ML17262A741

LER 91-010-00:on 911230,during Operator Review of Plant Process Computer Logs,Util Discovered That Invalid Data Had Been Used for Calculation.Caused by Invalid Input from Lefm.Lefm HP-85B Computer Was reinitialized.W/920129 Ltr

ML17262A741
Person / Time
Site:	Ginna
Issue date:	01/29/1992
From:	Backus W, Mecredy R ROCHESTER GAS & ELECTRIC CORP.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-91-010, LER-91-10, NUDOCS 9202060130
Download: ML17262A741 (20)

Text

ACCELERATED D TIUBUTION DEMONS TION SYSTEM

~"

k REGULATORY INFORMATION DISTRIBUTION SYSTEM (RIDS)'CCESSION NBR:9202060130 DOC.DATE: 92/01/29 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 6 Electric Corp.

MECREDY,R.C. Rochester Gas 6 Electric Corp.

RECIP.NAME RECIPIENT AFFILIATION R

SUBJECT:

LER 91-010-00:on 911230,during operator review of plant, process computer logs, util discovered that invalid data D had been used for calculation. Caused by invalid input from LEFM.LEFM HP-85B Computer was reinitialized.W/920129 ltr. S DISTRIBUTION CODE: IE22T COPIES RECEIVED:LTR ENCL SIZE:

.TITLE: 50.73/50.9 Licensee"Event Report (LER), Incident Rpt, etc.

NOTES:License Exp date in accordance with loCFR2,2.109(9/19/72). 05000244 A D

RECIPIENT COPIES RECIPIENT COPIES ID CODE/NAME LTTR ENCL ID CODE/NAME LTTR ENCL D PD1-3 LA 1 1. PD1-3 PD 1 1 JOHNSON,A 1 1 INTERNAL: ACNW 2 2 AEOD/DOA 1 1 AEOD/DSP/TPAB NRR/DET/ECMB 9H 1=;1 1 1 AEOD/ROAB/DS P NRR/DET/EMEB 7E 2

1 2

1 NRR/DLPQ/LHFB10 1 1 NRR/DLPQ/LPEB10 1 1 NRR/DOEA/OEAB 1 1 NRR/DREP/PRPB11 2 2 NRR/DST/SELB SD 1 1 NRR/DST/SICBSH3 1 1 NRR/D~SJS LB.S Dl 1 1 NRR/DST/SRXB SE 1 1 RE -'CE 2 1 1 RES/DS IR/EI B 1 1 RGN IL 01 1 1 EXTERNAL EGGG BRYCE g J ~ H 3 3 L ST LOBBY WARD 1 1 NRC PDR 1" 1 NSIC MURPHY,G.A 1 1 NSIC POORE,W. 1 1, NUDOCS FULL TXT 1 1 Q~~o D pgys7>//~ 7 D

D NOTE TO ALL "RIDS" RECIPIENTS:

S PLEASE HELP US TO REDUCE iVASTE! CONTACT THE DOCUMENT CONTROL DESK, ROOM P 1-37 (EXT. 20079) TO ELIMINATEYOUR NAME FROM DISTRIBUTION LISTS FOR DOCUMENTS YOU DON'T NEED!

FULL TEXT CONVERSION'REQUIRED TOTAL NUMBER OF COPIES REQUIRED: LTTR 31 ENCL 31

0 I ~ II~ I ~ 44 Jnc r TOOK 11KTC C'OCHESTER GAS AND ELECTRIC CORPORATION ~ 89 EAST AVENUE, ROCHESTER N.K 14649-0001 ROBERT C. MECREDY TELEPttONE Vice President AAEA CODE 7 1 6 546 2700 Cinna Nuclear Production January 29, 1992 U.S. Nuclear Regulatory Commission Document Control Desk Washington, DC 20555

'n

Subject:

LER 91-010, During Operator Review of Plant Process Computer Logs, It Was Discovered That Invalid Data Had Been Used For Calorimetric Calculation R.E. Ginna Nuclear Power Plant Docket No. 50-244 This event is being voluntarily reported using the guidance of NUREG-1022 (Licensee Event Report System), and Supplement No.

1 to NUREG-1022. This report is intended to alert other utilities which may have similar systems. This event is related to, but does not meet, the reporting requirements of 10CFR50.73, item (a)(2) (i) (B) . Item (a) (2)(i) (B) requires reporting of "any operation or condition prohibited by the plant's Technical Specifications". The attached Licensee Event Report LER 91-010 is hereby voluntarily submitted.

This event has in no way affected the public's health and safety.

Very truly yours, Robert C. Mecredy xc4 U.S. Nuclear Regulatory Commission Region I 475 Allendale Road King of Prussia, PA 19406 Ginna USNRC Senior Resident Inspector

~ P i gp g

~ ~

Lr i3 0 V 3 c.)

9202060130 920129 PDR ADOCK 05000244 8 PDR

0

>>AC y>><<SOO UW NUCLSA1 1IOULATOA1COO<<)~ON IO4$ )

AtyNOTIO ONS NO. $ 1IO 0IOV LICENSEE EVENT REPORT (LERl IXSYASI I/Svss yACILITYNAA)a Ill OOCXST NV<<tall Ql A R.E. Ginna Nuclear Power Plant o 6 o o 0 244 >0)0 8

<<I I)4 During Opera or Review of Plant Process Computer Logs, Invalid Data Had Be n U '

It Was Discovered That imetr TINT OATS ISI LSN NU<<SIN III AIAOAT oATI 1)I OTHI1 IACILITIISINYOLVIONl NO>>TH OAY TIAII YIA1 ) I0 V I >> 1 I A L "'VNJOV

-.I'WaasIA OAY TtA1 yAO4TT HAH)A OOCXIT NVNSIAISI HV>>IIA 0 6 0 0 0 12 30 OAIAATINO

<<OOI NI 9 1 1 010 0001 AlyOAT N SUSINTTIO yVASVANT T 0 THt AIOVIAININTSOy ISCSA 992 f: ICA>>e>>V W>>A>>ey <<A

~ O.T $ 4) Q I IN)

~ lll 0 6 0 0 0

'HIS )O AO) ~ I

~

I $ 0.AOOIN

~ ONI 1 SO.NN41 IIII I Stoa4II)l ~ O.T $ 41ll I lyl TL1)4)

LavtL 0 9 SOANNIllllll ~ OMNIQI SO.T $ 4) Ql NII X 'oTHa1 ISA>>>>y 4 A>>A>> AAO IA rAAL>>1C sa SO.AOS 4 I III I <<) ~ O.T$ 4lallt IO.T $ 4) 0 II>>XIIAI ESSA)

$ 0AOS41)) lr>>l ~ O.T$ 4)Q)lal ~ O.T$ 4)QII<<IINI

" a+ g)~ig@i. $ OAN 4) ll)ly) SO I $ 4)QI I<<I ~ O.T $ 4) Q I la I LICINIIICONTACT SON THN Lail 0$ )

NAUI TSLtyHONI NVHSIII ANIA COOS Wesley H. Backus Technical Assistant to the rations Mana er 315 CO<<yLITI ONI LINI IOA IACH CO<<AONSNT SAI LV1l OIICAI~ IO IN THN 1tyOAT 11$ )

wjv{IN+Jv wg CAVSI STSTIN COUAONINT UANUI'AC IAON TAS Ll Rh'I'.Pri ggi9%j) '..

Ig<qg&I> CAUII STITSLI CO<<yONINT NANUSAC lyONTAS L TVNI1 TO Ny10$ gpss TUNI1 TO NyAOS @4i-..)(;Q>~b)<

X I0 W12'0 N yg~t)4 p ge

~%A"'o'~<V~~

~ UyyLII4INTALAITOAT IXylCTIO 114 VONTH CAY TIA1 tXAICTIO LVSLVSSION OATS II II T tl or y>>, y>>AA>>>> sxyscrso svsa>>ss)o>> oArsl NO ASSTAACT ILAA>>>> WO>>>>>>, I 1, AA>>su>>vary Nf>>AA <<V>>~r ITA>>>>N>>A AA>>l lit)

On December 30, 1991 at approximately 1830 EST, with the reactor at approximately 984 full power, the Control Room Foreman, while performing a review of the hourly Plant Process Computer System (PPCS) logs, determined that the Leading Edge Flow Meter (LEFM}

data used by the PPCS to calculate calorimetrics, was invalid from 1800 EST, December 20, 1991 to approximately 1830 EST, December .30, 1991. This event is being voluntarily reported to alert other utilities which may have similar systems.

The Control Room operators notified the reactor engineer and reinitialized the LEFM computer. After reinitialization, the LEFM printout values and PPCS values returned to expected/normal indications.

Corrective actions include revi'sions to the affected procedures to require checking the operation of the LEFM prior to performing calorimetrics and the addition of a high'alarm on PPCS indicating a failure may have occurred.

The immediate cause was identified as a failure of the LEFM HP-85B computer.

>>1C I>>>> SOS ISO$ 1

IIAC lees WA U $ , ISUOLSAA ASOULAToAv coeeMISCIbll I$4$ I LICENSEE EVENT REPORT ILER) TEXT CONTINUATION S ASSAOV $ 0 OM$ IlO $ 1$ 0%104

$ SSIA$ 5 ~ I)I 1$

SACILITY IIAMC III OOCKlT IIUI&SA l1I LlA IIUSSOSII I~ I ~ AO$ 1$ I S I OI V SMIIT I1A L V IS IO M e

~ M TSIT Ils R.E. Nuclear

~ ~ s Ieeweel vee ~Power, Plant lVAC Sese ~ $ I I Ill o s o o o 24 49.1 01 0 00 020FO 8 PRE-EVENT PLANT CONDITIONS The plant was at approximately 98% steady state reactor power with no major activities in progress. The Control Room Foreman (CRF) was reviewing the hourly Plant Process Computer System .(PPCS) logs.

DESCRIPTION OF EVENT -

A. DATES AND APPROXIMATE TIMES OF MAZOR OCCURRENCES:

o December 20, 1991, 1800 EST: Event date and approximate time.

o December 30, 1991, 1830 EST: Discovery date and approximate time.

o December 30, 1991, 2055 EST: Performed calori-metric using Barton D/P cells option and plant thermal power was determined to be approximately the same as that calculated using the Leading Edge Flow'eter (LEFM) option.

B. EVENT:

On December 30, 1991 at approximately 1830 EST, with the reactor at approximately 98% full power, the CRF, while performing a review of the hourly PPCS logs, determined that the LEFM data used by the PPCS to calculate calorimetrics, was invalid from 1800 EST, December 20, 1991 to approximately 1830'EST, December 30, 1991. This determination was based on the following information:

o The LEFM Feedwater (FW) Flow confidence level was 100.0% for most of the time period in question (Zt should be less than 100% and some variation is expected).

~ IAC sOAes $ $ SA I$ 4$ l

IIAC $ ere I$4$ I M U.$ . IIUCLEAA AEOULATOAT COAIAIIION

/

LICENSEE EVENT REPORT ILER) TEXT CONTINUATION AftAOVEO 0+5 hO $ I$0&IOA EXP>AE5 ~ I$ I $ $

AACILITYIIAAIE Ill COCKET IIUAWEA IEI LEA IIIAACI1 I~ I AAOE 1$ I SAOVANTIAL VOW)<

~I U U R.E. Ginna Nuclear Power Plant 24 4 9.1 010 00 030FO TEXT ill' ~ ~. ~

A OtWarW NIIC /ebs ~ $ I I I Tl o s o o o o The LEFM Feed Water Flow (FFWH) was a constant 6399 KLB/Hr for most of the time period in question (some variation is expected).

o Local observation of the remotely located LEFM HP-85B computer indicated the last printout was on December 20, 1991.

o During the time period in question, the PPCS still indicated valid values for all data points used for the calorimetric.

The Control Room operators notified the reactor engineer and reinitialized the LEFM HP-85Bcomputer.

After reinitialization, the LEFM HP-85B computer printout values and PPCS values returned to ex-pected/normal indications. Subsequently, calorimetrics were performed using both the Barton D/P Cells and the LEFM for Feedwater Flow Data input. The observed difference in calculated thermal power, between these two calorimetrics was approximately 0.154.

n 1

C. ICONTRIBUTED NOPERABLE STRUCTURES g COMPONENTS 5 OR SYSTEMS THAT TO THE EVENT:

A constant feedwater flow output from the LEFM was the prime contributor to the event.

D. OTIC SYSTEMS OR SECONDARY FUNCTIONS AFFECTED None.

E. METHOD OF DISCOVERY:

The event was discovered during the review of hourly PPCS logs by the CRF.

<AC AOAAI $ $ $ A

<$ 4$ l

IIAC /earn

($451

~ LICENSEE EVENT REPORT (LER) TEXT CONTINUATION V.t. WJCltAA AEOVEATOAT COAItllttIOII

/

AtPAOV EO OVt NO 51$ 0&IOA

, t)tP<A 5 5 tltl 5$

EACIEITT IIAME Ill OOCKtT IIV~tIIItl I.IA IllANEAIII ~ Aot 151 TTAA $ 5QMteaTiAI, pEvISIO<

M A M TEXT llT asar ~ e eWnC ~ ~

R.E. Nuclear Power Plant JVJIC AamS ~ $ 11111 o s o o o 244 91 010 00 040'8 OPERATOR ACTION:

Immediate operator action was to notify the reactor engineer, reinitialize the LEFM HP-85B .computer and perform additional calorimetrics using the Barton D/P Cells to confirm that reactor thermal power was within limits.

G. SAFETY SYSTEM RESPONSES:

None.

III.s CAUSE OP ~9FP A. IMlKDIATE CAUSE:

The source of invalid data to the PPCS calorimetric calculation was invalid input from the LEFM.

INTERMEDIATE CAUSES:

When the LEFM HP-85B computer failed updating the Digital Analog (D/A) output card.

it stopped Therefore, the analog output to PPCS remained un-changed.

C ROOT CAUSE:

The invalid LEFM output was due to the unanticipated failure mode of the HP-85B computer.

ANALYSIS OR EVENT This event is being voluntarily reported using the guidance of NUREG-1022 (Licensee Event Report System), and Supplement

'No. 1 to NUREG-1022. This report is intended to alert other utilities which may have similar systems. This event is related to, but does not meet, the reporting requirements of 10CFR50.73, item (a)(2)(i)(B) which requires reporting of "any operation or condition prohibited by the plant's Technical Specifications".

VAC AOAII tttA

>$ 451

MAC ivv Wk U 5, MUCLTAA A54ULATOAYCOWMI55IOOI l045 I LICENSEE EVENT REPORT ILER) TEXT CONTINUATION AAAAOYCO OM5 W

/ 5I54&I04 5)tt<II55 ~ I)I '55 I'ACILITYMAMC III OOCIICT MUM45A ITI LLII MUMOl1 III ~ A45 15I S0 4 U iM T < A I. Y Ig lO M

~ I U M R.E. Ginna Nuclear Power Plant 2 449. 1 0 005 Qf0 8 TaXT W VVY ~ ~.e YV AAAMVMIeIIC ~~ LU I I TI

.o s o o o The plant Technical Specification related to this event is specification 4.1.1 which states, "calibration, testing, and checking of analog channel and testing of logic channel shall be performed as specified in Table 4.1-1".

Table 4.1-1 describes that the nuclear power ranges shall be calibrated daily (i.e. at least once per 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />) using a Heat Balance Calculation (i.e. calorimetric).

'he Plant Operating Review Committee determined that the intent to perform the required Technical Specification daily calibration of the Nuclear Power Ranges, using the Daily Heat Balance calculation, was met, but the Heat Balance Calculations were invalid due to the invalid LEFM input data.

In summary, it'as determined that while the daily heat balance calculations were not valid, 'due to the steady state power maintained during the period, the Nuclear Power Ranges continued to be within design tolerances and were capable of performing their intended functions. An assessment was performed considering both the safety consequences and implications of this event and it was concluded that there were no operational or safety consequences or implications attributed to the invalid LEFM input to the Heat Balance Calculation because:

During the period in question, the highest recorded total feedwater flow (from the Main Control Board Feedflow Transmitters as recorded on the Heat Balance Calculation Form) was 6497.29 KLB/Hr. Using this feedflow resulted in a calculated reactor thermal power of approximately 98.74. Previous to the LEFM HP-85B 'computer failure, reactor power as indicated from plant calorimetrics had been limited to a maximum of approximately 984 power due to degradation in secondary side performance. Therefore, reactor thermal power was below 100% for all calculated heat balances.

~ eAC AOAM 544A i54$ I

IIAC Setto ITS I

~ LICENSEE EVENT REPORT ILERI TEXT CONTINUATION V S. MVCLSAII AIOVL*SOAYCOSSMINIOM ASSAOYKO OMS NO IISOW104 SISS(AE5 SINAI'SS fACII.ICY ISASSI III OOCIIlY IIVl&lAQl LIA MLAQIA I ~ I ~ AOI ISI vSAA SSOVSMTtAL ASYOIOH M A U A R.E. Nuclear Power Plant

'fQP llY ~ ~ ~. ~ 4 r400onV IIIIC Aeltl ~'Sl I III o s o o o 2 4 4 91 010 00 060'8 o The lowest calculated reactor thermal power using the invalid LEFM input was 97.274. The highest that reactor thermal power was estimated to be was 98%.

Based on these indications, the Nuclear Power Ranges could have been calibrated as much as .74 low. The accuracy of the reactor thermal power 'calculation is determined to be + 2.0%. Therefore, the Nuclear Power Ranges were calibrated to within the accuracy of the reactor thermal power calculation and the Nuclear Power Range" trips would have actuated within their design tolerance.

Based on the above, the conclusion was that the public's health and safety was assured at all times.

An evaluation was performed to determine appropriate means to detect an invalid LEFM input to the Reactor Thermal Power Calculation. The following conclusions were reached:

0 Reactor power was steady state during the time in question (i.e. approximately 10 days) and all control board indications remained approximately the same.

0 When the LEFM HP-85B computer failed, the output to PPCS failed "as is", 'as opposed to a failure that would register in an expected obvious extreme mode.

0 This .failure mode. was not anticipated by the LEFM

vendor. As a result, verification functions and associated alarms for this failure mode. were not included in the system design.

0 In addition to the LEFM vendor, plant staff did not anticipate this failure mode. As a 'result, the awareness level for a problem of this type was low and the formal processes to detect and respond to this condition were not in place. Although specific procedure guidance was provided to verify LEFM accuracy, verifying operability of the LEFM HP-85B computer prior to performing the Reactor Thermal Power Calculation was not required.

MAC SOAM lt4ll

~

IIAC SVfff (545 I

~ uCENSEE BIENT REPORT ILERI TEXT CONTINUATION U.l, IIUCL5*A Al4ULATOAYCOMMISSIOII ASSAOY5O Ovl IVO 5150&104 5 >et1 A 5 5 ~ I5 I 5 5 SACILITY NAMS Ill OOCKST IIUS%511 ITI Llll IIUM454 Ill JA45 I51 S~ CIUSffTf AL V

IflV 4 lO AV V A V V R.E. Ginna Nuclear Power Plant 0 5 0 0 0 2 4 4 91 010 00 0 7o~ 08 TTI(T IIT fffVS MASS ~ fSSffSSA ffv SSSffVfV JVAC Afffff~$ 11171 CORRECTIVE ACTION A ACTION TAKEN TO RETURN AFFECTED SYSTEMS TO PRE-EVENT NORMAL STATUS:

o The LEFM HP-85B Computer was reinitialized and the LEFM printout and PPCS values returned to expected normal indications.

o All subsequent calorimetrics were performed using the Barton D/P Cells and the Nuclear Power Ranges were adjusted as required.

o A note was added to the Plan Of The Day (POD) to instruct the Control Room Operators,to use the Barton D/P cells data for calorimetrics 'until further notice.

B ACTION TAKEN OR PLANNED TO PREVENT RECUEGUBTCE:

The following corrective actions have been taken to assure valid data from the LEFM prior to its use in a PPCS reactor thermal power calculation (i.e. calori-metric) .

o An indication of failure of, the LEFM input to the PPCS is- that the LEFM Confidence Level (FFWCL) will approach 1004 indicating the LEFM output is not changing. A high alarm on PPCS for FFWCL has been added to indicate that this failure has occurred. If this alarm activates, the LEFM will not be used.

Operating Procedures 0-6.3 (Maximum Unit Power) and 0-6.3.2 (Maximum Unit Power Calculation Using The LEFM For Flow Measurement) have been revised to include the following requirements prior to using the LEFM for a calorimetric:

1. Verifying that the LEFM HP-85B computer is updating and has been operating for at least 10 minutes. This ensures the current value for the LEFM in the PPCS is not failed.

VAC SCAM SSSA 1545l

NAC

($ CS I Tee M U,S. NUCLSAN NSOULATOAV COMMISSION uCENSEE EVENT REPORT (LER) TEXT CONTINUATION AJfAOVKO OUI VO )ISOAIOI SXrlASS IISI %$

FACILITY NANS Ill OOCKST NULICCN ITI L$ 1 NlA&$1 I~ I AAOC ill StauevTsAL V 1 lO N v 1 R.E. Ginna Nuclear Power Plant:

TtxT IN ~ Nrrr ~ NrvrNI. wr rrrrew Nrc INrr ~'el I In o s o o o 2 4 4 91 01 0 00 08<<0 8

2. Verifying that the LEFM confidence level is not in alarm (FFWCL > 99;504 and FFWCL 99.904). This ensures that the LEFM is operating properly.

0 An identification and evaluation process focusing on other plant systems that may be susceptible to this type of failure mode is in progress.

ADDITIONAL INFORMATION A FAILED COMPONENTS:

The LEFM system was supplied by Westinghouse Electric Corporation, Model gLEFM-824, Serial gU001. The specific component that failed was the HP-85B computer with I/O ROM, Serial g2619A53804.

B PREVIOUS LERs ON SIMILAR EVENTS:

A similar LER event historical search was conducted with the following results: No documentation of similar LER events with an undetermined root cause could be identified.

C. SPECIAL COMMBFIS:

Due to the unanticipated nature of the computer failure, the industry will be notified via Nuclear Network.

vrC 101il SCCA

<$ 44 l