ML20085D825

Special Rept:On 910911,EDG 1B Shutdown Due to Excessive Vibration.On 910916,EDG 1B Shutdown Due to Oil Leak on Cylinder 8.On 910919,EDG 1B Shutdown Due to Loud Noises from Air Start Distributor.Distributor Shaft Reinstalled

ML20085D825
Person / Time
Site:	Seabrook
Issue date:	10/11/1991
From:	Feigenbaum T PUBLIC SERVICE CO. OF NEW HAMPSHIRE
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NYN-91165, NUDOCS 9110170197
Download: ML20085D825 (3)

Text

. .

New Hampshire Ted C. Feigenbaum Yhh h Preddent and Chief Enecutive OHicer NYN-91165 October 11, 1991 United States Nuclear Regulatory Commission Washington, D.C. 20555 .

Attention: Document Control Dcsk

References:

(a) Facility Operating Lict.tse No. NPF-86, Docket No.53-443 (b) Seabrook Station Technical Specification 4.8.1.1.3

Subject:

Diesel Generator Special Report Gentlemen:

The following Diesel Generator Special Report documents six failures of Emergency Diesel Generator 1B (EDG 1B), which occurred at Scabrook Station during the period Septemiser 11-25, 1991. This report is provided pursuant to Technical Specification Surveillance Requirement 4.8.1.1.3 and Technical Specification 6.8.2. Two of these failures were determined to be valid test fa;. lures and the four rcmaining failures were determined to be non-valid test failures. The test failures were determined as valid or non-valid in accordance with the criteria of Regulatory Guide 1.108 ' Periodic Testing of Diesel Generator Units Used as Onsite Electric Power Systems at Nueh.ar Power Plants." At the time of these events, the plant was shutdown for refueling (in Mode 5) with the A train emergency diesel generator operabic.

Event # 1 - September 11.1991 at 1745 linart During the performance of 'he monthly operability surveillance, EDG 1B was shutdown after 213 minutes of run time due to escessive vibration and overheating in the air start left bank air start distributor manifold arca, it was discovered that the air start distributor lines to cylinders #3, #4, #6 and #7 had failed. The air start distributor was subscauently disassembled. Upon disassembly, it was discovered that there was evidence of rubbing between the rotor and housing mating surfaces. The acceptance eriteria of the surveillance test was met but this test was determined to be a valid failure due to the required vibration shutdown. This was the first valid failure of EDG 1B in the last 20 and last 100 valid tests.

1 Eve nt -# 2 - September 16.1991 at 0059 Hours During the performance of the 18-Month Operability and Engineered Safeguards Purr. . l and Valve Response Time Testing Surveillance, EDG 1B was intentionally shutdown after 28 ,

minutes of run time due to an oil leak on #8 cylinder. After the oil leak was repa: red, it j was determined that the oil leak was not of the magnitude that would have resulted in diesel '

generator damage or failure. As a result, it was determined that this was a non-va!id test j failure, l 9110170197 911011 PDR ADOCK 05000443 S PDR / L i

!j()

New Hampshire Yankee Division of Public Service Compon'y of New Hampshire z l

P.O. Box 300

• Seabr%, NH 03874

• Telephone (603) 474 9521 1

. . , _ . - _ - _ .. -._ . _=_.- --- - -m .--

- United States Nuclear Regulatory Comtnission October M1,1991 Attention: Document Control Desk Page two Ey. cat # 31 Sententpfr 16.1901 at 0721 Ilours During the performance of the 18 Month Operability and Engineered Safeguatds Pump and Valve Response Time Testinr Surveillance, EDG 1B was shutdown after $6 minutes of run time. The diesel generator was shutdown due to the severing of two air lines on the left bank air start manifold. Upon engine shutdown, an investigation identified broken air start -distributot lines to cylinders #4, # 5, # 6 and # 7. The air bank distributor was disassembled and additional ciidence of rubbing between the rotor and housing mating surfaces was observed. Additional engine disassembly was performed to deter.nine the cause of the air start distributor inuuced failures. As a result of the shutdown of EDO -1B due to abnormal conditions that would.have resulted in er;uipment damage, it was determined that this was a valid test failure. This we.s the second valid failure of EDG IB in the last 20 and last 100 valid tests.

Event #4 ~- Scotember 10.1001 nt 055? Ilours .

During the performance of a maintenance engine run for troubleshootir.g and d u collection, EDO 1B was shutdown after 18 hours2.083333e-4 days <br />0.005 hours <br />2.97619e-5 weeks <br />6.849e-6 months <br /> and 47 minutes of run time due to loud -

ncises from the air start distributor. Since this engine run was performed in the process of troubleshooting, it was determined that this was a non-valid test failure.

Event #5 - Scotember 21.1991 at 2113 If oug During'the_ performance of a maintenance engine rnn for troubleshooting and data collection, EDG 111 was shutdown after 21 hours2.430556e-4 days <br />0.00583 hours <br />3.472222e-5 weeks <br />7.9905e-6 months <br /> and 8 minutes of run time due to 1(,ad noises from the air start distributor. Since this enginc run was performed in the process of troubleshooting, it was determined that this was a non-valld- test failure.

Event #6 - September 25.1901 at 0327 Hours During-the performance of a maintenance engine run for troubleshooting and data collection EDG IB failed to start. Starting circuit monitoring equipment, that had bec :

, _ temporarily installed for troubleshooting purposes, shorted to ground. The short to ground L

_ simulated a- failed potential -transforn.cr fuse and caused an automatie engine shutdown.

This was, the first start attempt after rebuilding the air start distributor following the failure reported in Event #5. Since this temporary mon:toring equipment is not part of the defined diesel generator unit desige and because this engine run was for troubleshooting and data collection, it = was determined that this was a non-valid test failure, t-Root Cause A number of possible root causes were identified and evaluated The most probable cause D of the test -failures was determined to be excessive vibration of -the left bank air start

- distributor. The vibration caused rubbing between the rotor and air distributor housing.

This rubbing eventually resulted in~ noisy operation and overheating of- the left air start distributor (Event #4 and #5), the severing of the left air start distrib.w air lines (Event

1. 1 and #3) and leaking oil line (Event #2).

The ro_ot cause of Evect -#6 was determined to be the use of faulty test equipment which caused an unexpected short to ground. This electrical short circuiting simulated a failed potential transformer fuse causing the engine to automatically shutdown.

L i

4 United States Nuclear Regulatory Commission October 11., 1991 Attenilon: Document Control Desk Pye three Corrective Action The original air start distributo: shaft was reinstalled and the rotor aswmbly was replaced.

The severed air lines and oi! leaks were repaired. EDG IB is cu:4cntly operable and is curtently being tested on a weekly basis pursuant to Tt.chnical 5pecification Surveillance Requiremont 4.8.1.1.2.a. This test frequency will be maintained until seven consecutive failure-free demands have been perfornied and the number af failures in the last 20 valid d:mrnds has been reduced to one or less. Additianally, to enhance EDG performance, actions are be'ng taken to reduce the total Emergency Diesel Generator vibrations.

Shciuld you have ai.y questions regarding this report, please contact Mr. James M.

Peschel, Regulatory Ct 1.pliance Manager, at (603) 474-9521 extension 3772. _

Very truly 6 Irs, e

(,)/s,g #.A y Tad C. Feighan- /

TCF:JWC/jwc tc: Mr. Thomas T. Martin Regional Administrator United States Nuclear Rep latory Commission Region 1 475 Allendale Road King of Prussia, PA 19496 Mr. Gordon E. Edisua, Senior Project Manager Pioject Directorate 1 Division of Reactor Projects United States Nuclea- Regulatory Commission Washington, DC 20555 Mr. Ncel Dudley NRC Senior Resident Inspector P. O. Box 1149 Seabrook, Nil 03874 s