ML17256A648

Rev 0 to Fracture Mechanics Evaluation of High Energy Piping Lines at Re Ginna Nuclear Power Plant.

ML17256A648
Person / Time
Site:	Ginna
Issue date:	04/08/1983
From:	COPELAND J F, WU Y S NUTECH ENGINEERS, INC.
To:
Shared Package
ML17256A647	List: ML17256A646 ML17256A648
References
TASK-03-05.A, TASK-3-5.A, TASK-RR RGE-02-004-R00, RGE-2-4-R, NUDOCS 8304290006
Download: ML17256A648 (59)

Text

RGE-02-004 Revision0April8,1983100.2602.0200 FRACTUREMECHANICS EVALUATION OFHIGHENERGYPIPINGLINESATTHER.E.GINNANUCLEARPOWERPLANTPreparedfor:Rochester Gas&ElectricCorporation Preparedby:NUTECHENGINEERS SanJose,California Preparedby:ZF.Dr.J.F.CopelandProjectEngineerReviewedandApprovedby:g/Dr.LECDHsu'anager,AppliedMechanics Issuedby:Dr.Y.S.WuConsultant ID.K.McWilliams ProjectManagerDate:,>",REmjATORY OOCKETFILEMPVnutech REVISIONCONTROLSHEETFracture-Mechanics Evaluation RgpoRTNUyBgRRGE-02-004 ofHighEnergyPipingLinesatRevision0theR.E.GinnaNuclearPower'lant NAME/TITLEINITIALSDr.L.C.HsuEnineerinManaerNAME/TITLEINITIALSDr.Y.S.Wu/Consultant INAMEITITLEYSWINITIALSNAMEITITLEINITIALSPAGEIS)REVPREPAREDBY/OATEACCURACYCRITERIACHECKBY/OATECHECKBY/OATEREMARKSit,hruv0Tf-~%+6)SW@~-F3-lthru53CP"S'-S3Pstk'ag-Q 3.3.QEP34.1.1nutech TABLEOFCONTENTS

1.0INTRODUCTION

/EXECUTIVE SUMMARYPacae1.1Background 11.2Objectives andTechnical Approaches forGinna'1.3Conclusions andRecommendations 52.0FRACTUREMECHANICS LEAK-BEFORE-BREAK ANALYSIS2.1CriticalCrackSizesforInstability 2.1.1J-Integral andTearingModulusllAnalysis2.1.2NetSectionPlasticCollapseCriterion 162.2LeakRates193.02.2.1Accumulator Line2.2.2Pressurizer SurgeLine2.2.3GinnaLeakDetection Capabilities 2.3Subcritical CrackGrowthRates2.3.1StressProfiles2.3.2CyclingRate2.3.3CrackGrowthAnalysisCONCLUSIONS ANDRECOMMENDATIONS 20222324252627294'REFERENCES 52RGE-02-004 Revision0nutech Ie-LISTOFTABLESNumberTitlePacae2-1Parameters forLeak-Before-Break AnalysisofPressurizer SurgeandAccumulator Lines312-2LevelDASMECodeMaximumAllowable StressesUsedforAnalysisofCrackInstability forPressurizer Surge(PSL)andAccumulator Lines(AL)322-3AppliedJ-Integral andTearingModulusValuesasFunctions ofThrough-Wall HalfCrackLength,a,FortheGinnaAccumulator LineWithAnAppliedStressof30,133psi332-4AppliedJ-Integral andTearingModulusValuesasFunctions ofAppliedStressfortheGinnaAccumulator LinewithaThrough-Wall HalfCrackLengthof3.5in.342-52-62-72-82-9AppliedJ-Integral andTearingModulusValuesasFunctions ofThrough-Wall HalfCrackLength,a,fortheGinnaPressurizer SurgeLinewithanAppliedStressof37,600psiFailureCrackSizesforPostulated CompoundCrackinGinnaAccumulator Line,BasedonNetSectionPlasticCollapseCriterion FailureCrackSizesforPostulated CompoundCrackinGinnaPressurizer SurgeLine,BasedonNetSectionPlasticCollapseCriterion LeakRateResultsforCircumferential Through-WallCracks(CTWC)andLongitudinal Through-WallCracks(LTWC)inthePressurizer SurgeandAccumulator Lines(PSLandAL)forNormalOperation PressureStressesTransients Considered inSubcritical CrackGrowthRateAnalysesforPressurizer SurgeandAccumulator Lines(Reference 8)35363738'39RGE-02-004 Revision0ivnutechIINOINQIIRB LISTOFFIGURESFiciure2-12-2TitlePressurizer SurgeLineSafetyInjection FromAccumulator APacae40412-3Representation ofPostulated CracksinPipesforFractureMechanics Leak-Before-Break Analysis422-4J-Integral Resistance CurvesforAustenitic Stainless Steel(Reference 7)432-5J-Integral/Tearing ModulusStability DiagramforGinnaAccumulator LinewithThrough-Wall Cracks442-6J-Integral/Tearing ModulusStability DiagramforGinnaPressurizer SurgeLinewithThrough-WallCracks2-7FailureAnalysisDiagramforPostulated Com-poundCrackinGinnaAccumulator LineLine,BasedonNetSectionPlasticCollapseCriterion 462-8FailureAnalysisDiagramforPostulated Com-poundCrackinGinnaPressurizer SurgeLine,BasedonNetSectionPlasticCollapseCriterion 472-9DiagramforMaximumSteam/Water FlowRatetoDetermine FlowRateforSaturated LiquidinthePressurizer SurgeLine(MoodyModel-Reference 15)482-10StressProfilesforBoundingCase(Reference 3)forSubcritical CrackGrowthPredictions 492-11CyclicLoadingConditions AssumedforConserva-51tiveSubcritical CrackGrowthRateAnalysisofPressurizer SurgeandAccumulator Lines2-12Predicted Subcritical CrackGrowthRatesforCircumferential andLongitudinal CrackswithAssumedInitialDepths(ai)of0.02inchesand0.10inchesforthePressurizer SurgeandAccumulator Lines51RGE-02-004 Revision0nutech

1.0INTRODUCTION

/EXECUTIVE SUMMARYHighenergylinebreak(HELB)analyseswerecompleted fortheresolution ofopenitemsfortheNRCSystematic Evaluation Program(SEP)TopicIII-5.AfortheR.E.GinnaNuclearPowerPlant.Thisreportaddresses leak-before-break fracturemechanics evaluations oftheGinnapressurizer surgeandaccumulator pipinglines.BackroundTheSEPwasinitiated bytheNRCtoreviewthedesignsofolderoperating nuclearreactorplantstoreconfirm anddocumenttheirsafety.Thereviewcomparedtheas-builtplantdesignwithcurrentcriteriain137different areasdefinedas"topics"(Reference 1).Manyofthesetopics'et currentcriteriaorwereacceptable onanotherdefinedbasisforGinna.Theobjective ofthisstudyistheresolution ofSEPTopicIII-5.AforGinna,asdefinedinReference 1.AppendixAof10CFRPart50reguiresthatstructures, systemsandcomponents important tosafety(Engineered SafetyFeatures, ESFs)beappropriately protected againstthedynamiceffectsofpostulated pipebreaks.ThegoalistoprotecttheseESFssotheplantcanbeRGE-02-004 Revision0nutechGNOINQGRG shutdownandmaintained inasafeshutdowncondition intheeventofapostulated ruptureofapipingsystemcontaining highenergyfluid.CurrentdesignsprotectESFsagainsttheconsequences ofhighenergylinebreaks(HELBs)throughtheuseofpipewhiprestraints, jetimpingement shields,physicalseparation andothermethods.However,plantsdesignedbeforetheexistence ofcurrentrequirements generally donothavethefullcomplement ofsuchfeatures.

Furthermore, inmanycasesmodifications toincorporate thesefeaturesmaybeimpractical duetophysicalplantconfigurations orotherconsiderations.

Therefore, theNRChasgivenguidanceonotheracceptable methodsfortheresolution ofSystematic Evaluation Program(SEP)TopicIII-5.A,forHighEnergyLineBreaksInsideContainment.

1.2Ob'ectives:

andTechnical AroachesforGinnaInReference 1,theNRCadvisesthatbreaksintheaccumulator lineorpressurizer surgelinecouldadversely affectnearbysafety-related equipment.

Additionally, guidanceforperforming fracturemechanics leak-before-break evaluations toresolvethisissuewasforwarded toRochester GasandElectricCorporation RGE-02-004 Revision0nutech (RGE)bytheNRC(Reference 2).Thisapproachwasemployedforthesepipinglines.Itisbasedonacombination ofinservice inspection (ISI)andleakdetection, todetectthepresenceofcracks,andoffracturemechanics analysistoassurethatcrackinstability willnotoccurforcrackssmallerthanthosedetectable bythesemethods.Thesedetection methodscomplement eachother,sinceISIisespecially suitedtofindinglongcracks,andleakmonitorsdetectshort,through-thickness cracks.Reference 2providesthemethodology tocomputecrackopeningareasfordetermining leakratesforcomparison withdetection limits.TheISIinvolvesvolumetric inspection inaccordance withASMESectionXIforaClass1system,regardless ofactualsystemclassification.

Thegoalistodetectandlimitanyserviceinducedflawstoallowable sizesprescribed bytheASMECode,SectionXI(crackdepthlimitedtolessthanapproximately 10%ofpipewallthickness).

Fracturemechanics subcritical crackgrowthanalysesareemployedtoassurethatthisgoalforlimitingcrackgrowthismet.TheselimitsoncracksizeimposedbyleakmonitorsandISIarecomparedtothecriticalcracksizespredicted forinstability andpiperupture,computedinaccordance withReference 2.Adequatemarginbetweencrackdetection andthecracksizeforrupturemustexist.Inthisway,crackRGE-02-004 Revision0nutech detection andcorrective actionswillprecedeanychanceforHELBsandsubsequent postulated effectsonESFs.Inaccordance withthelatestNRCguidance(Reference 2),theleak-before-break technique wasevaluated fortheGinnapressurizer surgeandaccumulator lines.Theelementsofthisevaluation includethedefinition ofthefollowing:

a)Largestcracksizewhichwillremainstableb)Leakrateresulting fromacrackoflength2t(twicethepipewallthickness) c)Sizeofcrackwhichwillleakatarategreaterthanlgpm,ifb)resultsinlessthanlgpm.d)Analysisofpart-through-thickness cracksforsubcritical crackgrowthratestoestablish ISIintervals.

Veryconservative analyseswereperformed topredictthelargeststablecracksizes,byusingworstcasestresses(References 3and4),aswellasASMECodemaximumallowable stresses.

Inthisway,setsofanalyseswereperformed toenvelopealllocations ineachline,asRGE-02-004 Revision0nutechGNOINGGAU wellastocompensate forpotential futureloadincreases.

Thesubcritical crackgrowthrateanalyseswerealsoenveloped inasimilarmanner,byusingaconservative loadcyclingspectrumbasedonGinnadesigntransients andthemostseveretransient loadsinthestressreports(References 3and4).1.3Conclusions Conclusions resulting fromthepreceding analysesarereflected intermsofcriticalthrough-wall cracklengthsforinstability andcracklengthsfor1gpmleakratesinthefollowing table:GINNASTRESSREPORTWORSTCASENSTRESSESCrackLenthforInstabilit (in.)LineCrackOrien-tationNetSectionCollaseCrackLengthTearingfor1gpmModulusLeakRatePressurizer SurgeAccumulator circ~long~ClrC~long.7.810>20>1222162.6<23.52.66BothNetSectionPlasticCollapseandTearingModulusapproaches areusedtopredictcriticalcracklengthsforinstability, basedonworst-case stresses.

Theworstcase(Ginnastressreport)stressesusedfortheRGE-02-004 Revision0nutechIINOINCGAS

~'

pressurizer surgelineanalysesarePm+Pb=34,747psi.Thecorresponding stressesfortheaccumulator linearePm+PI=30,133psi.Onlynormaloperating pressurestresseswereusedtocomputeleakrates.Thus,thisanalysisisconsidered tobeconservative.

RGSE(References 5and6)hascurrentbulkleakdetection capabilities todetect1gpmleakratesfortheselinesinatleast6.4hr.Sinceamarginofatleastafactorof2existsbetweenthecracklengthsfora1gpmleakandthe"worstactualstress"lengthsforinstability (consistent withReference 2guidance),

thesecurrentleakdetection systemsareconsidered adequate.

Furthermore, subcritical crackgrowthrateanalysesshowthatinservice inspection intervals of10yearsareappropriate todetectpart-through-thickness cracksbeforetheyapproachinstability.

RGE-02-004 Revision0nutech

2.0 FRACTUREMECHANICS

LEAK-BEFORE-BREAK ANALYSISPostulated breaksintheaccumulator lineorpressurizer surgelinecouldadversely affectnearbysafety-related equipment.

TheselinesattheGinnaPlantareshowninFigures2-1and2-2,andthepipingsystemparameters aregiveninTable2-1.Additionally, guidanceforperforming fracturemechanics leak-before-break evaluations toresolvethisissuewasforwarded toRochester GasandElectricCorporation (RGE)bytheNRC(Reference 2).Thisapproachwasemployedforthesepipinglines.Itisbasedonacombination ofinservice inspection (ISI)andleakdetection, todetectthepresenceofcracks,andoffracturemechanics analysestoassurethatpiperupturewillnotoccurforcrackssmallerthanthosedetectable bythesemethods.Thesedetection methodscomplement eachother,sinceISIisespecially suitedtofindinglongcracks,andleakmonitorsdetectshort,through-thickness cracks.Thesetypesofcracksarerepresented inFigure2-3.Inaccordance withthelatestNRCguidance(Reference 2),theleak-before-break technique wasevaluated fortheGinnapressurizer surgeandaccumulator lines.Theelementsofthisevaluation includethedefinition ofthefollowing:

RGE-02-004 Revision0nutechGNOINCRRU a)Largestcracksizewhichwillremainstable;b)Leakrateresulting fromacrackoflength2t(twicethepipewallthickness);

c)Sizeofcrackwhichwillleakatarategreaterthanlgpm,ifb)resultsinlessthan1gpm;d)Analysisofpart-through-thickness cracksforsub-criticalcrackgrowthratestoestablish ISIintervals.

Veryconservative analyseswereperformed topredictthelargeststablecracksizes,byusingASMECodemaximumallowable

stresses, andalso,byusingthemaximumstressesinthepipingstressreports(References 3and4).Inthisway,setsofanalyseswereperformed toenvelopealllocations ineachline,aswellastocompensate forpotential futureloadincreases.

Thesubcritical crackgrowthrateanalyseswerealsoenveloped inasimilarmanner,byusingaconservative loadcyclingspectrumbasedonGinnadesigntransients (Reference 8)andthemostseveretransient loadsinthestressreports(References 3and4).RGE-02-004 Revision0 2.1CriticalCrackSizesforInstabilit ThreemethodsofanalysistopredictcriticalcracksizesfortheGinnaaccumulator lineandpressurizer surgelinewereconsidered according totheguidancegivenbytheNRCinReference 2.Thesemethodsare:a)linearelasticfracturemechanics; b)J-Integral andTearingModulusapproaches (Reference 9);andIIc)thenetsectionplasticcollapsecriterion (References 10and11).AsseeninTable2-1,thesepipingmaterials areType316austenitic stainless steel,whichhasaveryhighleveloftoughness.

Reference 7reportsacriticalJ-Integralvalue,JIc,forfractureinitiation ofType316at600'F,of5260-2inaoneinchthicktestin-lbinspecimen.

OtherresultsforType304stainless steel(asimilarmaterial) areshowninFigure2-4andarealsoatahightoughness level.Figure2-4showscrackextension asafunctionofappliedJ-Integral loading.Insomecases,JIcvaluescanbesimplyconverted toKIc(linearelasticfracturetoughness) andusedforaRGE-02-004 Revision0nuteclh conservative analysisoffractureresistance inacomponent.

However,thisconversion issuspectformaterials whichdonotmeetthefollowing validitycriterion (Reference 12):B>25Icywhere,B=Specimenthickness (in.)JZc=CriticalJ-Integral Value(2)in-lbine=Materialyieldstrength(psi)Bysolvingthepreceding equationforB,usingJIc=5,260anda=30,000psi(theroomtemperature 2yieldstrength),

itcanbeseenthatBmustbegreaterthan4.4inchesforavalidconversion tolinearelasticfracturemechanics parameters (KIc)andanalysis.

Athighertemperature, astheyieldstrengthdecreases, thethickness requirement becomesevengreater.Thus,forthiscombination ofhighmaterialtoughness andlowyieldstrength, theapproachoflinearelasticfractureImechanics isnotconsidered valid.Otherapproaches, basedonelastic-plastic fracturemechanics arerequired.

Theapproaches ofJ-Integral, TearingModulus,andnetsectionplasticcollapsecriterion areusedinthisprogram,asdetailedinthefollowing subsections.

RGE-02-004Revision010nute@4 2.1.1J-Integral andTearingModulusAnalysesTheseanalysesfollowthemethodology ofReference 9forthestainless steelaccumulator andpressurizer surgelinesatGinna.Inaccordance withReference 2,LevelDstresseswereusedforthisanalysis.

ASMECodeallowable stressesfortheselinesaregiveninTable2-2.Forthecaseoftheseelastic-plastic crackstability

analyses, onlytheprimarystressesareconsidered, becauseoftherelatively largedeformations accompanying

fracture, whichwouldrelieveanysecondary stresses.

Thisisconsistent withReference 10,whereitisrecognized thatsecondary andpeakstresseshavenoeffectonthelimitloadbecausetheyareproducedbytheactionofimposedstrainsorarelocallyconf'ined andself-limiting.

2.1.1.1StressesTheCodemaximumallowable primarystresses(membrane plusbending)fromTable2-2areequaltoShyFortheaccumulator line,thisgivesPm+Pb=37,600psi.Forthepressurizer surgeline,Pm+Pb=56,400psi.RGE-02-004 Revision0nutscdKNOINIKAS Conservative analyseswerealsodoneforthemostseverestressesattheworstlocationineachline,according tothestressreports(References 3and4).Thesestressesalsoincludethermalstressesandassumealleventsoccurring simultaneously forconservatism.

Fortheaccumulator line,theworstcasePm+Pb=30,133psiatnode8,400(Reference 4),consisting ofthesumofdeadweight,RHRmalfunction, lossofload,andseismic(SSE)stresses, alongwiththeprimarymembranepressurestress@Pm5gl60psi.Forthepressurizer surgeline,theworstcasePm+Pb=34,747psiatnode690(Reference 3),consisting ofthesumofdeadweight,controlrodejectionandseismic(SSE)stresses, alongwiththeprimarymembranestressof5,919psi.2.1.1.2Accumulator LineBothcircumferential andlongitudinal through-wall Icracks,asshowninFigure2-3a)wereevaluated.

FromReference 9,theJ-integral (J)andTearingModulus(T)arecalculated from:20a(Stress)[2]EFactorandT=(Stress)[Y2+2]FactorRGE-02-004 Revision012nutschIKNOINGIXRG where,EFlowstressCrackhalf-length Elasticmodulus()=Stressfactor(Reference 9)[]=Geometryfactor,definedbygeometryparameters X,Y,andY'Reference 9)FromtheASMECodeSectionIIIAppendices, giscom-putedastheaverageofSyandSu(minimumexpectedyieldandtensilestrengths) tobe52.5ksiat100'Fand48.8ksiat300'F.Avalueof50ksiis,thus,usedforthisanalysis.

Similarly, E=28.3X10psi,at100'Fand27X10psi,at300'F.Thus,avalueof27.5X10psiwasselectedforthisanalysis.

Theparameter XisusedinReference 9todetermine thegeometryfactors.Itisdefinedas:a(Rt>/where,R=Piperadius,5inchest=Pipewallthickness, 1inchRGE-02-004 Revision013nuteelhj ValuesofJandTforastressof30,133psiandvaryingcracklengthswerecomputed, asshowninTable2-3,forcircumferential andlongitudinal cracks.Anothercasewascomputedforaflaw7incheslongwithvaryingstresses, inTable2-4.The7inchflawsizewasselectedtobeafactoroftwogreaterthanthelargestflawwhichwillgiveapredicted leakrateof1gpm(Section2.2ofthisreport).Thefactoroftwoisconsistent withthatgivenintheReference 2guidanceforthemarginbetweenaleakingandanunstablecrack.Thiswasdonetodefinethemaximumstressesatwhichthecracksarepredicted toremainstable.TheresultsoftheanalysesinTables2-3and2-4areplottedinFigure2-5forcomparison witharepresen-tativematerialJ/Tcurve(Reference 7).AppliedJandTvaluestotheleftofthematerialcurvearecon-sideredtoresultinstablecrackbehavior(Reference 9).AceilingofappliedJ=24,000in-lb/in.

wasalsoplacedonthisanalysis, sincethisisthehighestvaluerepresented bythematerialresistance curveinFigure2-5.Itshouldbenotedthattheanalysesforlongitu-dinalcracksareespecially conservative, sincestressesotherthanpressurestressesarenotaslikelytoaffectthisorientation offlaw;yettheyareincludedinthisevaluation.

RGE-02-004 Revision014 Conclusions fromTables2-3and2-4andFigure2-5arethatlargevaluesofstressesandcracksizesaretolerated bythesepipesbeforeinstability ispredicted.

Insomecases,thevalidityofthisanalysisisexceededbeforecrackinstability ispredicted.

Thelargestthrough-wall cracksizes,evaluated for,theworstcasestressof30,133psi.,whichremainstablearea22inchlongcircumferential crackanda16inchlonglongitudinal crack.Stressesofatleast55,000psi,and50,000psi,aretolerated for7inchlongthrough-wall circumferential andlongitudinal cracks,respectively.

ThesestressesarejustbelowthemaximumASMECodeallowable primarystressof56,400psi,butarewellabovetheworstcasestressof30,133psi.2.1.1.3Pressurizer SurgeLineAnalysessimilartothosedonefortheaccumulator linewerealsoperformed fortheGinnapressurizer surgeline.AppliedJandTvalueswerecomputedforastressof37,600psi,themaximumASMECodeallowable primarystresswhichisgreaterthantheworstcasestressof34,747psi(Reference 3),forvaryingcracksizes(Table2-5).Sincehigherprimarystressesarenotpermitted bytheASMECode,thecalculations forhigherstressesRGE-02-004 Revision015nutec4 werenotperformed inthiscase.Valuesofg=45ksiandE=25X10psiareused.Theresultsoftheseanalysesareplottedonthestability diagraminFigure2-6,aswasdonefortheaccumulator line.Conclusions fromFigure2-6arethatthrough-wall crackslongerthan20inchesand12inchesforcircumferential andlongitudinal orientations, respectively, arestablefortheappliedstressof37,600psi.2.1.2NetSectionPlasticCollapseCriterion Thenetsectioncollapsecriterion (NSCC)followsReference 10,withseveralminorchangesintheequations used.Specifically, thiscriterion assumesthatfailureisdefinedbyplasticinstability whichoccurswhenthestressinthenetsectionatthecrackreachesthematerialflowstress.Thisapproachiscon-servative foraustenitic stainless steel,sincetheflowstressistakenastheaveragebetweentheminimumexpectedyieldandtensilestrengths.

Inreality,strainhardening ofthismaterialwouldcontinuebeyondthisflowstressandwouldgiveincreased resistance tocollapse, whichisnottakenintoaccountinthisanalysis.

RGE-02-004 Revision016nutechQNQINQtERG Whenaconservative compoundcrack(acombination ofthrough-wall andpart-through wallcracks,asseeninFigures2-7and2-8)isassumed,theanalysisbecomesslightlymorecomplexbecauseofshiftingthepipeneutralaxisbytheangleg(Figures 2-7and2-8).Thiseffectisincludedinthefollowing equations tocomputecriticalcompoundcracksizesforinstability:

d"Pm(m-v)(1--)-(-)ta2(1--)dt2pb=-(1--)[2sing-sinv]mtwhere:aPmPbShiftoftheneutralaxisHalf-crack angleforthrough-wall crackDepthofpart-through wallcrackPipewallthickness FlowstressPrimarymembranestressPrimarybendingstressTheseparameters aredefinedfurtherinFigures2-7and2-8RGE-02-004 Revision017nutechGNOINCGRQ UsingthesamePmandPbstressesdiscussed insub-section2.1.1.1ofthisreport,thepreceding equations weresolvedsimultaneously toproducethefailurediagramsinFigures2-7and2-8.Numerical valuesofthecriticalcracksizesthuscomputedarealsogiveninTables2-6and2-7fortheaccumulator andpressurizer surgelinesatGinna.IConclusions fortheaccumulator linearethatcircum-ferential through-wall cracksof0.318and0.116fractions ofthecircumference arestablefortheworstcasestressof30,133psiandthemaximumASMECodeallowable primarystressof56,400psi,respectively.

Thesearecracks10and3.6incheslong,whicharemoreconservative thantheTearingModulusresultsofsubsection 2.1.1.Suchconservatism hasalsobeenshowninReference 7,wherenetsectioncollapsewasanalyzedbytheJ/Tapproach.

Thus,thenetsectioncollaspecriterion istrulyaconservative estimateofaustenitic stainless steelflawtolerance.

Part-through wallcracksequalto59.61%and21.83%ofthewallthickness weredefinedfortheonsetofinstability fortheaccumulator linewithstressesof30,133psiand56,400psi,respectively.

RGE-02-004 Revision018nutech Conclusions basedonthenetsectioncollaspecriterion fortheGinnapressurizer surgelinearethatcircumferential through-wall flawsequal'to0.248and0.223fractions ofthecircumference arestablefortheworstcasestressof34,747psiandthemaximumASMECodeallowable primarystressof37,600psi.Thesearecracksgreaterthan7incheslong.Part-through wallcracksequalto47.5%and43.5%ofthewallthickness weredefinedfortheonsetofinstability forthepressurizer surgelinewithstressesof34,747psiand37,600psi,respectively.

2.2LeakRatesInaccordance withtheguidanceofReference 2,crackopeningareasandleakrateswerecomputedforthrough-wallcracks2inches(2t)long.Cracksizesrequiredtogiveleakratesof1gpmwerealsocalculated fortheGinnaaccumulator andpressurizer surgelines.Theleakratesarequiteconservative sinceonlypressurestressesareconsidered.

Inreality,otherstresseswillalsotendtoopenthecracksforleakage,especially forcircumferential flaws.RGE-02-004 Revision019nutech 2.2.1Accumulator LineCrackopeningareaswerecomputedforcircumferential throughwallcracksfromthefollowing equations (Reference 2):Ap~(2mRt)G(X)EP(Rt)~Gp(Z)=X+0.16Kg(0<X<1)G(X)=0.02+0.81X+0~3X+0.03Zi(l<)<5)where,Ap=crackopeningareaa=pressurestress,inaxialdirection p1/2~,forinternalpressure, ptTheminimumnormaloperating pressureof750psi(Table2-1)wasemployedinthiscalculation.

CrackopeningarearesultsaregiveninTable2-8.RGE-02-004 Revision020nutechRNOINGGRS Foralongitudinal through-wall crackthefollowing equations areusedtocomputeAp(Reference 2):aA=~(2mRt)G(),)PEpGp(X)+0.625),,(0<><1)G(g)=0.14+0.36g+0.72Z+0.405X,(1<X<5)where,aisthehooppressurestressandtheothertermspareasdefinedforthecircumferential through-wall crack.Theleakflowratesfortheaccumulator linearecomputedforthenon-saturated liquidat120'F,usingtheBernoulli equation(Reference 13):2gap1/2G=p[)Pwhere,G=Flowrate,(ibm/(sec.

-in.))2gc=Densityat120'F=61.71ibm./1728 in.(32.2)(12)in./sec.Pressuredifference, (750-14.7)psi.Theresults,ingpm,areshowninTable2-8.FortheAL,leakratesof0.236gpmand0.516gpmresultfrompressurestressesfor2tlongcircumferential andlongitudinal through-wall cracks.Circumferential andRGE-02-004 Revision021nutechGNOWCRRQ longitudinal through-wall crackhalf-lengths of1.752and1.328inchesarerequiredforleakratesof1gpm.AlltheseresultsinTable2-8includealeakfrictioncoefficient factorof0.6(Reference 14).2.2.2Pressurizer SurgeLineAsimilaranalysistothatfortheaccumulator lineforApvalueswasdoneforthepressurizer surgeline.ResultsaregiveninTable2-8.Forthesaturated liquidtheflowrate,G,isobtainedfromFigure2-9(Reference 15),forthepressureof2235psiandthetemperature of612.2'F(Table2-1).Again,aflowfrictioncoefficient of0.6isemployed(Reference 14).Theresults,ingpm,areshowninTable2-8.Forthepressurizer surgeline,leakratesof0.567gpmand1.245gpmresultfrompressurestressesfor2tlongcircumferential andlongitudinal through-wall cracks.Acircumferential through-wall half-crack lengthof1.314in.isrequiredforaleakrateof1gpm.RGE-02-004 Revision022nutech 2.2.3GinnaLeakDetection Capabilities Forprimarycoolantleakdetection (thepressurizer surgeline),RGaEhasprovidedtheNRCdocumentation (Reference 5)supporting a1gpmleakdetection in1hour.Discussion withRGGE(Reference 6)gavefurtherdetailsregarding theseleakdetection capabilities.

Themethodsconsistof1)anairborneparticulate radio-activitymonitor,whichcanideallydetect0.013gpmwithin20minutes,2)amonitorofcondensate flowratefromtheaircooler,whichcandetect1gpmwithin1hourand,3)achemicalvolumecontrolsystem(CVCS)monitor,whichcandetect0.25gpmwithin1hour.IRG&Ehastwosystemsofleakdetection fortheaccumu-latorline(Reference 6).Leveldetectors consistofhighandlowlevelalarmssetfor1108ftand1134ft~Thedifference is26ftor194gal~Thisresultsinatimeintervalof3.23hrtodetectaleakof1gpmfortheworst-case, wheretheinitiallevelisjustbelowthehighlevelalarm.Therealsoisasumppump("A"pump)whichisactivated fromalevelalarm30.5in.fromthefloorofthe4.5ftx4.5ftsumparea.Thisgivesarequiredvolumeoffluidofabout51ftor385gal.Thisresultsinatimeintervalof6.4hr.todetectaleakof1gpm.RGE-02-004 Revision023nutech Promthepreceding leakdetection systems,itisapparentthatRG&Ecurrently hasthecapability todetecta1gpmleakatGinnaforthepressurizer surgeandaccumulator lines.AsseeninTable2-8,thosecracklengthscorresponding toalgpmleakrategivesignificant marginsagainstcrackinstability, whenactualworst-case stressesareusedtopredictinstability.

Thesmallestmarginisafactorof2.86oncracklengthforacircumferential crackintheaccumulator line.Thismarginisabovethefactorof2givenbytheNRCguidance(Reference 2).Thus,theexistingRGSEleakdetection capabilities appearadequatetosupportthisleak-before-break approach.

2~3Subcritical CrackGrowthRatesThepreceding fracturemechanics analysesandleakrateanalysesprovideinformation forprotecting againstHELBbytheleak-before-break approach.

However,suchinformation mustalsobegenerated toprotectagainstpiperuptureresulting fromthegrowthoflongpart-throughwall,non-leaking cracks.Thisisaccomplished bydetecting andpreventing suchcrackswithaugmented inservice inspection (ISI).AdequateISIintervals topreventpostulated part-through wallcracksfrombecom-RGE-02-004 Revision024nutech ingcritical(unstable) areestablished

'bypredicting subcritical crackgrowthrates.Suchanalysesarepresented inthefollowing subsections, fortheGinnapressurizer surgeandaccumulator lines.2.3.1StressProfilesLoadingconditions forsubcritical fatiguecrackgrowthanalysesweredefinedbyconsidering theboundingcaseinthestressreports(References 3and4).Theboundingcase(mostseverestresses) isatnode690forthepressurizer surgeline(Reference 3).Thesestressprofileswereusedtoenvelopeboththepressurizer surgelineandaccumulator lineforGinna.Thestressprofilesthroughthepipewallthickness areshowninFigure2-10forbothcircumferential andlongitudinal cracks.Infatigueanalysis, thecyclingbetweenminimumandmaximumloadsisconsidered.

Theminimumloadcondition isforplantshutdown, whereloadingconsistsofpipingdead-weight andweldresidualstressesforcircumferential flaws,andisassumedaszeroforlonglongitudinal flaws.Themaximumloadcondition, forallcyclesisconservatively assumedtobethecaseforpressurizer surgelinestresseswithcontrolrodejection(Reference 3).Forcircumferential RGE-02-004 Revision025nutechENIRINGliRQ cracks,themaximumloadcondition consistsofdead-weight,weldresidualstresses, pressure(3015psistresses, andthermalstresses.

Forlongitudinal cracks,themaximumloadcondition iscomprised ofpressureloading.Again,thestressprofilesassociated withtheseloadingsareshowninFigure2-10.WeldingresidualstressesareincludedintheNUTECHcrackgrowthcomputermodel,NUTCRAK(Reference 16).2.3.2CyclingRateThetransients considered todevelopthefrequency ofcycling(betweenthepreceding maximum/minimum loads)areshowninTable2-9(Reference 8).Thesetransients wereonlyusedtoestimatethenumberofcyclesexpectedduringtheplantlife,sincetheassumedloadingsaremoreseverethanthoseassociated withthesetransients.

Onlytransients withsignificant loadsorwhichwereassociated withthesubjectpipinglineswerecon-sidered,asshowninTable2-9.Thisresultedinatotalofapproximately 1200significant cyclesina40yearplantdesignlife,or30cyclesperyear.Thus,theloadcyclingspectrumassumedforthesubcritical crackgrowthanalysesareshowninFigure2-11.RGE-02-004 Revision026nutech 2.3.3CrackGrowthRateAnalysisThepreceding information wasinputtothefollowing crackgrowthlaw(Reference 17):'da(g()ndNdawhere,d=Crackgrowthrate(in./cycle)C.=Materialconstant=2.74x10n=Materialconstant=3.97bK=Rangeinappliedstressintensity factorforeachcycle.TheequationwassolvedbytheNUTCRAKprogram(Reference 16)forcrackdepthasafunctionofnumberofcycles.Stressintensity factorsforcircumferential andlongitudinal cracksareapartofthecalculation output,andincludetheeffectsofmaximum-to-minimum loadingratios.TheresultsofthisanalysisareshowninFigure2-12.Twoinitialcracksizes(depths)wereassumedforpart-Ithroughwallcracksofinfinitelength:ai=0.02inchRGE-02-004 Revision027nutech andai=0.10inch.Ifitisassumedthatacrackofdepth0.02inchcanbedetectedbyISI,andthisisusedastheinitialflawsize,theninsignificant flawgrowthoccursover40years(1200cycles).Asanextremecase,itisassumedthat.aflaw0.10inchdeep(10%ofthewall)isthelimitofISIdetect-ability,andisassumedastheinitialcracksize.Evenforthislargecracksize,longitudinal crackgrowthisinsignificant.

However,circumferential crackextension doesoccur,growingfrom10%ofthepipewallthickness to20%inabout10years.Thisisstillwellbelowthecracksizeofalmost50%ofwallthickness (d/tinTables2-7and2-8forv/m=0)predicted forinstability onthebasesofworstcasestressesandthenetsectionplasticcollapsecriterion.

Thus,itisrecommended thatevenforthisextremecase,anISIintervalof10yearsshouldbeadequatetodetectpart-throughwallcrackspriortoapproaching piperupture.RGE-02-004 Revision028nutechGNOINCERG

3.0CONCLUSION

S ANDRECOMMENDATIONS Theleak-before-break approachforresolution ofHELBfortheGinnapressurizer surgeandaccumulator linesisshowntobefeasibleandpractical.

Criticalcracksizesforruptureofthepipeswerepredicted conservatively byTearingModulusandnetsectionplasticcollapsecriterion approaches.

Theappliedloadswerebasedontheworstcasestresses, usingthestressreports.Through-wall circumferential cracks24.7%ofthepipecircumference (7.75inches)areshowntobestablefortheseloads,basedonthenetsectionplasticcollapsecriterion.

Longitudinal through-wall cracks12incheslongareshowntobestablefortheseloads,usingtheTearingModulusapproach.

Circumferential part-through wallcracksequaltoalmost50%ofthepipewallthickness areshowntobestablebythenetsectioncollapsecriterion.

Thus,anamplemarginoffractureresistance existsinthesepipes.TheanalysesbasedonmaximumASMECodeallowable stressesstillpredictsignificant resistance tofracture, butwithlessmarginforleakdetection andISI.RGE-02-004 Revision029nutech Leakrates,basedoninternaloperating pressurestressesonly,werecomputedforthroughwallflawsoflength2t(2inches).Leakratesrangefrom0.236gpmto1.245gpmforcircumferential andlongitudinal through-wall cracksinbothlines.Theworstcasecracklengthtogiveaminimumleakrateof1gpm,is3.5inchesforacircumferential crackintheaccumulator line.Thisstillprovidesmarginagainstreachingthecracklengthrequiredforinstability (10inches,Table2-6)fortheaccumulator linewithworstcasestressreportstresses.

RGGEcurrently hasbulkleakdetection systemsatGinnacapableofdetecting 1gpmleaksfortheselinesinlessthan6.4hours,andareconsidered adequate.

ISIintervals of10yearsarefoundtobeadequatetopreventpart-through walllongitudinal andcircumfer-entialcracksfromreachinginstability.

Asubstantial marginagainstruptureexistsevenwhenlargeinitialflaws(10%ofwallthickness) areassumed.RGE-02-004 Revision030nutech TABLE2-1PARAMETERS FORLEAK-BEFORE-BREAK ANALYSISOFPRESSURIZER SURGEANDACCUMULATOR LINES00PRESSURIZER SURGELINE(REFERENCE 3)Size=OuterDia.=10.75in.,Thickness

=1in.Material=A376TP316NormalMode=612.2'F,2235psi.pressureControlRodEjectionMode=697.2'F,3015psi.pressureACCUMULATOR LINE(REFERENCE 4)Size=OuterDia.=10.75in.,Thickness

=1in.Material=A376TP316NormalMode=120'F,2235/750psi.pressureLossofLoadMode=120'F,2628/750psi.pressureRHRHxMalfunction Mode=120/300'F, 2235/750psi.pressureRGE-02-004 Revision031nutech TABLE2-2LEVELDASMECODEMAXIMUMALLOWABLE STRESSESUSEDFORANALYSISOFCRACKINSTABILITY FORPRESSURIZER SURGE(PSL)ANDACCUMULATOR LINES(AL)(ASMECODESECTIONIIIgNC3600g1980EDITION)~allowable

~x+~b~tPSL6003i015AL1002,62865,35084,6005,9195I16059,43179,44014,85312,947where:allowable hyA'Shysma11erof(3Sh,2Sy)Sh=17ksi.,Sy=18.8ksi.at600'FSh=18.8ksi.,S=30ksi.at100'FYSAf(1.25Sc+0.25Sh)Sc=18.8ksi.at100'Ff=1forthermalcycles(700aballowable papr/(r-r)(ri=insideradius,ro=outsideradius)P(r+r.)/(r-r.)RGE-02-004 Revision032nutechGNOINGGRtl TABLE2-3APPLIEDJ-INTEGRAL ANDTEARINGMODULUSVALUESASFUNCTIONS OFTHROUGH-WALL HALFCRACKLENGTH,a,FORTHEGINNAACCUMULATOR LINEWITHANAPPLIEDSTRESSOF30,133PSI.CIRCUMFERENTIAL

~a(ai)a(in)CRACK:aStress[Factor,[Y][Y+2'Y]J(-'"P)in30,13330,13330,13330,13330,13330,133246810ll0.890.61.790.62.680.63.570.64~460.64.910.61.381.381.381.381.381.381.21.82.63.34.14.51.83.24.86.28.08.43019041,9593,3155,i496,2162.54~46.68.611.011.6LONGITUDINAL CRACK:30,13330,13330,13330,13330,1332468100.891.792.683.574.460.60.60.60.60.61.381.381.381.381.382.04.899.0714.5321.224.112.23~38.58.5025.72,45616.66,83431.714,59752.426,64880.0RGE-02-004 Revision033nutech TABLE2-4APPLIEDj-INTEGRAL ANDTEARINGMODULUSVALUESASFUNCTIONS OFAPPLIEDSTRESSFORTHEGINNAACCUMULATOR LINEWITHATHROUGH-WALLHALFCRACKLENGTHOF3.5IN.CIRCUMFERENTIAL CRACK:a(ksi.)a(in.)StressFactor~Y1[Y2+2gY.Y'J(in-lb)inT20.71.01.11.2355055603.53.53.53.51.561.561.561.562.0516.137.989.11.651.651.651~652.852.852.852.851,0775.88,46145.919,917108.046,824253.9LONGITUDINAL CRACK:0.71.01.13550553.53.53.51.562.054.01.5616.14.01.5637.94.011.011.011.02i61222.620,511177.148,285416.9RGE-02-004 Revision034nutechGNOINGGRG TABLE2-5APPLIEDJ-INTEGRAL ANDTEARINGMODULUSVALUESASFUNCTIONS OFTHROUGH-WALL HALFCRACKLENGTH,a,FORTHEGINNAPRESSURIZER SURGELINEWITHANAPPLIEDSTRESSOF37,600PSI.CIRCUMFERENTIAL CRACK:~a(ai.)a(in.)aiStress~Factor)Y2)(Y2+2~YYi)J(in-1b)illT37,60037,60037,60037,60037i6002468100.891.792.683.574.460.840.840.840.840.844.64.64.64.64.61.21'2.63.34.11.83.24.86.28.08942i6835,8139,83715,2778.2814.7222.0828.5236.80LONGITUDINAL CRACK:37,60037,60037,60037,60020.890.844.641.790.844.662.680.844.683.570.844.62.04.14.8912.9.0723.14.5338.1,4907,28820,27743,31118.8655.20105.80174.80RGE-02-004Revision035notechIINQINI1GRQ TABLE2-6FAILURECRACKSIZES*FORPOSTULATED COMPOUNDCRACKINGINNAACCUMULATOR LINEg.BASEDONNETSECTIONPLASTICCOLLAPSECRITERION Pm+Pb30,133PS'm'b56,400PSI.O.l0.20.30.4v/m0.3180.2910.2570.2160.1650.05O.l0.150.2v/n0.1160.0940.0700.0430.0120.5'.0960.596100.2183*TERMSDEFINEDINFIGURE2-7RGE-02-004 Revision036nutechGNOINGGRG TABLE2-7FAILURECRACKSIZES*FORPOSTULATED COMPOUNDCRACKINGINNAPRESSURIZER SURGELINEgBASEDONNETSECTIONPLASTICCOLLAPSECRITERION P+Pb34747PSI.P+Pb37,600PSI.0.10.20.30.4v/n0.2480.2150.1760.1270.063O.l0.20.30.4v/m0.2230.1890.1480.0960.0290.474900.4350*TERMSDEFINEDINFIGURE2-8RGE-02-004 Revision037nutechGNQINEKGRQ SA8Q4WIIOlOIh)0IOOATABLE2-8LEAKRATERESULTSFORCIRCUMFERENTIAL THROUGH-WALL CRACKS(CTWC)ANDLONGITUDINAL THROUGH-WALL CRACKS(LTWC)INTHEPRESSURIZER SURGEANDACCUMULATOR LINES(PSLANDAL)FORNORMALOPERATION PRESSURESTRESSESLineCrackPressureStressa(psi)PCrackAreaAp(in.)forHalf-'Lengtha=1in.LeakRate***(gpm.)fora=1in.CrackHalf-length a(in.)for1gpmLeakRate***PSL*AL**CTWCLTWCCTWCLTWC4,8899,7781,6403,2810.001130.002480.000380.000830.5671.2450.2360.5161.314<11.752l.328Saturated LiguidNon-saturated Liquid***BasedonFrictionFactorCf-0.6C'0 Operating CcleTABLE2-9TRANSIENTS CONSIDERED INSUBCRITICAL CRACKGROWTHRATEANALYSESFORPRESSURIZER SURGEANDACCUMULATOR LINES(REFERENCE 8)Occurrences in40r.DesinLife1.StartupandShutdown2.LargeStepDecreaseinLoad(withsteamdump)3.LossofLoad(withoutimmediate turbineorreactortrip)4.LossofPower(blockout withnaturalcirculation inReactorCoolantSystem)5.LossofFlow(partiallossofflow,onepumponly)6.ReactorTripfromFullPower7.Hydrostatic Test(beforeinitialstartup,andpostoperation) 8.HighHeadSafetyInjection 20020080408040055501105Assume1200Significant Cyclesin40yr.DesignLife(30cycles/yr.)

RGE-02-004 Revision039nutech 10Ice@If<<oo OoILOOSOCLOCoa<<NeoCCLlarecitLfoCACO6l~Vl>'J0O8C)Ottf~esses~IwlI~eol<<t~ota~awaI<<~Ia&<<~Itws<<NfaarwCO<<SINO'aotfetc<<LcceAececaKcafIotaercasoctope,awOKIILI<<OIoAraloooloelslf QlILt'Clsaftso'tCIta'fora'co r--akca.coc IIL'.IOC'swwew<<llwS~fC)Issf<<f0MtaWII~~)g(eaaa<<fee<<ow Qcwswso<<<<oNoew~0CW<IO<<ae<<e 8<<OIWI+e<<t<<n~Iowa<<+Mls<<le+NIfSSc~4C.l~KIIII~,Iwoe<<IPOONNNWNCO<<lN~NSSgILCNI'IIONllollOINNILCOINCWII INNNONWNINWILINNINI~c<<ltNNNlINWIrolllsocICtltfLICLare<<saaec sstca<<L'sa aacctaaocto<<fccooLN<<cceocwof Iescaot0ll'Lo<<s<<Ito<<eciaIAea<<ILatCC.Ia<<esca.SO%COSOKIOLINO~ltsaoa<<o<<LIa<<etcoeoKa~taowasaL~soaeosoto<<.LML<<CCOOCIIIfels<<OWflfCat\1~PNIIOOv<<Csso~sssooew~oewas~aeCL1ucttLAtoLCEttOtILttO1'et3og.4yK)ltoLlt1OCCSICQCAOLLICLOC1ISIC.COATCteoILOCLteCsa~I(eLI~CONCNCACCCCIL1CS.OIC C7fjl44S24W3SI-353IFigure2-1PRESSURIZER SURGELINEFRGE82.01 eaP.lVlOIbJ0IpoO~OICR4rsor.r.(UCOssILIssles~I>>'rsawllMlleas*seep'rrsalle>>ase~l>>alall>>>>>>ar~It>>I~UeMI~I'I'ro~IKSWCseeOl~rwe~ot~50C405'leo440IllUhecootoe~OtOstlaT>>lo to>>wt>>>>LsosLl>>wc>>secLwoteeeta>>eacsar>>a Tsa>>LC>>Tjoocoatl~UrW4>>tt~I\arrlfSi5(cseerstlssIPPOIS>>law>>Ia"'Cisev~t~O.Irss400~<QSP~ar~>>r4~r~laa4'~st,'tILIIULtcllet~WIOUIllsteel~OM>>Wer~~~oseslUUOse)('ceerwaeew 9ce>>sos>>ear WUWwss'P~r+IMUU+leo>>srQ>>l~sso~P~ifQCDCrAllSCCCCarscccsm.sect sOtttetectl Iswtsa>>etwoe>>araMLcloasecsoetliltteltsse>>l>>TerMULI>>s>>

tlsealtteeel11OIteswecSale,>>tea>>ar>>l>>

stsosltlteetcslt~TUSteaw>>Lree>>IIwre'ec'1>>ow>>st UlaLlwsleeLwt'LIS100SRICRSJttetfC'rtREBEC;rEh UIIIII,~ISOICSQILISWIISSCUD0SSSg0IISIteSUWI0IUSCsISsSCIWISSS0SeelWISUSILss'Kss~IU>>sswswssslssss,54>>0la>>tIre.>>0~44w>>~Is>>rawse>>jwasaseateveocssstt~sit>>0cLttteaecoteOwl>>aSILT>>Moawcrwcl>>>>el40 eeoIMIC~raastsclsrAloccsert s.Ic'.rsctcssl555ccIIrI4IItCtFigure2-2SAFETYINJECTION FROMACCUMULATOR AFRGE82.02 2a26a)THROUGH-THICKNESS CIRCUMFERENTIAL ANDLONGITUDINAL CRACKSOFLENGTH2a.b)PART-THROUGH THICKNESS CIRCUMFERENTIAL CRACKOFDEPTHa.FGRE82.03 Figure2-3REPRESENTATION OFPOSTULATED CRACKSINPIPESFORFRACTUREMECHANICS LEAK-BEFORE-BREAK ANALYSISRGE-02-004 Revision042nutechSNOINGSRS 480400TP304J-RCURVES()ISREFERENCE FORDATAPTS(giveninreference 7)320g240160MIY9")12)(16)(17)g(10)80JIc=5260in-lb/in.

FOR2TP316at600F,FORCOMPARISON TOTP304DATABASE(reference 7)00.00.40.81.21~62.02.4CRACKEXTENSION ha,in.FR0E82.04Figure2-4J-INTEGRAL RESISTANCE CURVESFORAUSTENITIC STAINLESS STEEL(reference 7RGE-02-004 Revision043nutechGNOINGGAG 280240200x160~Q120IIVI~s8040ALLONG.CRACK,30,133PSI8n/////////a=10"/ALCIRCCRACK,//30,133PPSIALCIRC.MATERIALCURVE(ref.7)ALLONG./CRACKa=3.5"/'NSTABLEcr=50,000/PSIa=SS,OOOPSISTABLECONSTANTSTRESS(30,133psi),CRACKSIZEVARIES004080120160200240CRGE83.01CONSTANTCRACKSIZE(a.=3.5"),STRESS.VARIESFigure2-5J-INTEGRAL/TEARING MODULUSSTABILITY DIAGRAMFORGINNAACCUMULATOR LINEWITHTHROUGH-WALL CRACKSRGE-02-004 Revision044nutech 280240TP304'J-TMATERIALCURVE(reference 7)PSL,LONGITUDINAL'CRACK, a=37,600PSI200160PSLCIRCUMFERENTIAL CRACKa=37,600PSIa=6"STABLEUNSTABLEa=10"1208040o,b.-CONSTANTSTRESS(37,600psi),CRACKSIZEVARIES4080120160200240CRGE83.02 Figure2-6J-INTEGRAL/TEARING MODULUSSTABILITY DIAGRAMFORGINNAPRESSURIZER SURGELINEWITHTHROUGH-WALL CRACKSRGE-02-004 Revision04SnutechENOINOGRO CRACKN1.0Rg0.8~tQ'Q~HM~H0~6~R~Hr~o~0.4+OHgog0.2P~+Pb=30,133PSI(ACTUALWORSTCASES'ROM STRESSREPORT)Pm+Pb56/400PSI(ASMESCTIIICL2MAXIMUMALLOWABLE) 0~0.000.20.40.60.81.0FRACTION.

OFCIRCUMFERENCE, v/m(THROUGH-WALL CRACK)CRGE83.03 Figure2-7FAILUREANALYSISDIAGRAMFORPOSTULATED COMPOUNDCRACKINGINNAACCUMULATOR LINE,BASEDONNETSECTIONPLASTICCOLLAPSECRITERION RGE-02-004 Revision046nutech CRACKvN1.0RZUra0.8~N'0~HM~Hra~~R~ra0.6oHg+H~OPo0~4HgogRE0.2p~+pb=34s747PSI(ACTUALWORSTCASESFROMSTRESSREPORT)Pm+Pb37I600PSI(ASMESCTIIICL2MAXIMUMALLOWABLE) 0.00.00.20.40.60.81.0FRACTIONOFCIRCUMFERENCE, v/m(THROUGH-WALL CRACK)CR6E83.04Figure2-8FAILUREANALYSISDIAGRAMFORPOSTULATED COMPOUNDCRACKINGINNAPRESSURIZER SURGELINE,BASEDONNETSECTIONPLASTICCOLLAPSECRITERION RGE.-02-004 Revision047nutechGNCIN4GRG QTlltlTCD LPVNNOART5lllltlTlD YAtOIKV~lTISXO~~~o0lNZOXO4XIRON0MNOSCINkllNNOLÃ0Laxraaunoaawunq<~g>FRGE82.05 Figure2-9DIAGRAMFORMAXIMUMSTEAM/WATER FLOWRATETODETERMINE FLOWRATEFORSATURATED LIQUIDINTHEPRESSURIZER SURGELINE(MOODYMODEL-REFERENCE 15)RGE-02-004 Revision048nutechGNOINQGRD 8.103ksi1.485ksiIDODIDODa)PRESSURE(p=3015psi)CIRCUMFERENTIAL CRACK22.952ksib)DEADWEIGHT,CIRCUMFERENTIAL CRACK34ksiIDODIDODc)THERMAL,CIRCUMFERENTIAL CRACK-34KSId)RESIDUAL(reference16),CIRCUMFERENTIAL CRACK16.206.ksi IDOD')PRESSURE(p=3015psi)LONGITUDINAL CRACKFigure2-10STRESSPROFILESFORBOUNDINGCASE(reference 3)FORSUBCRITICAL CRACKGROWTHPREDICTIONS FRGE82,06 RGE-02-004 Revision049nutech DEADWEIGHT+RES1DUAL+PRESSURE+THERMALSTRESSDEADWEIGHT+RESIDUALSTRESSTIME(months)a)CIRCUMFERENTIAL CRACK(30cyclesperyear)PRESSURESTRESSTIME(months)FRGE82.07 b)LONGITUDINAL CRACK(30cyclesperyear)Figure2-11CYCLICLOADINGCONDITIONS ASSUMEDFORCONSERVATIVE SUBCRITICAL CRACKGROWTHRATEANALYSISOFPRESSURIZER SURGEANDACCUMULATOR LINESRGE-02-004 Revision050nutech 0.5~0.4CIRCUMFERENTIAL CRACKgl0.3WQo0.20.10.0a-002"l.CIRCUMFERENTIAL

-LONGITUDINAL a-.=0.10"3.a=0.1044"fAT1200CYCLESLONGITUDINAL CRACK01200CYCLES(10cycles/year) 100200300400CYCLES(10cycles/year) 500FRGE82.08 Figure2-12PREDICTED SUBCRITICAL CRACKGROWTHRATESFORCIRCUMFERENTIAL ANDLONGITUDINAL CRACKSWITHASSUMEDINITIALDEPTHS(ai)OF0.02INCHESAND0.10INCHESFORTHEPRESSURIZER SURGEANDACCUMULATOR LINES

4.0REFERENCES

2.3.4~5.6.NUREG-0821, "Integrated PlantSafetyAssessment, Systematic Evaluation Program,R.E.GinnaNuclearPowerPlant",RGEDocketNo.50-244,DraftReport,U.S.NRC,May1982.Enclosure 3oftheSafetyEvaluation ReportforR.E.Ginna,forwarded toRGEbyNRCletter,datedFebruary22,1982.W.A.MassieandM.J.Harper,"PipingStressAnalysisReport,Pressurizer SurgeRC-200",R.E.GinnaNuclearPowerPlantSeismicUpgrading Program,Westinghouse ReportSDTAR-80-05-10 Rev.1,January1981.W.A.MassieandV.H.Mehta,"PipingStressAnalysisReport,SafetyInjection System,Section200",R.E.GinnaNuclearPowerPlantSeismicUpgrading Program,Westinghouse ReportSDTAR-80 26,Rev.0,March1981.SEPTopicV-S,"ReactorCoolantPressureBoundaryLeakDetection,"

GinnaSER,February8,1982.Telecon,G.Wrobel(RGE)toJ.F.Copeland(NUTECH),

"RGE-02,LeakDetection Sensitivities/

ApproachatGinna,"Februaryll,1983,NUTECHFileNo.100.2602.0001.

7~8.9.10.K.H.Cotter,et.al.,"Application ofTearingModulusStability ConceptstoNuclearPiping,"EPRIReportNP-2261,February1982."ThermalTransients andCategories,"

GinnaNuclearPowerPlant,AppendixH,RGGE,July15,1975.R.Johnson,"Resolution oftheReactorVesselMaterials Toughness SafetyIssue,"Volume1,forcomment,AppendixB,NUREG-0744, September 1981.D.A.Hale,etal.,"TheGrowthandStability ofStressCorrosion CracksinLarge-Diameter BWRPiping,"Volume2,AppendixA,EPRINP-2472-SY, July1982.H.Tada,etal.,"Stability AnalysisofCircumferential CracksinReactorPipingSystems,"

NUREG/CR-0838, June1979.RGE-02-004 Revision052nutechENGINEERS 12.J.D.LandesandJ.A.Begley,"TheEffectofSpecimenGeometryonJIC,"FractureToughness, Proceedings ofthe1971NationalSymposium onFractureMechanics, PartII,ASTMSTP514,1972,pp.24-39'3.P.H.F.Pao,FluidDnamics,c.1967.14.I.H.Shames,Mechanics ofFluids,McGraw-Hill, NewYork,c.1962,p.162.15.F.J.Moody,"MaximumFlowRateofaSingleComponent, Two-Phase Mixture,"

Trans.ASME,February, 1965,pp.134-142.16."NUTCRAKUser'sManual,"NUTECHFileNo.08.039.0005.

17.R.Huet,etal.,"StressCorrosion CrackingofType304Stainless SteelinHigh-Purity Water:ACompilation ofCrackGrowthRates,"EPRINP-2423-LD,June1982,Figure2-15.RGE-02-004 Revision053nutechGNOINGGRG