ML17329A067

From kanterella
Revision as of 12:34, 4 June 2018 by StriderTol (talk | contribs) (Created page by program invented by StriderTol)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to navigation Jump to search
DC Cook Unit 2 Low Temp Overpressure Protection Sys Setpoint Evaluation.
ML17329A067
Person / Time
Site: Cook  American Electric Power icon.png
Issue date: 06/30/1989
From:
WESTINGHOUSE ELECTRIC COMPANY, DIV OF CBS CORP.
To:
Shared Package
ML17329A065 List:
References
NUDOCS 9106270146
Download: ML17329A067 (687)


Text

92062701469PF'DRADOCK05000315PDROA~>AMERICA'NELECTRICPOWERSERVICECORPORATIONgAPPROVEDtNGENERALC1APPROVEDEXCEPTASNOTEDClNOTAPPROVEDCIFORREFERENCEONLYD.C.COOKUNIT2LOWTEMPERATUREOVERPRESSUREPROTECTIONSYSTEM(LTOPS)SETPOINTE'IALUATIONJUNE1989MESTENGHOUSEELECTRICCORPORATION8921o:1d/OSQBB9

'k$picIM4P,v'IP40&M4jIr'7+4*stkakwt4elf,>}aE<4i,1t4' INDEXSection~PneNo.Introductlone~~~~~~~~~~~~~~~~~~~e~~~~~~~~e~~~~~e~~~~~~ee~~~~e~~~~~~~~~~eS~1ummaryofResults......................................................S.2S1.0DescriptionoftheLowTemperatureTransients.......................l.l1~e~.1GeneralOescription..................................,...............1.21.2Operation.......................................1.3.PotentialOverpressureTransients...............1.3.1SummaryofHassInputTransients..........1.3.1.1InadvertentSafetyInjection......1.3.1.2Charging/LetdownFlowHismatch....-,1.3.2SummaryofHeatInputTransients,........."1.3.2.1ActuationofPressurizerHeaters..~e~e~~~~~~~~~~~~~~~~1~2~~~~~~~~~~~~e~~~~~~~1~5~~~~e~e~~~~~~~~~~~~ele5~~~~~~~~~~~~~~~~~~~ele5~~~~~~~~~e~~~~~~~~~~le7~~~~~~~~~~~~~~~~~~~el~7~~~~~~~~~~~~~~~~~~~ele71.3.2.2LossofRHRSCooling..........................1.3.2.3RCPStartupWithTemperatureAsymmetry........1.3.2.4RelativeSeverityoftheHeatInputTransients1.4SummaryofTransientEvaluation..............................~~~~~~~~1e8~~en~~~~1~8~~~~e~~le10~~~~~~ele102.0DescriptionoftheLTOPSSetpointAlgorithm.2.1PressureLimitsSelection.....,.............~~~~~~~~~~~~e~~~~~~~~~~e2e1e~~e~~~~~~~~~~~~~~~~~~~~2~12.2HassInputConsiderations...,...............2.3HeatInputConsiderations...................2.4FinalSetpointSelection....................~~~~~~~~~~~2e7~~~~~~~~~~~~~~~~~~~~~~~e2e7~~~~~~~~~~~~~~~~~~~~~e~~2~83.0LTOPSSetpointAnalysisforD.C.CookUnitZ.3.1OperationalLimits..............,.............3.2PORVStrokeTime.............................~~~~~~~~~~~~~~~~~~~~~~e3~1~~~~~~~0~~~~~~~~~e~~~I1302~ee~~~~~~~~~~~~~e~~~~~e3~5~~~~~~~~~~~~~~~~e~~~~~e3e5~~~~~~~~~~~~~~~~~~~~~~~3~6~~~~e~~~~~~~e~~~~e~~~e3~11~~~~~'~~~~~~~~~~~~~~~~3~18~~~e~~~e~~~~~~~~~~~~~~3e19~~~~~~~~~~~~~~~~~~~~~e3~203.3PORVOperation........................,......3.4HassInputConsiderations....................3.5HeatInputConsiderations....................3.6SpecificationsforHassInputTransients.....3.7SpecificationsforHeatInputTransients.....3.8.SetpointEvaluation..........................89216:1d/080889 I'1~~

I'ectionINDEX(Cont'd)~PaeNo.4.0CorrelationtoHOGLTOPSSetpointHethodology...4.1HOGHethodologyLTOPSSetpoints.................4.2LOFTRAN/MOGCorrelation.........................4.3ImpactofSteamGeneratorTubePlugging.........~~~~~~~~~~~~~~~~~~~~4~1~~~~~~~~~~~~~~~~~~~~4~1~o~~~~~~~~~~~~~~~~~4~10~~~~~~~~~~~~~~~o~~~4~148S21e:1d/060S89 WggI~g~,SIAe~gr,I!.,=p.~4,--y>~~ip)fjg+g<~"%.ljLVt+;'YflQE,<+4)l'~"pl'W'ltk~k~~~++a<<~'IA(

D.C.COOKUNIT2LOWTEMPERATUREOVERPRESSURE'PROTECTIONSYSTEMSETPOINTANALYSISINTRODUCTIONUSNRCRegulatoryGuide1.99Revision2,"RadiationEmbrittlementofReactorVesselMaterials,"datedHay,1988becameofficialwithit'spublicationintheFederalRegisteronJune8,1988.TheguiderevisesthegeneralproceduresacceptabletotheNRCstaffforcalculatingtheeffectsofneutronradiationembrittlementofthelowalloysteelscurrentlyusedforlightwatercooledreactorvessels.0AppendixGof10CFRPart50providesthefracturetoughnessrequirementsfor'""reactor>pressurevesselsundercertainconditions.ToensurethattheAppendixGlimitsarenotexceededduringanyanticipatedoperationaloccurrence,technicalspecificationpressure-temperaturelimitsareprovidedduringlowtemperatureoperations.Theembrittlementalgorithmspecifiedbyrevision2ofRegulatoryGuide1.99ismoreconservativethanrevision1,andrequiresthattheselimitsbere-calculated.TheLowTemperatureOverpressureProtectionSystem(LTOPS)providesprotectionagainstexceedingthevesselductilitylimits,asexpressedbytheAppendixGpressure-temperaturelimits,duringcoldshutdown,heatup,andcooldown"""'operatio'ns."'The'limitsresultingfromimplementationofthenewrevisiontoRegulatoryGuide1.99,requiresthattheLTOPSsetpointsbere-evaluated.Thepurposeofthisreportis,inpart,todocumentthere-evaluation.ThisreportincludesastudyofthesensitivityoftheLTOPSsetpointonpressurizerPORVopeningtime,andacorrelationthatbenchmarkstheresultsoftheanalysistothatofthealgorithmdescribedinthereport"PressureMitigatingSystemsTransientAnalysisResults"(July1977).ThisreportwaspreparedfortheWestinghouseOwnersGroup(acronymed"WOG")onReactorCoolantSystemOverpressurizationbyWestinghouseElectricCorporation.ThepurposeofthecorrelationistoprovideAmericanElectricPowerCorporationa8921e:1d/060789S.1 kVsf'tffi 4'4meansofdeterminingLTOPSsetpointsequivalent'othoseobtainedfromthe~~~~~~~~~~~~~relativelysophisticatedLOFTRANbasedanalysisdescribedhere,byusinga...simplemethodology;i.e.,theHOGreport.Thecorrelationremainsvalidaslongascertainplantparametersareunchanged.Theseare:instrumentationtimedelays,pressurizerPORVflowcharacteristics,andpressurizerPORVfullflowC(v).Theorganizationofthisreport,apartfromtheintroductorycommentsandthesummarystatement,isinfoursections:thefirstandsecondsectionsdescribe,respectively,thejustificationforthedesignbasistransientsandthealgorithmusedforthesetpointanalysis;thethirdsectiondocumentstheanalysisspecificto0.C.Cookunit2,andthefourthsectionprovidesthecorrelationtotheHOGmethodologyreferencedintheintroductiontothisreport.SummarofResultsTheresultoftheanalysisissummarizedbyF1guresS.1andS.2,illustrating,respectively,thedependencyofthemaximumallowedLTOPSsetpointonPORVopeningtimeforreactorvessel'xposuresof12EFPYand32EFPY.Aminimumsetpointlimit,forRCPnumber1sealprotection,doesnotexist.Thisisduetothefactthat,giventhe4secondPORVclosuretime,thereisnotenough,separation(whitespace)betweenthesteady-state'pressure-temperaturelimitandtheminimumRCSpressurerequirementsforanRCPstart,toumbrellathepressureswingresultingfromeitheraheatinjectionormassinjectionevent.Astheclosuretimedecreases,thepressureundershootwouldbecomelesssevere.Atthecurrent435psigsetpoint,withsinglePORVoperation,thepeakpressurewouldremainbelowtheAppendixGlimitprovidedthePORVstrokeopentimesremainedbelow6.5seconds,withvesselexposuresto12EFPY;or3.5seconds,withvesselexposuresto32EFPY.FigureS.3providesthecorrelationthatbenchmarkstheresultsofthe"LOFTRAN"basedanalysistotheresultsobtainedfromapplicationofthealgorithmdescribedinthe"MOG"report.Thesecurvesareusedtocompensateforsomeoftheconservatismsornonconservatisms,dependingontheselectedPORVstrokeopentime,inherentintheHOGmethodology.Thesecorrelations8921e:1d/060989S.2

'i\'MllI1

~~~~~~~~~~~.areindependentof,AppendixGlimits,andwillremainvalidaslongascertainplantparameters(pressurizerPORVflowcharacteristicsandC(v.),andinstrumentationsignaldelays)areunchanged.UtilizationofthecurvesrequiresthattheLTOPSsetpointfirstbedeterminedusingthe"MOG"methodology.AtthePORVopeningtimecorrespondingtothatselectedforthe"MOG"calculation,determinethe"LOFTRAN"analysissetpointfromtheordinatebylinearlyinterpolatingbetweenthetwocurvesboundingthepredetermined"MOG"setpoint..8921e:1d/060789S.3

~LFl~goqg@141 Notest1)l4Pressut~lnstrueentError2)StnglePORVOperetton3)PORVClosureTtne~4.ISec.4)ReectorVesselExposure~12EFPY6sI58.8OeqFRCSTe~p.i85.8OegFFigureS.lPORYLTOPSetpointvs.Ya1veOpeningTimeat12EFPY8921~:1d/060289S.4 04A.liltt~I.pleiad5.Agf+tl1I' Roice-1)HoPressurelnstrunentErrof2)SinglePORVOperation3)PORVClosureTine~4.8Sec.4)ReactorVesselExposure~32EFPYICC)Sb.bOegFRCSTemp.=,E.,85.bOegFFigureS.2.PORVLTOPSetpointvs.VaIveOpeningTimeat32EFPYd921~:1d/0602S9S.S F(OIVgr.y5peVIW~ePi,,ittIpC~

M06LTOPSSetpnt(paiq)~FigureS.3LOFTRAN/MOGLTOPSSetpointCorrelationvs.PressurizerPORVOpeningTimeM21e:1dj060289S.G a0~r.

1.0OESCRIPTIQNOFTHELOWTEMPERATUREPRESSURETRANSIENTSOverpressureprotectionforthereactorcoolantsystem(RCS)isachievedbymeansofself-actuatedsteamsafetyvalveslocatedhighinthesteamspaceofthepressurizer.ThesesafetyvalveshaveasetpressurebasedontheRCSdesignpressureandareintendedtoprotectthesystemagainsttransientsinitiatedintheplantwhentheRCSisoperatingnearitsnormaltemperature.Toavoidbrittlefractureatlowreactorvesselmetaltemperatures,theallowablesystempressureisprogressivelyreducedfromthenominalsystemdesignpressureastemperatureisdecreased.Therefore,supplementaloverpressuremitigationprovisionsforthereactorvesselmustbeavailablewhentheRCSandhencethereactorvessel,isatreducedtemperatures.Thissupplementalprotection,utilizingthepoweroperatedreliefvalves(PORYs),.-isknown.astheLowTemperatureOverpressureProtectionSystem.(LTOPS).ThePORVsaredesignedto.limittheRCSpressureduringnormaloperational,transientswhenthereactorisatpowerbydischargingsteamtothepressurizerrelieftank(PRT),thusavoidingtheneedforthecodesafetyvalvesto.function.The,flowcapacityandstroketimeofthePORVsisselectedtoavoidareactortripduringalargesteploaddecrease.Inaddition,thevalvesareutilizedforpressurerelief(water,gas,oramixture).asapartof.theLTOPS,,andwhenperformingthisfunctionwillalso.dischargetothePRT.ThesetpointfortheLTOPSfunctionisselectedsothat.if.one,valvefails,.toactuatewhenrequired,the,secondvalvewillbeabletomitigatethetransient.Normally,whentheRCSisatatemperaturebelow350F,theRCSisopentotheResidualHeatRemovalSystem(RHRS)forthepurposesofremovingresidualheatfromthecore,providingapathforletdowntothepurificationsubsystem,andcontrollingtheRCSpressurewhentheplantisoperatingina.watersolidmode.TheRHRSisprovidedwithselfactuatingwaterreliefvalvestopreventoverpressurizingthisrelativelylowpressuresystemduetoeventsoriginatingeitherfromwithinthesystemitselforfromtransientstransmittedfromtheRCS.TheRHRSreliefvalveswillmitigatepressuretransientsoriginatingintheRCS,tomaximumpressurevaluesdeterminedbythereliefvalvesset89216:1d/0602S9 0,h>~$AV>f-$eH 4pressureplusapressureaccumulationabovethesetpressuredependentontheliquidvolumemagnitudeofthetransient.ThelowpressureRHRSisnormallyisolatedfromthehighdesignpressureRCSduringoperationattemperaturesabove350'Fbytwoisolationvalvesinseries.BecauseoftheNRCrequiredautomaticclosurefeaturedesignedtopreventinadvertentoverpressurizationoftheRHRS,spuriousclosureoftheRHRSisolationvalvesisacreditableevent.TheLTOPSisintendedtoprovideoverpressuremitigationfortheRCS,consideringthosetransientswhichmightoccurwhentheRHRSisolationvalvesareinadvertentlyclosed,thusisolatingtheRHRSwaterreliefvalvesfromtheRCS.1.1GENERALOESCRIPTION'~'The"LTOPS'"is'designed'oprovidethecapability,duringrelativelylowtemperaturereactorcoolantsystemoperation,topreventtheRCSpressurefromexceeding10CFR50AppendixGlimits.TheLTOPSisprovidedinadditiontotheadministrativecontrolstoprevent,overpressuretransientsandasasupplementtotheRCSoverpressuremitigatingfunctionoftheresidualheatremovalsystemwaterreliefvalves.Thesystemisdesignedwithredundantcomponentstoassureitsperformanceintheeventofthefailureofanysingleactivecomponent.Thepoweroperatedrelief.valveslocatednearthetopofthepressurizer,together'with'additionalactuationlogic.fromthewiderangepressurizerchannels,areutilizedtomitigatepotentialRCSoverpressuretransientswhichmightoccur-iftheRHRwaterreliefvalvesareisolatedfromtheRCS.LTOPSprovidestheadditional'reliefcapacityforthespecifictransientswhichwouldnotbemitigatedbytheRHRSreliefvalvesandtherebymaintainthesystempressurebelowthelimitsspecifiedbyAppendixGrequirements.,1.2OPERATIONDuringnormalplantheatup,theRCSisopentotheRHRSandisoperatedinawatersolidmodeuntiIthesteambubbleisformedinthepressurizer.During8821a:1d/060289 Ilt41~tt'IIC~*

l7r~IJ%r')>~.~)~PII--igtheselow-temperatureIow-pressureoperatingconditions,theLTOPSisarmedandinareadystatustomitigatepressuretransientswhichmightoccuriftheRHRSisinadvertentlyisolated.Afterthesteambubbleisformedandthepressurizerwaterlevelisatits'ormalvalueforno"loadoperation,theRHRSismanuallyisolatedfromtheRCSandtheplantheatupcontinues..Pressuresurgecontrolisprovidedbythesteambubble.tthenthereactorcoolanttemperaturehasincreasedaboveapresetvalue(152'F,inthecaseofD.C.CookUnit2)theLTOPSisdisarmed.DuringanormalplantcooJdown,theLTOPSisarmedasthereactorcoolantrtemperatureisdecreasedbelowthepresetvalue.Atthistimethereisasteambubbleinthepressurizerandthewaterlevelisatthenormallevelforno-loadoperation.,TheRHRShasbeenplacedinservicebyopeningthesuction~isolationvalves;thusmakingtheRHRSwaterreliefvalvesavailableto'itigate"ressuretransients.Rhenthecoolanttemperaturehasbeenreducedtoabou160'ytheoperationoftheRHRS,thesteambubblemaybequenchedandthereactorcoolantpumpsstopped.Fromthispointoninthecooldown,theplantiswatersolidandiftheRHRSbecomesinadvertentlyisolated,theLTOPSwillbeinanactivestatusreadytomitigatethosepressuretransientsthatmightoccur.MhentheRCS,isoperatedinthewatersolidmode,thepressureisautomati-..callycontrolledbythelowpressure'letdownvalve.in'the=Chemical-andVolumeControlSystem(CYCS).Thisvalvesensesthepressureintheletdownline(refFigure1.1)andmaintainsthepressureattheselectedcontrolvaluebythrottlingtheletdownflowfromtheRCS.Atthistime,thechargingflowintotheRCSissetataconstantvalue.andiscontrolledbythechargingflowcontrolvalve.Itshouldbenotedthat.thepressurebeingcontrolledisthatintheletdownlinewhichthenindirectlycontrolsthepressureintheRCS.However,ifthepressuredropthroughtheRHRSandthebypasslineintotheCVCSischangedbythrottlingofvalvesorchangingtheflowratethroughtheRHRS,theRCSpressurewillalsochangesincethelocationofthecontrolledpressureisin.theletdownline.8921a:1d/0602891.3

'~~'Ig;4I1r~j\

ChargingF1'Conlroll~PressurePurlflcatieP0~RIIRRellefYalveRIIRPumpIIXRIIRIIXIIXeCVIICfgI.ethel4PressuretontrolI'igure1.1TypicalLowPressureCoolant/PurificationFlogPath892Ie:1d/06078$

'EelII'IgI1;fi'~~-wg0;t 7~~1.3POTENTIALOVERPRESSURETRANSIENTS",Duringlowtemperatureoperations,reactorcoolantsystemoverpressurizationtransientscanbecausedbyeitheroftwotypesofevents:massinputorheatinput.BothtypesresultinmorerapidpressurechangeswhentheRCSiswatersolid.However,atreactortemperaturesbelow350'F,theRCSmustbealignedtotheResidualHeatRemovalSystemtoremovecoredecayheatandthewaterreliefvalveswillbeavailabletomitigatepressuretransientswhichmightoccur.Also,therewillgenerallybeasteambubbleinthepressurizerwhenthereactorcoolanttemperatureisaboveabout150'Fduringplantcooldownsothatwatersolidconditionsarelimitedtorelativelylowtemperatureconditions.Therefore,thedescriptionsofthetwotypesoftransientsimplythattheRCSiswatersolidatalowtemperature.1.3.1SummarofMassInputTransients1.3.1.1InadvertentSafetyInjectionInadvertentactuationofsafetyinjectioneventsincludefullsystem(bothtrains),singletrain,orsinglecomponentwithinatrainactuation.Eachofthethreetypesofeventsarediscussedseparately.FullsystemactuationwouldincludetheopeningoftheisolationvalvesonallSIaccumulators,startupofalllow-headandhigh-headsafetyinjectionpumpsandisolationof,=;thenormal.letdownpath<to-,theChemical.and.Volume.ControlSystem.SuchaneventwouldresultinunacceptablylargevolumesofwaterbeingforcedintotheRCS.Therefore,sucheventsmustbepreventedbystrictadministrativecontrols.ThesecontrolsrequiretheblockingoftheautomaticSIactuationcircuits,immobilizingtheSIaccumulatormotoroperatedisolationvalves,andlockingoutpowertothehigh-headSIpumps.Inatypicalwestinghousedesign,thelow-headsafetyinjectionpumps(theRHRpumps)arenormallyinoperationandalignedtotaketheirsuctionfromtheRCS,andnotfromtherefuelingwaterstoragetank,duringlowpressureandlowtemperatureplantoperations.Therefore,evenifaspuriousstartsignalwasreceived,thelow-headSIpumpswouldnotfunctiontoinjectRMSTwaterintotheRCS.8921e:1d/0602891.5 P,.jg~>Ãi~'II1l':p44Dg+Af-A1IItMC Theprobabilityofa'singletrainactuationisaboutthesameasafullsystemactuation,sincethesignalswhichcallforsafetyinjection,bothmanualand'automatic,arenormallyprocessedthroughtheengineeredsafetyfeaturelogiccircuitssuchthatasignal,whetherspuriousornot,willimpactbothtrains.Therefore,sincetheSIsystemisessentiallyimmobilizedatlowtemperature,singletraininadvertentactuationisconsiderednomorelikelythanfullsystemactuation.ForthoseplantdesignsinwhichthechargingpumpsservethesecondfunctionofhighheadSIpumps,e.g.theCookunits,aspurioussafetyinjectionsignalwouldrealignthevalvingtotransferfromthechargingfunctiontothesafetyinjectionfunction.Therefore,theoneoperatingchargingpumpwhichisnotlockedoutwilldeliverthroughthesafetyinjectionflowpathtotheRCS.~'-'~~'~',~-However,,'~the~RHRS.would.remainopentotheRCSatthistimeandtheRHRSreliefvalveswouldmitigatetheresultingRCSpressuretransient.Inadvertentactuationofasinglecomponentwouldrequirethatanoperatorselectivelyunlocktheelectricalpowertothecomponentandthencausethecomponenttobeenergized.Themostprobablewayforthiseventtooccurwouldbeduringperiodicsurveillancetestingrequiredbythetechnicalspecificationsorduringpostmaintenancecheckoutofthecomponent.Deliberateopening'fanSIaccumulator,isolationvalvewhile..theaccumulator..ispressurizedwithgasisnotconsideredprobablebecausethereisnotechnical-'specification",to.testtheisolationvalvesatshutdown,andprudentmaintenanceproceduresforthevalveswouldlikelyrequirethatthecompressedgasintheaccumulatorberemoved.Postmaintenancecheckoutorperiodicsurveillancetestshowever,mightbeattemptedonahighheadsafetyinjectionpumpprovidinganopportunityforoperatorerrortocauseaninadvertentsinglepumpinjectionevent.Therefore,asinglesafetyinjectionpumpstartupeventduringsurveillancetestingorfollowingmaintenance.isconsideredapotentialmassinputtransient.SincetheRHRSwouldbeopentotheRCSatthistime,theRHRSreliefvalveswouldmitigatetheresultingRCSpressuretransient.89216:1d/0602891.6

%or0eAl~vltosM~g~II%~Sm~inr&4ighltl>4ICa1.~Sled~pi~'*rw'swww<<j"iieeaI~

1.3.1.2Charging/LetdownFlowMismatch~~~Charging/letdownflowmismatcheventscanbepostulatedtooccurinanumberofways.Onewaywouldinvolvethecompleteterminationofletdown:closureoftheletdowncontrolvalve,isolationoftheRHRS/CVCScrossoverpath,orclosureoftheRHRSinletisolationvalvescausedbymalfunctionsofthecontrolsystems.Asecondwaywouldinvolveanincreaseinthechargingflowbyeitheroperatororinstrumenterrorsuchthatthechargingflowexceedstheprevailingletdownflow.Themostseveremassinputtransientwouldoccuriftheletdownflowcontrolsfailedtothe.zeroflowconditionwhilethechargingflowcontrolsfailedtothefullflowcondition.Thisfailuremodewouldresultinthemaximum'-"-"'""'"'"charging/'letdown'flow-mismatcheventbutdoesnotresultintheisolationoftheRHRSreliefvalvesfromtheRCS.Therefore,theRHRSreliefvalveswouldmitigatethistransientandpreventanoverpressureconditionineithertheRHRSortheRCS.Themostlikelywaythatacharging/letdownflowmismatchwouldoccurisfortheRHRS(andreliefvalves)tobeinadvertentlyisolatedfromtheRCSbyspuriousclosureoftheRHRSinletisolationvalves.Suchaspuriousclosureiscredible'due'othe-presence'f.::theautomaticclosing..signal,requiredbytheNRC.SincethisspuriousvalveclosureeventcausesanRCSpressuretransient-by"stopping"the'letdownflowandconcurrentlyisolatestheRHRSreliefvalvesfromtheRCS,itisconsidereda"designbasis"transientfortheLTOPS.1.3.2SummarofHeatInputTransients1.3.2.1ActuationofPressurizerHeatersTheinadvertentactuationofthepressurizerheaterswhenthepressurizerisfilledsolidwillcauseaslowriseinthewatertemperaturewithaconsequentincreaseinpressureoftheconstantvolumeRCS,iftheinstalledautomatic'ressurecontrolequipmentisnotinservice.Sincethepressuretransientis8921e:1d/0602691.7

~CP<<Son'lllilffv~aabC veryslow,.theoperatorshouldrecognizeandterminatethetransientbeforeanunacceptablepressureisreached.TheRHRSisopentotheRCSwheneverthepressurizerisfilledsolid,inaccordancewithadministrativecontrols.Therefore,theRHRSreliefvalveswillbeactuated,negatingtheneedfortheLTOPS,iftheoperatordoesnotinterveneandstopthetransient.Thiscase'snotconsideredsignificanttothedesignofeithertheRHRSreliefvalvesortheLTOPS.1.3.2.2LossofRHRCoolingAlossofresidualheatremovalcoolingwhilethepressurizerisfilledsolidcouldbecausedbyalossofflowmalfunctioninthecomponentcoolingwaterortheservicewatersystems,ortheclosureoftheRHRSinletisolationvalves."..Thecontinual.releaseofcoreresidualheatintothereactorcoolant,withnoheatrejectionintotheenvirons,wouldcauseaslowriseinthecoolanttemperatureandpressure.Sincethetransientisslow;-theoperatorshouldrespondandeitherrestoretheRHRSvalvestotheiropenposition,restorecooling,orlimittheRCSpressurebyventingthepressurizer.Thistransient.isnotconsideredsignificanttothedesign.of,theLTOPSsinceitisarelativelyslowtransientcomparedtotheheatinputtransientresultingfromthestartupofareactorcoolantpumpincombinationwithanRCS/SGtemperatureasymmetry...'.,1.3.2.3.,RCP.StartupNi.thTemperatureAsymmetryDuringplantheatupandcooldownoperations,typicaladministrativecontrolsrequirethatatleastonereactorcoolantpumpbemaintainedinoperationwheneverthereactorcoolanttemperatureisgreaterthan160'F.Therefore,the'argevolumetric=flowthroughouttheRCSwillmaintainanisothermalconditionintheRCS.Thesteamgeneratorsecondarysidewaterimmediatelysurroundingthetubeswillalsoremainatatemperaturenearthatofthecirculatingreactorcoolantontheprimaryside.Duringnormalcooldownoperations,whenthereactorcoolanttemperaturehasdecreased'below160'Fthereactorcoolantpumpsmaybestopped.Subsequently,8921e:1d/0607891.8 EvP+,vat,@~~+A~!I~,~~Ali~~~<k14eylip"~Itgl4QQi~WF~1 isothermal-conditionsmaynolongerexist.The-.reactor,coolanttemperaturewillbedecreasedbelow160'FbyheatrejectionthroughtheRHRS.Thesteam'generatorcontainedwater(bothprimaryandsecondary)mayremainatarelativelyconstanttemperature,greaterthantheRCStemperature,duetolittleornocirculationthroughihetubes.Therefore,asignificanttemperatureasymmetrymaydevelopbetweentheprimarywatercontainedinthesteamgeneratorandtherestoftheRCS.Ifareactorcoolantpumpweretobestarted,thesuddenheatinputintothereactorcoolantfromthesteamgeneratorwouldcausearapidincreaseinreactorcoolanttemperature.Iftheeventweretooccurwhilethepressurizerisfilledsolid,arapidlyincreasingpressuretransientwouldoccur.Inaccordancewithtypicaladministrativecontrols,theplantwillbeunder-water..sol;id;.conditions..onlywhiletheRHRS.is.inservice,andatleastonereactorcoolantpumpwillbeinoperationatreactorcoolanttemperaturesabove160'F.Therefore,thistypeofheatinputtransientwillbelimitedtoinitialcoolanttemperaturesbelow160'F.Sinceitisnotpracticaltodeterminearepresentativetemperatureforthelargestagnantvolumeof-<<secondarywaterinthesteamgenerator,,the,operatorwillnotbeawarethatthetemperaturemaybesubstantiallydifferentfromtheremainderofthereactorcoolant.Fromtheinitialisothermaltemperatureof160'FwhentheRCPis.stopped,thebulk.reactor.coolanttemperatureisunlikelytodecrease...below110'Fwithoutsomeextraordinarycoolingmeans,whilethesteam:-<'.=-':"~..gene'rator..water,.may,remain,near..160'F,...Therefore,.thedifferentialtemperatureisnotexpectedtobegreaterthan50'Fforthistypeofheatinputtransient.IftheRHRSisinadvertentlyisolatedfromtheRCSbyclosureoftheisolationvalves,asbjaspuriousoperationoftherequiredauto/close"interlock,while,theplantiswatersolidandinmode4,typicaltechnicalspecificationsrequirethatareactorcoolantpumpberestartedwithinonehourifanRHRloopcannotbereturnedtoservice.Duringthepotentialonehourdelayperiod,atemperatureasymmetryinthereactorcoolantlo'ops,duetothecontinuedinputofcoldsealinjectionwater,coulddevelopandnotbeS921e:1d/0602BB1.9 lP'g~,(sr~yp"y~wp~yIg"~44.=~>aI%~~.$4.~'CCQ+~su.,~mgyjpykpr,,aA

~4apparenttotheoperator.-Thenwhenthereactorcoolant,pump,is.restarted,an~~increasingpressuretransientwilloccurasthecoldwatercontainedinthe'steamgenerator/RCPcross"overpipemixeswiththerestoftheRCSwater..1.3.2.4RelativeSeverityoftheHeatInputTransientsFigure1.2comparestherelativeseverityofthepressuretransientsresultingfromtheheatinputcasesdiscussedinthepreviousparagraphs,asanalyzedfortheHestinghouseOwnersGroup.Fromaninspectionofthefigure,itisevidentthattheheatinputcasesfrompressurizerheatersanddecayheatarelesssignificantthanthoseforthecaseswithaloopasymmetry.Therefore,theselesssignificantcasesarenotconsideredforthesetpointanalysis.Similarly,theloopseal(cross-overpipe)asymmetrycaseisseentoresultin<-a"relative'ly,.smallpressuretransient-:compared-,tothepotenti'alexcursionpossiblefromthesteamgenerator/RCStemperatureasymmetrycases.The"designbasis"caseforthesetpointdeterminationisthereforethetemperatureasymmetrybetweenthesteamgeneratorandtheRCS.1.4SUMMARYOFTRANSIENTEVALUATIONBasedonthepreviousdiscussion,mostoftheidentifiedmassinputandheatinputtransientswhichmightoccurwhiletheplantiswatersolid,willbemitigatedbythewaterreliefvalvesintheRHRS.However,-forthoseremote"'-"'~"cases~which".occuror~.are.causedby..the.RHRShaving,.become-isolatedfromtheRCS,theLTOPSmaybecalledupontomitigatecertainincreasingpressuretransients.Specifically,theLTOPSdesignbasistransientsare:1)themassinputtransientcausedbyanormalcharging/letdownflow'mismatchafterterminationofletdownflow,and2).theheatinputtransientcausedbytherestartofaRCPwhentheRHRSisnotopento.theRCS.8921e:1d/0602891.10

~144III'5JIL~"44twups~ig+4l~E44+><y*'1",S"r>t,:,IIpP0'CJ"4)lYpmksa.pLslq.1

~~SingleRCPatartvitaenoreliefvalveacutataenRCS/S6~l56F/2BBF--4'tA+"fIg'%'whyLoopSaai/RCS'86F/218FCor~OacayHthddttionhtti"Figure1.2RelativeRCSPressureTransientsResultingFromHeatInputEvents$82le:1d/060289 AlIJ,'7 2.0DESCRIPTIONOFTHELTOPSSETPOINTALGORITHM4ThedeterminationofthelowtemperatureoverpressureprotectionsetpointisbasedonalocalversionoftheLOFTRANcode.TheLOFTRANcodepredictsplanttransientthermal/hydraulicbehaviorbymodelingthereactorcoolantsystem,includingthesteamgenerators,pressurizer(includingPORVs),andreactorcoolantpumps,aswellasthecontrolandprotectionsystems,selectedvalving,and,somebalanceofplantsystems.TwoversionsoftheLOFTRANcodewereutilized:thefirstversion,usedforthemassinputcalculations,collapsestheseveralRCSloopsintoasingle'loopmodel;thesecondversion,usedfortheheatinputcalculation,modelseachloopexplicitly.TheselectionoftheproperLTOPSsetpointrequirestheconsiderationofanumberofsystemparameters.Amongthesearethefollowing:1:'Volume'ofthereactorcoolantinvolvedinthetransient.,2.,RCSpressuresignaltransmissiondelay.3.Volumetriccapacityofthereliefvalvesvs.openingposition.4.Stroketimeofthereliefvalves(openingandclosing).Ifthepressureundershootisimportant,theclosuretimeisrequired.15.'assinputrateintotheRCS.,G.Heattransfercharacteristicsofthesteamgenerators.7.Initialtemperature.asymmetrybetweentheRCSand,steamgeneratorsecondarywater.8.Massofsteamgeneratorsecondary.water.9.RCPstartupdynamics.10,RCPNo.1sealdeltaPrequirements.ImportantifalowersetpointlimitistobespecifiedforRCPsealprotection.11.AppendixGpressure/temperaturelimitsforthereactorvessel.2-1PRESSURELIMITSSELECTIONThefunctionoftheLTOPSistopreventtheRCSpressurefromincreasingabovethePORVpiping/structuralanalysislimitsandthelimitsprescribedbytheallowablepressure-temperaturecharacteristicsforthespecificreactorvessel8921e:1d/0720892.1 C"Qt~1f2%;<<1CFEt~CEa materialin.,accordance,withthe;rules.givenin.Appendix..G.to10CFR50.FortheCookunits,aconstantpressuresetpoint,independentoftemperature,isemployedsothatthelimitsconsideredfor,thisparticularcasemustmeetthemostrestrictivesegmentoftheAppendixGcurveswhencomparedtotheoverpressuretransientasafunctionmf~perature.ThePORVpiping/structurallimitsarewellabovethosepressuresofconcernhere,'ndaregenerallyconsiderationsonlyforthoseplantswhosePORYsetpointsarefunctionsoftemperature.<Qlg~1j'tjlAcharacteristicpressure-temperaturerelationshipisshowninFigure2.1,illustratingtheallowablesystempressureincreaseswithincreasingtemperature.Thistypeofcurvesetsthenominalupperlimitonthepressure,whichshouldnotbeexceededduringJKSincreasingpressuretransients.When-.a'relief.val.ve".is..actuatedtomitigateanmcreasingpressuretransient,thereleaseofavolumeofcoolantthroughthevalvewillcausethepressureincreasetobeslowedandreversedasillustratedbyFigure2.2.Thesystempressurethendecreases,asthereliefvalvedischargescoolant,untilaresetpressureisreachedwherethevalveissignalledtoclose.Notethatthepressurecontinuestodecreasebelowtheresetassureasthevalvecloses.Thenominallowerlimitonthepressureduringthetransientisselectedbased.onarequirementofthereactorcoolantpumpNo.1.sealtomaintainanominal200psidifferentialpressureacrossthesealfaces.'<<'..~.:~>>..Thenominal;;upper.."l.hami,t,(basedonthe..minimum.of.AppendixG,requirementsorthePORVpipinglimitations)andthenominalRCPNo.1seallowerlimitIpressurevaluescreateanacceptablepressurerangeintowhichthePORVsetpointsmustbefit.Anillustrationoftheselimitsal'ongwiththesetpointselectionrangeisshowninFigures2.3and2.4.IntheeventthatthesetpointselectionrangeisinsufficienttoaccommodateboththeAppendixG,.andtheRCPNo.1seallimit,theAppendixGlimitwilltakeprecedence.S921e:1d/0607892.2

)II"*its.gy'I~r'id4,~'ii'Itkt4>$48'iick'laity%i"icar~*~iPi<

vg2SOO22502000...1'750,150012SO1000linecceptableOperationIIIIIIIIIIAcceptableOperationIIIIIII~IIIIIIIIIIIsIIIIII~II~IIIIIIIIIIIIIIiIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIII~IIIIIIIIIIIIIIIIIIIIIIIIIIIIIO4lII~,750500250'-C001dONlRates'F/Hr02040601001IIIIIIIII'II~IIII~IIIII~I050100'502002503001NOlCATEDTEK1'ERATURE(DEC.F')FigUre2.1TypicalAppendixGP/TCharacteristics8921e:1d/0602892.3

  • J!v'4rlr<<~it~.-'PA POver'SETPO?RT-RESETUnderTlNEFigure2.2"TypicalPressureTransientd921e:1d/0602692.4 APPENDIXGMAXIMUMliMITLYilNPMINLRCPflSEALMINIYlNLIYilTSETP">"RAN"~PDRVSKTPOlNT,l'SIGFigure2.3SetpointDetermination(MassInput)d$21e:1d/0602892.5 044/WPM-1lfD<<y AppendixGMaximumLimit'NAXBINRCPflSealNinimvmLimitSETPOINTRANGEPSPORVSETPOINT~PSIGFigure2.4SetpointDetermination(HeatInput)SS21e:1d/0602S92.6 0l1t,'"+'A'b"1tf1~IttQJ"Pl-VIAI'I<~g t2.2MASSINPUTCONSIDERATIONSForaparticularmassinputtransienttotheRCS,thereliefvalvewillbesignalledtoopenataspecificpressuresetpoint.However,asshowninFigure2.2,therewillbeapressureovershootduringthedelaytimebeforethevalvestartstomoveandduringthetimethevalveismovingtothefullopenposition.Thisovershootisdependentonthedynamicsofthesystemandtheinputparameters(e.g.massflowrate),andresultsina.maximumsystem.pressuresomewhathigherthanthesetpressure.Similarly,therewillbeanundershootwhilethevalveisrelieving,bothduetotheresetpressurebeingcbelowthesetpointandtothedelayinstrokingthevalveclosed.ThemaximumandminimumpressuresreachedinthetransientareafunctionoftheselectedsetpointandmustfallwithintheacceptablepressurerangeshowninFigure2.3.AnumberofmassinputcasesarerunatvariousLTOPSsetpoints"'"--."'"<'.*'-"se'1'ecte'd'Co"bound*theexpected'setpointrangeandoverarangeofmassinjection'ates.Fromtheseruns,alocusofthemaximumandminimumpressurevaluesisgeneratedovertheexpectedsetpointrange,asshowninFigure2.3.~~~Theshadedarearepresentstheacceptablerangefromwhichtoselectthesetpoint.Themassinjectioncasesareconservativelyanalyzedatlowtemperaturewhere"thebulkmodulusofthefluidisgreatest.Theresultingovershootisthereforworstcase,andevaluatingmassinjectioneventsathigher,temperaturesisnotrequired.2.3HEATINPUT.CONSIDERATIONSTheheatinputcaseisanalyzedinaboutthesamewayasthemassinputcaseexceptthatthelocusoftransientpressurevaluesvs.selectedsetpointsaredeterminedforseveralvaluesoftheinitialRCStemperature.Thisheatinput,evaluationprovidesarangeofacceptablesetpointsdependentonthereactorcoolanttemperature,whereasthemassinputcaseislimitedtothemostrestrictivelowtemperatureconditionsonly.TheshadedareaonFigure2.4highlightstheacceptablerangeforaheatinputtransientforaparticularinitialreactorcoolanttemperature.8921I:1d/0608892.7 tJI

',2.4FINALSETPOINTSELECTION.By,.superimposingtheresultsoftheseveralmassinputandheatinputcasesevaluated(fromaseriesoffiguressuchas2.3and:2.4),therangeofallow-ablesetpointscanbedeterminedthatwillsatisfybothmassinputandheat'nputconsiderations.Aspreviouslystated,theseTectionofthepressuresetpointsforthePORVsisbasedontheuseofnominalupperandlowerlimits.Useofnominalvaluesisjustifiedbasedontherecognizedhighdegreeofconservatisminherentinthe10CFR50AppendixGpressure-temperaturelimits.8921I:1d/0607892.8

'0p~,I'.~II't>~Pic.ALCE,\I/

4'3.0LTOPSSETPOINTANALYSESFORTHED.C.COOKUNIT2The.LTOPSsetpointanalysispresentedinthissectionwasdevelopedfortheAmericanElectricPowerCorporationsD.C.Cookunit2usingthealgorithmdescribedintheprevioussection.TheanalysisevaluatestheimpactofPORVopeningtimesupto10secondsonthevalueofthesetpointrequiredtomaintainoverpressuresbelowthe10CFR50AppendixGlimits.PORVclosuretimeisassumedtobe4secondsforallcases.Inaddition,thissectiondocumentsthedevelopmentofthecorrelationbenchmarkingtheresultsoftheanalysistothatofthealgorithmdescribedinMestinghouseOwnersGroup(MOG)reportreferencedintheintroductiontothisreport.TheLTOPScurrentlyinstalledatCookUnit2featuresaconstantvaluesetpointprogram(i.e.aprogramindependentoftemperature)withan"-.'"enabl'e/disabl'e"reactorcoolantsystemtemperatureof152'F.-Theprogramsetpointiscurrently435psigforbothPORV's.Theanalysisassumesoverpressurizationtransientsresultingfromeithermassinjectionorheatinputeventsunder4-loopwatersolidconditions,withthereplacementmodel51Fsteamgenerators.'hemassinjectiontransientoccursasaresultoftheoperationofasinglecentrifugalchargingpumpconcurrentwithaspuriouslossofletdown.Theallowablechargingconfigurationislimitedbytechnicalspecificationtoasingle.centrifugal=chargingpumpin',operationalmodes5and6,;Thesafetyinjectionpumpsarerequiredtoberackedoutinthese'odes.'heheatinjectioneventresultsfromthestartofareactorcoolantpumpassumingthattheprimarywaterinthesteamgeneratoris50'Fwarmerthanthewatercontainedintherestofthereactorcoolantsystem.ThetemperatureasymmetrycandevelopduringacooldownmaneuverfollowingtheshutoffofthereactorcoolantpumpsandcontinuedcoolingwiththeRHRsystem.NocreditistakenfortheRHRreliefvalvesforeitheroftheoverpressurescenarios.Thesetpointselectionisbasedonthemostrestrictiveofeitherthemassinjectionorheatinputcases.Aconstraint'requiredfortheanalysisistheassumptionofthefailureofoneofthePORV's.Thedesignbasisfortheoverpressureeventsrequirethat8921e:1d/06028S3;1 0iz~'fI

eitherPORVprovideadequaterelievingcapabilityintheeventofasingle~~~~valvefailure.CodeDescriptionTheevaluationofthecoldoverpressuremitigationsystemsetpointsisbasedon'ocalversionsoftheLOFTRANcode.Twoversionsofthecodewereutilized:LOFT12,usedforthemassinputcalculations,collapsestheseveralRCSloopsintoasingleloopmodel;and'OFT4,whichmodelseachloopexplicitly,fortheheatinputcalculation.3.1OPERATIONALLIMITSThepressure-temperaturelimitcurves(AppendixGcurves)basedonrevision2'""of'USNRCRegulatory*Guide1.99,havebeengeneratedforreactorvesselexposuresof12and32effectivefullpoweryears(EFPY).Thesetpointevaluationisbasedonthesteady-statecooldownlimit,wherethetechnical~~differencebetweenasteady-statecooldownlimitandasteady"stateheatuplimitistheassumedlocationoftheflaw;i.e.,insideoroutsideofthevessel.Thesteady-statelimitprovidesthegreatestoperationalmarginandhasbeenacceptedbytheNRCwiththejustificationthat"mostpressuretransientshaveoccurredduringisothermalmetalconditions."Thesteady-statelimits"areshowninFigure3.1and,indigitizedform,inTable.3.1.Thecurvesalsoincludethe800psigPORVpipingloadlimit.Theanalysisassumesnominalvalues;i.e.,pressureinstrumentationuncertaintiesarenotincluded.ThisisconsistentwithstandardRestinghousepractice.UseoftheAppendixGlimitswithoutinstrumentationuncertaintyisjustifiedonthebasisofthelargeamountofconservatism(recognizedbytheNRC)inherentinthedevelopmentofthelimits.The800psigpipipglimitresultsfromananalysisofwaterhammereffectsonreliefvalvepipihgforcertainclassesofrapidlyopeningreliefvalves(e.g.,Garrettvalves)underwatersolidconditions.TheGarrettvalvesaresolenoidoperatedwitharapidlyincreasingflowvs.stempositioncurve.8921e:1d/0607893.2 ltl'Io,tlr.~~'~1f'pl$$P P0RVP12EFPYIIIIIIIFigure3.1ReactorCoolantSystemSteady-StatePressureTemperatureLimitsat12and32EFPY89216:1d/0720893.3 0'W'tÃ1tC&a'we'44iIl~Wj4;,I1tgkb1tIl>%l<<~yp<<t TABLE3.1STEADY"STATECOOLDOMNPRESSURE/TEMPERATURELIMITSRCSPressuresiRCSPressuresiRCSTemdeFRCS12EFPY32EFPYTemdeF12EFPY32EFPY85.090.095.0100.0105.0110;0115.0120.0125.0130,0'35.0140.0145.0150.0155.0160.0509.2513.3517.8522.6527.7'"5332'39.0545.4552.2559.6567.5576.1585.1594'.9605.5616.9490.2492.8495.7498.8502.1505.7509.6513.8518.2523.0528.2533.8539.6546.0552.9560:4165.0170.0175.0180.0185.0190.0a95.o200.0205.0210.0215.0220.0225.0230.0235.0240.0245.0629.1642.1656.3671.5687.7705.2724.1744.2766.0789.3814.5841.4870.49oa.6935.0970.81009.6568.3576.9586.1595.9606.6618.0630.4643.5657.7673.0689.3707..0726.0.746.3768.2791.7817.08921e:1d/0607883.4 I0,",L)~IL'LIIe1WIlJ'i'W+(Lf.I'IVL Whenacharacteristiccurveofthistypeiscombined,withGarrett'stypicallyrapidstroke'time(lessthantwoseconds).the"valvebecomeseffectivelyfullopenorfullclosedwithinafewtenthsofasecond,thussettingtheconditionsforawaterhammer.'heflowthroughanairoperatedreliefvalve,whencomparedtosolenoidvalves,ismuchlesssensitivetostemposition.Hhencombinedwiththerelativelyslowopeningandclosingtimescharacteristicofthesetypesof'valves,thewaterhammereffectswillbemuchreduced,ifnoteffectivelyeliminated.EvaluationofwaterhammerforcesonthepipingofairoperatedvalveshasnotbeenperformedbyHestinghouse.Thepracticehasbeentoassumetheconservativepositionoftakingtheworstcaseresults(theGarrettanalysis)andapplyingthemtoallLTOPSsetpointevaluations,regardlessofthereliefvalveemployed.3.2-PORVSTROKETIMEAsinstructedbyAmericanElectricPowerCorporation,theLTOPSanalysisalso~~includesaparameterstudyonvalveopeningtimeinordertodeterminetherelationshipbetweenopeningtime.andLTOPSsetpoint.Theparametricstudywasperformedassuming6differentvalveopeningtimes,rangingfrom1.0sec.to10seconds.Thiswasconsideredabroadenoughrangetocoverallreasonable"openingtimemeasurements;.""""'""'""'The"PORVclo'singtimeselectedfor-theanalysiswassetat4.0seconds,independentoftheopeningtime.Theclosuretimeselectionhasnoimpactontheoverpressuretransient(anAppendixGconsideration),butdoesimpacttheunderpressuretransient(importantfortheprotectionofthereactorcoolantpumpNo.1seal).Thecharacteristicofthetransientissuchthatasclosuretimeincreases,theunderpressurebecomesmoresevere.3.3PORVOPERATIONThedesignbasisforanoverpressureeventrequiresthateitherPORVprovideadequaterelievingcapabilityintheeventofasinglevalvefailure.8921e:1d/0601893.5 L'I'~o4~ro

'herefore,thesetpointanalysisisbasedontheassumptionofsinglevalveoperation.Thesetpointanalysisincludestheeffectoftimedelaysassociatedwiththetransmissionofthewideranger'eactorcoolantsystempressuresignal.Aconservativevalueof0.95secondswasutilizedfortheanalysis.Thebreakdownofthetimedelayisasfollows:Pressuresensinglinetransportdelay.,PressuretransmitterdelayElectronicsdelay...........'SolenoidactuationdelayValveChamber(Pneumatic)',delay....Total....0.15sec...0.25sec...0.10sec..0.10sec.,0.35sec...0.95secWiththeexceptionofthepressuretransmitterdelay,thefactorscomprisingthetotal'elaytimeareWestinghousegenericestimatesforairoperatedpressurizerreliefvalves.'.C.CookUnit2featurestwoFoxborowiderange(ModelN-EIIGH"HIM2-A)pressuretransmitters.WestinghouseinstrumentationgrouphadnoinformationonFoxborotransmitters,sothedelaytimeof0.25secondswasobtaineddirectlyfromFoxboro'sProductInformationGroup.Theflowcharacteristics(valveC(v).'vs.valve'stroke)-forthe-pressurizer.poweroperatedreliefvalveswereobtainedfromAmericanElectricPowerCo.andareshownbyFigure3.2.ThepercentatfullflowC(v)correspondingtoselectedpercentofvalvestrokevaluesislistedinTable3.2.3.4MASSINPUTCONSIDERATIONSThemassinjectiontransientisassumedtooccurasaresultoftheoperationofasinglecentrifugalchargingpumpincombinationwithasuddenloss.ofletdown.Theallowablechargingconfigurationis-.limitedbytechnicalspecificationsinoperationalmodes5and6,wheretheCookunitsLTOPSisenabled.Thechargingflow,assumingthatthenormalandalternateflowpaths,.are.bothopen,,asafunctionofreactorcoolantsystempressureisshowninFigure3.3.ThedateintabularformisgiveninTable3.3.BSZle:1d/0607BS3.6

('4~i0Ipf~+'+1~4~'I4CtfWI'i',IIIl'lp~IIC~

IloaonoilonLinearhirOporotodReliefVolvoFullFlovC(v)~46Figure3.2PressurizerPORYFlo~CharacteristicCurve8921a:1d/0607893.7 t1~*

TABLE3.2PRESSURIZERPORVCHARACTERISTICS~PORVenin'AValve'AValve'AValve5Valve0.010.0.30.040.050.00.09.026.035.045.0'0.070.080.090.0100.056.066.077.088.0100.0PORVClosin5Valve'AValve%%dValve.5Valve0.010.020.030.040.0100.088.077.066.056.050.060.070.090.0100.045.035.026.09.0100.0"BasedonthecharacteristiccurveshowninFigure3.2.8921e:1d/0607893.8

,{;%VQ)'~W~r4'4fk'1P*t CharglnOuato~Slnl~CantrlfualggpChargingPunpFigure3.3MassInjectionFlowvs.RCSPressure892)a:1d/0607893.9 841~I%OeEII TABLE3.3MAXIMUMCHARGINGFLOWATRCSPRESSURE~RCSPress.+>~si~ChargingFlowmRCSPress.siChargingFlow'm400,500600700800900'='1000439.2429.7420.2410.7399.9386.2372.31100120013001400~1500160017001800358.3344.1329.8315.2300.5285.6268.9251.5"Basedonsinglecentrifugalchargingpumpoperation.SS21e:1d/0607&93.10

~tt~,>>I~t~>>~I~

b'hemassinputcasewasconservativelyanalyzedatlowtemperature,85'F,'herethe'pressuretransientresultingfromamassinputshowsthegreatestovershootsandundershoots.Noother,calculationswereperformedatdifferenttemperatures,forthereasonthatthisistheworstcasesituationandtheD.C.CookUnit2LTOPSsetpointisindependentoftemperature.Massinjectionratesrangingfrom100to600gpmwereselectedforthemassinputparameterstudyportionoftheanalysis.ThisrangewasbasedonanalysesperformedforunitssimilartoD.C.CookUnit2,andprovidesagooddefinitionoftherelationshipbetweenmassinputandtheresultingpressuretransient.TheresultsoftheLOFTRANrunsarepresentedinFigures3.4through3.9forthebothoverpressureandunderpressuretransients.3.5.HEATINPUT.CONSIDERATIONSTheheatinputmechanism,fromthediscussioninSection1.3.2,isbasedona.singlereactorcoolantpumpstartupwithatemperatureasymmetryexisting~~betweentheprimarywaterheldupinthesteamgeneratortubesandthewater,intheremainderofthereactorcoolantsystem.Themagnitudeofthe\asymmetrydependsonthepreviousplantoperationwhichallowedtheasymmetrytodevelop.Forthisstudy,itwasconsideredconservativetoassumeamaximumasymmetryof50'Fasthedesignbasis,sincemuchhigherdifferenceswouldbedifficulttodevelop.TheheatinputcaseswereanalyzedatRCStemperaturesof85'Fand150'F(steamgeneratortemperaturesof135'F'and200'Frespectively).ThecharacteristicbehavioroftheoverpressuretransientresultingfromaheatinputeventistobecomemoreseverewithincreasingRCStemperature.Thisrequired.theadditionalLOFTRANrunsatincreasedtemperaturesinordertoprovideassurancethattheincr'easeinpressureovershootwithtemperaturedidnotexceedthatoftheAppendixGlimit,astheheatinjectioneventsbegantodominate.;.S921e:1d/0607893.11 91Wq<c,pe~ver,~0>>4-v~*>C41p'ItV+I\e\fflhIIgl++lg)Ek4~we4J)

OIIODOAlo;CINeF00'0aI/1Figure3.4OvershootandUndershootdPVa1uesvs.MassInjectionRateatPORVStrokeopentimeof1.0Sec.$921e:1d/OSOSS93.12 0tgVs~I'4"4')tAIll'I1g%*g)~IIIIfpI1gC O0Olostiel.~eec.0ailla'igure3.5OvershootandUndershoothPYaluesvs.MassInject)onRateatPORYStrokeOpenTimeof2.0Sec.8921e:1d/0808893.13 0~~':mViw'4'-eCea~-i,4~'gjjellSP4g,iIQC'P

!)IsPOaiuro~CQaC0aIllUIUIIWIIUUIWWllilWUlilllUIIIIIUllllFigure3.6Overshoot.andUndershoothPValuesvs.MassInjectionRateatPORVStrokeOpenTimeof4.0Sec.8921e:Ud/OBOS893.14 III7Pt@p4XItIl1'"9'hkCQ<t.sp,e,w."~'4+'f'!iI4'1 loC~0D0aQelraC0aFigure3.7OvershootandUndershoothPValuesvs.HassInjectionRateatPORYStrokeOpenTimeof6.0Sec.8921e:id/080889 e4iIh1 I.VC10ORSULhO~0~C~t0aaDC0a.Figure3.8OvershootandUndershoothP,Valuesvs.MassInjectionRateatPORVStrokeOpenTimeof8.0Sec.8921e:1d/0608893.16 Cjl1ihw)s~c1Ipl/b\wt$-~"p'rI D0-POIIre~ee.C0aOeDC0aFigure3.9OvershootandUndershoothPYaluesvs.HassInjectionRateatPORVStrokeOpenTime'of10.0Sec.892)e:1d/0808893.11 plysI0P~~l%4'ligtl.+rpsinc'v"H

'.6SPECIFICATIONFORMASSINPUTTRANSIENTS~YtReactorCoolantSystemtemperature=85'FReactorCoolantSstemVolume:RCSvolume=12,509cu.ft.fortheD.C.CookunitsInitialReactorCoolantSstemPressure:TheinitialRCSpressurewasassumedtobe2QDpsilessthanthesetpoint....pressure.ThisisconservativeandassuresthatthetransientiswelldefinedbythetimethePORVsetpointisreached.Giventhisdefinition,theoverpressureisessentiallyinsensitivetotheinitialpressureselection.ReactorCoolantReliefCaacit:ThetransientisanalyzedassumingthefailureofonePORV.PORYCharacteristics:LTOPSSetpoints=400,500,600,700psigPORVflowcharacteristicsareshowninFigure3.2OpeningTime=1.0,2.0,4.0,6.0,8.0,10.0sec.Closingtime=4.0sec.C(v)=468921e:1d/060789 lflV~lU3(

4I,IMassIn'ectionFlowCaabilit:MassinputRates=100,250,400,600gpmMaximumchargingflowcorrespondingtoLTOPSsetpointpressure.(Ref.Table3.3)PressureSinalTransmissionCharacteristics:TimedelaytoPORYstemmotion=0.95sec.3.7SPECIFICATIONFORHEATINPUTTRANSIENTS-"'"','".;,lSstem;Temeratures:SG/RCStemperaturedifference=50'FSteamgenerator(heatsource)temperature=135'F,200'FReactorCoolantSstemVolume:Thesameasthatusedforthemassinjectioncases.'"Initial.Reactor'CoolantSstemPressure:TheinitialRCSpressure=200psilessthantheLTOPSsetpointpressureReactor-CoolantSstemReliefCapabilit:ThesameasthatforthemassinjectioncasesPORVReliefCharacteristics:Thesameas"thatforthemassinjectioncases.89210:1d/0607893.19 "1~AtkC 1SteamGeneratorDesinCharacteristics:"S/Gtubeheattransfer.surface.area=,54,500sq.ft.S/Gtype=51F(D.C.CookUnit2replacementsteamgenerator)ReactorCoolantPuRCPtype=93APressureSinalTransmissionCharacteristics:Thesameasthatforthemassinjectioncases.3.8SETPOINTEVALUATIONTheimpactofvariablemassinjectionratesandPORVopeningtimesonovershootandundershootvaluesisprovidedbyTable3.4,Theinformationprovidedbythistableisbasedonthepressureovershoot/undershootrelationshipwithmassinjectionrateshownincompositeFigures3.4through3.9.Theovershootvaluewasdeterminedfromtheovershootcurvespresentedinthesefiguresforthemaximummassinjectionratesconsistentwithsinglecentrifugalchargingpumpinjectionflow(fromFigure3.3,or.thetabulation<<>>"".~shownin,Table3.3),for.the.indicated.,PORV.setpoint.,TheundershootvaluewasconservativelyestimatedfromthemaximumdeltaPvaluebelowtheindicatedsetpoint.Insomecases(PORVopeningtimesof1sec.and2sec.)theminimumdeltaPwasestimatedbyextrapolatingtheundershoottozeromassinjection.Thisprovidesabasisforestimatingthesetpointrequirementforreactorcoolantnumber1sealprotection.Theovershootandundershootpressuresresultingfromtheheatinjectioneventsare.showninTables3.5and3.6asfunctionsofsetpointpressureandPORVopeningtime.ThetablessummarizetheresultsofLOFTRANrunsperformedforRCStemper'aturesof85and150'F,respectively.Theevaluationwas8921e:1d/0607893.20 0C, DDt4aass'4DtOTABLE3.4,IMaxOvershoot/UndershootDeltaPValuesvsPORV0eninTimeDuetoMassInectionD.C.CookUnit2Overshoot/UndershootValuessivs.PORVeninTimesec*SetptPress.MassInj.st~Ratee400.0439.21.02.04.0Over.UnderOverUnderOver439.0231.0446.0232.0460.0Under282.06.08.010.0OverUnderOverUnderOverUnder473.0296.0487.0310.0499.0318.0500.0600.0700.0429.7420.2410.7538.0302.0542.0303.0555.0372.0566.0388,0578.0403.0588.0407.0635.0381.0641.0.388.0651.0460.0560.0482.0671.0492.0681.0493.0733.0465.0738.0479.0747.0547.0755.0576.0765.0577.0773.0591.0*PORVClosuretime=4.0sec.Notes:1.HassinjectionrateobtainedfromRCSpressurevs.singlepumpchargingflow.2.OvershootobtainedfrommaxdeltaPovershootVs.massinjectionflowcurve,3.UndershootobtainedfrommaxdeltaPundershootvs.massinjectionflow, 10'l~~IVE',"~u~J'I,~r>

tOaTABLE,3.5IMaxOvershoot/UndershootDeltaPValuesvsPORVeninTimeDdetoHeatInectionD.C,CookUnit2Overshoot/UndershootValuessivs..PORVeninTimesec*Setpoint1.02.04.0.6;08.010.0vOver,UnderUnder330.0UnderOver323.0435.0OverUnderOverUnderOverOver,Under431.0427.0311.0525.0400.0420.0266.0423.0289.0dIOO.O418.0265.0528.0416.0531.0426.0500.0517.0329.0518.0335.0521.0372.0600.0617.0422.0.618.0426.0621.0463.0624.0496.0627.0513.0629.0523.0700,0716.0502.0717.0510.0720.0557.0723.0593.0726.0610.0728.0621.0"-ValveClosuretime=4.0sec.RCStemperature=85.0'FS/Gtemperature=135.0'F 40IIA/4j' TABLE3.6MaxOvershoot/UndershootOeltaPValuesvsPORVeninTimeDueto}featInection0.C.CookUnit2Overshoot/UndershootValuessivs.PORVenintimesec*Setpoint1,02.04.06.08.010.0OverUnder400.0438.0278.0OverUnder445.0279.0OverUnderOver459.0295.0475.0Under311..0OverUnderOverUnder490.0311,05D5.0324.050D.O540.0338.0547.0342.0562.0381.0576.0395.0589.0407.0602.0408.0600.0-639,0419,0646.0433.0659.0468.0672.0487.0685.0493,0697.0495.0700;0"'739.'0512.0745.0516.0757.0563.0770.0579.0782.0581.0793.0585.0*ValveClosuretime~4.0sec.RCStemperature=150.0'FS/Gtemperature200,0'F lcnf'Iawn~>>i'd%Shg)*'eVq~-tR~i;e

~~~~~~~~~~~stoppedatthispoint(150'F)sinceitbecameevidentthatthemostlimitingtemperaturewithrespecttoAppendixGcriteriawasat85'Fandthatthemass...,..injectioneventsaredominant.Thedatafromthesetables(bothmassinjectionandheatinjection)ispre-sentedinFigures3.10through:3.21,showingthemaximaandminimasystempressureasafunctionofsetpointpressureforthevalveopeningtimesselectedforthestudy.Figures3.10through3.15showthepressureextremaatanRCStemperatureof85'F,andFigures3.16through3'.21showtheextrema-at150'F.Also.includedonthesefiguresistheAppendixGpressure.limitat12and32EFPY,andtheminimumsystempressureforanRCPstart.TheAppendixGlimits(ref.Figure3.1)correspondtotheRCStemperatureassumedforthecalculationshownbythefigure.TheRCPseallimitwasselectedto'"...correspond,.to.theminimum.systempressurespecifiedbyD.C.Cookforreactorcoolantfillandventoperations(325psig).Atothertimes,"thesystempressureisgovernedbytherequirementtomaintain200psidacrossthenumber,1RCPseal."Itisassumedthatthedifferentialpressurerequirementacrossthesealplusthevolumecontroltankpressureandthestaticheadinthenumber1sealleakofflinewillbenomorethan325psig;thevaluespecifiedforfillandvent.Theintersectionoftheselimits(AppendixGandRCPseal)withthemostlimitingofthemaximaandminima-curvesduetomassand/orheatinjection,~=;.,'..eventsformsthe..basis.fortheconstructionof..Figures3.22and3.23.Thesefiguresshow,respectively,thedependencyofthemaximumallowedLTOPSsetpointonvalveopeningtimeforreactorvesselexposuresof12EFPYand32EFPY.Theminimumsetpointlimit(RCPprotection)hasnotbeenshown,sincethe"whitespace"isgenerallynotlargeenoughtomeetboththeRCPnumber1,seal'equirementsand.theAppendixGlimit.8921e:1d/0606693.24

."ling"C~'V5IVgr.VL=1A~CC4-Z4't 0$$$$$III3$$t$$IIE$$$$$$HeatIn)ectionExtreaaPressurixcrPORVOpeningTine1.1Sec.PressurixorPORVClosingTine~4.6Sec.RCSTemperature~BS.IOegFSteam6cneratorTesperature~136.~OegFAppend6Lieat12EFPYAppend6Lieat32EFPYcRCPSealLimitprrFigure3.10$$RCSPressureExtremavs.SetpointPressureatPORVStrokeOpenTimeof1.0Sec.andRCSTemperatureof85'F8921c:1d/0607893$25 k;0F>>+g"IIIt0>>Ite4~'I tlassIntestateEetrenaHesttntaettenEeireaaPreseurliatPORVOpeningTice2.0Sec.PteaaurizerPORVClaalngTine~4.8Sec.RCSTenperature>85.0DegFSteam6eneratorTenperatur~~135.IDegFhppend6Lln~t12EFPYhppend6Llnat32EFPYRCPSealLimitFigure3e11RCSPressureExtremavs.SetpointPressureatPORVStrokeOPenTimeof2a0Sec.andRCSTemPeratureof85'F$921e:1d/0607893.26 lWI~0)g+C,g'Wg'I'gI"1i,li.s'III<<lkvg' ttassZntscttonExtras>toastIotasiionExtrsxaPressurixerPORVOpeningTineaa4.1Sec.PressurizerPORVClosingTine4.bSec.RCSTenperatura~85.~DegFStean6enaratorTenperatur~sa135.~DegFhppand6Linat12EFPYhppand6Lie"at32EFPYRCPSealLimitFigure3a12RCSPressureExtrema.vs.SetpointPressureatPORVx~StrokeOpenTimeof4.0Sec.andRCSTemperatureof85'FBS21e:1d/0607BS3.27 10gtt~'gtiJ'II0lt44I4'e+t.'

IlatsIntacttcnEntrantHeatIn>ectionExtrenaPteaeuriaerPORVOpen}ngTine6.8Sec.Preaauri?erPORVClosingTine~4.6Sec.RCSTenperature~85.SOepFStean6eneratOrTeEEperatureEs135.1DegFhppend6'Linat12EFPYhppend6LiEEat.32EFPYIE~RCPSealLimitFigure,.3..13.,RCS.PressureExtrema,vs.SetpointPressureatPORVStrokeOpenTimeof6.0Sec.andRCSTemperatureof85'F8921e:1d/0607893a28 I4gttCJ+\g$,1NtVg'AD.

tlaaaIntaattnnEatraaaHastZnjaattnnEatraaaPressurizerPORVOpeningTine~B.bSeenPressurizerPOAVClosingTine~4.8Sec.RCSTenperatut~~86.bDegFStean6eneratorTenper~ture~l3S.~DegFi.tataa'lt>>an"'tahppend6Lieat12EFPYhppend6Linat32EFPYRCPSealLimit""""Figure-3.14""RCS"Pressure.Extremavs.SetpointPressureatPORVStrokeOpenTimeofSa0Sec.andRCSTemperatureof85'FBB21a:1d/0607893.29 4tr1'tJ,rk1,ji RaaeInleatlanEntrantHeatInfectionExtremePressurizerPORVOpeningTine~IB.BSacPresaurixerPORVCloaingTine~4.BSec.RCSTenperature~85eBOegFSteanGeneratorTeRRperature~135.BOegFhppend6Linat12EFPYhppend6Linat32EFPY,RCPSealLtnttFigure3.15RCSPressureExtremavs.SetpointPressureatPORVStrokeOpenTimeof10.0Sec.andRCSTemperatureof85'FSS21e:1d/0607893.30

>>,II~I~II>>>>'1yg>>p~ping'(>>>>'">>I>>.

nessinteottonEntreesHestIntenttonEatreaaPressurizerPORVOpeningTinean1.6Sec.PreaaurizerPORllClosingTine~4.8Sec.RCSTellperature~150.8DegFStean6eneratorTemperature288.0DegFAppend6Lieat12EFPYAppend6Lieat32EFPYRCPSealLimit'igure3.16RCSPressureExtremavs.SetpointPressureatPORVStrokeOpenTimeofla0Sec.andRCSTemperatureof150F892le:1d/0607893.31

~aarah~arKfa I!axeInteattnnExtreneHeatIniaatinnExiranePreaaurixetPORVOpeningTine~2.8Sec.PreasurizerPORVClosingTine~4.8SeenRCSTemperature~150.8.8OegFStean6eneratorTenperature288.8.8OegFAppend6Llaat12EFPYAppend6Li~at32EFPYnCptealLtnti"Figure3.17RCSPressureExtremavs.SetpointPressureatPORVStrokeOpenTimeof2.0SectandRCSTemperatureof150'F8921e:1d/0607893.32 II0"\"\j4l lianaIntaattanEntranaHastIntanitanEntrantPressurizerVORVOpeningTine~4.8Sec.PrassurizerPORVClosingTine4.8Sec.RCSTenperature~158.8.8DegFStean6anaratorTenper~tur~~28B~8.8DegFhppend6Lillat12EFPYhppend6Linat32EFPYRCPSealLimitnPrr(Figure3.18RCSPressureExtremavs.SetpointPressureatPORVStrokeOpenTimeof4.0Sec.andRCSTemperatureat1SO'F8921c:1d/0607893.33 0sl~(VIP*P4h~*t naaaIntentionExiraxeHantIntentionExtranaPreasurizerPORVOpeninpTine~6.8Sec.PreeeurizerPORVCloainpTine~4.SSec.RCSTemperature~15I.IOepFhStean6eneratorTemperature2BB.BOepF~,ahAppend6Linat12EFPYAppend6Lieat32EFPYRCPSealLinttFigure3.19RCSPressureExtremavs.SetpointPressureatPORVStrokeOpenTimeof6;0Sec.andRCSTemperatureof150'F8921e:1d/D607893.34 e~~

n'ianaIntaattenEntreatHeatInteetienEetlaaaPressurizerPORVOpeningTine~8.8Sec.PressurizerPORVClosingTime~4.8Sac.RCSTenparatur~i150.8DegFStean6enersterTemperatua~288.~DegFAppend6Lieat12EFPYAppend6Lieat32EFPYRCPSealLimit"-Figure"3;20RCSPressure."Extremavs.SetpointPressureatPORVStrokeOpenTimeof8.0Sec;andRCSTemperatureof150'F8621e:1d/0607893.35

~QQ.0I4ee1P' hansIniaoiionExtras'astInianiionExisannPreaeuriverPORVOpeninpTine~1B.BSec.PressurizerPORVCloainpTineaa4.BSecsRCSTemperature~~l5I.~DepFStean6eneratorTenperatur~2BB.BOepFAppend6Lieat12EFPYAppend6Lieat32EFPYRCpOaalLsnst~Figure3a21RCSPressureExtremavs.SetpointPressureatPORVStrokeOpenTimeof10.0Sec.andRCSTemperatureof150'F892le:ld/0607893.36 JI0IIr5'II1rw~~'Yv4Md6%

Notea.1)NoPreaau~InatrunantError2)SinglePORVOperation5)PORVCloauraTtne~i.bSec.i)ReactorVeaaelExpoaur~~12EFPY)SB.BOepFRCSTeno.]BS.BOegFFigure3.22.PORYLTOPSetpointat12EFPYvs.ValveOpeningTimeB921e:1d/060/893.37

  • g~C Roice:1)NoPressureInstrumentError2)SinglePORVOperation3)PORVCloaur~Tine~i.bSec.4)ReactorVesselExposure~32EFPY1a158.$OegFRCSTemp.~&S.ODegFFigure'3.23"PORVLTOPSetpointat32EFPYvs.ValveOpeningTime8921e:1d/0607893.38

~%!la0'(I%IV14r>>

'.0CORRELATION'OMOGLTOPSSETPOINTMETHODOLOGYAspartoftheLTOPS'etpointanalysis"performedforD.C.CookUnit2,AmericanElectricPowerCorporationrequestedacorrelationthatbenchmarkstheresultsoftheanalysistothatofthealgorithmdescribedinthereport.preparedfortheMestinghouseOwnersGroup(acronymedMOG)onReactorCoolantSystemOver-pressurizationbyMestinghouseElectricCorporation.ThereasonfortherequestwastoprovideameanstodeterminetheequivalentLOFTRANderivedsetpointsfromtheexecutionofarelativelysimplealgorithm;.i.e.,'heMOGreport.Thecorrelationassumesamas'sinjectionevent,only.Thisisjustifiedonthefollowingbases:1)theLTOPS>>atD.C.Cookunit2featuresasetpoint'!'-~"::-~>'independent,"of".,temperature,')'he-mostlimiting'conditionisatlowtemperature,and3)themassinjectioneventdominatesatlowtemperatures.,Undertheseconditions,aswillbeshown,thecorrelationtakestheformofaseriesofcurvesof"LOFTRAN"setpointsplottedagainstpressurizerPORVopeningtime,with;the"MOG"setpointsasaparametric.4.1MOGMETHODOLOGYLTOPSSETPOINTS,The,MOGmethodology,formass.injectionevents,determinesthe.resulting.over-,pressurefroma4-factorformula;thederivation,ofwhich,isprovidedinthe","'"MOGreport.'referencedin"theintroduction.tothisreport:DeltaP=DeltaP(Ref)~F(v)"F(s)~F(z)Thefactorscomprisingtheformulaare-determinedfromaseriesoflinearrela-tionshipswithknown(orassumed)plantparameters.Forconvenience,theserelationshipsarereproducedherefromMOG:report,andareshownasFigures4.1through4.4.Fortheparameterstudyperformedhere,itisconvenienttoexpressthesefiguresanalytically:DeltaP(Ref)=~"(HassInjectionRate,ibm/sec)82B921e:1d/060ZSB4.1

~\'11~aC'4k,W.',Stt>>'P"tt"q*."tt~~~r1~.~~~*ti4>>'I+

mme=.A..~RESSIIIthPREF~RefcftneeOvershootFI6UREi.t.1Ik+-j-"~i"8:."...xtkssInputRate-lb/see"".FFigure4.1OeltaP(Ref)vs.MassInjectionRate8921e:1d/0607894.2 0)ITr.tr,sI14~y~4>>~I<<gs M~~~~SSIn)ItFVRCSVolteFactorIF16UREi.2.2,'4:AlMM=--2~\~~~KK-~~~~~~~~~r~~~'I~t~'I~"I~OOO~I'il.:".Bi:.!iiV.TotalRCSVol~-ft'SOOOi:-Ii:.:.-'z-.:Figure4.2VolumeFactor,F(v),vs.RCSTotalVo1ume8921e:1d/06D7894.3 0A1141~t~Clhtpd amassInputF>-kcllefHiveOpeningTksIeFactorFICHEa,2~~~I'~>>I'"If~~0~~~~~~::::LII~~~r>>~~I)I'I~~~~~I>><<~--ll~~%11>>as>>I>><<I<<y4I,>>>>1<<IL-~.L~~~~~~~~I<<I~~>>>>IMP~I-<<0<<nq<<1g>>~M">>~>>.I...~L1-i=~>>I=Z,kelfefValveOpeningTfaIe,secondsIiiiiI:'III0Figure4.3PORYOpeningTimeFactor,F(z),vs.ValveOpeningTime8921e:1d/0608894.4 0~w~C~at+I"'8IraM'"Ye1'fp~'4'.erEt)irlg~k~

~llnnnInnt=I:.:II::.".:.FS-ReliefValveSetpo)ntfattnp'll."'I:-'-:.-,~-.I::".~~.:i--'i:-:I.;::i:~'.IWn.-J.-~~-w4~~>>I+fl+YII.~l>>nnY~>>Lfn+ng+}QIfI---->>'IY$Ii.~~~~M=~:.-~>>.II~A~d;)IIYNNW~nYn~\~"-:IL',Rel1afValveSetpoknt-palyI:.:.'.:.I:.:.:.:.-:.I.':I-I.:i~::~I:.:.'::.:IIiFigure4.4LTOPSSetpointFactor,F(s),vs.LTOPSSetpoint8821~:1d/0608894.5

~Cf?fJCWfCV~C??~C*'f*L~jfftf

'(v)=1.435-(0.725E"4)~(RCSVolume,cu.ft.)=0.53forCookunit2.(Cookunit2coldRCSvolume=12509.2cu.ft.)F(s)=1.810-(0.00135)~(PORVSetpoint,psig)F(z)=0.200+(0.267)~(PORYOpeningTime,sec.)Theoverpressuresresultingfromtheapplicationofthesefactorsaredocumentedintabularformbelowandthroughpage4.9,fortheseveralPORV'opening-'-times'electedfortheanalysis.,Asummaryoftheoverpressuresisgivenonpage4.9.MassInectionOveressuresatPORVOneninTime=1.0sec.LTOPSSetointsi400500600700Inj.Mass(refTable3.3)(gpm)(ibm/sec.)DeltaP(Ref)......,.4'(v)0~~~I-~~0.~~.i~F(s)o~~~~~~~~~F(z)oo~~~~~~o~DeltaP(psig).....Overpressure(psig)439.260.9683.31..0.531.2700.46726.20426.2429.759.6481.51-.0.531.1350.46722.90522.9420.258.3279.700.531.0000.46719.70619.7410.757.0177.910.530.8650.46716.70716.78921e:1d/0608894.6 S(E/E+EE)healEEtI~QE$IIII'E~r~liE5lgI

,MassIn'ectionOveroressuresatPORVOoeninTime=2.0sec.LTOPSSetpointsi400500600700Inj.Mass(refTable(gpm)~~~~e~(ibm/sec.).OeltaP(Ref)....F(v)~~.~~~~~~F(s)e~~oo~~~F(x)o~~~~~~~-,6=,-.'Oe1.ta,P",.(psig),...,.Overpressure(psig)3.3)439.260.9683.310.531.2700.73441.20441.2429.759.6481.510.531.1350.734,'36.00536.0420.258.3279.700.531.0000,73431.00631.0410.757.0177.910.530.8650.73426.20726.2MassIn'ectionOveroressuresatPORVeninTime=4.0sec.LTOPSSet@ointsi400500600700Inj.Mass(refTable(gPlll)o~~~~~(ibm/sec.).DeltaP(Ref).'..F(V)~~~~~~~~F(s)o~~~~~e~F(Z)~~~~~~~~OeltaP(psig)Overpressure(psig).3.3)439.260.9683.310.531.2701.26871.10471.1429.759.6481.510.531.1351.268.62.20562.2420.258.3279.700.53;1.0001.26853.60-"653.6410.757,0177.910.530.8651.26845.30745.36921e:ld/0607894.7 0t~~r'v.svv:~hELf~rII'r.)

.MassInectionOverressuresatPORVOneninTime=6.0sec.LTOPSSetoointsi400500600700Inj.Mass(refTable(gPlll)~~~~~,~(ibm/sec.)DeltaP(Ref)F(v)J~~~~~~~f(s)J~~~~~~o.F(z)i~~~~~~~.~<<'Del,ta,.P.(psig).........Overpressure(psig).3.3)439.260.9683,31~~~~0J531.2701.802.......,..101.0501.0429.759.6481.510.531.1351.802,...,88.40588.4420.258.3279.700.531.0001.80276.10676.1410.757.0177.910.530.8651.80264.40764.4MassInectionOveroressuresatPORYeninTime=8.0sec.~~LTOPSSetoointsi400.500600700Inj.Mass(refTable(gpm)J~~~~~(ibm/sec.)OeltaP(Ref)F(v)a~~~~~~~F(s)J~~~~~~~f(I)~~~~~~~~'eltaP(psig).Overpressure(psig).3.3)439.260.9683.310.531.2702.336131.0531.0'429.759.64.'1.510.531.1352.33611'4.5614.5420.258.3279.700.531.0002.33698.70698.7410.757.0177.910.530.8652.33683.40783.48921e:1d/0601894.8

~~

MassInectionOveressuresatPORVeninTime=10.0sec.LTOPSSetoointsi400500500700Inj.Mass(refTable(gpm)0~~~~~(ibm/sec.)DeltaP(Ref)F(Y)F(s)i~~~~o~~F(x)o~~~~~~~.',DeltaP,,(psig).Overpressure(psig).3.3)439.260.9683.310.531.2702.870160.9560.9429.759.6481.510.531.1352.870140.7640.7420.258.3279.700.531.0002.870121.2721.2410.757.0177.910.530.8652.870102.5802.5RCSOveroressureSummarLTOPSSetpointsiPORVOpeningTimesec400500600700..1.02.04.06.08.010.0426.2.....,522.9441.2536.0471.1S62.2501.0588.4531.0614.5'560'.9640.7619.7631.0653.6676.1698.7721.Z716.7726.2745.3764.4783.4805.5&921e:1d/0807&94.9 C.4~VI~4H.(0~W1~~riel4~-~eP~t4H1VM~4~I"htlt+40'I&t~'4+4%EsW10 ThesummaryisshowninFigure4.5,illustratingthehighdegreeoflinearityIofthepeaksystempressureasafunctionofsetpointpressure.Thepeaksystempressure,resultingfromimplementationoftheHOGmethodology,can,thereforbeexpressedasalinearfunctionofsetpointpressure:1)PeakSystemPress=A+B~(P)MOSandthepeaksystempressuremustbelessthantheAppendixGlimit.Thecoefficientsoftheequationhavebeendeterminedbyperformingaleast-squaresfitonthepeakpressures(fromtheabovetable)asafunctionofsetpointpressure:PORVOpeningTimesecCoefficients1.02.04.06.08.010.038.8161.10105.35149.63194.13234.210.96830.95000.91400.87790.84140.81434.2LOFTRAN/MOGCorrelationThesystemoverpressuresdeterminedfromtheLOFTRANbasedanalysis'realsoquitelinear(referenceFigures3.10through3.15),andcanbeexpressedaslinearfunctionsofsetpointpressureforeachofthePORVopeningtimes:2)PeakSystemPress=C:~D"(Pi)LOFT89216:1d/0601894.10 0Qc+lCl~M4' 18.88.8-~PORVOpeningTine(sec)q6.82.81.8Figure4.5MaximumRCSOverpressrueSasedon"MOG"Methodologyvs.LTOPSSetpointPressure6921~:1d/0607694.11

$f);>+0ra1L awiththepeaksystempressuresrequiredtobe'lessthantheAppendixGlimit.~~Aleast-squaresfitofthedatain,Table3.4resultsin.thefollowingcoefficients:PORVOpeningTimesecCoefficientsD1.02.04.06.08.010.047.8055.5076.9096.50115.40132.000.97900.97500.95700.94000.92700.9150ThepeaksystempressuredeterminedbyeithertheMOGalgorithmortheLOFTRANbasedanalysesmustbelessthantheAppendixGlimit.Therefore,atthelimit,theoverpressuredeterminedfromequation1mustbeequaltotheoverpressuredeterminedfromequation2:+D~(PLoFT~~A+B~~PMos)wIJL0FTA'-C'Mos8921e:)d/0608894.12 wPs0~LA%19UOk, Thecoefficientsr'esultingfromthecombinedequations(1)and(2)areas.Cfollows:1CoefficientsPORVOpeningTimesecA-C8'1.02.04.06.08.010.0"9.185.7429.7356.5284.93111.700.98910.97440.95510.93390.90770.8899Usingtheabovetableofcoefficients,theLOFTRANequivalentLTOPSsetpointscorrespondingtoaseriesofselectedHOGsetpointsistabulatedbelow:SummarofLOFTRANEuivalentLTOPSSetoointssiPORYOpening'ime:secLTOPSSetointsiSasedonMOGAlorithm3004005006007008001.02.04.06.08.010.0287.6298.1316.3336.7357.2378.7386.5395.5411.8430.1.448.0467.7485.4584.3492.9590.4507,3602.8523.5616.9-538.8-629.6556.7645.6683.2687.8698.3710.3...720.3734.6782.1785.3793.8803.6.,811.1.823.68921e:1d/OB07B94.13 V<f0~lAgO~t,"%4*k=I 4'hetabulation"isshowngraphicallyinFigure4.6,andplotstheLOFTRAN'derivedLTOPSsetpointasafunctionofPORVopeningtime;parametricwiththetWOGLTOPSsetpoint.ThisfigureisusedtotranslatetheLTOPSsetpointderivedfromapplicationoftheWOGmethodologytotheLOFTRANequivalentvalue.UtilizationofthefigurerequiresthattheLTOPSsetpointfirstbedeterminedusingtheWOGmethodology,AtthePORVopeningtimecorrespondingtothatselectedfortheWOGcalculation,determinetheLOFTRANanalysissetpointfromtheordinatebylinearlyinterpolatingbetweenthetwocurvesboundingtheWOGsetpoint.4.3IMPACTOFSTEAMGENERATORTUBEPLUGGINGBoththeLOFTRANandtheWOGbasedanalyseswereperformedassumingnotubes'plugged"inthesteamgenerators.The-impactoftubepluggingisasmallreductioninRCSvolume;theconsequence,ofwhich,isslightlyhigherover-pressuresasaresultofmassinjectionevents,'andreducedoverpressuresfromheatinputevents.Theheatinputeventsarereducedinimportancebecauseofthereducti'oninheattransfersurfaceareaofthesteamgenerators.TheimportanceoftubepluggingwithrespecttoitsimpactonLTOPSsetpointsisaccountedforbytheF(v)termintheWOGmethodology(referenceFigure4.2),andisdirectlytranslatabletotheLOFTRANbasedanalysisthroughthecorrelationdevelopedinthischapter.~Asaworstcaseexample,reducingthenumberofsteamgeneratortubesby154overallfoursteamgenerators,'resultsinareductioninRCScoldvolumeof492..6cu.ft.(basedonanaveragetubelengthof69.77ft.,atubeO.D.of0.875inches,andawallthicknessof0.050inches).ThisrepresentsafractionalreductionininitialRCSvolumeofabitlessthan4A(i.eee0.0394)..TheimpactonoverpressurecanbedeterminedfromtheF(v)equationin:section4.1.Withouttubeplugging(RCS:volume=12509.2cu.ft.),F(v)=0.53.'With15Kofthetubesplugged(RCS-volume=12016.6cu.ft.),F(v)increasesto0.56.Thisrepresentsanincreaseinthedeltaoverpressoreof1.068,almosta7%increase.'9216:1d/0607894.14 O~~

MOSLTOPSSetpnt(yahoo)Figure4.6LOFTRAN/HOGSetpointCorrelationvs.PORVOpeningTime89216:1d/0607894.15 l~40Wkc4't\~4,,tt-4WL4J4+4I-4'PtP$l+~[h1'i')rI'A\I'4l0-

~hSECTICN4(KRATIN6INSTRUCTINSIIITAIDTlGIHl&SH93TI%FiOW4'+lipAOuSWAH.uc:0c~~uop~<<byBa"kCp~-o+s)falNORMA4MININNpNDMAXIMLMOPERATINGVALUES4.1.1Pump1.IIo.ISea1ITEMUNITNORMALMINIMUMMAXIMUMNOTESPlowgpmSeeFigure4-00.25.01,2,3TemperaturedegreesF100-19060235Pressurepsig22503252485fvl$4iHIA4)4'1@Co't%~rl%~hPpsi22172002470HOTOPERATIONALRECOMMENOKOALARMSETTINGSHIGHfLOW-5.0GPMLOWFLOW-.TGPMO3C7I4lIQg2.::SAFE'PERATINGRANGK.',~7~20III0200400600RAAIA>0I."CIAl400IRAA1800200022002400I'"'502'IOOSgg.DlmlNrmllnlI~HIp,>@IllI!Ql.1$AOOOROIFIGURE4-0tlo.1SealPerformanceParameters4-1Rev.1 NO-C0~5POl~.0

  • NoresDu'ringheatuporcooldown,vhenthesystemvacerpressureis1000'psigorbelow,leak~ffElowmaybeinsufficienttocaolthebearingandsealcompo-nents.Whencheleak-offflowisbelov1gpm,thesealbypassvalveshouldbe-opened.ThispermitsalimiredflovtobypasstheNo.1sealthroughanon-ad)useableorificeblock(externaltothepumpitself).2~Iftheleak-offflowislessthan0.2gpm,icisprobablechar.minorforeignmacterisrestrictingcheElovacthesealfaceinlet.Thesealfacesactlikeafilter.Foreignparticlesintheorderaf25-40micronscancollectbetweenthefacesandrestricr.theflow.Increasingthesealdif-ferentialpressure(hP)mayclearcheseal.Ificdoesnoc,decreasechesealdifferentialpressureto100psiandturnchepumprocorbyhand.Donotstartthepumpmotorifthesealflowisbelovchespecifiedminimum.3~IfaslowincreaseincheNo.1sealleak-offElavisobserved,i.e.overaperiodofseveraLweeksormore,Wescinghouseshouldbenotifiedcoprovideguidance.DuringthisperiodchepumpmaybeoperacedandtheNo.1sealLeak~ffvalveshouldbeLefr.open.Thesealleak-offElovshouldnotbepermittedtoexceedchelimitsofthesafeoperatingrangeofFigure4W.Iftheseconditionsdonocexist,theproceduresforemergencyoperationshownundertheNo.2sealshouldbefollowed.4.,Thistemperatureismeasuredbyachermocoupleac,theoutletofcheNo.1seal.Themaximumvalueshownshouldnotbeexceededeitherinnormalserviceorduringaloss-of-injectioncondition.Theminimumlooppressure(325psig)appliesonlyduringthefillingandventingoperation.Theminimumlooppressureforsubsequentoperationsiscancrolledby>theminimum4PacrosscheNo.1seal(200psi).2.No.2Seal,MIHlt%Mf"AXINNNOTES;"FlowTemperature%LetPressurgphdegreesFpsiN/A33..NegligibleN/A15N/A75psi3073Flow12TemperaturedegreesFN/AN/Apsi2235N/Aif/A*NotesNormaloperation-No.1sealoperacive.2.TheHo.2seeltemperaturewillvaryviththeNo.1sealtemperacureandisnocconsideredi,nfozmntivc.4-LaM(v.I

~~t~'

3.Establishedbyprevailingsystemconditions.4;Emergencyoperation-No.lsealinoperative,withaprimarypressuredropoccurringacrosstheNo.2seal.Thefollowingactionshouldbetaken:a.ClosetheNo.1leak-offvalvewithinfiveminutes.b.Prepareforpumpshutdown.Thepumpmaybeoperatedforaperiodnottoexceedanadditional30minutes.Duringthisperiod,thereactorpowershouldberampeddowntotheN-1allowablepowerlevel,whereNisthenumberofoperatingRCpumps.c.Securethepump.d.Donotrestartthepumpuntilthecauseofthesealmalfunctionhasbeendetermined.3.No.3SealPlowTemperatureUNITcc/hrdegreesF100N/AMINIMlN.NegligibleN/ANXINN200N/ApsiFlowTemperaturegphdegreesF-N/AN/AN/AN/AN/Apsi15N/AN/A*Notes1.,Normal.operation-No.1sealoperative.2.TheNo.3sealtemperaturesarenotconsideredinformative.3.Thesea1differentialpressureLsestablishedbytheheadtank.4.Emergencyoperation-No.1seal,inoperative,withaprimarypressuredropoccurringacrosstheNo.2seal.Refertonote4ofparagraph4.1.1.1.4.lnjectlonWaterITEMUNITMINIMLNNXIMLHNOTESPlow12TemperaturePressuredegreesF130N/A601'50N/A4-lbRev.1

"~0's4~-qp)i<q~~~k>~t.pi~et~e~@ewa>~tWcI'

~<*Notes..1....The,normalflowdistributionis5gpm'intothesystemand3gpmforthesealsupply.2.Ifthein]ection~aterisincreasedto150'F,~thereactorcoolanttempera-tureshouldbead)ustedtoatemperaturenottoexceed400'F.3.Infectionwaterpressureisad)ustedtoobtaintherequiredflow.5.ThermalBarrierCooiingWaterMINION/AXINGNOTESFlow403560TemperaturePressuredegreesFpsi8015060N/A105200*Notes1.Watershallbesuppliedfromthenon-radioactivecomponentcoolingsystem.Pressureshallbeadequaretoensuretherequiredflow.6.BearingMaterITEMUN[TMININNTemperaturedegreesF160Ambient225*Note1.Refertoparagraph4-5-1,items1and2(LossofIn]ectionWaterandHighTemperatureofInjectionWater).7.SealPurgeWaterTheNo.3sealisfittedwithaconnectionforpurgewatertominimizethebuildupofboricacidcrystalsacthetopoftheseal.Ifpurgeflowisconsiderednecessary,2to4gphofcool,clean,deminerali-ed,non-boratedwatershouldbeinjected.ThepurgewaterdrainsoutthroughthenormalNo.3sealleak-offline.Whenpurge~aterisused,theflowsofthetableinpara-graph4-1.1-3donotapply.8.AlarmSettings,1Alarmsettingsshouldbesetinaccordancewiththemaximumorminimumvaluesown.on"theprecedingcharrs.4-lcRev.1

'00 LICENSINGREPORTFORSTORAGEDENSIFICATIONOFD.C.COOKSPENTFUELPOOLINDIANAMICHIGANPOKERCOMPTE&byHoltecInternational<<bLAEPSCContractNo.C-7926HoltecProject00480 I'IIII DOCUMENTNAME:HOLTECINTERNATIONALREVIEWANDCERTIFICATIONLOGLICENSINGREPORTFORSTORAGEDENSIFICATIONOFD.C.COOKSPENTFUELPOOLHOLTECDOCUMENTIAD.NO.HOLTECPROJECTNO.CUSTOMER/CLIENT:HZ-90488AMERICANELECTRICPOWER(INDIANAMZCHZCANPOWERCO.)REVISIONBLOCKISSUENO.ORIGINALREVISION1REVISION2REVISION3REVISION4AUTHOR&DATE57mC'~g((q@Cg~<>+resoAc~y9~8'tYHl5'sl9~REVIEWER&DATEQSiIarrvCOflr~Cs9/~t$9(lgg.lnfQ.A.MANAGER.&DATEgl~ClV+,5o(rgl.c>>1~~rc~..r'FirCMn.APPROVEDBY&DATEM~l~+c+Cc((9'lV4+ggCi0<REVISION5REVISION6Ylv@4Rid8So&~PPEIrb]~IC/g~~GO~'/NOTE:Signaturesandprintednamesarerequiredinthereviewblock.MustbeProjectManagerorhisDesignee.Thisdocumentconformstotherequirementofthedesignspecificationandtheapplicablesectionsofthegoverningcodes.Thisdocumentbearstheinkstampoftheprofessionalengineerwhoiscertifyingthisdocument.SEALrggs}ggPALSlHGHt",12G906ElJ>g"3PProesszonalEngineer I

SUMMARYOFREVISZONSRevision1containsthefollowingnumberofpagesoftextincludintablesbutexcudinfiuresTitlePageReviewandCertificationLogSummaryofRevisionsPageTableofContentsListofFiguresSection1Section2Section3Section4Section5Section6Section7Section8.Section9Section10111438178(later)2348513(later)11Revision2containsthefollowingnumberofpagesoftextincud'ntablesand'esTitlePageReviewandCertificationLogSummaryofRevisionsPageTableofContentsListofFiguresSection1Section2Section3Section4Section5Section6Section7Section8Section9Section101114381814NotincludedinRevision23378517NotincludedinRevision2NotincludedinRevision2 i)~

SUMMARYOFREVISIONSRevision3containsthefollowingnumberofpagesoftextinclud'ntablesandfiuresTitlePageReviewandCertificationLogSummaryofRevisionsPageTableofContentsListofFiguresSection1Section2Section3Section4AppendixAtoSection4Section5Section6Section7Section8Section9Section10Sectionll11243919143493376517ll54Revision4containsthesamenumberofpagesasRevision3withtheseexcetions:ListofTables(addedtoRev.4):3SectionsrevisedinRev.4nowcontainthefollowingnumberofpages:Section1Section2Section4Section5Section99183534llIndividualpagesrevisedandtransmittedinRevision4are:Pages3-1,3-3<3-4,6>>6,6-15,6-18,6-28,6-30,7-4,7-5,7-6,8,7and10-2.

SUMMARYOFREVISIONSHoltecReportHZ-90488Revision5ThefollowingisrevisedinRevision5:Pages4-8and4-9Section9,AppendixAPagevofTableofContentsRevision6ThefollowinaesarerevisedinRevision6:ListofFiguresTableofContents(pagev)2-14-15,4-1652I53I55I56I58~59I510I512I515I516I517~518'-19,5-20,5-21,5-24through5-386-3,6-4,6-357-2,7-48-6,8-7,8-1310-411-3 4f~lI~~~~lI~t TABLEOFCONTENTS

1.0INTRODUCTION

2.0MODULEDATA2.1SynopsisofNewModules2.2MixedZoneTwoRegionStorage(MZTR)2.3MaterialConsiderations2.3.1Introduction2.3.2StructuralMaterials2.3.3PoisonMaterial2.3.4CompatibilitywithCoolant2.4ExistingRackModulesandProposedRerackingOperation3.0CONSTRUCTIONOFRACKMODULES3.1FabricationObjective3.2MixedZoneTwoRegionStorage3.3AnatomyofRackModules3.4Codes,StandardsandPracticesfortheD.C.CookSpentFuelPoolRacks3.5MaterialsofConstruction2-12-12-12-42-42-42-42-72~73-13-13-23-23-53-94.0CRITICALITYSAFETYANALYSES4.1DesignBasis4-14-14.2SummaryofCriticalityAnalyses4.2.1NormalOperatingConditions4.2.2AbnormalandAccidentConditions4-4444-64.3ReferenceFuelStorageCells4.3.1ReferenceFuelAssembly4.3.2HighDensityFuelStorageCells4-84-84-9

~~~f TABLEOFCONTENTS(continued)4.4AnalyticalMethodology4.4.1ReferenceDesignCalculations4.4.2FuelBurnupCalculationsandUncertainties4.4.3EffectofAxialBurnupDistribution4-104-104-124-134.5CriticalityAnalysesand'olerances4.5.1NominalDesign4.5.2UncertaintiesduetoManufacturingTolerances4.5.2.1BoronLoadingTolerances4.5.2.2BoralWidthTolerance4.5.2.3ToleranceinCellLatticeSpacing4.5.2.4StainlessSteelThicknessTolerances4.5.2.5FuelEnrichmentandDensityTolerances4.5.3Water-gapSpacingBetweenModules4.5.4EccentricFuelPositioning4-154-154-154-154-164-164-164-164-174-174.6Abnormal4.6.14.6.24.6.34.6.4andAccidentConditionsTemperatureandWaterDensityEffectsDroppedFuelAssemblyLateralRackMovementAbnormalLocationofaFuelAssembly4-174-174-184-184-194.7ExistingSpentFuel4.8References5.0THERMAL-HYDRAULICCONSIDERATIONS5.1Introduction5.2SpentFuelCoolingSystemDescription5.2.1SystemFunctions5.2.2SystemDescription5.2.3PerformanceRequirements5.3DecayHeatLoadCalculations4-194-215-15-25-25-35-45-4

~~~~lI TABLEOFCONTENTS(continued)5.4DischargeScenarios5.5BulkPoolTemperatures5.6LocalPoolWaterTemperature5.6.1Basis5.6.2ModelDescription5.7CladdingTemperature5.8BlockedCellAnalysis5.9ReferencesforSection56.0RACKSTRUCTURALCONSIDERATIONS6.1Introduction6.2AnalysisOutline6.3ArtificialSlabMotions6.4OutlineofSingleRack3-DAnalysis6.5DynamicModelfortheSingleRackAnalysis6.5.1Assumptions6.5.2ModelDescription6.5.3FluidCoupling6.5.4Damping6.5.5Impact6.6AssemblyoftheDynamicModel6.7TimeIntegrationoftheEquationsofMotion6.7.1TimeHistoryAnalysisUsingMulti-DegreeofFreedomRackModel~6.7.2EvaluationofPotentialforInter-RackImpact.6.8StructuralAcceptanceCriteria6.9MaterialProperties5-55-65-115-115-125-135-165-166-16-16-26-36-56<<76-96-116-126-136-136-146-176-176-196-196-21 c'l TABLEOFCONTENTS(continued)6.10StressLimitsforVariousConditions6.10.1NormalandUpsetConditions(LevelAorLevelB)LevelDServiceLimits6.11ResultsfortheAnalysisofSpentFuelRacksUsingaSingleRackModeland3-DSeismicMotion6.12ImpactAnalyses6.12.1ImpactLoadingbetweenFuelAssemblyandCellWall6.12.2ImpactsbetweenAdjacentRacks6.13WeldStresses6.13.1BaseplatetoRackWeldsandCell-to-CellWelds6.13.2HeatingofanIsolatedCell6.14WholePoolMulti-RackAnalysis6.14.1Multi-RackModel6.14.2ResultsofMulti-RackAnalysis6.15BearingPadAnalysis6.16ReferencesforSection67.0ACCIDENTANALYSISANDMISCELLANEOUSSTRUCTURALEVALUATIONS7.1Introduction7.2RefuelingAccidents7.2.1DroppedFuelAssembly7.3LocalBucklingofFuelCellWalls7.4AnalysisofWeldingJointsinRackduetoIsolatedHotCell7.5CraneUpliftLoadof3000lbs.7.6ReferencesforSection76-226-226-256-256-286-286-286-316-296-306-306-326-346-366-377-17-17-17-17-27~37-47-4

TABLEOFCONTENTS8.0STATICANDDYNAMICANALYSESOFFUELPOOLSTRUCTURE8.1Introduction8.2GeneralFeaturesoftheModel8.3LoadingConditions8.4ResultsofAnalyses8.5PoolLiner8.6Conclusions8.7ReferencesforSection89.0RADIOLOGICALEVALUATION8-18-38-68-108-118-118-129-19.1FuelHandlingAccident9.1.1AssumptionsandSourceTermCalculations9.1-2Results9-19-19.2SolidRadwaste9.3GaseousReleases9.4PersonnelExposures9.5AnticipatedExposureduringReracking9.6ReferencesforSection910.0IN-SERVICESURVEILLANCEPROGRAM10.1Purpose9-59-59-59-69-810-110-110.2Coupon10'.110.2.210.2.310.2.410.2.510.2.6SurveillanceDescriptionofTestCouponsBenchmarkDataCouponReferenceDataAcceleratedSurveillancePost-IrradiationTestsAcceptanceCriteria10-210-210-310-310-410-410-410.3ReferencesforSection1010-5 Ii1~~g~~~~

TABLEOFCONTENTS11.0COST/BENEFITANALYSIS,11.1Introduction11.2ProjectCostAssessment11.3ResourceCommitment11.4EnvironmentAssessment 5I LISTOFTABLESTable1.1.1Table1.1.,2Table1.1.3Table2.1.1Table2.1.2Table2.1.3Table2.3.1Table2.3.2Table2.3.3Table2.3.4Table4.1Table4.2Table4.3Table4.4Table4.5Table4.6Table5.4.1Table5.4.2DischargeScheduleAvailableStorageintheDonaldC.CookPoolRackModuleData,ExistingandProposedRacksModuleDataCommonModuleDataModuleDataBoralExperienceList(DomesticandForeign)1100AlloyAluminumPhysicalandMechanicalPropertiesChemicalComposition(byweight)-Aluminum(1100Alloy)BoronCarbideChemicalComposition,Weight%BoronCarbidePhysicalPropertiesSummaryofCriticalitySafetyAnalysesNormalStorageConfigurationsSummaryofCriticalitySafetyAnalysesInterimCheckerboardLoadingReactivityEffectsofAbnormalandAccidentConditionsDesignBasisFuelAssemblySpecificationsReactivityEffectsofManufacturingTolerancesEffectofTemperatureandVoidonCalculatedReactivityofStorageRackFuelSpecificPowerandPoolCapacityDataDataforScenarios1through3 l~~~-ll LISTOFTABLES(continued)Table5.4.3Table5.5.1DataforScenarios1through3PoolBulkTemperatureandHeatGenerationRateDataTable5.5.2Table5.6.1Table5.6.2Table5.7.1Table6.3.1Table6.5.1Table6.6.1Time-to-BoilforVariousDischargeScenariosPeakingFactorDataDataforLocalTemperaturesLocalandCladdingTemperatureOutputDatafortheMaximumPoolWaterCondition(Case1)CorrelationCoefficientDegreesofFreedomNumberingSystemforGapElementsandFrictionElementsTable6.6.2Table6.9.1Table6.11.1Table6.11.2Table6.14.1Table6.14.2Table6.14.3Table6.14.4Table6.14.5Table8.4.1TypicalInputDataforRackAnalyses(lb-inchunits)RackMaterialData(200'F)SupportMaterialData(200'F)StressFactorsandRack-to-FuelImpactLoadRackDisplacementsandSupportLoadsRackNumberingandWeightInformationMaximumDisplacementsfromWPMRRunMPlMaximumDisplacementsfromWPMRRunMP2MaximumDisplacementsfromWPMRRunMP3MaximumRackDisplacementsandFootLoadSafetyFactorsforBendingofPoolStructureRegionsViL3.

I LISTOFTABLES(continued)InventoriesandConstantsofFissionProductRadionuclidesSignificantDataandAssumptionsfortheEvaluationoftheFuelHandlingAccidentTypicalConcentrationsofRadionuclidesintheSpentFuelPoolWaterPreliminaryEstimateofPerson-RemExposuresDuringRerackingDonaldC.CookNuclearPlantWorstCaseSpentFuelInventory I

LISTOFFIGURESFigure2.l.1Figure3.3.1Figure3.3.2Figure3.3.3Figure3.3.4Figure3.3.5Figure4.1Figure4.2Figure4.3Figure4.4Figure4.5Figure4.6CookSpentFuelPoolLayout(upperboundcellcount3616cells)SeamWeldingPrecisionFormedChannelsCompositeBoxAssemblyArrayofCellsforNon-FluxTrapModulesAdjustableSupportLegElevationViewofRackModuleNormalStoragePattern(MixedThreeZone)InterimStoragePattern(Checkerboard)AcceptableBurnupDomaininRegions2&3FuelStorageCellCrossSectionKENOCalculationalModelEquivalentEnrichmentforSpentFuelatVariousBurnupsforInitialEnrichmentof4.95%Figure4.7Figure4.8Figure5.5.1Figure5.5.2Figure5.5.3Figure5.5.4Figure5.5.5Figure5.5.6Figure5.5.7EffectofWater-GapSpacingBetweenModulesonSystemReactivityAcceptableBurnupDomaininRegions263ShowingExistingSpentFuelAssemblies.PoolBulkTemperatureModelDonaldC.CookSFPNormalDischarge,OneCoolingTrain,CaselaDonaldC.CookSFPNormalDischarge,OneCoolingTrain,Case1bDonaldC.CookSFPNormalDischarge,TwoCoolingTrains,Case2DonaldC.CookSFPFullCoreOffloadTwoCoolingTrains,Case3DonaldC.CookSFPFullCoreOffloadOneCoolingTrain,Case4CookSFPLossofCoolingScenario,Casela Il FigureFigure5.6.25.6.3Figure6.2.1Figure6.3.1Figure6.3;2Figure6.3.3Figure5.5.8Figure5.5.9Figure5.5.10Figure5.5.11Figure5.6.1LISTOFFIGURES(continued)CookSFPLossofCoolingScenario,Case1bCookSFPLossofCoolingScenario,Case2CookSFPLossofCoolingScenario,Case3CookSFPLossofCoolingScenario,Case4Idealization'ofRackAssemblyThermalChimneyFlowModelConvectionCurrentsinthePoolPictorialViewofRackStructureDBE-N-SAccelerationTimeHistoryDBE-E-WAccelerationTimeHistoryDBE-VerticalAccelerationTimeHistoryLISTOFFIGURES(continued)Figure6.3.4Figure6.3.5Figure6.3.6Figure6.3.7Figure6.3.8Figure6.3.9Figure6.3.10Figure6.3.11Figure6.3.12HorizontalDesignSpectrumandN-STimeHistorySpectrum(5%damping)HorizontalDesignSpectrumandE-WTimeHistorySpectrum(5%damping)VerticalDesignandTimeHistoryDerivedSpectra(5%damping)OBE-N-SAccelerationTimeHistoryOBE-E-WAccelerationTimeHistoryOBE-VerticalAccelerationTimeHistory'HorizontalDesignSpectrumandTimeHistoryDerivedN-SSpectrum(2%damping)HorizontalDesignSpectrumandE-WTimeHistoryDerivedSpectrum(2%damping)VerticalDesignandTimeHistoryDerivedSpectra(2%damping)

LISTOFFIGURES(continued)Figure6.5.1Figure6.5.2Figure6.5.3Figure6.5.4Figure6.5.5Figure6.5.6Figure6.6.1Figure6.14.1Figure6.14.2SchematicModelforDYNARACKRack-to-RackImpactSpringsImpactSpringArrangementatNodeiDegreesofFreedomModellingRackMotionRackDegreesofFreedomforX-ZPlaneBendingRackDegreesofFreedomforY-2PlaneBending2-DViewofRackModuleRackandFootPedestalNumberingforCookMulti-RackModelCookPoolMulti-RackSeismicAnalysis,RunMP2Rack16toRack17SouthCornerDynamicGapatRackTopFigure6.14.3Figure6.14.4CookRackRackCookRackRackPoolMulti-RackSeismicAnalysis,RunMP216toRack17SouthCornerDynamicGapatTopPoolMulti-RackSeismicAnalysis,RunMP312toRack18WEstCornerDynamicGapatTopFigure6.14.5Figure7.3.1Figure7.4.1Figure8.2.1Figure8.2.2Figure8.2.3Figure8.3.1CookPoolMulti-RackSeismicAnalysis,RunMP3Rack12toRack18EastCornerDynamicGapatRackTopLoadingonRackWallWeldedJointinRackIsometricViewofCookSpentFuelPoolOverallFiniteModelofCookPoolTopViewOverallFiniteModelofCookPoolBottomViewPedestalLoadvs.Time I

1.0INTRODUCTION

DonaldC.Cookisatwinunitpressurizedwaternuclearpowerreactorinstallationownedandoperated,byIndianaMichiganPowerCompany.DonaldC.CookreceiveditsconstructionpermitfromtheAECinMarch,1969,anditsoperatingLicenseinOctober,1974forUnit1andDecember1977forUnit2.ThetworeactorswentintocommercialoperationinAugust,1975(Unit1)andJuly,1978(Unit2),respectively.TheDonaldC.Cookfuelstoragesystemismadeupofafuelpool58'-31/8"longx39'-19/16"widewithanintegralcasklaydownarea.Thepoolpresentlycontains1367spentfuelstorageassembliesand36miscellaneoushardwareitems.Thus,outofthetotalinstalledstoragecapacityof2050storagecells,1403storagecellsarepresentlyoccupied.Sincethefullcorehas193fuelassembliesforbothDonaldC.Cookreactors,maintainingfullcoreoffloadcapabilityfromonereactorimpliesthat1857storagecells(2050minus193)areavailablefornormaloffloadstorage.Table1.1.1providesthedataonpreviousandprojectedfuelassemblydischargeintheDonaldC.Cookspentfuelpool.Table1.1.2,constructedfromTable1.1.1data,indicatesthatDonaldC.Cookwilllosefullcoredischargecapability(foronereactor)in1995.ThisprojectedlossoffullcoredischargecapabilitypromptedthepresentundertakingtoincreasespentfuelstoragecapabilityintheDonaldC.Cookpool.

IIIl ThepurposeofthislicensingsubmittalistoreracktheDonaldC.Cookpoolandequipitwithnewpoisonedhighdensitystoragerackscontaining3613storagecells.Thererackingalsoentailsrelocationofthethimbleplugtool,spentfuelhandlingtool,RodClusterControlAssembly(RCCA)changetool,andBurnablePoisonRodAssembly(BPRA)toolbracketstotheSouthwalladjacenttothecaskpit.Twentythreefree-standingpoisonedrackmodulespositionedwithaprescribedandgeometricallycontrolledgapbetweenthemwillcontainatotalof3613storagecells(including3trianglecellslocatedattheSW,NWandNEcornersofthepool).Outofthesecells,theperipheralcellslocatedineachrackmoduleareflux-trapcells*,andtheinterioronesareoftheso-callednon-fluxtraptype.Thestoragecellssuitableforstoringfreshfuel(upto5%enrichment)areuniquelyidentified(seeSection4.0,Figures4.1and4.2),andaresurroundedbynon-fluxtrapcellswhichhaveaburnuprestrictiononthefuelwhichtheycanstore.Consistentwiththeconceptoftworegionstorage,theplacementoffuelwithagivenburnupintheallowablelocationisadministrativelycontrolled.Nocreditistakenforsolubleboroninnormalrefuelingandfullcoreoffloadstorageconditions.Afluxtrapconstructionimpliesthatthereisawatergapbetweenadjacentstoragecellssuchthattheneutronsemanatingfromafuelassemblyarethermalizedbeforereachinganadjacentfuelassembly.

l Itisnotedthat,theproposedrerackingeffortwillincreasethenumberoflicensedstoragelocationsto3613and,.asindicatedinTable1'.2,willextendthedateoflossoffullcoredischarge,capabilitythroughtheyear2008.Table1.1~3presentskeycomparisondataforexistingandproposedrackmodulesforDonald'.Cook.Thenewspentfuelstorageracksarefree-standingandselfsupporting.TheprincipalconstructionmaterialsforthenewracksareSA240-Type304stainlesssteelsheetandplatestock,andSA564-630(precipitationhardenedstainlesssteel)fortheadjust-ablesupportspindles.Theonlynon-stainlessmaterialutilizedintherackistheneutronabsorbermaterialwhichisboroncarbideandaluminum-compositesandwichavailableunderthepatentedproductname"Boral".ThenewracksaredesignedandanalyzedinaccordancewithSectionIII,Division1,SubsectionNFoftheASMEBoilerandPressureVessel(B&PV)Code.Thematerialprocurement,analysis,andfabricationoftherackmodulesconformto10CFR50AppendixBrequirements.ThisLicensingReportdocuments.thedesignandanalysesperformedtodemonstratethatthenewspentfuelrackssatisfyallgoverningrequirementsoftheapplicablecodesandstandards,inparticular/"OTPositionforReviewandAcceptanceofSpentFuelStorageandHandlingApplications",USNRC(1978)and1979Addendumthereto.

IIN(l Thesafetyassessmentoftheproposedrackmodulesinvolveddemonstrationofitsthermal-hydraulic,criticalityandstructuraladequacy.Hydrothermaladequacyrequiresthatfuelcladdingwillnotfailduetoexcessivethermalstress,andthatthesteadystatepoolbulktemperaturewillremainwithinthelimitsprescribedforthespentfuelpooltosatisfythepoolstructuralstrengthconstraints.Demonstrationofstructuraladequacyprimarilyinvolvesanalysisshowingthatthefree-standingrackmoduleswillnotimpact.witheachotherorwiththepoolwallsunderthepostulatedDesignBasisEarthquake(DBE)andOperatingBasisEarthquake(OBE)events,andthattheprimarystressesintherackmodulestructurewillremainbelowtheASMEB&PVCodeallowables.Thestructuralqualificationalsoincludesanalyticaldemonstrationthatthesubcriticalityofthestoredfuelwillbemaintainedunderaccidentscenariossuchasfuelassemblydrop,accidentalmisplacementoffueloutsidearack,etc.ThecriticalitysafetyanalysisshowsthattheneutronmultiplicationfactorforthestoredfuelarrayisboundedbytheUSNRClimitof0.95(OTPositionPaper)underassumptionsof95%probabilityand95%confidence.Consequencesoftheinadvertentplacementofafuelassemblyarealsoevaluatedaspartofthecriticalityanalysis.ThecriticalityanalysisalsosetstherequirementsonthelengthoftheB-10screenandthearealB-10density.ThisLicensingReportcontainsdocumentationoftheanalysesperformedtodemonstratethelargemarginsofsafetywithrespecttoallUSNRCspecifiedcriteria.Thisreportalsocontainstheresultsoftheanalysisperformedtodemonstratetheintegrityofthefuelpoolreinforcedconcretestructure,andanappraisalof1-4 I

radiologicalconsiderations.Acost/benefitanaysisdemonstratingrerackingasthemostcosteffectiveapproachtoincreasetheon-sitestoragecapacityoftheDonaldC.CookNuclearPlanthasalsobeenperformedandsynopsizedinthisreport.Allcomputerprogramsutilizedinperformingtheanalysesdocumentedinthislicensingreportareidentifiedintheappropriatesections.AllcomputercodesarebenchmarkedandverifiedinaccordancewithHoltecInternational'snuclearQualityProgrameTheanalysespresentedhereinclearlydemonstratethattherackmodulearrayspossesswidemarginsofsafetyfromallthreethermal-,hydraulic,criticality,andstructural-vantagepoints.TheNoSignificantHazardConsiderationevaluationsubmittedtotheCommissionalongwiththisLicensingReportisbasedonthedescriptionsandanalysessynopsizedinthesubsequentsectionsofthisreport.1-5 IIIIIII TableDISCHARGESCHEDULE~CelelA+2A3A1B**4A2B5A6A3B7A4B8ASB9A6B10A7B11A8B12A9B13A10B14A11B15A12B16A13BMonth/Year12/19764/19784/197910/19795/19805/19815/19817/198211/19827/19833/19844/19852/19866/19875/19883/19896/199010/199011/19912/19923/19936/19937/199410/199411/19954/19963/19978/19977/1998NumberofAssemblies6564648065926464728092808880808077807680808080808080808080CumulativeInventoryInthePool651291932733384304945586307108028829701050113012101282136214381518159816781758183819181998207821582238*'A-**ReactorUnit1Reactor.Unit21>>6 Table1.1.1(continued)DISCHARGESCHEDULE~Cele17A*14B**18A15B19A16B20A17B21A18B22A19B23A20B24A21BMonth/~e12/19981/20014/20005/2001.8/20019/200212/20021/20046/20045/200510/20059/20062/20071/20087/20087/2009I80808080808080808080808080808080NumberofAssembliesD'schareCumulativeInventorynthePoo2318239824782558263827182798287829583038311831983278335834383518A-ReactorUnit1**B-ReactorUnit21-7 IIIIIII Table1.1.2AVAILABLESTORAGEINTHEDONALDCCOOKPOOLNUMBEROFSTORAGELOCATIONSAVAILABLE~Cc1eMonth/YearWithPresentLicensedCapacity2050LocationsAfterReracking3616Locations7B11A8B12A9B13A10B14A11B15A12B16A13B17A14B18A15B19A16B20A17B21A18B19B23A20B24A21B25A768688612532452.372292*6/199010/199011/19912/19923/19936/19937/19948/19977/199812/19981/20004/20005/20018/20019/200212/20021/20046/20045/20059/20062/20071/20087/20087/200911/200910/199421211/1995132**4/199652***3/199723342254217820982018193818581778169816181538145813781298121811381058978898818738658578418338*258178**9818***fromboth*Dateoflossoffullcoreoffloadcapabilityreactors.**Dateoflossoffullcoreoffloadcapabilityforonereactor.***Dateoflossofnormaldischargecapability1>>8 IIIIIII Table1.1.3RACKMODULEDATA'XISTINGANDPROPOSEDRACKSITEMNumberofcellsNumberofmodulesNeutronAbsorber(Nom.)cellpitch,inch(Nom.)cellopeningsize,inchEXISTINGRACKS205020Boral10.5"8.884+0'24PROPOSEDRACKS3616*23Boral897n8.75"+0.04Includethreetriangularcornerstoragecells.1-9 IIIIIIII 2.0MODULEDATA2.1SnosisofNewModulesTheDonaldC.Cookspentfuelpoolconsistsofa39'-19/16"x58'-31/8"rectangularpitwitha10'-4"x10'-6"spacedesignatedforcaskhandlingoperations.Thepoolisconnected'tothefueltransfercanalthroughaweirgateontheWestwall.Thisgateisnormallyclosed.Atthepresenttime,theDonaldC.Cookpoolcontainsmediumdensityrackswitha10.5"nominalassemblycenter-to-centerpitch.Thereisatotalof2050storagecellsinthepool.Therearetwosizesofmodules,10xl0and10xll.The10x10moduleweighs33,800lb.andthe10xllmoduleweighs37,200lb.Figure2.1.1showsaftertheproposedandtabulatedincontainingatotalpitch.themodulelayoutfortheDonaldC.Cookpoolrerackingcampaign.AsshowninFigure2.1.1Table2.1.1,therearetwenty-threeracksof3613storagecellswitha8.97"nominalTheessentialcelldataforallstoragecellsisgiveninTable2.1.2.ThephysicalsizeandweightdataonthemodulesmaybefoundinTable2.l.3.Insummary,thepresentrerackingapplicationwillincreasethelicensedstoragecapacityoftheDonaldC.Cookpoolfrom2050to3613cells.2~2MixedZoneThreeReionStoraeMZTRThehighdensityspentfuelstorageracksintheDonaldC.Cookpoolwillprovidestoragelocationsforupto3613fuelassembliesandwillbedesignedtomaintainthestoredfuel,havinganinitialenrichmentofup=to5wt%U-235,inasafe,eoolable,andsubcriticalconfigurationduringnormaldischargeandfullcoreoffloadstoragesandpostulatedaccidentconditions.2-1 IIIII AllrackmodulesforDonaldC.Cookspentfuelpoolareoftheso-called"free-standing"typesuchthatthemodulesarenotattachedtothepoolfloornordotheyrequireanylateralbracesorrestraints.Theserackmoduleswillbeplacedinthepoolintheirdesignatedlocationsusingaspecificallydesignedliftingdevice,andthesupportlegsremotelyleveled(usingatelescopicremovablehandlingtool)byanoperatoronthefuelhandlingbridge.Thelevelingoperationsaredonewhenthesupportlegsareliftedoffthefloor.Exceptforthecrane,noadditionalliftingequipmentisneededwhilelevelingisbeingperformed.AsdescribedindetailinSection3,allmodulesintheDonaldC.Cookpoolareof"non-fluxtrap"construction.However,themodulebaseplatesextendoutby7/8"(nominal),suchthatthenom'inalgapbetweentheadjacentwallsoftwoneighboringracksis2"(nom.).Thus,althoughthereisasinglescreenofneutronabsorberpanelbetweentwofuelassembliesstoredinthesamerack,therearetwopoisonpanelswithawaterfluxtrap(2"wide)betweenthemforfuelassemblieslocatedincellsintwofacingmodules.Outofthesefluxtraplocations,andperipheralcelllocations(cellsadjacenttopoolwalls)acertainnumberofstoragecellsaredesignatedforstoringfreshfuel.Inaddition,asdescribedinSection4,acertainnumberofinteriorcellsineachrackaredesignatedforstoringfreshfuelof5%wt.U-235(max.)enrichment.Inthismanner,asufficientnumberoflocationswithoutanyburnuprestriction(RegionIcells)areidentifiedtoenableunrestrictedfullcoreoffloadoftheDonaldC.Cookreactorinthespentfuelpool.Theseso-calledRegionIcellsareidentifiedinSection4ofthisreport.Theremainingstoragecellshaveenrichment/burnuprestrictions.Appropriaterestrictionsontheenrichment/burnupofthestoredfuelinRegionIIandRegionIXIcellsarepresentedinSection4.2~2 IOlII Eachrackmoduleissupportedbyatleastfourlegswhichareremotelyadjustable.Thus,therackscanbemadeverticalandthetopoftherackscaneasilybemadeco-planarwitheachother.Therackmodulesupportlegsareengineeredtoaccommodatevariationsofthepoolfloor.Thesupportlegsalsoprovideanunderrackplenum'fornaturalcirculationofwaterthroughthestoragecells.Theplacementoftheracksinthespentfuelpoolhasbeendesignedtoprecludeanysupportlegsfrombeinglocatedoverexistingobstructionsonthepoolfloor.TheDonaldC.CookracksaresubjectedtomandatedseismicloadingspertheplantUFSAR.TheDesignBasisEarthquake(DBE)andOperatingBasisEarthquake(OBE)seismicresponsespectraareprovidedandsynthetictimehistoriesaregenerated.Theseaccelerationtimehistoriesareappliedasinertialoads(seeSection6.3).Undertheseseismicevents,therackmoduleshavefourdesignatedlocationsofpotentialimpact:(i)(ii)(iii)(iv)SupportlegtobearingpadStoragecelltofuelassemblycontactsurfacesBaseplateedgesRacktopcornersThesupportlegtopoolslabbearingpadimpactwouldoccurwhenevertheracksupportfootliftsoffthepoolfloorduringaseismicevent.The"rattling"ofthefuelassembliesinthestoragecellisanaturalphenomenonassociatedwithseismicconditions.ThebaseplateandracktopcornersimpactswouldoccuriftherackmodulestendtoslideortilttowardseachotherduringthepostulatedDBEorOBEseismicevents.Section6ofthisreportpresentstheanalysismethodologyandresultsforallthreelocationsofimpact,andestablishesthestructuralintegrityoftheracksundertheloadcombinationsspecifiedforplantconditionsrequiredbytheNRC.2~3 rIl~i~III Abearingpad,madeofausteniticstainlesssteel,isinterposedbetweenthesupportfootandthelinersuchthattheloadstransmittedtotheslabbytherackmoduleundersteadystateaswellasseismicconditionsarediffusedintothepoolslab,andallowablelocalconcretesurfacepressuresarenotexceeded.Section8ofthisreportpresentsthedetailedpoolstructureanalysis.2'MaterialConsiderations2.3.1IntroductionSafestorageofnuclearfuelintheDonaldC.Cookspentfuelpoolrequiresthatthematerialsutilizedinthefabricationofracksbeofprovendurabilityandbecompatiblewiththepoolwaterenvironment.Thissectionprovidesthenecessaryinformationonthissubject.2.3.2StructuralMaterialsThefollowingstructuralmaterialsareutilizedinthefabricationofthespentfuelracks:a.ASMESA240-304forallsheetmetalstock.b.c~d~Internallythreadedsupportlegs:ASMESA240-304.Externallythreadedsupportspindle:ASMESA564-630precipitationhardenedstainlesssteel.Weldmaterial-perthefollowingASMEspecification:SPA5.9ER308.2.3.3oisonMateriaInadditiontothestructuralandnon-structuralstainlessmaterial,theracksemployBoral,apatentedproductofAARBrooksSPerkins,asthethermalneutronabsorbermaterial.AbriefdescriptionofBoral,anditsfuelpoolexperiencelistfollows.Boralisathermalneutronabsorbingmaterialcomposedofboroncarbideand1100alloyaluminum.Boroncarbideisacompound2-4 II havingahighboroncontentinaphysicallystableandchemicalinertform.The1100alloyaluminumisalight-weightmetalwithhightensilestrengthwhichisprotectedfromcorrosionbyahighlyzesistantoxidefilm.Thetwomaterials,boroncarbideandaluminum,arechemicallycompatibleandideallysuitedforlong-,termuseintheradiation,thermalandchemicalenvironmentofaspentfuelpool.Boral'suseinthespentfuelpoolastheneutronabsorbingmaterialcanbeattributedtothefollowingreasons:Thecontentandplacementofboroncarbideprovidesaveryhighremovalcrosssectionforthermalneutrons.(ii)Boroncarbide,intheformoffineparticles,ishomogenouslydispersedthroughoutthecentrallayeroftheBoral.(iii)(iv)TheboroncarbideandaluminummaterialsinBoraldonotdegradeasaresultoflong-termexposuretogammaradiation.ThethermalneutronabsorbingcentrallayerofBoraliscladwithpermanentlybondedsurfacesofaluminum.(v)Boralisstable,strong,durable,andcorrosionresistant.ThepassivationprocessofBoralinanaqueousenvironmentresultsinthegenerationofhydrogengas.Ifthegenerationrateofhydrogenistoorapid,thenswellingofBoralmayoccur.LaboratorystudiesbyBoral'ssupplierindicatethattherateofhydrogengenerationisastrongfunctionoftheso-calledimpuritiesin,thechemicalcompositionoftheboroncarbidepowder,namelysodiumhydroxideandboronoxide.AARBrooksPerkinshasinstitutedastrictprogramofmonitoringofthechemistryofboroncarbideusedinthe,manufacturingofBoraltoensurethatnoswellingofthepanelswilloccur.Furthermore,2-5 II randomlyselectedcouponsofBoralpanelsfromproductionrunsaresubjectedtoswellingtestcheckstoprecludeanypossibilityofswellingofBoral.BoralismanufacturedbyAARBrooks&Perkinsunderthecontrolandsurveillanceofacomputer-aidedQualityAssurance/QualityControlProgramthatconformstotherequirementsof10CFR50AppendixB,"QualityAssuranceCriteriaforNuclearPowerPlantsandFuelReprocessingPlants".AsindicatedinTable2.3.1,BoralhasbeenlicensedbytheUSNRCforuseinnumerousBWRandPWRspentfuelstorageracksandhasbeenextensivelyusedinoverseasnuclearinstallations.BoralMaterialCharacteristicsAluminum:Aluminumisasilvery-white,ductilemetallicelementthatisabundantintheearth'scrust.The1100alloyaluminumisusedextensivelyinheatexchangers,pressureandstoragetanks,chemicalequipment,reflectorsandsheet,metal'work.Ithashighresistancetocorrosioninindustrialandmarineatmospheres.Aluminumhasatomicnumberof13,atomicweightof26.98,specificgravityof'.69andvalenceof3.Thephysical/mechanicalpropertiesandchemicalcompositionofthe1100alloyaluminumarelistedinTables2.3.2and2.3.3.Theexcellentcorrosionresistanceofthe1100alloyaluminumisprovidedbytheprotectiveoxidefilmthatdevelopsonitssurfacefromexposuretotheatmosphereorwater.ThisfilmpreventsthelossofmetalfromgeneralcorrosionorpittingcorrosionandthefilmremainsstablebetweenapHrangeof4.5to8.5.2-6 II BoronCarbide:TheboroncarbidecontainedinBoralisafinegranulatedpowderthatconformstoASTMC-750-80nucleargradeTypeIII'heparticlesrangeinsizebetween60and200meshandthematerialconformstothechemicalcompositionandpropertieslistedinTable2.3.4.2.3.4ComatibilitwithCoolantAllmaterialsusedintheconstructionoftheDonaldC.Cookrackshaveanestablishedhistoryofin-poolusage.Theirphysical,chemicalandradiologicalcompatibilitywiththepoolenvironmentisanestablishedfactatthistime.AsnotedinTable2.3.1,Boralhasbeenusedinbothventedandunventedconfigurationsinfuelpoolswithequalsuccess.Consistentwiththerecentpractice,theDonaldC-CookrackconstructionallowsfullventingoftheBoralspace.Austeniticstainlesssteel(304)iswidelyusedinnuclearpowerplants.2.4ExistinRackModulesandProosedRerackin0erationTheDonaldC.Cookfuelpoolcurrentlyhasmediumdensityrackmodules'ontainingatotalof2050storagecellsintwentymodules.Atthetimeoftheproposedrerackingoperation,approximately1678cells(between6/1993and7/1994)outof2050locationswillbeoccupiedwithspentfuel.Thereissufficientnumberofopen(unoccupied)cellsinthepooltopermitrelocationofallfuelsuchthattheexistingmodulescanbeemptiedandremovedfromthepool,andnewmodulesinstalledinaprogrammedmanner.Aremotelyengagableliftrig,whichisdesignedtomeetthecriteriaofNUREG-0612"ControlofHeavyLoadsofNuclearPowerPlants",willbeusedtolifttheemptymodules.AuxiliaryBuildingCraneswillbeusedforthispurpose.Amodulechange-out2-7 IIIII schemeandprocedurewillbedevelopedwhichensuresthatallmodulesbeinghandledareemptywhenthemoduleismovingataheightwhichismorethan12"abovethepoolfloor.TheAuxiliaryBuildinghastwooverheadcraneswhichrideonrailsthattraversetheentirefuelhandlingareaofthebuilding.Eachcranehasamainhookratedat150tons.Thesehooksaresinglefailureproof(SFP)(upto60tons).InadditionthereisanauxiliaryhoistontheEastCraneratedat20tons.Pursuanttothedefense-in-depthapproachofNUREG-0612,thefollowingadditionalmeasuresofsafetywillbeundertakenforthererackingoperation.(ii)(iii)Thecraneandhoistwillbegivenapreventivemaintenancecheckupandinspectionwithin3monthsofthebeginningofthererackingoperation.Thecranehookwillbeusedtoliftnomorethan50%ofitssinglefailureproofcapacityof60tonsatanytimeduringthererackingoperation.(Themaximumweightofanymoduleanditsassociatedhandlingtoolis24tons).Theoldfuelrackswillbeliftednomorethan6"abovethepoolfloorandheldinthatelevationforapproximately10minutesbeforebeginningtheverticallift.(iv)Therateofverticalliftwillnotexceed6'erminute.(v)(vi)(vii)(viii)Therateofhorizontalmovementwillnotexceed6'erminute.Preliminarysafeloadpathshavebeendeveloped.The"old"or"new"rackswillnotbecarriedoveranyregionofthepoolcontainingfuel.Therackupendingorlayingdownwillbecarriedoutinanareawhichisnotoverlappingtoanysafetyrelatedcomponent.Allcrewmembersinvolvedinthererackingoperationwillbegiventrainingintheuseoftheliftingandupendingequipment.Thetraining2-8 I

(ix)seminarwillutilizevideotaoesoftheactualliftingandupendingrigsontyoicalmodulestobeinstalledinthepoolsEverycrewmemberwillberequiredtopassawrittenexaminationintheuseofliftingandupendingapparatusadministeredbytherackdesigner.ReferringtoFigure2.1.1,itisnotedthatthefuelhandlingbridgecranecannotaccessstoragecellsfacingtheeastwallandseverallocationsinthesouthwestcorner.Therefore,itwillbenecessarytoloadtheinaccessiblecellswithfuelwhentherackisstaged'certaindistance(approximately20inches)fromthepoolwall.Havingloadedthesecells,themodulewillbeliftedapproximately4inchesabovethepoolliner,andlaterallytransportedtoitsfinaldesignatedlocations.AfuelshufflingandrackinstallationsequencehasbeendevelopedtoensurethatallheavyloadhandlingcriteriaofNUREG-0612aresatisfied.Therackhandlingrigisdesignedwithconsiderationoftherackmoduleweightalongwiththecontainedfuelassemblymass.Thefuelracks'willbebroughtdirectlyintotheAuxiliaryBuildingthroughtheaccessdoorwhichisatgroundlevel(609'levation).Thisdirectaccesstothebuildinggreatlyfacilitatestherackremovalandinstallationeffort.The"old"rackswillbedecontaminatedtotheextentpracticalon-siteandapprovedforshippingpertherequirementsof10CFR71and49CFR171-178,behousedinshippingcontainers,andtransportedtoaprocessingfacilityforvolumereduction.Non-decontaminatableportionsoftherackswillbeshippedtoalicensedradioactivewasteburialsiteorreturnedtositeforstorageifdisposalaccessisunavailable.Thevolumereductionisexpectedtoreducetheoverallvolumeoftherackstoabout1/10thoftheiroriginalvalue.AllphasesofthererackingactivitywillbeconductedinaccordancewithwrittenprocedureswhichwillbereviewedandapprovedbyX6M.2-9 I

ModuleI.D.A**BCDEFGH*Total~nantit5442421123Table21.1MODULEDATAArrayCellSize13x1412x1413x1212xl213xll12xll12x1013x14-(8x2)TotalCellCountforthisModuleTe9106726242885722641201663616Non-rectangularmodule.**ThreeoftheAmoduleshaveonetrianglecelltoaccommodatepoolcornercurvature.2-10 IIII Table2.1.2COMMONMODULEDATAStoragecellinsidedimension:8.75"+0.04"Storagecellheight(abovethebaseplate):168+1/16"Baseplatethickness:Supportlegheight:Supportlegtype:Numberofsupportlegs:Remoteliftingandhandlingprovision:Poisonmaterial:Poisonlength:Poisonwidth:Cell'Pitch:0.75"(nominal)5.25"(nominal)Remotelyadjustablelegs4(minimum)YesBoral144"7.5"8.97"(nominal)2-11 Table2.1.3MODULEDATADimensions(inch)*ModuleI.D.ABCDEFGHEast-West117-3/16108-1/8117-3/16108-1/8117-3/16108-1/8108-1/8117-3/16North-South126-3/16126-3/16108-1/8108-1/899-1/1699-1/1690-1/8126-3/16ShippingWeight~kis25.723.722.520.920.819.317'23.9*Alldimensionsareboundingrectangularenvelopesroundedtothenearestonesixteenthofaninch.2-12 III Table2.3.1BORALEXPERIENCELIST(DomesticandForeign)Pressurized.WaterReactorsPlantUtilityVentedConstruc-tionMfg.YearBellefont1,2DonaldC.Cook1,2IndianPoint3MaineYankeeSalem1,2SeabrookSequoyah1,2YankeeRoweZion1,2Byron1,2Braidwood1,2YankeeRoweThreeMileIslandITennesseeValleyAuthorityIndiana&MichiganElectricNYPowerAuthorityMaineYankeeAtomicPowerPublicServiceElec&GasNewHampshireYankeeTennesseeValleyAuthorityYankeeAtomicPowerCommonwealthEdisonCo.CommonwealthEdisonCo.CommonwealthEdisonCo.YankeeAtomicElectricGPUNuclearNoNoYesYesNoNoNo198119791987197719801979Yes1990Yes1964/1983Yes1980Yes1988Yes1988Yes1988BoilingWaterReactorsBrownsFerry1,2,3Brunswick1,2ClintonCooperDresden2,3DuaneArnoldJ.A.FitzpatrickE.I.Hatch1,2HopeCreekHumboldtBayLaCrosseLimerick1,2MonticelloPeachbottom2,3Perry,1,2PilgrimShorehamSusquehanna1,2VermontYankeeHopeCreekTennesseeValleyAuthorityCarolinaPower&LightIllinoisPowerNebraskaPublicPowerCommonwealthEdisonCo.IowaElec.Light&PowerNYPowerAuthorityGeorgiaPowerPublicServiceElec&GasPacificGas&ElectricDairylandPowerPhiladelphiaElectricNorthernStatesPowerPhiladelphiaElectricClevelandElec.IlluminatingBostonEdisonLongIslandLightingPennsylvaniaPower&LightVermontYankeeAtomicPowerPublicServiceElec&GasYesYesYesYesYesNoNoYesYesYesYesNoYesNoNoNoYesNoYesYes198019811981197919811979197819811985198619761980197819801979197819791978/198619892-13 IIIIIII Table2.3.1(continued)ForeignInstallationsUsingBoralFrance12PHRPlantsSouthAfricaKoeberg1,2SwitzerlandElectricitedeFranceESCOMBeznau1,2GosgenNordostschweizerischeKraftwerkeAGKernkraftwerkGosgen-DanikenAGTaiWBIlChin-Shan1,2Kuosheng1,2TaiwanPowerCompanyTaiwanPowerCompany2-14 IIIIIIIII Table2.3.21100ALLOYALUMINUMPHYSICALANDMECHANICALPROPERTIESDensityMeltingRangeThermalConductivity(77deg.F)Coef.ofThermalExpansion(68-212deg.F)Specificheat(221deg.F)ModulusofElasticityTensileStrength(75deg.F)YieldStrength(75deg.F)Elongation(75deg.F)Hardness(Brinell)AnnealingTemperature0.098lb/cu.in.2.713gm/cc1190-1215deg.F643-657deg.C128BTU/hr/sqft/deg.F/ft0.53cal/sec/sqcm/deg.C/cm13.1x10/deg.F23.6x10/deg.C0.22BTU/lb/deg.F0.23cal/gm/deg'.C10xl06psi13,000psiannealed18,000psiasrolled5,000psiannealed17,000psiasrolled35-45%annealed9-20%asrolled23annealed32asrolled650deg.F343deg.C2-15 IIIIIII Table2.3.3CHEMICALCOMPOSITION(byweight.)-ALUMINUM(1100Alloy)99.00%min.1.00%max.0.05-0.20%max..05%max..10%max..15%max.AluminumSiliconeandIronCopperManganeseZincotherseach2-16 IIIIIII Table2.3.4BORONCARBIDECHEMICALCOMPOSITIONWeihtTotalboronBisotopiccontentinnaturalboronBoricoxideIronTotalboronplustotalcarbon70.0min.18.03.0max.2.0max.94.0min.BORONCARBIDEPHYSICALPROPERTIESChemicalformulaBoroncontent(weight)Carboncontent(weight)CrystalStructureDensityMeltingPointBoilingPointMicroscopicthermal-neutroncross-sectionB4C78.28%21.72%rombohedral2.51gm./cc-0.0907lb/cu.in.2450C(4442F)3500C(6332F)600barn2-17 IIIIt:IIII l.eI~1Wa'eos+a~:~I'I~~~~~0~~~'

III 3.0CONSTRUCTIONOFRACKMODULESTheobjectof'thissectionistoprovideadescriptionofrackmoduleconstructionfortheDonaldC.Cookspentfuelpooltoenableanindependentappraisaloftheadequacyofthedesign.SimilarrackstructuredesignshaverecentlybeenusedinpreviouslicensingeffortsforKuoshengUnits1&2(TaiwanPowerCompany);J.A.FitzPatrick(NewYorkPowerAuthority);IndianPoint2(ConsolidatedEdisonCompanyofNewYork,Inc.);ThreeMileIslandUnit1(GPUNuclear);andHopeCreek1(PublicServiceElectricGasCompany).Alistofapplicablecodesandstandardsisalsopresented.3.1FabricationOb'ectiveTherequirementsinmanufacturingthehighdensitystorageracksfortheDonaldC.Cookfuelpoolmaybestatedinfourinterrelatedpoints:-(1)Therackmodulewillbefabricatedinsuchamannerthatthereisnoweldsplatteronthestoragecellsurfaceswhichwouldcomeincontactwiththefuelassembly.(2)Thestoragelocationswillbeconstructedsothatredundantflowpathsforthecoolantareavailable.(3)(4)Thefabricationprocessinvolvesoperationalsequenceswhichpermitimmediateverificationbytheinspectionstaff.Thestoragecellsareconnectedtoeachotherbyausteniticstainlesssteelcornerweldswhichleadstoahoneycomblatticeconstruction.Theextentofweldingisselectedto"detune"theracksfromtheseismicinputmotionoftheOperatingBasisEarthquake(OBE)andDesignBasisEarthquake(DBE).3-1 II 3.2MixedZoneTwoReionStoraeAllrackmodulesdesignedandfabricatedfortheDonaldC.Cookspentfuelpoolareoftheso-called"non-fluxtrap"type.Znthenon-fluxtrapmodules,asinglescreenofthepoisonpanelisinterposedbetweentwofuelassemblies.ThepoisonmaterialutilizedinthisprojectisBoral,whichdoesnotrequirelateralsupporttopreventslumpingduetotheinherentstiffness.However,accuratedimensionalcontrolofthepoisonlocationisessentialfornuclearcriticalityandthermal-hydraulic.considerations.Thedesignandfabricationapproachto,realizethisobjectiveispresentedinthenextsub-section.3.3AnatomofRackModulesAsstatedearlier,thestoragecelll'ocationshaveasinglepoisonpanelbetweenadjacentausteniticstainlesssteelsurfaces.Thesignificantcomponentsoftherackmoduleare:(1)thestorageboxsubassembly(2)thebaseplate,(3)thethermalneutronabsorbermaterial,(4)pictureframesheathing,and(5)supportlegs.Therackmodulemanufacturingbeginswithfabricationofthebox.The"boxes"arefabricatedfromtwoprecisionformedchannelsbyseamweldinginamachineequippedwith.copperchillbarsandpneumaticclampstominimizedistortionduetoweldingheatinput.Figure3.3.1showsthebox.Theminimumweldpenetrationwillbe80%ofthe-boxmetalgagewhichis0.075"(14gage).Theboxesaremanufacturedto8.75"X.D.(nominalinsidedimension).ThedesignobjectivecallsforinstallingBoralwithminimalsurfaceloading.Thisisaccomplishedbydieforminga"pictureframesheathing"asshowninFigure3-2 I4II 3.3.2.Thissheathingis0.035"thickandismadetoprecisedimensionssuchthattheoffsetis.010"to.005"greaterthanthepoisonmaterialthickness.AsshowninFigure3.3.1,eachboxhasfourlateral1"diameterholespunchednearitsbottomedgetoprovideauxiliaryflowholes.Theedgesofthesheathingandtheboxareweldedtogethertoformasmoothedge.Thebox,withintegrallyconnectedsheathing,isreferredtoasthe"compositebox".The"compositeboxes"arearrangedinacheckerboardarraytoformanassemblageofstoragecelllocations(Figure3.3.3).The'nter-boxweldingandpitchadjustmentareaccomplishedbysmalllongitudinalconnectors.Furtherdetailsaregivenlaterinthissection.Thisassemblageofboxassembliesisweldededge-to-edgeasshowninFigure3.3.3,resultinginahoneycombstructurewithaxial,flexuralandtorsionalrigiditydependingontheextentofintercellweldingprovided.ZtcanbeseenfromFigure3.3.3thattheedgesofeachinteriorboxareconnectedtothecontiguousboxesresultinginawelldefinedpathtoresistshear.hbpl'horizontalsurfaceforsupportingthefuelassemblies.Thebaseplateisattachedtothecellassemblagebyfilletwelds.Thebaseplateineachstoragecellhasa5"diameterflowhole.Thebaseplateis3/4"thicktowithstandaccidentfuelassemblydroploadspostulatedanddiscussedinSection7ofthisreport.(3)Thethermalneutronabsorbermaterial:Asmentionedintheprecedingsection,Boralisusedasthethermalneutronabsorbermaterial.(4)PictureFrameSheathin:Asdescribedearlier,thesheathingservesasthelocatorandretainerofthepoisonmaterial.Figure3.3.2showsaschematicofthesheathing.3~3 IlII Thepoisonmaterialisplacedinthecustomizedflatdepressionregionofthesheathing,whichisnextlaidonasideofthe"box".Theprecisionoftheshapeofthesheathingobtainedbydieformingguaranteesthat.thepoisonsheetinstalledinitwillnotbesubjecttosurfacecompression.Theflangesofthesheathing(onallfoursides)areattachedtotheboxusingskipwelds.Thesheathingservestolocateandpositionthepoisonsheetaccurately,andtoprecludeitsmovementunderseismicconditions.SuortLes:Allsupportlegsaretheadjustabletype(Figure3.3.4).Thetopportionismadeofausteniticsteelmaterial.ThebottompartismadeofSA564-630stainlesssteeltoavoidgallingproblems.Eachsupportlegisequippedwithareadilyaccessiblesockettoenableremotelevelingoftherackafteritsplacementinthepool.Lateralholesinthesupportlegprovidetherequisitecoolantflowpath.Anelevationcross-sectionoftherackmoduleshowninFigure3.3.5showstwoboxcells,andadevelopedcellinelevation.TheBoralpanelsandtheirlocationarealsoindicatedinthisfigure.Theboralpanelsarepositionedsuchthattheentireenrichedfu'elportionofthefuelassemblyisenvelopedbythethermalneutronabsorbermaterial.Thejointbetweenthecompositeboxarraysandthebaseplateismadebysinglefilletweldswhichprovideaminimumof7"ofconnectivitybetweeneachcellwallandthebaseplatesurface.AsshowninFigure3.3.4,thesupportlegisgussetedtoprovideanincreasedsectionforloadtransferbetweenthesupportlegsandthecellularstructureabovethebaseplate.Useofthegussetsalsominimizesheatinputinduceddistortionsofthesupport/baseplatecontactregions3-4 III 3.4CodesStandardsandPracticesfortheDonaldC.CookSentFuelPoolRacksThefabricationoftherackmodulesfortheDonaldC.Cookspentfuelpoolisperformedunderastrictqualityassurancesystemsuitableformanufacturingandcomplyingwiththeprovisionsof10CFR50AppendixB.Thefollowingcodes,standardsandpracticeswillbeusedasapplicableforthedesign,construction,andassemblyofthespentfuelstorageracks.Additionalspecificreferencesrelatedtodetailedanalysesaregivenineachsection.a~CodesandStandardsforDesinandTestin(1)AZSCManualofSteelConstruction,8thEdition,1980.(2)ANSIN210-1976,"DesignObjectivesforLightWaterReactorSpentFuelStorageFacilitiesatNuclearPowerStations".(3)AmericanSocietyofMechanicalEngineers(ASME),BoilerandPressureVesselCode,SectionIII/SubsectionNF,1989.(4)ASNT-TC-1A,June,1980AmericanSocietyforNondestructiveTesting(RecommendedPracticeforPersonnelQualifications).(5)ASMESectionV-NondestructiveExamination(6)ASMESectionZX-WeldingandBrazingQualifications(7)BuildingCodeRequirementsforReinforcedConcrete,ACI318-89/ACI318R-89.3-5 IIOtII (8)CodeRequirementsforNuclearSafetyRelatedConcreteStructures,ACI349-85andCommentaryACI349R-85(9)ReinforcedConcreteDesignforThermalEffectsonNuclearPowerPlantStructures,ACI349.1R-80(10)ACIDetailingManual-1980(11)ASMENQA-2,Part2.7"QualityAssuranceRequirementsofComputerSoftwareforNuclearFacilityApplications(draft).(12)ANSI/ASME,QualificationandDutiesofPersonnelEngagedinASMEBoilerandPressureVesselCodeSectionIII,Div.1,CertifyingActivities,N626-3-1977.Mate'aCodes(1)AmericanSocietyforTestingandMaterials(ASTM)Standards-A-240.(2)AmericanSocietyofMechanicalEngineers(ASME),BoilerandPressureVesselCode,SectionII-PartsAandC,1989.ASMEBoilerandPressureVesselCode,SectionIX-WeldingandBrazingQualifications(1986)orlatexissueacceptedbyUSNRC.ualitAssuranceCealinessPacka'nShiReceivinStoraeandHandlinReuirements(1)ANSIN45.2.2-Packaging,Shipping,Receiving,StorageandHandlingofItemsforNuclearPowerPlants.(2)ANSI45.2.1-CleaningofFluidSystemsandAssociatedComponentsduringConstructionPhaseofNuclearPowerPlants."3-6 IIIII (3)ASMEBoilerandPressureVessel,SectionV,NondestructiveExamination,1983Edition,includingSummerandWinterAddenda,1983.(4)ANSI-N16.1-75NuclearCriticalitySafetyOperationswithFissionableMaterialsOutsideReactors.(5)ANSI-N16.9-75ValidationofCalculationMethodsforNuclearCriticalitySafety.(6)ANSI-N45.2.11,1974QualityAssuranceRequirementsfortheDesignofNuclearPowerPlants.(7)ANSI14.6-1978,"SpecialLiftingDevicesforShippingContainersweighing10',000lbs.ormoreforNuclearMaterials".(8)ANSIN45.2',QualificationofInspectionandTestingPersonnel.(9)ANSIN45.2.8,Installation,Inspection.(10)ANSIN45.2.9,Records.(ll)ANSIN45.2..10,Definitions.(12)ANSIN45~212,QAAudits.(13)ANSIN45.2.13,Procurement.(14)ANSI45.2.23,QAAuditPersonnel.OtherReferences(Inthereferencesbelow,RGisNRCRegulatoryGuide)(1)RG1.13-SpentFuelStorageFacilityDesignBasis,Rev.2(proposed).(2)RG1.123-(endorsesANSIN45.2.13)QualityAssuranceRequirementsforControlofProcurementofItemsandServicesforNuclearPowerPlants.(3)RG1.124-ServiceLimitsandLoadingCombinationsforClass1LinearTypeComponentSupports,Rev.1.3-7 III (4)RG1.25-AssumptionsUsedforEvaluatingthePotentialRadiologicalConsequencesofaFuelHandlingAccidentintheFuelHandlingand"StorageFacilityofBoilingandPressurizedWaterReactors.(5)RG1.28-(endorsesANSIN45.2)-QualityAssuranceProgramRequirements,June'972.(6)RG1.29-SeismicDesignClassification,Rev.3.(7)RG1.31-ControlofFerriteContentinStainlessSteelWeldMetal,'ev.3.(8)RG1.38-(endorsesANSIN45.2.2)QualityAssuranceRequirementsforPackaging,Shipping,Receiving,StorageandHandlingofItemsforWater-CooledNuclearPowerPlants,March,1973.(9)RG1.44-ControloftheUseofSensitizedStainlessSteel.(10)RG.1.58-(endorsesANSIN45.2.2)QualificationofNuclearPowerPlantInspection,Examination,andTestingPersonnel,Rev.1,September,1980.(ll)RG1.64-(endorsesANSIN45..2.11)QualityAssuranceRequirementsfortheDesignofNuclearPowerPlants,October,1973.(12)RG1.71-WelderQualificationsforAreasofLimitedAccessibility.(13)RG1'4-(endorsesANSIN45.2.10)QualityAssuranceTermsandDefinitions,February,1974.(14)RG1.85-MaterialsCodeCaseAcceptabilityASMESectionIII,Division1.(15)RG1.88-(endorsesANSIN45.2.9)Collection,StorageandMaintenanceofNuclearPowerPlantQualityAssuranceRecords,Rev.2,October,1976.(16)RG1.'92-CombiningModalResponsesandSpatialComponentsinSeismicResponseAnalysis.3-8 IIIIII (17).RG3.41-ValidationofCalculationMethodsforNuclearCriticalitySafety.(18)GeneralDesignCriteriaforNuclearPowerPlants,CodeofFederalRegulations,Title10,Part50,AppendixA(GDCNos.1,2,61,62,and63).(19)NUREG-0800,StandardReviewPlan,Sections3.2.1,313.F1(3/3(38'(20)"OTPositionforReviewandAcceptanceofSpentFuelStorageandHandlingApplications,"datedApril14,1978,andthemodificationstothisdocumentofJanuary18,1979.(Note:OTstandsforOfficeofTechnology).(21)NUREG-0612,"ControlofHeavyLoadsatNuclearPowerPlants".(22)RegulatoryGuide8.8,"ZnformationRelativetoEnsuringthatOccupationalRadiationExposureatNuclearPowerPlantswillbeasLowasReasonablyAchievable(ALtGQ.).(23)10CFR50AppendixB,QualityAssuranceCriteriaforNuclearPowerPlantsandFuelReprocessingPlants(24)10CFR21-ReportingofDefectsandNon-Compliance3'MaterialsofConstruct'on.StorageCell:Baseplate:SupportLeg(female):SupportLeg(male):Poison:ASMESA240-304ASMESA240-304ASMESA240-304Ferriticstainlesssteel(anti-gallingmaterial)ASMESA564-630Boral3-9 IIII Lateralflowholes'.75"WeldSeam.075"Figure3.3.1SEAMWELDINGPRECISIONFORMEDCHANNELS3-10 IlI SheathingFigure3.3.2COMPOSITEBOXASSEMBLY Ie~II Figure3.3.3ARRAYOFCELLSFORNON-FLUXTRAPMODULES3-12 IIII BaseplateGussetFigure3.3.4ADJUSTABLESUPPORTIEG3-13 III BoxCellsCellPitchDeveloped)cellBoralpanelSheathingBaseplateOneInchLateralFlowHole(Typical)Figure3.3.5ELEVATIONVIENOFRACKNODULE3-14 IIIlI 4.0CRITICALITYSAFETYANALYSES4.1DesiBasisThehighdensityspentfuelstorageracksforDonaldC.cookNuclearPlantaredesignedtoassurethattheeffectiveneutronmultiplicationfactor(k~ff)isequaltoorlessthan0.95withtheracksfullyloadedwithfuelofthehighestanticipatedreactivity,andfloodedwithunboratedwateratthetemperaturewithintheoperatingrangecorrespondingtothehighestreactivity.Themaximumcalculatedreactivityincludesamarginforuncertaintyinreactivitycalculationsincludingmechanicaltolerances.Alluncertaintiesarestatisticallycombined,suchthatthefinalk,<<willbeequaltoorlessthan0.95witha954probabilityata954confidencelevel.Applicablecodes,standards,andregulationsorpertinentsectionsthereof,includethefollowing:oGeneralDesignCriteria62,PreventionofCriticalityinFuelStorageandHandling.oUSNRCStandardReviewPlan,NUREG-0800,Section9.1.2,SpentFuelStorage,Rev.3-July1981oUSNRCletterofApril14,1978,toallPowerReactorLicensees-OTPositionforReviewandAcceptanceofSpentFuelStorageandHandlingApplications,includingmodificationletterdatedJanuary18,1979.USNRCRegulatoryGuide1.13,SpentFuelStorageFacilityDesignBasis,Rev.2(proposed),December1981.ANSIANS-8.17-1984,CriticalitySafetyCriteriafortheHandling,StorageandTransportationofLWRFuelOutsideReactors.4-1 III Toassurethetruereactivitywillalwaysbelessthanthecalculatedreactivity,thefollowingconservativeassumptionsweremade:Moderatorisassumedtobeunboratedwateratatemperaturewithintheoperatingrangethatresultsinthehighestreactivity(determinedtobe204C).Theeffectivemultiplicationfactorofaninfiniteradialarrayoffuelassemblieswasused(seesection4.4.1)exceptfortheboundarystoragecellswhereleakageisinherent.Neutronabsorptioninminorstructuralmembersisneglected,i.e.,spacergridsareanalyticallyreplacedbywater.'IiThedesignbasisfuelassemblyisa15x15(Standard)WestinghousecontainingUO>atamaximuminitialenrichmentof4.95+0.05wt%U-235byweight.ForfuelassemblieswithnaturalUO)blanketsgtheenrichmentisthatofthecentralenrichedzone.Calculationsconfirmedthatthisreferencedesignfuelassemblywasthemost.reactiveoftheassemblytypesexpectedtobestoredintheracks.Threeseparatestorageregionsareprovidedinthespentfuelstoragepool,withindependentcriteriadefiningthehighest.potentialreactivityineachofthetworegions'sfollows:Region1isdesignedtoaccommodatenewfuelwithamaximumenrichmentof4.95+0.05wt%U-235,orspentfuelregardlessofthedischargefuelburnup.Region2isdesignedtoaccommodatefuelof4.954initialenrichmentburnedtoatleast50,000MWD/MtU(assemblyaverage),orfuelofotherenrichmentswithaburnupyieldinganequivalentreactivity.Region3isdesignedtoaccommodatefuelof4.954initialenrichmentburnedtoatleast38,000MWD/MtU(assemblyaverage),orfuelofotherenrichmentswithaburnupyieldinganequivalentreactivity.4-2 IIIII Thewaterinthespentfuelstoragepoolnormallycontainssolubleboronwhichwouldresultinlargesubcriticalitymarginsunderactualoperatingconditions.However,theNRCguidelines,basedupontheaccidentconditioninwhichallsolublepoisonisassumedtohavebeenlost,specifythatthelimitingkgffof0.9Sfornormalstoragebeevaluatedintheabsenceofsolubleboron.ThedoublecontingencyprincipleofANSIN-16.1-1975andoftheApril1978NRCletterallowscreditforsolubleboronunderotherabnormaloraccidentconditionssinceonlyasingleindependentaccidentneedbeconsideredatonetime.Consecgxencesofabnormalandaccidentconditionshavealsobeenevaluated,where"abnormal"referstoconditionswhichmayreasonablybeexpectedtooccurduringthelifetimeoftheplantand"accident"referstoconditionswhicharenotexpectedtooccurbutneverthelessmustbeprot'ectedagainst.4-3 lIII 4.24.2.1SummarofCriticalitAnalsesNormal0eratinConditionsThedesignbasislayoutofstoragecellsforthethreeregionsisshowninFigure4.1.Xnthisconfiguration,thefreshfuelcells(Region1)arelocatedalternatelyalongtherackperiphery(whereneutronleakagereducesreactivity)oralongtheboundarybetweentwostoragemodules(wherethewatergapprovidesaflux-trapwhichreducesreactivity).HighburnupfuelinRegion2affordsalow-reactivitybarrierbetweenfreshfuelassembliesandRegion3fuelofintermediateburnup.Thereareatthepresenttime,anadequatenumberofspentfuelassembliestonearlyfilland"blockoff"theRegion2barrierlocations(seeSection4.7).Thus,theadministrativecontrolsrequiredarecomparabletoaconventionaltwo-regionstoragerackdesign.Priortoapproachingthereactorend-of-life,notallstoragecellsareneededforspentfuel.Therefore,analternativeconfigurationmaybeusedinwhichtheinternalcellsareloadedinacheckerboardpatternoffreshfuel(orfuelofanyburnup)with'mptycells,asindicatedinFigure4.2.Thisconfigurationisintendedprimarilytofacilitateafullcoreunloadwhenneeded,priortothetimetheracksarebeginningtofillup.Figure4.3definetheacceptableburnupdomainsandillustratesthelimitingburnupforfuelofvariousinitialenrichmentsforbothRegion2(uppercurve)orRegion3(lowercurve),bothofwhichassumethatthefreshfuel(Region1)isenrichedto4.954U-235.Criticalityanalysesshowthatthemostreactiveconfigurationoccursalongtheboundarybetweenmoduleswiththereactivityof4-4 4$1I Itheedgeconfigurationbeingslightlylower.TheboundingcriticalityanalysesaresummarizedinTable4.1forthedesignbasisstoragecondition(whichassumesthesingleaccidentconditionofthelossofallsolubleboron)andinTable4.2fortheinterimcheckerboardloadingarrangement.Thecalculatedmaximumreactivityof0.940(sameforboththenormalstorageconditionandtheinterimcheckerboardarrangement)iswithintheregulatorylimitofak,<<of0.95.Thismaximumreactivity.includescalculationaluncertaintiesandmanufacturingtolerances(954probabilityatthe954confidencelevel),anallowanceforuncertaintyindepletioncalculationsandtheevaluatedeffectoftheaxialdistributioninburnup.Freshfueloflessthan4.954enrichmentwouldresultinlowerreactivities.Ascoolingtimeincreasesinlong-termstorage,decay'ofPu-241resultsinacontinuousdecreaseinreactivity,whichprovidesanincreasingsubcriticalitymarginwithtime.Nocreditistakenforthisdecreaseinreactivityotherthantoindicateconservatisminthecalculations.Theburnupcriteriaidentifiedabove(Figure4-3)foracceptablestorageinRegion2andRegion3canbeimplementedinappropriateadministrativeprocedurestoassureverifiedburnupasspecifiedintheproposedRegulatoryGuide1.13,Revision2.Administrativeprocedureswillalsobeemployedtoconfirmandassurethepresenceofsolublepoisoninthepoolwaterduringfuelhandlingoperations,asafurthermarginofsafetyandasaprecautionintheeventoffuelmisplacementduringfuelhandlingoperations.Thethickbase-plateontherackmodulesextendbeyondthestoragecellsandprovideassurancethatthenecessarywater-gapbetweenmodulesismaintained.4-5

~I~l~~l Forconvenience,theminimum(limiting)burnupdatainFigure4.3forunrestrictedstoragemaybedescribedasafunctionoftheinitialenrichment,E,inweightpercentU-235byfittedpolynomialexpressionsasfollows;ForReion2StoraeMinimumBurnupinMWD/MTU22'70+22I220E2I260E+149ForReion3StoraeMinimumBurnupinMWD/MTU26,745+18,746E-1,631E+98.4E4.2.2AbnormalandAccidentConditionsAlthoughcreditforthesolublepoisonnormallypresentinthespentfuelpoolwaterispermittedunderabnormaloraccidentconditions,mostabnormaloraccidentconditionswillnotresultinexceedingthelimitingreactivity(k,<<of0.95)evenintheabsenceofsolublepoison.TheeffectsonreactivityofcredibleabnormalandaccidentconditionsarediscussedinSection4.7andsummarizedinTable4.3.Oftheseabnormaloraccidentconditions,onlyonehasthepotentialforamorethannegligiblepositivereactivityeffect.4-6

~tl~~~~~~(I~

Theinadvertentmisplacementofafreshfuelassemblyhasthepotentialforexceedingthelimitingreactivity,shouldtherebeaconcurrentandindependentaccidentconditionresultinginthelossofallsolublepoison.Administrativeprocedurestoassurethepresenceofsolublepoisonduringfuelhandlingoperationswillprecludethepossibilityofthesimultaneousoccurrenceofthetwoindependentaccidentconditions.Thelargestreactivityincrease(+0.065Sk)wouldoccurifanewfuelassemblyof4.954enrichmentweretobepositionedinaRegion2locationwiththeremainderoftherackfullyloadedwithfuelofthehighestpermissiblereactivity.Underthisaccidentcondition,creditforthepresenceofsolublepoisonispermittedbyNRCguidelines,andcalculationsindicatethat550ppmsolubleboronwouldbeadecpxatetoreducethek,<<tothecalculatedk,<<(0.940)andapproximately450ppmwouldbesufficienttoassurethatthelimitingkoffof0.95isnotexceeded.DoublecontingencyprincipleofANSIN16.1-1975,asspecifiedintheApril14,1978NRCletter(Section1.2)andimpliedintheproposedrevisiontoReg.Guide1.13(Section1.4,AppendixA).4-7 Il~~~~Ii~~

4'4.3.1ReferenceFuelStoraeCellsReferenceFuelAssemblThedesignbasisfuelassembly,describedinFigure4.4,isa15x15arrayoffuelrodswith21rodsreplacedby20controlrodguidetubesand1instrumentthimble.Table4.4summarizesthefuelassemblydesignspecificationsandtheexpectedrangeofsignificantmanufacturingtolerances.Asshownbelow,initialcellcalculationswithCASMO-3indicatedthattheW15x15fuelexhibitedaslightlyhigherreactivityinthestoragerackcellthaneithertheW17x17standardoroptimized(OFA)fuelortheANFfuelassemblydesigns.FuelteW15x15W15x15Enrichment:2.52.5Burnup~D/DKU010Cell~k~1.02610.9210W17x17OFAW17x17OFAW17x17StndW17x17StndANF15x15ANF17x17W15x15W15x152.52.52.52.52.52.54.954.95010010000401.02050.91441.02170.91881.01481.01261.19410.9204W17xW17xW17xANF15ANF1717OFA17OFA17Stndx15x174.954.954.954.954.950400001.19330.91491.18801.18571.1883Highestvalues4-8

~~l~

Baseduponthecalculationslistedabove,theWestinghouse15x15roddesignwasusedasthebasisforthecriticalitycalculations.4.3.2HihDensitFuelStoraeCellsThenominalspentfuelstoragecellusedforthecriticalityanalysesoftheDonaldC.CookspentfuelstoragecellsisshowninFigure4.4.EachstoragecelliscomposedofBoralabsorberpanelspositionedbetweena8.75-inchI.D.,0..075-inchthickinnerstainlesssteelbox,anda0.035-inchouterstainlesssteelsheathwhichformsthewalloftheadjacentcell.Thefuelassembliesarenormallylocatedinthecenterofeachstoragecellonanominallatticespacingof8.97+0.04inches.TheBoralabsorberhasathicknessof0.101+0.005inchandanominalB-10arealdensityof0.0345g/cm.4-9 f'll'lI1wIKI 4'4.4.1AnalicalMethodoloReferenceDesinCalculationsInthefuelrackanalyses,theprimarycriticalityanalysesofthehighdensityspentfuelstoragerackswereperformedwiththeKENO-(1).~*5acomputercodepackage,'singthe27-groupSCALEcross-sectionlibraryandtheNITAWLsubroutineforU-238resonanceshieldingeffects(Nordheimintegraltreatment).Depletionanalysesanddeterminationofequivalentenrichmentsweremadewiththetwo-dimensionaltransporttheorycode,CASMO-3.Benchmark/calculations,presentedinAppendixA,indicateabiasof0.0000withanuncertaintyof+0.0024forCASMO-3and0.0090+0.0021(954/954)forNITAWL-KENO-Sa.Intrackinglong-term(30-year)reactivityeffectsofspentfuelstored'inRegion2ofthefuelstoragerack,previousCASMOcalculationsconfirmedacontinuousreductioninreactivitywithtime(afterXedecay)dueprimarilytoPu-241decayandAm-241growth.KENO-SaMonteCarlocalculationsinherentlyincludeastatisticaluncertaintyduetotherandomnatureofneutrontracking.TominimizethestatisticaluncertaintyoftheKENO-calculatedreactivity,aminimumof500,000neutronhistoriesin1000generationsof500neutronseach,areaccumulatedineachcalculation.Forthedesigncalculationfortheracks,1,250,000historieswereusedtoconfirmconvergenceoftheKENO-5acalculation.Figure4.5representsthebasicgeometricmodelusedintheKENO-5acalculations.Thismodeleffectivelydescribesarepeatingarrayof10storagecellsintheX-directionseparatedbya2-inchwater"SCALE"isanacronymforStandardizedComputerAnalysisforlicensing/valuation,astandardcross-sectionsetdevelopedbyORNLfortheUSNRC.4-10 LIItII gapbetweenmodulesandaninfinitearrayofcellsintheY-direction(periodicboundaryconditions).Intheaxial(Z)direction,thefulllength144-inchfuelassemblywasdescribedwitha30-cmwaterreflector.Asimiliarmodelwasusedforcalculationsoftherackperipheralcellswherethecalculationsweremadewithbothwaterandconcretereflectors(aconcretereflectorgaveaslightlyhigherreactivityby0.004Sk).Largermodels,encompassinganentirestoragemodule(halfofan11x11array,runfor1,250,000neutronhistoriestoassureconvergence)confirmedresultsobtainedwiththesmallerinfinitearraymodel.ThelargermodelwasalsousedtoconfirmthereactivitycalculationforthecheckerboardarrangementwithfreshfuelandemptycellsinRegion3andintheinvestigationoftheconsequencesofpotentialaccidentconditionswithamisplacedfreshfuelassembly.Inaddition,thecornerintersectionwasexplicitlymodeledand,as.expected,gavealowerreactivitythanthereferencedesigncalculation.IntheCASMO-3geometricmodel(cell),eachfuelrodanditscladdingweredescribedexplicitlyandreflectingboundaryconditions(zeroneutroncurrent)wereusedintheaxialdirectionandatthecenterlineoftheBoralandsteelplatesbetweenstoragecells.Theseboundaryconditionshavetheeffectofcreatinganinfinitearrayofstoragecellsinalldirectionsandprovideaconservativeestimateoftheuncertaintiesinreactivityattributedtomanufacturingtolerances.BecauseNITAWL-KENO-5adoesnothaveburnupcapability,burnedfuelwasrepresentedbyfuelofequivalentenrichmentasdeterminedbyCASMO-3calculationsinthestoragecell(i.e.anenrichmentwhichyieldsthesamereactivityinthestoraecellastheburnedfuel).4-11 lIII Figure4.6showsthisequivalentenrichment.forfuelof4.954initialenrichmentatvariousdischargeburnups,evaluatedinthestoragecell.4.4.2FuelBurnuCalculationsandUncertaintiesCASMO-3wasusedforburnupcalculationsinthehotoperatingcondition.CASMO-3hasbeenextensivelybenchmarked(AppendixAandRefs.2and7)againstcriticalexperiments(includingplutonium-bearingfuel).Inadditiontoburnupcalculations,CASMO-3wasusedforevaluatingthesmallreactivityincrements(bydifferentialcalculations)associatedwithmanufacturingtolerances.Sincetherearenocriticalexperimentdatawithspentfuelfordeterminingtheuncertaintyinburnup-dependentreactivitycalculations,anallowanceforuncertaintyinreactivitywasassignedbasedupontheassumptionof54uncertaintyinburnup.Thisisapproximatelyequivalentto54ofthetotalreactivitydecrement.Atthedesignbasisburnupsof38and50MWD/KgU,theuncertaintiesinburnupare+1.9and+2.5=MWD/KgUrespectively.Toevaluatethereactivityconsequencesoftheuncertaintiesinburnup,independentcalculationsweremadewithfuelof36,100and47,500MWD/MtUburnupinRegions2and3,andtheincrementalchangefrom.thereferenceburnupsassumedtorepresentthenetuncertaintiesinreactivity.Thesecalculationsresultedinanincrementalreactivityuncertaintyof+0.0047SkinRegion2(isolationbarrierat50MWD/KgUburnup)and+0.0019forRegion3(at38MWD/KgUburnup)..Intheracks,thefreshunburnedfuelinRegion1stronglydominatethereactivitywhichtendstominimizethereactivityconsequencesofuncertaintiesinburnup.The*Onlythatportionoftheuncertaintyduetoburnup.Otheruncertaintiesareaccountedforelsewhere.4-12 I4e allowanceforuncertaintyinburnupcalculationsis'conservativeestimate,particularlyinviewofthesubstantialreactivitydecreasewithtimeasthespentfuelages.4.4.3EffectofAxialBurnuDistributionInitially,fuelloadedintothereactorwillburnwithaslightlyskewedcosinepowerdistribution.Asburnupprogresses,theburnupdistributionwilltendtoflatten,becomingmorehighlyburnedinthecentralregionsthanintheupperandlowerends,asmaybeseeninthecurvescompiledinRef.4.Athighburnup,themorereactivefuelneartheendsofthefuelassembly(lessthanaverageburnup)occursinregionsoflowerreactivityworthduetoneutronleakage.Consequently,itwouldbeexpectedthatovermostoftheburnuphistory,distributedburnupfuelassemblieswouldexhibitaslightlylowerreactivitythanthatcalculatedfortheaverageburnup.Asburnupprogresses,thedistribution,tosomeextent,tendstobeself-regulatingascontrolledbytheaxialpowerdistribution,precludingtheexistenceoflargeregionsofsignificantlyreducedburnup.Amongothers,Turnerhasprovidedgenericanalyticresultsoftheaxialburnupeffectbaseduponcalculatedandmeasuredaxialburnupdistributions.Theseanalysesconfirmtheminorandgenerallynegativereactivityeffectoftheaxiallydistributedburnup.Thetrendsobserved,however,suggestthepossibilityofasmallpositivereactivityeffectathighburnup.CalculationsweremadewithKENO-5ainthreedimensions,baseduponthetypicalaxialburnupdistributionofspentfuel(thatobservedattheSurreyplantwastakenasrepresentative).Xnthesecalculations,theaxialheightoftheburnedfuelwasdividedintoanumberofaxialzones(6-inchintervalsnearthemoresignificanttopofthefuel),eachwithanenrichmentequivalenttotheburnup4-13 IOlIl ofthatzone.Thesecalculationsresultedinanincrementalreactivityincreaseof0.0037b'kforthereferencedesigncase.Fueloflowerinitialenrichments(andlowerburnup)wouldhaveasmaller(ornegative)reactivityeffectasaresultoftheaxialvariationinburnup.Theseestimatesareconservativesincesmalleraxialincrementsinthecalculationshavebeenshowntoresultinlowerincrementalreactivities4-14 II 4.54.5.1CriticalitAnalses'andTolerancesNominalDesinForthenominalstoragecelldesign,theNlTAWL-KENO-Sacalculationresultedinabias-correctedkof0.9250+0.0012(9S%/95%),which,whencombinedwithallknownuncertaintiesandtheaxialburnupeffect,resultsinakof0.929+0.011oramaximumkof0.940witha954probabilityatthe95:confidencelevelFortheinterimloadingpatternofcheckerboardedfuelandemptycellsinRegion3,calculationsresultedinessentiallythesamereactivityasthereferencedesignwithinthenormalKENO-5astatistics(maximumkof0.940,includingallallowancesanduncertainties,seeTable4.2).4UncertaintiesDuetoManufacturinTolerancesTheuncertaintiesduetomanufacturingtolerancesaresummarizedinTable4-5anddiscussedbelow.4.5.2.1BoronLoadinTolerancesTheBoralabsorberpanelsusedinthestoragecellsarenominally0.101inchthick,7.50-inchwideand144-inchlong,withanominalB-10arealdensityof0.034Sg/cm.Thevendorsmanufacturingtolerancelimitis+0.004Sg/cminB-10contentwhichassuresthatatanypoint,theminimumB-10arealdensitywillnotbelessthan0.030g/cm.DifferentialKENO-5acalculationsforthereferencedesignwiththeminimumtoleranceB-10loadingresultsinanincrementalreactivityof+0.00614Skuncertainty.4-15 II1ll 4.5.2.2BoralWidthToleranceThereferencestoragecelldesignusesaBoralpanelwithaninitialwidthof7.50+0.06inches.Forthemaximumtoleranceof0.06inch,thedifferentialCASMO-3calculatedreactivityuncertaintyis+0.0009Sk.4.5.2.3TolerancesinCellLatticeSacinThemanufacturingtoleranceontheinnerboxdimension,whichdirectlyaffectsthestoragecelllatticespacingbetweenfuelassemblies,is+0.06inches.Thiscorrespondstoanuncertaintyinreactivityof+0.0015SkdeterminedbydifferentialCASNO-3calculations.4.5.2.4StainlessSteelThicknessTolerancesThenominalstainlesssteelthicknessis0.075+0.005inchfortheinnerstainlesssteelboxand0.035+0.003inchfortheBoralcoverplate.Themaximumpositivereactivityeffectoftheexpectedstainlesssteelthicknesstoleranceswascalculated(CASMO-3)tobe+0.0009Sk.4.5.2.5FuelEnrichmentandDensitTolerancesThedesignmaximumenrichmentis4.95+0.05wt%U-235.SeparateCASMO-3burnupcalculationsweremadeforfuelofthemaximumenrichment(5.004)andforthemaximumUO>density(10.50g/cc).ReactivitiesinthestoragecellwerethencalculatedusingtherestartcapabilityinCASMO-3andequivalentenrichmentsdeterminedforthereferencefuelburnupsof38and50MWD/KgU.TheincrementalreactivitiesbetweenthesecalculationsandthereferenceCASMO-3cases,wereconservativelytakenasthe4-16 lI sensitivity'tosmallenrichmentanddensityvariations.ForthetoleranceonU-235enrichment,theuncertaintyinkis+0.00346kandforfueldensityis+0.0035.4.5.3Water-aSacinBetweenModulesThewater-gapbetweenmodulesconstituteaneutronflux-trapfortheouter(peripheral)rowofstoragecells.CalculationswithKENO-5aweremadeforvariouswater-gapspacings(Figure4.7).Fromthesedata,itwasdeterminedthattheincrementalreactivityconsequence(uncertainty)fortheminimumwater-gaptoleranceof+1/4inchis+0.00456k.Theracksaresonstructedwiththebaseplateextendingbeyondtheedgeofthecells.Thisassuresthataminimumspacingof1.75inchbetweenstoragemodulesismaintainedunderallcredibleconditions.4.5.4ccentricFuelPositioninThefuelassemblyisassumedtobenormallylocatedinthecenterofthestoragerackcell.InfinitearraycalculationsweremadeusingKENO-5aforasinglecellwiththefuelassembliescenteredandwiththeassembliesassumedtobeinthecornerofthestoragerackcell(four-assemblyclusteratclosestapproach).Thesecalculationsindicatedthatthereactivityuncertaintycouldbeasmuchas+0.00196k.4.64~6~1AbnormalandAccidentConditionsTemeratureandWaterDensitEffectsThemoderatortemperaturecoefficientofreactivityisnegative;amoderatortemperatureof20'C(68'F)wasassumedforthereferencedesigns,whichassuresthatthetruereactivitywillalwaysbelowerovertheexpectedrangeofwatertemperatures.Temperature4-17 IIIVIII effectsonreactivityhavebeencalculatedandtheresultsareshowninTable4.6.Withsolublepoisonpresent,thetemperaturecoefficientsofreactivitywoulddifferfromthoseinferredfromthedatainTable4.6.However,thereactivitieswouldalsobesubstantiallyloweratalltemperatureswithsolubleboronpresent,andthedatainTable4.6ispertinenttothehigher-reactivityunboratedcase.4.6'DoedFuelAssemblForadropontopoftherack,thefuelassemblywillcometoresthorizontallyontopoftherackwithaminimumseparationdistancefromthefuelintherackofmorethan12inches,including'thepotentialdeformationunderseismicoraccidentconditions.Atthisseparationdistance,theeffectonreactivityisinsignificant(<0.00016'k).Furthermore,solubleboroninthepoolwaterwouldsubstantiallyreducethereactivityandassurethatthetruereactivityisalwayslessthanthelimitingvalueforanyconceivabledroppedfuelaccident.4.6.3ateralRackMovementLateralmotionoftherackmodulesunderseismicconditionscouldpotentiallyalterthespacingbetweenrackmodules.However,themaximumrackmovementhasbeendeterminedtobelessthanthetoleranceonthewater-gapspacing.Furthermore,solublepoisonwouldassurethatareactivitylessthanthedesignlimitationismaintainedunderallaccidentorabnormalconditions.4-18 IIIII 4.6.4AbnormalLocationofaFuelAssemblTheabnormallocationofafreshunirradiatedfuelassemblyof4.95wt4enrichmentcould,intheabsenceofsolublepoison,resultinexceedingthedesignreactivitylimitation(kof0.95).Thiscouldoccurifafreshfuel'ssemblyofthehighestpermissibleenrichmentweretobeeitherpositionedoutsideandadjacenttoastoragerackmoduleorinadvertentlyloadedintoeitheraRegion2orRegion3storagecell.Calculations(KENO-5a)showedthatthehighestreactivity,includinguncertainties,fortheworstcasepostulatedaccident,condition(freshfuelassemblyinRegion2)wouldexceedthelimitonreactivityintheabsenceofsolubleboron.Solubleboroninthespentfuelpoolwater,forwhichcreditispermittedundertheseaccidentconditions,wouldassurethatthereactivityismaintainedsubstantiallylessthanthedesignlimitation.Itisestimatedthatasolublepoisonconcentrationof550ppmboronwouldbesufficienttomaintainkatthereferencedesignvalueof0.940underthemaximumpostulatedaccidentcondition.Approximately450ppmboronwouldberequiredtolimitthemaximumreactivitytoak~ffof0.95.4.7ExistinSentFuelAsofMay1990,therewere1596spentfuelassembliesinstorageattheDonaldC.Cookplant,includingthosenowinthereactorandtheirprojectedburnupsatdischarge.Figure4.8superimposestheenrichment-burnupcombinationofthesefuelassembliesonthecurvesdefiningtheacceptableburnupdomains.Asmaybeseeninthisfigure,mostofthespentfuelnowinstoragefallswellintotheacceptabledomainforthebarrierfuel(Region2).Thenumberoffuelassembliesmeetingtheenrichment-burnupcriteriaforstorageinRegion2is1390whichwillnearlyfillthe1447Region2storagelocations.Twelvefuelassemblies(discharged4-19 IIII prematurelyforvariousreasons)willneedtobekeptinaRegion1storagelocation,andtheremaining194assembliesmaybes'toredinRegion3locations.FuturedischargebatchesmayreasonablybeexpectedtohaveapreponderanceofhighlyburnedfuelcapableofbeingstoredinRegion2(orinRegion3onceRegion2isfilled).Anappreciablenumberofspentfuelassemblieshaveenrichment-burnupcombinationswellinexcessofthedesignbasisand,thisprovidesfurtherconservatisminthecriticalitysafetyofthespentfuelstoragerackdesign.4-20 IlII 4.8ReferencesGreen,Lucious,Petrie,Ford,White,andWright,"PSR-63-/NXTAWL-1(codepackage)NITAWLModularCodeSystemForGeneratingCoupledMultigroupNeutron-GammaLibrariesfromENDF/B",ORNL-TM-3706,OakRidgeNationalLaboratory,November1975.2~3.4~5.6.R.M.Westfallet.al.,"SCALE:AModularSystemforPerformingStandardizedComputerAnalysisforLicensingEvaluation",NUREG/CR-0200,1979.Volume2,SectionF11,"KENO-5aAnImprovedMonteCarloCriticalityProgramwithSupergrouping".A.Ahlin,M.Edenius,andH.Haggblom,"CASMO-AFuelAssemblyBurnupProgram",AE-RF-76-4158,Studsvikreport.A.AhlinandM.Edenius,"CASMO-AFastTransportTheoryDepletionCodeforLWRAnalysis",ANSTransactions,Vol.26~p604I1977~"CASMO-3AFuelAssemblyBurnupProgram,UsersManual",Studsvik/NFA-87/7,StudsvikEnergitechnikAB,November1986H.Richings,SomeNotesonPWR(W)PowerDistributionProbabilitiesforLOCAProbabilisticAnalyses,NRCMemorandumtoP.S.Check,datedJuly5,1977.S.E.Turner,"UncertaintyAnalysis-BurnupDistribu-tions",presentedattheDOE/SANDIATechnicalMeetingonFuelBurnupCredit,SpecialSession,ANS/ENSConference,Washington,D.C.,November2,1988M.G.Natrella,ExperimentalStatisticsNationalBureauofStandards,Handbook91,August1963.4-21 IIIIII Table4.1SUMMARYOFCRITICALITYSAFETYANALYSESNORMALSTORAGECONFIGURATIONDesignBasisburnupsat4.954+0.054initialenrichmentTemperatureforanalysisReferencek(KENO-5a)Calculationalbias,SkUncertaintiesBiasstatistics(954/954)KENO-5astatistics(954/95%)ManufacturingTolerancesWater-gapFuelenrichmentFueldensityBurnup(38MWD/KgU)Burnup(50MWD/KgU)EccentricityinpositionStatisticalcombingionofuncertaintiesAxialBurnupEffectTotalMaximumReactivity(k)0inRegion150inRegion238inRegion320C(68F)0.91600.0090+0.0021+0.0012+0.0064+0.0045+0.0034+0.0035+0.0019+0.0047+0.0019+0.01100.00370.9287+0.01100.940SeeAppendixASquarerootofsumofsquares.(2)4-22 III Table4.2SUMMARYOFCRITICALITYSAFETYANALYSESINTERIMCHECKERBOARDLOADINGDesignBasisburnupsat4.954+0.054initialenrichmentTemperatureforanalysisReferencek(KENO-5a)Calculationalbias,6k0inRegion150inRegion2Region3-CHECKERBOARD(FRESHFUELANDEMPTY)20~C(684F)0.91680.0090Uncertainties(Assumedsameasthereferencecase)Biasstatistics(954/954)KENO-5astatistics(954/954)ManufacturingTolerancesWater-gapFuelenrichmentFueldensityBurnup(38MWD/KgU)Burnup(50MWD/KgU)EccentricityStatisticalcombinationofuncertaintiesAxialBurnupEffectTotalMaximumReactivity(k)+,0.0021+0.0012+0.0064'0.0045+0.0034+0.0035NA+,0.0047+0.0019+0.01080.00370.9295+0.01080.940SeeAppendixASquarerootofsumofsquares.(2)4-23 IIIIIIII Table4.3REACTIVITYEFFECTSOFABNORMALANDACCIDENTCONDITIONSAccident/AbnormalConditionsReactivityEffectTemperatureincrease(above684F)Void(boiling)AssemblydroppedontopofrackLateralrackmodulemovementMisplacementofafuelassemblyNegative(Table4.6)Negative(Table4.6)Negligible(<0.0001Sk)(IncludedinTolerances)Positive(0.065Max6k)(controlledbysolublepoison)4-24 IIII Table4.4DESIGNBASISFUELASSEMBLYSPECIFICATIONSFUELRODDATAOutsidediameter,in.Claddingthickness,in.Claddinginsidediameter,in.CladdingmaterialPelletdensity,4T.D.Stackdensity,gUO>/ccPelletdiameter,in.Maximumenrichment,wt4U-2350.4220.02430.3734Zr-495.010.29+0.200.3659495+0.05FUELASSEMBLYDATAFuelrodarrayNumberoffuelrodsFuelrodpitch,in.NumberofcontrolrodguideandinstrumentthimblesThimbleO.D.,in.(nominal)ThimbleI.D.,in.(nominal)15x152040'630.5330.4994-25 IIIIII Table4.5ReactivityEffectsofManufacturingTolerancesToleranceIncrementalReactivitSkBoron-10loading(+0.0045g/cm)BoralWidth(+1/16inch)Latticespacing(+0.04inch)StainlessThickness(+0.005inch)0.00610.00090.00150.0009Total(statisticalsum)+0.00644-26 I

Table4.6EFFECTOFTEMPERATUREANDVOIDONCALCULATEDREACTIVITYOFSTORAGERACKCaseIncrementalReactivityChange,SkRegion1Region220oC(68oF)40oC(104oF)66C(150F)90C(194F)122C(252F)Reference-0.003-0.009-0013-0.024122C(252F)+204void-0.071Reference-0.002-0.005-0.010-0.015-0.0614-27 2g' IOf8lftmfskmf888%~attfftfNeftce@$$NSSSINefsNu$$NNESS%f8NmssSM388%~kmSNSNQSNftkrSPSOtftif?fNN$$NS$ftfN~SNPRESNSSNRft$~algllkak)IIISII5akmIarlakrEIONR'ciPNRSSIkfshtt"NamN88%I-omsff88$$hmmsk~SNettfmc$$NK$$kr~SNSNQNNNSRSS%8QR)%4lSSQI8%~%Nst'tf'NR$$NRQSaON$$fftNNf?INNtCISa%%assr%ask%aSOOSS~SSSC'ICSCIgN$$M<W$;5g%$8PRmmlf"mmmkrsN$$eua'eweesa%8SfsSISIS$NNN$%~SssRmf8sttmf8I88%a~'Earl%ma%a%askO~O~O~~IrlaiaCrkrsr-kmsff88$5"38$'?lsasNst.st$NiftfN&esoss8%888s)98%a%8%m?8$c'M%iftamSNRS$NNgNN$tf~rsssfsmssI88mt$smlr08Ksflms$%8KSSSSNlRstfNR$$SNfttkr%%8@Nh@ssRmsSSAI@t88$$%?8$$%%aoar%sara%aso~oo%%OS%SOS~4SIgi'5%IgiaamISNEQr'5~NIQS8tafsaIif?8$8QtifkkaSkkktftmfNt$IPSM4%88I888f8i38a%$$N$$N$$tft~IS5gg5gggI5)gg55g"5's?$8sssmssh8%r%%f88$$18$tflsOrSRtf$NANNA$I$%rSsmssSmf8bmoassff8msffsesfis%aRNNNNA$$%asar%%I%saskS~%%%%%O~%%OO~krarka%%srOI88$MII$58mk~ssNRftfRRN$$$$III%8$I88ISII8$8)ik%%8matQmf8%$$Skra$$NstfftfNtttf$$Naa%SfQRSfSNRISgrk5mf8$mf8884%$$NStf$$NNftfiQO%41INSfISS~kraNSIIIakr%tarn.ItfmmIm5rS'$tstfNfflttfNRft$%k%%$$Nssfslkstf~QRSS%$$fsmsfISmssfskOSRSNRHf8Smmra%aNIIIImaa~gggs5gt,gASSISIeOSNR$$NNSr%85$NRsfNRolsLSSI8sa%$$IR$$N5f$$akn5$NeisfefftmkNSSaCr5%%5aaSI%$$4mmi85ÃcSQS$MtIISNN$8%s$%ra%f?f588amANN(asmMmOtfttftfNIg4SIaki$t5fNlRWNN$$kr%%$tfmssftfSNSNSS~SfSSSS$SS8KSSS'SNwgssNQRSr~Sr%as%Is%SrOOSOO~g)5g5gggar'1158"LMM'r~emsseSI8emrr%0Ntttf$$NQ$$fft%rOhlkRI88QI8%rsBsssgmf8ismkrrrasaklsaagrPS%SO%oaa)IIlrSr5Srkrga'h11%$888SII88aNas$N0$$a8$cl8NSgOII88$8isf?I$886kk$$R?N$$8tf$$$?~%Nssl$$NsSSNRQomssss8mfssmkmsfnmae4~gnggugetm5gO~OO'Oalaar%pirl)CI%$N'$$NN$I$NQmk~SQftwmsssmsssRSIomml883immsslslrkNBSNRNQftfr~SN$$NQft?ItWNSSSksks1ISrgaRr~agggft5ggg5IOC~OOIkkaaaCS'LImkaoms$$88$8I8rsNstsNlR$$NRlslaSSQ$88MIS8%mI8$$tmNl88$$hglassB@st88RSNgstfNlftflftQsga8$$Imm8emsslsklIgiwggg541am%hakNQE$8tssfsms$I8%rSEemLIPamNIIr%$hSS@8i$SSQI8%rIf8l88$888$888ISaas'""ISIFSNIrIj)5w4Nra8$?i))maINEIN%Nrfftstfkrmsffsmkf8%$$NSSssftfm@kk8SfRRON888fQI8mf8SNia$$$9%1~OOOOO'INGISWISSiSSII$$$ftmamm5$$fftmkala%IpQ8%1I?$SNIaoaaNQQNSSf8Rmssktff$$8tffkrSlQftfNO%%%askOOkaM88$$%'5ala'5SCrIls5cv85~5ssfsmss8tef8samsfssmsfi88%NRstfNRltfNSrf888Msmssk88$f88s$%msaNEW4N4$tfNSa%a%a%a%a'Ig~aQgO)$NassfssellaOOOOam%Ca'5sksaCaSINNNNIeNIIISak%)sgsfI88%lskaIggg)5)ggg)LaCSCgN)SNgII~alaBSNN$?INrkssmfftssSSRlSS$$%%Qkarkf88ffiff8t$$8$PRN$$4tt$NQNRSNO~Iaaa4karISI~OOOO%OOOOIIaaakIEarmark'NNRQNrsmsslsmssOf8msf!88SSIPNifI8IN~SN@bmNOSlmS$SS8'aSSNMft$tONNINNasar%aOONNg86%LIaCasalmgCy'5la~iNNIS@luma~Omf8SINBSSMSSSNRlft$NAftfNQft$%rOES58f888tl88S$%akms3f88$si88$8isgr~CakakafimkaseOOOO~~~~g,o JiI Sfy'5PSyCqCy'ESQPiQQ~SNSSSSSSSSSS~5~~50hagi'iflISPSM~EN'Ihi8MBASSiilaENESOaPaIIe'~5~)agI3Q8~L'$S@A45AS0~SSS~iala)$SStCCSSSml88~iaaagaga5~:i0gEhaiaaNmlPaPSSS88I88SNIN8SSSNSCCNgSFsPamrea55I$ASSPASAS4~giaiaa~S8NISWRSSSCCW)'055ASAll~SglAlgaASQCdQQSASSQQQQQSSS.QQ.Q.Q.~~WQQQQQSSAQQQgSS54I¹S¹Ã¹g¹gA~M<8~~~X5$IQNCIINgiACSSSSSSSNSSAS~IimIImimi,SSNISrgSSSS~'>>i:hl~SCM8551S48A~i.i,,s,.SSSSSSmSSSSSSSS~iagamaaa/aaa'ghee8@8ImRSANNSSSgSSSSSS00~~~8@8858SI3858QASSS~0ASRSASQAASSSAQMQAS.Q.QQCd~SAQQf3QSSAASASSSASSSSSSSS0~SSSS~ASASASASASASS~SSAXSASSSSS~~S5SS~ENDS5SCA0l)C~aag+paaa~hiSSPRSSAND)aSCa8BASSA00SgNAS4SSPA5)A%ASSNamNSNA)III5lg'a.eiSOg8~a55iSMSSSgS8NCaprie'88585ggNNCCISSS5SS5tSNICCN$8%SSS5S'SSSSSSASSSSllS8SSNNNISSSSASNBN~CCRC~aii~58N~PSkmN5~IN'SSpiaS3AEASSSN~SNiaF$IRSS<<SICIgNRNNIIIISSS4S0SSSSASASSSSSSSSSCPCSSSIICSSS5NNN4SSSSSSSASS500005SNSSSSNSSSSNLCN$50SSSS5%85%85ISSSSRSNR8$88S~ICISNLNICINSI8m8im~Sag88g388RSSNNPRENSM4SNINNSN88INSSSaII388fI38SI38SSNICINCCWNN55NS~'II%!IIi:MaSSSSSSSSSSSSS0050 n4I~~I~

5QQOO45000hoaaa~35000a>30000CLK~25aaaCO~2aaaa4.1500010000ACBUREPTALEUPDMAIN(RequNACCRNUPresSterPTABLDOMAIeinRe1an1).01.52.0283.03.5h.oh.55.0INITIALENRICHMENT,mU-235F1g.4-3ACCEPTABLEBURNUPDOMAININREGIONS28c34-30

V4+g~p~7.50+0.06"BORALIN0.109"GAP0.101+.0.005"THK0.03459mB-10/crrPcQFUELOD0.3659IN.DENSITY10.29g/ccoogNUMBER204(15X15)CLADID0.3734IN.OOOO.RODOD0.422IN.00004I'ITCH0.563IN.+++++,THIMBLESOD0,546INOOOOOO<ID0.'512IN.'0000004ooooooooe00ioooooo.0000000000E,00000000000K00000000000040.035"SSSHEATH0.219"OOOiIS~@@e8.75-0.04"BOXI.D.PITCH8.969"NOMINALoo~-'075"SSBOXINOTTOSCALEIFIG.4-4FUELSTORAGECELLCROSSSECTION4-31 I

2"WATERGAPWITHREFLECTINGBOUNDARYCONDITIONSORREFLECTOR(WATERORCONCRETE)WITHZEROFLUXBOUNDARYCONDITIONPERIODICBOUMDARYCOHDITIOHC~CREGION3rREGIOH3REGION3REGION2FRESHFUEL.0:xo'.x>>'oREGION3REGION3REGION3REGION2REGION2PERIODICBOUMDARYCONDITIONFIG.4-5KENOCALCULATIONALMODEL I

I85.004.754.504.254.003.753.503.253.'002.752.502.252.001.751.501.251.000.75051015202530354045505560BURNUP,MWD/KgUFig.4-6EQUIVALENTENRICHMENTFORSPENTFUELATVARIOUSBURNUPSFORINITIALENRICHMENTOF4.95%4-33 l

0.9400.9350,9300-g0.925hlZ0.920~~0,91540.9100.9050.9000.5WATERGAPBETWEENMODULES,inches101520253.0Fig.4-7EFFECTOFVIATER-GAPSPACINGBETWEENMODULESONSYSTEMREACTIVITY'-34 lIl.IlI 55000saaoo4SOOO40000+35000CI+30000CL~2500QCQ~2ooaa4tsaaotaaaaACBURNEPTALEPDOAIN(RequNACCPTABLRNUPOOMAIresStoreInReIont)50000t.52.02DXO3.54.04.55.0INITIALENRICHMENT,mU-235Fig.4-8ACCEPTABLEBURNUPDOMAININREGIONS28c3SHOWINGEXISTINGSPENTFUELASSEMBLIES4-35 I

2CPPENDIXABENCHMARKCALCULATIONSbyStanleyE.Turner,PhD,PEHOLTECINTERNATIONALJanuary1991 I

1.0INTRODUCTION

ANDSUM~YThe.objectiveofthisbenchmarkingstudyistoverifyboththeNITAWL-KENO-5a(~)methodologywiththe27-groupSCALEcross-sectionlibraryandtheCASMO-3code()foruseincriticalitysafetycalculationsofhighdensityspentfuelstorageracks.Bothcalculationalmethodsarebasedupontransporttheoryandhavebeenbenchmarkedagainstcriticalexperimentsthatsimulatetypicalspent.fuelstoragerackdesignsasrealisticallyaspossible.Resultsofthesebenchmarkcalculationswithbothmethodologiesareconsistentwithcorrespondingcalculationsreportedintheliterature.Resultsofthebenchmarkcalculationsshowthatthe27-group(SCALE)NITAWL-KENO-5acalculationsconsistentlyunder-predictthecriticaleigenvalueby0.0090+0.0021Sk(witha95<probabilityata954confidencelevel)forcriticalexperiments<)thatareasrepresentativeaspossibleofrealisticspentfuelstoragerackconfigurationsandpoisonworths.Extensivebenchmarkingcalculationsofcriticalexperi-mentswithCASMO-3havealsobeenreported<),givingameank,<of1.0004+0.0011for37cases.WithaK-factorof2.14<@for954probabilityata954confidencelevel,andconservativelyneglect-ingthesmalloverprediction,theCASMO-3biasthenbecomes0.0000+0.0024.CASMO-3andNITAWL-KENO-5aintercomparisoncalculationsofinfinitearraysofpoisonedcellconfigurations(representativeoftypicalspentfuelstoragerackdesigns)showverygoodagreement,confirmingthat0.0000+0.0024isareasonablebiasanduncertaintyforCASMO-3calculations.Reference5alsodocumentsgoodagreementofheavynuclideconcentrationsfortheYankeecoreisotopics,agreeingwiththemeasuredvalueswithinexperimentalA-1 lIIII Thebenchmarkcalculationsreportedhereconfirmthateitherthe27-group(SCALE)NITAWL-KENOorCASMO-3calculationsareacceptableforcriticalityanalysisofhigh-densityspentfuelstorageracks.Referencecalculationsfortherackdesignsshouldbeperformedwithbothcodepackagestoprovideindependentverification.2.0NITAWL-KENO5aBENCHMARKCALCULATIONSAnalysisofaseriesofBabcock&Wilcoxcriticalexperiments(,includingsomewithabsorberpanelstypicalofapoisonedspentfuelrack,issummarizedinTable1,ascalculatedwithNITAWL-KENO-5ausingthe27-groupSCALEcross-sectionlibrayandtheNordheimresonanceintegraltreatmentinNITAWL.DancofffactorsforinputtoNITAWLwerecalculatedwiththeOakRidgeSUPERDANroutine(fromtheSCALE()systemofcodes).Themeanforthesecalculationsis0.9910+0.0033(1ostandarddeviationofthepopulation).Withaone-sidedtolerancefactorcorrespondingto95<probabilityata954confidencelevel(@,thecalculationalbiasis+0.0090withanuncertaintyof+0.0021forthesixteencriticalexperimentsanalyzed.SimilarcalculationaldeviationshavebeenreportedbyORNLforsome54criticalexperiments(mostlycleancriticalwithoutstrongabsorbers),obtainingameanbiasof0.0100+0.0013(954/954).Thesepublishedresultsareingoodagreementwiththeresultsobtainedinthepresentanalysisandlendfurthercredencetothevalidityofthe27-groupNITAWL-KENO-5acalculationalmodelforuseincriticalityanalysisofhighdensityspentfuelstorageracks.Notrendsink,zwithintra-assemblywatergap,withabsorberpanelreactivityworth,withenrichmentorwithpoisonconcentrationwereidentified.A-2

)

AdditionalbenchmarkingcalculationswerealsomadeforaseriesofFrenchcriticalexperiments)at4.754enrichmentandforseveraloftheBNWLcriticalswith4.264enrichedfuel.AnalysisoftheFrenchcriticals(Table2)showedatendencytooverpredictthereactivity,aresultalsoobtainedbyORNL.Thecalculatedk,<valuesshowedatrendtowardhighervalueswithdecreasingcoresize.Intheabsenceofasignificantenrichmenteffect(seeSection3below),thistrendandtheoverpredictionisattributedtoasmallinadequacyinNITAWL-KENO-5aincalculatingneutronleakagefromverysmallassemblies.SimilaroverpredictionwasalsoobservedfortheBR<Lseriesofcriticalexperiments<>,whichalsoaresmallassemblies(althoughsignificantlylargerthantheFrenchcriticals).Inthiscase(Table2),theoverpredictionappearstobesmall,givingameank<<of0.9990+0.0037(1crpopulationstandarddeviation).BecauseofthesmallsizeoftheBNWLcriticalexperimentsandtheabsenceofanysignificantenrichmenteffect,theoverpredictionisalsoattributedtothefailureofNITAWL-KENO-5atoadequatelytreatneutronleakageinverysmallassemblies.Sincetheanalysisofhigh-densityspentfuelstorageracksgenerallydoesnotentailneutronleakage,theobservedinadequacyofNITAWL-KENO-5aisnotsignificant.Furthermore,omittingresultsoftheFrenchandBNWLcriticalexperimentanalysesfromthedeterminationofbiasisconservativesinceanyleakagethatmightenterintotheanalysiswouldtendtoresultinoverpredictionofthereactivity.A-3 I

3.CASMO-3BENCHMARKCALCULATIONSTheCASMO-3codeisamultigrouptransporttheorycodeutilizingtransmissionprobabilitiestoaccomplishtwo-dimensionalcalculationsofreactivityanddepletionforBWRandPWRfuelassemblies.Assuch,CASMO-3iswell-suitedtothecriticalityanalysisofspentfuelstorageracks,sincegeneralpracticeistotreattheracksasaninfinitemediumofstoragecells,neglectingleakageeffects.CASMO-3isamodificationoftheCASMO-2Ecodeandhasbeenextensively'benchmarkedagainstbothmixedoxideandhotandcoldcriticalexperimentsbyStudsvikEnergiteknik).Reportedana-lyses@of37criticalexperimentsindicateameank<<of1.0004+0.0011(le).ToindependentlyconfirmthevalidityofCASHO-3(andtoinvestigateanyeffectofenrichment),aseriesofcalculationsweremadewithCASMO-3andwithNITAWL-KENO-5aonidenticalpoisonedstoragecellsrepresentativeofhigh-densityspentfuelstorageracks.Resultsoftheseintercomparisoncalculations(showninTable3)arewithinthenormalstatisticalvariationofKENOcalculationsandconfirmthebiasof0.0000+0.0024(954/954)forCASM0-3.Sincetwoindependentmethodsofanalysiswouldnotbeexpectedtohavethesameerrorfunctionwithenrichment,resultsoftheintercomparisonanalyses(Table3)indicatethatthereisnosignificanteffectoffuelenrichmentovertherangeofenrich-mentsinvolvedinpowerreactorfuel.Furthermore,neglectingtheFrenchandBNWLcriticalbenchmarkinginthedeterminationofbiasisaconservativeapproach.Intercomparisonbetween,analyticalmethodsisatechniqueendorsedbyReg.Guide5.14,"ValidationofCalculationalMethodsforNuclearCriticalitySafety".A-4 AEI REFERENCESTOAPPENDIXAGreen,Lucious,Petrie,Ford,White,andWright,>>PSR-63-/NITAWL-1(codepackage)NITAWLModularCodeSystemForGeneratingCoupledMultigroupNeutron-GAmmaLibrariesfromENDF/B>>,ORNL-TM-3706,OakRidgeNationalLaboratory,November1975.2.R.M.Westfallet.al.,"SCALE:AModularSystemforPerform-ingStandardizedComputerAnalysisforLicensingEvaluation",NUREG/CR0200'979'~A.Ahlin,M.Edenius,andH.Haggblom,"CASMO-AFuelAssemblyBurnupProgram",AE-RF-76-4158,Studsvikreport.A.AhlinandM.Edenius,>>CASMO-AFastTransportTheoryDepletionCodeforLWRAnalysis",ANSTransactions,Vol.26,p.604,1977.>>CASMO-3AFuelAssemblyBurnupProgram,UsersManual>>,Studsvik/NFA-87/7,StudsvikEnergitechnikAB,November19864.M.N.Baldwinetal.,"CriticalExperimentsSupportingCloseProximityWaterStorageofPowerReactorFuel",BAW-1484-7,TheBabcock&WilcoxCo.,July1979.5.M.EdeniusandA.Ahlin,>>CASMO-3:NewFeatures,Benchmarking,andAdvancedApplications",NuclearScienceandEnineerin100/342-351/(1988)6.M.G.Natrella,EerimentalStatistics,NationalBureauofStandards,Handbook91,August1963.7~8.R.W.WestfallandJ.H.Knight,"SCALESystemCross-sectionValidationwithShipping-caskCriticalExperiments",QSTransactions,Vol.33,p.368,November1979S.E.TurnerandM.K.Gurley,>>EvaluationofNITAWL-KENOBenchmarkCalculationsforHighDensitySpentFuelStorageRacks",ucleaScienceandEnineerinp80(2)230237,February1982.A-5 I

9.J.C.Manaranche,et.al.,"DissolutionandStorageExgerimentwith4.754U-235EnrichedUOzRods",NuclearTechnolo,Vol.50,pp148,September198010.S.R.Bierman,et.al.,"CriticalSeparationbetweenSub-criticalClustersof4.29Wt.4UEnrichedUO>RodsinWaterwithFixedNeutronPoisons",BatellePacificNorthwestLabora-tories,NUREG/CR/0073,May1978(withAugust1979errata).11.A.M.Hathout,et.al.,"ValidationofThreeCross-sectionLibrariesUsedwiththeSCALESystemforCriticalityAnaly-sis",OakRidgeNationalLaboratory,NUREG/CR-1917,1981.A-6 Il Table1RESULTSOF27-GROUP(SCALE)NZTAWL-KENO-5aCALCULATIONSOFB&WCRITICALEXPERIMENTSExperimentNumberCalculatedkerrZZI"'XXIXZZZIZZXIVXZXMeanBiasBias(954/954)0.99320.99150.99160.99180.99230.99190.99610.99600.98170.9843,0.99120.98660.99040.98610.99340.98740'9100.00900.0090+0.00161+0.0015+0.0013+0.0014+0.0015+0.0014+0.0015+0.0015+0.0015+0.0014+0.0015+0.0013+0.0014+0.0013+0.0013+0.0014+0.0014<'>+0.0033@+0.0021Calculatedfromindividualstandarddeviations.Calculatedfromk,<valuesandusedasreference.A-7 II Table2RESULTSOF27-GROUP(SCALE)NITAWL-KENO-5aCALCULATIONSOFFRENCHandBNWLCRITICALEXPERIMENTSSeparationDistance,cmFrenchExperimentsCriticalHeight,'cmCalculatedk,g02.55.010.023.824.4831.4764.341.0231+0.00361.0252+0.00431.0073+0.00130.9944+0.0014CaseBNWLExperimentsExpt.No.CalculatedkeaNoAbsorberSSPlates(1.05B)SSPlates(1.62B)SSPlates(1.62B)SSPlatesSSPlatesZrPlates004/0320090110120130300.9964+0.00340.9988+0.00381.0032+0.00330.9986+0.00360.9980+0.00380.9936+0.00361.0044+0.0035Mean0.9990+0.0037A-8 Il Table3RESULTSOFCASMO-3ANDNITAWL-KENO-5aBENCEBGQK(INTERCOMPARISON)CALCULATIONSEnrichment+Wt.cU-235NITAWL-KENO-5a~CASMO-3iSkl2.53.03.54.04.55.00.8385+0.00160.8808+0.00160.9074+0.00160.9311+0.00160.9546+0.00180'743+0.00180.90900.93460.95590.00160.00350.00130.97410.0002Mean0.00170.83790.00060.87760.0032Infinitearrayofassembliestypicalofhigh-densityspent:fuelstorageracks.kfromNITAWL-KENO-5acorrectedforbiasof0.0090Sk.A-9 IIIIl 5.0THERMAL-HYDRAULICCONSIDERATIONS5.1IntroductionAprimaryobjectiveinthedesignofthehighdensityspentfuelstorageracksfortheDonaldC.Cookspentfuelpoolistoensureadequatecoolingofthefuelassemblycladding.Inthefollowingsectionabriefsynopsisofthedesignbasis,themethodofanalysis,andthenumericalresultsisprovided.Similarmethodsofthermal-hydraulicanalysishavebeenusedinpreviouslicensingeffortsonhighdensityspentfuelracksforFermi2(Docket50-341),QuadCities1and2(Dockets50-254and50-265)gRanchoSeco(Docket50-312),GrandGulfUnit1(Docket50-416),OysterCreek(Docket50-219),VirgilC.Summer(Docket50-395),DiabloCanyon1and2(DocketNos.50-275and50-323),ByronUnits1and2(Docket50-454,455),St.LucieUnitOne(Docket50-335),MillstonePointI(50-245),Vogtle.Unit2(50-425),KuoshengUnits1&2(TaiwanPowerCompany),UlchinUnit2(KoreaElectricPowerCompany),J.A.FitzPatrick(NewYorkPowerAuthority)andTMIUnit1(GPUNuclear).wedoutfor-thethermal-hydraulicarraymaybebrokendownintotheandpoolbulkTheanalysestobecarr'ualificationoftherackfollowingcategories:(i)Pooldecayheatevaluationtemperaturevariationwithtime.(ii)Determinationofthemaximumpoollocaltemperatureattheinstantwhenthebulktemperaturereachesitsmaximumvalue.5-1 IIIII (iii)Evaluationofthemaximumfuelcladdingtemperaturetoestablishthatbulknucleateboilingatanylocationresultingintwophaseconditionsenvironmentaroundthefueldoesnotoccur.(iv)Evaluationofthetime-to-boilifallheatrejectionpathsthroughthecoolingandcleanuparelost.(v)Computetheeffectofablockedfuelcellopeningonthelocalwaterandmaximumcladdingtemperature.Thefollowingsectionspresentasynopsisofthemethodsemployedtoperformsuchanalysesandafinalsummaryoftheresults.5.2SentFuelCoolinSstemDescritionTheprincipalfunctionsoftheSpentFuelCoolingSystemaretheremovalofdecayheatfromthespentfuelstoredinthepoolitservesandmaintainingtheclarityof,andalowactivitylevelin,thewaterofthepool.Cleanup,ofpoolwaterisaccomplishedbydivertingpartoftheflow,maintainedforremovalofdecayheat,throughfiltersand/ordemineralizersasdescribedinSection9.4ofUFSAR.52.1SstemFunctionsTheSpentFuelPoolCoolingSystemisspentfuelpooltheheatgeneratedbyThesystemservesthespentfuelpooltwounits.designedtoremovefromthestoredspentfuelelements.whichissharedbetweenthe5-2 IIIII Thesystemdesignincorporatestwoseparatecoolingtrains.Systempipingisarrangedso,thatfailureofanypipelinedoesnotdrainthespentfuelpoolbelowthetopofthestoredfuelelements.5.2.2SstemDescritionEachofthetwocoolingloopsintheSpentFuelPoolCoolingSystemconsistsofapump,heatexchanger,strainer,piping,associatedvalvesandinstrumentation.Thepumpdrawswaterfromthepool,circulatesitthroughtheheatexchangerandreturnsittothepool.Componentcoolingwatercoolstheheatexchanger.Theclarityandpurityofthespentfuelpoolwaterismaintainedbypassingapproximately100gpmofthecoolingflowthroughafilteranddemineralizer.Skimmersareprovidedtopreventdustanddebrisfromaccumulatingonthesurfaceofthewater.Therefuelingwaterpurificationpumpandfiltercanbeusedseparatelyorinconjunctionwiththespentfuelpooldemineralizertoregainrefuelingwaterclarityafteracrudburstineitherunit.Thiscanpreventlossoftimeduringrefuelingduetopoorvisibility.ThesystemisalsousedtomaintainwaterqualityintheRefuelingWaterStorageTanksofbothunits.Thespentfuelpoolfilter/demineralizerisdownstreamofthespentfuelpoolcooler.Asaresult,thepoolpurificationcomponentsaresubjectedtowatertemperatureswhichcorrespondtothecooleroutlets(lessthan140'F).Allelementsofthepurificationsystem,includingtheresins,arequalifiedfor200Fdesigntemperature.5-3 IIIIII Thespentfuelpoolpumpsuctionlinespenetratethespentfuelpoolwallabovethefuelassembliesstoredinthepooltopreventlossofwaterasaresultofasuctionlinerupture.Thepoolisinitiallyfilledwithwateratthesameboronconcentration(2400ppm)asintherefuelingwaterstoragetank.Thespentfuelpoolislocatedoutsidethereactorcontainment.Duringrefuelingthewaterinthepoolcanbeisolatedfromthatinthere-fuelingcanalbyaweirgatesothatthereisonlyaverysmallamountofinterchangeofwaterasfuelassembliesaretransferred.5.2.3PerformanceReuirementsThefirstdesignbasisofthesystemisbasedonthenormalrefuelingoperationwithanormalbatchof80assembliesbeingremovedfromtheuniteachtime.Theseconddesignbasisforthesystemconsidersthatitispossibletounloadthereactorvessel(193fuelassemblies)formaintenanceorinspectionatatimewhenamaximumof3518spentfuelassembliesareassumedalreadystoredinthespentfuelpool.5.3DecaHeatLoadCalculationsThedecayheatloadcalculationisperformedinaccordancewiththeprovisionsofUSNRCBranchTechnicalPositionASB9-2,"ResidualDecayEnergyforLightWaterReactorsforLongTermCooling",Rev.2,July,1981.Forpurposesofthislicensing IIII application,itisassumedthatthepoolcontainsaninventoryaccumulatedthroughscheduleddischargesfrom1975to2009(Table1.1.1).Further,sincethedecayheatloadincreasesmonotonicallywithreactorexposuretime,anupperboundof1260fullpoweroperationdays(approximately3.5years)isassumedforallstoredfuel.Thecumulativedecayheatloadiscomputedfortheinstanceofhypotheticalnormaldischarge(21BinTable1.1.1)intheyear2009.AsshowninTable5.4.1,theratioofthisdecayheatloadduetotheinventoryofpreviouslystoredfueltotheaverageassemblyoperatingpowerPis0.3303.Thisdecayheatloadisassumedtoremaininvariantforthedurationofthepooltemperatureevaluationsperformedinthewakeofnormalandfullcoreoffloadsdiscussedbelow.5.4DischareScenariosThefollowingdischargescenariosareexamined:Case1:NormaldischargeAnormalbatchof80assemblieswith1260daysofreactorexposuretimeatfullpowerisdischargedinthepoolattheendofanormal18monthoperatingcycle.Thereare43previouslydischargedbatchesinthepool.Asdescribedlater,thenormaldischargeisassumedtooccurattherateofapproximately4assembliesperhourafter168hoursofdecayinthereactor.Onefuelpoolcoolingtrainisactiveandrunning.Onecoolingtraincontainsoneheatexchangerandonefuelpoolpump.Thiscaseisrunwiththedesignfuelpoolwaterflowrate(2300gpm)andactualavailableflowrate(2800gpm).ThesetwocasesarelabelledasCaselaandCase1b,respectively.Case2:NormaldischargeSameasCase1excepttwofuelpoolcoolingtrainsareoperating.5-5 IIIII Case3:Back-to-BackFullcoreoffloadThefullcoreoffloadconditioncorrespondstotheemergencyreactoroffloadconditionwhereintheshutdownofareactoroccurs30daysaftertheotherreactorshutdownforanormalrefueling.Twocoolingtrainsareassumedtobeoperatinginparallelaftertheshutdown.Thedecaytimeofthecoreinthereactorandtherateofdischargetothepoolare.thesameasinCasel.Case4:SameasCase3exceptonlyonecoolingtrainThiscaseislistedforreferenceonly;itbasiscasebytheDonaldC.CookTechnicaltheUSNRCguidelines(NUREG-0800).isinoperation.isnotadesignSpecificationorDetaileddataonthethreecasesaregiveninTable5.4.1to5.4.3.5.5BulkPoolTemeratureInordertoperformtheanalysisconservatively,theheatexchangersareassumedtobefouledtotheirdesignmaximum.Thus,thetemperatureeffectiveness,p,fortheheatexchangerutilizedintheanalysisisthelowestpostulatedvaluecalculatedfromheatexchangerthermalhydrauliccodes.pisassumedconstantinthecalculation.ThemathematicalformulationcanbeexplainedwithreferencetothesimplifiedheatexchangeralignmentofFigure5.5.1.Referringtothespentfuelpool/coolersystem,thegoverningdifferentialequationcanbewrittenbyutilizingconservationofenergy:dT=QL-QHx(5-1)QL=Pcons+Q(r)-QEV(T,t'a)5-6 IIIIIII where:C:Qz,-Thermalcapacitanceofthepool(netwatervolumetimeswaterdensityandtimesheatcapacity),Btu/'F.Heatloadtotheheatexchanger,Btu/hr.Heatgenerationratefromrecentlydischargedfuel,whichisaspecifiedfunctionoftime,z,Btu/hr.Pcons=PPo'eatgenerationratefrom"old"fuel,Btu/hr.(Po=averageassemblyoperatingpower,Btu/hr.)QHX:HeatremovalratebytheheatexchangerIBtu/hr.QEy(T(ta)'eatlosstothesurroundings,whichisafunctionofpooltemperatureTandambienttemperatureta,Btu/hr.QHXisanon-linearfunctionoftimeifweassumethetemperatureeffectivenesspisconstantduringthecalculation.QHXcan,however,bewrittenintermsofeffectivenesspasfollows:QHX+tCtp(T-ti)(5-2)where:Wt.Coolantflowrate,lb./hr.Ct.Coolantspecificheat,Btu/lb.'F.p:Temperatureeffectivenessofheatexchanger.5-7 IIII T:Poolwatertemperature,'Fti:Coolantinlettemperature,'Fto.CoolantoutlettemperatureIF.pisobtainedbyratingtheheatexchangeronaHoltecproprietarythermal/hydrauliccomputercode.Q(r)isspecifiedaccordingtotheprovisionsof"USNRCBranchTechnicalPositionASB9-2,"ResidualDecayEnergyforLightWaterReactorsforLongTermCooling",Rev.2,July,1981.Q(r)isafunctionofdecaytime,numberofassemblies,andin-coreexposuretime.Duringthefueltransfer,theheatloadinthepoolwillincreasewithrespecttotherateoffueltransferandequalstoQ(r)afterthefueltransfer.QEVisanon-linearfunctionofpooltemperatureandambienttemperature.QEVcontainstheheatevaporationlossthroughthepoolsurface,naturalconvectionfromthepoolsurfaceandheatconductionthroughthepoolwallsandslab.Experimentsshowthattheheatconductiontakesonlyabout4%ofthetotalheatlossI'5.5.1],therefore,canbeneglected.Theevaporationheatandnatureconvectionheatlosscanbeexpressedas:QEVm1's+hCAs8(5-~)where:m:As'assevaporationrateIlb/hrft.Latentheatofpoolwater,Btu/lb.Poolsurfacearea,ft..2hc-Convectionheattransfercoefficientatpoolsurface,Btu/ft.,hr.'F5-8

-I~~

8=T-ta.Thetemperaturedifferencebetweenpoolwaterandambientair,'FThemassevaporationratemcanbeobtainedasanon-linearfunctionof8.We,therefore,havem=hD(8)(Wps-Was)(5-4)where:Wps~HumidityratioofsaturatedmoistairatpoolwatersurfacetemperatureT.Was.hD(8):HumidityratioofsaturatedmoistairatambienttemperaturetaDiffusioncoefficientatpoolwatersurface.hDisanon-linearfunctionof8,lb./hr.ft.'FThenon-linearsingleorderdifferentialequation(5-1)issolvedusingHoltec'sQ.A.validatednumericalintegrationcode"ONEPOOL".Figures5.5.2through5.5.6providethebulkpooltemperatureprofilesforthenormaldischarge,andfullcoreoffloadscenariosrespectively.Table5.5.1givesthepeakwatertemperature,'Icoincidenttime,andcoincidentheatloadtothecoolerandcoincidentheatlosstotheambientforthreecases.Thenextstepintheanalysisistodeterminethetemperatureriseprofileofthepoolwaterifallforcedindirectcoolingmodesaresuddenlylost.Make-upwaterisprovidedwithafirehose.Clearly,themostcriticalinstantofloss-of-coolingiswhenpoolwaterhasreacheditsmaximumvalue.EtisassumedthatcoolingwaterisaddedthroughafirehoseattherateofGlb./hr.The5-9 I~~~~~~~LI coolingwaterisattemperature,tcool.The.governingenthalpybalanceequationforthisconditioncanbewrittenasdT[C+G(Ct)(zzo)]=Pcons+Q(~+rins)+G(Ct)(tcool-T)d'r-')EVwherewaterisassumed.tohavespecificheatofunity,andthetimecoordinatezismeasuredfromtheinstantmaximumpoolwatertemperatureisreached.-zoisthetimecoordinatewhenthedirectaddition(firehose)coolingwaterapplicationisbegun.tinsisthetimecoordinatemeasuredfromtheinstantofreactorshutdowntowhenmaximumpoolwatertemperatureisreached.Tisthedependentvariable(poolwatertemperature).Forconservatism,QE~isassumedtoremainconstantafterpoolwatertemperaturereachesandrisesabove170'F.AQ.A.validated.numericalquadraturecodeisusedtointegratetheforegoingequation.Thepoolwaterheatuprate,time-to-boil,andsubsequentwaterevaporation-timeprofilearegeneratedandcompiledforsafetyevaluation.Assumingnomake-upwater(G=0),thetime-to-boiloutputresultsarepresentedinTable5.5.2.Figures5.5.6through5.5.10showtheplotoftheinventoryofwaterinthepoolafterloss-of-coolant-to-the-poolconditionbegins.ZtisseenfromTable5.5.2thatsufficienttimetointroducemanualcoolingmeasuresexistsandtheavailabletimeisconsistentwithotherPWRreactorinstallations.5-10 I1~i~

5.6LocalPoolWaterTemeratureZnthissection,asummaryofthemethodology,calculationsandresultsforlocalpoolwatertemperatureispresented.5.6.1BasisZnordertodetermineanupperboundonthemaximumfuelcladdingtemperature,aseriesofconservativeassumptionsaremade.Themostimportantassumptionsarelistedbelow:0Thefuelpoolwillcontainspentfuelwithvaryingtime-after-shutdown(rs).Sincetheheatemissionfallsoffrapidlywithincreasingrs,itisconservativetoassumethatallfuelassembliesarefromthelatestbatchdischargedsimultaneouslyintheshortestpossibletimeandtheyallhavehadthemaximumpostulatedyearsofoperatingtimeinthereactor.Theheatemissionrateofeachfuelassemblyisassumedtobeequaland'aximum.0Asshowninthepoollayoutdrawings,themodulesoccupyanirregularfloorspaceinthepool.Forthe'ydrothermalanalysis,acirclecircumscribingtheactualrackfloorspaceisdrawn(Fig.5.6.1).Ztisfurtherassumedthatthecylinderwiththiscircleasitsbaseispackedwithfuelassembliesatthenominallayoutpitch.0Theactualdowncomerspacearoundtherackmodulegroupvaries.Thenominaldowncomergapavailableinthepoolisassumedtobethetotalgapavailablearoundtheidealizedcylindricalrack;thus,themaximumresistancetodownwardflowisincorporatedintotheanalysis(Figs.5.6.2and5.6.3)(i.e.minimumgapbetweenthepoolwallandrackmodule,includingseismickinematiceffect).0Nodowncomerflowisassumedtoexistbetweentherackmodules.5-11

~~~I~~~~I'~~I 0TheANF17x17fuelassemblyhasbeenusedintheanalysiswhich,fromthethermal-hydraulicstandpoint,boundsalltypesoffuelbundlesutilizedintheDonaldC.Cookreactor.0Noheattransferisassumedtooccurbetweenpoolwaterandthesurroundings(wall,etc.)5.6.2ModelDescritionXnthismanner,aconservativeidealizedmodelfortherackassemblageisobtained.Thewaterflowisaxisymmetricabouttheverticalaxisofthecircularrackassemblage,andthus,theflowistwo-dimensional(axisymmetricthree-dimensional).Fig.5.6.2showsatypical"flowchimney"renderingofthethermalhydraulicsmodel.Thegoverningequationtocharacterizetheflowfieldinthepoolcannowbewritten.Theresultingintegralequationcanbesolvedforthelowerplenumvelocityfield(intheradialdirection)andaxialvelocity(in-cellvelocityfield),byusingthemethodofcollocation.Thehydrodynamiclosscoefficientswhichenterintotheformulationoftheintegralequationarealsotakenfromwell-recognizedsources(Ref.5.6.1)andwhereverdiscrepanciesinreportedvaluesexist,theconservativevaluesareconsistentlyused.Reference5.6.2givesthedetailsofmathematicalanalysisusedinthissolutionprocess.Aftertheaxialvelocityfieldisevaluated,itisastraight-forwardmattertocomputethefuelassemblycladdingtemperature.Theknowledgeoftheoverallflowfieldenablespinpointingofthestoragelocationwiththeminimumaxialflow(i.e,maximumwateroutlettemperatures).Thisiscalledthemost"choked"location.Znordertofindanupperboundonthetemperatureinatypicalcell,it,isassumedthatitislocatedatthemostchokedlocation.Knowingtheglobalplenumvelocityfield,therevised5-12

axialflowthroughthischokedcellcanbecalculatedbysolvingtheBernoulli'sequationfortheflowcircuitthroughthiscell.Thus,anabsoluteupperboundonthewaterexittemperatureandmaximumfuelcladdingtemperatureisobtained.Inviewoftheseaforementionedassumptions,thetemperaturescalculatedinthismanneroverestimatethetemperaturerisethatwillactuallyoccurinthepool.,Holtec'scomputercodeTHERPOOL*,basedonthetheoryofRef.5.6.2,automatesthiscalculation.TheanalysisrprocedureembodiedinTHERPOOLhasbeenacceptedbytheNuclearRegulatoryCommissiononseveraldockets.TheCodeTHERPOOLforlocaltemperatureanalysesincludesthecalculationofvoidgenerations.Theeffectofvoidontheconservationequation,crudlayerintheclad,fluxtraptemperatureduetogammaheating,andthecladstresscalculationwhenavoidexists,areallincorporatedinTHERPOOL.ThepeakingfactorsaregiveninTable5.6.1.5.7CladdinTemeratureThemaximumspecificpowerofafuelarray~canbegivenby:M=qFxywhere:Fxy=radialpeakingfactorq=averagefuelassemblyspecificpower*THERPOOLhasbeenusedinqualifyingthespentfuelpoolsforEnricoFermiUnit2(1980),QuadCitiesIandII(1981)gOysterCreek(1984),V.C.Summer(1984),RanchoSeco(1983)fGrandGulfI(1985),DiabloCanyonIandII(1986),amongothers.5-13 1l Themaximumtemperatureriseofpoolwaterinthemostdisadvantageouslyplacedfuelassemblyiscomputedforallloadingcases.Havingdeterminedthemaximumlocalwatertemperatureinthepool,itisnowpossibletodeterminethemaximumfuelcladdingtemperature.AfuelrodcanproduceFztimestheaverageheatemissionrateoverasmalllength,whereFzistheaxialrodpeakingfactor.Theaxialheatdistributioninarodisgenerallyamaximuminthecentralregion,andtapersoffatitstwoextremities.Ztcanbeshownthatthepowerdistributioncorrespondingtothechoppedcosinepoweremissionrateisgivenbyq(x)=gAsinn(a+x)h+2awhere:h:activefuellengtha:choppedlengthatbothextremitiesinthepowercurvex:axialcoordinatewithoriginatthebottomoftheactivefuelregionThevalueofaisgivenbywhere:hza1-2z1/2mFz112++2F2nFzz25-14 l

whereFzistheaxialpeakingfactor.)ThecladdingtemperatureTcisgovernedbydifferentialequationwhichhastheformofathirdorderd3Td2TdT+al--a2-=f(x)dxdx2dxwhereal,a2andf(x)arefunctionsofx,andfuelassemblygeometricproperties.Thesolutionofthisdifferentialequationwithappropriateboundaryconditionsprovidesthefuelcladdingtemperatureandlocalwatertemperatureprofile.Inordertointroducesomeadditionalconservatismintheanalysis,weassumethatthefuelcladdinghasacruddepositwhichresultsin.005F-sq.ft.-hr/Btuofcrudresistance,whichcoverstheentiresurface.Table5.6.2providesthe.keyinputdataforlocaltemperatureanalysis.TheresultsofmaximumlocalpoolwatertemperatureandminimumlocalfuelcladdingtemperaturearepresentedinTable5.7.1.Thelocalboilingtemperatureofwaterisapproximately242'Fat26'elowthefreewatersurfaceandhigheratlowerelevations.Thelocationwherethelocalwatertemperaturereachesitsmaximumvalueisdeeperthan26'elowthefreewatersurface,wherethecoincidentboilingtemperatureofwaterisgreaterthan242'P.Itisshownthatthelocalpoolwatertemperatureislowerthanthelocalboilingpointandtherefore,nucleateboilingwillnotoccurs5-15

Finally,itisnotedthatthefuelcladdingconsiderablylowerthanthetemperaturestowhichsubjectedinsidethereactor.Therefore,itisthereissufficientmarginagainstfuelcladdingspentfuelpool.temperatureisthecladdingisconcludedthatfailureinthe5.8BlockedCellAnalsisCalculationsarealsoperformedassumingthat50%ofthetopopeninginthethermallylimitingstoragecellisblockedduetoahorizontallyplaced(misplaced)fuelassembly.ThecorrespondingmaximumlocalpoolwatertemperatureandlocalfuelcladdingtemperaturedataarealsopresentedinTable5.7.1.Thereisalsonoincidenceoflocalizednucleateboilingofthepoolwaterorpotentialforfuelcladdingdamage.5.9ReferencesforSection55.6.1GeneralElectricCorporation,R&DDataBooks,"HeatTransferandFluidFlow",1974andupdates.5.6.2Singh,K.P.etal.,"MethodforComputingtheMaximumWaterTemperatureinaFuelPoolContainingSpentNuclearFuel",HeatTransferEngineering,Vol.7,No.1-2Ipp.72-82(1986)~5-16 IlI Table5.4.1=-FUELSPECIFICPOWERANDPOOLCAPACITYDATATotalwatervolumeofPool:SpecificOperatingPowerofaFuelAssembly:Dimensionlessdecaypowerof"old"discharges:635645gallons60.3E+06Btu/hr.0.33035-17

Table5.4.2DATAFORSCENARXOS1through4CASENO.IaIbPoolthermalcapacityCxl0,Btu/F4.2414.2414.2414.2414.241No.ofCoolingTrainsNo.ofDischargesconsideredfortheAnalysisTimebetweenshutdowns,hr.720720CoolerInletTemp.,OF108.4108.4103.9101.4104'coolantFlowRate/cooler,10lb./hr.1.49149l.491.49l.49FuelPoolWaterFlowRate,10lb./hr.l.141.40l.14l.14l.14TemperatureEffectiveness/cooler,p0.39700430.39750.3979'.39875-18 II Table5.4.3DATAFORSCENARIOS1THROUGH4CaseDischergeNo.ofNo.lDAssembliesTimeAftershutdownwhenTransferBegins(hrs)OffloadExpo.T~eT'me(hrs)(hrs)laor1bDischarge18016819.0730240Discharge1Discharge13(orDischarge24(Fullcore)80808011316816816819.073024019.073024046.0010080302405-19 IIII Table5.5.1POOLBULKTEMPERATUREANDHEATLOADDATACaseNo.CoincidentCoolerDuty106Btu/hr.TmaxMax.PoolBulkTemp.,OpTimeCoincidento(afterreactorshutdown)CoincidentEvaporationHeptLoss,10~Btu/hr.lalb30.24130.6932.78750.69045.04159.54156'1131.57143'4176.912072061982222253.002.5780.6891.3956.8875-20 I

Table5.5.2TIME-TO-BOILFORVARIOUSDISCHARGESCENARIOSCaseNumberTime-to-Boil(hours)G=0GPMla1b7.828.2711.525.743.025-21 IIII Table5.6.1PEAKINGFACTORDATARadialBundlePeakingFactorTotalpeakingfactor1.652.405-22 IIII Table5.6.2DATAFORLOCALTEMPERATURETypeofFuelAssemblyFuelCladdingOuterDiameter,inchesFuelCladdingInsideDiameter,inchesStorageCellinsideDimension,inchesActivefuellength,inchesNo.ofFuelRods/AssemblyOperatingPowerperFuelAssemblyPox6pBtu/hrCellpitch,inchesCellheight,inchesPlenumradius,feetMin.Bottomheight,inchesMin.gapbetweenpoolwallandouterrackperiphery,inches0.360.318.7514426460.38.9716829.34.751.55-23 IIIII Table5.7.1LOCALANDCLADDINGTEMPERATUREOUTPUTDATAFORTHEMAXIMUMPOOLWATERCONDITION(Casea)ConditionWaterTem.'FTem.'FNoblockage50%blockage168.0219.2212.9246.95-24 lIII W,TSPERTFUE'QQL.l,,7Ã,CP~,Tj4~FIGURE5.5.1PoolBulkTemperatureModel5-25 IIII 165163DONALDC.COOKSFPNORMALDISCHARGE,ONECOOLINGTRAIN,CASE1a160158~155IJJ1530150ca14814514013850100150200250300350TIMEAFTERFIRSTSHUTDOWNOFTHEREACTOR,I.IRFIGURE5.5.2400450500 IIII 165163DONALDC.COOKSFPNORMALDISCHARGE,ONECOOLINGTRAIN,CASE1b160158155a.153~150o.Og14814514314013850100150200250300350TIMEAFTERFIRSTSHUTDOWNOFTHEREACTOR,HR400450500FIGURE5.5.3 III 135133DONALDC.COOKSFPNORMALDISCHARGE,TWOCOOLINGTRAINS,CASE2130~128OO12512312011850100150200250300350TIMEAFfERFIRSTSHUTDOWNOFTHEREACTOR,HRFIGURE5.5.4400450500 150145DONALDC.COOKSFPFULLCOREOFFLOAD,TWOCOOLINGTRAINS,CASE3140135LBIUJ13000~125120115110100200300400500600700800900100011001200TIMEAFTERFIRSTSHUTDOWNOFTHEREACTOR.HRFIGURE5.5.5 IIIIIII 185180DONALDC.COOKSFPFULLCOREOFFLOAD,ONECOOLINGTRAIN,CASE4175170165o160~155OOQ.g150145140135130125100200300400500600700800900100011001200TIMEAFTERFIRSTSHUTDOWNOFTHEREACTOR,HRFIGURE5.5.6 II 5.50E+0055.00E+0054,50E+005COOKSFPLOSSOFCOOLINGSCENARIO,CASE1a~4.00E+005cc3.50E+0050E~3.00E+005IL<2.50E+00502,00E+005E.o150E+0050~01.00E+0055.00E+004.O,OOE+00004080120TimeAfterMax,temp.hasbeenreached,hrFIGURE5.5.7160 I

5.50E+0055.00E+0054.50E+005COOKSFPLOSSOFCOOLINGSCENARIO,CASE1bID4.OOE+OO5Cc3.50E+005oE~3.00E+005VlIO2.50E+00502,00E+005o1.50E+0050o1.00E+0055.00E+004O.OOE+00004080120TimeAfterMax.temp.hasbeenreached,hrFIGURE5.5.8160 IIIIIII 5.50E+0055.00E+0054.50E+005COOKSFPLOSSOFCOOLINGSCENARIO,CASE2>4.00E+005c3.50E+0050E~3.00E+005I4J<2.50E+00502.00E+00501.50E+0050~01.00E+0055.00E+004O.OOE+00004080120TimeAfterMax.temp.hasbeenreached,hrFIGURE5.5.9160 IIIII 5.50E+0055.00E+0054.50E+005COOKSFPLOSSOFCOOLINGSCENARIO,CASE3>4.00E+005cc3.50E+0050E~3.00E+005>2.50E+00502.00E+005o1.50E+0050~()1.00E+005I-5.00E+004O.OOE+0000204060TimeAfterMax.temp.hasbeenreached,hrFIGURE5.5.j.o80 III~~I 5.50E+0055.00E+0054.50E+005COOKSFPLOSSOFCOOLINGSCENARIO,CASE4u4.OOE+OO5Cc3.50E+0050ECL3.00E+00502.50E+00502.00E+00501.50E+0050~~1.00E+005I-5.00E+004.O,OOE+0000204060TimeAfterMax.temp.hasbeenreached,hrFIGURE5.5.11.80 II I'I I

OUiP()D000ZI0QV~t~tZP+I~~0HEAiADDtiIGNtN5-37THERMALCHIMNEYFLOWMODELFlGLlRE5~6~2 III i$~~~~.44.4~r~~~0~~~RACK.40~e~~~~e4~g~~rh~~go~~rrd~~~~r~cf~V~~$~mgrrm~~~r'oS~rroawNCOMER5-38CONVECTIONCVRRENTSINTHEPOOLFIGURE5.6.3 IIIIIII 6.0STATICANDDYNAMICANALYSIS0RACKSTRUCTURE6.1IntroductionThepurposeofthissectionistopresentanalyseswhichdemonstratethestructuraladequacyoftheDonaldC.CookspentfuelhighdensityrackdesignundernormalstorageandthepostulatedaccidentloadingconditionsasdefinedbyandfollowingtheguidelinesoftheUSNRCStandardReviewPlan(Ref.6.1.1).Themethodofanalysispresentedusesatime-historyintegrationmethodsimilartothatpreviouslyusedinthelicensingreportsonhighdensityspentfuelracksforEnricoFermiUnit2(USNRCDocketNo.50-341),QuadCities1and2(USNRCDocketNos.50-254and50-265),RanchoSeco(USNRCDocketNo.50-312),GrandGulfUnit1(USNRCDocketNo.50-416),Oyster'Creek(USNRCDocketNo.50-219),V.C.Summer(USNRCDocketNo.50-395),DiabloCanyonUnits1and2(USNRCDocketNos.50-275and50-323),VogtleUnit2(USNRCDocketNo.50-425)andMillstonePointUnit1(USNRCDocketNo.50-245).TheanalysescarriedoutfortheDonaldC.Cookracksareconsiderablymoreelaborateandexhaustiveinscopeandsubstancethanthoseperformedintheaforementioneddockets,andreflectadvancesin3-Dfuelracksimulationtechnologyinthepasttwoyears.Thedetailsarepresentedlaterinthissection,aftertheessentialelementsofthedynamicmodelarefullyexplained.TheresultsshowthatthehighdensityspentfuelracksarestructurallyadequatetoresistthepostulatedstresscombinationsassociatedwithlevelA,B,C,andDconditionsasdefinedinRefs.6.1.1,6.1.2,and6.1.3.6-1 I

6.2AnalsisOutlineTheprincipalstepsinperformingtheseismicanalysisofDonaldC.Cookracksaresummarizedbelow:a0DevelopstatisticallyindependentsynthetictimehistoriesforthreeorthogonaldirectionswhichsatisfyUSNRCSRP3.8.4.Twotimehistoriesareconsideredtobestatisticallyindependentiftheirnormalizedcorrelationcoefficientislessthan0.15.b.Prepareathree-dimensionaldynamicmodelofthefuelrackwhichembodiesallelastostaticcharacteristicsandstructuralnonlinearitiesoftheDonaldC.Cookrackmodules.c~d.e.Performaseriesof3-DdynamicanalysesonalimitingmodulegeometrytypefromthoselistedinTables2.1.1and2'.3andforvaryingphysicalconditions(suchascoefficientoffriction,extent.ofcellscontainingfuelassemblies,andproximityofotherracks).Performstressanalysisforthecriticalcasefromthedynamicanalysisrunsmadeintheforegoingsteps.DemonstratecompliancewithASMECodeSectionIII,sub-sectionNF(Ref.6.1.2)limits.Carryoutadegree-of-freedom(DOF)reductionprocedureonthesinglerack3-DmodelsuchthatthekinematicresponsescalculatedbythereducedDOF(modelRDOFM)areinagreementwiththebaselinemodelofstep(b)above.ThisreducedDOFmodelisalsotrulythree-,dimensional.Prepareawholepoolmulti-rackdynamicmodelbycompilingtheRDOFM'sof~alrackmodulesinthepool,andbyincludingallfluidcouplinginteractionsamongthem,aswellasthosebetweentheracksandpoolwalls.This3-Dmulti-modulesimulationisreferredtoasaWholePoolMulti-Rack(WPMR)model.6-2 lIl goPerforma3-DWholePoolMulti-Rack(WPMR)analysistodemonstratethatallkinematiccriteriaforDonaldC.Cookrackmodulesaresatisfied(seeSection6.8),andthatpedestalcompressiveloadsarecomparabletotheloadsusedforstructuralqualificationperitemdabove.Section6.8givesthecriteriawhichneedtobechecked.FortheDonaldC.Cookracks,theprincipalkinematiccriteriaare(i)noracktopoolwallimpact,and(ii)norack-to-rackimpactinthecellularregionoftheracks.Figure6.2.1showsapictorialviewoftherackmodule.Itisnotedthatthebaseplateextendsbeyondthecellularregion,envelope,thusensuringthattheinter-rackimpact,ifany,wouldfirstoccuratthebaseplateelevation.Thebaseplateoftherackmodulesisstructurallyqualifiabletowithstandlargein-planeimpactloads.Wedescribeeachoftheaboveanalysisstepsinsomedetail,inthefollowingsub-sectionswithspecialemphasisonthebaseline3-Ddynamicmodelwhichisthebuildingblockforallsubsequentanalyses.Wealsopresenttheresultsoftheanalysisintheconcludingsub-section.6.3ArtificialSlabMotionsTheUFSARprovidesasingleresponsespectruminthehorizontaldirectionandasingleresponsespectrumintheverticaldirection(2/3ofthehorizontal)fortheDesignBasisEarthquake(DBE).AcorrespondingpairofspectraareprovidedfortheOperatingBasisEarthquake(OBE).6-3 IIIII Holtec'sQ.A.validatedtimehistorygenerationcodeGENEQ[6.3.1]wasusedtogeneratethreesyntheticstatisticallyindependenttimehistoriesfortheNorth-South,East-Westandverticaldirections,respectively,fromthetworesponsespectra.5%dampingisusedfortheDBEcondition.Figures6.3.1-6.3.3showtheDBEtimehistoryplots.ResponsespectracorrespondingtothesetimehistorieswerealsogeneratedandareshownoverlaidonthedesignspectrainFigures6.3.4-6.3.6.Thenormalizedcorrelationcoefficientspijbetweentimehistoriesiandj(1=N-S,2=E-W,3=vertical)areprovidedinTable6.3.1.TheaboveanalyseswererepeatedfortheOBEspectrausing2%damping.Figures6.3.7-6.3.9presentthetimehistoryplots,andFigures6.3.10-6.3.12showthecomparisonbetweenthedesignspectraandthederivedspectra.Table6.3.1alsoprovidespijfortheOBEtimehistories.Xtisnotedthattheenvelopingrequirementonthederivedspectraandstatisticalnon-coherenceofartificialmotionsareunconditional'lysatisfied.6-4 III 6.4OutlineofSineRack3-DAnalsisThespentfuelstorageracksareSeismicClassIequipment.TheyarerequiredtoremainfunctionalduringandafteraDesignBasisEarthquake(Ref.6.1.3).Theseracksareneitheranchoredtothepoolfloornorattachedtothesidewalls.Theindividualrackmodulesarenotinterconnected.Furthermore,aparticularrackmaybecompletelyloadedwithfuelassemblies(whichcorrespondstogreatestrackinertia),oritmaybecompletelyempty.Thecoefficientoffriction,p,betweenthesupportsandpoolfloorisanotherindeterminatefactor.AccordingtoRabinowicz(Ref.6.4.1),theresultsof199testsperformedonausteniticstainlesssteelplatessubmergedinwatershowameanvalueofptobe0.503withastandarddeviationof0.125.Theupperandlowerbounds(basedontwicethestandarddeviation)arethus0.753and0.253,respectively.Analysesarethereforeperformedforsingleracksimulationsusingvaluesofthecoefficientoffrictionequalto0.2(lowerlimit)and0.8(upperlimit),respectively.Theboundingvaluesofp~0.2and0.8havebeenfoundtobrackettheupperlimitofthemoduleresponseinpreviousrerackprojects.Asinglerack3-Danalysisrequiresanotherkeymodellingassumption.Thisrelatestothelocationandrelativemotionofneighboringracks.Thegapbetweenaperipheralrackandanadjacentpoolwallisknown,andthemotionofthepoolwallisprescribedwithoutanyambiguity.However,anotherrackadjacenttotherackbeinganalyzedisalsofree-standingandsubjecttomotionduringaseismicevent.Toconducttheseismicanalysisofagivenrackitsphysicalinterfacewithneighboringmodulesmustbespecified.Thestandardprocedureinthesinglerackanalysisistospecifythattheneighboring.racksmove180'ut-of-phasein6-5 II4II relationtothesubjectrack.Thus,theavailablegapbeforeinter-rackimpactoccursisonehalfofthephysicalgap.This"opposedphasemotion"assumptionincreasesthelikelihoodofpredictingintra-rackimpactsandisthusaconservativeassumption.However,italsoincreasestherelativecontributionoffluidcouplingterms,whichdependonfluidgapsandrelativemovementsofbodies,makingtheoutrightconservatismalesscertainassertion.Infact,3-DWholePoolMulti-rackanalysescarriedoutforTaiwanPowerCompany'sChinShanStation,andforGPUNuclear'sOysterCreekNuclearStationshowthatthesingleracksimulationspredictsmallerrackdisplacementduringseismicresponses.Nevertheless,singlerackanalysespermitdetailedevaluationofstressfields,andserveasabenchmarkcheckforthemuchmoreinvolved,WPMRanalysisresults.Inordertopredictthelimitingconditionsofrackmoduleseismicresponsewithintheframeworkofsinglerackanalysis,moduleA4(13x14)isanalyzed.Thisistypicalofthelargestmodule,andisalsoacornermodule.Thecornermodulehaslargerrack-to-wallgapswhichwillminimizethefluidcoupling.Therackisconsideredfullyloadedorhalfloaded,withlimitingcoefficientsoffriction;thesesimulationsidentifytheworstcaseresponseforrackmovementandforrackstructuralintegrity.Aftercompletionofreracking,thegapsbetweentherackmodulesIandthosebetweentheracksandwallswillbeinthemannerofFigure2.1.1.Weshowinthisreportthatallsinglerack3-Dsimulationspredictthatnorack-to-rackorrack-to-wallimpactswilloccurinthecellularregionoftheracks.Theseismicanalyseswereperformedutilizingthetime-historymethod.Poolslabaccelerationdatapresentedintheprecedingsub-sectionwasused.6-6 IlWSIII Theobjectiveoftheseismicanalysisofsingleracksistodeterminethestructuralresponse(stresses,deformation,rigidbodymotion,etc.)duetosimultaneousapplicationofthethreestatisticallyindependent,orthogonalseismicexcitations.Thus,recoursetoapproximatestatisticalsummationtechniquessuchasthe"Square-Root-of-the-Sum-of-the-Squares"method(Ref.6.4-2)isavoided.Fornonlinearanalysis,theonlypracticalmethodissimultaneousapplicationoftheseismicloadingtoanonlinearmodelofthestructure.Theseismicanalysisofasinglerackisperformedinthreesteps,namely:Developmentofanonlineardynamicmodelconsistingofinertialmasselements,spring,gap,andfrictionelements.2~3~Generationoftheequationsofmotionandinertialcouplingandsolutionoftheequationsusingthe"componentelementmethod"(Refs.6.4.3and6.4.4)todeterminenodalforcesanddisplacements.TheHolteccomputercodeDYNARACKisusedtosolvethesystemofequations[6.4.5].Computationofthedetailedstressfieldintherackjustabovethebaseplateandinthesupportlegsismadeusingthenodalforcescalculatedinthepreviousstep.Thesestressesarecheckedagainstthedesignlimitsgiveninalatersub-section.Abriefdescriptionofthedynamicmodelfollows.6.5DnamicModeforTheS'RacAnals'sSincetheracksarenotanchoredtothepoolslaborattachedtothepoolwallsortoeachother,theycanexecuteawidevarietyofmotions.Forexample,therackmayslideonthepoolfloor6-7 IIIIIII (so-called"slidingcondition");one'rmorelegsmaymomentarilylosecontactwiththeliner("tippingcondition");ortherackmayexperienceacombinationofslidingandtippingconditions.Thestructuralmodelshouldpermitsimulationofthesekinematiceventswithinherentbuilt-inconservatisms.Sincethemodulesaredesignedtoprecludetheincidenceofinter-rackimpactinthecellularregion,itisalsonecessarytoincludethepotentialforinter-rackimpactphenomenaintheanalysistodemonstratethatsuchimpactsdonotoccur.Lift-offofthesupportlegsandsubsequentlinerimpactsmustbemodelledusingappropriateimpact(gap)elements,andCoulombfrictionbetweentherackandthepoollinermustbesimulatedbyappropriatepiecewiselinearsprings.The'elasticityoftherackstructure,relativetothebase,mustalsobeincludedinthemodeleventhoughtherackmaybenearlyrigid.Thesespecialattributesofrackdynamicsrequireastrongemphasisonthemodelingofthelinearandnonlinearsprings,dampers,andcompressiononlystopelements.Thetermnon-linearspringisthegenerictermtodenotethemathematicalelementrepresentingthesituationwheretherestoringforceexertedbytheelementisnotlinearlyproportionaltothedisplacement.InthefuelracksimulationtheCoulombfrictioninterfacebetweentheracksupportlegandthelinerisatypicalexampleofanon-linearspring.Themodeloutlineintheremainderofthissub-section,andthemodeldescriptioninthefollowingsub-section,describethe*detailedmodelingtechniquetosimulatetheseeffects,withconsiderableemphasisplacedonthenonlinearityoftherackseismicresponse.6-8 IIIIIII 6.5.1ssumt'onsa."Thefuelrackstructureisafoldedmetalplateassemblageweldedtoabaseplateandsupportedonfourlegs.Anodd-shapedmodulemayhavemorethanfourlegs.Therackstructureitselfisaveryrigidstructure.Dynamicanalysisoftypicalmulti-cellrackshasshownthatthemotionofthestructureiscapturedalmostcompletelybymodellingtherackasatwelvedegree-of-freedomstructure,wherethemovementoftherackcross-sectionatanyheightisdescribedintermsofsixdegrees-of-freedomoftherackbaseandsixdegreesoffreedomdefinedattheracktop.TherattlingfuelismodelledbyfivelumpedmasseslocatedatH,.75H,.5Hg.25H,andattherackbase,whereHistherackheightasmeasuredfromthebase.b.Theseismicmotionofafuelrackischaracterizedbyrandomrattlingoffuelassembliesintheirindividualstoragelocations.Assumingacertainstatisticalcoherence(i.e.assumingthatallfuelelementsmovein-phasewithinarack)inthevibrationofthe.fuelassembliesexaggeratesthecomputeddynamicloadingontherackstructure.Thisassumption,however,greatlyreducestherequireddegrees-of-freedomneededtomodelthefuelassemblieswhicharerepresentedbyfivelumpedmasseslocatedatdifferentlevelsoftherack.Thecentroidofeachfuelassemblymasscanbelocated,relativetotherackstructurecentroidatthatlevel,soastosimulateapartiallyloadedrack.c.Thelocalflexibilityofthepedestalismodelledsoastoaccountforfloorelasticity,andlocalrackelasticityjustabove'hepedestal.d.Therackbasesupportmayslideorlift-offthepoolfloor.e.Thepoolfloorhasaspecifiedtime-historyofseismicaccelerationsalongthethreeorthogonaldirections.if.Fluidcouplingbetweenrackandfuelassemblies,andbetweenrackandwall,issimulatedbyintroducingappropriateinertialcouplingintothesystemkineticenergy.InclusionoftheseeffectsusesthemethodsofRefs.6.5.1and6.5.2forrack/assemblycouplingandforrack/rackcoupling.6-9 g.Potentialimpactsbetweenrackandfuelassembliesareaccountedforbyappropriate"compressiononly"gapelementsbetweenmassesinvolved.h.Fluiddampingduetoviscouseffectsbetweenrackandassemblies,andbetweenrackandadjacentrack,isconservativelyneglected.i.Thesupportsaremodeledas"compressiononly"elementsfortheverticaldirectionandas"rigidlinks"fortransferringhorizontalstress.Thebottomofasupportlegisattachedtoafrictionalspringasdescribedinsub-section6.6.Thecross-sectioninertialpropertiesofthesupportlegsarecomputedandusedinthefinalcomputationstodeterminesupportlegstresses.j.Theeffectofsloshingisnegligibleatthelevelofthetopoftherackandishenceneglected.k.Thepossibleincidenceofrack-to-wallorrack-to-rackimpactissimulatedbygapelementsatthetopandbottomoftherackinthetwohorizontaldirections.Thebottomelementsarelocatedatthebaseplateelevation.1.Rattlingoffuelassembliesinsidethestoragelocationscausesthe"gap"betweenthefuelassembliesandthecellwalltochangefromamaximumoftwicethenominalgaptoatheoreticalzerogap.Fluidcouplingcoefficientsarebasedonthenominalgap.m.Theformdragduetomotionofthefuelassemblyinthestoragecell,orthatduetomovementofarackinthepool,hasbeenneglectedinthisanalysisforaddedconservatism.n.Thefluidcouplingtermsarebasedonopposedphasemotionofadjacentmodules.FigurefreedomFiguresspringsbetween6.5.1showsaschematicofthemodel.Twelvedegreesofareusedtotrackthemotion'ftherackstructure.6.5.2and6.5.3,respectively,showtheinter-rackimpact(totrackthepotentialforimpactbetweenracksorrackandwall)andfuelassembly/storagecellimpact6-10 IIIIII springsataparticularlevel.Si(i=1,4)representsupportlocations,pirepresentabsolutedegrees-of-freedom,andqirepresentdegrees-of-freedomrelativetotheslab.Histheheightoftherackabovethebaseplate.AsshowninFigure6.5.1,themodelforsimulatingfuelassemblymotionincorporatesfiverattlinglumpedmasses.Thefiverattlingmassesarelocatedatthebaseplate,atquarterheight,athalfheight,atthreequarterheight,andatthetopoftherack.Twodegrees-of-freedomareusedtotrackthemotionofeachrattlingmassinthehorizontalplane.Theverticalmotionofeachrattlingmassisassumedtobethesameastherackbase.Figures6.5.4,6.5.'5,and6.5.6showthemodellingschemeforincludingrackelasticityandthedegreesoffreedomassociatedwithrackelasticity.IneachplaneofbendingashearandabendingspringareusedtosimulateelasticeffectsinaccordancewithRef.6.5.1.Table6.6.2givesspringconstantsforthesebendingspringsaswellascorrespondingconstantsforextensionalandtorsionalrackelasticity.6.5.2ModeDescritioTheabsolutedegrees-of-freedomassociatedwitheachofthemasslocationsareidentifiedinFigure6.5.1andinTable6.5.1.Therattlingmasses(nodes1*,2*,3*,4*,5*)aredescribedbytranslationaldegrees-of-freedomq7-q16.Ui(t)isthepoolfloorslabdisplacementseismictime-history.Thus,therearetwenty-twodegreesoffreedominthesystem.NotshowninFig.6.5.1arethegapelementsusedtomodelthesupportlegsandtheimpactswithadjacentracks.6-11 III1,.yIII 6.5.3FluidCoulinAneffectofsomesignificancerequiringcarefulmodelingisthe"fluidcouplingeffect"(Refs.6.5.1and6.5.2).Ifonebodyofmass(m1)vibratesadjacenttoanotherbody(massm2),andbothbodiesaresubmergedinafrictionlessfluidmedium,thenNewton'sequationsofmotionforthetwobodieshavetheform:NN(ml+Mll)Xl+M12X2appliedforcesonmassml+0(xl)M21Xl+(m2+M22)X2=appliedforcesonmassm2+0(x2)X1,X2denoteabsoluteaccelerationsofmassesm1andm2,respectivelyandthenotation0(x)denotesnon-lineartermswhichariseinthederivation.M11~M12M21,andM22arefluidcouplingcoefficientswhichdependontheshapeofthetwobodies,theirrelativedisposition,etc.Fritz(Ref.6.5.2)givesdataforMijforvariousbodyshapesandarrangements.Theaboveequationsindicatethattheeffectofthefluidistoaddacertainamountofmasstothebody(M11tobody1),andanexternalforcewhichisproportionaltotheaccelerationoftheadjacentbody(massm2).Thus,theaccelerationofonebodyaffectstheforcefieldonanother.Thisforceisastrongfunctionoftheinterbodygap,reachinglargevaluesforverysmallgaps.Thisinertialcouplingiscalledfluidcoupling.Ithasanimportanteffectinrackdynamics.Thelateralmotionofafuelassemblyinsidethestoragelocationwillencounterthiseffect.Sowillthemotionofarackadjacenttoanotherrackiftheracksarecloselyspaced.Theseeffectsareincludedintheequationsofmotion.Forexample,thefluid6-12 I

couplingisbetweennodes2and2*inFigure6.5.1.Furthermore,therackequationscontaincouplingtermswhichmodeltheeffectoffluidinthegapsbetweenadjacentracks.Thecouplingtermsmodelingtheeffectsoffluidflowingbetweenadjacentracksarecomputedassumingthatalladjacentracksarevibrating180outofphasefromtherackbeinganalyzed.Therefore,onlyonerackisconsideredsurroundedbyahydrodynamicmasscomputedasiftherewereaplaneofsymmetrylocatedinthemiddleofthegapregion.Finally,fluidvirtualmassisincludedintheverticaldirectionvibrationequationsoftherack;virtualinertiaisalsoaddedtothegoverningequationcorrespondingtotherotationaldegreeoffreedom,q6(t)andq22(t).6.5a4~DaminZnreality,damping(Ref.6.5.3)oftherackmotionarisesfrommaterialhysteresis(materialdamping),relativeintercomponentmotioninstructures(structuraldamping),andfluidviscouseffects(fluiddamping).Zntheanalysis,amaximumof1%structuraldampingisimposedonelementsoftherackstructureduringOBEandDBEsimulations.Materialandfluiddampingduetofluidviscosityareconservativelyneglected.Thedynamicmodelhastheprovisiontoincorporateformdrageffects;however,noformdraghasbeenusedforthisanalysis.6~5.5~ImactAnyfuelassemblynode(e.g.,2*)mayimpactthecorrespondingstructuralmassnode2.Tosimulatethisimpact,fourcompression-onlygapelementsaroundeachrattlingfuelassembly6-13 II LLnodeareprovided(seeFigure6.5.3).Thecompressiveloadsdevelopedinthesespringsprovidethenecessarydatatoevaluatetheintegrityofthecellwallstructureandstoredarrayduringtheseismicevent.Figure6.5.2showsthelocationoftheimpactspringsusedtosimulateanypotentialforinter-rackorrack-to-wallimpacts.Sub-section6.6givesmoredetailsontheseadditionalimpactsprings.Sincetherearefiverattlingmasses,atotalof20impactspringsareusedtomodelfuelassembly-cellwallimpact.6.6AssemblotheDnamicModelThe.cartesiancoordinatesystemassociatedwiththerackhasthefollowingnomenclature:x~Horizontalcoordinatealongtheshortdirectionofrackrectangularplanformy=Horizontalcoordinatealongthelongdirectionoftherackrectangularplanformz=VerticalcoordinateupwardfromtherackbaseLLTable6.6.1listsallspringelementsusedinthe3-Dsinglerackanalysis.Ifthesimulationmodelisrestrictedtotwodimensions(onehorizontalmotionplusverticalmotion,forexample)gortheurosesofodcar'f'cat'oonlthenadescriptivemodelofthesimulatedstructurewhichincludesgapandfrictionelementsisshowninFigure6.6.1.Theimpactsbetweenfuelassembliesandrackshowupinthegapelements,havinglocalstiffnessKI,inFigure6.6.1.InTable6.6.1,gapelements5through8areforthevibratingmassatthe6-14

'topoftherack.ThesupportlegspringratesKSaremodeledbyelements1through4inTable6.6.1.NotethatthelocalcomplianceoftheconcretefloorisincludedinKS.Tosimulateslidingpotential,frictionelements2plus8and4plus6(Table6.6.1)areshowninFigure6.6.1.Thefrictionofthesupport/linerinterfaceismodeledbyapiecewiselinearspringwithasuitablylargestiffnessKfuptothelimitinglateralload,pN,whereNisthecurrentcompressionloadattheinterfacebetweensupportandliner.Ateverytimestepduringthetransientanalysis,thecurrentvalueofN.(eitherzerofor'ift-offcondition,oracompressivefinitevalue)iscomputed.Finally,thesupportrotationalfrictionspringsKRreflectanyrotationalrestraintthatmaybeofferedbythefoundation.ThisspringrateiscalculatedusingamodifiedBousinesqequationandisincludedtosimulatetheresistivemomentofthesupporttocounteractrotationoftherackleginaverticalplane.Thisrotationspringisalsononlinear,withazerospringconstantvalueassignedafteracertainlimitingconditionofslabmomentloadingisreached.Thenonlinearityofthesesprings(frictionelements9,11,13iand15inTable6.6.1)reflectstheedginglimitationimposedonthebaseoftheracksupportlegsandtheshiftsinthecentroidofloadapplicationastherackrotates.Ifthiseffectisneglected,anysupportlegbending,inducedbyliner/baseplatefrictionforces,isresistedbythelegactingasabeamcantileveredfromtherackbaseplate.Thisleadstohigherpredictedloadsatthesupportleg-baseplatejunctionthanifthemomentresistingcapacityduetofloorelasticityatthefloorisincludedinthemodel.ThespringrateKS,modelingtheeffectivecompressionstiffnessofthestructureinthevicinityofthesupport,iscomputedfromtheequation:6-15 IIWII ThespringrateKS,modelingtheeffectivecompressionstiffnessofthestructureinthevicinityofthesupport,iscomputedfromtheequation:1111-+-+KSK1K2K3where:K1=springrateof,thesupportlegtreatedasatension-compressionmemberK2=localspringrateofpoolslabK3=springrateoffoldedplatecellstructureabovesupportlegAsdescribedintheprecedingsection,therack,alongwiththebase,supports,andstoredfuelassemblies,ismodeledforthegeneralthree-dimensional(3-D)motionsimulationbyatwenty-twodegreeoffreedommodel.Tosimulatetheimpactandslidingphenomenaexpected,upto64nonlineargapelementsand16nonlinearfrictionelementsareused.Gapandfrictionelements,withtheirconnectivityandpurpose,arealsopresentedinTable6.6.1.Table6.6.2listsrepresentativevaluesforamoduleusedinthesinglerackdynamicsimulations.Forthe3-Dsimulationofasinglerack,allsupportelements(describedinTable6.6.1)areincludedinthemodel.Couplingbetweenthetwohorizontalseismicmotionsisprovidedbothbyanyoffsetofthefuelassemblygroupcentroidwhichcausestherotationoftheentirerackand/orbythepossibilityoflift-offofoneormoresupportlegs.Thepotentialexistsfortheracktobesupportedononeormoresupportlegsduringanyinstantofacomplex3-Dseismicevent.'llofthesepotentialeventsmaybe6-16 IIIIII simulatedduringa3-Dmotionsothatamechanismexistsinthemodeltosimulatetherealbehavior.6.7'me'eat'onofthetonsoMoto6.7.1e-'stoAnasisUs'nuO'-DereeofFreedoRackModeHaving'assembledthestructuralmodel,thedynamicequationsofmotioncorrespondingtoeachdegreeoffreedomarewrittenbyusingLagrange'sPormulation.Thesystemkineticenergycanbeconstructedincludingcontributionsfromthesolidstructuresandfromthetrappedandsurroundingfluid.Asinglerackismodelledindetail.Thesystemofequationscanberepresentedinmatrixnotationas:[Ml(q")-(Q)+(G)where:[M](q)totalmassmatrix;thenodaldisplacementvectorrelativetothepoolslabdisplacement;doubleprimestandsforsecondaryderivations;avectordependentonthegivengroundacceleration;avectordependentonthespringforces(linearandnon-linear)andthecouplingbetweenmasses.Theequationcanberewrittenas(q<<)~[M~1(Q)+[M)-1(G)Asnotedearlier,inthenumericalsimulationsruntoverifystructuralintegrityduringaseismicevent,therattlingfuelassembliesareassumedtomoveinphase.Thiswillprovidemaximumimpactforcelevel,andinduceadditionalconservatisminthetime-historyanalysis.6-17 I~\IIII Thisequationsetismassuncoupled,displacementcoupledateachinstantintime,andisideallysuitedfornumericalsolutionusingacentraldifferencescheme.Theproprietary,USNRCaccepted,computerprogram"DYNARACK"*isutilizedforthispurpose.Stressesinvariousportionsofthestructurearecomputedfromknownelementforcesateachinstantoftimeandthemaximumvalueofcriticalstressesovertheentiresimulationisreportedinsummaryformattheendof.eachrun.Insummary,dynamicanalysisoftypicalmulti-cellrackshasshownthatthemotionofthestructureiscapturedalmostcompletelybythebehaviorofatwenty-twodegreeoffreedomstructure;therefore,inthisanalysismodel,themovementoftherackcross-sectionatanyheightisdescribedintermsoftherackdegreesoffreedom(ql(t),...q6(t)andq17-q22(t)).Theremainingdegreesoffreedomareassociatedwithhorizontalmovementsofthefuelassemblymasses.In'thisdynamicmodel,fiverattlingmassesareusedtorepresentfuelassemblymovementinthehorizontalThiscodehasbeenpreviouslyutilizedinlicensingofsimilarracksforEnricoFermiUnit2(USNRCDocketNo.50-341),QuadCities1and2(USNRCDocketNos.50-254and265),RanchoSeco(USNRCDocketNo.50-312),OysterCreek(USNRCDocketNo.50-219),V.C.Summer(USNRCDocketNo.50-395),andDiabloCanyon1and2(USNRCDocketNos.50-275and50-323),St.LucieUnitI(USNRCDocketNo.50-335),ByronUnitsIandII(USNRCDocketNos.50-454,50-455),Vogtle2(USNRCDocket50-425),andMillstoneUnit1(USNRCDocket50-245),IndianPointUnit2(USNRCDocketNo.50-247),amongothers.6-18 IIIlI plane.Therefore,thefinaldynamicmodelconsistsoftwelvedegreesoffreedomfortherackplustenadditionalmassdegreesoffreedomforthefiverattlingmasses.Thetotalityoffuelmassisincludedinthesimulationandisdistributedamongthefiverattlingmasses.6.7.2EvaluationofPotentialforInter-RackImactSinceracksareusuallycloselyspaced,thesimulationincludesimpactspringstomodelthepotentialforinter-rackimpact.Toaccountforthispotential,yetstillretainthesimplicityofsimulatingonlyasinglerack,gapelementsarelocatedontherackatthetopandatthebaseplatelevel.Fig.6.5.2showsthelocationofthesegapelements.Thebaseplatelocationisadesignatedpotentialimpactregion,andtheimpactspringslocatedinthisregionareexpectedtoregisterimpactloads.However,theimpactisdisallowedinthecellularregionoftheracks.Therefore,theimpactspringslocatedatthetopmustnotindicateanyloadsatanytimeduringtheseismicevent.6.8StructuralAccetanceCiteriaTherearetwosetsofcriteriatobesatisfiedbytherackmodules:a0KinematicCiter'oThis'riterionseekstoensurethattherackisaphysicallystablestructure.Theracksaredesignedtoprecludeinter-rackimpactsinthecellularregion.Therefore,physicalstabilityoftherackisconsideredalongwiththecriterionthatinter-rackimpactorrack-to-wallimpactsinthecellularregiondonotoccur.6-19 IlI b.StressLimitsThestresslimitsoftheASMECode,SectionIII,SubsectionNFg1989Editionareused.Thefollowingloadingcombinationsareapplicable(Ref.6.1.2)andareconsistentwiththeplantUFSARcommitments.LoadinCombinatioStressLim'tD+LD+L+ToD+L+To+LevelAservicelimitsD+L+Ta+ED+L+To+PfD+L+Ta+E'+L+FdLeveBservicematsLevelDservicelmxtsThefunctionalcapabilityofthefuelracksshouldbedemonstrated.Theabbreviationsinthetableare'hoseusedinSection3.8.4oftheStandardReviewPlanandthe"ReviewandAcceptanceofSpentFuelStorageandHandlingApplications":D=Deadweight-inducedinternalmoments(includingfuelassemblyweight)L=LiveLoad(notapplicablefor.thefuelrack,sincetherearenomovingobjectsintherackloadpath).FdForcecausedbytheaccidentaldropoftheheaviestloadfromthemaximumpossibleheight.Pf~UpwardforceontherackscausedbypostulatedstuckfuelassemblyE~OperatingBasisEarthquake(OBE)E'DesignBasisEarthquake(DBE)6-20 IIIII Differentialtemperatureinducedloads(normaloperatingorshutdownconditionbasedonthemostcriticaltransientorsteadystatecondition).Ta~Differentialtemperatureinducedloads(thehighesttemperatureassociatedwiththepostulatedabnormaldesignconditions).TheconditionsTaandTocauselocalthermalstressestobeproduced.Forfuelrackanalysis,onlyonescenarioneedbeexamined.Theworstsituationwillbeobtainedwhenanisolatedstoragelocationhasafuelassemblywhichisgeneratingheatatthemaximumpostulatedrate.Thesurroundingstoragelocationsareassumedtocontainnofuel.Theheatedwatermakesunobstructedcontactwiththeinsideofthestoragewalls,therebyproducingthemaximumpossibletemperaturedifferencebetweentheadjacentcells.Thesecondarystressesthusproducedarelimitedtothebodyoftherack;thatis,thesupportlegsdonotexperiencethesecondary(thermal)stresses.Forrackqualification,To,Taarethesame.6.9MaterialProertiesThedataonthephysicalpropertiesoftherackandsupportmaterials,obtainedfromtheASMEBoiler6PressureVesselCode,SectionZZZ,appendices,arelisted=in.Table6.9.1.Sincethemaximumpoolbulktemperatureislessthan200F,thisisusedasthereferencedesigntemperatureforevaluationofmaterialproperties.6-21 IIIII 6.10StressLimitsforVariousConditionsThefollowingstresslimitsarederivedfromtheguidelinesoftheASMECode,SectionIZZ,SubsectionNF[6.1.2],inconjunctionwiththematerialpropertiesdataoftheprecedingsection.AllparametersandterminologyareinaccordancewiththeCode.6.10.1NoaandUsetCond't'onsLeveAorLe~elBa~AllowablestressintensiononanetsectionFt~0'Sy(Sy~yieldstressattemperature)Ft=(0~6)(25I000)=15~000psi(rackmaterial)Ft=isequivalenttoprimarymembranestressesFt=(.6)(25,000)=15,000psi(upperpartofsupportfeet)(~6)(106/300)63@780psi(lowerpartofsupportfeet)b.Onthegrosssection,allowablestressinshearisoFv=.4Sy(.4)(25,000),10,000psi(mainrackbody)Ft=(.4)(25,000)=~10,000psi(upperpartofsupportfeet)(.4)(106,300)~42,520psi(lowerpartofsupportfeet)6-22 IIIIIII c.AllowablestressincompressionIFa.(kl)22[1--/2Ccr2SyF5klkl33E(-)+[3(-)/<<)-[(-)/8Cc3rrwhere:(2gz2E)1/2Syl=unsupportedlengthofcomponentk~lengthcoefficientwhichgivesinfluenceofboundaryconditions;e.g.kI1(simplesupportbothends)=1/2(cantileverbeam)=2(clampedatbothends)E~Young'sModulusr=radiusofgyrationofcomponentkl/rforthemainrackbodyisbasedonthefullheightandcrosssectionofthehoneycombregion.Substitutingnumbers,weobtain,forbothsupportlegandhoneycombregion:Fa=15,000psi(mainrackbody)Fa~15,000psi(upperpartofsupportfeet)~63,780psi(lowerpartofsupportfeet)d0Maximumallowablebendingstressattheoutermostfiberduetoflexureaboutoneplaneofsymmetry:Fb~0.60Sy=15,000Psi(rackbody)Fb~15,000psi(upperpartofsupportfeet)~63,780psi(lowerpartofsupportfeet)6-23 IIIIIII"n"II e.Combinedflexureandcompression:faCmxfbxCmyfby++<1FaDxFbxDyFbywhere:faDirectcompressivestressinthesectionfbxMaximumflexuralaxisstressalongx-fbyMaximumflexuralaxisstressalongy-Cmy=0.85Dx1faF'exDy~112n2faF'eyF'exiey=kl23(-)x,yandthesubscriptsx,yreflecttheparticularbendingplaneofinterest.f.Combinedflexureandcompression(ortension):fafbx+-+0.6SyFbxfby10Fby6-24 IIIII Theaboverequirementshouldbemetforboththedirecttensionorcompressioncase.6.'10.2LeveDServce'm'tsSectionF-1370(ASMESectionIII,AppendixF),statesthatthelimitsfortheLevelDconditionaretheminimumof1.2(Sy/Ft)or(0.7Su/Ft)timesthecorrespondinglimitsforLevelAcondition.Suistheultimatetensile'stressat200'FperTable6.9.1.Since1.2Syisgreaterthan0.7Suforthelowerpartofthesupportfeet,thelimitis1.54forthelowersectionunderDBEconditions.Thelimitfortheupperportionofthesupportfootis2.0underDBEconditions.Insteadoftabulatingtheresultsofthedifferentstressesasdimensionedvalues,theyarepresentedinadimensionlessform.Thesedimensionlessstressfactorsaredefinedastheratiooftheactualdevelopedstresstoitsspecifiedlimitingvalue.Withthisdefinition,thelimitingvalueofeachstressfactoris1.0fortheOBEand2.0(or1.54)fortheDBEcondition.6.11ResultsfortheAnalysisofSpentFuelRacksUsinaSinleRackModeland3-DSeismicMotionAcompletesynopsisoftheanalysisofthesinglerack,subjecttothepostulatedearthquakemotions,ispresentedinasummaryTable6.11.1whichgivestheboundingvaluesofstressfactorsRi(il...7).Thestressfactorsaredefinedas:R1=Ratioofdirecttensileorcompressivestressonanetsectiontoitsallowablevalue(notesupportfeetonlysupportcompression)R2Ratioofgrossshearonanetsectioninthex-directiontoitsallowablevalue6-25 IIIIII R3R4Ratioofmaximumbendingstressduetobendingaboutthex-axistoitsallowablevalueforthesectionRatioofmaximumbendingstressduetobendingaboutthey-axistoitsallowablevalueR5R6R7Combinedflexureandcompressivefactor(asdefinedin6.10.leabove)Combinedflexureandtension(orcompression)factor(asdefinedin6.10.1f)Ratioofgrossshearonanetsectioninthey-directiontoitsallowablevalue.Asstatedbefore,theallowablevalueofRi(i=1,2,3,4g5g6g7)is1fortheOBEconditionand2forteDBEexcetfothelowersectionofthesuortwherethefactoris1.54Thedynamicanalysisgivesthemaximax(maximumintimeandinspace)valuesofthestressfactorsatcriticallocationsintherackmodule.Ualuesarealsoobtainedformaximumrackdisplacementsandforcriticalimpactloads.Table6.11.1presentscriticalresultsforthestressfactors,andforrack-to-fuelimpactload.Table6.11.2presentsmaximumresultsforhorizontaldisplacementsatthetopandbottomoftherackinthexandydirection.Forsingleracksimulations"x"isalwaystheshortdirectionoftherack.InTable6.11.2,foreachrun,boththemaximumvalueofthesumofallsupportfootloadings(4supports)aswellasthemaximumvalueonanysinglefootisreported.Thetablealsogivesvaluesforthemaximumverticalloadandthecorrespondingnetshearforceatthelineratessentiallythesametimeinstant,andforthemaximumnetshearloadandthecorrespondingverticalforceatasupportfootatessentiallythesametimeinstant.6-26 III TheresultspresentedinTables6.11.1and6.11.2representthetotalityofsinglerackrunscarriedout.Thecriticalcaseforstructuralintegritycalculationsisincluded.Displacementsatthebaseplatelevelareminimal.ThesinglerackanalysisforrunA04gavethehigheststressfactorsforsubsequentstructuralintegritycalculations.Subsequenttothedetailedanalysis,pedestalsadjacenttothepoolwallswererelocatedfromthecornercelltonewlocations2cellsinboardfromtheedge.Sincethisrelocationcouldaffecttheconclusionsconcerningrackstructuralintegiity,thecriticalcaseofrunA04wasre-consideredusingthenewpedestallocations.Theresultsofthatre-analysisarepresentedinthetablesasrunA94.Thedetailedstructuralintegritycomputationsreportedhereinarebasedonthecriticalcasefortheloadingscenarioinvestigated.SubsequentWholePoolMulti-Rackanalysesarealsobasedonthefinalpedestallocations.TheresultscorrespondingtoDBEgivethehighestloadfactors.ThecriticalloadfactorsreportedforthesupportfeetareallfortheuppersegmentofthefootforDBEsimulationsandaretobecomparedwiththelimitingvalueof2'.Resultsforthelowerportionofthesupportfootarenotcriticalandarenotreportedinthetables.Analysesshowthatsignificantmarginsofsafetyexistagainstlocaldeformationofthefuelstoragecellduetorattlingimpactoffuelassemblies.6-27 II Overturninghasalsobeenconsidered.Thishasbeendonebyassumingamultiplierof1.5ontheDBEhorizontalearthquakesk(moreconservativethanrequiredbytheUSNRCStandardReviewPlan)andcheckingpredicteddisplacements.Thehorizontaldisplacementsdonotgrowtosuchanextentastoimplyanypossibilityforoverturning.ItisnotedthattheanalysesoftheDonaldC.Cookplantfuelrackshaveincludedsomeasymetricallyloadedracks.Theresultsofthesestudiescanbeusedasboundinganalysesforthecasewhenarackmoduleispickedupandrelocatedwhenloadedasymmetricallywithfuelassemblies.Theresultspresentedhereinindicatethattwistingordeformationthatwouldcauselossoffunctionorviolationofsafetymarginswillnotoccurduringaplannedrackrelocation.6.12ImactAnalses6.12.1ImactLoadinBetweenFuelAssembandCeWallThelocalstressinacellwallisconservativelyestimatedfromthepeakimpactloadsobtainedfromthedynamicsimulations.Plasticanalysisisusedtoobtainthelimitingimpactload.Thelimitloadiscalculatedas3125lbs.percellwhichismuchgreaterthantheloadsobtainedfromanyofthesimulations.6.12.2ImactsBetweenAd'acentRacksAllofthedynamicanalysesassume,conservatively,thattheracksareisolated.However,thedisplacementsobtainedfromthedynamicanalysesarelessthan50%oftherack-to-rackspacingorrack-to-wallspacingifthepoolisassumedfullypopulated.6-28 l~I Therefore,weconcludethatnoimpactsbetweenracksorbetweenracksandwallsoccurduringtheDBEevent.ThishasbeenfurtherprovenbytheWholePoolMulti-RackAnalysisdiscussedinSection6'4.6.13WeldStressesCriticalweldlocationsunderseismicloadingareatthebottomoftherackatthebaseplateconnectionandattheweldsonthesupportlegs.Resultsfromthedynamicanalysisusingthesimulationcodesaresurveyedandthemaximumloadingisusedtoqualifytheweldsontheselocations.6.13.1BaselatetoRackWeldsandCell-to-CellWeldsRef.[6.1.2](ASMECodeSectionIII,SubsectionNF)theDBEcondition,anallowableweldstresst=.42psi.Basedontheworstcaseofallrunsreported/weldstressforthebaseplateto'ackweldsis7605conditions.permits,forSu29/820themaximumpsiforDBETheweldbetweenbaseplateandsupportlegischeckedusinglimitanalysistechniques.ThestructuralweldatthatlocationisconsideredsafeiftheinteractioncurvebetweennetforceandmomentissuchthataderivedfunctionofF/FyandM/Myisbelowalimitingvalueof1.0.6-29

,/

FyIMyarethelimitloadandmomentunderdirectloadonlyanddirectmomentonly.F,Maretheabsolutevaluesoftheactualforceandmomentsappliedtotheweldsection.Thecalculatedvalueis.637(1.0based'ntheinstantaneouspeakloading.Thisvalueconservativelyneglectsanygussetsinplacetoincreasepedestalareaandinertia.Thecriticalareathatmustbeconsideredforcell-to-cellweldsistheweldbetweenthecells.Thisweldisdiscontinuousasweproceedalongthecelllength.Stressesinthestoragecelltostoragecellweldsdevelopalongthelengthofeachstoragecellduetofuel'ssemblyimpactwiththecellwall.Thisoccursiffuelassembliesinadjacentcellsaremovingoutofphasewithoneanothersothatimpactloadsintwoadjacentcellsareinoppositedirectionswhichwouldtendtoseparatethechannelfromthecellattheweld.ThecriticalloadthatcanbetransferredinthisweldregionfortheDBEconditioniscalculatedas5273lbs.ateveryfuelcellconnectiontoadjacent,cells.Anupperboundtotheloadrequiredtobetransferredis593lbs.Wherewehaveusedamaximumimpactloadof210lbs.(obtainedfromTable6.11.1),wehaveassumedtwoimpactlocationsaresupportedbyeachweldregion,andwehaveincreasedtheloadbyV'2toaccountfor3-Deffects.6.13.2HeatinofanIsolatedCellWeldstressesduetoheatingofanisolatedhotcellarealsocomputed.Theassumptionusedisthatasinglecellisheated,overitsentirelength,toatemperatureabovethevalueassociatedwithallsurroundingcells.Nothermalgradientinthe6-30 IlL~~~~

verticaldirectionisassumedsothattheresultsareconservative.Usingthetemperaturesassociatedwiththisunit,analysisshowsthattheweldstressesalongtheentirecelllengthdonotexceedtheallowablevalueforathermalloadingcondition.Section7reportsthevalueforthisthermalstress.6.14WholePooMuti-RackWPMRAnals'sThesinglerack3-Dsimulationspresentedintheprecedingsectionsdemonstratethestructuralintegrity,physicalstability,andkinematiccompliance(norack-to-rackimpactinthecellularregion)oftherackmodules.However,asnotedbefore,prescribingthemotionoftheracksadjacenttothemodulebeinganalyzedintroducesanassumptionofunpredictableimportinthesinglerackmodules.Forcloselyspacedracks,itispossibletodemonstratekinematiccomplianceonlybymodellingallrackmodulesinonecomprehensivesimulationwhichisreferredtoasWholePoolMulti-Rack(WPMR)model.IntheWPMRanalysis,DBEseismicloadisapplied(Ref.6.1.3)-andallracksareassumedfullyloadedwithfuelassemblies.Theprimaryintentoftheanalysisistoconfirmstructuralintegrityconclusionsfrom3-Dsinglerackanalysisandtoensurethathydrodynamiceffectsnotabletobemodelledinasinglerackanalysisdonotcauseunanticipatedstructuralimpacts.Thecrosscouplingeffectsduetothemovementoffluidfromoneinterstitial(inter-rack)spacetotheadjacentoneismodelledusingclassicalpotentialflowtheoryandKelvin'scirculationtheorem.ThisformulationhasbeenreviewedandapprovedbytheNuclearRegulatoryCommission,duringthepost-licensingmulti-rackanalysisforDiabloCanyonUnitIandIIrerackingproject.Thecouplingcoefficientsarebasedonaconsistentmodellingof6-31

thefluidflow.Whileupdatingofthefluidflowcoefficients,basedonthecurrentgap,ispermittedinthealgorithm,theanalyseshereareconservativelycarriedoutusingtheconstantnominalgapsthatexistatthestartoftheseismicevent.SuchacomprehensiveWPMRmodelwaspreparedfortheracksshowninthemodulelayoutdrawing(Fig.6.4.1).ComputercodeDYNARACKwasusedtoperformthesimulations.Znordertoeliminatethelastsignificantelementofuncertaintyinrackdynamicanalyses,thefrictioncoefficientwasalsoascribedtothesupportleg/poolbearingpadinterfaceinamannerconsistentwithRabinowicz'sexperimentaldata[6.4.1].AsetoffrictioncoefficientsweredevelopedbyarandomnumbergeneratorwithGaussiannormaldistributioncharacteristics.Theserandomderivedcoefficientsareimposedoneachpedestalofeachrackinthepool.Theassignedvaluesarethenheldconstantduringtheentiresimulationinorderthattheresultsarereproducible.6.14.1Multi-RackModelFigure6.14.1showsaplanformviewoftheDonaldC.'ookspentfuelpool.Arackandpedestalnumberingschemeissetupinthefigure.WesetupaglobalxaxistowardstheEast.Table6.14.1givesinformationonthenumberofcellsperrack,andontherackandfuelweights.Allracksareassumedloadedwithregularfuel.Therearetwenty-threeracksinthepool.Thecaskareainthepoolismodelledasafictitiousrack(Rackf24inFigure6.14.1).Asnotedpreviously,thepresenceofafluidmovinginthenarrowgapsbetweenracksandbetweenracksandpoolwallscausesfluidcouplingeffectswhichcannotbemodelledwithasimulationusing6-32

onlyasinglerack.Verysimply,asingleracksimulationcaneffectivelyincludeonlythehydrodynamiceffectsduetocontiguousrackswhenacertainsetofassumptionsisusedforthemotionofcontiguousracks.Inamulti-rackanalysisthefarfieldfluidcouplingeffectsofallracksisaccountedforusinganappropriatemodelofthepool-rackfluidmechanics.ForDonaldC.Cook,thecaskareawasmodelledassumingverylargefluidgapsbetweenracks18and24andbetweenracks23and24.IntheWholePoolMulti-Rackanalysis,usedtoinvestigatetheinteractioneffectsofallracks,weemployareduceddegree-of-freedom(RDOF)setforeachrackplusitscontainedfuel.Thepurposeofthewholepooldynamicanalysis,includingthecompletesetofracksinthepool,istodeterminewhethereffects,notabletobeconsideredinasinglerackanalysis,alteranyoftheconclusionsthatarebasedontheresultsofthe22DOFsinglerackanalysis.Inparticular,themulti-rackanalysisfocussesondisplacementexcursionsofeachrackandonpedestalcompressiveloads.TheWholePoolMulti-Rackanalysisisalsoutilizedtoinvestigatethepossibilityofimpactsbetweenracksorbetweenracksandpoolwalls.Thereduceddegree-of-freedomstructuralmodelforeachrackisdevelopedinasystematicwaysothattheimportantkinematicresultsfromadynamicanalysisareinagreementwithsimilarresultsfromasolutionobtainedusingthe22DOFsinglerackmodel.Theexternalhydrodynamicmassduetothepresenceofwallsoradjacentracksiscomputedinamannerconsistentwithfundamentalfluidmechanicsprinciplesandtheuseofareduced6-33 I

DOFfuelrackmodel[6.14.1].Thefluidflowmodel,usedtoobtainthewholepoolhydrodynamiceffectissitespecificandreflectsactualgapsandracklocations.Thewholepoolmulti-rackmodelincludesmanynon-linearcompressiononlygapelements.'herearegapelementsrepresentingcompressiononlypedestals(normallyfourpedestalsareassumedforeachrack),gapelementsdescribingtheimpactpotentialofthefuelassembly-fuelrackinterface,andgapelementstrackingrack-to-rackorrack-to-wallimpactpotentialat,thetopandbottomcornersoftherackcellstructure.Znadditiontothecompressiononlygapelements,,eachpedestalhastwofrictionspringsassociatedwiththecompressionspring.Asnotedpreviously,arandomnumbergeneratorisusedtoestablishafrictioncoefficient,foreachpedestalateachinstantwhenthepedestalisincontact,withtheliner.TheseismicexcitationdirectionsXandYareshowninFigure6.14.1.ThecriticalDBEeventthatgovernsthebehaviorofthesinglerackanalysisisappliedtothe3-Dmulti-rackmodelintheappropriatedirections.Threesimulationshavebeencarriedoutusingcoefficientsoffrictionassumedtobe0.2,toberandomwithameanof0.5atallpedestals,andtobe0.8,respectively.6.14.2--ResultsofMult'-RackAnalsisTables6.14.2-6.14.4showthemaximumcornerabsolutedisplacementsatboththetopandbottomofeachrackinxandydirectionsfromthreemulti-rackruns.ZnTable6.14.5,themaximumdisplacementsobtainedfromthethreemulti-racksimulationsarecomparedwithasinglerackanalysis.Znallof6-34

~~5rl thesetables,theresultsforfuelrack24canbeignoredsinethereisnorealrackatthatLocation."Theabsolutedisplacementvaluesarehigherthanthoseobtainedfromsinglerackanalysis.Thus,itappearsessentialtoperformWholePoolMulti-Rackanalysestoverifythatracksdonotimpactorhitthewall.Pigures6.14.2-6.14.5showthetimehistoryozrack-to-rackgapsforthecriticalracks.Ztisshownthattherack-to-rackdynamicgapsaregreaterthan1.65"duringa15secondearthcpxake.Detailedexaminationoftherack-to-rackdynamicgapsshowthattheracksprimar'ymovein-phaseinallthreesimulations.Thatis,theentireassemblageozrackstendstomoveandminimizechangesinrack-to-rackgaps.Table6.14.5alsopresentspeakpedestalcompressiveloadsofallpedestalsonthetwenty-threerealracks.l:nadditiontoareportofmaximumpedestalloads,thetimehistoryofeachpedestalLoadforeachrackisarchivedforuseinthestructuralevaluationofthefuelpoolslaband,theenvelopingwallsozthefuelpool,.Ztisnotedthatpredicted,maximumpedestalzorcefromthemulti-racksimulationgivingtheLazgestpedestalload(RunMP3inTabLe6-14.5)islowerthantheresultobtainedfromsinglerackanalysis.Themacimuminstantaneousverticalfootloadobtainedfromsinglerackanalysisis183300Lbs.PromtheWholePoolMulti-RackRunMP3,wefindapeaksinglepedestalLoadoz180900lbs.Because,detailedrackstesscalculationsarebasedonthesinglerackanalysisresults,zonewstructureconcernsareidentifiedbythescopingWholePoolanalysisandtheoverallstructuralintegrityconclusionsareconf~ed.6-35

6.15BearinPadAnalsisToprotect,theslabfromhighlocalizeddynamicloadings,bearingpadsareplacedbetweenthepedestalbaseandtheslab.Fuelrackpedestalsimpactonthesebearingpadsduringaseismiceventandtheverticalpedestalloadingistransferredtotheliner.ThebearingpaddimensionsaresettoensurethattheaveragepressureimpactedtotheslabsurfaceduetoastaticloadplusadynamicimpactloaddoesnotexceedtheAmericanConcreteInstitute[6.15.1]limitonbearingpressures.Thetimehistoryresultsfromthedynamicsimulationsforeachpedestalareusedtogenerateappropriatestaticanddynamicpedestalloadswhichareusedtodevelopthebearingpadsize.Fromthewholepoolmulti-rackanalysis,theworstcaseloadingonapedestal(instantaneouspeakload)is183,300lbs.(seeTable6.14.5).Fora12"x12"pad,thisgivesanaverageinstanteouspressurepa=1273psi.Section10.15of[6.15F1]givesthedesignbearingstrengthasfb=P(.85fc')CwhereP~.7andfcŽ3500psiforDonaldC.Cook.6=1exceptwhenthesupportingsurfaceiswideronallsidesthantheloadedarea.Inthatcase,C~(A2/A1)'butnotmorethan2.A1istheactualloadedarea,andA2isanareagreaterthanA1whichisdefinedpictoriallyintheACIcommentaryonSection10.15.ForDonaldC.Cook,1~6~2;ifweconservativelyuse6=1,thenfb2083psiwhichisinexcessofthecalculatedpressurepa.Thus,significantmarginisprovidedbythebearingpads.6-36 I

ReferencesforSection6USNRCStandardReviewPlan,NUREG-0800(1981).ASMEBoiler&PressureVesselCode,SectionIII,SubsectionNF,appendices(1989).USNRCRegulatory.Guide1.29,"SeismicDesignClassification,"Rev.3,1978.HoltecProprietaryReport-VerificationandUser'sManual,ReportHI-89364,January,1990."FrictionCoefficientsofWaterLubricatedStainlessSteelsforaSpentFuelRackFacility,"Prof.ErnestRabinowicz,MIT,areportforBostonEdisonCompany,1976.USNRCRegulatoryGuide1.92,"CombiningModalResponsesandSpatialComponentsinSeismicResponseAnalysis,"Rev.1,February,1976."TheComponentElementMethodinDynamicswithApplicationtoEarthquakeandVehicleEngineeringgS.LevyandJ.P.D.Wilkinson,McGrawHill,1976."DynamicsofStructures,"R.W.CloughandJ.PenziengMcGrawHill(1975).HoltecProprietaryReports:User'sManual,ReportHI-89343,Revision0;Theory,ReportsHI-87162,Revision1,andHI-90439,Revision0;Verification,ReportHI-87161,Revision2."DynamicCouplinginaCloselySpacedTwo-BodySystemVibratinginLiquidMedium:TheCaseofFuelRacks,"K.P.SinghandA.I.Soler,3rdInternationalConferenceonNuclearPowerSafety,Keswick,England,May1982.R.J.Fritz,"TheEffectsofLiquidsontheDynamicMotionsofImmersedSolids,"JournalofEngineeringforIndustry,Trans.oftheASME,February1972,pp167-172.USNRCRegulatoryGuide1.61,"DampingValuesforSeismicDesignofNuclearPowerPlants,"1973.6-37 I

"FluidCouplinginFuelRacks:CorrelationofTheoryandExperiment",byB.Paul,HoltecReportHI-88243.ACI318-89,ACI318R-89,BuildingCodeRequirementsforReinforcedConcrete,AmericanConcreteInstitutegDetroit,Michigan,1989.6<<38 II~~4'4IlI Table6.3.1CORRELATIONCOEFFICIENTTimeHistoGrouN-S"andE-W(1,2)N-StoVertical(1,3)E-WtoVertical(2,3)DBB0.01460.12690.01016OBE0.10560.09560.10606-39

Table6.5.1DEGREESOFFREEDOMLocation(Node)Dz.spacementUxUyUzRotationexeyezPlP2P3P17P18P19q4q5q6q20q21q22.where:Point2isassumedattachedtorigidrackatthetopmostpo'int.P7P8P9P10P11P12P13P14P15P16Pi'qi(t)+U1(t)qi(t)+U2(t)qi(t)+U3(t)i~1,7,9,11,13,15,17i~2,8,10,12g14I16g18i~3,19Ui(t)arethe3knownearthquakedisplacements.6-40 IIIIIOlI Table6.6.1NUMBERINGSYSTEMFORGAPELEMENTSANDFRICTIONELEMENTSZ.NonlinearSinsGaEements(64Total)NumberNodeLocatioDescr'ionSupportS1SupportS2SupportS3SupportS42I2ZcompressionZcompressionZcompressionZcompressionXrack/fuelelementonlyelementonlyelementonlyelementonlyelementassemblyimpact2I2X'rack/fuelassemblyimpactelement2I2Yrack/fuelassemblyimpactelement212Yrack/fuelassemblyimpactelement9-24Otherrattlingmassesfornodes1*i3*i4*and5*2544Bottomcross-sectionofrack(aroundedge)Inter-rackInter-rackInter-rackInter-rackInter-rackInter-rackInter-rackimpactimpactimpactimpactimpactimpactimpactelementselementselementselementselementselementselementsInter-rackimpactelements4564Topcross-sectionofrack(aroundedge)Inter-rackInter-rackInter-rackInter-rackInter-rackInter-rack-"Inter-rackInter-rackimpactimpactimpactimpactimpactimpactimpactimpactelementselementselementselementselementselementselementselements6-41 II Table6.6.1(continued)NUMBERINGSYSTEMFORGAPELEMENTSANDFRICTIONELEMENTSII.Fr'ctionElements(16total)Number123'4~5678910111213141516NodeLocat'onSupportS1SupportSlSupportS2SupportS2Support.S3SupportS3SupportS4SupportS4SlS1S2S2S3S3S4S4DescritionXdirectionfrictionYdirectionfrictionXdirectionfrictionYdirectionfrictionXdirectionfrictionYdirectionfrictionXdirectionfrictionYdirectionfrictionXSlabmoment,YSlabmomentXSlabmomentYSlabmomentXSlabmomentYSlabmomentXSlabmomentYSlabmoment6-42 IIlI Table6.6.2TYPICALINPUTDATAFORRACKANALYSES(lb-inchunits)SupportFootSpringConstantKs(0/in.)FrictionalSpringConstantKf(0/in.)RacktoFuelAssemblyImpactSpringConstant(0/in.)ElasticShearSpringforRack(4/in-)ElasticBendingSpringforRack(0-in/in.)ElasticExtensionalSpring(5/in.)ElasticTorsionalSpring(g-in./in.)IGaps(in.)(forhydrodynamiccalculations)4.91x1061.837x1091.38x105(x-direction)1.61x10(y-direction)5.986x10(x-direction)4.866x10(y-direction)5.458x10(x-zplane)4.71x1010(y-zplane)4.074x1071.322x1096-43 IIII Table6.9.1RACKMATERIALDATA(200'F)Material304S.S.Young'sModulusE(psi)27.9x106YieldStrengthSy(psi)25000UltimateStrengthSu(psi)71000SectionIXIReferenceTableX-6.0TableI-2.2TableX-3.2SUPPORTMATERXALDATA(200F)Material1SA-240,Type304(upperpartofsupportfeet)27.9x106Psi.25,000PSi71000PSi2SA-564-630(agehardenedat1100'F)279x106PSl106,300140,000PSiPSl6-44 IIk Table6.11.1STRESSFACTORSANDRACK-TO-FUELIMPACTLOADRunRemarks-1RRack/FuelImpactLoadPerCellatWorstLocationAlongHeightCriticalLocation~lbs-2R-3R-5R-TRa03DBEp=o2182cellsloadedwithreg.fuela04DBEp=o8182cellsloadedwithreg.fuela30p=0.291cellsloadedwithreg.fuela32p=0.291cellsloadedwithreg.fuel180.2179.6190209.8~018.274.018.284.012.181.012.176.023.074.025.079.012.046.011.049.159.167.172.214.090.118.094~112.166.161.178.172.073.106.090.110.198.417.204.431.109.281.109.271~231.442.239.460.127.299.127.289.027*.078*.033.095.013.052.015.049a94Sameasa04174.8withreloca-tedpedestals.018.325.018.056.168.250~121.118.187.483.219.522.032.113*Uppervaluesareforrackcellcross-sectionjustabovebaseplate.Lowervaluesareforsupportfootfemalecross-sectionjustbelowattachmenttobaseplate.

IIIIiIII Table6.11.2RackDisplacementsandSupportLoads(allloadsareinlbs.)FLOORLOAD(sumofallsupportfeet)inarack~lbs.a03Fullload3.510x10p=0'DBE,Reg.FuelMAXIMUMVERTICALLOAD(1foot)~lbs.1.549x10MAXIMUMSHEARLOADANDCOINCIDENTVERTICALLOAD30212(1.511x10)DX~in.~0609.0084DY**~in..0562.0105a04Fullload3.510x10p=0.8~DBE,Reg.Fuel1.605x1035832(9.791x10).0679.0015.0583.0012a30HalfloadinPos.xp=0.2DBE,Reg.Fuela32HalfloadinPos.yp=0'DBE,Reg.Fuel1.883x101.883x101.021x109.973x1020108(1.005x10)19389(9.71x10).0520.0010.0482.0055.0450.0008.0515.0080a94Sameasa043.508x10withreloca-pedestals1.833x1044406(1.4829x10).0678.0014.0778.0018Thevalueinparenthesisistheverticalloadattheinstatwhentheshearloadismaximum.Themaximumverticalandshearloadsgenerallydonotoccuratthesameinstant.Uppervaluesaretopmovements;lowervaluesarebaseplatemovements(notnecessarilyatthesametime).

III Table6.14.1RACKNUMBERINGANDWEIGHTINFORMATIONRackNo.123456789101112131415161718192021222324*No.of~Ces1821681681821821821561441441561561561431321321431431431821681681661200Weightof2570023700237002570025700257002250020900209002250022500'2250020800193001930020800208002080025700237002370023900177000WeightofFuelAssemblb.155015501550155015501550155015501550155015501550155015501550155015501550155015501550155015500fictitious6-47 III Table6.14.2MAXIMUMDISPLACEMENTSFROMWPMRRUNMP1(FrictionCoefficient=0.2)rack123567891011121314151617181920212223uxt.7004E-01.7506E-01.8464E-01.5943E-01.5131E-01.6793E-01.4783E-01.4856E-Ol.4533E-01.3830E-01.4224E-01.6411E-01.7253E-01.4602E-01.3557E-01~3467E-01.5755E-Ol.1011E+00~6980E-01.8202E-01.8404E-01.8173E-01,.5647E-01uyt.7756E-01.5227E-01.7521E-01~5218E-01.5306E-01.9512E-01.8928E-01.7065E-01.6377E-01.5754E-01.5336E-01.9620E-01.1079E+00.1114E+00.1079E+00~9211E-01.4429E-01.1301E+00~1125E+00.8680E-011455E+00.1057E+00.6598E-01uxb.6235E-01.6494E-01.6897E-01.4960E-01.4290E-01.5135E-01.3978E-01.3607E-01.3196E-01.2848E-01..3659E-01.4885E-01.6568E-01.3650E-01.2634E-01~2817E-01.5326E-01.8596E-01~6341E-01.6878E-01~6800E-01.7111E-'01~4812E-01uyb.7303E-01.3936E-01.6619E-01.3597E-01.4496E-01.9095E-01.7830E-01.5917E-01.5192E-01.4354E-01.4329E-01.8429E-01.9505E-01.9847E-01.9325E-01.8608E-01.3140E-01.9693E-01.8575E-01.6048E-01.1229E+00.9050E-01.6156E-01uxt=absolutevaluex-directionatuyt=absolutevaluey-directionatuxb=absolutevaluex-directionatuyb=absolutevaluey-directionatofmaximumrackracktop;ofmaximumrackracktop;ofmaximumrackrackbaseplate;ofmaximumrackrackbaseplate.cornerdisplacementincornerdisplacementincornerdisplacement,incornerdisplacementin6-48 IIIII Table6.14.3MAXIMUMDISPLACEMENTSFROMWPMRRUNMP2'(RandomFrictionCoefficient)rackuxtuyt'uxbuyb12345678910111213~14151617181920212223.6524E-01.1423E+00.1247E+00.1860E+00.1106E+00.9642E-01.4742E-01.1801E+00.1275E+00.2336E+00.1710E+00.4015E-Ol.1088E+00.1439E+00.6218E-01.3322E+00.1727E+00.1269E+00.8411E-01~8402E-01.1280E+008427E-01.2389E+00.4772E-01.5829E-01.4122E-01.6628E-01.6379E-Ol.7250E-01.6267E-01.5755E-01.3974E-01.7640E-01.8644E-01.4740E-01.1034E+00.4029E-01.5620E-01.5413E-01.5385E-01.1958E+00.8106E-01.6480E-01.4742E-01.4951E-01.6471E-01.3373E-01.1442E+00.1161E+00.1859E+00.1091E+00.8330E-01.3334E-01.1819E+00.1207E+00.2336E+00.1712E+00.2869E-01.1030E+00.1282E+00.6029E-01.3374E+00.1727E+00.1223E+00.6365E-01.5976E-01.1281E+00.7430E-01.2388E+00.2303E-01~4598E-01.2566E-01.3161E-01.2673E-01.6348E-01.5443E-01.4534E-01.2115E-01.5527E-01.6245E-01.2678E-01.1040E+00.1865E-01.3386E-01.3677E-01.4896E-01.1913E+00.7508E-01.4419E-01.3530E-01.2335E-01.5758E-01uxt=absolutevaluex-directionatuyt=absolutevaluey-directionatuxb=absolutevaluex-directionatuyb=absolutevaluey-directionatofmaximumrackracktop;ofmaximumrackracktop;ofmaximumrackrackbaseplate;ofmaximumrackrackbaseplate.cornerdisplacementincornerdisplacementincornerdisplacementincornerdisplacementin6-49 IIII rackTable6.14.4MAXXMUMDZSPLACEMENTSFROMWPMRRUNMP3(FrictionCoefficient=0.8}uxtuytuxbuyb1234567891011121314151617181920212223.2035E+00.2751E+00.2637E+00.1363E+00.1333E+00.1720E+00.2425E+00.1785E+00.1519E+00.8112E-01.1146E+00.1005E+00.1604E+00.7786E-01.8616E-01.9843E-01.8975E-01.1418E+00.1959E+00.2741E+00.2117E+00.2361E+00.1016E+00.1702E+00.5173E-01.5740E-01.5449E-01.8237E-01.1514E+00.8747E-01.6039E-01.4434E;01.5007E-01.7975E-01.1602E+00.1310E+00.7618E-01~5521E-01~4/80E-01.7115E-01~4416E+00.1720E+00.5563E-01.5159E-01.6081E-01.7703E-01.1987E+00-2732E+00-2638E+00.1321E+001273E+00.1609E+00.2461E+00.1784E+00.1506E+00.7887E-01.1117E+00~9143E-01.1633E+00.7823E-01.8214E-01.1024E+00.9063E-01.1089E+00.1922E+00.2727E+00.2120E+00.2242E+00.1033E+00.1774E+00.1658E-01.4010E-01.2788E-01.6876E-01.1617E+00.8782E-01.4260E-01.3129E-01.2883E-01.5071E-01.1601E+00.1073E+00.5953E-01.3148E-01.2903E-01.7056E-01.4526E+00.1806E+00.3118E-01-2287E-01.3986E-01.7364E-01uxt=absolutevaluex-directionatuyt=absolutevaluey-directionatuxb=absolutevaluex-directionatuyb=absolutevaluey-directionatofmaximumrackracktop;ofmaximumrackracktop;ofmaximumrackrackbaseplate;ofmaximumrackrackbaseplate.cornerdisplacementincornerdisplacementincornerdisplacementincornerdisplacementin6-50 II Table6.14.5MAXIMUMRACKDISPLACEMENTANDFOOTLOADRuna94RemarksSingleRackAnalysisWPMR,p~0.2MaximumRackCornerDisplacementinch0.07780.1455(Rack$21iny)MaximumFootPedestalForcelbs.183,300157,400(Rack$19,Foot4)WPMR,Randomp0.3322(Racki16inx)170,900(Rack$19IFoot4)WPMR,p=0'0.4416(Rack$18iny)180,900(Rack$5,Foot2)pfrictioncoefficient6-51 II TYPICALCELLWALLSBASEPLATEBASEPLATEBEARINGPADFigure6.2.1PictorialViewofRackStructure6-52 III 0.200.10RQ-0.00LdLaJC3C3-0.10l'lI>III-O.-100100300500700TIME*0.0lsec.900110013001500FIGURE6.3.lDOE-t~-SACCELERATIONTIME-flISTORY I~~~~I 0.200.10-O.00LdLdC3C3-0.10-0-'1OO10030050070090TIME*0.01sec.110013001500FIGURE6.3.2DBEE-WACCELERATIONTIMEHISTORY II 0.200.10o-0.00UlC3-0.10-O.-100100300500700900110013001500TIME*0.0lsec.FIGURE6.3.3DBE-VERTICALACCELERATIONTIMEHISTORY r~(~~l 0.8003O.60eOI-CL0.40C3~0.20O.10FREQUENCY,IIz10FIGURE6.3.4HORIZONTAI'.DESIGNSPECTRUMANDN-STIMEHISTORYSPECTRUM(5%damping)

~~i~~I4 0.80-~0.60IoI-~0.40LLIC3C30.200.10FREQUENCY,Hz10FIGURE6.3..5HORIZOtlTAI',DESIGNSPECTRUMANDE-WTIMEHISTORYSPECTRUM(5%,damping)

~~~~~~~~I~L~~I O.600.40COLLIC3~0.200.10FREQUENCY,Hz10Figure6.3.6VERTICALDESIGNANDTINEHISTORYDERIVEDSPECTRA(5>damping)

I 0.08.oi0.03ChIZ,LJI~~-O.02LLJC3C3-007-O.-10010030050070090011001300TIME*0.01sec.1500FIGORE6.3.7OBE-N-'CCELERATIONTIMEHISTORY I

0.08cri0.03ChOw~-0.02LLJ-0.07-O.-100100300500700900TI&1E*G.Olsec.110013001500FIGURE6.3.8OBE-E-NACCELERATIONTINEHISTORY 14 0.05-0.05100300500700900110013001500TItlE*0.0lsec.FIGURE6.3.9OBE-VERTICALACCELERATIONTIMEHISTORY

0.60FREQUENCY,Hz10Figure6.3.10HORIZONTALDESIGNSPECTRUMANDTIMEHISTORYDERIVEDN-SSPECTRUM(2%damping) l 0.60Cf)C3~"0.40ICtLdLd~0.200.10FRE(FLUENCY,Hz10Figure6.3.llHORIZONTALDESIGNSPECTRUMANDE-WTIMEHISTORY'SDERIVEDSPECTRUM(2%damping)

I O.500.40(/)0.30C)~0.20LLI0.10cnpFREQUENCY,Hz10Figure6.3.12VERTICALDESIGNANDTIMEHISTORYDERIVEDSPECTRA(2%damping)

B,ackGeometricCenterlinePieH/qpizp,YLongDirectionSupportf'IP16PsTypicalFrictionElementFigure6.5.lSCHEMATXCMODELFORDYNARACK6-65

!)

TypicalTopImpactElement0IIIRackStructureTypicalBottomImpactElementFigure6.5.2RACK-TO-RACKIMPACTSPRINGS6-66 I

~CFLLWALLxstt>~~~KIFU"='SS:-!)'=LY!C"='IMPACTh."-RlNGIYsFIGURE6.5.3IMPACTSPRINGARRANGEMENTATNODEi6-67 lI 1720LFigure6.5.4DEGREESOFPREEDOHHODELLXHGRACKHOTXON I

L2FIGURE6.5.5RACKDEGREESOFFREEDOMFORX-ZPLANEBENDING6-69 II 18L2L2FIGURE6.5.6RACKDEGREESOFFREEDOMFORY-ZPLANEBENDING6-70 llI FiJELASSY/C-->>llFACT'SPRINGy,~I0.25HPJ2XEz::c<,C.G.P2%gH/2T'PIC):Pi~aT.'iGEL<S~0.25MFRCTZO.'fIHxrZZ.-:C"-SPRZ,'tG,Kfs>>.o."-zt:-"S?REYG,.KFOUNDATIONROTATIONALCRK.EAt>C-SPREiVG,KFIGURE6.6.12-DVIEWOFRACKMODULE6-71 II 2RACK6321RACK521RACK42RACK12342RACKll421RACK1021RACK16421RACK1732RACK16RACK24(NOTREAL)21RACK2321RACK22421RACK~2RACK9212RACK15RACK2134321RACK221RACK12RACK82RACK72'IRACK1421RACKIS21RACK2042RACK194FIGURE6.14.lRACKANDFOOTPEDESDALNUMBERINGFORD.C.COOKMULTI-RACKMODEL6-72 IIlll 2.03-2.00COOKPOOLMULTI-RACKSEISMICANALYSIS,RUNMP2(RandomCof.)RACK16TORACK17NORTHCORNERDYNAMICGAPATRACKTOP1.98x'195o1.93CI1.90OD1881.851.831.80012345678910111213141516TIME,SEC.FIGURE6.14.2 II 2.032.00COOKPOOLMULTI-RACKSEISMICANALYSIS,RUNMP2(RandomCof.)RACK16TORACK17SOUTHCORNERDYNAMICGAPATRACKTOP1.9819591.93zO1.90O~o188I1.851.831.80012345678910111213141516TIME,SEC.FIGURE6.14.3 l'IILt~lili 2.052.00COOKPOOLMULTI-RACKSEISMICANALYSIS,RUNMP3(Cof.=O.S)RACK12TORACK18WESTCORNERDYNAMICGAPATRACKTOP1.95-1.90u1.851.80o17I1.701.651.60012345678910111213141516TIME.SEC.FIGURE6.14.4 IIIIIIIIIII 2.052.00COOKPOOLMULTI-RACKSEISMICANALYSIS,RUNMP3(Cof.=O.S)RACK12TORACK18EASTCORNERDYNAMICGAPATRACKTOP1.95xO190~1.85D1.80C3o1.751.701.651.60012345678910111213141516TIME,SEC.FIGURE6.14.5 III 7.0CCIDENTANALYSISANDMISCELLANEOUSSTRUCTURALEVALUATIONS7.1IntroductionThissectionprovidesresultsofaccidentanalysesperformedtodemonstrateregulatorycomplianceofthenewfuelracks.Thereareseveraltypesofaccidentswhichcouldpotentiallyaffectthespentfuelstoragepool.InstallationoftheproposedhighdensityrackswillenablethestorageofincreasedamountsofspentfuelintheDonaldC.Cookspentfuelpool.Accordingly,accidentsinvolvingthespentfuelpoolhavebeenevaluatedtoensurethattheproposedspentfuelpoolmodificationdoesnotchangethepresentdegreeofassurancetopublichealthandsafety.Thefollowingaccidentsandmiscellaneousstructuralevaluationshavebeenconsidered:Refuelingaccident-DroppedFuelLocalCellWallBucklingAnalysisofWeldedJointsduetoIsolatedHotCellCraneUpliftLoad7~2efuelinAccidentsThissectionconsidersthree(3)accidentsassociatedwiththehandlingoffuelassemblies.7~2~1DroedFuelAssemblTheconsequencesofdroppinganeworspentfuelassemblyasitisbeingmovedoverstoredfuelisdiscussedbelow.a~DoedFueAssembAcc'dentAfuelassemblyisdroppedfrom36"abovethetopofastoragelocationandimpactsthebaseofthemodule.Localfailureofthebaseplateisacceptable;however,therackdesignshouldensurethatgrossstructuralfailuredoesnotoccurandthesubcriticalityoftheadjacentfuelassembliesisnotviolated.Calculated7-1 IIIIII b.resultsshowthattherewillbenochangeinthespacingbetweencells.Localdeformationofthebaseplateintheneighborhoodoftheimpactwilloccur,butthedroppedassemblywillbecontainedandnotimpacttheliner.Weshowthatthemaximummovementofthebaseplatetowardthelineraftertheimpactislessthan1.52".Theloadtransmittedtothelinerthroughthesupportbysuchanaccidentiswellbelowthatcausedbyseismicloads.DroaedFuelAssemblAccidentllOnefuelassemblyis(assumeddryweight.=1550lbs.)droppedfrom36"abovethetopoftherackandimpactsthetopoftherack.Thisisamoresevereconditionthanthecurrentlypostulateddropof1616lbs.fromaheightof15"abovethetopoftherack.Permanentdeformationoftherackisacceptable,butisrequiredtobelimitedtothetopregionsuchthattherackcross-sectionalgeometryatthelevelofthetopoftheactivefuel(andbelow)isnotaltered.Analysisshowsthatalthoughlocaldeformationoccurs,itisconfinedtoaregionabovetheactivefuelarea.Theregionofpermanentdeformationistoadepth5.34"below.thetopoftherack.C~DroaaedFuelAssemblyAccidentXIX.Thispostulatedaccidentisidenticalto(b)aboveexceptthatthefuelassemblyisassumedtodropinaninclinedmannerontopoftherack.Analysesshowthatthestraightdropcase(casebabove)boundstheresults.7.3LocalBuckl'nofFuelCelWallsThissubsectionandthenextonepresentsdetailsonthesecondarystressesproducedbybucklingandbytemperatureeffects.Theallowablelocalbuckling'tressesinthefuelcellwallsareobtainedbyusingclassicalplatebucklinganalysis.Thefollowingformulaforthecriticalstresshasbeenusedbasedonawidthofcell"b":(SeeFigure7.3.1.)pn2Et212b2(1-p2)7-2 III whereE=27.9x10psi,p~0.3,(Poison'sratio),t~075gb=8.75".Thefactorpissuggestedin(Ref.7.3.1)tobe4.0foralongpanel.Forthegiven'datacrcr~7411psiItshouldbenotedthatthisstabilitycalculationisbasedontheappliedstressbeinguniformalongtheentirelengthofthecellwall.-Intheactualfuelrack,thecompressivestresscomesfromconsiderationofoverallbendingoftherackstructuresduringaseismiceventandassuchisnegligibleattheracktopandmaximumattherackbottom.Itisconservativetoapplytheaboveequationtotherackcellwallifwecomparecrcrwiththemaximumcompressivestressanywhereinthecellwall.AsshowninSection6,thelocalbucklingstresslimitof7411psiisnotviolatedanywhereinthebodyoftherackmodules,sincethemaximumcompressivestressintheoutermostcellisa3585psi.(FromTable6.11.1forR6~.239,thestressatthebaseoftherackundercombineddirectplusbendingloadsiscr~R6xallowablestress).7.4AnalsisoWeldedJointsinRackduetoIsolatedHotCellInthissubsection,in-rackweldedjointsareexaminedundertheloadingconditionsarisingfromthermaleffectsduetoanisolatedhotcell.Athermalgradientbetweencellswilldevelopwhenanisolatedstoragelocationcontainsafuelassemblyemittingmaximumpostulatedheat,whilethesurroundinglocationsareempty.Wecanobtainaconservativeestimateofweldstressesalongthelengthofanisolatedhotcellbyconsideringabeamstrip(acellwall)uniformlyheatedandrestrainedfromgrowthalongonelong7~3 IIIIII

'Iedge.ThestripissubjecttoauniformtemperaturerisehT59.66F.Thetemperaturerisehasbeencalculatedfromthedifferenceofthemaximumlocalwatertemperatureandbulkwatertemperatureinthespentfuelpool.(seeTables5.5.1and5.7.1).Then,usingashearbeamtheory,wecancalculateanestimateofthemaximumvalueoftheaverageshearstressinthestrip(seeFigure7.4.1).Thefinalresultforwallmaximumshearstress,underconservativerestraintassumptionsisgivenasrmax=tEaT.931wherea=9.5x106in/in'FTherefore,'weobtainanestimateofmaximumweldshearstressinanisolatedhotcellas16984'Sincethisisasecondarythermalstress,itisappropriatetocomparethistotheallowableweldshearstressforafaultedeventr<.42Su=29820psi.Inthefuelrack,thismaximumstressoccursnearthetopoftherackanddoesnotinteractwithanyothercriticalstress.7.5CraneUliftLoadof3000lb.Alocalupliftloadof3000lb.(UFSARlimitis2950lb.)willnotinduceanyupliftstressesintherackwhicharemoreseverethanthelimitingconditionsdiscussedintheforegoing.Thischoiceofloadshouldbeanupperboundloadonthemaximumloadthatcanbeappliedtoastruckfuelassemblyduringremoval.7-4 IIIlIt~-I 7.6ReferencesforSection77.3.1"StrengthofMaterials",S.P.Timoshenko,3rdEdition,PartIZ,pp194-197(1956).7-5 I

-)cf,abFIGURE7~3.1LOADINGONRACKWALLHeatedCellWallKAWdF&WWW&WWWA'WeldLine'FIGURE7.4.1VfELDEDJOINTINRACK'-6 I

8'STATICANDDYNAMICANALYSISOFFUELPOOLSTRUCTURE8.1IntoductionTheDonaldC.Cookspentfuelpoolisasafetyrelated,seismiccategoryI,reinforcedconcretestructure.Inthissectionanabstractoftheanalysistodemonstratethestructuraladequacyofthepoolstructureispresented.TheobjectoftheanalysisistodemonstratethecomplianceofthepoolslabandconfiningwallstotheapplicabledesigncodesandtoNRCregulationsfortheconditionofincreasedloadingsduetohighdensityfuelstorage.Theloadingonthepoolstructureisproducedbythefollowingdiscretecomponents:a)StaticLoad'n1)Deadweightofpoolstructurepluspoolwater(includinghydraulicpressureonthepoolwalls).2)Deadweightoftherackmodulesandfuelassembliesstoredtherein.b).DnamicLoadin1)VerticalloadstransmittedbytheracksupportpedestalstotheslabduringaDBEorOBEevent.2)Inertialoadsduetotheslab,poolwallsandcontainedwatermasswhichariseduringaDBEorOBEevent.c)ealLoad'n1)Meantemperatureriseandtemperaturegradientacrossthepoolslabandthepoolwallsduetotemperaturedifferentialbetweenthepool,waterandtheatmosphereexternaltotheslabandwalls.8-1 II Thespentfuelpoolisanalyzedusingthefiniteelementmethod.TheresultsfortheaboveloadcomponentsarecombinedusingfactoredloadcombinationsmandatedbyNUREG-0800,theStandardReviewPlan(SRP),Section3.8.4(Ref.8.1.1).Itisdemonstratedthatforthecriticalfactoredloadcombinations,structuralintegrityismaintainedwhenthefuelpoolisassumedtobefullyloadedwithhighdensityfuelrackswithallstoragelocationsoccupiedbyfuelassemblies.ThegeneralpurposefiniteelementcodeANSYS(Ref.8.1.2)isutilizedtoperformtheanalysis.Thecriticalregionsexaminedarethefuelpoolslabandthemostcriticalwallsectionsadjoiningthepoolslab.Bothmomentandshearcapacitiesofthecriticalregionsarecheckedforstructuralintegrity.Alsoevaluatedislocalpunchingintegrityinthevicinityofafuelrackbearingpad.StructuralcapacityevaluationsarecarriedoutinaccordancewiththerequirementsoftheAmericanConcreteInstitute(ACI)(Refs.8.1.3and8.1.4).Inthisanalysis,theloadfactorsofSRPSection3.8.4havebeenusedtogetherwiththeallowableconcreteandreinforcementloadsascalledforbytheAmericanConcreteInstitute.Thisconstitutesthemostconservativeapproachtothestructuralqualificationofthepoolstructurebasedonastaticloadqualificationmethod.8.2GenealFeaturesotheModeThefuelpoolmodelisconstructedusinginformationfromdesignbasisDonaldC.Cookauxiliarybuildingstructuraldrawings.Adescriptionoftheportionofthepoolmodelledforanalysisisgiveninthefollowing.8-2 II~~~I~IL~~~

Thefuelpoolslabisa5'-21/2"thickreinforcedconcreteslabwithinsidedimensions39'-19/16"wideand58'-31/8"long.Theslabislocatedatelevation600'-605'-21/2"anditslongdirectionisalignedalongtheplantEast-Westdirection.TheEastedgeoftheslabhasa5'-2"thickverticalreinforcedwallwhichextendsabovetheslabandismodeledtolevel650'.TheWestedgeoftheslabhasa6'hickwallfromlevel,605'-21/2"tolevel650'.TheWestwallseparatesthefuelpoolfromthefueltransfercanalwhichisnotmodelled;however,thediscontinuityinthewallstructureinthecenteroftheWestwallisincluded.Allwallmodelingisdonetolevel650',andweassumefreeedgesatthislevel.TheNorthwallisa6'hickwallextendingfromtheslabtolevel650'.TheSouthedgeoftheslabhasa5'hickwallextendinguptolevel650'.ItisclearfromtheabovedescriptionthattheSouthwallhasthelargestlengthtothicknessratio,andtherefore,mayrepresentalimitingconditionofstructuralstrength.Thefoundationmatisatelevation584'ndthepoolslabandupperwallsaresupportedonthefoundationmatbywallsandcolumnsaroundtheperiphery.TheNorthedgeoftheslabissupportedbyacontinuous3'-0"thickwall,whiletheEastedgeispartiallysupportedalongitslengthbya2'-6"thickwall.TherearethreeverticalcolumnslocatedattheSoutheastandSouthwestcorneroftheslab,andintermediatealongtheSouthedge;ThereisalsoaportionofawallbelowtheSouthedgeatonelocation.Thefloorslabhasinteriorverticalsupportprovidedbya2'-0"thick'erticalwallprovidingverticalrestraintinboththeNorth-SouthandEast-Westdirectionoverasubstantiallengthofslab.Inaddition,thereisa25'panstandardW14x158wideflangebeamfromtheslabNorthedgesupportingwalltogiveadditionalpoolslabsupport.ThisproppedbeamisskewedtowardtheEast16'romtheNorthedge.8-3 lI Theentirebeam(thestraightpartplustheskewedpart)issupportedverticallybyfourTS10"x10"squaretubes.Eachtubularcolumn,hasalsobeenstiffenedbyfour8"x3/8"plates.Figure8.2.1showsaschematicoftheabovegeometry.Thepoolslabisassumedtobeloadedwith23highdensityfuelrackshavingatotalof3616cells.Foranalysispurposes,eachcellisassumedtocontaina1550lb.weightfuelassembly.Asnotedpreviously,all,fuelpoolwallsabovethepoolslabareassumedtohaveafreeedgeatlevel650'.Lateralrestraintisprovidedtotheverticalwallsatcertainlocationsabovethe605'evel.Thisrestraintsimulatestheeffectofadjacentstructurewhichisnotincludedinthemodelledenvelope.Figures8.2.2and8.2.3showlayoutsoftheentire3-Dfiniteelementmodel.Thegridworkindifferentregionsshowsthetotalityofelementsused.Shellelementsareusedtomodeltheslabandwalls,whilebeamelementsareusedtomodelthecolumns.ThefiniteelementmodelisconstructedusingtheANSYSclassicalshellelementSTIF63andthebeamelementSTIF44oftheANSYSfiniteelementcode.Theshellelementthicknessinthevariousregionsofthestructureistheactualthicknessofthestructureatthelocation.Thefiniteelementmodelispreparedfortheanalysisofbothmechanicalloadandthermalload.Theeffectsofthereinforcedconcrete(crackedoruncracked)areaccountedforinthefiniteelementmodelbyestablishinganappropriateeffectivemodulusforeachshellelementandeffectiveinertiasforthecolumnelements.Effectivemoduliaredefinedforeachlocalin-planeaxisfortheshellelements.Thedifferentmodulireflectthefactthatdifferentreinforcementgeometriesmaybeusedinperpendiculardirectionsoftheplate-likesectionswhen8-4 I!~I thedifferentconcretesectionassumptions(crackedoruncracked)areappliedtotheslabandwalls.Onlymajorreinforcementwhichaffectstheplateandshell-likebehaviorofthestructureisincorporatedintothedefinitionoftheeffectivemoduli;additional,localreinforcementinvariousareasofthepoolstructureareneglectedinthedefiningoftheeffectivemoduli.However,suchlocalreinforcementisaccountedforinthestrengthevaluationafterresultsareobtained.Thenon-homogeneousnatureofthereinforcementistakenintoaccountbydefiningdifferentmaterialtypesasnecessarytoreflectthevaryingvaluesofeffectivemoduliindifferentregions.Theconcretesectionassumptions(crackedoruncracked)arefullyinaccordancewiththe'equirementsofAmericanConcreteInstitute(Refs.8.1.3and8.1.4).InaccordancewithRef.8.1.4,weassumeuncrackedsectionpropertiesforthemechanicalloadanalyses(includingloadfactors).Forthethermalanalyses,itisshownthatthethermalgradientswillalwaysyieldacrackedsectioniftheuncrackedstiffnessisused;therefore,aniterativesolutionisusedtoshowthatcrackedsectionpropertiesshouldbeusedforthethermalanalyses.Theeffectivepropertiesfortheelementsusedinthefiniteelementmodelarecalculatedusingstandardproceduresforreinforcedconcretesectionstodefineequivalenteffectivehomogeneousmaterialshavingtheappropriatestiffnessandstrength.8-5 l

8.3LoadinConditionsZnordertoevaluatetheresponseduetothedifferentloadmechanismsoutlinedinSection8.1,thefollowingfiniteelementanalysesarecarriedout.Sixloadingcasesaredefinedbelowwhichenableustoobtainthemomentsandshearsforfactoredloadingsbylinearcombination.1.Deadloadingfromconcrete,reinforcementand40'fhydrostatichead.Theloadingisappliedasa1.0gverticalgravitationalloadforthestructureandasurfacepressureontheslabandwallsforthehydrostatichead.2.Deadloadingduetoweightsofrackplusfullfuelload.Theseloadsareappliedasauniformstaticpressureappliedtotheslab.3.Seismicverticalloadingduetoracksplusfuelloadappliedasaneffectivesustainedpressureonthefloorslabpedestals.Theloadingappliedisobtainedfromthe'-Dwholepoolmulti-rackanalysisdescribedinSection6ofthisreport.Fromtheresultsofthatanalysis,we.takethestoredtimehistoryofeachpedestalloadanddefineaneffectivesustainedpoolpressureloadwhichyieldsthesametotalimpulseoverthetimedurationoftheseismicevent.Thedetailsofdevelopingthiseffectivesustainedpressureloadarepresentedlater.WedevelopeffectivesustainedverticalpressureloadsforbothOBEandDBEeventsandthenperformappropriatefiniteelementanalyses.4~Seismichorizontalloadingduetostructureweight(includingreinforcement).Theloadingisappliedasa1ghorizontalandverticalaccelerationappliedtothestructureplusahydrodynamicpressureequivalenttoanaccelerationofallofthewatermassagainsttheweakestwall.Theaccelerationlevelisobtainedfromtheapplicableresponsespectraandistakenasthepeakglevelonthespectraatfrequenciesabovethelowestnaturalfrequencyforthestructure.AseparateANSYSfrequencyanalysissimulationiscarriedouttoestablishthedynamiccharacteristicsofthestructure.'8-6 IIerI Seismichorizontalloadduetoshearloadsfromeachofthepedestals.Thisloadingisobtainedbyusingthestatic+effectivedynamicloadsdevelopedforcase3aboveandassumingacoefficientoffriction=.8.Thedirectionoftheseloadsissetsoastodevelopstressesthatmaximizetheloadcombinationsnecessarytosatisfystructuralintegrityrequirementsdiscussedbelow.Znthisloadcasewealsoimposealateralpressureontheweakestpoolwalltosimulatehydrodynamiceffectsfromfluidcouplingduetorackmotionrelativetothewall.6.Ameantemperatureriseplusathermalgradientisappliedacrossthewallsandfloorslabtosimulatetheheatingeffectofthewaterinthepool.ThisgradientiscalculatedbasedonthemaximumwalltemperaturededucedfromthepoolbulktemperaturecalculationsforthelicensingbasisscenariospresentedinSection5ofthisreport.Forsubsequentdiscussionofstructuralintegritychecksusingvariousmandatedloadcombinations,werefertotheaboveindividualfiniteelementloadcasesas"case1-6",respectively.Asnotedabove,inadditiontothestaticanalysesusingthedevelopedfiniteelementmodel,wealsoperformafrequencyanalysisofthepoolstructureassumingthatallcontainedfluidisattachedtothepoolslab.Uncrackedsectionpropertiesareusedhere.Thisfrequencyanalysisisusedtodeterminethelowestpoolstructuralfrequenciessoastoestablishappropriateseismicamplifierstoapplytoloadcases1and4.Theseseismic8-7 lII amplifiersareobtainedfromtheresponsespectraoftheseismiceventandmultiplytheresultsofloadcases1and4whenformingthemandatedloadcombinations.Asnotedabove,thecase3loadinginvolvesthedeterminationofaneffectivepressureloadtorepresenttheseismicloadontheslabduetotheracksplusfuel.Themethodofdeterminationofthiseffectivepressureisdescribedbelow.Asnotedpreviously,theHoltec3<<DdynamicsimulationcodeDYNARACKisusedtosimulatetheseismicresponseoftheentirefuelpoolcontainingmultipleracks.Theverticalloadtimehistoryfromeachpedestaloneachrackissavedinanarchivalfile.Forthepoolslabstructuralanalysis,whichisbasedonstaticanalyses,wecomputeaneffectivestaticloadincrementbasedonaveragingofthetimehistory.Figure8.3.1isusedtoillustratetheconceptwherethetotalpedestalloadisconsideredasthestaticload(FsinFigure8.3.1plusatimevaryingcomponent).NotethatinFigure8.3.1azeroloadduringaportionofthetimemeansthatthepedestalhasliftedoff.Wedefineaneffectivestaticloadforthepurposesofpoolstaticanalysisandstructuralqualificationasfollows:a0b.C~Promthearchivalpedestalloadtimehistorywemay,ateachpointintime,determinethetotalpoolloadFTbysummingthetotalloadsforeachpedestal.Ateachpointintimei,wecandefinethedynamicloadincrementforthepoolasFT-FS=DFiwhereFSnowrepresentsthetotalstaticloadontheslab.WekeeptrackofthenumberoftimepointsiwhereDFi)0.Anequivalent.staticpoolload(seismicaddertothestaticpoolload)isdefinedasSEISMICADDERSUMDFI/SUMNIi8-8

whereSUMDFZisthesumofallofthenonzeroDF1andSUMNiisthetotalnumberofpointsinthetimehistorywherethedynamicpoolloadincrementisgreaterthanzero.d~Znformingtheappropriateloadcombinationsmandatedforstructuralintegritychecks,thecalculated"seismicadder"dividedbythepoolarea,isusedastheeffectiveseismicpressureontheslab.OfallloadingconditionsmandatedinRef.8.1.1,thefactoredloadswhichapplytothisstructureandaredeemedcriticalare:A.1'D+1'EB.~75(1.4D+1.9E+1.7To)CDD+E'Towhere:D~DeadloadE'DesignBasisEarthquake(DBE)E~OperatingBasisEarthquake(OBE)To=SteadyStateThermalLoadTheappropriateloadcasesareformedfromtheindividualfiniteelementanalysesasfollows:D~case1+case2E'DBEamplifierxcase1+DBEamplifierxcase4+case3(forDBE)+case5(forDBE)E~OBEamplifierxcase1+OBEamplifier+case4+case3(forOBE)+case5(forOBE)Tocase6Loadcombinationsareformedusingabsolutevalueswherenecessarysoastomaximizecriticalstressresultants.8-9 II Asnotedabove,foranalysisoffuelpoolstructuralintegrity,theseismicamplifiersarebasedonthepeakglevelresponsesatthelowestresonantfrequencythatareobtainedfromtheplantaccelerationresponsespectrum.Weshowthatthisisconservative.8.4ResultsoAalsesTheANSYSpostprocessingcapabilityisusedtoformtheappropriateloadcombinationsidentifiedaboveandtoestablishthecriticalbendingmomentsinvarioussectionsofthepoolstructure.TheultimatemomentsforeachsectionarecomputedusingallowablelimitstrengthlevelsasdescribedinRef.8.1.3.ForDonaldC.Cook,thefollowinglimitstrengthsforconcreteandforreinforcementareusedinthecomputationoflimit(ultimate)moments.concreteo'c~3500psi(compression)reinforcement~cry~40000psi(tension/compression)Zneachsection,wedefinethesafetymarginforbendingastheultimatebendingmomentdividedbythecalculatedbendingmoment(fromtheANSYSpostprocessingoftherequiredloadcases).Table8.4~1summarizestheresultsobtainedfromthefiniteelementanalysesandshowsminimumsafetymarginsoneachsectionofthestructure.NotethatthesearesafetymarginsbasedonthefactoredloadconditionsasmandatedinRef.8.1.1andneedonlysatisfyalimit~1.0.8-10 Il Thefloorslabperimeterisalsocheckedagainstgrossshearfailureunderfactoredloadconditions.Localbearingstrengthandpunchingshearcalculationsareperformedinaccordancewith(Reft8.1.3)~Thepoollinerissubjecttoin-planestrainsduetomovementoftheracksupportfeetduringtheseismicevent.Calculationsaremadetoestablishthatthelinerwillnotfailduetocyclicstrainingcausedbytherackfootloading.AnANSYSanalysisofalinerplatesectionsubjectedtoverticalandhorizontalstaticpedestalloadingiscarriedout.Thetimehistoryresultforthepedestalloadingisthenusedtoevaluatethenumberofstresscyclestobeexpectedinthelinerforeachevent.Thecumulativedamagefactor(CDF)iscomputedandshowntobelessthan1.0incriticalregionsofthelinerandattachmentlocations.The@umberofstresscycles'sedintheCDFevaluationisbasedon1DBEand20OBEevents.Criticalregionsaffectedbyloadingthefuelpoolcompletelywithhighdensityracksareexaminedforstructuralintegrityunderbendingandshearingaction.Itisdeterminedthatadequatesafetyfactorsexist.assumingthatallracksarefullyloadedwithnormal(unconsolidated)fuelandthatthefactoredloadcombinationsare'heckedagainsttheappropriatestructuraldesignstrengths.Itisalsoshownthatlocalfrictionalloadingonthelinerresultsinin-planestressesthatarelowenoughsothatlinerfatigueisnotaconcern.8-11 aI 8.7ReferencesfoSection88.F18.1.28.1.3F1.4NUREG-0800/SRPforReviewofSafetyAnalysisReportsforNuclearPowerPlants,Section3.8.4,July1981.ANSYSUser'sManual,SwansonAnalysisRev.4.3,1987.ACI318-89,ACI318R-89,BuildingCodeRequirementsforReinforcedConcrete,AmericanConcreteInstitute,Detroit,Michigan.ACI349.1R-80,ReinforcedConcreteDesignforThermalEffectsonNuclear.PowerPlantStructures,1981.8>>12 IIIIII Table8.4.1SAFETYFACTORSFORBENDINGOFPOOLSTRUCTUREREGIONSREGIONFACTOROFSAFETYSlabNorthWallEastWallSouthWallWestWall1.231.081.261.05Thefactorsofsafetyhavebeenobtainedusingconservativeassumptionsonmechanicalandthermalloaddistribution.TheyrepresentfactorsofsafetyoverthevaluesrequiredbyNUREG-0800.8-13 III EASTWALLNORALLSOUTALLIII~rrr~rrretWESTWALL~err~IrrrrrrrrrI~~I~Irt.~~~~~~~III~s~rrt~~rsr'~I~~~~~~rt~~I~srt~~aeseeespasoSesptagIV~~I~rr~~~r~~~r~,SrII~~~~sIIII~~~I~~~~~~~I~~Irrtr~r>r~r~~II~I~I~~~~~~I~I~~~~rFIGURE8.2.1ISOMETRICVIEWOFCOOKSPENTFUELPOOL8-14 IIIIIII FIGURE8.2.2OVERALLFINITEMODELOFCOOKPOOLTOPVIEN8-15 III FIGURE8.2.3OVERALLFINITEMODELOFCOOKPOOLBOTTOMVIEN8-16 IIII FIGURE8.3.1PEDESTALLOADVS.TIME(PositiveLoadMeansPedestalinContactwithLiner)8-l7 II 9.0RADIOLOGICALEVALUATION9.1FuelHandlinAccident9.1.1AssumtionsandSourceTermCalculationsAnevaluationoftheconsequencesofafuelhandlingaccidenthasbeenmadeforfuelof5.0wt4initial-enrichmentburnedto60,000MWD/MTUgwiththereactorconservativelyassumedtohavebeenoperatingat3411MWthermalpower(38.8MWD/KgUspecificpower)priortoreactorshutdown.Exceptforthefuelenrichmentanddischargeburnup,theassumptionsusedintheevaluationarethesameasthosepreviouslyreviewedandacceptedbytheUSNRC.Asr,,inthe,previousevaluation,thefuelhandlingaccidentwasconservativelyassumedtoresultinthereleaseofthegaseousfissionproductscontainedinthefuel-rodgapsofalltherodsinthepeak-powerfuelassemblyatthetimeoftheaccident.GapinventoriesoffissionproductsavailableforreleasewereestimatedusingboththeassumptionsidentifiedinRegulatoryGuide1.25+andthoseinNUREG/CR-5009().NUREG/CR-5009"'hasconfirmedthattheRegGuide1.25assumptionsremainconservativeforextendedburnupexceptforI-131,forwhichthereleasefractionwasreportedtobe20%higher.Mostofthegaseousfissionproductshavingasignificantimpactontheoff-sitedosesaretheshort-livednuclidesofIodineandXenonwhichreachsaturationinventori'esduringin-coreoperation.Theseinventoriesdependprimarilyonthefuelspecificpoweroverthefewmonthsimmediatelyprecedingreactorshutdown.Inthehighestpowerassembly,thespecificpowerandhencetheinventoryofIodineandXenonwillbedirectlyrelatedtothepeakingfactor(assumedtobe1.65perReg.Guide1.25).9-1 I1lII Theinventoryoflong-livedKr-85(10.73yearhalf-life),however,isnearlyproportionaltotheaccumulatedfueldischargeburnupandhenceisindependentofthepeakingfactor.BecauseKr-85isaweakbetaemitter,ithasonlyaminorimpactonoff-sitedoses,primarilyaffectingthewhole-bodybetadose.Theoff-siteradiologicalconsequencesaredominatedbytheshort-livedradionuclides(whichareatsaturationconcentrationindependentoffuelburnup).Inthepresentanalysis,thecalculateddosesarehigherandmorecoservativethanthoseofthepreviousevaluationbecause(1)theanalysesreportedhereusehighergapinventoriesbasedonRegGuide1.25assumptionsand(2)theuseoftheup-datedORIGEN-2code<>forcalculatingthefissionproductinventories.Resultsoftheevaluationconfirmthattheoff-sitedosesremainwithintheregulatorylimits.Thepresentevaluationusesvaluesforthe2-houratmosphericdispersionfactor(X/Q)andfilterefficienciesthathavepreviouslybeenreviewedandaccepted.CoreinventoriesoffissionproductswereestimatedwiththeORIGEN-2codebaseduponareactorpowerof3411MWtandfuelwithaninitialenrichmentof5.0:U-235burnedto60,000NWD/MTU.Calculationsweremadefor100hourscoolingtimeasthesourcetermforthefuelhandlingaccident.ThereleasefractionofthecoreinventoriesassumedtobeinthegapbyboththeRegGuide1.25andNUREG/CR5009assumptionsarelistedinTable9.1.Thefollowingequation,from-RegGuide1.25,wasusedtocalculatethethyroiddose(D)fromtheinhalationofradioiodine,D=ZFgI;FPBR;(x/Q)DFpDFgRadssummedoverallIodineradionuclides.9-2.

~~~I~~~I FSfractionoffuelrodIodineinventoryingapspacecoreIodineradio-nu-clideinventoryattimeoftheaccident(cu-ries)fractionofcoredam-agedsoastoreleaseIodineintherodgap(1/193)Breathingrate=3.47x10cubicmeterspersecondDoseconversionfactor(rads/curie)fromReg.Guide1.25(X/Q)=atmosphericdiffusionfactor(3.15x10sec/m)Corepeakingfactor(1.65)DF<=effectiveIodinedecontaminationfactorforfilters(=10)DFp=effectiveIodinedecontaminationfactorforpoolwater(=150)ThegapinventorieslistedinTable9-1aretheproductofI;(coreinventory)andFz(thefractionexistinginthegap).Thefunctionusedtocalculatetheexternalwholebodydosefrombeta(D,)orgamma(D<)radiationinthecloudusesmanyofthetermsdefinedaboveandisgivenby:D~=Z0.23(x/Q)FPG<E~andDfZ0~25(X/Q)FPGiEwhereG<isthegapinventoryofthegaseousradionuclidesofXeandKrandthefunctionsabovearesummedoverallthenoblegases.E>andE<aretheaverageenergiesofdecay(betaandgammaradiationrespectively)forthevariousradionuclides.Thesefunctionsassumethenoblegasdecontaminationfactorsinwaterandthecharcoalfiltersare1.0.Thegapinventoriesofradioiodine9-3 C~l~~~I makeanegligiblecontributiontothewholebodydoses,D~orD<,becauseofthelargedecontaminationfactorsappropriatetotheiodines.9.1.2ResultsAsummaryoftheassumptionsusedtoevaluatethefuelhandlingaccidentisgiveninTable9-2.Theminimumtimeaftershutdownwhenfuelassemblieswouldbemovedwasconservativelyassumedtobe100hoursasidentifiedintheTechnicalSpecifications.At100hoursaftershutdown,thetwo-hourdoseatthesiteboundary,forafuelhandlingaccidentreleasingallofthegaseousfissionproductradioactivityinthegapsofallrodsinthehighestpowerassembly,areasfollows:Two-HourSiteBoundarDoseNUREG/CR-5009MethodReg.GuidePrevious1.25A~nalsisInhalationthyroiddose=7.07RadsWholebodybetadose,Dp=0.36RadsWholebodygammadose,D<=0.31Rads5.97Rads0.70Rads0.58Rads2.150.51Thesedosesarewellwithinthelimitsof10CFRPart100inconformancewiththeacceptancecriteriaofSRP15.7.4.(Rev.1,July1981)9-4

9.2SolidRadwasteThenecessityforresinreplacementisdeterminedprimarilybytherequirementforwaterclarityandtheresinisnormallychangedaboutonceayear.Nosignificantincreaseinthevolumeofsolidradioactivewastesisexpectedwiththeexpandedstoragecapacity.Duringrerackingoperations,acertainamountofadditionalresinsmaybegeneratedbythepoolcleanupsystemonaone-timebasis(perhaps10to30cubicfeet).,9.3GaseousReleasesGaseousreleasesfromthefuelstorageareaoftheauxiliarybuildingarecombinedwithotherplantexhausts.Normally,thecontributionfromthefuelstorageareaoftheauxiliarybuildingisnegligiblecomparedtotheotherreleasesandnosignificantincreasesareexpectedasaresultoftheexpandedstoragecapacity.9.4PersonnelExosuresDuringnormaloperations,personnelworkinginthefuelstorageareamaybeexposedtoradiationfromthespentfuelpool.Operatingexperiencehasshownthatthearearadiationdoserates,whichoriginateprimarilyfromradionuclidesinthepoolwater,aregenerallylessthan1mrem/hrbutmaytemporarilyincreaseto2.5-3mrem/hrduringrefuelingoperations.Noevidencehasbeenobservedofanycruddepositionaroundtheedgesofthepoolthatmightcauselocalareasofhighradiation.9-5

Radiationlevelsinzonessurroundingthepoolarenotexpectedtobesignificantlyaffected.Existingshieldingaroundthepool(waterdepthandconcretewalls)providemorethanadequateprotec-tion,despitetheslightlycloserapproachtothewallsofthepool+TypicalconcentrationsofradionuclidesinthepoolwaterareshowninTable9.3.Duringfuelreloadoperations,theconcentrationswillincreasedue.tocruddepositsspallingfromspentfuelassembliesandtoactivitiescarriedintothepoolfromtheprimarysystem.Whiletheseeffectsmayincreasetheconcentrations(asmuchasafactorof10),thepoolcleanupsystemsoonreducestheconcentrationstothenormaloperatingrange.Noevidencehasbeenseenofanysignificantlyhigherradiationdosesneartheedgeofthepoolthatmightsuggesttheaccumulationofcruddeposits.Operatingexperiencehasshownthattherehavebeennegligibleconcentrationsofairborneradioactivityandnoincreasesareexpectedasaresultoftheexpandedstoragecapacity.Areamonitorsforairborneactivitiesareavailableintheimmediatevicinityofthespentfuelpool.Noincreaseinradiationexposuretooperatingpersonnelisexpectedand,thereforeneitherthecurrenthealthphysicsprogramnortheareamonitoringsystemsneedtobemodified.9.5AnticiatedEosureDurinRerackinTotaloccupationalexposureforthererackingoperationisestimatedtobebetween6and11person-rem,asindicatedinTable9.4.WhileindividualtaskeffortsandexposuresmaydifferfromthoseinTable9.4,thetotalisbelievedtobeareasonableestimateforplanningpurposes.Diverswillbenecessarytoremove9-6 l1,J~.

certainunderwaterappurtenances.Theseappurtenancesarewellremovedforthestoredfuelwhichminimizestheradiationdoseratetothedivers.Carefulmonitoringandadherencetopre-preparedprocedureswillassurethattheradiationdosetothediverswillbemaintainedALARA.AllofthererackingoperationwillutilizedetailedprocedurespreparedwithfullconsiderationofALARAprinciples.SimilaroperationshavebeenperformedinanumberoffacilitiesinthepastandthereiseveryreasontobelievethatrerackingcanbesafelyandefficientlyaccomplishedattheDonaldC.CookNuclearPlant,withminimumradiationexposuretopersonnel.TheexistingradiationprotectionprogramattheCookNuclearPlantisadequateforthererackingoperations.Wherethereisapotentialforsignificantairborneactivity,continuousairsamplerswillbeinoperation.Personnelwearprotectiveclothingand,ifnecessary,respiratoryprotectiveequipment.ActivitiesaregovernedbyaRadiationWorkPermitandpersonnelmonitoringequipmentwillbeassignedtoeachindividual.Asaminimum,thisincludesthermoluminescentdosimetersandpocketdosimeters.~Additionalpersonnelmonitoringequipment(i.e.,extremitybadgesoralarmingdosimetersmaybeutilizedasrequired.Work,personneltraffic,andthemovementofequipmentwillbemonitoredandcontrolledtominimizecontaminationandtoassurethatexposuresaremaintainedALARA.Inreracking,theexistingstoragerackswillberemoved,decon-taminatedasmuchaspossiblebywashingandwipe-downs,packagedandshippedtoalicensedprocessing/disposalfacility.ShippingcontainersandprocedureswillconformtoFederalDOTregulationsandtherequirementsofanyStateDOTofficethroughwhichtheshipmentmaypass.9-7

9.6ReferencesforSection9Reg.Guide1.25(AECSafetyGuide25),"Assumptionsusedforevaluatingthepotentialradiologicalconsequencesofafuelhandlingaccidentinthefuelhandlingandstoragefacilityforboilingandpressurizedwaterreactors".2~C.E.Beyer,etal.,"AssessmentoftheUseofExtendedBurnupFuelinLightWaterPowerReactors",NUREG/CR-5009,PacificNorthwestLaboratory(PNL-6258).3.A.G.Croff,"AUser'sManualfortheORIGEN2ComputerCode",ORNL/TM-7175,July1980(ORIGEN=ORNLIsotopeGenerationandDepletion)Section15.7.4,"RadiologicalConsequencesofFuelHandlingAccidents"NUREG-0800,Section15.7.4,Rev.1July19819-8 III Table9-1INVENTORIESANDCONSTANTSOFSIGNIFICANTFISSIONPRODUCTRADIONUCLIDESNUCLIDESHUTDOWNCOREINVENTORYCURIESDECAYCONST.X,1/HRS100hrs100hrsTOTALGAPINVENTORY,CURIESNUREG/CR-5009Reg.Guide1.25DOSECONVERSIONRiE(MEV)PE(MEV)7l-131l-1329.0E+71.3B.83.591E-33.013E-17.5E+6Negligible~6.3E+6Negligible1.48E+65.35E+40.1860.389I-1331.8E+83.332E-26.3B56.3B540E+50.4190.597I-134I-1351.9E+81.7E+87.905E-11.048E-1NegligibleNegligibleNegligibleNegligible2.5E+41.24E+50.3941.456Kr-85MKr-85Kr-87Kr-881.9E+71.4E+63.6E+75.0E+71.547E-17.376E-65.451E-12.442E-1Negligible2.0E+5NegligibleNegligibleNegligible4.2E+5NegligibleNegligible0.2510.002Xe-131MXe-133MXe-133Xe-1351.0E+65.6E+61.8E+83.9E+72.427E-31.319E-25.506E-37.626E-27.9E+41.5E+55.1E+6Negligible7.9E+41.5E+51.0E+7Negligible0.1630.2330.1020.0810.3090.262'ORELEASEFRACTIONGIVEN-ASSUMEDSAMEASREG.GUIDE1.25 II Table9.2DATAANDASSUMPTIONSFORTHEEVALUATIONOFTHEFUELHANDLINGACCIDENTSourceTermAssumtionsVALUESCorepowerlevel,MWTFuelburnup,MHD/MTUAnalyticalmethodReleaseAssumtions341160,000ORIGENNumberoffailedfuelrodsFractionofcoreinventoryreleasedtogap(NUREG/CR-5009releaseofIodine-131isreportedtobe20%higher)AssumedpowerpeakingfactorInventoryingapavailableforreleasePooldecontaminationfactorsallrodsin1of193assembliesRe.Guide1.254oftheIodine-104oftheXenon-104ofKr-85301.65Table9.1ForIodinesFornoblegasesFilterdecontaminationfactorsForIodinesFornoblegasesAtmosphericDispersion,(x/Q)Breathingrate15011013.15x10sec/m3.47x10m/sec9-10 IlI Table9.3TypicalConcentrationsofRadionuclidesintheSpentFuelPoolWaterConcentrationNuclideAg-110MCo-58Co-60Cs-134Cs-137~C;~ml4.6x101'x104.4x103.2x106.4x109-11 I-II Table9.4PRELIMINARYESTIMATEOFPERSON-REMEXPOSURESDURINGRERACKING~SteNumberofPersonnelHoursEstimatedExosure<>RemoveemptyracksWashandDeconracksCleanandVacuumPoolRemoveunderwaterappurtencesPartialinstallationofnewrackmodules401025200.5to1.00.08to0.20.3to0.60.4to0.80.25to0.5MovefueltonewracksRemoveremainingracksWashandDeconracks150120300.8to1.51.5to3.00.2to0.4InstallremainingnewrackmodulesPrepareoldracksforshipment35800.4to0.81.0to2.0<'iTotalExposure,person-rem6to12Assumesminimumdoserateof21/2mR/hr(expected)toamaximumof5mR/hr,exceptforpoolvacuumingoperationswhichassumes4to8mR/hranddivingoperationswhichassume20to40mR/hr.Maximumexpectedexposure,althoughdetailsofpreparationandpackagingofoldracksforshipmenthavenotyetbeendeter-mined.9-12 lI 10.0IN-SERVICESURVEILLANCEPROGRAM10.1~PuroseThissectiondescribestheprogrammaticcommitmentsmadebyIndianaMichiganPowerCompany(I&M)forin-servicesurveillanceoftheBoralneutronabsorptionmaterialtocomplywiththeprovisionsofSectionIV(8)oftheOTPositionPaper(Ref.10.1.1).Allmaterialusedwithinastoragesystemforspentnuclearfuelarequalifiedtoalevelofperformancepredicateduponcalculatedworstcaseenvironmentalconditionsandarebasedonacceleratedtestingofthematerialstolevelsofservicelifecorrespondingtothatenvironment.Becausesuchenvironmentalcompatibilitytestinginthelaboratoryconditionsisaccelerated,itisprudentthateachofthesystemcomponentsbemonitoredtosomeextentthroughouttheservicelifetoassurethattheactualin-serviceperformanceremainswithinacceptableparametersasdefinedbytheacceleratedtesting.Formanyofthematerials,monitoringthroughouttheservicelifeisrelativelyeasy,however,theneutronabsorbingmaterialisencasedinastainlesssteeljacketprecludingadirectvisualorphysicalexaminationduringthein-servicecondition.Thecouponsurveillanceprogrampresentedhereinisintendedtoprovideadefinitiveassessmentofthepresentphysicalintegrityoftheneutronabsorber',aswellasinferentialinformationtodetectfuturedegradation.10-1 i

Thecouponsurveillanceprocedureconsistsofpreparingtwelveneutronabsorbercouponscarefully,encasedinastainlesssteelmetaljacket,andsuspendingthemfroma"coupontree".Thecoupontreeisplacedinthecenterofagroupoffreshlydischargedfuelassemblieseachtimeanewbatchisdischargedtothepool.Thegroupofassembliessurroundingthecoupontreeshallbethosewhichhavetheabove-averagevaluesofradialpeakingfactor.Theobject,ofcourse,istosubjectthis"tree"tothemaximumradiationexposureinthefuelpoolintheminimumamountoftime.Furtherdetailsareprovidedinthefollowing.10.2CouonSurveillance10.2.1DescritionofTestCouonsTheneutronabsorberusedinthesurveillanceprogramshallberepresentativeofthematerialusedwithinthestoragesystem.Itshallbeofthesamecomposition,producedbythesamemethod,andcertifiedtothesamecriteriaastheproductionlotneutronabsorber.ThesamplecouponshallbethesamethicknessastheneutronabsorberusedwithinthestoragesystemandshallmeetthereferencedHoltecdrawingdimensionalrequirements.Eachneutronabsorberspecimenshallbeencasedinastainlesssteeljacketofanalloyidenticaltothatusedinthestoragesystem,formedsoastoencasetheneutronabsorbingmaterialandfixitinapositionandwithtolerancessimilartothatforthestorageracks.Thejacketwouldbesimilartothatforthestorageracks.10-2 Il'lI~~I1 Thejacketwouldbeclosedbyquickdisconnectclampsorscrewswithlocknutsinsuchamannerastoretainitsformthroughouttheuseperiodandalsoallowrapidandeasyopeningwithout.contributingmechanicaldamagetotheneutronabsorberspecimencontainedtherein.Consistent.withtheUSNRCOTPositionPaper[reference10.1.1],requirementsofastatisticallyacceptablesamplesize,atotaloftwelvejacketedneutronabsorberspecimens,shallbeused.10.2.2BenchmarkDatThefollowingbenchmarktestsshallbeperformedontestcouponsderivedfromthesameproductionrunastheactualneutronabsorberpanels.(i)(ii)(iii)Length,width,thicknessandweight.measurementsWetchemistryNeutronattenuationmeasurement(optional)10.2.3CouonReferenceDataPriortoencasingthecoupons,eachcouponshallbecarefullycalibrated.Theirwidth,thickness,lengthandweightshallbecarefullymeasuredandrecorded.ThewetchemistrywillbeperformedonastriptakenfromthesameBoralplatesfromwhichthecouponsaremadetoprovideabenchmarkB-10loadingdata.Threepointsoneachcouponwillbedesignatedforneutronattenuationmeasurement.Neutronattenuationmeasurementsatthosethreepointswillbemadeandrecorded.10-3 IIiIIII 10.2.4AcceleratedSurveillanceAtthetimeofthefirstoff-loadofspentfuel,thecoupontreeissurroundedbystoragecellscontainingfuelassembliesfromthepeakpowerregionofthereactorcore.Atthetimeofthesecondoff-loadofthefuelassemblies,thetreeiswithdrawnfromthefuelpoolandonecouponistakenforevaluation.Thespecimenstripisreplacedinthefuelpoolinanewlocation,whereitisagainsurroundedbypeakpowerregionfuelassemblies.Thestoragecellthatwasvacatedmaynowbeusedtostoreafuelassembly.Thisarrangementisrepeatedatthefirsttwooff-loadsoffuelandafterthat,everythirdoutage.Byevaluationofthespecimens,anacceleratedmonitorofenvironmentaleffectsontheneutronabsorberwillbeobtained.10.2.5Post-ErradiationTestsCouponsremovedfrom,thepoolwillbet'estedfordimensional,neutronattenuation,andwetchemistrychangesusingthesameprocedureswhichwereusedininitialbenchmarkingtominimizethepotentialforinstrumenterrors.10.2.6AccetanceCriteriaAplantprocedurewillbedevelopedtoexecutethecommitmentsmadeinthislicensingsubmittal.Equipmentrequirements,step-by-stepinstructionsforexecutinginspectionsandacceptancecriteriawillbedescribedinthatprocedureforusebyplantpersonnel.10-4 II5l ReferencesforSecton10OTPositionforReviewandAcceptanceofSpentFuelStorageandHandlingApplications",byBrianK.Grimes,USNRC,April14,1978,andRevisiondatedJanuary18,1979.10-5 IIIOI 11.011~1ENVIRONMENTALCOSTBENEFITASSESSMENTIntr'oductionThespecificneedtoincreasetheexistingstoragecapacityofthespentfuelpoolattheDonaldC.CookNuclearPlantisbasedonthecontinually-increasinginventoryinthepool,theprudentrequirementtomaintainfull-coreoffloadcapability,andalackofviableeconomicalternatives.TheinventoryincreasecanbeinferredfromthefuelassemblydischargeschedulecontainedinTable-11.1.Theproposedprojectcontemplatesthererackingofspentfuelpoolwithfree-standing,highdensity,poisonedspentfuelracks.Theengineeringdesignandlicensingwillbecompletedforafullrerackingofthepool,whichiscurrentlyonlypartiallyracked.Engineeringanddesignwillalsobecompletedtoaccommodateconsolidatedfuel.Thelicensingeffortforconsolidatedfuelwill,however,bepursuedatalaterdateifconsolidationischosentoaccommodatefuturestorageneeds.11'ProectCostAssessmetThetotalcapitalcostforthererackproject.isestimatedtobeapproximately$14.1million.Manyalternativeswereconsideredpriortoproceedingwithreracking,whichisnottheonlytechnicaloptionavailabletoincreaseon-sitestoragecapacity.Rerackingdoes,however,enjoyacostadvantageoverothertechnologies,asshown:11-1 II TeofStoaeRerackFuelconsolidationDrycaskstorageStoragevaultNewpoolCapitalCostsKUg20(1)S20-34(')$45-110(')$40-90()$115(')Therearenoacceptablealternativestodevelopoff-sitespentfuelstorag~capacityfortheCookNuclearPlant.First,therearenocommercialindependentspentfuelstoragefacilitiesoperatingintheU.S.Second,theadoptionoftheNuclearWastePolicyAct(NWPA)createdadefactothrow-awaynuclearfuelcycle.Sincethecostofspentfuelreprocessingisnotoffsetbythesalvagevalueoftheresidualuranium,reprocessingrepresentsanaddedcostforthenuclearfuelcyclewhichalreadyincludestheNWPANuclearWasteFundfees.Inanyevent,therearenodomesticreprocessingfacilities.Third,I&Mhasnootheroperatingpowerplant;therefore,shipmentofspentfuelfromtheCookNuclearPlanttoothersystemnuclearpowerplantsisnotpossible.Fourth,at$600,000perdayreplacementpowercost,shuttingdowntheCookNuclearPlantismanytimesmoreexpensivethansimplyrerackingtheexistingspentfuelpools.FromEPRINF-3580,May1984FromDOERW-0220,"FinalVersionDryCaskStorageStudy,"February1989ActualestimatedcostperKgUofstoragespacegainedforthisproject11-2 IIIII ResourceCommitmentTheexpansionofthespentfuelpoolcapacityisexpectedtorequirethefollowingprimaryresources:Stainlesssteel360tons.BoralNeutronAbsorber30tons,ofwhich30tonsareBoronCarbidePowderand20tonsarealuminum.Therequirementsforstainlesssteelandaluminumrepresentasmallfractionoftotalworldoutputofthesemetals(lessthan.0001.).AlthoughthefractionofworldproductionofBoronCarbiderequiredforthefabricationissomewhathigherthanthato'fstainlesssteeloraluminum,itisunlikelythatthecommitmentofBoronCarbidetothisprojectwillaffectotheralternatives.ExperiencehasshownthattheproductionofBoronCarbideishighlyvariableanddependsuponneed,andcaneasilybeexpandedtoaccommodateworldwideneeds.11.4EnvironmentAssessmentDuetotheadditionalheat-loadarisingfromincreasedspentfuelpoolinventory,theanticipatedmaximumbulkpooltemperatureincreasesfromapreviously-licensed140'Ftoapproximately160'F,asdetailedinthecalculationsdescribedinSection5.0ofthisreport.Theresultanttotalheat-load(worst,case)is35.5millionBTU/HR,whichislessthan0.54ofthetotalplantheatlosstotheenvironment.Thenetresultoftheincreasedheatlossandwatervaporemission(duetoincreasedevaporation)totheenvironmentisnegligible.11-3 IIIIIII Table11.1DONALDC.COOKNUCLEARPLANTHORSTCASESPENTFUELINVENTORYASSEMBLIESItLssTORAB',19911992199319941995199619971998199920002001200220032004200520062007200820092010201120122013201420152016201713621518167818381918LoseMlcoredischargecapabilitywithcurrentcapacity19982158Losenormaldischargecapabilitywithcurrentcapacity23182318247826382798279829583118319832783438Losefullcoredischargecapabilitywithproposedrerack35983678Losenormaldischargecapabilitywithproposedrerack375839184078415843514431462411-4 IIIIIIII