'AC CELE RATED D1~BUT1 ON DE MON STIl ON SYSTgg\

REGULATORY INFORMATION DISTRIBUTION SYSTEM (RIDS)

ACCESSION NBR:8905310098 DOC.DATE: 89/05/19 NOTARIZED: NO DOCKETFACIL:50-323 Diablo Canyon Nuclear Power Plant, Unit 2, Pacific Ga 05000323

AUTH.NAME AUTHOR AFFILIATIONWILSON,S.D. Pacific Gas & Electric Co.SHIFFER,J.D. Pacific Gas & Electric Co.

RECIP.NAME RECIPIENT AFFILIATION

SUBJECT: LER 88-024-01:on 881231,auxiliary feedwater pump potentiallyinoperable due to pressure gauge tap root valve being open.

W/8 . ltr.DISTRIBUTION CODE: IE22D COPIES RECEIVED:LTR ENCL SIZE-TITLE: 50.73/50.9 Licensee Event Report (LER), Inciden Rpt, etc.

NOTES:

RECIPIENTID CODE/NAME

PD5 LAROOD,H

INTERNAL: ACRS MICHELSONACRS WYLIEAEOD/DSP/TPABDEDRONRR/DEST/ADE 8HNRR/DEST/CEB 8HNRR/DEST/ICSB 7NRR/DEST/MTB 9HNRR/DEST/RSB 8ENRR/DLPQ/HFB 10NRR/DOEA/EAB 11NUDOCS-ABSTRACTRES/DSIR/EIBRGN5 FILE 01

EXTERNAL: EG&G WILLIAMS,SL ST LOBBY WARDNRC PDRNSIC MURPHY,G.A

COPIESLTTR ENCL

1 11 1

1 11 11 11 11 11 11 11 11 11 11 11 11 11 1

4 41 11 11 1

RECIPIENTID CODE/NAME

PD5 PD

ACRS MOELLERAEOD/DOAAEOD/ROAB/DSPIRM/DCTS/DABNRR/DEST/ADS 7ENRR/DEST/ESB 8DNRR/DEST/MEB 9HNRR/DEST/PSB 8DNRR/DEST/SGB 8DNRR/DLPQ/PEB 10

PB 10G

RES/DSR/PRAB

FORD BLDG HOY,ALPDRNSIC MAYS,G

COPIESLTTR ENCL

1 1

2 21 12 21 11 01 11 11 11 11 12 21 11 1

1 11 11 1

A.

D

D

R

I

NOTTE 'IO ALL "RIDS" RECZPZENIS.

PIZASE HELP US IO REMJCE HASTE! CGHI'ACI'IHE DOCUMENT CORI'.ROL DESK,RGCN Pl-37 (EÃZ. 20079) IO EIJKQCQZ YOUR NAME PSYCH DISTiGBUTIGNLISTS H)R DOCQKNZS YOU DOH~T NEED)

h

D

S

TOTAL NUMBER OF COPIES REQUIRED: LTTR 43 ENCL 42

t,

Pacific Gas and Electric Company 77 Beaie Street

San Francisco, CA 94106

415I972.7000MX910 372 6587

James D. Shiffer

Vice President

Nuctear Power Generation

May 19, 1989

PG&E Letter No. DCL-89-140

U.S. Nuclear Regulatory CommissionATTN: Document Control DeskWashington, D.C. 20555

Re: Docket No. 50-323, OL-DPR-82Diablo Canyon Unit 2Licensee Event Report 2-88-024-01Auxiliary Feedwater Pump Potentially InoperableDue to Pressure Gauge Tap Root Valve Being Open

Gentlemen:

PGhE is submitting the enclosed revision to Licensee EventReport 2-88-024 concerning a pressure gauge tap root valve on thesteam supply to an auxiliary feedwater (AFW) pump that was foundopen during surveillance testing. It has been determined that thisevent most likely did not cause the AFW pump to be inoperable. Thisrevision provides corrective actions resulting from PGhE'sinvestigation of the causes of valve mispositioning events at DiabloCanyon Power Plant.

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

Kindly acknowledge receipt of this material on the enclosed copy of thisletter and return it in the enclosed addressed envelope.

Sincerely,

/"J. D. S fer

cc: J. B. MartinM. M. MendoncaP. P. NarbutH. RoodB. H. VoglerCPUCDiablo DistributionINPO

Enclosure

DC2-89-OP-N005DCO-89-OP-N003

2593S/0069K/ALN/2263

8905310098 890519PDR ADOCK 05000323

PDC

~ACILITTIWIH lfI

IJ WDIABLO CANYON UNIT 2

GAUGE TAP ROOT VALVE BEING OPEN

LICENSEE EVENT IIEPOIIT ILEIII 6,

~EKEET ~la Ql

0 5 000~4CII I

T OF

~sEOIET OETE «I Lla Iaa«la IEs aEEOaf 04TE ITI Offlla EACILITNENIVOLVEO «IOkf Vl44 VE44 Iastf »I ~

4»s»4 5 ~~» tODst ~ ~s 4 04» Vila ~4CSL f 44WE OOCKET INNEEEasl ~

0 5 0 0 0

~ aaf laomaK «I

Taa alroaT «wasITTEO EvaEOaln TO Tl«aEOOlalwNTE OE TE CEa g nll0 S 0 0 0

~OEIE ~LEVEL

H OTNE1 NEEENV «Ahaet~EseeWEI T~ NC Sea

(1) (B)

L«EI«EE CONTaCT EOa TN«Lta ILEI

STEPHEN D. NILSON, REGULATORY GNPLIANCE ENGINEER

COE«LETE IWt LWE EOa EECfs COE«ONENT EalLLlal OE~«EO la TII«aEEONT lllI

TELE~Ossl NMWE~

ala COOE

9 7 1

«alsIIE 4CTgola

«EleVE 4CTvala

altONTAELlTO N44DE

I's5;;5.,C~s5 I 5I

~LEIEENTaL aEEOaf EE4ECTEO Ill

TEE II»~ ~ C~CTlp E~lltt04 OETEI

~aaCT nil

EEET'OOSIE+IEESOSI

OETE sl ~

~ss f O4» fE44

On December 31, 1988, at 0129 PST, operations personnel were performing STP P-6B onAFH pump 2-1. The 1n1.tial portion of the STP requires warming the pump turb1ne byclosing the turbine throttle-trip valve (FCV-l52) and then opening the turbinesteam supply valve (FCV-95) to warm the steam inlet line to the turbine. 4non-licensed aux111ary operator closed FCV-152 locally at the turbine, and thenrequested that FCV-95 be opened from the control room. Hhen FCV-95 was opened,steam began blowing into the pump room. The auxiliary operator immediately exitedthe room, called the control room, and requested that FCV-95 be closed. A

subsequent investigat1on revealed that a pressure test connection (PX-213) wasuncapped with its root valve, NS-2-923, open, creating a d1rect steam. path into thepump room from the 5/8 inch 1nside diameter pressure gauge connect1on.

The root cause of the valve being open could not be determ1ned.

PGLE has investigated several recent valve oispositioning events on a generIc basisand has concluded that these events can be attributed to several possible causes:(l) personnel error, (2) inadequate or 1ncorrect procedures or drawings, and(3) unauthorized operation of plant components. To prevent recurrence, appropr1ateadministrative procedures have been and will be strengthened, operating procedureswill be rev1ewed against OVID drawings, an Operat1ng Order has been developed, anOperat1ons Policy has been revised, and a work planning policy memo has beenissued. The procedural changes will be reviewed with all plant operators.

2593S/0069K

1

0+C ~ %4+%4J I llCENSE NT hEPORT LER) TEXT CONTINU

)N

Uk ~L~ MosaaToaT geaaecaiose

~ao oae ee g>go~pe

0kCILfTT~ Ill

DIABLO CANYOIt UNIT 2vev r~eeeeeereC ~~eseas maSnn

o s o 0 0 3 3

LNO OIA&t+I4lOVla»Aaa ~

I1D%(0

1 0 2 0 8

The unit was $ n Hode 1 (Power Operation) at approximately 100 percent power.

A. Event:

B.

On December 31, 1988, at 0129 PST, operations personnel were performingSurveillance Test Procedure (STP) P-6B, Routine Surveillance Test ofSteam-Driven Auxiliary Feedwater Pump" on auxiliary feedwater (AFH) pump(BA)(TRB)(P) 2-1. The initial portion of the STP requires warming thepump turbine by closing the turbine .throttle trip valve (FCV-152)(SB)(FCV) and then opening the turbine steam supply valve (FCV-95)(SB)(ISV) to warm the steam inlet line to the turbine. A non-licensedauxiliary operator closed FCV-152 locally at the AFW turbine, and thenrequested that FCV-95 be opened from the control room. Hhen FCV-95 wasopened, steam began blowing into the pump room. The auxiliary operatorimmediately exited the room, called the control room, and requested thatFCV-95 be closed. A subsequent investigation revealed that a pressuretest connection (PX-213) (SB)(TV) was uncapped with its root valve,MS-2-923 (SB)(RTV), open, creating a direct steam path into the pump roomfrom the 5/8 inch inside diameter pressure gauge connection.

A review of operating records indicated that on November 30, 1988, AFHpump 2-1 was taken out of service for a bearing oil change following theroutine pump surveillance STP P-6B. PX-213 was used as the required"visible vent" for that clearance. The root valve for PX-213, HS-2-923,was documented as being closed and independently verified as closed onDecember 1, 1988, following completion of the pump bearing oil change. Apost maintenance test after the bearing oil change was not performedbecause the oil change was not considered by operations and maintenancepersonnel to have any potential impact on the operability of the AFH

pump. The independent valve alignment verification was consideredadequate to verify operability.

Inoperable structures, components, or systems that contributed to theevent:

C.

Kone

Dates and approximate times for ~or occurrences:

1. November 30, 1988 at 0511 PST:

2. November 30, 1988 at 1349 PST:

AFH pump 2-1 routine surveillanceis performed (STP P-6B)satisfactorily.

AFH pump 2-1 is taken out ofservice for a bearing oil change.

2593S/0069K~sac rose waOi)

ij

NNC fons 99EAl949 I LICENSEE NT REPORT tLER) TEXT CONTINUA N

U$ , NUCLEAII NEOULATOIIYCOMMISSION

AftNOVEDOMO NO. 9ISOWIOINIfNILS:OI91/%

fACILITYNAME lll

DIABLO CANYON UNIT 2TUCT W~ Mfa e~ ~~ lfIICAwe~&IITI

OOCIIET IDDEN Ql

0 5 0 0 0 2

LEO NIPNEII IEI

EEOVINnAL .'; VaneNVV IA V": NVNSIN

2 —0 1 30F

3. December 1, 1988 at 0700 PST: AFH.pump 2-1 is returned to.service following the oil change.

4. December 31, 1988 at 0129 PST: AFH pump 2-1 STP P-68 identifiesan open pressure tap PX-213.

D. Other systems or secondary functions affected:

None

E. Method of discovery:

The open pressure tap ms iamediat+)y obvious to the operator in the pumproom due to steam blowing out of the pressure gauge connection (PX-213).

F. Operator actions:

The steam supply valve (FCV-95) to AFH pump 2-1 was closed, PX-213 was

properly isolated and capped.

G. Safety system responses:

None

A. Ianediate Cause:

The ianediate cause of this event 1s that PX-213 was uncapped andunisolated, creat1ng a steam leakage path from the AFH pump turbine steamsupply line.

B. Root Cause:

The root cause of th1s specific event could not be positivelydetermined. However, during the 1nvestigation PGLE has reviewed severalrecent valve mispositioning events on a generic bas1s and has concludedthat these problems can be attributed to several possible root causes:

1. Personnel error due to a failure to follow established procedures orpolicies or failure to use estab11shed operating techn1ques duringvalve positioning.

2. Inadequate or incorrect operating procedures or OVID drawings.

3. Unauthorized operation of plant components.

2593S/0069K '

IIIIC fena 3NA(9431 LICENSEE NT REPORT (LER) TEXT CONTINUAggN

QS, IAICLEAAAKOULATOIIVCOMMISCIOII

Att1lOVEO OIIS IIO. llSOLI(HSCALES: NSI/%

DIABLO CANYON UNIT 2mar u~~i~ ~~aeC a ~~11n

0 5 0 0 0 2

VEAA

LCA II~EA IQ

S C 0 V1 IIT I ALNVU IA

0 2

VaaIIWVMQIA

0 1 4 OF

The root cause for the valve be1ng open could not be positivelydeterained. The root valve, l6-2-923, was closed, and verified closed onDecember 1, 1988. KS-2-923 1s located approximately at wa1st he1ght 1nthe AFH pump 2-1 room. Lighting 1n the area allows the valve tag forKS-2-923 to be read without the aid of a flashlight and is thereforeconsidered adequate. The pressure gauge connection consisted of a pipecoupling that should have had a plug screwed 1nto the open end. It ispossible that an operator could look at the pipe coupling while operatingKS-2-923 and conclude that the pipe coupling was plugged or mistake thecoupling for a p1pe cap, although th1s is considered unlikely due to theorientation of the valve (the valve protrudes directly into the walkway)and the 11ghting in the area. MS-2-923 was 1nspected after the event tosee if its operational characteristics could have misled operators intobelieving it was shut. The valve was found to be difficult to operate,but not so difficult as to lead the operators to believe that the valvewas closed. This is considered especially unlikely as the 1nitialoperator expected to find the valve open and antic1pated having to shutthe valve in order to place it in the pos1tion required by the clearancerequest form. In an effort to determine the cause of th1s problem, anextens1ve 1nvestigation was conducted wh1ch included the following steps:

l. Operating logs and maintenance records applicable to the November 30to December 31, 1988 time periods were rev1ewed, no maintenance ortesting act1vities involving this equipment were found. Clearancerequests and applicable work orders were reviewed in an effort todetermine any activities that may have led to the pressure gaugeconnection being open. Additionally, Unit 1 records were reviewed todetermine if maintenance had inadvertently been performed on thewrong un1t, and again no activities were 1dentified which would haverequired use of the Unit 1 PX-213 gauge connection.

2. The operations personnel directly 1nvolved with this evolution were1nterviewed. No additional informat1on was obtained from theinterviews that contradicted the valve positions recorded on theclearance request form.

3. All operations personnel were questioned to attempt to obtainadditional information. No additional information was obtained.

It 1s postulated that the valve was either left open by the twonon-licensed operators who were assigned to shut and verify the valveshut and capped or the valve was opened and uncapped during undocumentedmaintenance activities. There is no-conclusive evidence to supporteither postulation.

2593S/0069K

~IIIC I@IIII~AamI

NIIC fena ~(945) UCENSEE NT REPORT (LER) TEXT CONTINU '

US. NUCLCAA ACOUIATOIIYCOCNIICSION

AftAOVCDOTIC NO SICO&10iCXMCSI CICI~

DIABLO CANYON UNIT 2maI~~~N~~~NNca ~WIm

IV.

0 6 0 0 0 2

LCII NLSNCII lil~ 'IOUIATsAL

NVQ Is

02 4

IVIONHUMII0

0 1 5 OF

An init1al engineer1ng analysis of this event concluded that the temperaturein the turb1ne-dr1ven AFH pump room due to steam discharge from the uncappedpressure test connection would have exceeded the specified turbine operatingrange. Based on th1s analys1s, 1t was.conservatively concluded that the AFH

pump turbine say have failed if required to operate for an extended period oftime (over two hours) under the h1gh temperature condition. However,operating records indicated that both motor-driven AFH pumps were operableduring that time period and the physical size and orientation of themotor-driven pump room is such that the steam would not have rendered them1noperable. Therefore, it was concluded that no unanalyzed safetyimplications or consequences had resulted from this event.

Further evaluation following PG&E's initial report ind1cates that theturbine-driven AFH pump most 11kely would not have been disabled by the steamdischarge from the uncapped pressure test connection if called upon tooperate. PG&E's aux111ary steam header crack analysis shows that the durationof high room temperature resulting from a header crack is less than one hour.The temperature r1se is terminated when pressure buildup due to the steam leakcauses a door to fail. Following the door failure, the room temperaturerapidly decreases. Because the blowdown rate from the uncapped pressureconnect1on would exceed that for the header crack, the pressure bu1ldup andconsequent door failure would occur sooner. Also, the actual door integrityis less than that conservatively assumed in the header crack analysis.Therefore, PG&E concludes that the high temperature condition in theturbine-driven AFH pump room resulting from the uncapped pressure connectionleak would last s1gnificantly less than one hour.

In con)unction with th1s, actuation of the fire suppress1on system spr1nklerheads 1n the turbine-driven AFH pump room occurs at a temperature between160-180'F. These fusible-link actuated spr1nklers would quench thetemperature rise, and thus provide an additional 11kelihood that the peak roomtemperature for either the postulated header crack or the sub)ect pressureconnect1on leak would not approach the 341'F peak temperature calculated 1nthe steam header crack analys1s.

A vendor assessment of equ1pment operability at 350'F 1ndicated that pumpbearing failure is possible, but the pump is likely to perform its designfunction for several days. Turbine bearing failure due to increased ambienttemperature 1s also possible, but the turb1ne is 11kely to be operable for atleast one to two hours. Khen the above mitigat1ng events are coupled with theaanufacturer's assessment of equipment operab111ty, it is )udged that theturb1ne-driven AFH pump would have been operable if required.

2593S/0069K

51AC fona ~<9451 LICENSEE NT REPORT {LER) TEXT CONTINU, N

US. ffUCllAAAEOULATOIIYCOAfffNQOII

AttltOVE0 OMl ffO NiOM1OffXfNES:NllI

OIABLO CANYON UNIT 2TUCI/~~e~ ~~hefCFeea~slf171

1SA ffUSNlIIIQ

55OU5%5IAI 1"j VI5ION~IVM55 W '~ NIIM 5 A

o 5 o o o 2 88 —02 4 —0 l 6 QF

In con)unction with the potential inoperability of the turbine-driven AFH

pump, the following conditions which could disable the ootor-driven AFH pumpswere evaluated:

o Actuation of the sprinkler system in the motor-driven AFH pump room,or

o A loss of all AC power.

A qualitative evaluation based on the auxiliary steam header crack analysis1ndicates that the temperatures 1n the motor-driven AFH pump room result1ngfrom the pressure connection blowdown will not reach the sprinkler actuationtemperature. Add1tionally, the sprinkler heads 1n the motor-driven pump roomare actuated by fusible-links and therefore operate 1ndependently from and arenot actuated by the sprinklers 1n the turb1ne-driven pump room. Therefore, 1tcan be concluded that sprinkler actuation in the motor-driven pump room wouldnot occur during the sub)ect event.

V.

Hith regard to a loss of all AC power, it 1s noted that this is beyond thedesign bases of the Diablo Canyon Power Plant. Nevertheless, themanufacturer's estimate of one hour pump/turbine operab111ty provides asign1ficant time period to restore AC power using one of two offsite powersources or the onsite emergency generators.

Therefore, as prev1ously concluded, no unanalyzed safety implications orconsequences resulted from this event, and the health and safety of the publictherefore were not adversely affected.

A. Iaaedi ate Correcti ve Actions:

The steam leakage path was isolated by 1solating PX-213, and the pump wasverified operable by completing STP P-6B.

B. Corrective Actions to Prevent Recurrence:

Because of this- event and other recent recurring valve aispositioningevents, PG&E has investigated these events on a generic bas1s and hasdetermined the following corrective act1ons to prevent recurrence:

l. Administrative Procedure (AP) C-6Sl, 'Clearances," has been revisedto ensure that all vents and drains not connected to a closed systemhave a pipe cap 1nstalled whenever the vent and drain valve 1sclosed. Additionally, a work planning policy letter has been 1ssued

2593S/0069K

~IIIC 5 0AM 5554$ 45 I

IIIIC fena %CAl$4ll L1CENSEE NT REPORT {LER) TEXT CONTINU, N

UJ. IAICLCAIIIIKOULATOIIVONQAINIOII

AtffIOVKOOMO IIO. NSOWIOIEXES: I/RIN

LCA IIIANEII Nl

DIABLO CANYON UNIT 2TUCt H~~ 4 aysea4 ~ OOAaed AIICAna~ AlIlll

0 5 0 0 0 2

VCAA "g SLOVIHTIALN VI> IA

VI5IOHIVVIIA

0 1 7 OF

2.

3.

4.

5.

6.

7.

8.

to requ1re documentation of pipe cap removal and reinstallation inthe applicable clearance request or other controlling work document.

Operating procedures containing system alignment checklists will bereviewed and cross-checked with the OVID drawings to ensure that allsystem valves (excluding instrument valves downstream of Instrumentroot valves) are 1ncluded in the alignment checklists. Checklistswill require that appropriate vents nd drains be capped.

AP C-6S1 hand AP C-53, "-Authorization for Equipment Operation andHaintenance." have been rev1sed to strengthen requirements fornon-operator plant personnel to operate plant components. Theserevis1ons include spec1fic requirements to control those valves whichmay be operated by persons other than operations personnel.

Operating Order 0-17, "Hanual Valve Alignment," has been developed toprov1de explicit generic 1nstructions for performing valvingoperat1ons. This procedure 1ncorporates expected operator techn1quesand established "good practice" policies to be used as guidance forperforming valving type operat1ons.

AP C-9S1, "Sealed Valves," has been revised to strengthenrequirements for sealing valves and breaking valve seals andimplement a periodic walkdown of sealed valved checklists.

AP C-104Sl, "Independent Verification of Operating Activ1ties," willbe rev1sed to clarify independent verification requirements andresponsib111ties.

Operations Policy B-l, "Conduct of Operations," has been rev1sed toinclude a section providing requ1red techniques to be used whenevermanual valve man1pulations are to be performed.

The procedural changes discused above will be reviewed with all plantoperators 1n on-shift training sess1ons. In addition, changes toAP C-6Sl and AP C-53 will be reviewed with all affected personnel intailboards or management meetings.

Corrective actIons for equ1pment status control, including genericvalve aispositioning, were d1scussed at an April 25, 1989, meetingwith NRC management.

2593S/0069K

NAC IOIIV$4/AI44ll

IIAC feIIA~IfATI UCENSEE NT REPORT LER) TEXT CONTINUA t N

LIS. NJCLTAII AKOULATOIIYCOMMCQOII

AffIIOVTD OMO MO. SI %OMIT

%XMAS�'.~ IH

fACILITYIIAMLIll\

OOCIICT NAHKIIQl LtA IIIAMCAIOI

DIABLO CANYON UNIT 2mCrrr~MAMe~~~ffACff ~Ylnn

VI.

A. Failed Components:

None

0 6 0 0 0

YNAII v.i $ (OVCIITIALVIIV I II ~

i'2

VOIOIIVIIV fA

0 1 8 OF

B. Previous LERs on similar events:

LER 1-88-026-01 Reactor Trip due to Turbine Trip from Anti-NotoringRelay Caused by a Closed Root Valve on the LowPressure Side Sensing Line.

This event was a turbine trip and subsequent reactor trip that occurred asa result of a mispositioned valve associated with the main turbineanti-eotoring relay. The corrective act1ons 1ncluded 1ssuance ofAdministrative Procedure AP C-53, "Author)zation for Equipment Operationand Haintenance," and AP C-153, 'Plant Status Controls,'o provide1mproved documentation of changes in component status. S1nce the rootcause of the event in LER 2-88-024 could not be determ1ned, it can not bestated whether these correct1ve actions would have prevented this event.However, it is believed that these adm1nistrative procedure changes shouldreduce the frequency of events related to loss of plant component status.

LER 2-88-013-01 Reactor Cavity Sump Level Instrument Channel Out ofService Due to a Closed Valve on the Instrument A1rL1ne

This event involved exceed1ng the t1me requirements of a TS actionstatement when an air supply valve was left closed making the sump levelinstrument channel 1noperable. The root cause of this event was

. determined to be lack of a requirement 1n the appropriate operatingprocedures to verify the alignment of this air supply valve after extendedmaintenance periods. This valve was added to the app11cable valvealignment checkoff list to ensure that its pos1tion is verified to becorrect. This corrective action did not prevent the occurrence of theevent 1n LER 2-88-024 because the AFW pump had not been in an extendedNaintenance period and a complete system lineup check was therefore notperformed nor requ1red.

2593S/0069K

VAC fQAV~Atf43 I

lt <Ig