ML17332A790

From kanterella
Revision as of 08:02, 29 June 2018 by StriderTol (talk | contribs) (Created page by program invented by StriderTol)
Jump to navigation Jump to search
Proposed Tech Specs Re Analyses Supported Increases in Unit 1 SG Tube Plugging Limit & Maintenance of Consistency of Unit 2 Acceptance Criteria
ML17332A790
Person / Time
Site: Cook  American Electric Power icon.png
Issue date: 05/26/1995
From:
INDIANA MICHIGAN POWER CO. (FORMERLY INDIANA & MICHIG
To:
Shared Package
ML17332A791 List:
References
NUDOCS 9506130053
Download: ML17332A790 (889)


Text

2.0SAHsTYLIMITSANDLIMITINGSAIKIYSYSTEMSETTINGS6606502400aUNACCtIPrhBLE 64063021palaU-6206i06005905805700'.20.40.60.8Power(fraction ofratedthermalpower)1.2PRESSUREQ%hl1840200021002400(0.02,62086)>>(0.02,632.?9),(0.02,63985)>>(0.02,64986)>>(0.02,65982)>>(1.136,586.17),(1.094,60021),(1.068,608.72),(1.031>>62083)>>(0.996,632.42),(1Q57784)(12,58682)(1+591.77)(1+599AO)(1Z,606.63)Figure2.1-1ReactorCoreSafetyLimitspage2-2COOKNUCLEARPLANT-UNIT 19506l30053 950526PDRADQCK050003l5'PDRAMENDMENT W>>4>>~>>~

lA('~

2.0SMARTYLIMITSANDLIMITINGSAFETYSYSTEMSETTINGSTABL'E2.2-1REACTORTRIPSYSTEMINSTRUMENTATION TRIPSETPOINTS FUNCTIONAL UNIT1.ManualReactorTrip2.PowerRange,NeutronFlux3.PowerRange,NeutronFlux,HighPositiveRate.4.PowerRange,NeutronFlux,HighNegativeRateTRIPSETPOINTNotApplicable LowSetpoint-lessthanorequalto25%ofRATEDTHERMALPOWER-HighSetpoint-lessthanorequal.to109%ofRATEDTHERMALPOWERLessthanorequalto5%ofRATEDTHERMALPOWERwithatimeconstantgreaterthanorequalto2secondsLessthanorequalto5%ofRATEDTHERMALPOWERwithatimeconstantgreaterthanorequalto2secondsALLOWABLE VALUESNotApplicable LowSetpoint-lessthanorequalto26%ofRATEDTHERMALPOWERHighSetpoint-lessthanorequalto110%ofRATEDTHERMALPOWERLessthanorequalto5.5%ofRATEDTHERMALPOWERwithatimeconstantgreaterthanorequalto2secondsLessthanorequalto5.5%ofRATEDTHERMALPOWERwithatimeconstantgreaterthanorequalto2seconds5.Intermediate Range,NeutronFluxSourceRange,NeutronFluxLessthanorequalto25%ofRATEDTHERMALPOWERLessthanorequalto10'ountspersecondLessthanorequalto30%ofRATEDTHERMALPOWERLessthanorequalto1.3x10'ountspersecond7.Overtemperature DeltaT8.Overpower DeltaTSeeNote1SeeNote2SeeNote3SeeNote49.Pressurizer Pressure-LowGreaterthanorequalto1875psigGreaterthanorequalto1865psig10.Pressurizer Pressure-High11.Pressurizer WaterLevel--High12.LossofFlowLessthanorequalto2385psigLessthanorequalto92%ofinstrument spanGreaterthanorequalto90%ofdesignflowperloop*Lessthanorequalto2395psigLessthanorequalto93%ofinstrument spanGreaterthanorequalto89.1%ofdesignflowperloop**DesignflowisI/4ReactorCoolantSystemtotalfiowratefromTable3.2-1.COOKNUCLEARPLANT-UNIT IPage2-5AMENDMENT

$4,426,483

OOTABLE2.2-1(Continued)

REACTORTRIPSYSTEMINSTRUMENTATION TRIPSETPOINTS NOTATION1+~,sNote1:OvertemperaturehT 6dT,[K,-K-'T-T')+K(P-Pg-f,(41)]1+~pwhere:hT,Indicated hTatRATEDTHERMALPOWERp/1+,s1+stAveragetemperature,

'FIndicated T,~atRATEDTHERMALPOWER((576.3'F)Pressurizer

pressure, psigIndicated RCSnominaloperating pressure(2235psigor2085psig)Thefunctiongenerated bythelead-lagcontroller forT,~dynamiccompensation Timeconstants utilizedinthelead-lagcontroller forT,~Tt=22secs.v>=4secs.Laplacetransform operator I

TABLE2.2-1(Continued)

REACTORTRIPSYSTEMINSTRUMENTATION TRIPSETPOINTS NOTATIONS Continued Operation with4LoopsK,=1.17Ki=0.0230K3=0.00110andf,(hl)isafunctionoftheindicated difference betweentopandbottomdetectors ofthepower-range nuclearionchambers; withgainstobeselectedbasedonmeasuredinstrument responseduringplantstartuptestssuchthat:(i)Forq,-q,between-37percentand+3percent,f,(hl)=0(whereq,andq,arepercentRATEDTHERMALPOWERinthetopandbottomhalvesofthecorerespectively, andq,+q,istotalTHERMALPOWERinpercentofRATEDTHERMALPOWER).(ii)Foreachpercentthatthemagnitude of(q,-rgexceeds-37percent,thed,Ttripsetpointshallbeautomatically reducedby0.33percentofitsvalueatRATEDTHERMALPOWER.(iii)Foreachpercentthatthemagnitude of(q,-qQexceeds+3percent,thehTtripsetpointshallbeautomatically reducedby2.34percentofitsvalueatRATEDTHERMALPOWER.

TABLE2.2-1(Continued)

REACTORTRIPSYSTEMINSTRUMENTATION TRIPSETPOINTS NOTATIONContinued Note2:Overpowers,T KhT,[K-K-T-K(T-T)-f(hi)]wsSII+iS3where:Indicated d,TatRATEDTHERMALPOWERAveragetemperature,

'FIndicated T,~atRATEDTHERMALPOWER((563.0'F)1.083Ks0.0177/'F forincreasing averagetemperature and0fordecreasing averagetemperature 0.0015forT>T";+=0forT6T"~~SI+~qS~3Thefunctiongenerated bytheratelagcontroller forT,~dynamiccompensation Timeconstantutilizedintheratelagcontroller forT,~vs=10secs.S=Laplacetransform operatorf,(BI)=0spNote3:Thechannel's maximumtrippointshallnotexceeditscomputedtrippointbymorethan3.4percentd,Tspan.Note4:Thechannel's maximumtrippointshallnotexceeditscomputedtrippointbymorethan2.5percenthTspan.

InInI0 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVITY CONTROLSYSTEMS3/4.1.1BORATIONCONTROLSHUTDOWNMARGIN-TAVGGREATERTHAN200'FLIMITINGCONDITION FOROPERATION 3.1.1.1TheSHUTDOWNMARGINshallbegreaterthanorequalto1.3%Deltak/k.APPLICABILITY:

MODES1,2',3,and4.ACTION:WiththeSHUTDOWNMARGINlessthan1.3%Deltak/k,immediately initiateandcontinueborationatgreaterthanorequalto10gpmofasolutioncontaining greaterthanorequalto20,000ppmboronorequivalent untiltherequiredSHUTDOWNMARGINisrestored.

SURVEILLANCE REUIREMENTS 4.1.1.1.1 TheSHUTDOWNMARGINshallbedetermined tobegreaterthanorequalto1.3%Deltak/k:Withinonehourafterdetection ofaninoperable controlrods(s)andatleastonceper12hoursthereafter whiletherod(s)isinoperable.

Iftheinoperable controlrodisimmovable oruntrippable, theaboverequiredSHUTDOWNMARGINshallbeverifiedacceptable withanincreased allowance forthewithdrawn worthoftheimmovable oruntrippable controlrod(s).b.WheninMODE1orMODE2withKeffgreaterthanorequalto1.0,atleastonceper12hoursbyverifying thatcontrolbankwithdrawal iswithinthelimitsofSpecification 3.1.3.5.C.WheninMODE2withKefflessthan1.0,within4hourspriortoachieving reactorcriticality byverifying thatthepredicted criticalcontrolrodpositioniswithinthelimitsofSpecification 3.1.3.5.Priortoinitialoperation above5%RATEDTHERMALPOWERaftereachfuelloading,byconsideration ofthefactorsofebelow,withthecontrolbanksatthemaximuminsertion limitofSpecification 3.1.3.5.SeeSpecialTestException 3.10.1.COOKNUCLEARPLANT-UNIT 1Page3/41-1AMENDMENT

&,4A,448

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVITY CONTROLSYSTEMSCHARGINGPUMP-SHUTDOWNLIMITINGCONDITION FOROPERATION 3.1.2.3a.Onechargingpumpintheboroninjection flowpathrequiredbySpecification 3.1.2.1shallbeOPERABLEandcapableofbeingpoweredfromanOPERABLEemergency bus.b.Onechargingflowpathassociated withsupportofUnit2shutdownfunctions shallbeavailable.

APPLICABILITY:

Specification 3.1.2.3.a.

-MODES5and6Specification 3.1.2.3.b.

-AtalltimeswhenUnit2isinMODES1,2,3,or4.ACTION:WithnochargingpumpOPERABLE, suspendalloperations involving COREALTERATIONS orpositivereactivity changes."

WithmorethanonechargingpumpOPERABLEorwithasafetyinjection pump(s)OPERABLEwhenthetemperature ofanyRCScoldlegislessthanorequalto152'F,unlessthereactorvesselheadisremoved,removetheadditional chargingpump(s)andthesafetyinjection pump(s)motorcircuitbreakersfromtheelectrical powercircuitwithinonehour.Theprovisions ofSpecification 3.0.3arenotapplicablc.

Inadditiontotheabove,whenSpecification 3.1.2.3.b isapplicable andtherequiredflowpathisnotavailable, returntherequiredflowpathtoavailable statuswithin7days,orprovideequivalent shutdowncapability inUnit2andreturntherequiredflowpathtoavailable statuswithinthenext60days,orhaveUnit2inHOTSTANDBYwithinthenext12hoursandHOTSHUTDOWNwithinthefollowing 24hours.e.Therequirements ofSpecification

3.0. 4arenotapplicable

whenSpecification 3.1.2.3.b applies.SURVEILLANCE REUIREMENTS 4.1.2.3.1 Theaboverequiredchargingpumpshallbedemonstrated OPERABLEbyverifying, thatonrecirculation flow,thepumpdevelopsadifferential pressureofgreaterthanorequalto2290psidwhentestedpursuanttoSpecification 4.0.5.Amaximumofonecentrifugal chargingpumpshallbeOPERABLEwheneverthetemperature ofoneormoreoftheRCScoldlegsislessthanorequalto152'F."Forpurposesofthisspecification, additionofwaterfromtheRWSTdoesnotconstitute apositivereactivity additionprovidedtheboronconcentration ontheRWSTisgreaterthantheminimumrequiredbySpecification 3.1.2.7.b.2.

COOKNUCLEARPLANT-UNIT 1Page3/41-11AMENDMENT 98,420,434,444,447 0

3/4LIMITINGCONDITIONS FOROPE'RATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVITY CONTROLSYSTEMSCHARGINGPUMPS-OPERATING LIMITINGCONDITION FOROPERATION 3.1.2AAtleasttwochargingpumpsshallbeOPERABLE.

APPLICABILITY:

MODES1,2,3and4.ACTION:WithonlyonechargingpumpOPERABLE, restoreatleasttwochargingpumpstoOPERABLEstatuswithin72hoursorbeinHOTSTANDBYwithinthenext6hours;restoreatleasttwochargingpumpstoOPERABLEstatuswithinthenext48hoursorbeinCOLDSHUTDOWNwithinthefollowing 30hours.SURVEILLANCE REUIREMENTS 4.1.2.4Atleasttwochargingpumpsshallbedemonstrated OPERABLEbyverifying thatonrecirculation flow,eachpumpdevelopsadifferential pressureofgreaterthanorequalto2290psidwhentestedpursuanttoSpecification 4.0.5.,COOKNUCLEARPLANT-UNIT 1Page3/41-12 lf0 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVITY CONTROLSYSTEMSBORATEDWATERSOURCES-SHUTDOWNLIMITINGCONDITION FOROPERATION 3.1.2.7Asaminimum,oneofthefollowing boratedwatersourcesshallbeOPERABLE:

Aboricacidstoragesystemandassociated heattracingwith:1.Aminimumusableboratedwatervolumeof4300gallons,2.Between20,000and22,500ppmofboron,and3.Aminimumsolutiontemperature of145'F.Therefueling waterstoragetankwith:1.Aminimumusableboratedwatervolumeof90,000gallons,2.Aminimumboronconcentration of2400ppm,and3.Aminimumsolutiontemperature of70'F.APPLICABILITY:

ACTION:MODES5and6.WithnoboratedwatersourceOPERABLE, suspendalloperations involving COREALTERATIONS orpositivereactivity changes'ntil atleastoneboratedwatersourceisrestoredtoOPERABLEstatus.SURVEILLANCE REUIREMENTS 4.1.2.7Theaboverequiredboratedwatersourceshallbedemonstrated OPERABLE:

Atleastonceper7daysby:1.2.3.Verifying theboronconcentration ofthewater,Verifying thewaterlevelvolumeofthetank,andVerifying theboricacidstoragetanksolutiontemperature whenitisthesourceofboratedwater.Atleastonceper24hoursbyverifying theRWSTtemperature whenitisthesourceofboratedwater.Forpurposesofthisspecification, additionofwaterfromtheRWSTdoesnotconstitute apositivereactivity additionprovidedtheboronconcentration intheRWSTisgreaterthantheminimumrequiredbySpecification 3.1.2.7.b.2.

fCOOKNUCLEARPLANT-UNIT 1Page3/41-15 f:,!

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVITY CONTROLSYSTEMSBORATEDWATERSOURCES-OPERATIONS LIMITINGCONDITION FOROPERATION 3.1.2.8Eachofthefollowing boratedwatersourcesshallbeOPERABLE:

b.Aboricacidstoragesystemandassociated heattracingwith:1.Aminimumusableboratedwatervolumeof5650gallons,2.Between20,000and22,500ppmofboron,and3.Aminimumsolutiontemperature of145'F.Therefueling waterstoragetankwith:1.Aminimumcontained volumeof350,000gallonsofwater,2.Between2400and2600ppmofboron,and3.Aminimumsolutiontemperature of70'F.APPLICABILITY:

MODES1,2,3and4.ACTION:Withtheboricacidstoragesysteminoperable, restorethestoragesystemtoOPERABLEstatuswithin72hoursorbeinatleastHOTSTANDBYwithinthenext6hoursandboratedtoaSHUTDOWNMARGINequivalent toatleast1%d,k/kat200'F;restoretheboricacidstoragesystemtoOPERABLEstatuswithinthenext7daysorbeinCOLDSHUTDOWNwithinthenext30hours.b.Withtherefueling waterstoragetankinoperable, restorethetanktoOPERABLEstatuswithinonehourorbeinatleastHOTSTANDBYwithinthenext6hoursandinCOLDSHUTDOWNwithinthefollowing 30hours.SURVEILLANCE REUIREMENTS 4.1.2.8Eachboratedwatersourceshallbedemonstrated OPERABLE:

COOKNUCLEARPLANT-UNIT 1Page3/41-16 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.2POWERDISTRIBUTION LIMITSTABLE3.2-1DNBPARAMETERS PARAMETER ReactorCoolantSystemTavgPressurizer PressureReactorCoolantSystemTotalFlowRateLIMITS4LoopsinOperation atRATEDTHERMALPOWER6579.3'FR2050psig">341,100gpm"Indicated averageofatleastthreeOPERABLEinstrument loops."Limitnotapplicable duringeitheraTHERMALPOWERrampincreaseinexcessof5percentRATEDTHERMALPOWERperminuteoraTHERMALPOWERstepincreaseinexcessof10percentRATEDTHERMALPOWER."'Indicated value.COOKNUCLEARPLANT-UNIT 1Page3/42-14AMENDMENT 94,420,426,~

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.3INSTRUMENTATION TABLE3.3-3Continued ENGINEERED SAFETYFEATUREACTUATION SYSTEMINSTRUMENTATION FUNCTIONAL UNITf.SteamLinePressure-LowMINIMUMTOTALNO.OFCHANNELSCHANNELSAPPLICABLE CHANNELSTOTRIPOPERABLEMODESACTIONFourLoopsOperating ThreeLoopsOperating 1pressure/loop 1pressure/

operating loop2pressures anyloops1"'ressure inanyoperating loop1pressure1,2,3"any3loops1pressurein3"QIly2operating loops1415COOKNUCLEARPLANT-UNIT 1Page3/43-17AMENDMENT

$4,420,~

t 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.3INSTRVMENTATION TABLE3.3-3Continued ENGINEERED SAFETYFEATUREACTUATION SYSTEMINSTRUMENTATION FUNCTIONAL UNITCOINCIDENT WITHMINIMUMTOTALNO.OFCHANNELSCHANNELSAPPLICABLE CHANNELSTOTRIPOPERABLEMODESACTIONT,~-Low-LowFourLoopsOperating 1-T,+oop2Ta<anyloops1T,~any31,2,3'~loops14ThreeLoopsOperating 1T,+operating loop1"T,~inIT,~inany3'nyoperating twooperating looploops15e.SteamLinePressure-Low FourLoopsOperating 1pressure/loop 2pressures anyloops1pressure1,2,'3"any3loops14'hreeLoopsOperating 5.TURBINETRIP8cFEEDWATER ISOLATION 1pressure/

operating loop1"'ressure lnanyoperating loop1pressurein3"any2operating loops15a.SteamGenerator WaterLevel-High-High3/loop2/loopinany2/loopinoperating eachloopoperating loop1,2,314COOKNUCLEARPLANT-UNIT 1Page3/43-21AMENDMENT

$4,440,4' I'

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.3INSTRUMENTATION ENGINEERED SAFETYFEATURESINTERLOCKS DESIGNATION P-11P-12CONDITION ANDSETPOINTWith2of3pressurizer pressurechannelsgreaterthanorequalto1915psig.With2of4T,~channelslessthanorequaltoSetpoint.

Setpointgreaterthanorequalto541'FFUNCTIONP-11preventsordefeatsmanualblockofsafetyinjection actuation onlowpressurizer pressure.

P-12allowsthemanualblockofsafetyinjection actuation onlowsteamlinepressurecausessteamlineisolation onhighsteamflow.Affectssteamdumpblocks.With3of4T,~channelsabovetheresetpoint,preventsordefeatsthemanualblockofsafetyinjection actuation onlowsteamlineprcssure.

COOKNUCLEARPLANT-UNIT 1Page3/43-23a 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4,3INSTRUMENTATION TABLE3.3CENGINEERED SAFETYFEATUREACTUATION SYSTEMINSTRUMENTATION TRIPSETPOINTS FUNCTIONAL UNIT1.SAFETYINJECTION, TURBINETRIP,FEEDWATER ISOLATION, ANDMOTORDRIVENFEEDWATER PUMPSTRIPSETPOINTALLOWABLE VALUESa.ManualInitiation


SeeFunctional Unit9-b.Automatic Actuation Logicc.Containment Pressure-HighNotApplicable Lessthanorequalto1.1psigNotApplicable Lessthanorequalto1.2psigd.Pressurizer Pressure-LowGreaterthanorequalto1815psigGreaterthanorequalto1805psige.Differential PressureBetweenSteamLines-Highf.SteamLinePressure-LowLessthanorequalto100psiGreaterthanorequalto500psigsteamlinepressureLessthanorequalto112psiGreaterthanorequalto480psigsteamlinepressureCOOKNUCLEARPLANT-UNIT 1Page3/43-24AMENDMENT 40,426,~

3/4LIMI'HNGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.3INSTRUMENTATION TABLE3.3PContinued ENGINEERED SAFETYFEATUREACTUATION SYSTEMINSTRUMENTATION TRIPSETPOINTS FUNCTIONAL UNIT2.Containment Radio-activity-HighTrainA(VRS-1101, ERS-1301, ERS-1305) 3.Containment Radio-activity-High TrainB(VRS-1201, ERS-1401, ERS-1405) 4.STEAMLINEISOLATION TRIPSETPOINTSeeTable3.34SeeTable3.3-6ALLOWABLE VALUESNotApplicable NotApplicable a.Manual---------SeeFunctional Unit9-b.Automatic Actuation LogicIc.Containment Pressure-High-High d.SteamFlowinTwoSteamLines-High Coincident withT,~-Low-Low e.SteamLinePressure-Low5.TURBINETRIPANDFEEDWATER ISOLATION NotApplicable Lessthanorequalto2.9psigLessthanorequalto1.42x10~lbs/hrfrom0%loadto20%load.Linearfrom1.42x10'bs/hrat20%loadto3.88x10'bs/hrat100%load.T,~greaterthanorequalto541'FGreaterthanorequalto500psigsteamlinepressureNotApplicable Lessthanorequalto3psigLessthanorequalto1.56x10lbs/hrfrom0%loadto20%load.Linearfrom1.56x10'bs/hrat20%loadto3.9310'bs/hrat100%load.T,~greaterthanorequalto539'FGreaterthanorequalto480psigsteamlinepressurea.SteamGenerator WaterLevel-High-High Lessthanorequalto67%ofnarrow-range instrument spaneachsteamgenerator Lessthanorequalto68%ofnarrow-range instrument spaneachsteamgenerator COOKNUCLEARPLANT-UNIT 1Page3/43-26AMENDMENT 94,436,~

iI 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.3INSTRUMENTATION TABLE4.3-2ENGINEERED SAFETYFEATUREACTUATION SYSTEMINSTRUMENTATION SURVEILLANCE REUIREMENTS FUNCTIONAL UNIT1.SAFETYINJECTION, TURBINETRIP,FEEDWATER ISOLATION, ANDMOTORDRIVENAUXILIARY FEEDWATER PUMPSTRIPACTUATING MODESINCHANNELDEVICEWHICHCHANNELCHANNELFUNCTIONAL OPERATIONAL SURVEILLANCE CHECKCALIBRATION TESTa.ManualInitiation b.Automatic Actuation LogicN.A.c.Containment Pressure-Highd.Pressurizer Pressure-Low Se.Differential PressureBetweenSteamLines-'ighf.SteamLinePressure-Low S2.CONTAINMENT SPRAYa.ManualInitiation N.A.SeeFunctional Unit9M(2)M(3)MSeeFunctional Unit9N.A.N.A.N.A.N.A.N.A.1,2,3,41,2,31,2,31,2,31,2,3b.Automatic Actuation LogicN.A.c.Containment Pressure-High-High N.A.M(2)M(3)N.A.N.A.I,2,3,41,2,3COOKNUCLEARPLANT-UNIT 1Page3/43-31AMENDMENT 400)420)XRk) 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.3INSTRUMENTATION TABLE4.3-2Continued ENGINEERED SAFETYFEATUREACTUATION SYSTEMINSTRUMENTATION SURVEILLANCE REUIREMENTS FUNCTIONAL UNIT4.STEAMLINEISOLATION TRIPACTUATING MODESINCHANNELDEVICEWHICHCHANNELCHANNELFUNCTIONAL OPERATIONAL SURVEILLANCE CHECKCALIBRATION TESTa.Manualb.Automatic Actuation LogicN.A.c.Containment Pressure-High-High d.SteamFlowinTwoSteamSLines-High Coincident withT~-Low-Low e.SteamLinePressure-Low S5.TURBINETRIPANDFEEDWATER ISOLATION N.A.SeeFunctional Unit9M(2)M(3)MN.A.N.A.N.A.N.A.1,2,3,1,2,31,2,3I,2,3a.SteamGenerator WaterLevel-High-High 6.MOTORDRIVENAUXILIARY FEEDWATER PUMPSN.A.1,2,3c.SafetyInjection N.A.d.LossofMainFeedPumpsN.A.a.SteamGenerator WaterLevel-Low-Low b.4kvBusLossofVoltageSN.A.N.A.M(2)N.A.N.A.N.A.N.A.1,2,31,2,31,2,31,2COOKNUCLEARPLANT-UNIT 1Page3/43-33AMENDMENT 400,420,424,444,4' t

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.4REACTORCOOLANTSYSTEMSAFETYVALVES-SHUTDOWNLIMITINGCONDITION FOROPERATION 3.4.2Aminimumofonepressurizer codesafetyvalveshallbeOPERABLEwithaliftsettingof2485PSIGJ3%.IAPPLICABILITY:

ACTION:MODES4and5.Withnopressurizer codesafetyvalveOPERABLE:

a.Immediately suspendalloperations involving positivereactivity changes"andplaceanOPERABLERHRloopintooperation intheshutdowncoolingmode.b.Immediately renderallSafetyInjection pumpsandallbutonechargingpumpinoperable byremovingtheapplicable motorcircuitbreakersfromtheelectricpowercircuitwithinonehour.SURVEILLANCE REUIREMENTS 4.4.2Thepressurizer codesafetyvalveshallbedemonstrated OPERABLEperSurveillance Requirement 4.4.3.Theliftsettingpressureshallcorrespond toambientconditions ofthevalveatnominaloperating temperature andpressure.

"Forpurposesofthisspecification, additionofwaterfromtheRWSTdoesnotconstitute apositivereactivity additionprovidedtheboronconcentration intheRWSTisgreaterthantheminimumrequiredbySpecification 3.1.2.8.b.2 (MODE4)or3.1.2.7.b.2 (MODE5).COOKNUCLEARPLANT-UNIT 1Page3/44Q

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.4REACTORCOOLANTSYSTEMSAFETYVALVES-OPERATING LIMITINGCONDITION FOROPERATION 3.4.3Allpressurizer codesafetyvalvesshallbeOPERABLEwithaliftsettingof%85PSIG+3%.APPLICABILITY:

MODES1;2and3.ACTION:Withonepressurizer codesafetyvalveinoperable, eitherrestoretheinoperable valvetoOPERABLEstatuswithin15minutesorbeinHOTSHUTDOWNwithin12hours.SURVEILLANCE REUIREMENTS 4.4.3Noadditional surveillance requirements otherthanthoserequiredbySpecification 4.0.5.'Theliftsettingpressureshallcorrespond toambientconditions ofthevalveatnominaloperating temperature andpressure.

COOKNUCLEARPLANT-UNIT 1Page3/44-5

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.5EMERGENCY CORECOOLINGSYSTEMS(ECCS)SURVEILLANCE REUIREMENTS Continued d.Atleastonceper18monthsby:1.Verifying automatic isolation andinterlock actionoftheRHRsystemfromtheReactorCoolantSystemwhentheReactorCoolantSystempressureisabove600psig.2.Avisualinspection ofthecontainment sumpandverifying thatthesubsystem suctioninletsarenotrestricted bydebrisandthatthesumpcomponents (trashracks,screens,etc.)shownoevidenceofstructural distressorabnormalcorrosion.

e.Atleastonceper18months,duringshutdown, by:1.Verifying thateachautomatic valveintheflowpathactuatestoitscorrectpositiononaSafetyInjection testsignal.2.Verifying thateachofthefollowing pumpsstartautomatically uponreceiptofasafetyinjection signal:a)Centrifugal chargingpumpb)Safetyinjection pumpc)Residualheatremovalpumpf.Byverifying thateachofthefollowing pumpsdevelopstheindicated differential pressureonrecirculation flowwhentestedpursuanttoSpecification 4.0.5.1.Centrifugal chargingpumpgreaterthanorequalto2290psid2.Safetyinjection pumpgreaterthanorequalto1326psid3.Residualheatremovalpumpgreaterthanorequalto150psidByverifying thecorrectpositionofeachmechanical stopforthefollowing Emergency CoreCoolingSystemthrottlevalves:1.Within4hoursfollowing completion ofeachvalvestrokingoperation ormaintenance onthevalvewhentheECCSsubsystems arerequiredtobeOPERABLE.

COOKNUCLEARPLANT-UNIT 1Page3/45-5AMENDMENT 4P,426,444,448,444

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4,5EMERGENCY CORECOOLINGSYSTEMS(ECCS)REFUELING WATERSTORAGETANKLIMITINGCONDITION FOROPERATION 3.5.5Therefueling waterstoragetank(RWST)shallbeOPERABLEwith:Aminimumcontained volumeof350,000gallonsofboratedwater.Between2400and2600ppmofboron,andAminimumwatertemperature of70'F.APPLICABILITY:

ACTION:MODES1,2,3and4.Withtherefueling waterstoragetankinoperable, restorethetanktoOPERABLEstatuswithin1hourorbeinatleastHOTSTANDBYwithin6hoursandinCOLDSHUTDOWNwithinthefollowing 30hours.SURVEILLANCE REUIREMENTS 4.5.5TheRWSTshallbedemonstrated OPERABLE:

a.Atleastonceper7daysby:1.Verifying thecontained boratedwaterlevelinthetank,andVerifying theboronconcentration ofthewater.b.Atleastonceper24hoursbyverifying theRWSTtemperature.

COOKNUCLEARPLANT-UNIT 1Page3/45-11AMENDMENT 5B,444 f

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.7PLANTSYSTEMSSURVEILLANCE REUIREMENTS Continued 4.7.1.2Eachauxiliary feedwater pumpshallbedemonstrated OPERABLEwhentestedpursuanttoSpecification 4.0.5by:a.Verifying thateachmotordrivenpumpdevelopsanequivalent discharge pressureofgreaterthanorequalto1240psigat60'Finrecirculation flow.Verifying thatthesteamturbinedrivenpumpdevelopsanequivalent discharge pressureofgreaterthanorequalto1180psigat60'Fandataflowofgreaterthanorequalto700gpmwhenthesecondary steamsupplypressureisgreaterthan310psig.Theprovisions ofSpecification

4.0. 4arenotapplicable

forentryintoMODE3.Verifying thateachnon-automatic valveintheflowpaththatisnotlocked,sealed,orotherwise securedinpositionisinitscorrectposition.

Verifying thateachautomatic valveintheflowpathisinthefullyopenpositionwhenevertheauxiliary feedwater systemisplacedinautomatic controlorwhenabove10%RATEDTHERMALPOWER.Thisrequirement isnotapplicable forthoseportionsoftheauxiliary feedwater systembeingusedintermittently tomaintainsteamgenerator waterlevel.e.Verifying atleastonceper18monthsduringshutdownthateachautomatic valveintheflowpathactuatestoitscorrectpositionuponreceiptoftheappropriate engineered safetyfeaturesactuation testsignalrequiredbySpecification 3/4.3.2.Verifying atleastonceper18.monthsduringshutdownthateachauxiliary feedwater pumpstartsasdesignedautomatically uponreceiptoftheappropriate engineered safetyfeaturesactuation testsignalrequiredbySpecification 3/4.3.2.Verifying atleastonceper18monthsduringshutdownthattheunitcross-tie valvescancyclefulltravel.Following cycling,thevalveswillbeverifiedtobeintheirclosedpositions.

COOKNUCLEARPLANT-UNIT 1Page3/47-6AMENDMENT 400,43k,444,464

5.0DESIGNFEATURES5.4REACTORCOOLANTSYSTEMContinued a.Inaccordance withthecoderequirements specified inSection4.1.6oftheFSAR,withallowance fornormaldegradation pursuanttotheapplicable Surveillance Requirements, b.Forapressureof2485psig,andc.Foratemperature of650'F,exceptforthepressurizer whichis680'F.VOLUME5.4.2Thetotalcontained volumeofthereactorcoolantsystemisapproximately 12,466cubicfeetat0%steamgenerator tubepluggingand11,551cubicfeetat30%steamgenerator tubepluggingatanominalT,,of70'F.5.5EMERGENCY CORECOOLINGSYSTEMS5.5.1Theemergency corecoolingsystemsaredesignedandshallbemaintained inaccordance withtheoriginaldesignprovisions contained inSection6.2oftheFSARwithallowance fornormaldegradation pursuanttotheapplicable Surveillance Requirements, withoneexception.

Thisexception istheCVCSboronmakeupsystemandtheBIT.5.6FUELSTORAGECRITICALITY

-SPENTFUEL5.6.1.1Thespentfuelstorageracksaredesignedandshallbemaintained with:aeAk,itequivalent tolessthan0.95whenfloodedwithunborated water.Anominal8.97inchcenter-to~nter distancebetweenfuelassemblies placedinthestorageracks.Thefuelassemblies willbeclassified asacceptable forRegion1,Region2,orRegion3storagebasedupontheirassemblyaverageburnupversusinitialnominalenrichment.

Cellsacceptable forRegion1,Region2,andRegion3assemblystorageareindicated inFigures5.6-1and5.6-2.Assemblies thatareacceptable forstorageinRegion1,Region2,andRegion3mustmeetthedesigncriteriathatdefinetheregionsasfollows:COOKNUCLEARPLANT-UNIT 1Page5-5AMENDMENT 468,44$,469

BASES2.0SAFETYLIMITSANDLIMITINGSAFETYSYSTEMSETTINGS2.1SAFETYLIMITSBASES4LoopOperation Westinghouse Fuel(15x15OFA)(WRB-1Correlation)

Correlation LimitDesignLimitDNBRSafetyAnalysisLimitDNBRTypicalCell1.171.231.40ThimbleCell"1.171.221.42ThecurvesofFigure2.1-1showthelociofpointsofTHERMALPOWER,ReactorCoolantSystempressureandaveragetemperature forwhichtheminimumDNBRisnolessthantheapplicable designDNBRlimit,ortheaverageenthalpyatthevesselexitisequaltotheenthalpyofsaturated liquid.represents typicalfuelrod"represents fuelrodsnearguidetubeCOOKNUCLEARPLANT-UNIT 1PageB2-1(a)AMENDMENT 74,4%,4%

I'I'l, BASES2.0SALTYLIMITSANDLIMITINGSAFETYSYSTEMSETTINGSSAFETYLIMITSBASESThePowerRangeNegativeRateTripprovidesprotection forcontrolroddropaccidents.

Athighpower,aroddropaccidentcouldcauselocalfluxpeakingwhichcouldcauseanunconservative localDNBRtoexist.ThePowerRangeNegativeRateTripwillpreventthisfromoccumngbytrippingthereactor.Nocreditistakenforoperation ofthePowerRangeNegativeRateTripforthosecontrolroddropaccidents forwhichtheDNBRswillbegreaterthantheapplicable designlimitDNBR-value foreachfueltype.Intermediate andSourceRaneNuclearFluxTheIntermediate andSourceRange,NuclearFluxtripsprovidereactorcoreprotection duringreactorstartup.Thesetripsprovideredundant protection tothelowsetpointtripofthePowerRange,NeutronFluxchannels.

ThesourceRangeChannelswillinitiateareactortripatabout10+'ounts persecond,unlessmanuallyblockedwhenP-6becomesactive.TheIntermediate RangeChannelswillinitiateareactortripatacurrentlevelproportional toapproximately 25percentofRATEDTHERMALPOWERunlessmanuallyblockedwhenP-10becomesactive.Nocreditwastakenforoperation ofthetripsassociated witheithertheIntermediate orSourceRangeChannelsintheaccidentanalyses; however,theirfunctional capability atthespecified tripsettingsisrequiredbythisspecification toenhancetheoverallreliability oftheReactorProtection System.OvertemeratureDeltaTTheOvertemperature deltaTtripprovidescoreprotection topreventDNBforallcombinations ofpressure, power,coolanttemperature, andaxialpowerdistribution, providedthatthetransient isslowwithrespecttopipingtransitdelaysfromthecoretothetemperature detectors (about4seconds),

andpressureiswithintherangebetweentheHighandLowPressurereactortrips.Thissetpointincludescorrections forchangesindensityandheatcapacityofwaterwithtemperature anddynamiccompensation forpipingdelaysfromthecoretothelooptemperature detectors.

Withnormalaxialpowerdistribution, thisreactortriplimitisalwaysbelowthecoresafetylimitasshowninFigure2.1-1.Ifaxialpeaksaregreaterthandesign,asindicated bythedifference betweentopandbottompowerrangenucleardetectors, thereactortripisautomatically reducedaccording tothenotations inTable2.2-1.COOKNUCLEARPLANT-UNIT IPageB2QAMENDMENT V4,426

BASES2.0SAHHYLIMITSANDLIMITINGSAFETYSYSTEMSETTINGSLIMITINGSAFETYSYSTEMSETTINGSBASESOveowerDeltaTTheOverpower deltaTreactortripprovidesassurance offuelintegrity, e.g.,nomelting,underallpossibleoverpower conditions, limitstherequiredrangeforOvertemperature deltaTprotection, andprovidesabackuptotheHighNeutronFluxtrip.Thesetpointincludescorrections forchangesindensityandheatcapacityofwaterwithtemperature, anddynamiccompensation forpipingdelaysfromthecoretothelooptemperature detectors.

Theoverpower deltaTreactortripprovidesprotection orback-upprotection foratpowersteamline breakevents.Creditwastakenforoperation ofthistripinthesteamline breakmass/energy releasesoutsidecontainment analysis.

Inaddition, itsfunctional capability atthespecified tripsettingisrequiredbythisspecification toenhancetheoverallreliability ofthereactorprotection system.Pressurizer PressureThePressurizer HighandLowPressuretripsareprovidedtolimitthepressurerangeinwhichreactoroperation ispermitted.

TheHighPressuretripisbackedupbythepressurizer codesafetyvalvesforRCSoverpressure protection, andistherefore setlowerthanthesetpressureforthesevalves(2485psig).TheHighPressuretripprovidesprotection foraLossofExternalLoadevent.TheLowPressuretrip'provides protection bytrippingthereactorintheeventofalossofreactorcoolantpressure.

Pressurizer WaterLevelThePressurizer HighWaterLeveltripensuresprotection againstReactorCoolantSystemoverpressurization bylimitingthewaterleveltoavolumesufficient toretainasteambubbleandpreventwaterreliefthroughthepressurizer safetyvalves.Thepressurizer highwaterleveltripprecludes waterrelieffortheUncontrolled RCCAWithdrawal atPowerevent.COOKNUCLEARPLANT-UNIT IPageB2-5AMENDMENT 420,4%i,4$S 3/4BASES3/4.1REACTIVITY CONTROLSYSTEMS3/4.1.1BORATIONCONTROL3/4.1.1.1 and3/4.1.1.2 SHUTDOWNMARGINAsufficient SHUTDOWNMARGINensuresthat1)thereactorcanbemadesubcritical fromalloperating conditions, 2)thereactivity transients associated withpostulated accidentconditions arecontrollable withinacceptable limits,and3)thereactorwillbemaintained sufficiently subcritical toprecludeinadvertent criticality intheshutdowncondition.

SHUTDOWNMARGINrequirements varythroughout corelifeasafunctionoffueldepletion, RCSboronconcentration, andRCST,~.Themostrestrictive condition occursatEOL,withT,~atnoloadoperating temperature, andisassociated withapostulated steamlinebreakaccidentandresulting uncontrolled RCScooldown.

Intheanalysisofthisaccident, aminimumSHUTDOWNMARGINof1.3%Deltak/kisinitially requiredtocontrolthereactivity transient andautomatic ESFisassumedtobeavailable.

WithT,~lessthan200'F,the'eactivity transients resulting fromapostulated steamlinebreakcooldownareminimalanda1%Deltak/kSHUTDOWNMARGINprovidesadequateprotection forthisevent.TheSHUTDOWNMARGINrequirements arebaseduponthelimitingconditions described aboveandareconsistent withFSARsafetyanalysisassumptions.

3/4.1.1.3 BORONDILUTIONAminimumflowrateofatleast2000GPMprovidesadequatemixing,preventsstratification andensuresthatreactivity changeswillbegradualduringboronconcentration reductions intheReactorCoolantSystem.Aflowrateofatleast2000GPMwillcirculate anequivalent ReactorCoolantSystemvolumeof12,612plusorminus100cubicfeetinapproximately 30minutes.Thereactivity changerateassociated withboronreductions willtherefore bewithinthecapability foroperatorrecognition andcontrol.3/4.1.1.4 MODERATOR TEMPERATURE COEFFICIENT TCThelimitations onMTCareprovidedtoensurethattheassumptions usedintheaccidentandtransient analysesremainvalidthrougheachfuelcycle.Thesurveillance requirement formeasurement oftheMTCatthebeginning, andneartheendofeachfuelcycleisadequatetoconfirmtheMTCvaluesincethiscoefficient changesslowlydueprincipally tothereduction inRCSboronCOOKNUCLEARPLANT-UNIT 1PageB3/41-1AMENDMENT V4,4A,448 3/4BASES3/4.4REACTORCOOLANTSYSTEM3/4.4.1REACTORCOOLANTLOOPSTheplantisdesignedtooperatewithallreactorcoolantloopsinoperation, andmaintainDNBRabovethesafetyanalysislimitduringallnormaloperations andanticipated transients.

Alossofflowintwoloopswillcauseareactortripifoperating aboveP-7(11percentofRATEDTHERMALPOWER)whilealossofflowinoneloopwillcauseareactortripifoperating aboveP-8(31percentofRATEDTHERMALPOWER).InMODE3,asinglereactorcoolantloopprovidessufficient heatremovalcapability forremovingdecayheat;however,singlefailureconsiderations requirethattwoloopsbeOPERABLE.

ThreeloopsarerequiredtobeOPERABLEandtooperateifthecontrolrodsarecapableofwithdrawal andthereactortripbreakersareclosed.Therequirement assuresadequateDNBRmarginintheeventofanuncontrolled rodwithdrawal inthismode.InMODES4and5,asinglereactorcoolantlooporRHRloopprovidessufficient heatremovalcapability forremovingdecayheat;butsinglefailureconsiderations requirethatatleasttwoloopsbeOPERABLE.

Thus,ifthereactorcoolantloopsarenotOPERABLE, thisspecification requirestwoRHRloopstobeOPERABLE.

Theoperation ofoneReactorCoolantPumporoneRHRpumpprovidesadequateflowtoensuremixing,preventstratification andproducegradualreactivity changesduringboronconcentration reductions intheReactorCoolantSystem.Thereactivity changerateassociated withboronreduction will,therefore, bewithinthecapability ofoperatorrecognition andcontrol.Therestrictions onstartingaReactorCoolantPumpbelowP-7withoneormoreRCScoldlegslessthanorequalto152'FareprovidedtopreventRCSpressuretransients, causedbyenergyadditions fromthesecondary system,whichcouldexceedthelimitsofAppendixGto10CFRPart50.TheRCSwillbeprotected againstoverpressure transients andwillnotexceedthelimitsofAppendixGbyeither(1)restricting thewatervolumeinthepressurizer andtherebyproviding avolumefortheprimarycoolanttoexpandintoor(2)byrestricting startingoftheRCPstowhenthesecondary watertemperature ofeachsteamgenerator islessthan50'FaboveeachoftheRCScoldlegtemperatures.

COOKNUCLEARPLANT-UNIT IPageB3/44-1AMENDMENT 88,420,467.

If 3/4BASES3/4.5EMERGENCY CORECOOLINGSYSTEMS3/4.5.5REFUELING WATERSTORAGETANKTheOPERABILITY oftheRWSTaspartoftheECCSensuresthatsufficient negativereactivity isinjectedintothecoretocounteract anypositiveincreaseinreactivity causedbyRCSsystemcooldown, andensuresthatasufficient supplyofboratedwaterisavailable forinjection bytheECCSintheeventofaLOCA.Reactorcoolantsystemcooldowncanbecausedbyinadvertent depressurization, alossofcoolantaccidentorasteamlinerupture.ThelimitsonRWSTminimumvolumeandboronconcentration ensurethat1)sufficient waterisavailable withincontainment topermitrecirculation coolingflowtothecore,and2)thereactorwillremainsubcritical inthecoldcondition following aLOCAassumingmixingoftheRWST,RCS,ECCSwater,andothersourcesofwaterthatmayeventually resideinthesump,withallcontrolrodsassumedtobeout.Theseassumptions areconsistent withtheLOCAanalyses.

Thecontained watervolumelimitincludesanallowance forwaternotusablebecauseoftankdischarge linelocationorotherphysicalcharacteristics.

Thelimitsoncontained watervolumeandboronconcentration oftheRWSTalsoensureapHvalueofbetween7.6and9.5forthesolutionrecirculated withincontainment afteraLOCA.ThispHbandminimizes theevolution ofiodineandminimizes theeffectofchlorideandcausticstresscorrosion onmechanical systemsandcomponents.

TheECCSanalysestodetermine F<limitsinSpecifications 3.2.2and3.2.6assumedaRWSTwatertemperature of70'F.Thistemperature valueoftheRWSTwaterdetermines thatofthespraywaterinitially delivered tothecontainment following LOCA.Itisoneofthefactorswhichdetermines thecontainment back-pressure intheECCSanalyses, performed inaccordance withtheprovisions of10CFR50.46andAppendixKto10CFR50.COOKNUCLEARPLANT-UNIT 1PageB3/45-3AMENDMENT

$3,420,4$8 l,.0 3/4BASES3/4.6CONTAINMENT SYSTEMS3/4.6.1.4 INTERNALPRESSUREThelimitations oncontainment internalpressureensurethat1)thecontainment structure isprevented fromexceeding itsdesignnegativepressuredifferential withrespecttotheoutsideatmosphere of8psigand2)thecontainment peakpressuredoesnotexceedthedesignpressureof12psigduringLOCAconditions.

Themaximumpeakpressureresulting fromaLOCAeventiscalculated tobe11.49psig,whichincludes0.3psigforinitialpositivecontainment pressure.

3/4.6.1.5 AIRTEMPERATURE Thelimitations oncontainment averageairtemperature ensurethat1)thecontainment airmassislimitedtoaninitialmasssufficiently lowtopreventexceeding thedesignpressureduringLOCAconditions and2)theambientairtemperature doesnotexceedthattemperature allowable forthecontinuous dutyratingspecified forequipment andinstrumentation locatedwithincontainment.

Thecontainment pressuretransient issensitive totheinitially contained airmassduringaLOCA.Thecontained airmassincreases withdecreasing temperature.

Thelowertemperature limitof60'Fwilllimitthepeakpressureto11.49psigwhichislessthanthecontainment designpressureof12psig.Theuppertemperature limitinfluences thepeakaccidenttemperature slightlyduringaLOCA;however,thislimitisbasedprimarily uponequipment protection andanticipated operating conditions.

Boththeupperandlowertemperature limitsareconsistent withtheparameters usedintheaccidentanalyses.

3/4.6.1.6 CONTAINMENT VESSELSTRUCTURAL INTEGRITY Thislimitation ensuresthatthestructural integrity ofthecontainment steelvesselwillbemaintained comparable totheoriginaldesignstandards forthelifeofthefacility.

Structural integrity isrequiredtoensurethat(1)thesteellinerremainsleaktightand(2)theconcretesurrounding thesteellinerremainscapableofproviding externalmissileprotection forthesteellinerandradiation shielding intheeventofaLOCA.Avisualinspection inconjunction withTypeAleakagetestsissufficient todemonstrate thiscapability.

COOKNUCLEARPLANT-UNIT 1PageB3/46-2

PROPOSEDCHANGESTOTHEDONALDC.COOKNUCLEARPLANTUNITNO.2TECHNICAL SPECIFICATIONS

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVITY CONTROLSYSTEMS3/4.1.1BORATIONCONTROLSHUTDOWNMARGINTvGREATERTHAN200FLIMITINGCONDITION FOROPERATION 3.1.1.1TheSHUTDOWNMARGINshallbegreaterthanorequalto1.3%Deltak/k.APPLICABILITY:

MODES1,2,3,and4.ACTION:WiththeSHUTDOWNMARGINlessthan1.3%Deltak/k,immediately initiateandcontinueborationatgreaterthanorequalto10gpmofasolutioncontaining greaterthanorequalto20,000ppmboronorequivalent untiltherequiredSHUTDOWNMARGINisrestored.

SURVEILLANCE REUIREMENTS 4.1.1.1.1 TheSHUTDOWNMARGINshallbedetermined tobegreaterthanorequalto1.3%Deltak/k:4a.Withinonehourafterdetection ofaninoperable controlrod(s)andatleastonceper12hoursthereafter whiletherod(s)isinoperable.

Iftheinoperable controlrodisimmovable oruntrippable, theaboverequiredSHUTDOWNMARGINshallbeverifiedacceptable withanincreased allowance forthewithdrawn worthoftheimmovable oruntrippable controlrod(s).WheninMODE1orMODE2withK,>>greaterthanorequalto1.0,atleastonceper12hoursbyverifying thatcontrolbankwithdrawal iswithinthelimitsofSpecification 3.1.3.6.C.WheninMODE2withK,lessthan1.0,within4hourspriortoachieving reactorcriticality byverifying thatthepredicted criticalcontrolrodpositioniswithinthelimitsofSpecification 3.1.3.6.Priortoinitialoperation above5%RATEDTHERMALPOWERaftereachfuelloading,byconsideration ofthefactorsofebelow,withthecontrolbanksatthemaximuminsertion limitofSpecification 3.1.3.6.SeeSpecialTestException 3.10.1.COOKNUCLEARPLANT-UNIT 2Page3/4I-IAMENDMENT 8R,40S,434 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVITY CONTROLSYSTEMSCHARGINGPUMP-SHUTDOWNLIMITINGCONDITION FOROPERATION 3.1.2.3a.Onechargingpumpintheboroninjection flowpathrequiredbySpecification 3.1.2.1shallbeOPERABLEandcapableofbeingpoweredfromanOPERABLEemergency bus.Onechargingflowpathassociated withsupportofUnit1shutdownfunctions shallbeavailable.'PPLICABILITY:

ACTION:Specification 3.1.2.3.a.

-MODES5and6Specification 3.1.2.3.b.

-AtalltimeswhenUnit1isinMODES1,2,3,or4.a.WithnochargingpumpOPERABLE, suspendalloperations involving COREALTERATIONS orpositivereactivity changes."

WithmorethanonechargingpumpOPERABLEorwithasafetyinjection pump(s)OPERABLEwhenthetemperature ofanyRCScoldlegislessthanorequalto152'F,unlessthereactorvesselheadisremoved,removetheadditional chargingpump(s)andthesafetyinjection pump(s)motorcircuitbreakersfromtheelectrical powercircuitwithinonehour.c.Theprovisions ofSpecification

3.0. 3arenotapplicable

Inadditiontotheabove,whenSpecification 3.1.2.3.b isapplicable andtherequiredflowpathisnotavailable, returntherequiredflowpathtoavailable statuswithin7days,orprovideequivalent shutdowncapability inUnit1andreturntherequiredflowpathtoavailable statuswithinthenext60days,orhaveUnit1inHOTSTANDBYwithinthenext12hoursandHOTSHUTDOWNwithinthefollowing 24hours.Therequirements ofSpecification

3.0. 4arenotapplicable

whenSpecification 3.1.2.3.b applies.SURVEILLANCE REUIREMENTS 4.1.2.3.1 Theabove-required chargingpumpshallbedemonstrated OPERABLEbyverifying, thatonrecirculation flow,thepumpdevelopsadifferential pressureofgreaterthanorequalto2290psidwhentestedpursuanttoSpecification 4.0.5.Amaximumofonecentrifugal chargingpumpshallbeOPERABLEwheneverthetemperature ofoneormoreoftheRCScoldlegsislessthanorequalto152'F."Forpurposesofthisspecification, additionofwaterfromtheRWSTdoesnotconstitute apositivereactivity additionprovidedtheboronconcentration intheRWSTisgreaterthantheminimumrequiredbySpecification 3.1.2.7.b.2.

ICOOKNUCLEARPLANT-UNIT 2Page3/41-11AMENDMENT 8$,407,446 i'

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVITY CONTROLSYSTEMSCHARGINGPUMPS-OPERATING LIMITINGCONDITION FOROPERATION 3.1.2.4AtleasttwochargingpumpsshallbeOPERABLE.

APPLICABILITY:

MODES1,2,3Gild4.ACTION:WithonlyonechargingpumpOPERABLE, restoreatleasttwochargingpumpstoOPERABLEstatuswithin72hoursorbeinatleastHOTSTANDBYandboratedtoaSHUTDOWNMARGINequivalent toatleast1%hk/kat200'Fwithinthenext6hours;restoreatleasttwochargingpumpstoOPERABLEstatuswithinthenext7daysorbeinCOLDSHUTDOWNwithinthenext30hours.SURVEILLANCE REUIREMENTS 4.1.2.4Atleasttwochargingpumpsshallbedemonstrated OPERABLEbyverifying, thatonrecirculation flow,eachpumpdevelopsadifferential pressureofR2290psidwhentestedpursuanttoSpecification 4.0.5.COOKNUCLEARPLANT-UNIT 2Page3/41-12 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVI'IY CONTROLSYSTEMSBORATEDWATERSOURCES-SHUTDOWNLIMITINGCONDITION FOROPERATION 3.1.2.7Asaminimum,oneofthefollowing boratedwatersourcesshallbeOPERABLE:

a.Aboricacidstoragesystemandassociated heattractingwith:1.Aminimumusableboratedwatervolumeof4300gallons,2.Between20,000and22,500ppmofboron,and3.Aminimumsolutiontemperature of145'F.b.Therefueling waterstoragetankwith:1.Aminimumusableboratedwatervolumeof90,000gallons,2.Aminimumboronconcentration of2400ppm,and3.Aminimumsolutiontemperature of70'F.APPLICABILITY:

MODES5and6.ACTION:WithnoboratedwatersourceOPERABLE, suspendalloperations involving COREALTERATIONS orpositivereactivity changes'ntil atleastoneboratedwatersourceisrestoredtoOPERABLEstatus.SURVEILLANCE REUIREMENTS 4.1.2.7Theaboverequiredboratedwatersourceshallbedemonstrated OPERABLE:

Atleastonceper7daysby:1.Verifying theboronconcentration ofthewater,Verifying thecontained boratedwatervolume,andVerifying theboricacidstoragetanksolutiontemperature whenitisthesourceofboratedwater.Atleastonceper24hoursbyverifying theRWSTtemperature whenitisthesourceofboratedwater.Forpurposesofthisspecification, additionofwaterfromtheRWSTdoesnotconstitute adilutionactivityprovidedtheboronconcentration intheRWSTisgreaterthanorequaltotheminimumrequiredbySpecification 3.1.2.7.b.2.

ICOOKNUCLEARPLANT-UNIT 2Page3/41-15 ill 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.1REACTIVITY CONTROLSYSTEMSBORATEDWATERSOURCES-OPERATING LIMITINGCONDITION FOROPERATION 3.1.2.8Eachofthefollowing boratcdwatersourcesshallbeOPERABLE:

a.Aboricacidstoragesystemandassociated heattracingwith:1.Aminimumcontained boratedwatervolumeof5650gallons,2.Between20,000and22,500ppmofboron,and3.Aminimumsolutiontemperature of145'F.b.Therefueling waterstoragetankwith:APPLICABILITY:

ACTION:1.Aminimumcontained boratedwatervolumeof350,000gallonsofwater,2.Between2400and2600ppmofboron,and3.Aminimumsolutiontemperature of70'F.MODES1,2,3and4.Withtheboricacidstoragesysteminoperable, restorethestoragesystemtoOPERABLEstatuswithin72hoursorbeinatleastHOTSTANDBYwithinthenext6hoursandboratedtoaSHUTDOWNMARGINequivalent toatleast1%Deltak/kat200'F;restoretheboricacidstoragesystemtoOPERABLEstatuswithinthenext7daysorbeinCOLDSHUTDOWNwithinthenext30hours.b.Withtherefueling waterstoragetankinoperable, restorethetanktoOPERABLEstatuswithinonehourorbeinatleastHOTSTANDBYwithinthenext6hoursandinCOLDSHUTDOWNwithinthefollowing 30hours.SURVEILLANCE REUIREMENTS 4.1.2.8Eachboratedwatersourceshallbedemonstrated OPERABLE:

COOKNUCLEARPLANT-UNIT 2Page3/41-16AMENDMENT 04,434,AS 3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.5EMERGENCY CORECOOLINGSYSTEMS(ECCS)SURVEILLANCE REUIREMENTS Continued Atleastonceper18monthsby:1.Verifying automatic isolation andinterlock actionoftheRHRsystemfromtheReactorCoolantSystemwhentheReactorCoolantSystempressureisabove600psig.Avisualinspection ofthecontainment sumpandverifying thatthesubsystem suctioninletsarenotrestricted bydebrisandthatthesumpcomponents (trashracks,screens,etc.)shownoevidenceofstructural distressorcorrosion.

'.Atleastonceper18months,duringshutdown, by:t1.Verifying thateachautomatic valveintheflowpathactuatestoitscorrectpositiononaSafetyInjection testsignal.Verifying thateachofthefollowing pumpsstartautomatically uponreceiptofasafetyinjection signal:a)Centrifugal chargingpumpb)Safetyinjection pumpc)ResidualheatremovalpumpByverifying thateachofthefollowing pumpsdevelopstheindicated differential pressureonrecirculation flowwhentestedpursuanttoSpecification 4.0.5:1.Centrifugal chargingpump-Greaterthanorequalto2290psid2.SafetyInjection pumpGreaterthanorequalto1385psid3.ResidualheatremovalpumpGreaterthanorequalto160psidByverifying thecorrectpositionofeachmechanical stopforthefollowing Emergency CoreCoolingSystemthrottlevalves:1.Within4hoursfollowing completion ofeachvalvestrokingoperation ormaintenance onthevalvewhentheECCSsubsystems arerequiredtobeOPERABLE.

'Theprovisions ofTechnical Specification 4.0.8arcapplicable.

COOKNUCLEARPLANT-UNIT 2Page3/45-5AMENDMENT 43k,434,488,480

3/4LIMITINGCONDITIONS FOROPERATION ANDSURVEILLANCE REQUIREMENTS 3/4.5EMERGENCY CORECOOLINGSYSTEMS(ECCS)REFUELING WATERSTORAGETANKLIMITINGCONDITION FOROPERATION 3.5.5Therefueling waterstoragetank(RWST)shallbeOPERABLEwith:a.Aminimumcontained volumeof350,000gallonsofboratedwater,b.Between2400and2600ppmofboron,andc.Aminimumwatertemperature of70'F.APPLICABILITY:

MODES1,2,3and4.ACTION:Withtherefueling waterstoragetankinoperable, restorethetanktoOPERABLEstatuswithin1hourorbeinatleastHOTSTANDBYwithin6hoursandinCOLDSHUTDOWNwithinthefollowing 30hours.SURVEILLANCE REUIREMENTS 4.5.5TheRWSTshallbedemonstrated OPERABLE:

a.Atleastonceper7daysby:1.Verifying thecontained boratedwaterlevelinthetank,and2.Verifying theboronconcentration ofthewater.b.Atleastonceper24hoursbyverifying theRWSTtemperature.

COOKNUCLEARPLANT-UNIT 2Page3/45-11AMENDMENT BQ,94

3/4BASES3/4.1REACTIVlTY CONTROLSYSTEMS3/4.1.1BORATIONCONTROL3/4.1.1.1 Gild3/4.1.1.2 SHUTDOWNMARGINAsufficient SHUTDOWNMARGINensuresthat1)thereactorcanbemadesubcritical fromalloperating conditions, 2)thereactivity transients associated withpostulated accidentconditions arecontrollable withinacceptable limits,and3)thereactorwillbemaintained sufficiently subcritical toprecludeinadvertent criticality intheshutdowncondition.

SHUTDOWNMARGINrequirements varythroughout corelifeasafunctionoffueldepletion, RCSboronconcentration, andRCST,~.Themostrestrictive condition occursatEOL,withT,~atnoloadoperating temperature, andisassociated withapostulated steamlinebreakaccidentandresulting uncontrolled RCScooldown.

Intheanalysisofthisaccident, aminimumSHUTDOWNMARGINof1.3%Deltak/kisinitially requiredtocontrolthereactivity transient andautomatic ESFisassumedtobeavailable.

WithT,~lessthan200'F,thereactivity transients resulting fromapostulated steamlinebreakcooldownareminimalanda1%Deltak/kSHUTDOWNMARGINprovidesadequateprotection forthisevent.TheSHUTDOWNMARGINrequirements arebaseduponthelimitingconditions described aboveandareconsistent'ith FSARsafetyanalysisassumptions.

3/4.1.1.3 BORONDILUTIONAminimumflowrateofatleast2000GPMprovidesadequatemixing,preventsstratification andensuresthatreactivity changeswillbegradualduringboronconcentration reductions intheReactorCoolantSystem.Aflowrateofatleast2000GPMwillcirculate anequivalent ReactorCoolantSystemvolumeof12,612cubicfeetinapproximately 30minutes.Thereactivity changerateassociated withboronreductions willtherefore bewithinthecapability foroperatorrecognition andcontrol.COOKNUCLEARPLANT-UNIT 2PageB3/41-1AMENDMENT S2,408,434 I

3/4BASES3/4.5EMERGENCY CORECOOLINGSYSTEMS3/4.5.5REFUELING WATERSTORAGETANKTheOPERABILITY oftheRWSTaspartoftheECCSensuresthatsufficient negativereactivity isinjectedintothecoretocounteract anypositiveincreaseinreactivity causedbyRCSsystemcooldown, andensuresthatasufficient supplyofboratedwaterisavailable forinjection bytheECCSintheeventofaLOCA.Reactorcoolantsystemcooldowncanbecausedbyinadvertent depressurization, aLOCAorsteamlinerupture.ThelimitsofRWSTminimumvolumeandboronconcentration ensurethat1)sufficient waterisavailable withincontainment topermitrecirculation coolingflowtothecore,and2)thereactorwillremainsubcritical inthecoldcondition following aLOCAassumingmixingoftheRWST,RCS,ECCSwater,andothersourcesofwaterthatmayeventually resideinthesump,withallcontrolrodsassumedtobeout.Theseassumptions areconsistent withtheLOCAanalyses.

Thecontained watervolumelimitincludesanallowance forwaternotusablebecauseoftankdischarge linelocationorotherphysicalcharacteristics.

Thelimitsoncontained watervolumeandboronconcentration oftheRWSTalsoensureapHvalueofbetween7.6and9.5forthesolutionrecirculated withincontainment af'teraLOCA.ThispHbandminimizes theevolution ofiodineandminimizes theeffectofchlorideandcausticstresscorrosion onmechanical systemsandcomponents.

TheECCSanalysestodetermine F<limitsinSpecifications 3.2.2and3.2.6assumedaRWSTwatertemperature of80'F.Thistemperature valueoftheRWSTwaterdetermines thatofthespraywaterinitially delivered tothecontainment following LOCA.Itisoneofthefactorswhichdetermines thecontainment back-pressure intheECCSanalyses, performed inaccordance withtheprovisions of10CFR50.46andAppendixKto10CFR50.COOKNUCLEARPLANT-UNIT 2PageB3/45-3AMENDMENT 407,44k

3/4BASES3/4.6CONTAINMENT SYSTEMS3/4.6.1.4 INTERNALPRESSUREThelimitations oncontainment internalpressureensurethat1)thecontainment structure isprevented fromexceeding itsdesignnegativepressuredifferential withrespecttotheoutsideatmosphere of8psigand2)thecontainment peakpressuredoesnotexceedthedesignpressureof12psigduringLOCAconditions.

Themaximumpeakpressureresulting fromaLOCAeventiscalculated tobeIIA9psig,whichincludes0.3psigforinitialpositivecontainment pressure.

3/4.6.1.5 AIRTEMPERATURE Thelimitations oncontainment averageairtemperature ensurethat1)thecontainment airmassislimitedtoaninitialmasssufficiently lowtopreventexceeding thedesignpressureduringLOCAconditions and2)theambientairtemperature doesnotexceedthattemperature allowable forthecontinuous dutyratingspecified forequipment andinstrumentation locatedwithincontainment.

Thecontainment pressuretransient issensitive totheinitially contained airmassduringaLOCA.Thecontained airmassincreases withdecreasing temperature.

Thelowertemperature limitof60'Fwilllimitthepeakpressureto11.49psigwhichislessthanthecontainment designpressureof12psig.Theuppertemperature limitinfluences thepeakaccidenttemperature slightlyduringaLOCA;however,thislimitisbasedprimarily uponequipment protection andanticipated operating conditions.

Boththeupperandlowertemperature limitsareconsistent withtheparameters usedintheaccidentanalyses.

Thislinutation ensuresthatthestructural mtegrityofthecontauunent willbemaintained comparable totheoriginaldesignstandards forthelifeofthefacility.

Structural integrity isrequiredtoensurethat(1)thesteellinerremainsleaktightand(2)theconcretesurrounding thesteellinerremainscapableofproviding externalmissileprotection forthesteellinerandradiation shielding intheeventofaLOCA.Avisualinspection inconjunction withTypeAleakagetestsissufficient todemonstrate thiscapability.

COOKNUCLEARPLANT-UNIT 2PageB3/46-2AMENDMENT 0

ATTACHMENT 3TOAEP:NRC:1207 CURRENTPAGESMARKED-UP TOREFLECTPROPOSEDCHANGESTOTHEDONALDC.COOKNUCLEARPLANTUNITNOS.1AND2TECHNICAL SPECIFICATIONS

CURRENTPAGESMARKED-UP TOREFLECTPROPOSEDCHANGESTOTHEDONALDC.COOKNUCLEARPLANTUNITNO.ITECHNICAL SPECIFICATIONS gE!'le}Cc.-

mcusqua6=-a-6iac(gOPgiySrr~Q'r~~IrggCP~%7'.QN6c8-'loodg(~00'ye6ts0>s]<Al:CEPTABt.E QPERAT10f

~5~45~IS~RtftoCtjgrsOfnOecngl't t.t.tt.zPREHURE(PS(A)SREAKNlHTS (FRAC7LQN RATEOTHER'.PQMER,T-AVQDOEGREES)?8CQ200QQ22SQ24QQ(0.0,622.1),(0.0,633.8),(0.0,6m.8),(0.0,650.7),(0.0,660.1),(1.13,581.3),(1.08,601.<).(1.06,609.8),(1.02,621.9),(0.98,633.j),(1.20,D7.5)(1.20,586.0)(1.20,591.3)(1.20,598.9)(1.ZO,606.2)F16URE2.11REACTORCOREmEnt.rzrTSCOOKÃLC~WP~VC-4fX7L~~~~wo.7NXS2.168

'~RHR~~raiRSlra

.,rr~%aai~~aara L%EcNcrsssM

,55Rorhaa5855

'RRHhRRrRSRk%8

,rrRRRSSaERR%

~~%%%%L%%rrmumara~~arz

'rraarssash~%%

"raMaraasRR%

.,rararah~aLiraRRSra%%~%,,rrrarrrr%%rH

'arara%%ra,nrarrraaaaa TABLE2.2-1REACTORTRXPSYSTENINSTRUMENTATION TRIPSETPOINTS FUNCTIONAL UNITTRXPSETPOINT',ALLOWABLE

'VALUES1.ManualReactorTri.pNocAppli.cable NocApplicable 2.PowerRange,NeutronFluxLowSecpoint-less'chan orequaleo25%ofRATEDTHERE'LLPOWERHighSetpoinc-lessthanorequal,co109%ofRATEDTHER'iALPOWERLowSetpoint-lessthanorequalco26%ofRATEDTHERtfALPOWERHighSecpoine-lesschanorequalto110%ofBATEDTHERHALPOi:ER.3.Po~erRange,NeutronFlux,Hi.ghPositiveRaceLessthanorequalco5%ofRATEDTHERMALPOWERwithaci.Ieconstantgreaterchanorequalto2secondsLessthanorequalco5.5%ofRATEDTHEKfALPOWERwi.chacimeconscancgreaterchanorequalco2seconds(oPowerRange,NeutronFlux,Hi.ghNegativeRateXncermedf.ate Range,NeutronFluxLessthanorequal.co5SofRATEDTHER'iALPOWERwithacimeconstantgreacerthanorequalco2secondsLessthanorequa3to25%ofRATEDTHERiALPOWERLessthanorequalco5.5%ofRATEDTHER.'fAL POWERwichacimeconscanegreaeerchanorequalco2secondsLesschanorequalto30$ofRATEDTHER%6.POWER6.SourceRange,NeutronFluxLesschanorequalco105councspersecondLesschanorequalco1.3x10countspersecond7.Overtemperacure DeltaTSeeNoce1SeeNote38.Overpower Del.caTSeeNote2SeeNote49.Pressurizer Pressure--LowGreaterthanorequaleo1875psigGreacerchanorequalco1.865.psigLO.Pressurizer Pressure--High11.Pressurizer WaterLevel--HighLesschanorequaleo2385psigLessthanorequalco92%ofinstrument spanLessthanorequal.co2395psig,Lessthanorequalto93%ofinstrument span12.LossofFlowGteacerthanorequalco905ofdesignflowperloop*0reacerchanorequalco89.1.%ofdesignflowperCOOKNUCLEARPLANT-UNITL2-5loop*~/~VR.64cdatElo~~zLco~7'pll~gqgAiiENOSENT NO.9ifN151

REACTORTltlVSYS'1'DIINI'ItttHI,"N'I'ATIt)H

'I'IIIVSt".I'I'OlH'1'S IIl.ta>'I'A'I'I ttNHotel:Overtemperature AT4ATQ-K.-o121IT.>~Hl>erehI0~Indicated h'I'LIth'I'Itt'I'Itltetht.tiiHI'.I<

U=AVeragetemperat.uru I:Indicated

'I'LIlail'Ill)

TliHINAI.

VilHII((~c'Y)=Pressurizer prvu"urv, l+tlS1+vS1'-'I'imecutiLa>>L'LxIx...a:.Ii>>Lltv.lead-lagcontroller forTavg~l4st'cs~2~Laplacetransform oiterator 22sectt~IndicaLed Itt'S>>omi>>al opuraLilliJ harv""urc

('235psigor2085psig)~Thefunctiongetterated 1>yL)wlead-lagco>>troller forTdynatttiynatttccompensation

2~)Lg+~co~~~TaliIMJLJNHII&IaaLH~Operation uilh4LoopsI17lhK.~().()~)oK."UaOUllUthepcta-andf>(AI)isafunctionoftheindicated difference betweentopandbottoadttofepeter-range nuclearionchaabers; Mithgainstobaselectedbasedonaeaiury4oa~ecrgofinotnuaant responseduringplantstartuptestssuchthat:(i)Forq-qbstvaan-37ParcantandiRpercent, f(kf)0(share~andaraprcanRA77>>Tilli..i.(RRinthatopandbhttoahalvesofrespctively,andqtfqb~stotalTHEBNLLpoMERinpercentofamatiTHERMALPOMER).(ii)ForeachPercentthatthemagnitude of(q-q)oxcaoda-37porc0,,~lgtr)paatPcintshallbooutsoar(calif rsdu5adb$0.33percentofftevalueetIaavmTHERMhLPOMER.(iii)Foreachpercentthatthaaagnituda of(g-g)arcasds+kpercent, tbaagtripaatpointshallbaautoaatically rsdubadbIw~percentcflteValueetmrsTfKQNLLPOMER.C2~8 0

o4SSSU0N~Pg0CoueOverpower AT<AT(K-Ko45S1+1T-K(T-T")-f(AI)]62~here:AT~Itidic:ale.d ATatlych')'II)T~Averagetemperature, Indicated T.dtIth'l'L'D dVgTNERHALPOWER.563',DTllEENAEE'OWERK51.00300.0l77/Fforincreasing averagetexperature and0fordecreasing averagetemperature K600015forT>T";K0forT<T"61+r3SThafunctiongenerated bytharatalagcontroller torTdynaxiccompensation kg~~~kIT3Tixeconstantutilizedintheratelagcontroller torT~3~10sacs.4VQLaplacetransforx operatorI)it.e)cl.2(AI)~0Thachannel's xaxixuxxorethan~percentThechannel's xaxixuxxorathan2,1,percenttrippointATspan.Qh411notgxcooditscoxputadtX'ippQQQtrippointshallnotexceeditscoxputedtrippoint.bATspan.In.y

34.1REAcTzvzTY coNTROLsYsTBts.34.L.1IORATIONCONTROLSNJTDCNNNARCZNTAVCCREATERTHAN200FLDQTINCCONDITION fOROPERATION e)$3el.leltheSHUTDOWNMARCIAahallbegreatershanorequalto~DeltaQ/k.APB.ZnSnXTY:

&DES1,2>>,3,andS,.l.3WishsheSHtJTDOWNARCZNleeashan~+A%DeltaR/k,haaediately initiateandcontinueborationat'greaser shanorequalso10gpaofasolutionconsainigreasershanorequaLto20,000peaboronorequivalent untilsherequiredSHUTDONNARCZMtareatored.

SURVEILLANCE UIREKXNTS

(>g~4.1.1.1.1 qhaSimTDOMNhalOINshallbedacezaiuad cabedzeacza,hca'aazequalso~tDelta~a.withinonehourafterdetecsion ofantnoyerabl

~controlzod(a)and.'sleaatonceper12hourathereafter vhilethezod(a)iainoperable.

Zftheinoperable contzolrodiaiaIIovabLe orunsrippable, theaboverequiredNUXDOQWMhIQZNahallbeverifiedaycepsabLe vishanincreaeed allovanca forthevithdravn vorshofth>>fundableorunsrippable controLzod(a).b.%heninMDC1oz'DC2vithXetfgreaterthanorequaLto1.0,asleastonceyer12hourabyverifying thatcontrolbaakvishdravaL iavithintheLtaitaofSpecification 3.1.3.5.c.%heninMDC2vithXeffleaashan1.0,vithin4houzapriortoachieving reactorcriticality byverifying thatsheyred'icsed czittcalcontzoLzo4poaisioniavithintheLiILLtaof5pecificasion 3.1.3.5.d.?rior.teinitialoperation above5%RATEDTHHQtAZ.PtÃEXaftereachfaeLLoading.byconai4eration ofthefactorsofebelov,vishsheconszoLbanIcaattheaaabauaineertion LinisofSpecification 3.1.3.5.+SeeSpecialTeatException

$.10.1.,COOKNUCLEARBLASTUNXT13/41-1mme'O71tcV>c)48

REACTIVITY CONTROLSYSTEMSCHARCINCPUMP-SHUTDOWNLLGTINCCONDITION FOROPERATION 3.1.2.3a.Onechargingpumpintheboronin]ection flo~pathrequiredbySpecification 3.1.2.1shallbaOPERABLEandcapableofbeingpoweredfromanOPERABLEemergency bus.b.Onechargingflovpachassociated vithsupportofUnit2shucdoMnfunctions shallbeavailabla.*

APPLICABILITY:

Specification 3.1.2.3.a.

-MODES5and6Specification 3.1.2.3.b.

-AcaQtimesvhenUnit2isinMODES1,2,3,or4.ACT'KON:a.Mi.chnochargingpumpOPERABLE, suspendalloparacions involving COREALTERATIONS orpositiveraaccivicy changes.~

b.Qi.chmorachanonechargingpumpOPERABLEorarithasafetyinfection pump(s)OPERABLEvhanthatemperature ofanyRCScoldlegislesschan0orequalco152F,uxor.assthareactorvassalheadi.sremoved,removetheaddi.tional chargingpump(s)andchesafetyinfection pump(s)mocorcircuicbreakersfromcheelaccrical povercircuitvichinonehour.c.Thaprovisions ofSpecification 3.0.3aranotapplicable.

d.Inaddi.tion cocheabo~e,shenSpecificacion 3.1.2.3.b isipplicable andcherequiredflovpathisnocavailable, recurncherequiredflo~pathcoavailable scacusvichin7days,orprovideequivalenc shutdowncapability inUnit2andreturntherequiredflovpathcoavailable statuswithinthenext60days,orhaveUnic2inHOTSTANDBYwithinchenext12hoursandHOTSHUTDOWNvithinchefolloving 24hours.e.Tharaquiremencs ofSpecification

3.0. 4arenocapplicable

<henSpacifi.cacion 3.1.2.3.b appli.as.

SURVEILLANCE UI~ENTSoir~sC~'~~

4.1.2.3.1.

TheaboverequiredchargingpumpshallbademocratedOPERABL'E byveri.fying, chatonrecirculation flov,chapumpdevelopsa44si4a~pressureofgreacarchanorequalco+3&9ps+MhancascadpursuantcoSpeci.ficacion 4.0.5.~Z,go*Amaximumofonecentrifugal chargingpumpshallbeOPERABL'E wheneverchetemperature ofoneormoreofcheRCScoldlegsi.slasschanorequalco152F.o~forpurposasofthisspecification, additionofvatarfromchaRUSTdoesnotconsti.cuca apositivereaccivicy additi.on providadthaboronconcentration incheRUSTis.graacarchantheminimumrequiredbySpecification 3.1.2.7.b.2.

COOKNUCLEARPLANT-UNIT13/41-11AMENDMENT No.yg,ZAP.Z7Z,ZN

3.1.2.4AtleasttvochargingpumpsshallbeOPERASLE.

MODES1,2,3,and4.hRDQH:arithonlyonechargingpumpOPERhSIZ, restoreatleasttvochargingpumpstoOPERhhLEstatusvithin72hoursorbeinHOTSThHDbYvithinthenext6hours;restoreatleasttvochargingpumpstoOPEMLEstatusvithinthenext48hoursorbeinCOLDSHUTDOWNvithinthefolloving 30hours.pygmy5+hlYHf44.1.2.4htleasttvochargingpumpsshaQbedeonstrated OPEMLEbyverifying, thatonrecirculation flov,eachpumpdevelopsapg&aekmge-pressure ofgreaterthanorequalto~paidvhenteatedpursuanttoSpecification 4.0.S.x~soJ(0(COOKNUCLEARPLurr-UHZTl3/4l-l2~mmHo.os,164

IRcACTIVITY CONTROLSYSTEHS1BORATEOWATERSOURCES-SHL'TXWNI<<LIHITING CONDITION FORO~ciRAT;ON

<I~3.1.2.7Asaminimum,oneofthefollowing boratedwatersourcesshallbe:iOPERABLi:

a.Aboricacidstoragesystemandassociated heattracingwith:2.3.Aminimumusab'.eboratedwatervolumeof4300gallons,'etween20,000and22,500ppmofboron,andAminimumsolutiontemperature of145'F.b.Therefueling waterstoragetankwith:l.Aminimumusableborated~atervolumeof90,000gallons,2.Aminimumboronconcentration of2400ppm.and3.Aminimumsolutiontemperature of~F.0APPLICABILITY:

HOOKS=.'."."6.ACitON:Withnoboratedwatersor"eOPERABLE, suspendaloperations involving COREi"ALTiRATIONS orpositive~eacivitychangesuntilatleastoneboratedwater'.sourceisrestored.oOPERABLEstatus.~~'SURVEILLANCE REUIRcvENTS 4.1.2.7Theaboverequiredboratedwatersourceshallbedemonstrated OPiRABLi:

Ia.Atleastonceper7daysby:2.3.Verifying theboronconcentration ofthewater,Verifying the.waterlevelvolumeofthetank,andVerifying theboricacidstoragetanksolutiontemperature whenitistnesourceofboratedwater.b.Atleastonceper24hoursbyverifying theRWSTtemperature whenitistnesource.ofborptedwater.4"Forpurposesofthisspeci.ica ion,additionofwaterfromtheRMSTdoes11notconstitute apositivereactivity additionprovidedtheboronconcentra-

~;tionintheRMSTisgreatertnantheminimumrequiredbySpecification ii~.i.~.7.b.Z.

9.C.COOK-UNIT13/41-15Amendment No.$2.ill

II<EACIIVITYCONTROLSYSTENS"BORATEOMATERSOURCES-OPERATIONS

~~I1;:LIH:T:NGCONDITION FOROPEPATION

~~~~~~::3.1.2.8Eachofthefol/owing beratedwatersourcesshallbeOPERABLE:

~Ia.Aboricacidstoragesystemandassociated heattracingwith:l.Aminimumusableboratedwatervolumeof5650gallons,II1'~~I~~~,I~1~~Ii~~I~I2.Between20.000and22.500ppmofboron,and3.Aminimumsolutiontemperature of145'F.b.Therefueling waterstorage'tank with:I.Aminiru.-.

contained volumeof350,000gallonsofwater,2.Between-'".3and2600ppmofboron,and~~~IAPPLŽ.'8IlI:Y:I:I'ACION:3.Amini-.-solutiontemperature ofi9'F.7dDOGESI,",3and4.~~~~~~Miththeboricacidstoragesysteminoperable.

restorethestoragesystemtoOPERABLEstatuswithin72hoursorbeinatleastHOTSTANOBYwithinthenext6hoursandboratedtoaSHUTDOMNHARGINequivalent toatleast1"~k/kat200'F;restoretheboricacidstoragesystemtoOPERABLEstatuswithinthenext7daysorbeinCOLOSHUTOOMNwihinthenext30hours.~0II~~~~b.Miththerefueling waterstoragetankinoperable, restorethetanktoOPERABLEstatuswithinonehourorbeinatleastHOTSTANOBYwithinthenext6hoursandinCOLOSHUT-00MNwithinthefollowing 30hours.:SU'1"ILLANCEREUIRE1".E" 5~4.1.2.5Eachborateca:=r;=rceshallbedemons-rated OPERABLE:

CsQK-UNIT13/+1-10Amendment No.58,

TABLE3.2-1DNBPARAMETERS PARAMETER ReactorCoolantSystemTavgPressurizer PressureLLKZTS4LoopsinOperation atRATEDTHE'-hfAL POWERo*~59,73oF>2050psig*ReactorCoolantSystemTotalPlovRate++*>HEHMOO.gpm*Indicated averageofatleastthreeOPERABLEinstrument loops.Limitnotapplicagle duringeitheraTHELMLPOWERrampincreaseinexcesof5percentRATEDTHERMALPOWER.perminuteoraTHER.'tAL POWERstepincreaseinexcessof10percentRATED.'THEKf~

POWER.Indicated value.COOKNUCLEARPLANT-UNIT13/42-14AMENDMENT NO.~~if/',152 I,t'I

'(TABLE3.3-3Concinuad EHCKHEERED SAFETYFEATUREACTUhTION SYSTEMINSTRUMENTATION FUNCTIONAL UNITDEl.ETZ,f.SceamFLovinTerSceamLines-High TOTALNO.OFCHANNELSRININJHCHANNELSCHANNELSAPPLICABLE TOTRIPOPERABLEMODESACTIONFourLoopsOperacing 2/steamline1/samI/steamliney2Linesteam1nes2,3~14*reeLoops0racing2/opracIngsteamineL~/any1/operacing 3~operating steamlinesteamLineL5COZNCZDENT ZTHEITHER(.(T~~Lov~LovavgFourLoopsOperating ThreeLoopsOperating OR,COINCIDPITHSceamLinePressure-Lov1-T/LoopavgTvtoera8i,ngLoopTanylo'sgL~inanyoperacinLoopLTin3anyNooperating loopsL5I.Tany,1,2,3ERERLa*3LoofsFourLoopsOperating ThreeLoopsOperating 1pressure/

I.oop1pressure/

operating loop2pressures anyloopsL~presssure inanyoperacing loop1pressureany3loops1pressureinany2operating loops12,3sEeI.5,.INUCLEARPLANT-UNIT13/43-17eu2tDHENT NO91,fgg,153

r(TABLE3.3-3Continued ENCZNEEREO SAFETYFEATUREACTUATION SYSTEMINSTRUMENTATION FUNCTIONAL UNITNININJNTOTALNO.CHANNELSCHANNELSAPPLICABLE OFCNANNELSTOTRIPOPERABLEKODESACTIONCOINCIDENT METH%%%BBR-Tu~LovuLovavgFourLoopsOperating ThreeLoopsOperating 1T/loopavg1Tavoperating loop2T'any'Y5opa 1¹¹¹Tinanyoperating loop1TanyYgloops1Tinan/cvooperacing loops1,2,3¹¹'5SteanLinePresaure-Lo~j4FourtoopoOperating 1pressure/

loop2pressures anyloops1pressureI.,2,3¹¹14any3loopsThreeLoopsOperating 5.TURBINETRIP&FEEDMATER ISOLATION 1pressure/

operating loop1¹pressureinanyoperating loop1pressure3¹>>inany2operacing

'oops15a.SteamGenerator WacerLevel--High-High3/loop2/loopin2/loopin1,2.3anyoper-eachoper-atingloopacingloop,COOKNUCLEARPLANT-UNIT13/43-21ANENDNENT NO.9f,f29,153 00 ENGINEERED ShFETYFEATURESINTERLOCKS DESICNATIONF-12ComrrZONumSETFoaeVith2of3pressuriser pressurechannelsgreaterthan.orequalto1915psig.Vi,th2of4TchannelslessthanoreqqaltoSetpoint.

Setpointgreaterthanorequalto541FFUNCTZON2ILpreventsordefeatsmanualblockofsafetyinfection actuation onlospressurizer pressure.

2-12allovsthemanualblockofsafetyin]ection gcrubriou odlovateaalinepressure.

~~c~c-5s74'~AIR/8i~~j7~N~/A/A'5f~+

AM.4-Lgigfects stecadumpblocks.Vith3of4Tchannelsabovethereset%44+)augPoint@Pgc,~c~io(08F&~iyttE.hafuwWc.~dFShFBTy/~J8c77on/A~gg~pyp~

~<A~Ag8+csr~PRcssu/Z, 3/43-23a ThSLE3.3-4ENCQiEERED SAFETYFEhTUREhCTUhTION SYSTEMINSTRUMENTATION TRIPSETPOINTS FUNCTIONhL UNITTRIPSETPOIHTl.ShFETYINJECTION, TURSINETRIPeFEEDVhTERISOLhTIOH eANDMOTORDRXVENFEEDVhTER PUMPSa.ManualIatttation b.huteaattc Actuation Logtc-------------

SeeFunctional Unit9-------------

Hothpplfcabla Hoehpplicabla coContatIcant PressureHighd.Pressurtser Pressure--

Lmra.DSMerenttaL PressureSecveanStecaLtnes--HigheaaFloetnSteaas-HighCocidanteha-Lcw-rSteaaLinePressura-LmrLesschanorequalto1.LpsigGreaterchanorequaleoLdlSpaigLesschanorequalto100pstssthan6orqualto1.x10lbsfroa0%dto20%1d.ILpfrofa1.42xLOlbsa)20%1dto3.d!x0Ibs/hr00%loadTgreatertoraeo541PLesschanorequalto1.2psigCreatarthanorequaltoLd05psigLessthanorequaLeoJ112pstssthan6orqualtoL.6x10LbaCroa0%dto20%ad.Ideafrom1.5610lbag20%adto3.9310lbs/hrt100%loadTgreateroreto539F'+eaterthenorequal/+eaterthanorequaleo'to-500paigstaaaLine4d0pggsteaalinepressurepressureCOOXHUCmaPLhHT-UNITJ.3/43-24hMEHDMEHT HO.49,1ZS153

TARLK5.5-4Continued ENCINEERED ShFETYPEATQREhCTUATION SYSTEMINSTRUNENTATION TRIPSETPOIBTS FUNCTIONAL UNITZ.Contatamcnt Radio-acttvtty--High TrainA(VRS-LLOL.

ERS-1301, ERS-1305) 3.Containment Radio-activity--High Tratn5(VRS-1201, ERS-L40L, KRS-N05)TRIPSETPOINTSeeTable3.3-6SeeTable3.3-6NotApplicable NotApplicable 4.SmxLINEISOLATION a.manualSeefunctional Unit9b.Automatic Actuation Logicc.Containment Pressure--

Htgh-High d.SteamllovtnTvoSteamLines--High Coincident vithT--Lov-Lov

~NotApplicable LessthanorequaLto2.9patgLessrhan6orequaltoL.42x10Lbs/hrfroaOtloadto20%load.LtparfroaL.42x10lbs/hrag20%loadtoS.bbx10lbe/hratLOOtLoad.NotApplicable Lessthanorequalto3psigLessthaa6orequalto1.56x10Lbs/hrfrom0%Loadto20%load.ILparfrom1.56xLOLbs/hr6at 20%loadto3.9310Lbs/hrat100%load.5(gagc.(PcPNM~~"Lou)5~TURBINETRIPAHDPEEDVATER ISOLATION Tgreatertheaorc@lto54LPCreaterthanorequalto500patgsteaaLtaepressureTgreaterthanorc~Lto539FCreaterthanorcquaLto48Q'stgsteamLinepressurea.SteamCeacrator VaterLevel--Htgh-High Lessthanorequalto67%ofnarrov-range tnstrmacat spaneachste4%generator Lessthanorequalto6Nofaarrov-raagc instrument spaneachsteaagenerator COOXNUCLEARPLANTUNIT13/43-26JQKHDNEHT NO.gg,f/),153 TABLE4.32ENCAGE?DAD SAPETTPEATmEACTUATION SYSTEMINSTRUMENTATION FUNCTIONAL UNITTRZ?ACTUATING CHAHHELDEVICECHQWELCHAHNELTUMCTIOHAL OPERATIONAL CEECEClEEEIUTECH'TESTTESTNODESINQHZCHSURVEILLANCE L.SAPETTINJECTION, TURSZNETRIP,FEEDVATER ISOLATION, AHDMOTORDRIVEHAUXILIARY FEEDVATER PUMPSa.NanuaLZnitiation b.Automatic Actuation H.A.H.A.H(2)H.A.1,2.3.4-------------------

SeetuactionaL Unit9----~-------------

LogiccoContainment Pteaaeux'e-High d.Preaautiaet Pt'eas-useed%H.A.H.A.,1,2.3L,2,3e.Dtffet'entiaL Preaa-ut'SecweenSteamLineaHighH.A.H.A1,2~31.2~3<Pc'eaacceeeLNI'5~/A*<wc,2.CONTAQQKHT SPRlTa.KaaualInitiation aeeeeeeaeeeeeeeaeee SeefunctionaL Unit9aaeeaeaeeaeaaaaeaa b.Automatic Actuation LogicH.A.H.A.H(2)K.AT1,2,3,4c.Containment Preaa-uz'e-High-High

',M(5)H.AL,2,3Caor.NUCLEARPLurr-mIT1,3/43a51SEEECEEEE EC.gg,(P, FUNCTIONAL UNITTABLE4.3-2Continued EHCZNEERED SAPETTTEA'CURED ACTUATION STSTEMIHSTRQlKHTATZON SURVEZLIANCE REUIRDQBiTS Tr"'ZPACTUATZNC MODESINCHANKf"~DEVICE%HIGHRGLHHELCHANNELTUNCTZONALOPERATZOHAL SURVEILLANCE CZZCZCALZZlULTZCII TZZTTZST4.STEP~<<LINE ISOLATION a.Manualeaaaaaeaaeaeaeaaeaea SeeTuactioaal Unit9b.Automatic Actuation H.h.LogicH.A.M(2)N.A.-I.,2,3c.Containmeat'Preaa>>

ure--High-High d.SteamTlovia(,TvoSteamLiaea-HighCoiacident vithTavgaLoveLovSvhgmNuCPQGSRtPc/

5.TURBINETRZPAHDTEEDVATER ISOLATION a.SteamCene'rator MaterLevel--High-High6.MOTORDRIVENMXZLZARTFEEDWATERPUMPSa~SteamCeaerator PaterLevel--Lov-Lovb.4kvEuaLoaaofVoltagec.SafetyIa]ectioa d.LocaofMainTeedPumpaH.A.H.A.kN.AHA.M(3)M(2)H.A.H.A.H.A.H.A.H.A.H.A.H.A.1,2,31,2,3>i2>Z1,2~3lc2c31,2,31,:2,32coor.NUCLEARplurr-UHZT13/43-33Jl!ZHCHZBT IIC~Zgg)fZf,

3.4.2hminimumofone,pressurizer codesafecyvalveshallbeOPERQLEarithaliftsettingof2485PSIC++a.*NODES4and5.LEONE:Pithnopressurf,e'er codesafecyvalveOPERABLE:

a.'Immediately'uspend alloperations involving positivereactivity changeable'nd placeanOPEMLERHRloopintooperation intheshutdowncoolingmode.b.Imiediately renderallSafetyIn)ection pumpsandallbutonecharqui.ng pumpinoperable byremovingtheapplicable mocorcircuicbreakersfromtheelectricpovercircuit%thinonehour,.r4.4.2Thepressurizer codesafetyvalveshallbedemonstrated OPEMLEperS'urveillance Requirement 4.4.3.*Theliftsettingpressureshallcorrespond toambientconditions ofthevalveatnominaloperating temperature andpressure.

~Forpurposesofthisspecification, additionofvaterfromtheRUSTdoesnocconstitute aposicivereactivity additionprovidedtheboronconcentracion intheRUSTisgreaterthantheminimumrequiredbySpecification 3.1..2.8.b.2 (MODE4)or3.1.2.7.b.2 (MODES).D.C.COOK-.ORXT13/44-4~MEHTttoSB,1,20

3.4.3Al.lpreaauriaer code,aafetyvalveaahallbeOZXMZXwithalifeaettiagotX4aSZSZCg~.hQXIQE:Withonepresauriaez codesafetyvalveinoperable, eitherreatorethe..inoperable valvetoOPEMLEstatusvithinlSrLinutesorbeinHOTSHUTDOVNvitharr12hours.4.4.3Hoadditional surveillance requirements otherthanthoserepairedbySpecificatioa 4.0.S.Thel.haftsettingpressureshallcorrespond toambientconditions ofthevalveatnoaineloperating teaparatura andpressure.

d.AtleastonceperLgmonthsby:1.Verifying aueomaetc isolation andinterlock actionofeheRHXsystemfromtheReactorCoolantSystemvhentheReactorCoolantSystempressureisaboveC00psig.2.Avisualinspection ofthecoataiameat sumpandverifying ehaethesubsystem suctioninletsareaotrestricted bydebrisandchatehesumpcomponents (trashracksscreens,etc.)shovnoevidenceofstructural distzessorabnormaLcorrosion+

eohtleaseonceper'8months,duringshuedovn, by:1.Verifying chateachautomatic valveintheflepaehactuateseoitscorrectpositiononaSafetyInjection testsignal.'2.Verifying theeeachofthefoLLovtag pumpsseaztautomaeically uponreceiptofasafetyinjeceioa signal:a)Centrifugal chargingpumpb)Safeeyinjection pumpc)Residualheatremovalpumpg]QPEg~i~Syverifying ehaeeofthefolloviag pumpsdeveLopstheindicated 44aahe~>pressuza onrecirculation flovvhentestedpursuaattoSpecificatioa4.0.5.22yO~1.Centrifugal chargingpumpgreaterthanorequalto%405psig-I3A,&2.Safetyinjection pumpgreaterthanorequalto+09psig-IS'O3.keaiduaLheatremovalpumpgreaterthanorequalto404psig-Syverifying thecorrectpositionof<<achmechanicaL stopforthefoLLovtag Eiaergency CozeCoolingSysteathrottlevalves:1.Viehin4hoursfoLLovtng completion ofeachvalvestroking, operation ormaintenance oaehevalvevhentheECCSsub-systemsarerequiredtobeOFKRASLX.

COOKNUCLEhRPLANT-UNZTL3/4$-5

EMERGENCY CORECOOLINGSYS""=.'iS REFUEL!HG MATERSTOPPAGETklsi(LIHITINGCOHQI;IOHF".R'3P-""=-'".:"H 3.5.5Therefueling wa:erstoragetank(RMST)shallbeOPERABLEwith:1a.Aminimumcon'tained volumeof350,000gallonsofboratedwater.b.Between24OOand26OOppmofboron,andc.Aminimumwater=e~pera"ure ofPPF.APPLICABILITY:

HOOES1,2.3and4.7'0ACTION:Withtherefueling wa-.e~s=oragetankinoperable.

restorethetanktoOPERABLEstatuswithin1hour".roinatleastHOTSTANOBYwithin6hoursandinCOLOSi{UTNMNwithinthef"i:.~ng30hours..SURVEILLANCE REUIRE.".-N".S i4.5.5TheRMSTshallbede.znstrated OP=P~BLEa.Atleastonceper7daysby:Verifying trecontained boratedwatervolumeinthetank.and2.Verifying theboronconcentration ofthewater.b.Atleastonceper2-'oursbyverifying theRMSTtemperature.

IrjlII.tIID.C.COOV,-UNIT1Amendment No.$3.III

HTSConcued4.7.I..2Ecchaus:iliaxy feedvaeer pumpshaLlbedemonstrated OPERABLEvheatestedpursuantcoSpecification 4.0.5by:aib.Verifying thateachmotordrivenpumpdevelopsanequivaleac discharge pressureofgreaterthanorequaleoQR-psigae60Pinrecirculation flovc,-c.-

o+~kROi@0Verifying that'hesteamcarbinedrivenpumpdevelsanequivalent discharge pressureofgreaterthaaorequaleopsigac60Pandataflovofgreaterthanorequaleo700gpmvhenchesecondary sceamsupplypressureisgreaterthan310psig.Theprovisions ofSpeciiicacioa

4.0. 4arenotapplicable

forencryineoHOOK3.c.Verifying thateachnoa-automatic vaIveintheflovpaththatisnoclocked,sealed,orothervfse securedinpositionfsinitscorrectposition.

d.Verf.fyfng theeeachautomatic vaLveintheflovpathisinehefullyopenpositionvhenevercheauxiLf.azy feedvater systemisplacedinautomatic concroLorvhenaboveLO\RATEDTHECALPOQc3..Thisrequirement isnocapplicabl.e forthoseporcioasoftheauxiliary feedvaeer systembeingusedintermittently tomaintainsteamgenerator vaterlevel.e,Verifying aeleaseonceperI.8monehsduxingshuedovnthaeeachautomatic valveinthefIovpaehactuatestoiescorrectposI.cion uponrecefpcoftheapproprfate engineered safeeyfeaturesaccuaeioa testsignalrequiredbySpeciffcatioa 3(4.3.2.Verifying atleaseonceper18moachsduringshucdovathateachaus:i.Liaxy feedvaeer pumpscaresasdesf.gned aucomacically upoareceiptof.theappropriate engineexed safetyfeaturesactuation cescsignalrequixedbySpecification 3/4.3.2.Verifying aeleastonceper18moaehsduringshutdovnthattheunf.tcross-cie valvescaacyclefulltravel.FoLLovtng cycling,thevalvesvillbeverifiedtobeintheMclosedpositfons.

NATItEInaccordance withthecoderequirements specified inSection4.1.6oftheFSAR,withallowance fornormaldegradation pursuanttotheapplicable Surveillance Requirements, b.C.~VLUME5.4.2Forapressureof2485psig,andForatemperature of650'F,exceptforthepressurizer whichis680'F.approximately 12.466cubicfeetat0%steamgenerator tubepluggingand11,551cubicfeetat30%steamgenerator tubepluggingTh*K"aIPd*Ieeubte-feet atanominalT,,of70'F.5.5MERENYORELTNYTEMS5.5.1Theemergency corecoolingsystemsaredesignedandshallbemaintained inaccordance withtheoriginaldesignprovisions contained inSection6.2oftheFSARwithallowance fornormaldegradation pursuanttotheapplicable Surveillance Requirements, withoneexception.

Thisexception istheCVCSboronmakeupsystemandtheBIT.5.6LTRARITILTTY-PENTL5.6.1.1Thespent.fuelstorageracksaredesignedandshallbemaintained with:A~equivalent toless.than0.95whenQoodedwithunborated water.Anominal8.97inchcenter-to-center distancebetweenfuelassemblies placedinthestorage'racks.C.Thefuelassemblies willbeclassified asacceptable

'forRegion1,Region2,orRegion3storagebasedupontheirassemblyaverageburnupversusinitialnominalenrichment.

Cellsacceptable forRegion1,Region2,andRegion3assemblystorageareindicated inFigures5.6-1and5.6-2.Assemblies thatareacceptable forstorageinRegion1,Region2,andRegion3mustmeetthedesigncriteriathatdefinetheregionsasfollows:COOKNUCLEARPLAiNT-UNITIAMENDMENT NO.I,,169CORRECTED PAGE I

12.lSAFETYBASES4LoopOperation VescinghoQse POOL(15xL5OFh)(WRS-1Correlation)

  • TypicalCellThimbleCellCorrelation LimitDesignLimicDNBRSafecyAnalysisLimitDNBR'.L7~3-/,g3I.VO1.17~~~J.zxl.VaThecurvesofFigure2.1-1sho<<chelociofpointsofTHEK<ALPOt:ER,ReactorCooLancSystempressureandaveragetemperature for<<hichche='ni"umDNBRisnoess='.".an"'.".e-ppL'cable designDNBRlimit,orcheaveragee..chalpy ac"he:essel exicisequalcocheenchalpyofsacuraced Quidtrepresents typicalfuelrodrepresents fuelrodsnearguidetube,0CookNuclearPlantUnit1B2-L(a)g~I@4'

~oTheparerlac~Negati~mateTripprovidesprotacti.on forcontrolroddropaccidence" At:hf.ghpovec4roddropaccidentcouldcaus~1ocalf1~peWagWi~ornacalleanuxLco~arativelocalDNhkcoexist.ThaPover'Range NegativeRaceTripvf11preventthisfroaoccuziing bytrippingthereactor.Nocreditistakenforoperation ofthePoverRangeNegativeRataTripforthosecontrolroddropaccidents forvhichDNfffLjvillbegreaterthantheapplicable deafgnlimitNQvalueforeachfueltype.Intermediate andSourceRaneNuclearFlux7heIntermediate andSourceRange.NuclearFluxtripsprovidereactor-eoreprotection durf.ngreactorstart'p.Thesetripsprovideredundant protection tot..elovsetpointripofthePoverRange,NeutronFluxchannels..he sourceRangeChannels<<LLLinit'ateareactortripac~r5about10countspersecond.blessmanuallyblockedvhenP-6becomesactive.heInter=ediatc RangeChannelsvill,'niciataareaccortripateve,oporanal0approxima ey25percentofRAZHI~aKJLL2:;-R..n'asscanalyb'octad'-..en2-'0beco=esactive.Nocreditvas=<en="."operatono"=he="'psassociated ci"heithertheIntermediate I1cr.""-=eRangeha".zesin"heac"'Centa..a'ysas; hovever,theftfnttonalckoabI(attnespecifie'dtipsettingsisrequiredbythiss=ec=icstiontoe."".ance

=.".ec;era'e.Lab':yo=--.ReactorS'sem(4..".eCvertemperature delta.t"'pprovidescoreproteiontoprevenc"YSa"""bina"'ons ofpressure, pover,coolanttemperature, andaxfaL"o-crCist"'but'on.

providedthacthetransient isslovvithrespecttopipingtrinsi.tCelaysfromthacoretothetemperatu eCececcors (abouc4seconds),

andpressuref.svithintherangebecveantheHighandLovPressurereactortrips.Thissetpointincludescorrections forchangesildensityandheatcapacityofvacervichcemperacure anddynamiccompensacion foriindelasfromthecoretotheLooptemperature Catectors.

Thereferceaverageemperacure an~dreerencecratingpssure(P')reseceqtothefupoverindaced7avgathenomfnRCSoperangpressurrespecciv y,:oensuproctf.onoftcorelimf.andtoprervetheattiodcimoftheOvertperaturelta7triportherangof'fullpo"averagetemperauresassumeinthesatanaLsestno:=aaxapoMerdistribution, thisreactortripfmicf.saLvaysbelow=hecoresafetyLimi<<sshovninFigure2.1-1.IfaxiaLpeaksaregreaterchandesign,asindfcatedbythedifference betveentopandbottompokierrang>>nucleardetectors, thereactortripf.sautomatically reducedaccording tothenotations Ln7abla2.2-L.DELE(8COOK~~CLM%PLAÃEUNI7L52-4h~EÃM~7NO~7)L26 TheOverpover deltaTreactorertpprovtdeiassuraace offueLtneegriey,

~.g.~aomeletng,underaLLpossibleoverpover coadtttoas, Ltittstherequiredrangefot'vertenyeracure deltaTprotection, aadyrovtdes'abackuptoeheHighHeueroaPlus:,erty.

Theseepotnttncludaacorrections forchangestndensityandheatcapacityofvacervithtewperaeure, anddyaaatoconpeasattoa forttadelesfroathecoracothelooytemperature detectors.

Therefereaaveragetperaeura()isseeetoehe1overtndtcadTavgtoesurefuelegrttydurgoverpoveconditions orP~T~ehrangeofLpontavag>>thereesaeeaetn'heaaeyaaalsis.~overpover etareactorcrpprovesprotectonorc-uyprotection foracpoverstaaultne breakevents.Credttveatakenforoperation ofthistriptnthest~anLfnebreaknasa/energy releasesoutsidecoaeatnnene aaalysts.

Znaddttton, itsfunctional capability aethesyecified tripseettngisrequtredbythisspecification coenhanceeheoverallreliability ofehereactorproeeceioa syseea,.?resaurtser PreueThetressurtaer HighandLovPressuretripsareprovtdedtoLtntethepressurerangetnvhichreactoropecatioa tspemftted.

TheHightressuretriptsbackedupbyeheyrassuriaer codesafacyvalvesforPCSoverpressure protection.

andistherefore seeloverthaneheseepressureforthesevalves(2485pstg).TheHighPressuretripprovidesproeection foraLossofhccernalLoadevene.TheLovtressureeriyyrovtdesprotection bytrfyptngchereactorincheeveaeofalossofreactorcoolaatpressure.

PressutxerVaeerLaveThePressurizer HighQatarLevelcripensuresprocectioa agatascreactorCoolaneSystemoverpressurtxaetoa byLtmttiagthevaeerleveleoavolunesufficient toretainaaceanbubbleendpraveatvaterreliefthroughthepressurtrer safetyvalves.Theprassurtaer highvaterlevelcripprecludes vaeerrelieffortheUncontrolled RCCAN.thdraval aePoverevent.cooKNUCLuk?err-UNITlI2-5AHzHuKNTNo-XSSMltd,158 00 3/4~L~a~~CQHTRQLSY51!IASES34.L.LEmVTTQHecmZOL34.L.L.Land3/4.L.L.2 SH~iVHNARC'sufficient SHUZDQRfMMtCDf<<nsuzestheeI)thereactorcanb>>lsadesubczitioaL fzoaalloperating conditions, 2}thereactivity tzansiencs associated vithpostulated accidentconditions azecontrollable vithinacceptable Iinits,and3}thereactorvtLLbeIaincained sufQciently subczitioal toprecludeinadvertent criticality intheshutdovncondition.

/.3SjglTDdt'5?QRCXH recpLx'aae vazydcooghout coeLifeasafunctionoffueLdeplecion, RCSboroncentzation, andg5T.Theaoatrestrictive condition occursttOt.,vith'7atnoHidoperating tenperatuze, andisassociaavithaposeZiked stelaLinebreakaccidentandresulting unconczolled Rcooldovn.

ZatheanalysisofMaccident, anin~SHUTÃtVNhRCZHof~~DeLtaklkisinitially zeqhzedtoconc=oLthereacivitytransient andautoaatic'SP isassumedtobeavailable.

Vf.chTivglessthan200F,thereactivity czansients resulting fxoaapostulated steanLinebreakcooldovnazalintelandaLiDelta~SRZDONMH'-iprovidesadecpzata protection forthisevent.Tha%KEN%QRCIMreqxheaents arebaseduponthelimitingconditions descibadaboveandaraconsistent vichgQsafetyanalysisassunptions

~34.L.L.3hORQHDILIOH~0hainu+floerataofatLeast2000CCKprovidesade~ia~g.pzevencsstzatf.ication andensuresthatreactivity changesvillbegrsduaLduringboronconcentration reductions intheReactorCoolantSystaa.kflovrataofatleast2000CPIvQLcirculate anequivalent ReactorCoolantSystaavoluneofI2,02plusorainusI00cubicfeetinapproximately 30ainutas.Thereaccivity changerataassociated vithboronreductions

<LIthezefora bevithinthecapabg,ity foroperatorrecognition andcontzoL.34.L.'L.4MOOERATOR TBPKULTURE COEFFICIENT TheIinitations on8XCareprovidedtoensurethattheassunptions usedintheaccidentandtransient analysesremainvalidthrougheachfuelcycle.Thesurveillance requirement formeasuzeaent oftheETCatthebeginning, andnearth>>endofeachfuelcycleiaadequatetoconfizntheNCvaluesincethiscoefficient changesslovlyduepzincipaLly toth>>reduction inKCSboronPlAPi,-UHZTIh3/4I-LNO.1$,723,t48

4.4R:-~:"RCOO&v-SvS.=M%ASKS3/4.4.1R~aCTQRCOO'>M~~OPSst~~puhxv$lsl.imTheplaneisdesk~edtooperaceAallreactorcoolantLoopsfnoperacion, aadmaintainONERabove++}-during allnormaloperations andanccpatedtransients.

ALossoffloviatvoloopsvQLcauseareactorcr'pifoperating aboveP-7(llpercentofRATEDTHERNhLPQQKR)vhileaLossofilawinoneLoopvillcausaareactortripifoperating aboveP-8(31perceacof.ReTEDTHHQfALPOWER).,ZnNODE3,asi'aglereactorcoolancLoopprovidessuffic'eac heacremovalcapability forremovf.ag decayheat;hovever,singlefailureconsiderations requirethatc~oloopsbeOPERON.Threeloopsarerequired-obeOPHASL~an4tooperateifthecontroLrodsare.capableofvithdraval and-".ereactortr'pbreakersaeclosed.nerequirameat assuresadequateDNRmargin'n=".ea~eatofanuncontroLLed rodvi-Crave'"..

"h'X=ode.XaNODES4and5,asinglereactorcoolantlooporRHRLoopprovidessufzicianc heatremovalcapability forremovtagdecayheat;butsinglefailureconsidaratf.ons requirechataLeasttvoloopsbcOPMLE.Thus,ithereaco-coo1xatloopsarenocOPML=,thisspecification requirestvoRHRloopstobeOPRVJLL~.

Theoperation ofoneReactorCoolantPumporoaeRKLpumpprovidesadequateilovcoensuremincing,preveacstratiQcatioa aadproducegadualreactivity changesduringboronconcentration reductions intheReactorCoolantSystem.Thereaccivity changeraceassociace4 vithboroareduction vill.therefore, b>>vithinthecapability ofoperacorracogxitioa an4coacrol.Therestriccions onstart~myaReaccor.CoolancPumpbclovP-7vithoacormoreRCScoldlegslesstheaorequalcoL5ZcareprovidedtopreventRCSpressuretransiaacs, causedbyeaergadditions fromthcsecondary syscem,vhichcouldexceedchclimitsofhppeadixCto10CPRPar~50.TheRCSvil'>>prateccad againstoverpressure t=ansieat>>

aadvillaocexceedtheLimitsofAppendsCbyeither(1)restricting thavatarvolumeinthepressuriser aadtherebyproviding avolumefortheprimarycoolanttoexpandintoor(X)byrestricting startingoftheRCp'stovheathesecondary vacartemperacure ofeachsteamgenerator isLassthan50FaboveeachoftheRCScoldLegtemperatures.

COKhJCL"=<8, P'v7-4.lTTL3/441~".ZWO.".ES.

WO-PS.f4 BhSES34.5.5REZt~LIZC WATERSTOHACZmaCTheOPERhBILITY oftheESTaspartoftheECCSensuresthatsufficient negativereactivity isinfectedintothecoretocounteract any'ositive increaseinreactivity causedbyRCSsystemcooldovn, andensuresthatasufficient supplyof.,boratedvaterisavailable forinjection bytheECCSintheeventofaLOCA.,Reactorcoolantsystemcooldovncanbecausedbyinadvertent depressuzization, alossofcoolantaccidentorasteamlinerupture.ThelimitsonRUSTminimumvolumeandboronconcentration ensurethat1)sufficient vaterisavailable vt.thincontainment topermitrecirculation coolingflovtothecore,and2)thereactorvf.llremainsubczitical inthecoldcondition following assumptions areconsistent viththeLOCAanalyses.

These4xosF.~AThecontained

~atervolumelimitincludesanallovance forvaternotusablebecauseoftankdischarge linelocationorotherphysicalcharacteristics; Thelimitsoncontained vatez-volume andboronconcentration oftheRUSTalsoensureapHvalueofbeareen7.6and9.5forthesolutionrecirculated vithincontainment afteraLOCA.ThispHbandminimizes theevolution ofiodineandminimizes theeffectofchlorideandcausticstresscorx'osion onmechanical systemsandcomponents.

TheECCSanalysestodetermine FlimitsinSpecifications 3.2.2and3.2.6assumedaRUST~atertemperature of70F.Thistemperatux'e valueoftheRUSTeaterdecermines chatofthesprayvaterinitially delivered tothecontainment.

folloving LOCA.Itisoneofthefactorsvhichdetermines thecontainment back-pressuze intheECCSanalyses, performed inaccordance with.therovisions of10CFR50.46andAendQcKto10CFR50.vaeoeminRTteerareiTecicalecicatn3..5hbensetive'chgedo80toczeethcoseenbeeennitan.eloerRTteerarerultsinlerntaenreseomntaentpraandaferdsovsumetoitebk.veconitpssurestsincascfloMesianceofsearnitithecerebslog.roodcreinINSERTAaLOCAassumingmixingoftheRWST,RCSECCSwater,andothersourcesofwaterthatmayeventually resideinthesump,withallcontrolrodsassumedtobeout.

S~L~ggfjCO3A.&.1.<PRX55URIl Thelfafcaciona ooconcaiasent internalpressureensurathatl)thecontaiamenc struccura ispravanca4 fromexceeding itadesignnagLcivgpressuredifferential withrespecccotheoutsideatmosphere of8paidand2)theconcainmenc peakpressuredoesnocexceedthedesignpressureof12psfgduringLOCAcondi.tions.

,II.+9The/maximum peakpressureresulting fromaLOCAeventiscalculace4 tobe~.-.89-psf.g, vhichincludes0.3psigforinitial'positive containment pressure.

34.6.1.5AIRT~~PERATCRE eLimf.tat:

"..soncont@i"."enc averageaf,rtempezacure ensurethatl.)theconaf..Deni'ssisedoanini.ialmasssufcenclyLovto""eventexceeding thedesi'ress reduin'C~~conditions and2)"he-"biencair=a="era=.

ra='ces.".""excee4=hat=e"peracure allowable forthecont'nuous dutytati-..gspecified forequipmenc andinstrumentatfon Located'<<i=hf.ncents're..t.py.+9.hecontainment pressre"snsient'.ssensitive to--e'n'c'alLv con"a'ned air"ass-""i"..ga'A..;.ecotta'."ed a'ass'ncreases

<<ithecreasing

=a~era=...".e o<<ertempera~are

'a'-=60F<<iLLLimit"hepea,pressureto++psig"h'ch'slessthan;-contain<<sent desippressureof12psig.the."pertemperature limic1:.antesthepeakaccidenttempers"'e sli~hclyd'ingaMCA:however:his limf.tisbased<<itp'=ar'uponequ'pmenc pro"ect'on andanticipated operating conditions.

Bochtheupperandlo<<ertemperacure Limitsareconsistent Michtheparameters sedintheaccfdencanaLyses.

34.6.L.6C"NTAI'.iNHlT VESSELSTRUCTURAL IiTECRITYThisLf.=itacion ensuresthacchescructural integrity ofcheconcainmenc sceelvessel.viLLbemaincained comparable tocheotic'naLdesignstandards.

forthelifeofchefacility.

Structural integrity isrequiredtoensurethat(1)chesteellinerremainsleaktightsnd(2)checoncretesurrounding chesteeLLf.nerremainscapable=providing

.externaLmissileprocection forthesceellinerandr=f.*cionshf,e".-.gintheevencofaLOCh,.hvisualinspection incon]unction vf.thTypeA,Leakagetestsissufficient todemonstrate thfscapabi'ty, (0COOKNUCLEARPLANTUNIT1B3/L6-2Alfie)gENT NO.1"-6 CURRENTPAGESKQQKD-UPTOREFLECTPROPOSEDCHANGESTOTHEDONALDC.COOKNUCLEARPLANTUNIT'NO.2TECHNICAL SPECIFICATIONS 34.1REACTIVITY CONTROLSYSTEMS34.1.1BORATIONCONTROLSHUTDOWNMARGIN-TGREATERTHAN200FLIMITINGCONDITION FOROPERATION 3.1.1.1TheSHUTDOWNMARGINshallbegreaterthanorequalto~Deltak/k.APPLICABILITY:

MODES1,2*,3,and4.ACTION:J(2%yPiththeSHUTDORfMARGINlessChan~Deltak/k,immediately initiateandcontinueborationatgreaterthanorequalto10gpmofasolutioncontaining greaterthanorequalto20,000ppmboronorequivalent untiltherequiredSHUTDOWNMARGINisrestored.

SURVEILLANCE REUIREMENTS 4.1.1.1.1TheSHUTDOWNMARGINshallbedetermined tobegreaterthanorequalto~aDeltak/k:izooa.Vithinonehourafterdetection of.aninoperable controlrod(s)~andatleastonceper12hoursthereafter vhiletherod(s)isinoperable.

Iftheinoperable controlrodisimmovable or~untrippable, theaboverequiredSHUTNMNMARGINshallbeverifiedacceptable vithanincreased allovance forthevithdravn vorthoftheimmovable oruntrippable controlrod(s).b.KeninMODE1orMODE2vithKffgreaterthanorequalto1.0,atleastonceper12hoursbyverifying thatcontrolbankvithdraval isvithinChelimitsofSpecification 3.1.3.6.c~@heninMODE2vithKflessthan1.0,vithin4hourspriortoachieving reactorcri8fcali.ty byverifying thatthepredicted criticalcontrolrodpositionisvithinthelimitsofSpecification 3.1.3.6.PriorCoinitialoperation above5%RATEDTHERMALPOMERaftereachfuelloading,byconsideration ofthefactorsofebelov,viththecontrolbanksatthemaximuminsertion limitofSpecification 3.1.3.6.*SeeSpecialTestException 3,10.1COOKNUCLEARPLANT-UNIT23/41-1maeMENTNO.82,158,13a

REACTIUITY CONTROLSYSTEMSCHARGINGPUMP-SHUTDOWNLXMXTXNGCONDXTXON FOROPERATION 3.1.2.3a.Onechargingpumpintheboroninfection flowpathrequiredbySpecification 3.1.2.1shaDbeOPERABLEandcapableofbeingpoweredfromanOPERABLEemergency bus.b.Onechargingflowpathassociated withsupportofUnit1shutdownfunctions shallbeavailable..+

APPLICABXLXTY:

Specification 3.1.2.3.a.

-MODES5and6Specification 3.1.2.3.b.

-AtalltimeswhenUnit1isinMODES1,2,3,or4.ACTION:a.Wi.thnochargingpumpOPERABLE, suspendalloperations involving COREALTERATIONS orpositivereactivity changes.~

b.'i.thmorethanonechargingpumpOPERABLEorwithasafetyinfection pump(s)OPERABLEwhenthetemperature ofanyRCScoldlegislessthanorequalto0152F,unlessthereactorvesselheadisremoved,removetheadditional chargingpump(s)andthesafetyinjection pump(s)motorcircuitbreakersfromtheelectxical powercixcuitwithinonehour.c.Theprovisions ofSpecification 3.0.3arenotapplicable.

d.Inadditiontotheabove,whenSpecification 3.1.2.3.b isapplicable andtherequiredflowpathisnotavailable, returntherequiredflowpathtoavailable statuswithin7days,orprovideequivalent shutdowncapability'n Unit1andreturntherequiredflowpathtoavailable statuswithinthenext60days,orhaveUnit1inHOTSTANDBY~ithinthenext12hoursandHOTSHUTDOQHwithinthefollowing 24hours.e.Therequirements ofSpecification

3.0. 4arenotappLicable

whenSpecification 3.1.2.3.bapplies.SURVEILLANCE REUXREMENTS 4.1.2.3.1 Theabove-required chargingpumpshaDbedemonstrated OPERABLEbyverifying, thatonrecirculation f.ow,thepumpdevelopsa-44.-sehaege pressureofgreaterthanorequalto4%0-pekgwhentestedpursuanttogpecificacion 4.0.5.gg,fgp5I8*Amaximumofonecentrifugal chargingpumpshallbeOpERABLEwheneverthetemperature

!ofoneormoreoftheRCScoldlegsislessthanorequal.to152F.~<<Forpurposesofthisspecification, additionofwaterfromtheRUSTdoesnot(~ionstituce apositivereactivity additionprovidedtheboronconcentration inthe:t4STisgreaterthantheminimumrequiredbySpecification 3.1.2.T.b.2.

COOKNUCLEARPLANT-UNIT23/41-11AMENDMENT NO.B510T116 ICITIITYCalltlaLSffIEflS~a*pemeS-OPERATI%LIMITSCO~OITIO~

POROPERATIO~

3.1.2.4AtIeast~chargingpumpssha11beOPERABtc.APPLICABIUTY:

MOQESI,2,3and4.ACTION:NthonIyonechargingpumpOPERABLE, restoreatIeasttwochargingpumpstoOPHABLKsituswfthfn72hoursorbefnatIeastHOTSTAHOBYandboratedtoaSHUTDOWNMARQIHequfvaIent toatIeastI~~kat200Fwithfnthenext6hours;restoreatIeastt'~ochargfngpumpstoOPBNBl~situswithinthenex7daysorbefnCOLOSRUTUOMHwithinthenex30hours.SURYEIL>MHCE REOUIREHEHTi 4.1.2.4AtIeas--~ocha.~:rgpumpssha11bedemonstra zdOPBABLE"yverifying, thatonrecfrcuIatfon fIow,eachpvxodeveIopsa'o.>H&psfgwhenwsta6.pursuant4Speciffcatfon 4.0.5.~haJgqfOpsiq REACTIVITY CONTROLSYSTEMSBORATEDMATERSOURCES-SHUTDOWNLIMITINGCONDITION FOROPERATION 3.1.2.7Asaminimum,oneofthefollowing boratedwatersourcesshallbeOPERABLE:

a.Aboricacidstoragesystemandassociated heattractingwith:l.Aminimum.usableboratedwatervoIumeof4300gallons,2.Between20,000and22,500ppmofboron,and3.Aminimumsolutiontemperature of145'F.b.Therefueling waterstoragetankwith:l.Aminimumusableborated~pter, yo.lumeof.90,000gallons,2.Aminimumboronconcentration of2400ppm,and3.Aminimumsolutiontemperature of-SPf=.0~9A>FAPPLICABILITY:

MODES5and6.ACTION:MithnoboratedwatersourceOPERABLE, suspendalloperations involving COREALTFRATIONS orpositivereactivity changes"untilatleastoneboratedwatersourceisrestoredtoOPERABLEstatus.SURVEILLANCE REUIREMENTS 4.1.2.7TheaboverequiredboratedwatersourceshalTbe.demonstrated OPERABLE:

a.Atleastonceper7daysby:2.3.Verifying theboronconcentration ofthewater,Verifying thecontained boratedwatervolume,andVerifying theboricacidstoragetanksolutiontemperature whenitisthesourceofboratedwater.b.Atleastonceper24hoursbyverifying theRWSTtemperature when"itisthesourceofboratedwater.*Forpurposesofthisspecification, additionofwaterfromtheRMSTdoesnotconstitute adilutionactivityprovidedtheboronconcentration in.theRMSTisgreaterthanorequaltotheminimumrequiredbySpecification 3.1.2.7.b.2.

D..C,COOK-UNIT23I41-15Amendment No.8~.94 REACTIVITY CONTROLSYSTENSBORATEDWATERSOURCES-OPERATING LIHITINGCONDITION FOROPERATION 3.1.2.8Eachofthefollowing boratedvatersourcesshallbeOPERABLE:

Aboricacidstoragesyseemandassociaeed heateracingvith:1.Aminimumconcained boraeedvatervolumeof5650gallons,2.Between20,000and22,500ppmofboron,and3,Ami.ni.mum soluci.on temperature of145F.0b.Therefueling waterscoragetankvt.th:Aminimumcontained boratedvatervolumeofwater,350,000gallonsof~2.Between2400and2600ppmofboron,and3.APPLICABILITY:

NODES1,2,3and4.hminimumsoluciontemperature of8(A.<~o'F.ACTION:a~Wieheheboricacidstoragesysteminooerable, restoreehestoragesystemcoOPERABLEseatusvithi,n72hoursorbeinaeleastHOTSTANDBYvithinthenext6hoursandboratedtoaSHUTDOWNNARGINequivalent toaelease1XDeltak/kat200F;restoretheboricacidscoragesystemtoOPERABLEstatusvt.thinthenext7daysorbeinCOLDSHUTDOWNvithinthenext30hours.Wi.ththerefueling vaterseorageeankinoperable, restorethetankeoOPERABL'E statusvt.thinonehourorbeinatleaseHOTSTANDBYvithinthenext6hoursandinCOLDSHUT-DOWNvi.chi.nthefolloving 30hours.SURVEILLANCE REUIREHEVES 4.1.2.8Eachbo'ratedvatersourceshallbedemonstrated OPERABLE:

COOKNUCLEARPLANT-UNIT23/41-16\AMENDMENT NO.N,ZN,lw

EM-RCEN~Y CORECOOL:NCSYSiMS'ESURVETLLANCE REUTR~~NTS(Cortinuedl d.Atleastonceper18monthsby:1.Verifying automati.c i.solat'on andinterlock ac"'onoftheR~systemfromtheReactorCoolantSystemwhentheReactorCoolantSystempressureS.sabove600psig.Avisualinspection ofthecontainment sumpandthesubsystem suctioninletsarenotrest"icted thatthesumpcomponents (tashracks,sceens,evidenceofstructural distressorcorrosion.t v.ri.fying thatbydebrisandetc.)shownoAtleastonceper18months,duringshutdown, by:t1.Verifying thateachautomatic valvei.ntheflowpathactuatestoitscorrectpositiononaSafetyZnjection test~signal2~Verifying thateachofthefollowing pumpsstartauto-matically uponreceiptofasafetyinjection testsignal:a)Centrifugal chargingpumpb)Safetyinjection pumpc)ResidualheatremovalpumpByverifying thateachofthefollowing pumpsdevelopstheindicated

~Kaehmgepressureonrecirculation flowwhentestedpursuanttoSpecification 4.0.5:pg~su8>~Centrifugal chargingpump2.Safetyinjection pump3.Residualheatremovalpumpf0psiaCreaterthanoreaualto~~ah@"f38'SPSidICc'Ps(gCreaterthanorecpaalto~Q-pe4g-giByverS.fying thecorrectpositionofeachmechanical stopforthefollowing Emergency CoreCoolingSystemthrottlevalves:W.thin4hoursfollowing completion ofeachvalvestokingoperation ormaintenance onthevalvewhentheECCSsub-systemsarerequiredtobeOPERABLE.

l'heprovisions ofTechnical Soecification 4.0.8areapplicable.

COOKNUCLEONSPLZ~XT-UN7723/45-5a=-NOV:-NT NO."',~,-'o,159 EMERGENCY CORECOOLINGSYSTEMSREFUELING WATERSTORAGETANKLIMITINGCONDITION FOROPERATION 3.5.5Therefueling waterstoragetank(RWST)shallbeOPERABLEwith:a.Aminimumcontained volumeof350,000gallonsofboratedwater,Ib.Between2400and2600ppmofboron,andc.Aminimumwatertemperature of49-F-.APPLICABILITY:

MQOES1,2,3and4.M')0'FACTION:Withtherefueling waterstoragetankinoperable, restorethetanktoOPERABLEstatuswithin1hourorbeinatleastHOTSTANDBYwithin6hoursandinCOLDSHUTDOWNwithinthefollowing 30hours.SURVEILLANCE REUIREMENTS 4.5.5TheRWSTshallbedemonstrated OPERABLE:

a.Atleastonceper7daysby:l.Verifying thecontained boratedwatervolumeinthetank,and2.Verifying theboronconcentration ofthewater.b.Atleastonceper24hoursbyverifying theRWSTtemperature.

0.C.COOK-UNIT23/45-11Amendment No.3984 34.1REACTIVITY CONTROLSYSTEMS'34-1.1BORATIONCONTROL34.1.1.1and34.1.1.2SHUTDOWNMARGINhsufficient SHUTDOWNMARGINensuresthat1)thareactor'an bemadesubcritical fromalloperating conditions, 2)thex'eactivity transients associated vithpostulated accidentconditions arecontrollable vithinacceptable limits,and3)thereactorvillbemaintained sufficiently subcritical toprecludeinadvertent criticality.

intheshutdovncondition.

l.3%SHUTDOWNMARGINrequirements varythroughout corelifeasafunctionoffueldepletion, RCSboronconcentration, andRCST.Themostxestrictive condition occurs'tEOL,vithTatno)oadoperkfngtemperature, andisassociated vithapostulated steallinebreakaccidentandresulting avuncontrolled RCScooldovn.

Intheanalysiiofthisaccident, aminimumSHUTDOWNMARGINof~Deltak/kisinitially requiredto~control thereactivity transient andautomatic ESFisassumedtobeavailable.

QithTlessthan200F,thereactivity transients resulting froma0aupostulated st5amlinebreakcooldovnaz'eminimalanda1%Deltak/kSHUTDOQNMARGINprovidesadequateprotection forthisevent.~rTheSHUTDOWNMARGINrequirements arebaseduponthelimitingconditions described aboveandaraconsistent vithFSARsafetyanalysisassumptions.

34.1.1.3BORONDILUTIONhminimumR.ovrateofatleast2000CPMprovidesadequatemixing,preventsstratification andensuresthatreactivity changesvillbegradualduringboronconcentration reductions intheReactorCoolantSystem.hflovxataofatleast2000GPMvillcirculate anequivalent ReactorCoolantSystemvolumeof12,612cubicfeetinapproximately 30minutes.Thereactivity changexateassociated vithboronreductions villtherefore bevithinthecapability foroperatorrecognition andcontxol.0COOKNUCLEARPLANT-UNIT2B3/41-1AMENDMENT NO.HZiXSSs$34(

Il EMERGENCY COOLINGSYSTEMSBASES34.5.5REFUELING VhTERSTORAGETANKTheOPERABILITY ofcheRVSTaapareoftheECCSensuresthatsufficient negativereactivity iain]ectedintothecoratocounteract anypositiveincreaseinfeactivity causedbyRCSayacamcooldovn, andensuresthatasuffi.cient supplyofboracedvateriaavailabla forinjection bychaECCSintheeventofaLOCh.Reactorcoolantsystemcooldovncanbecausedby.inadvertent depreaauriration, aLOChorsteamlinerupture.ThelimitsofRVSTminimumvolumeindboronconcentration ensurethat1)sufficient vaterisavailable vithincontainment copermitrecirculation coolingflovtothecore,and2)chereactorvillremainaubcritical inthecoldcondition folloving Theseassumpti.ons

<xone,kvAareconsistent viththeLOChanalyses.

Thecontained vatervolumelimitincludesanallovance forvaternotusablebecauseoftankdischarge linelocation.

orotherphysicalcharacteristi.cs.

Thelimitsoncontained vatervolumeandboronconcentration oftheRVSTalsoensureapHvalu~ofbetveen7.6and9.5forchesolutionrecirculated vithincontainment afceraLOCh,ThispHbandminimizes theevolution ofiodineandmlnlmires cheeffect.ofchlorideandcausticstresscorrosion onmechanical systemsandcomponents.

~~FTheECCSanalysestodetermine FlimiinSpecifications 3.2.2and3.2.6assumedaRVSTvacertemperature of.Thistemperature valueoftheRVSTvaterdetermines thatofchesprayvaterini.cially delivered tocheconcainmenc folloving LOCh.Itisoneofchefactorsvhichdetermines thecontainment back'-pressure incheECCSanalyses, performed inaccordance vith.-."

theprovisions ofl0CFR50.46andhppendhcKto10CFR50.1NSERTAaLOCAassumingmixingoftheRWST,RCSECCSwater,andothersourcesofwaterthatmayeventually resideinthesump,withallcontrolrodsassumedtobeout./ACOOKNUCLEARPLANT-UNIT253/45-3AMENDMENT NO.797,14Z

CONTAINMEHT SYSTEHSBASES3/4.6.1;4 INTERNAL.PRESSURE Thelimitations on,containment internalpressureensurethat1)thecontainment structure isprevented fromexceeding itsdesignnegativepressuredifferential withrespecttotheoutsideatmosphere of8psigand2)thecontainment peakpressuredoesnotexceedthedesignpressureof12psigduringLOCAconditions.

rasul6>>ii.qqThemaximumpeakpressureis4-.4psigfromaLOCAevent0.5ps'~taittwL~aH~covl~A~

pvcssvhc.

~3/4.6.1.5 AIRTEHPERATURE (Qg~t~4&5Thelimitations oncontainment averageairtemperature ensurethat1),thecontainment airmassislimitedtoaninitialmasssufficiently lowtopreventexceeding thedesignpressureduringLOCA'conditions and2)theambientairtemperature doesnotexceedthattemperature allowable forthecontinuous dutyratingspecified forequipment and'nstrumentation located,withincontainment.

I~'Thecontainment presuretransient.

issensitive totheinitially contained airmassdurigaLOCA.Thecontai'ned aiimassincreases withdecreasing temperature.

Thelowertemperature limitof60'Fwilllimitthepeakpressureto.psigwhichislessthanthecontainment designpressureof12psig.Theuppertemperature limitinfluences thepeakaccidenttemperature slightlyduringaLOCA;however,thislimitisbasedprimarily uponequipment protection andanticipated operating conditions.

Boththeupperandlowertemperature limitsareconsistent withthepara-metersusedintheaccidentanalyses.

3/4.6.1.6CONTAINMENT STRUCTURAL INTEGRITY Thislimitation ensuresthatthestructural integrity ofthecon-tainmentwillbemaintained comparable totheoriginaldesignstandards forthelifeofthefacility.

Structural integrity isrequiredtoensurethat(1)thesteellinerremainsleaktightand(2)theconcretesurround-ingthesteellinerremainscapableofproviding externalmissileprotec-tionforthesteellinerandradiation shielding intheeventofaLOCA.Avisualinspection inconjunction withTypeAleakagetestsissufficient todemonstrate thiscapability.

D.C.COOK-UNIT2B3/46-2

ATTACHMENT 4TOAEP:NRC:1207 SUMMARYDESCRIPTION OFPROPOSEDINCREASED STEAMGENERATOR TUBEPLUGGINGTECHNICAL SPECIFICATIONS Attachment 4toAEP:NRC:1207 Page1KeyforSummaryTablePageSectionGroupTechnical Specification PageTechnical Specification RelatedGroupsDiscussed inAttachment 1,Description ofProposedChangesand10CFR50.92Significant HazardsConsideration AnalysisSGTPMarginBothGroup1,ChangesDirectlyRelatedtoIncreased SteamGenerator TubePluggingGroup2,ChangesProposedtoIncreaseunit1Operating MarginGroup3,ChangesProposedtoIncreasetheOperating MarginofBothunits.Description RemarksAdminGroup4,Administrative ChangeABriefDescription ofEachProposedChangeBriefCommentswithaCrossReference totheAnalyses 0

Attachment 4toAEP:NRC:1207 Page2Page2-22-52-72-82<<82-92-9SectionFigure2.1-1Table2.2-1FootnoteTable2.2-1Table2.2-1Table2.2-1Table2.2-1Table2.2-1GroupMarginSGTPMarginMarginMarginMarginMarginDescription ReviseReactorCoreSafetLimitsRedefinedesignflowinfootnoteofTable2.2-1tobe1/4MMF.TheupperlimitonT'ncreased toreflectanalyses.

DecreaseK1from1.32to1.17.Changef(n,I)toincreasetheregionofpositivealwhichiswithoutenaltDecreasetheupperlimitonTtoreflectanalyses.

Changetheallowable valuesinnote2and3.RemarksThenewthermaldesignisdiscussed inSection3.3.2.1ofAttachment 6,WCAP14285.MMFforDNBisdiscussed inSection3.3.2.1ofAttachment 6,NCAP14285.T.S.MMFis1.025timesthermaldesignflow(TDF).TDFisspecified inSection3.3.3.1ofNCAP14285.TheMMFemployedintheDNBanalysisis1.019timesTDF.ThiswasdonetosupportarangeofMMF'sfrom1.019to1.025timesTDFasindicated inSection2.1ofWCAP14285.Designflowincurrenttechnical specification Table2.2-1isMMF/4.TheOTDTtripisdiscussed inSection3.3.2.1ofNCAP14285.Details,including T',oftheanalyzedsetpointareinTable3.3-3ofWCAP14285.Thischangeandthenextchangetof(al)arebeingrequested tooptimizeoperating margin.Someloadrejection capability issacrificed forinstrumentation margin,increased allowance forcoreburndowneffectsonhotlegstreaming, andanincreaseinthepositivedlbreakpointforthef(~I)penalty.TheOTDTtripisdiscussed inSection3.3.2.1ofWCAP14285.DetailsoftheanalyzedsetpointareinTable3'-3ofNCAP14285.Seepreviousdiscussion ofK1decrease.

CookNuclearPlantunit1isoperatedinalowtemperature, lowpressuremodetoextendthelifeofthesteamgenerators.

Therefore, theanalysisoftheOPDTsetpointwasanalyzedwithalowupperlimitonTtoconvertunusedmargintooperating margin.TheOPDTtripisdiscussed inSection3.3'.1ofNCAP14285.Details,including T,oftheanalyzedsetpointareinTable3.3-3ofWCAP14285.Thevaluesindicated inthemarkupsofAttachment 3andintheproposedtechnical specifications ofAttachment 2werecalculated byourorganization.

Attachment 4toAEP:NRC:1207 Page3PageUNIT13/41-1SectionUNZT1Section3.1.1.14;1.1.1.1 GroupBothDescription Reducerequiredshutdownmargin.RemarksThenewvalueissupported byanalyses.

Unit1:Forcoreresponsesteambreak(CRSB),seeSection3.3.5.6ofWCAP14285~Forsteamline massandenergyrelease(SM&E)insidecontainment, seeSection3.5.4.2ofWCAP14285.ForSM&Eoutsidecontainment, seeSection3.3.4.7ofWCAP14285.UNIT23/41-1UNIT13/41-11UNIT2Section3.1.1.14.1.1.1UNIT1Section4.1.2.3.1 BothChangeCCPsurveillances tobeconsistent with10%degradation.

Changepumpsurveillance requirements fromdischarge pressuretodifferential pressure.

Unit2:ForCRSB,seeSectionB.3.11oftheVantage5ReloadTransition SafetyReportforCookNuclearPlantunit2(RTSR).ForSM&Einsidecontainment, seeSection3.5.4.2ofWCAP14285.ForSM&Eoutsidecontainment, seeSection3.3.4.7ofWCAP14285.AcopyofSectionB.3.11hasbeenincludedinAttachment 7tothissubmittal.

Thenewsurveillance criterion issupported byanalyses.

Thesurveillance criteriaarepresented inSection3.10.1.1ofWCAP14285.ThevaluegivenforCCPappliestoboth-unitsUnit1:ForLOCAseeSections3.1.1and3.1.2ofWCAP14285.Forcoreresponsesteambreak(CRSB),seeSection3.3.5.6ofWCAP14285.Forsteamline massandenergyrelease(SM&E)insidecontainment, seeSection3.5.4.2ofWCAP14285.ForSM&Eoutsidecontainment, seeSection3.3.4.7ofWCAP14285.UNIT23/41-11UNIT2Section4.1.2.3.1 Unit2:Thetechnical specification changessupported bytheRTSRaredelineated inTable6.1oftheRTSR.ForLOCAseeSectionsC.3.1.2andC.3.2oftheRTSR.ForCRSB,seeSectionBE3.11oftheRTSR.ForSM&Einsidecontainment, seeSection3'.4.2ofWCAP14285.ForSM&Eoutsidecontainment, seeSection3.3.4.7ofWCAP14285.CopiesofTable6.1andSectionsBE3.11,C.3.1.2,andC.3.2havebeenincludedinAttachment 7tothissubmittal.

Attachment 4toAEP:NRC:1207 Page4PageUNIT13/41-12UNIT23/41-12SectionUNIT1Section4.1.2.4UNIT2Section4.1.2.4GroupBothDescription ChangeCCPsurveillances tobeconsistent with10%degradation.

Changepumpsurveillance requirements fromdischarge pressuretodifferential RemarksSeepreviousdiscussion ofCCPdegradation increaseforpage3/41-11.UNIT13/41-15UNIT23/41-15UNIT13/41-16UNIT23/41-163/42-143/42-14UNIT1Section3.1.2.7UNIT2Section3.1.2.7UNIT1Section3.1.2~8UNIT2Section3.1.2.8Table3.2-1Table3.2-1BothBothSGTPSGTPReducetheMinimumRWSTtemperature to70F.ReducetheMinimumRWSTtemperature to70F.IncreaseDNBtemperature limitReduceMMFlimitThemode5and6minimumRWSTtemperature isconservatively maintained atthesamevalueasthatrequiredformodes1,2,3,and4.Thenewvalueissupported bythecoreresponseLBLOCA.Unit1:SeeSection3.1.1,Table3.1-2,ofWCAP14285'nit2:SeeTableC.3.1-2oftheRTSR.AcopyofthistableisincludedinAttachment 7.Thecalculation ofthenewDNBtemperature limitisdescribed inSection1.2ofWCAP14285undertheheading"DNBParameters, RCSTavgandRCSFlow".Thereadability erroris2.14F.Theresulting DNBtemperature limitis579.30F.Seethediscussion forpage2-5.

Attachment 4toAEP:NRC:1207 Page5Page3/43-173/43-213/43-23a3/43-243/43-263/43-31SectionTable3.3-3Table3.3-3Table3.3-3Table3.3-4Table3.3-4Table4.3-2GroupMarginMarginMarginMarginMarginMarginDescription ChangeESFactuation logictosupport12%AFWPumpdegradation ChangeESFactuation logictosupport12%AFWPumpderadationChangeESFactuation logictosupport12%AFWPumpdegradation ChangeESFactuation logictosupport12\AFWPumpderadationChangeESFactuation logictosupport12%AFWPumpdegradation ChangeESFactuation logictosupport12kAFWPumpdegradation RemarksTherevisedpartofTable3.3-3incorporates thesafeguards logicusedinCookNuclearPlantunit2.Thiswillallowfortheuseof12%auxiliary feedwater headdegradation (AFW)inunitl.Allanalyses, otherthanan"information only"feedlinebreakanalyses, havebeenperformed usingtheflowfromanAFWpumpwith12%headdegradation.

Thesafeguards logicitselfwillbemodifiedviadesignchangepriortoimplementation oftheserevisedT/Spages(i.e.beforeunit1,cycle16).Afterthismodification, theunit2feedlinebreakanalysisusing12%degradedflowwillboundunit1.Theevaluation whichshowsthattheunit2feedwater linebreakwillboundunit1isdiscussed inSection3.3.4.8ofWCAP14285.Seediscussion forpage3/43-17Seediscussion forpage3/43-17Seediscussion forpage3/43-17Seediscussion forpage3/43-17Seediscussion forpage3/43-17

Attachment 4toAEP:NRC:1207 Page6PageSectionGroupDescription Remarks3/43-33Table4.3-2MarginChangeESFactuation logictosupport12kAFNPumpdegradation Seediscussion forpage3/43-173/44-43/44-5Unit13/45-5Unit23/45-5UNIT13/45-5UNIT23/45-5Section3.4.2Section3.4.3Section4.5.2.f.2 4.5.2.f.3 Section4.5.2.f.2 4.5.2.f.3 UNIT1Section4.5.2.f.1 UNIT2Section4.5.2.f.l MarginMarginMarginAdminBothIncreasePressurizer ValveTolerance IncreasePressurizer ValveTolerance ChangeRHR/SIpumpsurveillances tobeconsistent with15%degradation.

ChangeRHR/SZpumpsurveillance requirements fromdischarge pressuretodifferential pressure.

ChangeRHR/SIpumpsurveillance requirements fromdischarge pressuretodifferential pressure.

ChangeCCPsurveillance tobeconsistent with10%degradation.

Changepumpsurveillance requirements fromdischarge pressuretodifferential pressure.

TheNon-LOCAaccidents werereanalyzed orreevaluated basedonapressurizer valvesetpointtolerance of3%.Thisisnotedinsection1.1and3.3.2.3ofNCAP14285.Seediscussion forpage3/44-5.15%degradation oftheSZandRHRpumpsisdiscussed inSections1.1and1.2ofWCAP14285.LBLOCAisdiscussed inSection3.1.1gSBLOCAisdiscussed Section3.1.2;andLOCAmassandenergyrelease(M&E)isdiscussed inSection3.5.2.1ofWCAP14285.Thenewsurveillance criteriaaresupported byanalyses.

Thesurveillance criteriaarepresented inSection3.10.1ofNCAP14285'hisanadministrative change.Thedischarge pressurecriteriainthecurrenttechnical specifications correspond tothesamepumpperformance characteristics astheproposeddifferential pressurecriteria.

Thechangeensuresthatsurveillance criteriausesimilaracceptance criteria.

Thesurveillance acceptance criteriafor10%degradedpumpswereprovidedinthetechnical specification mark-upsoftheRTSR.Acopyofthemark-upofpage3/45-5isincludedinAttachment 7.Refertothediscussion givenforpage3/41-11forinformation concerning the10%CCPdegradation.

Attachment 4toAEP:NRC:1207 Page7PageSectionGroupDescription RemarksUNIT13/45-11UNIT1Section3.5.5BothReduceminimumRWSTtemperature to70oFSeediscussion forpage3/41-16.UNIT23/45-113/47-65-5B2-1(a)B2-4B2-5UNIT2Section3.5.5Sections4.7.1.2.a and4.7.1.2.b Section5.4.2BasesSection2.1.1BasesSection2.2.1BasesSection2.2.1MarginSGTPMarginMarginMarginChangeAFWPumpsurveillance tobeconsistent with12%degradation Reducesystemvolumetoaccountforpluggedsteameneratortubes.ChangeDNBValuesforFuelRemovedetailfromthediscussion oftheOTDTrotection triRemovedetailfromthediscussion oftheOPDTprotection trip.Seediscussion forpage3/43-17.Theproposedsurveillance criteriaisidentical tothecriteriaintheunit2technical specifications.

Thesecriteriacorrespond totheauxiliary feedwater flowsusedinallanalysesforbothunitsexceptthe"information only"unit1feedwater linebreak.Asnotedinthediscussion forpage3/43-17,afterthechangestotheunit1safeguards actuation logic,unit1willbeboundedbtheunit2feedwater linebreak.Avolumerangecorresponding to0%to30%pluggingisspecified.

Seesection1.2ofWCAP14285.ThevaluesforDNBRfortypicalandthimblecellsarebeingrevised.TherevisedvaluesarenotedinSections1.2and3.3.2'ofWCAP14285.ThischangeisrelatedtothenewthermaldesignandthenewOTDTandOPDTprotection tripsetpoints.

Therefore, seealsodiscussions forpages2-2,2-7,2-8,and2-9.Thediscussion ofthepropernormalization ofT'ndP'sbeingremoved.Thisinformation isdocumented inSection3.3.2.1ofWCAP14285andwillbecontrolled administrativel Thediscussion ofthepropernormalization ofTisbeingremoved.Thisinformation isdocumented inSection3.3.2.1ofWCAP14285andwillbecontrolled administratively.

Attachment 4toAEP:NRC:1207 Page8PageUNZT1B3/41-1UNIT2B3/41-1B3/44-1UNIT1B3/45-3UNIT2B3/45-3UNIT1B3/46-2UNIT2B3/46-2SectionUNIT1BasesSection3/4.1.1.1 and3/4.1.1.2 UNIT2BasesSections3/4.1.1.1 and3/4.1.1.2 BasesSection3/4.4.1UNIT1BasesSection3/4.5.5UNIT2BasesSection3/4.5.5UNIT1BasesSections3/4.6.1.4 3/4.6.1.5 UNIT2BasesSections3/4.6.1.4 3/4.6.1.5 GroupBothMarginBothBothDescription ReducerequiredshutdownmarginChangeDNBValuesforFuelReducetheminimumRWSTtemperature to70'F.Changepeakcontainment pressuretoreflectanalysisresultRemarksSeediscussion forpage3/41-1.Change"1.69"to"thesafetyanalysislimit".Clarifyconditions underwhichthereactorwillremainsubcritical.

Specifically, LBLOCAiscalledoutastheinitiating condition andthecontrolrodsareassumedtobeoutinsteadofbeinginsertedexceptforthemostreactiveassembly.

Inaddition, theexplanation thataconservatively highvalueoftheRWSTtemperature isincludedinthetechnical specifications forunit1isbeingremovedbecausetheproposedvalueof704Fisbasedontheanalyses.

Seethediscussion forae3/41-16.Discussion ofmaximumcalculated containment pressureisgiveninSections1.2and3.5.3.4ofWCAP14285.

ATTACHMENT 5TOAEP:NRC:1207 DISCUSSION OPPREVIOUSRELATEDSUBMISSIONS 0

Attachment 5toAEP:NRC:1207 Page1Introduction Attachment 6tothissubmittal isWCAP14285.Itdescribes theanalysesandevaluations performed byWestinghouse ElectricCorporation inordertosupportareducedthermaldesignflowandareducedminimummeasuredflowwhichareexpectedtoresultfromincreased steamgenerator tubepluggingtothelevelof30%intheunit1steamgenerators.

Ztalsodescribes analysesandevaluations performed simultaneously tosupportcertainincreases inoperating marginsuchasincreased setpointtolerance forthepressurizer safetyvalves.Asdiscussed inSection2.0ofWCAP14285,thenewanalysesreplaceanalysesperformed earliertosupporttheoperation ofCookNuclearPlantandtheevaluations described inWCAP14285arebasedonthoseearlieranalyses.

Theearlieranalysesaredescribed inWCAP11902andWCAP11902Supplement 1,references 3and10.Theyarereferredtoasthe"Rerating Program"inWCAP14285.Thepurposesofthisattachment areto:1.indicatethoseaspectsofearlieranalyseswhichhavebeensubmitted forNRCreviewandapproved, 2.indicatethoseportionsoftheseanalyseswhichhavenotpreviously beensubmitted forreview.3.describetheearlieranalyses, 4.providereferences forprevioussubmittals fortheconvenience ofthereviewer, andThissubmittal includessomeproposedtechnical specification changesforbothunits.Therefore, thediscussion ofthisattachment describes submittals forbothunits.Thediscussion ofthisattachment describes theapplications madebyustoimplement thefeaturessupported bytheearlieranalysesandtheapprovals received.

Thisissignificant becauseitwillassistinclarifying available marginandbecausethereareincreased operating marginssupported bytheearlieranalyseswhichwehavenotpreviously implemented.

Znsomecases,wehavenotsubmitted arequestduetothedesiretomaintainthetechnical specifications forthetwoCookunitsasnearlyalikeaspossible.

Thefollowing listssummarizes thestatusofanalysisfeaturesofearlieranalyses:

PrincialFeaturesoftheEarlierAnalsesWhichHavebeenReviewedandAroved2.3.4~5.6.Reducedtemperature andpressureoperation forunit1.Reducedtemperature operation forunit2.10%degradation fortheRHRandHHSIpumpsforbothunits.Increased MSZVresponsetimeforbothunits.BIT0ppmboricacidconcentration forbothunits.ReducedMMFforunit1.

Attachment 5toAEP:NRC:1207 Page2PrincialFeaturesoftheEarlierAnalsesWhichHaveNotBeenSubmitted forReview2.3.4~5.unit1rerateto3413MWt.Theavailable powermarginisallocated inthissubmittal toallowforincreased steamgenerator tubeplugging.

unit2rerateto3588MWt.Additional analyticworkremainstobecompleted.

10%degradation forthecentrifugal chargingpumpsforbothunits.Approvaltoimplement thisfeatureisrequested inthissubmittal.

MinimumRWSTtemperature of704F.Approvaltoimplement thisfeatureisrequested inthissubmittal.

SDMrequirement of1.3%.Approvaltoimplement thisfeatureisrequested inthissubmittal.

PuoseoftheEarlierAnalsesReratinProramTheearlieranalyseswereperformed toaccomplish anumberofgoals.Themosturgentofthesewastopermitoperation ofunit1atreducedprimarytemperature andpressure.

Thebenefitofoperating inareducedprimarytemperature andpressuremodewastoslowthedegradation oftheunit1steamgenerators.

Inaddition, sinceessentially alloftheanalyticbasisoftheCookunitshadtobereviewedorrevised,alltheanalyseswereperformed topositionunit1forsubsequent upratingto3413MWtcorepowerandunit2to3588MWtcorepower.Asofthistime,Wehavenotrequested NRCreviewofupratingeitherunit.Thissubmission proposestousethemarginbetweentheunit1analyzed, upratedpowerandlicensed, ratedthermalpowertoaccommodate theincreased tubeplugging.

Finally,theearlieranalysessupported increased operating marginsinselectedareas.Amongthesewereincreased allowable ECCSpumpdegradation, reduction ofrequiredshutdownmargin(SDM),areduction intheminimumtemperature oftherefueling waterstoragetanks(RWST),removaloftheboroninjection tanks(BIT),andslowerresponsetimesforcertaincomponents andsystems.Thissubmittal requestsapprovalfortheimplementation ofanallowed10%degradation fortheECCScentrifugal chargingpumps,reduction ofrequiredSDM,andareduction, intheminimumtemperature oftheRWST'sforbothunits,whichissupported inpartbytheearlieranalyses.

DescritionandReviewHistoofPriorSubmittals Thefirstoftheearlieranalysesisdescribed inreference 1,WCAP-11908, Containment Integrity AnalysisforCookNuclearPlantunits1and2.Itwassubmitted forNRCreviewbyreference 2.Reference 1presented alongtermcontainment analysiswhichboundedbothunitsatacorepowerof3413MWt,operation atareducedtemperature andpressure, andoperation oftheECCSwithresidualheatremoval(RHR)crossties closed.Reference 2requested approvalforoperation withRHRcrosstiesclosed.Thenextgroupofanalysesisdescribed inreference 3,WCAP-11902, ReducedTemperature andPressureOperation forCookNuclearPlantunit1Licensing Report.Reference 3presented theremainder oftheanalysesandevaluations necessary tosupportoperation ofunit1atreducedtemperature andpressure.

The Attachment 5toAEP:NRC:1207 Page3analysespresented inreference 3wereperformed atacorepowerof3413MWt.However,theevaluations described inreference 3supported operation atacorepowerof3250MWt.Reference 3alsosupported 10%degradation oftheunit1RHRandhighheadsafetyinjection (HHSZ)pumpsandaminimumRWSTtemperature of704F.Reference (3)wassubmitted forNRCreviewbyreference 4.Thelettersofreferences 5,6,7,and8providedsupplementary information tothestaffrelatedtotherequestforapproval(references 2and4)tooperateunit1atreducedtemperature andpressurewith10%degradedRHRandHHSZpumps.Therequesttooperateunit1inthismannerwasapprovedbyreference 9.Reference 10,WCAP11902,Supplement 1,ReratedPowerandRevisedTemperature andPressureOperation forCookNuclearPlantunits1a2Licensing Report,describes thebalanceoftheanalyseswhichwereperformed byWestinghouse ElectricCorporation tosupporttheoperation ofunit1at3413MWt.Znparticular, ananalysisofthesteammassandenergyrelease(SMaE)tocontainment, theassociated containment

analysis, andtheSM&Eoutsidecontainment areincludedinthisreport.Thesetwoanalyseswereperformed toboundbothunitsattheunit2upratedcorepowerof3588MWt.Togetherwithreference 3,theanalysesofreference 10completed theWestinghouse scopeofanalysestosupportanadditional 3secondsfortheresponsetimeofthemainsteamisolation valves(MSIV),0ppmboricacidconcentration intheBIT,and10%degradation oftheCCP'sforunit1.ThetwoSMaEanalyseswereperformed assumingaSDMof1.3%.However,thecoreresponsesteambreakanalysisreportedinreference 3assumedaSDMof1.6%.Reference 11,Vantage5ReloadTransition SafetyReportforCookNuclearPlantunit2(RTSR),togetherwithreference 1,Containment Integrity
Analysis, reference 3,WCAP11902,Reduce'dTemperature andPressureOperation, andreference 10,WCAP11902,Supplement 1,ReratedPowerandRevisedTemperature andPressureOperation, supportreducedtemperature andpressureoperation forunit2atanupratedcorepowerof3588MWt.However,reference 1andtheRHRandHHSZcrosstieclosedLOCAcasesofreference 11onlysupportaunit2corepowerof3413MWt.Theanalysesreportedinreferences 1,10,and11support10%degradation oftheCCP's,HHSIpumps,andRHRpumps,anincreaseof3secondsinMSZVresponsetimeforunit2,0ppmboricacidconcentration intheBZTforunit2,aminimumRWSTtemperature of70~Fforunit2,andaSDMof1.3%forunit2.Theletterofreference 13submitted reference 11,RTSR,andtheportionsofreference 10,WCAP11902,Supplement 1,whichaddressed theSM&Etothecontainment.

Thelettersofreferences 14,15,and16providedsupplementary information tothestaffxelatedtoreference 13.Operation ofunit2atreducedtemperature with10%degradation oftheRHRandHHSZpumpswasapprovedbyreference 17.Somechangestoboththeunit1andunit2technical specifications whichreturnedcertainactivities toadministrative controlwerealsomade.

Attachment 5toAEP:NRC:1207 Page4Thelettersofreferences 18and19proposedtechnical specifications thatimplemented anincreaseof3secondsintheMSIVresponsetimes.Theseproposals weresupported byreference 3,WCAP-11902, reference 10,WCAP-11902, Supplement 1,reference 11,RTSR,andevaluations performed byus.Thelettersinreferences 18and19submitted theportionsofreference 10,WCAP11902,Supplement 1,whichaddressed theSM&Etothecontainment.

Theproposals toincreasetheMSIVresponsetimesby3secondswereapprovedbyreferences 20and21.Theletterofreference 22proposedtoreducetheprimarysystemminimummeasuredflow(MMF)forunit1.Anevaluation, performed byWestinghouse ElectricCorporation, allocated available margininMMFtotheflowreduction.

Theevaluation wasincludedinthesubmittal.

Thisproposalwasapprovedbyreference 23.Theletterofreference 24proposedtoreducetheboronconcentration intheBIT'sofbothunitsto0ppm.Thisproposalwassupported byreference 3,WCAP-11902, reference 10,WCAP-11902,Supplement 1,reference 11,RTSR,andanalysesperformed byus.Reference 10,WCAP11902,Supplement 1,wassubmitted initsentiretyinsupportofthisproposal.

Theproposalwasapprovedbyreference 25.Thelettersofreferences 26and27proposedtorelaxthetolerance ofthemainsteamsafetyvalvesetpoints forbothCookunits.Theproposalwasbasedonnewanalysesandonevaluations performed byWestinghouse ElectricCorporation.

Theevaluations werebasedontheanalysesdescribed inreference 1WCAP-11908, Containment Integrity

Analysis, reference 3,WCAP-11902, reference 10,WCAP-11902, Supplement 1,andreference 11,RTSR.Thedescriptions ofthenewanalysesandevaluations wereincludedasattachments totheseletters.Thisproposalwasapprovedbyreference 28.References 2.3.4~5.6.WCAP-11908, Containment Integrity AnalysisforCookNuclearPlantunits1and2,M.E.Wills,July1988.LetterAEP:NRC:1024D, Containment LongTermPressureAnalysistoSupportRHRCrossTieClosure,fromM.P.AlexichtoT.E.Murley,August22,1988.WCAP-11902, ReducedTemperature andPressureOperation forCookNuclearPlantunit1Licensing Report,D.L.CecchettandD.B.Augustine, October1988.LetterAEP:NRC:1067, ReducedTemperature andPressureProgramAnalysesandtechnical specification Changes,fromM.P.AlexichtoT.E.Murley,October14,1988.LetterAEP:NRC:1067A, Supplemental technical specification ChangesforReducedTemperature andPressureProgram,fromM.,P.AlexichtoT.E.Murley,December30,1988.LetterAEP:NRC:1067B, Additional Information onReducedTemperature andPressureSubmittal:

BoronDilutionAccident, fromM.P.AlexichtoT.E.Murley,February6,1989.

Attachment 5toAEP:NRC:1207 Page57.8.9.10.12.13.14.15.16~17.18.19.20.21.22.23.24LetterAEP:NRC:1067C, unit1RTPProgram:Additional Information onContainment Structural

Analysis, fromM.P.AlexichtoT.E.Murley,March14,1989.LetterAEP:NRC:1067D, Modification ofReducedTemperature andPressureProgramtechnical specification Changes,fromM.P.AlexichtoT.E.Murley,June5,1989.Amendment No.126toFacilityOperating LicenseNo.DPR-58.WCAP11902,Supplement 1,ReratedPowerandRevisedTemperature andPressureOperation forCookNuclearPlantunits162Licensing Report,September 1989.Vantage5ReloadTransition SafetyReportforCookNuclearPlantunit2,B.W.Gergos,Editor,January1990.Noreference 12.LetterAEP:NRC:1071E,unit2Cycle8ReloadLicensing, Proposedtechnical specifications forunit2Cycle8,andRelatedunit1Proposals, fromM.P.AlexichtoT.E.Murley,February6,1990.LetterAEP:NRC:1071H, Modification toOurPreviousSubmittal AEP:NRC:1071E; RevisedFiguresfortheLossofLoadEvent,fromM.P.AlexichtoT.E.Murley,April6,1990.LetterAEP:NRC:1071I, Information toSupplement OurPreviousSubmittals AEP:NRC:1071E and1071H,fromM.P.AlexichtoT.E.Murley,May29,1990.LetterAEP:NRC:1071K, OffsiteDoseCalculation fortheReactorCoolantPumpLockedRotorEventforunit2Cycle8,fromM.P.AlexichtoT.E.Murley,July23,1990.Amendment No.148toFacilityOperating LicenseNo.DPR-58andAmendment No.134toFacilityOperating LicenseNo.DPR-74.LetterAEP:NRC:1120, Expedited technical specification ChangeRequestSteamGenerator StopValves,fromM.P.AlexichtoT.E.Murley,January31,1990.LetterAEP:NRC:1123, technical specification ChangeRequest,SteamGenerator StopValves,fromM.P.AlexichtoT.E.Murley,May14,1990.Amendment No.147toFacilityOperating LicenseNo.DPR-58.Amendment No.135toFacilityOperating LicenseNo.DPR-74.LetterAEP:NRC:1130, technical specification Changeforunit1Cycle11,fromM.P.AlexichtoT.E.Murley,July23,1990.Amendment No.152toFacilityOperating LicenseNo.DPR-58.LetterAEP:NRC:1140, technical specification ChangeRequest,BITBoronConcentration Reduction, fromM.P.AlexichtoT.E.Murley,March26,1991.

t Attachment 5toAEP:NRC:1207 Page625.Amendment No.158toFacilityOperating LicenseNo.DPR-58andAmendment No.142toFacilityOperating LicenseNo.DPR-74.26LetterAEP:NRC:1169, technical specifications ChangetoIncreasetheAllowable Tolerance forMainSteamSafetyValves,fromE.E.Fitzpatrick toT.E.Murley,Novemberll,1992.27.LetterAEP:NRC:1169A, Updatefortechnical specification ChangetoIncreasetheAllowable Tolerances forMainSteamSafetyValves,fromE.E.Fitzpatrick toT.E.Murley,December17,1993.28.Amendment No.182toFacilityOperating LicenseNo.DPR-58andAmendment No.167toFacilityOperating LicenseNo.DPR-74.

1ItI ATTACHMENT 6TOAEP:NRC:1207 DESCRIPTION OFANALYSESPERFORMED BYWESTINGHOUSE ELECTRICCORPORATION FORCOOKNUCLEARPLANTUNIT1

Attachment 6toAEP:NRC:1207 Page1WCAP14285

ATTACHMENT 7TOAEP:NRC:1207 DESCRIPTION OFANALYSESPERFORMED BYWESTINGHOUSE ELECTRICCORPORATION FORCOOKNUCLEARPLANTUNIT2

ATTACHMENT 6TOAEP:NRC:1207 DESCRIPTION OFANALYSESPERFORMED BYWESTINGHOUSE ELECTRICCORPORATION FORDONALDC.COOKNUCLEARPLANTUNIT1 Attachment 6toAEP:NRC:1207 Page1WCAP14285 WESTINGHOUSE NONPROPRIETARY CLASS3WCAP-14285, Revision1DONALDC.COOKNUCLEARPLANTUNIT1STEAMGENERATOR TUBEPLUGGINGPROGRAMLICENSING REPORTMAY1995WESTINGHOUSE ELECTRICCORPORATION EnergySystemsBusinessUnitP.O.Box355Pittsburgh, Pennsylvania 152301995,Westinghouse ElectricCorporation, AllRightsReservedmA1944-1w.wpf:1d%51895 TABLEOFCONTENTSTitle~PaeListofTablesListofFiguresListofAcronymsandAbbreviations Definitions VllXXIIISummaryandConclusions XXIV

1.0INTRODUCTION

-DESCRIPTION OFLICENSEAMENDMENT REQUEST1.1PurposeforChange1.2CurrentLicenseBasisandFunctionofIdentified Technical Specification andDescription ofProposedChange1.1-11.1-11.2-12.0BASISFOREVALUATIONS/ANALYSES PERFORMED 2.1DesignPowerCapability Parameters 2.2NSSSDesignTransients 2.3ControVProtection SystemSetpoints 2.0-12.1-12.2-12.3-13.0SAFETYEVALUATIONS/ANALYSES PERFORMED 3.1LossofCoolantAccidentAnalyses3.2LOCAHydraulic Forces3.3Non-LOCAAnalyses3.4Post-LOCA HydrogenProduction 3.5Containment Analyses3.6SteamGenerator TubeRuptureAccidentAnalysis3.7Post-LOCA HotLegRecirculation Time3.8ReactorCavityPressureAnalysis3.9Radiological Analysis3.10FluidandAuxiliary SystemsEvaluations 3.11PrimalyComponents Evaluations 3.11.1SteamGenerators 3.11.2ReactorVessel3.11.3ReactorInternals 3.11.4ControlRodDriveMechanisms 3.11.5ReactorCoolantPumps3.1-13.1-13.2-13.3-13.4-13.5-13.6-13.7-13.8-13.9-13.10-13.11-13.11-13.11-43.11-83.11-113.11-12m51944-1w.wpf:1d~1295 TABLEOFCONTENTS(continued)

SectionTitlePacae3.11.6Pressurizer 3.11.7ReactorCoolantLoopPipingandSupports3.11.8Auxiliary Components 3.12FuelStructural Evaluation 3.11-133.11-153.11-163.12-

14.0CONCLUSION

S APPENDIXAProposedTechnical Specification Changes4-1m61944-1w.tNpf:1d~1295

LISTOFTABLESTitle1.2-12.1-13.1-13.1-23.1-33.1-43.1-53.1-6Summa'fTechnical Specification ChangesNSSSPerformance Parameters forSGTPProgramLargeBreakLOCAResultsPlantInputParameters UsedinLargeBreakLOCAAnalysisLargeBreakContainment Data(IceCondenser Containment)

MassandEnergyReleaseRates,MinimumSlNitrogenMassandEnergyReleaseRatesSafetyInjection FlowRate:ReratingProgramAnalysis3.1-7PlantInputParameters UsedinSmallBreakLOCAAnalyses:

ReratingProgramAnalysis3.1-83.1-93.1-103.1-11SmallBreakLOCACalculation:

ReratingProgramAnalysisTimeSequenceofEventsforCondition IIIEvents:ReratingProgramAnalysisSmallBreakLOCACalculation:

ReratingProgramAnalysisTimeSequenceofEventsforCondition IIIEvents:ReratingProgramAnalysis3.1-12PlantInputParameters UsedinSmallBreakLOCAAnalysis:

+/-3%MainSteamSafetyValveSetpointTolerance Analysis3.1-13TimeSequenceofEventsforCondition IIIEvents:+/-3%MainSteamSafetyValveSetpointTolerance Analysis3.1-14SmallBreakLOCACalculations:

+/-3%MainSteamSafetyValveSetpointTolerance Analysis3.1-15PlantInputParameters UsedinSmallBreak.LOCAAnalysis:

30%SGTPProgramAnalysiswithHHSICross-Ties Closedmf1944-1w.wpf:1d~1195 f!0 LISTOFTABLES(continued)

Title3.1-16TimeSequenceofEventsforCondition IIIEvents:30%SGTPProgramAnalysiswithHHSICross-Ties Closed3.1-17SmallBreakLOCACalculations:

30%SGTPProgramAnalysiswithHHSICross-Ties Closed3.3-13.3-2NSSSPerformance Parameters UsedinNon-LOCASafetyAnalysesTripPointsandTimeDelaystoTripAssumedinNon-LOCAAccidentAnalysis3.3-3OTbTandOPbTSetpointEquationandSafetyAnalysisLimitCoefficient Values3.3-43.3-53.3-6SummaryofInitialConditions andComputerCodesUsedSequenceofEventsforLossofFlowandLockedRotorAccidents SequenceofEventsforLossofExternalElectrical Load3.3-7LimitingSteamline BreakStatepoint DoubleEndedRuptureInsideContainment withOffsitePowerAvailable 3.3-8TimeSequenceofEvents-DoubleEndedRuptureInsideContainment withOffsitePowerAvailable 3.3-9Parameters UsedintheAnalysisoftheRodClusterControlAssemblyEjectionAccident3.5-13.5-2SystemParameters, InitialConditions SafetyInjection Flow,MinimumSlm:51944-1w.wpf:1d441195 IV LISTOFTABLES(continued)

Title3.5-3Double-Ended PumpSuctionGuillotine MinimumSlBlowdownMassandEnergyRelease3.5-4Double-Ended PumpSuctionGuillotine MinimumSlRefloodMassandEnergyRelease3.5-5Double-Ended PumpSuctionGuillotine MinimumSlPrincipal Parameters DuringReflood3.5-6Double-Ended PumpSuctionGuillotine MinimumSlPostRefloodMassandEnergyRelease3.5-73.5-8DoubleEndedPumpSuctionGuillotine MinimumSlMassBalanceDoubleEndedPumpSuctionGuillotine MinimumSIEnergyBalance3.5-93.5-103.5-113.5-12EnergyAccounting inMillionsofBTUEnergyAccounting inMillionsofBTUStructural HeatSinkTableMaterialProperties Table3.5-13-3.11-1Steamline BreakMass/Energy ReleasesInsideContainment Performance Characteristics at3262MWt3.11-2AssumedOperating Parameters forReactorVesselStructural Evaluation forCookNuclearPlantUnit13.11-33.12-13.12-2Pressurizer Components Calculated FatigueUsagesConsidering 30%SGTPMaximumLOCAandDBEGridLoadResultsFuelRodDesignAnalysisParameters m%1944-1w.wpf:1d 441195 LISTOFTABLES(continued)

Title3.12-33.12-430%SGTPProgramThermalHydraulic DesignParameters DNBRLimitsandMarginSummarym:$1944-1w.wpf:1d~1195 Vl

LISTOFFIGURES~FiereTitle3.1-1a-f3.1-2a-f3.1-3a-f3.1-4a-f3.1-5a-f3.1-6a-f3.1-7a-f3.1-8a-f3.1-9a-fReactorCoolantSystemPressure.

CasesA-FBreakFlowDuringBlowdown, CasesA-FCorePressureDrop,CasesA-FCoreFlowrate, CasesA-FAccumulator FlowDuringBlowdown, CasesA-FVesselLiquidLevelsDuringReflood,CasesA-FCoreInletFlowDuringRefloodAccumulator andSlFlowDuringReflood,CasesA-FIntegralofCoreInletFlow,CasesA-F3.1-10a-f MassFluxatPeakTemperature Elevation, CasesA-F3.1-11a-f RodH.T.C.atPeakTemperature Elevation, CasesA-F3.1-12a-f VaporTemperature, CasesA-F3.1-13a-f FuelRodPeakCladTemperature, CasesA-F3.1-14Containment

Pressure, CD=0.4,Min.Sl3.1-153.1-163.1-173.1-183.1-19UpperCompartment Structural HeatRemovalRate,CD=0.4,Min.SlLowerCompartment Structural HeatRemovalRate,CD&.4,MinSlHeatRemovalbySump,CD=0.4,Min.SlHeatRemovalbyLowerCompartment Spray,CD=0.4,Min.SlContainment Temperature, CD=0.4,Min.Slm%1944-1w.wpf:1d441195 VII LISTOFFIGURES(continued)

~Fiere3.1-203.1-21TitleSafetyInjection FlowRateDonaldC.CookUnit1HotRodPowerDistribution DonaldC.CookUnit13.1-22RCSPressure(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-23CoreMixtureHeight(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-24HotSpotCladTemperature (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-25CoreSteamFlowrate(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-26HotSpotHeatTransferCoefficient (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-27HotSpotFluidTemperature (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-28TotalBreakFlow(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-29IntactLoopPumpedSlFlow(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-30RCSPressure(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-31CoreMixtureHeight(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-32HotSpotCladTemperature (2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-33CoreSteamFlowrate(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mal944-1w.wpl:1d 441195Vill

LISTOFFIGURES(continued)

Ficiure3.1-34TitleHotSpotHeatTransferCoefficient (2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-35HotSpotFLuidTemperature (2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-36TotalBreakFlow(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-37IntactLoopPumpedSlFlow(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-38RCSPressure(rInch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-39CoreMixtureHeight(4Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-40HotSpotCladTemperature (4Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-41CoreSteamFlowrate(4Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-42HotSpotHeatTransferCoefficient (4Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-43HotSpotFluidTemperature (4Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-44RCSPressure(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-45CoreMixtureHeight(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-46HotSpotCladTemperature (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m&1944-1w.wpf:1d~l 195IX LISTOFFIGURES(continued)

~FiureTitle3.1-47CoreSteamFlowRate(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-48HotSpotHeatTransferCoefficient (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-49HotSpotFluidTemperature (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-50TotalBreakFlow(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-51IntactLoopPumpedSlFlow(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.1-52RCSPressure(3Inch)HighTemperature, HighPressureDonaldC.CookUnit13.1-53CoreMixtureHeight(3Inch)HighTemperature, HighPressureDonaldC.CookUnit13.1-54HotSpotCladTemperature (3Inch)HighTemperature, HighPressureDonaldC.CookUnit13.1-55CoreSteamFlowrate(3Inch)HighTemperature, HighPressureDonaldC.CookUnit13.1-56HotSpotHeatTransferCoefficient (3Inch)HighTemperature, HighPressureDonaldC.CookUnit13.1-57HotSpotFluidTemperature (3Inch)HighTemperature, HighPressureDonaldC.CookUnit13.1-58TotalBreakFlow(3Inch)HighTemperature, HighPressureDonaldC.CookUnit13.1-59IntactLoopPumpedSlFlow(3Inch)High.Temperature, HighPressureDonaldC.CookUnit1m:i'll944.1w.wpf:1d441195 LISTOFFIGURES(continued)

~Fiere3.1-86TitleHotRodPowerDistribution (3Inch,30%SGTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit13.3-1Illustration ofOvertemperature andOverpower hTProtection, NominalTavg=576.3'F,NominalPressure=2100psia3.3-2Illustration ofOvertemperature andOverpower ATProtection, NominalTavg=576.3'F,NominalPressure2250psia3.3-3Illustration ofOvertemperature andOverpower dTProtection, NominalTavg=553.0'F,NominalPressure=2250psia3.3-43.3-53.3-6Illustration ofOvertemperature andOverpower bTProtection, NominalTavg=553.0'F,NominalPressure=2100psiaINuclearPowerandHotChannelHeatFluxvs.TimefortheRodWithdrawal FromSubcritical EventFuelAverageandCladTemperature vs.TimefortheRodWithdrawal fromSubcritical Event3.3-7NuclearPowervs.TimefortheRCCAWithdrawal atPowerEvent,FullPower,80PCM/secInsertion Rate,MaximumReactivity Feedback3.3-8Pressurizer PressureandPressurizer WaterVolumevs.TimefortheRCCAWithdrawal atPowerEvent,FullPower,80PCM/secInsertion Rate,MaximumReactivity Feedback3.3-9CoreAverageTemperature andDNBRvs.TimefortheRCCAWithdrawal atPowerEvent,FullPower,80PCM/secInsertion Rate,MaximumReactivity Feedback3.3-10NuclearPowervs.TimefortheRCCAWithdrawal atPowerEvent,FullPower,4PCM/secInsertion Rate,MaximumReactivity Feedbackmal944-1w.wpf:1d~1195 XIII LISTOFFIGURES(continued)

~FirrreTitle3.3-11Pressurizer PressureandPressurizer WaterVolumevs.TimefortheRCCAWithdrawal atPowerEvent,FullPower,4PCM/secInsertion Rate,MaximumReactivity Feedback3.3-12CoreAverageTemperature andDNBRvs.TimefortheRCCAWithdrawal atPowerEvent,FullPower,4PCM/secInsertion.

Rate,MaximumReactivity Feedback3.3-13MinimumDNBRvsReactivity Insertion RatefortheRCCAWithdrawal atPowerEvent,100%Power3.3-14MinimumDNBRvs.Reactivity Insertion RatefortheRCCAWithdrawal atPowerEvent,60%Power3.3-15MinimumDNBRvs.Reactivity Insertion RatefortheRCCAWithdrawal atPowerEvent,10%Power3.3-163.3-17NuclearPowerandCoreHeatFluxvs.TimeforaTypicalResponsetoaDroppedRCCA(s)inAutomatic ControlAverageCoolantTemperature andPressurizer Pressurevs.TimeforaTypicalResponsetoaDroppedRCCA(s)inAutomatic Control3.3-18TotalCoreFlowvs.TimefortheCompleteLossofFlowEvent3.3-19NuclearPowerandPressurizer Pressurevs.TimefortheCompleteLossofFlowEvent3.3-20AverageandHotChannelHeatFluxesandDNBRvs.TimefortheCompleteLossofFlowEvent3.3-21TotalCoreFlowandFaultedLoopFlowvs.TimeforthePartialLossofFlowEventm."I1944-1w.wpf:1d~1195 XIV LISTOFFIGURES(continued)

~Fiere3.3-22TitleNuclearPowerandPressurizer Pressurevs.TimeforthePartialLossofFlowEvent3.3-23AverageandHotChannelHeatFluxesandDNBRvs.TimeforthePartialLossofFlowEvent3.3-243.3-253.3-26TotalCoreFlowandFaultedLoopFlowvs.TimefortheLockedRotorEventNuclearPowerandRCSPressurevs.TimefortheLockedRotorEventAverageandHotChannelHeatFluxesvs.TimeandCladInnerTemperature vs.TimefortheLockedRotorEvent3.3-27NuclearPowerandDNBRvs.TimeforLossofLoad,MinimumReactivity FeedbackwithPressurizer SprayandPORVs3.3-28Pressurizer PressureandPressurizer WaterVolumevs.TimeforLossofLoad,MinimumReactivity FeedbackwithPressurizer SprayandPORVs3.3-29CoreAverageandLoop1Temperature vs.TimeforLossofLoad,MinimumReactivity FeedbackwithPressurizer SprayandPORVs3.3-30TotalReactivity andPressurizer SteamReliefvs.TimeforLossofLoad,MinimumReactivity FeedbackwithPressurizer SprayandPORVs3.3-31SteamGenerator MassandSafetyValveReliefvs.TimeforLossofLoad,MinimumReactivity FeedbackwithPressurizer SprayandPORVs3.3-32NuclearPowerandDNBRvs.TimeforLossofLoad,MaximumReactivity FeedbackwithPressurizer SprayandPORVs3.3-33Pressurizer PressureandPressurizer WaterVolumevs.TimeforLossofLoad,MaximumReactivity FeedbackwithPressurizer SprayandPORVsm:11944-1w.wpt:1d441195 LISTOFFIGURES(continued)

Ficiure3.3-34TitleCoreAverageandLoop1Temperatures vs.TimeforLossofLoad,MaximumReactivity FeedbackwithPressurizer SprayandPORVs3.3-35TotalReactivity andPressurizer SteamReliefvs.TimeforLossofLoad,MaximumReactivity FeedbackwithPressurizer SprayandPORVs3.3-36SteamGenerator MassandSafetyValveReliefvs.TimeforLossofLoad,MaximumReactivity FeedbackwithPressurizer SprayandPORVs3.3-37NuclearPowerandDNBRvs.TimeforLossofLoad,MinimumReactivity FeedbackwithoutPressurizer SprayandPORVs3.3-38Pressurizer PressureandPressurizer WaterVolumevs.TimeforLossofLoad,MinimumReactivity FeedbackwithoutPressurizer SprayandPORVs3.3-3t."CoreAverageandLoop1Temperatures vs.TimeforLossofLoad,MinimumReactivity FeedbackwithoutPressurizer SprayandPORVs3.3-40TotalReactivity andPressurizer SteamReliefvs.TimeforLossofLoad,MinimumReactivity FeedbackwithoutPressurizer SprayandPORVs3.3-41SteamGenerator MassandSafetyValveReliefvs.TimeforLossofLoad,MinimumReactivity FeedbackwithoutPressurizer SprayandPORVs3.3-42NuclearPowerandDNBRvs.TimeforLossofLoad,MaximumReactivity FeedbackwithoutPressurizer SprayandPORVs3.3-43Pressurizer PressureandPressurizer WaterVolumevs.TimeforLossofLoad,MaximumReactivity FeedbackwithoutPressurizer SprayandPORVs3.3-44CoreAverageandLoop1Temperatures vs.TimeforLossofLoad,MaximumReactivity FeedbackwithoutPressurizer SprayandPORVsm31944-1w.wpf:1d~1195 xvl

LISTOFFIGURES(continued)

~Fiure3.3-45TitleTotalReactivity andPressurizer SteamReliefvs.TimeforLossofLoad,MaximumReactivity FeedbackwithoutPressurizer SprayandPORVs3.3-46SteamGenerator MassandSafetyValveReliefvs.TimeforLossofLoad,MaximumReactivity FeedbackwithoutPressurizer SprayandPORVs3.3-47Variation ofReactivity withCoreTemperature at1050psiafortheEndofLifeRoddedCorewithOneControlRodAssemblyStuck(ZeroPower)fortheSteamline BreakDoubleEndedRuptureEvent3.3-48DopplerPowerFeedbackfortheSteamline BreakDoubleEndedRuptureEvent3.3-49SafetyInjection FlowSuppliedbyOneChargingPumpfortheSteamline BreakDoubleEndedRuptureEvent3.3-50NuclearPowerandCoreHeatFluxvs.TimefortheSteamline BreakDoubleEndedRuptureEvent(InsideContainment withPower)3.3-51CoreAverageTemperature andRCSPressurevs.TimefortheSteam!inc BreakDoubleEndedRuptureEvent(InsideContainment withPower)3.3-52Pressurizer WaterVolumevs.TimefortheSteamline BreakDoubleEndedRuptureEvent(InsideContainment withPower)3.3-53Reactivity andCoreBoronConcentration vs.TimefortheSteamline BreakDoubleEndedRuptureEvent(InsideContainment WithPower)3.3-54NuclearPowervs.TimefortheRodEjectionEvent,HotZeroPower,EndofLife3.3-55FuelCenterline, FuelAverageandCladOuterSurfaceTemperature vs.TimefortheRodEjectionEvent,HotZeroPower,EndofLife3.3-56NuclearPowervs.TimefortheRodEjectionEvent,HotFullPower,EndofLifem&1944-1w.wpf:1d441195XVII LISTOFFIGURES(continued)

~Fiure3.3-57TitleFuelCenterline, FuelAverage,andCladOuterSurfaceTemperature vs.TimefortheRodEjectionEvent,HotFullPower,EndofLife,3.5-1LOCAMassandEnergyReleaseContainment Integrity, Containment PressureTransient 3.5-2LOCAMassandEnergyReleaseContainment Integrity, UpperCompartment Temperature Transient 3.5-3LOCAMassandEnergyReleaseContainment Integrity, LowerCompartment Temperature Transient 3.5-4LOCAMassandEnergyReleaseContainment Integrity, ActiveandInactiveSumpTemperature Transient 3.5-5LOCAMassandEnergyReleaseContainment Integrity, IceMeltTransient 3.5-61.4ft'ouble-Ended Rupture,102%Power,MSIVFailure,UpperCompartment Temperature 3.5-71.4ft'ouble-Ended Rupture,102%Power,MSIVFailure,LowerCompartment Temperature 3.5-81.4ft'ouble-Ended Rupture,102%Power,MSIVFailure,UpperCompartment Pressure3.5-91.4ft'ouble-Ended Rupture,102%Power,MSIVFailure,LowerCompartment Pressure3.5-100.942ft'plitBreak,30%Power,MSIVFailure,UpperCompartment Temperature m31944.1w.wpf:1dM1195 XVIII LISTOFFIGURES(continued)

~FitjreTitle3.5-110.942ft'plitBreak,30%Power,MSIVFailure,LowerCompartment Temperature 3.5-123.5-130.942ft'plitBreak,30%Power,MSIVFailure,UpperCompartment Pressure0.942ft'plitBreak,30%Power,MSIVFailure,LowerCompartment Pressurem:$1944.1w.wpf:1d441195 XIX LISTOFACRONYMSANDABBREVIATIONS AFWPRANSASMEAPCBITBOPCCWSCHG/SlCOLRCRDMCSCVCSDBEDECLDEHLDEPSDFDNBDNBREABECCECCSECTEDGEFPMEOPESFESFASESWFd,HF~FHAFSARGPMHELBHFPHZPIFBAIFMITDPLBAuxiliary Feedwater PumpRunoutAmericanNuclearSocietyAmericanSocietyofMechanical Engineers Alternate PluggingCnteriaBoronInjection TankBalanceofPlantComponent CoolingWaterSystemCharging/Safety Injection CoreOperating LimitsReportControlRodDriveMechanism Condensate SystemChemicalandVolumeControlSystemDesignBasisEarthquake Double-Ended ColdLegDouble-Ended HotLegDouble-Ended PumpSuctionDecontamination FactorDeparture fromNucleateBoilingDeparture fromNucleateBoilingRatioExclusion AreaBoundaryEmergency CoreCoolingEmergency CoreCoolingSystemEddyCurrentTestingEmergency DieselGenerator Effective FullPowerMonthsEmergency Operating Procedure Engineered SafetyFeaturesEngineered SafetyFeatureActuation SystemEssential ServiceWaterHotChannelEnthalpyRiseFactorTotalPeakingFactorFuelHandlingAccidentFinalSafetyAnalysisReportGallonsperMinuteHighEnergyLineBreakHotFullPowerHotZeroPowerIntegralFuelBurnableAbsorbers Intermediate FlowMixingImprovedThermalDesignProcedure LargeBreakm%1944-1w.wpf:1d441295 LISTOFACRONYMSANDABBREVIATIONS (continued)

LCOLOCALO~LOOPLPZM/EorM8EMMFMSLBMwtNRCNSSSOPBTOTATPCTPLOFPORVPTSPSSMPWRRCRCCARCLRCPRCPBRCSRHRRHRSRPSRSERSRRTDPRTPRTSRWSTRWFSSALSDMSERSBSFPCSSILimitingCondition forOperation LossofCoolantAccidentLossofLoad/Turbine TripLossofAllACPowertotheStationAuxiliaries LowPopulation ZoneMassandEnergyMinimumMeasuredFlowMainSteamLineBreakMegawattThermalNuclearRegulatory Commission NuclearSteamSupplySystemOverpower DeltaTOvertemperature DeltaTPeakCladTemperature PartialLossofReactorCoolantFlowPowerOperatedReliefValvePressurized ThermalShockPowerShapeSensitivity ModelPressurized WaterReactorReactorCoolantRodClusterControlAssemblyReactorCoolantLoopReactorCoolantPumpReactorCoolantPressureBoundangReactorCoolantSystemResidualHeatRemovalResidualHeatRemovalSystemReactorProtection SystemReloadSafetyEvaluation RelativeStability RatioRevisedThermalDesignProcedure RatedThermalPowerReactorTripSystemRefueling WaterStorageTankRCCABankWithdrawal fromaSubcritical Condition SafetyAnalysisLimitShutdownMarginSafetyEvaluation ReportSmallBreakSpentFuelPoolCoolingSystemSafetyInjection m%1944-1w.wpf:

1d441295XXI sosSGSGTPSGTRSLBSLB-CRSRTATAUGTHQTTCOLOTDFLISTOFACRONYMSANDABBREVIATIONS (continued)

SafetyInjection SystemSteam'enerator SteamGenerator TubePluggingSteamGenerator TubeRuptureSteamLineBreakSteamLineBreakCoreResponseSurveillance Requirement TotalAllowance RCSAverageTemperature VesselOutletTemperature VesselInletTemperature ThermalDesignFlowm.11944-1w.wpf:1d~1 295XXII DEFINITIONS ReratingProgram:WCAP-11902 documented theanalysesandevaluations performed tosupportreducedtemperature andpressureoperation ofDonaldC.CookNuclearPlantUnits1and2.Subsequently, asupplement toWCAP-11902 wasissuedtosummarize theadditional effortsperformed tosupportareratingofCookNuclearPlantUnit1andtoprovidepartofthesupportforaUnit2rerating.

Theseanalysesandevaluations aredescribed inSections2.0andaredocumented inReferences 1and2ofSection2.0.Throughout thisreport,theanalysesandevaluations documented inWCAP-11902 andSupplement arereferredtoastheReratingProgram.SteamGenerator TubePluggingProgram:Analysesandevaluations tosupportoperation ofCookNuclearPlantUnit1withuptoalevelof30%steamgenerator tubeplugging.

Inadditiontotheincreased tubeplugginglevel(andcorresponding reducedthermaldesignflow),severalincreased operating marginswerealsoaddressed intheSteamGenerator TubePluggingProgram.Theseoperating marginsaredescribed inSection1.0.mA1944-1w.wpt:1d441295 XXIII

SUMMARYANDCONCLUSIONS PROGRAMSUMMARYThepurposeofthisdocumentistoprovidethesafetyanalysisandevaluation resultstosupportoperation ofDonaldC.CookNuclearPlantUnit1withuptoalevelof30%steamgenerator tubeplugging(SGTP).Inadditiontotheincreased levelofsteamgenerator tubeplugging, theanalysesandevaluations supportacorresponding reduction inThermalDesignFlow(TDF)anda5%loopflowasymmetry.

Theanalysesandevaluations wereperformed overarangeofprimarytemperatures (553'Fand576.3'F)andfortwovaluesofprimarypressure(2100psiaand2250psia).Theevaluations inthisreportarebasedonanalysesperformed fortheReratingProgram.Thisprogramisdiscussed inmoredetailinSection2.0.Inadditiontoaddressing anincreased SGTPlevelof30%,thefollowing increased operating marginswerealsoaddressed:

(1)Reduction ofSlandRHRdischarge pressureonrecirculation

-TheRHRandSIminimumsafeguards pumpheadcurveswerereducedby15%,anadditional 5%reduction fromthecurrentanalysisdegradation of10%.Thechargingpumpheadcurvedegradation ismaintained atthecurrentvalueof10%.(2)Theemergency dieselgenerator (EDG)starttimewasincreased from10secondsto30seconds(3)Tosupportincreased dTdrift,themarginbetweenthesafetyanalysislimits(SAL)andthenominalvaluesoftheK,andK,gainsoftheDonaldC.CookNuclearPlantUnit1OTATandOPbTsetpointequations wereadjusted.

(4)Anincreaseinthepressurizer codesafetyvalve(PSV)setpointtolerance from+/-1%to+/-3%(5)Decreased shutdownmarginforT,greaterthan200'F.Theanalysesandevaluations inthisdocumentsupportallofthesechanges.Discussions ofspecificanalysesaddressissuesmostrelevanttothoseanalyses.

Theoperating parameters fortheincreaseinsteamgenerator tubeplugginglevelandtheadditional operating marginslistedabovewillbereferredtothroughout thisreportasthe"SteamGenerator TubePlugging(SGTP)Program".

Thisreportprovidesthenecessary documentation tosupporttheTechnical Specification changesassociated withtheSteamGenerator TubePluggingProgram.Thetopicsaddressed inthisreportareasfollows:m."i1944-1w.wpf:1d~1295 XXIV Description ofLicenseAmendment SummaryofTechnical Specification ChangesBasisforEvaluations/Analyses Performed LossofCoolantAccidentAnalysesPost-LOCA HydrogenProduction Post-LOCA HotLegRecirculation TimeLOCAHydraulic ForcesNon-LOCAAnalysesContainment AnalysesSteamGenerator TubeRuptureAnalysesReactorCavityPressureEvaluation Radiological AnalysisPrimaryComponents Evaluations FluidandAuxiliary SystemsEvaluations FuelStructural Evaluation AlsoprovidedintheAppendixtothisreportaretheproposedTechnical Specification changes.Abriefsummaryoftheresultsofeachanalysisandevaluation isprovidedbelow.ACCIDENTANALYSISCONCLUSIONS Theresultsoftheaccidentanalysesandevaluations performed fortheSGTPProgramdemonstrate thatsafeoperation oftheDonaldC.CookNuclearPlantUnit1ismaintained.

Thebasesfortheevaluations andanalysesperformed areprovidedinSection2.1Asummaryoftheconclusions ofeachoftheaccidentanalysesisprovidedbelow.LareBreakIOCA(Section3.1.1)ThelargebreakLOCAanalysiswasreanalyzed fortheimpactoftheincreased tubeplugginglevel,reducedTDF,loopflowasymmetry, revisedECCSflows,andtheincreased EDGstarttime.ThelargebreakLOCAanalysiswasnotimpactedbythepressurizer codesafetyvalvetolerance

increase, therevisedK1/K4values,orthedecreased shutdownmargin.ThelargebreakLOCAanalysiswasperformed withthe1981versionoftheWestinghouse ECCSEvaluation ModelusingtheBASHcomputercode.Analysisassumptions includedECCSflowwiththeRHRcross-tie valvesclosed,atotalpeakingfactorof2.15,ahotchannelenthalpyrisepeakingfactorof1.55,andanaccumulator temperature of100'F.Afullspectrumbreakanalysiswasperformed atthenominalRCSconditions (initialRCSpressureof2250psiaandinitialhotlegtemperature of609.1'F)fromwhichthelimitingbreakdischarge coefficient wasdetermined.

Thelimitingbreakwasthenreanalyzed atthereducedhotlegtemperature andnominalRCSpressureof2250psia,andalsoatnominalhotlegtemperature andaninitialRCSpressureof2100psia.Theabovecaseswereallanalyzedwithminimumsafetyinjection flow,whichwasdetermined tobelimiting.

Thelimitingbreakwasdetermined tobeC,=0.4atthenominalhotlegtemperature (T>>=609.1'F)andapressureof2100psiawithm:51944-iw.wpf:1d441295 minimumsafetyinjection flow.Thepeakcladdingtemperature wascalculated tobe2164'F,whichislessthanthe2200'Flimitin10CFR50.46.

SmallBreakLOCA(Section3.1.2)ThesmallbreakLOCAanalysiswasreanalyzed fortheimpactoftheincreased tubeplugginglevel,reducedTDF,loopflowasymmetry, revisedECCSflows,andtheincreased EDGstarttime.ThesmallbreakLOCAanalysiswasnotimpactedbythepressurizer codesafetyvalvetolerance

increase, therevisedK1/K4values,orthedecreased shutdownmargin.Thesmall.breakLOCAanalysiswasperformed withtheWestinghouse smallbreakLOCAECCSEvaluation ModelusingtheNOTRUMPcode(including therecentmodelchangessubmitted inWCAP-10054-P, Addendum2andWCAP-10081-NP, Addendum2).Thekeyanalysisinputassumptions includedECCSflowswiththeHHSIcross-tie discharge valvesclosed,atotalpeakingfactorof2.32andhotchannelenthalpyrisepeakingfactorof1.55.Otheranalysisinputassumptions incorporated inthesmallbreakLOCAanalysisarereducedhotassemblyaveragepower(P)andapowershapebasedonareducedaxialoffsetof+20%.Asinglebreaksizeanalysiswasperformed atthepreviously-limiting breaksizeofthreeinches.Thecalculation usedthereducedtemperature, reducedpressureoperating condition.

Anevaluation ofthebreakspectrumandtherangeofoperating conditions concluded thattheanalyzedcasewouldremainboundingwithrespecttopeakcladtemperature.

Thecalculation wasperformed withminimumsafetyinjection flow,whichwaslimiting.

Thepeakcladdingtemperature wascalculated tobe1443'F,whichislessthanthe2200'Flimitin10CFR50.46.

LOCAHdraulicForcinFunctions (Section3.2)LOCAhydraulic forcesarerelatively insensitive tospecificSGTPlevels.TheDonaldC.CookNuclearPlantLOCAhydraulic forcesweremostrecentlyanalyzedfortheReratingProgram.TheRCSparameters usedintheexistinganalysis-of-record conservatively boundtheconditions at30%tubeplugging.

Therefore, theexistingLOCAforcesanalysesremainconservative relativetotheSGTPProgram.Non-LOCAAnalses(Section3.3)Thenon-LOCAeventswereaddressed byacombination ofevaluations andanalysesfortheimpactoftheincreased tubeplugginglevel,reducedTDF,loopflowasymmetry, revisedECCSflows,pressurizer codesafetyvalvetolerance

increase, increased EDGstarttime,revisedK1/K4values,anddecreased shutdownmargin.Thecomputercodesandmethodsusedforthenon-LOCAanalyseshavebeenpreviously approvedbytheNRC.Thenon-LOCAsafetyanalyseswerereviewedonthebasisofbothDNBandnon-DNBacceptance criteria.

AllDNBeventreanalyses werefoundtoyieldaminimumDNBRwhichremainsabovethelimitvalue.Theanalysesdemonstrate thatalllicensing basiscriteriacontinuetobemetandtheconclusions presented intheUFSARremainvalid.mh19441w.wpf:1d~1295xxvl Post-LOCA HdroenGeneration (Section3.4)Thepost-LOCA hydrogengeneration ratesthatwerereviewedaspartoftheReratingProgramweredetermined toremainapplicable totheSGTPProgram.~Ci'SThecontainment integrity analyseswereaddressed fortheimpactoftheincreased leveloftubeplugging, reducedthermaldesignflow,loopflowasymmetry, revisedECCSflows,andtheincreased EDGstarttime.Thecontainment analyseswerenotimpactedbythepressurizer codesafetyvalvetolerance

increase, therevisedK1/K4values,ortheincreased shutdownmargin.Theincreaseinthecontainment pressureandtemperature following aLOCAwasanalyzed.

Themassandenergyreleaseratescalculated aspartoftheSGTPProgramformedthebasistoevaluatethestructural integrity ofthecontainment following apostulated accidenttosatisfytheacceptance

criteria, GeneralDesignCriterion 38.EventhoughCookNuclearPlantislicensedtoGDC'sinAppendixHoftheoriginalFSAR,moreconservative acceptance criteriawereused.Thecontainment integrity analysisforthemostlimitingcase(i.e.,RHRcrosstievalveclosed)resultedinamaximumcalculated containment pressureof11.49psig,forthedouble-ended pumpsuctionminimumsafeguards breakcase.Sincethecalculated pressureisbelowthedesignpressureof12.0psig,theresultsoftheLOCAcontainment integrity analysisareacceptable.

TheMainsteam LineBreak(MSLB)massandenergyreleaseswereusedasinputintothecontainment integrity analysistodemonstrate thatthepeakcontainment temperature resulting fromadesignbasisMSLBwillnotexceedtheequipment qualification criterion fortheplant.Thecontainment pressureresponsedetermined fortheLOCAcontainment integrity analysisiscalculated tobemoreseverethanfortheMSLB,andtherefore, boundstheMSLBanalysis.

Forthelargebreakcase,thelimitingcaseamongthedouble-ended rupturesisthe1.4ft'ouble-ended rupture,102%power,MSIVfailurecase.Thiscaseyieldedacalculated peaktemperature of322.7'F.Forthesmallbreakcase,themostlimitingcaseintermsofpeakcalculated temperature isthe0.942ft'plitbreak,30%powerwithanMSIVfailure.Thiscaseresultedinacalculated peaktemperature of326'F.BothcasesarewithintheEnvironmental Acceptance Criteria.

Therefore, theanalysisdemonstrates thatthecontainment heatremovalsystemsfunctiontorapidlyreducethecontainment pressureandtemperature intheeventofaMSLB.GeneralDesignCriterion 50andAppendixKaresatisfied.

ShortTermContainment Analsis(Section3.5.1)Theshorttermcontainment analysisthatwasperformed fortheReratingProgramwasreviewedanditwasdetermined thattheconclusions providedfortheReratingProgramremainvalidfortheSGTPProgram.Thatis,theresulting peakpressures remainbelowtheallowable designpeakpressures forthepressurizer enclosure, thefanaccumulator roomandthesteamgenerator enclosure.

mA1944-1w.wpf:1d441295 xxvil SteamGenerator TubeRuture(Section3.6)TheSGTReventwasanalyzedfortheimpactoftheincreased tubeplugginglevelandassociated reducedTDFandloopflowasymmetry.

TheSGTRanalysiswasnotimpactedbyanyoftheSGTPProgramincreased operating margins.Thethyroidandwholebodydosesestimated forCookNuclearPlantUnit1,basedonthe30%SGTPevaluation, remainwithina"smallfraction" (10%)ofthe10CFR100exposurelimitguidelines.

Smallfractionisthesmallestoftheexposureguidelines definedinNUREG-0800.

Therefore, theconclusions oftheUFSARremainvalid.Post-LOCA HotLeRecirculation Time(Section3.7)Thehotlegswitchover toprecludeboronprecipitation andpost-LOCA longtermcoolingaren'otadversely affectedbythe30%SGTPProgram.Theproposedchangesdonotsignificantly affectthenormalplantoperating parameters, thesafeguards systemsactuations, theaccidentmitigation capabilities important totheseevents,ortheassumptions usedintheanalysisoftheseevents.Theproposedchangesdonotcreateconditions morelimitingthanthoseassumedintheLOCA-related analyses.

ReactorCaviPressureAnalsis(Section3.8)TheReactorCavityPressureAnalysisthatwasperformed fortheReratingProgramwasreviewedanditwasdetermined thattheconclusions providedfortheReratingProgramremainvalidfortheSGTPProgram.TheSGTPProgramparameters affecttheReactorCavityPressureAnalysisthroughthemassandenergyreleasesprovidedasinputintotheanalysis.

ThereisnodirectimpactofSGTPlevelonshort-term massandenergyreleaseratecalculations andcontainment subcompartment responseanalysis.

ThemassandenergyreleasesusedasinputfortheReactorCavityPressureAnalysisreflected limitingconditions andtherefore, theNSSSperformance parameters fortheSGTPProgramdidnotimpacttheresults.RadioloicalDoses(Section3.9)Areanalysis oftheoffsitedosesfollowing alargebreakLOCAwasperformed fortheincreaseinemergency dieselgenerator starttimeto30seconds.Whiletherewasaslightincreaseintheoffsitethyroiddoses,thedosesarewithintheapplicable limits.ThesourcetermsforLOCAandthefuelhandlingaccidentareunaffected bytheincreaseinSGTPleveloranyoftheotherSGTPProgramadders.mA1944-1w.wpf:1d441295 XXVIII

FLUIDANDAUXILIARY SYSTEMSEVALUATION CONCLUSIONS (Section3.10)Thefluidsystemsproofofdesigncalculations werereviewedfortheSGTPconditions.

Thisreviewdemonstrated'that theNSSSfluidsystemswillcontinuetofunctionadequately asdesignedforallconditions oftheSGTPProgram.ECCSflowrates wererevisedaspartoftheSGTPProgramandwereusedinthesafetyanalysesandevaluations.

IntheNSSS/BOPinterface area,theproposedNSSSPerformance Parameters fortheSGTPProgramwerecomparedwiththoseoftheReratingProgram.Theresultsoftheevaluation showthataSGTPlevelof30%willhavenoadverseeffectsontheBalanceOfPlant(BOP)systemsperformance (MainSteamSystem,Condensate andFeedwater System,Auxiliary Feedwater System,SteamGenerator andBlowdownSystem).Theywillcontinuetoperformacceptably attheconditions associated with30%SGTP.Theevaluations forthefluidandauxiliary systemsaredescribed inmoredetailinSection3.10.PRIMARYCOMPONENTS EVALUATION CONCLUSIONS SteamGenerators (Section3.11.1)Inthethermal-hydraulic areas:themodifiedmoistureseparator packagesonCookNuclearPlantUnit1willpermitoperation atsteampressures downto700psiaandbelowwithoutexceeding 0.25%moisturecarryover.

SteamGenerator operating characteristics willbeacceptable downtotheminimumsteampressureof589psia.Theevaluation ofthermal-hydraulic stability indicates satisfactory resultsforallSGTPcases.Theevaluation performed fortheeffectsoftheSGTPprogramonU-bendtubefatigueforCookUnit1isdocumented inWCAP-13814.

Itwasconcluded inWCAP-13814 thatfourtubesweresusceptible tohighcyclefatigueatthe30%SGTPProgramconditions andwouldrequirepreventative action.Structural analysesandevaluations performed fortheCookUnit1steamgenerators indicatethatthesteamgenerator components remainincompliance withtheapplicable ASMECoderequirements undertheSGTPconditions.

ReactorVessel(Section3.11.2)Theresultsofthestructural.

evaluations performed forthereactorvesseldemonstrate thatoperation ofCookNuclearPlantUnit1withintheparameters oftheSGTPprogramdoesnotresultinstressintensities orfatigueusagefactorswhichexceedtheacceptance criteriaoftheapplicable ASMECodeversions.

TheSGTPProgramdoesnotresultinanincreaseinthefastneutronfluencevaluescalculated fortheReratingProgram.Therefore, thereactorvesselrn31944-1w.wpf:1d441295 xxlx

integrity analysesperformed aspartoftheReratingProgramwillremainapplicable after30%SGTP.ReactorInternals (Section3.11.3).Resultsofthethermal-hydraulic analysesperformed forthereactorinternals indicatethattheSGTPProgramforCookNuclearPlantUnit1resultsinacceptable valuesofcorebypassflow,pressuredrops,component liftforces,andmomentumfluxvalues.Itwasalsoconfirmed thatthecontrolroddroptimelimitof2.4secondsremainsapplicable fortheSGTPProgramconditions.

Fromthecomponent stressanalysisandtheflowinducedvibration evaluations, itisconcluded thatthemarginsofsafetyarewithinacceptable limitspertheoriginaldesignbasis.ControlRodDriveMechanisms (Section3.11.4)Theconclusion ofstructural evaluations performed forthe30%SGTPconditions fortheCRDMsdemonstrate thattheoperability, servicelife,andstructural integrity oftheCRDMlatchassembly, driverod,andcoilstackwillnotbeadversely affected.

ReactorCoolantPums(Section3.11.5)Thereviewperformed ofthereactorcoolantpumpsforthe30%SGTPconditions demonstrate thattheconditions areacceptable forthe93ARCP,andnoadditional thermalorstructural analysesarerequiredtodemonstrate compliance withtheapplicable codesandstandards.

TheRCPmotorevaluation revealedthatthemotorsareacceptable foroperation atthe30%SGTPconditions.

Pressurizer (Section3.11.6)Afatigueanalysisperformed fortheCookUnit1pressurizer, incorporating themostconservative conditions oftheSGTPprogram,demonstrated thatthepressurizer remainsincompliance withtheapplicable ASMECodecriteria.

ReactorCoolantPiinandSurts(Section3.11.7)Anevaluation wasperformed todetermine theeffectsofthe30%SGTPconditions ontheprimarylooppiping,primaryequipment

supports, andtheprimaryequipment nozzles.Operation attheSGTPProgramconditions wasfoundtobeacceptable becausetheseconditions arealreadyenveloped bytheReratingProgram.Inaddition, thereratingtransients andplantparameters associated withtheReratingandSGTPProgramsforDonaldC.CookNuclearPlantUnit1havebeenreviewed.

Theimpactonthedesignbasisanalysisforthemh1944.1w.wpf:1d441295 NRCBulletin88-08evaluation oftheauxiliary spraypipingandtheNRCBulletin88-11evaluation ofthepressurizer surgelinepipingisinsignificant.

AuxiliaComonents(Section3.1f.8)Evaluations wereperformed fortheauxiliary tanks,pumps,valves,andheatexchangers todetermine theeffectsoftherevisedRCSparameters duetotheSGTPProgram.Theresultsoftheseevaluations demonstrated that,duetoconservatively specified parameters usedintheprocurement oftheauxiliary equipment, the30%SGTPparameters arenotexpectedtoadversely affectthefunctionorstructural integrity ofthisequipment, FuelStructural Evaluation (Section3.12)Evaluations wereperformed ofthefuelforCookNuclearPlantUnit1fortheSGTPProgramconditions intheareasoffuelrodandfuelassemblystructural integrity, coredesign,andthermal-hydraulic design.Thefuelassemblystructural integrity isnotaffectedbytheSGTPProgram,andthecoreeoolablegeometryismaintained forthe15x15OFAfuelintheUnit1core.Theevaluation ofthefuelrodstructural integrity indicates theseconditions willbeacceptable, althoughitisnotedthatcycle-specific verification duringthenormalreloadwillstillbeperformed.

Theresultsofthecoredesignevaluation indicated thattheSGTPProgramconditions resultinnoimpacttothecoredesignexceptforthevaluesofthestatepoint fortheSteamline BreakAnalysis, andtheDroppedRodAnalysis.

Thermal-hydraulic analysesweremadeforthefuelforthelimiting30%SGTPparameters usingRTDPmethodology.

TheanalysisshowedthattheDNBRdesignbasiswasmetforthelimitingDNBevents.Thisanalysiscausedtheavailable DNBmargintoincrease.

Thismargincanbeusedforflexibility ofdesignandtooffsetunanticipated DNBRpenalties.

m:51944-1 w.wpf:1d441295 xxxl

2.1DESIGNPOWERCAPABILITY PARAMETERS Thissectiondescribes theparameters whichwereusedasthebasisfortheevaluations andanalysesperformed tosupportthe$GTPProgramforCookNuclearPlantUnit1.TheNSSSperformance parameters featurethecurrentlicensedNSSSpowerof3262MWt,aT,~temperature rangefrom553'Fto576.3'F,twoprimarypressurevaluesof2250psiaor2100psia,amaximumaverageandpeakSGTPlevelof30%,reducedTDF,and5%loopflowasymmetry.

TheRCStemperature rangeisboundedbytheReratingProgram.Alsoincorporated intotheSGTPProgramwas'anRCSflowmeasurement uncertainty rangeof1.9%to2.5%.TheTechnical Specifications willberevisedtoincorporate theMinimumMeasuredFlow(MMF)corresponding toaflowmeasurement uncertainty of2.5%.Thisincreasewillprovideadditional marginforinstrumentation.

Ifadditional flowmarginisneededinthefuture.themargincanbereallocated frominstrumentation margintoRCSflowmarginbyrevisingtheMMFTechnical Specification becausealloftheaccidentanalysesevaluated aMMFof339,100gpm.AMMFof339,100gpmtotalreflectsa1.9%flowmeasurement uncertainty.

AMMFof341,100gpmtotalreflectsa2.5%flowmeasurement uncertainty.

TheRCSflowmarginhasbeenreviewedandsufficient marginexiststomaintaintheTDFof83,200gpm/loopwiththe2.5%flowmeasurement uncertainty.

Abriefdescription ofeachsetofparameters isprovidedbelow:Case1:ThesearetheoriginalNSSSperformance parameters forUnit1andareshownforcomparison withtherevisedparameters.

TheNSSSpowerlevelof3250MWtdoesnotaccountforreactorcoolantpumpheat;atthetimethatUnit1wasdesigned, itwasthecustomtoindicateonlythecorepowerlevelvalue.Case2:Theseparameters incorporate acorepowerlevelof3250MWt,anNSSSpowerlevelof3262MWt(whichincludes12MWtofreactorcoolantpumpheat),anaveragesteamgenerator tubeplugginglevelof30%,primarypressures ofeither2250psiaor2100psia,andalowerboundvesselaveragetemperature of553.0'F.Case3:Theseparameters incorporate thesamefeaturesascase2,exceptthattheprimarytemperatures andresulting secondary parameters incorporate anupperboundvesselaveragetemperature of576.3'F.Thiscasewasusedasthebasisforselectedanalyses, wherehighpnmaiytemperatures werelimiting.

Case4:Theseparameters incorporate thesamefeaturesascase2,exceptthattheTDFwasreducedto79,000gpm/looptobound5%loopflowasymmetry.

Case5:Theseparameters incorporate thesamefeaturesascase4,exceptthattheprimarytemperatures andresulting secondary parameters incorporate anupperboundvesselaveragetemperature of576.3'F(thehighestvesselaveragetemperature considered fortheSGTP'31944-1w.wpf:1d~11952.1-1

,

1.0INTRODUCTION

-DESCRIPTION OFLICENSEAMENDMENT REQUEST1.1PURPOSEFORCHANGETheDonaldC.CookNuclearPlantUnit1hasexperienced tubecorrosion problemsinitssteamgenerators and,asaresult,anincreasing numberoftubeshavebeenpluggedduringthelastseveraloutages.Steamgenerator tubepluggingpotentially decreases reactorcoolantsystemflowduetoincreased flowresistances throughthesteamgenerators.

Asthenumberofpluggedtubesincreases, theRCSflowmaybereducedtoavaluebelowthatwhichiscurrently analyzedinthelicensing basis.Currently, thelicensing basisanalysesfortheDonaldC.CookNuclearPlantUnit1aredocumented intheUpdatedFinalSafetyAnalysisReport{UFSAR).Theseanalysesareboundingforamaximumaveragesteamgenerator tubeplugging(SGTP)levelofupto10%withapeaklevelof15%inanyonesteamgenerator.

Thisamendment requestreflectsthechangestothesafetyanalysisassumptions andresultsduetotherevisedoperating conditions resulting fromanincreased levelofsteamgenerator tubeplugging.

Whiletheanalysesandevaluations werebeingperformed fortheincreased leveloftubeplugging, severaloperating marginswereincreased andincorporated intotheanalysesandevaluations inordertomaximizethebenefitofthereanalysis.

Therefore, inadditiontoaddressing anincreased SGTPlevelof30%,thefollowing increased operating marginswerealsoaddressed:

(1)Reduction ofSlandRHRdischarge pressureonrecirculation

-TheRHRandSlminimumsafeguards pumpheadcurveswerereducedby15%,anadditional 5%reduction fromthecurrentanalysisdegradation of10%.Thechargingpumpheadcurvedegradation ismaintained atthecurrentvalueof10%.(2)Theemergency dieselgenerator startuptimewasincreased from10secondsto30seconds(3)Tosupportincreased hTdrift,themarginbetweenthesafetyanalysislimits{SAL)andthenominalvaluesoftheK,andK,gainsoftheDonaldC.CookNuclearPlantUnit1OTZTandOPbTsetpointequations wereadjusted.

{4)Anincreaseinthepressurizer codesafetyvalve(PSV)setpointtolerance from+/-1%to+/-3%(5)Decreased shutdownmarginforT,~greaterthan200'F.Therevisedparameters associated withthe,increaseintubeplugginglevelto30%SGTP,bothdirectlyandindirectly, arereferredtothroughout thisreportasthe"SteamGenerator m%1944.1w.wpf:1d~1295 TubePluggingProgram".

ThisprogramresultedinchangestotheDonaldC.CookNuclearPlantUnit1Technical Specifications, including:

CoreSafetyLimitsOPb,T/OTbT Setpoints ShutdownMarginforModes1,2,3,and4DNBParameters RTSResponseTimesESFASInstrumentation LogicPressurizer CodeSafetyValveLiftSettingTolerance ECCSPumpDischarge PressureonRecirculation MinimumRWSTTemperature Containment InternalPressureRCSVolumeEmergency DieselGenerator StartTimem:11944.1w.wpt:

1d4412951.1-2 beenbasedonthenewcoresafetylimitsandaccountforinstrument uncertainties.

Thereference temperatures arenowindicated valuesandthetemperature rangethatwasanalyzedisspecified.

ShutdownMarinforMODES123and4Shutdownmargin(SDM)requirements providesufficient reactivity margintoensurethatacceptable fueldesignlimitswillnotbeexceededfornormalshutdownandanticipated operational occurrences.

Assuch,theSDMdefinesthedegreeofsubcriticality thatwouldbeobtainedimmediately following theinsertion orscramofallshutdownandcontrolrods,assumingthatthesinglerodclusterassemblyofthehighestreactivity worthisfullywithdrawn.

TheLimitingCondition forOperation (LCO)3.1.1.1ShutdownMarginhasbeenrevisedfrom1.6'lodk/kto1.3'kAk/k.Thischangeisalsoincludedintheassociated Bases.Theacceptability ofthedecreaseintheSDMisbasedonre-analysis ofthemostlimitingaccident, coreresponsetoasteamlinebreak.DNBParameters RCSTavandRCSFlowLCO3.2.5,DNBParameters specifies RCSparameters assumedasinitialconditions inthetransient andaccidentanalyses.

InTable3.2-1,DNBParameters, theRCST,~hasbeenchangedfrom<570.9'Fto<(576.3+5.1)-(readability error)'F.

The576.3'Fvaluehasbeenverifiedinthere-analyses.

Theadditional 5.1'Fisincludedtoaccountfortemperature uncertainty factorssuchascoldlegstreaming, asdocumented inWCAP-12568, Rev.1.Thereadability errorwillbedetermined byAEPSC.Additionally, inTable3.2-1,theRCSTotalFlowRatehasbeenreducedfrom2361,600to>341,100gpm.Thisreduction isbasedontheincreaseintheSGTPlimitto30/oandincludesa2.5/oinstrument uncertainty.

RTSResonseTimesReactorTripSystem(RTS)Instrumentation responsetimesareassumedinaccidentanalysesforthetimeintervalfromwhenthemonitored parameter (level,pressure, temperature, etc.)exceedsitssetpointatthesensoruntillossofstationary grippercoilvoltage.RTSInstrumentation responsetimesspecified inTable3.3-2forFunctions 9,10,14,and16(Pressurizer Pressure-Low, Pressurizer Pressure-High,.

SteamGenerator WaterLevel-Low-Low,andUndervoltage-Reactor CoolantPumps)relaxedfrom1.0secondto2.0secondsm."i1944-1w.wpf:1d 4412951.2-2 (Functions 9,10,and14)orto1.5seconds(Function 16).Theacceptability oftheserelaxations hasbeenverifiedbyaccidentanalyses.

ESFASInstrumentation LoicTheEngineered SafetyFeatureActuation System(ESFAS)instrumentation initiates necessary safetysystems,basedonthevaluesofselectedunitparameters, toprotectagainstviolating coredesignlimits,protecttheRCSpressureboundary, andtomitigateaccidents.

ESFASInstrumentation logicforFunctional Units1and4andfortheP-12interlock havebeenrevised.Theserevisions reflecttheUnit1implementation ofthe"hybrid"steamline breakprotection logicthatispresently usedonUnit2.Function1-SafetyInjection, TurbineTrip,Feedwater Isolation andMotor-driven Feedwater Pumps:Actuation onSteamFlowinTwoSteamLines-High coincident witheitherT,-Low-LoworSteamLinePressure-Low hasbeenreplacedbyactuation onSteamLinePressure-LowFunction4-SteamLineIsolation:

Actuation onSteamFlowinTwoSteamLines-High coincident witheitherT,~-Low-Low orSteamLinePressure-Low hasbeenreplacedbyactuation onSteamFlowinTwoSteamLines-High coincident withT,-Low-LowandonSteamLinePressure-Low Pressurizer CodeSafeValveLiftSettinTolerance Thepressurizer safetyvalvesprovide,inconjunction withtheReactorProtection System,overpressure protection fortheRCS.Accidentandsafetyanalyseswhichrequiresafetyvalveactuation assumeoperation ofthreepressurizer safetyvalvestolimitincreases inRCSpressure.

Thelimitprotected bythisspecification isthereactorcoolantpressureboundarySafetyLimitof110%ofdesignpressure.

TheLCO3.4.2(MODES4and5)andLCO3.4.3(MODES1,2,and3)pressurizer codesafetyvalveliftsettingtolerance hasbeenincreased froma1%toa3%.Theacceptability oftheincreased safetyvalvetolerance hasbeenestablished byevaluation oranalysisofapplicable eventsincluding lossofload,turbinetrip,lockedrotor,lossofnormalfeedwater, feedwater linebreak,andlossofallpowertostationauxiliaries.

m:51944-1w.wpf:1d 4412951.2-3 ECCSPumFlowsLCO3.5.2,ECCSSubsystems

-T,~t350'F,specifies requirements applicable topumps.heatexchangers, andflowpathscreditedwithcorecoolingfollowing anaccident.

TheSurveillance Requirement (SR)4.5.2.f(centrifugal chargingpump,safetyinjection pumpandresidualheatremovalpumptests)requirements havebeenrevisedtospecifymeasurement ofdifferential pressurealongwithrevisedacceptance criteriapressures.

Therevisedcentrifugal chargingpumpdifferential pressurecriteriarepresents a10%pumphead-degradation.

Thedifferential pressurecriteriaspecified forthesafetyinjection pumpandtheresidualheatremovalpumpreflectsrelaxations intheassociated pumpcurveswhichrepresent 15%pumpheaddegradation.

Therelaxation forcentrifugal chargingpumpsisalsoapplicable toLCOs3.1.2.3and3.1.2.4.Theacceptability oftherelaxedpumpcurveshasbeenverifiedforallapplicable accidents.

MinimumRWSTTemeratureTheminimumRefueling WaterStorageTank(RWST)temperature hasbeenchangedfrom80'Fto70'FinLCOs3.1.2.7,3.1.2.8,and3.5.5andBases3.5.5.The70'Fminimumtemperature isacceptable basedontheLOCAandnon-LOCAanalysesperformed fortheCookNuclearPlantUnit1licensing basis.IContainment InternalPressureThemaximumcalculated post-accident containment pressuremustremainbelowthecontainment designpressureof12.0psig.Theresultsofthecontainment integrity analysesperformed fortheSGTPprogramresultedinamaximumcalculated containment pressureof11.49psig.Thus,thevalueintheBasesforLCO3.6.1.4(11.89psig),InternalPressure, isbeingrevisedtoreflecttheanalysisresults.EmerenDieselGenerator EDGStartTimeLCO3.8.1,ACSources-Operating, specifies requirements foroff-siteandon-site(dieselgenerator)

ACsources,including EDGtestingrequirements todemonstrate thecapability toachievetherequiredvoltageandfrequency withinthespecified time.TheEDGstarttime,plustheloadsequencer loadingtimes,plustheequipment actuation/start times,establishes thetotaltimeuntilthefunctionofESFequipment isassumedinaccidentanalyses.

TheSR4.8.1.1.2.a.4, SR4.8.1.1.2.e.4.b, SR4.8.1.1.2.e,6.b, andSR4.8.1.1.2.f.3 requirements havebeenrevisedtospecifyarelaxedEDGstarttimeof30seconds.Inthesafetyanalysis, the30secondtimeisthetimeatwhichtheloadsequencer isassumedtom:51944-1w.wpt:1d441295 1.2<

startloading.Additionally, theSRshavebeenrevisedtospecifythatvoltageandfrequency shallbeachievedratherthanengineRPM,consistent withthesafetyanalysis, Regulatory Guide1.9,andNUREG-1431.

Relaxation oftheEDGstarttimeto30secondshasbeenshowntobeacceptable basedonre-analysis oflimitingaccidents.

Onlytheresponsetimesspecified inTable3.3-5,Engineered SafetyFeaturesResponseTimeswhichincludethedieselgenerator starttimeareaffectedbythischange.Thelonger,responsetimesassumedinLOCA,Non-LOCA, andcontainment analyseshavebeenspecified inTable3.3-5.RCSVolumeDesignFeaturesSection5.4.2specifies thetotalcontained volumeoftheRCS.WiththeincreaseoftheSGTPlimitto30%,acorresponding reduction inRCSvolumemustbespecified.

Sincetheactualleveloftubepluggingmaychangeeachoutage,arangeofRCSvolumecorresponding totherangeof0%to30%tubeplugginghasbeenspecified:

approximately 12,466ft'o11,551ft'.m:41944.1w.wpf:

1d441295'.2-5 TABLE1.2-1SUMMARYOFTECHNICAL SPECIFICATION CHANGESPAGESECTIONDESCRIPTION OFCHANGEBASIS2-2B2-1{a)2.1Bases2.1CoreSafetyLimitsReplaceFigure2.1-1,ReactorCoreSafetyLimitsThesafetyanalysislimitDNBRspecified intheBasesforSection2.1hasbeenrevisedfrom1.45to1.42.Figure2.1-1hasbeenreplacedwithanewfigurebasedonthelatestanalyses, reflecting 30%SGTP,reducedratedthermalpower,reducedRCSflow,etc.Section3.3.2.12-72-82-9B2-4B2-52.2B2.2.1OTbT&OPbTSetpoints ReviseOTATandOPETTripSetpointandAllowable Valuenotes.Proposedsettingsbasedonnewcoresafetylimitsandaccountforinstrument uncertainties.

Section3.3.2.18Table3.3-33/41-1B3/41-13.1.1.1ShutdownMarginRelaxation basedonre-ShutdownMarginlimitrelaxedanalysisoflimitingaccidentfrom1.6to1.3%hk/k-coreresponsetosteamlinebreak.Sections3.3.4.783.3.5.63/42-143.2.5Table3.2-1DNBParameters ReviseTable3.2-1,DNBParameters, RCST,from570.9'Fto{581.4-readability error)'FReadability errorisresponsibility ofAEPSC.T,inputassumption verifiedbyreanalyses.

Sections2.1,3.3.2.183.12.4m'81944.1w.wpt:1d441295 1.2-6

TABLE1.2-1(continued)

SUMMARYOFTECHNICAL SPECIFICATION CHANGESPAGESECTIONDESCRIPTION OFCHANGEBASIS2-53/42-142.2Table2.2-13.2.5Table3.2-1DNBParameters ReviseTable3.2-1,DNBParameters, RCSTotalFlowfrom>361,600toa341,100gpm.Changebasedon30%SGTPlimit.Thevalueof341,100includesa2.5%instrument uncertainty.

Analysisused339,100whichincludesa1.9%instrument uncertainty.

Sections2.1,3.3.2.183.12.43/43-103/43-113.3.1Table3.3-2RTSResponseTimesTheacceptability oftheseRTSInstrumentation responserelaxations verifiedbytimesforFunctions 9,10,14,accidentanalyses.

and16(Pressunzer Pressure-Low,Pressurizer Pressure-Sections3.183.3High,SteamGenerator WaterLevel-Low-Low, andUndervoltage-Reactor CoolantPumps)relaxedfrom1.0secondto2.0seconds(Functions 9,10,and14)orto1.5seconds(Function 16)~3/43-173/43-213/43-23a3/43-243/43-263/43-283/43-313/43-33Table3.3-3Table3.3-4Table3.3-5Table4.3-2ESFASLOGICESFASInstrumentation logicforFunctional Units1and4andfortheP-12interlock havebeenrevised.ESFASlogicchangehasbeenshowntobeacceptable bynon-LOCAanalyses.

Section3.3m:11944-1w.wpf:

1d4412951.2-7 TABLE1.2-1(continued)

SUMMARYOFTECHNICAL SPECIFICATION CHANGESPAGESECTIONDESCRIPTION OFCHANGEBASISB3/44-13/44.1DNBRLimitTheDNBRlimitspecified inBases3/44.1isnolongerapplicable.

"1.69"replacedwith"thesafetyanalysislimit"Nobasisrequired.

3/44-43/44-53.4.23.4.3SafetyValveLiftSettingPressurizer codesafetyvalveliftsettingpressuretolerance increased to3%Relaxation basedonevaluation oranalysisofseverallimitingevents.Sections1.183.33/41-113.1.2.33/41-123.1.2,43/45-53.5.2ECCSPumpFlowsCentrifugal

charging, safetyinjection andresidualheatremovalpumps'est acceptance criteriarelaxed.CCP-10%RHR8Sl-15%Relaxedpumpcurveshavebeenverifiedtobeacceptable forallapplicable accidents.

Section3.3.4.7&3.103/41-153.1.2.73/41-163.1.2.83/45-113.5.5B3/45-3B3.5.5RWSTTemperature MinimumRWSTtemperature reducedfrom80'Fto70'FThe70'Fminimumtemperature isacceptable basedontheLOCAandnon-LOCAanalyses.

Section3.1.1,W-letter92AE-G-074,datedJune12,1992mA1944-1w.wpf:1d~1295 1.2-8 TABLE1.2-1(continued)

SUMMARYOFTECHNICAL SPECIFICATION CHANGESPAGEB3/46-23.6.1.4InternalPressureThus,thevalueintheBasesforLCO3.6.1.4(11.89psig),InternalPressure, isbeingrevisedtoreflecttheanalysisresults.SECTIONDESCRIPTION OFCHANGEBASISTheresultsofthecontainment integrity analysesperformed fortheSGTPprogramresultedinamaximumcalculated containment pressureof11.49psig.Section3.53/48-33/48-53/48-73/43-273/43-283/43-293/43-303.8.1.13.3.2Table3,3-5EDGStartTime/ESFASResponseTimesEDGstarttimerelaxedto30seconds.ESFASresponsetimesaffectedbyEDGstarttimerelaxation revised.Theincreaseof20secondsintheEDGstarttimehasbeenshowntobeacceptable forlimitingaccidents.

Sections3.383.55.4.2RCSVolumeChangebasedon30%DesignFeature,RCSvolumeSGTPlimit.reduced.Section3.10m:51944-1w.wpf:1d

~12951.2-9 1.2CURRENTLICENSEBASISANDFUNCTIONOFIDENTIFIED TECHNICAL SPECIFICATIONS ANDDESCRIPTION OFPROPOSEDCHANGETheproposedchangestotheDonaldC.CookNuclearPlantUnit1Technical Specifications aresummarized inTable1.2-1.Thesechangesreflecttheimpactonthedesign,analytical methodology, andsafetyanalysisassumptions outlinedinthisamendment request.TheproposedTechnical Specification changesareincludedinAppendixAofthisreport.Abriefoverviewofthesignificant Technical Specification changesfollows.Thechangesarebasedonanalysesandevaluations associated withtheSGTPProgram.Sincenewanalysesandevaluations wererequiredtoestablish theacceptability oftheSGTPlevel,severalrelatedTechnical Specification relaxations wereveriTied.

CoreSafeLimitsTechnical Specification Figure2.1-1,ReactorCoreSafetyLimits,showsthelociofpointsofthermalpower,RCSpressureandaveragetemperature belowwhichthecalculated DNBRisnolessthanthedesignDNBRvalueandtheaverageenthalpyatthevesselexitislessthantheenthalpyofsaturated liquid.Thefigureisbasedontheenthalpyhotchannelfactor.Figure2.1-1hasbeenreplacedwithanewfigurebasedonthelatestanalyses, reflecting 30%tubeplugging, reducedratedthermalpower,reducedRCSflow,etc.OPbT/OTbT SetpintsTechnical Specification Table2.2-1liststhereactorprotection systeminstrumentation tripsetpoints forthevarioustripfunctions.

Thereactortripsetpointlimitsspecified inTable2.2-1arethenominalvaluesatwhichthereactortripsaresetforeachfunctional unit.TheThermalOverpower bT(OPAT)tripfunctionprovidesassurance offuelintegrity (e.g.,nofuelmeltingandlessthan1%claddingstrain)underallpossibleconditions, limitstherequiredrangeforThermalOvertemperature hT(OTBT)protection, andprovidesabackuptotheHighNeutronFluxtrip.TheOTbTtripfunctionprovidessufficient coreprotection toprecludedeparture fromnucleateboiling(DNB)overarangeofoperating andtransient conditions.

Thesetpointisautomatically variedwithtemperature.

pressure, andtheaxialpowerdistribution.

TheF(d,l)penaltyfunctionadjuststhetripsetpointforaxialpeaksgreaterthandesign.Revisions tothelimitingsafetysystemsettingsfortheOTZTandOPbTtripfunctions (Table2.2-1,Notes1,2,3,and4)areproposedtomaintaioconsistency withthenon-LOCAAccidentAnalysis.

Thesetripfunctions provideprimaryprotection againstDNBandfuelcenterline melting(excessive kw/ft)duringpostulated transients.

Theproposedsettingshavem31944-1w.wpt:1d441295 1.2-1

2.2NSSSDESIGNTRANSIENTS TheNSSSdesigntransients evaluation fortheDonaldC.CookNuclearPlantUnit1SGTPProgramwascompleted andconfirmed thattheNSSSdesigntransients developed aspartoftheDonaldC.CookNuclearPlantUnits1and2ReratingProgramcontinuetoapplytoDonaldC.CookNuclearPlantUnit1attheincreased SGTPconditions.

Theevaluation consisted ofacomparison oftheNSSSperformance parameters fortheSGTPProgramwiththeparameters fortheReratingProgram.Thecomparison concluded thattheSGTPProgramparameters thathavethepotential toimpacttheNSSSdesigntransients (i.e.,temperatures, pressures, andpowerlevels)areboundedbytheparameters usedintheReratingProgram.TheNSSSperformance parameters thatwerenotboundedinthismanner(i.e.,SGTPlevelandRCSflow)wereevaluated anddetermined tonothaveasignificant impactontheNSSSdesigntransients.

Overall,theevaluation concluded thattheNSSSdesigntransients developed aspartoftheReratingProgramcontinuetoapplytoDonaldC.CookNuclearPlantUnit1.m:51944.1w.wpf:1d441195 2.2-1 Program).

Theseparameters wereusedforselectedanalyses, wherehighprimarytemperatures werelimiting, andsensitive toRCSloopflow.m&1944-1w.wpf:1d~1195 2.1-2 TABLE2.1-1COOKNUCLEARPLANTUNIT1NSSSPERFORMANCE PARAMETERS FORSGTPPROGRAMParameter NSSSPower,MWtCorePower,MWtRCSFlow,(gpm/loop)

MinimumMeasuredFlow,(totalgpm)RCSTemperatures,

'FCoreOutletVesselOutletCoreAverageVesselAverageVessel/Core InletSteamGenerator OutletZeroLoadRCSPressure, psiaSteamPressure, psiaSteamFlow,(10'b/hr.tot.)

Feedwater Temperature,

'F%SGTubePlugging'Unit1,Original)

Case13250325088,500361,600602.0599.3570.5567.8536.3536.3547.0225075814.12434.8FlowDefinitions:

RCSFlow(ThermalDesignFlow)-Theconservatively lowflowusedforthermal/hydraulic design.Thedesignparameters listedabovearebasedonthisflow.mal944.1w.wpf:1d~1195 2.1-3

TABLE2.1-1(continued)

COOKNUCLEARPLANTUNIT1NSSSPERFORMANCE PARAMETERS FORSGTPPROGRAMParameter (Revised)

(Revised)

Case2Case3(Revised)

(Revised)

Case4Case5'SSSPower,MWtCorePower,MWtRCSFlow,(gpm/loop)

MinimumMeasuredFlow,(totalgpm)-RCSTemperatures,

'FCoreOutletVesselOutletCoreAverageVesselAverageVessel/Core InletSteamGenerator OutletZeroLoadRCSPressure, psiaSteamPressure, psiaSteamFlow,(10'b/hr.tot.)

Feedwater Temperature,

'F%SGTubePlugging3262325083,200339,100589.7586.8555.8553.0519.2518.9547.02250or210059514.12434.8303262325083,200339,100611.9609.1579.4576.3543.5543.2547.02250of210074914.17434.8303262325079,000339,100591.5588.5556.0553.0517.5517.2547.02250of210058914.12434.8303262325079,000339,100613.6610.8579.7576.3541.8541.6547.02250or210074214.17434.830FlowDefinitions:

RCSFlow(ThermalDesignFlow)-Theconservatively lowflowusedforthermal/hydraulic design.Thedesignparameters listedabovearebasedonthisflow.-MinimumMeasuredFlow-Theflowspecified intheTech.Specs.whichmustbeconfirmed orexceededbytheflowmeasurements obtainedduringplantstaftupandistheflowusedinreactorcoreDNBanalysesforplantsapplyingtheRevisedThermalDesignProcedure.

MMFbasedonaflowmeasurement uncertainty of1.9%.AnalysesalsoboundaMMFof341,100gpmtotalwhichreflectsaflowmeasurement uncertainty of2.5%.m%1944-1w.wpf:1d441195 2.1-4

2.3CQNTROl/PROTECTION SYSTEMSETPQINTS ControlSystemswereevaluated andfoundtobeboundedbytheanalysesperformed aspartoftheReratingProgram.Theseanalysesreflected theobjective ofoptimizing controlparameters, primarily withrespecttotwoaspectsofplantbehavior:

stability ofthecontrolsystemsandoperating marginstothevariousreactorprotection systemtrips.Theflexibility identified duringtheReratingProgramtoadjustthefullloadvesselaveragetemperature andprimarypressureasnecessary onacycle-to-cycle basisremainsapplicable totheSGTPProgram.Controlsystemssetpoints areselectedforeachfuelcyclefromthoseanalyzedfortheReratingProgram.Therefore, theplantwillbeadequately stableforallSGTPProgramoperating conditions andoperateswithadequatemargintoreactorprotection systemsetpoints.

m81944-1w.wpf:1d~1195 2.3-1 rodinthethreephases.TheRevisedPADFuelThermalSafetyModel,described inReferences 13,generates theinitialfuelrodconditions inputtoLOCBART.SATAN-Vlcalculates theRCSpressure,

enthalpy, density;andthemassandenergyflowratesintheRCS,aswellassteamgenerator energytransferbetweentheprimaryandsecondary systemsasafunctionoftimeduringtheblowdownphaseoftheLOCA.SATAN-Vlalsocalculates theaccumulator watermassandinternalpressureandthepipebreakmassandenergyflowratesthatareassumedtobeventedtothecontainment duringblowdown.

Attheendoftheblowdown, information onthestateofthesystemistransferred totheREFILLcodewhichperformsthecalculation oftherefillperiodtobottomofcorerecoverytime.Oncethevesselhasrefilledtothebottomofthecore,therefloodportionofthetransient begins.TheBASHcodeisusedtocalculate thethermal-hydraulic simulation oftheRCSfortherefloodphase.Information concerning thecoreboundaryconditions istakenfromalloftheabovecodesandinputtotheLOCBARTcodeforthepurposeofcalculating thecorefuelrodthermalresponsefortheentiretransient.

Fromtheboundaryconditions, LOCBARTcomputesthefluidconditions andheattransfercoefficient forthefulllengthofthefuelrodbyemploying mechanistic modelsappropriate totheactualflowandheattransferregimes.Conservative assumptions ensurethatthefuelrodsmodeledinthecalculation represent thehottestrodsintheentirecore.Thelargebreakanalysiswasperformed withtheDecember1981versionoftheEvaluation Modelmodifiedtoincorporate theBASH(Reference 7)computercode.InputParameters andInitialConditions:

Theanalysispresented inthissectionwasperformed withareactorvesselupperheadtemperature equaltotheRCShotlegtemperature andauniformSGTPlevelof30%.Theanalysisisalsobasedonplantoperation withtheRHRcross-tie valvesclosed,andanEDGstarttimeof30secondswhichresultsinasafetyinjection delaytimeof47seconds.Alistofplantinputparameters usedinthelargebreakLOCAanalysisisprovidedinTable3.1-2.Arangeofreactoroperating temperatures wasanalyzedinordertojustifyplantoperation atareactorpowerlevelof3250MWtbetween609.1'Fto586.8'Finthehotlegsand543.5'Fand519.2'Finthecoldlegs.Inadditiontothetemperature rangeanalyzed, initial.RCS pressurewasalsovariedtojustifyplantoperation at2250and2100psia.Afullspectrumbreakanalysiswasperformed fornominalRCSconditions (initialRCSpressureof2250psia'andhotlegtemperature of609.1'F)fromwhichthelimitingbreaksizewasdetermined.

Thelimitingbreakwasthenreanalyzed forthereducedhotlegtemperature of586.8'FandnominalRCSpressureof2250psia.Thelimitingbreakwasalsoreanalyzed forthenominalhotlegm:4)944-2w.wpt:1d~1195 3.1-6 2.0BASISFOREVALUATIONS/ANALYSES PERFORMED ThepurposeoftheSGTPProgramwastoperformthenecessary NSSS-related effortstosupportanincrease'in thelevelofSGTPtoashighas30%andcontinueoperational flexibility intermsofprimarytemperature andpressure.

Inadditiontothechangeinparameters associated withtheincreased SGTPlevel,additional changeswereincorporated intotheanalyses, asdescribed inSection1.0(e.g.,EDGdelaytime,pressurizer safetyvalvetolerance, etc.).Previously, AEPSCsubmitted areportforNRCreviewinOctober1988,whichprovidedthenecessary

analysis, documentation, andlicensing efforttosupportoperation atreducedprimarytemperatures andpressures.

Theseanalyseswereperformed inanefforttoreducethepropensity fortheinitiation andpropagation ofcorrosion intheCookNuclearPlantUnit1Series51steamgenerator tubes.TheWestinghouse inputforthissubmittal wasprovidedinWCAP11902(Reference 1).Theeffortsperformed forWCAP11902supported 100%thermalpoweroperation (3250MWtcorepower)intherangeofvesselaveragetemperatures between547'Fand576.3'F,atprimarypressurevaluesof2100psiaand2250psia.Thepnmarypressures wereintendedastwodiscretevalues;theprogramwasnotstructured tosupportacontinuous rangeofprimarypressures.

Theintentofthereducedprimarypressurevalueistominimizetheprimarytosecondary pressuredropacrossthesteamgenerator tubesatreducedtemperature operation.

Inaddition, theanalysesandevaluations performed supportamaximumaveragetubeplugginglevelof10%,withapeaksteamgenerator tubeplugginglevelof15%.Subsequently, asupplement toWCAP11902wasissuedastheWestinghouse inputforasecondsubmittal totheNRCtosummarize theadditional effortsperformed tosupportareratingofCookNuclearPlantUnit1andtoprovidepartofthesupportforaUnit2rerating(Reference 2).TheimpactofthisdocumentonCookNuclearPlantUnit1istosupportthelicensing ofapoweruprating(inadditiontosupporting therangeofoperating conditions described above)to3425MWtNSSS.Onlythereducedtemperature andpressureportionofthisprogramandassociated operational improvements havebeenapprovedandimplemented atthistime.AEPSCcurrently selectsthedesiredoperating conditions fromwithintherangeaddressed intheReratingProgramonacycle-to-cycle basis.Theeffortsdocumented inWCAP-11902 andSupplement arereferredtothroughout thisreportasthe"Rerating Program".

TheRCStemperatures oftheSGTPProgramwerechosentobewithintheboundsoftheReratingProgram.Thetwoprimappressurevaluesof2100psiaand2250psiawereevaluated.

ThemaximumaverageandpeakSGTPlevelwasincreased to30%.Acorresponding reduction inthermaldesignflow(TDF)andalsoa5%loopflowasymmetry werealsoevaluated.

BecausetherangeofNSSSparameters waschosentobewithintheboundsoftheReratingProgram,manyoftheanalysesperformed fortheReratingProgram(Reference 1and2)remainapplicable totheSGTPProgram.Uponapprovaloftheanalysesm31944-1w.wpf:1d441295 2.0-1 andevaluations inthisreport,AEPSCwillselectthedesiredoperating conditions fromwithintherangeaddressed intheSGTPProgramonacycletocyclebasis.References 1.WCAP-11902, "ReducedTemperature andPressureOperation forDonaldC.CookNuclearPlantUnit1Licensing Report",October19882.WCAP-11902, Supplement, "ReratedPowerandRevisedTemperature andPressureOperation forDonaldC.CookNuclearPlantUnits1and2Licensing Report,"September 1989m:11944.1w.wpf:1d441295 2.0-2 TABLE3.1-14SMALL-BREAK LOCACALCULATIONS k3%MAINSTEAMSAFETYVALVESETPOINTTOLERANCE ANALYSISRESULTSNOTRUMPPeakCladTemperature

('F)PeakCladTemperature Location(ft)PeakCladTemperature Time(sec)LocalZr/H,OReactionMaximum(%)LocalZr/H,OReactionLocation(ft)TotalZr/H,OReaction(%)RodBurstBurstandBlockagePenalty('F)TotalPeakCladTemperature

('F)12.012.0189040425.063.7512.012.0<1.0<1.0NoneNone117152068ReducedPressure, ReducedTemperature 3-Inch2-inch19511833Mainsteamsafetyvalvesetpointtolerance increasecaseat3250MWtcorepower.m%1944-2w.wpt:1d441195 3.1-35 3.0SAFETYEVALUATIONS/ANALYSES PERFORMED 3.1LOSSOFCOOLANTACCIDENT(LARGEBREAKANDSMALLBREAK)3.1.1LargeBreakLOCAAlargebreakLOCAanalysiswasperformed forDonaldC.CookNuclearPlantUnit1tosupportanincreaseinthesteamgenerator tubeplugginglevelto30%,whilemaintaining theoperational flexibility oftheplantbyanalyzing arangeofinitialRCStemperature conditions andtwodiscreteRCSpressures.

Thelargebreakanalysiswasperformed withthe1981versionoftheWestinghouse ECCSEvaluation ModelusingtheBASHcomputercode.Analysisassumptions included30%steamgenerator tubeplugging, operation atareactorpowerlevelof3250MWtwiththeRHRcross-tie valvesclosed,atotalpeakingfactorof2.15,andahotchannelenthalpyrisepeakingfactorof1.55.Safetyinjection flowswerebasedonpumpheaddegradation of15%forthehighheadsafetyinjection pumpsandRHRpumps,and10%forthecentrifugal pumps.TheEDGstarttimewasalsoincreased to30seconds.Theanalysisassumedarangeofoperating temperatures inordertojustifyplantoperation between609.1'Fand586.8'Finthehotlegsand543.5'Fand519.2'Finthecoldlegs.Thesetemperature rangesrepresent theUnit1powercapability parameters for30%peakuniformsteamgenerator tubepluggingdisplayed inTable2.1-1.InitialRCSpressurewasalsovariedtojustifyplantoperation at2100and2250psia.Afullspectrumbreakanalysiswasperformed fornominalRCSconditions (initialRCSpressureof2250psiaandhotlegtemperature of609.1'F)fromwhichthelimitingbreakdischarge coefficient wasdetermined.

Thelimitingbreakwasthenreanalyzed forthereducedhotlegtemperature andnominalRCSpressureof2250psia,andalsoforthenominalhotlegtemperature andRCSpressureof2100psia.Theabovecaseswereallanalyzedwithminimumsafetyinjection flow.Thelimitingbreakwasalsoanalyzedwithmaximumsafetyinjection flow.Thelimitingbreaksizewasdetermined tobeC~=0.4atthenominalhotlegtemperature (Type=609.1'F)andapressureof2100psiawithminimumsafetyinjection flow.Thepeakcladdingtemperature wascalculated tobe2164'Fwhichislessthanthe2200'limit in10CFR50.46.Adetaileddescription ofthelargebreakLOCAanalysisispresented below.Identification ofCausesandFreuenClassification ALOCAistheresultofapiperuptureoftheRCSpressureboundary.

Fortheanalysesreportedhere,amajorpipebreak(largebreak)isdefinedasarupturewithatotalcross-sectional areaequaltoorgreaterthan1.0ft'.Thiseventisconsidered anANSCondition IVevent,alimitingfault,inthatitisnotexpectedtooccurduringthelifetimeofCookNuclearPlantUnit1,butispostulated asaconservative designbasis.mal944-2w.wpf:1d~1195 3.1-1 TheAcceptance CriteriafortheLOCAaredescribed in10CFR50.46(10CFR50.46andAppendixKof10CFR50,1974-Reference 1)asfollows:1.Thecalculated peakfuelelementcladtemperature isbelowtherequirement of2200'F.2.Theamountoffuelelementcladdingthatreactschemically withwaterorsteamdoesnotexceed1percentofthetotalamountofZircaloyinthereactor.3.Thelocalized claddingoxidation limitof17percentisnotexceededduringorafterquenching.

4.Thecoreremainsamenabletocoolingduringandafterthebreak.5.Thecoretemperature isreducedanddecayheatisremovedforanextendedperiodoftime,asrequiredbythelong-lived radioactivity remaining inthecore.Thesecriteriawereestablished toprovideasignificant margininemergency corecoolingsystem(ECCS)performance following aLOCA.WASH-1400 (USNRC1975),Reference 2,presentsastudyinregardstotheprobability ofoccurrence ofRCSpiperuptures.

IISeuenceofEventsandSstems0erationsShouldamajorbreakoccur,depressurization oftheRCSresultsinapressuredecreaseinthepressurizer.

Thereactortripsignalsubsequently occurswhenthepressurizer lowpressuretripsetpointisreached.Asafetyinjection signalisgenerated whentheappropriate setpointisreached.Thesecountermeasures willlimittheconsequences oftheaccidentintwoways:Reactortripandboratedwaterinjection supplement voidformation incausingrapidreduction ofpowertoaresiduallevelcorresponding tofissionproductdecayheat.NocreditistakenintheLOCAanalysisfortheboroncontentoftheinjection water.However,anaverageRCS/sumpmixedboronconcentration iscalculated toensurethatthecoreremainssubcritical.

Inaddition, theinsertion ofcontrolrodstoshutdownthereactorisneglected inthelargebreakanalysis.

2.Injection ofboratedwaterprovidesforheattransferfromthecoreandpreventsexcessive cladtemperatures.

InthepresentWestinghouse design,thelargebreaksingiedailure isthelossofoneRHR(lowhead)pump.Thismeansthatcreditcouldbetakenfortwochargingpumps,twosafetyinjection pumps,andonelowheadpump.Thefollowing isadiscussion ofthemodelingm:51944.2w.wpt:1d441195 3.1-2 procedure fortheminimumsafeguards andtheflowspillingfromabreakofanECCSbranchinjection line{i.e.,thespillinglineassumptions).

Thecurrentprocedure forlargebreakanalysesassumesthatatleastonetrainofECCSisavailable fordeliveryofwatertotheRCS.AlthoughthesinglefailureisanRHRpump,onlyonepumpineachsubsystem isassumedtodelivertotheprimaryloops.However,bothEDGsareassumedtostartinthemodelingofthecontainment deckfansandsprays.Modelingfullcontainment heatremovalsystemsoperation isrequiredbyBranchTechnical PositionCSB6-1andisconservative forthelargebreakLOCA.Thechargingpumpstartsanddeliversflowthroughtheinjection lines{oneperloop)withonebranchinjection linespillingtothecontainment backpressure.

Tominimizedeliverytothereactor,thebranchlinechosentospillisselectedastheonewiththeminimumresistance.

Whenonesafetyinjection pumpandonelowheadresidualheatremovalpumpstart,flowisdelivered totheRCSthroughtheaccumulator injection lines.Again,oneline,withtheminimumresistance, isassumedtospilltocontainment backpressure.

Inaddition, thesafetyinjection pumpandlowheadRHRpumpperformance curvesweredegradedby15%.Forthechargingpumps,theperformance cuwesweredegradedby10%anda25gpmflowimbalance wasassumed.Therefore, inthelargebreakECCSanalysisperformed byWestinghouse, singlefailureisconservatively accounted forviathelossofanECCStrain,andthespillingoftheminimumresistance injection linedespitefullcontainment activeheatremovalsystemoperation

{i.e.,twoEDGs)~Thetimesequenceofeventsfollowing alargebreakLOCAispresented inTable3.1-1.Beforethebreakoccurs,theunitisinanequilibrium condition; thatis,theheatgenerated inthecoreisbeingremovedviathesecondary system.Duringblowdown, heatfromfissionproductdecay,hotinternals andthevessel,continues tobetransferred tothereactorcoolant.Atthebeginning oftheblowdownphase,theentireRCScontainssubcooled liquidwhichtransfers heatfromthecorebyforcedconvection withsomefullydeveloped nucleateboiling.Afterthebreakdevelops, thetimetodeparture fromnucleateboilingiscalculated, consistent withAppendixKof10CFR50(Reference 1).Thereafter, thecoreheattransferisunstable, withbothnucleateboilingandfilmboilingoccurring.

Asthecorebecomesuncovered, bothturbulent andlaminarforcedconvection andradiation areconsidered ascoreheattransfermechanisms.

TheheattransferbetweentheRCSand.thesecondary systemmaybeineitherdirection, depending ontherelativetemperatures.

Inthecaseofcontinued heatadditiontothesecondary system,thesecondary systempressureincreases andthemainsteamsafetyvalvesmayactuatetolimitthepressure.

Makeupwater.to.the secondary sideisautomatically providedbytheauxiliary feedwater system.Thesafetyinjection signalactuatesafeedwater isolation signalwhichisolatesnormalfeedwater flowbyclosingthemainfeedwater isolation mA1944-2w.wpf:1 d4411953.1-3 valves,andalsoinitiates e'mergency feedwater flowbystartingtheauxiliary feedwater pumps.Thesecondary flowaidsinthereduction ofRCSpressure.

WhentheRCSdepressurizes to600psia,theaccumulators begintoinjectboratedwaterintothereactorcoolantloops.Theconservative assumption ismadethataccumulator waterinjectedbypassesthecoreandgoesoutthroughthebreakuntilthetermination ofbypass.Thisconservatism isagainconsistent withAppendixKof10CFR50.Sincelossofoffsitepower(LOOP)isassumed,theRCPsareassumedtotripattheinception oftheaccident.

Theeffectsofpumpcoastdown areincludedintheblowdownanalysis.

Theblowdownphaseofthetransient endswhentheRCSpressure{initialvalueswithuncertainty assumedtobe2317psiaor2033psia)fallstoavalueapproaching thatofthecontainment atmosphere.

Priortoorattheendoftheblowdown, themechanisms thatareresponsible fortheemergency corecoolingwaterbypassing thecorearecalculated nottobeeffective.

Atthistime{calledend-of-bypass) refillofthereactorvessellowerplenumbegins.Refilliscompleted whenemergency corecoolingwaterhasfilledthelowerplenumofthereactorvessel,whichisboundedbythebottomofthefuelrods(calledbottomofcorerecoverytime).Therefloodphaseofthetransient isdefinedasthetimeperiodlastingfromtheend-of-refill untilthereactorvesselhasbeenfilledwithwatertotheextentthatthecoretemperature risehasbeenterminated.

Fromthelatterstageofblowdownandthenthebeginning-of-ref lood,thesafetyinjection accumulator tanksrapidlydischarge boratedcoolingwaterintotheRCS,contributing tothefillingofthereactorvesseldowncomer.

Thedowncomer waterelevation headprovidesthedrivingforcerequiredfortherefloodingofthereactorcore.Thelowheadandhighheadsafetyinjection pumpsaidinthefillingofthedowncomer andsubsequently supplywatertomaintainafulldowncomer andcompletetherefloodingprocess.Continued operation oftheECCSpumpssupplieswaterduringlong-term cooling.Coretemperatures havebeenreducedtolong-term steadystatelevelsassociated withthedissipation ofresidualheatgeneration.

Afterthewaterleveloftherefueling waterstoragetank(RWST)reachesaminimumallowable value,coolantforlong-term coolingofthecoreisobtainedbyswitching tothecoldlegrecirculation phaseofoperation inwhichspilledboratedwaterisdrawnfromtheengineered safetyfeatures(ESF)containment sumpsbythelowheadsafetyinjection (residual heatremoval)pumpsandreturnedtotheRCScoldlegs.Thecontainment spraysystemcontinues tooperatetofurtherreducecontainment pressure.

Approximately 12hoursaftertheinitiation oftheLOCA,theECCSisrealigned tosupplywatertotheRCShotlegsinordertocontroltheboricacidconcentration inthereactorvessel.Long-term coolingincludeslong-term criticality control.Griticality controlisachievedbydetermining theRWSTandaccumulator concentration necessary tomaintainsubcriticality withoutcreditforRCCAinsertion.

Thenecessary RWSTandaccumulator concentration isam:$1944.2w.wpf:1d~1195 3.1-4 functionofeachcoredesignandischeckedeachcycle.ThecurrentTechnical Specification valueis2400ppmto2600ppmboron(Reference 3).CoreandSstemPerformance Mathematical Model:Therequirements ofanacceptable ECCSevaluation modelarepresented inAppendixKof10CFR50(FederalRegister1974),Reference 1.LargeBreakLOCAEvaluation ModelTheanalysisofalargebreakLOCAtransient isdividedintothreephases:(1)blowdown, (2)refill,and(3)reflood.Therearethreedistincttransients analyzedineachphase,including thethermal-hydraulic transient intheRCS,thepressureandtemperature transient withinthecontainment, andthefuelandcladtemperature transient ofthehottestfuelrodinthecore.Basedontheseconsiderations, asystemofinterrelated computercodeshasbeendeveloped fortheanalysisoftheLOCA.Adescription ofthevariousaspectsoftheLOCAanalysismethodology isgivenbyBordelon, Massie,andZordan(Reference 4).Thisdocumentdescribes themajorphenomena modeled,theinterfaces amongthecomputercodes,andthefeaturesofthecodeswhichensurecompliance withtheAcceptance Criteria.

TheSATAN-VI,

WREFLOOD, BASHandLOCBARTcodes,whichareusedintheLOCAanalysis, aredescribed indetailbyBordelonetal.(1974)"';

Kellyetal.(Reference 6);Youngetal.(Reference 7);andBordelonetal.(Reference 4).Codemodifications arespecified inReferences 8,9,10and11.ItisnotedthattheWREFLOODcode,whichwaspreviously usedtocalculate theRCSbehaviorduringvessellowerplenumrefill,hasbeenreplacedbytheREFILLcodeasreportedinReference 18.TheREFILLcodeisidentical tothesectionoftheWREFLOODcodethatmodeledtherefillphase.Thesecodesassessthecoreheattransfergeometryanddetermine ifthecoreremainsamenabletocoolingthroughout andsubsequent totheblowdown, refill,andrefloodphasesoftheLOCA.TheSATAN-Vlcomputercodeanalyzesthethermal-hydraulic transient intheRCSduringblowdownandtheREFILLcomputercodecalculates thistransient duringtherefillphaseoftheaccident.

TheBASHcodeisusedtodetermine thesystemresponseduringtherefloodphaseofthetransient.

TheLOTICcomputercode,described byHsiehandRaymundinWCAP-8355 andWCAP-8345 (Reference 12),calculates thecontainment pressure'ransient.

Thecontainment pressuretransient isinputtoBASHforthepurposeofcalculating therefloodtransient.

TheLOCBARTcomputercodecalculates thethermaltransient ofthehottestfuelmA1944.2w.wpl:1d~1 1953.1-5

temperature of609.1'FandRCSpressureof2100psia.Thecasesanalyzedareidentified inTable3.1-1.Thebasesusedtoselectthenumerical valuesthatareinputparameters totheanalysishavebeenconservatively determined fromextensive sensitivity studies(Westinghouse 1974(Reference 14);Salvatori 1974(Reference 15);Johnson,Massie,andThompson1975(Reference 16).Inaddition, therequirements ofAppendixKregarding specificmodelfeaturesweremetbyselecting modelswhichprovideasignificant overallconservatism intheanalysis.

Theassumptions whichweremadepertaintotheconditions ofthereactorandassociated safetysystemequipment atthetimethattheLOCAoccurs,andincludesuchitemsasthecorepeakingfactors,thecontainment

pressure, andtheperformance oftheECCS.Decayheatgenerated throughout thetransient isalsoconservatively calculated.

Anotherinputparameter thataffectsLOCAanalysisresultsistheassumedaxialpowershapeatthebeginning oftheaccident.

LargebreakLOCAanalyseshavebeentraditionally performed usingasymmetric, choppedcosineaxialpowershape.Recentcalculations haveshownthattherewasapotential fortop-skewed powerdistributions toresultinpeakcladdingtemperatures (PCT)greaterthanthosecalculated withachoppedcosineaxialpowerdistribution.

Westinghouse hasdeveloped aprocess,whichwasappliedtotheCycle13and14reloadsforCookNuclearPlantUnit1,thatreasonably ensuresthatthecosineremainsthelimitingpowerdistribution, bydefiningappropriate powerdistribution surveillance data.Thisprocess,calledthepowershapesensitivity model(PSSM),isdescribed inatopicalreport(WCAP-12909-P),

Reference 19,andfurtherclarified inET-NRC-91-3633, Reference 20,whicharecurrently underNRCreview.Withimplementation ofthePSSMinthereloaddesignprocess,topskewedaxialpowerdistributions thatarepotentially morelimitingthanthepowerdistribution usedintheECCSanalysisarereasonably precluded fromoccurring bythedesignandsurveillance dataprovidedtomonitorthepowerdistribution.

AmeetingwasheldattheWestinghouse Licensing OfficeinBethesdaonDecember17,1981,betweenmembersoftheU.S.NuclearRegulatory Commission andmembersoftheWestinghouse NuclearSafetyDepartment todiscusstheimpactofmaximumsafetyinjection onthelargebreakECCSanalysisonagenericbasis.Furtherdiscussion ofthisissueisprovidedinaletterfromE.P.Rahe,ManagerofWestinghouse NuclearSafetyDepartment, toRobertL.TedescooftheU.S.NuclearRegulatory Commission (Reference 17).Abriefdescription ofthisissueisgivenbelow.Westinghouse ECCSanalysescurrently assumeminimumsafeguards forthesafetyinjection flow,whichminimizes theamountofflowtotheRCSbyassumingmaximuminjection lineresistances, degradedECCSpumpperformance, andthelossofoneRHRpumpasthemostlimitingsinglefailure.Thisisconservatively modeledasalossofonetrainofsafetyinjection, including RHRpump,safetyinjection pumpandcentrifugal chargingpump.Bothcontainment spraypumpsareassumedoperable.

Thisisthelimitingsinglefailureassumption whenoffsitemA1944-2w.wpf:1d441195 3.1-7 powerisunavailable formostWestinghouse plants.However,forsomeWestinghouse plants,thecurrentnatureoftheAppendixKECCSevaluation modelsissuchthatitmaybemorelimitingtoassumethemaximumpossibleECCSflowdelivery.

Inthatcase,maximumsafeguards whichassumeminimuminjection lineresistances, enhancedECCSpumpperformance, andnosinglefailure,resultinthehighestamountofflowdelivered totheRCS.Therefore, theworstbreakforCookNuclearPlantUnit1(C~=0.4fornominalhotlegtemperature of609.1'FandRCSpressureof2100psia)wasreanalyzed assumingmaximumsafeguards.

Results:BasedontheresultsoftheLOCAsensitivity studies(Westinghouse 1974(Reference 14);Salvatori 1974(Reference 15);Johnson,Massie,andThompson1975(Reference 16)thelimitinglargebreakwasfoundtobethedouble-ended coldleg(DECL)guillotine.

Therefore, onlytheDECLguillotine breakisconsidered inthelargebreakECCSperformance analysis.

Calculations wereperformed forarangeofMoodybreakdischarge coefficients.

Theresultsofthesecalculations aresummarized inTable3.1-1.Thecontainment datausedtogeneratetheLOTICbackpressure transient areshowninTable3.1-3.Themassandenergyreleasedatausedforthelimitingminimumsafeguards caseareshowninTable3.1A.Nitrogenreleaseratestothecontainment aregiveninTable3.1-5.Figures3.1-1athrough3.1-19presenttheresultsofthecasesanalyzedforthelargebreakLOCA.Thealphadesignation inthefigurenumbercorresponds tothecasesasdescribed inTable3.1-1.Fiures3.1-1ato1fThesystempressureshownisthecalculated corepressure.

Fiures3.1-2ato2fTheflowratefromthebreakisplottedasthesumofbothendsoftheguillotine break.Fiures3.1-3ato3fThecorepressuredropshownisfromthelowerplenum,nearthecore,totheupperplenumatthecoreoutlet.Fiures3.1-4ato4fThecoreflowisshownduringtheblowdownphaseofthetransient.

Fiures3.1-5ato5fTheaccumulator flowduringblowdownisplottedasthesumofthatinjectedintotheintactcoldlegs.mal944-2w.wpf:1d~1195 3.1-8 Fiures3.1-6ato6fThecoreanddowncomer collapsed liquidwaterlevel,andthecorequenchfrontareplottedduringtherefloodphaseofthetransient.

Fiures3.1-7ato7fThecoreinletflowisshownasitiscalculated duringtherefloodphase.Fiures3.1-8ato8fThetotalaccumulator andpumpedECCSflowinjectedintotheintactcoldlegsduringrefloodisshown.Fiures3.1-9ato9fTheintegralofthecoreinletflowduringrefloodascalculated withBASHisplotted.Fiures3.1-10ato10fThemassfluxisplottedatthehotspot(thenodewhichproducedthepeakcladtemperature) onthehotrod.Fiures3.1-11ato11fTheheattransfercoefficient isplottedatthehotspotonthehotrod.Fiures3.1-12ato12fFiures3.1-13ato13fThevaportemperature atthehotspotonthehotrodisplotted.4Thecladtemperature atthehotspotisshownforthehotrod.Fiure3.1-14Thecontainment pressuretransient usedintheanalysisisprovidedfortheminimumSlcase.Fiures3.1-15to18Thesefiguresshowtheheatremovalratesoftheheatsinksfoundintheloweranduppercompartment andtheheatremovalbythesumpandlowercompartment spray.Fiure3.1-19Thisfigureshowsthetemperature transients inboththeloweranduppercompartments ofcontainment.

AsshowninTable3.1-1,thelimitingcaseforCookNuclearPlantUnit1isCaseE(C,=0.4fornominalhotlegtemperature of609.1'FandRCSpressureof2100psia).Themaximumcladtemperature calculated foralargebreakis2164'F,whichislessthantheAcceptance Criterialimitof2200'F.Themaximumlocalmetal-water reactionis14.30percent,whichiswellbelowtheembrittlement limitof17percentasrequiredby10CFR50.46.Thetotalcoremetal-water reactionforallbreaksislessthanthe1percentcriterion of10CFR50.46.Thecladtemperature transient isterminated atatimewhenthe-coregeometryisstillamenabletocooling.Asaresult,thecoretemperature willcontinuetodropandtheabilitytoremovedecayheatgenerated inthefuelforanextendedperiodoftimewillbeprovided.

mA1944-2w.wpt:1dM1195 3.1-9 References 1."Acceptance CriteriaforEmergency CoreCoolingSystemforLightWaterCooledNuclearPowerReactors,"

t0CFR50.46andAppendixKoft0CFR50,FederalVI39M13.2.U.S.NuclearRegulatory Commission 1975,"ReactorSafetyStudy-AnAssessment ofAccidentRisksinU.A.Commercial NuclearPowerPlants,"WASH-1400, NUREG-75/014.3.Attachment 13toletter,M.P.Alexich,IMECo,toH.R.Denton,NRC,March26,1987,AEP:NRC:0916W.

4.Bordelon, F.M.;Massie,H.W.;andZordan,T.A."Westinghouse ECCSEvaluation Model-Summary,"

WCAP-8339, 1974.5.Bordelon, F.M.etal~,"SATAN-Vl Program;Comprehensive Space,TimeDependent AnalysisofLoss-of-Coolant,"

WCAP-8302 (Proprietary) andWCAP-8306 (Non-Proprietary),

1974.~~e06.Kelly,R.D.etal.,"Calculation ModelforCoreReflooding AfteraLoss-of-Coolant Accident(WREFLOOD Code),"WCAP-8170 (Proprietafy) andWCAP-8171 (Non-proprietary),

1974.7.Young,M.Y.etal,"The1981VersionoftheWestinghouse ECCSEvaluation ModelUsingtheBASHCode,"WCAP-10266-P-A Revision2(Proprietary),

1987.8.Rahe,E.P.(Westinghouse),

lettertoJ.R.Miller(USNRC),LetterNo.NS-EPRS-2679, November1982.9.Rahe,E.P.,"Westinghouse ECCSEvaluation Model,1981Version,"

WCAP-9920-P-A (Proprietafy Version),

WCAP-9221-P-A (Non-Proprietafy version),

Revision1,1981.10.Bordelon, F.M.,etal.,"Westinghouse ECCSEvaluation Model-Supplementary Information,"

WCAP-8471 (Proprietary) andWCAP-8472 (Non-proprietary),

1975.11.Thomas,C.O.,(NRC),"Acceptance forReferencing ofLicensing TopicalReportWCAP-10484(P)/1 0485(NP),

'SpacerGridHeatTransferEffectsDuringReflood,'"LettertoE.P.Rahe(Westinghouse),

June21,1984.m:11944.2w.wpf:1d~1195 3.1-10

12.Hsieh,T.,andRaymund,M.,"Long-Term IceCondenser Containment LOTICCodeSupplement 1,"WCAP-8355, Supplement 1,May1975,WCAP-8345 (Proprietary),

July1974.13."Westinghouse RevisedPADCodeThermalSafetyModel,"WCAP-8720, Addendum2(Proprietary) andWCAP-8785 (Non-propnetary).

14."Westinghouse ECCS-Evaluation ModelSensitivity Studies,"

WCAP-8341 (Proprietary) andWCAP-8342 (Non-proprietary),

1974.15.Salvatori, R.,"Westinghouse ECCS-PlantSensitivity Studies,"

WCAP-8340 (Proprietary) andWCAP-8356 (Non-proprietary),

1974.16.Johnson,W.J.;Massie,H.W.;andThompson, C.M."Westinghouse ECCS-FourLoopPlant(17x17)Sensitivity Studies,"

WCAP-8565-P-A (Proprietary) andWCAP-8566-A

{Non-proprietary),

1975.17.Rahe,E.P.(Westinghouse).

LettertoRobertL.Tedesco(USNRC),LetterNo.NS-EPR-2538, December1981.18.Liparulo, N.J.(Westinghouse),

lettertoW.T.Russel(USNRC),LetterNo.~~~NTD-NRC-94-4143, May23,1994.19.Stucker,D.L.etal.,"Westinghouse ECCSEvaluation Model:RevisedLargeBreakLOCAPowerDistribution Methodology,"

WCAP-12909-P (Proprietary) andWCAP-12935-NP (Non-Proprietary),

May1991.20.Tritch,S.R.(Westinghouse),

lettertoR.C.Jones(USNRC),LetterNo.ET-NRC-91-3633, October25,1991.m&1944-2w.wpf:1d441195 3.1-11 3.1.2SmallBreakLOCAAsmallbreakLOCAanalysishasbeenperformed fortheDonaldC.CookUnit1NuclearPlanttosupportanincreaseinthesteamgenerator tubeplugginglevelto30%,whilemaintaining theoperational flexibility oftheplantbydemonstrating thatthe10CFR50.46Acceptance CriteriacanbemetforarangeofinitialRCSpressureandtemperature conditions.

ThesmallbreakLOCAanalysiswasperformed withtheWestinghouse smallbreakLOCAECCSEvaluation ModelusingtheNOTRUMPcode""",including therecentchangesinAddendum2"'oincorporate modelingofsafetyinjection intothebrokenloopandtheCOSIcondensation model.Thekeyanalysisinputassumptions included30%peakuniformsteamgenerator tubeplugging, operation atareactorpowerlevelof3250MWtwiththeHHSIcross-tie discharge valvesclosed,atotalpeakingfactorof2.32,andahotchannelenthalpyrisepeakingfactorof1.55.Alsoincorporated intheanalysisareareducedhotassemblyaveragepowerandapowershapebasedonareducedaxialoffsetof+20%.Safetyinjection flowsarebasedonpumpheaddegradation of15%forthehighheadsafetyinjection pumpsand10%forthecentrifugal chargingpumps,andtheemergency dieselgenerator starttimewasincreased to30seconds.Theanalysiswasperformed inordertoboundplantoperation withintherangeofRCStemperatures specified intheUnit1powercapability parameters for30%uniformsteamgenerator tubeplugginginTable2.1-1,andatRCSpressures of2100and2250psia.Asinglebreaksizeanalysiswasperformed atthepreviously-limiting breaksizeofthreeinches.Thecalculation usedthereducedtemperature, reducedpressureoperating condition whichwaspreviously demonstrated tobethelimitingoperating condition forthesmallbreakanalysis.

Basedonanevaluation ofthebreakspectrumandtherangeofoperating conditions, itwasconcluded thattheanalyzedcasewouldremainboundingwithrespecttopeakcladtemperature.

Thepeakcladdingtemperature wascalculated tobe1443'Fwhichislessthanthe2200'Flimitin10CFR50.46.Adetaileddescription oftheanalysisispresented inthefollowing sections.

Sincetheanalysistosupport30%steamgenerator tubepluggingisanextension ofprevioussmallbreakLOCAanalysesperformed forCookNuclearPlantUnit1,thedescription alsoincludesadiscussion ofthepreviousanalyses.

TheseincludetheReratingProgramanalysescurrently intheFSARwhichwereperformed forareactorpowerlevelof3588MWt,andtheanalysisperformed forareactorpowerlevelof3250MWttosupportanincreaseinthemainsteamsafetyvalve(MSSV)setpointtolerance tok3%.m:11944.2w.wpt:1d~1195 3.1-12 3.1.2.1ReratingProgramAnalysisAnalysisofEffectsandConsequences MethodofAnalysisForloss-of-coolant accidents duetosmallbreakslessthanonesquarefoot,theNOTRUMPcomputercode"~'susedtocalculate thetransient depressurization oftheRCSaswellastodescribethemassandenthalpyofflowthroughthebreak.TheNOTRUMPcomputercodeisastate-of-the-art one-dimensional generalnetworkcodeincorporating anumberofadvancedfeatures.

Amongthesearecalculation ofthermalnon-equilibrium inallfluidvolumes,flowregime-dependent driftfluxcalculations withcounter-current floodinglimitations, mixtureleveltrackinglogicinmultiple-stacked fluidnodesandregime-dependent heattransfercorrelations.

TheNOTRUMPsmall-break LOCAemergency corecoolingsystem(ECCS)evaluation modelwasdeveloped todetermine theRCSresponsetodesignbasissmallbreakLOCAs,andtoaddressNRCconcernsexpressed inNUREG-0611, "GenericEvaluation ofFeedwater Transients andSmallBreakLoss-of-Coolant Accidents inWestinghouse-Designed Operating Plants."Thereactorcoolantsystemmodelisnodalized intovolumesinterconnected byflowpaths.Thebrokenloopismodelledexplicitly, whilethethreeintactloopsarelumpedintoasecondloop.Transient behaviorofthesystemisdetermined fromthegoverning conservation equations ofmass,energy,andmomentum.

Themultinode capability oftheprogramenablesexplicit, detailedspatialrepresentation ofvarioussystemcomponents which,amongothercapabilities, enablesapropercalculation ofthebehavioroftheloopsealduringaloss-of-coolantaccident.

Thereactorcoreisrepresented asheatedcontrolvolumeswithassociated phaseseparation modelstopermit.transient mixtureheightcalculations.

Detaileddescriptions oftheNOTRUMPcodeandtheevaluation modelareprovidedinReferences 1and2.Safetyinjection systemsconsistofgaspressurized accumulator tanksandpumpedinjection systems.Minimumemergency corecoolingsystemavailability isassumedfortheanalysis.

Assumedpumpedsafetyinjection characteristics asafunctionofRCSpressureusedasboundaryconditions intheReratingProgramanalysisareshowninFigure3.1-20andinTable3.1-6.Thesafetyinjection flowratespresented arebasedonpumpperformance curvesdegraded10percentfromthedesignheadandareconsistent withclosureofthehighheadsafetyinjection systemcross-tie valve.TheeffectofflowfromtheRHRpumpsisnotconsidered inthesmallbreakanalysessincetheirshutoffheadislowerthantheRCSpressureduringthetimeportionofthetransient considered here.Safetyinjection isdelayed27secondsaftertheoccurrence oftheinjection signaltoaccountfordieselgenerator startupandemergency powerbusloadingincaseofalossofoffsitepowercoincident withanaccident.

mA1944-2w.wpf:1d~1195 3.1-13 Peakcladtemperature calculations areperformed withtheLOCTA-IV"'ode usingtheNOTRUMPcalculated corepressure, fuelrodpowerhistory,uncovered coresteamflowandmixtureheightasboundaryconditions.

Figure3.1-21depictsthehotrodaxialpowershapeusedtoperformthesmallbreakanalysisfortheReratingProgram.Thisshapewaschosenbecauseitrepresents adistribution withpowerconcentrated intheupperregionsofthecore.Suchadistribution islimitingforsmall-break LOCAsbecauseitminimizes coolantlevelswell,whilemaximizing vaporsuperheating andfuelrodheatgeneration intheuncovered elevations.

ThesmallbreakLOCAanalysisassumesthecorecontinues tooperateatfullratedpoweruntilthecontrolrodsarecompletely inserted.

ResultsThissectionpresentsresultsofthelimitingbreakanalysis(asdetermined bythehighestcalculated peakfuelrodcladtemperature) forarangeofbreaksizesandRCSpressures andtemperatures forareactorpowerlevelof3588MWt.Thelimitingbreakwasfoundtobea3-inchdiametercoldlegbreakinitiated atreducedRCSpressureandtemperature conditions.

Themaximumtemperature attainedduringthetransient was2122'F.Alistofinputassumptions usedintheReratingProgramanalysisforreducedpressureandtemperature conditions isprovidedinTable3.1-7.Theresultsofathreebreakspectrumanalysisperformed atreducedRCSpressureandtemperature conditions aresummarized inTable3.1-8,whilethekeytransient eventtimesarelistedinTable3.1-9.Figures3.1-22through3:1-29showthelimitingthree-inch breaktransient, respectively:

-RCSpressure,

-Coremixturelevel,-Peakcladtemperature,

-Coreoutletsteamflow,-Hotspotrodsurfaceheattransfercoefficient,

-Hotspotfluidtemperature,

-Coldlegbreakmassflowrate,and-Safetyinjection massflowrate.Duringtheinitialperiodofthesmall-break transient, theeffectofthebreakflowisnotstrongenoughtoovercometheflowmaintained bythereactorrecirculation coolingpumpsastheycoastdown.Normalupwardflowismaintained throughthecoreandcoreheatisadequately removed.Atthelowheatgeneration ratesfollowing

shutdown, thefuelrodscontinuetobewellcooledaslongasthecoreiscoveredbyatwo-phase mixturelevel~Fromthecladtemperature transient forthe3-inchbreakcalculation showninFigure3.1-24,itisseenthatthepeakcladtemperature occursnearthetimeatwhichthecoreismostdeeplyuncovered whenthetopofthecoreissteamcooled.Thistimeisalsoaccompanied bythehighestvaporsuperheating abovethemixturelevel.Acomparison ofthetotalbreakflowtocontainment showninFigure3.1-28tothesafetyinjection flowrateshowninFigure3.1-29showsthatatm31944-2w.wpf:1d~

11953.1-14 thetimethetransient wasterminated, thesafetyinjection flowbeingdelivered totheRCSexceededthemassflowoutthebreak.Althoughtheinnervesselcoremixturelevelhasnotyetcoveredtheentirecore,thereisnolongeraconcernofexceeding the10CFR50.46criteriasincethepressureisgradually decayingandthereisanetmassinventory gain.AstheRCSinventory continues togradually

increase, thecoremixturelevelwillcontinuetoincreaseandthefuelcladtemperatures willcontinuetodecline.ReratinProramBreakSectrumCasesStudiesdocumented inReference 4determined thatthelimitingsmall-break sizeoccurredforbreakslessthan10inchesindiameter.

Toensurethatthe3-inchdiameterbreakwaslimitingforthereducedtemperature andpressureRCSconditions, calculations werealsorunwithbreaksof2inchesand4inches.Theresultsofthesecalculations areshownintheResultsTable3.1-8andSequenceofEventsTable3.1-9.Plotsofthefollowing parameters areshowninFigures3.1-30through37forthe2-inchbreak,andFigures3.1-38through43forthe4-inchbreak.-RCSpressure,

-Coremixturelevel,-Peakcladtemperature,

-Coreoutletsteamflow,-Hotspotrodsurfaceheattransfercoefficient,

-Hotspotfluidtemperature,

-Coldlegbreakmassflowrate,(forthe2-inchcaseonly),and-Safetyinjection massflowrate(forthe2-inchcaseonly).AsseeninTable3.1-8,themaximumcladtemperatures werecalculated tobelessthanthatforthe3-inchbreak.Additional ReratinProramAnalsesCalculations werealsoperformed forCookNuclearPlantUnit1withtheNOTRUMP"~'ndLOCTA-IV"'odes toexaminetheinfluence ofinitialloopfluidoperating temperatures andoperating pressures onsmallbreakLOCApeakcladtemperature.

Theseadditional analysesconfirmed thatthemostlimitingPCTresultwasthatfromthereducedtemperature andpressurelimiting3-inchdiameterbreakdescribed previously.

Tosupportoperation oftheGookNuclearPlantUnit1atRCSpressures of2100psiaand2250psiaforarangeofloopoperating temperatures, twoadditional analyseswereperformed.

Calculations wereperformed fora3-inchdiameterbreakforaninitialRCSpressureof2250psiaatinitialloopfluidoperating temperatures corresponding toT,~programsetpoints of547'Fand578'F.Theresultsofthesecalculations areshownintheResultsTable3.1-10m31944-2w.wpf:1d441195 3.1-15 andtheSequenceofEventsTable3.1-11.Plotsofthefollowing parameters areshowninFigures3.1~through51forthereducedtemperature andhighpressurecase,andFigures3.1-52through59foithehightemperature andhighpressurecase.-RCSpressure,

-Coremixturelevel,-Peakcladtemperature,

-Coreoutletsteamflow,-Hotspotrodsurfaceheattransfercoefficient,

-Hotspotfluidtemperature,

-Coldlegbreakmassflowrate,and-Safetyinjection massflowrate.AsseeninTable3.1-10,themaximumcladtemperatures werecalculated tobelessthanthatforthe3-inchbreakinitiated atreducedtemperature andpressureconditions.

NUREG-0737<",

SectionII.K.3.31, requiredplant-specific smallbreakLOCAanalysisusinganEvaluation ModelrevisedperSectionII.K.3.30.

Inaccordance withNRCGenericLetter83-65"',genericanalysesusingNOTRUMP"~'ere performed andarepresented inWCAP-11145"'.Thoseresultsdemonstrate thatinacomparison ofcoldleg,hotlegandpumpsuctionlegbreaklocations, thecoldlegbreaklocationislimiting.

3.1.2.2MainSteamSafetyValveSetpointTolerance Relaxation AnalysisAdditional smallbreakLOCAanalyseswereperformed atareactorpowerlevelof3250MWttosupportanincreaseintheMSSVliftsetpointtolerance from~1%to~3%.Priortotheanalysisperformed tosupportMSSVtolerance relaxation, thismodeofoperation wassupported byanevaluation limitingcorepowerto3250MWt.TheMSSVanalyseswere.performed foroperation withtheHHSIcross-tie valvesclosedandassuminga25gpmchargingpumpflowimbalance.

Thisresultedinareduction inthechargingpumpflow,andthusareduction inthetotalsafetyinjection flowraterelativetotheReratingProgramanalysis.

Thelimiting3-inchbreakforreducedpressureandreducedtemperature operating conditions wasanalyzed, sincetheReratingProgramanalysisdemonstrates thatthiscaseresultsinthemostlimitingcladtemperature.

Sincethebasisforthelimitingcasedetermination remainsvalid,itwasnotnecessary toanalyzethefullspectrumofcases.However,ananalysiswasalsoperformed fora2-inchbreaksinceareduction insafetyinjection flowratecanpotentially shiftthelimitingbreaktoasmallerbreaksize.Theanalysisforthe2-inchbreakconfirmed thatthelimitingbreakdidnotshifttoasmallerbreaksize.-m:51944-2w.wpf:1d441195 3.1-16 Alistoftheplantinputparameters forthet3%MSSVsetpointtolerance analysisisprovidedinTable3.1-12.Theresultsofthelimiting3-inchbreakanalysisarepresented intheSequenceofEventsTable3.1-13andtheResultsTable3.1-14.Resultsofthenon-limiting 2-inchcasearealsoprovidedinTables3.1-13and3.1-14.Plotsofthefollowing parameters forthe3-inchbreakanalysisareshowninFigures3.1-60through3.1-67,andforthe2-inchbreakinFigures3.1-69through3.1.2-76:

-RCSpressure-Coremixturelevel-Peakcladtemperature

-Coreoutletsteamflow-Hotspotrodsurfaceheattransfercoefficient

-Hotspotfluidtemperature

-Coldlegbreakmassflowrate,and-Safetyinjection massflowrateFigure3.1-68containsthepowershapeforacorepowerlevelof3250MWtwhichisapplicable tobothcases.The3-inchbreakwithHHSIcross-ties closed,initiated atreducedpressureandtemperature operating conditions andacorepowerlevelof3250MWt,represents thelicensing basissmallbreakanalysisforanincreased MSSVsetpointtolerance ofk3%.Application ofaburstandblockagepenaltyresultedinapeakcladtemperature of2068'F,whichwaslessthanthe2200'Flimit.3.1.2.330%SteamGenerator TubePluggingAnalysisAnadditional smallbreakLOCAanalysishasbeenperformed tosupportanincreaseinsteamgenerator tubeplugginglevelfrom15%toamaximumof30%ineachsteamgenerator.

Theanalysiswasperformed inordertoboundplantoperation between609.1'Fand586.8'Finthehotlegsand543.5'Fand519.2'Finthecoldlegs.Thesetemperature rangesaredefinedintheUnit1powercapability parameters for30%peakuniformsteamgenerator tubepluggingdisplayed inTable2.1-1.Theanalysisalsosupportsplantoperation atRCSpressures of2100and2250psia.Theanalysiswasperformed forthelimiting3-inchbreakwithHHSIcross-tie valvesclosedatreducedpressureandreducedtemperature operating conditions andacorepowerlevelof3250MWt,whichwaspreviously demonstrated toresultinthemostlimitingcladtemperature.

Anevaluation ofthebasisforthelimitingcasedetermination wasperformed anditwasconcluded thatitwasnotnecessary toperformafullbreakspectrumforthiscase.mh1944.2w.wpf:1d441195 3.1-17

Theanalysisincorporates a20secondincreaseinemergency dieselgenerator startingdelayto30seconds,whichresultsinatotalsafetyinjection delayof47secondsaftertheoccurrence oftheinjection signal.Thesafetyinjectioh flowratesusedintheanalysisincludea5%increaseinhighheadsafetyinjection pumpdegradation, foratotalof15%degradation.

Forthehighheadchargingpumps,theperformance curvedegradation remainedat10%anda25gpmflowimbalance wasassumed.Theanalysisalsoincludesareduction inthemaximumaxialoffsetfrom+30%to+20%andareduction inthemaximumhotassemblypeakingfactorfrom1.433to1.38,withacorresponding changeintheaxialpowershapeusedintheanalysis.

TheuseoftherevisedaxialoffsetandhotassemblypowerfactorinthesmallbreakLOCAanalysisareconsistent withthecurrentcoredesignandoperation limits.Anevaluation ofupto5%RCSloopflowasymmetry wasalsoperformed tosupporttheanalysis.

Alistoftheplantinputparameters forthe30%SGTPanalysisisprovidedinTable3.1-15.Previously, safetyinjection intothebrokenloopwasnotmodeledintheWestinghouse smallbreakLOCAanalysessinceitwasassumedthattheadditional safetyinjection wouldbeabenefit.Becauserecentstudieshaveshownthattheresponsetobrokenloopsafetyinjection canresultinanincreaseinthecalculated PCT,modelingofsafetyinjection intothebrokenloophasnowbeenincorporated intotheNOTRUMPsmallbreakevaluation model.Amorerealistic modelforcondensation ofsteambypumpedsafetyinjection basedondatafromtheCOSItestfacilityhasalsobeenincorporated, whichprovidesabenefitlargerthanthepenaltyforsafetyinjection inthebrokenloop.Themethodology formodelingsafetyinjection tothebrokenloopinsmallbreakLOCAanalysesandapplication oftheCOSIcondensation modelarepresented intheNOTRUMPSmallBreakECCSEvaluation Model,Addendum2"'.Theanalysisfor30%steamgenerator tubepluggingmodeledthepumpedsafetyinjection andanaccumulator inthebrokenloop,andusedthemorerealistic COSIcondensation modelinReference 8.Theresultsofthe3-inchbreakanalysisarepresented intheSequenceofEventsTable3.1-16andtheResultsTable3.1-17.Plotsofthefollowing parameters forthe3-inchbreakanalysisareshowninFigures3.1-77through3.1-85:-RCSpressure-Coremixturelevel-HotSpotCladTemperature

-CoreOutletSteamFlow-Hotspotrodsurfaceheattransfercoefficient

-Hotspotfluidtemperature

-Coldlegbreakmassflowrate-Brokenloopsafetyinjection massflowrate,and-Lumpedintactloopsafetyinjection massflowrateFigure3.1-86containsthepowershapeusedintheanalysis.,

mal944-2w.wpf:1d441195 3.1-18

Duetothemodelingofsafetyinjection inthebrokenloopwiththeCOSIcondensation modelchange,inconjunction withthereducedpeakingfactors,thePCTforthe30%steamgenerator tubepluggingsmallbreakLOCAanalysisislowerthanforprevioussmallbreakanalyses.

Becausenorodburstwascalculated tooccurandthebeginning oflifecalculated peakcladtemperature islowenoughtoprecludeaZr/H,Oreactiontemperature excursion following burst,noburstandblockagepenaltyisapplied.Theresulting totalpeakcladtemperature of1443'Fislessthanthe2200'Flimit.SmallBreakLOCAAnalysisConclusions Analysespresented inthissectionshowthatthehighheadportionoftheemergency corecoolingsystem,togetherwiththeaccumulators, providesufficient corefloodingtokeepthecalculated peakcladtemperatures belowrequiredlimitsof10CFR50.46.Hence,adequateprotection isaffordedbytheemergency corecoolingsystemintheeventofasmallbreakloss-of-coolant

accident, References 1.Meyer,P.E.,"NOTRUMP-ANodalTransient SmallBreakandGeneralNetworkCode,"WCAP-10079-P-A, August1985.2.Lee,N.et.al.,"Westinghouse SmallBreakECCSEvaluation ModelUsingTheNOTRUMPCode,"WCAP-10054-P-A, August1985.3.Bordelon, F.M.,etal~,"LOCTA-IV Program:Loss-of-Coolant Transient Analysis,"

WCAP-8305,June1974,WCAP-8301, (Proprietary),

June1974.4."ReportonSmallBreakAccidents forWestinghouse NSSSSystem,"Vols.ItoIII,WCAP-9600,June1979.5."Clarification ofTMIActionPlanRequirements,"

NUREG-0737, November1980.6.NRCGenericLetter83-35fromD.G,Eisenhut, "Clarification ofTMIActionPlanItemII.K.3.31,"

November2,1983.7.Rupprecht, S.D.,et.al.,"Westinghouse SmallBreakLOCAECCSEvaluation ModelGenericStudyWiththeNOTRUMPCode;"WCAP-11145-P-A, October1986.8.Thompson, C.M.,et.al.,"Addendum totheWestinghouse SmallBreakECCSEvaluation ModelUsingtheNOTRUMPCode:SafetyInjection IntOtheBrokenLoopandCOSICondensation Model",WCAP-10054-P, Addendum2(Proprietary) andWCAP-10081-NP, Addendum2(Non-Proprietary)),

August1994.mal944-2w.wpf:1d441195 3.1-19 TABLE3.1-1LARGEBREAKLOCARESULTSPeakCladTemperature

('F)PeakCladLocation{ft)LocalZr/H,OReaction(Max%)LocalZr/H,OLocation(ft)TotalZr/H,OReaction(%)CaseAC0=0.4THOT609.1'FP=2250psiaMin.Sl20695.757.595.75CaseBCo=0.6THOT609.1'FP=2250psiaMin.Sl19936.258.196.00<1.0CaseCCO=0.8THQT609.1'FP=2250psiaMin.Sl19656.256,626.00<1.0CaseDC0=0.4THOT586.8'FP=2250psiaMin.Sl2036'.008.456.00(1.0CaseEC0=0.4THOT609.1'FP=2100psiaMin.Sl21646.2514.306.25<1.0CaseFC,=0.4THOT609.1'FP=2100psiaMax.Sl21496.2512.016.25<1.0HotRodBurstTime(s)43.641.845.746.542.042.0HotRodBurstLoc.(ft)5.756.006.006.006.256.25m31944.2w.wpf:

Id4411953.1-20 A

TABLE3.1-1(continued)

LARGEBREAKLOCARESULTSStartReactorTripSignalSafetyInjection SignalAccumulator Injection CaseAC,=0.4THQT609.1'FP=2250psiaMin.Sl0.00.644.8018.70CaseBCo=0.6THOT-609.1'FP=2250psiaMin.Sl0.00.644.6013.90CaseCCo=0.8THOT609.1'FP=2250psiaMin.Sl0.00.634.5011.60CaseDCO=0.4THOT-586.8'FP=2250psiaMin.Sl,0.00.554.4017.80CaseECo=0.4THQT-609.1'FP=2100psiaMin.SI0.0~0.494.1018.70CaseFCo=0.4HOT609.1'FP=2100psiaMax.SI0.00.494.1018.70EndofBlowdown40.7531.7728.0540.6139.9639.96PumpInjection BottomofCoreRecovery51.8054.3051.6044.6051.5041.8051.5055.3051.1054.2051.1054.00Accumulator Empty69.0962.3048.7570.0968.9669.78m%1944.2w.wpf:1d441195 3.1-21 TABLE3.1-2PLANTINPUTPARAMETERS USEDINLARGEBREAKLOCAANALYSISCorePower(MWt)PeakLinearPower(kW/ft)TotalCorePeakingFactor,F~HotChannelEnthalpyRiseFactor,F,MaximumAssemblyAveragePower,P-FuelAssemblyArraySteamGenerator TubePluggingLevel(%)Accumulator WaterVolume(ft'/tank)

Accumulator TankVolume(ft'/tank)

MinimumAccumulator GasPressure(psia)Accumulator WaterTemperature

('F)Refueling WaterStorageTankTemperature

('F)ThermalDesignFlowrate(gpm/loop)

RCSLoopAverageTemperature

('F)NominalInitialRCSPressure(psia)NominalSteamPressure(psia)SafetyInjection DelayTime(sec)RHRPumpHeadDegradation

(%)HHSIPumpHeadDegradation

(%)ChargingPumpHeadDegradation

(%)ChargingPumpFlowImbalance (gpm)RHRCross-Tie ValvePosition102%of3250102%of14.4342.151.551.3815X15OFA30946135060010070-10583,200553.0and576.32100and2250595and7494715151025Closedm%1944.2w.wpf:

1d~12953.1-22 NETFREEVOLUMETABLE3.1-3LARGEBREAKCONTAINMENT DATA(ICECONDENSER CONTAINMENT)

(Includes Distribution BetweenUpper,Lower,andDead-Ended Compartments)

InitialConditions PressureMaximumTemperature fortheUpper,Lower,andDead-Ended Compartments MinimumTemperature fortheUpper,Lower,andDead-Ended Compartments UC746,829ft'C249446ft'E116,168ftIC163,713ft'4.7psiaUC100'FLC120'FDE120'FUC60'FLC60'FDE60'FRWSTTemperature Temperature OutsideContainment InitialSprayTemperature SpraySystemRunoutFlowforaSprayPumpNumberofSprayPumpsOperating Post-Accident Initiation ofSpraySystemDistribution ofSprayFlowtotheUpperandLowerCompartments DeckFanPost-Accident Initiation ofDeckFansFlowRateperFanAssumedSprayEfficiency ofWaterfromIceCondenser Drains70oF22oF70oF3600gpm36secLC2700gpmUC4500gpm480sec43,890cfmperfan100%m:11944-2w.wpf:1d441195 3.1-23 TABLE3.1-3(continued)

STRUCTURAL HEATSINKS1LC2LC3LC4LCLCLCLC8LC9LC10LC11LC12LC13UC14UC~Areatr'2,10511,70165,9795,4625,273290'4,8964,5155,77557,3179,4042,62337834,895Thickness ft0.0469/2.0 2.04.00.08330.01030.250.00780.10420.0090.008330.03130.03130.0365/0.1667 0.0078MaterialSteel/concrete ConcreteConcreteSteelSteelLeadSteelSteelSteelSteelSteelSteelSteeVconcrete Steel1516171819UCUCUCUCUC8,06042029,33234,1254200.02080.00522.00.0469/2.0 0.0052SteelSteelConcreteSteel/concrete SteelUC:UpperCompartment LC:LowerCompartment DE:Dead-Ended Compartment IC:IceCompartment m:519442w.wpf:1d441195 3.1-24 TABLE31-4MASSANDENERGYRELEASERATES,MINIMUMSITimesec101212.41416182024283236405265758695124294MassIbm/sec5791048870335002526022660195801698016000145301214010410917070106750564035804390230280390810420430330EneBTU/sec3.081(10)2.542(10')

1.762(10')

1.357(10')

1.223(10')

1.096(10)9.838(10)9.346(10)8.608(10')

7.313(10')

6,254(10')

5.472{10')

3.871(10')

2.839(10')

1.757(10')

7.951(10')

9.057(10')

1.267(10')

6.321(10')

2.073(10')

2.884{10')

2.464(10')

1.666(10')

1.452(10')

1.314(10')

m'A1944-2w.wpt:

1d~11953.1-25 TABLE3.1-5NITROGENMASSANDENERGYRELEASERATES~Timeeec69.273.277.281.285.289.293.297.2101.2105.2109.2113.2117.2121.2125.2129.2137.2141.2145.2153.2161.2169.2177.2FlowRateIbm/sec231.8166.4120.887.362.142.928.819.514.111.19.07.35.94.83.93.22.11.81.41.00.70.50.3mal944-2'.wpt:1d441195 3.1-26 TABLE3.1-6SAFETYINJECTION FLOWRATERERATINGPROGRAMANALYSISRCSPRESSURE(psia)415515615715815915101511151215131514151515HHSIFLOW(Ib/sec)20.4819.3818.1816.8915.5414.0912.5010.467.994.480.000.00CHARGINGFLOW(Ib/sec)37.3235.2933.2931.2629.0626.8024.4822.1619.7417.2714.7012.02TOTALFLOW(Ib/sec)57.8054.6751.4748.1544.6040.8936.9832.6227.7322.051¹.7012.02m:51944-2w.wpf:

1d4411953.1-27 TABLE3.1-7PLANTINPUTPARAMETERS USEDINSMALLBREAKLOCAANALYSISRERATINGPROGRAMANALYSISCorePowerTotalCorePeakingFactorSteamGenerator TubePluggingLevel102%of3588MWt2.3215%(peakuniform)Accumulator Conditions:

CoverGasPressureWaterVolumeTotalVolume600psia946.0ft1350ftRCSInitialConditions:

ReducedTemperature, ReducedPressureCaseLoopTemperatures Consistent WithT,ProgramSetpointof,PressureVesselFlowrate547OF2100psia354000gpmm%1944-2w.wpf:1d441295 3.1-28 TABLE3.1-7PLANTINPUTPARAMETERS USEDINSMALLBREAKLOCAANALYSISRERATINGPROGRAMANALYSISCorePowerTotalCorePeakingFactorSteamGenerator TubePluggingLevel102%of3588MWt2.3215%(peakuniform)Accumulator Conditions:

CoverGasPressureWaterVolumeTotalVolume600psia946.0ft'350ft'CSInitialConditions:

ReducedTemperature, ReducedPressureCaseLoopTemperatures Consistent WithT,ProgramSetpointof,PressureVesselFlowrate547'F2100psia354000gpmm&1944-2w.wpl:1d~1195 3.1-28

TABLE3.1-8SMALL-BREAK LOCACALCULATION RERATINGPROGRAMANALYSISRESULTSPARAMETER VALUEReducedTemeratureReducedPressureBreakSIze:2-Inch3-Inch4-InchPeakcladtemperature

('F)Elevation (ft)1899212212.0012.00141411.25Zr/H,Ocumulative reactionMaximumlocal(%)Elevation (ft)Totalcore(%)7.167.700.2512.0012.0011.50<0.3<0.3<0.3RodBurstNoneNoneNoneCALCULATION:

NSSSPowerMWt102%ofPeakLinearPowerkW/ft102%ofHotRodPowerDistribution (kW/ft)Accumulator WaterVolume,cu.ft.358816.426SeeFigure3.1-21946Doesnotincludepumpheat.mA1944-2w.wpf:1d441195 3.1-29 TABLE3.1-8SMALL-BREAK LOCACALCULATION RERATINGPROGRAMANALYSISPARAMETER RESULTSVALUEReducedTemeratureReducedPressureBreakSize:2-Inch3-Inch4-InchPeakcladtemperature

('F)Elevation (ft)Zr/H,Ocumulative reactionMaximumlocal(%)Elevation (ft)189912.007.1612.002122141412.0011.257.700.2512.0011.50Totalcore(%)RodBurst(0.3None(0.3None(0.3NoneCALCULATION:

NSSSPowerMWt102%ofPeakLinearPowerkW/ft102%ofHotRodPowerDistribution (kW/ft)Accumulator WaterVolume,cu.ft.358816.426SeeFigure3.1-21946Doesnotincludepumpheat.m'31944-2w.wpf:1d441295 3.1-29 TABLE3.1-9TIMESEQUENCEOFEVENTSFORCONDITION IIIEVENTSRERATINGPROGRAMANALYSISSmall-Break LossofCoolantAccidentEventTime(s)ReducedTemeratureReducedPressureBreakSIze:2-Inch3-Inch4-InchBreakoccursReactortripsignalSafetyinjection signalStartofsafetyinjection deliveryLoopsealventing25.3736.5463.541634.417.1044.1010.7437.74652.1420A11.246.85LoopsealcoreuncoveryLoopsealcorerecoveryBoil-offcoreuncoveryAccumulator injection beginsPeakcladtemperature occursTopofcorecoveredSlflowexceedsbreakflowN/AN/A2216.7N/A4143.8N/A4587.5645.8680.31045.71711.51958.7N/A2197.1424.6439.2696.5901.0969.51982.7N/Am:II1944-2NI'.WPf:1d441295 3.1-30 TABLE3.1-9TIMESEQUENCEOFEVENTSFORCONDITION IIIEVENTSRERATINGPROGRAMANALYSISSmall-Break LossofCoolantAccidentEventTime(s)ReducedTemeratureReducedPressureBreakSize:2-Inch3-Inch4-InchBreakoccursReactortripsignalSafetyinjection signalStartofsafetyinjection deliveryLoopsealventing25.3736.5463.541634.411,2417.1044.10652.1420.46.8510.7437.74LoopsealcoreuncoveryLoopsealcorerecoveryN/A645.8424.6N/A680.3439.2Boil-offcoreuncoveryAccumulator injection beginsPeakcladtemperature occursTopofcorecoveredSlflowexceedsbreakflow2216.7N/A1711.51045.71958.74587.52197.14143.8N/AN/A696.5901.0969.51982.7N/Amal944.2w.wpf:

1d4411953.1-30 TABLE3.1-10SMALL-BREAK LOCACALCULATION RERATINGPROGRAMANALYSISRESULTSPARAMETER VALUEHighTemp.ReducedTemp.HihPressureHihressure3-Inch3-InchPeakcladtemperature

('F)Elevation (ft)Zr/H0cumulative reactionaximumlocal(%)Elevation (ft)Totalcore(%)RodBurst175611.751.9911.75<0.3None188711.754.6411.75<0.3NoneCALCULATION'SSS PowerMWt102%ofPeakLinearPowerkW/ft102%ofHotRodPowerDistribution (kW/ft)Accumulator WaterVolume,cu.ft.358816.426SeeFigure3.1-21946Doesnotincludepumpheat.m(1944-2w.wpf:1d441295 3.1-31 TABLE3.1-10SMALL-BREAK LOCACALCULATION RERATINGPROGRAMANALYSISRESULTSPARAMETER VALUEHighTemp.ReducedTemp.HihPressureHihressure3-Inch3-InchPeakcladtemperature

('F)Elevation (ft)Zr/H0cumulative reactionaximumlocal(%)Elevation (lt)Totalcore(%)RodBurst175611.751.9911.75<0.3None188711.754.6411.75<0.3NoneCALCULATION:

NSSSPowerMWt102%ofPeakLinearPowerkW/ft102%ofHotRodPowerDistribution (kW/ft)Accumulator WaterVolume,cu.ft.358816.426SeeFigure3.1-21946Doesnotincludepumpheat.m%19442)N.wpf:1d~1195 3.1-31 TABLE3.1-11TIME,SEQUENCEOFEVENTSFORCONDITION IIIEVENTSRERATINGPROGRAMANALYSISSmall-Break LossofCoolantAccidentEventTime(s)HighTemp.ReducedTemp.HihPressureHihPressureBreakoccursReactortripsignaISafetyinjection signalStartofsafetyinjection deliveryLoopsealventingLoopsealcoreuncoverLoopsealcorerecoveryBoil-offcoreuncoveryAccumulator injection beginsPeakcladtemperature occursTopofcorecoveredSlflowexceedsbreakflow3-Inch19.0329.7451.74666.96.N/AN/A1070.41672.01793.7N/A2022.03-Inch15.9720.9547.95698.78691.54726.851166.81855.21986.2N/A2282.7mh1944-2IN.WPf:1d~1195 3.1-32 TABLE3.1-11TIMESEQUENCEOFEVENTSFORCONDITION IIIEVENTSRERATINGPROGRAMANALYSISSmall-Break LossofCoolantAccidentEventTime(s)'ighTemp.ReducedTemp.HihPressureHihPressureBreakoccursReactortripsignalSafetyinjection signalStartofsafetyinjection deliveryLoopsealventingLoopsealcoreuncoveryLoopsealcorerecoveryBoil-offcoreuncoveryAccumulator injection beginsPeakcladtemperature occursTopofcorecoveredSlflowexceedsbreakflow3-Inch19.0329.7451.74666.96N/AN/A1070.41672.01793.7N/A2022.03-Inch15.9720.9547.95698.78691.54726.851166,81855.21986.2N/A2282.7mh1944-2w.wpf:1dM1295 3.1-32

TABLE3.1-12PLANTINPUTPARAMETERS USEDINSMALLBREAKLOCAANALYSISk3%MAINSTEAMSAFETYVALVESETPOINTTOLERANCE ANALYSISCorePower(MWt)PeakLinearPower(kW/ft)TotalCorePeakingFactor,F~HotChannelEnthalpyRiseFactor,F,MaximumAssemblyAveragePower,PAxialOffset(%)FuelAssemblyArraySteamGenerator TubePluggingLevel(%)Accumulator WaterVolume(ft'/tank)

Accumulator TankVolume(ft'/tank)

MinimumAccumulator GasPressure(psia)Accumulator WaterTemperature

('F)Refueling WaterStorageTankTemperature

('F)ThermalDesignFlowrate(gpm/loop)

RCSLoopAverageTemperature

('F)NominalInitialRCSPressure(psia)NominalSteamPressure(psia)SafetyInjection DelayTime(sec)HHSIPumpHeadDegradation

(%)ChargingPumpHeadDegradation

(%)ChargingPumpFlowImbalance (gpm)HHSICross-Tie ValvePositionAuxiliary Feedwater TotalFlowrate(gpm)102%of3250102%of14.9212.321.551.433+3015X15OFA15946135060013012088,500547.0210060727101025Closed750m61944-2w.wpf:1d~1295 3.1-33 TABLE3.1-12PLANTINPUTPARAMETERS USEDINSMALLBREAKLOCAANALYSISk3%MAINSTEAMSAFETYVALVESETPOINTTOLERANCE ANALYSISCorePower(MWt)PeakLinearPower(kW/ft)TotalCorePeakingFactor,FaHotChannelEnthalpyRiseFactor,F,MaximumAssemblyAveragePower,PAxialOffset(%)FuelAssemblyArraySteamGenerator TubePluggingLevel(%)Accumulator WaterVolume(ft'/tank)

Accumulator TankVolume(ft'/tank)

MinimumAccumulator GasPressure(psia)Accumulator WaterTemperature

('F)Refueling WaterStorageTankTemperature

('F)ThermalDesignFlowrate(gpm/loop)

RCSLoopAverageTemperature

('F)NominalInitialRCSPressure(psia)NominalSteamPressure(psia)SafetyInjection DelayTime(sec)HHSIPumpHeadDegradatlon

(%)ChargingPumpHeadDegradation

(%)ChargingPumpFlowImbalance (gpm)HHSICross-Tie ValvePositionAuxiliary Feedwater TotalFlowrate(gpm)21001013012088,500547.06072710750Closed102%of3250102%of14.9212.321.551.433+3015X15OFA159461350mal944.2w.wpf:1d441195 3.1-33 TABLE3.1-13TIMESEQUENCEOFEVENTSFORCONDITION IIIEVENTS%MAINSTEAMSAFETYVALVESETPOINTTOLERANCE ANALYSISmall-break LossofCoolantAccidentTime(s)EventReducedPressure, ReducedTemperature 3-Inch2-InchBreakOccursReactortripsignalSafetyinjection signalStartofsafetyinjection Startofauxiliafy feedwater deliveryLoopsealventingLoopsealcoreuncovefyLoopsealcorerecovefyBoil-offcoreuncoveryAccumulator injection beginsPeakcladtemperature occursTopofcorecoveredSlflowrateexceedsbreakflowrate0.08.6417.1344.1368.6592N/AN/A16801890N/A18900.019.0337.1164.1179.11390N/AN/A2312N/A4042N/A4091Mainsteamsafetyvalvesetpointtolerance increasecaseat3250MWtcorepower.m%1944-2IN.WPf:1d441195 3.1-34 TABLE3.1-13TIMESEQUENCEOFEVENTSFORCONDITION IIIEVENTS%3%MAINSTEAMSAFETYVALVESETPOINTTOLERANCE ANALYSISSmall-break LossofCoolantAccidentTime(s)'ventReducedPressure, ReducedTemperature 3-Inch2-InchBreakOccursReactortripsignalSafetyinjection signalStartofsafetyinjection Startofauxiliary feedwater deliveryLoopsealventingLoopsealcoreuncoveryLoopsealcorerecove1yBoil-offcoreuncoveryAccumulator injection beginsPeakcladtemperature occursTopofcorecoveredSlflowrateexceedsbreakflowrate0.08.6417.1368.6592N/A'6801890N/A18900.019.0337.1164.1179.11390N/AN/A2312N/A4042N/A4091'ainsteamsafetyvalvesetpointtolerance increasecaseat3250MWtcorepower.m&1944.2W.Wpf:

1d~11953.1-34 TABLE3.1-14SMALL-BREAK LOCACALCULATIONS

%MAINSTEAMSAFETYVALVESETPOINTTOLERANCE ANALYSISRESULTSNOTRUMPPeakCladTemperature

('F)PeakCladTemperature Location(ft)PeakCladTemperature Time(sec)LocalZr/H,OReactionMaximum(%)12.012,0189040425.063.75ReducedPressure, ReducedTemperature 3-Inch2-Inch19511833LocalZr/H,OReactionLocation(ft)TotalZr/H,OReaction(%)RodBurstBurstandBlockagePenalty('F)TotalPeakCladTemperature

{F)12.0<1.0None117206812.0<1.0None151848Mainsteamsafetyvalvesetpointtolerance increasecaseat3250MWtcorepower.m%1944-2W.wpf:1d44'f f953.1-35

TABLE3.1-15PLANTINPUTPARAMETERS USEDINSMALLBREAKLOCAANALYSIS30%SGTPPROGRAMANALYSISWITHHHSICROSSTIESCLOSEDCorePower(MWt)PeakLinearPower(kW/ft)TotalCorePeakingFactor,F~HotChannelEnthalpyRiseFactor,F,MaximumAssemblyAveragePower,PAxialOffset(%)FuelAssemblyArraySteamGenerator TubePluggingLevel(%)Accumulator WaterVolume(ft'/tank)

Accumulator TankVolume(ft'/tank)

MinimumAccumulator GasPressure(psia)Accumulator WaterTemperature

('F)Refueling WaterStorageTankTemperature

('F)ThermalDesignFlowrate(gpm/loop)

RCSLoopAverageTemperature

('F)NominalInitialRCSPressure(psia)NominalSteamPressure(psia)SafetyInjection DelayTime(sec)HHSIPumpHeadDegradation

(%)ChargingPumpHeadDegradation

(%)ChargingPumpFlowImbalance (gpm)HHSICross-Tie ValvePositionAuxiliary Feedwater TotalFlowrate(gpm)102%of3250102%of14.122.321.551.38+2015X15OFA30946135060013012083,200553.0210059547151025Closed750m:$1944-2w.wpf:1dM1 1953.1-36

TABLE3.1-15PLANTINPUTPARAMETERS USEDINSMALLBREAKLOCAANALYSIS30%SGTPPROGRAMANALYSISWITHHHSICROSS-TIES CLOSEDCorePower(MWt)PeakLinearPower(kW/ft)TotalCorePeakingFactor,FaHotChannelEnthalpyRiseFactor,F,MaximumAssemblyAveragePower,PAxialOffset(%)FuelAssemblyArraySteamGenerator TubePluggingLevel(%)Accumulator WaterVolume(ft'/tank)

Accumulator TankVolume(ft'/tank)

MinimumAccumulator GasPressure(psia)Accumulator WaterTemperature

('F)Refueling WaterStorageTankTemperature

('F)ThermalDesignFlowrate(gpm/loop)

RCSLoopAverageTemperature

('F)NominalInitialRCSPressure(psia)NominalSteamPressure(psia)SafetyInjection DelayTime(sec)HHSIPumpHeadDegradation

(%)ChargingPumpHeadDegradation

(%)ChargingPumpFlowImbalance (gpm)HHSICross-Tie ValvePositionAuxiliary Feedwater TotalFlowrate(gpm)102%of3250102%of14.122.321.551.38+2015X15OFA30946135060013012083,200553.0210059547151025Closed750m31944-2w.wpf:1d441295 3.1-36 TABLE3.1-16TIMESEQUENCEOFEVENTSFORCONDITION IIIEVENTS30%SGTPPROGRAMANALYSISWITHHHSICROSS-TIES CLOSEDSmall-break LossofCoolantAccidentEventBreakoccursReactortripsignalSafetyinjection signalStartofsafetyinjection Startofauxiliary feedwater deliveryLoopsealventingLoopsealcoreuncovefyLoopsealcorerecoveryBoil-offcoreuncoveryAccumulator injection beginsPeakcladtemperature occursTopofcorerecovered CombinedpumpedSlflowrateexceedsbreakflowrateTime(s)ReducedPressure, ReducedTemperature 3-Inch0.08.817.464.488.8528N/AN/A1054164817482995185630%steamgenerator tubepluggingcaseat3250MWtcorepower.m&1944-2w.wpf:1d~1295 3.1-37 TABLE3.1-16TIME.SEQUENCEOFEVENTSFORCONDITION IIIEVENTS30%SGTPPROGRAMANALYSISWITHHHSICROSS-TIES CLOSEDSmall-break LossofCoolantAccidentEventTime(s)ReducedPressure, ReducedTemperature 3-InchBreakoccursReactortripsignalSafetyinjection signalStartofsafetyinjection Startofauxiliary feedwater deliveryLoopsealventingLoopsealcoreuncoveryLoopsealcorerecoveryBoil-offcoreuncoveryAccumulator injection beginsPeakcladtemperature occursTopofcorerecovered CombinedpumpedSlflowrateexceedsbreakflowrate88.80.08.817.464.4528N/AN/A1054164817482995185630%steamgenerator tubepluggingcaseat3250MWtcorepower.m:11944-2w.wpf:1d441195 3.1-37 TABLE3.1-17SMALL-BREAK LOSSOFCOOLANTACCIDENTCALCULATIONS 30%SGTPPROGRAMANALYSISWITHHHSICROSS-TIES CLOSEDRESULTSReducedPressure, ReducedTemperature 3-InchNOTRUMPPeakCladTemperature

('F)PeakCladTemperature Location(ft)PeakCladTemperature Time(sec)LocalZr/H,OReactionMaximum(%)LocalZr/H,OReactionLocation(ft)TotalZr/H,OReaction(%)RodBurstBurstandBlockagePenaltyTotalPeakCladTemperature

('F)1443'F11.51748(1.011.5(1.0NoneNone1443'F30%steamgenerator tubepluggingcaseat3250MWtcorepower.mh1944-2w.wpf:1d441295 3.1-38 TABLE3.1-17SMALL-BREAK LOSSOFCOOLANTACCIDENTCALCULATIONS 30%SGTPPROGRAMANALYSISWITHHHSICROSS-TIES CLOSEDRESULTSReducedPressure, ReducedTemperature 3-InchNOTRUMPPeakCladTemperature

('F)PeakCladTemperature Location(ft)PeakCladTemperature Time(sec)LocalZr/H,OReactionMaximum(%)LocalZr/H,OReactionLocation(ft)TotalZr/H,OReaction(%)RodBurstBurstandBlockagePenaltyTotalPeakCladTemperature

('F)11.51443'F11.51748(1.0(1.0NoneNone1443'F30%steamgenerator tubepluggingcaseat3250MWtcorepower.m:51944-2w.wpf:1d441195 3.1-38

2500200015001000500002030TIME(S)4050'gure3.1-1aReactorCoolantSystemPressureCaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1I%1944-2w.wpf:1d~1195 3.1-39 250020001500LLI1000LIJ500101520TIME(S)253035igure3.1-1bReactorCoolantSystemPressureCaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m&1944-2w.wpf:1

&)411953.140

250D2000150D1000101520TIME(S)2530igure3.1-1cReactorCoolantSystemPressureCaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mh1944-2w.wpf:1d~1195 3.1<1 250D200015001000500102D30TIME(S)4050igure3.1-1dReactorCoolantSystemPressureCaseD,CD=0,4,Thot=586.8'F, P=2250psiaDonaldC.CookUnit1mA1944-2w.wpf:1d~1195 3.1-42

25002000150010005001020TIME(S)30igure3.1-1eReactorCoolantSystemPressureCaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1m&1944-2w.Wpf:1d441195 3.1' 250020001500LU100050020TlhlE(S)igure3.1-1fReactorCoolantSystemPressureCaseA,CD=0.4,Thot=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1m&1944-2w.wpf:1d441195 E6000050000<000030000LIJ20000410000102030TIME(S)4050igure3.1-2aBreakFlowDuringBlowdownCaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC,CookUnit1m&1944-2w.wpf:1d441195 3.145 70000600005000040000I-30000C72000010000)0152025TIME(S)5035Figure3.1-2bBreakFlowDuringBlowdownCaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mA1944.2w.wpf:1d441195 3.1<6 SOOOOCD4JMCQ600004000020000101520TIME(S)2530Figure3.1-2cBreakFlowDuringBlowdownCaseC,CD=0.8,That=609.1'F, P=2250psiaDonaldC.CookUnit1m:51944-2w.wpf:1dM1195 3.1-47 700006000050000LJ4000030000C)20000$000010203DTIME(S)Figure3.1-2dBreakFlowDuringBlowdownCaseD,CD=0.4,Thot&86.8'F, P=2250psiaDonaldC.CookUnit1mal944-2W.wpf:1d441195 3.1<8 6000050000400003000020000100001020TIME(S)Figure3.1-2eBreakFlowDuringBlowdown~~CaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1mh1944-2w.wpt:1dC41195 3.149

6000050000<00003000020000100001020TIME(S)30Figure3.1-2fBreakFlowDuringBlowdownCaseF,CD=0.4,Thot=609.1'F, P=2250psia,maxSlDonaldC.CookUnit1m%1944.2w.wpf:1d~1195 3.1-50

2010C/)cn]0CL-20I-4J-3044C5-40-50102030TIME(S)4050Figure3.1-3aCorePressureDropCaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m61944-2w.wpf:1d441195 3.1-51 100-5010152025TIME(S)30Figure3.1-3bCorePressureDropCaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mA1944-2w.wpf:1d441195 3.1-52' 10CL-10LIJCLC/l-20LLlLXCL-30I-LQ-40L4-50-60101520TIME(5)25CorePressureDropCaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m%1944-2w.wpf:1d441195 3.1-53 4020CACL-60102030TIME(S)Figure3.1-3dCorePressureDrop~~CaseD,CD=0.4,Thot=586.8'F, P=2250psiaDonaldC.CookUnit1m(1944-2w.wpf:1dM1195 3.1-54 2010I-QJ204J4-301020TIME(S)30Figure3.1-3eCorePressureDropCaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1mA1944.2w.wpf:1 d~11953.1-55

20100C/)n--10I-QJ204J4-30-401020TIME(S)30Figure3.1-3fCorePressureDropCaseF,CD=0.4,That=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1m&1944-2w.wpf:1d441195 3.1-56

-COREINLET~COREOUTLET40000300002000010000-10000102030TIME(S)4050Figure3.1<aCoreFlowrateCaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m&1944-2w.wpt:1d~1195 3.1-57

-COREINLET~COREOUTLET40000300002000010000-10000-20000101520TIME(S)253035Figure3.1MbCoreFlowrateCaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m%1944-2w.wpf:1d~1195 3.1-58

-COREINLETaaCOREOUTLET40000300002000010000CO-10000-20000-30000101520TIME(S)-2530Figure3.1<cCoreFlowrateCaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mal944-2w.wpf:1d441195 3.1-59

-COREINLET~COREOUTLET40000300002000010000-10000102030TIME(S)4050Figure3.1<dCoreFlowrateCaseD,CD=0.4,Thot&86.8'F, P~OpsiaDonaldC.CookUnit1m&1944-2w.wpf:1d~1195 3.1-60 I'IlailIIII4%tIIIIt4f~~~I~~~I'tl~~~.~~~~~~

-COREINLET~COREOUTLET40000300002000010000-100001020TIME(S)30Figure3.1<fCoreFlowrateCaseF,CD=OA,Thot=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1mA1944-2w.wpf:1d~1195 3.1-62 60005000~4000300020001000102030TIME(S)4050eFigure3.1-5aAccumulator FlowDuringBlowdownCaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m&1944-2w.wpf:1d~1195 3.1-63 60005000~<0003000I-2000100010152025TIME(S)30Figure3.1-5bAccumulator FlowDuringBlowdownCaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m:11944-2w.wpf:1d~1195 3.1-64 600050004000300020001000101520TIME(S)2530Figure3.1-5cAccumulator FlowDuringBlowdownCaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mA1944.2w.wpf:1dM1195 3.1-65 500040003000I-20001000102030TIME(S)4050Figure3.1-5dAccumulator FlowDuringBlowdownCaseD,CD=0.4,Thot=586.8'F, P=2250psiaDonaldC.CookUnit1m&1944-2w.wpf:1d~1 1953.1-66 60005000~<00030002000100020TIME(S)30Figure3.1-5eAccumulator FlowDuringBlowdownCaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1m81944-2w.wpt:1d~1195 3.1-67 60005000~<000m3000200010001020TIME(5)30Figure3.1-5fAccumulator FlowDuringBlowdownCaseF,CD=0.4,Thot=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1mA1944-2w.wpf:1d~1195 3.1-68 COREIllXTURELEVELOUENCHFRONTLOCATION~DOWNCOIIER LEVEL20u1510501015020250TIMEAFTERREFLOOD(S)30Figure3.1-6aVesselLiquidLevelsDuringRefloodCaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m&1944-2w.wpf:1d441195 3.1-69

-COREMIXTURELEVEL~QUENCHFRONTLOCATION~OOWNCOMER LEVEL201510005010150200250TtMEAFTERREFLOOD(S)300Figure3,1-6bVesselLiquidLevelsDuringRefloodCaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mh1944-2w.wpf:1d441195 3.1-70

-COREMIXTURELEVELa-aOUENCHFRONTLOCATION~OOWNCOMER LEVEL2520I-u1510501015020250TIMEAFTERREFLOOD(S)30Figure3.1-6cVesselLiquidLevelsDuringRefloodCaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mA1944-2w.wpf:1d441195 3.1-71 COREMIXTURELEVEL~QUENCHFRONTLOCAT'ION

~OOWNCOIIER LEVEL20I-1050101020250TIMEAFTERREFLOOD(S)30Figure3.1-6dVesselLiquidLevelsDuringRefloodCaseD,CD=0.4,Thot=586.8'F, P=2250psiaDonaldC.CookUnit1mA1944-2w.wpf:1dM1195 3.1-72 COREMIXTURELEVEL~QUENCHFRONTLOCATION~OOWNCOMER LEVEL25201510501015020250TIMEAFTERREFLOOO($)300Figure3.1-6eYesselLiquidLevelsDuringRefloodCaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1mh1944-2W.Wpt:1 d4411953.1-73 COREMIXTURELEVELe-aQUENCHFRONTLOCATION~DOWNCOIIER LEVEL2520501015020250TIMEAFTERREFLDDD(S)Figure3.1-6fVesselLiquidLevelsDuringRefloadCaseF,CD=0.4,That=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1m&1944-2w.Wpf:1d~1195 3.1-74 1~1F9LLJI0'C9CDo0~7o40'0'0'501001502TIMEAFTERREFLOOD(S)250Figure3.1-7aCoreInletFlowDuringRefloodCaseA,CD=0.4,Thot=609,1'F, P=2250psiaDonaldC.CookUnit1m%1944-2w.wpf:1d441195 3.1-75 Cf)0'I-0'C9CDo0'CO40'C)CO0'0'50015020TIMEAFTERREFI.OOD(S)250Figure3.1-7bCoreInletFlowDuringRefloodCaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnitim%1944-2w.wpf:1d~1195 3.1-76 0'CQC)o0'C)4F6C)tD0'0'50015020TIMEAFTERREFLOOD(S)Figure3.1-7cCoreInletFlowDuringRefloodCaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m%1944-2w.wpt:1d~1195 3.1-77 1~2F9C9z,'~8C)C)0'oF6C)0'A@5010150200TIMEAFTERREFLOOD(S)250Figure3.1-7dCoreInletFlowDuringRefloadCaseD,CD=0.4,Thot=586.8'F, P=2250psiaDonaldC.CookUnit1m31944-2W.wpf:1d~1195 3,1-78 0'OIC90ASC5C)0'I-o0.6C)0~50~45010150200TIMEAFTERREFLOOD(S)250Figure3.1-7eCoreInletFlowDuringRefloodCaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1mA1944-2w.wpf:1d~1195 3.1-79 1~2CQ0~8CDCDCD40.6CD0~450,1015020()TIMEAFTERREFLOODS250Figure3.1-7fCoreInletFlowDuringRefloodCaseF,CD=0.4,That=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1m:(1944-2w.wpt:1dM1195 3.1-80 50004000LIJ30002000100005010150202030TIMEAFTERREFLOODSFigure3.1-8aAccumulator andSlFlowDuringRefloodCaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mA1944-2w.wpf:1d441195 3.1-81 50004000LLI3000LIJI-2000100050010200250300TIMEAFTERREFLOODSFigure3.1-8bAccumulator andSlFlowDuringRefloodCaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mA1944-2w.wpt:1d441195 3.1-82 50004000300020001000501015020250TIMEAFTERREFLOOD(S)300Figure3.1-8cAccumulator andSlFlowDuringRefloodCaseC,CD=0.8,Thot=609.1 F,P=2250psiaDonaldC.CookUnit1m%1944-2w.wpf:1d441195 3.1-83 50004000300020001000501015020250300TIMEAFTERREFLOODSFigure3.1-8dAccumulator andSlFlowDuringRefloodCaseD,CD=0.4,Thot=586,8'F, P=2250psiaDonaldC.CookUnit1mA1944-2w.wpf:1dM1195 3.1-84 50004000300020001000501015020250TIMEAFTERREFLOOD(S)Figure3.1-8eAccumulator andSlFlowDuringRefloodCaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1m31944.2w.wpf:1d441195 3.1-85

'50,0040003000200010000050100150200250TIMEAFTERREFLOOD(S)300Figure3.1-8fAccumulator andSlFlowDuringRefloodoCaseF,CD=0.4,Thot=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1m%1944-2w.wpf:1d441195 3.1-86 CDCV15II1050101502020TIMEAFTERREFLOOD(S)Figure3.1-9aIntegralofCoreInletFlow~~CaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m31944-2w.wpf:1 d~11953.1-87 2010LK0I5010150200250TIMEAFTERREFLOOD(S)Figure3.1-9bIntegralofCoreInletFlowCaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m&1944-2w.wpf:1d441195 3.1-88' 20C)C415II-10CL0I-5010150200250TIMEAFTERREFLOOD(S)300Figure3.1-9cIntegralofCoreInletFlowCaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m%1944-2w.wpf:1d~1195 3.1-89 20tDz15I10CQLDC)o5CLCP0I-5010102020TIMEAFTERREFLOOD(S)30Figure3.1-9dIntegralofCoreInletFlowCaseD,CD=0.4,Thot=586.8'F, P=2250psiaDonaldC.CookUnit1m:41944-2W.wpf:1d~1 1953.1-90 20C)CVx15I501015020250TIMEAFTERREFLOOD(S)300eFigure3.1-9eIntegralofCoreInletFlowCaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1m51944.2w.wpf:1dM1195 3,1-91 20I10C9CDCDo5CL0I-501015020250TIMEAFTERREFLOOD(S)300eFigure3.1-9fIntegralofCoreInletFlowCaseF,CD&.4,Thot=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1mh1944-2w.wpt:1dM1199 3.1-92 800600400X200-200501015020250T.lME(5)0350Figure3.1-10aMassFluxatPeakTemperature Elevation CaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m&1944-2w.wpf:1d441195 3.1-93 SOO600400OC200-2005000150TIME(S}2503DFigure3.1-10bMassFluxatPeakTemperature Elevation CaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC,CookUnit1mh1944-2w.wpt:1d441195 3.1-94 800600400200-200501015020TIME(S)25030Figure3.1-10cMassFluxatPeakTemperature Elevation CaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mA1944-2w.wpf:1d441195 3.1-95 800600I-400X200-200501015020250TIME(S)3030Figure3.1-10dMassFluxatPeakTemperature Elevation CaseD,CD=0.4,Thot=586.8'F, P=2250psiaDonaldC.CookUnit1m51944-2w.wpf:1d~1 1953.1-96 800600400X200-2005010$502025030350TIME(S)Figure3.1-10eMassFluxatPeakTemperature Elevation CaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1mh1944-2w.wpf:1d441195 3.1-97

800600I-400X200-2005010015020250300350TIMESFigure3.1-1OfMassFluxatPeakTemperature Elevation CaseF,CD=0.4,Thot=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1m%1944-2w.wpf:1d441195 3.1-98 ilf 4JCf)~10I4J$050100150200250TIME(S)300350Figure3.1-11aRodH.T.C.atPeakTemperature Elevation CaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mh1944<w.wpf:1d~1195 3.1-99 10LaJCO~10I--'I$0050100150200TIME(S)250300Figure3.1-11bRodH.T.C.atPeakTemperature Elevation CaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m:L19444w.wpf:1dM1195 3.1-100 10~)0I--110050100150200TIME(S)250300Figure3.1-11cRodH.T.C.atPeakTemperature Elevation CaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mh19444w.wpf:1d~1195 3.1-101 10CQ~10100501001502DD250TIME(S)300350Figure3.1-11dRodH.T.C.atPeakTemperature Elevation CaseD,CD=0.4,Thot&S6,8'F, P=2250psiaDonaldC.CookUnit1m&19444w.wpt:1dM1195 3.1-102 10~10~10LJC71o104JCf)~10-110050100150200250TIME(S)300350Figure3.1-11eRodH.T.C.atPeakTemperature Elevation CaseE,CD=0.4,Thot=609.1'F, P=2100psiaDonaldC.CookUnit1mh1944<w.wpf:1d441195 3.1-103 410~10cn~104~~10C/7~10I-1050100150200250'00TIME(S)350Figure3.1-11fRodH.T.C.atPeakTemperature Elevation CaseF,CD&.4,Thot=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1mA1944<w.wpt:1d441195 3.1-104 1800160014001200LLJ10008006004002005010102020300350()TIMESFigure3.1-12aVaporTemperature CaseA,CD=0.4,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m:519444w,wpf:1d~1195 3.1-105 18001600140012001000800IJJ600400200500150TIME(S)20250300Figure3.1-12bVaporTemperature CaseB,CD=0.6,That=609.1'F, P=2250psiaDonaldC.CookUnit1mh19444w.wpf:1d~1 1953.1-106

18001600140012001000800600400200501015020TIME(S)250300Figure3.1-12cVaporTemperature CaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mA19444w.wpf:1d441195 3.1-107 180016001400120010DO800600400200501015020025030350TIME(S)Figure3.1-12dVaporTemperature CaseD,CD=0.4,Thot=586.8'F, P=2250psiaDonaldC.CookUnit1mh19444w.wpf:1d441195 3.1-108 20001500LJ1000LJ500501015020025030350TtME(5)Figure3.1-12eVaporTemperature CaseE,CD=0.4,Thot=609.1'F, PM100psiaDonaldC.CookUnit1m%1944<w.wpf:1d~1 1953.1-109 fi/i 200015001000500501015020250300350TIMESFigure3.1-12fVaporTemperature CaseF,CD=0.4,Thot=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1mh1944<w.wpt:1dM1195 3.1-110 22002000180016001140012001000800600501015020250TIME(S)300350Figure3.1-13aFuelRodPeakCladTemperature CaseA,CD=0.4,That=609.1'F, P=2250psiaDonaldC.CookUnit1mh1944<w.wpf:1d441195 3.1-111 20001800160014001200I-100080060050100150200TIME(S}250300Figure3.1-13bFuelRodPeakCladTemperature CaseB,CD=0.6,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1m:L19444w.wpf:1d441195 31-112 20001800160014001200I-100080060050150TIME(S)200250Figure3.1-13cFuelRodPeakCladTemperature CaseC,CD=0.8,Thot=609.1'F, P=2250psiaDonaldC.CookUnit1mh1944<w.wpf:1d441195 3,1-113 220020001800160014001200I-100080060050101502020TIME(S)30350Figure3.1-13dFuelRodPeakCladTemperature CaseD,CD=0.4,Thot&86.8'F, P=2250psiaDonaldC.CookUnit1mh1944<w.wpf:1d441195 3.1-114 5~4l5II 22002000180016001400LJ1200I-10008006005010102025030350TIME(5)Figure3.1-13eFuelRodPeakCladTemperature CaseE,CD=0.4,Thot=609.1'F, P&1OOpsiaDonaldC.CookUnit1mh1944<w.wpf:1d441195 3.1-<<S

2200200D1B0016001400120DLLI10008006005001502025030TIME(S)350Figure3.1-13fFuelRodPeakCladTemperature CaseF,CD&.4,Thot=609.1'F, P=2100psia,maxSlDonaldC.CookUnit1m:L19444w.wpf:1 d4411953.1-116

-UPPERCOIIPARTMEHT

~LOWERCOIIPARTMEHT 501020TIME(SEC)250300Figure3.1-14Containment PressureCD=0.4,MinSlDonaldC.CookUnit1m51944<w.wpf:1d~1 1953.1-117

1400012000>0000ISOOOO.C)6000<0002000500102TlhlE(SEC)250Figure3.1-15UpperCompartment Structural HeatRemovalRateCD=0.4,MinSlDonaldC.CookUnit1m319444w.wpf:1dM1195 3.1-118 10CDCA~10C)S~104$050100$50200TIME(SEC)250300Figure3.1-16LowerCompartment Structural HeatRemovalRateCD=0.4,MinSlDonaldC.CookUnit1mh19444w.wpf:1d~1195 3.1-119

$600001400001200001000008000060000O.tD400002000D-2DOOD5010015020TIME(SEC)2030Figure3.1-17HeatRemovalbySumpCD=0.4,MinSlDonaldC.CookUnit1mal944<w.wpf:1d~1185 3.1-120

10000080000m60000I40000OLIJ2000050101020TIME(SEC)25030Figure3.1-18HeatREmovalbyLowerCompartment SprayCD=0.4,MinSlDonaldC.CookUnit1I&19444w.wpf:1dM1195 3.1-121 UPPERCOMPARTMENT aaLOWERCOMPARTMENT 250200150LIJI-100'0501015020TIME(SEtl)250300Figure3.1-19Containment Temperature CD=0.4,MinSIDonaldC.CookUnit1mh1944<w.wpf:1d~

11953.1-122 6050io53020104006008001000PRESSURE(PRA)12001ioo1600Figure3.1-20SafetyInjection FlowRateDonaldC.CookUnit1mA1944<w.wpf:

1d4411953.1-123 18.016.014.012~010.0F0llI6~04~02~00'2'4'6.08-0ZLVTLTION (PT)Figure3.1-21HotRodPowerDistribution DonaldC.CookUnit1mL1944<w.wpt:1d~1195 3.1-124 2288.2888.1888.K7c)1688.4Jg1488.K~1288.~1888.888.688.488@1888,1588.TINE(SECl2888.2688.Figure3,1-22RCSPressure(3Inch}ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpf:1d441195 3.1-125 f

36.34.32.I4w28.LJ)Ldu26.OCIz24~4JOCC)22.TOPQFCORE28.18.16@588.1888.'1588.TINE(SEC)2888.Figure3.1-23CoreMixtureHeight(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpf:1d441195 3.1-126

~2588.2888.Cl1588.o~)888.L5K5.588.~~688.$88.1888.1288~$488.l688.l888.2888.2288.2488.26BB.TlHEtSEC)Figure3.1-24HotSpotCladTemperature (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wpf:1d441195 3.1-127

'258.Lalz288.ECD~158.CIwl88~CDl888.1588.TINE(SEC)2888.2688.Figure3.1-25CoreSteamFlowrate(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpf:1d~1195 3.1-12B 18>cD1824cn181188688.888.1888.1288.1488,1688.1888.2888.2288.2488.2688.TlfC(SECfFigure3.1-26HotSpotHeatTransferCoefficient (3inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpf:1d441 1953.1-129 4~2588.CI-2888.1588.y1888.o588.S.688.SBS.1888.1288.1488.1688.ISSS.2888.2288.2488.2688.TltK1SEC)Figure3.1-27HotSpotFluidTemperature (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mh19444w.wpf:1d441195 3.1-130 1B88.1688.1488.~1288.c1888.B88.688.488.288.258.588.758.1888.1258.1588.1758.2888.2258.2588.TINE(SEC)Figure3.1-28TotalBreakFlow(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mal9444w.wpf:1d441195 3.1-131 45,~~48.X:~55.o~38.D~25.x:0a.28.C)C)iS.18.2SB.588.7SB.1888.1258.1588.1758.2888.2258.2588.TINE(SEC)Figure3.1-29IntactLoopPumpedSlFlow(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpf:1d441195 3.1-132 2288.2888.-1888.4JP)1688.Kf)1488.+1288.188gl.888@1888.2888.5888.4888.5888.6888.TINE(SEC)Figure3.1-30RCSPressure(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m:51944<w.wpf:1d~1195 3.1-133 56.34,52.38.I4a28.4JLJ~26.OC2244lC)4722.TOPOFCORE28.18.1888.2888.5888.4888.TINE(SEC)5888.6888.Figure3.1-31CoreMixtureHeight(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mh19444w.wpf:1dM1195 3.1-134 4~2588.2888.ClCl1588.~1888.588.RS.2888.2588.5888.5588.48BB.4588.5888.5588.6888..T1NEtSEC)Figure3.1-32HotSpotCladTemperature (2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpf:1dM1195 3.1-135

168.148.CJ4Jz128.Zg1884C)88.IX4J68.C)48.28.1888.2888.5888.4888.5888.TINEtSEC)Figure3.1-33CoreSteamFlowrate(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpf:1dM1195 3.1-136 0

CVcD]82cn]8]2888.2588.5888.5588.4888.4588.5888.5588.6888.7lHElSEC)Figure3.1-34HotSpotHeatTransferCoefficient (2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m:11944<w.wpf:1dM1195 3.1-137

~2588.2888.l588.5tZo.1BSS.CIB.2888.2588.5888.5588..4888.l588.SISS.5588.6888.'Tlirt(SEClFigure3.1-35HotSpotFLuidTemperature (2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1S44<w.wpf:1d441195 3.1-138 888.788,688.Kw588.288.188.1888.2888.3888.4888.5888.TIME(SEC)Figure3.1-36TotalBreakFlow(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA19444w.wpf:1d441195 3.1-139 45.48.CD4JCO~38.C7L25.C7a28.815.18.2888.3888.4888.TINEtSEC)5888.6888.Figure3.1-37IntactLoopPumpedSlFlow(2Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944-3w.wpf:1d441195 3.1-140 2288.2888.'888.~1688.ac1488.a.1288,K~~1888.688.688.488.288~588.1888.1588.TINEfSEC)2588.Figure3.1-38RCSPressure(rInch)ReducedTemperature, ReducedPressureDonafdC.CookUnit1mh1944<w.wpf:1d~1195 3.1-141 56.54.52.4a28.4lLLJu26.GCIx:24.LLlOCC)CJ22.TOPOFCORE28.18'6@588.1888'.1588.TINEfSEC)2588.Figure3.1-39CoreMixtureHeight(4Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mh1944<w.wpf:1d~1195 3.1-142

~2588.-2888.ClCl1588.CI~1888,L5I:O5588.e.i88.688.888.1888.1288.fi88.1688.1888.2888.TlNElSEC1Figure3.1-40HotSpotCladTemperature (4Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpf:1dM1195 3.1-143 275.258.225.-288.LJg175.~~168.cr125.~l88.CI75.25.588.i888.1588.TiME(SEC)2588.Figure3.1<1CoreSteamFlowrate(4Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpt:1d~1 1953.1-144 Cg4.cDl82ChC/ltXIocol8lCIIICf.l88488.$88.888.l888.l288.1488.T1NC1SECIl688,l888,2888.Figure3.1-42HotSpotHeatTransferCoefficient (4inch)eReducedTemperature, ReducedPressureOonaldC.CookUnit1mA1944<N.Wpl:1d441195 3.1-145 4~2588.D2888.1588.5a.1888.KO588.8.488.688.888.1888.1288.1488.168B.1888.2888.71NEtSEC)Figure3.1-43HotSpotFluidTemperature (4Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m&1944<w.wpt:1dM1195 3.t-<46 2488.2288.2888'1888.ar.1688~a-1488.X~~1288.~1888.688.488@1888.1688.2888.2688.5888.TINE(SEC)Figure3.1-44RCSPressure(3Inch)ReducedTemperature, ReducedPressureDonaidC.CookUnit1m:51944<w.wpf:1d~

f1953.1-147 36.34,32,I4a28.4JLl~26.OCIOCE24'JC)4722.TOP0FCORE28.js.i888.1588.2888.2588.TINE(SEC)Figure3.1-45CoreMixtureHeight(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wpf:1d441195 3.1-148 cI,2SSB.2888.ClCD1SSS.R~1888.5SSB.688.888."1888.1288.1488.1688.1888.2888.2288.2488.2888.Tlat%1SKC)Figure3.1<6HotSpotCladTemperature (3Inch)ReducedTemperature, ReducedPressureDonaldC,CookUnit1m&1944<w.wpf:1d 441195'3.1-149 258.225.288'17S.X-158.o4~125.CL~188.LJ5CJ25.588.1888.1588.2888.TINE(SEC)Figure3.1<7CoreSteamFlowRate(3Inch)~~ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA19444w.wpf:1d441195 3.1-150 58~CkIOJ4582tCWCIcA585I688.888.l888.5288.5488.l688.l688.2888.2288.2488.2688.TlNElSEClFigure3.1-48HotSpotHeatTransferCoefficient (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA19444w.wpf:1d441195 3.1-161 IlIIt

~2588.2888.1588.gg~ICIa1888.KIOSSS.QSS.SSS.1888..1288.1488.1688.1888.2888.22882488.2688.71NEtSEC)Figure3.1<9HotSpotFluidTemperature (3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mh19444w.wpt:td441195 3.1-152 1888.1688.1488.~1288.~1888.hC888.688.488.288.S88.1888.1588.2888.2S88.5888.TINE(SEC)Figure3.1-50TotalBreakFlow(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m:51944<w.wpf:1d~1195 3.t-153 55.58.45.48,CDKw35.C)~S8.CDCl~25.X:CLa28.C)C)~~~5.i8.588.1888.1588.2888.TIME(SEC)Figure3.1-51IntactLoopPumpedSlFlow(3Inch)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m:11944<w.wpf:1d441195 3.1-154 0

2488,2288.2888.~1B88.ac1688.~1488.K~~1288.1888.888.688.488~1888.1S88.2888.2S88.5888.TINEtSEC)Figure3.1-52RCSPressure(3Inch)HighTemperature, HighPressureDonaldC.CookUnit1m:51944<w.wpf:1d~1195 3.1-155

36.34,32.38.I4a28.4J4J426.I~24.4JC)22.TOPOFCORE28.18.16@1888.1588.2888.TINE(SEC)Figure3.1-53CoreMixtureHeight(3Inch)HighTemperature, HighPressureDonaldC.CookUnit1m519444w.wpf:1d441 1953.1-156

5888.~2888.4J4J4lCl2888.o1688.F888.5.688.l888.f288.1488.3688~ISSS.2888.2288.2488.2688.2888.SBSS.TlNE(SEC)Figure3.1-54HotSpotCladTemperature (3Inch)HighTemperature, HighPressureDonaldC.CookUnit1m51944<w.wpf:1d441195 3.1-157 hrllHf 228.288.188.-168.CJ~148.~~128.cr188.I88.4JC788.48.28.588.1888.1588.2888.2588.TINE(SEC)Figure3.1-55CoreSteamFlowrate(3Inch)HighTemperature, HighPressureDonaldC.CookUnit1m:L1944<w.wpt:1d441195 3.1-158

le>OCAJIcD182CLCIcn1elCZlee1888.1288.1488.l688.1888.2888.2288.2488.2688.2888.5888.TllKtSEC)Figure3.1-56HotSpotHeatTransferCoefficient (3Inch)HighTemperature, HighPressureDonaldC.CookUnit1mal944<w.wpt:1 d~11953.1-159 4.~2588.Ch2888.1588.y1888.la)P588.1888..1288.1488.1688.1QSS.2888.2288.2488.2888.2888.5888.Tll1ClSEClFigure3.1-57HotSpotFluidTemperature (3Inch)HighTemperature, HighPressureOonaldG.GookUnit1mh1944-3w.wpf:1d441195 3.1-160 1688.1488.1288.Kg1888.o888.g688.I488.288.588.1888.1588.2888.TINE(SEC)2S88.5888,Figure3.1-58TotalBreakFlow(3Inch)HighTemperature, HighPressureDonaldC.CookUnit1mA19444w.wpf:1dC41195 3.1-161 I,

58.45.~48.KzKS.o~S8.o~~2S.X:lLa28.C)CI15.28.1888.1588.2888.2588.TINE(SEC)Figure3.1-59IntactLoopPumpedSlFlow(3Inch)HighTemperature, HighPressureDonaldC.CookUnit1mh1944<w.wpf:1d~1195 3.1-162 2DC315COLJlJllOCO5000100020003000TIME(SEC)Figure3.1-60RCSPressure(3Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1400050(mA1944<w.wpf:1d441195 3.1-163 4~30oc25TopofCnre20150f00020003000TIME(SEC)~0005CCFigure3.1-61CoreMixtureLevel(3Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wpf:1dC41195 3.1-164

'8"G.16"0.1400.Ll12".0.IOC4JK)000.I800.600.400500.1000.1500.2000.(s)2500.3000.350Figure3.1-62PeakCladTemperature (3Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wpf:1d~1195 3.1-165

(

531005000100020003000TIME(SEC)40005COOFigure3.1-63CoreOutletSteamFLowRate(3Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1mL1944<w.wpf:1d441195 3.1-166

oc103)01500.1000.i500.2000.2500.T:ME:(S)3000.35GGFigure3.1-64HotSpotRodSurfaceHeatTransferCoefficient (3Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1mh1944<w.wpf:1d441195 3.'I-167 2ODO.1800.1600.1400.1200.K1000.SOO.600.400500.1000.1500.2000.2500.3000.TIME(S)Figure3.1-65HotSpotFLuidTemperature (3Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1mh1944<w.wpf:1d 4411953.1-168

>5COECD~tOCO50000100020003000-IMK(SEC)4000Figure3.1-66ColdLegBreakMassFlowRate(3Inch,3%MSSYTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wpf:1d~1195 3.1-169 6:Vi>2000100020003000TIME(SEC)40005CGCeFigure3.1-67SafetyInjection MassFlowRate(3Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpf:1d441195 3.1-170 141210o'8CL468Elevation (ft)Figure3.1-68HotRodPowerDistribution ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wp1:1d441195 3.1-171 22002000180016001400120010008006001000200030004000500060007000TIME(SEC)Figure3.1-69RCSPressure(2Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1mh1944<w.wpf:1d441195 3.1-172 40LLJ3025ITopofCo20151000200030004000500060007000TIME(SEC)Figure3.1-70CoreMixtureLevel(2Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wpf:1d441195 3.1-173

2000180016001400120010008006004002000300040005000TIME(S)600070001Figure3.1-71PeakCladTemperature (2Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<w.wpt:1d441195 3.1-174

200CAcn150100o50-501000200030004000500060007000TIME(SEC)Figure3.1-72CoreOutletSteamFlowRate(2Inch,3%MSSYTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1mht9444w.wpf:td44t t953.1-175

510I4L10IiLZI~10UJCOV210WI-100102000300040005000TIME(S)60007000Figure3.1-73HotSpotRodSurfaceHeatTransferCoefficient (2Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wpf:1d441195 3.1-176 1800160014001200I-1000LUI8006004002000300040005000TIME(S)60007000Figure3.1-74HotSpotFluidTemperature (2Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA19444w.wpf:1dM1195 301177 700600500m400300C)2001001000200030004000500060007000TIME(SEC)Figure3.1-75ColdLegBreakMassFlowRate(2Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1m:119444w.wpf:1d441195 3.1-178 504030o20101000200030004000500060007000TIME(SEC)Figure3.1-76SafetyInjection MassFlowRate(2Inch,3%MSSVTolerance)

ReducedTemperature, ReducedPressureDonaldC.CookUnit1m319444w.wpf:1d441195 3.1-179

220020001800160014004J1200LaJ100080060040010002000TIME(S}30004000Figure3.1-77RCSPressure(3Inch,30IoSGTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA19444W.wpf:1d441195 3.1-180 4035LaJ4J30LJ25I2015010002000TIME(S)30004000Figure3.1-78CoreMixtureLevel(3Inch,30%SGTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mh19444w.wpf:1d441195 3.1-181 16001400l2001000LJ8006004001000150020002500TIME(Sj30003500tFigure3.1-79HotSpotCladTemperature (3Inch,30%SGTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m&1944<w.wpf:1d

~11953.1-182 250200150o10000l0002000TIME(Sj30004000Figure3.1-80CoreOutletSteamFlow(3Inch,30%SFTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA1944<wwpf:1d441 1953.1-183 510I4110ICQ3~10C)V2101i-100101000150Q20002500TIME(Sj30003500Figure3.1-81HotSpotRodSurfaceHeatTransferCoefficient (3Inch,30%SGTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA19444w.wpf:1dM1195 3.1-184 11001QQQ900800I-700CLI-600500400010002000TIME(S)30004000IFigure3.1-82HotSpotFluidTemperature (3Inch,3%SGTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mA19444w.wpf:1d441195 3,1-185 16001400~cn120010004JI800CO60040020010002000TIME(Sj30004000Figure3.1-83ColdLegBreakMassFlowRate(3inch,30%SGTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wpf:1d~1 1953.1-186 3530254JICL2015CD4cnI010002000TIME(S)30004000Figure3.1-84BrokenLoopSafetyInjection MassFlowRate(3Inch,30%SGTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit1mh1944<w.wpf:1d~1195 3.1-187 605040201010002000TIME(S)30004000Figure3.1-85LumpedIntactLoopSlMassFlowRate(3Inch,30%SGTP)ReducedTemperature, ReducedPressureDonaldC.Cook'Unit 1m%19444w.wpf:1d441195 3.1-188 3.2LOCAHYDRAULIC FORCESLOCAhydraulic forcesarerelatively insensitive tospecificsteamgenerator tubeplugginglevelsandtheassociated changesinthermaldesignflow,providedtheRCStemperatures remainunchanged.

TheLOCAhydraulic forcesanalyzedfortheReratingProgramaredocumented inSection3.2ofWCAP-11902 andusedconservative valuesof582.3'Fand511.7'FforTandT~,respectively.

TheLOCAhydraulic forcingfunctions inSection3.2ofWCAP11902conservatively boundtheRCSparameters inTable2.1-1ofthisreport,evenwithconsideration of5%asymmetric flow.WCAP-11902, Section3.2,remainsvalidandnochangesand/oradditions arerequired.

m%1944<w.wpf:1d441195 3.2-1 3.3NON-LOCAANALYSES3.3.1Introduction Thissectionevaluates theeffectsofreducedtemperature andpressureoperation withamaximumaverageSGTPlevelof30%forDonaldC.CookNuclearPlantUnit1withrespecttothenon-LOCAsafetyanalyses.

Theeffortperformed istosupportUnit1operation withacorepowerof3250MWtintherangeoffull-power reactorvesselaveragetemperatures between553'Fand576.3'Fatprimarypressurevaluesof2100psiaor2250psia(Cases1and2ofTable3.3-1,whichareidentical tocases2and3ofTable2.1-1).Thecurrentnon-LOCAanalysesofrecordforUnit1supportareratedcorethermalpowerof3411MWt(3425MWtNSSS)withafullpowervesselaveragetemperature between547'Fand578.7'Fataprimarysystempressureof2100psiaor2250psia.Cases3and4ofTable3.3-1presenttherangeofconditions supported bythecurrentnon-LOCAanalysesofrecord.It.isimportant tonotethatthecurrentnon-LOCAsafetyanalysesofrecordsupportthereratingofDonaldC.CookNuclearPlantUnit1.However,theunithasneverbeenlicensedtooperateinaccordance withtheparameters definedasCases3and4ofTable3.3-1.TheDonaldC.CookNuclearPlantUnit1licensing basis,asreportedintheUFSAR(Reference 14)includesanalysesandevaluations ofsixteennon-LOCAevents,whicharedelineated onthenexttwopages.Thislicensing-basis hasbeenreviewedtoassesstheimpactassociated withtheSGTPProgram.Thefollowing eventswere~re-analzedaspartoitheSGTPProgram:Unit1UFSARSectionAccident14.1.114.1.2Uncontrolled RCCABankwithdrawal fromaSubcritical Condition Uncontrolled RCCABankwithdrawal AtPower14.1.314.1.414.1.614.1.814.2.5RodClusterControlAssemblyMisalignment RCCADropLossofReactorCoolantFlow(Including LockedRotor)LossofExternalElectrical Loadand/orTurbineTripRuptureofaSteamPipe(coreresponseanalysis) m:11944<w.wpt:1d441195 3.3-1

~,

14.2.6RuptureofaCRDMHousing(RodEjection) 14.3.4.4MassandEnergyReleaseAnalysisforPostulated Secondafy SystemPipeRupturesInsideContainment.

Thefollowing events/analyses havebeenevaluated tosupporttheoperating conditions associated withtheSGTPprogram:Unit1UFSARSectionAccident14.1.514.1.714.1.9ChemicalandVolumeControlSystemMalfunction Start-upofanInactiveLoopLossofNormalFeedwater 14.1.1014.1.11Excessive HeatRemovalDuetoFeedwater SystemMalfunctions Excessive LoadIncreaseIncident14.1.12LossofAllACPowertothePlantAuxiliaries 14.2.8(Unit2)RuptureofaMainFeedwater Pipe14.4.11.3 Steamline BreakMass/Energy ReleaseOutsideContainment 3.3.2Non-LOCASafetyAnalysisAssumptions Requiring Technical Specification ChangesToenhanceoperating flexibility forRCSreducedtemperature andpressureoperation withamaximumaveragesteamgenerator tubepluggingof30%,certainreactorprotection systemsetpoints wererevised.Thefollowing requirements wererelaxedtoenhanceoperating flexibility aswell:EDGstarttimefromambientconditions; pressurizer codesafetyvalvesetpointtolerance; andshutdownmarginforT,~greaterthan200'F.TherevisedRPSsetpoints includetheovertemperature bT(OTET)andtheoverpower AT(OPAT)reactortrips.Thegeneralequations fortheOTATandOPBTreactortripsetpoints andthesafetyanalysislimitcoefficient valuesarepresented inTable3.3-3:AdetailedThe30%SGTPprogramalsoincludedanevaluation oftheMajorRuptureofaFeedwater Pipeevent(UFSARSection14.2.8),whichisnotpartoftheUnit1licensing basisandisprovidedforinformational purposesonly.mA1944<w.wpf:1d441195 3.3-2 discussion oftherevisedsetpointequations forthesereactortripfunctions isprovidedinSection3.3.2.1.Discussions oftheEDGstarttimerequirement, thepressurizer codesafetyvalvesetpointtolerance adjustment, andshutdownmarginrelaxation arealsopresented inthesectionsthatfollow.Theapplicable Technical Specification updatesfortheserevisions/relaxations areprovidedinAppendixA.3.3.2.1ReactorProtection SystemTripSetpoints RevisedOTETandOPATsetpoints arebaseduponnewcorethermalsafetylimits,whichaccountfortheeffectsoftheRCSparameter changesassociated withtheincreased levelofsteamgenerator tubeplugging, usingthemethodology described inReference 1.Thesesetpoints wererevisedtoincreasetheavailable marginbetweenthesafetyanalysissetpointvaluesandthenominal,orTechnical Specification values,suchthatmorehT-driftcouldbeaccommodated betweeninstrumentation calibrations duringthefuelcycle.Presently, thepowermarginassociated withtheReratingProgramisbeingutilizedtooffsetthebT-driftthatisbeingexperienced duringcorebumup(i.e.,thecorepowerof3411MWtissupported bytheanalyses, buttheplantisactuallyoperatedwithacorefull-power valueof3250MWt).However,sincethe30%SGTPparameters donothavethispowermarginavailable, therewasaneedtorevisetheOTBTandOPETsetpoints aspartoftheSGTPProgram.Figures3.3-1through3.3-4presenttheallowable reactorcoolantloopaveragetemperature andbTconditions asafunctionofprimarycoolantpressure, baseduponaminimummeasuredflow(MMF)of339,100gpmanda1.55choppedcosineaxialpowerdistribution.

Figure3.3-1represents themostlimiting30%SGTPoperating configuration (nominalfull-powerT,~=576.3'F,nominalpressure=2100psia)oftherangeofconditions described inTable3.3-1(Cases1and2)forthecalculation oftheOTLTandOPETsetpoints.

Theboundaries ofoperation definedbytheOTBTandOPITtripsarerepresented as"protection lines"onthisdiagram.Theprotection linesaredrawntoincludealladverseinstrumentation andsetpointerrorssothatundernominalconditions, atripwouldoccurwellwithintheareaboundedbytheselines.TheutilityofthisdiagramisinthefactthatthelimitimposedbyanygivenDNBRcanberepresented asaline.TheDNBlinesrepresent thelocusofconditions forwhichtheDNBRequalsthelimitvalue(1.40and1.42fortypicalandthimblecells,respectively; seeTable3.12-3).AllpointsbelowandtotheleftofaDNBlineforagivenpressurehaveaDNBRgreaterthantheSafetyAnalysisLimitDNBRvalue.ThediagramshowsthatDNBisprevented forallcasesiftheareaenclosedwiththemaximumprotection linesisnottraversed bytheapplicable SafetyAnalysisLimitDNBRatanypoint.Theareaofpermissible operation (power,pressure, andtemperature) isboundedbythecombination ofreactortrips:highneutronflux(fixedsetpoint);

highandlowRCSpressure(fixedsetpoints);

overpower andovertemperature hT(variable setpoints),

andtheopeningofm:51944<w.wpf:Id~1195 3133 tll thesteamgenerator safetyvalves,whichlimitthemaximumRCSaveragetemperature.

TheSafetyAnalysisLimitDNBRvalue(1.40typicaland1.42thimble),

whichwasusedastheDNBRlimitforallaccidents analyzedwiththeRevisedThermalDesignProcedure (RTDP;Reference 2),isconservative comparedtotheactualDesignLimitDNBRvalue(1.23and1.22fortypicalandthimblecells,respectively),

requiredtomeettheDNBdesignbasis.Table3.3-2presentsthelimitingtripsetpoints assumedintheaccidentanalysesandthetimedelayvaluesassumedforeachtripfunction.

Thedifference betweenthelimitingtrippointassumedfortheanalysisandthenormaltrippointrepresents anallowance forinstrumentation channelerrorandsetpointerror.Nominaltripsetpoints arespecified intheplantTechnical Specifications.

Duringplantstart-uptests,itisdemonstrated thatactualinstrument timedelaysareequaltoorlessthantheassumedvalues.Additionally, protection systemchannelsarecalibrated andinstrument responsetimesaredetermined periodically inaccordance withtheTechnical Specifications.

The30%SGTPeffortassumedthatthereference averagetemperatures (T'ndT")usedintheOTATandOPATsetpointequations arescaledtothefull-power averageRCStemperature eachtimethecycleaveragetemperature ischanged.Itisalsoassumedthatthereference pressure(P')intheOTZTequationissetequaltotheappropriate nominalprimarysystempressureforaparticular cycle(either2100psiaor2250psia).Theseassumptions arekeytoensurethattheactualplantconditions requiredtoresultinanOTETand/orOPATtripsignaltobegenerated areconservative withrespecttoassumptions madeinthesafetyanalyses.

Figures3.3-1through3.3-4illustrate theOTETandOPATprotection setpoints fortheendpoints oftherangeoffull-power vesselaveragetemperatures fortheSGTPProgramateither2100psiaor2250psia.Thecalibration oftheNISexcoredetectors, tocompensate forthechangesincoolantdensityeachtimethecycleoperating conditions arechanged,isalsoassumedintheanalyses.

TheOTATandOPbTreactortripfunctions provideprimaryprotection againstfuelcenterline melting,amongotherconcerns(i.e.,DNBandhot-legboiling).

Thecriterion fornofuelmeltis,theuraniumdioxidemeltingtemperature shallnotbeexceededforatleast95percentofthelimitingfuelrodsata95percentconfidence level(Reference 1).Thiscriterion ismetbylimitingthecalculated fuelcenterline temperature to4700'F(validfor60000MWD/MTUbumupperReference 16).Inmanycases,fuelcenterline meltingcanbeprevented bylimitinggrosscorethermalpowertoaprescribed limit(historically 118%ofnominalpower)independent ofaxialpowerdistribution.

Aspartofthereloadprocess(viatheReloadSafetyAnalysisChecklist, orRSAC),thepeaklinearheatgeneration rateofthecore(i.e.,peakkw/ft)isdetermined specifically forfuelcenterline meltingconcerns.

EventhoughtherevisedOTZTandOPETreactortripsetpointequations allowthetypicalgrosscoreaveragethermalpowertoslightlyexceedthehistorical valueof118%(CookUnit1-analysesindicatethatapeakoverpower of119.03%canbeachievedwiththerevisedsetpoints),

fuelcenterline meltm%1944<w.wpf:1d441195 3.3A l7Lf concernsarespecifically evaluated onacycle-by-cycle basisaspartoftheformalreloadprocesstoensurefuelcenterline meltingdoesnotoccur.SincetherevisedOTBTandOPATreactortripsetpointequations allowthetypicalgrosscoreaveragethermalpowertoslightlyexceedthis118%value,asnotedabove,thisfactwasaddressed withrespecttothesteamline break-coreresponse(SLB-CR)methodology.

Thefull-power steamline breakanalysisforcoreresponseconsiderations isnotintheDonaldC.CookNuclearPlantlicensing basis.Nevertheless, ithasbeendetermined thattherevisedOTBTandOPATsetpointequations, withtheOPZTreference averagetemperature (T")restricted tovaluesnogreaterthan563.0'F,providessufficient assurance thatminimumDNBRwillbeprotected duringaHFPSLB.3.3.2.2'mergency DieselGenerator Start-upTimeRelaxation Thoseeventsthatmustconsideralossofoffsitepower(i.e.,LossofAllACPowertotheStationAuxiliaries, andSteamlineBreakforcoreresponse) havebeenevaluated withrespecttoanincreaseintheEDGstarttimefrom10secondsto30seconds,Thisstarttimerelaxation oftheEDGhasbeenfoundtobeacceptable.

ThisfeaturealsoaffectstheMainSteamline BreakMass/Energy ReleasesInsideContainment analysis.

However,theimpactislimitedtothecontainment responseportionoftheanalysis, asthesteammassandenergyreleasecalculations arebasedontheconservative assumption thatoffsitepowerisavailable forthedurationoftheblowdown.

3.3.2.3Pressurizer CodeSafetyValveSetpointTolerance IncreaseThefollowing events,whicharepotentially impactedbyanincreaseinthepressurizer codesafetyvalvesetpointtolerance, havebeenshowntosupportanincreasefrom+1%to+3%setpointtolerance:

LossofExternalElectrical Load;LossofNormalFeedwater; LossofAllACPowertotheStationAuxiliaries; andLockedRotor/Shaft Breakevents.Thus,asetpointtolerance of+3%forthepressurizer codesafetyvalvesisacceptable.

3.3.2.4ShutdownMarginRelaxation AllofthecurrentDonaldC.CookNuclearPlantUnit1licensing-basis analyses(i.e.,theanalysessupporting theReratingProgram)thatmodelshutdownmargin(SDM)assume1.3%6k/k, exceptfortheSteamline BreakforCoreResponse(SLB-CR)event.There-analysisoftheSLB-CReventfortheSGTPProgramwasperformed withaSDMassumption of1.3%6k/k.

Assuch,allofthenon-LOCAsafetyanalysesthatmodelSDMsupportthereducedSDMvalueof1.3%6k/k.

mA1944<w.wpt:1d441195 3.3-5

'Ht]IIIf 3.3.2:5Steamline BreakProtection SystemModification Thecoincidence logic.currently requiredforsafetyinjection initiation andsteamline isolation onhighsteamflowandlowsteampressureorlow-lowT,~forUnit1istobemodifiedtomatchthatinstalled atUnit2.Thislogicispartofthesteamline breakprotection system.Adetaileddescription ofbothofthesteamline breakprotection systemscurrently installed ineachoftheunitsispresented inSection3.5.4.TheproposedUnit1modification, whichwillresultinthetwounitshavingidentical steamline breakprotection systems,consistsofreplacing Slactuation onhighsteamflowcoincident withlowsteampressure, orhighsteamflowcoincident withlow-lowT,~,withSlactuation onlowsteampressureonly.TheproposedUnit1modification alsoreplacessteamline isolation onhighsteamflowcoincident withlowsteampressurewithSLIonlowsteampressureonly.Thecoincidence requirement forhighsteamflowwithlowsteampressureofthecurrentUnit1designincreases thelikelihood thatsafeguards initiation mightbedelayedcomparedtotheproposedUnit1modifieddesign,whereonlyalowsteampressuresignalisrequired.

Inthecasewherethecoincidence logicprohibits safetyinjection andsteamline isolation onhighsteamflowwithlowsteampressure, oneoftheothersignalsmustbereceivedbeforethesafeguards areinitiated.

Assuch,thecurrently installed Unit1steamline breakprotection systemdesignassumedintheSteamline BreakMass/Energy ReleasesOutsideContainment calculations (Section3.3.4.7),

theRuptureofaSteamPipeanalysis(Section3.3.5.6),

andtheSteamline BreakMass/Energy ReleasesInsideContainment analysis(Section3.5.4)boundstheproposedmodifications totheUnit1steamline breakprotection system,astherequirement tosatisfythecoincidence, discussed above,canresultinSland/orsteamline isolation laterinatransient thanhadSland/orSLIactuatedbylowsteampressurealone.Adelayintheinitiation ofsafeguards isconservative forallthreeofthepreviously listedevents.Also,anevaluation regarding theMajorRuptureofaFeedwater PipeeventforUnit1hasbeenperformed (Section3.3.4.8),

whichspecifically addresses theproposedUnit1steamline breakprotection systemmodifications.

ItshouldbenotedthatthedeletedSlactuation

function, i.e.,highsteamflowcoincident withlow-lowT,~,isnotmodeledinanyofthenon-LOCAsafetyanalyses.

3.3.3Methodology TheUnit1non-LOCAsafetyevaluation fortheSGTPProgramwasperformed usingcurrentWestinghouse methodology andcomputercodes.Thefollowing foursub-sections discuss:theInitialConditions assumed,whichreflectsthechangefromtheImprovedThermalDesignProcedure (ITDP)(utilized fortheReratingProgramofUnit1)totheRevisedThermalDesignProcedure (RTDP)formostoftheeventsthatareDNBlimited;theComputerCodesUtilized; andthe5%RCSFlowAsymmetry.

m51944<w.wpt:

id~11953.3-6 3.3.3.1InitialConditions Formostaccidents whichareDNBlimited,nominalvaluesofinitialconditions andminimummeasuredflow(339,100gpm)areassumed.Theallowances onreactorpower,RCStemperature andpressurearedetermined onastatistical basisandareincludedinthelimitDNBRasdescribed inWCAP-11397 (Reference 2).Thisprocedure isknownasthe"RevisedThermalDesignProcedure".

Foraccidents thatarenotDNBlimitedorinwhichRTDPisnotemployedtheinitialconditions areobtainedbyaddingthemaximumsteady-state errorstoratedvalues.Thefollowing steady-state errorsareconsidered:

a.CorePower+2%calorimetric errorallowance b.AverageRCSTemperature

+4.1'Fcontroller deadbandandmeasurement errorallowance; alsoa+1.0'Fbiasforcold-legstreaming c.Pressurizer Pressure+67psisteady-state fluctuations andmeasurement errorallowance (seeparagraph below)d.ReactorFlowThermalDesignFlow(332,800gpm)Itshouldbenotedthatthepressurizer pressureuncertainty includesanallowance of23psifor"readability,"

whichisonlyapplicable forDNBconsiderations.

However,the67psiu'ncertainty wasconservatively appliedtoallnon-LOCAanalysesforsimplicity.

Thus,thereisanadditional 23psiofpressuremarginthatcanberealized, ifnecessafy, fornon-DNBevents.Table3.3-4summarizes initialconditions andcomputercodesusedintheaccidentanalysis, andshowswhichaccidents employedaDNBanalysisusingtheRTDP.3.3.3.2ComputerCodesUtilizedSummaries oftheprincipal computercodesusedinthetransient analysesaregivenbelow.Thecodesusedintheanalysisofeachtransient havebeenlistedinTable3.3-4.FACTRANFACTRANcalculates thetransient temperature distribution inacross-section ofametalcladUO,fuelrodandthetransient heatfluxatthesurfaceofthecladusingasinputthenuclearm%1944<w.wpf:1d441195 3.3-7 powerandthetime-dependent coolantparameters (pressure, flow,temperature, anddensity).

Thecodeusesafuelmodelwhichsimultaneously exhibitsthefollowing features:

A.Asufficiently largenumberofradialspaceincrements tohandlefasttransients suchasrodejectionaccidents.

B.Materialproperties whicharefunctions oftemperature andasophisticated fuel-to-cladgapheattransfercalculation.

C.Thenecessafy calculations tohandlepost-departure fromnucleateboilingtransients:

filmboilingheattransfercorrelations, Zircaloy-water

reaction, andpartialmeltingofthematerials.

.FACTRANisfurtherdiscussed inReference 3.LOFTRANTheLOFTRANprogramisusedfortransient responsestudiesofapressurized waterreactor(PWR)systemtospecified perturbations inprocessparameters.

LOFTRANsimulates amultiloop systembyamodelcontaining thereactorvessel,hotandcoldlegpiping,steamgenerators (tubeandshellsides),andthepressurizer.

Thepressurizer heaters,spray,relief,andsafetyvalvesarealsoconsidered intheprogram.Pointmodelneutronkinetics, andreactivity effectsofthemoderator, fuel,boron,androdsareincluded.

Thesecondary sideofthesteamgenerator utilizesahomogenous, saturated mixtureforthethermaltransients andawaterlevelcorrelation forindication andcontrol.Thereactorprotection systemissimulated toincludereactortripsonhighneutronflux,overpower hT,overtemperature dT,highandlowpressure, lowflow,andhighpressurizer level.Controlsystemsarealsosimulated including rodcontrol,steamdump,feedwater control,andpressurizer pressurecontrol.TheECCS,including theaccumulators, isalsomodeled.LOFTRANalsohasthecapability ofcalculating thetransient valueofDNBRbasedontheinputfromthecorelimits.Thecorelimitsrepresent theminimumvalueofDNBRascalculated fortypicalorthimblecell.LOFTRANisfurtherdiscussed inReference 4.TWINKLETheTWINKLEprogramisamulti-dimensional spatialneutronkineticscode,whichwaspatterned aftersteady-state codespresently usedforreactorcoredesign.Thecodeusesanimplicitfinite-difference methodtosolvethetwo-grouptransient neutrondiffusion equations inone,two,andthreedimensions.

ThecodeusessixdelayedneutrongroupsandcontainsamA1944<w.wpf:1d 4411953.3-8

detailedmulti-region fuel-clad-coolant heattransfermodelforcalculating pointwise Dopplerandmoderator feedbackeffects.Thecodehandlesupto2000spatialpointsandperformsitsownsteady-state initialization.

Asidefrombasiccross-section dataandthermal-hydraulic parameters, thecodeacceptsasinputbasicdrivingfunctions suchasinlettemperatures,

pressure, flow,boronconcentration, controlrodmotion,andothers.Variouseditsareprovided; e.g.,channelwisepower,axialoffset,enthalpy, volumetric surge,pointwise power,andfueltemperatures.

TheTWINKLEcodeisusedtopredictthekineticbehaviorofareactorfortransients whichcauseamajorperturbation inthespatialneutronfluxdistribution.

TWINKLEisfurtherdescribed inReference 5.THINGIVTheTHINGIVcomputerprogram,asapprovedbytheNRC,isusedtodetermine coolantdensity,massvelocity,

enthalpy, vaporvoid,staticpressure, andDNBRdistributions alongparallelflowchannelswithinareactorcoreunderallexpectedoperating conditions.

TheTHINGIVcodeisdescribed indetailinReference 6.3.3.3.35%RCSFlowAsymmetry A5%RCSflowasymmetry issupported bythenon-LOCAsafetyanalyses.

Specifically, areduction ofRCSflowinoneloopupto5%belowthenominalaverageperloopflowrateisacceptable, aslongasthetotalminimummeasuredRCSflowisequaltoorgreaterthan339,100gpm.Shouldmorethanoneloopbebelowthe84,775gpm/loopflowrate,thesumoftheloopflowshortfalls canbenogreaterthan5%ofoneloop.Thenon-LOCAeventsthatpotentially aresensitive toasymmetric RCSflowinclude:Uncontrolled RCCABankWithdrawal fromaSubcritical Condition (RWFS),PartialLossofForcedReactorCoolantFlow(PLOF),ReactorCoolantPumpLockedRotor/Shaft Break(LockedRotor),LossofNormalFeedwater, Excessive HeatRemovalDuetoFeedwater SystemMalfunctions, LossofAllACPowertotheStationAuxiliaries, Steamline BreakforCoreResponse, RodEjectionatzeropowerconditions (HZPRodEjection),

andRuptureofaMainFeedwater Pipe.Thebalanceofthenon-LOCAeventsarenotsensitive toRCSflowasymmetry.

Thefollowing eventsexplicitly accounted fortheeffectsofasymmetric RCSflowaspartoftheSGTPProgramanalyses:

RWFS,PLOF,LockedRotor,andHZPRodEjection.

Specifically, thePLOFandLockedRotoranalysesmodelthefaulttooocurintheloopwiththehighestflow(i.e.,5%abovethenominalperloopminimummeasuredflow).Thus,thelowflowreactortrip(conservatively assumedasapercentage ofnominalflowasopposedtoapercentage ofmh1944<w.wpf:

1d4411953.3-9 normalized flow)isdelayedasmuchaspossible, alsothelargestoverallflowreduction isobtainedwiththismodel.FortheRWFSandRodEjectionanalyses, aRCSflowcorresponding totwooutoffourreactorcoolantpumpsinservice(Mode3flow)isassumed.Aconservative flowfractionisused,whichboundsaworstcase5%flowasymmetry scenarioinMode3wheretheloopsthatwouldbeproviding themostflowareoutofservice.Thesafetyanalysiscriteriaforalloftheaforementioned reanalyses continued tobemetafterexplicitly accounting fortheasymmetric RCSflow.Theremainder oftheeventsthatarepotentially sensitive toasymmetric RCSflow,butdidnotexplicitly accountfortheeffectsinthespecificanalysis, havebeenevaluated andwerefoundtobeabletoaccommodate aRCSflowasymmetry of5%.Thus,itcanbeconcluded that5%RCSflowasymmetry issupported (eitherdirectlyorindirectly) bytheCookNuclearPlantUnit1non-LOCAsafetyanalysesandevaluations.

3.3.4Non-LOCASafetyEvaluation:

Transients Evaluated Thesectionsthatfollowcontainthedetaileddescriptions oftheimpactoftheSGTPProgramontheapplicable non-LOCAtransients.

Thisfirstgroupingoftransients arethosewhichcouldbeevaluated andthesecondgrouping(Section3.3.5)aretransients whichrequiredre-analysis.

Inallcasestheappropriate UFSARacceptance criteriaaresatisfied.

3.3.4.1ChemicalandVolumeControlSystemMalfunction Reactivity canbeaddedtothecorebyfeedingpnmarygradewaterintotheRCSviathereactormakeupportionoftheCVCS.Borondilutionisamanualoperation.

Aboricacidblendsystemisprovidedtopermittheoperatortomatchtheboronconcentration ofreactorcoolantmakeupwaterduringnormalchargingtothatintheRCS.TheCVCSisdesignedtolimit,evenundervariouspostulated failuremodes,thepotential rateofdilutiontoavaluewhichprovidestheoperatorsufficient timetocorrectthesituation inasafeandorderlymanner.TheopeningofthePrimaryWaterMakeupControlValvesupplieswatertotheRCSwhichcandilutethereactorcoolant.Inadvertent dilutioncanbereadilyterminated byclosingthisvalve.InorderformakeupwatertobeaddedtotheRCS,atleastonechargingpumpmustalsoberunninginadditiontotheprimarywaterpumps.Therateofadditionofunborated watermakeuptotheRCSislimitedbythecapacityoftheprimarywaterpumps.Themaximumadditionrateinthiscaseis225gpmwithbothprimarywaterpumpsrunning.The225gpmreactormakeupwaterdeliveryrateisbasedonapressuredropcalculation comparing thepumpcurveswith.thesystemresistance curve.Thisisthemaximumdeliverybasedontheunitpipinglayout.Normally, onlyoneprimarywatersupplypumpisoperating whitetheotherisonstandby.m:41944<w.wpf:1d4411953.3-10 Jt Theboricacidfromtheboricacidtankisblendedwithprimarygradewaterintheblenderandthecomposition isdetermined bythepresetflowratesofboricacidandprimarygradewateronthecontrolboard.,Inordertodilute,twoseparateoperations arerequired.

First,theoperatormustswitchfromtheautomatic makeupmodetothedilutemode;second,thestartbuttonmustbedepressed.

Omittingeitherstepwouldpreventdilution.

Thismakesthepossibility ofinadvertent dilutionveryremote.Information onthestatusreactorcoolantmakeupiscontinuously available totheoperator.

Lightsareprovidedonthecontrolboardtoindicatetheoperating condition ofpumpsintheCVCS.Alarmsareactuatedtowarntheoperatorifboricacidordemineralized waterflowratesdeviatefrompresetvaluesasaresultofsystemmalfunction.

Tocoverthephasesoftheplantoperation andtoaccountforthereduction inthevolumeoftheRCSduetotheincreaseinthelevelofSGTPupto30%,borondilutionduringstartupandpoweroperation wereexamined.

Includedintheevaluation wastheeffectofthedifference inthedensityofunborated makeupwaterandthedensityofthereactorcoolant.Theevaluation istoshowthat,frominitiation oftheevent,sufficient timeisavailable toallowtheoperatortodetermine thecauseoftheadditionandtakecorrective actionbeforeexcessive shutdownmarginislost.ResultsandConclusions Becauseofthestepsinvolvedinthedilutionprocess,anerroneous dilutionisconsidered highlyunlikely.

Nevertheless, ifitdoesoccur,numerousalarmsandindications areavailable toalerttheoperatortothecondition.

Themaximumreactivity additionduetothedilutionisslowenoughtoallowtheoperatortodetermine thecauseoftheadditionandtakecorrective actionbeforeexcessive shutdownmarginislostforthephasesoftheplantoperation (start-up andat-power) affectedbytheSGTPProgram.3.3.4.2StartupofanInactiveLoopInaccordance withTechnical Specification 3/4.4.1,CookNuclearPlantUnit1operation duringModes1and2withlessthanfourreactorcoolantloopsisnotpermitted.

Sincethreeloopoperation duringModes1and2isprohibited, theStartupofanInactiveLoopeventdoesnothavetobeconsidered aspartofthe30%SGTPProgram.3.3.4.3LossofNormalFeedwater Alossofnormalfeedwater (frompumpfailures, valvemalfunctions, orlossofoffsiteACpower)resultsinareduction incapability ofthesecondary systemtoremovetheheatmA1944<w.wpf:

1d~11953.3-11

generated inthereactorcore.Ifanalternative supplyoffeedwater werenotsuppliedtotheplant,coreresidualheatfollowing reactortripwouldheattheprimarysystemwatertothepointwherewaterrelieffromthepressurizer wouldoccur,resulting inasubstantial lossofwaterfromtheRCS.Sincetheplantistrippedwellbeforethesteamgenerator heattransfercapability isreduced,theprimarysystemvariables neverapproachaDNBcondition.

Thereactortriponlow-lowwaterlevelinanysteamgenerator providesthenecessary protection againstalossofnormalfeedwater.

Theauxiliary feedwater systemisstartedautomatically.

Theturbinedrivenauxiliary feedwater pumputilizessteamfromthesecondary systemandexhauststotheatmosphere.

Themotordrivenauxiliary feedwater pumpsaresuppliedbypowerfromthedieselgenerators ifalossofoffsitepoweroccurs.Thepumpstakesuctiondirectlyfromthecondensate storagetankfordeliverytothesteamgenerators.

Anevaluation ofthesystemtransient hasbeenperformed toshowthatfollowing alossofnormalfeedwater, withiriitialplantconditions consistent withthosedefinedintheSGTPProgram,theauxiliary feedwater systemiscapableofreturning theplanttoasafecondition byremovingthestoredandresidualheat,thuspreventing eitheroverpressurization oftheRCSoruncoveryofthecore.Theresultsoftheevaluation demonstrate thattheLossofNormalFeedwater eventcansupportthe30%SGTPconditions.

Thelimitingpeakpressurizer leveloccursunderlowT,~(553'F)conditions.

Thisisconsistent withthecurrentLossofNormalFeedwater analysisofrecordperformed aspartoftheReratingProgram.Theconclusions presented intheDonaldC.CookNuclearPlantUnit1UFSAR(Reference 14)remainapplicable for30%SGTPconditions, sincetheLossofNormalFeedwater analysisunderreratedconditions (i.e.,Cases3and4ofTable3.3-1)yieldmoresevereresultsthanthoseobtainedfromthesensitivity casesinvestigated fortheSGTPProgram.Thisisduetothebenefitsfromthepowerlevelreduction (3411MWt~3250MWt)andtheincreaseinthelowerboundT,~(547'F~553'F)morethanoffsetstheheatremovalpenalties causedbytheincreaseinSGTPlevel(15%~30%)andthethermaldesignflowreduction (354,000gpm-+332,800gpm).3.3.4.4Excessive HeatRemovalduetoFeedwater SystemMalfunctions Reductions infeedwater temperature oradditions ofexcessive feedwater aremeansofincreasing corepowerabovefullpower.Suchtransients areattenuated bythethermalcapacityofthesecondary plantandoftheRCS.TheOverpower-Overtemperature Protection (highneutronflux,overpower dT,andovertemperature hTAnps)preventsanypowerincreasewhichcouldleadtoDNBRlessthanminimumallowable valueintheeventthatthesteamgenerator High-High LevelProtection hasnotbeenactuated.

m:11944<w.wpf:

1d~11953.3-12 Excessive feedwater flowmaybecausedbyfullopeningofafeedwater controlvalveduetoaFeedwater ControlSystemmalfunction oranoperatorerror.Atpowerconditions, thisexcessflowcausesagreater.loaddemandontheRCSduetoincreased subcooling inthesteamgenerator.

Withtheplantatnoloadconditions, theadditionofcoldfeedwater maycauseadecreaseinRCStemperature andthusareactivity insertion duetotheeffectsofthenegativemoderator coefficient ofreactivity.

Theexcessive heatremovalduetoFeedwater SystemMalfunction eventsareexaminedprimarily todemonstrate coreprotection.

For.theSGTPProgram,anevaluation ofthesystemtransient hasbeenperformed toshowthatacceptable consequences willoccurintheeventofanexcessive feedwater

addition, duetocontrolsystemmalfunction oroperatorerrorwhichallowsoneormorefeedwater controlvalve(s)toopenfully.Thisevaluation considered bothatpowerandzeropowerscenarios withthereactorbeingoperatedunderbothautomatic andmanualrodcontrolconditions.

Afeedwater malfunction eventasdescribed aboveresultsinanincreaseintherateatwhichheatisremovedfromthereactorcoolant.Anincreaseintheleveloftubeplugginginthesteamgenerators resultsinareduction intheheattransfercharacteristics betweentheprimarycoolantandthesteamsystem.Thus,alessseverecooldownwouldbeexperienced forthiseventunderthe30%SGTPconditions.

However,theRCSflowreduction duetothelargernumberoftubesbeingpluggedisaDNBpenaltyfortheatpowerevents.Furthermore, thereduction incorepowerfromthereratedvalueof3411MWtto3250MWtprovidesaDNBbenefit.Theevaluation performed fortheSGTPProgramconservatively ignoredthebenefitassociated withthereducedabilityoftheexcessive feedwater flowtocooltheprimarycoolant.Theevaluation conservatively minimized thebenefitassociated withtheratedthermalpowerreduction andconservatively maximized thepenaltyduetotheRCSflowreduction withrespecttothepowerandflowvaluesassumedintheanalysesofrecord.Theparameters assumedintheanalysesofrecordforthefeedwater malfunction eventsareconsistent withthosepresented asCases3and4ofTable3.3-1.Theevaluation concluded thatboththeatpowerandzeropowerfeedwater malfunction transients cansupport30%SGTPconditions.

Thereactivity insertion rateassumedinthecurrentUFSARanalysis(120pcm/sec)forthezeropowereventcontinues tobeconservative forthe30%SGTPconditions.

ItshouldbenotedthattherevisedOTbTandOPbTsetpointequations donotimpactthefeedwater malfunction events,duetothefactthattheanalysesdonottakecreditfortheprotection offeredbythesetripfunctions.

Theconclusions presented intheDonaldC.CookNuclearPlantUnit1UFSARfortheExcessive HeatRemovalDuetoFeedwater SystemMalfunctions (UFSARSection14.1.10)remainapplicable.

m:11944<w.wpf:1d~1195 3.3-13 3.3.4.5Excessive IncreaseinSecondary SteamFlowAnexcessive loadincreaseincidentisdefinedasarapidincreaseinsteamflowthatcausesapowermismatchbetweenthereactorcorepowerandthesteamgenerator loaddemand.Thereactorcontrolsystemisdesignedtoaccommodate a10%steploadincreaseanda5%perminuteramploadincreaseintherangeof15to100%offullpower.Anyloadingrateinexcessofthesevaluesmaycauseareactortripactuatedbythereactorprotection system.Thisaccidentcouldresultfromeitheranadministrative violation suchasexcessive loadingbytheoperatororanequipment malfunction inthesteamdumpcontrolorturbinespeedcontrol.Duringpoweroperation, steamdumptothecondenser iscontrolled byreactorcoolantcondition signals,i.e.,highreactorcoolanttemperature indicates aneedforsteamdump.Asinglecontroller malfunction doesnotcausesteamdump;aninterlock isprovidedwhichblockstheopeningofthevalvesunlessalargeturbineloaddecreaseorturbinetriphasoccurred.

Protection againstanexcessive loadincreaseaccidentisprovidedbythefollowing reactorprotection systemsignals:Overpower ATOvertemperature d,TPowerrangehighneutronfluxLowpressurizer pressureAnexcessive increaseinsteamloadresultsinanincreaseintherateatwhichheatisremovedfromthereactorcoolant.Anincreaseintheleveloftubeplugginginthesteamgenerators resultsinareduction intheheattransfercharacteristics betweentheprimarycoolantandthesteamsystem.Thus,alessseverecooldownwouldbeexperienced forthiseventunderthe30%SGTPconditions.

However,theRCSflowreduction duetothelargenumberoftubesbeingpluggedisaDNBpenaltyforthisevent.Conversely, thereduction inthecorepowerfromthereratedvalueof3411MWtto3250MWtprovidesaDNBbenefit.Theevaluation performed fortheSGTPProgramconservatively ignoredthebenefitassociated withthereducedabilityoftheexcessive steamflowtocooltheprimarycoolant.Theevaluation conservatively minimized thebenefitassociated withtheratedthermalpowerreduction andconservatively maximized thepenaltyduetotheRCSflowreduction withrespecttothepowerandflowvaluesassumedintheanalysesofrecord.Theparameters assumedintheanalysesofrecordfortheExcessive LoadincreaseIncidentareconsistent withthosepresented asCases3and4ofTable3.3-1....

m%1944<w.wpf:1d~1195 3.3-14 Theevaluation concluded thattheExcessive LoadIncreaseIncidentcansupportthe30%SGTPconditions.

TherevisedOTETandOPBTsetpointequations donotimpactthisevent,asthecurrentanalysisofrecordresultedintheplantreachingastabilized condition atthehigherpowerlevel,i.e.,noreactortripoccurredforthisevent.Theconclusions presented intheDonaldC.CookNuclearPlantUnit1UFSARfortheExcessive LoadIncreaseIncident(UFSARSection14.1.11)remainapplicable.

3.3.4.6LossofAllACPowertotheStationAuxiliaries ThelossofallACpowertothestationauxiliaries event,aswiththelossofnormalfeedwater

incident, isalimitingtransient withrespecttopressurizer overfill~Thedecreaseinprimarytosecondary heattransferability,duetotheincreaseinSGTP,aggravates theheatupportionofthetransient, andincreases thepotential forfillingthepressurizer.

Assuch,thelossofallACpowertothestationauxiliaries isevaluated fortheSGTPProgram.Acompletelossofall(non-emergency)

ACpower(e.g.offsitepower)mayresultinthelossofallpowertotheplantauxiliaries, i.e.,theRCPs,condensate pumps,etc.Thelossofpowermaybecausedbyacompletelossoftheoffsitegridaccomplished byaturbinegenerator tripatthestation,orbyalossoftheonsiteACdistribution system.Thistransient isanalyzedtoshowtheadequacyoftheheatremovalcapability oftheauxiliary feedwater system.Thetransient ismoreseverethanthelossofloadeventanalyzedbecauseinthiscasethedecreaseinheatremovalbythesecondary systemisaccomplished byaflowcoastdown whichfurtherreducesthecapacityoftheprimarycoolanttoremoveheatfromthecore.Thereactorwilltripdueto:(1)turbinetrip;(2)uponreachingoneofthetripsetpoints intheprimaryandsecondary systemsasaresultoftheflowcoastdown anddecreaseinsecondary heatremoval;or(3)duetolossofpowertothecontrolroddrivemechanisms asaresultofthelossofpowertotheplant.Following alossofpowerwithturbineandreactortrips,thesequencedescribed belowwilloccur:A.Plantvitalinstruments aresuppliedfromemergency DCpowersources.B.Asthesteamsystempressurerisesfollowing thetrip,thesteamgenerator power-operated reliefvalvesmaybeautomatically openedtotheatmosphere.

Thecondenser isassumednottobeavailable forsteamdump.Ifthesteamflowratethroughthepowerreliefvalvesisnotavailable, thesteamgenerator safetyvalvesmaylifttodissipate thesensibleheatofthefuelandcoolantplustheresidualdecayheatproducedinthereactor.mA1944<w.wpf:1d~1195 3.3-15

Asthenoloadtemperature isapproached, thesteamgenerator power-operated reliefvalves(orsafetyvalves,ifthepoweroperatedreliefvalvesarenotavailable) areusedtodissipate theresidualdecayheatandtomaintaintheplantatthehotstandbycondition.

Thestandbydieselgenerators, startedonlossofvoltageontheplantemergency busses,begintosupplyplantvitalloads.Themotordrivenauxiliary feedwater pumpsaresuppliedpowerbythedieselsandtheturbine-driven pumputilizessteamfromthemainsteamsystem.Bothtypepumpsaredesignedtosupplyratedflowwithin80secondsoftheinitiating signalevenifalossofallnon-emergency ACpoweroccurssimultaneously withlossofnormalfeedwater.

Theturbineexhauststheusedsteamtotheatmosphere.

Theauxiliary feedwater pumpstakesuctionfromthecondensate storagetankfordelivertothesteamgenerators.

Following theRCPcoastdown causedbythelossofACpower,thenaturalcirculation capability oftheRCSwillremovedecayheatfromthecore,aidedbyauxiliang feedwater inthesecondang system.Theresultsofanevaluation ispresented heretoshowthatthenaturalcirculation flowintheRCS,following alossofallACpowertothestationauxiliaries withinitialplantconditions consistent withthosedefinedintheSGTPProgramissufficient toremovedecayheatfromthecore.Theresultsoftheevaluation demonstrate thattheLossofAIIACPowertotheStationAuxiliaries (LOOP)eventcansupportthe30%SGTPconditions.

Thelimitingpeakpressurizer leveloccursunderlowT,(553'F)conditions.

Thisisconsistent withthecurrentLOOPanalysisofrecord,whichwasperformed fortheReratingProgram.Theconclusions presented intheDonaldC.CookNuclearPlantUnit1UFSAR(Reference 14)remainapplicable forthe30%SGTPconditions, sincetheLOOPanalysisunderreratedconditions (i.e,Cases3and4ofTable3.3-1)yieldmoresevereresultsthanthoseobtainedfromthesensitivity casesinvestigated fortheSGTPProgram.Thisisduetothebenefitsfromthe'owerlevelreduction (3411MWT-+3250MWt)andtheincreaseinthelowerboundT,~(547'F~553'F)morethanoffsetstheheatremovalpenalties causedbytheincreaseinSGTPlevel(15%~30%),theadditional delayinAFWdelivery(60seconds-+80seconds)duetotherelaxedEDGstarttimedelay(10seconds~30seconds),

andthethermaldesignflowreduction (354,000gpm~332,800gpm).3.3.4.7Steamline BreakMass/Energy ReleasesOutsideContainment Theexistingmassandenergy(M/E)releasesfollowing asteamline break(SLB)outsidecontainment wereperformed tosupporttherangeofconditions possiblefortheReratingProgramofUnit1(Cases3and4ofTable3.3-1),aswellastopositionUnit2forapotential futureuprating(i.e.,3600MWtNSSS).Thus,theM/Ereleasesarebaseduponaratedm51944<w.wpf:1d441195 3.3-16 thermalpowerof3600MWt,Thecorereactivity parameters werechosentoconservatively maximizethe'reactivity feedbackeffectsofthecooldownresulting fromablowdownfromeitherDonaldC.CookNuclearPlantunit.Thechangesassociated withtheSGTPProgramforUnit1,i.e.,RCSflowreduction, reducedprimary-to-secondary heattransfercapability, andreduction intheratedthermalpower,arelesslimitingparameters relativetotheassumptions currently madefortheM/Ereleasecalculations following aSLBoutsidecontainment.

Furthermore, theadjustment intheK,safetyanalysisvalueoftheOPbTsetpointequation(discussed inSection3.3.2.1)doesnotimpacttheSLBM/EReleaseOutsideContainment

analysis, whichistheonlynon-LOCAsafetyanalysisthatreliesonthistripfunctionforprimaryprotection.

Thisisbecauseaconservatively largerK,valueof1.18wasassumedintheSLBM/EReleaseOutsideContainment analysis.

TherevisedsafetyanalysisK,valueforUnit1is1.172.TheincreaseintheEDGstarttimedelayfrom10secondsto30secondshasnoeffectonthisanalysisaswell,sinceitisconservative tomaintainoffsitepowersuchthatreactorcoolantpumpoperation ismaintained (whichaidsinmaximizing thesteamreleases).

Therefore, itcanbeconcluded thatthecurrentlicensing basisoutsidecontainment SLBM/Ereleases(UFSARSection14.4.11.3) continuetoboundUnit1operation asdefinedbytheSGTPProgram.Itiskeytonoticethattheexistingoutsidecontainment SLBM/Ereleaseanalysis-of-record becamepartoftheCookNuclearPlantUnit1licensing basis(andUnit2forthatmatter)following theapprovaloftheBoronInjection Tank(BIT)Removalsubmittal (Reference 18).TheexistingSLBM/EReleaseOutsideContainment analysisassumed:a.End-of-life shutdownmarginof1.3IDk/katno-load,equilibrium xenonconditions, andthemostreactiveRCCAstuckinitsfullywithdrawn position.

Minimumcapability fortheinjection ofboricacidsolutioncorresponding tothemostrestrictive singlefailureinthesafetyinjection system.TheECCSconsistsofthefollowing systems:1)thepassiveaccumulators, 2)thelowheadsafetyinjection (residual heatremoval)system,3)thehighheadsafetyinjection system,and4)thechargingsystem.Onlythechargingsystemandthepassiveaccumulators aremodeledforthesteamline breakaccidentanalysisforM/Ereleasesoutsidecontainment.

Centrifugal Chargingpumpheaddegradation of10%wasassumed.Coincidence logicrequiredforSlandSLIconsistent withthecurrentUnit1steamline breakprotection system.Aproposedmodification willchangethelogicassociated withthissystem.However,asdiscussed inSection3.3.2.5,thecurrentanalysis, whichassumesthecqrrantUnit1steamline breakprotection system,boundstheproposedmodifications totheUnit1steamline breakprotection system.mh1944<w.wpf:1d441195 3.3-17 3.3.4.8MajorRuptureofaFeedwater PipeThefeedlinebreakeventiscurrently presented intheUnit1UFSAR(Section14.2.8)for"informational purposesonly,"asthiseventisnotpartoftheUnit1licensing-basis.

However,CookUnit2doeshavethefeedlinebreakeventinitslicensing scope.TheSGTPProgramforUnit1includesanevaluation todemonstrate thattheresponseofUnit1toafeedlinebreakisboundedbytheexistingUnit2feedlinebreakanalysis.

Akeystipulation forthisevaluation isthattheUnit1steamline breakprotection logic,whichiscurrently classified asthe"Old"system,mustbemodifiedtomatchthe"Hybrid"steamline breakprotection logicthatisinplaceatCookNuclearPlantUnit2.Adetailedevaluation wasperformed, whichincludedsensitivity casesusingtheLOFTRANcode.Theevaluation specifically assessedtheplantparameter changesassociated withtheSGTPProgram(Cases1and2ofTable3.3-1)relativetotheUnit2parameters.

corresponding toreratedconditions (i.e.,3600MWtNSSS).Sensitivity casesinvestigated theeffectsofincreasing thepressurizer safetyvalvesetpointtolerance froma1%toa3%andincreasing theEDGstarttimedelayfrom10secondsto30seconds.Thesensitivity casesalsoincludedtheeffectsassociated withthesetpointtolerance increasing fromt1%tot3%,whichhasbeenpreviously evaluated (Reference 19).Theevaluation concluded thattheresultspresented intheDonaldC.CookNuclearPlantUnit2UFSARfortheMajorRuptureofMainFeedwater Pipeevent(UFSARSection14.2.8)areapplicable toUnit1,providedthatthesteamline breakprotection logicinstalled atUnit1ismodifiedtomatchthatinstalled inUnit2.Furthermore, theevaluation concluded thatanincreaseintheEDGstarttimefrom10secondsto30seconds;anincreaseinthepressurizer safetyvalvesetpointtolerance froma1%toa3%,aswellastheinclusion ofthe1.0'Fbiastoaccountforthecold-legstreaming phenomenon canbeaccommodated.

mA1944<w.wpf:1d441195 3.3-18

3.3.5Non-LOCASafetyEvaluation:

Transients AnalyzedThesubsections thatfollowcontainthedetailsoftheaccidents re-analyzed tosupport30%SGTPoperation ofUnit1.Inallcases,theapplicable UFSARacceptance criteriaaresatisfied.

3.3.5.1Uncontrolled RCCAWithdrawal FromASubcritical Condition Theuncontrolled RCCAwithdrawal fromasubcritical condition eventisanalyzedtodetermine theimpactofthereducedRCSflowasaresultoftheincreased steamgenerator tubeplugginglevelof30%.Thiseventisanalyzedtodemonstrate coreprotection.

Althoughtheno-loadtemperature doesnotchangefortheSGTPProgram,thereduction innominalRCSflowisnon-conservative withrespecttotheDNBtransient.

AnRCCAwithdrawal incidentisdefinedasanuncontrolled additionofreactivity tothereactorcorebywithdrawal ofRCCAbanksresulting inapowerexcursion.

Whiletheoccurrence ofatransient ofthistypeishighlyunlikely, suchatransient couldbecausedbyamalfunction oftheReactorControlorControlRodDriveSystems.Thiscouldoccurwiththereactoreithersubcritical oratpower.The"atpower"caseisdiscussed inSection3.3.5.2.Reactivity isaddedataprescribed andcontrolled rateinbringingthereactorfromashutdowncondition toalowpowerlevelduringstartupbyRCCAbankwithdrawal.

Althoughtheinitialstartupprocedure usesthemethodofborondilution, thenormalstartupiswithRCCAbankwithdrawal

~RCCAbankmotioncancausemuchfasterchangesinreactivity thancanbemadebychangingboronconcentration.

Thecontrolroddrivemechanisms arewiredintopreselected banks,andthesebankconfigurations arenotalteredduringthecorelife.TheRCCA'saretherefore physically prevented frombeingwithdrawn inotherthantheirrespective banks.Powersuppliedtotherodbanksiscontrolled suchthatnomorethantwobankscanbewithdrawn atanytime.TheRCCAdrivemechanism isofthemagneticlatchtypeandthecoilactuation issequenced toprovidevariablespeedrodtravel.Themaximumreactivity insertion rateisanalyzedbyassumingthesimultaneous withdrawal ofthecombination ofthetwobanksofthemaximumcombinedworthatmaximumspeed.Shouldacontinuous controlrodassemblywithdrawal beinitiated, thetransient willbeterminated bythefollowing reactortripfunctions.

Sourcerangeneutronfluxleveltrip-actuatedwheneitheroftwosourcerangechannelsindicates afluxlevelaboveapreselected, manuallyadjustable value.Thistripfunctionmaybemanuallybypassedwheneitherintermediate rangefluxchannelindicates afluxlevelabovethesourcerangecutofflevel.Itismh1944<w.wpf:1d~1195 3.3-19

automatically reinstated whenbothintermediate rangechannelsindicateafluxlevelbelowthesourcerangecutofflevel.Intermediate rangeneutronfluxleveltrip-actuatedwheneitheroftwointermediate rangechannelsindicates afluxlevelaboveapreselected, manuallyadjustable value.Thistripfunctionmaybemanuallybypassedwhentwoofthefourpowerrangechannelarereadingaboveapproximately 10percentoffullpowerfluxandisautomatically reinstated whenthreeofthefourpowerrangechannelsindicateafluxlevelbelowthisvalue.Powerrangeneutronfluxleveltrip(lowsetting)-actuatedwhentwooutofthefourpowerchannelsindicateafluxlevelaboveapproximately 25percentoffullpowerflux.Thistripfunctionmaybemanuallybypassedwhentwoofthefourpowerrangechannelsindicateafluxlevelaboveapproximately 10percentoffullpowerfluxandisautomatically reinstated whenthreeofthefourchannelsindicateafluxlevelbelowthisvalue.Powerrangeneutronfluxleveltrip(highsetting)-actuatedwhentwooutofthefourpowerrangechannelsindicateafluxlevelaboveapresetsetpoint.

Thistripfunctionisalwaysactive.Inaddition, controlrodstopsonhighintermediate rangefluxlevelandhighpowerrangefluxlevelservetodiscontinue rodwithdrawal andpreventtheneedtoactuatetheintermediate rangefluxleveltripandthepowerrangefluxleveltrip,respectively.

Theneutronfluxresponsetoacontinuous reactivity insertion ischaracterized byaverfastpowerriseterminated bythereactivity feedbackeffectofthenegativefueltemperature coefficient.

Thisself-limitation oftheinitialpowerburstresultsfromafastnegativefueltemperature feedback(Dopplereffect)andisofprimeimportance duringastartupincidentsinceitlimitsthepowertoatolerable levelpriortoprotective action.Aftertheinitialpowerburst,theneutronfluxismomentarily reducedandthen,iftheincidentisnotterminated byareactortrip,theneutronfluxincreases again,butatamuchslowerrate.Termination ofthestartupincidentbythepreviously discussed protection channelspreventscoredamage.Inaddition, thereactortripfrompressurizer highpressureservesasabackuptoterminate theincidentbeforeanoverpressure condition couldoccur.MethodofAnalsisTheanalysisoftheuncontrolled RCCAbankwithdrawal fromsubcritical accidentisperformed inthreestages:firstanaveragecorenuclearpowertransient calculation, thenanaveragecoreheattransfercalculation, andfinallythedeparture fromnucleateboilingratio(DNBR)m:<1944<w.wpf:

td~11953.3-20 calculation.

Theaveragecorenuclearcalculation isperformed usingspatialneutronkineticsmethods(TWINKLE) todetermine theaveragepowergeneration withtimeincluding thevarioustotalcorefeedbackeffects,i.e.,Dopplerreactivity andmoderator reactivity.

Theaverageheatfluxandtemperature transients aredetermined byperforming afuelrodtransient heattransfercalculation inFACTRAN.TheaverageheatfluxisnextusedinTHINGIVfortransient DNBRcalculations.

Analysisofthistransient incorporates theneutronkinetics, including sixdelayedneutrongroupsandthecorethermalandhydraulic equations.

Inadditiontotheneutronfluxresponse, theaveragefuel,cladandwatertemperature, andalsotheheatfluxresponse, arecomputed.

Inordertogiveconservative resultsforastartupincident, thefollowing additional assumptions aremadeconcerning theinitialreactorconditions:

Sincethemagnitude oftheneutronfluxpeakreachedduringtheinitialpartofthetransient, foranygivenrateofreactivity insertion, isstronglydependent ontheDopplerpowerreactivity coefficient, aconservatively lowvalue(i.e.,smallinabsolutevalue)isusedforthestartupincident(-0.9x10Ak/%power).Thecontribution ofthemoderator reactivity coefficient isnegligible duringtheinitialpartofthetransient becausetheheattransfertimeconstantbetweenthefuelandthemoderator ismuchlongerthantheneutronfluxresponsetimeconstant.

However,aftertheinitialneutronfluxpeak,thesucceeding rateofpowerincreaseisaffectedbythemoderator temperature reactivity coefficient.

Althoughduringnormaloperation (100%ratedpower)themoderator coefficient willnotbepositiveatanytimeincorelife,ahighlyconservative valuehasbeenusedintheanalysistoyieldthemaximumpeakcoreheatflux.Theanalysisisbasedonamoderator coefficient whichwasatleast+5pcm/'Fatthezeropowernominalaveragetemperature, andwhichbecamelesspositiveforhighertemperatures.

Thiswasnecessary sincetheTWINKLEcomputercodeusedintheanalysisisadiffusion theorycoderatherthanapointkineticsapproximation andthemoderator temperature feedbackcannotbeartificially heldconstantwithtemperature.

3.Thereactorisassumedtobeathotzeropower(547'F).Thisassumption ismoreconservative thanthatofalowerinitialsystemtemperature.

Thehigherinitialsystemtemperature yieldsalargerfueltowaterheattransfer, alargerfuelthermalcapacity, andalessnegative(smallerabsolutemagnitude)

Dopplercoefficient.

ThelessnegativeDopplercoefficient reducestheDopplerfeedbackeffecttherebyincreasing theneutronfluxpeak.Thehighneutronfluxpeakcombinedwithahighfuelthermalcapacityandlargerthermalconductivity m:51944<w.wpi:1d~11953.3-21

yieldsalargerpeakheatflux.Initialmultiplication factor(k,)isassumedtobecloselyapproaching 1.0sincethisresultsinthemaximumneutronfluxpeak.Tworeactorcoolantpumpsareassumedtobeinoperation.

Thisisconservative withrespecttotheDNBtransient.

Themostadversecombination ofinstrumentation andsetpointerrors,aswellasdelaysfortripsignalactuation andcontrolrodassemblyrelease,aretakenintoaccount.A10%increasehasbeenassumedforthepowerrangefluxtripsetpointraisingitfromthenominalvalueof25%toavalueof35%inadditiontotakingnocreditforthesourceandintermediate rangeprotection.

Reference toFigure3.3-5,however,showsthattheriseinnuclearfluxissorapidthattheeffectoferrorsinthetripsetpointontheactualtimeatwhichtherodsarereleasedisnegligible.

Inadditiontotheabove,therateofnegativereactivity insertion corresponding tothetripactionisbasedontheassumption thatthehighestworthcontrolrodassemblyisstuckinitsfullywithdrawn position.

TheaccidentisanalyzedusingtheStandardThermalDesignProcedure withtheinitialconditions listedinTable3.3-4.Theanalysiswasperformed forareactivity insertion rateof75pcm/sec.Thisreactivity insertion rateassumedisgreaterthanthatforthesimultaneous withdrawal ofthecombination ofthetwosequential controlbankshavingthegreatestcombinedworthatmaximumspeed(45inches/minute).

1pcm=10'hk/kResultsandConclusions Thenuclearpower,heatflux,fuelaveragetemperature, andcladtemperature versustimefora75pcm/secinsertion rateareshowninFigures3.3-5and3.3-6.Thisinsertion rate,coupledwiththe30%SGTPconditions, yieldsaminimumDNBRwhichremainsabovethelimitvalue.FortheRodWithdrawal fromsubcritical event,thecoreaxialpowerdistribution isseverelypeakedtothebottomofthecore.TheW-3DNBcorrelation isusedtoevaluateDNBRinthespanbetweenthelowernon-mixing vanegridandthefirstmixingvanegrid.TheWRB-1correlation remainsapplicable fortherestofthefuelassembly.

ForallregionsofthecoretheDNBdesignbasesaremet.mA1944<w.wpf:1d441195 3.3-22 3.3.5.2Uncontrolled ControlRodAssemblyBankWithdrawal AtPowerAnuncontrolled RodClusterControlAssembly(RCCA)withdrawal atpowerresultsinanincreaseincoreheatflux.Sincetheheatextraction fromthesteamgenerator lagsbehindthepowergeneration untilthesteamgenerator pressurereachestherelieforsafetyvalvesetpoint, thereisanetincreaseinreactorcoolanttemperature.

Unlessterminated bymanualorautomatic action,thepowermismatchandresultant coolanttemperature risewouldeventually resultinDNB.Therefore, tominimizethepossibility ofbreaching thecladding, theReactorProtection Systemisdesignedtoterminate anysuchtransient beforetheDNBRfallsbelowthelimitvalue.Theautomatic featuresoftheReactorProtection SystemwhichminimizeadverseeffectstothecoreinanRCCABankWithdrawal incidentatpowerincludethefollowing:

Nuclearpowerrangeinstrumentation actuatesareactortriponhighneutronfluxiftwooutoffourchannelsexceedanoverpower setpoint.

ReactortripisactuatedifanytwooutoffourhTchannelsexceedanovertemperature bTsetpoint.

Thissetpointisautomatically variedwithaxialpowerdistribution, coolantaveragetemperature andpressuretoprotectagainstDNB.3.ReactortripisactuatedifanytwooutoffourhTchannelsexceedanoverpower bTsetpoint.

Thissetpointisautomatically variedwithcoolantaveragetemperature sothattheallowable fuelpowerratingisnotexceeded.

4.Ahighpressurereactortrip,actuatedfromanytwooutoffourpressurechannels, issetatafixedpoint.Thissetpressureislessthanthesetpressureforthepressurizer safetyvalves.5.Ahighpressurizer waterlevelreactortrip,actuatedfromanytwooutofthreelevelchannels, issetatafixedpoint.Inadditiontotheabovelistedreactortrips,therearethefollowing RCCAWithdrawal blocks.a.Highneutronflux(oneoutoffour)b.Overpower AT(twooutoffour)c.Overtemperature dT(twooutoffour)Themannerinwhichthecombination ofoverpower and.overtemperature bTtripsprovideprotection overthefullrangeofReactorCoolantSystemconditions isillustrated inFigures3.3-1through3.3-4.Thesefiguresrepresent theallowable conditions ofreactormal944<w.wpf:1d441195 3.3-23 coolantloopaveragetemperature andpowerwiththedesignpowerdistribution inatwo-dimensional plot.Thepurposeofthisanalysisistodemonstrate themannerinwhichtheaboveprotective systemsfunctionforvariousreactivity insertion ratesfromdifferent initialconditions.

Reactivity insertion ratesandinitialconditions governwhichprotective functionoccursfirst.MethodofAnalsisThistransient isanalyzedbytheLOFTRANcode.Thecorelimitsasillustrated inFigure3.3-1through3.3-4areusedasinputtoLOFTRANtodetermine theminimumDNBRduringthetransient.

Theanalysisisperformed toboundtheconditions ofhighandlowaveragetemperature withhighandlowRCSpressures forUnit1.ThisaccidentisanalyzedwiththeRTDPdescribed inReference 2.Plantcharacteristics andinitialconditions arelistedinTable3.3-4.Foranuncontrolled rodwithdrawal atpoweraccident, thefollowing conservative assumptions aremade:A.Nominalvaluesareassumedfortheinitialreactorpower,pressure, andRCStemperatures.

Uncertainties ininitialconditions areincludedinthelimitDNBRasdescribed inReference 2.B.Reactivity coefficients

-twocasesareanalyzed:

MinimumReactivity Feedback.

A+5pcmPFmoderator temperature coefficient ofreactivity andaleastnegativeDoppleronlypowercoefficient (seeTable3.3A)areassumed.2.MaximumReactivity Feedback.

Aconservatively largenegativemoderator temperature coefficient andamostnegativeDoppleronlypowercoefficient (SeeTable3.3-4)areassumed.Thereactortriponhighneutronfluxisassumedtobeactuatedataconservative valueof118percentofnominalfullpower.ThebTtripsincludealladverseinstrumentation andsetpointerrors,whilethedelaysforthetripsignalactuation areassumedattheirmaximumvalues.D.'TheRCCAtripinsertion characteristic isbasedontheassumption thatthehighestworthassemblyisstuckinitsfullywithdrawn position.

mA1944<w.wpf:1d441195 3.3-24 E.Themaximumpositivereactivity insertion rateisgreaterthanthatforthesimultaneous withdrawal ofthecombinations ofthetwocontrolbankshavingthemaximumcombinedworthatmaximumspeed.ResultsFigures3.3-7through3.3-9showthetransient responseforarapidRCCAbankwithdrawal incidentstartingfromfullpower.Reactortriponhighneutronfluxoccursshortlyafterthestartoftheaccident.

Sincethisisrapidwithrespecttothethermaltimeconstants oftheplant,smallchangesinT,~andpressureresultandmargintoDNBismaintained.

Thetransient responseforaslowRCCAbankwithdrawal fromfullpowerisshowninFigures3.3-10through3.3-12.Reactortriponovertemperature dToccursafteralongerperiodandtheriseintemperature andpressureisconsequently largerthanforrapidRCCAbankwithdrawal

~Again,theminimumDNBRisgreaterthanthelimitvalue.Figure3.3-13showstheminimumDNBRasafunctionofreactivity insertion ratefrominitialfullpoweroperation forminimumandmaximumreactivity feedback.

Itcanbeseenthattworeactortripfunctions provideprotection overthewholerangeofreactivity insertion rates.Thesearethehighneutronfluxandovertemperature bTfunctions.

TheminimumDNBRisalwaysgreaterthanthelimitvalue.Figures3.3-14and3.3-15showtheminimumDNBRasafunctionofreactivity insertion rateforRCCAbankwithdrawal incidents startingat60and10percentpowerrespectively.

Theresultsaresimilartothe100percentpowercase,exceptastheinitialpowerisdecreased, therangeoverwhichtheovertemperature hTtripiseffective isincreased.

InneithercasedoestheDNBRfallbelowthelimitvalue.Conclusions Thehighneutronfluxandovertemperature bTtripchannelsprovideadequateprotection overtheentirerangeofpossiblereactivity insertion rates,i.e.,theminimumvalueofDNBRisalwayslargerthanthelimitvalueforallfueltypes.Also,thepressurizer doesnotfill.mal944<w.wpf:1d~11953.3-25 3.3.5.3RodClusterControlAssemblyMisalignment Therodclustercontrolassemblymisalignment eventsareprimarily examinedtodemonstrate coreprotection.

Althoughthereduction inratedthermalpowerisabenefitfortheDNBevaluation, thereduction inRCSflowisnon-conservative withrespecttotheDNBtransient.

Assuch,therodclustercontrolassemblymisalignment eventsareanalyzedtodetermine theimpactoftheSGTPProgram.Rodclustercontrolassemblymisalignment accidents include:A.AdroppedRCCAB.AdroppedRCCAbankC.Statically misaligned RCCAEachRCCAhasapositionindicator channelwhichdisplayspositionoftheassembly.

Thedisplaysofassemblypositions aregroupedfortheoperator's convenience.

Fullyinsertedassemblies arefurtherindicated byrodbottomlight.Groupdemandpositionisalsoindicated.

RCCAsarealwaysmovedinpreselected banks,andthebanksarealwaysmovedinthesamepreselected sequence.

Therodscomprising agroupoperateinparallelthroughmultiplexing thynstors.

Thetwogroupsinabankmovesequentially suchthatthefirstgroupisalwayswithinonestepofthesecondgroupinthebank.Adefinitescheduleofactuation (ordeactuation ofthesecondary gripper,movablegripper,andliftcoilsofamechanism) isrequiredtowithdrawtheRCCAattachedtothemechanism.

Sincethestationaiy gripper,movablegripper,andliftcoilsassociated withtheRCCAsofarodgrouparedriveninparallel, anysinglemalfunction whichwouldcauserodwithdrawal wouldaffectaminimumofonegroup.Mechanical malfunctions areinthedirection ofinsertion, orimmobility.

AdroppedRCCAorRCCAbankisdetectedby:a.Suddendropinthecorepowerlevelasseenbythenuclearinstrumentation system;b.Asymmetric powerdistribution asseenonout-of-core neutrondetectors orcoreexitthermocouples; c.Rodatbottomsignal;d.Rodpositiondeviation monitor,e.Rodpositionindication.

m%1944<w.wpf:

1d4411953.3-26 Misaligned RCCAaredetectedby:a.Asymmetric powerdistribution asseenonout-of-core neutrondetectors orcoreI'xitthermocouples; b.Rodpositiondeviation monitor;c.Rodpositionindicators.

Theresolution oftherodpositionindicator channelis+5percent(+12steps).Deviation ofanyassemblyfromitsgroupbytwicethisdistancewillnotcausepowerdistributions worsethanthedesignlimits.Therodpositiondeviation monitoralertstheoperatortoroddeviation beforeitcanexceedtenpercentofspan(+24steps).Iftherodpositiondeviation monitorisnotoperable, theoperatorisrequiredtotakeactionasrequiredbytheTechnical Specifications.

MethodofAnalsisA.OneormoredroppedRCCAsfromthesamegroup.Forevaluation ofthedroppedRCCAevent,thetransient systemresponseiscalculated usingtheLOFTRANcode.Thecodesimulates theneutronkinetics, RCS,pressurizer, pressurizer reliefandsafetyvalves,pressurizer spray,steamgenerator, andsteamgenerator safetyvalves.Thecodecomputespertinent plantvariables including temperatures, pressures, andpowerlevel.Nominalvaluesforinitialreactorpower,temperature, andRCSpressureareassumedtoboundtheoperation ofUnit1with30%SGTP.Theinitialconditions arepresented inTable3.3Q.Uncertainties forinitialconditions areincludedinthelimitDNBR.Statepoints arecalculated andnuclearmodelsareusedtoobtainahotchannelfactorconsistent withtheprimarysystemconditions andreactorpower.Byincorporating theprimaryconditions fromthetransient andthehotchannelfactorfromthenuclearanalysis, theDNBdesignbasisisshowntobemetusingtheTHINGIVcode.Thetransient

response, nuclearpeakingfactoranalysis, andDNBdesignbasisconfirmation areperformed inaccordance withthemethodology described inReference 9.Notethatoperation withautomatic rodcontrolisassumedfortheanalysis.

Alsonotethattheanalysisdoesnottakecreditforthenegativefluxratereactortrip.mh1944<w.wpf:

1d~11953.3-27 Statically Misaligned RCCASteadystatepowerdistributions areanalyzedusingthemethodology described inReference 9.ThepeakingfactorsarethenusedasinputtotheTHINGIVcodetocalculate theDNBR.ResultsOneormoreDroppedRCCAsSingleormultipledroppedRCCAswithinthesamegroupresultinanegativereactivity insertion.

Thecoreisnotadversely affectedduringthisperiod,sincepowerisdecreasing rapidly.Following plantstabilization, normalrodretrieval orshutdownprocedures arefollowed.

TheoperatormaymanuallyretrievetheRCCAbyfollowing approvedoperating procedures.

Powermaybereestablished eitherbyreactivity feedbackorcontrolbankwithdrawal.

Following adroppedrodeventinmanualrodcontrol,theplantwillestablish anewequilibrium condition.

Theequilibrium processwithoutcontrol-system interaction ismonotonic, thusremovingpowerovershoot asaconcern,andestablishing theautomatic rodcontrolmodeofoperation asthelimitingcase.ForadroppedRCCAeventintheautomatic rodcontrolmode,theRodControlSystemdetectsthedropinpowerandinitiates controlbankwithdrawal.

Powerovershoot mayoccurduetothisactionbytheautomatic rodcontroller afterwhichthecontrolsystemwillinsertthecontrolbanktorestorenominalpower.Figures3.3-16and3.3-17showatypicaltransient responsetoadroppedRCCA(orRCCAs)inautomatic control.Uncertainties intheinitialcondition areincludedintheDNBevaluation asdescribed inReference 9.Inallcases,theminimumDNBRremainsabovethelimitvalue.DroppedRCCAbankAdroppedRCCAbanktypically resultsinareactivity insertion greaterthan500pcm.Thecoreisnotadversely affectedduringtheinsertion period,sincepowerisdecreasing rapidly.Thetransient willproceedasdescribed in"A"above;however,thereturntopowerwillbelessduetothegreaterworthofanentirebank.Following plantstabilization, normalrodretrieval orshutdownprocedures arefollowedtofurthercooldowntheplant.m:41944<w.wpt:1d441195 3.3-28 C.Statically Misaligned RCCAThemostseveremisalignment situations withrespecttoDNBRatsignificant powerlevelsarisefromcasesinwhichoneRCCAisfullyinserted, orwherebankDisfullyinsertedwithoneRCCAfullywithdrawn.

Multipleindependent alarms,including abankinsertion limitalarm,alerttheoperatorwellbeforethepostulated conditions areapproached.

Thebankcanbeinsertedtoitsinsertion limitwithanyoneassemblyfullywithdrawn withouttheDNBRfallingbelowthelimitvalue.Theinsertion limitsintheTechnical Specifications mayvaryfromtimetotimedepending onanumberoflimitingcriteria, Itispreferable, therefore, toanalyzethemisaligned RCCAcaseatfullpowerforapositionofthecontrolbankasdeeplyinsertedasthecriteriaonminimumDNBRandpowerpeakingfactorwillallow.Thefullpowerinsertion limitsoncontrolbankDmustthenbechosentobeabovethatpositionandwillusuallybedictatedbyothercriteria.

Detailedresultswillvanefromcycletocycledepending onfuelarrangements.

WithbankDinsertedtoitsfullinsertion limitandoneRCCAfullywithdrawn, DNBRdoesnotfallbelowthelimitvalue.Thiscaseisanalyzedassumingtheinitialreactorpower,pressure, andRCStemperatures areattheirnominalvalues(asgiveninTable3.3A)butwiththeincreased radialpeakingfactorassociated withthemisaligned RCCA.DNBcalculations havenotbeenperformed specifically forRCCAsmissingfromotherbanks;however,powershapecalculations havebeendoneasrequiredforRCCAejectionanalysis.

Inspection ofthepowershapesshowsthattheDNBandpeakkw/ftsituation islessseverethanthebankDcasediscussed aboveassuminginsertion limitsontheotherbanksequivalent toabankDinsertion limit.ForRCCAmisalignments withoneRCCAfullyinserted, theDNBRdoesnotfallbelowthelimitvalue.Thiscaseisanalyzedassumingtheinitialreactorpower,pressure, andRCStemperatures areattheirnominalvalues,(asgiveninTable3.34)butwiththeincreased radialpeakingfactorassociated withthemisaligned RCCA.DNBdoesnotoccurfortheRCCAmisalignment incidentandthustheabilityoftheprimarycoolanttoremoveheatfromthefuelrodisnotreduced.Thepeakfueltemperature corresponds toalinearheatgeneration ratebasedontheradialpeakingfactorpenaltyassociated withthemisaligned RCCAandthedesignaxialpowerdistribution.

Theresulting linearheatgeneration iswellbelowthatwhichwouldcausefuelmelting.mh1944<w.wpt:1d441195 3:3-29 Following theidentification ofaRCCAgroupmisalignment condition bytheoperator, theoperatorisrequiredtotakeactionasrequiredbytheplantTechnical Specifications andoperating instructions.

m:51944<w.wpt:1d4411953,3-30 3.3.5.4LossofReactorCoolantFlow(Including LockedRotorAnalysis)

Alossofforcedreactorcoolantflowmayresultfromasimultaneous lossofelectrical suppliestoallreactorcoolantpumps.Ifthereactorisatpoweratthetimeoftheaccident, theimmediate effectoflossofcoolantflowisarapidincreaseinthecoolanttemperature whichismagnified byapositiveMTC.ThisincreasecouldresultinDNBwithsubsequent adverseeffectstothefuelifthereactorwerenottrippedpromptly.

Thetripsystemsavailable tomitigatetheconsequence ofthisaccidentarediscussed intheUFSAR.Simultaneous lossofelectrical powertoallreactorcoolantpumpsatfullpoweristhemostseverecrediblelossofflowcondition.

Forthiscondition reactortriptogetherwithflowsustained bytheinertiaofthecoolantandrotatingpumppartswillbesufficient topreventRCSoverpressurization andtheDNBratiofromexceeding thelimitvalues.Thedecreaseinreactorcoolantsystemflowrateeventsareprimarily examinedtodemonstrate coreprotection.

Thereduction inRCSflow,asaresultoftheincreaseinsteamgenerator tubepluggingto30%,isnon-conservative withrespecttotheDNBtransient.

Assuch,analysesarepresented todiscusstheimpactofthischange.MethodofAnalsisThefollowing lossofflowcasesareanalyzed:

1.Lossoffourpumpsfromnominalfullpowerconditions withfourloopsoperating.

2.Lossofonepumpfromnominalfullpowerconditions withfourloopsoperating.

Thenormalpowersuppliesforthepumpsarefourbusesconnected tothegenerator.

Eachbussuppliespowertoonepump.Whenagenerator tripoccurs,thepumpsareautomatically transferred toabussuppliedfromexternalpowerlines,andthepumpswillcontinuetosupplycoolantflowtothecore.Thesimultaneous lossofpowertoallreactorcoolantpumpsisahighlyunlikelyevent.Sinceeachpumpisonaseparatebus,asinglebusfaultwouldnotresultinthelossofmorethanonepump.Afullplantsimulation isusedintheanalysistocomputethecoreaverageandhotspotheatfluxtransient responses, including flowcoastdown, temperature, reactivity andcontrolrodinsertion effects.Thesedataarethenusedinadetailedthermal-hydraulic.

computation tocomputethemargintoDNBusingtheRTDP.Thiscomputation solvesthecontinuity, momentumandenergyequations offluidflowtogetherwiththeWRB-1DNBcorrelation.

mA1944<w.wpf:1dO41195 3.3-31

Theanalysesareperformed toboundtheconditions oftheSGTPProgram.Uncertainties ininitialconditions areincludedinthelimitDNBRasdescribed inReference 2.Nominalvaluesareassumedfortheinitialreactorpower,pressure, andRCStemperatures.

Theinitialconditions usedarelistedinTable3.3-4.Thistransient isanalyzedbythreedigitalcomputercodes.FirsttheLOFTRANcodeisusedtocalculate theloopandcoreflowduringthetransient, thetimeofreactortripbasedonthecalculated flows,thenuclearpowertransient, andtheprimarysystempressureandtemperature transients.

TheFACTRANcodeisthenusedtocalculate theheatfluxtransient basedonthenuclearpowerandflowfromLOFTRAN.Finally,theTHINGIVcodeisusedtocalculate theDNBRduringthetransient basedontheheatfluxfromFACTRANandflowfromLOFTRAN.TheDNBRtransients presented represent theminimumofthetypicalorthimblecellforeachtypeoffuel.ResultsFigures3.3-18through3.3-20showthetransient responseforthelossofpowertoallRCPswithfourloopsinoperation.

Thereactorisassumedtobetrippedonundervoltage signal.Figure3.3-20showstheDNBRtobealwaysgreaterthanthelimitvalueforthemostlimitingfuelassemblycell.Figures3.3-21through3.3-23showthetransient responseforthelossofoneRCPwithfourloopoperation.

Thereactorisassumedtobetrippedonlowflowsignal.Figure3.3.-23showstheDNBtobealwaysgreaterthanthelimitvalueforthemostlimitingfuelassemblycell.Thesequenceofeventsfollowing eachofthesetransients isincludedinTable3.3-5..SinceDNBdoesnotoccur,theabilityoftheprimarycoolanttoremoveheatfromthefuelrodisnotsignificantly reduced.Thus,theaveragefuelandcladtemperature donotincreasesignificantly abovetheirrespective initialvalues.Conclusions TheanalysisshowsthattheDNBRwillnotdecreasebelowthelimitvalueatanytimeduringthetransient.

Thus,nofueladverseeffectsorcladruptureispredicted, andallapplicable acceptance criteriaaremet.m:11944<w.wpf:1d441195 3.3-32 LockedRotorAccidentAtransient analysishasbeenperformed fortheinstantaneous seizureofareactorcoolantpumprotor.Flowthroughtheaffectedreactorcoolantloopisrapidlyreduced,leadingtoareactortriponalowflowsignal.Following thetrip,heatstoredinthefuelrodscontinues topassintothecorecoolant,causingthecoolanttoexpand.Atthesametime,heattransfertotheshellsideofthesteamgenerator isreduced,firstbecausethereducedflowresultsinadecreased tubesidefilmcoefficient andthenbecausethereactorcoolantinthetubescoolsdownwhiletheshellsidetemperature increases (turbinesteamflowisreducedtozerouponplanttrip).Therapidexpansion ofthecoolantinthereactorcore,combinedwiththereducedheattransferinthesteamgenerator causesaninsurgeintothepressurizer andapressureincreasethroughout theRCS.Theinsurgeintothepressurizer causesapressureincreasewhichinturnactuatestheautomatic spraysystem,opensthepower-operated reliefvalves,andopensthepressurizer safetyvalves,inasequencedependent ontherateofinsurgeandpressureincrease.

Thepower-operated reliefvalvesaredesignedforreliableoperation andwouldbeexpectedtofunctionpropertyduringtheaccident.

However,forconservatism, theirpressure-reducing effectaswellasthepressure-reducing effectofthesprayarenotincludedinthisanalysis.

Thelockedrotoreventisexaminedtodetermine theDNBtransient andtodemonstrate thatthepeakRCSpressureandpeakcladtemperature remainbelowthelimitvalues.Thereduction inRCSflow,duetotheincreaseintheSGTPlevel,isnon-conservative withrespecttotheDNBevaluation.

Assuch,thelockedrotoreventwasre-analyzed.

MethodofAnalsisTwodigital-computer codesareusedtoanalyzethistransient.

TheLOFTRANcodeisusedtocalculate theresulting loopandcoreflowtransients following thepumpseizure,thetimeofreactortripbasedontheloopflowtransients, thenuclearpowerfollowing reactortrip,andtodetermine thepeakpressure.

Thethermalbehaviorofthefuellocatedatthecorehotspotisinvestigated usingtheFACTRANcode,usingthecoreflowandthenuclearpowercalculated byLOFTRAN.TheFACTRANcodeincludestheuseofafilmboilingheattransfercoefficient.

Theanalysisisperformed toboundtheconditions associated withtheSGTPProgram.AsinthepreviousUFSARanalysis, theanalysisassumesoffsitepowerisavailable following thereactortripandturbinetrip.Evaluation ofthePressureTransient Afterpumpseizure,theneutronfluxisrapidlyreducedbyGontrolrodinsertion.

Rodmotionbegins1secondaftertheflowintheaffectedloopreaches87percentofnominalflow.Nom:51944<w.wpf:1d441195

~3.3-33 creditistakenforthepressurereducingeffectofthepressurizer reliefvalves,pressurizer spray,steamdumporcontrolled feedwater flowafterplanttrip.Althoughtheseoperations areexpectedtooccurandwouldresultinalowerpeakRCSpressure, anadditional degreeofconservatism isprovidedbyignoringtheireffect.Thepressurizer safetyvalvesareassumedtoinitially openat2575psiaandachieveratedflowat2580psia.Thisanalysisassumedaninitialpressurizer pressureof2317psia.Table3.3-4presentstheinitialconditions assumedforthepeakpressuretransient.

Evaluation ofthePeakCladTemperature Forthisaccident, DNBisassumedtooccurinthecore;therefore anevaluation oftheconsequences withrespecttofuelrodthermaltransients isperformed.

Theassumption ofrodsgoingintoDNBasaconservative initialcondition ismadeinordertodetermine thecladtemperature andzirconium waterreaction.

Thisanalysisassumedaninitialpressurizer pressureof2100psia.Resultsobtainedfromanalysisofthishotspotcondition represent theupperlimitwithrespecttocladtemperature andzirconium waterreaction.

Intheevaluation, therodpoweratthehotspotisassumedtobe2.5timestheaveragerodpower(i.e.,FQ=2.5)attheinitialcorepowerlevel.Table3.3-4presentstheinitialconditions assumedforthepeakcladtemperature transient.

FilmBoi%ngCoefficient Thefilmboilingcoefficient iscalculated intheFACTRANcodeusingtheBishop-Sandberg-Tongfilmboilingcorrelation (Reference 13).Thefluidproperties areevaluated atfilmtemperatures (averagebetweenwallandbulktemperatures).

Theprogramcalculates thefilmcoefficient ateverytimestepbasedupontheactualheattransferconditions atthetime.Theneutronflux,systempressure, bulkdensity,andmassflowrateasafunctionoftimeareusedasprograminput.Forthepeakcladtemperature

analysis, theinitialvaluesofthepressureandthebulkdensityareusedthroughout thetransient sincetheyarethemostconservative withrespecttocladtemperature response.

Forconservatism, DNBwasassumedtostartatthebeginning oftheaccident.

FuelCladGapCoefficient Themagnitude andtimedependence oftheheattransfercoefficient betweenfuelandclad(gapcoefficient) hasapronounced influence onthethermalresults.Thelargerthevalueofthegapcoefficient, themoreheatistransferred betweenpelletandclad.Basedoninvestigations ontheeffectofthegapcoefficient uponthemaximumcladtemperature duringm:51944<w.wpf:1d~1195 3.3-34 thetransient, thegapcoefficient wasassumedtoincreasefromasteadystatevalueconsistent withinitialfueltemperature to10,000BTU/hr-ft'-'F attheinitiation ofthetransient.

Thusthelargeamountofenergystoredinthefuelbecauseofthesmallinitialvalueisreleasedtothecladattheinitiation ofthetransient.

Zirconium-Steam ReactionThezirconium-steam reactioncanbecomesignificant above1800'F(cladtemperature).

Inordertotakethisphenomenon intoaccount,thefollowing correlation, whichdefinestherateofthezirconium-steam

reaction, wasintroduced intothemodels(Reference 10).=33.3xttPexp(-'1.366where:w=amountreacted,mg/cm'=time,secondsT=temperature, KThereactionheatis1510cal/gEvaluation ofRods-in-ONB Anevaluation ismadetodetermine whatpercentage, ifany,ofrodsareexpectedtobeinDNBduringthetransient.

Forthisevaluation, thepredicted coreconditions areusedasinputtoaTHINGIVcalculation oftheminimumDNBRduringthetransient.

ResultsoftheTHINCIVevaluation arethenusedtodetermine thepercentage offuelrodswhichexperience DNB.Table3.3-4presentstheinitialconditions assumedfortherods-in-DNB evaluation.

ReeeiteThetransient resultsforthelockedrotoraccidentareshowninFigures3.3-24through3.3-26.ThepeakRCSpressure(2641psia)reachedduringthetransient islessthanthatwhichwouldcausestressestoexceedthefaultedcondition stresslimits(thispeakpressureisalsobelow110%ofthedesignpressure).

ThepressureresponseshowninFigure3.3-25istheresponseatthepointintheReactorCoolantSystemhavingthemaximumpressure.

Also,thepeakcladsurfacetemperature (1934'F,showninFigure3.3-26)isconsiderably lessthan2700'F.ThesequenceofeventsisincludedinTable3.3-5.Forthemostlimitingfuelassembly, lessthan7%ofthe.rodsreachaDNBRvaluelessthanthelimitvalueforthe30%SGTPconditions.

mA1944<w.wpf:1d441295 3.3-35 Conclusions A.SincethepeakRCSpressurereachedduringanyofthetransients islessthanthatwhichwouldcausestressestoexceedthefaultedconditions stresslimits,theintegrity oftheprimarycoolantsystemisnotendangered.

Sincethepeakcladsurfacetemperature calculated forthehotspotduringtheworsttransient remainsconsiderably lessthan2700'F(thetemperature atwhichcladembrittlement maybeexpected),

thecorewillremaininplaceandintactwithnolossofcorecoolingcapability.

mA1944<w.wpt:

1d4411953.3-36 3.3.5.5LossofExternalElectrical LoadThecompletelossofsteamloadfromfullpowerisexaminedprimarily toshowtheadequacyofthepressurerelieving devicesandalsotodemonstrate coreprotection.

Thereduction inRCSflow,asresultofincreasing thelevelofSGTP,isnon-conservative withrespecttotheDNBbehavior.

Primaryprotection forthiseventisprovidedbythehighpressurizer

pressure, OTLT,highpressurizer waterlevel,andlow-lowsteamgenerator waterlevelreactortrips.Thelossofexternalelectrical loadmayresultfromanabnormalvariation innetworkfrequency orotheradversenetworkoperating conditions.

Itmayalsoresultfromatripoftheturbinegenerator orinanunlikelyopeningofthemainbreakerfromthegenerator whichfailstocauseaturbinetripbutcausesarapidlargeNSSSloadreduction bytheactionoftheturbinecontrol.MethodofAnalsisThelossofloadtransients areanalyzedbyemploying thedetaileddigitalcomputerprogramLOFTRAN.Theprogramsimulates theneutronkinetics, RCS,pressurizer, pressurizer reliefandsafetyvalves,pressurizer spray,steamgenerator, andsteamgenerator safetyvalves.Theprogramcomputespertinent plantvariables including temperatures, pressures, andpowerlevel.Ananalysisisperformed toboundtheconditions oftheSGTPProgram.Nominalvaluesareassumedfortheinitialreactorpower,temperature, andpressure.

ThisaccidentisanalyzedwiththeRTDP.Plantcharacteristics andinitialconditions arelistedinTable3.3-4.Majorassumptions aresummarized below:A.InitialOperating Conditions

-nominalconditions forreactorpower,pressure, andRCStemperatures areassumedforstatistical DNBanalyses.

B.Moderator andDopplerCoefficients ofReactivity

-thelossofloadisanalyzedwithbothmaximumandminimumreactivity feedback.

Themaximumfeedbackcasesassumealargenegativemoderator temperature coefficient andthemostnegativeDopplerpowercoefficient.

Theminimumfeedbackcasesassumeapositivemoderator temperature coefficient andtheleastnegativeDopplercoefficients.

C.ReactorControl-fromthestandpoint ofthemaximumpressures attaineditisconservative toassumethatthereactorisi'anualcontrol.Ifthereactorwereinautomatic control,thecontrolrodbankswouldmovepriortotripandreducetheseverityofthetransient.

mal944<w.wpf:1d~1195 3.3-37 D.Pressurizer SprayandPower-Operated ReliefValves-twocasesforboththeminimumandmaximummoderator feedbackcasesareanalyzed:

Fullcreditistakenfortheeffectofpressurizer sprayandpower-operated reliefvalvesinreducingorlimitingthecoolantpressure.

Safetyvalvesarealsoavailable.

2.Nocreditistakenfortheeffectofpressurizer sprayandpower-operated reliefvalvesinreducingorlimitingthecoolantpressure.

Safetyvalvesareoperable.

E.SteamRelease-nocreditistakenfortheoperation ofthesteamdumpsystemorsteamgenerator power-operated reliefvalves.Thesteamgenerator pressurerisestothesafetyvalvesetpointwheresteamreleasethroughthesafetyvalveslimitsthesecondary steampressure.

F.Feedwater Flow-mainfeedwater flowtothesteamgenerators isassumedtobelostatthetimeofturbinetrip.Nocreditistakenforauxiliary feedwater flowsinceastabilized plantcondition willbereachedbeforeauxiliary feedwater initiation isnormallyassumedtooccur;however,theauxiliary feedwater pumpswouldbeexpectedtostartonatripofthemainfeedwater pumps.Theauxiliary feedwater flowwouldremovecoredecayheatfollowing plantstabilization.

G.ReactortripisactuatedbythefirstReactorProtection Systemtripsetpointreached.Tripsignalsareexpectedduetohighpressurizer

pressure, overtemperature bT,highpressurizer waterlevel,andlow-lowsteamgenerator waterlevel.ResultsThetransient responses foralossofloadfromfullpoweroperation areshownforfourcases:minimumandmaximumreactivity
feedback, withandwithoutpressurecontrol(Figures3.3-27through3.3-46).Figures3.3-27through3.3-31showthetransient responses forthelossofloadwithminimumreactivity feedbackassumingfullcreditforthepressurizer sprayandpressurizer power-operated reliefvalves.Nocreditistakenforthesteamdump.Thereactoristrippedbytheovertemperature ATtripsignal.TheminimumDNBRremainswellabovethelimitvalue.Thepressurizer reliefandsafetyvalvespreventoverpressurization ofthepnmarysystem.Thesteamgenerator safetyvalvesmA1944<w.wpf:1d441195 3.3-38

preventoverpressurization ofthesecondary system,maintaining pressurebelow110percentofdesignvalue.Figures3.3-32through3.3-36showtheresponses forthetotallossofsteamloadwithmaximumreactivity feedback.

Allotherplantparameters arethesameastheabove.TheDNBRincreases throughout thetransient andneverdropsbelowitsinitialvalue.Pressurizer reliefvalvesandsteamgenerator safetyvalvespreventoverpressurization inprimaryandsecondary systems,respectively.

Thereactoristrippedbythelow-lowsteamgenerator waterlevelsignal.Thepressurizer safetyvalvesarenotactuatedforthiscase.Intheeventthatfeedwater flowisnotterminated atthetimeofturbinetripforthiscase.flowwouldcontinueunderautomatic controlwiththereactoratareducedpower.Theoperatorwouldtakeactiontoterminate thetransient andbringtheplanttoastabilized condition.

Ifnoactionweretakenbytheoperatorthereducedpoweroperation wouldcontinueuntilthecondenser hotwellwasemptied.Alow-lowsteamgenerator waterlevelreactortripwouldbegenerated alongwithauxiliary feedwater initiation signals.Auxiliary feedwater wouldthenbeusedtoremovedecayheatwiththeresultslessseverethanthosepresented inSection3.3.4.3,LossofNormalFeedwater Flow.Thelossofloadaccidentwasalsostudiedassumingtheplanttobeinitially operating atfullpowerwithnocredittakenforthepressurizer spray,pressurizer power-operated reliefvalves,orsteamdump.Thereactoristrippedonthehighpressurizer pressuresignal.Figures3.3-37through3.3-41showthetransient responses withminimumreactivity feedback.

Theneutronfluxremainsessentially constantatfullpoweruntilthereactoristripped.TheDNBRnevergoesbelowitsinitialvaluethroughout thetransient.

Inthiscasethepressurizer safetyvalvesareactuated, andmaintainsystempressurebelow110percentofthedesignvalue.Figures3.3-42through3.3-46showthetransient responses withmaximumreactivity feedbackwiththeotherassumptions beingthesameasinthepreceding case.Again,theDNBRincreases throughout thetransient andthepressurizer safetyvalvesareactuatedtolimitprimarypressure.

Thesequenceofeventsfollowing eachofthesetransients isincludedinTable3.3-6.Conclusions Resultsoftheanalysesshowthattheplantdesignissuchthatalossofloadwithoutadirectorimmediate reactortrippresentsnohazardtotheintegrity oftheRCSorthemainsteamsystem.Pressurerelieving devicesincorporated inthetwosystemsareadequatetolimitthemaximumpressures towithinthedesignlimits.Theintegrity ofthecoreismaintained bymhl944<w.wpf:1d441195 3.3-39 operation ofthereactorprotection system,i.e.,theDNBRwillbemaintained abovethelimitvalue.Thustheconclusions presented istheUFSARremainvalidfor30%SGTP.3.3.5.6RuptureofaSteamPipeAlthoughtheno-loadtemperature doesnotchangefortheSGTPProgram,andareduction intheheattransfercapability, duetotheincreased numberofpluggedsteamgenerator tubes,wouldresultinalessseverecooldown, theimpactoftheRCSflowreduction needstobeaddressed forthesteamline breakaccident.

Thereanalysis alsoassumedareduction intheavailable shutdownmarginfrom1.60to1.30%d,k/katno-loadconditions.

Anevaluation hasbeenperformed forthosecasesthatmodelacoincident lossofoffsitepowerinordertoaddresstheincreaseintheEDGstarttimefrom10to30seconds.Thisanalysiswasperformed assumingthecoincidence logicrequiredforSIandSLIconsistent withthecurrentUnit1steamline breakprotection system.Aproposedmodification totheUnit1steamline breakprotection systemwillchangethislogic.However,thisanalysisboundstheproposedmodifications totheUnit1steamline breakprotection system,asdiscussed inSection3.3.2.5.Aruptureofasteampiperesultsinanuncontrolled steamreleasefromasteamgenerator.

Thesteamreleaseresultsinaninitialincreaseinsteamflowwhichdecreases duringtheaccidentasthesteampressurefalls.Theenergyremovalfromthecausesareduction incoolanttemperature andpressure.

Inthepresenceofanegativemoderator temperature coefficient (MTC),thecooldownresultsinareduction ofcoreshutdownmargin.IfthemostreactiveRCCAisassumedstuckinitsfullywithdrawn

position, thereisanincreased possibility thatthecorewillbecomecriticalandreturntopower.Areturntopowerfollowing asteampiperuptureisapotential concernmainlybecauseofthehighhotchannelfactorswhichexistwhenthemostreactiveassemblyisassumedstuckinitsfullywithdrawn position.

Thecoreisultimately shutdownbyboricaciddelivered bytheECCS.Theanalysisofasteampiperuptureisperformed todemonstrate that:Assumingastuckassembly, withorwithoutoffsitepower,andassumingasinglefailureintheengineered safetyfeatures, thereisnoconsequential damagetotheprimarysystemandthecoreremainsinplaceandintact.AlthoughDNBandpossiblecladperforation following asteampiperupturearenotnecessarily unacceptable, thefollowing

analysis, infact,showsthatnoDNBoccursforanyruptureassumingthemostreactiveassemblystuckinitsfullywithdrawn position.

m:51944<w.wpf:1d

~11953.3<0

MethodofAnalsisTheanalysisofthesteampiperupturehasbeenperformed todetermine:

A.ThecoreheatfluxandRCStemperature andpressureresulting fromthecooldownfollowing thesteamlinebreak.TheLOFTRANcodehasbeenused.Thethermalandhydraulic behaviorofthecorefollowing asteamlinebreak.Adetailedthermalandhydraulic digital-computer code,THINGIV,hasbeenusedtodetermine ifDNBoccursforthecoreconditions computedinitemAabove.Thefollowing conditions wereassumedtoexistatthetimeofamainsteamlinebreakaccident:

End-of-life shutdownmargin(1.30%hk/k)atnoload,equilibrium xenonconditions, andthemostreactiveRCCAstuckinitsfullywithdrawn position:

Operation ofthecontrolrodbanksduringcorebumupisrestricted insuchaway(tonotviolatetherodinsertion limitspresented intheTechnical Specifications) thatadditionofpositivereactivity inasteamlinebreakaccidentwillnotleadtoamoreadversecondition thanthecaseanalyzed.

Anegativemoderator coefficient corresponding totheend-of-life roddedcorewiththemostreactiveRCCAinthefullywithdrawn position:

Thevariation ofthecoefficient withtemperature andpressurehasbeenincluded.

Thek,versustemperature at1050psiacorresponding tothenegativemoderator temperature coefficient usedisshowninFigure3.3-47.TheDopplerpowerfeedbackassumedforthisanalysisispresented inFigure3.3-48.Thecoreproperties associated withthesectornearesttheaffectedsteamgenerator andthoseassociated withtheremaining sectorwereconservatively combinedtoobtainaveragecoreproperties forreactivity feedbackcalculation.

Further,itwasconservatively assumedthatthecorepowerdistribution wasuniform.Thesetwoconditions causeunderprediction ofthereactivity feedbackinthehighpowerregionnearthestuckrod.Toverifytheconservatism ofthismethod,thereactivity aswellasthepowerdistribution wascheckedforthelimitingconditions forthecasesanalyzed.

Thiscoreanalysisconsidered theDopplerreactivity fromthehighfueltemperature nearthestuckRCCA,moderator feedbackfromthehighwaterenthalpynearthestuckRCCA,powerredistribution andnon-uniform coreinlettemperature effects.Forcasesinwhichsteamgeneration occursinthehigh,fluxregionsofthecore,theeffectofvoidformation wasalsoincluded.

Itwasdetermined thatthereactivity employedinthekineticsanalysiswasalwayslargerthanthereactivity mh1944<w.wpf:1d441195 3.341

calculated including theabovelocaleffectsforthestatepoints.

Theseresultsverifyconservatism; i.e.,underprediction ofnegativereactivity feedbackfrompowergeneration.

Minimumcapability forinjection ofboricacid(2400ppm)solutioncorresponding tothemostrestrictive singlefailureinthesafetyinjection system.TheECCSconsistsofthefollowing systems:1)thepassiveaccumulators, 2)thelowheadsafetyinjection (residual heatremoval)system,3)thehighheadsafetyinjection system,and4)thechargingsystem.Onlythechargingsystemandthepassiveaccumulators aremodeledforthesteamlinebreakaccidentanalysis.

Centrifugal Chargingpumpheaddegradation of10%wasassumed.Themodelingofthesafetyinjection systeminLOFTRANisdescribed inReference 4.Figure3.3-49presentsthesafetyinjection flowratesasafunctionofRCSpressureassumedintheanalysis.

Theflowcorresponds tothatdelivered byonechargingpumpdelivering itsfullflowtothecoldlegheader.Nocredithasbeentakenforthelowconcentration boratedwater,whichmustbesweptfromthelinesdownstream oftheRWSTpriortothedeliveryofboricacidtothereactorcoolantloops.Forthisanalysis, aboronconcentration of0ppmfortheboroninjection tankisassumed.Itshouldbenotedthatthisanalysisalsoconsiders theoperation oftheCentrifugal ChargingPumpMinimumFlowIsolation Valves.Thesevalvesareassumedtoclosefollowing thereceiptofaSIsignalandreopenwhenRCSpressurerisesabove2000psig.TheSlflowratesassumedinthesteamline breakanalysis, graphically showninFigure3.349,correspond toCentrifugal ChargingPumpMinimumFlowIsolation Valvesbeingintheclosedposition.

Forthecaseswhereoffsitepowerisassumed,thesequenceofeventsinthesafetyinjection systemisthefollowing.

Afterthegeneration ofthesafetyinjection signal(appropnate delaysforinstrumentation, logic,andsignaltransport included),

theappropriate valvesbegintooperateandthechargingpumpstarts.In27seconds,thevalvesareassumedtobeintheirfinalpositionandthepumpisassumedtobeatfullspeedandtodrawsuctionfromtheRWST.Thevolumecontaining thelowconcentration boratedwaterissweptintocorebeforethe2400ppmboratedwaterreachesthecore.Thisdelay,described above,isinherently includedinthemodeling.

Incaseswhereoffsitepowerisnotavailable, a30seconddelayisassumedtostarttheEDGsandtocommenceloadingthenecessary safetyinjection equipment ontothem.mal944<w.wpf:1d~1195 3.3<2 D.Designvalueofthesteamgenerator heattransfercoefficient including allowance forfoulingfactor.E.Fourcombinations ofbreaksizesandinitialplantconditions havebeenconsidered indetermining thecorepowertransient whichcanresultfromlargeareapipebreaks.a.Completeseverance ofapipedownstream ofthesteamflowrestrictor withtheplantinitially atnoloadconditions andallreactorcoolantpumpsrunning.b.Completeseverance ofapipeinsidethecontainment attheoutletofthesteamgenerator withthesameplantconditions asabove.c.Case(a)abovewithlossofoff-sitepowersimultaneous withthegeneration oftheSafetyInjection Signal(lossofACpowerresultsincoolantpumpcoastdown).

d.Case(b)abovewiththelossofoff-sitepowersimultaneous withtheSafetyInjection Signal.Afifthcase,inwhichthespuriousopeningofasteamdump,relief,orsafetyvalveoccurs,wasconsidered.

Anevaluation concluded thattheDNBRremainsabovethelimitvalueforthiscase.e.Abreakequivalent toasteamflowof247Ibspersecondat1100psifromonesteamgenerator withoff-sitepoweravailable.

F.Powerpeakingfactorscorresponding toonestuckRCCAandnonuniform coreinletcoolanttemperatures aredetermined atendofcorelife.Thecoldestcoreinlettemperatures areassumedtooccurinthesectorwiththestuckrod.Thepowerpeakingfactorsaccountfortheeffectofthelocalvoidintheregionofthestuckcontrolassemblyduringthereturntopowerphasefollowing thesteamlinebreak.Thisvoidinconjunction withthelargenegativemoderator coefficient partially offsetstheeffectofthestuckassembly.

Thepowerpeakingfactorsdependuponthecorepower,temperature,

pressure, andflow,and,thus,aredifferent foreachcasestudied.Theanalysesassumedinitialhotshutdownconditions attimezerosincethisrepresents themostpessimistic initialcondition.

Shouldthereactorbejustcriticaloroperating atpoweratthetimeofasteamlinebreak,thereactorwillbetrippedbythenormaloverpower protection systemwhenpowerlevelmal944<w.wpf:1d441195 3.3-43

reachesatrippoint.Following atripatpowertheRCScontainsmorestoredenergythanatno-load,theaveragecoolanttemperature ishigherthanatno-loadandthereisappreciable energystoredinthefuel.Thus,theadditional storedenergyisremovedviathecooldowncausedbythesteamlinebreakbeforetheno-loadconditions ofRCStemperature andshutdownmarginassumedintheanalysesarereached.Aftertheadditional storedenergyhasbeenremoved,thecooldownandreactivity insertions proceedinthesamemannerasintheanalysiswhichassumesno-loadcondition attimezero.Inaddition, sincetheinitialsteamgenerator waterinventory isgreatestatno-load,themagnitude anddurationoftheRCScooldownaremoreseverethansteamlinebreaksoccurring atpower.G.Incomputing thesteamflowduringasteamlinebreak,theMoodyCurve(Reference 11)forfl/D=0isused.Thefastactingsteamline isolation valvesareassumedtocloseinlessthan11secondsfromreceiptofactuation signal.The11secondclosuretimeoftheisolation valvesisbasedupontheactuating signalbeinggenerated bythesteamflowintwosteamlines-highcoincident withsteamlinepressure-lowfunctions.

Forbreaksdownstream oftheisolation valves,closureofallvalveswouldcompletely terminate theblowdown.

Foranybreak,inanylocation, nomorethanonesteamgenerator wouldexperience anuncontrolled blowdownevenifoneoftheisolation valvesfailstoclose.ResultsThelimitingcaseforCasesathroughewasshowntobethedouble-ended rupturelocatedupstreamoftheflowrestrictor withoffsitepoweravailable.

Table3.3-7liststhelimitingstatepoint forthisworstcase.Theresultspresented areconservative indication oftheeventswhichwouldoccurassumingasteamlinerupturesinceitispostulated thatalloftheconditions described aboveoccursimultaneously.

Figures3.3-50through3.3-53showtheRCStransients andcoreheatfluxfollowing amainsteamlinerupture(complete severance ofapipe)upstreamoftheflowrestrictor atinitialno-loadcondition.

Thesequenceofeventsforthistransient ispresented inTable3.3-8.Offsitepowerisassumedavailable sothatfullreactorcoolantflowexists.Thetransient shownassumesanuncontrolled steamreleasefromonlyonesteamgenerator.

Shouldthecorebecriticalatnearzeropowerwhentheruptureoccurstheinitiation ofsafetyinjection byhighdifferential pressurebetweenanysteamline andtheremaining steamlines orbyhighsteamflowsignalsincoincidence witheitherlow-lowRCStemperature orlowsteamlinemA1944<w.wpt:1d441195 pressurewilltripthereactor.Steamreleasefrommorethanonesteamgenerator willbeprevented byautomatic tripofthefastactingisolation valvesinthesteamlinesbyhighcontainment pressuresignalsorhigh.steamflowcoincident withlowsteamline pressureorlow-lowT,.Evenwiththefailureofonevalve,releaseislimitedtoapproximately 13secondsfortheothersteamgenerators whiletheonegenerator blowsdown.Thesteamlinestopvalvesareassumedtobefullyclosedinlessthan11secondsfromreceiptofaclosuresignal(steamflowintwosteamlines-highcoincident withsteamlinepressure-low).

AsshowninFigure3.3-53,thecoreattainscriticality withtheRCCAsinserted(withthedesignshutdownassumingonestuckRCCA)beforeboronsolutionat2400ppmenterstheRCS.Apeakcorepowerlessthanthenominalfullpowervalueisattained.

Thecalculation assumestheboricacidismixedwith,anddilutedbythewaterflowingintheRCSpriortoenteringthereactorcore.Theconcentration aftermixingdependsupontherelativeflowratesintheRCSandinthesafetyinjection system.Thevariation ofmassflowrateintheRCSduetowaterdensitychangesisincludedinthecalculation asisthevariation offlowrateinthesafetyinjection systemduetochangesintheRCSpressure.

Thesafetyinjection systemflowcalculation includesthelinelossesinthesystemaswellasthepumpheadcurve.NotethatsincetheRCSpressure(Figure3.3-51)dropsbelow2015psiaandneverrepressurizes abovethatvalue,theautomatic operation toopentheCentrifugal ChargingPumpMinimumFlowIsolation Valveswouldnotoccurduringthisevent.Therefore, therewouldbenotreduction inSlflowbelowthatassumedinthesafetyanalysis.

Theassumedsteamreleaseforanaccidental depressurization ofthemainsteamsystem(Casee)isthemaximumcapacityofanysinglesteamdump,relief,orsafetyvalve.Safetyinjection isinitiated automatically bylowpressurizer pressure.

Operation ofonecentrifugal chargingpumpisassumed.Boronsolutionat2400ppmenterstheRCSproviding sufficient negativereactivity topreventcoredamage.Thetransient isquiteconservative withrespecttocooldown, sincenocreditistakenfortheenergystoredinthesystemmetalotherthanthatofthefuelelementsortheenergystoredintheothersteamgenerators.

Sincethetransient occursoveraperiodofabout5minutes,theneglected storedenergyislikelyforthiseventtohaveasignificant effectinslowingthecooldown.

TheDNBtransient isboundedbythelimitingcaseforasteamline rupture.TheDNBanalysisforthelimitingcase(double-ended rupturelocatedupstreamoftheflowrestrictor) showedthattheminimumDNBRremainedabovethelimitvalue.Conclusions Theanalysishasshownthatthecriteriastatedearlieraresatisfied.

mA1944<w.wpf:1d~1195 3.3-45 AlthoughDNBandpossiblecladperforation following asteampiperupturearenotnecessarily unacceptable andnotprecluded bythecriteria, theaboveanalysis, infact,showsthatnoDNBoccursfortherupture(includiog.an accidental depressurization ofthemainsteamsystem)assumingthemostreactiveRCCAstuckinitsfullywithdrawn position.

3.3.5.7RuptureofControlRodDriveMechanism Housing(RCCAEjection)

Thisaccidentisdefinedasthemechanical failureofacontrolrodmechanism pressurehousingresulting intheejectionofaRCCAanddriveshaft.Theconsequence ofthismechanical failure,inadditiontobeingasmallbreakloss-of-coolant

accident, isarapidpositivereactivity insertion togetherwithanadversecorepowerdistribution, possiblyleadingtolocalized fuelroddamage.Thiseventhasbeenanalyzedaspartofthe30%SGTPProgramtoaddressthereduction inRCSflowduetotheincreaseintheSGTPlevel.IfanRCCAejectionaccidentweretooccur,afuelrodthermaltransient whichcouldcauseDNBmayoccurtogetherwithlimitedfueldamage.TheamountoffueldamagethatcanresultfromsuchanaccidentwillbegovernedmainlybytheworthoftheejectedRCCAandthepowerdistribution attainedwiththeremaining controlrodpattern.Thetransient islimitedbytheDopplerreactivity effectsoftheincreaseinthefueltemperature andisterminated byreactortripactuatedbyneutronfluxsignals,beforeconditions arereachedthatcanresultindamagetothereactorcoolantpressureboundafy, orsignificant disturbances inthecore,itssupportstructures orotherreactorpressurevesselinternals whichwouldimpairthecapability tocoolthecore.Theneutronfluxresponsetoacontinuous reactivity insertion ischaracterized byaveryfastfluxincreaseterminated bythereactivity feedbackeffectoftheDopplercoefficient, Thisselflimitation ofthepowerburstisofprimaryimportance sinceitlimitsthepowertoatolerable levelduringthedelaytimeforprotective action.ShouldaRCCAEjectionaccidentoccur,thefollowing automatic featuresoftheRPSareavailable toterminate thetransient, Thesource-range highneutronfluxreactortripisactuatedwheneitheroftheindependent source-range channelsindicates aneutronfluxlevelaboveapreselected manuallyadjustable setpoint.

Thistripfunctionmaybemanuallybypassedwheneitherintermediate-range fluxchannelindicates afluxlevelaboveaspecified level.Itisautomatically reinstated whenbothintermediate-range channelsindicateafluxlevelbelowaspecified level.Theintermediate-range highneutronfluxreactortripisactuatedwheneitheroftwoindependent intermediate-range channelsindicates afluxlevelaboveapreselected manuallyadjustable setpoint."7his tripfunctionmaybemanuallybypassedwhentwoofthefourpower-range channelsgivereadingsabovemA1944<w.wpf:1d~1195 3.3<6

approximately 10%offullpowerandisautomatically reinstated whenthreeofthefourchannelsindicateapowerbelowthisvalue.Thepower-range highneutronfluxreactortrip(lowsetting)isactuatedwhentwo-out-of-four power-range channelsindicateapowerlevelaboveapproximately 25%offullpower.Thistripfunctionmaybemanuallybypassedwhentwoofthefourpower-range channelsindicateapowerlevelaboveapproximately 10%offullpowerandisautomatically reinstated whenthreeofthefourchannelsindicateapowerlevelbelowthisvalue.d.Thepower-range highneutronfluxreactortrip(highsetting)isactuatedwhentwo-out-of

-fourpower-range channelsindicateapowerlevelaboveapresetsetpoint(typically 109%offullpower).Thistripfunctionisalwaysactive.e.Thehighnuclearfluxratereactortripiscalculated whenthepositiverateofchangeofneutronfluxontwo-out-of-four nuclearpower-range channelsindicates arateabovethepresetsetpoint.

Thistripfunctionisalwaysactive.Duetotheextremely lowprobability ofaRCCAEjectionaccident, thiseventisclassified asanANSCondition IVevent(Limiting Fault).Theultimateacceptance criteriaforthiseventisthatanyconsequential damagetoeitherthecoreortheRCSmustnotpreventlong-term cooling,andthatanyoffsitedoseconsequences mustbewithintheguidelines of10CFR100.Todemonstrate compliance withtheserequirements, itissufficient toshowthattheRCSpressureboundaryremainsintact,andthatnofueldispersal intothecoolant,grosslatticedistortions, orsevereshockwaveswilloccurinthecore.Therefore, thelimitingcriteriaisdescribed inReference 12,andsummarized below:A.Averagefuelpelletenthalpyathotspotbelow225caVgforunirradiated fueland200caVgforirradiated fuel.B.Averagecladtemperature atthehotspotbelowthetemperature atwhichcladembrittlement maybeexpected(3000'F).

C.Peakreactorcoolantpressurelessthanthatwhichcouldcausestressestoexceedthefaultedcondition stresslimits.D.Fuelmeltingwillbelimitedtolessthantenpercent10%ofthefuelvolumeatthehotspoteveniftheaveragefuelpelletenthalpyisbelow'the limitsofcriterion Aabove.Theanalysisperformed istoboundtheparameters associated withtheincreased SGTPlevelof30%.mA1944<w.wpf:1d~1 1853.3<7 MethodofAnalsisThecalculation oftheRCCAejectiontransient isperformed intwostages,firstanaveragecorechannelcalculation andthenahotregioncalculation.

Theaveragecorecalculation isperformed usingspatialneutronkineticsmethodstodetermine theaveragepowergeneration withtimeincluding thevarioustotalcorefeedbackeffects,i.e.,Dopplerreactivity andmoderator reactivity.

Enthalpyandtemperature transients inthehotspotarethendetermined bymultiplying theaveragecoreenergygeneration bythehotchannelfactorandperforming afuelrodtransient heattransfercalculation.

Thepowerdistribution calculated withoutfeedbackispessimistically assumedtopersistthroughout thetransient.

Adetaileddiscussion ofthemethodofanalysiscanbefoundinReference 12.AverageCoreAnalysisThespatialkineticscomputercode,TWINKLE,isusedfortheaveragecoretransient analysis.

Thiscodesolvesthetwogroupneutrondiffusion theorykineticequationinone,twoorthreespatialdimensions (rectangular coordinates) forsixdelayedneutrongroupsandupto2000spatialpoints.Thecomputercodeincludesadetailedmulti-region, transient fuel-clad-coolant heattransfermodelforcalculation ofpointwise Dopplerandmoderator feedbackeffects.Inthisanalysis, thecodeisusedasaonedimensional axialkineticscodesinceitallowsamorerealistic representation ofthespatialeffectsofaxialmoderator feedbackandRCCAmovement.

However,sincetheradialdimension ismissing,itisstillnecessary toemployveryconservative methods(described below)ofcalculating theejectedrodworthandhotchannelfactor.Furtherdescription ofTWINKLEappearsinSection3.3.3.2.HotSpotAnalysisInthehotspotanalysis, theinitialheatfluxisequaltothenominaltimesthedesignhotchannelfactor.Duringthetransient, theheatfluxhotchannelfactorislinearlyincreased tothetransient valuein0.1second,thetimeforfullejectionoftherod.Therefore, theassumption ismadethatthehotspotbeforeandafterejectionarecoincident.

Thisisveryconservative sincethepeakafterejectionwilloccurinoradjacenttotheassemblywiththeejectedrod,andpriortoejectionthepowerinthisregionwillnecessarily bedepressed.

Thehotspotanalysisisperformed usingthedetailedfuelandcladtransient heattransfercomputercode,FACTRAN.Thiscomputercodecalculates thetransient temperature distribution inacrosssectionofametalcladUO,fuelrod,andtheheatfluxatthesurfaceoftherod,usingasinputthenuclearpowerversustimeandthelocalcoolant'conditions.

Thezirconium-water reactionisexplicitly represented, andallmaterialproperties arerepresented asfunctions oftemperature.

Aconservative radialpower.distribution isused.withinthefuelrod.m:51944<w.wpf:1d441195 3.3<8

FACTRANusestheJens-Lottes orDittus-Boelter correlation (References 8and15,respectively) todetermine thefilmheattransferbeforeDNB,andtheBishop-Sandberg-Tong correlation (seeReference 13)todetermine thefilmboilingcoefficient afterDNB.TheBishop-Sandberg-Tong correlation is'consewatively usedassumingzerobulkfluidquality.TheDNBratioisnotcalculated, insteadthecodeisforcedintoDNBbyspecifying aconservative DNBheatflux.Thegapheattransfercoefficient canbecalculated bythecode;however,itisadjustedinordertoforcethefullpowersteady-state temperature distribution toagreewiththefuelheattransferdesigncodes.Furtherdescription ofFACTRANappearsinSection3.3.3.2.Adetailedthree-dimensional calculation ofaworstcasescenario(Reference 12)demonstrates anupperlimittothenumberofrods-in-DNB fortheRCCAEjectionaccidentas10%.SincetheseverityoftheCookNuclearPlantUnit1analysisdoesnotexceedthisworstcaseanalysis, themaximumnumberofrodsinDNBfollowing aRCCAEjectionwillbelessthan10%,althoughneitherthenumberofrodsinDNBnortheminimumDNBRvalueisexplicitly calculated intheCookNuclearPlantUnit1analysis.

Themostlimitingbreaksizeresulting fromaRCCAEjectionwillnotbesufficient touncoverthecoreorcauseDNBatanylatertime.Sincethemaximumnumberoffuelrodsexperiencing DNBislimitedto10%,thefissionproductreleasewillnotexceedthatassociated withtheguidelines of10CFR100.SystemOverpressure AnalysisBecausesafetylimitsforfueldamagespecifled earlierarenotexceeded, thereislittlelikelihood offueldispersal intothecoolant.Thepressuresurgemaytherefore bycalculated onthebasisofconventional heattransferfromthefuelandpromptheatgeneration inthecoolant.Thepressuresurgeiscalculated byfirstperforming thefuelheattransfercalculation todetermine theaverageandhotspotheatfluxversustime.Usingthisheatfluxdata,aTHINGIVcalculation isconducted todetermine thevolumesurge.Finally,thevolumesurgeissimulated intheLOFTRANcomputercode.Thiscodecalculates thepressuretransient takingintoaccountfluidtransport intheRCSandheattransfertothesteamgenerators.

Nocreditistakenforthepossiblepressurereduction causedbytheassumedfailureofthecontrolrodpressurehousing.Inputparameters fortheanalysisareconservatively selectedonthebasisofvaluescalculated forthistypeofcore.Themoreimportant parameters arediscussed below.Table3.3-9presentstheparameters usedinthisanalysis.

mA1944<w.wpf:1d441195 3.3-49

EjectedRodWorthsandHotChannelFactorsThevaluesforejectedrodworthsandhotchannelfactorsarecalculated usingeitherthreedimensional staticmethodsorbyasynthesis methodemploying onedimensional andtwodimensional calculations.

Standardnucleardesigncodesareusedintheanalysis.

Nocreditistakenforthefluxflattening effectsofreactivity feedback.

Thecalculation isperformed forthemaximumallowedbankinsertion atagivenpowerlevel,asdetermined bytherodinsertion limits.Adversexenondistributions areconsidered inthecalculation toprovideworstcaseresults.Appropriate marginsareaddedtotheejectedrodworthandhotchannelfactorstoaccountforanycalculational uncertainties, including anallowance fornuclearpowerpeakingduetodensification.

Powerdistribution beforeandafterejectionforaworstcasecanbefoundinReference 12.Duringplantstartupphysicstesting,ejectedrodworthsandpowerdistributions aremeasuredinthezeroandfullpowerconfigurations andcomparedtovaluesusedintheanalysis.

Experience hasshownthattheejectedrodworthandpowerpeakingfactorsareconsistently overpredicted intheanalysis.

Reactivity FeedbackWeighting FactorsThelargesttemperature rises,andhencethelargestreactivity feedbacks occurinchannelswherethepowerishigherthanaverage.Sincetheweightofaregionisdependent onflux,theseregionshavehighweights.Thismeansthatthereactivity feedbackislargerthanthatindicated byasimplechannelanalysis.

Physicscalculations havebeencarriedoutfortemperature changeswithaflattemperature distribution, andwithalargenumberofaxialandradialtemperature distributions.

Reactivity changeswerecomparedandeffective weighting factorsdetermined.

Theseweighting factorstaketheformofmultipliers which,whenappliedtosinglechannelfeedbacks, correctthemtoeffective wholecorefeedbacks fortheappropriate fluxshape.Inthisanalysis, sinceaonedimensional (axial)spatialkineticsmethodisemployed, axialweighting isnotnecessary iftheinitialcondition ismadetomatchtheejectedrodconfiguration.

Inaddition, noweighting isappliedtothemoderator feedback.

Aconservative radialweighting factorisappliedtothetransient fueltemperature toobtainaneffective fueltemperature asafunctionoftimeaccounting forthemissingspatialdimension.

Theseweighting factorshavealsobeenshowntobeconservative comparedtothreedimensional analysis(Reference 12).Moderator andDopplerCoefficient Thecriticalboronconcentrations atthebeginning oflifeandendoflifeareadjustedinthenuclearcodeinordertoobtainmoderator densitycoefficient curveswhichareconservative m%1944<w.wpl:

1d4411953.3-50 comparedtoactualdesignconditions fortheplant.Asdiscussed above,noweighting factorisappliedtotheseresults.Theresulting moderator temperature coefficient isatleast+5pcm/'Fattheappropriate zeroor,fullpowernominalaveragetemperature, andbecomeslesspositiveforhighertemperatures.

Thisisnecessary sincetheTWINKLEcomputercodeutilizedintheanalysesisadiffusion-theory coderatherthanapoint-kinetics approximation andthemoderator temperature feedbackcannotbeartificially heldconstantwithtemperature.

TheDopplerreactivity defectisdetermined asafunctionofpowerlevelusingaonedimensional steady-state computercodewithaDopplerweighting factorof1.0.TheDopplerweighting factorwillincreaseunderaccidentconditions, asdiscussed above.DelayedNeutronFraction, P,Calculations oftheeffective delayedneutronfraction(P,)typically yieldvaluesnolessthan0.70%atbeginning oflifeand0.50%atendoflife.Theaccidentissensitive toP,iftheejectedrodworthisequaltoorgreaterthanp,asinzeropowertransients.

Inordertoallowforfuturecycles,pessimistic estimates ofp,of0.50%atbeginning ofacycleand0.40%atendofacyclewereusedintheanalysis.

TripReactivity Insertion Thetripreactivity insertion assumedisgiveninTable3.3-9andincludestheeffectofonestuckRCCAadjacenttotheejectedrod.Thesevaluesarereducedbytheejectedrodreactivity.

Theshutdownreactivity wassimulated bydroppingarodoftherequiredworthintothecore.Thestartofrodmotionoccurred0.5secondsafterthehighneutronfluxtrippointsisreachedbeforesignificant shutdownreactivity isinsertedintothecore.Thisisparticularly important conservatism forhotfullpoweraccidents.

Theminimumdesignshutdownmarginavailable forthisplantatHZPmaybereachedonlyatendoflifeintheequilibrium cycle.Thisvalueincludesanallowance fortheworststuckrod,anadversexenondistribution, conservative Dopplerandmoderator defects,andanallowance forcalculational uncertainties.

Physicscalculations haveshownthattheeffectoftwostuckRCCAs(oneofwhichistheworstejectedrod)istoreducetheshutdownbyaboutanadditional 1%hk/k.Therefore, following areactortripresulting fromanRCCAejectionaccident, thereactorwillbesubcrltical whenthecorereturnstoHZP..&Depressurization calculations havebeenperformed assumingthemaximumpossiblesizebreak(2.75inchdiameter) locatedinthereactorpressurevesselhead.Theresultsshowarapidpressuredropandadecreaseinsystemwatermassduetothebreak.TheECCSisactuatedonlowpressurizer pressurewithinoneminuteafterthebreak.TheRCSpressurecontinues todropandreachessaturation (1100to1300psidepending onthesystemtemperature) inabouttwotothreeminutes.DuetothelargethermalinertiaofprimaryandrnA1944<w.wpf:1d441 1953.3-51

secondary system,therehasbeennosignificant decreaseintheRCStemperature belowno-loadbythistime,andthedepressurization itselfhascausedanincreaseinshutdownmarginbyabout0.2%d,k/kduetothe,pressurecoefficient.

Thecooldowntransient couldnotabsorbtheavailable shutdownmarginuntilmorethan10minutesafterthebreak.Theadditionofboratedsafetyinjection flow(supplied fromtheRWST)startingoneminuteafterthebreakismuchmorethansufficient toensurethatthecoreremainssubcritical duringthecooldown.

ReactorProtection Reactorprotection forarodejectionisprovidedbyhighneutronfluxtrip(highandlowsetting)andhighrateofneutronfluxincreasetripalthoughtheanalysismodeledthehighneutronfluxtrip(highandlowsetting)only.Theseprotection functions arepartofthereactortripsystem.Nosinglefailureofthereactortripsystemwillnegatetheprotection functions requiredfortherodejectionaccident, oradversely affecttheconsequences oftheaccident.

ResultsTable3.3-9summarizes theresults.Casesarepresented forbothbeginning andendoflifeatzeroandfullpower.A.Beginning ofCycle,FullPowerControlBankDwasassumedtobeinsertedtoitsinsertion limit.Theworstejectedrodworthandhotchannelfactorwereconservatively calculated tobe0.15%b,k/kand6.8respectively.

Thepeakcladaveragetemperature was2299'F.Thepeakspotfuelcentertemperature reachedmelting,conservatively assumedat4900'F.However,meltingwasrestricted tolessthan10%ofthepellet.B.Beginning ofCycle,ZeroPowerForthiscondition, ControlBankDwasassumedtobefullyinsertedandbanksBandCwereattheirinsertion limits.TheworstejectedrodislocatedinControlBankDandhasaworthof0.65%d,k/kandahotchannelfactorof12.0.Thepeakcladaveragetemperature reached2130'F,thefuelcentertemperature was3120'F.C.EndofCycle,FullPowerControlBankDwasassumedtobeinsertedtoitsinsertion limit.Theejectedrodworthandhotchannelfactorswereconservatively calculated tobe0.19%m:0944<w.wpf:1d~1195 3.3-52 d,k/kand7.1respectively.

Thisresultedinapeakcladaveragetemperature of2245'F.Thepeakhotspotfuelcentertemperature reachedmeltingat4800'F.However,meltingwas.restricted tolessthan10%ofthepellet.D.EndofCycle,ZeroPowerTheejectedrodworthandhotchannelfactorforthiscasewereobtainedassumingControlBankDtobefullyinsertedandbanksBandCattheirinsertion limits.Theresultswere0.75%d,k/kand19.0respectively.

Thepeakcladaverageandfuelcentertemperatures were2322'Fand3258'F.The,Dopplerweighting factorforthiscaseissignificantly higherthanfortheothercasesduetothevefylargetransient hotchannelfactor.Forallthecasesanalyzed, averagefuelpelletenthalpyatthehotspotremainsbelow200cal/g.Thenuclearpowerandhotspotfuelandcladtemperature transients fortwocases(endoflifezeropowerandendoflifefullpower)arepresented inFigures3.3-54through3.3-57.TheejectionofanRCCAconstitutes abreakintheRCS,locatedinthereactorpressurevesselhead.Following theRCCAejection, theoperatorwouldfollowthesameemergency instructions asforanyotherLOCAtorecoverfromtheevent.PressureSurgeAdetailedcalculation ofthepressuresurgeforanejectionworthofonedollaratbeginning oflife,hotfullpower,indicates thatthepeakpressuredoesnotexceedthatwhichwouldcausestresstoexceedthefaultedcondition stresslimits(Reference 12).Sincetheseverityofthepresentanalysisdoesnotexceedthe"worstcase"analysis, theaccidentforCookNuclearPlantUnit1willnotresultinanexcessive pressureriseorfurtheradverseeffectstotheRCS.LatticeDeformatf'ons Alargetemperature gradientwillexistintheregionofthehotspot.Sincethefuelrodsarefreetomoveintheverticaldirection, differential expansion betweenseparaterodscannotproducedistortion.

However,thetemperature gradients acrossindividual rodsmayproduceadifferential expansion tendingtobowthemidpointoftherodstowardthehottersideoftherod.Calculations haveindicated thatthisbowingwouldresultinanegativereactivity effectatthehotspotsinceWestinghouse coresareunder-moderated, andbowingwilltendtoincreasetheunder-moderation atthehotspot.Inpractice, nosignificant bowingisanticipated, sincethestructural rigidityofthecoreismorethansufficient towithstand theforcesproduced.

Boilinginthehotspotregionwouldproduceanetflowawayfromthatregion.However,theheatmA1944<w.wpf:1d~1195 3.3-53

fromthefuelisreleasedtothewaterrelatively slowly,anditisconsidered inconceivable thatcrossflowwillbesufficient toproducesignificant latticeforces.Evenifmassiveandrapidboiling,sufficient todistortthelattice,ishypothetically postulated, thelargevoidfractioninthehotspotregionwouldproduceareduction inthetotalcoremoderator tofuelratio,andalargereduction inthisratioatthehotspot.Theneteffectwouldtherefore beanegativefeedback.

ltcanbeconcluded thatnoconceivable mechanism existsforanetpositivefeedbackresulting fromlatticedeformation.

Infact,asmallnegativefeedbackmayresult.Theeffectisconservatively ignoredintheanalysis.

Conclusions Evenonapessimistic basis,theanalysesindicatethatthedescribed fuelandcladlimitsarenotexceeded.

Itisconcluded thatthereisnolikelihood ofsuddenfueldispersal intothecoolant.Sincethepeakpressuredoesnotexceedthatwhichwouldcausestressestoexceedthefaultedcondition stresslimits,itisconcluded thatthereisnolikelihood offurtherconsequence totheRCS.Theanalyseshavedemonstrated thefissionproductreleaseasaresultoffuelrodsenteringDNBislimitedtolessthan10%ofthefuelrodsinthecore.3.3.5.8Steamline BreakMass/Energy ReleasesInsideContainment Thenon-LOCAdiscussion regarding thereanalysis ofthesteamline breakmassandenergyreleasesinsidecontainment canbefoundinSections3.5.4and3.5.5.Itshouldbenotedthatthechangesassociated withtheSGTPProgramforUnit1,i.e.,RCSflowreduction, reducedpnmary-to-seconda1y heattransfercapability, andreduction intheratedthermalpower,arelesslimitingparameters relativetotheassumptions currently madefortheM/Ereleasecalculations following aSLBinsidecont'ainment.

Theparameter changesassociated withtheSGTPprogramdonotwarrantreanalysis ofthisevent.However,evaluations arecurrently inplace(References 7and17)toaddressseveralnon-conservative assumptions intheanalysis.

Areanalysis effortwasundertaken forthesteamline breakmassandenergyreleasesinsidecontainment aspartoftheSGTPProgram,suchthattheReference 7and17evaluations willnolongerberequired.

3.3.6Conclusions oftheNon-LOCASafetyEvaluation Thenon-LOCAsafetyanalysesandevaluations presented inthissectionsupporttheoperation ofDonaldC.CookNuclearPlantUnit1withSGTP,asdescribed inTable3.3-1(Cases1and2).References 1.Ellenberger S.L.etal.,"DesignBasesfortheThermalOverpower hTandThermalOvertemperature dTTripFunctions,"

WCAP-8746, March,1977.m:11944<w.wpf:1d~1195 3.3-54 2.Friedland, A.J.,Ray,S.,"RevisedThermalDesignProcedure,"

WCAP-11397-A, April,1989.3.Hargrove, H.G.,"FACTRAN-AFORTRAN-IV CodeforThermalTransients inaUO,FuelRod,"WCAP-7908, June,1972.4.Burnett,T,W.T.,etal.,"LOFTRANCodeDescription,"

WCAP-7907-A, April1,1984.5.Risher,D.H.,Jr.,andBarry,R.F.,"TWINKLE-aMulti-Dimensional NeutronKineticsComputerCode,"WCAP-8028-A, January,1975.6.Friedland, A.J.andRay,S.,"Improved THINC-IVModelingforPWRCoreDesign,"WCAP-12330-P, August1989.7."American ElectricPowerServiceCorporation, DonaldC.CookNuclearPowerPlantUnits1and2,Increased Ur8LowerComartmentSraDeliveTimes,"WLetterAEP-94-712, June13,1994.8.W.H.Jens,P.A.Lottes,"Analysis ofHeatTransfer, Burnout.PressureDrop,andDensityDataforHigh-Pressure Water,"U.S.AECReportANL<627(1951).9.Haessler, R.L.,et.al.,"Methodology fortheAnalysisoftheDroppedRodEvent,"WCAP-11394-P-A andWCAP-11395-NP-A, January1990.10.Baker,L.,andJust,L.,"StudiesofMetalWaterReactions ofHighTemperatures, IIIExperimental andTheoretical StudiesoftheZirconium-Water Reaction,"

ANL-6548, ArgonneNationalLaboratory, May,1962.11.Moody,F.S.,"Transitions oftheASME,JournalofHeatTransfer,"

Figure3,Page134,February1965.12.Risher,D.H.,Jr.,"AnEvaluation oftheRodEjectionAccidentofWestinghouse Pressurized WaterReactorsUsingSpatialKineticsMethods,"

WCAP-7588, Revision1A.13.Bishop,A.A.,Sandberg, R.O.,andTong,L.S.,"ForcedConvection HeatTransferatHighPressureAftertheCriticalHeatFlux,"ASME65-HT-31, August,1965.14.DonaldC.CookNuclearPlantUnit1UpdatedFinalSafetyAnalysisReport,USNRCDocketNumber50-315,updatedthrough1993.mA1944<w.wpf:

1d4411953.3-55 15.F.W.Dittus,C.N.Boelter,"HeatTransferinAutomobile Radiators oftheTubularType,"Calif.Univ.Publication inEng.,2of13,4433-461(1930).16.Christensen, J.A.,et.al.,"MeltingPointofIrradiated UraniumDioxide,"

Transactions oftheAmericanNuclearSocie7,1964.17."American ElectricPowerServiceCorporation, DonaldC.CookNuclearPlantUnits1and2,Feedwater Isolation ValveEvaluation Suort"WLetterAEP-93-528, April8,1993.18.LetterfromWilliamO.Long,Sr.(USNRC)toEugeneE.Fitzpatrick (AEPSC),

Subject:

Amendment Nos.158and142toFacilityOperating LicenseNos.DPR-58andDPR-74(TACNos.80262and80263),datedNovember20,1991.19.SECL-91-429, "DonaldC.CookUnits1and2MainSteamSafetyValveLiftTolerance Relaxation,"

March1992.m:51944<w.wpt:1d441195 3.3-56 TABLE3.3-1DONALDC.COOKNUCLEARPLANTUNIT1NSSSPERFORMANCE PARAMETERS USEDINNON-LOCASAFETYANALYSESParameter NSSSPower,MWt(30%SGTPProgram)Case1Case232623262Case33425Case43425(Rerating Program)"'orePower,MWtRCSFlow,gpm/loop"'inimumMeasuredFlow,totalgpm@RCSTemerature'FCoreOutlet325083200339,100589.7325083200339,100611.93413366,400t')583.6341388500366,400t'~614.0VesselOutlet586.8CoreAverage555.8VesselAverage553.0Vessel/Core Inlet519.2SteamGenerator Outlet518.9ZeroLoad547.0609.1579.4576.3543.5543.2547.0580.7549.7547.0513.3513.1547.0611.2581.8578.7546.2546.0547.0RCSPressure, psia2250or21002250or21002250or21002250or2100SteamPressure, psiaSteamFlow(10'b/hrtotal)Feedwater Temp.,'FSGTubePlugging,

%59514.12434.83074914.17434.83060314.981082015.07442.~10CookUnit1isnotlicensedtooperateatthereratedconditions specified byCases3and4with30%steamgenerator tubeplugging(SGTP)levels.However,severaleventsthatwerepreviously performed usingtheseconditions weresubsequently evaluated tosupportthe30%SGTPprogram.Hence,thereratedconditions areahospecified inthistableforcompleteness.

RCSFlow(ThermalDesignFhw)-Theconservatively lowfhwusedforthermal/hydraulic design.Thedesignparameters listedabovearebaseduponthisflow.MinimumMeasuredFlow-Theflowspecified intheTechnical Specifications whichmustbeconfirmed orexceededbytheflowmeasurements obtainedduringphntstartupandistheflowusedinreactorcoreDNBanalysesforphntsapplyingtheRevisedThermalDesignProcedure.

MMFbasedupona1.9%flowmeasurement uncertainty..Analyses alsoboundaMMFof341,100gpmwhichreflectsa2.5%flowmeasurement uncertainty.

AMMFof366,400gpmwasassumedintheReratingProgramanalyses.

Asafetyevaluation wasperformed tosupportareduction ofMMFto361,600gpm(SECL-90-280).

m:51944<w.wpf:11441195 3.3-57 fi1IIII TABLE3.3-2TRIPPOINTSANDTIMEDELAYSTOTRIPASSUMEDINNON-LOCAACCIDENT, ANALYSIS'ri FunctionLimitingTripPointAssumedTimeDelay~SecondsPowerrangehighneutronflux,highsettingPowerrangehighneutronflux,lowsettingOvertemperature dTOverpower hTHighpressurizer pressureLowpressurizer pressureHighpressurizer waterlevelLowreactorcoolantflow(Fromloopflowdetectors)

Undervoltage tripLow-lowsteamgenerator levelHighsteamgenerator levelTurbineTripFeedwater Isolation 118percent35percentVariable, seeFigure3.3-1through3.3-4andTable3.3-3Variable, seeFigure3.3-1through3.3-4andTable3.3-32420psig1825psig100%NRS87percentloopflow0.0percentofnarrowrangelevelspan82percentofnarrowrangelevelspan0.50.58.0'.02.02.02.01.01.52.011.0Toutetimedelay(including RTDbypassloopfluidtransport delayeffect,bypasslooppipingthermalcapacity, RTDtimeresponse, andtripcircurt,channelelectronics dehy)fromthetimethetemperature difference inthecoolanthopsexceedsthetripsetpointuntiltherodsarefreetofall.Thetimedelayassumedintheanalysissupportsthe6secondresponsetimeoftheRTDtimeresponse, tripcircuitdelays,andchannelelectronics dehypresented intheTechnical Specifications.

Noexplicitvalueassumedintheanalysis.

Undervoltage tripsetpointassumedreachedatinitiation ofanalysis.

Thecontrolrodscramtimetodashpotis2.4secondsOverpower hTreactortripwasassumedinthesteamline breakmasslenergy releaseoutsidecontainment calculations.

mA1944<w.wpt:1d 4411953.3-58 t,l,It TABLE3.3-3OTLTANDOPbTSETPOINTEQUATIONANDSAFETYANALYSISLIMITCOEFFICIENT VALUESOvertemperature hTequation:

OT~T54TO[K)-K2[](T-T)+(P-P)fi(>l)1+t~swhere,K,=1.35K=0.02322seconds4secondss=Laplacetransform operatorT'553.0to576.3'F0.0011P=2100or2250psiaf,(dI):Dead-band:

from-37to+3%6,lPositiveWing:2.34/J'/od,l foreachpercentAl>+3%6,lNegativeWing:0.33'/d/&I foreachpercentbl<-37%5,lOverpower dTequation:

OPBT-~To[K4K,['T-K,(T-T")-f,(~l)1+cswhere,K,73sTllPSf(bl):1.1720.0177;thisgainisnotmodeledinthenon-LOCAsafetyanalyses10secondsLaplacetransform operator553.0to563.0'F0.00152100or2250psia0mal944<w.wpf:1d~1195 3.3-59 TABLE3.3.4SUMMARYOFINITIALCONDITIONS ANDCOMPUTERCODESUSEDFaultsUncontrolled RodClusterAssemblyBankWithdrawal fromaSubcritical CondbonComputerCodesUtilizedTWINKLEFACTRANTHINGIVModerator Temperature J~P~FModerator Density~hK/ml~cc~DoerRefertoSection3.3.5.1Min(I)ReactiviCoefficients AssumedDNBCorrelation WQIWRB-ISeeSection3.3.4.3RevisedThormalDesignProcedure NoReactorInitialNSSSThermalPower~OoolMWlVesselVesselCoolant~FlowGPM146,432AverageTemperature

~FPressurizer Pressure~PGIAr2033Uncontrolled RodClusterAssemblyBankWithdrawal AtPower(2)RodClusterControlAssemblyMisalignment LOFTRANLOFTRANTHINGIV+5NA'inandMax(3)NAWRB-1WRB-1YesYes327019623273270339,100339,100576.3564.58549.93576.321002100Uncontrolled BoronDilutionNANANANANA34250NANANALossofForcedReactorCoolantFlowLOFTRANFACTRANTHINGIVt5NAWRB-1Yes3270339.100576.32100LockedRotor(PeakPressure)

LockedRotor(PeakCladTemp)LockedRotor(Rods-in-DN8)

LOFTRANLOFTRANFACTRANLOFTRANFACTRANTHINGIV+5+5+5NANANANANAWRB-1NANAYes333533353270332,800332,800339,100581.4581.4576.323172100'A-NotApplicable (1)MinimumDopplerpowerdefect(porn//.power)

=-9.55+0.0350where0isin%power.(2)Multiplepowerlevels,Tavg,andreactivity feedbackcaseswereexamined.

(3)MaximumDopplerpowerdefect(pcmP/Gpower)=-19.4+0.0650.(4)MinimumandMaximumreacbvity feedbackcaseswereexamined.

mtt944.4w.wpf:1d-041195 3.3-60 TABLE3.3.4(continued)

SUMMARYOFINITIALCONDITIONS ANDCOMPUTERCODESUSEDFaultsComputerCodesUtilizedReactiviCoefficients AssumedModerator Moderator Temperature DensityQc~ml'F~alt'Jr~Ice DoODlerDNBCorrelation RevisedThermalDesignPlooedule InitialNSSSThermalPower~OuulMWtReactorVesselCoolant~FlowGPMVosselAverageTemperature

~FPressurizer Pressure~PSIALossofElectrical LoadandforTurbineTrip(4)LOFTRAN+5.54MaxandMinWRB-1Yes3262339,100576.32100Excessive HeatRemovalDuetoFeedwater SystemMalfunction (5)LOFTRANLossofNormalFeedwator (5)LOFTRAN+5.54MinNAWRB-1Yes349434250354,000366,400551.5578.754722852100ExcessLoadIncreaseIncident(5)LOFTRANNA0and.54MaxandWRB-1MinYes3425366,400578.72100LossofOlfsitePowerlotheStationAuxiliaries (5)LOFTRAN+5NANA354,000542.52285RuptureolaSteamPipeRuptureofaControlRodDriveMechanism HousingLOFTRANTHINGIVTWINKLEFACTRANSeeFigureNA3.3w47SeeSectionNA3.3.5.7SeeFigureW.33.3-48NA33350332,800332,800146,432547581.45472100'A-NotApplicable (1)MinimumDopplerpowerdefecl(pcs%%dpower)

=-9.55+0.035QwhereQisin%power.(2)Multiplepowerlevels,Tavg,andreactivity feedbackcaseswereexamined.

(3)MaximumDopplerpowerdefect(pcml%%dpower)=~19.4+0.065Q.(4)MinimumandMaximumreactivity feedbackcaseswereexamined.

(5)Valuespresented conespond totherespective reratinganalysis.

Subsequent evaluabons supportthe30%%dSGTPparameters givenasCases1and2ofTable3.3-1.m'.11944.4 w.wpf:1d.041195 3.3-61 TABLE3.3-5SEQUENCEOFEVENTSFORLOSSOFFLOWANDLOCKEDROTORACCIDENTS AccidentCompleteLossofFlowPartialLossofFlowLockedRotorEventAllpumpslosepowerandbegincoastingdown,undervoltage tripsignalgenerated RodsbegintodropMinimumDNBRoccursOneoperating pumplosespowerandbeginscoastingdownLowreactorcoolantflowtripsetpointreachedinfaultedloopRodsbegintodropMinimumDNBRoccursOnepumprotorseizesLowreactorcoolantflowtripsetpointreachedinfaultedloopRodsbegintodropMaximumpercentage ofrodsinDNBpredicted MaximumRCSpressureoccursMaximumcladtemperature occursTimesec.0.01.503.400.01.742.743.900.00.041.042.63.203.49m:51944<w.wpf:1d441195 3.3-62 TABLE3.3-6SEQUENCEOFEVENTSFORLOSSOFEXTERNALELECTRICAL LOADCaseMinimumFeedbackwithPressureControlMaximumFeedbackwithPressureControlMinimumFeedbackwithoutPressureControlMaximumFeedbackwithoutPressureControlEventLossofexternalelectrical loadOTBTtripsetpointreachedPeakRCSpressureoccursRodsbegintodropMinimumDNBRoccursLossofexternalelectrical loadMinimumDNBRoccursPeakRCSpressureoccursLow-lowsteamgenerator leveltripsetpointreachedRodsbegintodropLossofexternalelectrical loadMinimumDNBRoccursHighpressurizer pressuretripsetpointreachedRodsbegintodropPeakRCSpressureoccursLossofexternalelectrical loadMinimumDNBRoccursHigh.pressurizer pressuretripsetpointreachedRodsbegintodropPeakRCSpressureoccursTimesec.0.014.215.516.218.00.00.010.068.170.10.00.08.410.412.00.00.08.910.912.5mA1944<w.wpf:1d~1195 3.3-63

TABLE3.3-7LIMITINGSTEAMLINE BREAKSTATEPOINT DOUBLEENDEDRUPTUREINSIDECONTAINMENT WITHOFFSITEPOWERAVAILABLE HeatInletTimePressureFluxColdTemp.FlowBoronReactivity secpsiaFraction'FHot'FFractionPPMPercentDensitygm/cc180.2601.93.228336.6463.31.07.13.001.849mA1944<w.wpf:1d~1 1953.3-64 TABLE3.3-8TIMESEQUENCEOFEVENTSDOUBLEENDEDRUPTUREINSIDECONTAINMENT WITHOFFSITEPOWERAVAILABLE EventSteamlineruptureoccursLowsteamlinepressurecoincident withhighsteamflowintwosteamlinesreachedFeedwater Isolation (Allloops)Criticality attainedSteamline Isolation (Loops2,3and4)Pressurizer emptiesSlflowstartsBoronfromSlreachesthecorePeakheatfluxattainedCorebecomessubcritical Timesec0.002.0610.0612.4013.0613.2029.0639.80179.2180.0m:11944<w.wpf:1d~1195 3.3-65

TABLE3.3-9PARAMETERS USEDINTHEANALYSISOFTHERODCLUSTERCONTROLASSEMBLYEJECTIONACCIDENTTimeinLifePowerLevel(%)EjectedRodWorth(%b,k)DelayedNeutronFractionFeedbackReactivity Weighting TripReactivity

('iMk)F,BeforeRodEjectionF,AfterRodEjectionNumberofOperational PumpsMaximumFuelPelletAverageTemperature

('F)MaximumFuelCenterTemperature

('F)MaximumCladAverageTemperature

('F)MaximumFuelStoredEnergy(cal/gm)FuelMeltinHotPellet,%HZP~Beinnin0.650.00502.0712.2.5012.276431202130112.7HFPBe<einning 1020.150.00501.302.506.8405649682299<10HZPEnd0.750.00402.7552.5019.296332582322122.2HFPEnd1020.190.00401.302.507.1396948722245172.7<10m%1944<w.wpf:1d441195 3.3-66 1IIt 858075706560~55CI5045i1840psia~oooooo~oooo~ooooo2400psiax2100psia~,"'toooo~~~ooo\OPETTrip~ooooo~oooo~oooo~oooooooo'oooooooooSGSafetyOpen35CareLimitsOTBTTrip%~30560580600T('F)620DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-1Illustration ofOvertcntperature andOverpower dTProtection NominalTavg=576.3'FNominalPressure=2100psiam:51944-5w.wpt:1d441195 3.3%7 8580757065~55CI504535l840psia~kT'oooooOPhTTrip~o~oo'ooooooo~oo~ooo~oo'ooCoreLimits2400psia%50psia~ooooooo~oooo//>rSGSafetyrxOTBTTrip~ValvesOpen30560580600T,('F)620DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-2Blustration ofOvertemperature andOverpower ATprotection NominalTave=576.3'FNominalPressure~2250psiam:11944.5w.wpt:1dM1195 3.348

8580757065~60+55CI501840psia%oooooooooooooo 2250psia%oooooooo~oooo\2250psia1840psiaS\2400psiaoooo~".....OphTTrip~ooooo~ooo~ooooooo~oo4535CoreLimitsOTETTrip30560580600T,s('8620DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-3Illustration ofOvertemperature andOverpower 4TProteeuon NominalTavg=553.0'FNominalPressure=2250psiam:11944.5w.wpf:1dO41195 3.3<9 8580757065~60~55504535~sosbassocrossSGSafetyValvesOpenCoreLimitsOTETTrip2100psia2400s~sss~oYosooso'ti840psra~oooo\2400psiati2100psiaii840psut\'\\\'\OPETTrip~osscross~sos~~~~oos30580600T,g('8620DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3<Illustration ofOvertemperature andOverpower hTProtection NominalTavg=553.0'FNominalPressure=2l00psiam:$1944-5w.wpt:1d441195 3.3-70 1.0E+1CUCEC0C0C$OnC5c3z0.01.0E-I1.0E-2010Time(s)1520250.5IEg0.400.3um0.2ZCC~0.1OOx0.0010Time(s]152025DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-5NuclearPowerandHotChannelHeatFluxvs.TiaraFor,TheRodWithdrawal FromSubcritical Eventm:119444w.wpf:1d441195 3.3-71

2,400o2,QQQc1,600E~IDeg1,200PCl80010Time[s]152025750u~700K~650~I~6000550500010Time[s]152025DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-6FuelAverageandCladTemperature vs.TimeForTheRodWithdrawal FromSubcritical Eventm:11944.5w.wpt:1d~1195 3.3-72 1.4+1.2~1.0QI-0.8~0.6Q0.4z0.20.00Time[s)10DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-7NuclearPowervs.TimeForTheRCCAWithdrawal AtPowerEvent,FullPower,80PCM/Sec.Insertion RateMaxiamnReactivity Feedbackm:$1944-5w.wpf:

1d44119530373 2,3002,250o2300n2,150Ng2,100n.2,0502,0000Time[s]101,200o1,150EIm1,100lI1,050Q.1,0000Time[s)10DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-8Pressurizer PressureandPressurizer WaterVolumevs.TimeForTheRCCAWithdrawal AtPowerEvent,FullPower,80PCM/Sec.Insertion RateMaximumReactivity Feedbackm&1944-5w.wpf:1d441195 3.3-74

590uo580f-570I560OO5500Time[s]104.03.53.02.52.01.50Timefs]10DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-9CoreAverageTemperature andDNBRvs.TiaraForTbeRCCAWahdrawal AtPowerEvent,FullPower,80PCM/Sec.Insertion RateMaximumReactivity Feedbackm:11944-5w.wpf:1d

~11953.3-75 1.41.2I~1.00.I-0.8~0.6O~0.4z0.20.0050150200250300350Time[s]DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-10NuclearPowervs.TimeForTheRCCAWithdrawal AtPowerEvent,FullPower,4PCM/Sec.Insertion RateMaxittmnReactivity Feedbackm&1944-5w.wpf:1dM1195 3.3-76

2,200~2,160g2,120n..a2,0805n-2,0402,000050100150200250300350Timefs]1,5001,400N>1,300'5e1,200N'C8n1,1001,000050100150200250300350Time(s]DONALDC.COOKNUCLEARI'LAM'NIT 1FIGURE3.3-11Pressurizer PressureandPressurizer WaterVolumevs.TimeForTheRCCAWithdrawal AtPowerEvent,FullPower,4PCM/Sec.Insertion RateMaximumReactivity Feedbackm:11944.5w.wpf:1d~1195 30377 o595cL.590E~$m585o580575050100150200250300350Time[s]4.03.53.02.52.01.5050150200250300350Time[s]DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-12CoreAverageTemperature andDNBRvs.TuneForTheRCCAWithdrawal AtPowerEvent,FullPower,4PCM/Sec.Insertion RateMaximumReactivity Feedbackm:$1944-5w.wpl:1dO41195 3.3-78

2.0II//mz1.80E'a1.70venemperature 4TTripHighNeutronFluxTrip1.6Min.FeedbackMax.Feedback1.50.331030Reactivity Insertion Rate[PCM/Sec) 100DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-13MiniinimDNBRvs.Reactivity Insertion RateForTheRCCAWithdrawal AtPowerEvent,1009oPowermi1944-5w.wpt:1d441195 3.3-79

2.42.2tLz2.00Ea1.80vertemperature hTTripKghNeutronFluxTrip1.6Min.FeecbackMax.Feedback1.40.3131030Reactivity Insertion Rate[PCM/Sec)

DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-14MinitmunDNBRvs.Reactivity Insertion RateForTheRCCAWithdrawal AtPowerEvent,60%Powerm:11944-5w.wpt:1d441195 3.3-80 2.8Min.Feetataek Max.Feedback2.6Kz2.40E22HighNeutronRuxTrip~2.0Overtemperarure ETTrip1.80.331030Reactivity Insertion Rate(PCM/Sec) 100DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-15MinitmtmDNBRvs.Reactivity Insertion RateForTheRCCAWithdrawal AtPowerEvent,10%Powerm119444w.wph1d4141195 3.3-81

l.t000l.1000l.0000QCO.SOOOO.00000O.70000.COOOOR.50000o8.Slet000l1000CEc1.0000QI,SOOOO.00000u.70000Z~C0000.5000088Time[IjDONALDC;COOKNUCLEARPLANTUNIT1FIGURE3.3-16NuclearPo~andCoreHeatFluxvs.TineforaTypicalResponsetoaDtoppedRCCA(s)inAutomatic Controlm:(1944.5w.wpf:1d~1195 3.3-82 550.0$10.0550.0I-530.0eo$10.0O490.0888tVf200.02100,0g2000.0ne1NN.Oge1000+0n1100o08Timefs]DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-17AverageCoohntTempertaure andPressurizer Pressurevs.Tinefora7ypicalResponsetoaDroppedRCCA(s)inAuromancControlmal944-5w.wpf:1d441195 3.3-83 1.41.2c~1.0c0c0.80.6Oi"0.400.20.004Time[s]10DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-18TotalCoreFlowvs.TimeforTheCompleteLossOfFlowEventm:i1944-5w.wpf:1d441195 3.3-84 fi0 1.4~1.2S~1.00C'=0.8~0.6.Q~0.4O~0.20.00Time[s)102,600EU'v)2,400CLC7g2.200lN2,000CL1,8000Time[s]10DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-19NuclearPowerandPressurizer Pressurevs.TimeforTheCompleteLossOfFlowEventm:11944.5w.wpt:

id~11953.3-85 1.6C5~1.2I.OC=0.8XG:~0.4ITAverageChannelHotChannel0.00Time[s]104.0'LzO3.02.01.004Time[s]'eatfluxesareshownasafractionofthenominalaveragechannelheatfluxDONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-20AverageandHotChannelHeatFluxesandDNBRvs.TimefortheCompleteLossOfFlowEventm:$1944-5w.wpf:1d441 1953.3-86 f'l 1.41.2CE1.O0cpg0.6OKp40.20.00Time[s]101.4~1.2Co1.0C0~o.se'.6~0.43D~0.20.00Time[s]10DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-21TotalCoreFlowandFaultedLoopFlowvs.TimeforThePartialLossOfFlowEventm:51944-5w.wpi:1d~1195 3.3-87 1.4~1.2~1.00C0'=0.8~0.6O~0.4~0.20.00Time(s]102,600C5~2,400CLIg2,200IN'C~2,000L,1,8000Time[s]10DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-22NuclearPoweramlPressurizer Pressurevs.TiaraforThePartialLossOfFlowEventm:$1944.5w.wpt:1d441195 3.3-88 1.6~1.2Er0=0.88tdI)CLL~0.4zAverageChannel0.00Time[s)104.03.02.01.004Time[s]'eatfluxesareshownasafractionofthenominalaveragechannelheatfluxDONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-23AverageandHotChannelHeatFluxesandDNBRvs.TimeforthePartialLossOfFlowEventmA1944-5w.wpf:1d~1195 3.3-89 1.41.2C51.O0C080.6u-O.4I00.20.00Time[s]101.41.2I1.O0.8CO0.60.4u.0.28o.o'De-0.2cl~4-0.60Time(s]10DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-24TotalCoreFlowandFaultedLoopFlowvs.TiaraForTheLockedRotorEventm:(1944.5w.wpt:1d441195 3.3-90

1.4a$~1.00LO'=0.8~0.6O~0.4z0.20.00Time[s]102,8002,600.5I2,400o-2,200CoOK2,0001,8000Timets]10DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-25NuclearPowerandRCSPressurevs.TimeForTheLockedRotorEventmf1944-5w.wpf:id%41195 3.3-91 1.6.I1.2HaClannelr0r.=0.8XLL~0.4rDAverageChannel0.00Time(s]103,000.02,500.00~~2,000.0~rnr-1,500.0O1,000.0500.00Time[s]10'eatfluxesareshownasafractionofthenominalaveragechannelheatfluxDONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-26AverageandHotChannelHeatFluxesvs.TineandCladInnerTemperature vs.TineForTheLockedRotorEventmA1944.5w.wpf:1d441195 3.3-92 1.4ClI+1.,0QCO'=0.8~0.6O~0.4z0.20.0020Timefs)60805.04.0P3.0A2.01.00204060Time[s]80DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-27NuciearPowerandDNBRvs.TiaraForLossofLoad,MinimumReactivity FeedbackWithPressurizer SprayandPORVsm:51944.5w.wpt:1d 4411953.3-93 2.8002,600Cto2;400g.2,200Ng2,000G.1,8001,600020Time[s]802,0001,800I~~1,600)Im1,400'~~1,200e-1,000800020Time[s]80DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-28Pressurizer PtessuteandPressurizer WaterVolundvs.Tun:ForLossofLoad,MinimumReactivity FeedbackWithPressurizer SprayandPORVsm:11944.5w.wpt:1d041195 3.3-94 uel650COCLEi-600I550OO500020Time[s]80100G.:650n.600E~Q~5508~ThOfToold500020Timefs]80100DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-29CoreAverageandLoop1Temperatures vs.Time.ForLossofLoad,MinimumReactivity FeedbackWithPressurizer SprayandPORVsm:11944.5w.wpt:1d~1195 3.3-95 1,0005ci-1,0000u-2.000CUK9-3,000~O-5,0000204060Time[s]8050~~30E~20NI10CL20Time[s]80DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-30TotalReactivity andPressurizer SteamReliefvs.TineForLossofLoad,MinimumReactivity FeedbackWithPressurizer SprayandPORVsm."tl944-5w.wpf:1d441 1953.3-96

1i0,000E100,000tO,90.000C)m80,000UEto70,00060.000020Time[s]80~~300Kg200@100(9V)0020lime(s)80DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-31SteamGenerator MassandSafetyValveReliefvs.Tirn:ForLossofLoad,MinimumReactivity FeedbackWithPressurizer SprayandPORVsm:11944-5w.wpf:1d~1195 3.3-97 1.4<1.2+1.0.OC'=0.8~0.6O~0.4z0.20.020Time(s]805.04.02.01.0020Time[s]80DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-32NuclearPowerandDNBRvs.TimeForLossofLoad,MaximumReactivity FeedbackWithPressurizer SprayandPORVsm:11944-5w.wpt:1d441 1953.3-98 2,8002,600m2;4002.200INg2,000Itt.1,8001,600020Time(s]802,0001,800I~~1,600Im1,400'~~1,200Pn-1,000800020Time[s]80DONALDC.COOKNUCIXmPLANTUNITIFIGURE3.3-33Pressurizer PressureandPressurizer WaterVolumevs.TimeForLossofLoad,Maxitrtum Reactivity FeedbackWithPressurizer SprayandPORVsm:t1944-5w.wpt:

1d4411953.3-99 u0o650f-600ro550OO500020Time[s]80<650Pa.600E~8~550yThotTcold500020Time[s]80DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-34CoreAverageandLoopITemperatures vs.TimeForLossofLoad,MaximumReactivity FeedbackWithPressurizer SprayandPORVsmal944.5w.wpf:1 d4411953.3-100 j

1.0005~-1;000)v2)000C$Cl9-3,000~O-5,0000204060Time[s]8050$40(t-30E~20n10O.0020Time[s)80DONALDC.COOK'UCLEARPLANTUNITIFIGURE3.3-35TotalReactivity andPressurizer SteamReliefvs.TineForLossofLoad,MaximumReactivity FeedbackWithPressurizer SprayandPORVsm:$1944.5w.wpf:ld441195 3.3-101

110,000E100,000~90,000m80,000QEto70,00060,000020Time[s]80400.0~300.0Ko200.0~100.0(9to0.0020Time[s]80100DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-36SteamGenerator MassandSafetyValveReliefvs.Tirade ForLossofLoad,MaximumReactivity FeedbackWithPressurizer SprayandPORVsm:51944-5w.wpf:1d~1195 3.3-102 tII 1.4co]2I~1.0-0.88~0.60~0.4~0.20.0020Time[s]805.04.0g3.0O2.01.0020Time[s]80DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-37NuclearPowerandDNBRvs.TimeForLossofLoad,MioimumReactivity FeedbackWithoutPtessurizer SprayandPORVsm:51944.5w.wpf:1d~1195 3.3-103

2,8002,600cao2;4Mg.2,200INg2,000Q.1,8001,600020Time[s]802,000-1,800~~1,600I~1,400I'~1,200PL1,000800020Time[s]80100DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-38Pressurizer PressureandPressurizer WaterVolumevs.TineForLossofLoad,MiainnunReactivity FeedbackWithoutPressurizer SprayandPORVsm:51944-5w.wpktd441 1953.3-104

uI650t-600o55000500020Time(s]6080100G.:650a.600E~ID~550Tcold500020Time[sj80DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-39CoreAverageandLoopITemperatures vs.TineForLossofLoad,MinimumReactivity FeedbackWithoutPressurizer SprayandPORVsmA19445w.wpt:1d441195 3.3-105

1,000Ec-1,000u-2.0006$tr.9-3,000~O-5,000020Time(s]8010050rr-30E~20N'CI10n0020Timefs]80DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3~TotalReactivity andPressurizer SteamReliefvs.TimeForLossofLoad,MinimumReactivity FeedbackWithoutPressurizer SprayandPORVsm:11944.5w.wpf:

1dC411953.3-106 0

110.000E100,000Cl90.000OS80,000QEto70,00060,000020Time[s]6080~300K>>200to100(9to20Time[s]80100DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3<1SteamGenerator MassandSafetyValveReliefvs.Tine,'orLossofLoad,MinimumReactivity FeedbackWithoutPressurizer SprayandPORVsm:$1944.5w.wpk 1d4411953.3-107

1.4g1.2100CO'=0.8~0.6O~0.4Oz0.20.0020Time[s]805.04.0~3.0O2.01.0020Time[s]80DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3Q2NuclearPowerandDNBRvs.TimeForLossofLoad,MaximumReactivity FeedbackWithoutPressurizer SprayandPORVsm:11944.5w.wpf:1d~1195 3.3-108

2,8002,600(0e2;4002,200INg2,00001,8001,600020Time[s]60802,000n-1,800I~~1,600Lm1,400I'~1,200PL1.000800020Time[s]80100DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3Q3Pressurizer PtessuteandPressurizer WaterVolumevs.Ttm.ForLossofLoad,MaximumReactivity FeedbackWithoutPressurizer SprayandPORVsm:51944.5w.wpf:

1d4411953.3-109

u0m650tOt-600I550OO500020Time(s]80t.650n.600E~Q~5503Tcold500020Time[s]80DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3~CoreAverageandLoopITemperatures vs.TimeForLossofLoad,MaximumReactivity FeedbackWithoutPressurizer SprayandPORVsm."t1944-5w.wpt:1d441195 3.3-110

1,000Eg-1,000o-2.000C5IDtr.m-3.000~O-5,000020Time[s]8050I~30E~20QhlN10Q.20Time[s]80DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3A5TotalReactivity andPressurizer SteamReliefvs.Tire.ForLossofLoad,MaximumReactivity FeedbackWithoutPressurizer SprayandPORVsmht944.5w.wpf:

1d~t1953.3-111

110,000E100,000~90,000580,000(9Ero70,00060,000020Time[s]6080~300Kg200@100(00020Time[s]80DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3MSteamGenerator MassandSaferyValveReliefvs.TineForLossofLoad,MaximumReactivity FeedbackWithoutPressurizer SprayandPORVsmal944.5w.wpf:1d~1195 3.3-112

1.0301.0201.010QQQo~~~~~~r0.990II0.980III240280320360400,440480520560CoreAverageTemperature

['F]DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3%7Variation ofReactivity WithCoreTemperature At1050psiaForTheEndOfLifeRoddedCoreWithOneControlRodAssemblyStuck(ZeroPovver)ForTheSteamline BreakDoubleEndedRuptureEventmht944-5w.wpf:id~1195 3.3-113 3.60Eu3.202.80X2.40+~o2.00o3"-1.60~o12000.800.400.000101520253035404550CorePower[PrecentofNominal]DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3A8DopplerPowerFeedbackForTheSteamline BreakDoubleEndedRuptureEventm:51944-5w.wpf:

1d~11953.3-114

2,4002,000~1,600g1,200Q.o800K////005101520253035404550ColdLegSafetyInjection tlbm/sec]

DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3<9SafetyIn~ecaonFlowSuppltedByOneChargtngPumpForTheSteamline BreakDoubleEndedRuptureEventm$1944.5w.wpf:1d44]

1953.3-115 0.4Eo030C00.2l0oI0.1z0.0050150Time[s]2500.4I0.3OCO0.2LLK0.1PO00.0050150Time[s]250DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-50NuclearPowerandCoreHeatFluxvs.TineForTheSteamline BreakDoubleEndedRuptureEventfInsideContainment WithPower]m:51944-5w.wpt:1d~1195 3.3-116

uo520CLEt-480I440OO400050100.150Time[s]2502,5002,000Iu)1.5000MOtt:1,000500050150'Time[s]250DONALDC.COOKNUCLEARPLANrUNIT1FIGURE3.3-51CoreAverageTemperature andRCSPressurevs.TimeForTheSteamline BreakDoubleEndedRuptureEvent[insideContainment WithPower]m:11944-5w.wpt:1d~1195 3.3-117

500~~400Le300I'~200n-10050150Time(s]250DONALDC.COOKNUCLEARE'LAN'I'NIT IFIGURE3.3-52Pressurizer WaterVolumevs.TimForTheSteamline BreakDoubleEndedRuptureEventPnsideContainment WithPower]mA1944.5w.wpf:1d~1195 3.3-118

5CL-500tt:-1,000-1,500050100150Time[s]25040I.30COCa20OO0~10POO50150Time[s]250DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.53Reactivity andCoreBoronConcentration vs.TimeForTheSteamline BreakDoubleEndedRuptureEventPnsideContainment WithPower]mh1944.5w.wpf:1d

~11953.3-119 1.0E+2CEo1.0E+10.80.0O0g1.0E-1Oz1.0E-202Time(s]DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-54NuclearPowervs.TimeForTheRodEjectionEvent,HotZeroPower,EndOfLifem%1944-5w.wpf:1d~1195 3.3-120 u3000t5i-2000D0D-1000LLFuelCenterline FuelAverageCladOuterSurfaceTime[s]10DONALDC.COOKNUCLEARPLANTUNITIFIGURE3.3-55FuelCenterline, FuelAverage,andCladOuterSurfaceTetnperature vs.TiaraForTkieRodEjectionEvent,HotZeroPower,EndOfLifem:51944-5w.wpi:

1d4411953.3-121 3.025Io2.0IG"->.5~+1.0~0.5z0.002Time[s]DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-56NuclearPowervs.TimeForTheRodEjectionEvent,HotFullPower,EndOfLifem%1944-5w.wpf:1d 4411953.3-122 u5000ca4000O.Ei-3000DO~2000u1000FuelCenteriinc FuelAverageCladOuterSurfaceTime[s)10DONALDC.COOKNUCLEARPLANTUNIT1FIGURE3.3-57FuelCenterline, FuelAverage,andCladOuterSurfaceTemperature vs.TiceForTheRodEjectionEvent,HotFullPower,EndOfLifemA1944<w.wpt:Ld441195 3.3-123 r3.4POSTLOCAHYDROGENPRODUCTION AspartoftheReratingProgram,Westinghouse providedhydrogengeneration ratesandinventories insidecontainment fromthevarioussources;including coreradiolysis, sumpradiolysis, corrosion-generated hydrogenandthezirconium/water reaction.

Acomparison ofthekeyparameters andassumptions employedintheseanalyseswerecomparedtothedataprovidedfortheSGTPProgram.Thiscomparison indicates thatthemaximumnormaloperation containment temperature of120'(maximum) remainsunchanged.

Sinceacontainment temperature of120'priortotheaccidentisthebasesforthecurrentanalysisofrecord,theReratingProgramresultsremainapplicable.

However,thepost-accident time-temperature profileinsidecontainment andthefractionofthecorethatundergoes azirc-water reactionresulting fromthisanalysishavebeenreviewedinordertoensurethatthevaluesemployedintheReratingProgramanalysisremainboundingfortheSGTPProgram.Thisreviewhasbeencompleted withthefollowing conclusions:

TheanalysisfortheReratingProgramconsidered zirconium-water reactions of1.5%,3.0%,and5.0%.ThelimitingPCTcalculations forthesmallbreakLOCAoccurring withinthepressurizer doghouseshowsthatonly0.128%ofthecorecladisoxidized.

Applyingthe10CFR50.44(d)(1) factorof5increaseresultsinazirc-water reactionpercentage of0.06%.Theanalysesremainhighlyconservative, sincethevalueislessthantheminimumvalueconsidered inthecalculations (i.e.,1.5%).Also,theupdatedvaluesdonotcontradict theUFSARstatement that"..zirconium-water reactioniscalculated tobeamaximumof0.1%byweightofthetotalquantityofzirconium inthecore."(UFSAR,page14.3.6-2datedJuly,1982)ForthelargebreakLOCA,thezirc-water reactionis4.93%.Sincethisvalueislessthanthevaluethatwasconsidered intheReratingProgram(i.e.,5%),theresultsfromtheanalysisremainapplicable.

2.Theimpactof,therevisedpost-LOCA temperatures onthepost-accident hydrogengeneration hasbeenreviewedandfoundtobenegligible.

3.Anotherconsideration isthecurrentinventory ofcorrodible materials insidecontainment.

AEPSCprovidedthechangestotheinventory ofcorrodible materials insidecontainment andconfirmed thatUFSARTables14.3.6-3and14.3.6-7remainvalid.Therefore, theReratingProgramanalysisremainsboundingfortheSGTPProgram.I'A1944<w.wpf:1d441 1953.4-1 16141210w20468ELEVATION (FT)Figure3.1-86HotRodPowerDistribution (3Inch,30%SGTP)ReducedTemperature, ReducedPressureDonaldC.CookUnit1m%1944<w.wpf:1d~1195 3.1-189 3.5CONTAINMENT ANALYSES3.5.1Short-Term Containment AnalysisTheshorttermcontainment integrity analysisisusedtoverifytheadequacyofinteriorstructures andwallsbydemonstrating thatcalculated differential pressures arelessthandesignlimits.Thefunctionality oftheicecondenser isdemonstrated andcontainment integrity isalsoverified.

Theeffortsperformed fortheshorttermcontainment

analysis, applicable tothePressurizer Enclosure, asdescribed inSection3.4.1ofWCAP-11902, supportoperation ofCookNuclearPlantUnits182overthefullrangeofreratedparameters.

ThemajorimpactsonLOCAshorttermmassandenergyreleaseratecalculations andcontainment subcompartment responseanalysis, aretheeffectsduetoRCStemperature changes.Forthesteamgenerator enclosure, massandenergyreleasesandthesubsequent containment responseareperformed atzeropower,whichmaximizes effectsbecausesteampressureismaximum.Allrelevantanalysesandevaluations performed fortheReratingProgramassumedvalueswhichwouldboundbothUnits182atthereratedpowerlevelsandrevisedtemperatures andpressures described inTable2.1-1ofWCAP-11902, Supplement 1.Theresultsoftheshorttermcontainment analysesandevaluations fortheSGTPProgramdemonstrate that,forthepressurizer enclosure, thefanaccumulator roomandthesteamgenerator enclosure, theresulting peakpressures remainbelowtheallowable designpeakpressures.

Sincethecalculated pressures inWCAP-11902, Supplement 1fortheloopcompartments exceededthedesignpressure, demonstration ofstructural adequacywasrequired.

Thisissuewasaddressed byAEPSCandisdocumented inUFSARSection14.3.4.2.3.4.

3.5.2Loss-of-Coolant MassandEnergyRelease3.5.2.1PurposeThepurposeofthisanalysiswastocalculate thelongtermLOCAmassandenergyreleaseswiththeproposedrevisedplantconditions andincreased operating margins.Theincreased operating marginsincludeincreased EDGstarttimeto30secondsandtherevisedRHRandHHSIpumpflowrates.Thissectionprovidestheanalytical basiswithrespecttotheLOCAcontainment massandenergyreleasefortheoperation oftheDonaldC.CookNuclearPlantUnits1and2attheSGTPProgramconditions.

Thiscontainment integrity analysesboundsbothunits.Ruptureofanyofthepipingcarryingpressurized hightemperature reactorcoolant,termedaLOCA,willresultinreleaseofsteamandwaterintothecontainment.

This,inturn,willresultinanincreaseinthecontainment pressureandtemperature.

Themassandenergyreleaseratesdescribed inthisdocumentformthebasisoffurthercomputations toevaluatethestructural integrity ofthecontainment following apostulated accidenttosatisfytheNuclearmAl9444w.wpf:1d441195 3,5-1 Regulatory acceptance

criteria, GeneralDesignCriterion 38,whichismorerestrictive thantheGDCcriteriainAppendixHoftheoriginalFSAR,towhichtheDonaldC.CookNuclearPlantsarelicensed.

Section3.5.3presentsthelongtermcontainment integrity analysisforcontainment pressurization evaluations.

3.5.2.2SystemCharacteristics andModelingAssumptions Themassandenergyreleaseanalysisissensitive totheassumedcharacteristics ofvariousplantsystems,inadditiontootherkeymodelingassumptions.

SomeofthemostcriticalitemsaretheRCSinitialconditions, coredecayheat,safetyinjection flow,andmetalandsteamgenerator heatreleasemodeling.

Specificassumptions concerning eachoftheseitemsarediscussed next.Tables3.5-1and3.5-2presentkeydataassumedintheanalysis.

Forthelongtermmassandenergyreleasecalculations, operating temperatures toboundthehighestaveragecoolanttemperature rangewereusedasboundinganalysisconditions.

Themodeledcoreratedpowerof3413MWtadjustedforcalorimetric error(+2percentofpower)wasthebasisintheanalysis.

Theuseofhighertemperatures isconservative becausetheinitialfluidenergyisbasedoncoolanttemperatures whichareatthemaximumlevelsattainedinsteadystateoperation.

Additionally, anallowance of+5.1'Fisreflected inthetemperatures inordertoaccountforinstrument erroranddeadband.

TheinitialRCSpressureinthisanalysisisbasedonanominalvalueof2250psia.Alsoincludedisanallowance of+67psi,whichaccountsforthemeasurement uncertainty onpressurizer pressure.

Theselection of2250psiaasthelimitingpressureisconsidered toaffecttheblowdownphaseresultsonly,sincethisrepresents theinitialpressureoftheRCS.TheRCSrapidlydepressurizes fromthisvalueuntilthepointatwhichitequilibrates withcontainment pressure.

TherateatwhichtheRCSblowsdownisinitially moresevereatthehigherRCSpressure.

Additionally theRCShasahigherfluiddensityatthehigherpressure(assuming aconstanttemperature) andsubsequently hasahigherRCSmassavailable forreleases.

Thus,2317psiainitialpressurewasselectedasthelimitingcaseforthelongtermmassandenergyreleasecalculations.

Theseassumptions conservatively maximizethemassandenergyintheRCS.Theselection ofthefueldesignfeaturesforthelongtermmassandenergycalculation isbasedontheneedtoconservatively maximizethecorestoredenergy.Themarginincorestoredenergywaschosentobe+15percent.Thus,theanalysisveryconservatively accountsforthestoredenergyinthecore.Thefuelconditions wereadjustedtoprovideaboundinganalysisforcurrentCookNuclearPlantUnits1and2fueldesignfeatures..The following itemsserveasthebasistoensureconservatism inthecorestoredenergycalculation:

timeofmaximumfueldensification andhighestBOLtemperatures.

MargininRCSvolumeof3%(whichiscomposedof1.6%allowance forthermalexpansion and1.4%foruncertainty) ismodeled.m319444w.wpt:1d~1195 3.5-2 Regarding safetyinjection flow,themassandenergycalculation considered thelimitingscenarioofminimumsafetyinjection flow,withtheRHRcrosstievalveisassumedtobeclosed,inconjunction witha15%pumpheaddegradation fortheRHRandSlpumpsand10%pumpheaddegradation forthechargingpumps.Thisconfiguration conservatively boundsotherrespective alignments.

ClosureoftheRHRcrosstiewasconsidered overtheHHSIcrosstiebecausethiswouldhaveamoresevereimpactontheanalysis(i.e.,closureoftheRHRcrosstiewouldboundclosureoftheHHSIcrosstie).

Thisresultsintheconservative minimumsafetyinjection flowrateused.Thefollowing assumptions wereemployedtoensurethatthemassandenergyreleasesareconservatively calculated, therebymaximizing energyreleasetocontainment:

Maximumexpectedoperating temperature oftheRCS(100%fullpowerconditions) 2.Anallowance intemperature forinstrument erroranddeadband(+5.1'F)13.Margininvolumeof3%(whichiscomposedof1.6%allowance forthermalexpansion, and1.4%foruncertainty) 4.Coreratedpowerof3413MWt5.Allowance forcalorimetric error(+2percentofpower)6.Conservative coefficient ofheattransfer(i.e.,steamgenerator primary/secondary heattransferandRCSmetalheattransfer) 7.Allowance incorestoreenergyforeffectoffueldensification 8.Amarginincorestoredenergy(+15percentincludedtoaccountformanufacturing tolerances) 9.Anallowance forRCSinitialpressureuncertainty

(+67psi)10.Steamgenerator tubepluggingleveling(0%uniform)~Maximizes reactorcoolantvolumeandfluidrelease~Maximizes heattransferareaacrosstheSGtubes~Reducescoolantloopresistance, whichreducesdelta-pupstreamofbreakandincreases breakflowmA1944+w.wpf:1d441195 3.5-3 Thus,basedontheaboveconditions andassumptions, aboundinganalysisofCookNuclearPlantUnits1and2ismadeforthereleaseofmassandenergyfromtheRCSintheeventofaLOCAtosupporttheSGTPProgram.3.5.2.3LongTermMassandEnergyReleaseAnalysis3.5.2.3.1 Introduction Theevaluation modelusedforthelongtermLOCAmassandenergyreleasecalculations wastheMarch1979modeldescribed inReference 1.Thisevaluation modelhasbeenreviewedandapprovedbytheNRC,andhasbeenusedintheanalysisofothericecondenser plants.ThisreportsectionpresentsthelongtermLOCAmassandenergyreleasesthatweregenerated insupportoftheSGTPProgram.Thesemassandenergyreleasesarethensubsequently usedintheLOTIC-1containment integrity analysispeakpressurecalculation.

3.5.2.3.2 LOCAMassandEnergyReleasePhasesThecontainment systemreceivesmassandenergyreleasesfollowing apostulated ruptureintheRCS.Thesereleasescontinueoveratimeperiod,which,fortheLOCAmassandenergyanalysis, istypically dividedintofourphases:Blowdown-theperiodoftimefromaccidentinitiation (whenthereactorisatsteadystateoperation) tothetimethattheRCSandcontainment reachanequilibrium stateatcontainment designpressure.

2.Refill-theperiodoftimewhenthelowerplenumisbeingfilledbyaccumulator andECCSwater.Attheendofblowdown, alargeamountofwaterremainsinthecoldlegs,downcomer, andlowerplenum.Toconservatively considertherefillperiodforthepurposeofcontainment massandenergyreleases, itisassumedthatthiswaterisinstantaneously transferred tothelowerplenumalongwithsufficient accumulator watertocompletely fillthelowerplenum.Thisallowsanuninterrupted releaseofmassandenergytocontainment.

Thus,therefillperiodisconservatively neglected inthemassandenergyreleasecalculation.

3.Reflood-beginswhenthewaterfromthelowerplenumentersthecoreandendswhenthecoreiscompletely quenched.

4.Post-reflood(Froth)-describes theperiod.following therefloodtransient.

Forthepumpsuctionbreak,atwo-phase mixtureexitsthecore,passesthroughthehotlegs,andissuperheated inthesteamgenerators.

Afterthebrokenloopsteamgenerator cools,thebreakflowbecomestwophase.m:41944+w.wpf:1d441195 3.5-4 3.5.2.3.3 ComputerCodesTheReference 1massandenergyreleaseevaluation modeliscomprised ofmassandenergyreleaseversionsofthefollowing codes:SATANVl,WREFLOOD, andFROTH.Thesecodeswereusedtocalculate thelongtermLOCAmassandenergyreleasesforCookNuclearPlantUnits1and2.SATANcalculates

blowdown, thefirstportionofthethermal-hydraulic transient following breakinitiation, including
pressure, enthalpy, density,massandenergyflowrates, andenergytransferbetweenprimaryandsecondary systemsasafunctionoftime.TheWREFLOODcodeaddresses theportionoftheLOCAtransient wherethecorerefloodingphaseoccursafterthepnmarycoolantsystemhasdepressurized (blowdown) duetothelossofwaterthroughthebreakandwhenwatersuppliedbytheEmergency CoreCoolingrefillsthereactorvesselandprovidescoolingtothecore.Themostimportant featureisthesteam/water mixingmodel(SeeSection3.5.2.6.2).

FROTHmodelsthepost-refloodportionofthetransient.

TheFROTHcodeisusedforthesteamgenerator heatadditioncalculation fromthebrokenandintactloopsteamgenerators.

3.5.2.4BreakSizeandLocationGenericstudieshavebeenperformed withrespecttotheeffectofpostulated breaksizeontheLOCAmassandenergyreleases.

Thedoubleendedguillotine breakhasbeenfoundtobelimitingduetolargermassflowratesduringtheblowdownphaseofthetransient.

Duringtherefloodandfrothphases,thebreaksizehaslittleeffectonthereleases.

Threedistinctlocations intheRCSloopcanbepostulated forpiperupture:1.Hotleg(betweenvesselandsteamgenerator) 2.Coldleg(betweenpumpandvessel)3.Pumpsuction(betweensteamgenerator andpump)ThebreaklocationanalyzedfortheSGTPProgramisthepumpsuctiondoubleendedrupture,DEPS(10.48ft').Breakmassandenergyreleaseshavebeencalculated fortheblowdown, reflood,andpost-reflood phasesoftheLOCAforeachcaseanalyzed.

Thefollowing information providesadiscussion oneachbreaklocation.

Thehotlegdoubleendedrupturehasbeenshowninpreviousstudiestoresultinthehighestblowdownmassandenergyreleaserates.Althoughthe.ce'efloodingratewouldbethehighestforthisbreaklocation, theamountofenergyreleasedfromthesteamgenerator secondary isminimalbecausethemajorityofthefluidwhichexitsthecorebypassesthesteamgenerators ventingdirectlytocontainment.

Asaresult,therefloodmassandenergym:51944+w.wpf:ad~1195 3.5-5 releasesarereducedsignificantly ascomparedtoeitherthepumpsuctionorcoldlegbreaklocations wherethecoreexitmixturemustpassthroughthesteamgenerators beforeventingthroughthebreak.Forthehotlegbreak,genericstudieshaveconfirmed thatthereisnorefloodpeak(i.e.,fromtheendoftheblowdownperiodthecontainment pressurewouldcontinually decrease).

Themassandenergyreleasesforthehotlegbreakhavenotbeenincludedinthescopeofthiscontainment integrity analysisbecauseforthehotlegbreakonlytheblowdownphaseofthetransient isofanysignificance.

Sincetherearenorefloodandpost-refloodphasestoconsider, thelimitingpeakpressurecalculated wouldbethecompression peakpressureandnotthepeakpressurefollowing icebedmeltout.Thecoldlegbreaklocationhasalsobeenfoundinpreviousstudiestobemuchlesslimitingintermsoftheoverallcontainment energyreleases.

Thecoldlegblowdownisfasterthanthatofthepumpsuctionbreak,andmoremassisreleasedintothecontainment.

However,thecoreheattransferisgreatlyreduced,andthisresultsinaconsiderably lowerenergyreleaseintocontainment.

Studieshavedetermined thattheblowdowntransient forthecoldlegis,ingeneral,lesslimitingthanthatforthepumpsuctionbreak.Duringreflood,thefloodingrateisgreatlyreducedandtheenergyreleaserateintothecontainment isreduced.Therefore, thecoldlegbreakisnotincludedinthescopeoftheSGTPProgram.Thepumpsuctionbreakcombinestheeffectsoftherelatively highcorefloodingrate,asinthehotlegbreak,andtheadditionofthestoredenergyinthesteamgenerator.

Asaresult,thepumpsuctionbreakyieldsthehighestenergyflowratesduringthepost-blowdown periodbyincluding alloftheavailable energyoftheRCSincalculating thereleasestocontainment.

Thisbreaklocationhasbeendetermined tobethelimitingbreakforallicecondenser plants.Insummary,theanalysisofthelimitingbreaklocationforanicecondenser containment hasbeenperformed andisshowninthisreport.Thedouble-ended pumpsuction(DEPS)guillotine breakhashistorically beenconsidered tobethelimitingbreaklocation, byvirtueofitsconsideration ofallenergysourcespresentintheRCS.Thisbreaklocationprovidesamechanism forthereleaseoftheavailable energyintheRCS,including boththebrokenandintactloopsteamgenerators.

3.5.2.5Application ofSingleFailureCriteriaAnanalysisoftheeffectsofthesinglefailurecriteriahasbeenperformed onthemassandenergyreleaseratesfortheDEPSbreak.Aninherentassumption inthegeneration ofthemassandenergyreleaseisthatoffsitepowerislost.Thisresultsintheactuation oftheemergency dieselgenerator, requiredtopowerthesafetyinjection system.Thisisnotanissuefortheblowdownperiodwhichislimitedbythecompression peakpressure.

Thelimitingminimumsafetyinjection casehasbeenanalyzedfortheeffectsofasinglefailure.Inthecaseofminimumsafeguards, thesinglefailurepostulated tooccuristhelossofanemergency dieselgenerator.

Thisresultsinthelossofonepumpedsafetyinjection train,mA19444w.wpf:1d441195 3.5-6 therebyminimizing thesafetyinjection flow.Anadditional conservatism hasbeenincludedinthisanalysisinthattheclosureoftheRHRcrosstievalvehasbeenconsidered becauseitresultsinafurtherreduction insafety.injection flow.Theanalysisfurtherconsiders theRHRandSlpumpheadcurvestobedegradedby15%andthechargingpumpheadcurvetobedegradedby10%.ThisresultsinthegreatestSlflowreduction fortheminimumsafeguards case.3.5.2.6MassandEnergyReleaseData3.5.2.6.1 BlowdownMassandEnergyReleaseDataAversionoftheSATAN-Vlcodeisusedforcomputing theblowdowntransient, whichisthecodeusedfortheECCScalculation inReference 2.Thecodeutilizesthecontrolvolume(element) approachwiththecapability formodelingalargevarietyofthermalfluidsystemconfigurations.

Thefluidproperties areconsidered uniformandthermodynamic equilibrium isassumedineachelement.Apointkineticsmodelisusedwithweightedfeedbackeffects.Themajorfeedbackeffectsincludemoderator density,moderator temperature, andDopplerbroadening.

Acriticalflowcalculation forsubcooled (modified 2aloudek),

two-phase (Moody),orsuperheated breakflowisincorporated intotheanalysis.

Themethodology fortheuseofthismodelisdescribed inReference 1.Table3.5-3presentsthecalculated massandenergyreleasesfortheblowdownphaseoftheDEPSbreak.Forthepumpsuctionbreaks,breakpath1inthemassandenergyreleasetablesreferstothemassandenergyexitingfromthesteamgenerator sideofthebreak;and,breakpath2referstothemassandenergyexitingfromthepumpsideofthebreak.3.5.2.6.2 RefloodMassandEnergyReleaseDataTheWREFLOODcodeusedforcomputing therefloodtransient, isamodifiedversionofthatusedinthe1981ECCSevaluation model(Reference 2).TheWREFLOODcodeconsistsoftwobasichydraulic models-oneforthecontentsofthereactorvessel,andoneforthecoolantloops.Thetwomodelsarecoupledthroughtheinterchange oftheboundaryconditions appliedatthevesseloutletnozzlesandatthetopofthedowncomer.

Additional transient phenomena suchaspumpedsafetyinjection andaccumulators, reactorcoolantpumpperformance, andsteamgenerator releaseareincludedasauxiliary equations whichinteractwiththebasicmodelsasrequired.

TheWREFLOODcodepermitsthecapability tocalculate variations duringthecorerefloodingtransient ofbasicparameters suchascorefloodingrate,coreanddowncomer waterlevels,fluidthermodynamic conditions (pressure,

enthalpy, density)throughout theprimarysystem,andmassflowratesthroughtheprimarysystem.Thecodepermitshydraulic modelingofthetwoflowpathsmal944+w.wpf:1d441195 35-7 available fordischarging steamandentrained waterfromthecoretothebreak;i.e.thepaththroughthebrokenloopandthepaththroughtheunbrokenloops.Acompletethermalequilibrium mixingcondition forthesteamandECCSinjection waterduringtherefloodphasehasbeenassumedforeachloopreceiving ECCSwater.Thisisconsistent withtheusageandapplication oftheReference 1massandenergyreleaseevaluation modelintheReratingProgramanalyses.

EventhoughtheReference 1modelcreditssteam/mixing onlyintheintactloopandnotinthebrokenloop,justification, applicability, andNRCapprovalforusingthemixingmodelinthebrokenloophasbeendocumented (Reference 3).Thisassumption isjustified andsupported bytestdata,andissummarized asfollows:Themodelassumesacompletemixingcondition (i.e.,thermalequilibrium) forthesteam/water interaction.

Thecompletemixingprocess,however,ismadeupoftwodistinctphysicalprocesses.

Thefirstisatwophaseinteraction withcondensation ofsteambycoldECCSwater.Thesecondisasinglephasemixingofcondensate andECCSwater.Sincethesteamreleaseisthemostimportant influence tothecontainment pressuretransient, thesteamcondensation partofthemixingprocessistheonlypartthatneedbeconsidered.

(Anyspillagedirectlyheatsonlythesump.)Themostapplicable steam/water mixingtestdatahasbeenreviewedforvalidation ofthecontainment integrity refloodsteam/water mixingmodel.Thisdataisthatgenerated in1/3scaletests(Reference 4),whicharethelargestscaledataavailable andthusmostcleadysimulates theflowregimesandgravitational effectsthatwouldoccurinaPWR.Thesetestsweredesignedspecifically tostudythesteam/water interaction forPWRrefloodconditions.

Fromtheentireseriesof1/3scaletests,agroupcorresponds almostdirectlytocontainment integrity refloodconditions.

Theinjection flowrates forthisgroupcoverallphasesandmixingconditions calculated duringtherefloodtransient.

Thedatafrom'these testswerereviewedanddiscussed indetailinReference 1.Forallofthesetests,thedataclearlyindicatetheoccurrence ofveryeffective mixingwithrapidsteamcondensation.

Themixingmodelusedinthecontainment integrity refloodcalculation istherefore whollysupported bythe1/3scalesteam/water mixingdata.Additionally, thefollowing justification isalsonoted.Thepost-blowdown limitingbreakforthecontainment integrity peakpressureanalysisisthepumpsuctiondoubleendedrupturebreak.Forthisbreak,therearetwoflowpaths available intheRCSbywhichmassandenergymaybereleasedtocontainment.

Oneisthroughtheoutletofthesteamgenerator, theotherviareverseflowthroughthereactorcoolantpump.Steamwhichisnotcondensed byECCSinjection intheintactRCSloopspassesaroundthedowncomer andthroughthebrokenloopcoldlegandpumpinventingtocontainment.

Thissteamalsoencounters ECCSinjection waterasitpassesthroughthebrokenloopcoldleg,completemixingoccursandaportionofitiscondensed.

Itisthisportionofsteamwhichiscondensed thatistakencreditforinthismA19444w.wpt:1d 4411953.5-8 analysis.

Thisassumption isjustified baseduponthepostulated breaklocation, andtheactualphysicalpresenceoftheECCSinjection nozzle.Adescription ofthetestandtestresultsiscontained inReferences 1and4.Table3.5-4presentsthecalculated massandenergyreleasefortherefloodphaseofthepumpsuctiondoubleendedrupturewithminimumsafetyinjection.

Thetransients oftheprincipal parameters duringrefloodareprovidedinTable3.5-5.3.5.2.6.3 Post-RefloodMassandEnergyReleaseDataTheFROTHcode(Reference 5)isusedforcomputing thepost-refloodtransient.

TheFROTHcodecalculates theheatreleaseratesresulting fromatwo-phase mixturelevelpresentinthesteamgenerator tubes.Themassandenergyreleasesthatoccurduringthisphasearetypically superheated duetothedepressurization andequilibration ofthebrokenloopandintactloopsteamgenerators.

Duringthisphaseofthetransient, theRCShasequilibrated withthecontainment

pressure, butthesteamgenerators containasecondary inventory atanenthalpythatismuchhigherthantheprimaryside.Therefore, thereisasignificant amountofreverseheattransferthatoccurs.Steamisproducedinthecoreduetocoredecayheat.Forapumpsuctionbreak,atwophasefluidexitsthecore,flowsthroughthehotlegsandbecomessuperheated asitpassesthroughthesteamgenerator.

Oncethebrokenloopcools,thebreakflowbecomestwophase.Themethodology fortheuseofthismodelisdescribed inReference 1.Aftercontainment depressurization, themassandenergyreleaseavailable tocontainment isgenerated directlyfromcoreboiloff/decay heat.Table3.5-6presentsthetwophasepost-reflood(froth)massandenergyreleasedataforthepumpsuctiondoubleendedcase.3.5.2.7DecayHeatOnNovember2,1978theNuclearPowerPlantStandards Committee (NUPPSCO) oftheAmericanNuclearSociety(ANS)approvedANSstandard5.1forthedetermination ofdecayheat.Thisstandardwasusedinthemassandenergyreleasemodelwiththefollowing input:Significant assumptions inthegeneration ofthedecayheatcuwe:Decayheatsourcesconsidered arefissionproductdecayandheavyelementdecayofU-239andN,-239.mal9444w.wpl:1d

~11953.5-9 2.Decayheatpowerfromfissioning isotopesotherthanU-235isassumedtobeidentical tothatofU-235.3.Fissionrateisconstantovertheoperating historyofmaximumpowerlevel.4.Thefactoraccounting forneutroncaptureinfissionproductshasbeentakenfromTable10ofReference 6.5.Operation timebeforeshutdownis3years.6.Thetotalrecoverable energyassociated withonefissionhasbeenassumedtobe200MeV/fission.

7.Twosigmauncertainty (twotimesthestandarddeviation) hasbeenappliedtothefissionproductdecay.3.5.2.8SteamGenerator Equilibration andDepressurization Steamgenerator equilibration anddepressurization istheprocessbywhichsecondary sideenergyisremovedfromthesteamgenerators instages.TheFROTHcomputercodecalculates theheatremovalfromthesecondary massuntilthesecondary temperature isthesaturation temperature (T,)atthecontainment designpressure.

AftertheFROTHcalculations, steamgenerator secondary energyisremovedbasedonfirstandsecondstagerates.Thefirststagerateisapplieduntilthesteamgenerator reachesT,attheuserspecified intermediate equilibration

pressure, whenthesecondary pressureisassumedtoreachtheactualcontainment pressure.

Then,thesecondstagerateisuseduntilthefinaldepressurization.

Theheatremovalofthebrokenloopandintactloopsteamgenerators arecalculated separately.

DuringtheFROTHcalculations, steamgenerator heatremovalratesarecalculated usingthesecondary sidetemperature, primarysidetemperature andasecondary sideheattransfercoefficient determined usingamodifiedMcAdam'scorrelation (Reference 7).Steamgenerator energyisremovedduringtheFROTHtransient untilthesecondary sidetemperature reachessaturation temperature atthecontainment designpressure.

Theconstantheatremovalrateusedduringthefirstheatremovalstageisbasedonthefinalheatremovalratecalculated byFROTH.TheSGenergyavailable tobereleasedduringthefirststageintervalisdetermined bycalculating thedifference insecondary energyavailable atthecontainment designpressureandthatatthe(lower)userspecified intermediate equilibration

pressure, assumingsaturated conditions.

Thisenergyisthendividedbythefirststageenergyremovalrate,resulting inanintermediate equilibration time.Atthistime,therateofenergyreleasedropssubstantially tothesecondstagerate.Thesecondstagerateisdetermined asthefractionofthedifference insecondary energyavailable betweentheintermediate equilibration andfinaldepressunzation.

mA19444w.wpi:1 d~11953.5-10 3.5.2.9SourcesofMassandEnergyThesourcesofmassconsidered intheLOCAmassandenergyreleaseanalysisaregiveninTable3.5-7.ThesesourcesaretheRCS,accumulators, andpumpedsafetyinjection.

Theenergyinventories considered intheLOCAmassandenergyreleaseanalysisaregiveninTable3.5-8.Theenergysourcesinclude:1.RCSWater2.Accumulator WaterPumpedInjection WaterDecayHeatCoreStoredEnergyRCSMetal-PrimaryMetal(includes SGtubes)7.SteamGenerator Metal(includes transition cone,shell,wrapper,andotherinternals) 8.SteamGenerator Secondary Energy(includes fluidmassandsteammass)Secondary TransferofEnergy(feedwater intoandsteamoutofthesteamgenerator secondary)

EnergyReference Points:Available Energy:212'F;14.7psiaTotalEnergyContent:32'F;14.7psiaItshouldbenotedthattheinconsistency intheenergybalancetablesfromtheendofRefloodto3600seconds,i.e.,"TotalAvailable" dataversus"TotalAccountable",

resultedfromtheomissionofthereactorupperheadintheanalysisfollowing blowdown.

Ithasbeenconcluded thattheresultsaremoreconservative whentheupperheadisneglected.

Thisdoesnotaffecttheinstantaneous massandenergyreleases, ortheintegrated values,butcausesanincreaseinthetotalaccountable energywithintheenergybalancetable.mA$944<w.wpt:1d441195 3.5-11

Themassandenergyinventories arepresented atthefollowing times,asappropriate:

1.2.3.4.5.6.Timezero(initialconditions)

EndofblowdowntimeEndofrefilltimeEndofrefloodtimeTimeofbrokenloopsteamgenerator equilibration topressuresetpointTimeofintactloopsteamgenerator equilibration topressuresetpointInthemassandenergyreleasedatapresented, noZirc-water reactionheatwasconsidered becausethecladtemperature didnotrisehighenoughfortherateoftheZirc-water reactionheattobeofanysignificance.

Theconsideration ofthevariousenergysourcesinthemassandenergyreleaseanalysisprovidesassurance thatallavailable sourcesofenergyhavebeenincludedinthisanalysis.

AlthoughCookNuclearPlantUnit1isnotaStandardReviewPlanPlant,thereviewguidelines presented inStandardReviewPlanSection6.2.1.3havebeensatisfied.

3.5.2.10References "Westinghouse LOCAMassandEnergyReleaseModelforContainment Design-March1979Version",

WCAP-10325-P-A, May1983(Proprietary),

WCAP-10326-A (Non-Propnetary).

2."Westinghouse ECCSEvaluation Model-1981Version",

WCAP-9220-P-A, Rev.1,February1982(Proprietary),

WCAP-9221-A, Rev.1(Non-Proprietary) 3.DocketNo.50-315,"Amendment No.126,FacilityOperating LicenseNo.DPR-58(TACNo.7106),forD.C.CookNuclearPlantUnit1",June9,1989.4.EPRI294-2,MixingofEmergency CoreCoolingWaterwithSteam;1/3ScaleTestandSummary,(WCAP-8423),

FinalReportJune1975.5."Westinghouse MassandEnergyReleaseDataForContainment Design",WCAP-8264-P-A, Rev.1,August1975(Proprietary),

WCAP-8312-A (Non-Proprietary).

6.ANSI/ANS-5.1 1979,AmericanNationalStandardforDecayHeatPowerinLightWaterReactors",

August1979.7.W.H.McAdam,HeatTransmission, McGraw-Hill 3rdedition,1954,p.172.mal944+w.wpf:1d441195 3.5-12 3.5.3LOCAContainment Integrity Analysis3.5.3.1Description ofLOTIC-1ModelEarlyintheicecondenser development program,itwasrecognized thattherewasaneedformodelingoflongtermicecondenser performance.

Itwasrealizedthatthemodelwouldhavetohavecapabilities comparable tothoseofthedrycontainment (COCO)model.Thesecapabilities wouldpermitthemodeltobeusedtosolveproblemsofcontainment designandoptimizethecontainment andsafeguards systems.Thishasbeenaccomplished inthedevelopment oftheLOTICcode,described inReference 1.Themodelofthecontainment consistsoffivedistinctcontrolvolumes,theuppercompartment, thelowercompartment, theportionoftheicebedfromwhichtheicehasmelted,theportionoftheicebedcontaining unmeltedice,andthedeadendedcompartment.

Theicecondenser controlvolumewithunmeltedandmeltediceisfurthersubdivided intosixsubcompartments toallowformaldistribution ofbreakflowtotheicebed.Theconditions inthesecompartments areobtainedasafunctionoftimebytheuseoffundamental equations solvedthroughnumerical techniques.

Theseequations aresolvedforthreephasesintime.Eachphasecorresponds toadistinctphysicalcharacteristic oftheproblem.Eachofthesephaseshasauniquesetofsimplifying assumptions basedontestresultsfromtheicecondenser testfacility.

Thesephasesaretheblowdownperiod,thedepressurization period,andthelongterm.Themostsignificant simplification oftheproblemistheassumption thatthetotalpressureinthecontainment isuniform.Thisassumption isjustified bythefactthataftertheinitialblowdownoftheRCS,theremaining massandenergyreleasedfromthissystemintothecontainment aresmallandveryslowlychanging.

Theresulting flowratesbetweenthecontrolvolumeswillalsoberelatively small.Theseflowratesthenareunabletomaintainsignificant pressuredifferences betweenthecompartments.

Inthecontrolvolumes,whicharealwaysassumedtobesaturated, steamandairareassumedtobeuniformly mixedandatthecontrolvolumetemperature.

Theairisconsidered aperfectgas,andthethermodynamic properties ofsteamaretakenfromtheASMEsteamtable.Thecondensation ofsteamisassumedtotakeplaceinacondensing nodelocated,forthepurposeofcalculation, betweenthetwocontrolvolumesintheicestoragecompartment.

Theexittemperature oftheairleavingthisnodeissetequaltoaspecificvaluewhichisequaltothetemperature oftheicefilledcontrolvolumeoftheicestoragecompartment.

Lowercompartment exittemperature isusediftheicebedsectionismelted.m:119444w.wpf:1d441195 3.5-13 l

3.5.3.2Containment PressureCalculation Thefollowing arethemajorinputassumptions usedintheLOTICanalysisforthepumpsuctionpiperupturecasewiththesteamgenerators considered asanactiveheatsourcefortheCookNuclearPlantContainment:

Minimumsafeguards areemployedinallcalculations, e.g.,oneoftwospraypumpsandoneoftwosprayheatexchangers; oneoftwoRHRpumpsandoneoftwoRHRheatexchangers providing flowtothecore;oneoftwosafetyinjection pumpsandoneoftwocentrifugal chargingpumps;andoneoftwoairreturnfans.2.2.11x10'bs.oficeinitially intheicecondenser.

3.Theblowdown, reflood,andpostrefloodmassandenergyreleasesdescribed inSection3.5.2areused.4.Blowdownandpost-blowdown icecondenser draintemperatures of190'Fand130'Fareused,respectively.

5.Nitrogenfromtheaccumulators intheamountof4510lbs.isincludedinthecalculations.

6.Essential setvicewatertemperature of87.5'Fisusedonthesprayheatexchanger andthecomponent coolingheatexchanger, 7.Theairreturnfaniseffective 10minutesafterthetransient isinitiated.

8.Nomaldistribution ofsteamflowtotheicebedisassumed.(Thisassumption isconservative sinceitcontributes toearlyicebedmeltouttime.)9.Noicecondenser bypassisassumed.(Thisassumption depletestheiceintheshortesttimeandisthusconservative.)

10.Theinitialconditions inthecontainment areatemperature of57'Fintheuppercompartment volume,and60'Fintheloweranddead-ended compartment volumes.Allvolumesareatapressureof0.3psig.11.Containment structural heatsinksareassumedwithconservatively lowheattransferrates.(SeeTables3.5-11and3.5-12)mA1944+w.wpf:1d%51895 3.5-14 12.Theoperation ofonecontainment sprayheatexchanger (UA=3.107x10'tu/hr-'F),

forcontainment coolingandtheoperation ofoneRHRheatexchanger.

(UA=2.22x10'tu/hr-'F) forcorecooling.Thecomponent coolingheatexchanger wasmodeledat3.58x10'tu/hr-'F.

13.Theairreturnfanreturnsairatarateof39,000cfmfromtheuppertothelowercompartment.

14.Anactivesumpvolumeof40,600ft'sused.15.102%of3413MWtpowerisusedinthecalculations.

16.Subcooling ofECCSwaterfromtheRHRheatexchanger isassumed.17.Essential servicewaterflowtothecontainment sprayheatexchanger wasmodeledas2000gpm.Alsothenuclearservicewaterflowtothecomponent coolingheatexchanger wasmodeledas5000gpm.18.RHRSprayinitiation isassumedafterswitchover frominjection torecirculation hasbeencompleted andcontainment pressureisgreaterthanorequalto8sl3.5.3.3Structural HeatRemovalProvision ismadeinthecontainment pressureanalysisforheatstorageininteriorandexteriorwalls.Eachwallisdividedintoanumberofnodes.Foreachnode,aconservation ofenergyequationexpressed infinitedifference formsaccountsfortransient conduction intoandoutofthecontainment structural heatsinksusedintheanalysis.

ThematerialpropertydatausedisfoundinTables3.5-11and3.5-12.Theheattransfercoefficient tothecontainment structure isbasedprimarily ontheworkofTagami(Reference 2).WhenapplyingtheTagamicorrelations, aconservative limitwas'lacedonthelowercompartment stagnantheattransfercoefficients.

Theywerelimitedtoasteam-air ratioof1.4according totheTagamicorrelation.

Theimposition ofthislimitation istorestricttheuseoftheTagamicorrelation withinthetestrangeofsteam-air ratioswherethecorrelation wasderived.Withtheseassumptions, theheatremovalcapability ofthecontainment issufficient toabsorbtheenergyreleasesandstillkeepthemaximumcalculated pressurebelowthedesignpressure.'al 944<w.wpf:1d441195 3.5-15 3.5.3.4AnalysisResultsTheresultsoftheanalysisshowsthatthemaximumcalculated containment pressureis11.49psigfortheDEPSminimumsafeguards breakcase.Thispressurepeakoccursatapproximately 7752seconds,withicebedmeltoutatapproximately 5423seconds.Thefollowing plotsshowthecontainment integrity transient, ascalculated bytheLOTIC-1code.Figure3.5-1:Containment PressureTransient Figure3.5-2:UpperCompartment Temperature Transient Figure3.5-3:LowerCompartment Temperature Transient Figure3.5-4:ActiveandInactiveSumpTemperature Transient Figure3.5-5:IceMeltTransient Tables3.5-9and3.5-10giveenergyaccountings atvariouspointsinthetransient.

3.5.3.5RelevantAcceptance CriteriaTheLOCAmassandenergyanalysishasbeenperformed inaccordance withthecriteriashownintheStandardReviewPlan(SRP)Section6,2.1.3.Inthisanalysis, therelevantrequirements ofGeneralDesignCriteria(GDC)50and10CFRPart50AppendixKhavebeenincludedbyconfirmation thatthecalculated pressureislessthanthedesignpressure, andbecauseallavailable sourcesofenergyhavebeenincluded, whichismorerestrictive thantheGDCcriteriainAppendixHoftheoriginalFSAR,towhichtheDonaldC.CookNuclearPlantsarelicensed.

Thesesourcesincludereactorpower,decayheat,corestoredenergy,energystoredinthereactorvesselandinternals, metal-water reactionenergy,andstoredenergyinthesecondary system.Thecontainment integrity peakpressureanalysishasbeenperformed inaccordance withthecriteriashownintheSRPSection6.2.1.1.b foricecondenser containments.

Conformance toGDC's16,38,and50isdemonstrated byshowingthatthecontainment designpressureisnotexceededatanytimeinthetransient.

Thisanalysisalsodemonstrates thatthecontainment heatremovalsystemsfunctiontorapidlyreducethecontainment pressureandtemperature intheeventofaLOCA.3.5.3.6Conclusions

/Basedupontheinformation presented, itmaybeconcluded thatoperation withtherevisedplantconditions andincreased operating marginsforDonaldC.CookNuclearPlantisacceptable.

Operation withtheRHRcrosstievalveclosedwasalsoshowntobemorelimitingthanoperation withthevalveopensincethereislesssafetyinjection wateravailable forsteamcondensation.

Operation withtherevisedplantconditions, increased operating marginsmA1944+w.wpf:1d~1195 3.5-16 andtheRHRcrosstievalveclosedresultsinacalculated peakcontainment pressureof11.49psig,ascomparedtothedesignpressureof12.0psig.Thus,themostlimitingcasehasbeenconsidered, andhasbeendemonstrated toyieldacceptable results.3.5.3.7References 1."LongTermIceCondenser Containment Code-LOTICCode",WCAP-8354-P-A, April1976(Proprietary),

WCAP-8355-A (Non-Proprietary).

2.Tagami,Takasi,InterimReportonSafetyAssessments andFacilities Establishment ProjectinJapanforPeriodEndingJune,1965(No.1).3.5.4Steamline BreakMass/Energy ReleasesInsideContainment Themass/energy releasesfortheinsidecontainment analysisofrecordatthetimethe30%SGTPProgramwasunderwayisbasedontheReratingProgram,whichwasperformed toboundbothunits.Thecalculation ofthemass/energy releasesfollowing asteamline breakisdescribed intheCookNuclearPlantUnit1UFSARSection14.3.4.4.

Steamline rupturesoccurring insideareactorcontainment structure mayresultinsignificant releasesofhighenergyfluidtothecontainment environment, possiblyresulting inhighcontainment temperatures andpressures.

Thequantitative natureofthereleasesfollowing asteamline ruptureisdependent uponthemanypossibleconfigurations oftheplantsteamsystemandcontainment designsaswellastheplantoperating conditions andthesizeoftherupture.Thesevariations makeitdifficult toreasonably determine thesingle"worstcase"forbothcontainment pressureandtemperature evaluations following asteambreak.

Theanalysisperformed aspartoftheReratingProgramdetermined thatthelimitingscenarioofthesteambreak casesanalyzedforthecontainment responseevaluation wereabreaksizeof0.86ft'ccurring atfullpowerforthesplitrupturescenarioandabreaksizeof4.6ft'ccurring atfullpowerforthedouble-ended rupturescenario.

Aspartofthe30%SGTPProgram,severalofthelimitingcasesofthesteamline breakmassandenergyreleasecalculations insidecontainment havebeenreanalyzed toassessalongerfeedwater motoroperated(FMO)valvestroketime,largerunisolatable feedlineandsteamline volumes,andrevisedmaximumAFWflowrates.Arelaxation intheEDGstart-uptimefrom10to30seconds,andanincreaseintheuppercontainment andlowercompartment spraydelaytimearealsoaddressed.

However,theselatteranalysisassumption changesonlyaffectthecontainment responseanalysis.

Itshouldbenotedthatthechangesassociated onlywiththeSGTPProgramforUnit1,i.e.,RCSflowreduction, reducedprimary-to-secondary heattransfercapability, andreduction intheratedthermalpower,arelesslimitingparameters relativetotheassumptions currently madeforthemass/energy releasecalculations following asteamline breakinsidemal9444w.wpf:1d441195 3.5-17 containment.

Theparameter changesassociated withtheSGTPProgramdonotwarrantreanalysis ofthisevent.However,evaluations arecurrently inplace(References 1and2)toaddressseveralnon-conservative assumptions intheanalysis.

Areanalysis effortwasundertaken forthesteamline breakmassandenergyreleasesinsidecontainment aspartoftheSGTPProgram,sothattheReference 1and2evaluations willnolongerberequiredandtosupporttheincreased operating marginsincludedinthediscussion ofthepreviousparagraph.

Thissectiondiscusses aseriesofsteamline breaks,consistent withthecasespresented intheUFSAR,whichwereanalyzedtodetermine themassandenergyreleasesfromavarietyofpostulated pipebreaksencompassing widevariations inplantoperation, safetysystemperformance, andbreaksizes.Themassandenergyreleasedataissubsequently usedasinputtothecontainment integrity analysisdiscussed inSection3.5.5.3.5.4.1MethodofAnalysisTheLOFTRANcomputercode(Reference 3)wasusedtocalculate thebreakflowsandenthalpyofthereleasesthroughthesteamline breakasafunctionoftime.Blowdownmasslenergy releasesdetermined usingLOFTRANincludetheeffectsofcorepowergeneration, mainandauxilianJ feedwater additions, engineered safeguards systems,reactorcoolantthickmetalheatstorage,andreversesteamgenerator heattransfer.

Aboundinganalysiswasperformed toaddresstherangeofconditions possibleforUnit1operation andthepotential Unit2uprating.

Theassumptions ontheinitialconditions aretakentomaximizethemassandtotalenergyreleased.

Thehigherprimarytemperatures alongwiththehigherupratedpowerlevelassociated withtheUnit2upratingparameters areconservative forthemass/energy releasecalculations.

Theupperboundtemperature ofTableS-2.1-1,Case8ofWCAP-11902, Supplement (Reference 4),wasused.Sincethemassblowdownrateisdependent onsteampressure, andthesteampressureislessforthelowerboundtemperature case,thesteampressureoftheupperboundtemperature caseislimitingfortherangeofoperating conditions possiblefortheupratingofUnit2.Thefunctions whichactuatesafetyinjection andsteamline isolation duringasteamline ruptureeventarecommonlyreferredtoastheSteamline BreakProtection System.Aplant'ssteamline breakprotection systemdesigncanhavealargeeffectonsteamline breakresults.Thecurrentsteamline breakprotection systemdesignsforUnit1andUnit2aredifferent.

ThecurrentUnit1designisreferredtoasan"OLD"steamline breakprotection systemdesign.TheUnit2design(andtheproposedmodifiedUnit1design;seeSection3;3.2.5)isreferredtoasa"HYBRID"steamline breakprotection systemdesign.Thetwosystemshavethefollowing characteristics:

mal9444w.wpf:

1d~11953.5-18 CurrentUnit1-"OLD"Steamline BreakProtection SafetyInjection Signals1.Highsteamflowcoincident withlowsteam(inc pressure(twooutoffourlines)2.Highsteamflowcoincident withlow-lowT,~(twooutoffourlines)3.Twooutofthreedifferential pressuresignalsbetweenasteamlineandtheremaining steamlines4.Twooutofthreelowpressurizer pressuresignals5.Twooutofthreehighcontainment pressuresignalsSteamline Isolation Signals1.Highsteamflowcoincident withlowsteamline pressure(twooutoffourlines)2.Highsteamflowcoincident withlow-lowT,~(twooutoffourlines)3.Twooutoffourhigh-high containment pressuresignalsUnit2-"HYBRID"Steamline BreakProtection SafetyInjection SignalsLowsteamline pressure(twooutoffourlines)2.Twooutofthreedifferential pressuresignalsbetweenasteamlineandtheremaining steamlines3.Twooutofthreelowpressurizer pressuresignals4.Twooutofthreehighcontainment pressuresignalsSteamline Isolation Signals1.Lowsteamline pressure(twooutoffourlines)2.Highsteamflowcoincident withlow-towT,(twooutoffourlines)3.Twooutoffourhigh-high containment pressuresignalsTheonlydifferences betweenthecurrentUnit1andUnit2steamline breakprotection logicdesignsaretheactuations fromahighsteamflowandlow-lowT,~signalandthelogicassociated withthelowsteamline pressuresignalrequiredtoactuatesafetyinjection andmA1944+w.wpf:1d~1195 3.5-19 steamline isolation.

Currently, forUnit1,ahighsteamflowcoincident withlow-lowT,~signalactuatesbothsafetyinjection andsteamline isolation.

ForUnit2,ahf'ghsteamflowcoincident withlow-lowT,signalactuatesonlysteamline isolation.

However,thedifference isnotsignificant forthecalculation ofthemass/energy releasessincetheanalysisdoesnottakecreditforanyESFactuations onahighsteamflowcoincident withlow-lowT,signal.ThecurrentUnit1designrequiresacoincidence betweenthelowsteamline pressureandhighsteamflowforprotection actuation.

TheUnit2designonlyrequiresthelowsteamline pressuresignalforprotection actuation; nocoincidence withsteamflowisrequired.

Thecoincidence logicrequiredforsafetyinjection initiation andsteamline isolation onhighsteamflowandlowsteampressureforthecurrentUnit1designismorelimitingforthecalculation ofmass/energy releasesinsidecontainment thanUnit2'sdesign.Actuation ofsafetyinjection andsteamline isolation willlimitthemass/energy releasedtothecontainment.

Delayingthesafeguards initiation willresultinaconservative calculation ofthemass/energy releasesforthecontainment pressureandtemperature evaluation.

Thecoincidence requirement forhighsteamflowwithlowsteampressureofthecurrentUnit1designincreases thelikelihood thatsafeguards initiation mightbedelayedcomparedtoUnit2'sdesignwhereonlyalowsteampressuresignalisrequired.

Inthecasewherethecoincidence logicprohibits safetyinjection andsteamline isolation onhighsteamflowwithlowsteampressure, oneoftheothersignalsmustbereceivedbeforethesafeguards areinitiated.

Assuch,thecurrentUnit1steamline breakprotection systemdesignwasassumedinthisboundinganalysisforthecalculation ofthemass/energy releasesinsidecontainment.

3.5.4.2Assumptions Severalsteamline breakswereanalyzedtodetermine alimitingbreakcondition forthecontainment temperature andpressureresponse.

Thefollowing assumptions wereusedintheanalysis:

Double-ended pipebreakswereassumedtooccuratthenozzleofonesteamgenerator andalsodownstream oftheflowrestrictor.

(SinceneitherUnit1norUnit2hasintegralflowrestnctors.)

Splitruptureswereassumedtooccuratthenozzleofonesteamgenerator.

b.Theblowdownisassumedtobedfysaturated steam.c.Asdiscussed previously, theUnit1steamline breakprotection systemdesignisassumed.However,creditwasnottakenforsafeguards actuation onhighsteamlinedifferential pressureorhighsteamflowcoincident withlow-lowT,.d.Steamline isolation isassumedcomplete11secondsafterthesetpointisreachedforeitherhighsteamflowcoincident withlowsteampressureorm:L1944+w.wpf:1d~1195 3.5-20 high-high containment pressure.

Theisolation timeallows8secondsforvalveclosureplus3secondsforelectronic delaysandsignalprocessing.

AspartoftheReratingProgram,4.6ft'nd1.4ft'ouble-ended pipebreakswereanalyzedat102,70,30,andzeropercentpowerlevels.FortheSGTPProgram,asub-setofthesedouble-ended pipebreakshavebeenre-analyzed.

Thissub-set,listedbelow,corresponds tothemostlimitingdouble-ended pipebreaks,asdetermined duringtheReratingProgrameffort.4.6ft',102%power,withanMSIVfailure;4.6ft',70%power,withanMSIVfailure;1.4ft',102%power,withanMSIVfailure;1.4ft',70%power,withanMSIVfailure.AspartoftheReratingProgram,splitpiperuptureswereanalyzedat0.86ft',102%power,0.908ft',70%power,0.942ft',30%power;and0.4ft',hotshutdown.

ThesesplitbreaksizesforeachpowerlevelweremodeledbecausetheyreflectthelargestbreaksforwhichESFactuations (i.e.,steamline isolation, feedwater isolation, andsafetyinjection) mustbegenerated byhighcontainment pressuretrips.Thehighsteamflowcoincident withlowsteampressureisnotreachedforthesebreaksizesorsmallerbreaksizesattherespective powerlevels(Reference 5).FortheSGTPProgram,asub-setofthesesplitbreakshavebeenre-analyzed.

Thissub-set,listedbelow,corresponds tothemostlimitingsplitbreaksizes,asdetermined duringtheReratingProgrameffort.0.86ft',102%power,withanauxiliary feedwater runoutprotection (AFWRP)failure;0.942ft',30%power,withanAFWRPfailure;0.942ft',at30%power,withanMSIVfailure.FailureofanMSIV,failureofafeedwater isolation valveormainfeedpumptrip,andfailureofauxiliary feedwater runoutcontrolwereconsidered.

AspartoftheReratingProgramtwocasesforeach>reaksizeandpowerlevelscenariowereanalyzedwithonecasemodelingtheMSIVfailureandtheothercasemodelingtheAFWrunoutcontrolfailure.Eachcaseassumedconservative mainfeedwater additiontoboundthefeedwater isolation valveormA1944+w.wpt:1d441195 3.5-21 mainfeedpumptripfailure.FortheSGTPProgram,thesub-setofcasere-analyzed assumedafailureaspreviously noted(itemse.andf.).Thecasere-analyzed fortheSGTPProgramalsoassumed.Feedlineisolation viaFMOvalvescomplete44secondsafterthesetpointisreachedforeitherhighsteamflowcoincident withlowsteampressureorhighcontainment pressure.

Theisolation timeallows41secondsforvalveclosureplus3secondsforelectronic delaysandsignalprocessing.

h.Theauxiliary feedwater systemismanuallyre-aligned bytheoperatorafter10minutes.Ashutdownmarginof1.3%Ak/kisassumed.Amoderator densitycoefficient of0.54Ak/gm/ccisassumed.Minimumcapability forinjection ofboricacid(2400ppm)solutioncorresponding tothemostrestrictive singlefailureinthesafetyinjection system.TheECCSconsistsofthefollowing systems:1)thepassiveaccumulators, 2)thelowheadsafetyinjection (residual heatremoval)system,3)thehighhead(intermediate head)safetyinjection system,and4)thechargingsafetyinjection system.Onlythechargingsafetyinjection systemandthepassiveaccumulators aremodeledforthesteamlinebreakaccidentanalysis.

Themodelingofthesafetyinjection systeminLOFTRANisdescribed inReference 3.Figure3.3-49ofthisreportpresentsthesafetyinjection flowratesasafunctionofRCSpressureassumedintheanalysis.

Theflowcorresponds tothatdelivered byonechargingpumpdelivering itsfullflowtothecoldlegs.Thesafetyinjection flowsassumedinthisanalysistakeintoaccountthe10%degradation ofthechargingpumpperformance.

Nocredithasbeentakenforanyboratedwaterthatmightexistintheinjection lines,whichmustbesweptfromthelinesdownstream oftheboroninjection tankisolation valvespriortothedeliveryofboricacidtotheRCSloops.Forthisanalysis, aboronconcentration of0ppmfortheboroninjection tankisassumed.Afterthegeneration ofthesafetyinjection signal(appropriate delaysforinstrumentation, logic,andsignaltransport included),

theappropriate valvesbegintooperateandthesafetyinjection chargingpumpstarts.In27seconds,thevalvesareassumedtobeintheirfinalposition(VCTchargingpumpsuctionvalvehasclosedfollowing openingofRWSTchargingpumpsuctionvalve)andthepumpisassumedtobeatfullspeedandtodrawsuctionfromtheRWST.Thevolumecontaining thelowconcentration boratedwaterissweptintothecorebeforethe2400ppmboratedwaterreachesthecore.Thisdelay,described above,isinherently includedinthemodeling.

Notethattherelaxedm%1944<w.wpf:1d441195 3.5-22 EDGstart-uptimeisnotreflected inthesteamline mass/energy

releases, asconservative releasesareobtainedifoffsitepowerismaintained (seeitemm.below):.Fortheat-powercases,reactortripisavailable bysafetyinjection signal,overpower protection signal(highneutronfluxreactortriporOPETreactortrip),andlowpressurizer pressurereactortripsignal.m.ForRCPoperation, offsitepowerisassumedavailable.

Continued operation oftheRCPsmaximizes theenergytransferred fromtheRCStothesteamgenerator.

n.Nosteamgenerator tubepluggingisassumedtomaximizetheheattransfercharacteristics.

3.5.4.3SingleFailureEffectsFailureofanMSIVincreases thevolumeofsteampipingwhichisnotisolatedfromthebreak.Whenallvalvesoperate,thepipingvolumecapableofblowingdownislocatedbetweenthesteamgenerator andthefirstisolation valve.Ifthisvalvefails,thevolumebetweenthebreakandupstreamoftheisolation valvesintheothersteamlines, including safetyandreliefvalveheadersandotherconnecting lines,willfeedthebreak.ForthecaseswhichmodeledafailureofanMSIV,thesteamline volumesassociated withUnit1wereassumedsincethevolumeavailable forblowdownforthisscenarioisgreaterthanthatforUnit2.ForthecaseswhichdidnotmodelafailureofanMSIV,thesteamline volumesassociated withUnit1wereassumedsincethevolumeavailable forblowdownforthisscenarioisgreaterthanthatforUnit2.b.Failureofadieselgenerator wouldresultinthelossofonecontainment safeguards trainresulting inminimumheatremovalcapability.

Failureofafeedwater isolation valvewouldresultinadditional inventory inthefeedwater linewhichwouldnotbeisolatedfromthefaultedsteamgenerator.

Themassinthisvolumecanflashintosteamandexitthroughthebreak.Forconsistency withtheUFSARsteamline breakmasslenergy releaseanalysis, allcasesconservatively assumedfailureofthefeedwater isolation valve,whichresultedintheadditional inventory available forreleasethroughthesteambreak andinhigherthannormalmainfeedwater flows.d.Failureoftheauxiliary feedwater runoutcontrolequipment wouldresultinahigherauxiliary feedwater flowenteringthefaultedsteamgenerator priortore-alignment oftheAFWsystem.Forcaseswheretherunoutcontroloperatesproperly, aboundingm:51944+w.wpf:Id~

11953.5-23 constantAFWflowof775gpmtothefaultedsteamgenerator wasassumed.Thisvaluewasincreased to1375gpmtosimulateafailureoftherunoutcontrol.3.5.4.4ResultsThesteamline breakmass/energy releasesinsidecontainment werecalculated toaccountfortherangeofconditions possibleforthepotential upratingofUnit2.Onesetofmass/energy releaseswerecalculated toboundbothUnitsincorporating thelimitingsteamline breakprotection designcurrently installed inUnit1.Section3.5.5presentsthecontainment integrity evaluation foramainsteamline breakusingthemass/energy releasescalculated here.Asdiscussed inSection3.5.5,thelimitingscenarios ofthesteambreak casesanalyzedforthecontainment responseevaluation wereabreaksizeof1.4ft'ccurring at102%powerwithanMSIVfailureforthedouble-ended rupturescenarioandabreaksizeof0.942ft'ccurring at30%powerwithanMSIVfailureforthesplitrupturescenario.

Table3.5.13presentsthemass/energy releasesfortheselimitingsteambreak casesofthecontainment responseevaluation.

3.5.4.5References 1."American ElectricPowerServiceCorporation, DonaldC.CookNuclearPowerPlantUnits1and2,Increased Upper8LowerCompartment SprayDeliveryTimes,"WLetterAEP-94-712, June13,1994.2."American ElectricPowerServiceCorporation, DonaldC.CookNuclearPowerPlantUnits1and2,Feedwater Isolation ValveEvaluation Support,"

WLetterAEP-93-528, April8,1993.3.Burnett,T,W.T.,etal.,"LOFTRANCodeDescription,"

WCAP-7907-A, April1,1984.4."ReratedPowerandRevisedTemperature andPressureOperation forDonaldC.CookNuclearPlantUnits1&2Licensing Report,"WCAP-11902, Supplement, September 1989.5.Land,R.E.,"MassandEnergyReleasesFollowing aSteamLineRupture,"

WCAP-8860, September 1976.mal944+w.wpf:1d441195 3.5-24 3.5.5MainSteamLineBreakContainment Integrity 3.5.5.1Introduction andBackground Aseriesofmainsteamlinesplitanddouble-ended breakswereanalyzedaspartoftheReratingProgramforCookNuclearPlantUnits1and2todetermine themostseverebreakcondition forcontainment temperature andpressureresponseforthisdesignbasisevent.Theanalysisandevaluation conducted arediscussed inReference 1.TheresultsfromtheReratingProgram,whicharedocumented intheFSAR,showthattheworstcaseofthedouble-ended breakswasa4.6squarefootbreak,occurring at102%powerwithamainsteamisolation valvefailure.Theworstcaseforthesplitbreakswasthe0.86squarefootbreak,occurring at102%power,withthefailureofauxiliary feedwater runoutprotection.

Thecalculated peakcontainment temperature was324.9'Fand324.4'F,respectively.

3.5.5.2PurposeofAnalysisAnanalysiswasperformed asapartoftheSGTPProgram,todemonstrate thatthepeakcontainment temperature resulting fromadesignbasismainsteamlinebreakwillnotexceedtheequipment qualification criterion fortheplant.Theanalysiswasperformed toboundCookNuclearPlantUnits1and2operation underupratedconditions (3600MWtNSSS).Thecontainment pressureresponsegenerated fortheLOCAdouble-ended pumpsuctionbreak(Section3.5.3)iscalculated tobemoresevere,andtherefore isnotaconcernhere.3.5.5.3MajorAnalytical Assumptions Ananalysisconsistent withtheReference 1analysiswasperformed.

Theanalytical effortprovidesboundingcalculations forbothUnits182atapowerlevelof3600MWt.AspectrumofthelimitingsplitbreaksfromReference 1wereanalyzed:

0.86ft',102%power,withanAFWRPfailure;0.942ft',30%power,withanAFWRPfailure;0.942ft',at30%power,withanMSIVfailure.Also,thefollowing double-ended breaksfromReference 1wereanalyzed:

4.6ft',102%power,withanMSIVfailure;4.6ft',70%power,withanMSIVfailure;1.4ft',102%power,withanMSIVfailure;1.4ft',70%power,withanMSIVfailure.aThemassandenergyreleasetocontainment asaresultofthepostulated steamlinebreakwerecalculated usingtheLOFTRANcomputercode(Reference 2).Consistent withthem:$19444w.wpf:

1d4411953.5-25 Reference 1analysis, nocreditwastakenforentrainment.

Section3.5.4presentsadditional detailsregarding thecalculation oftheinsidecontainment steamlinebreakmassandenergyreleases.

Theconsequences ofthesereleases; inparticular thepeakcontainment temperature, wascalculated usingtheLOTIC-3computercode(Reference 3).Thefollowing arethemajorinputassumptions usedinLOTIC-3:Thecontainment integrity calculations wereperformed withanadditional failureofoneofthecontainment safeguards trains,e.g.,oneoftwospraypumps,whichresultsintheminimumsprayflowandoneoftwoairreturnfans.Whereapplicable, plantdataconsistent withtheLOCAcontainment integrity analysis(Section3.5.3)wasused.2.Thetotalinitialicemassusedis2.11x10'bs.3.Theinitialconditions inthecontainment areatemperature of120'Fintheloweranddead-ended compartments, atemperature of27'Fintheicecondenser, andatemperature of57'Fintheuppercompartment.

Allvolumesareatapressureof0.3psigandarelativehumidityof15%.4.TheRWSTtemperature wasassumedtobe105'F.5.Acontainment spraypumpflowof2075gpmtotheuppercompartment and1006gpmtothelowercompartment wasassumed.6.Containment sprayresponsetimefollowing high-high containment pressuresetpointis115seconds.7.ThemassandenergyreleasesaregiveninSection3.5.4.3.5.5.4ResultsLareBreakThelimitingcaseamongthedouble-ended

ruptures, whichyieldedacalculated peaktemperature of322.7'F,is.the1.4ft'ouble-ended rupture,102%power,MSIVfailurecase.Figures3.5-6and3.5-7providetheupperandlowercompartment temperature profiles.

Figures3.5-8and3.5-9illustrate theupperandlowercompartment pressuretransients.

mh19444w.wpf:1d441195 3.5-26

SmallBreakThemostlimitingcaseintermsofpeakcalculated temperature isthe0.942ft'plitbreak,30%power,withanMSIVfailure.Thiscaseresultedinacalculated peaktemperature of326'F.Figures3.5-10and3.5-11providetheupperandlowercompartment temperature profiles.

Figures3.5-12and3.5-13presenttheupperandlowercompartment pressuretransients.

3.5.5.5Conclusion Themainsteamlinebreakcontainment integrity analysishasbeenperformed consistent withthecurrentlicensing basisanalysisandSGTPProgram,considering thepresentoperating plantconditions.

TheresultsofthisanalysisshowthattheEnvironmental Acceptance Criteria(Reference 4)applicable forCookNuclearPlantUnits1and2aremet.Thisanalysistherefore demonstrates thatthecontainment heatremovalsystemsfunctiontorapidlyreducethecontainment pressureandtemperature intheeventofamainsteamlinebreakaccident.

GDC50and10CFRPart50AppendixKaresatisfied, whichismorerestrictive thantheGDCcriteriainAppendixHoftheoriginalFSAR,towhichtheDonaldC.CookNuclearPlantsarelicensed.

3.5.5.6References WCAP-11902, Supplement 1,September 1989,"ReratedPowerandRevisedTemperature andPressureOperation forDonaldC.CookNuclearPlantUnits1&2Licensing Report."2.WCAP-7907-P-A (Proprietary),

"LOFTRANCodeDescription",

April1984.3.WCAP-8354-P-A (Proprietary),

Supplement 2,"LongTermIceCondenser Containment Code-LOTIC-3Code",February1979.4.AEP/WSGTP-19,"DonaldC.CookNuclearPlantSteamGenerator TubePluggingAnalysisTechnical Documentation Transmittal",

August10,1994.mA1944+w.wpt:1d~1195 3.5-27

TABLE3.5-1DONALDC.COOKNUCLEARPLANTUNITS1AND2SYSTEMPARAMETERS "INITIALCONDITIONS PARAMETERS CoreThermalPower(MWt)ReactorCoolantSystemFlowrate, perLoop(gpm)VesselOutletTemperature

('F)CoreInletTemperature

('F)VesselAverageTemperature

('F)InitialSteamGenerator SteamPressure(psia)SteamGenerator DesignSteamGenerator TubePlugging(%)InitialSteamGenerator Secondary SideMass(Ibm)Accumulator WaterVolume(ft')N,CoverGasPressure(psia)Temperature

('F)SafetyInjection Delay(sec)(includes timetoreachpressuresetpoint)

VALUE341379000615.2547.4581.3836.3Model5111407594660012048.0(analysis valueincludesanadditional

+5.1'Fallowance forinstrument erroranddeadband) mh1944+w.wpf:1d441195 3.5-28 TABLE3.5-2DONALDC.COOKNUCLEARPLANTUNITS1AND2SAFETYINJECTION FLOWMinimumSlRCSPressure(psig)020406080100120140160180200INJECTION MODETotalFlow(gpm)3635.53447.23235.33003.72738.02425.62041.31493.3889.5883.0876.4RECIRCULATION MODE(w/oRHRSpray)TotalFlow(gpm)3011RECIRCULATION MODE(w/RHRSpray)TotalFlow(gpm)mA1944<w.wpf:1d451895 3.5-29 TABLE3.54DONALDC.COOKNUCLEARPLANTUNITS1AND2DOUBLE-ENDED PUMPSUCTIONGUILLOTINE MINIMUMSISLOWDOWNMASSANDENERGYRELEASETIMEBREAKPATHNO.lFLOWBREAKPATHNO.2FLOWSECONDS.000.101.201.301.401.601.8001.101.402.302.803.003.403.904.605.206.206.608.008.408.809.4011.013.816.418.418.819.019.219.419.619.820.020.220.620.821.021.421.822.022.222.423.623.824.424.625.827.028.0LBM/SEC.040932.241809.946791.447304.844920.545062.141715.938993.831603.526248.921790.019301.618078.715511.613955.812621.112409.812926.312289.710124.19779.39523.67137.65492.84381.24172.24012.73961.13820.33734.83554.03322.73066.42642.12411.62237.31909.01705.51562.81445.61337.01176.0914.5833.4475.4413.0140.866.519.0THOUSANDBTU/SEC.022363.522984.525944.926505.725777.326395.225037.923888.120634.817695.214847.813347.212537.010751.89663.88669.38475.88608.28355.67526.07313.06842.35509.24593.43819.63729.53668.63685.63688.43730.03671.63582.13451.83188.32951.72755.62365.52122.81947.91805.11672.21475.11151.51050.2600.7523.1179.485.224.5,LBM/SEC.021842.523698.723586.223083.221588.619910.918871.618466.417983.417008.416632.615908.215034.413993.713348.712570.913180.512562.212442.312205.811841.310832.19219.17813.76705.59594.35136.18561.98618.35453.28394.04907.96807.84697.05832.63815.94740.72922.46501.64567.95030.533M.61329.92964.02402.32843.32309.9299.7126.0THOUSANDBTU/SEC.011892.712911.512864.812605.811803.010889.610327.010106.39842.59311.59107.58714.58239.87674.47323.96900.47237.46907.86843.76713.46514.65959.45074.54311.03474.54972.02661.24245.44403.82703.54148.02449.33152.72086.82567.91683.91974.21187.02457.31744.31926.81231.8459.4796.5633.4754.2648.093.945.3mA1944<w.wpI:1d~1195 3.5-30 TABLE3.&4DONALDC.COOKNUCLEARPLANTUNITS1AND2DOUBLEENDEDPUMPSUCTIONGUILLOTINE MINIMUMSlREFLGODMASSANDENERGYRELEASETIMESECONDS28.028.328.528.729.730.030.734.136.238.139.140.141.242.243.245.247.248.249.251.253.255.257.259.261.262.363.364.368.370.374.378.381.382.386.393.397.3101.3105.3107.3117.3127.3135.3143.3153.3163.3175.3193.3223.3249.7LBM/SEC.0143.2122.8110.9108.0119.6125.7148.1160.6170.9213.4351.3371.3368.0362.3350.6339.7355.6346.0337.3329.2321.6314.5307.8313.7456.3448.9414.7399.6372.0347.1329.9324.5304.0272.8257.3243.3230.9225.1201.1183.5173.0165.1158.1153.5150.'I148.0148.4150.3THOUSANDBTU/SEC.0168.1144.3130.3126.8140.4147.6174.1188.8200.9251.1414.5438.4434.6427.7413.8400.8419.7419.3408.2397.9388.3379.3370.8362.9369.5539.8531.0490.1472.1439.2409.5389.1382.6358.3321.1302.8286.3271.5264.7236.3215.4203.0193.7185.5180.0176.0173.5173.8175.9BREAKPATHNO.1FLOWBREAKPATHLBM/SEC.01818.31791.41778.91717.91696.21658.11508.01431.71371.52017.33809.94023.23989.73929.73807.43691.63902.63885.23787.03694.53607.33524.83446.83372.8238.7301.8298.4282.9276.1263.7252.5244.9242.6233.6220.1213.5207.6202.4200.0190.1183.0178.8175.8173.1171.4170.1169.4169.6170.5NO.2FLOWTHOUSANDBTU/SEC.0162.7160.3159.2153.8151.8148.4135.0128.1122.8223.3531.6581.6581.3574.9560.8547.3563.8561.6549.9538.9528.5518.7509.4500.5160.2244.5240.1219.4210.4194.1179.5169.7166.5154.9137.4128.9121.3114.6111.698.989.884.480.476.974.572.771.470.971.4mA19444w.wpf:1d441195 3.5-31 TABLE3.5-5DONALDC.COOKNUCLEARPLANTUNITSIAND2DOUBLE-ENDED PUMPSUCTIONGUILLOTINE MINIMUMSIPRINCIPAL PARAMETERS DURINGREFLOODTIMESECONDS28.028.528.929.229.430.533.'I36.241.243.044.247.248.250.158.261.262.363.366.g74.395.3106.3121.3136.3155.3171.8191.3210.0229.3231.3249.7DEGREEF219.92I7.1215.1214.8214.8215.1216.7218.7221.6222.6223.4225.3226.0227.2232.9235.1235.8236.5238.4242.8244.3242.3243.4244.0243.2244.3243.5244.3243.8244.3244.2244.1IN/SEC.00019.9628.4523.4124.3032.5702.1412.2203.5303.4013.3193.1493.2563.1772.8992.8173.0923.7153.5153.0352.6282.2241.9491.6941.5371.4231.3751.3471.3381.3341.3341.337FLOODINGTEMPRATECARRYOVER FRACTION.000.000.018.041.314.534.635.710.724.731.742.745.750.761.764.765.763.766.770.771.769.768.766.764.765.765.768.770.774.774.776COREHEIGHTFT.00.561.051.181.241.501.772.002.362.502.602.812.873.003.503.673.733.804.014.525.005.566.006.537.007.558.008.529.009.499.5410.00DOWNCOMER HEIGHTFT.00.35.55.971.273.007.0011.5515.9916.0016.0016.0016.0016.0016.0016.0015.9815.7915.2313.9512.9212.0211.5511.3111.3611.6812.0712.6113.1713.7513.8114.37FLOWFRACTION.2501.0001.0001.0001.000.598.485.445.578.575.573.566.576.575.562.558.612.628.626.62'I.614.604.594.581.571.563.560.559.581.563.563.565(POUNDSMASSPER.0.07168.87168.87070.47070.46999.36999.36953.16953.16672.96672.96194.76194.75728.85728.84713.34713.34549.34549.34455.04455.04240.34240.34472.34049.64397.43971.23981.53548.83851.33416.7437.7.0403.2.0408.8.0421.8.0431.8.0441.3.0447.3.0452.4.0455.8.0458.8.0460.6.0462.4.0463.9.0465.3.0465.5.0466.7.0SECOND).0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0.0INJECTION TOTALACCUMULATOR SPILLENTHALPYBTU/LBM..0089.5089.5089.5089.5089.5089.50I89.50:89.5089.5089.5089.5087.9487.9087.7187.6472.9972.9972.9972.9972.9972.9972.9972.9972.9972.9972.9972.9972.9972.9972.9972.99mAI9444w.wpI:1d 0411953.5-32 TABLE3.5+DONALDC.COOKNUCLEARPLANTUNITS1AND2DOUBLE.ENDEDPUMPSUCTIONGUILLOTINE MINIMUMSIPOSTREFL'OODMASSANDENERGYRELEASETIMESECONDS249.7254.7259.7264.7269.7274.7279.7284.7289.7299.7304.7309.7314.7324.7329.7339.7344.7359.7364.7399.7404.7424.7449.7454.7469.7474.7489.7499.7519.7529.7534.7544.7549.7554.7574.7579.7609.7614.7870.8870.9874.71979.71982.32222.22316.8LBM/SEC200.8201.2200.4200.8200.0200.3199.6199.9199.0199.6198.7199.0198.1198.5197.6197.9197.0197.2196.2195.8194.8194.4193.3193.8192.5192.8191.7191.7190.7190.9190.1190.1189.5189.9188.6188.6187.382.082.079.979.864.864.764.763.463.4THOUSANDBTU/SEC251.0251.5250.6251.0250.0250.5249.5249.8248.8249.5248.5248.7247.7248.1247.0247.4246.2246.5245.3244.7243.5243.1241.6242.3240.6241.0239.7239.7238.4238.7237.7237.7236.9237.4235.8235.8234.1102.5102.594.499.475.280.580.578.878.8BREAKPATHNO.1FLOWBREAKLBM/SEC285.3284.9285.7285.3286.1285.8286.6286.3287.1286.6287.4287.2288.0287.6288.5288.2289.2289.0289.9290.4291.4291.7292.9292.3293.7293.3294.4294.4295.5295.2296.0296.0296.7296.2297.5297.5298.8404.1404.1406.3406.3421.4103.3103.3332.2332.2103.3103,2103.2103.0103.1102.9103.0102.8102.9102.5102.6102.5102.5102.2102.3102.0102.1101.7101.8101.2101.3100.9100.7100.4100.4100.2100.299.999.799.599.599.399.499.299.098.998.5120.8120.888.7155.784.896.596.5158.0158.0PATHNO2FLOWTHOUSANDBTU/SECmal944+w.wpf:1d441195 3.5-33 TABLE3.5-7DONALDC.COOKNUCLEARPLANTUNITS1AND2DOUBLE-ENDED PUMPSUCTIONGUILLOTINE MINIMUMSI'ASSBALANCETIME(SECONDS)

.0028.0028.00249.66870.942316.82INITIALINRCSANDACCADDEDMASSPUMPEDINJECTION TOTALADDED-TOTALAVAILABLE

"'ASS(THOUSAND LBM)771.32771.32771.32771.32771.32771.32.00.00.0091.01393.021087.36.00.00.0091.01393.021087.36771.32771.32771.32862.331164.341858.68DISTRIBUTION REACTORCOOLANT537.3257.7467.87135.93135.93135.93ACCUMULATOR 234.00171.20161.07.00.00.00TOTAlCONTENTS771.32228.94228.94135.93135.93135.93EFFLUENTBREAKFLOWECCSSPILLTOTALEFFLUENT'-TOTALACCOUNTABLE-

.00542.36542.36726.391028.391722.73.00.00.00.00.00.00.00542.36542.36726.391028.391722.73771.32771.30771.30862.311164.321858.66m%1944-6w.wpt:td.04t 1953.5-34 TABLE3.5-8DONALDCCOOKNUCLEARPLANTUNITS1AND2DOUBLE-ENDED PUMPSUCTIONGUILLOTINE MINIMUMSlENERGYBALANCETIME(SECONDS)

.0028.0028.00249.66870.942316.82INITIALENERGYINRCS,ACC,S GENADDEDENERGYPUMPEDINJECTION DECAYHEATENERGY(MILLIONBTU)901.43901.43901.43.00.00.008.96.008.96-5.10TOTALADDED'""TOTALAVAILABLE

-'ISTRIBUTION REACTORCOOLANTACCUMULATOR CORESTOREDPRIMARYMETALSECONDARY METALEFFLUENTSTEAMGENERATOR TOTALCONTENTSBREAKFLOWECCSSPILLTOTALEFFLUENT'"TOTALACCOUNTABLE-'00 3.87901.43905.30318.0012.7420.9415.3228.0613.71178.97168.7484.1984.08271.26275.82901.43570.41.00334.41.00.00.00334.41901.43904.823.87905.3013.6414.4213.71168.7484.08275.82570.41334.41.00334.41904.82HEATFROMSECONDAR.00-5.10901.436.6434.20-5.1035.75937.1830.54.003.19143.6077.83252.03507.19421.73.00421.73928.92901.4328.6987.00-2.19113.491014.9230.54.003.1692.9459.46189.66375.76630.90.00630.901006.66901.4384.57181.864.03.270.461171.8930.54.002.9263.4135.57116.14248.57916.99.00916.991165.56rnA19444w.wpf:1d441195 3.5-35 TABLE3.5-9ENERGYACCOUNTING INMILLIONSOFBTUIceHeatRemovalStructural HeatSinksRHRHeatExchanger HeatRemovalSprayHeatExchanger HeatRemovalEnergyContentofSumpIceMelted(Pounds)(10')Approx.EndofBlowdown(t=10.0sec.)207.717.37188.940.67Approx.EndofReflood(t=249.7sec.)250.344.73250.00.84Integrated Energiesm%19446w.wpf:

1d6411953.5-36 TABLE3.5-10ENERGYACCOUNTING INMILLIONSOFBTUApprox.TimeofApprox.TimeofIceMeltOutPeakPressure(t=5423sec.)(t=7752sec.)IceHeatRemoval'tructural HeatSinksRHRHeatExchanger HeatRemoval'pray HeatExchanger HeatRemoval'nergy ContentofSumpIceMelted(Pounds)(10')

567.2182.5249.058.31583.62.11567.21112.6877.3192.3599.32.11Integrated Energiesm:519444w.wpf:

1d4411953.5-37 TABLE3.5-11STRUCTURAL HEATSINKTABLESURFACESUpperCompartment Material1.PaintCarbonSteelConcrete2.PaintConcrete3.PaintConcrete4.PaintConcreteLowerCompartment Material5.PaintConcrete6.PaintConcrete7.PaintConcrete8.PaintConcreteAREA(Ft)32500.32500.32500.10086.10086.5880.5880.11970.11970.5069.5069.13660.13660.16730.16730.8665.8665.THICKNESS (Ft)0.0010830.04692.00.0010832.00.001251.50.001251.00.001252.00.001251.50.001251.00.001252.0IceCondenser 9.Steel10.Steel11.Steel12.PaintConcrete180600.76650.28670.3336.3336.0.006630.0217,0.02670.0008330.333m:519444w.wpf:1d441195 3.5-38 TABLE3.5-11(continued)

STRUCTURAL HEATSINKTABLESURFACESIceCondenser 13.SteelandInsulation Steel14.SteelandInsulation ConcreteAREA(Ft')19100.19100.13055.13055.THICKNESS (Ft)1.00.06251.01.0mA1944+w.wpt:1d441195 3.5-39 TABLE3.5-12MATERIALPROPERTIES TABLEMaterialConcreteConductivity Volumetric HeatCapacityBtu/hr-'ft-'FBtu/ftF0.8Steel26.056.4m:$19444w.wpf:1d441195 3.5-40 TABLE3.5-13UNIT1/UNIT2STEAMLINE BREAKMASS/ENERGY RELEASESINSIDECONTAINMENT 30%Power,0.942ft'plitBreakFailure-MSIVTIME~sec.0000.20005.6007.00010.0013.0013.6014.8015.6016.0018.0020.0026.0035.0040.0045.0050.0060.0070.0080.0090.00100.0110.0120.0150.0200.0270.0290.0292.5295.0297.5320.0337.5352.5367.5395.0410.0432.5495.0605.0MASS~tbm/eec.00001873.1744.1734.1718.1703.1698.1688.1681.1677.1629.1522.1284.1061.974.2905.9853.1782.0741.1719.1707.4701.3698.3696.7695.1694.1692.9691.3667.1607.8554.0476.8403.4344.5296.7183.5136.6114.3106.6109.5ENERGY~MBtu/sec .00002.2342.0852.0732.0542.0362.0312.0202.0112.0071.9501.8251.5441.2771.1731.0911.028.9418.8925.8659.8517.8444.8408.8388.8369.8357.8342.8323.8028.7312.6658.5711.4820.4106.3531.2174.1609.1345.1252.1282mA19444w.wpf:1d~1195 3.5-41 TABLE3.5-13(continued)

UNIT1/UNIT2STEAMLINE BREAKMASS/ENERGY RELEASESINSIDECONTAINMENT 102%Power,1.4ft'oubleEndedRuptureFailure-MSIVTIME~sec.0000.20001.4003.8006.0008.00010.0011.6012.0012.2013.0014.2015.2016.0016.4017.0022.0024.0026.0028.0030.0032.0034.0036.0045.0075.00100.0200.0280.0282.5285.0287.5292.5300.0320.0350.0610.0MASS~Ibm/sec.00009753.8708.7436.7228.7069.6882.6658.6441.6224.5353.4047.2959.2090.1657.1482.1306.1253.1208.1169.1137.1110.1087.1068.1006.878.0851.6831.4825.3789.4695.6619.9455.2241.3112,0106.53.231ENERGY~MBtu/sec.000011.6810.458.9408.6938.5048.2818.0147.7527.4906.4434.8713.5622.5161.9951.7841.5721.5091.4551.4081.3691.3371.3091.2851.2111.0561.024.9998.9924.9485.8349.7431.5429.2850.1.308.1244m'319444w.wp1.'1d~

11953.5-42 10010110210Time(e)310Figure3.5-1LOCAMassEnergy.ReleaseContainment Integrity Containment Pressuremh19444w.wpt:1d~1 1953.5-43 SO010110210Time{I)Figure3.5-2LOCAMassEnergyReleaseContainment Integrity UpperContainment Temperature m:L1944+W.WPf:1d441195 3.5-44 220210O200I~~~~~~~~~~~~~~~~~~0~~I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~170~~~~~~~~~~~~~~~~~0~~~~~~~~~010t10210Time(e)1010Figure3.5-3LOCAMassEnergyReleaseContainment Integrity LowerContainment Temperature mA1944+w.wpf:1d~)

1953.5A5 1801?0LLQP160CL8I~~0$~\~010110210Time(sj31010AottveSumpTwnp.InaotfvaSumpTemp.Figure3.&4LOCAMassEnergyReieaseContainment integrity ActiveSumpandInactiveSumpTemperature Transient

,25E+07.2E+07.15f+Ol~~~~.1E+07~~~~~~~t~~~~~~~~~~~~~~0~~~~~500000~~~t~~~~0101010Time(e)Figure3.5-5LOCAMassEnergyReleaseContainment Integrity IceMeltTransient m:51944<w.wpt:1dM1195 3.5-47 u110ICOICL8Q+100~~~y~~~oo800Time(s)~PIRATURt (FjFigure3.5-6102%Power,1.4sq.ft.DoubleEndedRupture-MSIVFailure~~~UpperCompartment Temperature mal944<w.wpf:1d~1195 3.5-48

~II~I'.~~~~~~~~I~~~~

CO40CL>efhthCLTime(s)PRCSSURE(PSIO)Figure3.5-8102%Power,1.4sq.ft.DoubleEndedRupture-MSIVFailureUpperCompartment Pressurem%1944+w.wpf:1d441195 3.5-50 0PRSSSURR(PSIO)Figure3.5-9102%Power,1.4sq.ft.DoubleEndedRupture-MSIVFailure~~LowerCompartment Pressurem:519444w.wpf:1d441195 3.5-51 I~~~~-I'~I~~~~~~'

85082soKEI~~0~~~t0~Time(s)TEMPIRATURE P)Figure3.5-1'i30%Power,0.942sq.ft.SplitBreak-MSIVFailureLowerCompartment Temperature m:519444w.wpf:1dM1195 3.5-53

CRYl54~~~~~~0Tlma(I)PRRSSURR(tOIO)Figure3.5-1230%Power,0.942sq.ft.SplitBreak-MSIVFailure~~~~~~UpperCompartment Pressurem31944+w.wpf:1d~1195 3.5-54

~~~

'

3.6STEAMGENERATOR TUBERUPTUREACCIDENTANALYSISTheUFSARanalysisofasteamgenerator tuberupture(SGTR)transient isperformed toconservatively predicttheradiological consequences ofsuchanevent.Anevaluation ofthistransient, supporting anincreaseto30%SGTPforDonaldC.CookNuclearPlantUnit1,hasbeencompleted todetermine theimpactonthedosereleases.

Theprimarythermalhydraulic parameters whichaffectthecalculated offsiteradiation dosesforaSGTReventaretheassumedradioactivity levelinthereactorcoolant,thereactorcoolantreleasedthroughtherupturedtubetothesecondary steamgenerator volume,andthesteamreleasedfromtherupturedsteamgenerator totheatmosphere.

WithrespecttotheUFSARanalysis, achangeintheSGTPleveldoesnotimpactthereactivity levelofthereactorcoolant.However,boththepotential'primary coolantreleasetothesecondary andthesecondary steamreleasetotheatmosphere areimpactedbytheassumedtubeplugginglevel~Toevaluatetheeffectofa30%SGTPlevel,themassreleasesfromtheRCSandfromthesecondary volumetotheatmosphere werecalculated.

Fourcaseswereconsidered assuminganominalRCStemperature of533.0'Fand576.3'Fwithbothsymmetric andasymmetric RCSflowconditions.

Anominalfullpowerlevelof3262MWt(NSSS)wasalsoassumed.Thethyroidandwholebodydosesestimated forCookNuclearPlantUnit1,basedonthe30%SGTPevaluation, remainwithina"smallfraction" (10%)ofthe10CFR100exposurelimitguidelines.

SmallfractionisthesmallestoftheexposureguidelinesdefinedinNUREG-0800 (Standard ReviewPlan).Therefore, theconclusions oftheUFSARremainvalid.m:519444w.wpf:1d441195 3.6-1 3.7POST-LOCA HOTLEGRECIRCULATION TIMEThehotlegswitchover calculation performed toprecludeboronprecipitation duringcoldlegrecirculation following aLOCAisnotadversely affectedbythechangesassociated withtheSGTPProgram.Theanalysisisnotaffectedsincetheproposedchangesdonotaffectthenormalplantoperating parameters, thesafeguards systemsactuations, ortheaccidentmitigation capabilities important tothecalculation.

TheSGTPProgramdoesnotsignificantly changetheassumptions usedintheanalysis.

Anyvariation intheinitialRCSfluidinventory duetotubepluggingisjudgedtobeinsignificant withrespecttothecalculated timeforhotlegswitchover toprecludeboronprecipitation.

Otherfluidinventories andboronconcentrations remainunchanged.

Furthermore, theproposedchangesdonotcreateconditions morelimitingthanthoseassumedinthehotlegswitchover calculation.

m:51944+w.wpf:1d441195 3.71

!!I 3.8REACTORCAVITYPRESSUREANALYSISTheReactorCavityPressureAnalysisisperformed tocalculate theinitialpressureresponseinthereactorcavitytoalossofcoolantaccident.

TheReactorCavityPressureAnalysisthatwasperformed fortheReratingProgramwasreviewedanditwasdetermined thattheconclusions providedfortheReratingProgram(WCAP-11902) remainvalidfortheSGTPProgram.TheSGTPProgramparameters affecttheReactorCavityPressureAnalysisthroughthemassandenergyreleasesprovidedasinputintotheanalysis.

ThereisnodirectimpactofSGTPlevelonshort-term massandenergyreleaseratecalculations andcontainment subcompartment responseanalysis(SeeSection3.5.1).ThemajorimpactresultsfromchangestoRCStemperature.

Forshort-term effects,higherreleaseratestypically resultfromcoolerRCSconditions.

ThemassandenergyreleasesusedasinputfortheReactorCavityPressureAnalysisreflected limitingconditions andtherefore, theNSSSperformance parameters fortheSGTPProgramdidnotimpacttheresults.m%19444w.wpf:1d441195 3.8-1 3.9RADIOLOGICAL ANALYSISAreviewwasperformed oftheradielogical analysisintheUFSARforCookNuclearPlantUnit1todetermine theeffectsoftheSGTPProgram.ThesourcetermsforLOCAandthefuelhandlingaccidentareunaffected bytheincreaseinSGTPlevel.However,areanalysis oftheoffsitedosesfollowing alargebreakLOCAwasperformed fortheincreaseinEDGstarttimeto30seconds.TheEDGstarttimedelayresultedinadelayinsprayinjection flowtocontainment of115seconds,whereasthepreviousanalysisassumednodelayinsprayinjection flowtocontainment.

Whiletherewasaslightincreaseintheoffsitethyroiddoses,thedosesarewithintheapplicable limits.SourcetermsfortheSGTReventwererecalculated attheSGTPProgramconditions.

Theradiological consequences oftheSGTRaresummarized inSection3.6ofthisreport.Forthesteamline breakradiological

analysis, theoffsitethyroiddoseandthecorresponding steamgenerator primary-to-secondary leakratedetermined forthealternate steamgenerator tubepluggingcriteriaprogram(APC)is,bydesignofthemethodology, relatively insensitive totheamountofsteamreleasedtotheenvironment.

Forthesteamgenerator intherupturedloop,alloftheinitialiodineactivityalongwithalloftheprimary-to-secondary leakageactivityisreleasedtotheenvironment.

Anyadditional steamreleasefromthissteamgenerator wouldbeduetotheintroduction ofcleanauxfeedwater, whichwouldnotincreasetheactivityreleasedtotheenvironment (i.e.,can'treleasemorethan100%oftheactivityfromthissteamgenerator).

Steamreleasedformtheunaffected steamgenerators duetoboilingofthesecondary coolantaccountsforapproximately 1%ofthetotalactivityrelease.Additional steamreleasedfromthesesteamgenerators willhavenosignificant impactoneitherthecalculated offsitedoseortheallowable primary-to-secondary leakrate.Therefore, asteamline breakradiological analysiswasnotrequiredfortheSGTPProgram.mal944+w.wpf'.1d~

11953.9-1 3.10FLUIDANDAUXILIARY SYSTEMSEVALUATIONS 3.10.1FluidSystemsEvaluation 3.10.1.1Introduction Thissectionaddresses theimpactoftheSGTPProgramontheabilityoftheReactorCoolantSystemandauxiliary fluidsystemstoperformtheirrequiredfunctions.

Theparameters considered arelistedinTable2.1-1.Inordertosupporttheoperation ofCookNuclearPlantUnit1attheSGTPProgramconditions, thefollowing systemswereevaluated atthenewconditions:

1)ReactorCoolantSystem{RCS),2)ChemicalandVolumeControlSystem(CVCS),3)Emergency CoreCoolingSystem(ECCS),4)ResidualHeatRemovalSystem(RHR),and5)SpentFuelPoolCoolingSystem(SFPCS).Abriefdescription ofeachsystemisprovidedbelow.TheEmergency CoreCoolingSystemflowrates wererevisedaspartoftheSGTPProgram.TheseECCSflowrates reflected achargingpumpheaddegradation of10%{differential pressureof2290psidonrecirculation),

asafetyinjection andRHRpumpheaddegradation of15%(differentia pressures of1326psidand150psid,respectively, onrecirculation).

TheECCSflowrates wereusedinthesafetyanalysesandevaluations fortheSGTPProgram.3.10.1.2Description ofFluidSystemsReactorCoolantSstemTheRCSconsistsoffouridentical heattransferloopsconnected inparalleltothereactorvessel.Eachloopcontainsareactorcoolantpumpandasteamgenerator.

Inaddition, thesystemincludesapressurizer, apressurizer relieftank,interconnecting piping,andinstrumentation necessary foroperational control.Duringoperation, theRCPscirculate pressurized waterthroughthereactorvesselandthefourcoolantloops.Thewater,whichservesbothasacoolant,moderator, andsolventforboricacid(chemical shimcontrol),

isheatedasitpassesthroughthecore.Itthenflowstothesteamgenerators, wheretheheatistransferred tothesteamsystem,andreturnstotheRCPstorepeatthecycle.RCSpressureiscontrolled bytheuseofthepressurizer wherewaterandsteamaremaintained inequilibrium byelectrical heatersandwatersprays.Steamcanbeformed(bytheheaters)toincreaseRCSpressureorcondensed (by.thepressurizer spray)toreducethepressure.

Threespringloadedsafetyvalvesandthreepoweroperatedreliefvalvesareconnected tothepressurizer anddischarge tothepressurizer relieftank,wherethesteamiscondensed andcooledbymixingwithwater.m:51944<w.wpt:

1d4411953.10-1 ChemicalandVolumeControlSstemTheCVCSprovidesforboricacidaddition, chemicaladditions forcorrosion control,reactorcoolantclean-upanddegasification, reactorcoolantmake-up,reprocessing ofwaterletdownfromtheRCS,andRCPsealwaterinjection.

Duringplantoperation, reactorcoolantflowsthroughtheshellsideoftheregenerative heatexchanger andthenthroughaletdownorifice.Theregenerative heatexchanger reducesthetemperature ofthereactorcoolantandtheletdownorificereducesthepressure.

Thecooled,lowpressurewaterleavesthereactorcontainment andenterstheauxiliary building.

Asecondtemperature reduction occursinthetubesideoftheletdownheatexchanger followedbyasecondpressurereduction duetothelowpressureletdownvalve.Afterpassingthroughoneofthemixedbeddemineralizers, whereionicImpurities areremoved,coolantflowsthroughthereactorcoolantfilterandenterstheVolumeControlTank(VCT).EmerenCoreCoolinSstemTheECCSinjectsboratedwaterintothereactorfollowing abreakineitherthereactororsteamsystemsinordertocoolthecoreandpreventanuncontrolled returntocriticality.

Twoeachsafetyinjection pumpsandresidualheatremovalpumpstakesuctionfromtheRWSTanddeliverboratedwatertofourcoldlegconnections viatheaccumulator discharge lines.Inaddition, twocentrifugal chargingpumpstakesuctionfromtheRWSTonSlactuation andprovideflowtotheRCSviaseparateSlconnections oneachcoldleg.Thisarrangement ofSlpumpscanprovidesafetyinjection flowatanyRCSpressureuptothesetpressureofthepressurizer safetyvalves.ResidualHeatRemovalSstemTheRHRSisdesignedtoremovesensibleanddecayheatfromthecoreandreducesthetemperature oftheRCSduringthesecondphaseofplantcoo!down.

Asasecondary

function, theRHRSisusedtotransferrefueling waterbetweentheRWSTandtherefueling cavityatthebeginning andendofrefueling operations.

TheRHRSconsistsoftworesidualheatexchangers, twoRHRpumpsandassociated piping,valvesandinstrumentation.

Duringsystemoperation, coolantflowsfromonehotlegoftheRCStotheRHRpumps,throughthetubesideoftheresidualheatexchangers andbacktotwoRCScoldlegs.Theresidualheatexchangers areoftheshellandU-tubetype.Reactorcoolantcirculates throughthetubes,whilecomponent coolingwatercirculates throughtheshell.SentFuelPoolCoolinSstemTheSFPCSremovesthedecayheatgenerated byspentfuelelementsstoredinthespentfuelpool.Asecondary functionistomaintaintheclarityandpurityofthespentfuelpoolm:119444w.wpf:1d 4411953.10-2 water.TheSFPCSservesthespentfuelpool'which issharedbetweenthetwoCookNuclearPlantunits.Thesystemdesignincorporates twocoolingtrains.EachofthetwocoolingtrainsintheSFPCSconsistsofapump,heatexchanger,

strainer, associated piping,valvesandinstrumentation, andashareddischarge.

Thepumpdrawswaterfromthepool,circulates itthroughtheheatexchanger andreturnsittothepool.Theheatexchangers areoftheshellandU-tubetype;component coolingwatercirculates throughtheshell,andspentfuelpoolwatercirculates throughthetubes.3.10.1.3FluidSystemsEvaluation TheimpactoftheSGTPProgramontheabilityofthefollowing fluidsystemstoperformtheirrequiredfunctions hasbeenevaluated fortheRCS,CVCS,ECCS,RHRS,andSFPCS.ReactorCoolantSstemThecapability oftheRCStooperatewasevaluated attheSGTPProgramconditions listedinTable2.1-1.Thecapacities ofthepressurizer sprayandpoweroperatedreliefvalves,pressurizer surgeline,reliefline,RTDbypassdelaytimesandpressurizer relieftanksetpoints wereevaluated.

Itwasconcluded thatthedesignpressurizer sprayflowrateof750gpmcannotbeachievedatthe30%SGTPProgramconditions.

Thepressurizer sprayflowratewascalculated tobe596gpm.Thiswasdetermined tobeadequatesuchthatthePORVsarenotactuatedfollowing a10%steploaddecreasefromfullpower.ResultsofthisanalysisindicatethatthereducedsprayvalvecapacityisadequatetopreventPORVactuation forthenominaloperating pressures of2100and2250psiaaswellasovertherangeoffullloadaverageRCStemperatures between553'Fand576.3'F.Thepressurizer surgelinepressuredropwasevaluated duringadesignbasissurge.Thedesignbasissurgeresultsfromthreesafetyvalvesrelieving atthedesigncapacity.

Itwasdetermined thattheRCSmaximumpressureatthedischarge oftheRCPis2745psiawhichisbelowtheASMEmaximumallowable pressurefortheRCS.Thepressurizer relieflinepressuredropcalculation wasunaffected bytherevisedNSSSparameters fortheSGTPProgram.Theresultsoftheevaluations showedthattheinstalled PORVcapacityof630,000Ibm/hrisadequateforthedesignbasisloadswingsforoperation atallSGTPoperating conditions.

TheRTDBypassdelaytimecalculations indicatethatthefluidtransport delaytimesfortheexistingpipingnetworkremainbelow1.0secondinallloopsandare,therefore, acceptable.

m:119444w.wpt:1d~1195 3.10-3

Thepressurizer relieftanksetpoints werefoundtobeacceptable.

ThePRTpressurewillbemaintained belowtherupturediscsetpressurefollowing adesignbasisdischarge withthecurrentlevelsetpoints.

ChemicalandVolumeControlSstemTheregenerative andletdownheatexchangers aredesignedtocoolletdownflowfromTto115'F.Thisreduction intemperature isrequiredtoensurethatthenormalRCPsealinjection temperature requirement of130'Fwillbemaintained, including anallowance fora15'Ftemperature riseacrossthecentrifugal chargingpump.Thevariations inT~considered fortheSGTPProgramareboundedbythedesigninlettemperature of547'Ffortheregenerative heatexchanger.

Therefore, thecoolingrequirements oftheletdownfunctionaremetwiththerevisedoperating parametets.

TheletdownfunctionisdesignedtoreducethestaticpressureofthereactorletdownstreamfromtheRCPsuctionpressuretoVCToperating

pressure, suchthatthedesignpressureofintervening pipingandcomponents isnotexceededandfluidismaintained inasubcooled condition throughout thesystem.Themajorityofthepressurereduction istakenacrosstheletdownorifices.

Thepressurecontrolvalve,QRV-301,ensuresthatadequatebackpressureismaintained ontheletdownonficestoensuresubcooled fluidconditions.

Thepressurizer pressures considered (2100or2250psia)areboundedbythedesignpressurizer operating pressure.

Inaddition, ithasbeenverifiedthatQRV-301iscapableofmaintaining sufficient backpressure ontheletdownorificestoensuresubcooled fluidconditions whenthepressurizer pressureisreducedto2100psia.Therefore, thepressurereduction requirements oftheletdownfunctionaremetwiththerevisedoperating parameters.

EmerencCoreCoolinSstemTheprimarysystempressures considered forthisprogramarelessthanorequaltotheprimarysystempressureagainstwhichtheoriginalsystemwasdesignedtodeliver.Therequiredcorecoolingflowrateisproportional toreactorpowerlevelwhichhasnotchangedasaresultofthisprogram.Therefore, therevisedprimarysystemparameters donotrequireanincreaseineitherthemotivepressureorcorecoolingcapacityoftheECCS.ResidualHeatRemovalSstemTheRHRSisnormallyplacedinoperation approximately fourhoursafterreactorshutdownwhenthepressureandtemperature oftheRCSareapproximately 400psigand350'F,respectively.

Undernormaloperating conditions, theRHRSisdesignedtoreducethetemperature ofthereactorcoolantto140'Fwithin20hoursfollowing reactorshutdown, withbothtrainsoperating.

Intheeventofatrainfailure,theRHRSisdesignedtoreducethereactorcoolanttemperature to200'Fwithin36hoursafterreactorshutdown.

Sincetheinitiation temperature anddecayheatgeneration rates(powerlevel)havenotchangedfromm:519444w.wpf:1d441195 3.10-4 thosepreviously evaluated fortheReratingProgram,thedemandsontheRHRSarenotaffected.

Therefore, theRHRSisstillcapableofreducingthereactorcoolanttemperature to140'Fwithinthe20hourlimitfornormaloperating conditions, whenbothtrainsareoperating.

Intheeventofatrainfailure,theRHRSisstillcapableofreducingthereactorcoolanttemperature to200'Fwithinthe36hourlimit.SentFuelPoolCoolinSstemTheprimaryfunctionoftheSFPCSistoremovedecayheatwhichisgenerated bythespentfuelpoolelementsstoredinthepool.Decayheatgeneration isproportional toplantpowerlevel.Sincetheplantpowerlevelof3262MWtremainsunchanged fromthatpreviously evaluated fortheReratingProgram,thedemandsontheSFPCSarenotincreased.

Thepurification functioniscontrolled bySFPCSdemineralization andfiltration rates,whicharenotaffectedbytheSGTPProgram.3.10.2NSSS/Balance ofPlantInterface SystemsEvaluation TheproposedNSSSPerformance Parameters fortheSGTPProgramwerecomparedwiththoseoftheReratingProgram.Theresultsoftheevaluation showthataSGTPlevelof30%willhavenoadverseeffectsontheBalanceOfPlant(BOP)systemsperformance.

TheDonaldC.CookNuclearPlantUnit1BOPfluidsystemsandcomponents havebeenevaluated toassesstheeffectsofincreasing theSGTPlevelupto30%.Theevaluation comparedtheboundingNSSSperformance parameters withthecurrentboundingUnit1ReratingProgramparameters (Cases1,3,4,5,86ofWCAP-11902, Supplement 1)todetermine theimpactonthefollowing BOPsystems:MainSteamSystemCondensate andFeedwater SystemAuxiliary Feedwater SystemSteamGenerator andBlowdownSystemTheproposedperformance parameters whichaffecttheBOPsystemsandcomponents, comparedtotheUnit1ReratingProgramparameters, eitherdonotchange,orchangeinafavorable direction withincreased SGTPlevelsofupto30%.Forexample,theSGTPProgrampowerlevelof3262MWtcorresponds totheminimumpowerlevelevaluated fortheUnit1ReratingProgram.Thefinalfeedwater temperature remainsunchanged aswellasnoloadTavgandsecondary steampressure; Also,thesteammassflowrates areboundedbytheUnit1ReratingProgramparameter.

Onesignificant changeinparameters isthechangeinthefullpowersteampressurewherethelowerboundirig fullpowersteampressure(589psia-Case4ofTable2.1-1)isbelowalowerboundingreratingfullpowersteampressure(603psia-Case4,Table2.1-1ofWCAP11902,Supplement 1).Thiswouldresultinanincreaseinvolumetoc steamflow(cubicfeet/sec) forthesamepowerrating.However,m&19444w.wpf:

1d4411953.10-5 sincetheReratingProgramincorporated apoweruprating(3425MWt)andcorresponding massflowincrease, thevolumetric flowincreaseatthereducedpowerlevel(3262MWt)alsofallswithintheboundsoftheUnit1-Rerating Programparameters.

Consequently.

thechangesinsteam'flow ratesandthedesignconsiderations associated withsteamflowratesarenotsignificant.

Therefore, itwasconcluded thatanincreaseinSGTPlevelsofupto30%willhaveeithernoimpactoraninsignificant impactontheNSSS/BOPfluidsystems.Theywillcontinuetoperformacceptably attheconditions associated withtheSGTPProgram.m:$1944<w.wpf:1d441195 3.10-6 3.11PRIMARYCOMPONENT EVALUATIONS Evaluations wereperformed forallNSSSprimaryandauxiliary components tosupporttheSGTPProgramforCookNuclearPlantUnit1.Insomecases,structural reanalysis wasperformed.

Ingeneral,theevaluations andanalyseswereperformed assumingtheassociated NSSSperformance parameters case(s)(fromTable2.1-1)mostlimitingfortheparticular component.

TheNSSScomponents reviewedfortheSGTPProgramareasfollows:.SectionComponent 3.11.13.11.23.11.33.11.43.11.53.11.63.11.73.11.8SteamGenerators ReactorVesselReactorInternals ControlRodDriveMechanisms ReactorCoolantPumpsPressurizer ReactorCoolantLoopPipingandSupportsAuxiliary Components Asummaryoftheevaluations andanalysesisprovidedbelow.3.11.1SteamGenerators Thefollowing sectionsdescribetheanalysesandevaluations performed undertheCookNuclearPlantUnit1SGTPProgramfortheUnit1SteamGenerators.

TheSteamGenerators evaluated aretheoriginalModel51-series.

Threeseparateareasofevaluation areaddressed fortheSGTPProgram:Thermal-hydraulic performance characteristics (including moistureseparator performance)

U-bendtubefatigueStructural integrity Thermal-H draulicPerformance Evaluation Thefactorsgoverning thethermal-hydraulic performance ofsteamgenerators canbereducedtothethermalpowerandsteampressure.

Otherfactorssuchasprimarytemperature, primaryflowandplugginglevelareimportant onlyinsofarastheyaffectthesteampressure.

Primarypressure, intherangeunderconsideration, doesnotaffectthermalhydraulic performance.

m:519444w.wpf:1d441195 3.11-1 II' AspartoftheReratingProgram,thermalhydraulic performance parameters wereevaluated forarangeofthermalpowersandsteampressures.

ThepressurerangeisboundedbytheReratingProgramwhehallpowersareconsidered.

Atthe3262MWtpowerrating,applicable tothe30%pluggingevaluation, theloweststeampressureisslightlybelowthevalueanalyzedduringtheReratingProgram.Thiswillbeshowntobeofnoconsequence.

Theconclusion oftheReratingProgramwasthattheperformance characteristics ofthesteamgenerators, including moisturecarryover, continuetobeacceptable atalltheReratingProgramconditions.

Thisconclusion continues toapplyforthe30%SGTPconditions.

MoistureSeparator LimitsModifications tothemoistureseparators atUnit1werecompleted intheSpringof1989.Thesemodifications includethefollowing elements:

~primaryseparator "tophats"whichdiffusethejetissuingfromtheprimaryseparators, "steamchimneys" whichventsteamfrombelowthemiddeckplatewithoutentraining liquiddrops,andadditional uppertierdryerdrains.Withthesemodifications, moisturecarryover valuesmeasuredinthefieldhavebeennearorbelow0.1%overawiderangeofpowerlevelsandsteampressures.

Inearlierseparator systems,including theunmodified Model51,moisturestronglyincreased withpower.Thetophatseliminate aprimarycauseofthisdependence whichisthejetsissuingfromtheprimaryseparators.

Thetrendofincreasing moisturewithdecreasing steampressureremains,butitseffectissmallandthemoisturelevelremainslowtothelowestlimitofthedata,700psia.Basedontheavailable fielddata,moisturecarryover isexpectedtoremaincomfortably below0.25%forsteampressures downto700psiandbelow.OtherThermalHydraulic Characteristics Inadditiontomoisturecarryover, theReratingProgramevaluated circulation ratio,hydrodynamic stability, andsteamgenerator massasadditional indicators ofacceptable performance.

Thechangeintheseparameters fromthedesignvaluetothevaluesateachoftheReratingProgramconditions werecalculated.

Steampressures analyzedfortheratingof3262MWthadarangeof610to820psia.Variation ofthethreeparameters overthisrangeispresented inTable3.11-1alongwiththedesignvalueat<heratedpower.ItisevidentfromTable3.11-1thattheparameters listedareminimally affectedbysteampressureatconstantpower.Circulation ratioisessentially unaffected.

Dampingfactoristhem:hl944+w.wpf:1dM1195 3.11-2 measureofhydrodynamic stability, alargenegativevalueindicating astableunit.Thisparameter, too,isessentially unaffected bythesteampressure.

Steamgenerator massisslightlyaffectedbyreducedsteampressure.

Assteampressuredecreases, thevoidsinthebundleincreasereducingthemassinventory.

Theeffectissmallanddoesnotaffectoperability.

Table3.11-1displaystheparameter variation downtoasteampressureof610psia.Forthe30%pluggingconditions, theminimumsteampressureis589psia.The21psipressurechangerepresented inthetablewasshowntohaveminimalaffectontheparameters reviewed.

Theadditional 21psipressurechangeto589willalsobesmall.Steamgenerator operating characteristics willbeacceptable downtotheminimumsteampressureof589psia.U-bendFatiueEvaluation AcompleteU-bendfatigueevaluation isdocumented inWCAP-13814, December1993,"D.C.CookUnit1-Evaluation forTubeVibration InducedFatigue"(Reference 1).Theevaluation wasperformed todetermine thesusceptibility tofatigue-induced

cracking, consistent withNRCBulletin88-02.Theevaluations wereperformed forthecurrentoperating conditions aswellasforalevelof30%tubeplugging.

Theanalysisidentified preventative actionsfortubesidentified aspotentially susceptible toU-bendvibration inducedfatigue.Structural InteriEvaluation Structural integrity evaluation ofsteamgenerator components performed fortheDonaldC.CookNuclearPlantUnit1ReratingProgramincludedNSSSperformance parameter casesthatboundedsteamgenerator tubeplugging(SGTP)levelsofupto15%.TheNSSSdesigntransients developed fortheReratingProgramcontinuetoapplytoDonaldC.CookNuclearPlantUnit1atthe30%SGTPconditions.

Sincetheperformance parameters andthedesigntransients stillapply,theevaluations performed for15%SGTPwouldbeapplicable forallcomponents ofthesteamgenerator exceptthedividerplate.Anewevaluation ofthedividerplate,therefore, wasperformed forahigherpressuredifferential acrossthedividerplatecausedbyahigher(30%)tube-plugging level.Thisanalysisdemonstrated thestructural acceptability ofthedividerplate.Itwastherefore concluded thattheCookNuclearPlantUnit1steamgenerator structural integrity wouldbemaintained foroperations withatube-plugging levelofupto30%.Reference WCAP-13814, "DonaldC.CookUnit1Evaluatiorilor TubeVibration InducedFatigue,"

December1993mal944+w.wpf:1d 4411953.11-3

3.11.2ReactorVessel3.11.2.1Introduction" TheAddendumReportpreparedfortheUnit1reactorvessel(reference 1)toevaluatethestressandfatigueeffectsoftheoperating parameters andRCStransients associated withtheReratingProgramremainsapplicable.

Therefore, nonewreactorvesselstresscalculations wereperformed fortheSGTPProgram.Thisreportevaluates themaximumprimaryplussecondaiy stressintensity rangesandmaximumcumulative fatigueusagefactorsresulting fromtheReratingProgramconditions whichboundthe30%SGTPconditions.

Thecalculated stressintensity rangeandusagefactorvaluesarecomparedwiththeapplicable limitsofSectionIIIoftheASMEBoilerandPressureVesselCode.Theoperating parameters showninTable3.11-2wereusedasabasisfortheevaluation.

Theseparameters boundthe30%SGTPparameters contained inTable2.1-1ofthisreport.3.11.2.2SummaryofResultsTheresultsofthereactorvesselanalysesandevaluations aresummarized below.Basedontheseresults,allofthestressintensity andfatigueusagelimits(withtheexception ofthe3Smaximumrangeofprimaryplussecondary stressintensity limitforthecontrolroddrivemechanism housingsandoutletnozzlesafeend)oftheapplicable ASMECodeversionforUnit1(reference 2)aremet.Exceeding the3SlimitfortheCRDMhousingsandoutletnozzlesafeendisreconciled bysimplified elastic-plastic analysesinaccordance withreference 3.Therefore, thereactorvesselforCookNuclearPlantUnit1continues toremainincompliance withtheapplicable Codefortheconditions associated withtheReratingandSGTPPrograms.

ControlRodDriveMechanism HousinAdaterThemaximumrangeofprimaryplussecondary stressintensity iscalculated tobe77.76ksiwhichexceedsthe3Slimitof69.9ksi.However,asimplified elastic-plastic analysiswasperformed inaccordance withparagraph NB-3228.3 ofthe1971EditionofSectionIIIoftheASMEBoilerandPressureVesselCode,andthehigherrangeofstressintensity isreconciled.

Themaximumcumulative fatigueusagefactoris0.1687whichisbelowtheASMECodelimitof1.0.MainClosureReionThemainclosureregionofthereactorvesselconsistsof.thevesselflange,theclosureheadflangeandtheclosurestudassemblies whichcoupletheheadtothevessel.Themaximumrangesofstressintensity intheclosureheadflangeandthevesselflangeare65.26ksiandmA1944<w.wpf:1d~1195 3.11-4 61.04ksi,respectively, comparedtoanASMECode3Slimitof80.1ksi.Themaximumserviceintheclosurestudsis91.8ksiwhichcomparesfavorably tothe3Slimit107.7ksi.Themaximumcumulative fatigueusagefactorfortheclosureheadflange,vesselflangeandclosurestudsare0.018,0.029and0.99,respectively.

Theusagefactorsarealllessthanthe1.0ASMECodelimit.However,itshouldbenotedthattheclosurestudusagefactorof0.99wascalculated undertheassumption thatthefirst25percentofthe11,680occurrences ofplarltloadingandplantunloading at5percentoffullpowerperminute(2,920occurrences ofeach)occurredduringthefirst10yearsofoperation whenthevesseloutlettemperature (T)was599.3'F.Ifthe0.99usagefactorisunacceptably highorifcyclecountingindicates that1.00maybeexceeded, theclosurestudsarereadilyreplaceable.

OutletNozzleThemaximumrangeofprimaryplussecondary stressintensity intheoutletnozzlesafeendiscalculated tobe59.58ksicomparedtothe3Slimitfortheaustenitic stainless steelmaterialof50.1ksi.Sincethemaximumrangeexceeds3S,asimplified elastic-plastic analysisperparagraph NB-3228.3 ofthe1971EditionofSectionIIIoftheASMEBoilerandPressureVesselCodewasperformed whichjustified thehighermaximumrangeofstressintensity.

Themaximumusagefactoratthesafeendis0.021whichislessthan1.0.Themaximumrangeofstressintensity intheoutletnozzleandnozzle-to-shell junctureis57.09ksicomparedtothe3Sallowable of80.1ksi.Themaximumcumulative usagefactorinthenozzleandnozzle-to-shell junctureis0.0631whichisalsolessthan1.0.InletNozzleThemaximumrangeofstressintensity intheinletnozzlesafeendis49.65ksiwhichislessthan3S=50.1ksi.Themaximumrangeofstressintensity intheinletnozzleandnozzle-to-shell junctureis49.86ksiwhichcomparesfavorably witha3Slimitof80.1ksi.Themaximumcumulative usagefactorsinthenozzlesafeendandnozzle-to-shell junctureare0.0174and0.0977,respectively, whicharebothlessthan1.0.VesselWallTransition Themaximumrangeofstressintensity andcumulative fatigueusagefactorforthevesselwalltransitionbetween thenozzleshellandthevesselbeltline, are33.57ksiand0.0066.ThesevaluesarelessthantheASMECodelimitsof80.1ksiand1.0,respectively.

BottomHead-to-Shell JunctureThemaximumrangeofprimaryplussecondary stressintensity atthejuncturebetweenthevesselbottomhemispherical headandthevesselbeltlineshellis34.53ksicomparedtoa3Sm:i1944+w.wpi:

id~i1953.11-5 allowable of80.1ksi.Themaximumcumulative fatigueusagefactoratthejuncturewascalculated tobe0.0182whichislessthan1.0.BottomHeadInstrumentation Penetrations Thebottomheadinstrumentation penetrations areacceptable fortheSGTPProgrambaseduponamaximumrangeofprimaryplussecondary stressintensity of51.49ksiandamaximumcumulative fatigueusagefactorof0.1220.Thesevaluescomparefavorably withASMECodeallowables of69.9ksi(3S)and1.0,respectively.

CoreSuortPadsThecoresupportpadswereevaluated tohaveamaximumrangeofstressintensity of69.70ksicomparedtoa3Slimitof69.9ksi.Themaximumcumulative fatigueusagefactorwascalculated tobe0.693whichislessthanthe1.0ASMECodelimit.3.11.2.3Conclusions Theresultsoftheevaluations demonstrate thatoperation ofthereactorvesselinaccordance withtheconditions associated withthe30%SGTPProgramdoesnotresultinstressintensities orfatigueusagefactorswhichexceedtheacceptance criteriaoftheapplicable ASMECodeversionforCookNuclearPlantUnit1(reference 2).Someofthestressintensity rangesarehigherthantheoriginalstressreport.However,allofthestressintensity limitsspecified intheapplicable ASMECodeversionarestillsatisfied withtheincorporation of30%SGTPconditions, withtheexception ofthe3Smaximumstressintensity rangelimitfortheCRDMhousingsandoutletnozzlesafeends.Exceeding 3SintheCRDMhousingsandoutletnozzlesafeendsisreconciled bysimplified elastic-plastic analysesinaccordance withtherequirements ofparagraph NB-3228.3 ofthe1971EditionofSectionIIIoftheASMECode(reference 3).3.11.2.4ReactorVesselIntegrity Evaluation The30%SGTPconditions forDonaldC.CookNuclearPlantUnit1willnotresultinanincreaseinthefastneutronfluencevaluescalculated fortheReratingProgram.Basedonthisinformation, thereactorvesselintegrity analysesperformed perthemethodology ofRegulatory Guide1.99,Revision2,aspartoftheReratingProgramwillremainapplicable after30%SGTP.Theseanalysesareapplicable forreactorvesselinlettemperatures (Tcold)above525'F.Operation below525'Fdownto510'Fhasbeenfurtherevaluated andfoundtobeacceptable (Reference 4).TheincreaseinSGTPto30%doesnotimpactthetechnical supportprovidedinReference 4foroperation below525F-.m:El944<w.wpf:1d441195 3.11-6 3.11.2.5References 1.WCAP-11967, "TReduction/Rerating ReactorVesselEvaluation AddendumtoAnalytical ReportforIndianaandMichiganElectricCompany.DonaldCookNuclearPowerPlantUnitNo.1ReactorVessel",August,1988,byS.L.Abbott.2.ASMEBoilerandPressureVesselCodeSectionIII,AmericanSocietyofMechanical Engineers, NewYork."NuclearVessels",

1965EditionwithAddendathroughtheWinterof1966.3.ASMEBoilerandPressureVesselCode,SectionIII,AmericanSocietyofMechanical Engineers, NewYork."NuclearPowerPlantComponents",

1971Edition.4.AEP-93-582, "American ElectricPowerServiceCorporation, DonaldC.CookNuclearPlantUnit1,Operation Below525degreeF",KeithF.MatthewstoJohnJensen,October18,1993.m:41944+w.wpf:

1d4411953.11-7 3.11.3ReactorInternals

\3.11.3.1Introduction Thissectiondocuments theresultsandconclusions oftheevaluations performed toinvestigate theimpactoftheSGTPProgramontheCookNuclearPlantUnit1reactorvesselinternals.

InordertoassesstheimpactoftheSGTPProgram,thefollowing evaluations wereperformed.

~ReviewandEvaluation ofThermalTransients

~ReviewandEvaluation ofPowerLevelThermal-Hydraulic Analysis-Theanalysesincluded:

Evaluation oftheeffectsoncorebypassflowPressuredropdistribution inthereactorvesselComponent hydraulic liftforcesMechanical SystemEvaluations whichinclude:Asymmetric flowevaluation Flowinducedvibrational

~Components ThermalStressandFatigueEvaluations 3.11.3.2ThermalTransients andPowerLevelReviewThermalTransients PerSection2,2,thethermaltransients andnumberofoccurrences usedintheReratingProgramremainunchanged; therefore thethermaltransients evaluation fortheReratingProgramremainapplicable.

PowerLevelThepowerlevelfortheSGTPProgramperTable2.1-1is3250MWtreactorpowerwhichistheoriginaldesignbasisforCookNuclearPlantUnit1.TheReratingProgramusedapowerof3588MWtreactorpower.Achangeinpowerlevelwillaffectthethermalloadsonvariousreactorinternals components suchas:LowerCoreSupportStructure Baffle-Barrel RegionThermalShieldm:519444w.wpf:

1d4411953.11-8 Thedecreaseinpowerlevelfrom3588MWtto3250MWtwillnothaveanadverseeffectonthestructural evaluation performed fortheabovecomponents.

3.11.3.3Thermal-Hydraulic AnalysesThermal-hydraulic

analyses, aspartofthereactorinternals qualification fortheSGTPProgramwereperformed.

Thethermal-hydraulic analysesinputparameters weretakenfromTable2.1-1.Fourdifferent conditions wereevaluated usingtheinputparameters forCases2and3fromTable2.1-1(Case2O2250psi,Case2O2100psi,Case3O2250psi,andCase3O2100psi).Themostconservative resultsforpressuredrops,liftforces,andcorebypassflowwasobtainedusingCase2inputparameters O2250psifromTable2.1-1.CoreBassFlowCorebypassflowisdefinedasthetotalamountofreactorcoolantflowthatbypassesthecoreregion,andisnotconsidered effective inthecoreheattransferprocess.Consequently, theeffectofincreasing bypassflowisareduction incorepowercapability.

Evaluations showthattheinputparameters fromCase2ofTable2.1-1,provideresultswiththehighesttotalbypassflowof4.4%.Theresulting totalcorebypassflowisstillwithintheallowable limitof4.5%specified fortheSGTPProgram.SstemPressureLossesandHdraulicLiftForcesTheReratingProgramusedaconservative evaluation forthesystempressurelosses,andhydraulic liftforces.Thesystempressurelossesandhydraulic liftforcesfortheSGTPProgramareconsidered boundedbytheReratingProgram.Theinputparameters, whichinfluence thesystempressurelossesandhydraulic liftforcesfortheSGTPProgramarethesameorlowerthanthoseusedfortheReratingProgram.TheSGTPparameters willyieldsystempressurelossesandhydraulic liftforceswhichareboundedbythoseusedintheReratingProgram.Theonlyparameter whichwouldincreasethesystempressurelossesandhydraulic liftforcesisthechangeinpowerlevel,andthechangeinpowerlevelisconsidered tohaveaninsignificant effectfortheseparameters.

Therefore, theSGTPProgramdoesnothaveanadverseeffectonthesystempressurelossesandhydraulic liftforces.3.11.3.4Mechanical SystemEvaluation FlowInducedVibration Theparameters whichcaninfluence theflowinducedvibration characteristics ofthereactorvesselinternals istheflowandtemperature.

Themechanical designflowfortheSGTPProgramarenotchanging, onlythethermaldesignflowischanginganditisdecreasing.

Themechanical designflowisunchanged andthetemperature rangeisenveloped bythetemperature rangeintheReratingProgram.Therefore, flowinducedvibration willnotbem:519444w.wpf:1d441195 3.11-9 adversely impactedbytheSGTPProgramsincetheflowinducedvibration loadingsareenveloped bytheworkperformed fortheReratingProgram.AsmmetricFlowEvaluation Theeffectofasymmetric flowonthereactorvesselinternals wasevaluated.

Theasymmetric flowloadsfromTable2.1-1wereusedtoevaluatetheeffectofthenewflowcondition.

Testdatafromanotherplantwasalsousedsincetheinternals forCookNuclearPlantUnit1aresimilartotheinternals ofthatplant.Theevaluation concluded thatthemaximumdisplacements forthecorebarrelbeammoden=1,andshellmodesn=2andn=3wereenveloped bythetestdata.Therefore, theassymmetric flowcondition isconsidered acceptable.

3.11.3.5Component Evaluation ReactorInternals Thermal/Stress andFatiueEvaluation Thereactorinternals thermal/stress andfatigueevaluation wasperformed byusingtheReratingProgramevaluation asthelastqualified operating conditions andevaluating thechangeinloadingsduetotheSGTPProgramonreactorinternals.

Loadingswhichcanimpacttheevaluation performed are:ThermalTransients andtheNumberofOccurrences PowerLevelGammaHeatingRatesMechanical LoadingsFlowRatesSeismicLoads(OBE)Operating Temperature Thenewloadingsforthereactorvesselinternals areevaluated inthefollowing section.LoadEvaluation TheNSSSdesigntransients, Section2.2,fortheSGTPProgramremainthesameasthosepreviously analyzedfortheReratingProgram.Seismicloads,mechanical loads,andgammaheatingratesarenotaffectedbysteamgenerator tubeplugging.

Theoperating temperatures fromTable2.1-1werechosensuchthattheywouldbeenveloped bytheoperating temperatures usedintheReratingProgram.Therefore, forthisevaluation, powerlevelandflowratesaretheonlyparameters whicharechangingthafcanaffectthereactorvesselinternals evaluation.

Thereduction inpowerlevel,from3588MWtto3250MWt,willdecreasethegammaheatinglevelsforvariousreactorvesselcomponents.

Alowergammaheatingvaluewillcausethemetaltemperature duetogammaheatingtodecrease, whichwillbringm:$1944+w.wpt.'1d~1 1953.11-10

themetaltemperature closertothefluidtemperature.

Thiswillcauseasmallerthermalgradientonthevariousportionsofthereactorvesselinternals whichareaffectedbygammaheating.Thesmallerthermalgradientwillcausealowerstressintheaffectedcomponents.

Sincethereducedpowerlevelwillresultinalowerstressstate,thepreviousevaluation fortheReratingProgramisconsidered boundingforthermalstressandfatigue.Thereduction inthermaldesignflowisapproximately 6%forthenormaloperating conditions.

Thetemperature rangeforthe30%SGTParewithintherangeoftemperatures evaluated andareconsidered boundedbytheReratingProgram.Sincethereisonlyasmallchangeinflowrate,thiswillcausethepressuredropinthereactorvesseltodecrease.

Thedecreased pressuredropwillresultinasmallercalculated stresslevelforvariouscomponents, (corebarrelandbaffle-former plates).Thereduction inThermalDesignFlowtranslates a5%orlessreduction intheforcedconvection heattransfercoefficients.

A5%reduction inthefilmcoefficients isnotexpectedtosignificantly affecttheheatconvection betweenthereactorvesselinternals andthereactorcoolant.Therefore, thethermalevaluation performed fortheReratingProgramisconsidered applicable.

Conclusion Thereactorvesselinternals stressandfatigueevaluation isconsidered boundedbytheReratingProgramevaluation.

TheSGTPProgramdoesnothaveanadverseeffectonthereactorvesselinternals sincetheloadswhicharechangingareactuallyimproving themarginsforthereactorvesselinternals whencomparedtotheReratingProgramresults.RodDroTimeAnassessment wasmadetoconfirmthepresentRCCAdroptimelimitof2.4secondsremainsapplicable withtheSGTPProgramconditions.

Basedontheanalysisperformed, itisconcluded thatthe2.4secondRCCAdroptimeremainsapplicable.

3.11.4ControlRodDriveMechanisms Anevaluation wasperformed toevaluatetheeffectsof30%SGTPforDonaldC.CookNuclearPlantUnit1.AreviewoftheNSSSPerformance Parameters, giveninTable2.1-1,showsthattheseconditions areboundedandhavebeenevaluated.

SincetheNSSSPerformance Parameters andtheNSSSDesignTransients fortheSGTPProgramareboundedbythoseoftheReratingProgram,theconclusion ofthegenericanalysisperformed fortheReratingProgramremainsvalidfortheSGTPProgram.Varyingthehotlegreactorcoolanttemperature willhavenoimpactonthestructural andthermalanalysisoftheCRDM.Varyingthehotlegtemperature willaffectcertainCRDMmaterialproperties.

However,theeffectwillbeinsignificant.

Sincethehotlegtemperature rangefortheSGTPProgramiswithintheboundsofthehotlegtemperature rangeforthem:'11944+w.wpf:1d441 1953.11-11 ReratingProgram,theReratingProgramanalysiscontinues toapplyandthedesignrequirements fortheCROMpressureboundaryarestillmet.3.11.5ReactorCoolantPumpsandMotorsTheModel93Areactorcoolantpumps(RCPs)usedintheCookNuclearPlantUnit1NSSSwerereviewedtodetermine theimpactoftheNSSSparameters fortheSGTPProgramprovidedinTable2.1-1ofthisreport.BecausetheNSSSparameters fortheSGTPProgramareboundedbythoseoftheReratingProgramandtheNSSSdesigntransients arealsoboundedbytheReratingProgram,noadditional thermalorstructural analysiswererequiredtodemonstrate compliance withthecodesandstandards ineffectatthetimeoftheoriginalcontract.

Varyingthecoldlegtemperature from536.3'Fupto543.2'Fordownto517.2'Fwillhavenoimpactonthestructural andthermalanalysisofthereactorcoolantpump.Theonlydifference isrelatedtothematerialproperties usedintheanalysis.

Thedifference inmaterialproperties atthesubjecttemperature isconsidered negligible.

Therefore, thedesignrequirements fortheRCPpressureboundaryarestillmet.TheRCPmotorswereevaluated todetermine theworstcaseloading.Theperformance oftthemotorsattheseloadshasbeenevaluated andtheresultsareasfollows:1.Continuous operation atthenewhotloopratingat6420HP.Thisrepresents a7.0%increaseoverthenameplate ratingofthemotor.Thechangeinstatorwindingtemperature resulting fromtheincreasewillbelessthan5'C.Originaltestdataindicates thatwiththistemperature increaseincluded, theNEMAdesignlimitsforaClassBwindingwillnotbeexceeded.

Therefore, continuous operation ofthemotorsunderhotloopconditions with30%SGTPisacceptable.

2.Operation atthenewcoldloopratingof8020HP.Therevisedloadrepresents a6.9%increaseoverthenameplate ratingofthemotor.Analysisindicates thisloadincreasewillcausethestatorwindingtemperature toincreaseabout7'C.Theresulting windingtemperature willbelessthantheClassFNEMAdesignlimits.Therefore, operation ofthemotorsundercoldloopconditions with30%SGTPisacceptable.

mA1944<w.wpf:1d441195 3.11-12 3.Startingwithrevisedloadtorqueundertheworstcaseconditions (maximumreverseflow,coldloop,80%voltage).

Theincreaseinrotorcagewindingtemperature duetotheincreased loadissmallandthetotalwindingtemperature iswellbelowthedesignlimit.Therefore, startingundertheworstcasescenarioisacceptable.

ThereviewfortheSGTPProgramoftheCookNuclearPlantUnit1reactorcoolantpumpsdemonstrates thattheSGTPconditions areacceptable fortheModel93ARCP.Thedesignrequirements oftheRCPpressureboundaryarestillmet.TheRCPMotorevaluation determined thattheCookNuclearPlantUnit1motorsareacceptable foroperation withthe30%SGTPconditions.

3.11.6Pressurizer 3.11.6.1Introduction Thefunctions ofthepressurizer aretoabsorbanyexpansion orcontraction oftheprimaryreactorcoolantduetochangesintemperature andpressureandtokeeptheRCSatthedesiredpressure.

Thefirstfunctionisaccomplished bykeepingthepressurizer approximately halffullofwaterandhalffullofsteamatnormalconditions, connecting thepressurizer totheRCSatthehotlegofoneofthereactorcoolantloopsandallowinginfloworoutflowtoorfromthepressurizer asrequired.

Thesecondfunctionisaccomplished bykeepingthetemperature inthepressurizer atthewatersaturation temperature (T,,)corresponding tothedesiredpressure.

Thetemperature ofthewaterandsteam'nthepressurizer canberaisedbyoperating electricheatersatthebottomofthepressurizer andcanbeloweredbyintroducing relatively coolwatersprayintothesteamspaceatthetopofthepressurizer.

Thelimitinglocations fromastructural standpoint onthepressurizer arethesurgenozzle,thespraynozzle,andtheuppershellatthepointofsprayimpingement.

Thelimitingoperating condition (relative totheSGTPconditions) ofthepressurizer occurswhentheRCSpressureishighandtheRCShotlegtemperature (T>>)andcoldlegtemperature (T~,)arelow.Thisisexplained asfollows:Duetoinflowandoutflowtoandfromthepressurizer duringvarioustransients thesurgenozzlealternately seeswateratthepressurizer temperature (T,,)andwaterfromtheRCShotlegatT>>.IftheRCSpressureishigh(whichmeansthatT,,ishigh)andT>>islow,thenthesurgenozzlewillseemaximumthermalgradients andthusexperience themaximumthermalstress.Likewisethespraynozzleanduppershelltemperatures alternate betweensteamatT~,andspraywhichformanytransients isatT~p.Thus,ifRCSpressureishigh(T,,ishigh)andT~,islow,thenthespraynozzleanduppershellwillalsoexperience themaximumthermalgradients andthermalstresses.

m%1944<w.wpf:1d441195 3.11-13 3.11.6.2Description ofAnalysisandResultsTheupdatedanalysisperformed for.theDonaldC.CookNuclearPlantUnit1SGTPProgramforthepressurizer isbasedontheNSSSdesigntransients providedfortheReratingProgram.Thedesigntransients arealsoapplicable fortheSGTPProgram(seeSection2.2).Theanalysiswasperformed bymodifying theoriginalCookPressurizer analysis(Reference 1),whichwasperformed totherequirements oftheASMECode,1968Edition(Reference 2).Theoriginalanalysiswasperformed usingfiniteelementtechniques.

Finiteelementmodelswereconstructed forthevariouspartsofthepressurizer.

Thesewerethensubjected tothepressureloads,externalloads(suchaspipingloadsonthenozzles)andthermaltransients.

Themodelsthencalculate theprimary,secondary andpeakstressesforthevariousconditions.

The,pressurizer maximumpressureandmaximumexternalloadsdidnotincreaseduetotheSGTPProgram.Thus,theprimarystressesfromtheoriginalanalysisarestillvalid.Also,theconditions thatcausemaximumprimaryplussecondary stress(inadvertent auxiliary sprayforspraynozzleanduppershell,andDBEforthesurgenozzle)havenotchanged.Therefore, theonlyASMECoderequirement affectedbythetransient modifications wasfatigue.ThefatigueusagefactorsareshowninTable3.11-3forthecriticalcomponents.

3.11.6.3Conclusions Afatigueanalysiswasperformed fortheCookNuclearPlantUnit1pressurizer, incorporating themostconservative conditions oftheSGTPProgram.Theresultsofthisanalysisdemonstrate thatthepressurizer remainsincompliance withtheapplicable ASMECodecriteria.

3.11.6.4References 1.Model51SeriesPressurizer Report,Westinghouse ElectricCorporation, October1974.2.ASMEBoilerandPressureVesselCode,1968Edition,SectionIII,Article4.mA19444w.wpt:1d441195 3.11-14 3.11.7ReactorCoolantLoopPipingandSupportsAspartoftheReratingProgram,anevaluation ofthereactorcoolantlooppiping,primaryequipment nozzles,andtheprimaryequipment supportswasperformed forasetofthermalparameter casesthatincluded15%SGtubeplugging.

Theprogramreportedresultsforareratingandalsoaddressed anumberofadditional casestocovervarioustemperatures inthelooppiping.Inthatreporttheanalysisforthelooppiping,theprimaryequipment nozzles,andtheprimaryequipment supportswerereconciled totheReratingProgramaswellastheSGTPProgramconditions.

The30%SGTPlooppipingtemperatures areenveloped onboththelowerandupperendsbythelooppipingtemperatures alreadyconsidered intheReratingProgram.TheLOCAhydraulic forcingfunctions generated fortheReratingProgramboundtheproposed30%SGTPconditions.

TheNSSSthermaldesigntransients areapplicable forboththeReratingandSGTPProgramconditions.

TheReratingProgramtransients andtheplantparameters associated withboththeReratingandtheSGTPProgramswerereviewedforimpactontheWCAP-14070 (Reference 1)evaluation forNRCBulletin88-08,"ThermalStressesinPipingConnected totheReactorCoolantSystem".TheWCAPspecifically addressed theauxiliary spraypiping.Thedefinedtransients areprimarylooppipingtransients andarefarenoughremovedfromtheauxiliary spraypipingtohaveanegligible impact.Theoperating parameters forboththeReratingandtheSGTPProgramconditions havenormaloperating coldlegtemperatures thatdeviatefromtheexistingdesignbasisvalues.Therangeofnormaloperating coldlegtemperatures havebeenreviewedforimpactontheNRCBulletin88-08evaluation andwerefoundtohavenoimpactontheconclusions statedinWCAP-14070.

TheReratingProgramtransients forCookNuclearPlantUnit1werereviewedforpotential impactontheexistingevaluation forthepressurizer surgelinethermalstratification analysis.

Thereportthatwaspreparedtodemonstrate compliance withNRCBulletin88-11,"Pressurizer SurgeLineThermalStratification" isWCAP-12850 (Reference 2).Thereconciliation ofthereferenced transients appliestoboththeReratingProgramandtheSGTPProgrambecausethetransients coverbothprograms.

Theresultsoftheevaluation indicatethatthefatigueusagefactorincreases byasmallamount(from0.275to0.277).SincethemaximumusagefactorreportedintheWCAPwasroundedtoavalueof0.30,theresultdoesnotchange.Becausetheallowable fatigueusagefactoris1.0,theresultsareacceptable.

AspartofthesurgelinestratiTication

analysis, asetofpressurizer nozzleloadingsduetostratification wasusedasinputtothepressurizer evaluation.

Ourevaluation showsthatthechangestothepressurizer nozzleloadingsarenotsignificant, andneednotbeevaluated further(therearenoincreases greaterthat2%andloaddecreases wereignored).

Inconclusion, thereactorcoolantlooppiping,theprimaryequipment nozzles,andtheprimaryequipment supportloadsareacceptable fortheSGTPProgramconditions becausetheseconditions arealreadyenveloped bytheevaluation performed fortheReratingProgram.Alldesignbasisanalysisperformed forthesecomponents appliestothe30%SGTPcondition.

m%19444w.wpf:1d441195 3.11-15 II TheReratingProgramtransients andplantparameters associated withtheReratingandtheSGTPProgramsforDonaldC.CookNuclearPlanthavebeenreviewed, andtheimpactonthedesignbasisanalysisfortheNRCBulletin88-08evaluation oftheauxiliary spraypipingandNRCBulletin88-11evaluation ofthepressurizer surgelinepipingisinsignificant.

References WCAP-14070, "Evaluation ofDonaldC.CookUnits1and2Auxiliary SprayPipingperNRCBulletin88-08,"July1994.2.WCAP-12850, "Structural Evaluation ofDonaldC.CookNuclearPlantUnits1and2Pressurizer SurgeLines,Considering theEffectsofThermalStratification,"

January1991.3.11.8Auxiliary Components Theauxiliary components (pumps,valves,tanksandheatexchangers) werereviewedtodetermine theimpactoftheNSSSparameters fortheSGTPProgram,providedinTable2.1-1ofthisreport.BecausetheNSSSparameters oftheSGTPProgramareboundedbythoseoftheReratingProgramandtheAuxiliary Equipment Transients areeitherunchanged orstillbounded,thereisnoeffectontheauxiliary components ofCookNuclearPlantUnit1.m:11944<w.wpt:1d441 1953.11-16 TABLE3.11-1PERFORMANCE CHARACTERISTICS AT3262MWTParameter DesignValueMaximumSteamPressureMinimumSteamPressureSteamPressure(psia)Circulation RatioDampingFactor(hr')Secondary Mass(Ibmx10')8125.49-4531138205.45-4341136105.42-482108mh1944<w.wpf:1d~1195 3.11-17 II' TABLE3.11-2ASSUMEDOPERATING PARAMETERS FORREACTORVESSELSTRUCTURAL EVALUATION FORCOOKNUCLEARPLANTUNIT1DesignPressure(psig)NormalOperating Pressure(psig)2485UpperBoundLowerBound22351985DesignTemperature

('F)NormalOperating VesselInletTemperature

('F)NormalOperating VesselOutletTemperature

('F)650511.7615.2ZeroLoadTemperature,

('F)547-DesignLife:Thedesignlifeofthereactorvesselis40years.Thedesignlifeistheperiodofanticipated plantservicewhichisusedasabasisfordefiningthenumberofoccurrences ofdesigntransients andexternalloadstobeusedinthedesignfatigueanalysis.

Thedesignlifeisnottobeconsidered asawarrantybutisusedstrictlyfordetermining fatigueusagefactorsforthereactorvesselcomponents.

Thereactorvesselisanalyzedtooperatewithnormaloperating vesselinlettemperatures (T~)from511.7'Fto547'Fandnormaloperating vesseloutlettemperatures (T)from582.3'Fto615.2'F.Thereactorvesselclosurestudswereanalyzedforfatigueusageassuminganormalvesseloutlettemperature of599.3'Fforthefirst10yearsofoperation andthemaximumnormalvesseloutlettemperature of615.2'Ffortheremaining 30years.mA19444w.wpf:1d441195 3.11-18 TABLE3.11-3DONALDC.COOK1PRESSURIZER COMPONENTS, CALCULATED FATIGUEUSAGESCONSIDERING 30%SGTPCOMPONENT FATIGUEUSAGESurgeNozzleSprayNozzleSafetyandReliefNozzleSeismicAnalysisLowerHead-HeaterWellLowerHeadPerforations UpperHeadandShellSupportSkirt/Flange HeaterVibrations BaffleVibrations SupportLugManwayInstrument NozzleImmersion HeaterValveSupportBracket0.33230.990.1480.070.01650.970.0110.0480.00.10840.0040.01mal944+w.wpf:1d441195 3.11-19 3.12FUELSTRUCTURAL EVALUATION Evaluations wereperformed ofthefuelforCookNuclearPlantUnit1undertheReratingProgramintheareasfuelrodandfuelassemblystructural integrity, coredesignandthermal-hydraulic design.Theseevaluations assumedamaximumcorepowerlevelof3250MWtandtheassociated rangeofoperating conditions fromTable2.1-1.3.12.1FuelAssemblyStructural Evaluation Fuelassemblies aredesignedtoperformasdescribed intheTechnical Specifications.

Thecombinedeffectsofdesignbasisloadsareconsidered intheverification ofthefuelassemblyanditscomponents tomaintainthefuelassemblystructural integrity.

Thisisnecessary sothatthefuelassemblyfunctional requirements aremet,thecoreeoolablegeometryismaintained, andthereactorcorecanbeshutdownsafely.Astructural evaluation ofthefuelassemblywasperformed fortheSGTPProgramforCookNuclearPlantUnit1,considering therangeofoperating parameters described inTable2.1-1.Thisevaluation assumed15x15optimized fuelforUnit1:ThesummaryofthemaximumLOCAandDBEgridloadresultsarepresented inTable3.12-1withconsideration oftherequirement ofgridloadcombination, theSRSSoftheDBEandLOCAmaximumloadsislessthan2040lbs.Thismaximumloadis33.6%ofthegridstrengthforthe15x15OFAfuelassemblydesign.Thus,the15x15OFAdesignhasamplemarginforresisting faultedconditional loading.Thefuelassemblydesignisstructurally acceptable forDonaldC.CookNuclearPlantUnit1~Inconclusion, theSGTPProgramforCookNuclearPlantUnit1doesnotsignificantly increasetheoperating andpostulated transient loadssuchthattheywilladversely affectthefuelassemblyfunctional requirements.

Thefuelassemblystructural integrity isnotaffectedandthecoreeoolablegeometryismaintained fortheassumedfueltypeforCookNuclearPlantUnit1.3.12.2FuelRodStructural Evaluation Anevaluation wasperformed undertheSGTPProgramoftheimpactofNSSSperformance parameters inTable2.1-1ontheabilityoffueltosatisfyfuelroddesigncriteriaforCookNuclearPlantUnit1.Whilefuelroddesignanalysesarenotdirectlyimpactedbysteamgenerator tubeplugginglevels,theyaresensitive tocoreinlettemperature, massflowrates,andotherrelatedparameters.

Table3.13-2providesacomparison oftheparameters assumedfortheReratingProgramagainstthoseoftheSGTPProgram.Areviewofthethermalmodelsindicates thatthe-5%reduction inpower,orheatfluxwillgenerally offsetthe-6%reduction inmassflowrate,especially whencombinedwiththe2.7'FmA1944+w.wpf:1d441195 3.12-1 reduction inthemaximuminlettemperature.

Otherfuelperformance models,e.g.,fissiongasrelease,thermalcreep,etc.,dependent uponthecorepowerandfueltemperatures.

willalsobeoffsetbytheseeffects.Asaresult,fuelroddesignanalysesperformed forthe30%SGTPparameters wouldnotbeanticipated tobemorelimitingthantheReratingProgramanalysesforanyoftheimpactedfuelroddesigncriteria, andtheconclusions ofReratingProgramwillremainvalidforSGTPProgramforDonaldC.CookNuclearPlantUnit1.Finally,asinthepast,cycle-specific fuelperformance analyseswillcontinuetobeperformed foreachfuelregiontoconfirmthatthisassessment, andallfuelroddesigncriteria, aresatisfied fortheoperating conditions specifictoeachcycleofoperation.

Theseevaluations supporttheReloadSafetyEvaluation (RSE),whichistransmitted toAEPSCpriortoeachcycleofoperation.

3.12.3CoreDesignTheresultsofthecoredesignevaluation indicated thattheincreased steamgenerator tubeplugginglevelandreducedThermalDesignFlowresultinnoimpactstothecoredesignexceptforthevaluesofthestatepoint fortheSteamline BreakAnalysis, andtheDroppedRodAnalysis.

SeeSection3.3forthenewstatepoint valuesfortheSteamline BreakAnalysis.

SeeSection3.12.4forthenewlimitsconcerning DroppedRodLimitLines.3.12.4.1PurposeofAnalysisThepurposeofthissectionistodescribethethermal-hydraulic analysisnecessary tosupportthedecreaseinflowassociated withanincreaseinSGTPlevelto30%overarangeofRCStemperatures.

3.12.4.2Assumptions Table3.12-3summarizes thethermal-hydraulic designparameters usedinthisanalysis.

Thecoreinlettemperature isconsistent withthehightemperature 30%SGTPcase.Useofhighinlettemperature boundstherangeofRCSTavgwithregardtotheDeparture fromNucleateBoiling(DNB)analysis.

IncludedinTable3.12-3,forcomparison, arethethermal-hydraulic parameters currently intheDonaldC.CookNuclearPlantUnit1SafetyAnalysis.

mA1944+w.wpf:1d441195 3.12-2 3.12.4.3Discussion ofEvaluation 3.12.4.3.1 Calculation MethodsThethermalhydraulic designcriteriaandmethodsremainthesameasthosepresently intheDonaldC.CookNuclearPlantUnit1UFSARwiththeexceptions described inthefollowing paragraphs.

DNBMethodolo Theexistingthermal-hydraulic analysesusetheImprovedThermalDesignProcedure (ITDP)(Reference 1).Forthismethodology, uncertainties inplantoperating parameters, nuclearandthermalparameters, andfuelfabrication parameters areconsidered statistically, suchthatthereisatleasta95percentprobability ata95percentconfidence levelthattheminimumDNBRofthelimitingfuelrodisgreaterthanorequaltotheDNBRlimitoftheDNBcorrelation beingused.Plantparameter uncertainties areusedtodetermine theplantDNBRuncertainty.

ThisDNBRuncertainty, combinedwiththeDNBRlimit,establishes DesignLimitDNBRvalueswhichmustbemetinplantsafetyanalyses.

Sincetheparameter uncertainties areconsidered indetermining theDesignLimitDNBRvalues,theplantsafetyanalysesareperformed usingvaluesofinputparameters withoutuncertainties.

Inaddition, theDesignLimitDNBRvaluesareincreased tovaluesdesignated astheSafetyAnalysisLimitDNBRs.TheDNBRmarginavailable betweentheSafetyAnalysisLimitDNBRvaluesandtheDesignLimitDNBRvaluesisusedtooffsetDNBRpenalties.

Theanalysisofthe30%SGTPconditions usestheRevisedThermalDesignProcedure (RTDP)(Reference 2).Thismethodology givesimprovedDNBRperformance overITDPbystatistically combining theDNBcorrelation uncertainties withtheITDPuncertainties listedabove,i.e.,uncertainties inplantoperating parameters (vesselcoolantflow,corepower,coolanttemperature, systempressureandeffective coreflowfraction),

nuclearandthermalparameters (F~),fuelfabrication parameters (F~,),THINGIV,andtransient codes.Theuncertainty factorobtainedisusedtodefinetheDesignLimitDNBRwhichsatisfies theDNBdesigncriterion.

TheDNBdesigncriterion isthattheprobability thatDNBwillnotoccuronthemostlimitingfuelrodisatleast95percentata95percentconfidence levelduringnormaloperation andoperational transients (Condition Ievents)andduringtransient conditions arisingfromfaultsofmoderatefrequency (Condition IIevents).Condition IandIIeventsaredefinedinANSI18.2.AswasdonewithITDP,thedesignlimitDNBRvaluesareincreased tovaluesdesignated astheSafetyAnalysisLimitDNBRandtheDNBRmarginavailable betweentheselimitsisusedforflexibility ofdesignandoperation oftheplantandtooffsetDNBRpenalties suchasrodbow.TheDNBRlimits,currentpenalties, andmarginassociated withRTDPanalysisarelistedinTable3.12-4.mh1944<w.wp1:1d441195 3.12-3

~tllCMdiAnimprovedTHINGIVmodelwasusedintheDNBanalysisofthiscore.Thismodelisdescribed inReference 3andhasbeenapprovedforusebytheNRC.3.12.4.3.2 DesignEvaluation DNBPerformance Thechangeindesignparameters ingoingfromthecurrentanalysistothe30%SGTPconditions includeddecreasing thepower,flowandinlettemperatures asshowninTable3.12-3.ThisaffectstheDNBperformance ofthecore.TheDNBmethodology waschangedfromITDPtoRTDPtogenerateDNBRmargin.TheDNBRSafetyAnalysisLimits(Table3.12-4)weresettokeeptheDNBRlimitingportionofthecorelimitsunchanged.

Theassociated axialoffsetlimitswererecalculated.

TheDNBeventsnotprotected bycorelimitsthatwereanalyzedwereLossofFlow,LockedRotor,StaticRodMisalignment, DynamicDroppedRod(RCCA),andRCCABankWithdrawal fromsubcritical (RodWithdrawal fromSubcritical).

Theresultsoftheanalysesshowedthatthethermalhydraulic designcriteriaweremetforeachevent.FuelTemeraturesThelimitingvaluesofthefuelaverageandcenterline temperatures willnotchangeduetothe30%SGTPconditions.

3.12.4.4Conclusions Thermal-hydraulic analysesweremadeforthefuelforthelimiting30%SGTPparameters usingRTDPmethodology.

TheanalysisshowedthattheDNBRdesignbasiswasmetforthelimitingDNBevents.Thisanalysiscausedtheavailable DNBRmargintoincrease.

Thismargincanbeusedforflexibility ofdesignandtooffsetunanticipated DNBRpenalties.

3.12.4.5References WCAP-8567-P-A, "Improved ThermalDesignProcedure,"

H.Chelemer, L.H.Boman,D.L.Sharp,February, 1989.2.WCAP-8567-P-A, "RevisedThermalDesignProcedure,"

A.J.Friedland andS.Ray,April1989.3.WCAP-12330-P, "Improved THINGIVModelling forPWRCoreDesign,"A.J.Friedland andS.Ray,August,1989.mA1944+w.wpt:1d441195 3.12-4 TABLE3.12-1MAXIMUMLOCAANDDBEGRIDLOADRESULTSCaseAccumulator DBESRSS(DBE8LOCA)39475<2000<2040GridLoad(Ibs)XZGridStrength)6080lbs.<33.6%ofgridstrengthm51944%w.wpf:1d441195 3.12-5 TABLE3.12-2FUELRODDESIGNANALYSISPARAMETERS Parameter CorePower,MWtMinimumSystemPressure, psiaMaximumInletTemperature,

'FThermalDesignFlow,gpmBypassFlow,%FDHReratinProram34132100546.2354,0004.51.55SGTPProram32502100543.5332,8004.51.55m:11944<w.wpf:1'1953.12-6 TABLE3.12-3DONALDC.COOKNUCLEARUNIT130%SGTPPROGRAMTHERMALANDHYDRAULIC DESIGNPARAMETERS ReactorCoreHeatOutput,MWtReactorCoreHeatOutput,10'tu/hrHeatGenerator inFuel,%Pressurizer

Pressure, Nominal,psiaRadialPowerDistribution DesignParameters CurrentAnalysis30%SGTPProgram"'41 3f'1325011,64611,09097.497.4210021001.55[1+0.3(1-P)]

1.55[1+0.3(1-P)]

LimitDNBRforDesignTransients FlowChannelDNBCorrelation HFPNominalConditions VesselThermalDesignFlow,10'bm/hrCoreFlowRate,10'bm/hrBypassFlow,%NormalVessel/Core InletTemp,'FVesselAverageTemp,'FCoreAverageTemp,FVesselOutletTemp,'FAverageTempRiseinVessel,'FAverageTempRiseinCore,FHeatTransfer" AverageHeatTransferArea,ft'verage HeatFlux,Btu/hr-ftverage LinearPower,kw/ft"'eak LinearPowerforNormalOperation, kw/ft"Temperature atPeakLinearPowerforPrevention ofCenterline Melt,'FTypical1.45WRB-1133.4127.44.5546.4578.7581.8611.064.667.352,200217,4007.0416.5'700ThimbleTypicalThimble1.451.401.42WRB-1125.9120.34.5543.5576.3579.4609.165.668.452,200207,0006.7015.74700(a)(b)(c)(d)CookNuclearPlantUnit1iscurrently licensedtooperateat3250MWtHighinlettemperature boundstheproposedtemperature rangewithrespecttoDNBBasedonnominal144inchactivefuellengthBasedon2.35FPeakingFactormA19444W.wpf:1d441195 3.12-7

TABLE3.12MDONALDC.COOKNUCLEARPLANTUNIT1SGTPPROGRAMRTDPDNBRLIMITSANDMARGINSUMMARYDNBCorrelation CellTypeDesignLimitSafetyAnalysisLimitTotalDNBRMarginDNBRPenalty-RodBow1'FTemperature BiasNetRemaining DNBRMarginTypical1.231.4012.12.61.58.0WRB-1Thimble1.4214.12.61.510.0m%1944<w.wpf:1d~1195 3.12-8

4.0CONCLUSION

S Providedinthisdocumentaretheresultsandconclusions ofthesafetyanalysesandevaluations tosupporttheimplementation oftheSGTPProgramandtherevisedTechnical Specification changesforCookNuclearPlantUnit1.Thesafetyanalyses, evaluations, andsupporting documentation providedinthissubmittal demonstrate acceptable resultsineachcase,incorporating therevisedoperating conditions associated withtheSGTPProgram.Abriefsummaryoftheresultsofeachanalysisandevaluation isprovidedinthe'Summaryand~Conclusions" sectionofthisreport.mhl9444w.wpf:1d 4411954-1 (t

APPENDIXAProposedTechnical Specification Changes

~I~~I~~II~~I~II~II~~~II~I~,I~~~

6606502400galaUNACCQPTOPEAATlOH 6406302100paIaP-62061060059058057000.20.40.60.81Power(fraction ofratedthermalpower)1.2PRESSURELEihhl18402000210022502400(0.02,62086),(0.02,632.79),(0.02,63985))(0,02,64986),(0.02,65952),BREAKPOINTS (1.136,586.17),(1.094,60021),(1.068,608.72),(1.031,620.83),(0.996,632.42),(12,577.94)(12,58652)(12,591.77)(12,599.40)(12,606.63)FIGURE2.1-1REACTORCORESAFETYLIMITS TABLE2.2-1REACTORTRIPSYSTEÃINSTRUMENTATION TRIPSETPOINTS FUNCTIONAL UNITTRIPSETPOINTALLOWASLE

'VLUESl.ManualReactorTripNocApplicable Nor,Applicable 2.PoverRange,NeutronFlux3.PoverRange,NeuczonFlux,HighPositiveRateLovSerpoint-lessthanorequalto25\ofRATEDTHERMALPOWERHighSetpoinr-lesschanorequalto109\ofRATEDTHERMICPOWERLessthanorequalto5%ofRATEDTHER.QLPOWERvichatimeconscancgreaterthanorequalco2secondsLovSetpoint-lessthanorequalto26%ofRATEDTHER%0.POWERHighSetpoint-lesschanorequalto110%o"RATEDTHERMALPOWERLesschanorequalto5.5%ofRATEDTHER'.9J.

"POWERvithatimeconstancgreaterchanorequalro2seconds4.PoverRange,NeutronFlux,HighNegativeRace5.Intermediace Range,NeutronFlux6.SourceRange,NeutronFlux7.Overtemperature DeltaTLessthanorequalto5%ofRATFDTHERMO.POWERvithatimeconstantgreacezchanorequalro2secondsLessthanorequalto25%ofRATEDTHERMALPOWERLessthanorequalto105countspersecondSeeNore1Lessthanorequalto5.5%ofRATEDTHEB.fALPOWERvitharimeconscantgreaterchanozequalto2secondsLessthanorequalro30~ofRATEDTHERMALPOWER'esschanorequalto1.3x10councspersecondSeeNote38.Overpover DeltaTSeeNote2SeeNoce49.Pressurizer Pessure--Lovl.O.Pressurizer Pressure--High11.Pressuzirer WaterLevel--HighGreaterrhanorequalto1875psigLessthanorequalto2385psigLessthanorequalco92%ofinstrument spanGreaterchanorequalro1865psigLessthanorequalro2395psigLessthanorequalto93'ff.nstrument spanAMENDMENT NO.9i12515812.LossofFlovGreaterthanorequalto90%Greaterthanorequalcoof~ekgn-flovperloop+89.1%of-4askgs.flovpe"loop*85)%75*Se..~n.flovis~468.gpmpezloop.~~~iwuwNERSuCZDCOOKNUCLEARPLANT-UNIT12-5

REACKNTIIII'Y~'I'IJI Itl~"I'l<IIHI N'I'h'I'll>N

'I'IIII'lI'I'UIN'I~

N>>'I'A'I' l>NNote1sOvurteaperature ATlATg-K.oi2$>T.'IIlIl>-'III'-T'>Kll~-I~'-l{hl)]I'Iwl>urulA'I'-"lndiCated AI4LIth'II.I>

llll.lOIAI

~I'L>WI.I.

u~Averagete>II>orat>>l>>,

I'7(.9u,Indicated

'I'LIIA'I'I:I>

'I'IIII>I%I.I'>>wl;II(~~II')4V>JAPreeeurLxl.'II>ruu~uta:~I~:>lep>+~,s>tvS~l~II>dicated II>'5nu>>>i>>4Iul>U>4ti>>g l>c>:ule('235l>sigor2085psig)~Thefunction>J>.>>er4I>:>I l>yI.l>Ulua>ll4>Jcol>lrollerforTdynaaiccompensation avg'I>>><<~>>>>I4>>t~>>>>l.:..>>>>thulead-la>J cuntloller forTt~22sccavgSUCSel2=Laplacetra>>stor>>>

op<<rator

n'fhL~I.I."

2,2-j+c~ot nnngi1:Ct:-IMILJIUIgo~Aggo~~r.o~~~

Ol)er~tiunMichiloupelh,~4IK-i)u~luK.=-u.ouisv3Iq)exceeds-37percent,tbeifbII0.33percentnf1tnrainsat,Iq)exceeds+k>percent, tbeitbg~percent nfltnrainsatC2.N'ii)Foreachpercentthattheaagnituds oC(qtripsstpointshallbeautoaaklcally reducedaavaeTHERHhLPOMER.(iii)Foreachpercentthattheaagnituds og(qtripsstpointshallbeautosatically rsdu5aSga'rmTllERlQLPOMER.~Inlf(sf)isafunctionoftheindicated differencebatueentopanybottondetectors nfthser-rangenuclearionchasbers; uithgainstobeselectedbasedonnaasuryginstrunsnt responseduringplantstartuptestssuchthat:~3(i)Forqw%,between-37percentand+R.percent, C(hI)p(vhsroqandgarephrCentHA'i'I:.l~

Tll)I,i.lVRinthetOpandbhttOahalVeaOltheCOrlrdfspactively, a>>dqiqi"total'tHENlhLpOMERinpercentoCa@~TBERCaLPOMER).TQ%I aXMI.L2..~Locojtjudge~SS'Q~SII~~~

pg~o~~nca~g~~ii Mote2:Overpower nT<nTiK-Koi5<3SST-K(T-T")-f~nIii62where:nTolitic-'ala:8

.>'I'tlu'<'>:>>'I'IIVIIHAL PQWEIIAverageteapcrature,}'ndicated Tatiwra>>avg1.003'I'lVIU4hlI'OWERIi'-~>>l0~0.0171/1'orincreasing averageteaperature hand0fordecreasing averagetesperature fcl+So.uai5forT>T";K60forT<T"Thefunctiongenerated bytheratelagcontroller forTdynaaiccoapensation LVQMot~3$Mote4!Tfaeconstantutilizedintheratelagcontroller fort~l0secs.34VQLaplacetransfora operatorf>(nIl~0Thechannel's aaxfauatrippointshallnotexceeditscoaputedtrfppofnCQy~prethan~percentnTspan-~(PEP~RrsPowI8~rv)

Thechannel'aaxiauatrippointshallnotexceeditsdeputedtripponggorethan~percentnTspan.

~++h2.1SBASE4LoopOperation'estinghouse Fuel(15xl5OFh)(LR3-1Correlation)

TypicalCellThimbleCel].**Correlation LimitDesignLimitDNBRSafetyAnalysisLimitD.':BR1.17/~23I.WO~A1.17Jo22A/r+2.A~Reac"oroolantSystem""ess'es~NIA~o~volBRs0ess~~~oa":era"eenthalpva"the'se'.e:(itplociofpointsofi.-:"=K~AL PO~ER,x:etagete=perature forwhichthebedesin"'.lBRlimit,orthiseqaltotheen=halpyofsaturated oo~represents typicalfuelrod*represents fuelrodsneargidetubeCookNuclearPlantUnit1B2-1(a)Amendment No.7$$J?ZS~

P~gNsg~Iejetfve~teTz'fppravfdesprotectfon forcontrolraddrop~id'+:Athfjbpaver,aroddropaccidentcouldcauselocalf1~>+~ayv}6<cauldcauseanuncansezvatfve localDNRtoexist.PaverRsnjeNejatfveRateTripvillpreventthistramace~injbytrfppinjthereactor..

Nocredftfstakenforoperation ofthePorNegativeRateTzfpfor'hosecontro1roddropacefdentsfar~ichtheDNERisvfllbegreaterthantheapplicable design1faitDNRvaluefareachfueltype.Intermediate andSourceRan~NuclearFluxTheIntermedi,ate andSourceRange,NuclearFluxtripsprovidereactorcoreproectionduring.eactorstart"p.Thesetripsprovideredundant protection tothe1ovsetpointtr'pofthePoverRange,NeutronFluxchannc1s.

ThcsourceRange"hanna'svillin't'ateazcactortripatabou"1Ccoun"spcrsecond,nlessmanuallyblockedvhenP-6becomes=tive.hentc-.cd'a"cRangeChanne'svillinitiateareactortripat"=cnt.ere'."c"or='ai

=capproximately 25percentofRATEDK"=KQL?-':-Rn'css=anayb'ac~cd-hcn?-'.0bcco=csactive.4ocreditvas=kcn="=operation of=hctripsassociated vi=hcihertheIntermediate c"="==Rangehanncsn=hcac=cnana.yses; hovever,theirinotionalcapabi=r at=hcspeci'cdtz'psettingsisrequiredbvthiss=cci=ication toc-.'".ance

=hcc<<rcra:zciab'vo=--Reactor~y<<<<p<<p<<~Crcrtc-.pcraturcdelta.:"'pprov'Ccscoreprate:'ontoprevent"NSa'=""bina='ons o:pressure, pover,coolanttemps"aturc,andaxial-overCis=".'bu='on.

providedthatthetransient issovvithrespecttopi"'ng""a..sitCciaysfzomthecoretothetemperature detectors (about4seconds),

andpressureisvithintherangebetveenthcHighandLovPrcssurereactortzips.Thissetpointincludescorrec=iona forchangesindcnsiyandheatcapacityofvatezvithtemperature anddynamiccompensation foriindelasfromthecoretothelooptemperature detectors.

Thereferceaverageemperatuze an".ereerencecratingpssure(P')reseteqtothefupovcrindatedTavgathenomiRCSaperangpressurzespectiv y,:oensuproctionoftcorelimiandtoprervetheat.=iontimoftheOverperatureltaTtriportherangoffullpo'czaveragetemperauresassumeint'hesatanalsestnor=aaxapovcz4'stzibution, thisreactortzipimit,isalvaysbelov=hecoresafetylimitasshovninFigure2.1-1.Ifaxialpeaksaregreaterthandesign,asindicated bythedifference betveentopandbottom-overrangenucleardetectors, thereactortripisautomatically reducedaccording tothenotations inTable2.2-1.DnE-rZCOOKNUC~~PLANTUNIT1B2-4~MZNTNo,74,126

~rtaTheoverpover deltaTreactortripprovidesassurance offuelintegrity,

~.g..nomelting,underallpossibleoverpover conditions, liaitstherequiredrangeforOvertemperature deltaTprotection, andprovidesabackuptotheHighHeutronfluxtrip.Thesetpointincludeacorrections forchangesindensityandheatcapacityofvatervithtemperature, anddynasticcompensation foriindelasfroathecoretothelooptemperature detectors

~Thereferenaveragetperature()isse'cetoe1overindicadTavgtoesurefuelegritydurgoverpovaconditions orPRirangeof1poveraveagetemperaesassumein'hesatyana]sis,eoverpover etereactortrpp'rovasprotectonorc~upprotection foratpoversteamline breakevents.Creditvaatakenforoperation ofthistripNthesteamlinebreakmass/energy releasesoutsidecontainment ana]ysis.

1naddition, itsfunctional capability atthespecified tripsettingisrequiredbythisspecification toenhancetheoverallreliability ofthereactorprotection system.Pressuriser treueThetressuriser HighandLosPressuretripsareprovidedtolimitthepressurerangeinvhichreactoroperation ispermitted.

TheHighPressuretripisbackedupbythepressuriser codesafetyvalvesforLCSoverpressure protection, andistherefore setloverthanthesetpressureforthesevalves{2485psig),TheHighPressuretripprovidesprotection foraLossofKxternalLoadevent.TheLovPressuretripprovidesprotection bytrippingthereactorintheeventofalossofreactorcoolantpressure.

Pzessuriser UaterLevaThePressuriser HighPaterLeveltripensuresprotection againstReactorCoolantSystemoverpressurization bylimitingthevaterleveltoavolumesufficient toretainasteambubbleendpreventvaterreliefthroughthepressuriser safetyvalves.Thepressuriser highvaterleveltripprecludes vaterrelieffortheUncontrolled RCCAWithdrsval atPoverevent,COOKNUCLEARPESTUNZT1~2-5AMZgu~HO.XW.LN,15

'll 34.1REACTIVTTY CONTROLSYR~34.l.1ADORATION CONTROLSHUTDOQHMARCIN-TAUCCREATERTHAN200FLIMITINCCONDITION POROPERATION

/.33.1.1.1TheSHUTDOMNMARCINshallbegreaterchanorequalto~Deltak/k,APPLICASILITY:

NODESl.2+,3,and4.ACTION:1.3UichtheSHUTDOMHMARCIHlesschan~~lDeltak/k,%mediately tnitiaceandconcinueborationatgreaterthanorequalto10gpssotasolutioncontaining greaterthanorequalto20.000ppaboroaorequivalent unciltherequiredSMUTDORtMARCIHisreacored.

SURVEILIANCE UIREMENTS 4.1.1.1.1 ThsSIICTOOIOI IIASCINshallbedatartltsd tabedresserthatsitstealss+H\Deltah/h:aeb.C~Vithinonehourafterdetectioa ofaninoperable controlrod(a)andacleastonceper12hoursthereafter vhiletherod($)iainoperable.

Iftheinoperable concrolrodtshearableoruntrippablet theaboverequiredSttUTDOQN MARCIHshallbeverifiedacceptable vithanincreased allovance forthevichdzavn vorthotcheiaaovable oruntrippable controlrod(a).SheniaNODE1orMODE2vithjeffgreaterthanorequalto1.0,acjleastonceper12hoursbyverifying thatcontrolbankvithdraval isvithintheliaitsofSpecification 3.1.3.5.%heninMODE2vichXefflessthan1.0,vithia4hourspriortoachieving reactorcriticality byverifying thatthepredicted criticalcontrolrodpositioniavichintheliattaofSpecification 3.1.3.5.d.Priortoinitialoperation above50lATEDSHOALNMEkaftereachfuelloading,byconsideracion ofthefactorsofebelov,viththecoacrolbanksattheaaxituainsertion liaitofSpecificacioa 3.1.3.5.+SeeSpecialTossiizceptioa 3e1041eCOOKHUCLLCPLAHT~UNIT13/411AMEHDmrrNO.7k.128.148

REACTIVITY CONTROLSYS~CHARCINCPUMP-SHUTDO'4N LLNTINCCONDITION FOP.OPTATION3.1.2.3a.Onechargingpumpintheboroninjection flowpathrequiredbySpecification 3.1.2.1shallbeOPERABLEandcapableofbeingpoweredfromanOPERABLEemergency bus.b.Onechazgingflowpachassociated vithsupportofUnit2shutdownfunctions shallbeavailable.

>>APPLICABILITY:

Specification 3.1.2.3.a.

-MODES5and6Specification 3.1.2.3.b.

-AtalltimesvhenUnit2isinMODES1,2,3,or4.ACTION;a.Mi.thnochargingpumpOPERABLE, suspendalloperations involving COREALTERATIONS orposi.tive reactivity changes.>>>>

b.WithmorethanonechargingpumpOPERABLEorvithasafecyinjection pump(s)OPERABLEvhenthetemperature ofanyRCScoldlegislessthatorequalco152F,unlesschereaccorvesselheadisremoved,remove0theadditional chargingpump(s)andthesafetyinjection pump(s)motorcircuitbreakersizomtheelectrical povezcircuitwithinonehouz.c.Theprovisions ofSpecification 3.0.3arenotapplicable.

d.Inadditiontotheabove,vhenSpecif'cation 3.1.2.3.b isapplicable andcherequiredflovpathisnotavailable, returntherequiredflowpathtoavailable statuswithin7days,orprovideequivalent shutdowncapability inUnit2andreturntherequiredflovpathtoavailable statusvithinthenexc60days,orhaveUnit2inHOTSTANDBYwithinthenext12hoursandHOTSHUTDOWNvithinthefollowing 24hours.e.Therequiremencs ofSpecification

3.0. 4arenotapplicable

whenSpecification 3.1.2.3.b applies.SURVEILLANCE REUIMHENTSp>rrsg~'~<

4.1.2.3.1 TheabovezequizedchargingpumpshallbedemotratedOPERABLEbyverifying, thaconrecirculation flov,chepumpdevelopsa~hasp+pressureofgreaterthanorequalto~ps+whentestedpursuanttoSpecification 4.0.5.gZ,Vo+Amaximumofonecentrifugal chargingpumpshallbeOPERABLEvhcneverthetemperature ofoneormoreofcheRCScoldlegsislessthanorequalto152F.>>>>Fozpuzposesofthisspecification, additionofvatezfromtheRUSTdoesnocconsticuce apositivereactivity additionprovidedtheboronconcentration intheRUSTis.greaterthantheminimumrequiredbySpecification 3.1.2.7.b.2.

COOKNUCLFMLPLANT-UNIT13/41-11AMENDMENT NO.9'f,f/',167, 3.1.2.4AtleasttvochargingpumpsshallbeOPERABLE.

MODES1,2,3and4.hEZQH:WithonlyonechargingpumpOPERABLE, restoreatleasttvochargingpumpstoOPERABLEstatusvithin72hoursorbeinHOTSTANDBTvithinthenext6hours;restoreatleasttvochargingpumpstoOPERABLEstatusvithinthenext4ShoursorbeinCOLDSHUTDOWSvithinthefollovtng 30hours.piggyQEA/YNK4.1.2.4Atleasttvochargingpumpsshallbedeonstrated OPERABLEbyverifying, thatonrecirculation flov,eachpumpdevelopsagBaeherge-pressure ofgreaterthanorequalto~paidvhentestedpursuanttoSpecification 4.0.5.~~NoCOOKNUCLEARPIhHT-UNIT13/41-12AMENDMENT NO.04,164

[Rr>CTIVITY CONTROLSYST=MSl!BORATEOWATERSOURCESS~UT"OWN;>LIMIT!NGCONDITION FOR".o=~AION-;'I'.3.1.2.7Asaminimum,oneofthefollowing boratedwatersourcesshallbe.-'>OPER."BLE:

a.Aboricacids:oragesystemandassociated heattracingwith:l.Aminimumusab'.eboratedwatervolumeof4300gallons,2.Between20,0:0and22,500ppmofboron,and3.Aminimumsolutiontemperature of145'F.b.Therefueling waterstoragetankwith:l.Aminimumusableboratedwatervolumeof90,000gallons,2.Aminimumboronconcentration of2400ppm,and3.Aminimu.-.,

solutiontemperature of~F.OAPPLICABILITY MOO'S=an"6.ACTION:>Withnoboratedwatersor"eOPERABLE, suspenda>',operations involving COREi"ALTERATIONS orpositive~eactivity changesuntilatleastoneboratedwaterresourceisrestoredoGP=.=.'BLE status.i>>'URVEILLANCE REUIRE'".ENTS (sj>4.1.2.7 Theaboverequiredboratedwatersourceshallbedemonstrated

>OPERABLE:

a.Atleastonceper7daysby:2.3.Verifying theboronconcentration ofthe~ater,Verifying thewaterlevelvolumeofthetank,andVerifying theboricacidstoragetanksolutiontemperature whenitistnesourceofboratedwater.b.Atleastonceper24hoursbyverifying theRWSTtemperature whenitistnesource,ofborgtedwater.I>IIIiForpurposeso.tnisspeci.ica.ion,additionofwaterfromtheRWSTdoes!lnotconsituteapositivereactivity additionprovidedtheboronconcentra-

>;tionintheRMSTisgreatertnantheminimumrequiredbySpecification ii>I>Tb>.Q.C.COOK-UNIT13/41-15Amendment No.$2,Ill

",RE'CIVITYCONTROLSYSTEMS!I"BORATEDWATERSOURCES-OPERA.fONSi:LI.'1.'T:NG CONDITION FOROPEPAT!ON I::3.l.2.8~II'~~i~II~~III~~I~Eachofthefollowing boratedwatersourcesshallbeOPERABLE:

a.Aboricacidstoragesystemandassociated heattracingwith:l.Aminimumusableboratedwatervolumeof5650gallons,Between20I000and22,500ppmofboron,and~\3.Aminimumsolutiontemperature of145'F.b.Therefueling waterstoragetankwith:~g~~II~~l.AminiŽu.".

contained volumeof350,000gallonsofwater,2.3.'IAPPL:".<<9ILITY:<lf~AC~fQaIBetweenŽ3and2600ppmofboron,andAmin;--s"1Žiontemperature ofg&F.70MODESI,",3and4.Withthed'or'.cacidstoragesysteminoperable, restorethestoragesystemtoOPERABLEstatuswithin72hoursorbeinatleastHOTSTANDBYwithinthenext6hoursandboratedtoaSHUTOOWNMARGINequivalent toatleast1'k/kat200'F;restoretheboricacidstoragesystemtoOPERABLEsatuswithinthenex.7daysorbeinCOLDSHUTDOWNwithinthenext30hours.'I~~I~~Withtherefue'iing waterstoragetankinoperable, restorethetanktoOPERABLEstatuswithinonehourorbeinatleastHOTSTANDBYwithinthenext6hoursandinCOLDSHUT-DOWNwithinthefollowing 30hours.~.'UR'I'5iLLANCE"4.1.2.8Eachboratec~a:ers=urcshallbedemonsratedOPERABLE:

~~3.C.COOK-UNIT13/41-16Amendment No.gg,-I TABLE3.2-1DNBPARAMETERS LENITS4LoopsinOperation atRATEDTHEBKWLPOWERReactorCoolantSystemTavgPressurirer Pressure4(+7/'S+5'./)(g8~j)QL]/ifQggog~pfepscREsp~asuiTvg

>2050psig**ReactorCoolantSystemTotalFlovRate/CD/)/dg>~~gpm*Indicated averageofatleastthreeOPERABLEinstrument loops.Limitnotapplicable duringeitheraTHERMALPOWERrampincreaseinexceof5percentRATEDTHERMALPOWERperminuteoraTHELRQ.POWERstepincreaseinexcessof10percentRATEDTHFKM.POWER.Indicated value.COOKNUCLEARPLANT-UNIT13/42-14

TASKS3.3iIRZACTORTRIPSYST5tINSTNlMBtTATIOÃ RESPONStTIMESl.ManualReactorTrip2.PoverRange,'eutron Flux(HighandLovSecpoinc)

NOTAP?LICASLE Lesschanorequalco0.$seconds+3.PoverRange,NeutronFlux,HighPositiveRate4.PoverRange,NeutronFlux,HighNegaciveRate5.Incermediace Range,NeutronFlux6.SourceRange,NeutronFlux7.Overtemperature deltaT8.Overpover deltaTsthanorequalto0.5seconds'OT APPLICASLK NOTAP?LICASLE Lessthanorequalto6.0seconds*Lesschanorequalto6.0seconds+9.Pressurirer Pressure--Lov LessthanorequaltosecondsIO.Pressurizer Pressure-

-HighLesschanorequalto2,0~secondsII.Pressurizer MaterLevel-HighLessthanorequalto2.0secondsNeutrondececcors areexempcfromresponsetimetesting.Responserimeofcheneutronfluxsignalportionofthechannelshallbemeasuredfromdetectoroutput,orinputoffirscelectronic component inchannel.COOKNUCDhRPlhFTNITl3/4310AMEHDMENT NO.98.L1;sa TABLE3.3-2Continued REACTORTRIPSYSTEMINSTRUMENTATION RESPONSETIMESP>CTIONAL UNIT12.LossofFlov-SingleLoop(AboveP-8)RESPONSETIMELessthanorequalto1.0seconds13.LossofFlov~TvoLoops(AboveP-7andbelovP~8)Lessthanorequalto1.0seconds14.SteamGenerator VaterLevel~-Lov-Lov15.Steam/Feedvater Flo~MismatchandLosSteamGenerator MaterLeveL16.Undervoltage-Reactor CoolantPumps17.~Žderfrequency-Reactor CoolantPumpsLessthanorequalto2..0++secondsNOTAPPLICABLE LessthanorequaltoI<~secondsLessthanorequalto0.6seconds13.TurbineTripA.LouFluidOil,PressureB.TurbineStopValve'.9.5afe=y Injection InputfromESF20.Reactor CoolantPumpBreakerPositionTripNOTAPPLICABLE NOTAPPLICABLE NOTAPPLICABLE NOTAPPLICABLE COOKNUCLEARPLANT-UNIT13/43-11hMKHDMENT NO.ZlS,15S TABLE3.3-3Continued ENCINEERED SAFETYFEATUREACTUATION SYSTENINSTRPAENTATION FUNCTIONAL UNITDcAF7c.f.SteamFlowinTvSteamLines-Hi.gh KININJMTOTALNO.CHANNELSCHANNELSAPPLICABLE OFCHANNELSFourLoopsOperacing

'2/steaml.ine1/samliney2steam1ne51/steamlinereeLoops0ratingCOINCIDENT ITHEITHER2/opratingsteaminel~/any1/operating 3'peracing steamlinesteamlineT~~Lov~LovavgFourLoopsOperating 1T/loopavgTanyavg1Tany1,2,3~14*3loomisThreeLoopsOperati.ng OR,COINCIDVITHTavdocratingloopinanyoperatinloop1Tin.3anyc5ooperacing loops1.5SteamLinePressure-LovFourLoopsOperacing ThreeLoopsOperating 1pressure/

loop1pressure/

operating loop2pre55ures anyloops1~pre555ure inanyoperating loop1pressureany3loops1pressureinany2operati.ng loops1,2,3'4".OOKNUCLEARPLhNT-UNIT13/43-17AHEIHENTNO.Pf,f28153

TABLE3.3-3Continued ENGINEERED SAFETYFEATUREACTUATION SYSTEHINSTRUMEÃI'ATION FUNCTIONAL UNITCOINCIDE'C MITH~HBR-TOTALNO.OFCHANNELSHINIMUHCHANNELSCHANNELSAPPLICABLE TDTRIPOPERABLEMODESACTIONeeLogeLogavgFourLoopsOperating ThreeLoopsOperating 1T/loopavgopera8fng loop2Tany75ops1~Tinanyoperating loop1Tany'Ploops1Tinan)cooperating loops1,2,3'5SteamLinePressure-LovFourLoopsOperating ThreeLoopsOperacing 5.TURBINETRIP&FEEDWATER ISOLATION 1pressure/

loop1pressure/

operating loop2pressures anyloopslee>>pressureinanyoperating loop1pressure1,2,3e14any3loops1pressure3~inany2operacing'oops a.SteamGenerator MaterLevel--High-High3/loop2/loopin2/loopin1,2,3anyoper-eachoper-atingloopatingloopCOOKNUCLEARPLANT-UNIT13/43-21AHENDMENT NO.gf,f26,153

ENCZNEERED SAPETTFEATURESIN?ERLOCXS DESICNATIONF-11P-12CONDITION ANDSETPOINTVith2of3pressuriser pressurechannelsgreaterthanorequalto1915psig.Pith2of4TchannelsavelessthanorequaltoSetpoint.

Setpointgreaterthanorequalto541FFUNCTIONP-11preventsordefeatsmanualblockofsafetyin)ection actuation onloTpressurizer ptessure.

P-12allovsthemanualblockofsafetyin]ection gau87iouowllovsteamlinepressure.

~~CAccfc:S ST~gAzMZISdMD4oNHI4HStcam.FC~,~~factssteamdumpblocks.Pith3of4TchannelsavgabovetheresetWQev-Polw,P/ZPFPV/pgpgA~Iyh'grNfAJHhc8LdCCPFSrtr=BTyj~J6G770~Ac7&t97ppwo~w~~EA~r~ZPR~ss~g~COOKNUCLEALHAH'UHIT13/43-23aammmrrNO.153

TASLE3.3-4EHCINEERED SAFETYFEATUREACTUATION SYSTEHINSTRUHENTATION TRIPSETPOINTS FUNCTIONAL UNITTRIPSETPOINTALLOVASLE VALUES1.SAFETYINJECTION, TURBINETRIP,FEEDVATER ISOLATION, AHDHOTORDRIVENFEEDVATER PIPSa.HanualZniti.ation SeeFunctional Qnfc9b.Automatic Actuation Logicc.Contaizuaent Preaaure--

Highd.Preaauriser Pressure--

LcnrHotApplicable Lessthanorequalto1.1paigCreaterthanorequaltoldl5psigHotApplicable Lessthanorequalto1.2paigCreaterthanorequaltoId05paig~.Differential Presaur~hetveenSteamLines-Highf.eaaPlovtnSteama--Hi.ghCoci.dentso+arSteamLinePreasure--

LoeLessthanorequalto100psiaathan6orqualto1.x10lbsfromOtdto20%1d.Lineafrom1.42x10lbaa)20%1dto3.1$z0lbs/hr00%loadLessthanorequaLto112psiaathan6orqualto1.6x10lbsfrom0%dto20%,ad.Lipafrom1.5610Iba)20'dto3.9310lbs/hrt100%loadTgreacercoc'o541PTgreaceroc'to539FQg'eaterchanorequal(+eaterthanorequaltoto500yaigsteamline480p+gsteamlinepressurepressureCOOKNUCLEARKLFZUHZT13/43-24AHEIHEHTHO.49,128153 TABLE3.3-4Continued ENCINEERED SAFETYFEATUREACTUATION SYSTEMINSTRUMENTATION TRIPSETPOINTS FUNCTIONAL UNIT2.Containment Radio-activityy--High TrainA(VRS-1101, ERS1301,ERS-1305) 3.Contatnmeat Radio-activity--High TrainB(VRS-1201, ERS-1401, ERS-1405) 4.STEAMLINEISOIhTION TRIPSETPOINTSeeTable3.3-6SeeTable3.3-6ALMVABLEVALUESHotApplicable NotApplicable a.Manualb.Automatic Actuattoa Logic--------SeeruaeaaoaaX UnSe9------...

NotApplicable HotAppltcable c.Containment Pressure--

HighHighd.SteamFlovtnTvoSteamLinea--Htgh Cotncident vtthT--Lov-Lcnr

~Lessthanorequalto2.9patgLessthan6orequalto1.42z10lba/hrfrom0%loadto20%load.Linearfrom1.42x10lbs/hra(20%loadto3.4Bs10lba/hrat100%load.Lessthanorequalto3patgLeaathan6orequalto1.56x10lbs/hrfrom0%loadto20%load.LiIIearfrom1.56x10lbs/hr6at 20%loadto3.9310lba/hrat100%load.Q,5ig444.inc,MN54&~-Lusa)5.TURBZ?RTRIPASDFEEDVATER ZSOLATZOX Tgreatertheaora@ito54loFCreaterthanorequalto500pstgsteamlinepressureTgreaterthanor~qQ1to539FCreaterthanorequalto4BOpaigsteamltaepressurea.StaaaCenerator QatarLevel,--High-High Lessthanorequalto67%ofnarra'-range instrument spaneachsteamgenerator Lessthanorequalto6$%ofnazrcw-range instrument spaaeachsteamgenerator COOKNUCLEARPALS?-UNIT13/43-26AHKHDMEST EO.94,Qg,153 ThhLE3.35BfCNEELEDSAFETYFEATURESkESPONSETIMESINITIATINC SXCNALANDFUNCTIONRESPONSETIMEINSECONDS1.ManualaoSafetyIn)ection (ECCS)Feedvater Isolation ReactorTrip(SI)Containment Isolation-Phase

'AContainment PurgeandExhaustIsolation Auxiliary Feedvater PuipsEssential ServiceWaterSystemNot,Applicable NotApplicable NotApplicable NotApplicable NotApplicable NotApplicable NotApplicable b.Containment SprayContainment Isolation-Phase 5Containmenc PurgeandExhaustIsolation Containment hirRecirculation Fanc.Containment Isolation-Phase

'h'ontainment PurgeandExhaustIsolation d.SteamLineIsolation 2.Containment Pressure-Hi ha.SafetyIn]ection (ECCS)b.ReactorTrip(froaSI)c.Feedvater Isolation d.Containment Isolation-Phase h'.Containment PurgeandExhaustIsolation f.Auxiliary Fee@Water Pumpsg.Essential ServiceWaterSystemNotApplicable NotApplicable NotApplicable NotApplicable NotAp'placable NotApplicable NotApplicable

+7,0Lessthanorequal27.OQQ/~MLessthanorequalLessthanorequalLessthanorequal18.04/28.0H NotApplicable NotApplicable Lessthanorequal13.0f/48.0H toto3,0to8.0to'toCOOKNUCLEARPLANT-UNIT13/43-27AMnmENTgag,158 TABLE3.35ContinuEÃCZNEERED SAFETYFEATURESRE5PONSETZMZ5IHITZATINO 5ZCNALAHDFUNCTIONRE5POHSETINEINSECONDS3.pressurizer Pressure-t,ov a.SafetyIn)ection

{ECCS)b.C~d.~oReactorTrip(fromSI)Feedvater Zsolation Containment Isolation PhaseA~Containment PurgeandExhaustZsolation Auxiliary Feedwater PumpsEssential ServiceMaterSystemy7.0Lessthanorequalto21~000/9+vO++

Lessthanorequalto3.0Z,essthanorequalto8.0I.essthanorequalto18.0PNotApplicable NotApplicable I,essghanorequalto48.0/13.004.D!.fferential PressureBetweenSteamLines-Hiha.SafetyZn)ection

{ECCS)b.ReactorTrip(fromSI)c.Feedvater Isolation d.Containment Isolation-Phase "A"e.Containment PurgeandExhaustIsolation f.Auxiliary Feedvater Pumpsg.Essential ServiceMaterSystemS7,~I.essthanorequal27.088/~~

LessthanorequalLessthanorequalLessthanorequal18.08/28.OfINotApplicable NotApplicable Lessthanorequal13.0f/48.0tt toto3.0to8.0toto5.SteamFlowinTwoSteamLines-HihCoincident wit.".TavLowLovgd7R?Plcd8/-ra.SafetyIn)ection (ECCS)b.C~d.ReactorTrip(fromSI)Feedwater Isolation Containment Isolation-Phase A"e~f.goContainment Purgear'xhaustIsolation Auxiliary Feedvater PumpsEssential SeFficeMaterSys'emSteamI,ineIsolation NotApplica'".'~

NotApplicable HWPLessthanorequalto13."COOKNUCLEARPLANTUNIT13/4328AMENDMXNT NONeXI' TASL~3.3'Continued EHCINE~SAFETYFM~iSRESPONSETLKSINITIATINC SICNALANDFUNCTIQN6.SteamLinePressuz'LovRESPONSETO%INSECQNDSa.SafetyIa)ection (ECCS)b.ReactorTrip(fzomSI)c.Feedvater Isolatioa d.Containmeac Isolation-Phase

'Ae.Containment Pur$eandExhaustIsolation f.Auxiliary Feedvacer Pumps$.Essencial ServiceQatarSystemh.SteamLineIsolation 7.ConaizaentPessue--High-High Laggchanorequal27.~>.%~LegsthanorequalLaggthanorequalLegsthaaorequal1$,0e/2$.0>>NotApplicable NocApplicable Lassthaaorequal14.0e/48.0>>

Legsthaaorequaltoto3.0to$.0CoCoto11.0a~b.C.Concainmenc SprayContainment Isolation-Phase bSteamLineIsolation Containment AirRecirculation FanLesschanorequalNotApplicable LasschanorequalLessthanozequalg5.oto~to10.0o600~0S.SearnCenerator "aterLevel--Hi h-Hiha.TurbineTr'pb.Feedvater Isolatioa 9.SteamCenerator PacerLevel-Lov-Lova.MotorDrivenAuxiliary Feedvater Pu"psb.TurbineDrivenAuxiliary Feedvacer P"~s10.4160voleEmerercv$usLossofVolageLassthanorequalto2.5Lassthanorequalto11.0a/go.o~~Lessthanozequalco60.0Lassthanorequalto50.0~a.AotorDrivenAuxiliary Feedvatar Mps11.Lossof~ginFeedvacer MŽosLessthanorequalco60.0a.MotorDrivenAuxiliary Feedvater PumpsLessthanorequalto60.0>12.ReactorCoolantPm~SusUndcrvolta

~a.TurbineDrivenAuxiliary Feedvacar PumpsLassthanorequalto60.0>CQQKNUC~~PLANTUNIT1'3/4329AMZNDKBtT NQ.49t229.168 TASLE3~3-5Continued TA31ZNOTlTXONeDieselgenerator startingandsequenceloadingdelaysnosinclu4ed.

Otfsitepaveravailable.

Responsetimelimitincludeaopeningofvalvestoestablish SZpathandattainment ofdischarge pseaaureforcentrifugal chargingpumps.~Dieselgenerator startingandsequenceloadingdelaysincluded.

Responsetimelimitincludesopeningofvalvestoestablish SIpathand~astainmenc ofdischarge pressureforcentrifugal chargingpumps.++Dieselgenerator startingandsequenceloadingdelaysincluded.

Responsetimelimitincludeaopeningofvalvestoestablish SZpathandattainment

'fdischarge pressureforcentrifugal

charging, SZ.andMRpumps.Sequential transferofchargingpumpauctionfromtheVCTtotheRVST(RVSTvalvesopen.thenVCTvalvesclose)isNOTincluded.

Dieselgenerator startingandsequenceloadingdelaysincluded.

Responsesineli&tincludeaopeningofvalvestoestablish SZpathandattainment ofdischarge pressureforcensrifugal chargingpumps.Sequential transferofchargingpumpsucsionfromtheVCTtotheRUST(ESTvalvesopen,thenVCTvalvesclose)iainclude4.

98Dieselgenerator startingandsequenceloadingdelaysNOTincluded.

Offsitepoveravailable.

Responsetimelimitincludeaopeningofvalvestoestablish SIpathandattainment ofdischarge pressureforcentrifugal chargingpumps.Sequensial transferofchargingpumpauctionfromtheVCTsotheRVST(RVSTvalvesopen,thenVCTvalvesclose)isincluded.

REsP((<rTINFsu$cOF'o4ofFEiTEScowl(s'4afgcF A'%vs(5.~~(P(=~~/~C'~~&V.

g(Qg(=('(=~(=+~~+

5'7~+((mg+pLIOS(=+~su~c.uR(Cy~+~~/+5+((+0M(/9T+@Sparse7(4(=(eavp(=SeP'En(ecoFv'+<egg(ac+((yi~pggrv(

o(=Q<acpp)QZ PR~~4coW/~~M.CCOKHUC~~PLhBT-UHZT13/43-30hHENDMEPZ NO~7'li8 TAbLE4.3-2ENCINEHtED SAFETYFEATUREACTUATION SYSTEMINSTRUMENTATION SURVEIILANCE REUZKDKNTSFUNCTIONAL UNIT1.SAFESTINJECTZOH, TURSZNETRIPsPEEDVATER ZSOIATIOH, ASDMOTORDRZVEHAUXZIZhRY FEEDVATER PUMPSTRIPACTUATING CHANNELDEVICEC~lHELCHANNELTUHCTIONAL OPERATIONAL CHECKCBETBEATTOE

'TESTTESTMODESINMHZCHSURVEILLANCE REUIREDa.MenuelZnitfstfon eeooseeeeoeeoeesosoSeePunctfone1Unft9eeeoeseeeoeeeseeeob.iutomstfc Actuetfoa LogicS.A.S.A.M(2)H.A.12,3,4c.Containment Press-ure-Highd.Pressurfser Press-ureeiAKe.DDferentfs1 PressurebetveeaSteamLines--High M(3)H.A.H.A.H.A.H'A.l.2,31,2,31,2.31,2.3PressureeeLolf (57rg&LBa/22.COHTAZNMEHT SPRATa.Mam+1Initiation soosooooeosssoooosoSeePunctfoae1Unit9esoeoooeoeeooooseeob.hutomstfc Actuation LogicH.A.H.A.M(2)H.A.1,2.3B4~c.Containment Press-ureeHfgheHfghM(3)H.A.1,2,3CQOXNUCLEARPIASTUHZT13/43e31Sl!EEOI!EET EO.fII,/It, TABLE4.3-2Continued ENCINEERED SAFETYFEA'TURED ACTUATION SYSTEMINSTRUMENTATION SURVEILLANCE REUZREMENTS FUNCTIONAL UNITTZPACTUATINtH QRNNELDEVICECHANNELQGQiNELFUNCTIONAL OPHbLTIONAL CHECKCBIIBEBTICH TESTTESTMODESZNRiICHSURVEILLANCE 4~SThBLINEISOLATION a.Manual---SeePunctional Unit9--------

b.Automatic Actuation Logicc.Containment Preaa-ur~--High-High d.SteamFlovinTvoSteamLinea--HighCoincident vithTavg~~Lov~LovN.A.N.A.M(2)M(3)N.A.H.A.H.A.1,2,31,2,31,2.3SergeuuB=WgE":SSqee-~S~~5.TURBINE TRIPANDFEEDVATER ISOLATION a.SteamCenerator VaterLevel--High-High6.MOTORDRIVENAUXZLIART FEEDVATER PUMPSa.SteamCenerator QatarLevel-LovLovH.A.1.A.IH2>Zl.2,31,2.3b.4krEuaLoaaofVoltagec.SafetyInjection d.LoaaofMainPeedPumpaH.A.1.A.1.A.H.A.M(2)N.A.H.A.H.A.1,2.31,2,31,2COOKNUCLEARPLANT-UN'3/43-33JHEHBHEHT HO.TI,g],(ET 3.4.2hminimumofonepressurirer codesafetyvalveshallbeOPERABLEwithaliftsettingof2485PSIC<<+t.*h.3MODES4and5,MEZQE:Pithnopressuri.ter codesafetyvalveOPERABLE:

a.'Immediately'uspend alloperations involving positivereactivity changes~andplaceanOPERABLERHRloopintooperation intheshutdowncoolingmode.b.Immediately renderallSafetyInfection pumpsandallbutonechargingpumpinoperable byremovingtheapplicable motorcircuitbreakersfromtheelectricpowercircuitwithinonehour.4.4.2Thepressurizer codesafetyvalveshallbedemonstrated OPERABLEperSurveillance Requirement 4.4.3.~eliftsettingpressureshallcorrespond toambientconditions ofthevalveatnominaloperating temperature andpressure.

~Forpurposesofthisspecification, additionofwaterfromtheRUSTdoesnotconstitute apositivereactivity additionprovidedtheboronconcentration intheRWSTisgreaterthantheminimumrequiredbySpecification 3.1.2.S.b.2 (MODE4)or3.1.2.7.b.2 (MODE5).D.C.COOK-3hGT'13/444AMENDMENT NOSE~1; 3.4.3Allpressuriser codesafetyvalvesshallbeOPERASLXrithaliftsettinI,of24ISPSZCgm.TChEIJXt:

hQXZQI:Withonepressuriaer codesafetyvalveinoperable, eitherrestor~theinoperable valvetoOPERASLEstatusvithin15minutesorbeinHOTSHUTDOWNvithin12hours.4.4.3Noadditional surveillance requirements otherthanthoserequiredbySpecification 4.0.S.'Theliftsettingpressureshallcorrespond toambientconditions ofthevalveatnominaloperating temperature andpressure.

COOIIUCLZhkPLhHTUHIT13/44ejAMEHDHEIT 50.440,

itleastonceper11monthsby:1.Verifying automatic isolation andinterlock actionoftheRHRsystemfromtheReactorCoolantSystemshentheReactorCoolantSystempressureisabovei00paig.2.Lvfaualinspection ofthecontainment sumpandverifying thatthesubsystem auctioninletsarenotrestricted bydebrfaandthatthesumpcomponents (trashracks,screens,etc.)ahovnoevidenceofstructural distressorabnormalcorroafon.e.htleastonceper1$months,duringshutdown, by:1.Verifying thateachautomatic valveintheflovpathactuatestoitacorrectpositionon~SafetyInjection teatsignal.2.Verifying thateachofthefolloving pumpsatattautomatically uponreceiptofasafetyinjection signal:a)Centrifugal chargingpumpb)Safetyinjection pumpc)Residualheatremovalpumpg>FFggwphc.

f.byverifying thateofthefollcnrfng pumpsdevelopstheindicated 4kool~zpresaure onzecizculation flovvhentestedpuzsuanttoSpecification 4.0.5.g2PO1.Centrifugal chargingpumpgreaterthanorequalto4405paf~13k&2.Safetyinjection pumpgreaterthanorequalto+40%paig-l5O3.4aidualheatremovalpumpgreaterthanorequalto404pafg-g.byverffyfng thecorrectpositionofeachmechanfcal atopforthefollovfng Emergency CozeCoolingSystemthrottlevalves:1.within4houzsfollowing completion ofeachvalvestrokingoperation ormaintenance onthevalvevhentheECCSsub-systemsarerequiredtobeOPSNLIl.COOKNUCLEARHAUNTUNIT13/45-5AHESDMEHT 80.44%,404,444,444~g{

EMERGENCY CORECOOLINGSYST.".'AS REFUELING MATERSTOPPAGE."A'KLIMITINGCONOI".ION;

~ROPE-"..-"":3'<

3.5.5Therefueling wa:e.storagetank(RMST)shallbeOPERABLEwith:Aminimumcontained volumeof350,.000gallonsofboratedwater.b.Between2400and2600"pmofboron,andc.Aminimumwatertemperature of~F.7'0APPLICABILI; Y:NODES1,2,3anC4.ACTION:Withtherefueling w=e~storagetankinoperable, restorethetanktoOPERABLEstatuswithin1hour".r"einaleastHOTSTANOBYwithin6hoursandinCOLOSHUTDOWNwithinthe;":.Nin"30hours.,SURVEILLANCE REUIRE.".E".'l.5 TheRWSTshallbede:-.onstrated OPE?wBL!!4.5.5II2.Verifying tneboronconcentration ofthewater.b.Atleastonceper2'oursbyverifying theRMSTtemperature.

a.Atleastonceper7daysby:l.Verifying

-.necontained boratedwatervolumeinthetank,and0.C.COOK-UNIT1;,'-"5-11Amendment No.Q,lll ELECTRICAL POWERSYSTEMSSURVEILLANCE REUIREMENTS Continued 4.8.1.1.2Eachdieselgenerator shallbedemonstrated OPERABLE:

inaccordance withthefrequency specified inTable4.8-1onaSTAGGERED TESTBASISby:~lEVMIHi30M~OSiudge.P4tI9%'6D~%26(l>LyZ4+0RfEQC&cyRr44<<r.2da.b.C.1.2.3.Verifying thefuellevelinthedaytank,57W~OQyVerifying thefuellevelinthefuelstoragetankVerifying thatthefueltransferpumpcanbesttedandthatittransfers fuelfromthestoragesystemtotedaytank,Verifying thatthedieselstartsfrom~~4condition and5.Verifying thatthegenerator isloadedtogreaterthanorequalto1750kwandthatitoperatesforgreaterthanorequalto60minutesandverifying thatthegenerator outputbreakertotheemergency busisOPERABLE, and6..Verifying thatthedieselgenerator isalignedtoprovidestandbypowertotheassociated emergency busses.Byremovingaccumulated water"~:1)Fromthedaytankatleastonceper31.daysandaftereachoccasionwhenthedieselisoperatedforgreaterthan1hour,and2)Fromthestoragetanksatleastonceper31days.Bysamplingnewfueloil"~inaccordance withtheapplicable guidelines ofASTM04057-81priortoaddingnewfueltothestoragetanksand1)Byverifying, inaccordance withthetestsspecified inASTM0975-81andpriortoaddingthenewfueltothestoragetanks,thatthesamplehas:hed>esgenerarstart1secopeormedaleastoeper1daysenginstartsrtherposeothisactionbeateducedceleranthatmhi1tsandarontsrommbentndt>oshathesesveillantests.Allotrsueillanctestingndcompsatoryratesrecoendedbthemanactureodiesenineaminimed.*~Theactionstobetakenshouldanyoftheproperties befoundoutsideofspecified limitsaredefinedintheBases.D.C.COOK-UNIT13/48-3AMENDMENT NO.125 ELECTPICAL POWERSYSTEMSSURVEILLANCE REUIREMEMTS Continued 2.Verifying thegenerator capability torejectaloadgreater"thanorequalto600kwwhilemaintaining voltageat4160+420voltsandfrequency at60+1.2Hz,3.Verifying thegenerator capability torejectaloadof3500kwwithoutexceeding 75Xofthedifference betweennominalspeedandtheoverspeed tripsetpoint, 4.Simulating alossofoffsitepowerbyitself,and:a)Verifying de-energization oftheemergency bussesandloadsheddingfromtheemergency busses,b)Verifying thatthe.dieselstartsontheauto-start signal,energizes theemergency busseswitlipermanently connected loadswithinseconds,energizes theauto-connected shutdownloadsthroughtheloadsequencer andoperatesforgreaterthanorequalto5minuteswhileitsgenerator isloadedwiththeshutdownloads.Afterloadsequencing iscompleted, thesteadystatevoltageandfrequency oftheemergency bussesshallbemaintained at4160.+420voltsand60+1.2Hzduring.thetest.5.Ve~ifying that,onaSafetyIn)ection actuation testsignal(withoutlossofoffsitepower),thedieselgenerator startsontheauto-start signalandoperatesonstandbyforgreaterthanorequalto5minutes,6.Simulating alossofoffsitepowerincon)unction withaSafetyInjection actuation testsignal,andby:a)Verifying de-energizationoftheemergency bussesandloadsheddingfromtheemergency busses,Verifying thedieselstartsonth>>auto-start signal,eneizestheemergency busseswithpermanently connected oaswithinseconds,energizes theauto-connected emergency (accident) loadsthroughtheloadsequencer andoperatesforgreaterthanorequalto5minuteswhileitsgenerator isloadedwiththeemergency loads.Afterloadsequencing iscompleted, thesteadystatevoltageandfrequency oftheemergency bussesshallbe4160+420voltsand60+1.2Hz.Thevoltageandfrequency shalTbemaintained withintheselimitsfortheremainder ofthistest,andD.C.COOK-UNIT13/48-5NENOHENTNO.125 ELECTRICAL POWERSYSTEMSSURVEILLANCE REUIREMENTS Continued b)Agitatethefueloilinthestoragetankawhilepumpingtheoilfromthebottomofthetankthrougha5-micronfilter,andbacktotheoppositeendofthetank.Threesuccessive samplesshallbetakenandanalyzedaccording toASTM02276-83.

Ifthecontaminant levelinanyofthesamplesisgreaterthan10mgperliter,theagitation, filtration, andsamplingprocesses shallberepeated.

Ifthecontaminant levelremainsabove10mgperliterafter3iterations, thedrainingandcleaningmethoddescribed insurveillance requirement 4.8.1.1.2.f.l.a shallbeemployed.

2)Performing aprecision leakdetection testtoverifythattheleakageratefromthefueloilsystemislessthanor,equalto.05gallonsperhour.sZ,~s-.~~~ea~~~.~3)...Starting bothdieselgenerators simultaneouslyduring

shutdown, andverifying thatbothdieselgenerators It'lErw~pygp<wOS>vvcrhgE-yrwCni+toVspaz.g~8ocy prfog/.2fFg.taeperormeateranymo>scatsonswccouaectgeesegeneraorinterdependence.

O.C.COOK-UNIT13/48-7AMENDMENT NO.125

34.1REACTIVITY CONTROLSYSTEMShhSES34.1.1ADORATION epamOL34.1.1.1and34.1.1.2SHUTDOWNÃ MARCINAsufficienC SHUTD(RNMARCINensuresChaC1)thereaccorcanbemadesubcritical fromalloperating cond.tions, 2)thereaccivicy transieacs associated vithpostulated accidentconditioas azecontrollable vithinacceptable limits,and3)thereactorvillbemaintained sufficiently subcricical toprecludeinadvertent criticality iatheshutdovncondition.

/.3SjiUTDOQN MARCINrequireme varythroughout coelifeasafuactionoffueldepletion, RCSboroncncentration, andRCST~Themostrestrictive conditioa occurscEOL,vithTataoj~ofdoperating Itemperature, andisassociate vithapostdated steamlinebreakaccidentandresulting uncontrolled Rcooldova.

Iatheanalysisofthisaccident,

.aminimumSHUTDONMARCINof~+HHDeltak/kisiaitially requiredtocontrolthereactivity transient andaucomatic'F isassumedtobeavailable.

VithTavglessthan200F,Chereactivicy transients resulting fromapostulated steamlinebreakcooldovnareminimalanda1%Deltak/kSHUTDOWNMARCINtprovidesadequateproceccion forthisevent~TheSHUTDOWNÃ MARCINrequirements arebaseduponchelimiciagconditions described aboveandareconsistent vithFSARsafetyanalysisassumptions.

34.1.1.3EORONDITIONAainhaaaflovrateofatleast2000CPMprovidesadequatemiming,preventsstratification andensuresChatreactivicy changesvillbegradualduringboronconcentratioa reductions iatheReactorCoolantSystem.Aflovrateofatleast2000CPMvillcirculate aaequivalent ReactorCoolantSyscemvolumeof12,612plusorminus100cubicfeetiaapproximately 30minutes.Thefreactivicy changex'atcusociated vithboronreductions villtherefore bevithinthecapability foroperatorrecognition andcoatrol.34.1.1.4MODERATOR TEMPGULTURE COEFFICIBIT MTCThelimitations onMTCareprovidedtoensurethattheassumptions usedintheaccidentandtransient analysesremaiavalidthrougheachfuelcycle.ThesuzveQlancc requirement formeasurement oftheETCatthebeginning, andnearthecndofeachfuelcycl>>isadequatetocoafizmthcMTCvaluesincethiscoefficient changesslovlyducprincipally tothereductioa inRCSboronCOOKNUCLMPESTUNIT153/411ANQiDMVTNO.74i08~>->

3Ae3/44>qt'(CORC~~')'pmrgS5+gg+$&lfcV5/SI.ITheplantisdesigned=ooperate'2allreactorcoolantloopsinoperation, andmaintain5NRabove~during allno~1operations andant'cpatedtransients.

Alossofflouintvoloopsvillcauseareactort"ipLfoperating aboveP-7(11percentofRATEDTHER'OWER) vhilealossofflewinoneloopvillcauseareactortz'pifoperating aboveP-8(31percentof.RATEDTHEq.'LAI.

POWER).In!LODE3,as'nglareactorcoolantloopprovidessuffic'ant heatremovalcapability forremovingdecayheat;hovavez,singlefailureconsiderations requizethattvoloopsbeOPMLE.Threeloopsazaraqu'"edtobeOP~LEandtooperate'fthecontrolrodsazacapableofvithdzaval andt~ereactor-z'pbreakersarec'osed..ne requirement assuresadequateDh3Rmazgf.n'ncheeventofanuncontrolled rodv'=.".dza-a'.".

-".'s=ode.InLODES4and5,asinglereactorcoolantloopozRHRloopprovidessufficient heacremovalcapabiLity forremovingdecayheat;butsingleailuzaconsideracions requirethatatleasttvoloopsbeOPBABLE.Thus,ifthereactocoolantloopsarenotOPMI=,thisspeciicationzequiescvoRHRLoopstobeOPHABIZ.Thaoperation ofonaReactorCoolantPumporoneRBRpumpprovidesadequateflovtoensuzamixing,preventstratification andproducegradualreactivicy changesduringboronconcanczacion reductions intheReactorCoolantSystem.Thareaccivity changerataassociated vithboronreduction vi11,cherefoze, bevithinthecapabilicy ofoperatorrecognition andcontroLTherescrictions onstartingaReactor,CoolancPumpbelovP-7vithoneormoreRCScoldlegsLassthanorequalto1527azaprovidedtopravencRCSpressurecransiencs, causedbyenergyadditions fzomchasecondary system,vhichcouldexceedthelimitsofAppendixCtoLOCFRPaz-50.TheRCSvil'eprotected againstoverprassure transients andvillnocexceedthelimitsof,AppendixCbyeither(1)restzicting thavatazvolumeinthepressurizer andtherebyproviding avolumefortheprimarycooLanttoexpandintooz(2)byrestricting staztingofthaRCP'stovhen,thesecondazy vatartemperature ofbeachsteamgeneratoz's lessthan50FaboveeachoftheRCScoldlegtemperacuzas

~COOKh"QCI"-Qt, P~~Z-4.lTT13/44A~~'fDl".~'iT NO.N,fa3,167 ardEf34.5.5RE~ACVOTERSTOay:tmmgEPCACZghTtt1>skihTheOPERAbILITY oftheRVSTaspartoftheECCSensuresthatsufficient negativereactivity isin)ectedintothecoretocounteract anypositiveincreaseinreactivity causedbyRCSsysteacooldcnm, andensuresthatasufficient supplyof.borate4vaterisavailable forin)ection bytheECCSintheeventof~LOCA.ReactorcoolantsysteNcooldovacanbecausedbyinadvertent depressurisation, alossotcoolantaccidentorasteamlinerupture.ThelimitsonRQSTminhtuavolumeandboronconcentration ensur~t}1)sufficient vaterisavailable vithincontainment toermitrecirculation coolinflovtothecore.and2)~reaorvilremascticalntheoldcoditioolongaxngofeRQSandeRCSa'terlumesitha:ctroldsiertedcetorthestrtiveontroesse1.Theseassumptions areconsistent viththeIANNAanalyses.

Thecontained vatervolumelimitincludesanallovance forvaternotusablebecauseoftankdischarge linelocationorotherphysicalcharacteristics.

Thelimitsoncontained vatervolumean4boronconcentration oftheESTalsensureapHvalueofbetveen7.6and9.5forthesolutionrecirculated vithicontainment afteraLOCA.ThispHbandminimises theevolution otiodineanmi.nimires theeffectofchloridean4causticstresscorrosion onmechanical systemsandcomponents.

TheECCSanalysestodetermine F~limipinSpecifications 3.2.2and3.2.6assumedaRVSTvatertemperature of70F.Thistemperature valueoftheRMS'aterdetermines thatofthesprayvaterinitially delivere4 tothecontainment following LOCA.Itisoneofthefactorswhichdetermines thecontainment back-pressure intheECCSanalyses, performed inaccordance viththerovisions of10CFR50.46andhendixKto10CFR50.~aueothminiumRVtempaturenPccalecifition5.5hbeenconseativechae4to0Ftincrea~theonsisncybeeenits1~gZTEnd2.The1erRTtempratureresultinloerconainmenpressrefro,cntaintspyan4afe4sflassdtoitthbreak.Lovecotainmetpresurereultsincresedflvrestancefsteexitgtheorerebevin<<floeandireasiPCT.INSERTAthereactorwillremainsubcritical inthecoldcondition following aLOCAassumingmixingoftheRWST,RCS,ECCSwater,andothersourcesofwaterthatmayeventually resideinthesump,withallcontrolrodsassumedtobeout.COON.'asm PmN-VÃIT1I3/453AHRf5lcÃT M.fg,li CONTAI~~3a.g.1.4-PRESSVREThelimitations oacontaiaaant internalpressureensurethat1)th+containment structure isprevented froaexceeding itsdesignnegativere!suredifferential withrespect,totheoutsideatmosphere ofdpsigand2)thecontainment peakpressuredoesnotexceedthedesignpressureof12psigduringLOCAconditions.

<<.+9Themaximumpeakpressureresulting fromaLOCAeventiscalculated tobe~~psig,whichincludes0.3ps'gforinitialpositivecontainment pressure.

34615AIRTeuaERAL%E.he1'=isa"."..sonconta'"."cn averageairtemperature ensurethat1)thecontainment

."=ass's'=.=edtoani..i"ialmasssufficiently lowto"".eventexceeding t..edesign"ressredringL"CAcondi"'ons and2)=hea.".biens a'r=epara='e='.es.-.==ex"eed:hat temperature allowable orthecont'mucus dutyratingspecified fsrcqu'pment andinstrumentation locatedwithinccnaŽen".~eccntainment pressre"rsnsient

'.ssensit've:o

-'.~'ni='allv con=a'ned air-.ass""itga'"A...".e santa'ned a'ss'"..c"eases with"ecreasi.".z

=emperatu...-.e

'ower=e"perature

'-.=6Ocwilllipellpressure0++s.,'h'lesshan.~co'n'tainmen't designpressureof12"sig..he=pertemperature limitI;.encesthepeakac"identtemperatres'ighlydr'ga'CA;however=hislimitisbasedpr=ar'.vuponequ'pment "ro"ec='on andant'cipa=ed

==era"ing conditions.

Boththeupperandlovertemperature limitsarecons's"ent withtheparameters

.sedin=heaccidentanalyses.

3n.6.1.6C"NTAI>~ENT VESSELS.RUCT';RAL INTEGRITY Thisli='cation ensuresthatthestructural integrity ofthecontainment steelvesselwillbemaintained comparable totheori-naldesignstandards forthelifeofthefacility.

Structural i-..=egrity isrequi=edoensurethar,(1)thesteellinerremainsleaktightx-..d(2)theconcretesurrounding thesteellinerremainscapable.=providing externalmissileprotection forthesteellinerandr-"iation shielc;;.g intheeventofaLOCA.Avisualinspection incon]u"..stion withTypeAleakagetestsissufficient todemonstrate thiscapabiity.COOKNUCLEARPLANTUNIT1B3/46-2AMENDMENT NO,1"6

Inaccordance withthecoderequirements specified inSection4,1.6oftheFSAR,withallowance fornormaldegradation pursuanttotheapplicable Surveillance Requirements, b.C.M>LUhK5.4.25.5Forapressureof2485psig,andForatemperature of650'F,exceptforthepressurizer whichis680'F.HPF'Ro)OehVKLy IZ,,+&4C~Slc.FeE7APO%Sl&PAPQE~FRdmgMMPCu~z~AuldIl>55/Cg8ICF'EZ7AT3OL'ran~GEAIEA!lb@

P~~iud.Tha'MIfhIy~hie-fees atanominalT~of70'F.5.5.1Theemergency corecoolingsystemsaredesignedandshallbemaintained inaccordance withtheoriginaldesignprovisions contained inSection6.2oftheFSARwithallowance fornormaldegradation pursuanttotheapplicable Surveillance Requirements, withoneexception.

Thisexception istheCVCSboronmakeupsystemandtheBIT.5.65.6.1.1Thespentf'uelstorageracksaredesignedandshallbemaintained with:a.A~equivalent tolessthan0.95whenfloodedwithunborated water.b.Anominal8.97inchcenter-~nter distancebetweenfuelassemblies placedinthestorageracks.C.Thefuelassemblies will'beclassified asacceptable forRegion1,Region2,orRegion3storagebasedupontheirassembiyaverageburnupversusinitialnominalenrichment.

Cellsacceptable forRegion1,Region2,andRegion3assemblystorageareindicated inFigures5.6-1and5.6-2.Assemblies thatareacceptable forstorageinRegion1,Region2,andRegion3mustmeetthedesigncriteriathatdefinetheregionsasfollows:COOKNUCLEARPLANT-UNIT15-5AMENDMENT NO.l,,169CORRECTED PAGE ATTACHMENT 7TOAEP:NRC:1207 DESCRIPTION OFANALYSESPERFORMED BYWESTINGHOUSE ELECTRICCORPORATION FORDONALDC.COOKNUCLEARPLANTUNIT2 Vantage5ReloadTransition SafetyReportfoIDonaldC.CookNuclearPlantUnit2 6.0SUhB~YOPTECHNICAL SPECIFICATIONS CHANGESTable6.1presentsalistoftheTechnical SpeciGcations changes.ThechangesnotedinTable6.1aregivenintheproposedTechnical Specifications pagechangesinAppendixA.43 TABLE6.1SUMMARYOFTECHNICAL SPECIFICATIONS CHANGESSECIIONPAGECHANGEREASONFORCHANGE1.0,AddCOLRCOLRimplementation pgItoindex1.12a,pg1-3AddCOLRCOLRimplementation Figure2.1-1,pg2-2RevisedsafetylimitsReanalysis supportsVANTAGE5reload2.2.1,pg2-5DesignQowchange&tripsetpointChangeindesignQowduetoVANTAGE5fuelreload,RTDPimplementation Table2.2-1,pg2-7&24ReviseOvertemperature Reanalysis supportsVANTAGE5reloadhTlimitsTable2,2-1,pg2-9ReviseOverpower b.TlimitsReanalysis supportsVANTAGE5reload2.1.1Bases,pgB2-1&B2-2UpdatetobasesVANTAGE5fuelreloadandCOLRimplementation (relocation ofFNgH)2.1.1Bases,pgB2-4UpdatetobasesVANTAGE5fuelreloadanddeleteCycle6speciQcinformation 2.1.1Bases,pgB2-5RevisebasesReanalysis supportsVANTAGE5reload2.1.1Bases,pgB2-7RevisebasescircuitbreakertimeReanalysis supportsVANTAGE5reload3/4.1.1.1, pg3/41-1&1-2DecreaseshutdownmarginReanalysis withreducedSDM3/4.1.1.2, pg3/41-3&1-3bDecreaseshutdownmarginReanalysis withreducedSDM.ChangetoVfestinghouse dilutionaccidentmethodology TABLE6.1SUMMARYOFTECHNICAL SPECIFICATIONS CHANGES(continued)

SECIIONPAGECHANGEREASONFORCHANGE3/4.1.1.4, pg3/41-5&3/41-6MTCrelocated toCOLR&revisedEOLlimitVANTAGE5fuelreloadandCOLRimplementation (relocation ofMTC)3/4.1.1.5, pg3/41-7Minimumtemperature forsurveillance req.Reanalysis withreducedtemp3/4.1.2.3, pg3/41-11Changech.pumpdischarge headMakeconsistent withtheanalysis3/4.1.2.4, pg3/41-12Changech.pumpdischarge headMakeconsistent withtheanalysis3/4.1.2.7, (pg3/41-153/4.1.2.8, pg3/41-16Change80OFto70GFChangevolumefrom5650to7715gallons&change80Fto'70FMakespecconsistent withtheanalysislimitMakespecconsistent withtheVANTAGE5reloadanalysislimittoaccommodate reducedrodworthandmanagement flexibility 3/4.1.3.1, pg3/41-19Deletereference toFig.3.1-1COLRimplementation 3/4.1.3.4, pg3/41-23Changeroddroptimefrom2.2to2.7secRelocatestepswithdrawn toCOLRMakespecconsistent withtheanalysislimit&COLRimplementation 3/4.1.3.5, pg3/41-&Relocateshutdownrodinsertion limitstoCOLRCOLRimplementation (relocation ofshutdownrodinsertion limits)45

TABLE6.1SUMMARYOFTECHNICAL SPECIFICATIONS CHANGES(continued)

SECTIONPAGECHANGEREASONFORCHANGE3/4.19.6, pg3/41-25Relocatecontrolrodinsertion limitstoCOLRCOLRimplementation (relocation ofcontrolrodinsertion limits)3/4.19.6, pg3/41-26DeleteQgure3.1-1COLRimplementation 3/49.2.1, pg3/42-1&2-3RelocateaxialQuxdifference limitstoCOLRCOLRimplementation (relocation ofAFDlimits)3/4.3.2.1, pg3/42-4RelocateaxialQuxdifference allowable deviation Fig.toCOLRCOLRimplementation (relocation ofAFDallowable deviation) 3/43.2.2, pg3/42-5RelocateFqlimitstoCOLRCOLRimplementation (relocation ofFqlimit)3/4.3.2.2, pg3/424,2-8a&24bRelocateK(Z)&V(Z)QgurestoCOLRCOLRimplementation (relocation ofFqlimit)3/4.3.2.3, pg3/42-9RelocateFN~limitstoCOLR'OLRimplementation (relocation ofFNgHlimit)3/4.25.1, pg3/42-15ReformatDNBspecChangeDNBparameter valuesandaddlowTavgwindowAdoptplannedCookNuclearPlantUnit1specformatconsistent withVANTAGE5reload3/4.2.5.1, pg3/42-16&2-17&2-18Deletetables3.2-1and3.2-2Delete3.2.5.2AdoptplannedCookNuclearPlantUnit1specformatNotrequired TABLE6.1SUMMARYOFTECHNICAL SPECIFICATIONS CHANGES(continued)

SECTIONPAGECHANGEREASONFORCHANGE3/4.3.2.6, pg3/42-19RelocateFglimitsCOLRimplementation (relocation toCOLR-'fFglimit)ChangeddeGnition ofFgWestnghouse CAOCmethodology Table33-2,pg3/43-9&3-10ChangedandaddedRPSMakeconsistent withtheanalysisresponsetimeslimitsTable3.4-4,pg3/43-25ChangeESFASsetpointMakeconsistent withanalysisTable3.3-5,pg3/43-26&3/43-27&3/43-28ChangedESFresponsetimeMakeconsistent withtheanalysistimeslimits3/4.4.1.2, pg3/44-2&4-3ReducenumberofRCPsMakeconsistent withtheanalysisrequiredoperableinlimitsmode33/4.4.4,pg3.44-6Changewatervolumefrom62%to92%Makeconsistent withtheanalysislimit3/4.4.6.2, pg3/44-15&3/44-16Controlled leakageintermsofresistance Consistent withanalysis3/45.1b,pg3/45-1Reviseminimumcontained boratedwatervolume&min/maxcover-pressure Makeconsistent withanalysislimits(03/4.5.2.f, pg3/45-5RevisedSIpumpperformance Reanalysis withdegradedSIperformance 47 TABLE6.1SUMMMRYOFTECHNICAL SPECIFICATIONS CHANGES(continued)

SECTIONPAGECHANGEREASONFORCHANGE3/45.2.h, pg3.45-6RevisedSIpumpQowbalancelimitsAdoptlimitssimilartoCookNuclearPlantUnit13/4$.5,pg3/45-11ReduceRWSTmintempMakespecconsistent withto70'Fanalysislimit3/4.1.1.1, pgB3/4'1-1DecreaseshutdownmarginReanalysis withreducedshutdownmarginB3/4.1,pgB3/41-3Reviseconcentrations andvolumesMakespecconsistent withanalysislimitsB3/4.2.1,pgB3/42-1&2-2&2-3RevisetoreQectCOLRimplementation ChangedtoWCAP-8385 COLRimplementation

'relocation ofAFDlimits)'estinghouse methodology B3/4.2.2&3,pgB3/42-4thru2-4bRevisedtoreQectCOLRimplementation

&VANTAGE5reload'ANTAGE5reloadT-HanalysisandCOLRimplementation (relocation ofFgandFgHlimits)B3/4.M,pgB3/42-5B3/4.2.6,pgB3/42-5RevisetoreQectreducedtempDNBlimitRevisetoreQectCAOCcontrolReanalysis withreducedtemp\Makespecconsistent withanalysisB3/4.55,pgB3/45-3ReduceRWSTtempto70OFMakespecconsistent withtheanalysislimitB3/4.7.1,pgB3/47-1ReformatvalveliftcriteriaMakeconsistent withtheanalysislimit TABLE6.1SUMMARYOFTECHNICAL SPECIFICATIONS CHANGES(continued)

SECfIONPAGECHANGEREASONFORCHANGE3.4.9.1,pgB3/49-1Deletereference torefueling reactivity calcsat2000ppmReanalysis ofrefueling reactivity at2400ppmboron6.9.1.11, pg6-18AddCOLRtosection6COLRimplementation 49 EHERCENCY CORECOOLENCSYSTE.'fS SURVEILLANCE REUZREHENTS Continued d.Atleastonceper18monthsby:1.Verifying automat'c isolation andinterlock actionoftheRHRsystemfromcheReactorCoolantSystemMhentheReactorCoolantSystempressureisabove600psig,*2.hvisualinspeccion ofthecontainment sumpandverifying thacthesubsystem suctioninletsarenocrestricted bydebrisandtha-thesumpcomponents (trashracks,screens,etc.)shoenoev'denceofsttuctural distressorcorrosion.e.Atleastonceper18months,duringshucdovn, by:1.Vexifying thateachautomatic valveinthefloepathactuatescoicscorrectpositionqpaSafecyInfection testsignal.2,Verifying thateachofthefollowing pumps"startautomaticallv uponreceiptofasafetyXnJection testsignal:a)Centrifugal chargingpumpb)Safetyinjection pumpc)sidualheatrovalpumpJjg~fi4Byverifinggthateachothefollowing pumpsdevelopstheindicate4+ee~hssureonrecircu'ation fl.oushentestedpursuantticaton4.0.5:1.Centrifugal chargingpump2.SafecyIn)eccion pump3.ResidualheatremovaLpump~ZarOpseud$8'$foal~40Psig.Byverifying thecorrectposicionofeacmechanical stopfo"thefolloving Emergency CoreCoolingSystemthroctlevalves:1.within4hoursfolloving complecion ofeachvalvestrokingoperation ormaintenance onthevalveMhentheECCSsubsyste'ms arerequiredtobeOPERABLE,

  • Theprovisions ofSpecification 4.0.7areapplicable.

D.C.COOK-UNTT23/45-5Amendment 8.~89 APPENDIXBNON-LOCAANALYSESFORTHEDONALDC.COOKNUCLEARPLANTUNIT2TRANSITION TO17X17VANTAGE5FUEL B9.11RuptureofaSteamline (Steamline Break)B3.11.1Introduction Althoughthenoloadtemperature doesnot,change duetotheplantreratingandVANTAGE5fuel,theimpactofthevariousfuelparameter changesaswellasvarioustemperature andpressureoperation wasaddressed.

Also,thenominallowsteampressuresetpointforsteamline isolation andsafetyinjection actuation isrevised(loweredfrom600psigto500psig)toprovideoperating margin.Assuch,theruptureofasteampipeeventwasanalyzed.

Includedintheanalysisarethedesignchangesassociated withtheVANTAGE5transition andothermodifledsafetyanalysisassumptions asdiscussed inSectionB.1.Aruptureofasteampiperesultsinanuncontrolled steamreleasefromasteamgenerator.

Thesteamreleaseresultsinaninitialincreaseinsteamflowwhichdecreases duringtheaccidentasthesteampressurefalls.TheenergyremovalfromtheReactorCoolantSystemcausesareduction ofcoolanttemperature andpressure.

Inthepresenceofanegativecoolanttemperature coefficient, thecooldownresultsinareduction ofcore'shutdown margin.IfthemostreactiveRCCAisassumedstuckinitsfullywithdrawn

position, thereisanincreased possibility thatthecorewillbecomecriticalandreturntopower.Areturntopowerfollowing asteampiperuptureisapotential concernmainlybecauseofthehighhotchannelfactorswhichexistwhenthemostreactiveRCCAisassumedstuckinitsfullywithdrawn position.

Thecoreisultimately shutdownbyboricaciddelivered bytheEmergency CoreCoolingSystem.Theanalysisofasteampiperuptureisperformed todemonstrate that:A.AssumingastuckRCCA,withorwithoutoffsitepower,andassumingasinglefailureintheengineered safetyfeatures, thereisnoconsequential damagetothecoreandthecoreremainsinplaceandintact.B.AlthoughDNBandpossiblecladperforation following asteampiperupturearenotnecessarily unacceptable, thefollowing analysis,'n fact,showsthatnoDNBoccursforanyruptureassumingthemostreactiveRCCAstuckinitsfullywithdrawn position.

~~~B.3.11.2MethodofAnalysisTheanalysisofthesteampiperupturehasbeenperformed todetermine:

A.ThecoreheatQuxandRCStemperature.;and pressureresulting fromthecooldownfollowing thesteamlinebreak.TheLOFIRANCode(Reference 5)hasbeenused.B.Thethermalandhydraulic behaviorofthecorefollowing asteamlinebreak.Adetailedthermalandhydraulic digitalcomputercode,THINC,hasbeenusedtodetermine ifDNBoccursforthelimitingcoreconditions computedinitemAabove.Thefollowing conditions wereassumedtoexistatthetimeofamainsteamlinebreakaccident:

A.End-of-life shutdownmargin(19%6k/k)atnoload,equilibrium xenonconditions, andthemostreactiveRCCAstuckinitsfullywithdrawn position.

8-94 B.Anegativemoderator temperature coefficient corresponding totheendwf-life roddedcorewiththemostreactiveRCCAinthefullywithdrawn position:

Thevariation ofthecoefflcient withtemperature andpressurehasbeenincluded.

Thekeffversustemperature at1050psiacorresponding tothenegativemoderator temperature coefficien usedplustheDopplertemperature effect,isshowninFigureB3-55.TheDopplerpowerfeedbackassumedforthisanalysisispresented inFigureB3-56.Thecoreproperties associated withthesectornearesttheaffectedsteamgenerator'nd thoseassociated withtheremaining sectorwereconservatively combinedtoobtainaveragecoreproperties forreactivity feedbackcalculation.

Further,itwasconservatively assumedthatthecorepowerdistribution wasuniform.Thesetwoconditions causeunderprediction ofthereactivity feedbackinthe'highpowerregionnearthestuckrod.Toverifytheconservatism ofthismethod,thereactivity aswellasthepowerdistribution wascheckedforthelimitingconditions forthecasesanalyzed.

Thiscoreanalysisconsidered theDopplerreactivity fromthehighfueltemperature nearthestuckRCCA,moderator feedbackfromthehighenthalpywaternearthestuckRCCA,powerredistribution andnon-uniform coreinlettemperature effects.Forcases'nwhichsteamgeneration occursinthehighfluxregionsofthecore,theeffectofvoidformation wasalsoincluded.

Itwasdetermined thatthereactivity employedinthekineticsanalysiswasalwayslargerthanthereactivity calculated including theabovelocaleffectsforthestatepoints.

Theseresultsverifyconservatism; i.e.,underprediction ofknegativereactivity feedbackfrompowergeneration.

GMinimumcapability forinjection ofboricacid(2400ppm)solutionfromtheRWSTcorresponding tothemostrestrictive singlefailureinthesafetyinjection system.TheEmergency CoreCoolingSystem(ECCS)consistsofthefollowing systems:1)thepassiveaccumulators, 2)thelowhead.safetyinjection (residual heatremoval)system,3)theintermediate headsafetyinjection system,and4)thehighheadsafetyinjection (charging) system.Onlythehighheadsafetyinjection (charging) systemandthepassiveaccumulators aremodeledforthesteamlinebreakaccidentanalysis.

Centrifugal Chargingpumpflowdegradation of10%wasassumed.Themodelingofthesafetyinjection systeminLOFTRANisdescribed inReference 5.FigureB3-57presentsthesafetyinjection flowratesasafunctionofRCSpressureassumedinthe8-95 analysis.

TheQowcorresponds tothatdelivered byonechargingpumpdelivering itsfullQowtothecoldlegheader.Nocredithas,.been takenforthelowconcentration boratedwaterwhichmustbesweptfromthelinesdownstream oftheRWSTisolation valvespriortothedeliveryofboricacidtothereactorcoolantloops.Forthisanalysis, aboronconcentration of0ppmfortheboroninjection tankisassumed.ForthecaseswhereofBitepowerisassumed,thesequenceofeventsinthesafetyinjection systemisthefollowing.

Afterthegeneration ofthesafetyinjection signal(appropriate delaysforinstrumentation, logic,andsignaltransport included),

theappropriate valvesbegintooperateandthehighheadsafetyinjection pumpstarts.In27seconds,thevalvesareassumedtobeintheirGnalpositionandthepumpisassumedtobeatfullspeedandtodrawsuctionfromtheRWST.Thevolumecontaining thelowconcentration boratedwaterissweptintocorebeforethe2400ppmboratedwaterreachesthecore.Thisdelay,described above,isinherently includedinthemodeling.

Incaseswhereoffsitepowerisnotavailable, anadditional 10seconddelayisassumedtostartthedieselgenerators andtocommenceloadingthenecessary safetyinjection equipment ontothem.D.Designvalueofthesteamgenerator heattransfercoefQcient.

KFourcombinations ofbreaksizesandinitialplantconditions havebeenconsidered indetermining thecorepowertransient whichcanresultfromlargeareapipebreaks.aCompleteseverance ofapipedownstream ofthesteam"Qow restrictor withtheplantinitially atnoloadconditions andallreactorcoolantpumpsrunning.b.Completeseverance ofapipeinsidethecontainment attheoutletofthesteamgenerator (upstream ofthesteamQowrestrictor) withthesameplantconditions asabove.~c.Case(a)abovewithlossofoff-sitepowersimultaneous withthegeneration oftheSafetyInjection Signal(lossofACpowerresultsinreactorcoolantpumpcoastdown).

d.Case(b)abovewiththelossofoffsitepowersimultaneous'with theSafetyInjection Signal.AfifthcasewasanalyzedtoshowthattheDNBRremainsabove.thelimitvalueintheeventofthespuriousopeningofasteamdumporrelief,valve.e.Abreakequivalent toasteamQowof265lbspersecondat1100psiafromonesteamgenerator withoffsitepoweravailable.

F.Powerpeakingfactorscorresponding toonestuckRCCAaredetermined atendofcorelifeassumingnon-uniform coreinletcoolanttemperatures.

Thecoldestcoreinlettemperatures areassumedtooccurinthesectorwiththestuckrod.Thepowerpeakingfactorsaccountfortheeffectofthelocalvoidintheregionofthestuckcontrolassemblyduringthereturntopowerphasefollowing thesteamlinebreak.Thisvoidinconjunction withthelargenegativemoderator coeKcient partially offsetstheeffectofthestuckassembly.

Thepowerpeakingfactorsdependuponthecorepower,temperature,

pressure, andflow,andarethusdifferent foreachcasestudied.Theanalysesassumedinitialhotshutdownconditions attimezerosincethisrepresents themostpessimistic initialcondition.

Shouldthereactorbejustcriticaloroperating atpoweratthetimeofasteamlinebreak,-.'the reactorwillbetrippedbythenormaloverpower protection systemwhenpowerlevelreachesatrippoint.Following atripatpowerthereactorcoolantsystemcontainsmorestoredenergythanatno-load,theaveragecoolanttemperature ishigherthanatno-loadandthereisappreciable energystoredinthefuel.Thus,theadditional storedenergyisremovedviathecooldowncausedbythesteamlinebreakbefore.theno-loadconditions ofRCStemperature andshutdownmarginassumedintheanalysesarereached.Aftertheadditional storedenergyhasbeenremoved,thecooldownandreactivity insertions proceedinthesamemannerasintheanalysiswhichassumesno-loadconditions attimezero.(Inaddition, sincetheinitialsteamgenerator waterinventory isgreatestatno-load,themagnitude anddurationofRCScooldownaremoreseverethansteamlinebreaksoccurring atpower.8-97 G.Incomputing thesteamflowduringasteamlinebreak,theMoodyCurve(Reference 11)for~~fL/D=0isused.H.Thetotaldelaytimeassumedforthesteamline isolation is11secondsfromreceiptofactuation signaLThe11secondsteamline isolation timeincludesvalveclosuretime,andelectronics andsensordelay.TheTechnical Speciflcations requireamaximum8secondvalveclosuretime.Forbreaksdownstream oftheisolation valves,closureofallvalveswouldcompletely terminate theblowdown.

Foranybreak,inanylocationfollowing steamline isolation, nomorethanonesteamgenerator wouldexperience anuncontrolled blowdownevenifoneoftheisolation valvesfailstoclose.Plantcharacteristics andinitialconditions areshowninTableB.2.4.B.3.11.3ResultsThelimitingcaseforCasesathroughewasshowntobethedouble-ended rupturelocatedupstreamoftheflowrestrictor withoKsitepoweravailable (caseb).TableB.3-10liststhelimitingstatepoint forthisworstcase.Theresultspresented areaconservative indication oftheeventswhichwouldoccurassumingasteamlinerupture.FiguresB.3-58throughB3-60showtheRCStransient andcoreheatfluxfollowing amainsteamlinerupture(complete severance ofapipe),.upstream oftheflowrestrictor atinitialno-loadconditions.

Offsitepowerisassumedavailable sothatfullreactorcoolantflowexists.Thetransient shownassumesanuncontrolled steamreleasef'romonlyonesteamgenerator.

Shouldthecorebe'critical atnearzeropowerwhentheruptureoccurstheinitiation ofsafetyinjection byhighdifferential pressurebetweenanysteamline andtheremaining steamlines orbylowsteamlinepressureintwosteamlines willtripthereactor.Steamreleasefrommorethanonesteamgenerator willbeprevented byautomatic tripofthefastactingisolation valvesinthesteamlinesbyhigh-high

'ontainment pressuresignalsorlowsteamline pressureorhighsteamflowcoincident withlow-lowT-avg.Evenwiththefailureofonevalve,releasefromtheothersteamgenerators isterminated bysteamline isolation whiletheonegenerator blowsdown.Thesteamlinestopvalvesareassumedttobefullyclosedinlessthan11secondsfromreceiptofaclosuresignal.B-98 AsshowninFigureB3M,thecoreattainscriticality withtheRCCAsinserted(withthedesignshutdownmarginassumingonestuckRCCA)beforeboronsolution(2400ppmfromRWST)enterstheRCS.Apeakcorepowerlessthanthenominalfullpowervalueisattained.

Thecalculation assumestheboricacidismixedwith,anddilutedby,thewaterflowingintheRCSprior.toenteringthereactorcore.Theconcentration aftermixingdependsupontherelativeflowratesintheRCSandinthesafetyinjection system.Thevariation ofmassflowrateintheRCSduetowaterdensitychangesisincludedinthecalculation asisthevariation offlowrateinthesafetyinjection systemduetochangesintheRCSpressure.

Thesafetyinjection systemflowcalculation includesthelinelossesinthesystemaswellasthepumpheadcurve.Theassumedsteamreleaseforanaccidental depressurization ofthemainsteamsystem(casee)isthemaximumcapacityofanysinglesteamdump,relief,orsafetyvalve.Safetyinjection isinitiated automatically bylowpressurizer pressure.

Operation ofonecentrifugal chargingpumpisassumed.Boronsolutionat2400ppmenterstheRCSproviding sufficien negativereactivity topreventcoredamage.Thetransient isquiteconservative withrespecttocooldown, sincenocreditistakenfortheenergystoredinthesystemmetalotherthanthatofthefuelelementsortheenergystoredintheothersteamgenerators.

Sincethetransient occursoveraperiodofabout5minutes,theneglected storedenergyislikelyforthiseventtohaveasigniflcant effectinslowingthecooldown.

TheDNBtransient isboundedbythelimitingcaseforasteamline rupture.TheDNBanalysisforthelimitingcase(double-ended rupturelocatedupstreamoftheflowrestrictor) showedthattheminimumDNBRremainedabovethelimitvalue.TheDNBRdesignbasislimitforthehypothetical steamline breakeventis1.45.Thepressures forthiseventfallinthelowpressurerange(500-1000 psia)wheretheW-3basedDNBcorrelation isusedwitha1.45limitDNBR.Thisdesignlimitforlowpressureapplications oftheW-3correlation hasbeenapprovedbytheNRCinReference 15.Althoughthelowpressurelimitwasapprovedinconjunction withWCAP-9227-NP, whichisnotreferenced intheCookNuclearPlantUnit2UFSAR,theSERisanapplicable reference forreloaddesigns.Thecalculated sequenceofeventsforthelimitingcase(doublewnded rupturelocatedupstreamoftheflowrestrictor) areshowninTableB.3-11.8.99

B3.11.4Conclusions

~~~TheanalysishasshownthatthecriteriastatedearlieraresatisGed.

AlthoughDNBandpossiblecladperforation following asteampiperupturecanbeacceptable andisnotprecluded bythecriteria, theaboveanalysis, infact,showsthatnoDNBoccursfortherupture(including anaccidental depressurization ofthemainsteamsystem)assumingthemostreactiveRCCAstuckinitsfullywithdrawn position.

TABLEB.3-10LIMI'.HNG S'ZEAMLINE BREAKSTATEPOINT DOUBLEENDEDRUPTUREINSIDECONTAINMENT WITHOFFSITEPOWERAVAILABLE SecPsia100.2598.7Fraction~FOFFracPPMPercent0.107330.2441.81.01.510.044InletTempTimePressureHeatFluxColdHotFlowBoronReactivity Density~GMCC0.8638-132

(gTABLEB.3-11TIMESEQUENCEOFEVENTSAccidents RuptureofaSteamline EventsY~imesec1.InsideContainment WithOffsitePoweravailable Steamlineruptures0.0Lowsteamline pressuresetpointreached0.26Feedwater Isolation (Allloops)8.26Steamline Isolation (Loops2,3and4)11.26Pressurizer empties13.8SIfiowstarts27.26Criticality attained29.4BoronfromSIreachescores38.2Peakheatfluxattained100.2Corebecomessubcritical 116.28-133 TABLEB3-11'continued)

TIMESEQUENCEOFEVENTSAccidents RuptureofaSteamline EventsT~imesec2.InsideContainment WithoutOffsitePoweravailable Steamlineruptures0.0Lowsteamline pressuresetpointreached0.26Feedwater Isolation (Allloops)8.26Steamline Isolation (Loops2,3and4)s11.26Pressurizer empties15.4Criticality attained37.0SIflowstarts37.26BoronfromSIreachescores51.4Peakheatfluxattained236.0Corebecomessubcritical 291.78-134 1,051021.0151.010,995240280320%0400440480520560COREAVERAGETENPERATURE

('F)FigureB.3-55Variation ofReactivity withCoreTemperature at1050PSIAfortheEndofLifeRoddedCorewithOneControlRodAssemblyStuck(AssumesZeroPower)

gI-UJ1.4i22C/lOoED(~080.6COREPOWER(FRACTION OFNOMINAL)FigureB.3-56Dopp1erPowerFeedbackforSteamline Break8.227 1.60,6COLDLEGSAFETYINJECTION (LB/SEC)FigureB.3-57SafetyInjection FlowSuppliedbyOneChargingPumpB-228

.275.25.225.2.175U.15.'125CD~1cC~.875.65.82558.188.158.288.258.588.(y..275-~25.225VJ~2C/l.175.15.125~cD.875.8258.8,C:nvO~108.158.208.256.588.TINE(SEC)FigureB.3-58NuclearPower'andCoreHeatFluxVersusTimeSteamline BreakOERInsideContainment withPower

550.CLIP-5LUCDLP568L458.488.358.560.250.0,t25CG.2250.2000.1750.150C.1259.1908.7c6500.25C.-'L~56.IEG.156.260.250.300.IL...15~cCO.c56.300.rI)'5)5-.C))608.1460.LLJCDCCl)-ccCL'200.1600.680~698.400.0.50~180.158.208~250'CO.TIME(SEc)FigureB.3-59CoreAverageTemperature, RCSPressure, andPressurizer WaterVolumeVersusTimeSteamline BreakDERInsideContainment withPower8-230 2588.2888.1=08.1808.O568.8~I--588.-1668.CX1588-2888.-2568.50.168158288258388258.288.OO158.m188.ED58.c88.50.168.158.288.258.388.TINE(SEC)FigureB.3-60Reactivity andCoreBoronConcentration VersusTimeSteamline BreakOERInsideContainment withPower8-231 APPENDIXCLOCAANALYSESFORTHEDONALDC.COOKNUCLEARPLANTUNIT2'IRANSITION TO17x17VANTAGE5FUEL C.3.1.2MAJORLOCAANALYSESAPPLICABLE TOWFDTINGHOUSE FUELIdentification ofCausesandFreuenClassi6cation Alosswf-coolant accident(LOCA)istheresultofapiperuptureoftheRCSpressureboundary.

Fortheanalysesreportedhere,amajorpipebreak(largebreak)isdeGnedasarupturewithatotalcross-sectional areaequaltoorgreaterthan1.0ft2.Thiseventisconsidered anANSCondition IVevent,alimitingfault,inthatitisnotexpectedtooccurduringthelifetimeoftheDonaldC.CookNuclearPlantUnit2,butispostulated asaconservative designbasis.TheAcceptance CriteriafortheLOCAaredescribed in10CFR50.46(10CFR50.46andAppendixKof10CFR501974)(1)asfollows:Thecalculated peakfuelelementcladtemperature isbelowtherequirement of2200oF.<2.Theamountoffuelelementcladdingthatreactschemically withwaterorsteamtogeneratehydrogen, doesnotexceed1percentofthetotalamountofZircaloyinthefuelrodcladding.

3.Thecladtemperature transient isterminated atatimewhenthecoregeometryisstillamenabletocooling.Thelocalized claddingoxidation limitof17percentisnotexceededduringorafterquenching.

4.Thecoreremainsamenabletocoolingduringandafterthebreak.~5.Thecoretemperature isreducedanddecayheatisremovedforanextendedperiodoftime,asrequiredbythelong-lived radioactivity remaining inthecore.Thesecriteriawereestablished toprovideasignificant margininemergency corecoolingsystem(ECCS)performance following aLOCA.WASH-1400 (USNRC1975)(10) presentsastudyinregardstotheprobability ofoccurrence ofRCSpiperuptures.

li SeuenceofEventsandStems0erationsShouldamajorbreakoccur,depressurization oftheRCSresultsinapressuredecreaseinthepressurizer.

Loss-Of-OKsite Power(LOOP)isassumedcoincident withtheoccurence ofthcbreak.Thereactortripsignalsubsequently occurswhenthepressurizer lowpressuretripsctpointisreached.Asafetyinjection signalisgenerated whentheappropriate setpointisreached.Thesecountermeasures willlimittheconsequence oftheaccidentintwoways:Reactortripandboratedwaterinjection supplement voidformation incausingrapidreduction ofpowertotheresiduallevelcorresponding tofissionproductdecayheat.NocreditistakenintheLOCAanalysisfortheboroncontentoftheinjection water,howeveranaverageRCS/sumpmixedboronconcentration iscalculated toensurethatthccoreremainssubcritical.

Inaddition, theinsertion ofcontrolrodstoshutdownthereactorisneglected inthelargebreakanalysis.

2.Injection ofboratedwaterprovidesforheattransferfromthecoreandprcvcntscxccssivc cladtemperatures.

InthepresentWestinghouse design,thelargebreaksinglefailureisthelossofoncRHR(lowhead)pump.Thismeansthatcreditcouldbetakenfortwohighheadchargingpumps,twosafetyinjection pumps,andonelowheadpump.Thefollowing isadiscussion ofthemodelling proccdurc fortheminimumsafeguards andtheflowsplitting fromabreakofanECCSbranchinjection linc(i.e.,thespillinglineassumptions).

Thecurrentprocedure forlargebreakanalysesassumesthatatleastonetrainofECCSisavailahlc fordeliveryofwatertotheRCS.AlthoughthesinglefailureisanRHRpump,onlyoncpumpineachsubsystem isassumedtodelivertotheprimaryloops.However,bothEmcrgcncy DicsclGenerators (EDGs)areassumedtostartinthemodelling ofthecontainment deckt'ansandsprays.Modelling fullcontainment heatremovalsystemsoperation isrequiredbyBranchTcchnical PositionCSB6-1andisconservative forthelargebreakLOCA.Thehighheadchargingpumpstarts>>nd deliversOowthroughtheinjection lines(oneperloop)withonebranchinjection lincspillingtothecontainment backpressure.

Tominimizedeliverytothereactor,thebranchlincchosentnspill Cisselectedastheonewiththeminimumresistance.

Whenonesafetyinjection pumpandonclowheadresidualheatremovalpumpstart,Qowisdelivered tothereactorcoolantsystemthroughthcaccumulator injection lines.Again,oneline,withtheminimumresistance, isassumedtospill.torcontainment backpressure.

Inaddition, allsafetyinjection pumpperformance curvesweredegradedsby10%anda25gpmQowimbalance wasassumedforthehighheadchargingpumps.Therefore, inthelargebreakECCSanalysisperformed byWestinghouse, singlelailureisconservatively accounted forviathelossofanECCStrain,andthespillingoftheminimumresistance injection linedespitefullcontainment activeheatremovalsystemoperation (i.c.,twoEDGs).Thetimesequenceofeventsfollowing alargebreakLOCAispresented inTableC.3.1-5.Thcsafetyinjection performance, asmodelledforthelargebreakLOCA,ispresented inFiguresC.3.1.1andC.3.1.2.'(IBeforethebreakoccurs,theunitisinanequilibrium condition; thatis,theheatgenerated inthccoreisbeingremovedviathesecondary system,Duringblowdown, heatfromemissionproductdecay,hotinternals andthevessel,continues tobetransferred tothereactorcoolant.Atthcbeginning oftheblowdownphase,theentireRCScontainssubcooled liquidwhichtranslcrs heatl'romthccorebyforcedconvection withsomefullydeveloped nucleateboiling.Afterthcbrcakdcvclops, thetimetodeparture fromnucleateboilingiscalculated, consistent withAppendixKofl0CFR50(1).Thereafter, thecoreheattransferisunstable, withbothnucleateboilingandl>lmboilingoccurring.

Asthecorebecomesuncovered, bothturbulent andlaminarforcedconvection

<<ndradiation areconsidered ascoreheattransfermechanisms.

TheheattransferbetweentheRCSandthesecondary systemmaybeineitherdirection, dcpcnding ontherelativetemperatures.

Inthecaseofcontinued heatadditiontothesecondary system,thcsecondary systempressureincreases andthemainsteamsafetyvalvesmayactuatetolimitthcpressure.

Makeupwatertothesecondary sideisautomatically providedbythcauxiliary fccdwatcr system.Thesafetyinjection signalactuatesafeedwater isolation signalwhichisolatesmainfeedwater flowbyclosingthemainfeedwater isolation valves,andalsoinitiates auxiliary fccdwatcr flowbystartingtheauxiliary feedwater pumps.Thesecondary flowaidsinthereduction ofRCSipressure.

C-3 WhentheRCSdepressurizes to600psia,theaccumulators begintoinjectboratedwaterintothereactorcoolantloops.Theconservative assumption ismadethataccumulator waterinjectedbypassesthecoreandgoesoutthroughthebreakuntilthetermination ofbypass.Thisconservatism isagainconsistent withAppendixKof10CFR50.Sincelossofoffsitepower(LOOP)isassumed,theRCPsareassumedtotripattheinception oftheaccident.

Theeffectsofpumpcoastdown areincludedintheblowdownanalysis.

Theblowdownphaseofthetransient endswhentheRCSpressure(initialvalueswithuncertainty assumedtorangefrom2037to2313psia)fallstoavalueapproaching thatofthecontainment atmosphere.

Priortoorattheendoftheblowdown, themechanisms thatareresponsible fortheemergency corecoolingwaterbypassing thecore,arecalculated nottobeeffective.

Atthistime(calledend-of-bypass) refillofthereactorvessellowerplenumbegins.Refilliscompleted whenemergency corecoolingwaterhasfilledthelowerplenumofthereactorvessel,whichisboundedbythebottomofthefuelrods(calledbottomofcorerecoverytime).Therefloodphaseofthetransient isdefinedasthetimeperiodlastingfromthebottomofcorefrecoveryuntilthereactorvesselhasbeenfilleduiithwatertotheextengthat thecoretemperature risehasbeenterminated.

Fromthelatterstageofblowdownandthenthebeginning-of-reflood, thesafetyinjection accumulator tanksrapidlydischarge boratedcoolingwaterintotheRCS,contributing tothefillingofthereactorvesseldowncomer.

Thedowncomer waterelevation heapsprovidesthedrivingforcerequiredforthereflooding ofthereactorcore.TheRHR(lowhead),safetyinjection andhighheadchargingpumpsaidinthefillingofthedowncomer and,subsequently, supplywatertomaintainafulldowncomer andcompletethereflooding process.Continued operation oftheECCSpumpssupplieswaterduringlong-term cooling.Coretemperatures havebeenreducedtolong-term steadystatelevelsassociated withthedissipation ofresidualheatgeneration.

AfterthewaterleveloftheeasMuabwaterstoragetank(RW~reachesaminimumallowable value,coolantforlong-term coolingofthecoreisobtainedbyswitching tothecoldlegrecirculation phaseofoperation inwhichspilledboratedwaterisdrawnfromtheengineered safetyfeatures(ESF)containment sumpsbytheresidualheatremoval(lowhead)safetyinjection pumpsandreturnedtotheRCScoldlegs.Thecontainment spraysystemcontinues tooperatetofurtherreducecontainment pressure.

Approximately 12hoursaftertheinitiation oftheLOCA,theECCSisrealigned toinjectwatertotheRCShotlegsinordertocontroltheboricacidconcentration inthereactorvesseLLong-term coolingincludeslong-term criticality control:Criticality controlisachievedbydetermining theRWSTandaccumulator concentrations necessary tomaintainsubcriticality withoutcreditforRCCAinsertion.

Thenecessary RWSTandaccumulator concentrations areafunctionofeachcoredesignandarecheckedeachcycle.ThecurrentTechnical Specifications valueare2400to2600ppmboronfortheRWSTand2400to2600ppmfortheaccumulators.

Theaccumulators areconservatively modelledat2300ppmforthepost-LOCA subcriticality requirement.

CoreandSternPerformance Mathematical Model:Therequirements ofanacceptable ECCSevaluation modelarepresented inAppendixKof10CFRSo(1).LargeBreakLOCAEvaluation ModelTheanalysisofalargebreakLOCAtransient isdividedintothreephases:(1)blowdown, (2)refill,and(3)reQood.Therearethreedistincttransients analyzedineachphase,including thethermal-hydraulic transient intheRCS,thepressureandtemperature transient withinthecontainment, andthefuelandcladtemperature transient ofthehottestfuelrodinthecore.Basedontheseconsiderations, asystemofinterrelated computercodeshasbeendeveloped fortheanalysisoftheLOCA.Adescription ofthevariousaspectsoftheLOCAanalysismethodology isgivenbyBordelon,-

Massie,andZordan(1974)().Thisdocumentdescribes themajorphenomena modeled,theinterfaces amongthecomputercodes,andthefeaturesofthecodeswhichensurecompliance withtheAcceptance CriteriaTheSATAN-VI,

WREFLOOD, BASHandLOCBARTcodes,whichareusedintheLOCAanalysis, aredescribed indetailbyBordelonetal.(1974)(5);

Kellyetal.(1974)();Youngetal.(1987)(");

andBordelonetal.(1974)(6).

Codemodifications arespecified inReferences 2,7,13,and17.Thesecodesassessthecoreheattransfergeometryanddetermine ifthecoreremainsamenabletocoolingthroughandsubsequent totheblowdown, refill,andrefioodphasesoftheLOCA.TheSATAN-VIcomputercodeanalyzesthethermal-hydraulic C-5 transient intheRCSduringblowdownandtheWREFLOODcomputercodecalculates this~~transient duringtherefillphaseoftheaccident.

TheBASHcodeisusedtodetermine theRCSresponseduringtherefloodphaseofthetransient.

TheLOTICcomputercode,described byHsiehandRaymundinWCAP-8355 (1975)andWCAP-8345 (1974)(),calculates thecontainment backpressure transient.

Thecontainment backpressure transient isinputtoBASHforthepurposeofcalculating therefloodtransient.

TheLOCBARTcomputercodecalculates thethermaltransient ofthehottestfuelrodinthethreephases.Theimprovedfuelperformance model,described inReference 15,generates theinitialfuelrodconditions inputtoLOCBART.SATAN-VIcalculates theRCSpressure,

enthalpy, density,andthemassandenergyflowratesintheRCS,aswellassteamgenerator energytransferbetweentheprimaryandsecondaiy systemsasafunctionoftimeduringtheblowdownphaseoftheLOCA.SATAN-VIalsocalculates theaccumulator watermassandinternalpressureandthebreakmassandenergyflow.ratesthatareassumedtobeventedtothecontainment duringblowdown.

Attheendoftheblowdown, information onthestateofthesystemistransferred totheWREFLOODcodewhichperformsthecalculation oftherefillperiodtobottomofcore(BOC)recoverytime.Oncethevesselhasrefilledtothebottomofthecore,therefloodportionofthetransient begins.TheBASHcodeisusedtocalculate thethermal-hydraulic simulation oftheRCSfortherefloodphase.Information concerning thecoreboundaryconditions istakenfromalloftheabovecodesandinputtotheLOCBARTcodeforthepurposeofcalculating thecorefuelrodthermalresponsefortheentiretransient.

Fromtheboundaryconditions, LOCBARTcomputesthefluidconditions andheattransfercoefficient forthefulllengthofthefuelrodbyemploying mechanistic modelsappropriate totheactualflowandheattransferregimes.Conservative assumptions ensurethatthefuelrodsmodeledinthecalculation represent thehottestrodsintheentirecore.Thelargebreakanalysiswasperformed withtheDecember1981versionoftheEvaluation Modelmodifiedtoincorporate theBASH(d)computercode.

InputParameters andInitialConditions:

Theanalysispresented inthissectionwasperformed withareactorvesselupperheadtemperature equaltotheRCShotlegtemperature.

Arangeofreactoroperating temperatures wasanalyzedinordertojustifyplantoperation atareactorpowerlevelof3588Mwtbetween582.2Fto615;2Finthehotlegsand511.7'Fto547.6Finthecoldlegs.Inadditiontothetemperature rangeanalyzed, initialRCSprcssurizcr pressurewasalsovariedtojustifyplantoperation between2037and2313psia.Afullspectrumbreakanalysiswasdoneatthehighpressure/high temperature RCSconditions (initialRCSpressurizer

pressure, withuncertainty, of2313psiaandinitialhotlegtemperature of615.2F)fromwhichthelimitingbreaksizewasdetermined.

Thelimitingbreakwasthenreanalyzed lorlowtemperature andhighRCSpressure.

Thelimitingbreakwasalsoreanalyzed forthchightemperature andlowinitialRCSpressureof2037psia.Theanalysisalsoconsidered plantoperation atreducedpowerlevelwiththeRHRcrosstievalveclosed.Thereduction inpowerlevelwasconsidered necessary tooffsetthereduction insafetyinjection flowduetothcclosure.oftheRHRcrosstievalve.Thiscaseassumedareducedpowerlevelof3413.MWtandminimumsafeguards withtheRHRcrosstievalveclosedatthelimitingRCSconditions.

Allcasesconservatively assumed15%steamgenerator tubeplugginginallfoursteamgenerators.

Table.C.3.1-1describes thecasesanalyzed.

TablesC.3.1-2andC3.1-3summarize thckcyinputIparameters andsetpoints modelledintheCookNuclearPlantUnit2largebrcakLOCAanalysis.

Thebasesusedtoselectthenumerical valuesthatareinputparameters totheanalysishavebccnconservatively determined fromextensive sensitivity studies(Westinghouse 1974(12);

Salvatori 1974();Johnson,Massie,andThompson1975(8)).

Inaddition, therequirements ofAppendix(11).Kto10CFR50(1)regarding speciGcmodelfeaturesweremetbyselecting modelswhichprovideasignificant overallconservatism intheanalysis.

Theassumptions whichweremadepertaintothcconditions ofthereactorandassociated safetysystemequipment atthetimethattheLOCAoccurs,andincludesuchitemsasthecorepeakingfactors,thecontainment

pressure, andthcperl'ofillancc oftheECCS.Decayheatgenerated throughout thetransient isalsoconservatively calculated aspertherequirements ofAppendixKto10CFR50(1).C-7 Anotherinputparameter thataffectsLOCAanalysisresultsistheassumedaxialpowershapeat~~~thebeginning oftheaccident.

Powershapesensitivity studiesperformed withWestinghouse ECCSevaluation modelshavealwaysdemonstrated thechoppedcosineshapewiththepeakatthecoremidplanetobelimiting.

Westinghouse hasperformed "spotcheck"analysesusingtheBASHrefloodevaluation modelforpowershapesskewedtothetopofthecore.Resultsoftheseanalyseshavedemonstrated thechoppedcosinepeakedatthecoremidplaneremainsthelimitingpowershape(18)

AmeetingwasheldattheWestinghouse Licensing OfficeinBethesdaonDecember17,1981,betweenmembersoftheU.S.NuclearRegulatory Commission andmembersoftheWestinghouse NuclearSafetyDepartment todiscusstheimpactofmaximumsafetyinjection onthelargebreakECCSanalysisonagenericbasis.Furtherdiscussion ofthisissueisprovidedinaletterfromE.P.Rahe,ManagerofWestinghouse NuclearSafetyDepartment, toRobertITedescooftheU.S.NuclearRegulatory Commission(14).

Abriefdescription ofthisissueisgivenbelow.Westinghouse ECCSanalysescurrently assumeminimumsafeguards forthesafetyinjection flow,~~whichminimizes theamountofflowtotheRCSbyassumingmaximuminjection lineresistances, degradedECCSpumpperformance, andthelossofoneresidualheatremoval(RHR)pumpasthemostlimitingsinglefailure.Thisisthe'limiting singlefailureassumption whenoffsitepowerisunavailable formostWestinghouse plants.However,forsomeWestinghouse plants,including CookNuclearPlantUnit2,thecurrentnatureoftheAppendixKECCSevaluation modelsissuchthatitmaybemorelimitingtoassumethemaximumpossibleECCSflowdelivery.

Inthatcase,maximumsafeguards whichassumeminimuminjection lineresistances, enhancedECCSpumpperformance, andnosinglefailure,resultinthehighestamountofflowdelivered totheRCS.Therefore, theworstbreakforCookNuclearPlantUnit2(CD=Q.6)wasreanalyzed, assumingmaximumsafeguards (CaseAvs.CaseFofTableC.3.1-1).

Examination oftheLOCAanalysisresultsinTableC.3.1-6demonstrates thatminimumsafeguards assumptions resultinthehighestpeakcladtemperature forCookNuclearPlantUnit2.

Whenassessing theeffectoftransition coresonthelargebreakLOCAanalysis, itmustbedetermined whetherthetransition corecanhaveagreatercalculated peakcladdingtemperature (PCT)thaneitheracompletecoreofthe17x17ANFassemblydesignoracompletecoreoftheWestinghouse 17x17VANTAGE5design.Foragivenpeakingfactor,theonlymechanism available tocauseatransition coretohaveagreatercalculated PCTthanafullcoreofeitherfuelisthepossibility offlowredistribution duetofuelassemblyhydraulic resistance mismatch.

Hydraulic resistance mismatchwillexistonlyforatransition coreandistheonlyuniquedifference betweenacompletecoreofeitherfueltypeandthetransition core.The17x17ANFfuelassemblyisnearlyidentical totheWestinghouse 17x17OFAassemblyintermsofhydraulic andgeometric characteristics.

Therefore, theanalysesreportedinReference 19whichdemonstrate thatthe17x17VANTAGE5fuelfeaturesresultinafuelassemblythatismorelimitingthanaWestinghouse.17x17 OFAfuelassembly, withrespecttolargebreakLOCAECCSperformance, remainvalidasappliedatCookNuclearPlantVnit2.ThesamelargebreakLOCAtransition corepenaltyreported.

inSection5.23ofReference 19willbeappliedtothetransition from17x17ANFfuelassemblies toWestinghouse 17x17VANTAGE5fuelassemblies.

Westinghouse transition coredesigns,including specific17X17OFAto17x17VANTAGE5transition corecases,wereanalyzed.

Theincreaseinhydraulic resistance fortheVANTAGE5assemblywasshowntoproduceareduction inrefloodsteamflowratefortheVANTAGE5fuelatmixingvanegridelevations fortransition coreconfigurations.

Thevariousfuelassemblyspecifictransition coreanalysesperformed resultedinpeakcladdingtemperature increases ofupto50Fforcoreaxialelevations thatboundthelocationofthePCT.Therefore, themaximumPCI'enalty possibleforVANTAGE5fuelresidinginatransition coreis50F,Reference 19.Asstatedearlier,thistransition corepenaltycontinues toapplytothetransition from17x17ANFfuelassemblies toWestinghouse 17x17VANTAGE5fuelassemblies duetothenearidentical designof17x17ANFandWestinghouse 17x17OFAfuelassemblies.

OnceafullcoreofVANTAGE5fuelisachievedthelargebreakLOCAanalysiswillapplywithoutthetransition corepenalty.C-9

Results:BasedontheresultsoftheLOCAsensitivity studies(Westinghouse 1974(1);Satvatori 1974(11);

Johnson,Massie,andThompson1975(8)),

thelimitinglargebreakwasfoundtobethedoublewnded coldlegguillotine (DECLG).Therefore, onlytheDECLGbreakisconsidered inthelargebreakECCSperformance analysis.

Calculations wereperformed forarangeofMoodybreakdischarge coefficient.

Theresultsofthesecalculations aresummarized inTablesC3.1-5andC3.1-6.Thecontainment datausedtogeneratetheLOTICbackpressure transient areshowninTable.C3.1-4.Themassandenergyreleasedatafortheminimumandmaximumsafeguards cases(CaseAandF)areshowninTablesC.3.1-7andC.3.1-8respectively.

Inaddition, massandenergyreleasedataforCaseG.(3413Mwt,RHRcrosstievalveclosed)areshowninTableC3.1-9.Themassreleasesfortheremaining casesarenotpresented, sincetheydonotvarysigniflcantly fromthedatashowninTableC.3.1-7.Nitrogenreleaseratestothecontainment aregiveninTableC.3.1-10.

FiguresC.3.1-3athroughC.3.1-30presenttheresultsofthecasesanalyzedforthelargebreakLOCA,Thealphadesignation inthefigurenumbercorresponds tothecasesasdescribed inTableC.3.1-1.FiguresC3.1-3a-g Thesystempressureshownisthecalculated corepressure.

FiguresC.3.1-4a-g Theflowratefromthebreakisplottedasthesumofbothendsoftheguillotine break.FiguresC3.1-5a-g Thecorepressuredropshownisfromthelowerplenum,nearthecore,totheupperplenumatthecoreoutlet.FiguresC.3.1-6a-g Thecoreflowrateisshownduringtheblowdownphaseofthetransient.

FiguresC.3.1-7a-g Theaccumulator flowrateduringblowdownisplottedasthesumofthatinjectedintotheintactcoldlegs.

P FiguresC3.14a-gThecoreanddowncomer collapsed liquidwaterlevelsareplottedduringtherefloodphaseofthetransient.

FiguresC.3.1-9a-g Thecoreinletflowrateisshownasitiscalculated duringtherefloodphase.FiguresC3.1-10a-g ThetotalpumpedECCSfiowrateinjecting intotheintactcoldlegsisshown.FiguresC3.1-11a-g Theintegralofthecoreinletflowrateascalculated withBASHisplotted.FiguresC3.1-12a-g Themassfluxisplottedatthehotspot(thenodewhichproducedthepeakcladtemperature) onthehotrod.FiguresC.3.1-13a-g Theheattransfercoefficient isplottedatthehotspotonthehotrod.FiguresC.3.1-14a-g Thefluidtemperature atthehotspotonthehotrodisplotted.FiguresC3.1-16-18 Thecontainment backpressure transient usedintheanalysisisprovidedforCasesA,FandG(theminimumandmaximumSIflowcases,andthe3413Mwtcrosstievalveclosedcase).FiguresC.3.1-19-27 Theseflguresshowtheheatremovalratesoftheheatsinksfoundintheloweranduppercompartment andtheheatremovalbythesumpandlowercompartment sprayforCasesA,FandG.FiguresC.3.1-28-30 Thesefiguresshowthetemperature transients inboththeloweranduppercompartments ofcontainment andflowfromtheuppertolowercompartments forCasesA,FandG.Thepeakcladtemperature calculated foralargebreakis2140'F,whichislessthantheacceptance criterion limitof2200~F.Themaximumlocalmetal-water reactionis6.80percent,whichiswellbelowtheembrittlement limitof17percentasrequiredby10CFR50.46.Thetotal

metal-water reactionislessthan0.3percentforallbreaks,corresponding tolessthan0.3percenthydrogengeneration, ascomparedwiththe1percentcriterion of10CFR50A6.Thecladtemperature transient isterminated atatimewhenthecoregeometryisstillamenabletocooling.Asaresult,thecoretemperature willcontinuetodropandtheabilitytoremovedecayheatgenerated inthefuelforanextendedperiodoftimewillbeprovided.

DONALDC.COOKNUCLEARPLANTUNIT2TABLEC3.1-2INPUTPARAMETERS USEDINTHELARGEBREAKLOCAECCSANALYSISCrossT~ies0eRHRCrossTiesClosedLicenseCorePower(a),

(MWt)PeakLinearPower(),(kw/ft)TotalPeakingFactor,FgTAxialPeakingFactor,FZHotChannelEnthalpyRiseFactor,FgHPowerShape:FuelAssemblyArrayAccumulator WaterVolume,Nominal(ft3/accumulator)

Allowance Accumulator TankVolume,Nominal(ft/accumulator)

Accumulator-Gas

Pressure, Minimum(psia)SafetyInjection PumpedFlowRate(Allpumpsdegraded10'Yo,Chargingpumpflowrateimbalance

=25Containment Parameters InitialLoopHow(GPM)VesselInletTemperature

(~F)VesselOutletTemperature (F)AverageReactorCoolantPressure(psia)SteamPressure(psia)SteamGenerator TubePluggingLevel(%)Refueling WaterStorageTankTemperature

('F)358812.7142.2201.3701.6203413.12.7212.3351.4201.644ChoppedCosine17X17VANTAGE5946946+25+2513501350600'600SeeFiguresC.3.1.1gpm)andC3.1.2SeeTableC.3.1-488,50088,500511.7to513.3to547.6546.4582.2to580.6to615.2611.22037.4to2037.4to2312.62312.6587to603to820'20151570(b)70(b)(a)Twopercentisaddedtothispowertoaccountforcalorimetric error.(b)TheBASHcomputercodemodelsaverageRWSTtemperature duringcorereflooding (85~F).Othercomputercodesinthemodeluse70~F.

C.3.2LOSSOFREACTORCOOLANTFROMSMALLRUPTUREDPIPESORFROMCRACKSINLARGEPIPESWHICHACTUATESTHEEMERGENCY CORECOOLINGSYSTEMAnalisofEffectsandConsuencesMethodofAnalisForsmallbreaks(lessthan1.0ft2)theNOTRUMP()()digitalcomputercodeisemployedtocalculate thetransient depressurization oftheReactorCoolantSystemaswellastodescribethemassandenthalpyoftheQuidflowthroughthebreak.SmallBreakLOCAAnalisUsinNOTRUMPForloss-of-coolant accidents duetosmallbreakslessthan1squarefoot,theNOTRUMPcomputercode()()isusedtocalculate thetransient depressurization oftheRCSaswellastodescribe(10)(11)~themassandenthalpyoftheQuidflowthroughthebreak.TheNOTRUMPcomputercodeisastate-of-the-art one-dimensional generalnetworkcodeincorporating anumberofadvancedfeatures.

Amongthesearecalculation ofthermalnonwquilibrium inallQuidvolumes,flowregime-dependent driftfluxcalculations withcounter-current Qoodinglimitations, mixtureleveltrackinglogicinmultiple-stacked Quidnodesandregime4ependent heattransfercorrelations.

TheNOTRUMPsmall-break LOCAemergency corecoolingsystem(ECCS)evaluation modelwasdeveloped todetermine theRCSresponsetodesignbasissmallbreakLOCAs,andtoaddressNRCconcernsexpressed inNUREG-0611, "GenericEvaluation ofFeedwater Transients andSmallBreakLosswf-Coolant Accidents inWestinghouse-Designed Operating Plants".Thereactorcoolantsystemmodelisnodalized intovolumesinterconnected byQowpaths.

Thebrokenloopismodelledexplicitly, whilethethreeintactloopsarelumpedintoasecondloop.Transient behaviorofthesystemisdetermined fromthegoverning conservation equations ofmass,energy,andmomentum.

Themultinode capability oftheprogramenablesexplicit, detailedspatialrepresentation ofvarioussystemcomponents which,amongothercapabilities, enablesapropercalculation ofthebehavioroftheloopsealduringalosswf-coolant accident.

Thereactorcoreisrepresented asheatedcontrolvolumeswithassociated phaseseparation modelstopermittransient mixtureheightcalculations.

Detaileddescriptions oftheNOTRUMPcodeandtheevaluation modelareprovidedinReferences 10and11.

<0,AfterthesmallbreakLOCAisinitiated, reactortripoccursduetoalowpressurizer pressuresignal(1860psia).Soonafterthereactortripsignalisgenerated, thesafetyinjection signalisactuatedduetolowpressurizer pressure(1715psia).Safetyinjection systemsconsistofgaspressurized accumulator tanksandpumpedinjection systems.ThesmallbreakLOCAanalysisassumedanaccumulator watervolumecflualtotheaverageofthatallowedinthetechnical specification withacovergaspressureof600psia.Thisistheminimumofthecovergaspressure.

allowedintheTechnical Specifications.

Mni1numemergency corecoolingsystemavailability isassumedfortheanalysisatthemaximumRWSTtemperature.

Assumedpumpedsafetyinjection characteristics asafunctionof-RCSpressureusedasboundaryconditions intheanalysisareshowninFigureC.3.2-1andinTableC9.2-6.Thesafetyinjection flowratespresented arebasedonpumpperformance curvesdegraded10percentfromthedesignheadandanassumedchargingsystembranchlincimbalance of25gpm.TheeffectofflowfromtheRHRpumpsisnotconsidered inthesnmllbreakLOCAanalysessincetheirshutoffheadislowerthantheRCSpressureduringtbctimeportionofthetransient considered here.Safetyinjection (SI)isdelayed27secondsaftertheoccurrence oftheinjection signaltoaccountfordieselgenerator startupandemergency powerbusloadingincaseofalossofoffsitepowercoincident withaLOCA.ThesmallbreakLOCAanalysisalsoassumedthattheauxiliary feedwater pumpsweredegradedby15percentandthatthcroddroptimewas2.7seconds.Peakcladtemperature calculations areperformed withtheLOCTA-IV(2) codeusingtheNOTRUMPcalculated corepressure, fuelrodpowerhistory,uncovered coresteamflowandmixtureheightsasboundaryconditions.

FigureC3.2-10depictsthehotrodaxialpowershapeusedtoperformthesmallbreakLOCAanalysis.

Thisshapewaschosenbecauseitrepresents adistribution withpowerconcentrated intheupperregionsofthecore.Suchadistribution islimitingforsmall-break LOCAsbecauseitminimizes coolantlevelswell,whilemaximizing vaporsuperheating andfuel'rodheatgeneration attheuncovered elevations.

ThesmallbreakLOCAanalysisassumesthecorecontinues tooperateatfullpoweruntilthecontrolrodsarecompletely inserted.

Results'his sectionpresentsresultsofthelimitingsmaHbreakLOCAanalysis(asdetermined bythehighestcalculated peakcladtemperature) forarangeofbreaksizesandRCSpressures andtemperatures.

Thelimitingbreakwasfoundtobea4-inchdiametercoldlegbrcakinitiated at

reducedRCSpressurizer pressure(2100psia)andhightemperature (coreTavg=581.3OF)~~conditions.

Thepeakcladtemperature attainedduringthetransient was1357F.Alistofinputassumptions usedinthelowpressureandhightemperature analysisisprovidedinTableC.3.2-1.Theresultsofathreebreakspectrumanalysisperformed atthereducedRCSpressureandhightemperature conditions aresummarized inTableC3.24,whilethekeytransient eventtimesarelistedinTableC3.2-2.FiguresC.32-2through9showforthelimitingfour-inch breaktransient, respectively:

RCSpressureCoremixturelevelPeakcladtemperature CoreoutletsteamQowrateHotspotrodsurfaceheattransfercoefficient HotspotQuidtemperature ColdlegbreakmassQowrateSafety'injection massQowrateDuringtheinitialperiodofthesmall-break transient theeffectofthebreakflowrateisnotstrongenoughtoovercometheQowratemaintained bythereactorcoolantpumpsastheycoastdown.NormalupwardQowismaintained throughthecoreandcoreheatisadequately removed.Atthelowheatgeneration ratesfollowing reactortripthefuelrodscontinuetobewellcooledaslongasthecoreiscoveredbyatwo-phase mixturelevel.Fromthecladtemperature transient forthe4-inchbreakcalculation showninFigureC3.2-4,'it isseen'hatthepeakcladtemperature occursnearthetimewhenthecoreismostdeeplyuncovered andthetopofthecoreissteamcooled.Thistimeisalsoaccompanied bythehighestvaporsuperheating abovethemixturelevel.Acomparison ofthetotalbreakQowratetocont'ainment showninFigureC.3.2-8tothesafetyinjection flowrateshowninFigureC9.2/shows thatatthetimethetransient w'asterminated, eitherwhenthesafetyinjection Qowratethatwasdelivered totheRCSexceededthemassflowrateoutthebreakorwhenthecorewascoveredasinFigureC3.2-20.Althoughtheinnervesselcoremixturelevelhasnotyetcoveredtheentirecore,thereisnolongeraconcernofexceeding the10CFR50A6criteriasincethepressureisgradually decayingandthereisanetmassinventory gain.AstheRCSinventory continues togradually

increase, thecoremixturelevelwillcontinuetoincreaseandthefuelcladtemperatures willcontinuetodecline.

Conclusions Analysespresented inthissectionshowthatthehighheadchargingandsafetyinjection subsystems oftheEmergency CoreCoolingSystem,togetherwiththeaccumulators, providesufficient coreQoodingtokeepthecalculated peakcladtemperatures belowtherequiredlimitof10CFR50.46.Henceadequateprotection isaffordedbytheEmergency CoreCoolingSystemintheeventofasmailbreakloss-of~lant accident.

Additional BreakCasesStudiesdocumented inReference 3determined thatthelimitingsmall-break sizeoccurredforbreakslessthan10inchesindiameter.

Toinsurethatthe4-inchdiameterbreakwaslimiting, calculations wererunwithbreaksof3inchesand6inches.Theresultsofthesecalculations areshownintheSequenceofEventsTableC.3.2-2,andtheResultsTableC.3.2-4.Plotsofthefollowing parameters areshowninFiguresC3.2-11through18forthe3-inchbreak,andFiguresC.3.2-19through26forthe6-inchbreak.RCSpressure.

CoremixturelevelPeakcladtemperature CoreoutletsteamQowrateHotspotrodsurfaceheattransfercoefficient ColdlegbreakmassQowrateSafetyinjection massQowrateAsseeninTableC3.2Qthepeakcladtemperatures werecalculated tobelessthanthatforthe4-inchbreak.Additional AnalisCalculations werealsoperformed forCookNuclearPlantUnit2withtheNOTRUMP()(l)andLOCTA-IV(

)codestoexaminetheinQuenceofinitialRCScoolantoperating temperatures andoperating pressures onsmallbreakLOCApeakcladtemperature.

Theanalysesperformed demonstrated thatthereducedpressureandhightemperature conditions analyzedresultedinthelimitingPCI'orthe4-inchdiameterbreak.

Tosupportoperation ofCookNuclearPlantUnit2atRCSpressures of2100psiaand2250psiaforarangeofloopoperating temperatures, twoadditional analyseswereperformed.

Calculations wereperformed forafour-inch diameterbreakforaninitialRCSpressurizer pressureof2250psiaatinitialRCScoolantoperating temperatures corresponding tocoreTavgprogramsetpoints of5813~FandataTavgof547.0F.Theresultsofthesecalculations areshownintheSequenceofEventsTableC.3.2-3,andtheResultsTableC.3.2-5.Plotsofthefollowing parameters areshowninFiguresC3.2-27through34forthehighpressure, hightemperature case,andFiguresC3.2-35through42forthehighpressure, lowtemperature case.RCSpressureCoremixturelevel,Peakcladtemperature, Coreoutletsteamflowrate,Hotspotrodsurfaceheattransfercoefficient,

-HotspotQuidtemperature, ColdlegbreakmassQowrate,andSafetyinjection massQowrate.AsseeninTableC3.2-5,thepeakcladtemperatures werecalculated tobelessthanthat.forthe4-inchbreakinitiated atreducedpressureandhightemperature conditions.

Additional calculations weremadetosupportclosureofthehighheadsafetyinjection crosstievalves.Sincetheamountofpumpedinjection Qowwouldbereducedwiththehighheadcrosstievalvesclosed,itwasnecessary tolowercorepowerinordertomaintainthepeakcladtemperature withinthe10CFRSOA6limit.Thusthecalculation whichsupportsplantoperation withthehighheadcrosstievalvesclosedassumedaninitialRCSpressurizer pressureat2100psiaandcoreTavgat581.3Fatacorepowerlevelof3413Mwt.Thiscalculation alsoassumedachargingsystemQowrateimbalance of25gpm.TheassumedpumpedECCSQowperformance forthehighheadcrosstievalveclosedcaseislistedinTableC.3.2-9.

Pastanalyseshaveshownthatareduction inpumpedsafetyinjection flowrateincreases thepeakcladtemperature forsmallerbreaks(3inches)morethanlargersmallbreaks(4and6inches).Animportant parameter indetermining whatwillbethelimitingbreaksizeisthereactorpowertosafetyinjection flowrateratio.Forthehighheadcrosstieclosurecasethereactorpowertosafetyinjection flowrateratiowasreducedwhichshiftedthelimitingbreaksizetothe3-inchdiametercoldlegbreak.EvidenceofthiseffectistheCookNuclearPlantUnit1smallbreakLOCAanalysiswhichwasperformed withthehighheadcrosstievalvesclosedassumingachargingsystemflowrateimbalance of10gpmwithareactorpowerof3588Mwt.TheCookNuclearPlantUnit1smallbreakLOCAanalysishadareactorpowertosafetyinjection flowrateratioapproximately equaltoCookNuclearPlantUnit2withthehighheadcrosstievalvesclosedassuming25gpmchargingsystemfiowrateimbalance atareactorpowerlevelof3413Mwt.Itwasconcluded thatwiththehighheadcrosstievalvesclosedandwithreducedreactorpowerthelimitingbreakwouldbeshiftedfromthe4-inchdiametercoldlegbreaktothe3-inchdiameterbreaksize.Toverifythisconclusion, twocalculations wereperformed whichassumedbreaksizesof3-and4-inchdiameters atthereducedpressure, hightemperature initialconditions.

TableC.3.24liststheresultsofthecrosstieclosedcaseswhichshowthatwiththereducedsnl'etyinjection flowthe3-inchdiameterbreakislimiting.

Thesequence'f eventsforthesecalculations islistedinTableC.3.2-7.PastsmallbreakLOCAanalysesthatwereperformed forplantswhicharesimilartoCookNuclearPlantUnit2buthavepowertosafetyinjection flowrateratioslessthanthatofCookNuclearPlantUnit2,haveshownthatanassumedbreaksizeof2inchesdidnotresultinthelimitingpeakcladtemperature.

Thus,basedonthecomparision ofpowertosafetyinjection lowrateratio,itwasconcluded thata2-inchdiameterbreakwouldnotyieldapeakcladtemperature morelimitingthatthatofthe3-inchdiameterbreaksize.Plotsforthe3-and4-inchbreakwiththeHHSIcrosstievalvesclosedareshowninflguresC.3.2-43throughC.3.2-50andC.3.2-51throughC.3.2-58respectively.

NUREG4737(13),

SectionII.K3.31, requiredplant-speciflc smallbreakLOCAanalysisusinganEvaluation ModelrevisedperSectionILK.3.30.

Inaccordance withNRCGenericLettcr83-65(4),genericanalysesusingNOTRUMP()()wereperformed andarepresented inWCAP-11145(

).Thoseresultsdemonstrate thatinacomparison ofcoldleg,hotlegandpumpsuctionIcgbreaklocations, thecoldlegbreaklocationislimiting.

C-139

~'