ML17262A680

LER 91-009-00:on 911111,steam Generator Feedwater Isolations Occurred on Both Steam Generators.Caused by Perturbations of Advanced Digital Feedwater Control Sys.Feedwater Regulating Valves Manually controlled.W/911211 Ltr

ML17262A680
Person / Time
Site:	Ginna
Issue date:	12/11/1991
From:	BACKUS W H, MECREDY R C ROCHESTER GAS & ELECTRIC CORP.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-91-009, LER-91-9, NUDOCS 9112170532
Download: ML17262A680 (22)

Text

e!'CCELERATED DISTRIBUTION DEMONRATION SYSTEM'EGULATORY INFORMATION DISTRIBUTION SYSTEM (RIDS)ACCESSION NBR:9112170532 DOC.DATE: 91/12/11 NOTARIZED:

NO DOCKET FACIL:50-244 Robert Emmet Ginna Nuclear Plant, Unit 1, Rochester G 05000244 AUTH.NAME'AUTHOR AFFILIATION BACKUS,W.H.

Rochester Gas&Electric Corp.MECREDY,R.C.

Rochester Gas&Electric Corp..RECIP.NAME RECIPIENT AFFILIATION

SUBJECT:

LER 91-009-00:on 911111,steam generator feedwater isolations'ccurred on both steam generators.

Caused by perturbations of~advanced digital feedwater control sys.Feedwater regulating valves manually controlled.W/911211 ltr.DISTRIBUTION CODE: IE22T COPIES RECEIVED:LTR g ENCL g SIZE: i(7 TITLE: 50.73/50.9 Licensee Event Report (LER), Incident Rpt, etc.D NOTES:License Exp date in accordance with 10CFR2,2.109(9/19/72).

05000244 A RECIPIENT ID CODE/NAME PD1-3 LA JOHNSON,A INTERNAL: ACNW AEOD/DS P/TPAB.NRR/DET/ECMB 9H NRR/DLPQ/LHFB10 NRR/DOEA/OEAB NRR/DST/SELB 8D NRR~ST/S PLB8 Dl N1 LE 01 EXTERNAL: EG&G BRYCE,J.H NRC PDR NSIC POORE,W.COPIES LTTR ENCL 1 1 1 1 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1'1 3 3 1 1 1 1 RECIPIENT ID CODE/NAME PD1-3 PD AEOD/DOA AEOD/ROAB/DS P NRR/DET/EMEB 7E NRR/DLPQ/LPEB10 NRR/DREP/PRPB11 NRR/DST/SICB8H3 NRR/DST/SRXB 8E RES/DSIR/EIB L ST LOBBY WARD NSIC MURPHY,G.A NUDOCS FULL TXT COPIES LTTR ENCL 1 1 1 1 2 2 1.1 1 1 2 2 1 1 1 1 1 1 1 1 1 1 1 1 D D D NOTE TO ALL"RIDS'ECIPIENTS:

D D PLEASE HELP US TO REDUCE i'i'ASTE!CONTACT THE DOCUlii!EiiT CONTROL DESK, ROOli I Pl-37 (EXT.2M79)TQ LILIih!INAl'E YOUR NAiIF.FROii1 DISTRIBUTION LISTS FOR DOCUiiIENTS YOU DON'T NEED!FULL TEXT CONVERSION REQUIRED TOTAL NUMBER OF COPIES REQUIRED'TTR 31 ENCL 31 a ke 0 ROCHESTER GAS AND ELECTRIC CORPORATION ns~~~'w~~st , ss@ss e e9 EAST AVENUE, ROCHESTER N.K 14649-0001

", ROBERT C MECREDY Vice Psesidens Cfnnn s'ueiess PsodueBun TELEPiSONE AREA CUE 7 s 6 546 2700 December 11, 1991 U.S.Nuclear Regulatory Commission Document Control Desk Washington, DC 20555

Subject:

LER 91-009, Automatic Feedwater Control Perturbations, Due To'lectromagnetic Noise Spikes From Unrelated Relay Actuation, Caused Steam Generator Feedwater Isolation on High Level R.E.Ginna Nuclear Power Plant Docket No.50-244 In accordance with 10 CFR 50.73, Licensee Event Report System, item.(a)(2)(iv), which requires a report of,"any event or condition that resulted in manual or automatic actuation of any Engineered Safety Feature (ESF), including the Reactor Protection System (RPS)", the attached Event Report LER 91-009 is hereby submitted.

This event has in no way affected the public's health and safety.V ery truly yours, Robert C.Mecredy xco U.S.Nuclear Regulatory Commission Region I 475 Allendale Road King of Prussia, PA 19406 Ginna USNRC Senior Resident Inspector 0 I NAC tera 000 ISAS I UCENSEE EVENT REPORT (LER)U*NUOLCAR RICULATQR'v coeal~At<<IIOVID OMS NO,SIM OIOrr lxSIRIS I/Sr/tt f ACILITY NA<as09'" Automatic Feedwater Control Perturbations, Due To Electromagnetic Noise.Spikes From Unrelated Rela Actuation mused Steam Generator Feedwater Isolation on Hi h L vel IVINT OATI ISI LIR NVMICR IM AltORT OATS ITI OTHI1 fACILI'Till INVOLVIO Ql MONTrr OAY YCAR YCAJI~IOVIrlrl*L

")AIVRIQrr NVMIIA r lerrreI A MONTH OAY YCAR f ACr LIT'r HAMIL OOCXIT NVMIIRISI 0 6 0 0 0 9 1 9 1 009 0 0 1211 91 0 6 0 0 0 OtC1 AT INC MOO C III LIYIL 0 9 8~i>rio,.W)Wi$~g'.+THII 1lfORT II SVSMITTCO tURSVANT T S0.<<0SIII 00.<<00 4 I I I I III S0.0004IIIIIII SS.IM 4 I II I IMI SIACI 4 I II I IHI SO.IMle)llllrl SS.IMirl, MMlrl ll I MMNIQI M.TSUIQII4 M.T I le IQI I II M.TSNIQIIX4 M.T SV I Q I lle I M.T S4I QIIrl M.T 04 I Q llrll M.TSNIQIIHXIIAI M.TSNIQ)lralltl M.TSIIIQllel 0 THI 1 IOV I 1 Sir INTS Of 10 Cf 1 f<ICrrrei eee er erere er ar le<<eeeeN III TLTllel 0TH c 1/S<<ee<<V 4 AAreerr~erie rrre 4 Tr<<L NRC fare LICINSII CONTACT tO1 THIS LIR lltl NAMt TILltHONI NVMSIR Wesley H.Backus ARIA COOt Technical Assistant to the rations Mana er 31 5524-4446 COM<<LITC ONI LINC f01 IACH COM<<ONINT f AILURC OtICRIIIO IN THIS RltORT llll CAVSt SYSTSM COMtO HINT MANVf AO TURIR A Prr"""~">""+CAVSI SYSTSM COMtONCNT TO NtROI.5":ij>S.<~,~86: 'jjpQ%.(@~kg MANVf AO TURIN ltORTASL TO NtROS e~rg~@4~@IVtf LtMINTAL Rtt01T IXtlCTIO IIII Y Ct lll fer, reer<<rra IX<</CTIO SVS MISSION OATII NO LIITAAOT II<<err a Irof ireea, I~..rterer<<errNf INaee r<<vrre<<rrr trteerarre

<<ew ll~I CXflCTIO Lvll<<I SCION OATI I'III MOHTH CAY YIAR On November 11, 1991 at approximately 1214 EST, with the reactor at approximately 98%full power, steam generator feedwater isolations occurred on both steam generators.

These feedwater isolations were caused by perturbations of the advanced digital feedwater control system which increased feedwater flow to t1'team generators.

Immediate operator action was to manually control the feedwater regulating valves to reduce steam generator levels and stabilize the plant.The underlying cause of the event was determined to be electro-magnetic noise spikes affecting the advanced digital feedwater control system.Corrective action taken was to modify specific relay circuits that were causing these spikes.NAC faa SM II AS I I l IIAC See M I000 I LICENSEE EVENT REPORT (LER)TEXT CONTINUATION V.I.IIVCLCAA AlOVLATOAV

<<OMMHNISI/AtPAOV KO OMA M ll%0&I04 a>>tiACS ll)IWS I'ACILIT Y IIAMC III OOCKET IIVM~EA l11 Llll IIVMOIII III'NOVINT<Al,'ttV4%8 U A~AOI III R.E.Ginna Nuclear Power Plant TQTT III~~%M A~.~y480OAV NAC Anil~'il I ITI 2 4491-009-00 02or0 PRE-EVENT PLANT CONDITIONS The plant was at approximately 98%steady state reactor power with no major activities in progress.The Maintenance Department was performing troubleshooting, to determine the source of electromagnetic noise, spikes in the Advanced Digital Feedwater Control System (ADFCS).The troubleshoot-ing was being performed under the guidance of Work Order package f9122181.Unexplained electromagnetic

'noise spike problems were identified previously as coinciding with the start of the diesel fire pump, and which had minor effect on the ADFCS control functions.

The ADFCS was installed during the 1991"Annual Refueling and Maintenance Outage.These electromagnetic noise spikes were first noticed on June 4, 1991, when a minor feedwater perturbation occurred, following a diesel fire pump start.Since June 4, spikes have occurred almost every time the diesel'fire pump has started.The ADFCS.has handled spikes with no noticeable feedwater perturbations, except for two (2)occasions.

These occasions, the first on June 4, 1991 and the second on September 13, 1991, were handled by the ADFCS in automatic and no operator action was required.There has been an ongoing search for the possible source of this electromagnetic noise spike so that it could be corrected.

As part of this ongoing search, the Electrical Engineering Department evaluated their cable tray database and identified circuit E174 as a possible source.Circuit E174 is the 125 Volt DC power feed to the fire relay panel and shares some cable trays with ADFCS input cables, most notably, the feedwater header pressure inputs to ADFCS'P501 and P502).'IAC AOAM AAA ital

NIIC Pe<<<<1~044 I I.'ICENSEE EVENT REPORT ILER)TEXT CONTINUATION

<<U.S.NUCLIAA AIOULAT01T COMMITeION

/AAAAOV%0 OMI NO TISOWIOA T)eAt1%$~III 1$I'ACILITY NAMI III R.E.Ginna Nuclear Power Plant TQ(T IA'<<<<44 NMCO 4~.<<<<M MM41<<<<M NOMIC/Oral~T I I I TI DOC%IT NUNMtN ITI osooo244 Llll NUMOl1 III v TAA,'AQVANT<<AL 1(V4<<O<<<<4 1<<<<4 1 91-0 09 0 0~AOI IS 0 30'9 Westinghouse Electric Corporation (the manufacturer of the ADFCS)was contacted and could not explain the ADFCS excursions based on, data available.

In conjunction with the Electrical Engineering Department, Westinghouse had previously recommended that the shielding and grounding schemes for all ADFCS inputs be checked.These inputs were checked in August, 1991.This check indicated that all ADFCS inputs are correctly shielded and grounded.DESCRIPTION OF EVENT A.DATES AND APPROXIMATE TIMES OF MAZOR OCCURRENCES:

0 0 November 11, 1991, 1214 EST: Event Date and Approximate Time.November 11, 1991, 1214 EST: Discovery Date and Approximate Time.o November 15, 1991: Cause of EMP noise spike identified and suppressed to acceptable levels.WGBFZ: On November 11, 1991, at approximately 1214 EST, with the reactor at approximately 98%full power, the diesel fire pump was started, as required for ADFCS troubleshooting per Work.Order f9122181.Approximately thirty (30)seconds after the diesel fire pump was started an"ADFCS System Trouble" alarm (G-22)was received.The Control Room operator responsible for feedwater control had pre-positioned himself in front of the"A" and"B" S/G Main Feedwater Regulating Valves (FRV)control panel prior to the start of the diesel fire pump.~<<1C ADAM TQAA<<tel

1IIC Seam~I$4$l LICENSEE EVENT REPORT ILER)TEXT CONTINUATION V.S.IIVCLCA1 AIOVLATOAY COSSSSt%IOII r ASSAOYCO OU~IIO SI$0&IOS TIIAI1$$~ITI%$AACILITY NASIS I 1 I-OOCKST IIVIAOIA Ill L$1 NLAN$1 I~I SSOUS>S<AL ASYCK)U U 1'U A~A4$ISI R.E.Ginna Nuclear Power Plant TQ(T/IT mew AIASS 1~.~e4004HV JYAC AMID~Yl I Ill o s o o o2 44 9 1-0 09-0004 or0 9 At this time, the Control Room operator noticed that both the"A" and"B" Steam Generator (S/G)main feedwater flows were pegged high.with both"A" and"B" S/G Main Feedwater Regulating Valves continuing to open further.The condensate low pressure heater.bypass valve opened automatically and the standby condensate pump started automa'tically (to increase main feedwater pump suction pressure).

Main Feedwater pump suction pressure was decreasing due to the increased feedwater flow to the S/Gs.The"A" and"B" S/G levels continued to increase and before the Control Room operator could shift'the FRVs to manual, ADFCS automatically shifted the FRVs to manual.While the Control Room operator was manually lowering the setpoints for the FRV controllers, to control S/G level, the following alarms annunciated and feedwater isolation o'ccurred on both S/Gs;G-4 (S/G A HI LEVEL CHANNEL ALERT 67%)and G-6 (S/G B HI LEVEL CHANNEL, ALERT 67%).Immediately following the feedwater isolation, the condensate booster pumps tripped on high*pressure.A load de'crease was initiated at 10%/hour to lessen the impact of unstable S/G levels.Main feedwater to the S/Gs was controlled in manual in order to stop secondary system oscillations that were occurring due to the event.During the S/G level stabile.zation, S/G feedwater isolation occurred several times.The S/G levels were subsequently stabilized and main feedwater control was returned to automatic.

After main feedwater control was returned to automatic the load decrease was terminated.

Total load decrease was approximately 0.54 full power during the event.Subsequently, the condensate low pressure heater bypass valve was closed, the condensate booster pumps were restored, and the standby condensate pump was secured and realigned for automatic standby.~~VAC SOAV SSAA i$4$l

II1C I<e<ws~l043 I LICENSEE EVENT REPORT ILER1 TEXT CONTINUATION V S.1VCLSA1 1SOVLATO1Y COMMISSION ATT1DV CO OVS<<O SISO&10l S>et<1%$~IS I SS 1ACILITT IIAMt III DOC%ST IIVMOS1 lll LSII IIVMOI1 ISI SSOVT<<T<AL r r~AOI ISI R.E.Ginna Nuclear Power Plant TuT nr<<r<r r<Aee r<rr<rrr,~~A<rc I<<r<<<~Lv I I TI o 5 o o o 2 4491-0 Q9-0005 o>0 9 C INOPERABLE STRUCTURES, COMPONENTS', OR SYSTEMS THAT CONTRIBUTED TO THE EVENT: None.'.OTHER SYSTEMS OR SECONDARY FUNCTIONS AFFECTED: E.None.METHOD OF DISCOVERY:

The event was immediately apparent due to alarms and indications in the Control Room.OPERATOR.ACTION: The Control Room operators took immediate manual actions to control S/G levels, reduce power level, and stabilize the plant.Subsequently, the Control Room operators notified higher supervision and the Nuclear Regulatory Commission per 10 CFR 50.72, non-emergency, 4 hour4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> notification.

G.SAFETY SYSTEM RESPONSES:

The"A" and"B" FRVs closed automatically from the feedwater isolation signal.III.CAUSE OF>wryly A.IMMEDIATE CAUSE: The feedwater isolation of the"A" and"B" S/G was due to the"A" and"B" S/G narrow range levels being)/=67%.~<AC 101M)00A<040<

IIAC Pere M l044l LICENSEE EVENT REPORT (LER)TEXT CONTINUATION V.5.IIVCLTAW 150VLATOAT CISPAII55IISI APPAOv50 Ov5 rrO IITC~IOA P TlrprA55 5ITI,55 PACILITT rIAAI5 III OOCIIKT IIVAIOCA Ill L5A IILAPCI A I5!TTarrarrTrAL e eg V Ie IO rr e PAO5 I5I R.E.Ginna Nuclear Power Plant TSCT np rrrere~r reereep.~oeeoaw rTAC rrerrrr~VV I I TI o s o o o 2 4 4 1-0 0 9-0 0 06 av0 9 INTERMEDIATE CAUSES: The"A" and"B".S/G narrow range levels were>/=67%due to increased feedwater flow to both S/Gs caused by a perturbation of the ADFCS.The perturbation of th'e ADFCS was apparently due to electromagnetic noise spikes affecting the feedwater header pressure inputs to ADFCS, (i.e.P501 and P502).C.ROOT CAUSE:-After extensive troubleshooting, it was determined that the spikes that affected the ADFCS feedwater header pressure inputs were caused by the de-energiza-tion of Relay ARSO, located in the fire relay panel.This relay, which lights the diesel fire pump trouble light, de-energizes approximately 10 to 15 seconds after a diesel fire pump start.During this de-energization, inductive"kickback" causes an electro-magnetic noise spike to be generated and induced into the feedwater header pressure inputs.The signal cables carrying the feedwater header pressure transmitter (PT-.501 and PT-502)inputs share some common cable trays with the'DC power source for the ARSO relay, and a noise spike was induced from the ARSO relay cable to the feedwater header pressure input cables.ANALYSIS OF&TENT This event.is reportable in accordance with 10 CFR 50.73, Licensee Event Report System, item (a)(2)(iv), which requires reporting of,"any event or condition that resulted in manual cr automatic actuation of any Engineered Safety Feature (ESF)including the Reactor Protection System (RPS)".The feedwater isolation of the"A" and"B" S/Gs was an automatic actuation of an ESF system.~r A C~O 1 re r TATI

'I MAC Pere M 19441 LICENSEE EVENT REPORT (LERI TEXT CONTINUATION V.l.IIVCLCAA 1lQULATOAY COMMI5OQM P APPAOYlO OMl rrQ)IlQ~IQA l>ceIAlS lrll 1$I'ACILITY NAlll III QOCKlT HU%%$1 lll LIII MVeell1 Ill j~Ovlrrer AL rrl v le Io rr A~AOl lll R.E.Ginna Nuclear Power Plant TlxT IA'~r~.~ereereMI eIAC Perrrr~'ll I I TI o s o o o 24 49 1-0-9-0 07 os0 9 An assessment was performed considering both the safety consequences and implications of this event with the following results and conclusions:

There were no operational or safety consequences or implications attributed to the feedwater isolations because: o The feedwater isolations occurred at the required S/G levels.o The plant was quickly stabilized and manual control of the FRVs was accomplished to mitigate the transient.

o As the feedwater'solation occurred as designed, the assumptions of the FSAR for steam line break were met.Based on the above,-it can be concluded that the public's health and safety were assured at all times.V.CORRECTIVE ACTION A.ACTION TAKEN TO RETURN AFFECTED SYSTEMS TO PRE-EVIBFZ NORMAL STATUS: 0 The Diesel Fire Pump was temporarily removed from service pending the outcome of root cause troubleshooting and determination.(The pump was returned to service after a noise suppression diode was installed across relay AR80).0 When S/G levels were stabilized, subsequent to the ADFCS perturbation termination, the FRVs were placed in automatic control.0 After the plant had been stabilized and the FRVs returned to automatic control, the condensate low pressure heater bypass valve was closed, the condensate booster pumps were restored and the standby condensate pump was secured and realigned for automatic standby.rrAC POAM el4ll

NAC 1vv~(~I LICENSEE EVENT REPORT ILER)TEXT CONTINUATION-U,S, NUCLTAA AIOULATOAV COMAIITSION A>>AOVlO OUI NO)1$0&I04 T Ict<A CS~Il I'l$FACILITY NAIAC III OOCAtT NU~tA Ill LtA NLAA4IA III T~OVINTIAL v 1 AA V IS lO N A~AOI IlI R.E.Ginna Nuclear Power Plant Terr IA~>>A>>A~.~>>>>VV>>V NAC AV>>~SU I I TI o 5 o o'o 24 491-00 9-00 08 DFO 9 B.ACTION TAKEN OR PLANNED TO PREVENT RECURE&NCE:

o*A reverse-biased diode was temporarily installed across the coil of AR80 on November 15, 1991 and subsequent testing determined that the spikes from the AR80 circuit, affecting ADFCS feedwater pressure inputs following diesel fire pump , starts, were eliminated.

This noise suppression diode'was permanently installed on November 18, 1991.After reviewing the results of troubleshooting and the discussion with Westinghouse, the following is an outline of the corrective actions being taken or planned in response to the ADFCS noise spiking events: o Short Term Response a)Operations personnel were made aware that one source of spikes on ADFCS was eliminated, but that spikes from other sources, while reduced in frequency.

and magnitude, might occur.Operations will identify any new spikes on the ADFCS by submitting a Work Request/Trouble Report (WR/TR).b)A WR/TR was submitted for installation of a diode for the fire booster pump relay AR85 (which also produces small spikes on ADFCS).However, these spikes are not of the same magnitude as the noise spikes that were caused by the Diesel Fire Pump starts.o Intermediate Term Response Electrical Engineering will consult with Westinghouse concerning a database change to increase the ADFCS slew rate filter constant.This filter is used to dampen any abrupt changes to feedwater regulating valve demand in the event that feedwater header pressure input values are rejected due to noise spikes.It is thought NAC IOAV TAAA~T4ll

NIIC form SOSA (04S I SACILITY NASIE Ill LICENSEE EVENT REPORT ILERI TEXT CONTINUATION COCKET NUIAOEII (El U.E, NUCLEA(l AECULATOAY COSSSIIS3(ON ASSAOYEO OUI NO TI SOW(04 f Exs(RES'S(SI'ES LE(i NUIAOEA I~I$SCUSNTrAL osvrsrorr N O Nvm O~ACE (j(R.E.Ginna Nuclear Power Plant TExT (rf moro A(sos rs rosrrsos, om sNsooonor (Y(lc form~'ll (ill o s o o o 24 491-009-00 09 OF 0 9 that this filter can also"Lock In" an erroneous value following a feedwater headei pressure spike.Increasing this constant will more quickly restore a correct value for feedwater header pressure, after a spike has decayed.The spikes last less than five (5)seconds.o Long Term Response a)b)Electrical Engineering will check with Westinghouse for the results of their review of ADFCS arbitration error checking software.This review will determine if the error checking routine for the switching to arbitration values instead of feedwater header pressure field input values is substituting erroneous values for feedwater header pressure used in FRV demand calculations.

Electrical Engineering will evaluate the routing of feedwater header pressure input circuits (to the-ADFCS), and will identify any additional modifications that may be required to eliminate the electromagnetic noise spike concern.VX.ADDITXONAL NFORMATION A.FAILED COMPONENTS:

None B.PREVIOUS LERs ON SIMILAR EVENTS: A similar LER event historical search was conducted with the following results:.No documentation of imilar LER events with the same root cause could be.identified.

C.SPECIAL CO~BFZS: None.NAC s01lo SSOA rS4TI