. - -,.

a

*<

" '

Cp&Lm

U$r m &t!FJtGam e m m Ma*MfMt Car <2!ina Power & Light Company

y wommazarnwaamaswna

Brunswick Nuclear Plant .!

P.O. Box 10429 ;Southport, NC 28461-0429 '

1'

WR I i 1994

SERIAL: BSEP-94-0101,

10CFR50.73

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

BRUNSWICK NUCLEAR PLANT UNIT 1DOCKET NO. 50-325/ LICENSE NO. DRP-71LICENSEE EVENT REPORT 1-92-023 Supplement Three

Gentlemen:,

In accordance with the Code of Federal Regulations, Title 10, Part 50.73, Carolina Power &Light Company submits the enclosed Licensee Event Report Supplement. This report issubmitted in accordance with the format set forth in NUREG-1022, September 1983.

Please refer any questions regarding this submittal to Mr. Steve Tabor at (910) 457-2178.

Very truly yours,

|4 h & __aJ. Cowan, Director-Site Operations (acting)Brunswick Nuclear Plant

SFT/

Enclosures1. Licensee Event Report2. Summary of Commitments

cc: Mr. S. D. Ebneter, Regional Administrator, Region 11 -Mr. P. D. Milano, NRR Project Manager - Brunswick Units 1 and 2Mr. R. L. Prevatte, Brunswick NRC Senior Resident inspector

150051iMop,) , 'EBR3$886R888366 A

S PDR- k .l,

. . __ _

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

% ,.. .

*; s

EnclosureList of Regulatory Commitments

The following table identifies those actions committed to by Caro!!na Power & Light Company,

' in this document. Any other actions discussed in the submittal repressent intended or plannedacticos by Carolina Power & Light Company. They are described to the NRC for the NRC's.

information and are not regulatory commitments. Please notify the Manager-RegulatoryAff airs at the Brunswick Nuclear Plant of any questions regarding this document'or anyassociated regulatory commitments.

.- ..

CommittedCommitment date or

outage

NONE

|

|

|!

'Ii

'|

l,

.

.

NRC FDRM 366 U.S. NUCLEAR REGULATORY COMMISSION APPROVED OMB NO. ? 150 0104(5/92) EMPIRES: 5/3195

ESTIMATED BURDEN PER RESPONSE 'O COMPT Y WITH THISINFORMATION COLLECHON REQUEST: M (5 HHS. FORWARD

LICENSEE EVENT REPORT (LER) COMMENTS REC ARDING BURDEN ESUMATE TO THE MORMATIONI

AND RECORDS MANAGEMENT BRANCH (MN88 7714L U.S. NUCLEARREGULATORY COMM:S$10N, W ASHINGTON, DC20555-000), AND TOTHE PAPERWORK REDUCTION PROJECT (31500104L OFFICE OFMANAGEMENT AND BUDGET. WASHINGTON DC 20$O3

I ACIUTY NAME (1) DOCKET NUM8ER W PAGE (3)

Brunswick Steam Electric Plant, Unit 1 05000325 1 of 17_

TITLE 14)

Local Leak Rate Test Failure of Both RCIC Steam Line Isolation Valves

EVENT DATE (S) LER NUMBER (6) REPORT DATE (7) OTHER FACILITIES INVOLVED (8)

MONTH DAY YEAR YEAR MONTH DAY YEARNUMBER NUMBER BSEP Unit 2 05000324

09 27 92 92 - 23 - 003 03 11 94 F AcitrTY NAME OOC KE T NUMBER

05000 |IH6 MOHI i$ buBMilILD PUH5UANI TO THE REQuiHLMENib OF 10 CFR L (Che one er mue of We fuiiuwmg411)

OPERATINGMOM @ 20.402ib) 20.405(c) 50.73(aH2Hiv) 73. 71(b)

POWER00LEVEL 110) 20.405(a H 1 HW 50 36(cH2) 50. 73(a H 2Hvn) X oTHER

20.405(a H 1 Hm) 50.73(a H 2H O 50. 73(a H 2H vmHA) (Specify in Abstractand Text)

20.405(a H ilbv) X 50. 73(a H 2HW 50.73(aH 2HviuMB)

20.405(aH 1H vl 50.73(a H 2H m) 50. 7 3(aH 2H z)

LICENSEE CONTACT FOR THIS LER (12)

NAME TELEPHONE NUMBER

Steve F. Tabor, Regulatory Compliance Specialist (919) 457-2178

COMPLETE ONE LINE FOR FACH COMPONENT FAILURE DESCRIBED IN THIS REPORT (13)

CAUSE SYSTEM COMPONENT MANUF ACTURER CAUSE SYS TEM COMPONENT M ANUF AC TUHf.Rg PD

X JM ISV A391 Y

"0NT" "^* *" ISUPPLEMENTAL REPORT EXPECTED (14) EXPECTEDSUBMISSIONns y' W )

DATE (IS)m yn wrm EXPECTED sueMissiaN Don || J.

AJSTRACT (Urmt to 1400 spaces. i.e. apprommatefy fif teen sing!e space typewritten hnes) (16)

On September 26, 1992, with Unit 1 in Cold Shutdown, Local Leak Rate Testing (LLRT) of ;

the Reactor Core Isolation Cooling (RCIC) outboard steam isolation valve, 1-E51-F008, i

identified value seat leakage in excess of acceptable limits. An LLRT of the RCIC |inboard steam isolation valve, 1 -- E 51 - F0 0 7 , performed on August 3, 1992, similarly |identified excessive leakage past the valve seat. Consequently, with both Primary

|Containment Isolation (PCIS) valves failing to seat, the potential existed for loss ofthe Primary Containment Isolation capability for the affected penetration had the unitbeen operating. An inspection of the 1-E51-F007 was performed following the LLRT. Theinspection did not reveal the cause for the excessive leakage. The 1-E51-F007 valvediscs were replaced due to a previously identified concern with binding of this type at i

lvalve in the open position. A post maintenance LLRT was performed which verified 0 SCFHIleakage. Although initial inspection of the 1-E51-F007 did not reveal the cause of the

excessive leakage, a root cau;e evaluation of this failure is continuing. Additionally,an investigation into the cause of the 1-E51-F008 leakage is in progress. Failure ofboth PCIS valves to seat on the same 3 inch line is significant and represents a loss ofthe PCIS capability for that penetration. A previous similar event is addressed withinLER 1-91-016. Subsequent valve testing has identified additional failures of 3", 4", 6"

and 10" Anchor / Darling double-disc gate valves. Cumulatively, these failures areconsidered 10CFR21 reportable. A supplemental information section is included to address i

!the 10CFR21 notification requirements.I

Supplement 3 to this LER provides the results of the analysis performed to determine the fIcause of the 1-E51-F007 and 1-B21-F019 valve LLRT failures.-

J

-- . .

.

NRC FORM 366A U. S. HUCLEAR GEGULATORY COMMISSION APPROVED OMB NO. 3150-0104 ~{5/02) EXPIRES: 5/31/95

EST! MATED BURDEN PER RESPONSE TO COMPLY WTTH THl3INFORMATION COLLECTION REQUEST: SO O HRS. FORWARD

LICENSEE EVENT REPORT (LER) CouMENTS REcARo,No euRaEN ESTiu ATE TO THE woRu Anon 4No

TEXT CONTINUATION RECORDS MANAGEMENT BRANCH (MNBB 7714). U.S. NUCLEAftREGULATORY COMMISSION. WASHINGTON. DC 20556 000L AND TOTHE PAPERWORK REDUCTION PROJECT (3150-O t04L OFFICE OFMANAGEMENT AND BUDGM. WASHtNGTON. DC 20503.

m

FACILITY NAME (1) DOCKET NUMBER (2) LER NUMBER (6) PAGE (3)

SEQUENTIAL REVISION

Brunswick Steam Electric Plant " * " " " * *'"05000325 2 of 17

Unit 192 - 23 - 003

TEX T (If more space is required, use add tional' HC form 366Ksl (1 h

TITLE,

Local Leak Rate Test Failure of Both RCIC Steam Line Isolation Valves

INITIAL CONDITIONS

On September 26, 1992, Unit 1 was in Cold Shutdown in day 158 of a maintenance outage.To satisfy Technical Specification (TS) containment leakage rate surveillance requirements,Technical Support engineering personnel had commenced leak testing of the RCIC outboardsteam isolation valve, 1-E51-F008. The reference testing is accomplished by closing thevalve, pressurizing the disc / seat area to 50 psig, and measuring the makeup flow to thevalve. The normal acceptable leakage rate for the associated penetration is 3 scfh.

EVENT NARRATIVE

On August 3, 1992, a LLRT of the RCIC Steam Inboard Isolation Valve, 1-E51-F007, wasconducted. The test results indicated that the valve would not pressurize to the desired50 psig test pressure. On August 13, 1992, troubleshooting was initiated. Following valvedisassembly, the valve internals were inspected and the valve disc blue checked. Theinspection did not reveal a condition to which the excessive leakage could be attributed.The blue check verified 98% disc to seat contact. Due to a previously identified genericconcern with the Anchor Darling double-disc gate valve (i.e., valve disc sharp edgesresulting in jamming in the partially-open position when opened against differentialpressure), the valve disc was replaced. On September 25, 1992, a post maintenance LLRTverified the 1-E51-F007 leakage to be o sefh.

.)On September 26, 1992, a LLRT of the RCIC Steam outboard Isolation Valve, 1-E51-F008, wasconducted. The test results indicate that valve leakage was 252.07 scfh. |

|

Based on the results of the 1-E51-F008 LLRT and having previously identified an excessiveleakage rate while testing the 1-E51-F007, a ENS Notification was made at 1235 on September27, 1992.

Investigation into the cause of the 1-E51-F008 LLRT failure is currently in progress. Thevendor is assisting in this investigation. Preliminary results of the investigationindicate that disc wedge irregularities may be contributing to this problem. TechnicalSupport and Nuclear Engineering Department personnel are pursuing completion of the rootcause for failure of both of the RCIC steam isolation valves. Corrective actions toprevent recurrence will be established as part of the root cause investigation.

CAUSE OF EVENIAn investigation int'o the cause of the RCIC steam isolation valve leakage is in progress.An action item has been assigned to Technical Support to complete the investigation anddevelop / implement ;orrective actions to prevent recurrence.

CORRECTIVTdC_TJ MS_

The RCIC at2am Incoard Isolation Valve, 1-E51-F007 has been repaired based on the currentunderstaading or the cause of the problem. LLRT results verified that the valve leakagefollowing :uaintenance is within the acceptable TS limit. However, the on-goinginvestigation of the 1-E51-F008 may identify additional failure modes not recognized at

_

ce

i

.

.

NRC FORM 366A U. S. NUCLEAR REGULATORV COMMihHON APPROUED OMB NO. 3150-0104(5/92) EXPIRES: 5/31/95

,

E$ TIM AIED BUROEN PER RESPONSE TO COMPLY WITH THISINFORM AllON COLLECTION MOUESt ; SO O HAS. F ORWARO

LICENSEE EVENT REPORT (LER) COMMEN T$ MGARDWO BURDEN f SUMAM TO THE WFORM A00N AND

TEXT CONTINUATION MCORDS MAN AGEMEN T BRANCH (MN88 m 4L U S. NUCLEARREGULA FORY COMMIS$10N. WASHfMOTON DC 20655 000), AND TO

THE PAPE RWORK REDUCTION PROJECT 13160 01041. OFFICE OFMANAGEMENT AND BUDGET. WASHINGTON, DC 20503.

FACluTY NAME (Il DOCKET NUMBER (2) LER NUMBER (6) PAGE (3)

SE QUEN TIAL NEVISION

Brunswick Steam Electric Plant """"" * " ' ' "05000325 3 of 17

Unit 192 - 23 - 003

-

itXT fit more space is requved, use additionalNHC form 366A's1 (1 h

the time of the 1-E51-F007 repair. Consequently, additional 1-E51-F007 correctivemaintenance may be required.

Additional corrective actions resulting f rom the completion of the root cause investigationwill be documented ist a supplement to thic report.

SAFETY ASSESSMENT

The LLRT f ailure of both the 1-E51-F007 and 1-E51-F000 valves resulted in a leakage of 252SCFH. An analysis was performed to determine the safety significance of this conditionduring the High Energy Line Break (HELB) and the Loos of Coolant Accident (LOCA) scenarios.The following provides the results of this analysis.

HELD:

The RCIC steamline isolation valves are designed to close in the event of a RCIC steamlinebreak in secondary containment to prevent an uncontrolled loss of reactor coolant. Sincea source term is not created by a RCIC steam line pipe break, a leak tight isolation ofthese valves is not critical. CP&L has addretsed this position in a response to Supplement3 to Generic Letter 09-10.

During a RCIC steam line break or leak outoide of primary containment, the 1-E51-F007 and1-E51-F008 will automatically close due to high temperature in the RCIC steam line areaand/or 300% steam flow. These valves will immediately isolate when 300% steam flow isdetected; however, isolation due to high RCIC area temperature is delayed for 30 minutesin accordance with system design. As addressed by the Reactor Building EnvironmentalReport (RBER), the most limiting case for a RCIC steam line break or laak occurs duringa RCIC small break condition, For the RBER analysis, the RCIC steam line is assumed toleak at 299% of normal steam flow for 30 minutes prior to isolation. A steam flow of 25lb/sec (1500 lb/ min.) into the reactor building is assumed for the full 30 minutes.

Comparison of the HELB leakage and the expected leakage from the subject valves at a rateproportional to that experienced during the LLRT revealed that the amount of leakage fromthe valves is not significant. The 252 SCFH at 49 paid equates to a water leakage of 3.654gpm at 1010 paid. This amount of leakage is lower than the Technical Specification limitof 5 gpm unidentified drywell leakage. Additionally, this leakage is approximately twoorders of magnitude lower than the 30 minute RCIC leak assumed in the RBER and is expectedto have an insignificant impact on the post-HELB reactor building environment.

The preceding leakage comparison analysis was performed assuming a leak rate of 252 SCFHfollowing a HELB. Although the subject valves exhibited significant leakage during theLLRT, significant leakage is not expected following a HELB. The conditions contributingto the LLRT failure of the subject valves is expected to occur only when the valves areclosed against low differential pressure. During a HELB these valves would ue expectedto close against a differential pressure of at least 1000 psid. Under these conditionsthe differential pressure assists the valve discs in achieving a good seal. Thischaracteristic was demonstrated in June of 1991 when the subject valves were found leakingagainst a low differential pressure, and yet were capable of isolation under operationalconditions.

Based on the above, the LLRT leakage from the 1-E51-F007 and 1-E51-F008 valves is notconsidered safety significant during a RCIC steamline HELB.

I

I

- I

i

. . -

NRC FORM 366A - U. S. NUCLEAR REGULATORY COMMISSION APPROVED OMB NO. 3150-0104 |(5/92) ' EXPlHES: 5/31/95 |

ESTIMATED SURDf.N PER RESPONSE TO COMPLY WITH THISINFORM ATION COL LECTION REQUEST: 50 0 HR$. FORWARD

LICENSEE EVENT REPORT (LER) COMMENT 3 R,OARom UnoEN ES nu ATE TO Ts,,N,ORu AT,0N ANo

TEXT CONTINUATION RECOROS MANAGEMENT BRANCH (MN08 7714L U.S. NUCLEARREGULATORY COMMISSION, WASHINGTON, DC 20$55,000). AND TOTHE PAdRWORK REDUCTION PROJECT (3150-01045, OFFICE OFMANAGEMENT AND BUDGET, WASHINGTON. DC 20603.

FACntlTY NAME fil DOCKET NUMBER (2) LER NUMBER (6) PAGE (3)

SEQUENT |AL REVISION

Brunswick Steam Electric Plant " " " * ' " " " " ' ' "05000325 4 of 17Unit 1

92 - 23 - 003

TEXT ilt more space ,s requise ! use additionaHIRC Form 366A 's) (1 h

LOCA:

During a LOCA, RCIC will initially inject to the vessel in an attempt to raise vessel water Ilevel. When vessel pressure drops to approximately 62 psig, RCIC will trip and the subjectvalves will close to isolate the RCIC steam line. Since the drywell will be pressurizedand a sourra Tmrm present , the subject valves will be closing for PCIS purposes. Duringa LOCA the subject valves will clcse against low ditterential pressure resulting in apotential leak rate of 252 SCFH. The off-site dose calculations in section 15.6.4 of theUpdate Final Safety Analysis Report (UFSAR) assume that primary containment leaks 0.5 jpercent of the contained free volume per day to secondary containment. This equates to.a primar y containment leakage of approximately 266 SCFH into secondary containment. When

.

the cumu.1ative LLRT leakage from the other containment penetrations is added to the 252SCFH leakage resulting from the subject valves, the total leakage would exceed that assumedin the UFJAR analysis.

A reviu of the potential leak paths out of the RCIC system was performed to understandthe ar.fety significance of this event. Leakage through the subject valves could becontaMed within the RCIC piping, released into the reactor building, or released to the jcondenser through the E51-F025 and E51-F026 valves. The most safety significant case is jthat cf a release to the condenser. Assuming no line losses, the 252 SCFH could bereleaned to the condenser and then to the environment. The General Electric Company )

performed an evaluation of the environmental impact of a 400 SCFH total Main Steam Line i

Valve leak to the Condenser. The results of this analysis indicated that an additional400 SCFH leak would not increase the off-site doses above the 10CFR100 limits. Since the252 SCFH is less than the 400 SCFH assumed in the analysis, off-site effects are notconsidered safety significant. Although this analysis did indicate off-site dose within10CFR100 limits, the control room allowable thyroid dose limits as defined by the GeneralDesign Criteria (GDC) 19 could potentially be exceeded. Based on the potential forexceeding the GDC-19 limits, this event is considered potentially safety significant, |

PREVIOUS SIMILAR EVENTS 11

|

A previous similar event involving failure of both RCIC steam isolation valves to seat isdocumented in LER 1-91-016. The cause of that event was attributed to atypicalaccumulation of corrosion products.

ETIS COMPONEUT IDENTIFICATION

Svstem/ Component EIIS Code

PCIS JMRCIC BN1-E51-F007 BN/ISV1-E51-F008 BN/ISV

- - - - ._. --

| .

NRC FORM 360A U. S. NUCLEAR REGULATORY COMMISSION APPROVED OMB NO. 3150-0104(S/92) EXPtRES: 5/31/95

ESTIMATED BUROEN PER RESPONSE TO COMPtY WITH THISINFOHMATION COLLECTION REQUEST: 50.0 HRS FORWARD.lCENSEE EVENT REPORT (LER) COMMtNis REOARoiNo URoEN EsTiuATE TO THE iNronuATiON AND

|' TEXT CONTINUATION RECOROS MANAGEMENT BRANCH (MNB8 ??)4L U.S. NUCLEARREGULATORY COMMISSION, WASH!NGTON. DC 20555 0001. AND TO

THE PAPERWORK REDUCTION PROJEC T (31500iO4h OFFICE OFMANAGEMENT AND BUDGET, WASHINGTON. DC 20bO3.

FACIUTY NAME (1) DOCKET NUMBER (2) LER NUMBER (6) PAGE (3)

SEQUENTIAL REVISIONBrunswick Steam Electric Plant " " " " * " " # "'"

05000325 5 of 17Unit 192 - 23 - 003

TENT fit nwre space is required. use additiona1NRC form 366A's) (17)

SUPPLEMENTAL INFOPMATION

Both Units have been in cold shutdown since notification of the failure of the RCIC SteamIsolation Valves 1-E51-F007 and 1-E51-F008 to meet local leak rate testing requirements.Subsequent LLRTs of Anchor / Darling double-disc gate valves have identified four Unit 1 and

,

| three Unit 2 containment isolation valves which failed to meet acceptable leakage limits.| To da t e , the defects described herein have been identified in six double-disc gate valves.

Initial NRC notification of these defects was made on November 13, 1992. Attachment Oneprovides a list of the types of deficiencies identified.

The basic components identified as containing the defects contributing to the LLRT f ailuresare:

3-inch, 900 lb Class, Double-Disc Gate Valves4-inch, 900 lb Class, Double-Disc Gate Valves6-inch, 900 lb Class, Double-Disc Gate Valves10-inch, 600 lb Class, Double-Disc Gate Valves

| Although deficiencies have been identified primarily in the 3", 4", 6", and 10" valves,I these deficiencies are considered potentially present in all Anchor / Darling double-disc

gate valve sizes.

The basic components identified as containing the defects were supplied by:

Anchor / Darling Valve Company701 First StreetPO Box 3428Williamsport, PA 17701 i

The valves listed below identify the Anchor / Darling double-disc gate valves installed atthe Brunswick Nuclear Plant (BNP). With the exception of the 1/2-E41-F001 valves, thesevalves provide a primary containment isolation function.

B21-F016/F019 Main Steam Drain Valve (3-inch)E41-F002/F003 HPCI Steam Isolation Valves (10-inch)E41-F006 HPCI Injection Valves (14-inch)E41-F001 HPCI Steam Admission Valves (10- inch)E51-F007/F008 RCIC Steam Isolation Valves (3-inch)E51-F013 RCIC Injection Valves (4-inch)G31-F001/F004 RWCU Inlet Isolation Valves (6-inch)

The following is a discussion of the double-disc gate valve deficiencies which have beenidentified as contributing to the LLRT failures. Some of the deficiencies were identifiedby inspections occurring in September of 1991 and at that time were considered isolatedoccurrences. However, based on the number and types of deficiencies identified during the,

'

recent outage and the similarity of these deficiencies to those identified in 1991 asaddressed in the Brunswick Information Report dated December 27, 1991, BNP has recognizedthe need to report these deficiencies in accordance with the requirements of 10CFR21.

i

l

j' Uneven Stanchion Lencith On The Lower Wedere

The bottom of the lower wedge has two stanchions as shown on Figure 1. Thepurpose of these stanchions ic to stop lower wedge motion during valveclosure, so that the advance of the upper wadge into the lower wedge will

| result in normal load against the in-body seats. To provide even wedge- . - -

u

NRC FORM 366A U. S. MUCLEAR REGULAT W COMMISSION APPROVED OMB NO. 3150-0104(5/92) EXPIRES: 5/31/95

ESTIM A TED BURDEN PER RESPONSE TO COMPLY WITH THISINFORMATION COLLECTION REQUEST: SO O HRS, FORWARD

LICENSEE EVENT FEPORT (LER) COMMENTS REGARDING 8URDEN ESTIMATE TO THE INFORM ADON AND

TEXT CONTINUATION RECORDS MANAGEMUf7 BRA WH (MN8B 77141, U.S. NUCLEARREGULATORY COMMISSION. WASHINGTON, DC 20555 0001. AND TOTHE PAPERWM r@UP* ION PROJECT Q150 0104L OFFICE OFMANAGF MtNT AND BUDGET. WASHINGTON. DC 20503

FACILITY NAME (1! DOCKET NUMBER (2) LER NUMBER (6) PAGE (3)

SEQUENTIAL REVISION

Brunswick Steam Electric Plant " " " " * " """"'"05000325 6 of 17

Unit 192 - 23 - 003

TEX T Uf more space is required. use addisonalNRC form 366Ns) (1 h

expansion and consequently an even seating load distribution, the stanchionsshould be the same length.

During an inspection of the 1-G31-F001 in September of 1991 and the recentinspections of the 1-E51-F008 and the 2-G31-F001 valves, the associated lowerwedge stanchions were identified to be shorter than the other stanchion. Ifnot accounted for by adjusting the lower wedge geometry, uneven stanchionlengths cause the lower wedge to cock as the valve seats. Any cocking of thelower wedge results in uneven contact with the upper wedge during seating.This uneven contact causes the wedge pieces to load predominately on only oneincline which results in uneven transmission of the stem wedging force to thediscs. This contributed to the failure of the valves to seal when subjectedto the low LLRT pressure.

Casting Flaws In The Wedge Surfaces

Inspections of the 1-E51-F008, 2-G31-F001, and 2-G31-F004 identifieddiscontinuities in the cast surf ace of the upper wedges. The 2-G31-F001 and2-G31-F004 valves were found to have a casting flaw (raised metal) on theupper portion of the upper wedge (see Figure 2). These casting flawsprevented the upstream disc from sitting flush against the machined surfaceof the upper wedge. Although the upstream disc does not provide a sealagainst differential pressure, these flaws may have prevented proper wedgingand uniform loading of both the upstream and downstream discs. Additionally,these casting flaws caused gouges on the inside edge of the upstream discs.

The 1-E51-F008 valve was found to have a sharp " lip" on the upper wedge wherethe wedge begins to taper (See Figure 3), During wedging, this lip wouldcontact a slight protrusion on the lower wedge which would cause binding ofthe two wedges. This binding could cause an uneven distribution of theseating force to the valve discs and subsequent seat leakage.

Non-uniform Contact Of The Upper And Lower Wednes

Non-ur iform contact between the upper and lower wedges has been identified onthe 1- B21- F016, 1-E51-F008, 2 - E 51 - F013 , 1 -G 31 - F0 01 (in September 1991) , 2-G31-F001 and 2-G31-F004 valves (See Figure 4). This non-uniform contact resultsin an uneven distribution of the seat force and potential seat leakage.

The non-uni orm contact between the upper and lower wedges has resulted fromseveral di f ferent causes. The mating surfaces between the upper and lower jwedges are hand-ground surfaces. This process may result in low spots or '

slightly different wedpe angles which cause uneven contact and wedging. Theuneven cantact r the 1-E51-F008 valve may have been caused by a poor lowerwedge castir ;

1

Lmgroner ' ente .q of th' Valve Stem |

During recent inspection of the 1-E41-F002 valve, the valve stem was found notto be centered in the valve when the discs were seated. This condition was |

caused by the stem connection in the upper wedge not being properly centered. |On 4-inch and larger double-disc gate valves, the stem is rigidly threaded i

into the upper wedge. With a misaligned stem, the upper wedge and stem must |deflect when the valve is seated. This condition may cause seat leakage andgalling of the valve stem on the bonnet bore. Improper valve stem centeringwas also observed to exist on the 1-E41-F001 valve.

I|

NRC FORM 366A U. S. NUCLEAR REGULATORY COMMISSION APPROVED OMB NO. 3160-0104(5/92) EXPIRES: 5/31/95 |

ESTIMATED BURDEN PEN RESPONGE TO COMPtY WIT H THl$ I

INFORMATION COLLEC TION REQUESf: SO O HRS. FORWAROUCENSEE EVENT REPORT (LER) COMuENT$ REOARo,No eURoEN esriM ATE TO THE iNrOR,.tAf TON AND

TEXT CONTINUATION RECORDS M AN AGEMENT ORANCH (MN88 7714), U S. NUCLIARREGULATORY COMMISSION. WASHINGTON. DC 20555 0001 AND TO )THE PAPERWORK REDUCTION PROJEC T 0150 0104). OFFICE OF ]

MANAGEMENT ANO BUDGET. WASHINGTON. OC 20503 !:_

i'

FACILITY NAME (O DOCKET NUMBER (2) LER NUMBER (61 PAGE (3)

SEQUENTIAL REVISION

Brunswick Steam Electric Plant " " " " " " " " ' ' "05000325 7 of 17

Unit 192 - 23 - 003

--_

TEXT fit more space is required, use additional NRC form 366A 's) (l R,

Improper Wedge Orientation

During an inspection of the 2-G31-F001 valve in September of 1991, it wasdetermined that the valve was supplied with the upper wedge installed on thedownstream side of the valve. Although the Anchor / Darling Valve Companymaintains that these valves should provide bi-directional sealing, there isa preferred wedge orientation. Since the upper wedge is rigidly connected tothe stem. the wedge cannot self-align to the seat. Therefore, any non-parallelism between the line of stem action and the seat face or misalignmentbetween the upper wedge face and the seat face cannot be accommodated.Consequently, this results in a non-uniform distribution of contact pressureand potential leakage. Since the lower wedge is free to " float", this wedgeshould always be installed on the downstream or sealing side. Theinstallation of the upper wedge on the downstream side of the 2-G31-F001 valvealso resulted in the stem not being centered as described in the precedingdiscussion of improper valve stem centering.

In addition to the aforementioned defects, BNP has identified a concern with Anchor / Darlingdouble-disc gate valves larger than 6-inches. These valves have a stellite hardfacing onthe upper and lower wedge mating surf aces. Non-uniform contact between the upper and lowerwedges has resulted in localized loading and subsequent cracking of the stellite. Althoughthis stellite cracking has not resulted in LLRT failures, it is of concern since piecesof stellite could have broken off during valve operation. Cracking of the stellitehardf acing with the associated cracks propagating into the base metal has been observedon the wedges of the 1-E41-F001 and 1-E41-F006 valves.

These manufacturing defects in the Anchor / Darling double-disc gate valves may result insignificant seat leakage when subjected to low differential pressures. Excessive seatleakage on both isolation valves for a given containment penetration or excessive leakageon one valve coincident with a single f ailure of the redundant valve could result in a loss ,

of the PCIS uafety function.

Corrective Actionn

Each double-disc gate valve installed in a PCIS application has been or will be local leakrate tested at least twice during the current outages. Since the noted defects can resultin inconsistent seating and cealing of the valves under low-pressure conditions, the secondleakage test is performed to demonstrate repeatability. Valves which fail one of theselocal leak rate tests have been or will be disassembled and repaired as necessary. Forthe defects noted in this 10CFR21 report, the repair typically involves custom grindingof the valve wedges or replacement of the wedges to achieve proper contact and uniformdistribution of the seating load. Once the necessary corrective actions are implemented,each repaired valve will receive at least two post-maintenance local leak rate tests. Thesecond leakage test is performed to demonstrate repeatability and the adequacy of therepair. BNP believes that repeat testing, whether performed pre-or-post repair, providesadequate assurance that the defects noted to date would be identified.

The corrective actions to eliminate the noted defects will be completed prior to startupof Units 1 and 2. Startup of Unit 2 is currently scheduled for March 28, 1993. Startupof Unit 1 is expected to occur in the second quarter of 1993.

Technical Support will establish the appropriate controls for ensuring that futureinspection / repair of Anchor / Darling double-disc gate valves will include an inspection forthe deficiencies noted herein.

~,

1

___m _. . . _ .

*

rNRC FORM 366A U. S. NUCLEAR REOULATORY COMMISSION APPROVED OMB NO. 3150 0104 i(5/92) EMPIRES: 5/31/95 i

ES flM A TED BURDEN PER RESPON3E TO COMPLY WITH THIS |LNF ORM A TION COLLECTION REQUESf: 50.0 HRS. FORWARD I

UCENSEE EVENT REPORT (LER) COMMEN13,EoAno,NaRURotNE T,u TEToist,NEonyArlON ,No |TEXT CONTINUATION RECOROS MANAGEMENT RRANCH (MNP8 7714), U.S. NUCLEAR }

REGULATORY COMMISSION, W A$HtNGTON, DC 20555 000 f, AND TO{

THE PAPCRWORK REDUCTION PROKCT (3160 0104L OFFICE OF jM ANAGEMENT AND BUDGET. Y ASHINGTON. DC 20503.

.

!

fFACluTY NAME (1) DOCKET NUMBER (h LER N5%BER (6) PAGE (3)

SEQUENTIAL ltEVISIONBrunswick Steam Electric Plant """" " * "'"

05000325 8 of 17Unit 192 - 23 - 003 i

!

|TEX T W more space is required. use additionalNRC Form 366A's! (11| -

|

The final root cause for the failure of these LLRT failures including the 1-E51-F007 and1-E51-F008 will be completed by February 26, 1993. A supplement to this LER reflectingthe final root cause is expected to be submitted by 3/31/93.

t

Industry Recommendation

Based on BNP's experience to date with Anchor / Darling double-disc gate valves, BNP believes,

the noted defects can result in unpredictable and unrepeatable LLRT results. With the'

noted defects present, it may be possible to successfully perform one leakage test and faila subsequent test af ter the valve has been stroked. Consequently, to ensure repeatability,it may be necessary to perform redundant leak rate tests.

Summary of Root Cause Investicration

During the current unit outages, six of ten Unit 1 and three of ten Unit 2 Anchor Darlingdouble disc gate valves f ailed to meet LLRT acceptable leakage limits. Additionally, oneUnit 2 'talve demonscrated inconsistent LLRT resW 's. The root cause of these LLRT failuresis primarily attributed to manufacturing deficiencies in the valve wedge pieces.Additionally, low spots in the seats were identified as contributing to some of thefailures.

The following provides the problems identified, the corrective actions implemented, andthe root cause of the failure for each of the Anchor Darling double-disc gate valve LLRTfailures which occurred during the current Unit 1 and Unit 2 outages:

Blit 1 Failures: |

1-B21-F016:

Tne problems identified with the 1-B21-F016 valve included minor seat pitting, low spotson the downstream in-body seat, non-uniform point of contact between the upper and lowerwedges due to rough wedge finishen, and stem galling 2 to 3 inches above the backseat.The LLRT f ailure of the 1-B21-F016 valve is attributed to the low spots in the downstreamseat ud the uneven contact between the upper and lower wedges. The uneven contact betweenthe upper and lower wedges resulted from poor vendor manufacturing practices. The lowspots in ths in-body seats are believed to have resulted from thermally induced stressesincurred dur.ng reactor heatup and cooldown. Temperature changes f rom reactor heatup andcooldown may 1 ave provided some stress relief resulting in minor changes to the valve seatgeometry, i.e., low spots. The stress relief phenomenon represents a hypothesis of howthe low spots were formed. This hypothesis has not been confirmed by actual insitu orprototype toscing. Corrective actions taken include replacement of the galled stem andboth the upstream and downstream discs. The in-body seats were lightly lapped andnarrowed. The wedges were custom ground to achieve even contact. Two post maintenancelocal leak rate tests were successfully performed prior to returning the valve to service.

1-E41-F002:

The problems identified with the 1-E41-F002 valve included low spots in the downstream in-body seat, wedge centerline fit-up problems resulting in a non-centered stem in the bonnetbore, and spider web cracks in the upstream disc stellite overlay. The LLRT failure ofthe 1-E41-F002 is attributed to the low spots in downstream in-body seat. The cause ofthe low spots is the same as that addressed in the preceding paragraph. The impropercentering of the valve stem in the bonnet bore is a contributing f actor in the LLRT f ailureof this valve. The improper centering of the valve stem is due to a manufacturing

- ,

.

NRC FORM 366A U. S. NUCLEAR REGULATORY COMMISSION APPROVED OMB NO. '3150-0104(5/9 21 EXPlRES: 5/31/95-

ES TIMATED BURDEN PER RESPONSE TO COMPLY WITH THl$INFORM ATION COLLECT!ON HEQUES h $O 0 HR$ FORWARDLICENSEE EVENT REPORT (LER) COMMENTS RMARDINO SURDEN ESDMW TO THE INFORMAh0N AND

TEXT CONTINUATION RECORDS MAN AGEMEN T BRANCH WN80 m 41. U S. NUCLEARREGUL.ATORY COMMISSION, WASHINGTON, DC 20596 0001. AND TO

THE PAPEHWORK RFOUC TION PROJECT (3150 0104). OFFICE OFMANAGEMENT ANO BUDOET, WASHING 10N, DC 20603.

FACluTY NAME (O DOCKET NUMBER ( LER NUMBER 16) PAGE (3}

SEQUf N fl AL Rf. VISION

Brunswick Steam Electric Plant """ """ ~

05000325 9 of 17Unit 192 - 23 - 003

TEX t fit more space is required. use additionaltRC Form 366A's) (17)

deficiency resulting in misalignment of the valve body seats and the valve stem with thevalve center line. The cracking of the upstream disc stellite overlay is not considereda contributor ta the LLRT failure of this valve. Although a definitive reason for thecracking has nC t been established, recent examination supports the hypothesis that thecracking could aave been caused by overstressing of the discs into the seats or thermalshock. Correceive actions taken include lapping of the upstream and downstream seats toremove the lw spots. The cracked disc on the upstream side of the valve was replaced.To correctly center the valve stem the valva discs were switched. Two post maintenancelocal leak rate tests were successf ully performed prior to returning the valve to service.

1-E41-F003:

The problems idantified with the 1-E41-F003 valve included uneven contact between the uppera i lower wedges and cracking of the upatteam valve disc stellite overlay. The LLRTfn lure of the 1-E41-F003 valve is attributed to the uneven contact of the upper and lowerwadges. This uneven contact resulted in the tilting of the upper wedge and valve stem andcalling of the valve stem. Tilting of the upper wedge is believed to have caused the lowerweage to tilt during valve seating resulting in a loading of the wedge pieces on oneincline. This resulted in a non-uniform transmission of the seating force to the dic:csand, consequently, seat leakage. The cause of the stellite overlay cracking is the sameas that addressed in the preceding paragraph. Corrective actions for this valve are notcomplete. The procurement of new machined wedges to eliminate the uneven wedge contactis in progress. The cracked upstream disc will be replaced. The valve will be local leakrate tested at least twice prior to returning the valve to nervice.

1-E51-F008:

The problems identified with the 1-E51-F008 valve included a bent stem, the existence ofa sharp lip on the upper wedge, a skewed and slightly twisted lower wedge, and uneven lowerwedge stanchions. The LLRT failure of the 1-E51-F008 is attributed to the deficiencieswith the upper and lower wedges. The sharp lip on the upper wedge is believed to havecaused the upper wedge to bind with the lower wedge during seating. This binding resultedin an uneven distribution of the valve disc seating force and the subsequent leakage. Theskewed and twisted lower wedge and the lower wedge uneven stanchion lengths is believedto have resulted in an uneven distribution of the seating force and subsequent seatleakage. Corrective actions included replacement of the valve stem and wedges. The discswere machined and polished to eliminate the sharp edges. Two post maintenance local leakrate tests were successfully performed prior to returning the valve to service.

1-E51-FOO7/1-B21-F019:

Disassembly and investigation into the cause of the LLRT failures of the 1-E51-F007 and1-B21-F019 valves have not been performed. These valves will be disassembled and inspectedto support a root cause determination prior to the startup of Unit 1. The results of thisdetermination will be reported in a supplement to this LER.

Unit 2 Failures;

2-E51-F013:

The LLRT failure of the 2-E51-F013 valve is attributed to the uneven contact between theupper and lower wedges resulting in tilting of the lower wedge during seating. The unevencontact was caused by a shallower wedge angle on one side of the upper wedge. The tiltingresulted in the non-uniform transmission of the seating force to the discs and the

-.

|.

NRC FORM 306A U. S. NUCLEAR REGULATORY COMMISSION APPROVED OMB NO. 3150-0104i5/92) EMPIRES: 5/31/95

IST tM ATED BURDEN PER RESPONSE TO COMPLY WITH THIS |INFORM A TION COLUCTION REQUES'T; 50 ? HRS FORWARD |

LICENSEE EVENT REPORT (LER) COMMENn agcARoiNG EURDEN ESTIMAM TO haNFORM ADON AND |TEXT CONTINUATION RECORDS M AN AGEMEN T BRANCH iMNBR M t 4L U S. NUCLFAR I

REGULATOM COMMISSION WASHINGTON. DC 20555 0001, AND TO

THE PAPE RWORK REDUCTION PROJECT (3150 0104L OFFICE OFM ANAGEMEN'T AND OUOGET. WASHINGTON OC 20501

FACIUTY NAME IO DOCKET NUMBER (2) LER NUMBER (0) PAGE (3) j1

SEQUENTIAL REVi$ TON

Brunswick Steam Electric plant " # "*" " * "#"0500032S 10 of 17 |Unit 1 1

92 - 23 - 003 ;

|

TEXT fit rmue wace is re<suved, use additionalNRC form 3tMA 's] O h

inconsistent seating of the valve. Corrective actions to achieve uniform contact included i'

replacement of the original wedges with new custom fit wedges. Two post maintenance localleak rate tests were successfully performed prior to returning the valve to service.

I2-G31-F001: i

The problems identified with the 2-G31-F001 valve included downstream in-body seat low'spots, upper wedge casting flaws, and uneven stanchions on the bottom of the lower wedges.The 2-G31-F001 valve LLRT f ailure is attributed to the low spots in the downstream in-bodyseat. The cause of the low spots is addressed in the preceding discussion of the 1-B21-F016. Additional factors contributing to the 2-G31-F001 valve LLRT failure include theupper wedge casting flaws and the uneven lower wedge stanchion lengths. The effects ofthe uneven lower wedge stanchion lengths is addressed in the preceding 1-E51-F008 valve )discussion. The upper wedge casting flaws prevented the upstream disc from sitting flushagainst the machined surface of the upper wedge. Although the upstream disc does notprovide a seal against dif ferential pressure, these flaws may have prevented proper wedgingand uniform loading of the upntream and downstream discs. The casting flaws and unevenlower wedge stanchion lengths would result in inevnsictent wa t inct of the valve.Corrective actions included polishing of the dices and lapping of the downstream in-budyseat to remove the low spots, replacing the upstream disc damaged by the casting flaws onthe upstream wedge, and replacing and custom fitting the upper and lower wedges. Two postmaintenance local leak rate tests of the 2-G31-F001 were successfully performed prior toreturning the valve to service.

2-G31-F004:

The problems identified with the 7-G31-F004 valve included a galled stem above thebackseat, upper wedge casting flaws, and uneven contact between the upper and lower wedges.The LLRT failure of the 2-G31-F004 valve is attributed to the upper wedge casting flawsand the uneven contact between the upper and lower wedges. The ef fects of the upper wedgecasting flaws is discussed in the preceding paragraph. The effects of the uneven contactare addressed in the discussion of the 1-E51-F003. Corrective actions included replacementof the upstream disc, blending of the stem galling, and the custom grinding of the upperand lower wedges to achieve uniform contact. Two post maintenance local leak rate testswere successfully performed prior to returning the valves to service.

2-821-F019:

During initial local leak rate testing of the 2--B21-F019 valvo, the valve passed twoconsecutive local leak rate tests. However, the results of these tests were not consistentand consequently the valve received an internal inspection and root cause investigation.The problems identified with the 2-B21-F019 valve included poor mating of the wedgesresulting in the rocking of the lower wedge. The rocking is attributed to differences inthe thicknesses of the wedge sliding surfaces. This condition resulted in the unevendistribution of forces at the seata and can resulted in inconsistent local leak rate t estresults. The corrective actions included widening of the seats on the downstream matingsurface and the dressing of the upper wedge sliding surface until the top-to-footdimensions were equal.

Additional Corrective Actions:

During the investigation to determine the cause of the deficiencies contributing to the iLLRT failures of the Anchor-Darling doubie disc gate valves installed at BNP, the vendor Ihas assisted in the repair and trouble shooting of the defective valves. Based on the ;

problemn witnessed by the vendor, the vendor recognizes that their manuf acturing techniques |

|

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

NRC FORM 36SA U. S. NUCLEAR REGULATORY COMMISSION APPROVED OMB NO. 3150-0104'

(5/92) EXPIRES: 5/31195f $ flM ATED UURDEN PER fMSPONSE TO COMPLY WITH THISINFORM A TION COLUCTION REQUEST: 90 0 HRS. FORW ARD

LICENSEE EVENT REPORT (LER) COMuENrs o,OuOmumN EsTmn TO TNow0nM4riouMO

TEXT CONTINUATION REC RDS M AN AGCMENT BAANCH (MNtl0 7714L US NUCLEARREGUL ATORY COMMISSION W ASHINGTON, DC 20555 000t. AND TO

THE PAPERWORK REOUCitON PftOJEC T 0160 0104L OFFICE OFMANAGEMENT ANO BUDGET. W ASHtNGTON. DC 20503.

. _ _ _ . --

.-

FACluTY NAME (O DOCKET NUMBER (2) LER NUMBER (6) PAGE (3)

SEQUENYIAL REVtSION

Brunswick Steam Electric Plant " ' " " ' " ""'"05000325 11 of 17

Unit 192 - 23 - 003

TEXT ttf more space is requked, use additionalNRC form 366Nsl (17)

warrant improvement. The vendor is currently aesessing changes to the manufacturingprocess which could prevent recurrence of those problems experienced at BNP. An exampleof this is evidenced by the recent change in the manufacture of the upper and lower wedges.In the past, Anchor-Darling has custom fit each disc by hand grinding the mating surfacesof the upper and lower wedges. This process has resulted in inconsistent seating when thevalves are closed. Anchor / Darling has recently changed its manufacturing process such thatthe wedge mating surf ace are machined instead of hand ground. This enhancement shouldresult in more reliable valve operation. To ensure that future wedge replacements utilizethe enhanced wedges, BNP is currently procuring the new wedges.

To ensure cont inued reliability of the Anchor / Darling double disc gate valves, each doubledisc gate valve uned to satisfy a PCIS function will receive two local leak rate testsduring the next Unit 1 and Unit 2 refuel outages.

The repair of the 1-E41-F003 and the disassembly, inspection, and repair of the 1-E51-F007and 1-B21-F019 will be completed prior to Unit 1 startup.

The final supplement to this LER will be submitted following the Unit 1 startup to addressthe results of the inspection and repair of the 1-E51-F007 and 1-B21-F019 valves.

SUPPLEMENT THREE INFORMATION

This supplemental information provides the results of the analysis performed to identifythe cause of the 1-E51-F007 and 1-821-F019 leakage identified during B109R1 local leak ratetesting and the corrective actions taken.

1-B21-F019:

Innpection of the 1-B21-F019 valve internals identified uneven lower wedge stanchionlengths which can cause the wedge mating surfaces to load predominately on one inclineresulting in a non-uniform transmission of the coating force to the discs. Unevendistribution of the seating force can result in significant seat leakage at lowdifferential pressures. Additionally, the wedge mating surfaces were found rough.

To correct the cause of the 1-B21-F019 LLRT failure the rough spots on the wedge matingsurfaces were resurfaced to achieve an even profile and the length of the lower wedgestanchion was evened to ensure unifonn wedge contact. Two post-maintenance LLRTs weresuccessfully performed on the 1-B21-F019 valve.

1-E51-F007:

During the B109R1 Unit 1 refuel outage, the RCIC Steam Line Inboard Isolation Valve,1-E51-F007, failed local leak rate testing on two occasions. Although extensive inspection andtesting was performed to determine the cause of the LLRT failures, a conclusive reason forthe cause of the failures was not determined. The following factors are consideredpotential contributors to the failures of the valve.

After the initial 1-E51-F007 LLRT failure, the valve was disassembled for inspection.During the disassembly the valve bonnet pressure seal ganket dropped into the valve bodywhen the last bonnet retainer capscrew was removed. This is unusual for a pressure sealvalve of this type. The loose bonnet pressure seal may have resulted from loose bonnetretaine.c capscrews. The loose capscrews could have caused the bonnet to not be fully'

seated on the pressure seal gasket. With test pressure (50 psig) applied to the valveinternals the resultant upward force on the bonnet would be approximately 600 pounds whichis approximately the same load expected from the packing for this size valve. Since the

- _ _ _ , .- , __ ..

-

iD i

NRC FORM 366A U. S. NUCLEAR REGU:.ATORY COMMISSION APPROVED OMB NO. 3150-0104 |(5!92) EXPIRES: 5/31/95 i

EsilMATED BURDEN PE R RESPONS$ TO COMPLY WITH THl3INFORMATION COLLECTION RE QUE S T: 50 0 HRS. FORWARD

LICENSEE EVENT REPORT (LER) COu-rs REcARoiNO euRoEN Esw.uTE rO TnE ,mORMATION AND

TEXT CONTINUATION RECORos uANacEusNT eRANCs cuNea 77i4>. s,.s NUCLEAR l|REGULATORY COMMiS$10N, WASHINGTON, DC 20655-000), AND Tb

THE PAPERWORK REOUCTION PROJECT (3150 0104). OFFICE MMAN AGEMENT AND BUDGET, WA$mNGTON. OC 20503,

FACluTY NAME W ' DOCXET NUMBER (2) LER NUMBER (6) PAGE (3)

SEQUENilAL HEVIS:ON

Brunswick Steam Electric Plant " " " " " " ' " " ' "05000325 12 of 17

Unit 192 - 23 - 003

_.

Tt M T Ut more space is required, use additkws! MC form 366A 's) (1 h

initial LLRT results indicated the valve would not pressurize, the internal valve pressurewould have been lower than the required 50 psig test pressure and the upward force on the !

bonnet would have probably been less than the packing load. Consequently, with a gapbetween the bonnet and pressure seal gasket, the packing resistance could have preventedthe test pressure from seating the bonnet resulting in a leakage path. This leakage pathwould probably not have existed during normal operation since the normal line pressure ofapproximately 1005 psig would have creat.ed an upward fort e on the bonnet in excess of thepacking load.

The inspection performed at this time revealed no additional problems which could havecontributed to the LLRT failure. As a result of the wedge problems encountered on otherdouble-disc gate valves (af ter 1-E51-F007 was reassembled) , the 1-E51-F007 was disassembleda second time and the wedges inspected. Although the wedge contact surfaces were foundto be rough, the surface roughness was not considered significant enough to have causedthe valve to fail the LLRT. Following resurfacing of the wedge surfaces the valve wasreassembled and retected. The valve failed to pressurize during the post maintenancetesting. The valve was disassembled and inspected a third time. The cause of the secondLLRT failure was not identified. While debris in the seats may have caused the secondfailure no significant material was found in the valve body during the final inspection.A series of tests to ensure proper mating of the wedges and sealing of the discs wasperformed. Upon final reassembly the valve passed four consecutive informational LLRTsand two official LLRTs.

'l-t g

.

.

no _

NRC FORM 366A U. S. NUCLEAR REGULATORY COMMISSION APPROVED OMB NO, 3150-0104

(5/92) EXPlRES: 5/31/95ES TIMA T ED BURDEN KR RESPONSE TO COMPLY WITH THl3INFORM A TION Collf CiiON REQUEST: 60 0 HR$ F ORWAHO

LICENSEE EVENT REPORT (LER) COuytN,S REGARoiNO euRocN csriuAre rO tne woHuAriON ANo

TEXT CONTINUATION RECOROS M ANAGE MENT ORANCH IMN80 77141, US NUCl.lARNfGULA FORY COMMfSSION, WASHINGTON. DC 20S$6 0001, AND TOTHF PAPE RWORK REDUCTION PROJECr (3150 0104), OrriCf OFM ANAGEMEN T AND BUDGEI, WASHINGTON DC 20$03

.,_

F ACILITY NAME (1) DOCKET NUMBER (2) LER NUMBER (6) PAGE (31

$EOVENTIAL REVISIONBrunowick Steam Electric plant " # ''" " # 8'"

05000325 13 of 17Unit 192 - 23 - 003

iExi ut more space is regeed.1,se adda,ont NRC fwm 366KsI (1 h

ATTACIIMENT ONE

Summary of Deficiencies Identified Per Valve

(This table identifier the double disc gate valce aeficmacies identified to dateeither during the current outage or daring previous retuel outages.)

Valve Deficiencies Identified

3-B21-F016 3,7,8

1-B21-F019 1,7 |

1 C41 F001 4,6,9

1-E41-F002 4, 8, 10l

1-E41-F003 3, 10 ||

1-E41-F006 6

|||1-E51-F007 7

1-E51-F008 1,2,3,7 i

|1-G31-F001 1,3 ;

12-B21-F016 7 |

2-B21-F019 7

2-E51-F007 7

2-E51-F008 7

2-E51-F013 3

2-G31-F001 1,2,5,8

2-G31-F004 2,3

Deficiencien

1 - Uneven Stanchion Length on Lower Wedge2 - Casting Flaws in Wedge Surfaces3 - Non-uniform Contact of the Upper and Lower Wedges4 - Improper Centering of the Valve Stem5 - Improper Wedge Orientation6 - Cracked Ste111te on Wedges7 - Sharp Edges on Discs8 - Low Spots in Seats9 - Disc Scraping Body Casting10 - Stellite seat face cracking

- - _.. -. _

_-_ ,

. . . _ , ___ _ . . _ _ . _ . _ _ _ _ . . - _

*NRC f,ORM 366A U. S. NUCLLAR REGULATORY COMMISSION APPROVED OMB NO. 3150-0104,

,

15'9 2) EMPIRES: 5!31/95 '

(SYlM Af f D PVHDt N Pf N Mi$fGN5| Th (OMP(V WIT H T HIS.

' ' ' * " " ' ' * '*"' * * "'"'' ' "*' ' "***"LICENSEE EVENT REPORT (LER) COMME NTS ptGAROtNG pumDEN ESTillAf f TO Mf INF 1V4 ?'04 AND

TEXT CONTINUATION af cORos M ANAGEMENI SRANCH i'4N88 W in D UCd ARatGULATORY COMM.'SSION W ASHiNGTON. D: . C' h ' Q 's, AND TOTHE P apt RWORA RLDUCf rON PROJECT O MC E' .04. oF F KE OFM ANAGEMENT AND BUDGET WASHINGTON DC 20bvJ

F ACtLITY NAME (11 OOCKET NUMBER (2) LER NUMBER (61 PAGE(31

SL OVE N TIAL REvisONBrunswick Steam Electric Plant """ " * *'"05000325 14 of 17Unit 1

92 - 23 - 003

TEKT lit more soste is requeed. use additaonalNRC form 366A 'st (17)

Lower Wedge

r ,

~.

I

1

!

i<

,

.

Uneven Stanchianv iLength on LowerWedge

!r > 1

y

1

4

l

|

1* e

i

l.

1

I

!

|I|

Figure 1 |

|J

. _ . .

hRC$OHP.i366A U S. NUCLL AH HEGUL ATORY COMMIS$102 APPHOV[D OMB NO. 3150-0104'

'tb 9 p ' EXPIRES: 5!31/95t e.f rua tt o pumwN es a ni smNs to cooni, wnH t oi ,

IN6 OHM 41 K)N LOttiC f TON M( UUt 5 ? 60 0 HHb 4 0tiv% AMD

LICE'NSEE EVENT REPORT (LER) Cooo,N,s m Gano,Nu euHn N isi,o Ai,40 i,. ,Ooo Ai,0N ann

TEXT CONTINUATION *( "5 *^N'G'"'N' 8"*NC" m an nu,. us NuCti an

RtGUL A10nv COMhlintON W ASHINGTON. DC 20%b 0001 ANO 10T HE P apt f6WOnt R(DUC flON PHOJfC) (3 tbo 0104!. Of f 6Ct OF

M AN AGE MINT ANr) BUDG[1, W ASHsNGlON DC 20503

FACILITY NAME (t) DOCKET NUMBER 17) ( LER NUMBER (6) PAGE (3)

$( QUt NTI AL Rf. VISION

Brunswick Steam Electric Plant """ " """ "05000325 15 of 17

Unit 1 92 - 23 - 0031~~"

TE K1 111 more wace a rettuned, use acidut sm*I NRC form 366A 's! (1 h

Upper Wedge

mt

.

Casting Flaws ', Castina Flawsnon Upper Wedge 'V on Upper Wedge ;

!.

en

0

\

NN

./a

Figure 2

- - - -. . -.

.

. INHC 10%" 366 A U S. NUCLE AR HE GUL ATOHY COMMIS$10N APPHOVED OMB NO. 31bo 0104lb 9 21 ' EXPIAES: 5!31/%

isi.uAim ,,,,n N ri o m sro~. , n, c oM,.o W,i n i n,s.

W8 DMM A f t04 ( Ot t| CilON Mt OU( hl 40 0 H5th 6 0MW AHO

LICENSEE EVENT REPORT (LER) Couoi Nis o,GAno,Nu .o m, niiu A,, w im ,N,0 MA,c ann

TEXT CONTINUATION "'CodOS **NAGt*'N' '"'NCH '**"" '"'' us NUcit anMIGut ATORY COMMi$$10N W A$H1NGT ON DC 205% 0001- AND 70T HL P APE RWQH 4 REDUC 160N P OJf C1 0150($1046 Off tCE OsM AN AGf MIN 1 AND BUDGf f WASHtNG10N DC 2050.)

FACILITY NAMI 111 DOCEP fv0MBER (21 LEH NUMilER (6) PACE (3)

St OUt NilAL Rf Vr360N

Drunswick St u.m Electric Plant """""05000325 " ' " " " . 16 of 17Un1L 1

92 - 23 - 003 | _.TE 51 W more space is recturred, use addirronal NHC f orr" 36tiA 'si (17)

Sharp Edge onUppe. Wed0cMating Surface

Upper Wedge

1

i} f | !,F '

, , x| | ! N' r

,' : !N', -

l 'I \<

/I

Figure 3

- _____ --__ _

m w -

._ . . - . . . .

-. , --- -- -. - - - --_

INHC)ORY 366A U S. NUCLE AR REGUL ATORY COMMISSION APPROVED OMB NO. 3150 0100.

ps 92i tMPlHES: Sf31/95. s, A,i n m,.0, N . , ,.o~s, , o < n A.I, , W,H , Hi

' " ' " " ' ' * ' '""'#'** " ' * " ' "" ""' " ' " * ' " ' '

LICENSEE EVENT REPORT (LER) COMML NTS Rt(a AMOINie leUHOt h (Kl eM Ali IO T Ht INI OHM AllON ANL)

TEXT CONTINUATION "'(0"05 *^N*cd*'N' '" 'Nc " ''' N"' n u, os NuctAnRE GUL ATORY COMMIS$ TON. W ASHINGTON DC 705% 000) AND TOTHC P apt HWOHn REDUCTION PROJ[Cl {31 $0 010+ OM ICI OFM AN AG(Mt NT AND BUDGf f, W ASHINGTON DC 20$03

_ _ .

F ACill1Y NAME til DOCKET NUMBER (p LIR NUMBER (61 PAGE 01

SE GUI N TIAL REVISION

Brunswick St.eam Electric Plant "" " "" "05000325 17 Of 17

Unit 1 92 - 23 - 003

st x s or n,,,,e w.,ee os reases, use nas,rnmot tenc f orm 366ks! i t n

Uppur WedgeLower WedOO

r ,

)r ,

O

.

Uncven Contacton Lowet WedD0

./ \

4 / .

/tW,

< >

/[ \ 3r 7 l

kn m|

.. 1

1

COntDCtIWedge|

Figure 4

i

.

.

|1

_ _ _ _ _ _ _ _ ._ _ _. _.