ELV-01995, Special Rept 1-90-05:on 900711,valid Diesel Generator Failures Occurred.Caused by Initial Mfg Process.Starting Air Valves on Each of Four Diesel Generators Tested & Polished to Provide Adequate Clearance Between Pistons & Caps

Special Rept 1-90-05:on 900711,valid Diesel Generator Failures Occurred.Caused by Initial Mfg Process.Starting Air Valves on Each of Four Diesel Generators Tested & Polished to Provide Adequate Clearance Between Pistons & Caps

ML20058M278
Person / Time
Site:	Vogtle
Issue date:	08/07/1990
From:	Hairston W GEORGIA POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
1-90-05, 1-90-5, ELV-01995, ELV-1995, NUDOCS 9008100035
Download: ML20058M278 (3)

Text

. .

l- [

?-> Georg'a Power Company u)1 333 Piedrnont Amnue ~ '

Aharta Georgia 30F 4 Telephone 404 $26 3195 J <

• ( Maitire Addren : . . .

43 inwrness Cortest Pa3way '

l , . Post Oftce !kW 1295

. . Grrmnotiarn Alaoarra 3520I Totephone POS 808 5581 :

August 7, 1990 Ye sou%rn vectic systern W. G. Heireton,111 Sorvor Vee PreMent =

Nuclear Operat.ons 0535-

- Docket No. 50-424' 50-425 1

L'..- S. Nuclea'r Pagulatory Commission ATTN: Document Control-Desk Washington, D. C-. 20555 Gentlemen:-

t V0GTLE ELECTRIC GENERATING PLANT SPECIAL REPORT .

VALID DIESEL GENERATOR FAILURES.

In accordence with the requirements of the Vogtle Electric Generating Plant Technical- Specifications,; Sectlons 4.8.1.1.3 and 6.8.2, Georgia Power = Company .

Lhereby submits. tne enclosed Special Report concerning thr.ee valid diesel ~'~

. generator failures.

Sincerely, Q2. 4.:t v W..G. Hairston, 111 WGH,Ill/NJS/gm

Enclosure:

. Special Report 1-90-05 xc: Georaia Power Comoany-Mr. C. K. McCoy Mr. G. Bockhold, Jr. '

Mr. P. D. Rushton .

-Mri-R. M. Odom J

NORMS 4

. U. S. Nuclear Reaulatory Commission Mr. S. D. Ebneter, Regional Administrator Mr. T. A. Reed, licensing Project Manager, NRR Mr. B. R. Bonser, Senior Resident inspector, Vogtie s

nn 1

9008100'035 90o8o7 m

{fDR ADOCK 05000424

" PDC 5, g  ;

!i l

g 43 l'

ENCLOSURE V0GTLE ELECTRIC GENERATING PLANT - UNITS 1-AND 2

' TECHNICAL SPECIFICATION SPECIAL kEPORT l-90-05 YAl10 DIESEL GENERATOR-FAllURES A. REQUIREMENT FOR-REPORT This report is required in accordance with Technical Specification--(TS) 4.8.1.'l.3. This specification requires that all diesel generator (DG) failures, valid or non-valid, be reported to the Commission in a Special Report pursuant to TS 6.8.2.

B. DESCRIPTION OF EVENT On 7-11-90, Diesel Generator (DG) 2A was being tested during a-routine surveillance per procedure 14980-2,." Diesel Generator Operability Test".

The right air start bank was isolated to allcw testing of the left air start: bank. -The engine start button was pushed by the control room-onerator and-.the engine began to roll with starting air. - According to the. local operator in the diesel room, the engine rolled twice ar.d stopped.i The DG'was declared inoperable and the TS action statement was initiated. The DG was navailable for emergency operation for a period of-67 hours and 49 minut4s.

During the review of,this event, it was determined that similar events had occurred-on 4-12-90 and 7-5-90. These previous similar events had not been recognized as failures and therefore had not'bcen reported as such. These events are. described-as follows:-

-On 4-12-90, operators conducted a TS surveillance test of DG 2A per procedure 14980-2. The manual start button was pushed, but no start occurred. Operators decided that the pushbutton had not been depressed-long'enough and made another attempt which resulted in a successful start. 0n 7-5-90, a similar incident occurred on DG 18. and a successful

-start again resulted on'the second attempt. Neither DG'was considered to

. be unavailable for emergency operation as a result of there two events.

'C. CAUSE OF-EVENT An investigation into the 7-~11-90 event by utility and vendor personnel found that the starting air valve pistons could stick in their cap assemblies due to inadequate manufacturing tolerances. This condition was apparently the result of the initial manufacturing process which left insufficient clearances between some of the pistons and caps. A failure to start would occur only after the engine had been shut down from a .

previous run and the engine stopped with a particular alignment of faulty air start valves and crankshaft position.

v

p ,

g- m 1*y Jog ,

g

, i.

ENCLOSURE.

V0GTLE ELECTRIC GENERATING PLANT - UNITS 1 AND 2 TECHNICAL: SPECIFICATION SPECIAL REPORT l-90-05 VALID DIEML GENERATOR FAILURES ,

1 On a non emergency manual star. with the air start pilot valves malfunctioning,_ the' initial burst of air was not adequate to start the ,

@ engine. -The burst of air was adequate to change the alignment of the. l u crankshaft with respect to.the ?aulty air start pilot valves so that any Y subsejuent attempt:to start the engine could be successful. This problem l W is now believed to have been the cause of the DG failures on 1-24-90 and I

l-25-90, which were reported t9 the t,cemission on 2-19-90 as Special' Report 2-90-02. On 7-19-90, the manufacturer of the valves submitted a 10 CFR-21 report to the Commission as a result of the ahve findings. .1 The failure of the DGLoperators to recognize the initial start attempts

< of 4-12-90 and 7-5-90 as DG failures is partially attributed to-limitations of the simulator computer. The simulator requires operators-

-to hold.the DG. manual start pushbutton in order to have the proper control signals annunciate, creating the misconception thati the ~ , i pushbutton must remain depressed for a given period of time'in order for  !

,a DG start to occur. I D. CORRECTIVE ACTIONS 1

1. The sixteen starting air valves on each of the four DG's -were tested and polished where necessary to: provide adequate clearance between the pistons and caps. .

l

2. The appropriate maintenance procedures will be revised by the next refueling outages-to require testing of the starting air valves to

-demonstrate freedom of movement following 1 overhaul.

3._During shift briefings, operators were advised tha't-the DG should start when the manual pushbutton is depressed, any failure to manually start is a reportable event, and such information should be relayed to f the appropriate personnel so that a report can be initiated.

4. Operator training will be enhanced during the next training cycle to advise personnel-that a DG start should occur without having to .

. continue depressing the manual start pushbutton.

5. The DG IB and 2A test frequency is currently once per 7 days in accordance.with TS Table 4.8-1. This frequency will be continued u until 7 consecutive valid tests are completed with no more than one valid failure in the last 20 valid tests and/or no more than 4 valid q failures in-the last 100 valid tests. Up to-and including the 7-5-90 valid failure, there have been a total of 6 valid failures in 79 valid n tests of DGlB. Up to and including the 7-11-90 valid failure, there have been a-total of 5 valid failures in 43 valid tests of DG2A.

kY w

_h_.__-__ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . - - _ _ _ - --