API Standard 622

EMISSION CONTROL

www.valve-world.net June 2009 1

Experiences performingAPI standard 622Type testing of process valve packing for fugitive emissions

The publication of the API622 specification has markeda departure from the commonmethod of evaluating packingperformance as a part of valvequalification tests. Instead, API622 evaluates packingperformance in a standardizedfixture, outside the influenceof a particular valve’s design.This article will outline the useof API 622 in tests performedby Yarmouth Research andTechnology. The generalexperiences performing API622 emissions tests werepositive ones. This paperdescribes the test procedures,our results in performing thetests on a braided flexiblegraphite packing, and thelessons that were learned.

By Chris Boss and James Drago, P.E. from GarlockSealing Technologies and Matt Wasielewski, P.E.Yarmouth Research and Technology

Description of Test ProceduresSeveral important distinctions set API 622apart from other packing performancetests.The scope of the specificationhighlights several of these unique points:• API-622: specifies requirements forcomparative testing of block valve stempacking for process applications wherefugitive emissions are a consideration

• API-622: is not simply a performancetest, but rather a range of tests thatevaluate many factors that aredetermined to affect fugitive emissionsperformance.These include:temperature, pressure, thermal cycling,

mechanical cycling, corrosion, andmaterial makeup

• API-622: does not provide acceptancecriteria for valve packing, nor is itintended to replace proof testing ofvalve assemblies or valve productiontesting

The specification consists of three parts:fugitive emissions testing, corrosion testing,and materials testing. API-622 providesvery detailed specifications, descriptions,and guidelines for both the test fixturesand procedures.The following is a verybrief summary of these tests.

1. Fugitive emissions testingThe fugitive emissions portion of the testis performed on a test fixture thatsimulates valve packing gland conditions.The fixture uses 97 % pure, dry methaneas the detectable, pressurizing fluid. Overthe course of the three day test, thepacking gland is subjected to threethermal cycles from ambient to 500 deg F,three pressure cycles from 0 psig to 600psig, and 1500 actuation cycles. Each day,250 actuation cycles are performed atambient temperature and 250 cycles areperformed at 500 deg F. Methane leakageis monitored by a flame ionization vaporanalyzer. Leakage is collected from aroundthe stem/packing interface by way of aspecially designed gland follower to whichthe detector is connected. It should benoted that none of the leakage from thepacking OD is collected.

2. Corrosion testingThe corrosion testing portion of thespecification involves two different tests.One test is performed at ambienttemperature, and the other at 300 deg F.

• Ambient temperature corrosion testingIn the ambient test, a packing sample issoaked in demineralized water and theninserted into a small fixture thatcompresses the packing sample around a

metal specimen.These fixtures are storedin an enclosure that is equipped with ashelf, suspending the fixtures over a waterbath, and ensuring a moist environmentfor the 28 day duration of the test.Following the 28 day exposure period themetal samples are photographed. Afterthey undergo an x-ray corrosion analysis,the amount and severity of corrosion isquantified in terms of percentage ofsurface corroded, and mean andmaximum pitting depth.

• High temperature corrosion testingThe high temperature corrosion test usesa fixture similar to the ambient corrosiontest, except that the packing and metalsamples are subjected to demineralizedwater at a temperature of 300 deg F and650 psig for 35 days.The same post-exposure analysis and measurements areperformed.

3. Packing Materials Testing• Weight Loss at Elevated temperaturePacking samples are subjected toprogressively elevated temperatures (inair) at 300, 500, 600, 700, 800, 900, and1000 deg F for one hour. After eachexposure, the sample is removed from theoven and any resulting loss of mass ismeasured.

• DensitySimple density measurements areperformed on a packing sample bydividing the mass of the sample by itsvolume.

• Lubricant contentTwo tests are performed to determinelubricant content. First, the PTFE contentin the packing is determined byperforming a total Fluorine contentmeasurement per ASTM D129 andD1179.The PTFE content can be derivedfrom the total Fluorine content.The second test is a solvent extraction

EMISSION CONTROL

www.valve-world.net2 June 2009

test used to measure wet lubricantcontent in the sample.The procedure isoutlined in the specification.

• LeachablesTwo ASTM standards are referenced inAPI-622 to be used in measuringleachable Chloride and Fluoride content;ASTM D512 and ASTM D1179,respectively.

Lessons Learned

What was difficult?For the emissions portion of thespecification the primary difficulty is thedaily duration of the test.With a cyclingrate of 1 cycle per 60 seconds and a 2hour heat-up time, an average test day isabout 12 hours. If another more massivetest rig were to be created, the heat uptime would be even greater.For theWeight Loss portion of thematerials testing, it is important to insurethat the oven being used to heat thesamples will provide an accurate and eventemperature distribution throughout.Testing of several samples isrecommended in order to insure that hotor cold spots in the oven do not affectthe test results.

What went well?The emissions test fixture is well designedwith the leak probe at a fixed location.This provides repeatable results. Actuatingthe stem by means of a hydraulic cylinderprovides fairly smooth motion.The flowrate of hydraulic fluid needs to becontrolled so that actuation speed is keptwithin the required limits.The emissions test produced results thatare consistent in magnitude and in patternwith other emissions test proceduresperformed on the same materials. In manyemissions test procedures, using a widevariety of materials, it is common to see aleakage rate increase following the firstthermal cycle.The same behavior wasnoted in this procedure. Following a single,allowable packing adjustment, leakagecontrol was sustained at very low levelsfor the remainder of the test.This samebehavior has been noted in otheremissions test procedures with thismaterial.The tests for Density, Lubricant Content,and Leachables are very straightforward

tests with well defined procedures.

RecommendationsBased on the experience with thespecification there are several changes tothe test standard that we would like tosubmit for consideration.• The addition of a final leak

measurement point at 1500 cycles aftercooling to ambient temperature isrecommend.This will complete readingsfor the 3rd thermal cycle.

• Further it is recommended that theprimary and secondary thermocouplepositions be reversed.There are welldefined requirements for the

Test Parameter Results of test 1 Results of test 2

Average Test Pressure 602 psi (41.5 bar) 600 psi (41.4 bar)

Number of Actuation Cycles Completed 1500 1500

Number of Thermal Cycles 3 3

Number of Packing Adjustments RequiredCycle at which adjustment occurred

1 adjustmentat cycle 600at ambienttemperature

1 adjustmentat cycle 650at ambienttemperature

Average Leakage Throughout Test 58 ppmv 41 ppmv

Maximum Leakage Throughout 877 ppmv 598 ppmv

ResultsEmissions test results

Results observed were consistent for the two tests run, indicating repeatability of the method.

TemperatureCondition andMagnification

Metal not in contactwith packing

Metal in contact withpacking

Ambient100X magnification

Ambient200X magnification

300°F (149°C)100X magnification

300°F (149°C)200X magnification

Corrosion test results

No corrosion effects were noted on the metal samples in either the ambient or the high temperature corrosion tests.There were no notable visual differences between the surfaces in contact or not in contact with the packing. Because ofthis fact, quantification of corrosion products utilizing X-Ray diffraction was not performed.

EMISSION CONTROL

www.valve-world.net June 2009 3

temperature and tolerance for theprimary thermocouple, where thereference thermocouple has no suchrequirements. The current version ofthe standard has the primarythermocouple position about 12 inchesaway from the packing location, whereasthe reference probe is about 1 inchfrom the bottom of the packing.Withthe current configuration, thetemperature in the packing area canvary by 20 deg C or more dependingupon heater and insulation position andamount.This change would result inbetter control of the temperature towhich the packing is exposed. (SeeFigure bottom right)

• Although there is not a clearrecommendation, the current versionrequires a long test day for laboratorypersonnel, with a typical test runningabout 12-14 hours. Readings arerequired about every hour. Possiblemethods for reducing the daily durationof the test would include increasing thespeed of actuation or allowingactuations to continue during thetemperature increase from ambient to500 deg F

• The current standard does not measureactuation force.While the test fixturecan handle high loads, excessive packingfriction can cause operating problems invalves. Opening and closing forces canbe easily calculated from the actuatorpressure

• If the test standard is to be performedon various sized fixtures, the standardshould note that variations in thetemperature rise rates are allowed. Also,larger fixtures might require 2 days foreach thermal cycle

• The metal test specimens that are usedin the corrosion tests do not have a welldefined material specification. API-622simply states: “sample steel rings shall bemachined from metal rods having thesame properties as the finished valvestem”.While this does allow the user totailor the test to their ownrequirements, it does not ensure thatresults from various tests will be directlycomparable. In our test, 410 StainlessSteel was chosen for the metal samples

• The corrosion test specification calls forphotos to be taken at both 100X and200 X magnifications. It is recommendedthat a reduction in the magnification

level to 50X will help to provide imagesthat will show the overall area moreclearly, while still being able todifferentiate corroded areas. It is felt thatthe 600X magnification is much too high

• It is recommended that photos of thecorrosion test metal specimens be takenboth before and after the test and thatthe after photos should include areas ofthe metal specimen that were both incontact with the packing and not incontact with the packing

ConclusionsThe tests prescribed by API-622 providemethods for the end user to evaluatevalve stem packing in three key areas;emissions performance, corrosionbehavior, and general physical and

chemical attributes.When combined withfire safe test specification such as API-589,a very thorough assessment of packingmaterials can be achieved.The results can be used to effectivelycompare the relative performance ofpacking materials and configurations aswell as assess their strengths andweaknesses in the areas of chemical purityand physical properties.While thespecification itself does not impose anypass/fail criteria, it does provide aconsistent means of quantitativemeasurement.The user is free todetermine what measured variables holdthe most significance for their applicationand whether they will set their ownminimum requirements.

Cross sectional view of the API 622 Emissions test fixture.Thermocouple locations are noted as part 11(by permission of The American Petroleum Institute)

High Temperature Weightloss Results