ML20029B643

LER 91-007-00:on 910209,pressure Switch Instrument Drift Caused Automatic Start of Annulus Exhaust Gas Treatment Sys. Caused by Component Failure.Switch Recalibrated to Adjust Setpoints to Proper value.W/910308 Ltr

ML20029B643
Person / Time
Site:	Perry
Issue date:	03/08/1991
From:	Hegrat H, Lyster M CENTERIOR ENERGY
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-91-007, LER-91-7, PY-CEI-NRR-1331, NUDOCS 9103130254
Download: ML20029B643 (4)

Text

. . - - .. . ~. . - . - .

g -m_

PERRY NUCLEAR POWER PLANT Mall Address:

PO. BOX w . Michael D. Lyster 110 CENTER ROAD PERRY, OHIO 44081 )

. PERRY, OHIO 44001 VICE PRESIDERT - NUCLEAR -

(216) 259 3737 March 8, 1991..

PY-CEI/NRR-1331 L U.S. Nuclear Regulatory Commission Document. Control-Desk Washington, D.C. 20555 Perry Nuclear Power Plant Docket No. 50-440 LER ,9,1-007 Dear Sir Enclosed is Licensee Event Report 91-007 for the Perry Nuclear Power Platt.

Sincerely, it'. .

Michael D. Lyster MDL:CRE:njc j Enclosure LER 91-007 cc: NRC Project Manager NRC Sr. Resident Inspector U.S. Nuclear Regulatory Commission 799 Roosevelt. Road -

Glen Ellyn, Illinois 60137 ocerat to cewes C!med Peche thummarr g 130027 b ecofdson 910313o254 910308 foR Aoock osooo44o

- D'[ I PDR

, .~ ,-, ., - . . - . - . , . - -

b- w N, , C,y ,

u. xu AutoutA,ow eC mon ,,,,,,,,,o ,,,,o 3,,,,,,a

. i ...n m.2 Eff nu Af ED bunotN pga af sposait TO Cow *Lv wf M T Mtl LICENSEE EVENT REPORT ILER)  !"o'. "4 ,'iOo?.'o'J*o'.' u'no"I, Elr','u#l',U',.t .t c s"o s AND At#0M fl MANAGtW4NT OR ANCM iP 6301. u S NUCLE AR 7 en atD ft R J 13 to 04 0 IC Os wa NAct utNT ANO 800G47. vvASHINGT08: OC 2060J 8 ACILiTy Naast t)) 'pocalt.svueen421 8 AGt 'J' Perry Nuclear Power Plant, Unit I o 15 l 0101014 l 4l 0 1 l0Fl 013 Differential Pressure Switch Instrument Drift Causes Automatic Start of Annulus Exhaust Gas Treatment System, Train A IVSNT OATR ill Lt A stumastR 441 REPOAT Da f t 171 OYMin f AC1Liitts issv0Lvt0 848 MONTM Dav Vlam vtaa 0, W$ ',4 "['fy roNTM oav etAn f acitst v isawes DoCat t aeve.st aiss 015101010 ! 1 I

- ~

Oj2 0l9 9 1 9l l 0 l 0l7 0l0 0l 3 48 9 l1 o isto 1 0 1 0 1 1 1 O*t h MM TMit A GPORY 18 SutwfTIO PVaSUANT TO TMt ag ogen t Mc NT109 10 C8 A { 'Cher. ca. .* *er. .' fa# *uwf*****' (t il "00' 8'  ! 20 403:01 20 actiel y se MieHines 73 71141 JO .05tead Hd to 304111) to 73 ten 1H.I  ?) 714 5 i901 19( 29 Mt.lOHel 60 M4eH21 te M6entH, d MtA 5 a ,a y ..n n H

. le ,,-,

n ,uH.,

,,..H.,

.e n .u,.

e..H.,._,..,

H A, 2.w

• 9 40s tell1 H.i 40 734est2Heen 93 734eA2Hai LICIN&lt CONT ACT SOA YMis La m (13 NAW4 f t LEPwoN4 NUMet h Amt A C004 -

Henry L. Hegrat, Compliance Er.gineer. Extension 685? 21 1 e 6 , 2 ',19 i 1 37 1  ! 37 1 COMPLETE ONE LINE FOR S ACM CDupoNENT s Astgat DEManet018e THis aspont lim CAu$6 $vlf tw COUPON 4Nt ]"[, 'C C

"','g",T," O CAust ty gt t u Cou'O*e l 8e ?

g ,AC nt fa t X Vic PIDI 4 S t 2[6 14 Y l t 3 i  ! I I I I I ! I I I _

1 1 I l  ! I I SUP*ttWENT AL REPORT E xPtCTIO Mes uoseT M Day vtan io.. u.oM q ,u1,<. - amCnDsv.wwCNOcui )No , ,  ;

A n ACT ,- . . ... r. . ._ . ,-, .a , n .,

On February 9, 1991, at 2003 the AEGTS train A exhaust fan automatically started when an entry was made into the annulus. On February 10, 1991, at 0712, during the monthly required performance of Surveillance Instruction (SV1-M15-T1239) " Annulus Exhaust Gas Treatment System Operability Test", the A fan unexpectedly restarted after shifting from the A to the B fan. Instrumentation and Control technicians determined the problem to be with the calibration of a differential pressure (d/p) swltch [PDS} (Solon, Model Number 7PS2DS). Both the high and low differential pressure setpointa were found out of calibration low. The technicians readjusted the setpoints to the proper values and returned the switch to service on February 10, 1991, at 2225.

The cause of these events was component failure. This instrument has been replaced numerous times in the past and has a documented history of drifting out of calibration.

During this event, the instrument drifting out of calibration resulted in the automatic starting of exhaust fan A when minor disturbances were introduced to the annulus atmosphere during routine evolutions.

To prevent recurrence, the switch was recalibrated to adjust the setpoints to their proper value. Testing was performed to verify that the callbration of the d/p switch

corrected the problem. The calibration frequency of the d/p switch has been increased l- from six months to three conths as an interim measure while an engineering evaluation is in progress to determine a suitable instrument for potential replacement of the Solon d/p switch. As part of the established requalification training program, all plant licensed operators will be instructed on the lessons learned from this event.

NiG f erse 364 idrett

l l

y.. I u. a.m.m.om, co ,o.

i.,,......

LICENSEE EVENT REPORT (LER) '0",'Jl,*du"gd'" "E"',i/ .g"*!l ,*o'",'."41 TEXT CONTINUATION W,71.",'n',",lt3?M'.*d',l/,"^'Un"/ "En

/

Ei"IMOM"n"e$d',Ind.Me.eQTO.e.DC 30 01s'n?E5Y' UI.E Os mAseA M utN1 Asso tuDGif mA

,Acm ,vIbau.tu gga seuest.n ts 9400 (31 lDfEElf NUMesa (21 ma " U fJ',"t  ;'gil Perry Nuclear Power Plant, Unit 1 016101010 M l 410 9l1 -

01Ol7 -

0l 0 Ol2 M 0h inxi n . e, e4.nm On February 9, 1991 at 2003 and on February 10, 1991 at 0712 the Annulus Erhaust Oas Treatment System (AEGTS) (BA) train A exhaust fan [ FAN) automatically started. At the time of the events the plant was in Operational Condition 1 (Power Operation) at approximately ninety six percent of rated thermal power, with the Reactor Pressure Vessel (RPV) at saturated conditions at approximately 1000 psig.

On February 9, 1991, at 2003 the AEGTS train A exhaust fan automatically started when an entry was made into the Reactor Building annulus. Prior to the A fan starting, the A fan was in " STANDBY" and the AEGTS train B exhaust fan was running. At 2020 it was noted that the A fan was running with the control switch in " STANDBY" and no heater in operation. Operators reviewed Emergency Response Information System (ERIS) records and dete rmined that the A fan started at approximately the time that the entry to the annulus was made. Additionally, since the heater was not started, it was concluded that the A fan started due to a H1/Lo dif ferential pressure across the B f an, caused by the entry into the annulus. On February 9, 1991 at approximately 2025, operatorn returned the A fan to " STANDBY" in accordance with approved operating instructions and a work request was written to troubleshoot the problem.

On February 10, 1991, at 0712, during the monthly required performance of Surveillance Instruction (SVI-M15-T1239) "Annulun Exhaust Gas Treatment System Operability Test", the A fan again unexpectedly restarted after shifting from the A to the B fan. The operator held the A fan control switch in the "STOP" position until annulus pressure stabilized and then released the switch to the

" STANDBY" position. An designed the A fan remained off when the switch was released. Instrumentation and Controln (16C) technicians began troubleshooting the A fan automatic starting problem. The 160 techniciann determined the problem to be with the calibration of a differential pressure (d/p) switch (PDS) (Solon, Model Number 7PS2DS). Both the high and low differential pressure setpoints were found out of calibration low. The technicians readjusted the actpoints to the proper values and returned the switch to service on February 10, 1991, at 2225.

The cause of these events was component failure. This instrument han been replaced numerous times in the past and has a documented history of drifting out of calibration. During this event, the instrument drifting out of calibration resulted in the automatic starting of exhaust fan A when minor disturbances were introduced to the annulus atmosphere during routine evolutions.

The AECTS is designed to continunualy discharge filtered air from the reactor building (ngl annulus. This maintains the annulus pressuto negative with respect to the shield building and containment [NHl. The negative pressure in the annulus causes all leakage through the shield building and containment to flow ,

into the annulus, ensuring that any leakage from tne containment vessel will be 5 filtered through the AECTS. The AEGTS consists of two identical trains, one normally in standby. The exhaust fan [ FAN} of the standby train automatically j starts if the operating train air flow is low. A low flow condition la detected by either a low or high d/p across the exh.iust fans. Although an actual low flow 8 2C F.rne J44A 1649)

l s.

. .' s Dets 3044 U 4. mWCL EAR 40 SVLA70%T Cge.samsom

.- .A.,....-

- LICENSEE EVENT REPORT (LER) C',',',*,Aug'di"o"'!la**d/%c?"ll,,"o',".',".'j TEXT CONTINUAT10N f,'o"',,,"/4 '

"l"?,T..'A".1'd ',", ','.'2 '",' "M"f!

",'J",'#. ".!J ""4'"oa"e"'t" a "lFe?"st"& Of 41 '

Of tsaacAotester? A8eo tuCQtt,wAssesse0 Tom. DC 70003.

9 ACl46fy asAa.E tti 00Cattosvestem ut gg, g,ygggg, gg, pagg gy, ma " t'.,*,' V."  %"u: ,

Perry Nuclear Power Plant. Unit I o l6 l 0 l 0 l 0 l 41410 911 -

0l017 -

0 10 013 0F 0 13 l rixvn . - =acn asnawe condition did not exist, the instrument drift resulted in a false d/p signal which caused the A exhaust fan to start as designed.. Spurious actuations of the standby AEGTS train has no adverse safety consequences. Therefore, this' event is

-not' considered to be safety significant. .Two previous events of instrument drift ca'using the A exhaust fan to start, have been documented by LER 87-043 and LER '89-016.-- For the 1987 event, the tolerances were changed for the d/p switch to.

alleviate the problem. For the 1989 event, the d/p switch was replaced with a switch of the same make and model. After the 1989 event, the d/p switch continued to experience drift problems although no additional unexpected fan starts occurred,'and as a-result of the problems with the d/p switch, the calibration frequencyLwas increased from eighteen months to six months.

To prevent recurrence, the switch was recalibrated to adjust the setpoints to their proper value. Testing was performed, by simulating the conditions under

-which the unexpected fan starts occurred, to verify that the calibration of the

~

Ed/p switch corrected the problem. The calibration frequency of the d/p switch has been increased f rom six months to three' months as an-interim measure while an engineering evaluation is in progress to determino a suitable inst.rument- for potential replacement of the Solon d/p switch. As part of the established requalification training program, all plant licensed operators will be instructed on the' lessons learned from this event.

Energy Industry Identification System Codes are identifled in the-text as.[XX). ,

4 NIC Perm Je4A to498

, - , _ , , _ - . __ .,..m_ , , . , _ _ _ , _ . . - . . _ . _ . , _ - _ . , , . . _ . . _ . . _ . . -