!
! " ! # $ ! # %& ' ( )*
+
,---, . #/ 0 & 1 2 233
+ , # 4565
+ ,---..
, . #7 1 #/ 0 0/#
0 3 0 1 + 7#
! " ! # $ ! # %& ' ( )
3 $ ! #
2 ( + # #
i
! " " # $% " "
! $% & " " $ ' $%
( ) ) ( ** + ,)) (
-* , ./
0 , 122222222222
, 1 #$
) - 6 7) , * #, 8 ( 6 * - 0
, 6 ) ,9 $ ) 6 * 6 ) 6
, $ ) , ) 6 , 6 $ , ,** * 0, 0
iii
* 6 ( , 8 + ) 8*, ) 0 *
0: ) 0 ) ) , , ) $
0 0 , - * $ $ ; $< *, *
9, ) * ) 0 - * , * $ .
$ + ( - *, ,00 )
-. ( 8 ) , - 6 $ . $ .
, ,** ( 0, ) ,0 , ( , . 6
( , 8 8 ) * ( 0 - ) * 0 - 0
-0 - ) .
( , 8 ) ( ) , ) , )
( , ) )) ,00 . , 6
) - * ) * , ) , ) 0 .
& , - 6 ) 0 ) . 6 8
" * = * ; < * ( ,
) * * ) , *, * ) ,
) * ( 8 * ) . ) ( ) (
* ; < ( *
; < . - 0 * , - 0 6 - 0
0 0 ) * 6 ) - ) *
, > ,) , 6 ), , 6 8
8 ), , 6 , ) 8 7 * *
) 6 , , ) 8 )
, . , ) , ) - ) .
)) * ( * * . - 0
( * ( , ) * +
v
% 8 ) , , " * ? * ; <
0, 8 , ,8 ) ,8 08 , , 8 ) 8
) 8 ,) 0 ) * ) 8 ) 00, 0 , .
% 0 8,8 0 * 0 .
@ ), 0, 8 6 ,0 ) , 8 0 , , ,8
) 0 8 8 ), , ) ) 8 6 ,6 ) ) , 0
0 , 0 > ) 0 8 ,8 6
) 8,8 ) , , ) , 8 ) ,8 6 ,0
) 0, 08 ( 8 , 0 ,8 , ,8 ) 0 08 8 ) 6
) 08 8 0 , ) 0, 08 ) 0 ,8 , ,8
) 8 , , ,8. 8 0 ,
, ,8 8 8 ) , ) *. ,A ,
8 ) , ,8 ) ) 8 08 . @ 0 8
) , ,88 ( 0, 8 , ,8 ) ,8
-+
+
5
5.5 80 , * % 5
5.3 % ) ) B
5.B - C
5.C C
D
3.5 - - ( D
3.3 * ; < E
3.3.5 F, - E
3.3.5.5 ! 0 *
0 ;! <
vii
3.3.3 - ) 0 ; <
, A 55
3.3.B F, - * 5D
3.3.B.5 *
; < 5D
35
B.5 $ * * 35
B.3 * 3B
B.B $ * ! 3D
B.C $ * - ) 0 BB
B.D $ * *
; < BE
B.D.5 8 0 0
* BE
B.D.3 * , A *
$ BH
B.D.B - - ( * ) ) A *
" CB
B.D.C % ) ) CE
B.D.D " * ) CG
B.D.E 0 CG
D4
C.5 D4
C.5.5 ! 0 * 0
C.5.3 - ) 0 ; < D5
C.5.B * ; < DB
C.3 ) DI
E5
D.5 , E5
D.3 " )) E3
EB
ix
@ , 5H
* C3
* ) ) - ) & D3
0 * ) 6 , ) 5G
$ * 0 * 35
* , 3B
! 3D
, * + , ! ) . 3H
* BC
- 0 * ) 0 , BG
) ) + BI
! 0 ( ; ) < 0 - ( , *
) .
BH
" * , A * . C4
# ) * * - . ) C5
0 * & , *,
*, * ) .
CB
J * ) - . ) A
0.
CI
@ * , * DC
- * A ) * DE
CHAPTERI
INTRODUCTION
Thischaptercontainstheprojectbackgr oundoftheresearchstudywhichentitled“ Feasiblestudy
on characterizationandcomparisonofacceleratedlife testsandthecompatiblesoftwareforautomotive
component”. This chapter is including the background of the proj ect whereas a little bit of project
summary will be include, objectives of the project’ s research, scope, and problem statement of the
project.
1.1BackgroundofProject
Traditional life data analysis involves analyzing times-to- failure data (of a product, system or
component)obtainedundernormaloperatingconditio nsinordertoquantifythelifecharacteristicsof the
product, system or component. In many situations, a nd for many reasons, such life data (or
times-to-failuredata)isverydifficult,ifnotimpossible, toobtain.Thereasonsforthisdifficultycanincl udethe
longlifetimesoftoday'sproducts,thesmalltime periodbetweendesignandreleaseandthechallenge of
testingproductsthatareusedcontinuouslyundern ormalconditions.
Giventhisdifficulty,andtheneedtoobservefailu resofproductstobetterunderstandtheirfailure
modes and their life characteristics, reliability p ractitioners have attempted to devise methods to fo rce
these products to fail more quickly than they would under normal use conditions. In other words, they
haveattemptedtoacceleratetheirfailures.Overth eyears,thetermacceleratedlifetestinghasbeen used
todescribeallsuchpractices.
Avarietyofmethodsthatservedifferen tpurposeshavebeentermed“acceleratedlifetesti ng.” As
thetermisusedinthisreference,acceleratedlif etestinginvolvesaccelerationoffailureswiththe single
accelerated life testing, the engineer is intereste d in predicting the life of the product at normal u se
conditions, from data obtained in an accelerated li fe test. Each type of Accelerated Life Tests (ALT)
analyseswillbeindetailexplanationincludingits example.
1.2ProblemStatement
In the past, life test analysis is used to predict the life span of components. Normally, it takes
longer to estimate the life span. This can cause pr oblems to production line whereas it will reduce the
production rate and takes longer to complete a prod uct. With using accelerated life test analysis, the
usefulreliabilityinformationcouldbeobtainedin amatterofweeksinsteadofmonths.
Reliability is a very important measure of modern e lectronic products and its growth and
improvementarevitaltosafeguardingprogressinm anufacturingindustries.Reliabilityatfirstseems to
be just a number without any meaning to many who do not have a deep understanding of its proper
application. However, it is the only possible means b y which the robustness of modern manufacturing
andtestingproceduresmaybemeasured.
An integrated reliability program includes all relia bility testing programs such as Design
ValidationTesting,AcceleratedLifeTesting,HighlyAcce leratedLifetesting,ReliabilityDemonstration
Testing,andOn-Going ReliabilityTests.Thisshouldbeimplementedinsuch awaythatallthereliability
needsfortheproductarecoveredandthereisnoe xtraeffort.Thedesignofanoptimalreliabilityte sting
program will also help reduce the overall life cycl e costs of the product while improving the products
1.3Objective
Theseanalysisobjectivesare toanalyzethecharacteristicoftheacceleratedli fetestanalysisand
tocomparewithAcceleratedLifeTest(ALT)softwares. Understandandquantifytheeffectsofstress(or
otherfactors)onproductlife.
Designacceleratedteststhatwillbethemosteffec tivetoachievedesiredobjectives.Significantly
reducethetesttimerequiredtoobtainreliability metricsforaproduct,whichcanresultinfaster
time-to-market,lowerproductdevelopmentcostsandimprove ddesigns.
1.4Scope
Thescopeofthisanalysisis toobserveandstudythemethodology.Anothermajor scopeofthis
projectistomakecomparisonsoftheAcceleratedLi feTests(ALT).An AcceleratedLifeTests(ALT)
software must been chosen to find the compatible ALT software th at can be used on automotive
CHAPTERII
LITERATUREREVIEW
Inthischapter,thereisbrieflyexplanationregar dingtothepurposeofthisprojectwhichistofind
thecompatibleAcceleratedLifeTests (ALT)softwareforautomotivecomponent.EachofALTwill be
brieflyexplainedincludingitsexample.Thefunctio nsandcomparisonofallALTsoftwarewillbestated.
2.1Overview
In typical life data analysis, the practitioner ana lyzes life data of a product's sample operating
undernormalconditionsinordertoquantifytheli fecharacteristicsoftheproductandmakepredicti ons
about all of the products in the population. In lif e data analysis, the practitioner attempts to make
predictionsaboutthelifeofallproductsinthep opulationby"fitting"astatisticaldistributiont olifedata
fromarepresentativesampleofunits.Theparameter izeddistributionforthedatasetcanthenbeused to
estimate important life characteristics of the prod uct such as reliability or probability of failure a t a
specifictime,themeanlifefortheproductandfa ilurerate.
Lifedataanalysisrequiresthepractitionerto:
l Gatherlifedatafortheproduct.
l Selectalifetimedistributionthatwillfittheda taandmodelthelifeoftheproduct.
l Estimatetheparametersthatwillfitthedistributi ontothedata.
l Generateplotsandresultsthatestimatethelifech aracteristics,likereliabilityormeanlife,ofth e
2.2AcceleratedLifeTest(ALT)
Eachtypeoftestthathasbeencalledanaccelerate dtestprovidesdifferentinformationaboutthe
product and its failure mechanisms. Generally, accel erated tests can be divided into three types:
Qualitative Tests (Torture Tests or Shake and Bake Tests ), ESS and Burn-in and finally Quantitative
AcceleratedLifeTests.
2.2.1QualitativeTests
QualitativeTestsareteststhatyieldfailureinform ation(orfailuremodes)only.Theyhavebeen
referredtobymanynamesincluding:
• ElephantTests • TortureTests
• HALT(HighlyAcceleratedLifeTesting) • Shake&BakeTests
Qualitativetestsareperformedonsmallsampleswith thespecimenssubjectedtoasinglesevere
levelofstress,toanumberofstresses,ortoat ime-varyingstress(i.e.,stresscycling,coldtohot, etc.).If
the specimen survives, it passes the test. Otherwise, appropriate actions will be taken to improve the
product’sdesigninordertoeliminatethecause(s)offail ure.Qualitativetestsareusedprimarilytoreveal
probable failure modes. However, if not designed prop erly, they may cause the product to fail due to
modesthatwouldhaveneverbeenencounteredinreal life.
Agoodqualitativetestisonethatquicklyreveals thosefailuremodesthatwilloccurduringthe
lifeoftheproductundernormaluseconditions.In general,qualitativetestsarenotdesignedto yieldlife
data that can be used in subsequent analysis or for “Accelerated Life Test Analysis.” In general,
qualitative tests do not quantify the life (or reli ability) characteristics of the product under norma l use
conditions.HighlyAcceleratedLifeTesting(HALT)ischos enunderthistopicforfurtherdiscussion.
becoming recognized as powerful tools for improving product reliability, reducing warranty costs and increasingcustomersatisfaction.
AHighlyAcceleratedLifeTest(HALT)isastr esstestingmethodology foracceleratingproduct
reliabilityduringtheengineeringdesignprocess. Itiscommonlyappliedtoelectronicequipmentand is
performed to identify and thus help resolve design weaknesses in newly- developed equipment. Thus it
greatlyreducestheprobabilityofin-servicefailu resi.e.itincreasestheproduct'sreliability.
Progressively more severe environmental stresses are applied building to a level significantly
beyondwhattheequipmentwillseein-service.
Bythismethodweaknessescanbeidentifiedusinga smallnumberofsamples(sometimesoneor
twobutpreferablyatleastfive)intheshortestpo ssibletimeandatleastexpense.Asecondfunction of
HALT testing is that it characterizes the equipment un der test, and identifies the equipment's safe
operatinglimitsanddesignmargins.Individualcom ponents,populatedprintedcircuitboards,andwhole
electronicsystemscanbesubjectedtoHALTtesting.
The size of the test sample is governed by many fact ors including the number of samples
available, cost, type of stresses applied, and phys ical size. For example, component manufacturers can
typically test thousands of individual components a t one time whereas often it is not economically
feasibletowriteoffmorethanafewitemsofvery expensiveequipmentbecauseproductionquantitieso r
theapplicationdoesnotjustifythecost.
AgeneralprincipalisthatwhilstHALTtestcanandshou ldbeconductedatunitlevel,itisvery
desirable to conduct it at sub-assembly and piece- part level as well. Temperature cycling and random
vibration, power margining and power cycling are the most common form of failure acceleration for
electronic equipment. HALT does not measure or determi ne equipment reliability but it does serve to
improvethereliabilityofaproduct.Itisanempi ricalmethodusedacrossindustrytoidentifythel imiting
failuremodesofaproductandthestressesatwhich thesefailuresoccur.
On military design and development programs HALT is cond ucted before qualification testing.
Bysodoing,significantcostsavingscanaccrue,b ecausetheformalqualificationoftheequipmentan d
subsequentcustomeracceptancewillproceedmorerap idlyandatlowercost, andtheneedformultiple
Thereis no one recipeforaHALT test,but different st resses areappliedwith differentfailures occurringduringeach.Thetypesofstresstypically employedare: 1. ColdStep 2. HotStep 3. RapidTemperatureCycling 4. SteppedVibration(random)
5. Combined Environment stress (temperature cycling and random vibration plus power switching
andpowermargining)
InHALTthesestressesareappliedinacontrolled,in crementalfashionwhiletheunitundertestis
continuously monitored for failures. Once the weaknes ses of the product are uncovered and corrective
actions taken, the limits of the product are clearl y understood and the operating margins have been
extendedasfaraspossible.Amuchmorematurepro ductcanbeintroducedmuchmorequicklywitha
higherdegreeofreliability.
When HALT testing is applied during the design process , it can produce a very robust product
withoutunduecost,becauseimprovementsaretargete donlywheretheyareneeded.Asfailuremodesare
discoveredandunderstoodtheproductlifecanincr easesignificantly.Thismakestheproductmorerobu st
andriskoffailurereducesdrastically.
Individualcomponentssuchasresistors ,capacitors,anddiodes,printedcircuitboards ,andwhole
electronicproductssuchascellphones,PDAs,andTV s,eventuallyfailatdifferentratesunderdifferen t
end-user stress levels. For example, a typical cons umer-owned television will not likely be operated at
temperatures outside the range of normal living acc ommodations, or subjected to mechanical stress by
being repeatedly dropped. A cell phone, on the other hand, may be dropped from 3 or 4 feet off the
groundfairlyoften,andsubjectedtoavaryingran geofvibrations.
A commercialtelephoneswitch mayberequiredtooper ateinremoteinstallationsrangingfrom
Barrow, Alaska to Phoenix, Arizona, at an ambient tem perature range of from minus 50 to plus 120
degrees Fahrenheit. Components used in military and aerospace applications may be subjected to even
more severe operating temperature requirements as we ll as high G- forces and ionizing radiation,
sometimessimultaneously,tomeetspecificMIL-SPEC standards.
Therefore, failure rate data used to select any dev icein a product must correlatewith the stress
levelsintheproductorapplication.Toaccomplish this,therequiredlifetimeandoperatingcondition sfor
minutesofactualuse,buttheirfailureratewillb eexpectedtobezeroduringthatperiod.Devicesu sedin
satellitesorspacevehicleswherereplacementisn otpossibleareexpectedtohaveazerofailurerat efor
the lifetime of the vehicle. Knowing the required fai lure rate as determined by the application,
components can be selected based on the failure rat e data supplied by the manufacturer as described
above.
AtestchamberdesignedforHALTtestingisreallyrequ iredforsatisfactoryandsuccessfulHALT
testing.Atemperatureramprateofatleast45degr eesC/minuteisrequired,andsomeHALTchambers
canachievecloserto60degreesC/minute. Toachie vethesehighramp- rates,coolingby meansofthe
evaporationofliquidnitrogenisrequired.Chamber sequippedwithoneortwocompressorsforcooling,
suchasthosetypicallyemployedduringqualificati ontesting,arereallynotadequate.Suchchambers are
typically only capable of 15- 18 degrees C/minute, and are subject to frequent br eakdowns if always
pushedtotheirlimit.
Asuitablechamberalsohastobecapableofapplyin grandomvibrationwithasuitablespectral
density profileinrelationto frequency.Ithas to becapableofdoingthisatthesametimeas it ap plies
temperaturecyclingsoacombinedchamberisessent ialratherthanseparatechambersforvibrationand
temperaturecycling.Whereaschambersusedforqual ificationtestingoftenuseelectro- dynamicvibrators
(like a giant moving- coil loudspeaker). Chambers designed for HALT testing use pneumatic hammers
arrangedtostrikethebaseplateuponwhichtheequ ipmentismounted.Thehammersproducesixdegree
of freedom random vibration: they produce simultane ous vibration in each principle axis and rotations
abouteachaxis.Thiscompareswiththesingleaxiso fvibrationprovidedbyelectro-dynamicvibrators.
2.2.2EnvironmentalStressScreening(ESS)andBurn -In
The second type of accelerated test cons ists of ESS and Burn- In testing. ESS is a process
involving the application of environmental stimuli to products (usually electronic or electromechanica l
products) on an accelerated basis. The surviving population, upon completion of screeni ng, can be
assumedtohaveahigherreliabilitythanasimilar unscreenedpopulation.ThegoalofESSistoexpose,
identifyandeliminatelatentdefectswhichcannot bedetectedbyvisualinspectionorelectricaltest ingbut
which will cause failures in the field. ESS is perform ed on the entire population and does not involve
Developedtohelpelectronicsmanufacturersdetect productdefectsandproductionflaws,ESSis
widelyusedinmilitaryandaerospaceapplications, lesssoforcommercialproducts.Thetestsneednot be
elaborate,forexample,switchinganelectronicore lectricalsystemonandoffafewtimesmaybeenoug h
to catch some simple defects that would otherwise be encountered by the end user very soon after the
productwasfirstused. Teststypicallyincludethefollowing: · Temperaturevariations · Vibrationtests · Pressure · Flexibilitytests
ESS can be performed as part of the manufacturing pr ocess or it can be used in new product
qualificationtesting.AnESSsystemusuallyconsists ofatestchamber,controller,fixturing,intercon nect
andwiring,andafunctionaltester.Thesesystemsca nbepurchasedfromavarietyofcompaniesinthe
environmentaltestindustry.
Thestressscreeningfromthisprocesswillhelpfind infantmortalityintheproduct.Findingthese
failures before the product reaches the customer yi elds better quality and lower warranty expenses.
Associated military terminology includes an operatio nal requirements document (ORD) and on- going
reliabilitytesting(ORT).
To conduct an effective screen, the product must be capable of surviving the high stimulation
levelsneededtoacceleratethefailuremechanismo fassemblyrelateddefects.Theparticipationofdes ign
andreliabilityengineeringistodeterminethelim itsofenvironmentalstimulationwhichtheproductc an
endurebeforeitsperformanceispermanentlydegrad ed.Amechanically“weak” designmaybechanged
toimproveitsmarginswithrespecttoaspecificf ormofenvironmentalstimulus.
TheformofESSchosenbythefactoryisdependenton thefailuremechanismsfortherelevant
fieldfailures.AnESSprogramisfaultywhenitdoes notexposethelocusoffaultsseenbythecustomer .
TheESSprocessisadynamicprocesswhichmustchange asproductfailurebehaviorchanges.Forthis
reason, it is not appropriate to “spec” an ESS regimen and leave the regimen unchanged throu ghout
productlife.
Again,thepurposeofanyenvironmentalstressscree ning(ESS)regimenistoexposethehidden
defects that were introduced during the manufacturin g process. More succinctly, manufacturing defects
that the design is robust enough to meet its design goals. As the electronics industry has matured,
componenttechnologyandassemblytechniqueshavec hangedprofoundly.
Thefirstproducttobesubjectedtoenv ironmentalstressscreeningintheformof“burn-in ” was
productsdesignedwithvacuumtubetechnology.Burn- incanberegardedasaspecialcaseofESS.For
these products, high temperature burn- in was the best stimuli for precipitating latent def ects. Early
transistorandintegratedcircuittechnologydefect swerealsoefficientlystimulatedtofailureusing high
temperatureburn- in. During this time, most product defects werecompo nent related defects. Todaywe
seeamuch different failurebehaviorforelectroni cproducts. Components havebecomeso reliablethat
mostproductdefectsarerelatedtotheassemblypr ocess.Astheproductfaultspectrumhaschanged,th e
screening stimuli for rapid defect precipitation mu st also change to ensure that latent defects are
efficientlystimulatedtoanobservablefailure.
According to MIL-STD-883C, Burn- in is a test performed for the purpose of screening or
eliminatingmarginaldevices.Marginaldevicesare thosewithinherentdefectsordefectsresultingfro m
manufacturing aberrations which cause time and stre ss-dependent failures. As with ESS, Burn- in is
performedontheentirepopulation.Theintentionis todetectthoseparticularcomponentsthatwouldfa il
as a result of the initial, high-failure rate porti on of the bathtub curve of component reliability . If the
burn-in period is made sufficiently long (and, perhaps, artificially stressful), the system can then be
trustedtobemostlyfreeoffurtherearlyfailures oncetheburn-inprocessiscomplete.
Apreconditionforasuccessfulburn-in isabathtub- likefailurerate,thatis,therearenoticeable
early failures with a decreasing failure rate follow ing that period. By stressing all devices for a cer tain
burn-in time the devices with the highest failure rate fa il first and can be taken out of the cohort. The
devices that survive the stress have a later positi on in the bathtub curve (with an appropriately lower
ongoingfailurerate).
Thusbyapplyingaburn-in,earlyin- usesystemfailurescanbeavoidedattheexpense( tradeoff)
ofareducedyieldcausedbytheburn- inprocess.Whentheequivalentlifetimeofthestr essisextended
into the increasing part of the bathtub-like failur e-rate curve, the effect of the burn- in is a reduction of
productlifetime.Inamatureproductionitisnot easytodeterminewhetherthereisadecreasingfail ure
rate. To determine the failure time distribution for a very low percentage of the production, one would
havetodestroyaverylargenumberofdevices.
Becauseofthis,aprocessthatinitiallyusesburn -inmayeventuallyphaseitoutasthevariousroot causes
forfailuresareidentifiedand eliminated.Forele ctroniccomponents,burn- in is frequently conductedat
elevated temperature and perhaps elevated voltage. This process may also be called heat soaking . The
componentsmaybeundercontinuoustestorsimplyt estedattheendoftheburn-inperiod.
Thereisanotheruseofthetermbysome audiophiles ,wholeavenewaudioequipmentturnedon
for multiple days or weeks, to get the components to achieve optimal performance. However, many
debatesariseaboutthebeneficialeffectsofthis practice.
2.2.3QuantitativeAcceleratedLifeTests
QuantitativeAcceleratedLifeTesting,unlikethequali tativetestingmethods(i.e.,TortureTests,
Burn-in, etc.) described previously, consists of quantit ative tests designed to quantify the life
characteristics of the product, component or system under normal use conditions, and thereby provide
“Reliability Information.” Reliability information can include the determinati on of the probability of
failure of the product under use conditions, mean l ife under use conditions, and projected returns and
warrantycosts.Itcanalsobeusedtoassistinthe performanceofriskassessments,designcomparison s,
etc.AcceleratedLifeTestAnalysis(ALTA)softwarehasbeen chosenforQALT.
2.2.3.1AcceleratedLifetestAnalysis(ALTA)
Accelerated Life Testing can take the form of “Usage Rate Acceleration” or “ Overstress
Acceleration.” Both Accelerated Life Test methods are d escribed next. Because “ Usage Rate
Acceleration” test data can be analyzed with typical life data an alysis methods, the Overstress
Acceleration method is the testing method. For all l ife tests, some time-to- failure information for the
productisrequiredsincethefailureoftheproduc tistheeventwewanttounderstand.
In other words, if we wish to understand, measure, and predict any event. Most products,
components or systems are expected to perform their functions successfully for long periods of time,
competitive,thetimerequiredtoobtaintimes-to- failuredatamustbeconsiderablylessthantheexp ected
For products that do not operate continuously, one can accelerate the time it takes to induce
failures by continuously testing these products. Thi s is called “Usage Rate Acceleration.” For products
forwhich“UsageRateAcceleration” isimpractical,onecanapplystressatlevelsthat exceedthelevels
that a product will encounter under normal use condi tions and use the times- to failure data obtained in
thismannertoextrapolatetouseconditions.Thisi scalled“OverstressAcceleration.”
i)UsageRateAcceleration
For products that do not operate continuously under normal conditions, if the test units are
operatedcontinuously,failuresareencounteredear lierthaniftheunitsweretestedatnormalusage. For
example, a microwave oven operates for small periods of time every day. One can accelerate a test on
microwaveovensbyoperatingthemmorefrequentlyun tilfailure.Thesamecouldbesaidofwashers.If
weassumeanaveragewasheruseof6hoursaweek,on
ecouldconceivablyreducethetestingtime28-fold by testing these washers continuously. Data obta ined through usage acceleration can be analyzed
with the same methods used to analyze regular times- to-failure data. The limitation of “ Usage Rate
Acceleration” arises when products, such as computer servers and p eripherals, maintain a very high or
even continuous usage. In such cases, usage acceler ation, even though desirable, is not a feasible
alternative. In these cases the practitioner must s timulate the product to fail, usually through the
applicationofstress.Thismethodofacceleratedli fetestingiscalled“OverstressAcceleration”.
ii)OverstressAcceleration
For products with very high or continuo us usage, the accelerated life- testing practitioner must
stimulatetheproducttofailinalifetest.
