ML20044F132

LER 93-009-01:on 930217,TS 3.0.3 Entered Due to Potentially Undersized Fuses in Ssps.Caused by Random Age Related Failure of Ssps Fuse.Fuses in Selected Equipment Replaced During Fuse & Breaker Field Verification Effort

ML20044F132
Person / Time
Site:	South Texas
Issue date:	05/20/1993
From:	Pinzon J HOUSTON LIGHTING & POWER CO.
To:
Shared Package
ML20044F112	List: ML20044F111 ML20044F132
References
LER-93-009, LER-93-9, NUDOCS 9305260483
Download: ML20044F132 (5)

Text

f NRC FORG 366 U.S. OfCLEAE REGULAT(DY (IPMISSi(O ADR3v1D BT CDS NO. 3950-0104 (5 92) . EXPIRES 5/32/95 ESTIMATED BURDEN PER RESPokSE TO COMPLY KITH r6 tim tECitw R n:

) LICENSEE EVENT REPORT (LER) $5g9 [ cgRct g cw( tgNR c .,

THE 1kFORMATIDW AND RECORDS MAkAGEMENT BRANCH (MkEB 7714), U.S. WUCLE AR REGULATORY COMMISSIDW, (See reverse for regaired neber of digits / characters for each block) 6.A S H I NGTON , DC 20555-0001, AND TO THE PAPERWDRK j REDUCTION PROJECT (3150-0104), OFFICE Of i a MAkAGEMENT AND BUDGET, WASH 1WGTOW, DC 20503. j F ACILITY hAE (1) DOCKET EMBER (2) PAGE (3)

South Texas Unit 1 05000 498 1 OF 5 1

TITLE (4) Technical Specification 3.0.3 Entry Due to Potentially Undersized Punon in tho solid stato Prntor tinn svntom EVENT DATE (5) LER K MP.ER (6) REPORT DATE (7) OTHER F ACilITIES INVOLVID (B)

SEQUEWTIAL REVISION FACILITY KAME DOCKET NUMBER MOWTN DAY YEAR TEAR TH DAY TEAR South Texas, Unit 2 NJMBER NUMBER 05000-499

1. ^ "" '"I' 02 17 93 93 009 01 05 20 93 ["

OPE RAT !WG Tuts RrPoRT 15 SusMITir PuesuAwT T Tut RE aterMEw15 or to Cre 5: (Check one or more) (11)

ICDE (9) 5 20.402(b) 20.405(c) 50.73(a)(2)(i v) 73.71(b)

PCMER 20.405(a)(1)(i) 50.36(c)(1) 50.73(a)(2)(v) 73.71(c)

LEVEL (10) 0 20.405(a)(1)(ii) 50.36(c)(2) 50.73(a)(2)(vii) OTHER 20.405(a)(1)(iii) X 50.73(a)(2)(i) 50.73(a)(2)(vi i i)( A) (Specify in

^ #

20.405(a)(1)(iv) 50.73(a)(2)(ii) 50.73(a)(2)(viii)(B) 1 3

20.405(a)(1)(v) 50.73(a)(2)(iii) 50.73(a)(2)(m) NRC Form 366A) tICENSEE CIniTACT FOR THIS LER (12)

DAME TELEPHohE WUMBER (include Area Code)

Jairo Pinzon - Senior Engineer (512) 972-8027 COMPLETE ONE t lNE FOR EACH COMPOWENT F AltuRE DESCRIBED IN THIS REPORT (13)

CAUSE SYSTEM COMPOWENT MANUFACTURER p CAUSE SYSTEM COMPOWEkT MANUFACTURER p SUPPL E MENT AL RE PORT E XPE CTED (14) EXPECTED " I" UA* "##

7gg SUBMISSION (If yes, conplete EXPECTED SUBMISSION DATE). WO X DATE (15)

ABSTRACT (Limit to 1400 spaces, i.e., apprcximately 15 single-spaced typewritten lines) (16)

On February 17, 1993, Unit 1 was in Mode 5 and Unit 2 was in Mode 4, both at 0% power. Plant personnel determined that a potentially unanalyzed l '

a condition existed in both units related to a failed fuse event that occurred on February 13, 1993. This unanalyzed condition involved suspected undersized fuses found in the Solid State Protection System g (SSPS) in which an inrush current could cause the fuses to fail and prevent the fulfillment of the SSPS's intended safety function. In response to the inoperable SSPS actuation cabinets, Unit 2 entered Technical Specification

Section 3.0.3 and plant cooldown to Mode 5 was initiated at 1030 on

February 17, 1993. Unit 1 was already in Mode 5. The event was caused by j the random age related failure of a SSPS fuse. In response to the event, 4

the 10 amp fuses were replaced with 20 arp fuses in both Units, other 120 volt vital A.C. distribution and class 1E DC circuit panels were reviewed for similar conditions, field verification of selected protective devices .

was conducted, and a failure analysis of the blown fuse was performed by an i independent laboratory and reviewed by the fuse manufacturer. ,

Additionally, lessons learned will be formally factored into the design process.

i NRC FORM 366 (5-92) 93052604B3 930520 PDR ADOCK 05000498 S PDR

NRC FORM 366A U.S. BUCLEAR REGULATORY CD MISSIOt APPROVED Bf 05 00. 3150-0104 (5-02) EXPIRES 5/31/95

. ESTIMATED BURDEN PER RESPONSE TO COMPLV UITH TN!S 1hFDEMATION COLLECTION REQUEST: 50.0 HRS.

FORWARD COMMENTS RE(.ARDlWG BURDEN ESilMATE TO THE LICENBEE EVENT REPORT (LER) IwFORMATIOk AND RECORDS MAkAGEMENT BEAbCH fMkBB TEXT CONTINUATION m 4), u.S. NUCLEAR RE GUL ATORY COMMISSION, WASH 1kGTON, DC 20555-0001 AND TO THE PAPER.0FtK REDUCTION PROJECT (3150-0104), OFFICE OF MAkAGEMENT AND BUDGET, WASHINGTOW, DC 205C3, e ! F ACILITY KW (1) DOCLTT tLMBER (?) I LER NtmBER (69 PAT (3)

YEAR SEQUE hTI AL REVISION South Texas, Unit 1 05000 498 2 OF 5 93 009 01 TEri tif core space is reosired. use additionet copies of WRC Form 366A) (17)

DESCRIPTION OF EVENT:

On February 17, 1993, Unit 1 was in Mode 5 and Unit 2 was in Mode 4, both at 0% power. Plant personnel determined that an unanalyzed condition existed in both units related to a failed fuse event that occurred on February 13, 1993. This unanalyzed condition involved potentially ]

undersized fuses found in the Solid State Protection System (SSPS) in which an inrush current could possibly cause the fuses to fail and prevent the fulfillment of SSPS's intended safety function. [

On February 13, 1993, while performing SSPS Train S reactor trip breaker trip actuating device operational surveillance test in Unit 1, the power was lost to the Train A SSPS actuation cabinet. It was subsequently determined that a 10 amp fuse in the electrical distribution panel EDP 1201 which feeds the Train A SSPS actuation cabinet failed. The unit was in Mode 5 at 0% power at the time of the failure. Testing was suspended and the event was referred to engineering for investigation. Engineering concluded that the fuse, sized for steady state current conditions, may have been undersized based on inrush current. The review determined that Westinghouse had provided 20 amp fuses in the SSPS actuation cabinet but the fuses in the electrical distribution panel feeding this cabinet had been sized at 10 amps. The 10 amp fuses were also installed in Unit 2.

. An initial operability review was performed to determine the impact of l having 10 amp fuses feeding the SSPS actuation cabinet. The initial results concluded that all three SSPS actuation trains were inoperable, and as a result, Unit 2 entered Technical Specification 3.0.3. At 1030 on February 17, 1993, plant cooldown to Mode 5 was initiated in Unit 2. Unit 1 was already in Mode 5 so entry into Technical Specification 3.0.3 was not required. Concurrently, plant change forms were initiated to revise the fuse size from 10 amp to 20 amp. Unit 2 exited Technical Specification 3.0.3 when the 10 amp fuses were replaced with 20 amp fuses.

CAUSE OF EVENT: t

, The cause of this event was a random age related failure of the 10 amp fuse in the electrical distribution panel feeding Train A SSPS actuation cabinet.

ANALYSIS OF EVENT:

1 The SSPS contains three trains of Engincered Safety Features (ESP) actuation cabinets which actuate various ESF equipment via relays providing 1 protection to mitigate the consequences of postulated accidents. When the j correct logic requirements are met, master relays are energized which in turn energize a set of slave relays that operate the various ESF components.

I l

I

NRC FORM 366A U.S. KACLEAR REGJLATORY CDWISSION APPROVD BY OMB NO. 3150-0104 (5-92) EXPIRES 5/31/95 ESTIMATED BURDEN PER RESPONSE TO COMPLY ti1TH THIS IhTORMATION COLLECTION REQUE ST: 50.0 HRS.

' LICENSEE EVENT REPORT (LER) NTEAN RE$RD "NGEMENT BR NCH 7714), U.S. NUCtEAR REcuLATORT CcamlS$10W, TEXT CONTINUATION WASHINGTON, DC 20555-0001 AND TO THE PAPEPWORK REDUCTION PROJECT (3150-0104), OFFICE OF MAhACEME NT AND BUDGET, WASHINGTON, DC 20503.

F ACitITY KME (1) DOCKET EMBir (?) LER NUMRER (6'i PAM (3)

YEAR SEQUENTI AL REVISION South Texas, Unit 1 05000 498 3 OF 5 93 009 01 T0ri (If more space is reasired, use additionet copies of ARC Form 366A) (17)

ANALYSIS OF EVENT: (Cont'd)

The initial evaluation of the design was based on information available from Westinghouse (the actuation cabinets' vendor) that indicated a momentary inrush current of 46.5 amps may be experienced. Based on this, and the fact that Westinghouse had furnished 20 amp fuses in the SSPS actuation cabinets, it was concluded that the installed 10 amp fast acting fuses in the distribution panel feeding this equipment may blow on energization of the maximum number of relays during Modes 1 through 4.

Therefore, the SSPS actuation cabinets were conservatively declared inoperable in these modes.

Further evaluation of the design determined that the worst case accident scenario was a main steam line break, which would initiate the slave relays associated with steam line isolation and safety injection. This condition energizes 45 relays (43 latching and 2 non-latching) in the Train A or B actuation cabinet; 33 (30 latching and 3 non-latching) in the Train C actuation cabinet. Calculated maximum circuit currents during relay inrush for these cabinets is less than the published time-current characteristic for 10 amp fuses (Gould Shawmut type A60X10). Therefore, the SSPS actuation cabinets were, in fact, operable with the originally installed (unblown) 10 amp fuses.

It should be noted that the above evaluation and conclusion was based on Gould Shawmut product information for their A60X10 fuses which indicates an average melting current versus time value of approximately 29 amps at 10 msec. Recently received product information for these same fuses shows an average melting current versus time value of approximately 69 amps at 10 msec. Gould Shawmut attributes this change to improved equipment and techniques used in testing and developing fuse time-current characteristic curves. The results of this improved technology are particularly apparent at short time (millisecond) values. This curve significantly increases the SSPS fuse application design margin and supports that the SSPS actuation cabinets were operable with the originally installed (unblown) 10 amp fuses.

In addition to the design evaluation, Southwest Research Institute (SwRI),

an independent laboratory, was contracted to evaluate the blown fuse in order to determine the cause of failure, if possible. During this i evaluation SwRI examined the blown fuse, several unblown companion fuses and a new fuse. SwRI concluded:

o The blown fuse did not open as a result of a high current fault. ,

o It was not possible to determine whether the blown fuse link had a defect  !

that caused it to open. '

o Thermal damage to unopened links in both the blown fuse and the unblown i companion fuses indicate that they had been subjected to greater than I rated current.

1

NRC FORM 366A U.S. IEJCLEAR REGULATORY CGotIESION APPROVED BY 0 e No. 3150-0104 (5 92) EXPIRES 5/31/95

. ESTIMATED B'IDEN PER RESPONSE TO COMPLY If!TH THIS INFORMATION COLLECTION REQUEST: 50.0 HRS.

.

• FORWARD COMMENTS REGAPD]NG BURDEN ESTIMATE TO THE LICENSEE EVENT REPORT (LER) thf0RMAT10N AND RECORDS MANAGEMENT BRANCH (MNE,B TEXT CONTINUATION 7714), u.s. NUCLEAR REGutATORT COMNISSION, WASHINGTON, DC 20555-0001 AND TO THE PAPERWORK REDUCT!ON PROJECT (31dD-0104), OFFICE OF MANActMENT AND BUDGET, WASHINCTON, DC 20503.

FACillTT MAM (1) DOCKTT s 4BER (2) t ER IKMBER (6S PAGE (3)

TEAR SEQUENTIAL REVISION South Texas, Unit 1 05000 498 4 OF 5 93 009 01 TEXT (if we spece is reovired, use acrJitionet eccles of WRC form 366A) (17)

ANALYSIS OF EVENT: (Cont'd) o Cracks in ferrules of the blown fuse had no apparent impact on its electrical performance and no apparent role in the fuse opening. l l

Gould Shawmut reviewed the SwRI Failure Analysis Report and in general agreed with the SwRI report. Gould Shawmut agreed that the fuses had been subjected to greater than rated current (eg. 10 amps), however, this is not unusual and fuses are designed for such service. That is, fuses are designed to accept a certain amount of "overcurrent" due to conditions such j as inrushes. Gould Shawmut concl*.2ded that the thermal damage observed in '

both the blown SSPS fuse and the unblown companion fuses was indicative of stress cracking of the zinc element caused by thermal cycling. As such, the most probable cause of the fuse opening was mechanical stressing (aging) of the element. That is, the fuse reached the end Jf its life.

I STP has concluded that the SSPS fuse opening was a random age related l failure event. This conclusion is based on the circuit analysis discussed i above and is supported by the fact that the SSPS actuation system has in l the past operated properly on safety injection actuations wherein the currents experienced by the subject fuses closely approach those of a main steam line break scenario. The design of STP accounts for single random failures and, therefore, this fuse failure was within the plant's design basis.  ;

Entry into Technical Specification 3.0.3 is reportable pursuant to 10CFR50.73 (a) (i) (B) .

l l CORRECTIVE ACTIONS:

l

1. The electrical distribution panel fuses feeding the three trains of SSPS actuation cabinets in both units were replaced with 20 amp fuses.

l A 20 amp fuse provides adequate protection for this design and provides additional margin to reduce the probability of nuisance fuse blowing due to random age related failure mechanisms.

2. A comparative evaluation of vital 120 VAC distribution panel fuses and selected DC circuit breakers and the main incoming protective devices in the panels fed by them has been done to determine if other problems i

exist. Vendor panels were found in which the panel protection is l larger than the distribution panel protection but these cases were determined to be acceptable after review of the supplied load currents.

One case was identified (6 similar radiation monitors per unit) where

the increased margin provided by a larger size fuse was warranted.

I These 15 amp fuses have been replaced with 30 amp fuses in Unit 1 and an identical change has been designed for Unit 2.

{

• i l

NRC FORM 366A U.S. IUCLEAR REGULATORY CO*lSSION APPROWED BT QMB 183. 3150-0104 l (5-9i) EXP!RES 5/39/95

• ESTIMATED BURDEN PER RESPONSE VO COMPLY MITH TH15 INFORMATION COLLECTION REQUEST: 50.0 HRS.

' FORWARD COMMENTS REGARDING BURDEN ESTIMATE TO THE

. 'LICENBEE EVENT REPORT (LER) INFORMATION AND RECORDS MAhAGEMENT BRANCH (MWBB TEXT CONTINUATION m4), u.S. NUCLEAR REGULATORY COMMIS$10N, WASHINGTON, DC 20555-0001 AND TO THE PAPERW3RK I REDUCTION PROJECT (31$0-0104), OFFICE OF hWACEMENT AND BLOCET, WASHINGTON, DC ?O503.

I FACitITY EAME (1) DOCKET tKMIER (2) LER NLMPER (69 PAGE (3)

YEAR SEQUENTIAL REVISION South Texas, Unit 1 "*" "MR 05000 498 5 OF 5 93 009 01 V9T (If more space is remired. use additionel copies of kRC Form 366A) (17)

CORRECTIVE ACTIONS: (Cont'd) l l 3. A review has been performed by Engineering to determine the adequacy 1,t the station design related to the size selection of fuses and/or circuit breakers within the original architect engineer's design scope which interfaced with vendor designed safety systems. This review was initially conducted on a random sample basis and was later expanded in stages to include 100% of power distribution fuses in Class 1E distribution panels. This review resulted in the conclusion that the potential for undersized protective devices relative to inrush currents appears to be isolated t., fast acting fuses. As a result, fast acting power distribution fuses in Class 1E distribution panels have been reviewed and no operability impacts have been identified, and fast acting fuses feeding the SSPS sctuation cabinets and fast acting fuses in the radiation monitors have been upsized for increased margin.

4. A fuse and breaker field verification was performed to assess the accuracy of the documentation used in the engineering review. This effort was initially conducted on a random sample basis and was later expanded in stages to include 100% of the fast acting power distribution fuses in Class 1E distribution panels. The assessment resulted in the conclusion that the molded case circuit breakers agreed j with the design documentation while the fuses had some r;ize/ type I discrepancies with design documentation. However, all of the

! identified fuse discrepancies were minor in nature, were acceptable for use as is and presented no operability concerns.

5. To reduce the potential of random age related fuse failures affecting I critical systems, fuses in selected equipment were replaced during the fuse and breaker field verification effort.

6. Interim guidance has been issued to the STP Electrical Design Staff to sensitize them to several fuse sizing considerations highlighted by the investigation of this event. STP design practices will be revised to l

formally incorporate this interim guidance by December 31, 1993.

i l

ADDITIONAL INFORMATION:

There have been no previously reported events concerning fuses being undersized causing an unanalyzed event.

l l

1 I