._ _ _ . _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ - _ _ _ _ . _ _ _

,

Cp&L '

Carolina Power & Ll ht CompanyD

t

Brunswick Nuclear ProjectP. O. Box 10429

Southport, N.C. 28461 0429July 1, 1991

FILE: B09 135100 10CFR50.73

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

BRUNSWICK STEAM ELECTRIC PIANT UNIT 2DOCKET No. 50-324

LICENSE NO. DPR 62SUPPLEMENT TO LICENSEE EVENT REPORT 2 90 004

Gentlemen:

In accordance with Title 10 of the Code of Federal Regulations, the enclosedSupplemental Licensee Event Report is submitted. The original report fulfilledthe requirement for a written report within thirty (30) days of a reportableoccurrence and was submitted in accordance with the format set forth in NUREG-1022, September 1983.-

Very truly yours,

Y $kk V

J. V. Spencer, General ManagerBrunswick Nuclear Project

TMJ/

Enclosure

ec; Mr. S. D. EbneterM r . N . B , LeBSEP NRC Resident Office

;

f

[[|[K05000.;124 /|1 91070DS *

PDR { y

. _ _ ,, _ . _ ~ . .i -... ....

._ _ i .. . . _ . _ _ . . _ _ _ _ - - - - - -

- _ - _ _ _ - _ _ - _ _ _ _ _ _ _ _ .____ _ _ _ . _ _ _ _ _ _ _ -_. _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ - - _ _ _ _ . _ _ _ _ _ _ _ . - _ _ _ . _ _ - _ - _ _ _ _ _ _ _ -

, .-

NitC FCAM 3(4 U.S NUCtI Aft fif GULA10Hf CC#Mi$50N Af HOYt D (#ff N3. 31MeiD4*

(asuttS-4 M 2( STIMATLD flVHi4 N f t H htbfMbt *O COM8'tY WITH THtt lNFOftMA1K)4CoLLiCTION HL QUI L7: 50 0 NHL f ORWAFC C(#Mt NTS 544 GAf0NG,

'*N'5""*"'''"'"'*""5'N"""""'5"^*d*'*"""".LICENSEE EVENT REPORT (LER) (8'$Y4. U S Nutt1 AR fil OUL ATORY CLWM;55 TON, W AtH'NGTON. DC ?a5L5. I'

AND TO 1Ht I Ast HWD8* Id DUCTK)N Phall CT (3150 0104),of f Ki rpMANACd Mi N1 AND ltVDal1,6 AMtN'atDN, DC ?OM). 1

I# AciuTY NAMr o) Brunswick Steam Electric Plant Unit 2 cen suMra n c24 fant oi |

0$00032401 0F 05

-

nitt<o Manual Reactor SCRAM Due to Failure of Safety Relief Valve "G" to Close During Start upTesting

ivt NT DAit (5) li R NUM6f 31(Os fd f 04f DATE (7) Of H( H F ACIUTil S IN 01 Vt O (t')

MONTH DAY YT AR YiAR Sf0 NO Rf V NC MSNT H DAV YEAR F AClufY NAMf DOCpf f NUMill R

004 - 03 07 01 9103 13 90 90 -

TH8 h!MnRT IS $Ut'MITilD PURSUANT 10 n.t filOJ4I MLNT$ Of' 10 CF R S (Ched one or more of the folloeng)(14Of1 RAhNQ

MX'' (*) 2 20 *a?:to 20 4cNt) X M7)tadhN 7"'M

FOWi R 70 4:taH1;n M rex 1) M 7Ea HDM 737hr)

If dl 00) 007 ?o ** anum Mwann 50 7xaH?vei oTHE n (Steps in At>*vact and Tenti

20 405(aHb410 $07 hah 7H4 50 734sH?dvhmA),

70 4Wa1(hM $0 7FaH7Hiti $0 73taVDfvmHty1 <

20 40Sia H1)M SO T1f air?qhn m7ye d?Hs)

LKf NLEI CONTACT F04 THt$ tf R(12)I

|NAMt THERESA M. JONES, REGULATORY COMPLIANCE SPECIALIST TtuPHOnt NuMnt R

(919) 457-2039COMPLl7E ONE UNC FoR E ACH COWONE NT f AllU8d' D(bcR$t'D IN THl$ fitIOHY O 3)

CAU$( $YSTE M COMfCNENT M ANur ACTU84t R fif 80HT ABL E CAU$f $Y$1(M COMIONE NT MANUF ACTUf4t R Hf f 9RTAHLI v

70 NPRDS TO NPHD$

X AD RV 7020 Y

buPI'L1 MENTAL HENAT LKIT CTL D (14) f AFTCllD MONT H DAY YEARLUbM;5t>ON

YL5 ('f yet, complete tUTcTED SUBMiLLION DATO X NO DATL W

ABS f RACT (Umit to 1400 spaces, I e sporommatety fteen single spar _e tytiewritten unes) 06)

/.t 0536 on March 13, 1990, a manual scram was initiated due to the failure ofsafety / relief valve (SRV) B21-F013G to closo during startup testinc. Reactor

power was ap' proximately 7% and reactor pressure was approximately 256 psig. The11 unit SRV s were beinp cycled in accordance with plant procedures to verifyoperability per Technical Specification 3.5.2. Ten of the eleven SRV's had beensuccessfully tested prior to this failure. A normal scram recovery was conductedper plant procedures and no autwmatic safety actuations or isolations occurred.The investigation determined t'iat the solenoid valve which allows remote manualoperation of B21 F013G was iruperable. The solenoid would allow the SRV to beopened, but would not allow ;1mely closure of the SRV. The solenoid valve wasreplaced, the unit returned to the required testing conditions and the SRV wassuccessfully tested. The se lenoid valve .tas sent to Wyle Laboratory and the rootcause was determined to be failure of the solenoid disc to properly realign withits seat af ter de-energizacion. The report indicated that a potential cause was" dirt" embedded in the ruboer pad located on the backseat. An investigation intothe source of the material was conducted. A contamination source was not foundand the investigation concluded that the instrument air system does not have ageneric problem with contamination The air monitoring program developed inresponse to GL 88-14, the recently installed pneumatic nit ogen systems on eachunit, and the manufacturer's duvelopment of a more durabic solenoid for futureinstallation decreases the potential for a reoccurrence of this event. No

further corrective actions are planned. This event had minimal safetysignificance as the Unit is analyzed for a stuck open SRV at 1004 power.

_ _ _ _ _ _ _ ~ . _ -__ .- ____ _

. . . . .

.

NiC l-OHM d6A U s. NUCtb84 fit OULATOt1Y CoMM.$$KW AFPip.t 0 oMD 80 31% 0104

( niild $. 4/30W15fiMAn D livethi N f t it fit biW410 COMPLY wtTH 1His INIOstMATKW

LICENSEE EVENT REPORT (LER) (y,jg,'O','g,$Eu'|"''N ["s E$ so t>< a co,a ANo id ewis' '

TEXT CONTINUATION u^N^of = $n^Nen m U s. Nticti Ari,4routAioi41 coMM|ssioN.W ALH1NGToN, DC 20%\ AND TO 164 PAlt F4 A'OltK Iti DUCTK14 PlV14 C1

(31500104) of f 601 M MANAGt Mt Ni ANC. TODCdT. A ASHiNGioN.1C 70%)

f AOLitV NAMC (1) (KOft L1 H NUMDt R (6) LACd (3)NUMDI Ft (h

Brunswick Steam Electric Plant Unit 2 'tiouiNnAt swo-

05000324 " A '' NuMne n NuMai n~ 02 0F 05

004 0390 - -

it rt pr Masgt br ACL l$ Its Guild D. US[ A(OnoNAt. NIC f 0844 3%A*S)(17)

EVENTi

A Unit 2 manual reactor scram during safety relief velve (SRV) testing due to thefailure of SRV B21 F013G to close

1RITIAL CONDITIONS:

Unit 2 was at approximately 7% power and at 272 psig performing startup testingfollowing a refueling / recirculation piping replacement outage. Manual cyclingof the SRV's was being performed in accordance with plant procedere Periodic Test(PT) 11.1.2. This tesc manually cycles cach valve open and closed to verifyvalve operability. Ten of the eleven valves had been successfully tested, withonly B21- F013G remaining tc be tested.

EVENT DESCRIPTION

At approximately 0500 on March 13, 1990, SRV testing in accordaace with PT 11.1.2was initiated to verify SRV operability. This testing leach refueling outage per Technical Specification 3.5.I.is performed fo lowingby opening each valvefrom the control room using the remote manual control switch (one at a time),verifying steam flow, and then shutting the valve from the control room. Thefirst ten valves were successfully cycled between 0500 and 0534.

At 0535, SRV B21-F013G was opened from the RTGB, steam flow was verified, and theswitch was placed in the "closo" position. The Control Operator (CO) noted thatthe valve did not go closed and cycled the valve several times in accordance withAbnormal Operating Procedure (AOP)-30, Safety / Relief Valve Failure. Theseattempt- failed to close the valve and at 0536, a reactor scram was manuallyinserted in accordance with AOP 30 At approximately 0537, SRV B21 F013G closedat a pressure of approximately 230 psig. /. normal scram recovery was conductedper Emergency Operating Procedures Path 2, low Power Scram. During this event,the maximum indicated pressure was 279 psig and the icvel was maintained between172 inches and 193 inches (normal 182.5-192.5 inches). There were no automaticchallenges to any safety systems and level was maintained using the feedwatersystem.

EVENT INVESTICATION

Following the event, an investigation was initiated to determine the cause of theB21 F013G failure to close on a "close" signal from its remote manual controlswitch. Initial troubleshooting centered on the solenoid valve (Target RockModel 1/2 SMS-A-01-1) used to port air to the pilot. assembly for remote manualoperation. The solenoid valve was removed from the SRV and taken to themaintenance shop for investigation by plant maintenance and technical supportpersonnel.

Troubleshooting activities began by developing a testing inethodology. A newspare solenoid valve was obtained f rom stock and hooked up to a test rig. Testconditions simulated the installed corditions as near as possible, including a

-- _ - _ _ _ _ _ _ . _ _ .

,

we neu u s wra An ni uut Aum cwM ssou ma o cm o mnem4L WHL h. 4/XW2

,

!$TIMAll D ttU4D(I4 H H Hl hit 18$( 10 LOMP4 Y WI?H THib IN8 OfqMA1IDN

! LICENSEE EVENT REPORT (LER) "'$',",l','$,7,',Zy,'"L',$',^Mif, 3a ,n, ,m,ng, ,,,o nm,n,,,

TEXT CONTINUATION *^4*a MiNn= H m u swo An wtirom a-soHW ALHiNLioN. (C 20'& AND lo THE T'AM RWOhK Ht DUCnON NOJL 0 ?(31'stlem4) of f K t of MANAM Mi NT AND buDCd T. W At.HiNG1DN. tic Ft&J

f AOutY MM( (1) (EtKiT d R HUMDI H @) FAbf O,NUVN H th

Brunswick Steam Electric Plant Unit 2 blOtitNtiAL 4t estON

05000324 ' tan NaMNn NuMat n

004 0390 - -

T( KT (4 Ma41 L8'ACf Ib Hi Quint D. UL[ ADDIf K WAL NIC iohy VAA't;(17)

125 VDC power source and an air supply of approximately 113 psig. Air pressurewas applied to the inlet port of the test solenoid and no leakage was identifled.When power was applied to the solenoid coil, the solenoid valve repositioned,allowing air to enter the test volume (nimulat ing the SRV opening) . power wasthen removed f rota the coil and the solenoid valve repositioned again, venting airfrom the test. volume. If this had been installed on an actual SRV, this ventingof the air pressure would have allowed the spring pre loading to close the SRV.This test was conducted four times to verify test repeatability.

Following vnrification of the test ing methodology, the solenoid reinoved f rom SRVB21 F013G was connect ed to the test rig. The f irst. observation not ed was a Icakf rom the exhaust port. when air was applied t o the solenold. The solenoid coilwas then energized and the solenoid valve repositioned, allowing air to enter thetest volume (simulated valve opening), Air leakage was atill noted around theexhaust port, but the leakare did not appear to be affecting the testing..

Following verification that the solenoid correctly repositioned to the "open"position, power was removed from the solenoid coll. No immediate change wasobserved. The leak at the exhaust port continued and the pressure in the testvolume slowly decreased. This test was also performed several times with thesame results. Based on this testing it was determined that the solenoid was thecause of the fatture of SRV B21 F0i,3G to close.

The solenoid valve which was removed f rom the SRV was shipped to Wyle Laboratoryfor inspection and root cause determination. The root cause investigation was

conducted by Target Rock with assistance from Wyle and Carolina Power 6 Light.Initial test results confirmed the results found at BSEp in that, the solenoidwould open but would not resent correctly. The inability to resent wasattributed to the disc not realigning with its seat properly after de-energization. A visual inspection showed that " dirt" was embedded ln the rubberpad located on the backseat and that loose " dirt" was also f ound in the vicinity.Per the Target Rock personnel performing the inspection, the debris found in thesolenoid was " dirt" and not metal shavings, chips, etc... The solenoid wascleaned out and the dirt disposed of; therefore, it is not availabic forane.lys i s . A search of SRV work requests was made to determine if this was arepetitive failure problem and no other similar failures were identified.

CORRECTIVE ACT12 8

The solenoid valve for SRV B21 F0130 was replaced and the unit was restarted andreturned to test pressure (approximately 250 psig), where the SRV wassuccessfully tested. (Fee: 2-90 004-03 Supplemental Informattoo)

EVENT ASSESSMENT

Unit 2 is analyzed for a st uck open SRV at 100% power which f ails to shut,therefore chts event is bounded. The testing in progress when this failureoccurred is designed to assure SRV operability prior to reaching high powerlevels where one or more SRV failures could create a enore significant event. Assuch, operation within the analysis was assured. The test ing met its purpose andthe event had minimal safety significance

_ _ - _ _ - _ _ _ _ _ _ ____ _ _

'

NFC FOHM t$0A U $. NUCLI aft HE C,ULATOHY COMM$ TON A8WOWD oMB NO. 31SO,0104i KNtf s. 0/1M2

IsilVATID DU50lN F784 IIC broNtt To COMPLY W1TH TH$ INFCd1MahDNautcToN acomsTm Hns uwna

LICENSEE EVENT REPORT (LER) COMMt hT5 REGARD 1N3 Duf0CN isT1 MATE TO THE ufCorot AND tittopfg

TEXT CONTINUATION u'NwMcNT mCH e sm u s. Nucu An nroutamai com smWAEHINGTON. CC 20M$. AND TO THE FAPf fWo4K AfDUCTON PfC.llCT(3150 01D4) of fICE OF MANAGI Mt NT AND BUDGFT. WASHINGTON.oC ?JS03

F ADif Y NAME (1) DOCPD LIR NUM0tH (t) PAGE (3)NUMfit R (h

Bru'swick Steam Ele tric Plant Unit 205000324 vua Nuvern NuMei n

90 004 03--

Tt rT & uonc orACt is atownro, ust AtoTioNAL Not ronM wsmn

2-90-004-03 SUPPLEMENTAL INFORMATION

EVENT INVESTIGATION

In response to GL 8814, Unit 2 instrument air sampling data was collected (ie;prior to this event). A review of data collected on October 16, 1989, f roni themost representative point available, did not indicate that significant debris waspresent in the instrument air line at the time of the collection. SRV pilotsolenoid maintenance history was also reviewed for indications that past f ailuresmay have been caused instrument air contamination. No evidence of previous

problems of this kin were found. The air systems are identical on eachlirunswick Unit and the Unit 1 SRV pilot solenoids are sub ect to similar airquality conditions as the Unit 2 SRV pilots. On December 1, 1990, two Unit 1SRV pilot solenoids were removed from service and disassembled. The involvedpilot solenoids were specifically chosen because they are located on two separatemain steam lines on opposite sides of the drywell and are two of the first threeSRV's to be manually operated when required. No debris was found in either ofthe solenoids; each appeared clean and showed no signs of wear. On January 17,1991, the Unit 1 instrument air supply lines were blown down and mlnutequantitles of debris were collected. The volturie of debris collected was so smallthat it was impractical to measure its volume. The debris was determined toconsist of a gritty / sandy substance and there was no evidence that the smallspecimen of debris was metallic or a corrosion product. It is thought the debrisentered the air lines while sample connections were open.

A contamination source has not been found and the investigation concluded thatthe instrument air system does not have a generic problem with contamination.The potential for fc*ure contaminations and similar events is low because ofrecently installed instrument air nitrogen pneumatic supply (PNS) systems on eachunit and increased instrument air monitoring in accordance witn the programdeveloped in response to GL 88-14. Additionally, the solenoid manufacturer hasreplaced this type of solenoid with a tpc less susceptible to air systemcontamination. This more durable solenoid will by procured for futureinstallation as the original stock is depleted.

ORRECTIVE ACTIONS

The GL 88-14 air monitoring program, PNS and future installation of the moredurable solenoids decreases the potential for a reoccurrence of this event. Nofurther corrective actions are planned.

._- _ _ - _ _ _ _ _ - _ _ -

_ - _ _ - -

. .

NiC f Or# $,6A U $ NUCit AR N GULATOtty COVM%CN Al44cvtD CMf4 N3 3150 0104rova5 nam

L511MATtD BUNDt N ET R FM tion 5f TO CCM61Y WITH THi$ INF OltMATON.

C" '"" 'd out S' "*S ' o"** *LICENSEE EVENT REPORT (LER) COMMINTS 54 GAFONG tlVFtDE N EST)MAft TO THE InCORDS AND tilsORTSI

5m u s *ru^a w wt*= ccwM 5 son.TEXT CONTINUATION u^NA't ut ' t <*Nc" .("AND TO T HE l'AIT HWCnw Ftr DUCYON f 40,l[CTW ASH +NGTON, LC XB55(3150 0104) 05 iICE Or MANAGE MENT AND ttuDCET, W ASHtNGTON.CC 70503

F A01UTY NAMt (1) CKKMT LE R NUMDER (6) FAGE (3)NUMBth A

Brunswick Steam Electric Plant Unit 2 SE QUE NTi AL sqwcN

05000324 3rAa suMet R NuvntR

004 0390 --

it MT (if M0tt. $r ACC 15 Rf Q3:400, ULE ADCWTONAL NtC F04Y WNS)(17)

EIIS

Sys t em/Cottponent C2dt

Safety Relief Valve AC/RVFeedwater SJ

Solenoid Valve AC/V125 VDC Power El

LDAir SupplyPilot Assembly None Found

JCReactor ScramSteam Flow F1

IXPNS