Rak-43.3415 Building Physics Design 2ACOUSTICAL DESIGN

Autumn 2015

LECTURE 4Room acoustics

Matias Remes, M.Sc. FISE A acoustics

Room acoustics

”Huoneen hyvällä akustiikalla tarkoitetaan sellaisia äänisuhteitten ominaisuuksia, että huoneessa esitetty puhe ja musiikki kuuluu korvaan kauniina, luonnollisena ja selvänä

huoneen jokaisessa kohdassa.”

Diplomi-insinööri U. Varjo 1938

What is the significance of roomacoustics?

Significance of room acoustics

• The purpose of room acoustical design is to control the propagation, reflection and attenuation of sound within a space– Direct sound– Reverberant sound (reflections)– Useful and harmful reflections– Sound attenuation and absorption, diffusion

• Design goals according to the use of space, for example:– Speech Good speech distinction (e.g. auditorium) / good speech privacy

(e.g. open plan office)– Music Appropriate reverberation and sense of space in the audience,

stage acoustics which support music making

Significance of room acoustics

• Room acoustical design– Design of sound reflections– Design of sound absorption– Design of the shape and geometry of the space

• Room acoustical design ≠ maximising the amount of sound absorbing material– E.g. in a lecture hall the performer must be able to speak without

restarining ones voice and so that the audience can distinguish what is being said

Need for both sound absorbing and reflecting surfaces!• Succesful room acoustics is, thus, a combination of the

geometry of the space and the absorptive and reflective properties of materials

Significance of room acousticsExamples of design goals• Movie theater

– Hearing the sound track in the way the movie makers have intended it to be heard

• Concert hall– Good spatial impression (sound surrounds the listener), sense of intimacy,

”warm” sound color, adequate clarity, etc.• Restaurant

– Peacefull acoustical environment (communication from short distance)• Open plan office

– Speech sound distract concentration speech privacy between work places

• Factory– Noise level may cause hearing damage design of effective sound

absorption and noise blocking screens

Reflection of sound

• Sound reflects at simplest as light: angle of incidence = angle of reflection (specular reflection), this applies when– Sound wavelength is adequately

smaller than the dimensions of the object causing the reflection

– The reflecting surface is even (not sound scattering) and hard (not sound absorbing)

• Sound reflection is complicated phenomenon and depends on frequency and the properties of the reflective surface

Reflection of soundSignificance of reflections• Example: sound suddenly stops in a large concert hall

– Number of reflections occuring within the first 1 s is about 8000• As the number of reflections increases, there is a

reverberant sound field in the space in which the listener cannot distinguish single reflections from one another; in large spaces this occurs after about 100 ms

• Sound field in a space comprises of three distinguishable parts:– Direct sound– Early reflections– Reverberant sound field

Sound field in a room

1

1

1

3

1

1

1

3

1. Direct sound from source to listener

2. Early reflections within 20...50 ms after direct sound

3. Gradually attenuating reverberant sound field

Refelection of sound

Sound field in a room

Kuvat: Rossing et al . 2002, Beranek 2004

Direct sound, early reflections and reverberant sound and their relations determine how sound is perceived in a space

Reflection of soundSignificance of reflections• Sound perception is affected by the

level of reflections, their delay in relation to direct sound and the direction from which they reach the listener

• Strong reflections with adequate delay are heard as separate echoes (disturbance)

• If the delay between direct sound and early reflections is appropriate (about 50...80 ms), the reflections increase the loudness of sound (perceived sound level) important in the design of speech and music spaces

• Lateral reflections (reaching the listener`s ears from the sides) add to the sense spatial impressionand broadening of the sound source crucial in the design of concert halls

The effect of single lateral reflection to sound perception (Barron 2003)

Reflection of sound The effect of basic room geometry

Fan-shape:Reflections scatter and are directed mainly to the rear part of the space (not in the middle)

Rectangle:Lateral reflections occur in the entire space

Round:Reflections from concave surfaces cause sound to strongly focus on some parts of the space

Reflection of sound Example: sound focusing

1

3

4

5

6

7

8

9

1011

12

13

12

Reflection of sound Effect of surface geometry

Reflection of sound Example: reflecting surfaces in a concert hall

Reverberation timeSignificance of reverberation time• The reverberation time in a space correlates rather well

with the perceived clarity of speech or music: long reverberation the syllables in speech or separate musical notes attenuate slowly and mask each other

Reverberation timeSignificance of reverberation time• Too short a reverberation time is not desirable because in an

overly damped space there are no useful reflections!• In addition to appropriate reverberation time, good room

acoustics provides that– The space has appropriate size and shape– Sound absorbing and reflecting surfaces are positioned correctly

• Two viewpoints: room acoustics experienced by the audienceand by the performer– For the audience, it is important to have useful sound reflections

form the performer to the audience– For the performer, it is important that the stage acoustics supports

the performer`s activity

Reverberation time vs. use

Reverberation time vs. use

Tila V / hlö [m3]

Kokoustila 3…5

Auditorio, teatteri 4…6

Musiikkiteatteri, ooppera 5…8

Kamarimusiikkisali 6…10

Konserttisali 8…12

Kirkko 10…14

Reverberation timeSabine equation• Sabine equation:

• Sabine equation assumes that the sound field in the room is diffuse, i.e., at any point in the room sound can arrive from any direction and the field remains the same throughout the room

• This is an idealisation that does not hold perfectly true in real rooms

AVRT 161,060

i

n

1iinn2211 ... SSSSA

Reverberation timeNotes on Sabine equation• Sabine equation can be used with good accuracy in rooms which

are sufficiently reverberant– Sabine equation is most accurate in a reverberant room where the average

absorption coefficient is < 0,25 1)

– In very absorbent rooms the Sabine equation gives erroneous results• Additional requirements for Sabine equation to yield accurate

results:– The room geometry should be simple (cube-like) and the room should be

quite small– The absorption material should be evenly distributed on the room surfaces– Calculation error increases in large and complex spaces

• In rooms where all the absorption material is positioned only on one surface, Sabine equation yields shorter reverberation time than is the case in practice

1) F.A. Everest, K.C. Pohlman, Master Handbook of Acoustics, 2009

Reverberation timeEyring-Norris equation• Eyring-Norris equation:

where V is room volume, S is the total surface area of the room and αaverage is the average absorption coefficient:

averageSVRT

1ln

161,060

i

iiaverage S

S

Reverberation timeNotes on Eyring-Norris equation• Eyring-Norris equation can be used in more absorptive rooms

where average absorption coefficient is > 0,25, e.g., studios• However, sound absorption coefficients that are commonly

available and published by material manufacturers are Sabine coefficients (measured in a reverberation chamber and calculated using Sabine equation) and can, thus, be directly applied only to the Sabine equation

• For this reason, Sabine equation is the usual choice in acoustical design and is also used on this course

• Other researchers have suggested alternative reverberation formulas, e.g., Fitzroy, Millington, Hopkins-Striker...

Reverberation timeAir absorption• Taking account of air absorption, the Sabine and Eyring-

Norris equations can be written as:

mVAVRT4

161,060 mVS

VRTaverage 41ln

161,060

Sabine: Eyring-Norris:

where m is the air attenuation coefficient (some values: m = 0,009 at 2 kHz; m = 0,025 at 4 kHz; m = 0,080 at 8 kHz)

• Air attenuation– is only significant in large spaces above 2 kHz– depends on relative humidity of air, absorption increases at low

humidity

Air absorption

T = 20 °C, RH = 50 %

Room modes

• Sound field within a room is comprised of room resonances, called room modes

• Room mode = characteristic resonance of the room

• Three types of modes: axial, tangential, oblique• The amount and spacing of room modes changes

with frequency• At low frequencies there are only a few room modes

and the modes are sparsely spaced, as a result of which the reverberation time and sound level can vary considerably in different points in the room (consider the placement of a subwoofer in a living room)

• At high frequencies the number of room modes gets so high and their frequencies are so close to one another, that single room modes cannot be distinguished sound field approaches the idealisation of diffusivity

Room modes

• The frequency, below which the sound field in a room is not diffuse (so-called Schröder frequency) depends on reverberation time and volume:

• Example, typical dewlling room: T = 0,5 s ja V = 30 m3 fs = 260 Hz• The room modes of a rectangular room can be calculated based on the

dimensions of the room (Lx, Ly, Lz):

where l, m, n are integers

VTfs 2000

222

2 ZYXlmn L

nLm

Llcf

Room modes

• The number of modes (modedensity) increases with increasing frequency, at lowfrequencies mode density is low

• From room acoustics point of view, room modes are the moreproblematic the smaller the sizeof the room

• Room modes must beconsidered in the design of, e.g.– Control rooms– Recording studios

Room modes

• Standard deviations of reverberation times measured in 50 empty rooms in a dwelling

• Each point represents the reverberation time calculated from measurement conducted in 12 points in a room

• In small rooms sound field is not diffuse at low frequencies, thusthe variance in reverberation timeincreases towads lowfrequencies

Absorption vs. reflection

Sound absorbing materialsGeneral• Sound absorption –

three absorption mechanisms:– Porous materials (P)– Resonant absorbers (R)– Membrane / panel

absorbers (M)

• Typical absorption behaviour in the figure

Sound absorbing materialsPorous materials – effect of placement• Sound absorption of porous

materials is based on thermal losses caused by friction in the pores of the material

• At the surface of a rigidstructure (e.g. wall, roof) sound pressure is at maximum and particlevelocity at minimum

• Maximal particle velocityoccurs at 1/4λ distance fromthe surface of the structure to achieve effectiveabsorption, there should beabsorbing material at thisdistance

Sound absorbing materialsPorous materials

/4

/4

• Porous material absorbs sound most effectively when the thickness of the material is at least four times the sound wavelength: d ≥ 1/4λ

• Example:mineral wool 20 mm: 20 mm ≥ 1/4λ λ ≤ 80 mm f ≥ 4290 Hz

• Note: low-frequency sounds have long wavelength (e.g. 100 Hz 3,4 m) thin layers of porous material do not much absorb low frequencies!

Sound absorbing materialsPorous materials• Absorption

coefficients of mineral wool

• Note the effect of suspension height

Sound absorbing materialsPorous materials – examples of use

Sound absorbing materialsPorous materials – examples of use

Sound absorbing materialsPerforated panels (resonant absorbers)• Perforated panels act as resonant absorbers; the absorption is

based on mass-spring resonance caused by the air in the hole acting as mass and the air in the background airspace acting as spring

• Absorption is most effective at the resonance frequency of the mass-spring system

• Resonance frequency and absorption coefficient are affected by:– Thickness of the airspace and filling with porous material– Size, amount and geometry of holes– Thickness of the panel

• Resonance frequency is typically at mid frequencies (500-1000 Hz), below and above which absorption coefficient decreases

• Absorption coefficient of perforated panel absorbers can beincreased by filling the background airspace with porous absorption material

Sound absorbing materialsPerforated panels

0,0

0,2

0,4

0,6

0,8

1,0

125

250

500

1000

2000

4000

Keskitaajuus [Hz]

Abso

rptio

suhd

e

Reikäala 7 %, ilmaväli 30 mm

Reikäala 17 %, ilmaväli 30 mm

Reikäala 17 %, ilmaväli 200 mm

Reikäala 17 %, ilmaväli 200 mm,jossa 50 mm mineraalivilla

Sound absorbing materialsPanel absorbers• Structure of a panel (or membrane) absorber: a closed

airspace behind an impervious (not perforated) panelwhich can be filled with porous material

• Absorption coefficient is highest at low frequencies around the resonance frequency

• Resonance frequency depends on the surface mass of the panel and thickness of the airspace:

• Note: at high and mid frequencies panel absorbers aresound reflecting structures!

dmf

`60

0

Sound absorbing materialsPanel absorbers

0,0

0,2

0,4

0,6

0,8

1,0

125

250

500

1000

2000

4000

Keskitaajuus [Hz]

Abso

rptio

suhd

e

Kipsilevy 13 mm, mineraalivilla 50mm

Kipsilevy 13 mm, mineraalivilla 100mm

Kipsilevy 13 mm, ilmaväli 100 mm

2 x kipsilevy 13 mm, mineraalivilla50 mm

Sound absorbing materialsMiscellaneous materials and structures

Hard surfaces

0,0

0,2

0,4

0,6

0,8

1,0

125

250

500

1000

2000

4000

Keskitaajuus [Hz]

Abso

rptio

suhd

e

Lakattu puu 69 mm

Maalattu betoni

Muovimatolla päällystetty betoni

Rapattu tiili

Chairs and audience

0,0

0,2

0,4

0,6

0,8

1,0

125

250

500

1000

2000

4000

Keskitaajuus [Hz]

Abso

rptio

suhd

e

Puu- tai muovituolit

Pehmustetut istuimet

Yleisö pehmustetuilla istuimilla

Air absorpionEffect on reverberation time

0,0

0,5

1,0

1,5

2,0

2,5

125

250

500

1000

2000

4000

Keskitaajuus [Hz]

Jälk

ikai

unta

-aik

a T

[s]

• In large spacesreverberation timedecreases at highfrequencies because of air absorption

• Example curve: Savonlinna hall

Effect of vapour barrier on absorption

Effect of vapour barrier on absorption

Classification of absorption materials

• According to EN 11654 (classes A-E)

• Measured absorption coefficient is compared to a reference curve, the sum of unfavourabledeviations ≤ 0,10

• Note: definition of absorption class does not consider frequencies below 200 Hz!

Classification of absorption materials

Sound diffusing structures (diffusors)

[Salter et al. 1999]

Concert hallsSubjective and objective parameters• The characteristics of sound perceived in a hall can be

describes subjectively (e.g. ”warm sound”, ”three dimensional sound”)– Difficulty: how to define accurately what different subjective

descriptions mean– Common terminology to be used by acousticians and musicians

has been developed (”subjective parameters”) which define the different sound characteristics (e.g. Beranek)

• Objective parameters are measurable quantities whichtry to describe the subjective perceptions as well as possible

Concert hallsObjective parameters• Reverberance (Kaiuntaisuus)

– Reverberation time RT60– Early decay time (Varhainen jälkikaiunta-aika) EDT, has been found to correlate better with

subjective impression than RT60)• Clarity (Selvyys)

– Clarity C80 (describes how much of the sound arrives at the listener 0...80 ms after directsound)

– Definition D50• Spatial impression (tilan tuntu)

– Lateral Energy Fraction LF (describes hthe amount of reflections reaching the listener fromthe sides, laterally)

• Loudness or strength of sound (Äänen voimakkuus)– Strength G

• Warmth, brilliance (Äänen lämpimyys ja kirkkaus)– Bass Ratio– Treble Ratio

• Stage support (lavatuki)– Support ST1 (correlates with how the performers are able to hear each other on stage)

Concert hallsConnections between parameters

Subjektiivinen ominaisuus Objektiivinen mittaluku

Concert hallsSound field in a concert hall• Acoustics experienced by the audience

– Ideally the audience should be able to hear the performace in the same way in every point of the hall (nod acoustically badseats)

– The balance between different players / singers should be goodand not dependent on the listening position

• Acoustics experienced by the performer(s)– Performers should be able to hear each other– Room acoustics shoud act as a continuation of the performers

instrument: the performer should be able to hear the acousticsof the space properly and the acoustics should support the performers efforts

Concert hallsDesign issues• Volume and basic shape

– What is performed?• Rock / classical?• Full-sized syphony orchestra / chamber music?• Music theatre / opera?• Multi-functionality?

– Seating capacity (audience, performers)

• Also other than room acoustical issues must be considered!– Sound insulation to surrounding spaces– Noise control

• HVAC• Theater equipment, lighting etc.

Concert hallsExample: Wien, Musikverein (1870)

Concert hallsExample: Berlin Filharmonie (1963)

Concert hallsExample: Finlandiasali (1971)

Concert hallsExample: Sibeliussali (2000)

Auditoria

• Most important room acoustical consideration: speechintelligibility

• Essential to good speech intelligibility:– Direct sound path from speaker to listener (other audience

members must not be in the way to block the direct sound) raised floor on the audience area

– Directioning of early reflections towards the audience (to strengthen the direct sound and, thus, improve speechintelligibility) positioning of sound reflecting and absorbing surfaces

– Appropriate reverberation time (remember the masking effectfrom previous slides…)

AuditoriaObjective parameters• Objective parameters describing speech intelligibility:

– Speech Transmission Index STI– Rapid Speech Transmission Index RASTI

• STI depends on the sound level of speech, reverberationtime, reflections, distance between speaker and listener and background noise level

• STI is based on distinction of speech syllables in a space, value range 0...1– STI = 0 none of the syllables are heard correctly– STI = 1 all the syllables are heard correctly

• In auditoria and other speech spaces the goal is to achieve a high value of STI (good speech intelligibility), whereas in spaces such as open plan offices the goal is as low STI-valueas possible (good speech privacy between work stations)

AuditoriaSignificance of STI

AuditoriaExample of positioning surfaces• A surface above the speaker which

reflects sound to the front and rearparts of the audience (1)

• Sound reflecting surfaces to the midand rear parts of the audience sothat the entire audience areareceives reflections (2-5)

• Front surfaces of the hall sound reflecting (non-parallel walls on stage to avoid flutter echoes), lowerparts of the walls sound reflecting to achieve useful reflections from the sides as well as from the ceiling (6)

• Absorbing surfaces to the upperparts of side walls and rear part of the ceiling, rear wall sound absorbing or scattering to avoiddistracting reflections back to the front part of the hall and the stage(6)

1) 2)

3) 4)

5)

Heijastavaapintaa

Absorboivaa pintaaA

6)

ClassroomsOptimising speech intelligibility (SFS 5907)• In classrooms, it is important that speech

distinction from the treacher to pupils is good• Recommended reveberation time 0,5...0,8 s• When class A sound absorbing material is

used (absorption coefficient > 0,9), the centre part of the ceiling should be left sound reflecting (absorption coefficient 0...0.20), from which sound is reflected to the mid and rear part of the room this improves speech intelligibility

• If class C material is used, the material can be positioned on the whole ceiling surface because the material also reflects sound

• The amount of class A material needed is about 70 % of the floor area and for class C material correspondingly 100 %

• Absorption material should also be placed on the walls to avoid distracting flutter echoes

Screens

• Wall elements used for dividing space, do not extend to ceiling height (as opposed to walls), thus sound is diffracted over the screen

• Screens have many noise control applications: screens for blocking speech in open plan offices, noise blocking screens in factories, road noise barriers

• The sound attenuation of a screen, D [dB], is determined as insertion loss: D = Lp,1 – Lp,2where Lp,1 is the measured sound pressure level at listener position without the screen and Lp,2 is the SPL with the screen in place

Screens

• In the free-field (a space with no sound reflections, e.g. outdoors), the sound attenuation by screen of a point sound source is given by the equation

where z is the distance difference and λ is sound wavelength (λ = c0/f)

zD 201log10 10

Screens

• The effect of screen height Hon sound attenuation when x1= x2 = h1 = h2 = 1,2 m, calculated with equation on previous slide

• The mass of the screen is assumed to be infinitely large so the sound insulation of screen is insignificant

• Solid lines correspond to sound reduction index (mass law values) of two different walls (surface masses 5 kg/m2 and 20 kg/m2)

• Note: the sound attenuation of a solid wall is always significantly higher than that of a screen

ScreensEffect of sound reflections on attenuation• The sound attenuation values on the previous slide

assume a space with no sound reflections; the values can be achieved by, e.g., outdoor noise barriers

• In practical rooms the sound attenuation of screens is even considerably lower because of reflections; screens mainly attenuate the direct sound from source to listener but have little impact on reverberant sound

• Attenuation is especially affected by sound absorption of the ceiling and wall surfaces close to the screen

ScreensEffect of sound reflections on attenuation• Example: factory hall with

screen constructed around a work station with noisy activity

• Although the screen height was 6 m, the screen only attenuated sound by 1dB(A) to its surroundings

• Reasons:• Ceiling and wall surfaces

were sound reflecting, reverberation time of the hall was high

• Screen height was only about 1/3 of the room height

Open plan offices

[Haapakangas et al. 2007]

• Acoustics is a key component of indoor climate in open plan offices

• The most distracting sound is speechheard from surrounding work stations

• Acoustical design goal is to maximisespeech privacy between work stations, thus minimise speech intelligibility(note: opposite goal to e.g. classrooms)

• In open plan offices there are no sound insulating partitions, but only screens speech privacy must be mainlyachieved by means of room acoustics

Open plan offices

• Good speech privacy between work stations requiresthat three factors are considered:1. Direct sound path between work stations must be truncated

using screens, preferrable scren height is > 150 cm2. Sounds reflecting from room surfaces must be adequately

attenuated sufficient amount of sound absorbing materialcorrectly positioned on the surfaces of the room (ceiling, floor, walls, screens, furnishings)

3. Masking sound sufficiently high background sound level in the space to mask speech sounds, preferrably implementedartificially using loudspeakers to produce the masking sound, recommende sound level 40...45 dBA

Studios

Reflection-free zone at the mixer`sposition

Studios

Dolbysurround 7.1 speakerlayout

Studios

Requirements for early reflectionsand reverberationtime

[2005 Dolby 5.1-channel music production guidelines]