ML20091C085

Special Rept PIR 1-M91-0113:on 910616,diesel Engine Cooling Water Leak Discovered on Diesel Generator 1A at Discharge Flange on Intercooler End Bell Cover.Caused by Poor Casting Quality.New Covers Will Be Installed

ML20091C085
Person / Time
Site:	McGuire
Issue date:	08/01/1991
From:	Mcconnell T DUKE POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
PIR-1-M91-0113, PIR-1-M91-113, NUDOCS 9108050270
Download: ML20091C085 (3)

Text

. ~ . - _ _ . - ~ . . . . - . _ . - _ - _ . _ - . - - - . . - - . ~ . . - . - . . - . - _ - . _ _ .

ll 1 e fluke IUurt Cornpany (7041815 4000

~n bli UW!f NWlt'd! Statien

' . Ilith' Hayes hir) Road flunterndie NC?so!S 8.qM

t. DUNC POWCil August. 1, 1991 U.S. Nuclear Regulatory Commission Doctunent Control Desk Washington, D.C. 20555

Subject:

McGuiro Nuclear Station Unit 1 i Docket No. 50-369 Diesel Generator Special Report Gentlemen

-Pursuant to McGuire Technical Specifications 4.8.1.1.3.C and 6.9.2, find a special report concerning a valid failure of Diesel Generator lA.

Should there be any questions, please contact William Recalde at (704)875-4869. .

Very truly yours, Z }$ h

• T.L.McCo{nnell i ADJ/chl Attachment xc Mr. S. D. Ebneter -

Administrator, Region 11 U.S. Nuclear ReRulatory Commission  :

101 Marietta St., NW, Suite 2900 Atlanta, GA 30323 INPO Records Center Suite 1500 1100 circlo 75 Parkway Atlanta, GA 30339 Mr. Tim Reed U.S. Nuclear Regulatory Commission Offico of Nuclear Reactor Regulation Washington, D.C. 20555 Mr. P. K. Van Doorn-NRC Resident inspector  :

McGuire Nuclear Station l

• 108050270 910903 /

PDR

• ADOCR 0"000369 PDR

/[7,@ g 4

DUKE POWER COMPANY McGU1RE NUCLEAR STATION Diosol Generator Special Report P1R 1-M91-0113 July 29, 1991 [

On June 16, 1991 at 0930, a diosol engino cooling water look was discovered on diosol generator (DG) 1A during routino rounds by Operations personnol. The leak was at the dischargo flango on_the intercooler and bell cover. Coolant was noticed dripping (70 drops por minuto) from a 2.5 inch vertical crack on the loft sido of the water outlet flange. Based on a crack growth calculation by Dosign Engincoring personnel, a f ailuro was considered likely if the DG had boon required to-operate for 2 days at full load. This assumed that the leak would increase to the point that it surpassed the makeup capacity of the cooling water system. The leak is conservatively classiflod as a VALID FAILURE of DG 1A because of discovery of a condition that would have caused failure. Start attempt #844 was assigned for record kooping purposes. DG 1A was dociared Inoperablo at 1730 on 6/16/91 to replace the intercoolor tubo bundio (Work Roquest 145095).

This was the first Valid Failure in the last 20 Valid Tosts and the fifth Valid Failure in the last 100 Valid Tests for DG 1A. On a unit basis, this is the fourth valid Fa11uro in the last 100 Valid Tosts of DGs 1A and 1B combined. Accolorated survoillance tost frequency (weekly) has begun por Technical Specification Tablo 4.8-1 (Diosol Generator Tost Schedulo).

Operations personnel visually veriflod that none of the other throo DG intercoolers waro leaking and QA department personnol performed a dyo ponotrant chock (no cracks indicated) . Maintenance personnel removed and replaced the intercooler tube bundle. DG 1A was functionally _ operable for testing on 6/19/91 at 0139. Start attempt 845 was performed on 6/19/91 at 0707. The DG started and was-shu_tdown by operations due to a turbocharger oil pump priming problom:(Invalid Test). Start attempt 846 was mado at 0822 to run the surveillanco test (PT/1/A/4350/02A, Diopol Generator 1A operability _ Tost). The DG was successfully run at full load for an hour with no indication of water looks (Valid Success) . DG 1 A was declared operablo on 6/19/91 at 1155.

The intercoolor end bell cover was taken to the Applied Science Contor _for failure analysis. In summary, the Metallurgical Analysis Report (Samplo #1155, 6/26/91) indicated a brittle f racturo mode for tho . crack. There was no evidence of f atigue such as a vibration problem would indicato. The casting quality of the cover was not good and appears to be the root of the problem. The replacement intercoolor end bell cover was fabricated from shoot M

=e see e ,- -

e ew'w.-v= g ,,,v--w,ry -ww-- m+ ---,+---~,-,,--,,v .---<c -w - - , - - e< e'

metal. The shoot mo t.a 1 has greator ductility than the original grey cast iron and should exporlonce no brittle cracking. Efforts are underway to havo now and boll covers f abricated f or the 111 and 213 DGs which still have the cast iron covere. The now covers will be inst.alled as soon as possible.

i i

i f

L 4 4

, R.M. oberts l System Engincoring cc: M.D. Rains R.L. Browdor R.E. Ilunning W.W. Gallman G.L. Dorest S.D. 1lart R.J. D0080 B.T. liarkey S. Anderson' J.D. Motos l-s I

J' w ,^ +,wy, + -- ., ..,,,-,,,y-, , . , , , _ f...,y., ,,_,y,,, , , , , , , , , . , , , , , , , . , , , , ., , ,_ , , ,