ML17309A406

Responds to NRC Bulletin 88-004 Re Potential safety-related Pump Loss.Sizing of Min Flow Bypass Lines for RHR & Safety Injection Pumps Adequate to Provide Pump Protection for Short Term Operation in Min Flow Mode

ML17309A406
Person / Time
Site:	Ginna
Issue date:	07/07/1988
From:	SNOW B A ROCHESTER GAS & ELECTRIC CORP.
To:	STAHLE C NRC OFFICE OF ADMINISTRATION & RESOURCES MANAGEMENT (ARM), Office of Nuclear Reactor Regulation
References
IEB-88-004, IEB-88-4, IEIN-87-059, IEIN-87-59, NUDOCS 8807130187
Download: ML17309A406 (30)

Text

ROCHESTER GASANDELECTRICCORPORATION

~89EASTAVENUE.ROCHESTER, N.Y.14649-0001 July7,1988U.S.NuclearRegulatory Commission DocumentControlDeskAttn:Mr.CarlStahlePWRProjectDirectorate No.1Washington, D.C.20555

Subject:

ResponsetoNRCBulletin88-04Potential Safety-Related PumpLossR.E.GinnaNuclearPowerPlantDocketNo.50-244

DearMr.Stahle:

ThesubjectBulletinrequested licensees todetermine whetherconditions mayexistwhichcouldpotentially causesafety-related pumpdamageduetoparalleloperation.

Damagewouldbetheresultofoverheating causedbythepumpoperating deadheaded.

TheBulletinalsorequested anevaluation oftheadequacyoftheminimumrecirculation flowduringoperation, testing,andaccidentconditions considering thelargesttimespentinthesemodesandthecumulative effectsonthepumpoverthelifctimeoftheplant.RG&Ewasinitially notifiedoftheseconcernsbyWestinghouse ElectricCorporation, whichwasfollowedbyNRCInformation Notice87-59.RESIDUALHEATREMOVALANDSAFETYINJE~ONSYSTEMSTheresultsofourevaluations performed andtestsrunsubsequent tothesenotifications haveindicated thattheresidualheatremoval(RHR)pumps,whichalsoserveasthelowheadsafetyinjection pumps,wouldbesusceptible tothecondition described intheBulletinforthreereasons:1)TheRHRsystemutilizesaminimumflowbypasslinewhichiscommontobothpumps.2)Theminiflowpipingisconfigured suchthataninteraction betweenthetwopumpsisnotprecluded duringparalleloperation.

3)TheRHRpumpdesignischaracterized byahead-capacity curvewhichhasarelatively flatshapeatlowflows.

Thethreehighheadsafetyinjection (SI)pumpsareeachprovidedwithaminimumflowrecirculation linewhichislocatedupstreamofeachdischarge linecheckvalve.Therefore,eachSIpumpwillreceiveaminimumflowindependent oftheotherpumpsandarenotsusceptible totheparalleloperation concerndescribed inthesubjectBulletin.

Theexistingminimumflow(designed for30gpm)wasbasedonshorttermoperation and.limitingthetemperature riseofthepumpinternalfluid.Thissystemhasbeenre-evaluated assetforthinAttachment A.Baseduponourreviewofperiodicandspecialtestdata,andthepumpdesignsandoperational modes,wehaveconcluded thatthesizingoftheminimumflowbypasslinesforboththeRHRandSIpumpscouldbeconsidered adequatetoprovidepumpprotection forshorttermoperation intheminimumflowmode,forwhichtheselineswereintended.

However,theparalleloperation concernfortheRHRpumpsnecessitated corrective actionwhichwastakenonaninterimbasisbyprocedural stepsandwillbetakenforthelongertermbyaplantmodification.

Thesafetyinjection systemrecirculation systemisalsoundergoing aredesignforaplannedplantmodification forthepurposeofde-sensitizing thesafetyinjection pumpstothepotential forlongtermweardegradation causedbymonthlyperiodictestsinthelowflowmode.Thisplannedmodification isbeingcombinedwithapreviousmodificationcommitment madeinresponsetoIEBulletin86-03,Potential FailureofMultipleECCSPumpsDuetoSingleFailureofAirOperatedValveinMinimumFlowRecirculation Line,submitted inourletterdatedJanuary8,1987.Hardwaremodifications forbothRHRandSIsystemsareplannedduringour1989annualmaintenance andrefueling outage.FortheRHRsystem,aredesignisunderwaywhichwillprovideeachRHRpumpwithaminimumflowrecirculation linewhichisindependent oftheoppositetrain.Thelineisbeingsizedtoprovidesufficient recirculation flowwhenthepumpdischarge pathisisolated.

Intheinterim,procedures havebeenmodifiedwhichrequiresecuringoneRHRpumpiftwopumpsareoperating withacombinedflowlessthan1200gpm.Thisflowwasdetermined asaconservative valueabovewhichsufficient flowwouldbeguaranteed throughtheweakerpumpwhenthepumpsoperateinparallel.

FortheSIsystem,theplannedmodification willrequireadditional largerdiameterrecirculation pipingwithapressurebreakdown orificeineachseparateline,sizedtoallow100gpmorabout25%ofbestefficiency point(BEP)offlow.Eachofthesedesignshadtheobjective ofproviding sufficient recirculation flowconsistent withpumpmanufacturer recommendations withoutreduction intheinjectedflowdeliveryduringaccidentconditions usedinthesafetyanalysisreport.

Anevaluation oftheSIandRHRrecirculation redesigns andjustification forcontinued operation issetforthinAttachments AandB.AUXILIARY FEEDWATER SYSTEMSTheauxiliary feedwater pumpshavealsobeenevaluated relativetotheBulletin88-04concerns.

TheGinnaStationmainauxiliary feedwater systemconsistsoftwo1009ocapacitymotordrivenpumpsandone200%,capacityturbinedrivenpump.Twoadditional standbyauxiliary feedwater pumps,each100%capacity, arealsoinstalled inaseparatebuildingasabackuptothemainauxiliary feedwater pumps.Eachofthesepumpsisprovidedwithanautomatically controlled minimumflowrecirculation systemsizedandperiodically testedtoensurethatsufficient minimumflowwillbeprovidedunderallaccidentandnormaloperating conditions.

Ithasbeendetermined thattheminimumflowconcernsraisedinBulletin88-04havebeenadequately addressed inthe'designandtestingofthesesystemsatGinna.Anevaluation oftherecirculation systemsforthesepumpsissetforthinAttachment C.Therearenoothersafety-related pumpswhicharesusceptible totheBulletin88-04concer'ns.

Asrequiredbyparagraph 5of.thesubjectbulletin, awrittenresponsedescribing theactionstakenwillbeprovidedwithin30daysfollowing completion oftheplannedplantmodifications described inAttachments AandB.Asrequiredbyparagraph 6ofthesubjectbulletin, thisresponsetothebulletinandourevaluations setforthinAttachments AthroughCwillbemaintained atGinnaStationforaminimumoftwoyears.Verytrulyyours,'ruceA.SnowSuperintendent NuclearProduction Subscribed andsworntobeforemethis7thdayofJuly,1988.NotaryPublicLYNNJ.HAUCKNotaryPrrtraeintheStateoiNewYorkMONROECOUNTY Commission ExpiresiVov.30.l9R5 Attachments xcU.S.NuclearRegulatory Commission RegionI475Allendale RoadKingofPrussia,PA19406xc:GinnaSeniorResidentInspector ATTACHMENT AEvaluation.

ofSafetyInjection Recirculation Systemand,Justification forContinued rationReference (a):LetterURGE-87-670 fromS.P.Swigert(Westinghouse) toS.M.Spector(RG&E),datedDecember8,1987Reference (b):AgingandServiceWearofAuxiliary Feedwater PumpsforPWRNuclearPowerPlants,Volume1,Operating Experience andFailureIdentification, NUREG/CR-4597-VldatedJuly1986NRCBulletin88-04requested anevaluation oftwoconditions which,ifexisting, couldresultindamageorfailureofsafety-relatedpumps.First,theconcernexiststhatwhentwopumpsareoperatedinparallel, theweakerpumpmaybedeadheaded bythestrongerpumpwhenthepumpsareoperating intheminimumflowmode.Second,theconcernexiststhattheamountofminimumflowmaynotbeadequatetoprecludepumpdamageorfailureifevenasinglepumpisoperating.

Shouldeitheroftheseconditions exist,theplannedshorttermandlongtermmodifications aretobeidentifiedaswellasthejustificationforcontinued operation.

R.E.Ginnautilizesatwo-loopWestinghouse designedreactorcoolantsystemandhasthreehighheadsafetyinjection pumpsTheexistingsafetyinjection systemisdepictedonFigure6.3-1oftheUFSAR.Alinediagramdepicting thepresentSIrecirculation systemisalsoshownonFigureA-1ofthisattachment.

Eachsafetyinjection pumpisprovidedwitha3/4-inchlinecontaining apressurebreakdown orifice,acheckvalveandalockedopenmanualvalve.Thedischarge linesfromthe"A"pumpand"B"pumphavetrainseparation, leadingtotheRCSloop"B"andloop"A"coldlegs,respectively.

The"C"pumpisaswingpumpthatensuresdeliverytobothtrainswillexistintheeventthateitherthe"A"or"B"pumpfailstostart.Whenallthreepumpsareoperating, theflowfromthe"C"pumpisdirectedtobothtrains.The3/4-inchlinesfromthethreepumpsjoinandforma1-inchlinewhicheventually leads'backtotheRWST.However,asnotedinWestinghouse letter,Reference (a),theplacement oftheorificeupstreamofthisjunctionreducesthepressurelowenoughsothataweakerpumpwillnotbeprevented fromdelivering itsminimumflow.Theorificeintheindividual recirculation lineessentially desensitizes thesystemtostrong/weak pumpminiflowconcerns.

Sincethe3/4-inchrecirculation linesarelocatedupstreamofthepump'sdischarge checkvalves,andthefactthatanorificeislocatedineachminiflowline,theoperating pressureofastrongerpumprunninginparalleldoesnothavethepotential topreventrecirculation flowthroughthisline,anddeadheadtheotherpump.

AsshownonUFSARFigure15.6-12,duringtheinjection phaseSIflowwillbepumpedintotheRCScoldlegswhenpressuredropsbelow1400psia.Ofthebreaksizesrequiredtobeanalyzedinsection15.6oftheUFSAR,onlyabreaksizebelow4-inchdiameterwouldresultintheRCSremaining above1400psiaforanyappreciable time.The4-inchbreaktransient wasthesmallestsizeanalyzedandresultedintheRCSpressurefallingto1400psiainapproximately 20seconds.Theelapsedtimetogeneratethesafetyinjection signal,actuatethestartsequenceandbringthepumpstofullspeedisontheorderof20seconds(assuming lossofoffsitepower).Ifoffsitepowerremainsavailable, theSIpumpswouldbebroughtuptospeedin10seconds,sothatpotentially apumpmightexpecttooperatewiththedesignedrecirculation capacityforonly10secondsbeforethepumpdischarge pressurewouldovercomethesystempressure.

Therefore, forthebreaksizesanalyzed, representing theboundingcases,thepumpswouldnotbeexpectedtooperateforanappreciable timeintherecirculation mode.Forprimarysystembreakssmallerthan4-inchdiameter, secondary sidesteambreaks,steamgenerator tuberupture,rodejectionaccident, oraninadvertent SIsignal,thesafetyinjection pumpsmaybecalledupontooperateintheminimumrecirculation flowmodeforshortperiodsoftime.Procedure reviewandtrainingonthe:-ulatorconservatively hasshownthattheoperating timeinthe."irculation flowmodewill'otexceed60minutes.Thedesignedminimumflowis30gpmatshutoff,i.e.,whenthedischarge valveisclosed,30gpmrecirculates continuously throughthepump.Th'erearenoautomatic valvesintheindividual recirculation lines,sothateachlineisalwaysopen.Thesafetyinjection pumpsareWorthington Corporation, horizontal, 11stage,centrifugal pumpsdrivenbyWestinghouse 350hp,3600SRPM,A.C.motors.Theminimumflowrecirculation systemforthesafetyinjection pumpswasdesignedtoprovidepumpprotection forshorttermoperation andwasbasedonlimitingtheheatriseoftheinternalfluidwithinthepumptoacceptable values.Considering theboricacidstoragetanksasthesourceofwaterforthepumpsandtheheattracingapplied,thecalculated temperature riseis40F.Thisiswellwithinestablished limitsgiventheoperating conditions andthedesignbasis.Theallowable temperature risetosaturation conditions isover100F.Thedesignofthepumpprovidesforprocessfluidcoolingforthemechanical seals,usingseparatesealwaterheatexchangers, usingcomponent coolingwaterandservicewaterjacketcoolingforthepumpballthrustbearing.Althoughthevalueof30gpm(orabout7%ofbestefficiency point;BEP)wouldbeconsidered lowusingcurrentguidelines ofminimumflowdesign,RG&Econsiders itaconditionally acceptable valuebasedonthefollowing:

1)Itprovidespumpprotection forshorttermoperation only2)Demonstrated performance ofthesepumpsduringperiodictestingatlowflowssubstantiates acceptability.

A-2 tTrendsinminimumowdesignphilosophy and,moreimportantly, developments intheeffectsofsuctionrecirculation onpumphydraulic performance andimpellerweardegradation inthelongterm,hasledpumpmanufacturers torecommend significantly highervaluesofrecirculation flow.Theserecommendations areconsistent fromtheperspective that1)oftenthespecified designoperating conditions anddurationisnot,orcannot,bewelldefinedand,2)themanufacturers havenocontrolovertheusers'peration ofthepump,yetriskliability byassociation ifthecomponent sufferspremature degradation.

Consequently, minimumflowrecommendations fromthemanufacturers, unlessotherwise noted,applytocontinuous longtermoperation andareontheorderof25:ofBEP,depending onpumpparameters suchaspumpedwatertemperature, rotatingspeed,singleormulti-stage design,horizontal andverticalmounting, wearingringmaterials, pumpspecificspeed,brakehorsepower, suctionconditions, andothers.Operation intherecirculation modefor24hoursperdayorsomefinitenumberofhourspermonth,perhapsontheorderof100hourspermonth,hasthepotential tocauseaggravated impellerdamageandpremature internalwearifampleflowisnotprovided.

Quantifying theoperating timetothedegreeofanticipated wearcanonlybedetermined empirically.

Areportonagingandservicewearofauxiliary feedwater pumps,Reference (b),makesreference totheeffectsofsuctionrecirculation andcausesofinternalwear.Basedonthisreasoning, themanufacturerhasnowrecommended anominalvalueof1/3ofBEP(150gpm)forcontinuous operation, andseatedthatoperation atlowerflowsisafunctionofownerrelatedexperience.

Whilethistrendofproviding minimumflowrecommendations basedonlongtermoperation isinthebestinterestoftheownersandsuppliers, attheexpenseofoversizing thepumpsandrelatedcomponents, earlierdesigns,whichproducedminimumflowconsistent withshorttermoperation, canbeshowntobeadequate.

RG&E'sexperience withtheSXpumpperformance duringperiodictestinghasbeenexcellent.

Therehavenotbeenanymaintenance problemsassociated withthepumps.Monthlylowflowperiodictestsareperformedatessentially thesameconditions astherecirculation modeofoperation.

Procedure PT-2.1specifies operation ofeachpumpwithflowthrougha3/4-inchtestlinewiththeflowthrottled to53gpm(Technical Specification valuesetinSection4.5.2.1).

Thisrepresents theonlyflowthroughthepumpduringthetest,becausethemanualvalveintherecirculation lineisclosedduringtesting.Thevalueof53gpmisnotjudgedtorepresent asignificant increaseinflowover-thevalueof30gpm,whichisthedesignedflowthroughtherecirculation line.Asameansofsubstantiating this,aspecialtestwasconducted onApril6,1988.Duringthistest,valvealignments wereslightlymodifiedfromthestandardperiodictestprocedure inordertooperatethe"A"pumpwithaminimumflowof30gpmthroughthetestlineandcomparethepumpoperation andvibration withthatforaflowof53gpmaswellasaflowinexcessof100gpm*.AsummaryoftestresultsisincludedonpageA-5.

Reviewofthisdataandperiodictestdataoverthepastseveralyearsindicates thatthereisnoappreciable difference inpumpvibration levelsrelativetoflowsfrom30gpmto100gpm.Thevaluesrecordedforeachofthe3pumpsaretypicalofthoseseenduringmonthlyperiodictesting,andareconsistent (orlowerthan)the-referenced valuesestablished duringtheinitialtestprogramforthesepumps.Thevibration levelsarewellwithintheacceptable rangeof0to1mils.Testingofthesafetyinjection pumpsduringmonthlyperiodictestingrequireslowflowoperation forperiodsgenerally of30minutesto1hour.Thepumpsmayalsobeoperatedatlowflowsforbriefperiodstomaintainaccumulator levelorforrefilling theaccumulators whendrainedduringrefueling shutdowns.

Thetotaloperating timeforthe3pumpscombinedforthefirst18yearsofplantoperation hasbeenapproximately 600hours.The"C"pumphasloggedthemostnumberofhours,280.Therefore, continued plantoperation utilizing thepresentrecirculation systemdoesnotcreateacondition whichisoutsideoftheoriginaldesignbasis,forthesystemnordoesitrepresent anunacceptable condition intermsofpumpprotection forthemaximumdurationthatthepumpswouldbeexpectedtooperateintherecirculation mode.Thebasisforcontinued opera:isthattheminimumrecirculation flowpipingdesignw~...individual linesbranchedoffthepumpdischarge lineupstreamofthecheckvalveandanorificeplacedineachindividual recirculation linepreventsastrong/weak pumpinteraction.

Also,thetestresultsconducted forthefirst18yearsofplantoperation demonstrate thatpumpperformance, asreflected byvibration measurements, isnotsignificantly different between30gpmand100gpm,thelatterrepresenting about25tofBEP.Theremainder ofthisevaluation isintendedtofocusontheplannedplantmodification toprovideincreased pumprecirculation capacity.

Thebasisforthismodification istoreducethepotential forpumpinternalaccelerated wearasaresultoflowflowoperation andtherequiredperiodictesting.Thegreatmajorityofpumpoperating hourshavetodateoccurredduringlowflowoperation.

Operation ofthepumpsduringanyofthepostulated accidentconditions willnotresultinoperation inminimumflowmodeformorethan1hour,duetoprocedural SItermination.

Byinstalling newresizedrecirculation pipingpresently plannedtorecirculate 100gpmwhenthepumpoperatesatshutoff,apositivelongtermeffect,althoughunquantified, isexpectedoverthelifeofthepumps.Themodification willprovidethecapability toperformperiodictestsat150gpm."Existing flowmeter inthetestlineis70gpmfullscale.Duringthetesting,thethrottlevalvewasfullopenandindicated inexcessoffullscale.Combinedwithaflowofabout30gpmthroughtherecirculation line,theflowwasdetermined tobeinexcessof100gpm.A-4 ttSUMMARYOFTESTRESULTSFORSPECIALMINIMUMFLOWTESTCONDUCTED ON"A"SAFETYINJECTION PUMP30~i"A"SIPUMPFLOW53c3pm)100~mAcceptable Rangeat53~mReference Valuesat53~mDeveloped

pressure, psigVibration, milsIBVerticalIBHorizontal OBVerticalOBHorizontal 14460.0570'840.190.1514430.060.0880.180.1414130.0480.0820.170.1351420-1481 0-1mils0-1mils0-1mils0-1mils14520.0550.090.240.11Developed

pressure, psig"B"PUMP953GPM1457Acceptable Raneat1420-1493 Reference Valuesat1464Vibration, milsIBVerticalIBHorizontal OBVerticalOBHorizontal 0.0520.0960.120.110-1milsO-lmils0-1mils0-1mils0.0650.120.210.11Developed

pressure, psig"C"PUMP953GPM1459Acceptable 1420-1488 Reference Valuesat1459Vibration, milsIBVerticalIBHorizontal OBVerticalOBHorizontal 0.0810.170.0880.140-1mils0-1mils0-1milsO-lmils0'950.160.350.16IB=Inboard(coupling) endOB=OutboardA-5 Therecirculation rateof100gpmrepresents avalueof25%ofBEP.Thisvaluewasselectedbaseduponfourcriteria:

1)Thevalueapproached themanufacturer's recommendations forlongtermoperation, reducinghydraulic anddynamicforcesimposedonthepump.Themanufacturer recommended 150gpmforlongtermoperation, butmadenoexplicitrecommendation forshortertermoperation.

sinceitisdependent uponownerexperience.

Theproposedmodification willenabletestingat150gpm./2)Thevaluewas*considered themaximumrecirculation flowwhichwillstillallowtherequiredSIdeliverytotheRCSunderallaccidentconditions.

Therequireddeliverycurveispresented intheUFSARFigure15.6-12.3)Thevaluewillreducethecalculated pumpedfluidtemperature increasethroughthepumpfourfold, i.e.,toa10Ftemperature rise.4)Thevaluewillincreasetheoperating lifeofthepumpballbearingsbyreducingthebearingloadsduringperiodictesting.Thepresentperiodictestingisconducted usingaparatetestlinewiththerecirculation linesisolated.

Thetestflowisadjustedto53gpmandpumpperformance ismonitored inaccordance withASMESectionXIArticleIWP-3100, withoneexception asdiscussed below.Inordertodetermine theeffectofincreased recirculation flowoverthatoftheexistingsystem,30gpmperpump,ahydraulic modelusingtheKYPIPEcomputercodefortheECCSsystemhasbeendeveloped.

Thepumpmanufacturer, Worthington Corporation, doesnotrecommend bearingtemperature measurements atlowflowsasagoodindication ofbearingcondition.

Vibration readingsprovideabetterindication oftheircondition.

Atlowflows,bearingtemperature measurements areexpectedtobehigherduetoincreased axialandradialthrustloadsinthisregion.Thetemperature riseofthepumpedfluidisalsogreater,andthisheatisconducted tothebearingsthroughthepumpshaft.Thesefactorstypically resultedinbearingstabilization timebeingextendedbeyondthatwhichwouldbenormallyexpectedforthepumpatitsrated,flowcondition.

Stabilization undertheselowflowconditions duringpreviousperiodictestsoccurredtypically afterabout1hour,andsometimes 2hours.Retestswerenotuncommonduetotemperatures exceeding the160Fimposedlimit,whichwas,infact,themanufacturer's recommended limitattheratedcondition of300gpm.A-6 Ithasalsobeenshownthatbearingstabilization temperature anddurationisdependent onotherfactors,suchasambientairtemperature andthrustbearinghousingcoolingwatertemperature (servicewater).Thiswasconfirmed duringaseriesoftestsperformedoneachofthe3pumpsduringthewinterof1981andthewinterandsummerof1982.Consequently, itisnotpossibletoobtainrepeatable, consistent andmeaningful resultsrelatable tobearingcondition.

Obtaining thisonlymarginally usefuldatawasoutweighed bythenegativeimpactofsubjecting thepumptounnecessarily highloadsinthelowflowtestcondition forextendedperiodsoftimeeachmonth.Overall,thiswasviewedashavingadetrimental effectonbearinglife,andultimately pumpserviceability.

Consequently,

'bearingstabilization testswerediscontinued inJanuary1983.Stabilization temperature wasnotnoticeably changed.bytestflowsbetween50to80gpm.Thetestingperformed in1981and1982didconfirmthatthebearingtemperatures willstabilize inthelowflowmodeandnotreachunacceptable levelsconsidering thedesignlaketemperature of80Fforsummeroperation.

Theproposedmodification willallowaflowduringtestinginexcessofthe150gpmrecommended bythemanufacturer forcontinuous operation bycombining themodifiedrecirculation lineflowandexistingtestlineflow.Nearlydoublingtheflowfrom80gpmto150gpmwillplacethepumpinanoperating regionwhereresidualaxialthrustcanbeexpected.

tob.-:educed.Sincethevalueof150gpmrepresents aflowjudgedacceptable forlongtermoperation, thiscondition canbeviewedasanimprovement tooverallbearinglife.TheKYPIPEcomputercodehasbeenverifiedagainstactualplantdatatakeninMarch1971whichwasperformed byWestinghouse tore-evaluate thecapability ofthesafetyinjection system.Theplannedrecirculation systemwillconsistofasinglerecirculation linesizedfor100gpmforeachpumpwithpressurebreakdown floworifice,checkvalveandmanualvalve.Thisconfiguration waspreferred asitrepresents apassivetypesystemwithnoactivecomponents, andtherefore meetsthesingleactivefailurecriteria.

Itrepresents amorereliableoptionthantheotheractive/passive parallelpathsystemsconsidered whichwouldrequirecontrolsystemswithairand/orelectrical interfaces.

Aschematic oftheplannedsystemisdepictedonFigureA-2..Thecriteriausedtoestablish themaximumallowable recirculation flowwasthedeliverycurveshownonFigure15.6-12oftheUFSAR(attached FigureA-3).Thehydraulic modelisconservative forthefollowing reasons:a)Theexistingpumpperformance curvewasdegraded5percenttoevaluatetheabilityofthesafetyinjection systemtomeettheUFSARSIdeliverycurve.Periodictestprocedure PT-2.1containsactionrequirements whenthepumpperformance degrades3-o-4%ofthereference pressure.

A-7

(~b)Thehydraulic modelpredictslessSIfowthanthatwhichwasactuallymeasuredduringthefullflowtestingperformed.

inMarch1971.c)Thehydraulic modelassumedtheminimumSIsuctionheadforallrangesofSIflow.minimumsuctionheadisequivalent totheRWSTlevel,thepointatwhichtheSIsuctionistransferred tothecontainment viatheRHRpumps.pumpThe15'tpumpsumpd)Thedetermination oftheresultant SIdeliverycurve,utilizing themodified(100gpm)recirculation flow,wasbaseduponthehighestresistance flowpath(pump"C"deliverytolooplIAll)Theresultant safetyinjection deliverycurveexceedstherequirement shownonFigure15.6-12oftheUFSAR(attached FigureA-3),utilizing theproposedrecirculation designwhichwillallow100gpmrecirculation flowatshutoff.Inordertoreducevelocities toacceptable values,eachrecirculation linewillbeincreased from3/4-inchtol-inch,andtherecirculation manifoldlinewillincreasefrom1-inchto2-inch.Theexisting1-inchmanifoldlinecontains2airoperatedvalvesinseries.Duringthesumprecirculation mode,therecirculation manifoldlineisisolatedtoprecluderadioactive contamination oftheRWST.Currently these2valves,AOV897andAOV898,aremechanically blockedopentopreventclosureonlossofairsupply.Closurewouldblockofftheminiflowprotection forallthreeSIpumps.Theymustbemanuallyunblocked andclosedpriortotheinitiation ofthesumprecirculation phase.Aspreviously reportedunderRG&E'sresponsetoIEBulletin86-03,RG&Eplanstoreplacethesevalves.'PThecriteriausedforthereplacement valvesis1)toensurethatnosingleactivefailurewouldpreventtherecirculation linefromremaining openforpumpprotection, 2)toensurethatnosingleactivefailurewillpreventtherecirculation linetobeisolatedpriortothesumprecirculation phase,and3)thatvalvemanipulation canbeperformed fromtheControlRoom.Thepresentschedulecallsforthenewrecirculation piping,valvesandcontrolstobeinstalled duringthe1989refueling outage,withpre-operational testingandthemodification turnedoverforusebytheendof1989.A-8 FIGUREA-1.SISMINIFLOWRECIRCLATIONSYSTEM3/4'.O.RESTERKlltieORIFICE(IYP)IMAwiFoLDLINE.3/4'.O.~8973/4'.O.098RltSTSlPOPAREACTOf1LOOP,8eASISIPlNPCRllSTslRNpeI]RFACIORLOS'THROTTL'C VALVETESTLIREI FIGUREA-2MODIFICATION SCHEMATIC MRFLlMIMARYPFICPu~p9PO~Pl00GRILOORIFICE2'/i3/4'eLO3/i'3/4'QISIR.OITESTLCP3/4'OAY~~wrrcxaLOLESSCSPLY2'ESTLCOPLOjOPTIONAI

)SIRlOA(TYPIC'LE%%:Dllllt8Ll%llllEQlIRCÃlIEIlllHlGISTIl8LMAlOEml18%Rf%ÃFIEO%VLI%AIOEmJIP%N1IHSTlLLfD

>0RvST

~II>II~IIIIIIII'~IIliII~I~e00~II~

ae~~ATTACHMENT BEvaluation oftheResidualHeatRemoval(RHR)Recirculation Sstemand.Justification forContinued ration

Reference:

(a)FlowMeasurement EnineerinHandbook, R.W.Miller,McGrawHillBookCompany,NewYork,1983,Chapter4NRCBulletin88-04requested anevaluation oftwoconditions which,ifexisting, couldresultindamageorfailureofsafety-relatedpumps.First,theconcernexiststhatwhentwopumpsareoperatedinparallel, theweakerpumpmaybedeadheaded bythestrongerpumpwhenthepumpsareoperating intheminimumflowmode.Second,theconcernexiststhattheamountofminimumflowmaynotbeadequatetoprecludepumpdamageorfailureifevenasinglepumpisoperating.

Shouldeitheroftheseconditions exist,theplannedshorttermandlongterm-modifications aretobeidentified aswellasthejustification forcontinued operation.

TheRHRpumpsatR.E.Ginnaserveastheresidualheatremovalpumpsusedduringnormalplantcooldown, heatupandrefueling aswellasthelowheadsafetyinjection pumps,functioning aspartoftheemergency corecoolingsystem(ECCS).Thereare4modesduringwhichtheRHRpumpsareplacedinoperation:

1)Lowheadsafetyinjection (LHSI)aspartoftheECCS,withpumpstakingsuctionfromtherefueling waterstoragetank(RWST).Thisistheinjection phaseandrepresents thenormalplantalignment uponreceiptofanSIsignal.2)Lowheadsafetyinjection (LHSI)aspartoftheECCS,withpumpstakingsuctionfromthecontainment sump.Thisrepresents therecirculation phasefollowing depletion ofRWSTto15%level.3)Residualheatremovalmodeduringplantcooldownandduringrefueling operations.

TheRHRpumpstakesuctionfromtheRCSloop"A"hotleganddischarge backtotheRCStoloop"B"coldleg.DuringthismodeRCSpressureandtemperature arebelow350psigand350F.Pumpoperability testingisalsoconducted monthlyduringshutdowns atapumpflowof700gpm.

4)Testmode.Monthlyperiodictestingisperformed inaccordance withASMESectionXIArticleIWP-3100duringnormalplantoperation.

Pumpsareoperatedseparately andtakesuctionfromtheRWSTanddischarge throughtheminiflowrecirculation lineandbacktothepumpsuction.TheRHRsystemconfiguration isshownonFigure5.4-7oftheUFSAR.Thepumpsdischarge intoparallelpathseachwithacheckvalve,residualheatexchanger, andcontrolvalveinprimarily 8-inchpiping.Thedischarge linesmergeintoacommon10-inchsectionofpipingwhichleadsinsidecontainment.

This10-inchlinedirectstheflowthroughtwomotoroperated, valvestotheloop"B"coldleg.ThesenormallyclosedvalvesareopenedfornormalRHRoperation duringplantcooldown.

Two6-inchbranchlinesleadingtothereactorvessel,areautomatically alignedforlowheadsafetyinjection uponreceiving anSIsignal.A2-inchminiflowrecirculation linebranchesoffthe10-inchcommondischarge lineandleadstoasectionof10inchRHRsuctionpipingwhichsplitsandleadsbacktoeachoftheRHRpumps.'heminiflowrecirculation linecontainsamanualglobevalveandflowmeterFI-628.Theminiflowlineisalwaysopen.ThisminiflowlineservesbothpumpsandprovidestheneededflowthroughthepumpswhentheRCSpressureexceedsRHRpumpshutoffhead.Thesystemisdesignedtoprovide200gpmminimumflowwithonepumprunningor100gpmperpumpifoperatedinparallel.

Westinghouse initially notifiedRG&Eofapotential concernforthoseplantsthathadacommonrecirculation lineandidentified R.E.Ginnaasaplantthatwaspotentially affected(Westinghouse letterstoRG&EdatedOctober26,1987,November30,1987andDecember8,1987).NRCInformation Notice87-59datedNovember17,1987wasissuedidentifying theconcernswhichwereagaindocumented inNRCBulletin88-04datedMay5,1988.RG&Einitiated aCorrective ActionReport(CAR1828)following receiptoftheinitialWestinghouse correspondence.

TheRHRpumpsaresinglestage,horizontal, coupledcentrifugal pumps,Model6-SVC,manufactured byPacificPumpDivisiondrivenbyWestinghouse ElectricCorporation 200hpA.C.motors.Thepumpsareratedfor1560gpmat280ft.totalheadat1770rpm.Reviewofthemanufacturer's initialpumpperformance curvesindicated thatthepumpsarewellmatchedforparalleloperation, however,duetotheflatnessofthecurvesatlowflowswhichistypicalofpumpswithlowspecificspeedsuchasthese,smalldifferences intheactualpumptotaldeveloped headorsystemresistance cancauselargedifferences intheindividual pump'soperating flow.Areviewofseveralmonthsofperiodictestdataperformed intheApril-September 1987timeframe demonstrated thatthedifference betweenthetwopumps'eveloped pressurewasabout1or2psiatthe200gpmflowrate.

Theimpactofthiswasthen.evaluated.

B-2 TheRHRsystemcontainsoneflowmeter (FZ-626)locatedinthe10-inchcommondischarge line,whichmeasuresthetotalflowfrombothRHRlegs,therefore, confirmation oftheactualflowfromeachpumpduringparallellowflowoperation necessitated aspecialtest.Twotestswereconducted, onDecember16,1987andDecember22,1987,duringwhichanultrasonic typeflowmeter wasinstalled onthe>>B>>RHRdischarge leg.Byrecording simultaneously theflowontheminiflowrecirculation lineflowmeter FZ-628andtheultrasonic flowmeter inthe>>B>>pumpleg,flowinbothlegsoftheRHRpipingwasdetermined.

Pumpbearinghousingvibration, casingvibration, sealwaterpipingtemperature, casingtemperature andmotorcurrentwerealsomonitored andrecordedduringthetest.Thetestingdetermined thatwithbothpumpsoperating inparallelatatotalflowof200gpm,the>>A>>pumpreceivedonlyabout5gpmandwas,therefore, essentially deadheaded.

Operation intheparallelmodewasterminated 51/2minutesafterstart.Thetotaldischarge pressureconfirmed the>>A>>pumptobeonly2psilessthanthe>>B>>pumpat200gpmwheneachpumpoperatedseparately.

Basedupontheshapeofthehead-capacity curve,however,forthesepumps,a2psidifference wouldresultinthetotalflowwithbothpumpsoperating inparallelrequiredtobeabout500gpmbeforetheweaker>>A>>pumpwouldbeabletoovercometheheadatthisoperating point.Znordertomorecloselyestablish theoperational limitations whileinparallel, areviewwasmadeofalargersampleofmonthlyperiodictestdata.Thistes'tingisconducted duringnormalplantoperation utilizing theminiflowrecirculation line.Theflowrateissetat200gpmbythrottling manualglobevalve822B.Eachpumpistestedseparately.

Dataoverthelast24monthshasbeentrendedtoestablish thedifference betweenthedifferential pressures ofthe>>A>>and>>B>>pumpsconsidering thepressuregagerepeatability.

Thesamedischarge pressuregage,PI-629,isusedfortestingeachpump.Suctionpressureisdetermined byanelevation headcalculation based.onRWSTwaterlevel.Testingisdoneonbothpumpssequentially usingthesameprocedure.

Thesamepersonnel performed thetestingonthe>>A>>aswellas>>B>>pump,hence,gagereadingsareexpectedtobeconsistent foraparticular monthlytest.Sincethesamegageisused,thecalibration andbiaserrorofthegagewouldremainunchanged foraparticular monthlytest,andnotbeafactorindetermining thedifference indifferential pressures betweenpumps>>A>>and>>B>>.Thetrending, therefore, providesthebasisfordetermination ofthemeanvalueofthepressuredifference andstandarddeviation ofthedata.Thetrendingshowsthatthemeanvalueforthedifference indifferential pressures is1.76psi~Thedatapointsreviewedoverthelast24monthsofperiodictestingindicatethatthepressureindicator producesreasonably repeatable pressurevalues.Theaccuracyofthevalues,thatis,thecloseness tothetruevalue,whichincludestheeffectsofbothprecision andbiaserror,is+1.6%forthe>>A>>pumpdataand+1.2:forthe>>B>>pumpdata.Bydetermining thedifference B-3

'I'inpressurereadingsandaccounting fortwostandarddeviations, theadjustedpressuredifference wasdetermined tobe3.76psiforanalysispurposes(Reference a).Ananalysisofthelossesduetopipingandcomponents showsthattheheadlossinthe"B"pumpand"A"pumpdischarge sectionsofpipinguptothejunctionpointofthetwoloopsiswithin0.2feetofwateroressentially equivalent atthe200gpmflowrate.

Theheadlossencompassed pipingandcomponent lossessuchaselbows,reducers, heatexchangers, andvalvesaswellaselevation headlosses.Therefore, thedifference inpumpdifferential pressures recordedisduetoindividual pumpcharacteristics, notdifferences intheindividual RHRdischarge lineresistances.

Thepurposeofthedatareviewandtrendingwastoestablish avalueforthedifference ofpumpdeveloped pressurewhichwouldprovidethebasisforestablishing operating limitations forparallelpumpoperation.

Thisadjustedpressuredifference resultedinacondition wherebythetotalflowwithbothpumpsoperating inparallelwouldneedtobeabout700gpmbeforetheweaker"A"pumpwouldbeassured,ofovercoming theheadattheoperating point.Following RG&Enotification oftheminiflowconcernbyWestinghouse, calculations wereperformed todetermine thefollowing:

1)Thecalculated temperature

=riseinthefluidwithinthecriticalareasinternaltothepumpversustheflowratethroughthepump.2)Thetimetosaturation ofthewaterwithinthecriticalinternalareasofthepumpafterbeingstartedinthedeadheaded condition.

TheRHRpumpsarenormallyalignedtotheRWSTwhichismaintained ataminimumwatervolumebyTechnical Specification, therefore, theelevation headabovetheRHRpumpsuctioncenterline isknown.Themaximumwatertemperature of80FfortheRWSTwasusedinthecalculations.

Thevolumeofwaterwithinthepumpwasobtainedwiththehelpofthemanufacturer.

Thetemperature riseversuspumpflowcurvewasplottedwiththeresultsthata40gpmflowwouldproducea10Friseanda20gpmflowwouldproducea21Friseintheinternalfluid.Theresultsalsoindicated thatifstartedinadeadheaded condition, theinternalfluidcouldreachsaturation conditions in6.7minutes,assumingnoheatlossesfromthepump,noheatlossduetothecomponent coolingwaterjacketcoolingwhichisprovidedaroundthepumpcasingcover,andallthepumpbrakehorsepower isconverted immediately intoheatupofthepumpfluid.B-4

~~~Theminiflowrecirculation linewasoriginally sizedbaseduponpumpinternaltemperature riseandwasintendedtoprovidepumpprotection againstoperation intherecircmodeforshortperiodsoftime.Thecapacityoftheminiflowlineisinexcessof200gpmwithonepumpoperating, sinceV-822Brequiressomethrottling inordertoachievethisflow.Consequently, forparalleloperation, ifbothpumpsweretoequallysharethisflow(asistheidealcondition inparalleloperation),

eachpumpwoulddeliver100gpm.Baseduponthisandthecalculated temperature risecurve,theminiflowlinewasadequately sizedandwouldresultinatemperature riseofonly3.9F.Regarding the6.7minutestosaturation time,duringthespecialtestrunonDecember22,1987the"A"pumpwasdetermined tobeessentially deadheaded for5.5minuteswhilethepumpparameters weremonitored.

Sealreturnpipingsurfacetemperature andcasingsurfacetemperature weremonitored.

Thesealreturntemperature didnotshowanoticeable increase.

Thecasingsurfacetemperature indicated 85Fwhenthepumpwassecured.Bearinghousingvibration atthehighestlocation(inboardendvertical) increased from0.75mils(normal)to1.2mils(deadheaded)

.Casingvibration increased from0.20mils(normal)to0.32mils(deadheaded) onthe"A"pump.Ingeneral,thetrendinthebearinghousingvibration levelswasanincreaseof10-20%asflowwasreducedfrom200gpmto50gpm.Onthe"A"pump,whichwasinfactdeadheaded duringthe51/2minutespan,thevibration increased from0.80milsat50gpmto1.2mils(inboardvertical) or50%,yet,theselevelsthemselves aretypicalfortheequipment andarewithintheacceptedstandards.

Consequently, testdatarecordedintherangeof200gpmto50gpmdidnotindicateasignificant increaseinpumpvibration, orothermonitored parameters, althoughthecasingtemperature didindicatethefluidtemperature wasincreasing, aswouldbeexpected.

Basedupontheforegoing summaryofeventsandinformation, allprocedures whichrequireoperation oftheRHRpumpshavebeenchangedtorequirethatwhentwopumpsarerunningandthetotalflowindicated onflowindicator FI-626islessthan1200gpm,onepumpisstopped.Conversely, ifonepumpisrunningwithgreaterthanorequalto1200gpm,thesecondpumpisstartedifavailable.

Itisacceptable tostopanRHRpumpbecauseasinglepumpwillprovideallrequiredaccidentflow.Theseprocedural modifications havebeenmadeasaninterimmeasurejustifying continued operation untilaplannedplantmodification canbeinstalled whichwillprovideaseparateandindependent recirculation flowloopforeachpump.Additionally, anoperatoraidtaghasbeenplacedontheFI-626indicator intheControlRoomwhichidentifies thisoperating limitation.

Trainingontheseprocedure changesisbeingprovidedtotheoperators onanongoingbasis.Theprocedure changeshavebeeninsertedintheimmediate actionstepsoftheemergency procedures.

Monitoring thistripandrestartcriteriaisperformed onacontinual basisB-5 asrequiredontheprocedure foldoutpage.Combinedwiththeoperatortrainingandtheoperatoraidtag,ahighconfidence existsthatoperation ofthepumpsinparallelatacondition whichcouldresultinoverheating ordamagewillnotoccur.Thesetpointof1200gpmhasbeenchosentoaccountforinstrument uncertainty andtoprovideanacceptable flowsplitduringparalleloperation.

Utilizing theperiodictestdata,it.hasbeendetermined thattotalflowsaboveabout700gpmwillensurethattheweakerpumpwillovercometheoperating head.Considering theeffectsofinstrument uncertainty ofapproximately 3%,whichwouldresultinanadditional

required, marginof130gpm,thevalueof1200gpmwillensurethatbothpumpswilldeliveranadequateflowwhileoperating inparallel.

Ithasbeendetermined thatanindicated flowrateof1200gpmonFI-626intheControlRoomwouldresultinnolessthan1070gpmtotalflowconsidering themostlimitinginstrument uncertainty.

Ifthiswerethecase,the"A"pumpwouldreceive300gpmandthe"B"pump770gpm.Anactualtotalflowof1200gpmwouldbesplitsuchthatthe"A"pumpwoulddeliver390gpmandthe"B"pump810gpm.Therefore, thesetpointprovidesaflowsplitwherebytheweaker"A"pumpreceivesnominally about20percentofBEPflowandanamountconsistent withcurrentmanufacturer's recommendations.

Thepumpmanufacturer, PacificPumpDivision, hasrecommended arecirculation flowof100gpmperpumpforshortdurations upto30minutes.Therecommended valueforintermediate rangeoperation, thatof100hourspermonth,is260gpm(15%ofBEP)andlongtermcontinuous operation avalueof520gpm(30%ofBEP)isrecommended.

Themajorityofoperating timeforthepumpsisspentintheRHRmodeduringplantheatup,cooldown, andrefueling operations.

OneorbothRHRpumpsoperateforatotalofabout1000hoursperyearintheRHRmode.'heprecisetimesplitbetweenthe"A"pumpand"B"pumphasnotbeendetermined.

Operation inthetestmodeof200gpmisperformed monthlyexceptduringannualrefueling shutdowns whentheyaretestedat700gpm.Thiswouldamounttoadurationontheorderof25hoursperyear(orless)orabout5'tofthetotaloperating time(assuming 500hoursperpump).Sincethe200gpmvalueissufficiently closetotherecommended valueof260gpmrecommended bythemanufacturer foroperation upto1200hoursperyear(100hourspermonth),theapproximately 25hoursspentat200gpmwillnotresultinaccelerated wearonthepumpinternals.

Asafetyanalysiswasperformed whichsupported theprocedural modifications whichrequiresecuringoneRHRpumpasdescribed above.Parallellowflowoperation ofRHRpumpsatGinnamayoccurduring:B-6 1.SAFETYINJECTION (INJECTION PHASE)DuringtheECCSinjection phase1of2RHRpumps,2of3highheadsafetyinjection pumpsand1of2SIaccumulators areassumedtoinjectwater.2.SAFETYINJECTION (RECIRCULATION PHASE)DuringtheECCSrecirculation phase,1of2'RHRpumpsandheatexchanger willsatisfythelongtermcorecoolingrequirements of10CFR50.46.

3-RHRSYSTEMSOPERATION DURINGAPLANTCOOLDOWNBELOW350PDuringRCScooldownbelow350F,only1of2RHRpumpsareprocedurally operatedatlow-flowconditions precluding RHRpumpdamage.Areviewofprocedures showsthatatnotimedoprocedures usedtoinitiate, operate,orshutdowntheRHRsysteminthismodeofoperation requireorrecommend therunningofRHRpumpswiththedischarge pathisolated.

Additionally, therearenoautomatic isolation signalsintheGinnaconfiguration thatwouldplacetheRHRpumpsinthepa'rallel low-flowoperating mode.4.TESTMODEAreviewofperiodictestprocedures andrefueling shutdownsurveillance procedures revealedthatatnotimewouldtherebeparallellow-flowoperation.

Therefore, forallanticipated normaloperating conditions thepotential forRHRpumpdamageislimited,suchthattheRHRpumpsavailability isassuredtothemaximumextentpossiblesuchthattheprovisions of10CFR50AppendixAGDC34ResidualHeatRemovalaresatisfied.

Inaddition, themarginofsafetyandperformance oftheECCSasdescribed in10CFR50AppendixAGDC35aresatisfied.

Theprocedure changesareintendedasaninterimmeasure.Aplantmodification designedtoaffectlongtermcorrective actionispresently underway.

Criticalpipingtie-insforthemodification arescheduled tobeinstalled duringthe1989Refueling andMaintenance

shutdown, withnewhardwareandpipinginstallation, testing,andthemodification turnedoverforusebytheendof1989.Anewrecirculation systemisbeingdesignedwhichwillensureanincreased minimumflowcapacityconsistent withthemanufacturer'recommendations forintermediate termoperation underallanticipated normalandaccidentconditions.

Therequirements willbemetwheneveroneorbothRHRpumpsareoperating.

Thenewrecirculation capacitywillnotpreventtherequireddeliverytothereactorvessel,asdefinedonFigure15.6-13oftheUFSAR,frombeingmet.B-7

~~~Throughtheuseofthe.computermodelKYPIPEdeveloped fortheECCSsystematGinna,theexistingsystemandproposedmodification totherecirculation systemcanbeassessedandimpactonsystemdeliverydetermined.

Theoptimumdesignhasnotasyetbeenfinalized.

Sincethecriteriarequiresanincreaseinrecirculation capacitywhenthedischarge pathisisolatedaswellassufficient capacitytoensurethattheUFSARdeliverycurve(Figure15.6-13)ismet,arecirculation systemwithfloworpressurecontrolcomponents maybenecessary inlieuofapassivepressurebreakdown systemasisnowemployed.

Thesystemredesignoptionwhichisbeinggivenstrongconsideration forimplementation atthistimeisdepictedonFigureB-1.Asrequiredbyparagraph 5ofBulletin88-04,awrittenresponsedescribing theactionstakenwillbeprovidedwithin30daysfollowing completion oftheplantmodification.

Appropriate conservatism willbeappliedtotheKYPIPEcomputercode.Sufficientpumpperformance degradation willalsobeappliedinthemodelsothatpumpmaintenance willbeperformed priortosystemdeliveryrequirements beingreducedbelowtheacceptable limitsestablished intheUFSAR.Pumpdeveloped pressuremeasurements inthemonthlyperiodictestprocedures canbeusedtodictatewhencorrective actionwouldbewarranted.

Additionally, theincreased capacityoftherecirculation systemwillreducetoaminimumt"...possibility ofaccelerated wearduetooperation intheperiodtestmode.B-8 P1427IOB70%71188'00BP)IOIIFICEITYPICAL))HIHXA30704A)IP141710A709A711C8't3o3040~4509%NAXINHITVPlra.)LON'HOTLEB701P14010'"AMBI886FIGUREB-lRHRHINIHUHRECIRCULATION FLOW

~~~rATTACHMENT CEVALUATION OFRECIRCULATION SYSTEMSFORAUXILIARY FEEDWATER SYSTEMReference (a):LetterfromJ.E.Maier(RGGE)toD.M.Crutchfield (NRC),

Subject:

NRCRequirements forAuxiliary Feedwater Systems,datedJanuary8,1982Reference (b):LetterfromJ.E.Maier(RG&E)toD.M.Crutchfield (NRC),

Subject:

NRCRequirements forAuxiliary Feedwater Systems,datedJune8,1981Reference (c):LetterfromL.D.White(RGGE)toD.M.Crutchfield (NRC),

Subject:

NRCRequirements forAuxiliary Feedwater Systems,datedMay28,1980Thesesystemshavebeenevaluated underpreviousSystemati'c Evaluation Program(SEP)TopicXandTMI-2TaskActionPlanItemsII.E.l(fromNUREG-0660).

Aspartofthelatertask,thesepumpsunderwent a48hourendurance testperformedatminimumflowconditions.

Thesetestsdemonstrated continued operability andacceptability ofcriticalparameters atlowflowconditions baseduponlimitsestablished forthepumpsatratedconditions.

Monthlyperiodictestsrequirethateachauxiliary feedwater pumprecirculation systemoperatesasdesigned.

Eachpumpisprovidedwithaseparatelineequippedwithanautomatic airoperated(fail-open) valveandpressurebreakdown orifice.Theselinestieintoacommonlinewhichleadsbacktothetwo30,000galloncondensate storagetanksforthemainauxiliary feedwater system.Thestandbyauxiliary feedwater systemissimilarwiththecommonlineleadingbacktothe10,000galloncondensate supplytank.*MAINAUXILIARY F19KMATER SYSTEMThesystemisdepictedonFigure10.5-1oftheUFSAR.Eachofthethreepumpsisprovidedwitha2-inchrecirculation line,automatically controlled airoperatedvalve,andpressurebreakdown orifice.Thevalveisdesignedtofailtotheopenposition.

The"A"motordrivenAFWpumpiscapableofdischarging 200gpmtothe"A"steamgenerator andthe"B"motordrivenAFWpump200gpmtothe"B"steamgenerator.

TheturbinedrivenAFWpumpiscapableofproviding atotalof400gpmtobothsteamgenerators.

Bothmotordrivenpumpsautomatically startif1)eithersteamgenerator decreases toalow-lowlevelof17%,2)anSIsignalispresent,or3)tripofbothmainboilerfeedpumpsoccurs.Sinceeachmotordrivenpumpisalignedtoitssteamgenerator, aninteraction cannotoccurasdescribed inBulletin88-04.

a(~tTherecirculation valvesforthemotordrivenpumpsystemremaincloseduntilthedischarge pressureinthedischarge linereachestheestablished setpoint.

Recenttestsperformed onthepumpsdemonstrated thatthevalveswerefullyopenat,adischarge flowof80gpm(32%ofBEP).Therefore, thetotalflowthroughthepumpexceeds80gpm,sincetherecirculation lineisalsoopentoprovideadditional flow.Theexistingperiodictestprocedures requireverification thattherecirculation valveisfullyopenedwhentheflowthroughthedischarge isthrottled to50gpm,or20:ofBEP,whichisconsidered adequate.

Therefore, theestablished setpoints providemarginoverandabovetheacceptable valueof50gpm.Theperiodictestalsodemonstrates thattherecirculation valveisfullyclosedwhenthedischarge flowreaches125gpm.Theperiodictestingalsoverifiesthatthevalvewillfailtoopenpositionwhenairisremoved.Themotordrivenpumpsaredesignedtoprovideaminimumof200gpmduringaccidentconditions.

Thedischarge valvesareautomatically throttled toachievethisflow.Thedischarge valvesareonlythrottled tolowerflowsduringplantstartupandlowpoweroperation orduringperiodictesting.Duringthesemodesofoperation, theflowisprocedurally controlled sothataminimumof50gpmisalwaysprovided.

Sincetherecirculation valveopensautomatically, aminimumflowthroughthislineprovidesassurance thatthepumpwillalwaysreceivearecirculation flow.TheturbinedrivenAFWpumpisalsoequippedwitha2-inchseparaterecirculation line,automatic airoperatedvalve(fail-open)andpressurebreakdown orifice.Inthissystemthevalveiscontrolled byflowinthedischarge line,designedtoopenthevalveiftheflowdecreases to100gpm.Periodictestingisperformed onamonthlybasisanddemonstrates thatthevalveisfullyopenat100gpm.Thisflowrepresents 20%ofBEPandistherefore adequate.

Theturbinedrivenpumpwouldonlybethrottled backtothe100gpmcondition duringtestmodes.Theturbinedrivenpumpisautomatically startedonalow-lowlevelof17%inbothsteamgenerators, oronlossofboth4160Vbuses11Aand11B.Allthreemainauxiliary feedwaterpumpsstartonlossofoffsitepower.Inthiscase,eachofthe"A"and"B"motordrivenpumpswouldberunninginparallelwiththeturbinedrivenpumpsupplying watertothe"A"and"B"steamgenerators, respectively.

Sincethesepumps,areprovidedwithautomatic systems,therecirculation valveswouldopenintheunlikelyeventthatthedischarge valvesweretobethrottled back,therebyproviding adequateflowforeachpump.Additionally, anorificeisinstalled ineachminimumflowrecirculation lineupstreamofthecommonlinetie-insothatthestrong/weak pumpinteraction notedinBulletin88-04isde-sensitized.

C-2

~~~eMonthlyperiodictestingofthepumpsperfo'rmed atratedflowdemonstrates thatthedischarge pressurelimitsandbearinghousingvibration arewithintheboundsestablished.

Thisdataistrendedandprovidesameansofdetermining performance changes.The48hourendurance testsonthemotordrivenAFWpumpswereconducted.

onOctober9throughOctober11,1981("A"pump)andonNovember6throughNovember8,1981("B"pump).TestresultswerereportedinReference (a).Thetestwasrunintherecirculation modeof50gpmduringthetest.The48hourendurance testontheturbinedrivenAFWpumpwasperformed onMarch6throughMarch8,1981.TestresultswerereportedbyReference (b)anddemonstrated operability andacceptability ofcriticalparameters duringtherecirculation condition atabout100gpm.Consequently, theconcernsraisedinBulletin88-04havebeenadequately addressed inthedesignandaredemonstrated duringperiodictestingofthesystems.STANDBYAUXILIARY FEEDWATER PUMPThissystemisdepictedonFigure10.5-2oftheUFSAR.Eachofthepumpsisratedfor200gpm.Eachpumpisprovidedwitha1.5-inchrecirculation line,withautomatic flowcontrolled airoperatedvalveand,manufacturer-designed pressurebreakdown orifice.Thevalveisdesignedtofailintheopenposition.

Thevalveisdesignedtoopenwhenthedischarge flowdecreases to80gpm.Eachpumpfeedsaseparatesteamgenerator andparalleloperation asdescribed inBulletin88-04is,therefore, notaconcern.Therecirculation linesmergetoacommonlinewhichleadsbacktothe10,000galloncondensate supplytank.Thestandbyauxiliary feedwater pumpsareprovidedonlyforuseifallthreeofthemainauxiliary feedwater pumpswereunavailable, suchascertainsteamlinebreaksandunmitigated firescenarios.

Thepumpsaremanuallyactuatedand.controlled.

Therearenoknownmodesofoperation wherebythedischarge valveswouldbethrottled backwiththesepumpsrunninginparallel.

However,sincethecommonlinetie-inisdownstream ofeachorifice,thelineisde-sensitized tothepossibleeffectsofthestrong/weak pumpinteraction described inBulletin88-04.Periodictestingofthepumpsisperformed monthlyanddemonstrates thateachpumps'eveloped pressureremainswithinestablished limitsandthatataflowof50gpmtherecirculation valveisfullyopen.Testingalsoconfirmsthatthevalvefailstotheopenpositionwhenairisremoved.C-3 Thebest-efficiency pointofthepumpsis450gpmmakingtherecirculation flowof50gpmabout11%ofBEP.Pumptemperature riseoftheinternalfluidatthiscondition iscalculated tobeonly17.3F.Sincetheonlymodeofoperation whichwillresultinthislowflowisthetestmode,thisvalueisacceptable.

Thepumpsareoperatedinthismodeforbriefperiodsoftimeonly,inordertodemonstrate thattherecirculation valvefullyopens.The48hourendurance testwasconducted onbothpumpsattherecirculation condition inMarchandMay1980.TestresultswerereportedinReference (c)anddemonstrated operability andacceptability ofcriticalparameters.

Consequently, theconcernsraisedinBulletin88-04havebeenadequately addressed inthedesignandaredemonstrated duringperiodictestingofthepumps.C-4 P<~,vt'~

TABLE1DETAILEDTECHNICAL SPECIFICATION CHANGESLocationPage4.5-3Page4.5-3DescritionofChaneAdded450gpmat138psigforRHRpumps;addedsquarebracketsaround,200gpmand.140psigi.nTable4.5-1.Added.Note(1).Added150gpmat1356psigforSIpumps;addedsquarebracketsaround50gpmand1420psiginTable4.5-1.AddedNote(2).ReasonforChaneReducethepotential forpumpaccelerated wear..duetolowflowoperation byincreasing surveillance testflowequivalent to0heminimumflowrecommended forcontinuous operation.

Page4.5-8Elaborated thebasisfortheminimumdischarge pressureandpumpdegradation.

,AddedReference 2.ClarifybasisinkeepingwithUFSARanalyses.

Page4.5-9Nochange.Reformatted onpage.Page4.5-10ChangedReference 2toReference 3andReference 3toReference 4.Renumberreferences.

Page4.5-11UpdatedReferences asfollows:Reference (1)fromFSAR6.2toUFSAR6.3.5.2.rAddedReference (2),UFSARFigs.15.6-12&15.6-13.ChangedReference (2)toReference (3)andupdateditfromFSAR6.3toUFSAR6.5.1.2.4.

Updatethereferences fromtheoriginalFSARtotheappropriate sectionoftheUFSAR.ChangedReference (3)toReference (4)andupdateditfromFSAR14.3.5toUFSAR6.4.3.1.