ML20128M204

Summary of ACRS Subcommittee on Electrical Sys 850320 Meeting in Washington,Dc Re Ongoing NRC & Industry Work on Diesel Generator Reliability.Viewgraphs & Technical Rept A-3134 1-85 Encl

ML20128M204
Person / Time
Issue date:	03/31/1985
From:	Advisory Committee on Reactor Safeguards
To:	Advisory Committee on Reactor Safeguards
Shared Package
ML20128M206	List: ML20128M204
References
ACRS-2295, GL-84-15, NUDOCS 8507110609
Download: ML20128M204 (10)

Text

ktRS- D a C &

F 4 lg $b) ,

3 POR oca184 g[ g DATE ISSUED: 3/31/85 l

3\WSS ELECTRICAL SYSTEMS SUBCOMMITTEE MEETING MINUTES MARCH 20, 1985 WASHINGTON, DC The ACRS Subcommittee on Electrical Systems met on March 20, 1985 at 1717 H Street, NW, Washington, DC. The purpose of this meeting was to discuss the ongoing NRC Staff and Industry work on diesel generator reliability. The meeting was for information and no ACRS action is planned. The Subcommittee heard presentations from representatives of NRR, IE, and EPRI. The meeting was began at 8:30 am and was adjourned at 3:00 pm, and was heli entirely in open session. The principle attendees were as follows:

W. Kerr - Subconinittee Chairman J. C. Mark - ACRS Member C. Michelson - ACRS Member C. Wylie - ACRS Member R. Savio - ACRS Staff M. El-Zeftawy - ACRS Staff L. Rubinstein - NRR M. Srinivasan - NRR J. Knight - NRR C. Berlinger - NRR R. Hernan - NRR G. Zech - IE E. Herschoff - IE H. Wyckoff - EPRI HIGHLIGHTS

1. The NRC Staff is currently working to resolve Generic Issue B-56,

" Diesel Generator Reliability," Generic Letter 84-15 has been 8507110609 850331 5 PDR

Electrical Systems Minutes March 20, 1985 issued under this activity. The Generic Letter requests that the Licensees supply to the NRC the plant data on diesel generator reliability and a description of the plant's diesel generator reliability programs (if any) and Licensee comerts on the NRC Staff proposed diesel generator reliability goals as specified in the current NRC Staff proposal for the resolution of B-56 (See l Highlight 2). The Generic Letter also proposes that Licensees strive to reduce fast-start surveillance testing and supplies a sample Technical Specification change for implementing a modified test procedure. The response to this Generic Letter has not been overwhelming. Most Licensees have supplied data on diesel generator reliability and reliability programs as they exists.

Thirty-nine plants have indicated an interest in modifying Technical Specifications to allow for less frequent fast-start testing. The other Licensees appear to be reluctant to do this.

- Control hardware would have to be modified to accomodate the propose new testing procedures and the difficulties associated with changing Technical Specificatfons would have to be dealt with. The NRC Staff will be having additional discussions with all Licensees in the near future.

l

2. The data supplied by the Licensees generally indicates that the reliability of diesel generator is improving. Currently (based on thelast100 starts)89%ofthedieselgenerators(singleengines) have a reliability of 0.95/ demand or better and 2% have a reliability of less than 0.90/ demand. The current NRC Staff proposal for diesel generator reliability programs specifies that a reliability at'0.99/ demand is a goal, that no action is required if the reliability is 2. 0.95, that no immediate action will be taken for plants between 0.95 and 0.90 pending the resolution of A-44,

" Station Blackout," and that the plants below 0.90 will be dealt with on a case-by-case basis.

J

c s

• Electrical Sy*,tems Minutes March 20, 1985

3. EPRI has evaluated the existing US nuclear plant diesel generator experience and has compared this work with similar work performed by the NRC. EPRI has concluded that the data base used by the NRC was not complete and that the NRC's evaluation did not yield accurate results. In the NRC analysis, engine starts without loading the generator and deliberately terminated load runs (not long enough to qualify as a successful run) were not included in the data base. However, ail failures to start or load were included. EPRI concluded that this resulted in failures be counted against a artifically small data base. EPRI is proposing that the NRC's process for computing diesel generator reliability be modified to account for this. EPRI also believes that only the more recent diesel generator experience should be used to predict future performance. The NRC Staff appeared to be receptive to discussing these proposals.

4. The status of work being performed by the NRC's Task Force on the TDI diesel generator reliability was discussed. The crankshaft failure on a Shoreham TDI diesel engine led to the formation of this Task Force. TDI diesel experience was reviewed by the NRC and by the Industry. A significant number of problems were discovered and solutions have or are being developed. A sumary of some of this operating experience is given on Figure 1.

5. The NRC Staff gave a short presentation comparing Japanese and US diesel generator reliability practices. A summary comparing Japanese and US practice is given on pages 2 to 6. The reasons for It would the differences and the benefits are not well understood.

appear that more investigation is needed.

6. The Vendor Program Branch will be inspecting the implementation of QA programs at diesel generator, vendor facilities and at plant sites. Inspections have been perfortned at Cooper Energy Survey,

s

! Electrical Systems Minutes March 20, 1985 Morrison-Knudsen, Stesart-Stevenson, and Colt /Fairbanks Morse. A BNL review of diesel generator performance is to be used to identify inspection priorities. A copy of the BNL report is included as Attachment A.

NOTE: Additional meeting details can be obtained from a transcript of this meeting available in the NRC Public Document Room, 1717 H Street, N.W., Washington, D.C, or can be purchased from ACE-Federal Reporters, 444 North Capitol Street, Washington, D.C. 20001, (202) 347-3700.

I 4

HISTORY OF EVENTS WHICH LEAD TO TDI PP0 JECT GROUP BEING FnPMFD:

o CRANKSBAFT FAILIJRE AT SHOREHAM o REVIEW OF OTHER TDI OPERATING EXPEPIENCE PROBLEMS ,

A) PISTON-CROWN SEPARATION F) PISTON SKIRT CRACKS C) FUEL LINE FAILURES - FIRE D) CYLINDER HEAD CRACKS E) TURB0 CHARGER PROBLEM

- BEAPINGS

- INLET VANES

- MOUtiTING BOLTS F) PUSH R0D WELD CPACKS G) GENERATOR SHORT DUE TO FASTENER FAILURE H) AIR START VALVE PROBLEM I) JACKET WATER PUMP J) CRANKSHAFT CRACKS - DIL HOLES K) CONNECTING ROD BEARING FAILURES L) CYLINDER BLOCK CRACKS

- BLOCK TOP

- CAM GALLEPY M) UNQUALIFIED INSTPUMENT CABLES / CONNECTORS o TDI VENDOR INSPECTIONS (1979-1983)

IDENTIFIED NUMBER OF NONCONF0PMANCES/ VIOLATIONS

s yy l o

l eRELIABILITY G0AL U.S.

JAPAN CURRENT CRITERIA (RG 1.108)

HAS NO NUMERICAL RELIABILITY ESTABLISHES DG RELIABILITY G0AL FOR DG G0AL 0F 0.99 ePROT0 TYPE QUALIFICATION REQUIREMENTS eSPECIFYDESIGNREQUIREM$NTS, DESIGN ADEQUACY ASSURED BY:

AND PERFORMANCE AND RELIABILITY TESTS THROUGH CODES & STANDARDS

• IEEE 387, RG 1.9 AND OTHERS eCODES & STANDARDS EQUIV.

TO U.S.

eTESTS DEMONSTRATE DESIGN CAP- f eNO INDICATION OF TEST SERIES ABILITY, RESPONSE & RELIABILITY l

)

TO DEMONSTRATE RELIABILITY 300 START / LOAD TESTS l 99% RELIABILITY DEMONSTRATED

• 0A PROGRAM REQUIRED BY 10 CFR eQA PROGRAM APPENDIX B l

l

Jdr~ [3 PREOPERATIONAL PHASE U.S.

JAPAN eDURING PRE 0PERATIONAL TESTING eINSPECTION BEFORE USE THE INSITU RELIABILITY OF .99 IS DEMONSTRATED PER RG 1.108

• ACCEPTANCE TEST - MINISTRY BY 69 TESTS WITHOUT A FAILURE 0F INTERNATIONAL TRADE -

COVERS ALARM, INTERLOCKS, eFUNCT10NAL AND CAPABILITY OVERSPEED GOVERNOR AUTO PICKUP, HEAT RUN TESTS ACCOMPLISHED PER RG 1.68 (PRE 0P TESTING) eQA PROGRAM - INSTALLATION e0A PROGRAM - INSTALLATION eREGIONAL INSPECTION OF INSTALLATION AND TESTING

I-

~

xy OPERATIONAL PHASE

~

PERIODIC TESTING U.S.

JAPAN e AT LEAST MONTHLY;.EERIOD eMONTHLY ACCORDING TO FAILURES IN SLOW START (LOAD & UNLOAD) LAST 100 TESTS FUEL TRANSFER PUMP AND AIR FAST START & LOAD (RELIEF COMPRESSOR ON FAST START HAS BEEN GIVEN TO LICENSEES)

FUEL TRANSFER PUMP START EON DG FAILURE, REMAINING DGs e0N DG FAILURE REMAINING DG TESTED, IMMEDIATELY AND DAILY TESTED IN 1 HR. & EVERY 8 HR.

THEREAFTER - IF NOT OPERABLE WITHIN 10 DAYS OPERABLE DG RUN CONTINUQUSLY (MAX 30 DAYS) e72 HOUR PLANT SHUTDOWN IF NOT eNO PLANT SHUTDOWN INDICATED REPA! RED e92 DAY SAMPLE TEST FULL OIL

• 18MONTHDURINGREFUELING

• REFUELING

- LOAD REJECT CAPABILITY l AUTO START SIMULATION OF - SIMULATED LOOP & LOCA i

DESIGN BASIS AND LOOP SEPARATELY & COMBINED WITH AUTO LOAD ACCIDENT - 214 HOUR LOAD TEST LOADS ON SEQUENCE - AUTO LOAD SEQUENCER

\

_ _ __ ___ _ J

K{

_. OPERATIONAL PHASE INSPECTIONS & PREVENTATIVE MAINTENANCE JAPAN U.S.

• VARIES FROM PLANT-TO-PLANT

• DAILY CONFIRMATION OF STANDBY CONDITION DAILY OR EVERY SHIFT CHECKLIST WALK THROUGH EXAMPLES:

LUBE OIL TEMPERATURE, FUEL TANK LEVEL, VALVE AND SWITCH POSITIONS, AIR PRESSURE eINSPECTION REQUIRED AT LEAST eINSPECTIONS EVERY 18 MONTHS .

ONCE PER YEAR DURING REFUELING PER MFRS.

RECOMMENDATIONS

• 0VERWAUL PERIOD RANGES FROM *0VERHAUL PERIOD DEPENDS ON 1 f4 YEARS FOR VARIOUS MFR. RECOMMENDATION COMPONENTS DETAILED INSPECTION OF PARTS

' PARAMETERS MONITORED eTREND ANALYSIS

9 8 4 FAILURE BREAKDOWN JAPAN U.S.

SINCE JUNE 1979 SINCE 1980 30 MALFUNCTIONS 450 MALFUNCTIONS

• 66% 1. MECHANICAL EQUIPMENT 407.

1. MECHANICAL EQUIPMENT 30% 2. ELEC. EQUIPT AND 42%

2. ELEC. EQUIPT AND INST. CONT. SYSTEM INST. CONT. SYSTEM 4% 3. HUPAN ERROR 18%

3. HUMAN ERROR FINDING 0F MALFUNCTIONS ARE MALFUNCTIONS FOUND BY:

AS FOLLOWS:

1. FINDING BY PATROL 44% 1. ROUTINE DAILY CHECKS

2. SURVEILLANCE 26% 2. TESTING

3. BY ALARMS 8% 3. PERIODIC INSPECTIONS

• INPO SOER NO. 83-1