecom www.ti.com/communications

Data Converters dataconverter.ti.com Computers and Peripherals www.ti.com/computers

DLP® Products www.dlp.com Consumer Electronics www.ti.com/consumer-apps

DSP dsp.ti.com Energy and Lighting www.ti.com/energy

Clocks and Timers www.ti.com/clocks Industrial www.ti.com/industrial

Interface interface.ti.com Medical www.ti.com/medical

Logic logic.ti.com Security www.ti.com/security

Power Mgmt power.ti.com Space, Avionics and Defense www.ti.com/space-avionics-defense

Microcontrollers microcontroller.ti.com Video and Imaging www.ti.com/video

RFID www.ti-rfid.com

OMAP Applications Processors www.ti.com/omap TI E2E Community e2e.ti.com

Wireless Connectivity www.ti.com/wirelessconnectivity

Mailing Address: Texas Instruments, Post Office Box 655303, Dallas, Texas 75265Copyright © 2013, Texas Instruments Incorporated

B.4. Hoja de datos TLC277

757

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

1POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

Trimmed Offset Voltage:TLC277 . . . 500 µV Max at 25°C,VDD = 5 V

Input Offset Voltage Drift . . . Typically0.1 µV/Month, Including the First 30 Days

Wide Range of Supply Voltages OverSpecified Temperature Range:

0°C to 70°C . . . 3 V to 16 V–40°C to 85°C . . . 4 V to 16 V–55°C to 125°C . . . 4 V to 16 V

Single-Supply Operation

Common-Mode Input Voltage RangeExtends Below the Negative Rail (C-Suffix,I-Suffix types)

Low Noise . . . Typically 25 nV/√Hz at f = 1 kHz

Output Voltage Range Includes NegativeRail

High Input impedance . . . 1012 Ω Typ

ESD-Protection Circuitry

Small-Outline Package Option AlsoAvailable in Tape and Reel

Designed-In Latch-Up Immunity

description

The TLC272 and TLC277 precision dualoperational amplifiers combine a wide range ofinput offset voltage grades with low offset voltagedrift, high input impedance, low noise, and speedsapproaching those of general-purpose BiFETdevices.

These devices use Texas Instruments silicon-gate LinCMOS technology, which providesoffset voltage stability far exceeding the stabilityavailable with conventional metal-gate pro-cesses.

The extremely high input impedance, low biascurrents, and high slew rates make these cost-effective devices ideal for applications previouslyreserved for BiFET and NFET products. Fouroffset voltage grades are available (C-suffix andI-suffix types), ranging from the low-cost TLC272(10 mV) to the high-precision TLC277 (500 µV).These advantages, in combination with goodcommon-mode rejection and supply voltagerejection, make these devices a good choice fornew state-of-the-art designs as well as forupgrading existing designs.

Copyright 2002, Texas Instruments IncorporatedPRODUCTION DATA information is current as of publication date.Products conform to specifications per the terms of Texas Instrumentsstandard warranty. Production processing does not necessarily includetesting of all parameters.

1

2

3

4

8

7

6

5

1OUT1IN–1IN+GND

VDD2OUT2IN–2IN+

D, JG, P, OR PW PACKAGE

3 2 1 20 19

9 10 11 12 13

4

5

6

7

8

18

17

16

15

14

NC2OUTNC2IN–NC

NC1IN–

NC1IN+

NC

FK PACKAGE(TOP VIEW)

NC

1OU

TN

C2I

N +

NC

NC

NC

GN

DN

C

NC – No internal connection

P PackageTA = 25°C25

20

15

10

5

4000–4000

800

30

VIO – Input Offset Voltage – µV

Per

cen

tag

e o

f U

nit

s –

%

–800

DISTRIBUTION OF TLC277INPUT OFFSET VOLTAGE

VD

D

473 Units Tested From 2 Wafer LotsVDD = 5 V

(TOP VIEW)

LinCMOS is a trademark of Texas Instruments.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

2 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

description (continued)

AVAILABLE OPTIONS

PACKAGED DEVICESCHIP

TAVIOmaxAT 25°C

SMALLOUTLINE

(D)

CHIPCARRIER

(FK)

CERAMICDIP(JG)

PLASTICDIP(P)

TSSOP(PW)

CHIPFORM

(Y)

500 µV TLC277CD — — TLC277CP — —

0°C to 70°c

500 µV2 mV

TLC277CDTLC272BCD

——

——

TLC277CPTLC272BCP

——

——

0°C to 70°c2 mV5 mV

TLC272BCDTLC272ACD — —

TLC272BCPTLC272ACP — —5 mV

10mVTLC272ACDTLC272CD — —

TLC272ACPTLC272CP TLC272CPW TLC272Y

500 µV TLC277ID — — TLC277IP — —

40°C to 85°C

500 µV2 mV

TLC277IDTLC272BID

——

——

TLC277IPTLC272BIP

——

——

–40°C to 85°C2 mV5 mV

TLC272BIDTLC272AID — —

TLC272BIPTLC272AIP — —5 mV

10 mVTLC272AIDTLC272ID — —

TLC272AIPTLC272IP — —

The D package is available taped and reeled. Add R suffix to the device type (e.g., TLC277CDR).

In general, many features associated with bipolar technology are available on LinCMOS operational amplifierswithout the power penalties of bipolar technology. General applications such as transducer interfacing, analogcalculations, amplifier blocks, active filters, and signal buffering are easily designed with the TLC272 andTLC277. The devices also exhibit low voltage single-supply operation, making them ideally suited for remoteand inaccessible battery-powered applications. The common-mode input voltage range includes the negativerail.

A wide range of packaging options is available, including small-outline and chip carrier versions for high-densitysystem applications.

The device inputs and outputs are designed to withstand –100-mA surge currents without sustaining latch-up.

The TLC272 and TLC277 incorporate internal ESD-protection circuits that prevent functional failures at voltagesup to 2000 V as tested under MIL-STD-883C, Method 3015.2; however, care should be exercised in handlingthese devices as exposure to ESD may result in the degradation of the device parametric performance.

The C-suffix devices are characterized for operation from 0°C to 70°C. The I-suffix devices are characterizedfor operation from –40°C to 85°C. The M-suffix devices are characterized for operation over the full militarytemperature range of –55°C to 125°C.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

3POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

equivalent schematic (each amplifier)

P5 P6

OUT

N7N6

R7N4

C1R5

N3

GND

N2

D2R4D1R3

N1

IN+

IN–

P1

R1

P2

R2 N5

R6

P3 P4

VDD

TLC272Y chip information

This chip, when properly assembled, displays characteristics similar to the TLC272C. Thermal compression orultrasonic bonding may be used on the doped-aluminum bonding pads. Chips may be mounted with conductiveepoxy or a gold-silicon preform.

BONDING PAD ASSIGNMENTS

CHIP THICKNESS: 15 TYPICAL

BONDING PADS: 4 × 4 MINIMUM

TJmax = 150°C

TOLERANCES ARE ±10%.

ALL DIMENSIONS ARE IN MILS.

PIN (4) IS INTERNALLY CONNECTEDTO BACKSIDE OF CHIP.

+

–1OUT

1IN+

1IN–

VDD(8)

(6)

(3)

(2)

(5)

(1)

–

+(7) 2IN+

2IN–2OUT

(4)

GND

60

73

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

4 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

absolute maximum ratings over operating free-air temperature range (unless otherwise noted)†

Supply voltage, VDD (see Note 1) 18 V. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Differential input voltage, VID (see Note 2) ±VDD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Input voltage range, VI (any input) –0.3 V to VDD. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Input current, II ±5 mA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . output current, IO (each output) ±30 mA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Total current into VDD 45 mA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Total current out of GND 45 mA. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Duration of short-circuit current at (or below) 25°C (see Note 3) unlimited. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Continuous total dissipation See Dissipation Rating Table. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Operating free-air temperature, TA: C suffix 0°C to 70°C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

I suffix –40°C to 85°C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . M suffix –55°C to 125°C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Storage temperature range –65°C to 150°C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Case temperature for 60 seconds: FK package 260°C. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Lead temperature 1,6 mm (1/16 inch) from case for 10 seconds: D, P, or PW package 260°C. . . . . . . . . . . . Lead temperature 1,6 mm (1/16 inch) from case for 60 seconds: JG package 300°C. . . . . . . . . . . . . . . . . . . .

† Stresses beyond those listed under “absolute maximum ratings” may cause permanent damage to the device. These are stress ratings only, andfunctional operation of the device at these or any other conditions beyond those indicated under “recommended operating conditions” is notimplied. Exposure to absolute-maximum-rated conditions for extended periods may affect device reliability.

NOTES: 1. All voltage values, except differential voltages, are with respect to network ground.2. Differential voltages are at IN+ with respect to IN–.3. The output may be shorted to either supply. Temperature and/or supply voltages must be limited to ensure that the maximum

dissipation rating is not exceeded (see application section).

DISSIPATION RATING TABLE

PACKAGETA ≤ 25°C

POWER RATINGDERATING FACTORABOVE TA = 25°C

TA = 70°CPOWER RATING

TA = 85°CPOWER RATING

TA = 125°CPOWER RATING

D 725 mW 5.8 mW/°C 464 mW 377 mW N/A

FK 1375 mW 11 mW/°C 880 mW 715 mW 275 mW

JG 1050 mW 8.4 mW/°C 672 mW 546 mW 210 mW

P 1000 mW 8.0 mW/°C 640 mW 520 mW N/A

PW 525 mW 4.2 mW/°C 336 mW N/A N/A

recommended operating conditions

C SUFFIX I SUFFIX M SUFFIXUNIT

MIN MAX MIN MAX MIN MAXUNIT

Supply voltage, VDD 3 16 4 16 4 16 V

Common mode input voltage VVDD = 5 V –0.2 3.5 –0.2 3.5 0 3.5

VCommon-mode input voltage, VIC VDD = 10 V –0.2 8.5 –0.2 8.5 0 8.5V

Operating free-air temperature, TA 0 70 –40 85 –55 125 °C

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

5POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

electrical characteristics at specified free-air temperature, VDD = 5 V (unless otherwise noted)

PARAMETER TEST CONDITIONS TA†TLC272C, TLC272AC,TLC272BC, TLC277C UNITPARAMETER TEST CONDITIONS TA†

MIN TYP MAXUNIT

TLC272CVO = 1.4 V, VIC = 0, 25°C 1.1 10

TLC272CVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 12

mV

TLC272ACVO = 1.4 V, VIC = 0, 25°C 0.9 5

mV

V Input offset voltage

TLC272ACVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 6.5

VIO Input offset voltage

TLC272BCVO = 1.4 V, VIC = 0, 25°C 230 2000

TLC272BCVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 3000

V

TLC277CVO = 1.4 V, VIC = 0, 25°C 200 500

µV

TLC277CVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 1500

αVIO Temperature coefficient of input offset voltage25°C to

70°C 1.8 µV/°C

I Input offset current (see Note 4)25°C 0.1 60

pAIIO Input offset current (see Note 4)

V 2 5 V V 2 5 V70°C 7 300

pA

I Input bias current (see Note 4)

VO = 2.5 V, VIC = 2.5 V25°C 0.6 60

pAIIB Input bias current (see Note 4)70°C 40 600

pA

VCommon-mode input voltage range

25°C–0.2

to4

–0.3to

4.2V

VICRCommon mode in ut voltage range(see Note 5)

Full range–0.2

to3.5

V

25°C 3.2 3.8

VOH High-level output voltage VID = 100 mV, RL = 10 kΩ 0°C 3 3.8 VVOH High level out ut voltage VID 100 mV, RL 10 kΩ70°C 3 3.8

V

25°C 0 50

VOL Low-level output voltage VID = –100 mV, IOL = 0 0°C 0 50 mVVOL Low level out ut voltage VID 100 mV, IOL 0

70°C 0 50

mV

25°C 5 23

AVD Large-signal differential voltage amplification VO = 0.25 V to 2 V, RL = 10 kΩ 0°C 4 27 V/mVAVD Large signal differential voltage am lification VO 0.25 V to 2 V, RL 10 kΩ70°C 4 20

V/mV

25°C 65 80

CMRR Common-mode rejection ratio VIC = VICRmin 0°C 60 84 dBCMRR Common mode rejection ratio VIC VICRmin

70°C 60 85

dB

S l lt j ti ti25°C 65 95

kSVRSupply-voltage rejection ratio(∆VDD/∆VIO) VDD = 5 V to 10 V, VO = 1.4 V 0°C 60 94 dBkSVR (∆VDD/∆VIO) VDD 5 V to 10 V, VO 1.4 V

70°C 60 96

dB

V 2 5 V V 2 5 V25°C 1.4 3.2

IDD Supply current (two amplifiers)VO = 2.5 V,No load

VIC = 2.5 V,0°C 1.6 3.6 mADD y ( )

No load70°C 1.2 2.6

† Full range is 0°C to 70°C.NOTES: 4. The typical values of input bias current and input offset current below 5 pA were determined mathematically.

5. This range also applies to each input individually.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

6 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

electrical characteristics at specified free-air temperature, VDD = 10 V (unless otherwise noted)

PARAMETER TEST CONDITIONS TA†TLC272C, TLC272AC,TLC272BC, TLC277C UNITPARAMETER TEST CONDITIONS TA†

MIN TYP MAXUNIT

TLC272CVO = 1.4 V, VIC = 0, 25°C 1.1 10

TLC272CVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 12

mV

TLC272ACVO = 1.4 V, VIC = 0, 25°C 0.9 5

mV

V Input offset voltage

TLC272ACVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 6.5

VIO Input offset voltage

TLC272BCVO = 1.4 V, VIC = 0, 25°C 290 2000

TLC272BCVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 3000

V

TLC277CVO = 1.4 V, VIC = 0, 25°C 250 800

µV

TLC277CVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 1900

αVIO Temperature coefficient of input offset voltage25°C to

70°C 2 µV/°C

I Input offset current (see Note 4)25°C 0.1 60

pAIIO Input offset current (see Note 4)

V 5 V V 5 V70°C 7 300

pA

I Input bias current (see Note 4)

VO = 5 V, VIC = 5 V25°C 0.7 60

pAIIB Input bias current (see Note 4)70°C 50 600

pA

VCommon-mode input voltage range

25°C–0.2

to9

–0.3to

9.2V

VICRCommon mode in ut voltage range(see Note 5)

Full range–0.2

to8.5

V

25°C 8 8.5

VOH High-level output voltage VID = 100 mV, RL = 10 kΩ 0°C 7.8 8.5 VVOH High level out ut voltage VID 100 mV, RL 10 kΩ70°C 7.8 8.4

V

25°C 0 50

VOL Low-level output voltage VID = –100 mV, IOL = 0 0°C 0 50 mVVOL Low level out ut voltage VID 100 mV, IOL 0

70°C 0 50

mV

25°C 10 36

AVD Large-signal differential voltage amplification VO = 1 V to 6 V, RL = 10 kΩ 0°C 7.5 42 V/mVAVD Large signal differential voltage am lification VO 1 V to 6 V, RL 10 kΩ70°C 7.5 32

V/mV

25°C 65 85

CMRR Common-mode rejection ratio VIC = VICRmin 0°C 60 88 dBCMRR Common mode rejection ratio VIC VICRmin

70°C 60 88

dB

S l lt j ti ti25°C 65 95

kSVRSupply-voltage rejection ratio(∆VDD/∆VIO) VDD = 5 V to 10 V, VO = 1.4 V 0°C 60 94 dBkSVR (∆VDD/∆VIO) VDD 5 V to 10 V, VO 1.4 V

70°C 60 96

dB

V 5 V V 5 V25°C 1.9 4

IDD Supply current (two amplifiers)VO = 5 V,No load

VIC = 5 V,0°C 2.3 4.4 mADD y ( )

No load70°C 1.6 3.4

† Full range is 0°C to 70°C.NOTES: 4. The typical values of input bias current and input offset current below 5 pA were determined mathematically.

5. This range also applies to each input individually.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

7POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

electrical characteristics at specified free-air temperature, VDD = 5 V (unless otherwise noted)

PARAMETER TEST CONDITIONS TA†TLC272I, TLC272AI,TLC272BI, TLC277I UNITPARAMETER TEST CONDITIONS TA†

MIN TYP MAXUNIT

TLC272IVO = 1.4 V, VIC = 0, 25°C 1.1 10

TLC272IVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 13

mV

TLC272AIVO = 1.4 V, VIC = 0, 25°C 0.9 5

mV

V Input offset voltage

TLC272AIVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 7

VIO Input offset voltage

TLC272BIVO = 1.4 V, VIC = 0, 25°C 230 2000

TLC272BIVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 3500

V

TLC277IVO = 1.4 V, VIC = 0, 25°C 200 500

µV

TLC277IVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 2000

αVIO Temperature coefficient of input offset voltage25°C to

85°C1.8 µV/°C

I Input offset current (see Note 4)25°C 0.1 60

pAIIO Input offset current (see Note 4)

V 2 5 V V 2 5 V85°C 24 15

pA

I Input bias current (see Note 4)

VO = 2.5 V, VIC = 2.5 V25°C 0.6 60

pAIIB Input bias current (see Note 4)85°C 200 35

pA

VCommon-mode input voltage range

25°C–0.2

to4

–0.3to

4.2V

VICRCommon mode in ut voltage range(see Note 5)

Full range–0.2

to3.5

V

25°C 3.2 3.8

VOH High-level output voltage VID = 100 mV, RL = 10 kΩ –40°C 3 3.8 VVOH High level out ut voltage VID 100 mV, RL 10 kΩ85°C 3 3.8

V

25°C 0 50

VOL Low-level output voltage VID = –100 mV, IOL = 0 –40°C 0 50 mVVOL Low level out ut voltage VID 100 mV, IOL 0

85°C 0 50

mV

L i l diff ti l lt lifi ti25°C 5 23

AVDLarge-signal differential voltage amplification

VO = 1 V to 6 V, RL = 10 kΩ –40°C 3.5 32 V/mVAVD VO 1 V to 6 V, RL 10 kΩ85°C 3.5 19

V/mV

25°C 65 80

CMRR Common-mode rejection ratio VIC = VICRmin –40°C 60 81 dBCMRR Common mode rejection ratio VIC VICRmin

85°C 60 86

dB

S l lt j ti ti25°C 65 95

kSVR Supply-voltage rejection ratio(∆VDD/∆VIO)

VDD = 5 V to 10 V, VO = 1.4 V –40°C 60 92 dBkSVR (∆VDD/∆VIO)VDD 5 V to 10 V, VO 1.4 V

85°C 60 96

dB

V 2 5 V V 2 5 V25°C 1.4 3.2

IDD Supply current (two amplifiers)VO = 2.5 V,No load

VIC = 2.5 V,–40°C 1.9 4.4 mADD y ( )

No load85°C 1.1 2.4

† Full range is –40°C to 85°C.NOTES: 4. The typical values of input bias current and input offset current below 5 pA were determined mathematically.

5. This range also applies to each input individually.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

8 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

electrical characteristics at specified free-air temperature, VDD = 10 V (unless otherwise noted)

PARAMETER TEST CONDITIONS TA†TLC272I, TLC272AI,TLC272BI, TLC277I UNITPARAMETER TEST CONDITIONS TA†

MIN TYP MAXUNIT

TLC272IVO = 1.4 V, VIC = 0, 25°C 1.1 10

TLC272IVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 13

mV

TLC272AIVO = 1.4 V, VIC = 0, 25°C 0.9 5

mV

V Input offset voltage

TLC272AIVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 7

VIO Input offset voltage

TLC272BIVO = 1.4 V, VIC = 0, 25°C 290 2000

TLC272BIVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 3500

V

TLC277IVO = 1.4 V, VIC = 0, 25°C 250 800

µV

TLC277IVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 2900

αVIO Temperature coefficient of input offset voltage25°C to85°C

2 µV/°C

I Input offset current (see Note 4)25°C 0.1 60

pAIIO Input offset current (see Note 4)

V 5 V V 5 V85°C 26 1000

pA

I Input bias current (see Note 4)

VO = 5 V, VIC = 5 V25°C 0.7 60

pAIIB Input bias current (see Note 4)85°C 220 2000

pA

VCommon-mode input voltage range

25°C–0.2

to9

–0.3to

9.2V

VICRCommon mode in ut voltage range(see Note 5)

Full range–0.2

to8.5

V

25°C 8 8.5

VOH High-level output voltage VID = 100 mV, RL = 10 kΩ –40°C 7.8 8.5 VVOH High level out ut voltage VID 100 mV, RL 10 kΩ85°C 7.8 8.5

V

25°C 0 50

VOL Low-level output voltage VID = –100 mV, IOL = 0 –40°C 0 50 mVVOL Low level out ut voltage VID 100 mV, IOL 0

85°C 0 50

mV

25°C 10 36

AVD Large-signal differential voltage amplification VO = 1 V to 6 V, RL = 10 kΩ –40°C 7 46 V/mVAVD Large signal differential voltage am lification VO 1 V to 6 V, RL 10 kΩ85°C 7 31

V/mV

25°C 65 85

CMRR Common-mode rejection ratio VIC = VICRmin –40°C 60 87 dBCMRR Common mode rejection ratio VIC VICRmin

85°C 60 88

dB

S l lt j ti ti25°C 65 95

kSVR Supply-voltage rejection ratio(∆VDD/∆VIO)

VDD = 5 V to 10 V, VO = 1.4 V –40°C 60 92 dBkSVR (∆VDD/∆VIO)VDD 5 V to 10 V, VO 1.4 V

85°C 60 96

dB

V 5 V V 5 V25°C 1.4 4

IDD Supply current (two amplifiers)VO = 5 V,No load

VIC = 5 V,–40°C 2.8 5 mADD y ( )

No load85°C 1.5 3.2

† Full range is –40°C to 85°C.NOTES: 4. The typical values of input bias current and input offset current below 5 pA were determined mathematically.

5. This range also applies to each input individually.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

9POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

electrical characteristics at specified free-air temperature, VDD = 5 V (unless otherwise noted)

PARAMETER TEST CONDITIONS TA† TLC272M, TLC277MUNITPARAMETER TEST CONDITIONS TA†

MIN TYP MAXUNIT

TLC272MVO = 1.4 V, VIC = 0, 25°C 1.1 10

mV

V Input offset voltage

TLC272MVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 12

mV

VIO Input offset voltage

TLC277MVO = 1.4 V, VIC = 0, 25°C 200 500

VTLC277MVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ Full range 3750

µV

αVIOTemperature coefficient of input offsetvoltage

25°C to 125°C

2.1 µV/°C

I Input offset current (see Note 4)25°C 0.1 60 pA

IIO Input offset current (see Note 4)

V 2 5 V V 2 5 V125°C 1.4 15 nA

I Input bias current (see Note 4)

VO = 2.5 V VIC = 2.5 V25°C 0.6 60 pA

IIB Input bias current (see Note 4)125°C 9 35 nA

VCommon-mode input voltage range

25°C0to4

–0.3to

4.2V

VICRCommon mode in ut voltage range(see Note 5)

Full range0to

3.5V

25°C 3.2 3.8

VOH High-level output voltage VID = 100 mV, RL = 10 kΩ –55°C 3 3.8 VVOH High level out ut voltage VID 100 mV, RL 10 kΩ125°C 3 3.8

V

25°C 0 50

VOL Low-level output voltage VID = –100 mV, IOL = 0 –55°C 0 50 mVVOL Low level out ut voltage VID 100 mV, IOL 0

125°C 0 50

mV

25°C 5 23

AVD Large-signal differential voltage amplification VO = 0.25 V to 2 V RL = 10 kΩ –55°C 3.5 35 V/mVAVD Large signal differential voltage am lification VO 0.25 V to 2 V RL 10 kΩ125°C 3.5 16

V/mV

25°C 65 80

CMRR Common-mode rejection ratio VIC = VICRmin –55°C 60 81 dBCMRR Common mode rejection ratio VIC VICRmin

125°C 60 84

dB

S l lt j ti ti25°C 65 95

kSVRSupply-voltage rejection ratio(∆VDD/∆VIO)

VDD = 5 V to 10 V, VO = 1.4 V –55°C 60 90 dBkSVR (∆VDD/∆VIO)VDD 5 V to 10 V, VO 1.4 V

125°C 60 97

dB

V 2 5 V V 2 5 V25°C 1.4 3.2

IDD Supply current (two amplifiers)VO = 2.5 V,No load

VIC = 2.5 V,–55°C 2 5 mADD y ( )

No load125°C 1 2.2

† Full range is –55°C to 125°C.NOTES: 4. The typical values of input bias current and input offset current below 5 pA were determined mathematically.

5. This range also applies to each input individually.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

10 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

electrical characteristics at specified free-air temperature, VDD = 10 V (unless otherwise noted)

PARAMETER TEST CONDITIONS TA† TLC272M, TLC277MUNITPARAMETER TEST CONDITIONS TA†

MIN TYP MAXUNIT

TLC272MVO = 1.4 V, VIC = 0, 25°C 1.1 10

mV

V Input offset voltage

TLC272MRS = 50 Ω, RL = 10 kΩ Full range 12

mV

VIO Input offset voltage

TLC277MVO = 1.4 V, VIC = 0, 25°C 250 800

VTLC277MRS = 50 Ω, RL = 10 kΩ Full range 4300

µV

αVIOTemperature coefficient of input offsetvoltage

25°C to 125°C 2.2 µV/°C

I Input offset current (see Note 4)25°C 0.1 60 pA

IIO Input offset current (see Note 4)

V 5 V V 5 V125°C 1.8 15 nA

I Input bias current (see Note 4)

VO = 5 V, VIC = 5 V25°C 0.7 60 pA

IIB Input bias current (see Note 4)125°C 10 35 nA

VCommon-mode input voltage range

25°C0to9

–0.3to

9.2V

VICRCommon mode in ut voltage range(see Note 5)

Full range0to

8.5V

25°C 8 8.5

VOH High-level output voltage VID = 100 mV, RL = 10 kΩ –55°C 7.8 8.5 VVOH High level out ut voltage VID 100 mV, RL 10 kΩ125°C 7.8 8.4

V

25°C 0 50

VOL Low-level output voltage VID = –100 mV, IOL = 0 –55°C 0 50 mVVOL Low level out ut voltage VID 100 mV, IOL 0

125°C 0 50

mV

L i l diff ti l lt25°C 10 36

AVDLarge-signal differential voltageamplification VO = 1 V to 6 V, RL = 10 kΩ –55°C 7 50 V/mVAVD amplification VO 1 V to 6 V, RL 10 kΩ

125°C 7 27

V/mV

25°C 65 85

CMRR Common-mode rejection ratio VIC = VICRmin –55°C 60 87 dBCMRR Common mode rejection ratio VIC VICRmin

125°C 60 86

dB

S l lt j ti ti25°C 65 95

kSVRSupply-voltage rejection ratio(∆VDD/∆VIO) VDD = 5 V to 10 V, VO = 1.4 V –55°C 60 90 dBkSVR (∆VDD/∆VIO) VDD 5 V to 10 V, VO 1.4 V

125°C 60 97

dB

V 5 V V 5 V25°C 1.9 4

IDD Supply current (two amplifiers)VO = 5 V,No load

VIC = 5 V,–55°C 3 6 mADD y ( )

No load125°C 1.3 2.8

† Full range is –55°C to 125°C.NOTES: 4. The typical values of input bias current and input offset current below 5 pA were determined mathematically.

5. This range also applies to each input individually.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

11POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

electrical characteristics, VDD = 5 V, TA = 25°C (unless otherwise noted)

PARAMETER TEST CONDITIONSTLC272Y

UNITPARAMETER TEST CONDITIONSMIN TYP MAX

UNIT

V Input offset voltageVO = 1.4 V, VIC = 0,

1 1 10 mVVIO Input offset voltageVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ 1.1 10 mV

αVIO Temperature coefficient of input offset voltage 1.8 µV/°C

IIO Input offset current (see Note 4)V 2 5 V V 2 5 V

0.1 pA

IIB Input bias current (see Note 4)VO = 2.5 V, VIC = 2.5 V

0.6 pA

VICR Common-mode input voltage range (see Note 5)–0.2

to4

–0.3to

4.2V

VOH High-level output voltage VID = 100 mV, RL = 10 kΩ 3.2 3.8 V

VOL Low-level output voltage VID = –100 mV, IOL = 0 0 50 mV

AVD Large-signal differential voltage amplification VO = 0.25 V to 2 V RL = 10 kΩ 5 23 V/mV

CMRR Common-mode rejection ratio VIC = VICRmin 65 80 dB

kSVR Supply-voltage rejection ratio (∆VDD /∆VIO) VDD = 5 V to 10 V, VO = 1.4 V 65 95 dB

IDD Supply current (two amplifiers)VO = 2.5 V,No load

VIC = 2.5 V,1.4 3.2 mA

NOTES: 4. The typical values of input bias current and input offset current below 5 pA were determined mathematically.5. This range also applies to each input individually.

electrical characteristics, VDD = 10 V, TA = 25°C (unless otherwise noted)

PARAMETER TEST CONDITIONSTLC272Y

UNITPARAMETER TEST CONDITIONSMIN TYP MAX

UNIT

V Input offset voltageVO = 1.4 V, VIC = 0,

1 1 10 mVVIO Input offset voltageVO = 1.4 V,RS = 50 Ω,

VIC = 0,RL = 10 kΩ 1.1 10 mV

αVIO Temperature coefficient of input offset voltage 1.8 µV/°C

IIO Input offset current (see Note 4)V 5 V V 5 V

0.1 pA

IIB Input bias current (see Note 4)VO = 5 V, VIC = 5 V

0.7 pA

VICR Common-mode input voltage range (see Note 5)–0.2

to9

–0.3to

9.2V

VOH High-level output voltage VID = 100 mV, RL = 10 kΩ 8 8.5 V

VOL Low-level output voltage VID = –100 mV, IOL = 0 0 50 mV

AVD Large-signal differential voltage amplification VO = 1 V to 6 V, RL = 10 kΩ 10 36 V/mV

CMRR Common-mode rejection ratio VIC = VICRmin 65 85 dB

kSVR Supply-voltage rejection ratio (∆VDD/∆VIO) VDD = 5 V to 10 V, VO = 1.4 V 65 95 dB

IDD Supply current (two amplifiers)VO = 5 V, No load

VIC = 5 V,1.9 4 mA

NOTES: 4. The typical values of input bias current and input offset current below 5 pA were determined mathematically.5. This range also applies to each input individually.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

12 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

operating characteristics at specified free-air temperature, VDD = 5 V

PARAMETER TEST CONDITIONS TA

TLC272C, TLC272AC,TLC272BC, TLC277C UNITPARAMETER TEST CONDITIONS TA

MIN TYP MAXUNIT

25°C 3.6

VIPP = 1 V 0°C 4

SR Slew rate at unity gainRL = 10 kΩ,C 20 pF

VIPP 1 V

70°C 3V/ sSR Slew rate at unity gain CL = 20 pF,

See Figure 1 25°C 2.9V/µs

See Figure 1VIPP = 2.5 V 0°C 3.1VIPP 2.5 V

70°C 2.5

Vn Equivalent input noise voltagef = 1 kHz,See Figure 2

RS = 20 Ω,25°C 25 nV/√Hz

V V C 20 F25°C 320

BOM Maximum output-swing bandwidthVO = VOH,RL = 10 kΩ

CL = 20 pF,See Figure 1

0°C 340 kHzBOM Maximum out ut swing bandwidthRL = 10 kΩ, See Figure 1

70°C 260

kHz

V 10 V C 20 F25°C 1.7

B1 Unity-gain bandwidthVI = 10 mV,See Figure 3

CL = 20 pF,0°C 2 MHzB1 Unity gain bandwidth See Figure 3

70°C 1.3

MHz

V 10 V f B25°C 46°

φm Phase marginVI = 10 mV,CL = 20 pF

f = B1,See Figure 3

0°C 47°φm gCL = 20 pF, See Figure 3

70°C 43°

operating characteristics at specified free-air temperature, VDD = 10 V

PARAMETER TEST CONDITIONS TA

TLC272C, TLC272AC,TLC272BC, TLC277C UNITPARAMETER TEST CONDITIONS TA

MIN TYP MAXUNIT

25°C 5.3

VIPP = 1 V 0°C 5.9

SR Slew rate at unity gainRL = 10 kΩ,C 20 pF

VIPP 1 V

70°C 4.3V/ sSR Slew rate at unity gain CL = 20 pF,

See Figure 1 25°C 4.6V/µs

See Figure 1VIPP = 5.5 V 0°C 5.1VIPP 5.5 V

70°C 3.8

Vn Equivalent input noise voltagef = 1 kHz,See Figure 2

RS = 20 Ω,25°C 25 nV/√Hz

V V C 20 F25°C 200

BOM Maximum output-swing bandwidthVO = VOH,RL = 10 kΩ

CL = 20 pF,See Figure 1

0°C 220 kHzBOM Maximum out ut swing bandwidthRL = 10 kΩ, See Figure 1

70°C 140

kHz

V 10 V C 20 F25°C 2.2

B1 Unity-gain bandwidthVI = 10 mV,See Figure 3

CL = 20 pF,0°C 2.5 MHzB1 Unity gain bandwidth See Figure 3

70°C 1.8

MHz

V 10 V f B25°C 49°

φm Phase marginVI = 10 mV,CL = 20 pF

f = B1,See Figure 3

0°C 50°φm gCL = 20 pF, See Figure 3

70°C 46°

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

13POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

operating characteristics at specified free-air temperature, VDD = 5 V

PARAMETER TEST CONDITIONS TA

TLC272I, TLC272AI,TLC272BI, TLC277I UNITPARAMETER TEST CONDITIONS TAMIN TYP MAX

UNIT

25°C 3.6

VIPP = 1 V –40°C 4.5

SR Slew rate at unity gainRL = 10 kΩ,C 20 pF

VIPP 1 V

85°C 2.8V/ sSR Slew rate at unity gain CL = 20 pF,

See Figure 1 25°C 2.9V/µs

See Figure 1VIPP = 2.5 V –40°C 3.5VIPP 2.5 V

85°C 2.3

Vn Equivalent input noise voltagef = 1 kHz,See Figure 2

RS = 20 Ω,25°C 25 nV/√Hz

V V C 20 F25°C 320

BOM Maximum output-swing bandwidthVO = VOH,RL = 10 kΩ

CL = 20 pF,See Figure 1

–40°C 380 kHzBOM Maximum out ut swing bandwidthRL = 10 kΩ, See Figure 1

85°C 250

kHz

V 10 V C 20 F25°C 1.7

B1 Unity-gain bandwidthVI = 10 mV,See Figure 3

CL = 20 pF,–40°C 2.6 MHzB1 Unity gain bandwidth See Figure 3

85°C 1.2

MHz

V 10 V f B25°C 46°

φm Phase marginVI = 10 mV,CL = 20 pF

f = B1,See Figure 3

–40°C 49°φm gCL = 20 pF, See Figure 3

85°C 43°

operating characteristics at specified free-air temperature, VDD = 10 V

PARAMETER TEST CONDITIONS TA

TLC272I, TLC272AI,TLC272BI, TLC277I UNITPARAMETER TEST CONDITIONS TAMIN TYP MAX

UNIT

25°C 5.3

VIPP = 1 V –40°C 6.8

SR Slew rate at unity gainRL = 10 kΩ,C 20 pF

VIPP 1 V

85°C 4V/ sSR Slew rate at unity gain CL = 20 pF,

See Figure 1 25°C 4.6V/µs

See Figure 1VIPP = 5.5 V –40°C 5.8VIPP 5.5 V

85°C 3.5

Vn Equivalent input noise voltagef = 1 kHz,See Figure 2

RS = 20 Ω,25°C 25 nV/√Hz

V V C 20 F25°C 200

BOM Maximum output-swing bandwidthVO = VOH,RL = 10 kΩ

CL = 20 pF,See Figure 1

–40°C 260 kHzBOM Maximum out ut swing bandwidthRL = 10 kΩ, See Figure 1

85°C 130

kHz

V 10 V C 20 F25°C 2.2

B1 Unity-gain bandwidthVI = 10 mV,See Figure 3

CL = 20 pF,–40°C 3.1 MHzB1 Unity gain bandwidth See Figure 3

85°C 1.7

MHz

V 10 V f B25°C 49°

φm Phase marginVI = 10 mV,CL = 20 pF

f = B1,See Figure 3

–40°C 52°φm gCL = 20 pF, See Figure 3

85°C 46°

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

14 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

operating characteristics at specified free-air temperature, VDD = 5 V

PARAMETER TEST CONDITIONS TATLC272M, TLC277M

UNITPARAMETER TEST CONDITIONS TA MIN TYP MAXUNIT

25°C 3.6

VIPP = 1 V –55°C 4.7

SR Slew rate at unity gainRL = 10 kΩ,C 20 pF

VIPP 1 V

125°C 2.3V/ sSR Slew rate at unity gain CL = 20 pF,

See Figure 1 25°C 2.9V/µs

See Figure 1VIPP = 2.5 V –55°C 3.7VIPP 2.5 V

125°C 2

Vn Equivalent input noise voltagef = 1 kHz,See Figure 2

RS = 20 Ω,25°C 25 nV/√Hz

V V C 20 F25°C 320

BOM Maximum output-swing bandwidthVO = VOH,RL = 10 kΩ

CL = 20 pF,See Figure 1

–55°C 400 kHzBOM Maximum out ut swing bandwidthRL = 10 kΩ, See Figure 1

125°C 230

kHz

V 10 V C 20 F25°C 1.7

B1 Unity-gain bandwidthVI = 10 mV,See Figure 3

CL = 20 pF,–55°C 2.9 MHzB1 Unity gain bandwidth See Figure 3125°C 1.1

MHz

V 10 V f B25°C 46°

φm Phase marginVI = 10 mV,CL = 20 pF

f = B1,See Figure 3

–55°C 49°φm gCL = 20 pF, See Figure 3

125°C 41°

operating characteristics at specified free-air temperature, VDD = 10 V

PARAMETER TEST CONDITIONS TATLC272M, TLC277M

UNITPARAMETER TEST CONDITIONS TA MIN TYP MAXUNIT

25°C 5.3

VIPP = 1 V –55°C 7.1

SR Slew rate at unity gainRL = 10 kΩ,C 20 pF

VIPP 1 V

125°C 3.1V/ sSR Slew rate at unity gain CL = 20 pF,

See Figure 1 25°C 4.6V/µs

See Figure 1VIPP = 5.5 V –55°C 6.1VIPP 5.5 V

125°C 2.7

Vn Equivalent input noise voltagef = 1 kHz,See Figure 2

RS = 20 Ω,25°C 25 nV/√Hz

V V C 20 F25°C 200

BOM Maximum output-swing bandwidthVO = VOH,RL = 10 kΩ

CL = 20 pF,See Figure 1

–55°C 280 kHzBOM Maximum out ut swing bandwidthRL = 10 kΩ, See Figure 1

125°C 110

kHz

V 10 V C 20 F25°C 2.2

B1 Unity-gain bandwidthVI = 10 mV,See Figure 3

CL = 20 pF,–55°C 3.4 MHzB1 Unity gain bandwidth See Figure 3125°C 1.6

MHz

V 10 V f B25°C 49°

φm Phase marginVI = 10 mV,CL = 20 pF

f = B1,See Figure 3

–55°C 52°φm gCL = 20 pF, See Figure 3

125°C 44°

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

15POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

operating characteristics, VDD = 5 V, TA = 25°C

PARAMETER TEST CONDITIONSTLC272Y

UNITPARAMETER TEST CONDITIONSMIN TYP MAX

UNIT

SR Slew rate at unity gainRL = 10 kΩ, CL = 20 pF, VIPP = 1 V 3.6

V/ sSR Slew rate at unity gainRL = 10 kΩ,See Figure 1

CL = 20 F,

VIPP = 2.5 V 2.9V/µs

Vn Equivalent input noise voltage f = 1 kHz, RS = 20 Ω, See Figure 2 25 nV/√Hz

BOM Maximum output-swing bandwidthVO = VOH,See Figure 1

CL = 20 pF, RL = 10 kΩ,320 kHz

B1 Unity-gain bandwidth VI = 10 mV, CL = 20 pF, See Figure 3 1.7 MHz

φm Phase marginVI = 10 mV,See Figure 3

f = B1, CL = 20 pF,46°

operating characteristics, VDD = 10 V, TA = 25°C

PARAMETER TEST CONDITIONSTLC272Y

UNITPARAMETER TEST CONDITIONSMIN TYP MAX

UNIT

SR Slew rate at unity gainRL = 10 kΩ, CL = 20 pF, VIPP = 1 V 5.3

V/ sSR Slew rate at unity gainRL = 10 kΩ,See Figure 1

CL = 20 F,

VIPP = 5.5 V 4.6V/µs

Vn Equivalent input noise voltage f = 1 kHz, RS = 20 Ω, See Figure 2 25 nV/√Hz

BOM Maximum output-swing bandwidthVO = VOH,See Figure 1

CL = 20 pF, RL = 10 kΩ,200 kHz

B1 Unity-gain bandwidth VI = 10 mV, CL = 20 pF, See Figure 3 2.2 MHz

φm Phase marginVI = 10 mV,See Figure 3

f = B1, CL = 20 pF,49°

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

16 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

PARAMETER MEASUREMENT INFORMATION

single-supply versus split-supply test circuits

Because the TLC272 and TLC277 are optimized for single-supply operation, circuit configurations used for thevarious tests often present some inconvenience since the input signal, in many cases, must be offset fromground. This inconvenience can be avoided by testing the device with split supplies and the output load tied tothe negative rail. A comparison of single-supply versus split-supply test circuits is shown below. The use of eithercircuit gives the same result.

VDD–

VDD+

–

+

CL RL

VOVIVI

VO

RLCL

VDD

–

+

(a) SINGLE SUPPLY (b) SPLIT SUPPLY

Figure 1. Unity-Gain Amplifier

VO

2 kΩ

20 Ω20 Ω

VDD–

20 Ω

2 kΩ

VO

20 Ω

1/2 VDD

–

+

VDD+–

+

VDD

(b) SPLIT SUPPLY(a) SINGLE SUPPLY

Figure 2. Noise-Test Circuit

VDD–

VDD+

–

+

10 kΩ

VO

100 Ω

CL

VIVI

1/2 VDDCL

100 Ω

VO

10 kΩ

–

+

VDD

(a) SINGLE SUPPLY (b) SPLIT SUPPLY

Figure 3. Gain-of-100 Inverting Amplifier

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

17POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

PARAMETER MEASUREMENT INFORMATION

input bias current

Because of the high input impedance of the TLC272 and TLC277 operational amplifiers, attempts to measurethe input bias current can result in erroneous readings. The bias current at normal room ambient temperatureis typically less than 1 pA, a value that is easily exceeded by leakages on the test socket. Two suggestions areoffered to avoid erroneous measurements:

1. Isolate the device from other potential leakage sources. Use a grounded shield around and between thedevice inputs (see Figure 4). Leakages that would otherwise flow to the inputs are shunted away.

2. Compensate for the leakage of the test socket by actually performing an input bias current test (usinga picoammeter) with no device in the test socket. The actual input bias current can then be calculatedby subtracting the open-socket leakage readings from the readings obtained with a device in the testsocket.

One word of caution: many automatic testers as well as some bench-top operational amplifier testers use theservo-loop technique with a resistor in series with the device input to measure the input bias current (the voltagedrop across the series resistor is measured and the bias current is calculated). This method requires that adevice be inserted into the test socket to obtain a correct reading; therefore, an open-socket reading is notfeasible using this method.

8 5

1 4

V = VIC

Figure 4. Isolation Metal Around Device Inputs(JG and P packages)

low-level output voltage

To obtain low-supply-voltage operation, some compromise was necessary in the input stage. This compromiseresults in the device low-level output being dependent on the common-mode input voltage level as well as thedifferential input voltage level. When attempting to correlate low-level output readings with those quoted in theelectrical specifications, these two conditions should be observed. If conditions other than these are to be used,please refer to Figures 14 through 19 in the Typical Characteristics of this data sheet.

input offset voltage temperature coefficient

Erroneous readings often result from attempts to measure temperature coefficient of input offset voltage. Thisparameter is actually a calculation using input offset voltage measurements obtained at two differenttemperatures. When one (or both) of the temperatures is below freezing, moisture can collect on both the deviceand the test socket. This moisture results in leakage and contact resistance, which can cause erroneous inputoffset voltage readings. The isolation techniques previously mentioned have no effect on the leakage since themoisture also covers the isolation metal itself, thereby rendering it useless. It is suggested that thesemeasurements be performed at temperatures above freezing to minimize error.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

18 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

PARAMETER MEASUREMENT INFORMATION

full-power response

Full-power response, the frequency above which the operational amplifier slew rate limits the output voltageswing, is often specified two ways: full-linear response and full-peak response. The full-linear response isgenerally measured by monitoring the distortion level of the output while increasing the frequency of a sinusoidalinput signal until the maximum frequency is found above which the output contains significant distortion. Thefull-peak response is defined as the maximum output frequency, without regard to distortion, above which fullpeak-to-peak output swing cannot be maintained.

Because there is no industry-wide accepted value for significant distortion, the full-peak response is specifiedin this data sheet and is measured using the circuit of Figure 1. The initial setup involves the use of a sinusoidalinput to determine the maximum peak-to-peak output of the device (the amplitude of the sinusoidal wave isincreased until clipping occurs). The sinusoidal wave is then replaced with a square wave of the sameamplitude. The frequency is then increased until the maximum peak-to-peak output can no longer be maintained(Figure 5). A square wave is used to allow a more accurate determination of the point at which the maximumpeak-to-peak output is reached.

(d) f > BOM(c) f = BOM(b) BOM > f > 1 kHz(a) f = 1 kHz

Figure 5. Full-Power-Response Output Signal

test time

Inadequate test time is a frequent problem, especially when testing CMOS devices in a high-volume,short-test-time environment. Internal capacitances are inherently higher in CMOS than in bipolar and BiFETdevices and require longer test times than their bipolar and BiFET counterparts. The problem becomes morepronounced with reduced supply levels and lower temperatures.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

19POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS

Table of GraphsFIGURE

VIO Input offset voltage Distribution 6, 7

αVIO Temperature coefficient of input offset voltage Distribution 8, 9

vs High-level output current 10, 11VOH High-level output voltage

vs High-level out ut currentvs Supply voltage

10, 1112VOH High level out ut voltage vs Su ly voltage

vs Free-air temperature1213

vs Common-mode input voltage 14, 15

V Low level output voltage

vs Common-mode in ut voltagevs Differential input voltage

14, 1516

VOL Low-level output voltagevs Differential in ut voltagevs Free-air temperature

1617vs Free air tem erature

vs Low-level output current17

18, 19

vs Supply voltage 20AVD Large-signal differential voltage amplification

vs Su ly voltagevs Free-air temperature

2021AVD Large signal differential voltage am lification vs Free air tem erature

vs Frequency21

32, 33

IIB Input bias current vs Free-air temperature 22

IIO Input offset current vs Free-air temperature 22

VIC Common-mode input voltage vs Supply voltage 23

I Supply currentvs Supply voltage 24

IDD Supply currentvs Su ly voltagevs Free-air temperature

2425

SR Slew ratevs Supply voltage 26

SR Slew ratevs Su ly voltagevs Free-air temperature

2627

Normalized slew rate vs Free-air temperature 28

VO(PP) Maximum peak-to-peak output voltage vs Frequency 29

B Unity gain bandwidthvs Free-air temperature 30

B1 Unity-gain bandwidthvs Free air tem eraturevs Supply voltage

3031

vs Supply voltage 34φm Phase margin

vs Su ly voltagevs Free-air temperature

3435φm Phase margin vs Free air tem erature

vs Load capacitance3536

Vn Equivalent input noise voltage vs Frequency 37

Phase shift vs Frequency 32, 33

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

20 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS

Figure 6

–50

Per

cen

tag

e o

f U

nit

s –

%

VIO – Input Offset Voltage – mV5

60

–4 –3 –2 –1 0 1 2 3 4

10

20

30

40

50

ÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌ

753 Amplifiers Tested From 6 Wafer LotsÌÌÌÌÌÌÌÌ

VDD = 5 VÌÌÌÌÌÌÌÌÌÌ

TA = 25°CÌÌÌÌÌÌÌÌÌÌ

P Package

DISTRIBUTION OF TLC272INPUT OFFSET VOLTAGE

Figure 7

50

40

30

20

10

43210–1–2–3–4

60

5VIO – Input Offset Voltage – mV

Per

cen

tag

e o

f U

nit

s –

%

0–5

DISTRIBUTION OF TLC272INPUT OFFSET VOLTAGE

ÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌ

753 Amplifiers Tested From 6 Wafer Lots

ÌÌÌÌVDD = 10 VÌÌÌÌÌÌÌÌ

TA = 25°CÌÌÌÌÌÌÌÌ

P Package

Figure 8

ÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌ

324 Amplifiers Tested From 8 Wafer LotsVDD = 5 VTA = 25°C to 125°CP PackageOutliers:(1) 20.5 µV/°C

50

40

30

20

10

86420–2–4–6–8

60

10

Per

cen

tag

e o

f U

nit

s –

%

0–10

αVIO – Temperature Coefficient – µV/°C

DISTRIBUTION OF TLC272 AND TLC277INPUT OFFSET VOLTAGE

TEMPERATURE COEFFICIENT

Figure 9

ÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌÌ

324 Amplifiers Tested From 8 Wafer LotsVDD = 5 VTA = 25°C to 125°CP PackageOutliers:(1) 21.2 µV/°C

–100

Per

cen

tag

e o

f U

nit

s –

%

10

60

–8 –6 –4 –2 0 2 4 6 8

10

20

30

40

50

ÁαVIO – Temperature Coefficient – µV/°C

DISTRIBUTION OF TLC272 AND TLC277INPUT OFFSET VOLTAGE

TEMPERATURE COEFFICIENT

Á

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

21POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS†

Figure 10

VDD = 3 V

VDD = 4 V

VDD = 5 V

VID = 100 mVTA = 25°CSee Note A4

3

2

1

– 8– 6– 4– 2

5

– 10IOH – High-Level Output Current – mA

VO

H –

Hig

h-L

evel

Ou

tpu

t V

olt

age

– V

00

HIGH-LEVEL OUTPUT VOLTAGEvs

HIGH-LEVEL OUTPUT CURRENT

ÁÁÁÁÁÁ

V OH

NOTE A: The 3-V curve only applies to the C version.

Figure 11

TA = 25°CVID = 100 mV

VDD = 10 V

VDD = 16 V14

12

10

8

6

4

2

– 30– 20– 10

16

– 40IOH – High-Level Output Current – mA

VO

H –

Hig

h-L

evel

Ou

tpu

t V

olt

age

– V

00

HIGH-LEVEL OUTPUT VOLTAGEvs

HIGH-LEVEL OUTPUT CURRENT

ÁÁÁÁÁÁ

V OH

– 5 – 15 – 20 – 25 – 35

Figure 12

ÌÌÌÌÌTA = 25°CÌÌÌÌÌÌÌÌ

RL = 10 kΩÌÌÌÌÌVID = 100 mV

0

16

2

4

6

8

10

12

14

1412108642 16VDD – Supply Voltage – V

0

HIGH-LEVEL OUTPUT VOLTAGEvs

SUPPLY VOLTAGE

VO

H –

Hig

h-L

evel

Ou

tpu

t V

olt

age

– V

ÁÁÁÁÁÁ

V OH

Figure 13

VDD = 10 V

VDD = 5 VVID = 100 mAIOH = –5 mA

–75TA – Free-Air Temperature – °C

125

VDD –1.6

–50 –25 0 20 50 75 100

VDD –1.8

HIGH-LEVEL OUTPUT VOLTAGEvs

FREE-AIR TEMPERATURE

VDD –1.7

VDD –1.9

VDD –2.1

VDD –2

VDD –2.3

VDD –2.2

VDD –2.4

VO

H –

Hig

h-L

evel

Ou

tpu

t V

olt

age

– V

ÁÁÁÁÁÁ

V OH

† Data at high and low temperatures are applicable only within the rated operating free-air temperature ranges of the various devices.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

22 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS†

Figure 14

VID = –1 V

VID = –100 mV

VDD = 5 VIOL = 5 mA

TA = 25°C600

500

400

321

700

4VIC – Common-Mode Input Voltage – V

VO

L –

Lo

w-L

evel

Ou

tpu

t V

olt

age

– m

V

3000

LOW-LEVEL OUTPUT VOLTAGEvs

COMMON-MODE INPUT VOLTAGE

ÁÁÁÁÁÁ

V OL

0.5 1.5 2.5 3.5

650

550

450

350

Figure 15

VID = –100 mV

VID = –2.5 V

VID = –1 V

TA = 25°CIOL = 5 mAVDD = 10 V

108642

500

450

400

350

300

VIC – Common-Mode Input Voltage – V0

250

LOW-LEVEL OUTPUT VOLTAGEvs

COMMON-MODE INPUT VOLTAGE

VO

L –

Lo

w-L

evel

Ou

tpu

t V

olt

age

– m

V

ÁÁÁÁ

V OL

1 3 5 7 9

Figure 16

VDD = 10 V

VDD = 5 V

TA = 25°CVIC = |VID/2|IOL = 5 mA

0

100

200

300

400

500

600

700

800

–8–6–4–2 –10VID – Differential Input Voltage – V

0

LOW-LEVEL OUTPUT VOLTAGEvs

DIFFERENTIAL INPUT VOLTAGE

–1 –3 –5 –7 –9

VO

L –

Lo

w-L

evel

Ou

tpu

t V

olt

age

– m

V

ÁÁÁÁÁÁ

V OL

Figure 17

VDD = 10 V

VDD = 5 V

IOL = 5 mAVID = –1 VVIC = 0.5 V

800

700

600

500

400

300

200

100

1007550250–25–50

900

125TA – Free-Air Temperature – °C

0–75

LOW-LEVEL OUTPUT VOLTAGEvs

FREE-AIR TEMPERATURE

VO

L –

Lo

w-L

evel

Ou

tpu

t V

olt

age

– m

V

ÁÁÁÁÁÁ

V OL

† Data at high and low temperatures are applicable only within the rated operating free-air temperature ranges of the various devices.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

23POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS†

Figure 18

LOW-LEVEL OUTPUT VOLTAGEvs

LOW-LEVEL OUTPUT CURRENT

VDD = 5 V

VDD = 4 V

VDD = 3 V

ÌÌÌÌÌÌÌÌ

TA = 25°CSee Note A

ÌÌÌÌÌÌÌÌ

VIC = 0.5 VÌÌÌÌVID = –1 V0.9

0.8

0.7

0.6

0.5

0.4

0.3

0.2

0.1

76543210

8

1.0

IOL – Low-Level Output Current – mA

VO

L –

Lo

w-L

evel

Ou

tpu

t V

olt

age

– V

0

ÁÁÁÁV O

L

NOTE A: The 3-V curve only applies to the C version.Figure 19

LOW-LEVEL OUTPUT VOLTAGEvs

LOW-LEVEL OUTPUT CURRENT

VDD = 16 V

VDD = 10 V

ÌÌÌÌVID = –1 V

ÌÌÌÌÌÌÌÌÌÌ

VIC = 0.5 V

ÌÌÌÌÌÌÌÌ

TA = 25°C2.5

2.0

1.5

1.0

0.5

2520151050

30

3.0

IOL – Low-Level Output Current – mA

VO

L –

Lo

w-L

evel

Ou

tpu

t V

olt

age

– V

0

ÁÁÁÁÁÁ

V OL

Figure 20

0

60

160

2 4 6 8 10 12 14

10

20

30

40

50

VDD – Supply Voltage – V

TA = –55°C

RL = 10 kΩ

LARGE-SIGNALDIFFERENTIAL VOLTAGE AMPLIFICATION

vsSUPPLY VOLTAGE

ÌÌÌÌÌÌÌÌ

TA = 25°C

ÌÌÌÌTA = 85°C

ÌÌÌÌÌTA = 125°C

ÌÌÌÌTA = 0°C

AV

D –

Lar

ge-

Sig

nal

Dif

fere

nti

al

ÁÁÁÁÁÁ

AV

DV

olt

age

Am

plif

icat

ion

– V

/mV

Figure 21

–75

50

1250

–50 –25 0 25 50 75 100

5

10

15

20

25

30

35

40

45

VDD = 5 V

VDD = 10 V

RL = 10 kΩ

LARGE-SIGNALDIFFERENTIAL VOLTAGE AMPLIFICATION

vsFREE-AIR TEMPERATURE

TA – Free-Air Temperature – °C

AV

D –

Lar

ge-

Sig

nal

Dif

fere

nti

al

ÁÁÁÁÁÁ

AV

DV

olt

age

Am

plif

icat

ion

– V

/mV

† Data at high and low temperatures are applicable only within the rated operating free-air temperature ranges of the various devices.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

24 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS†

Figure 22

INPUT BIAS CURRENT AND INPUT OFFSET CURRENTvs

FREE-AIR TEMPERATURE

0.1125

10000

45 65 85 105

1

10

100

1000

25

– In

pu

t B

ias

and

Off

set

Cu

rren

ts –

pA

VDD = 10 VVIC = 5 VSee Note A

ÌÌIIB

I IB

I IO

and

TA – Free-Air Temperature – °C

ÌÌÌÌ

IIO

35 55 75 95 115

NOTE A: The typical values of input bias current and inputoffset current below 5 pA were determined mathematically.

Figure 23

COMMON-MODEINPUT VOLTAGE POSITIVE LIMIT

vsSUPPLY VOLTAGE

0VDD – Supply Voltage – V

16

160

2 4 6 8 10 12 14

2

4

6

8

10

12

14TA = 25°C

ICV

– C

om

mo

n-M

od

e In

pu

t V

olt

age

– V

Figure 24

SUPPLY CURRENTvs

SUPPLY VOLTAGE

0VDD – Supply Voltage – V

5

160

2 4 6 8 10 12 14

1

2

3

4

VO = VDD/2No Load

TA = –55°C

– S

up

ply

Cu

rren

t –

mA

I DD

0.5

1.5

2.5

3.5

4.5

ÌÌÌÌTA = 70°C

ÌÌÌÌÌÌÌÌ

TA = 125°C

ÌÌÌTA = 0°CÌÌÌÌÌÌÌÌ

TA = 25°C

Figure 25

SUPPLY CURRENTvs

FREE-AIR TEMPERATURE

–75

– S

up

ply

Cu

rren

t –

mA

2

1250

0.5

1

1.5

–50 –25 0 25 50 75 100

No LoadVO = VDD/2

VDD = 10 V

VDD = 5 V

2.5

3

3.5

4

I DD

TA – Free-Air Temperature – °C

† Data at high and low temperatures are applicable only within the rated operating free-air temperature ranges of the various devices.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

25POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS†

Figure 26

AV = 1VIPP = 1 VRL = 10 kΩCL = 20 pFTA = 25°CSee Figure 1

8

7

6

5

4

3

2

1

14121086420

16VDD – Supply Voltage – V

0

SLEW RATEvs

SUPPLY VOLTAGE

µsS

R –

Sle

w R

ate

– V

/

Figure 27

VIPP = 1 VVDD = 10 V

VIPP = 2.5 VVDD = 5 V

VIPP = 1 V

VDD = 5 V

VDD = 10 VVIPP = 5.5 V

–750

1

2

3

4

5

6

7

8

1007550250–25–50 125TA – Free-Air Temperature – °C

SLEW RATEvs

FREE-AIR TEMPERATURE

AV = 1RL = 10 kΩCL = 20 pFSee Figure 1

µsS

R –

Sle

w R

ate

– V

/

Figure 28

VDD = 5 V

VDD = 10 V

1.5

0.5

0.6

0.7

0.8

0.9

1.0

1.1

1.2

1.3

1.4

CL = 20 pFRL = 10 kΩVIPP = 1 VAV = 1

1007550250–25–50 125

TA – Free-Air Temperature – °C

No

rmal

ized

Sle

w R

ate

–75

NORMALIZED SLEW RATEvs

FREE-AIR TEMPERATURE

Figure 29

TA = –55°CTA = 25°CTA = 125°C

RL = 10 kΩSee Figure 1

VDD = 5 V

VDD = 10 V

1000100

9

8

7

6

5

4

3

2

1

010000

10

f – Frequency – kHz

10

MAXIMUM PEAK OUTPUT VOLTAGEvs

FREQUENCY

– M

axim

um

Pea

k-to

-Pea

k O

utp

ut

Vo

ltag

e –

VV O

(PP

)

† Data at high and low temperatures are applicable only within the rated operating free-air temperature ranges of the various devices.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

26 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS†

Figure 30

2.5

2.0

1.5

1007550250–25–501.0

3.0

TA – Free-Air Temperature – °C

–75

See Figure 3CL = 20 pFVI = 10 mVVDD = 5 V

UNITY-GAIN BANDWIDTHvs

FREE-AIR TEMPERATURE

– U

nit

y-G

ain

Ban

dw

idth

– M

Hz

B1

125

Figure 31

2.0

1.5

14121086421.0

16

2.5

VDD – Supply Voltage – V

0

VI = 10 mVCL = 20 pFTA = 25°CSee Figure 3

UNITY-GAIN BANDWIDTHvs

SUPPLY VOLTAGE

– U

nit

y-G

ain

Ban

dw

idth

– M

Hz

B1

Ph

ase

Sh

ift

LARGE-SIGNAL DIFFERENTIAL VOLTAGEAMPLIFICATION AND PHASE SHIFT

vsFREQUENCY

Phase Shift

AVD

VDD = 5 VRL = 10 kΩTA = 25°C

180°

0°

30°

60°

90°

120°

150°

106

105

104

103

102

101

1

1 M100 k10 k1 k1000.1

10 M

f – Frequency – Hz

10

107

AV

D –

Lar

ge-

Sig

nal

Dif

fere

nti

al

ÁÁ

AV

DV

olt

age

Am

plif

icat

ion

Figure 32

† Data at high and low temperatures are applicable only within the rated operating free-air temperature ranges of the various devices.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

27POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS†

Phase Shift

AVD

TA = 25°CRL = 10 kΩVDD = 10 V

Ph

ase

Sh

ift

150°

120°

90°

60°

30°

0°

180°

106

105

104

103

102

101

1

1 M100 k10 k1 k1000.1

10 M

f – Frequency – Hz

10

107

LARGE-SIGNAL DIFFERENTIAL VOLTAGEAMPLIFICATION AND PHASE SHIFT

vsFREQUENCY

AV

D –

Lar

ge-

Sig

nal

Dif

fere

nti

al

ÁÁÁÁ

AV

DV

olt

age

Am

plif

icat

ion

Figure 33

Figure 34

45°

See Figure 3

VI = 10 mV

TA = 25°CCL = 20 pF

51°

49°

47°

1412108642 16

53°

VDD – Supply Voltage – V

m –

Ph

ase

Mar

gin

0

PHASE MARGINvs

SUPPLY VOLTAGE

mφ

48°

46°

52°

50°

Figure 35

PHASE MARGINvs

FREE-AIR TEMPERATURE

TA – Free-Air Temperature – °C

See Figure 3

VI = 10 mVCL = 20 pF

VDD = 5 V

48°

46°

44°

42°

1007550250–25–5040°

125

50°

–75

m –

Ph

ase

Mar

gin

mφ

† Data at high and low temperatures are applicable only within the rated operating free-air temperature ranges of the various devices.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

28 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

TYPICAL CHARACTERISTICS

Figure 36

VDD = 5 V

TA = 25°CVI = 10 mV

See Figure 345°

40°

35°

30°

8060402025°

100

50°

CL – Capacitive Load – pF

0

PHASE MARGINvs

CAPACITIVE LOAD

m –

Ph

ase

Mar

gin

mφ

10 30 50 70 90

Figure 37

VN

– E

qu

ival

ent

Inp

ut

No

ise

Vo

ltag

e –

See Figure 2

RS = 20 ΩTA = 25°C

VDD = 5 V

10010

300

200

100

01000

400

f – Frequency – Hz

1

EQUIVALENT INPUT NOISE VOLTAGEvs

FREQUENCY

nV

/H

zV

n

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

29POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

APPLICATION INFORMATION

single-supply operation

While the TLC272 and TLC277 perform well using dual power supplies (also called balanced or split supplies),the design is optimized for single-supply operation. This design includes an input common-mode voltage rangethat encompasses ground as well as an output voltage range that pulls down to ground. The supply voltagerange extends down to 3 V (C-suffix types), thus allowing operation with supply levels commonly available forTTL and HCMOS; however, for maximum dynamic range, 16-V single-supply operation is recommended.

Many single-supply applications require that a voltage be applied to one input to establish a reference level thatis above ground. A resistive voltage divider is usually sufficient to establish this reference level (see Figure 38).The low input bias current of the TLC272 and TLC277 permits the use of very large resistive values to implementthe voltage divider, thus minimizing power consumption.

The TLC272 and TLC277 work well in conjunction with digital logic; however, when powering both linear devicesand digital logic from the same power supply, the following precautions are recommended:

1. Power the linear devices from separate bypassed supply lines (see Figure 39); otherwise, the lineardevice supply rails can fluctuate due to voltage drops caused by high switching currents in the digitallogic.

2. Use proper bypass techniques to reduce the probability of noise-induced errors. Single capacitivedecoupling is often adequate; however, high-frequency applications may require RC decoupling.

–

+

C0.01 µF

R3VREF

VI

R1R2

VDD

VO

R4

VREF VDDR3

R1 R3

VO (VREF VI)R4R2

VREF

Figure 38. Inverting Amplifier With Voltage Reference

(b) SEPARATE BYPASSED SUPPLY RAILS (preferred)

(a) COMMON SUPPLY RAILS

–

+

–

+

Logic Logic LogicPowerSupply

SupplyPower

LogicLogicLogicOUT

OUT

Figure 39. Common vs Separate Supply Rails

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

30 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

APPLICATION INFORMATION

input characteristics

The TLC272 and TLC277 are specified with a minimum and a maximum input voltage that, if exceeded at eitherinput, could cause the device to malfunction. Exceeding this specified range is a common problem, especiallyin single-supply operation. Note that the lower range limit includes the negative rail, while the upper range limitis specified at VDD – 1 V at TA = 25°C and at VDD – 1.5 V at all other temperatures.

The use of the polysilicon-gate process and the careful input circuit design gives the TLC272 and TLC277 verygood input offset voltage drift characteristics relative to conventional metal-gate processes. Offset voltage driftin CMOS devices is highly influenced by threshold voltage shifts caused by polarization of the phosphorusdopant implanted in the oxide. Placing the phosphorus dopant in a conductor (such as a polysilicon gate)alleviates the polarization problem, thus reducing threshold voltage shifts by more than an order of magnitude.The offset voltage drift with time has been calculated to be typically 0.1 µV/month, including the first month ofoperation.

Because of the extremely high input impedance and resulting low bias current requirements, the TLC272 andTLC277 are well suited for low-level signal processing; however, leakage currents on printed-circuit boards andsockets can easily exceed bias current requirements and cause a degradation in device performance. It is goodpractice to include guard rings around inputs (similar to those of Figure 4 in the Parameter MeasurementInformation section). These guards should be driven from a low-impedance source at the same voltage levelas the common-mode input (see Figure 40).

Unused amplifiers should be connected as grounded unity-gain followers to avoid possible oscillation.

noise performance

The noise specifications in operational amplifier circuits are greatly dependent on the current in the first-stagedifferential amplifier. The low input bias current requirements of the TLC272 and TLC277 result in a very lownoise current, which is insignificant in most applications. This feature makes the devices especially favorableover bipolar devices when using values of circuit impedance greater than 50 kΩ, since bipolar devices exhibitgreater noise currents.

VI

–

+

–

+

VI

(b) INVERTING AMPLIFIER

–

+

(c) UNITY-GAIN AMPLIFIER(a) NONINVERTING AMPLIFIER

VIOUT OUT OUT

Figure 40. Guard-Ring Schemes

output characteristics

The output stage of the TLC272 and TLC277 is designed to sink and source relatively high amounts of current(see typical characteristics). If the output is subjected to a short-circuit condition, this high current capability cancause device damage under certain conditions. Output current capability increases with supply voltage.

All operating characteristics of the TLC272 and TLC277 are measured using a 20-pF load. The devices candrive higher capacitive loads; however, as output load capacitance increases, the resulting response poleoccurs at lower frequencies, thereby causing ringing, peaking, or even oscillation (see Figure 41). In manycases, adding a small amount of resistance in series with the load capacitance alleviates the problem.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

31POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

APPLICATION INFORMATION

output characteristics (continued)

(c) CL = 150 pF, RL = NO LOAD

(b) CL = 130 pF, RL = NO LOAD(a) CL = 20 pF, RL = NO LOAD

VI

–2.5 V

CL

VO

2.5 V

–

+

TA = 25°Cf = 1 kHzVIPP = 1 V

(d) TEST CIRCUIT

Figure 41. Effect of Capacitive Loads and Test Circuit

Although the TLC272 and TLC277 possess excellent high-level output voltage and current capability, methodsfor boosting this capability are available, if needed. The simplest method involves the use of a pullup resistor(RP) connected from the output to the positive supply rail (see Figure 42). There are two disadvantages to theuse of this circuit. First, the NMOS pulldown transistor N4 (see equivalent schematic) must sink a comparativelylarge amount of current. In this circuit, N4 behaves like a linear resistor with an on resistance betweenapproximately 60 Ω and 180 Ω, depending on how hard the operational amplifier input is driven. With very lowvalues of RP, a voltage offset from 0 V at the output occurs. Second, pullup resistor RP acts as a drain load toN4 and the gain of the operational amplifier is reduced at output voltage levels where N5 is not supplying theoutput current.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

32 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

APPLICATION INFORMATION

output characteristics (continued)

Figure 42. Resistive Pullup to Increase VOH

VDD – VOIF + IL + IP

Rp =

IL

IF

IP

RLR1R2

VO

RP

VDD

VI

–

+

ÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁÁ

Ip = Pullup current required bythe operational amplifier(typically 500 µA)

Figure 43. Compensation for Input Capacitance

C

–

+

VO

feedback

Operational amplifier circuits almost always employ feedback, and since feedback is the first prerequisite foroscillation, some caution is appropriate. Most oscillation problems result from driving capacitive loads(discussed previously) and ignoring stray input capacitance. A small-value capacitor connected in parallel withthe feedback resistor is an effective remedy (see Figure 43). The value of this capacitor is optimized empirically.

electrostatic discharge protection

The TLC272 and TLC277 incorporate an internal electrostatic discharge (ESD) protection circuit that preventsfunctional failures at voltages up to 2000 V as tested under MIL-STD-883C, Method 3015.2. Care should beexercised, however, when handling these devices as exposure to ESD may result in the degradation of thedevice parametric performance. The protection circuit also causes the input bias currents to be temperaturedependent and have the characteristics of a reverse-biased diode.

latch-up

Because CMOS devices are susceptible to latch-up due to their inherent parasitic thyristors, the TLC272 andTLC277 inputs and outputs were designed to withstand –100-mA surge currents without sustaining latch-up;however, techniques should be used to reduce the chance of latch-up whenever possible. Internal protectiondiodes should not, by design, be forward biased. Applied input and output voltage should not exceed the supplyvoltage by more than 300 mV. Care should be exercised when using capacitive coupling on pulse generators.Supply transients should be shunted by the use of decoupling capacitors (0.1 µF typical) located across thesupply rails as close to the device as possible.

The current path established if latch-up occurs is usually between the positive supply rail and ground and canbe triggered by surges on the supply lines and/or voltages on either the output or inputs that exceed the supplyvoltage. Once latch-up occurs, the current flow is limited only by the impedance of the power supply and theforward resistance of the parasitic thyristor and usually results in the destruction of the device. The chance oflatch-up occurring increases with increasing temperature and supply voltages.

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

33POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

APPLICATION INFORMATION

10 kΩ

10 kΩ

5 kΩ

VI

–

+

10 kΩ

10 kΩ

10 kΩ

0.016 µF

–

+

R = 5 kΩ(3/d-1) (see Note A)Band Pass

High Pass

Low Pass

–

+

0.016 µF

5 V

1/2TLC272

1/2TLC272

1/2TLC272

NOTE A: d = damping factor, 1/Q

Figure 44. State-Variable Filter

1/2TLC272

VI

12 V

H.P.5082-2835

0.5 µFMylar

VO

100 kΩ

–

+

–

+

1/2TLC272

N.O.Reset

Figure 45. Positive-Peak Detector

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

34 POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

APPLICATION INFORMATION

1/2TLC272

110 Ω

VO(see Note B)

–

+250 µF,25 V

10 kΩ

TIP31TL431

20 kΩ 1 kΩ

100 kΩ0.47 µF

15 Ω

TIS193

0.01 µF

47 kΩ

22 kΩ

0.1 µF

4.7 kΩ

1.2 kΩ

VI(see Note A)

–

+

NOTES: A. VI = 3.5 to 15 VB. VO = 2 V, 0 to 1 A

Figure 46. Logic-Array Power Supply

9 V

100 kΩ

47 kΩ

TLC2721/2

1/2TLC272

R2

VO (see Note A)

VO (see Note B)10 kΩ

10 kΩ

R3

0.1 µF

100 kΩ

9 V

R1

–

+

fO 1

4C(R2)R1R3

C

NOTES: A. VO(PP) = 8 VB. VO(PP) = 4 V

Figure 47. Single-Supply Function Generator

TLC272, TLC272A, TLC272B, TLC272Y, TLC277LinCMOS PRECISION DUAL OPERATIONAL AMPLIFIERS

SLOS091E – OCTOBER 1987 – REVISED FEBRUARY 2002

35POST OFFICE BOX 655303 • DALLAS, TEXAS 75265

APPLICATION INFORMATION

10 kΩ

1/2TLC277

(see Note A)R1,10 kΩ

VO

95 kΩ10 kΩ

–

+

–5 V

VI+

VI–

5 V

10 kΩ 100 kΩ

–

+

–

+1/2

TLC277

1/2TLC277

NOTE B: CMRR adjustment must be noninductive.

Figure 48. Low-Power Instrumentation Amplifier

R/25 MΩ

C270 pF

2C540 pF

R10 MΩ

VO

VI

5 V–

+

fNOTCH 1

2RC

C270 pF

R10 MΩ

1/2TLC272

Figure 49. Single-Supply Twin-T Notch Filter

IMPORTANT NOTICE

Texas Instruments Incorporated and its subsidiaries (TI) reserve the right to make corrections, modifications,enhancements, improvements, and other changes to its products and services at any time and to discontinueany product or service without notice. Customers should obtain the latest relevant information before placingorders and should verify that such information is current and complete. All products are sold subject to TI’s termsand conditions of sale supplied at the time of order acknowledgment.

TI warrants performance of its hardware products to the specifications applicable at the time of sale inaccordance with TI’s standard warranty. Testing and other quality control techniques are used to the extent TIdeems necessary to support this warranty. Except where mandated by government requirements, testing of allparameters of each product is not necessarily performed.

TI assumes no liability for applications assistance or customer product design. Customers are responsible fortheir products and applications using TI components. To minimize the risks associated with customer productsand applications, customers should provide adequate design and operating safeguards.

TI does not warrant or represent that any license, either express or implied, is granted under any TI patent right,copyright, mask work right, or other TI intellectual property right relating to any combination, machine, or processin which TI products or services are used. Information published by TI regarding third–party products or servicesdoes not constitute a license from TI to use such products or services or a warranty or endorsement thereof.Use of such information may require a license from a third party under the patents or other intellectual propertyof the third party, or a license from TI under the patents or other intellectual property of TI.

Reproduction of information in TI data books or data sheets is permissible only if reproduction is withoutalteration and is accompanied by all associated warranties, conditions, limitations, and notices. Reproductionof this information with alteration is an unfair and deceptive business practice. TI is not responsible or liable forsuch altered documentation.

Resale of TI products or services with statements different from or beyond the parameters stated by TI for thatproduct or service voids all express and any implied warranties for the associated TI product or service andis an unfair and deceptive business practice. TI is not responsible or liable for any such statements.

Mailing Address:

Texas InstrumentsPost Office Box 655303Dallas, Texas 75265

Copyright 2002, Texas Instruments Incorporated

This datasheet has been download from:

www.datasheetcatalog.com

Datasheets for electronics components.

B.5. Hoja de datos HC05

795

HC-05 Bluetooth module iteadstudio.com 06.18.2010

1 Tech Support: info@iteadstudio.com

HC-05

-Bluetooth to Serial Port Module

Overview

HC-05 module is an easy to use Bluetooth SPP (Serial Port Protocol) module, designed for

transparent wireless serial connection setup.

Serial port Bluetooth module is fully qualified Bluetooth V2.0+EDR (Enhanced Data Rate) 3Mbps

Modulation with complete 2.4GHz radio transceiver and baseband. It uses CSR Bluecore

04-External single chip Bluetooth system with CMOS technology and with AFH(Adaptive

Frequency Hopping Feature). It has the footprint as small as 12.7mmx27mm. Hope it will simplify

your overall design/development cycle.

Specifications

Hardware features

Typical -80dBm sensitivity

Up to +4dBm RF transmit power

Low Power 1.8V Operation ,1.8 to 3.6V I/O

PIO control

UART interface with programmable baud rate

With integrated antenna

With edge connector

HC-05 Bluetooth module iteadstudio.com 06.18.2010

2 Tech Support: info@iteadstudio.com

Software features

Default Baud rate: 38400, Data bits:8, Stop bit:1,Parity:No parity, Data control: has.

Supported baud rate: 9600,19200,38400,57600,115200,230400,460800.

Given a rising pulse in PIO0, device will be disconnected.

Status instruction port PIO1: low-disconnected, high-connected;

PIO10 and PIO11 can be connected to red and blue led separately. When master and slave

are paired, red and blue led blinks 1time/2s in interval, while disconnected only blue led

blinks 2times/s.

Auto-connect to the last device on power as default.

Permit pairing device to connect as default.

Auto-pairing PINCODE:”0000” as default

Auto-reconnect in 30 min when disconnected as a result of beyond the range of connection.

Hardware

HC-05 Bluetooth module iteadstudio.com 06.18.2010

3 Tech Support: info@iteadstudio.com

HC-05 Bluetooth module iteadstudio.com 06.18.2010

4 Tech Support: info@iteadstudio.com

HC-05 Bluetooth module iteadstudio.com 06.18.2010

5 Tech Support: info@iteadstudio.com

AT command Default:

How to set the mode to server (master):

1. Connect PIO11 to high level.

2. Power on, module into command state.

3. Using baud rate 38400, sent the “AT+ROLE=1\r\n” to module, with “OK\r\n”

means setting successes.

4. Connect the PIO11 to low level, repower the module, the module work as server

(master).

AT commands: (all end with \r\n)

1. Test command:

Command Respond Parameter

AT OK -

2. Reset

Command Respond Parameter

AT+RESET OK -

3. Get firmware version

Command Respond Parameter

AT+VERSION? +VERSION:<Param>

OK

Param : firmware version

Example:

AT+VERSION?\r\n

+VERSION:2.0-20100601

OK

HC-05 Bluetooth module iteadstudio.com 06.18.2010

6 Tech Support: info@iteadstudio.com

4. Restore default

Command Respond Parameter

AT+ORGL OK -

Default state:

Slave mode, pin code :1234, device name: H-C-2010-06-01 ,Baud 38400bits/s.

5. Get module address

Command Respond Parameter

AT+ADDR? +ADDR:<Param>

OK

Param: address of Bluetooth

module

Bluetooth address: NAP: UAP : LAP

Example:

AT+ADDR?\r\n

+ADDR:1234:56:abcdef

OK

6. Set/Check module name:

Command Respond Parameter

AT+NAME=<Param> OK Param: Bluetooth module

name

(Default :HC-05)

AT+NAME? +NAME:<Param>

OK (/FAIL)

Example:

AT+NAME=HC-05\r\n set the module name to “HC-05”

OK

AT+NAME=ITeadStudio\r\n

OK

AT+NAME?\r\n

+NAME: ITeadStudio

OK

7. Get the Bluetooth device name:

Command Respond Parameter

AT+RNAME?<Param1> 1. +NAME:<Param2>

OK

2. FAIL

Param1,Param 2 : the address

of Bluetooth device

Example: (Device address 00:02:72:od:22:24，name：ITead)

AT+RNAME? 0002，72，od2224\r\n

+RNAME:ITead

OK

8. Set/Check module mode:

Command Respond Parameter

AT+ROLE=<Param> OK Param:

0- Slave AT+ ROLE? +ROLE:<Param>

HC-05 Bluetooth module iteadstudio.com 06.18.2010

7 Tech Support: info@iteadstudio.com

OK 1-Master

2-Slave-Loop

9. Set/Check device class

Command Respond Parameter

AT+CLASS=<Param> OK Param: Device Class

AT+ CLASS? 1. +CLASS:<Param>

OK

2. FAIL

10. Set/Check GIAC (General Inquire Access Code)

Command Respond Parameter

AT+IAC=<Param> 1.OK

2. FAIL

Param: GIAC

(Default : 9e8b33)

AT+IAC +IAC:<Param>

OK

Example:

AT+IAC=9e8b3f\r\n

OK

AT+IAC?\r\n

+IAC: 9e8b3f

OK

11. Set/Check -- Query access patterns

Command Respond Parameter

AT+INQM=<Param>,<Param2>,

<Param3>

1.OK

2. FAIL

Param:

0——inquiry_mode_standard

1——inquiry_mode_rssi

Param2: Maximum number of

Bluetooth devices to respond

to

Param3:

Timeout (1-48 : 1.28s to

61.44s)

AT+ INQM? +INQM： <Param>,<Param2>,

<Param3>

OK

Example:

AT+INQM=1,9,48\r\n

OK

AT+INQM\r\n

+INQM:1, 9, 48

OK

HC-05 Bluetooth module iteadstudio.com 06.18.2010

8 Tech Support: info@iteadstudio.com

12. Set/Check PIN code:

Command Respond Parameter

AT+PSWD=<Param> OK Param: PIN code

(Default 1234)

AT+ PSWD? + PSWD ：<Param>

OK

13. Set/Check serial parameter:

Command Respond Parameter

AT+UART=<Param>,<Param2>,<

Param3>

OK Param1: Baud

Param2: Stop bit

Param3: Parity

AT+ UART? +UART=<Param>,<Param2>,

<Param3>

OK

Example:

AT+UART=115200，1,2,\r\n

OK

AT+UART?

+UART:115200,1,2

OK

14. Set/Check connect mode:

Command Respond Parameter

AT+CMODE=<Param> OK Param:

0 - connect fixed address

1 - connect any address

2 - slave-Loop

AT+ CMODE? + CMODE:<Param>

OK

15. Set/Check fixed address:

Command Respond Parameter

AT+BIND=<Param> OK Param: Fixed address

(Default

00:00:00:00:00:00)

AT+ BIND? + BIND:<Param>

OK

Example:

AT+BIND=1234，56，abcdef\r\n

OK

AT+BIND?\r\n

+BIND:1234:56:abcdef

OK

16. Set/Check LED I/O

Command Respond Parameter

AT+POLAR=<Param1,<Param2> OK Param1:

0- PIO8 low drive LED

1- PIO8 high drive LED

AT+ POLAR? + POLAR=<Param1>,<Param2>

OK

HC-05 Bluetooth module iteadstudio.com 06.18.2010

9 Tech Support: info@iteadstudio.com

Param2:

0- PIO9 low drive LED

1- PIO9 high drive LED

17. Set PIO output

Command Respond Parameter

AT+PIO=<Param1>,<Param2> OK Param1: PIO number

Param2: PIO level

0- low

1- high

Example:

1. PIO10 output high level

AT+PI0=10，1\r\n

OK

18. Set/Check – scan parameter

Command Respond Parameter

AT+IPSCAN=<Param1>,<Param2

>,<Param3>,<Param4>

OK Param1: Query time

interval

Param2：Query duration

Param3：Paging interval

Param4：Call duration

AT+IPSCAN? +IPSCAN:<Param1>,<Param2>,<P

aram3>,<Param4>

OK

Example:

AT+IPSCAN =1234,500,1200,250\r\n

OK

AT+IPSCAN?

+IPSCAN:1234,500,1200,250

19. Set/Check – SHIFF parameter

Command Respond Parameter

AT+SNIFF=<Param1>,<Param2>,

<Param3>,<Param4>

OK Param1: Max time

Param2: Min time

Param3: Retry time

Param4: Time out

AT+ SNIFF? +SNIFF:<Param1>,<Param2>,<Par

am3>,<Param4>

OK

20. Set/Check security mode

Command Respond Parameter

AT+SENM=<Param1>,<Param2> 1. OK

2. FAIL

Param1:

0——sec_mode0+off

1——sec_mode1+non_seAT+ SENM? + SENM:<Param1>,<Param2>

HC-05 Bluetooth module iteadstudio.com 06.18.2010

10 Tech Support: info@iteadstudio.com

OK cure

2——sec_mode2_service

3——sec_mode3_link

4——sec_mode_unknow

n

Param2:

0——hci_enc_mode_off

1——hci_enc_mode_pt_t

o_pt

2——hci_enc_mode_pt_t

o_pt_and_bcast

21. Delete Authenticated Device

Command Respond Parameter

AT+PMSAD=<Param> OK Param:

Authenticated Device

Address

Example:

AT+PMSAD =1234,56,abcdef\r\n

OK

22. Delete All Authenticated Device

Command Respond Parameter

AT+ RMAAD OK -

23. Search Authenticated Device

Command Respond Parameter

AT+FSAD=<Param> 1. OK

2. FAIL

Param: Device address

24. Get Authenticated Device Count

Command Respond Parameter

AT+ADCN? +ADCN：<Param>

OK

Param: Device Count

25. Most Recently Used Authenticated Device

Command Respond Parameter

AT+MRAD? + MRAD：<Param>

OK

Param: Recently

Authenticated Device

Address

26. Get the module working state

Command Respond Parameter

HC-05 Bluetooth module iteadstudio.com 06.18.2010

11 Tech Support: info@iteadstudio.com

AT+ STATE? + STATE：<Param>

OK

Param:

“INITIALIZED”

“READY”

“PAIRABLE”

“PAIRED”

“INQUIRING”

“CONNECTING”

“CONNECTED”

“DISCONNECTED”

“NUKNOW”

27. Initialize the SPP profile lib

Command Respond Parameter

AT+INIT 1. OK

2. FAIL

-

28. Inquiry Bluetooth Device

Command Respond Parameter

AT+INQ +INQ: <Param1>， <Param2>，

<Param3>

….

OK

Param1：Address

Param2：Device Class

Param3 ： RSSI Signal

strength

Example:

AT+INIT\r\n

OK

AT+IAC=9e8b33\r\n

OK

AT+CLASS=0\r\n

AT+INQM=1,9,48\r\n

At+INQ\r\n

+INQ:2:72:D2224,3E0104,FFBC

+INQ:1234:56:0,1F1F,FFC1

+INQ:1234:56:0,1F1F,FFC0

+INQ:1234:56:0,1F1F,FFC1

+INQ:2:72:D2224,3F0104,FFAD

+INQ:1234:56:0,1F1F,FFBE

+INQ:1234:56:0,1F1F,FFC2

+INQ:1234:56:0,1F1F,FFBE

+INQ:2:72:D2224,3F0104,FFBC

OK

28. Cancel Inquiring Bluetooth Device

Command Respond Parameter

AT+ INQC OK -

HC-05 Bluetooth module iteadstudio.com 06.18.2010

12 Tech Support: info@iteadstudio.com

29. Equipment Matching

Command Respond Parameter

AT+PAIR=<Param1>,<Param2> 1. OK

2. FAIL

Param1：Device Address

Param2：Time out

30. Connect Device

Command Respond Parameter

AT+LINK=<Param> 1. OK

2. FAIL

Param：Device Address

Example:

AT+FSAD=1234,56,abcdef\r\n

OK

AT+LINK=1234,56,abcdef\r\n

OK

31. Disconnect

Command Respond Parameter

AT+DISC 1. +DISC:SUCCESS

OK

2. +DISC:LINK_LOSS

OK

3. +DISC:NO_SLC

OK

4. +DISC:TIMEOUT

OK

5. +DISC:ERROR

OK

Param：Device Address

32. Energy-saving mode

Command Respond Parameter

AT+ENSNIFF=<Param> OK Param：Device Address

33. Exerts Energy-saving mode

Command Respond Parameter

AT+ EXSNIFF =<Param> OK Param：Device Address

HC-05 Bluetooth module iteadstudio.com 06.18.2010

13 Tech Support: info@iteadstudio.com

Revision History

Rev. Description Release date v1.0 Initial version 7/18/2010

B.6. Hoja de datos HC06

809

Electronic brick of HC-06 serial port Bluetooth iteadstudio.com 19th

, April, 2013

1 Tech Support: support@iteadstudio.com

Electronic Brick of HC-06 Serial Port

Bluetooth

Overview

What is an electronic brick? An electronic brick is an electronic module which can be assembled like Lego bricks

simply by plugging in and pulling out. Compared to traditional universal boards and circuit modules assembled with

various electronic components, electronic brick has standardized interfaces, plug and play, simplifying construction of

prototype circuit on one’s own. There are many types of electronic bricks, and we provide more than twenty types with

different functions including buttons, sensors, Bluetooth modules, etc, whose functions cover from sensor to motor drive,

from Ethernet to wireless communication via Bluetooth, and so on. We will continue to add more types to meet the various

needs of different projects.

Electronic brick of HC-06 serial port Bluetooth can be connected to hardware UART or analog UART on the control

board. With Bluetooth communication, it can achieve wireless transmission which can be applied in various kinds of remote

communication occasions.

Features

1. Plug and play, easy to use. Compatible with the mainstream 2.54 buckled interfaces in the market.

Electronic brick of HC-06 serial port Bluetooth iteadstudio.com 19th

, April, 2013

2 Tech Support: support@iteadstudio.com

2. With resetting button and status indicator

Specifications

PCB Size 39.5mm X 20.5mm X 1.6mm

Working voltage 3.3V DC

Operating voltage 3.3V DC

Compatible interfaces 2.544-pin buckled interface(1)

Communication protocols Bluetooth 2.0

Communication range <20m

Note 1: DO for data output (TX) port of UART, DI for data input (RX) port of UART, V and G for voltage at the common

collector and ground respectively.

Electronic brick of HC-06 serial port Bluetooth iteadstudio.com 19th

, April, 2013

3 Tech Support: support@iteadstudio.com

Electrical characteristics

Parameter Min. Typical Max. Unit

Working voltage 3.0 3.3 4.2 VDC

Digital output voltage（VCC=3.3V） 0 - 3.3 V

Working current（VCC=3.3V） - 20 40 mA

Switch and Indicator

1. Resetting button

After pressing down resetting button, module will re-enter standby mode and the status lamp will flash.

2. Status lamp

When the module is under standby mode, status lamp will keep flashing; when the module is under connection mode,

status lamp will keep being ON.

AT Command

Default：

Slave，9600 bit/s baud rate, Pincode 1234。

AT command：

1. Communications test：

Sent： AT

Receive： OK

2. Change baud rate

Sent： AT+BAUD1

Receive： OK1200

Sent： AT+BAUD2

Receive： OK2400

1---------1200

2---------2400

3---------4800

4---------9600

5---------19200

6---------38400

7---------57600

Electronic brick of HC-06 serial port Bluetooth iteadstudio.com 19th

, April, 2013

4 Tech Support: support@iteadstudio.com

8---------115200

Baud rate setting can be saved even if power is off.

3. Change name of Bluetooth device

Sent： AT+NAMEdevicename

Receive： OKname

（devicename is the name you want it to be, which will be shown when searched）

Name setting can be saved even if power is off.

4. Change Pincode

Sent： AT+PINxxxx

Receive： OKsetpin

（xxxx is the pin code that you want to set）

Pin code can be saved even if power is off.

DEMO

Connect D0 port of electronic brick of serial port Bluetooth to D0 port of Arduino board, D1 port to D1 port of Arduino

board, and we will use the following program to keep sending character string of HELLO and have them displayed via

Bluetooth serial port software of cellphone.

void setup()

Serial.begin(9600);

void loop()

Serial.println("HELLO"); //Send the "HELLO" character string to HC-06

delay(1000); // wait for a second

Revision record

Version Description Date Written by

v1.0 Initial edition 19th, April, 2013 Stan Lee