ML20086F979

Special Rept:On 910920,22,27,1001 & 02,invalid Failures of Diesel Generators 1A & 1B Occurred.Redundant Diesel Generator Operable During Each Event.Diesel Generator 1B Placed on Weekly Testing Frequency

ML20086F979
Person / Time
Site:	Byron
Issue date:	11/22/1991
From:	Pleniewicz R COMMONWEALTH EDISON CO.
To:	Davis A NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
BYRON-91-0922, BYRON-91-922, NUDOCS 9112040103
Download: ML20086F979 (7)

Text

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -

e Novemb2r 22, 1991 LTR Dyron 91-092?

TILE 2.07.400 Mr. A. Itert Davis Regional Administrator Region Ill U.S. Nuclear Regulatory Commiision 799 Roosevelt Road Glen Ellyn, Illinois 60137 Subjects Byron Unit 1 Operating License HPr-37, Docket Hu. 50 454 Reporting of Deergency Diesel Generator railures Dear Mr. Davist This report is submitted in accordance with Byron Station Unit 1 Operating i.lcense NPT-37 Appendix A, Technical Specification 4.8.1.1.3, Reporting of Diesel Generator Failures.

The attached reports address two valid failures and three invalid failures l experienced on the ID Diesel Generator, and three invalid failures experienced I on the 1A Diesel Generator. The esitoria to determine valid and invalid tests i and failures in in accorda..cc with Section C.2.e of Regulatory Guide 1.108. l Only one Diesel Generator was required for operation at the time of the tallures. The redundant Diesel Generator was operable during each event.

Since the Diesel Generators were not in service at the time of the failures, they were not considered to be unavailable at any time as a result of these )

failures. The valid failurer were the first and second failures within the  !

previous twenty valid tests for the 1D Diesel Generator. Per Technical l Specification Table 4.8-1, the ID Diese? Generator was placed on a weekly

( testing frequency. These failures were tha fourth and fifth valid failures within the preceding one hundred Unit 1 Diesel Gererator valid tests.

Sa~cerely, l

~~~~

, i )

ff d$) A ~d t R. Plenlewicz StationManager(

Byron Nuclear Power Station RP/TGB/ph Attachments cc Byron St.ation NRC Senior R 'ident Inspector U.~3. HRCgDocument Contic.'. Desk. Washington D.C.

INPO Record Center l

CECO Distribution List (3421z.VS)

$ifN

( n,94 t'

u, ;

e I

i -

DIESEL GENERATOR 1B INVALID TAILURE SEPTD4 DER 20,1991 On September 20, 1991, following a scheduled 18 month inspection, a start attempt was made on the IB Diesel Generator for a maintenance run. The Diesel Generator came up to rated speed, then immediately tripped. No alarms annunciated. Three more unsuccessful start attempts were mado. The cause of the start failures was test equipment installed in preparation for the calibration of the vibration trip device. The alarm for the trip was bypassed with a j umper. Once the cause was identified and corrected, the Diesel Generator was successfully started. The failures occurred during maintenance runs, due to test. equipment installed on a circuit that would be bypassed during emergency mode operation. Therefore,'the failure was classified as invalid.

t (3421z.VS)

l

'R -

DIESEL GENERATOR 1H VALID FAILURE SEPTEHBER 22, 1991 On September 22, 1991, while performing IBVS 8.1.1.2.f-16, IB Diesel Generator Undervoltage Sequeater Test, the IB Diesel Generator experienced a delay of approximately two ms etos prior to starting after receiving a start signal. A second start atten'.pc f or troubleshooting was also unsuccessful.

The delay in the Diesel Generator start was caused by a failure of the turning gear interlock valves. The function of the turning gear interlock valves is to block the air supply to the starting air valves when the turning gear is engaged, thereby preventing the diesel generr. tor from starting. During this event, the valves were found to be leaking excessively. The leakage limited the air supply to the starting air valves. After approximately two minutes enough air pressure was available to actuate the starting air valves, an, the diesel generator started promptly. The cause of the leakage is attributt6 'o normal degradation of the plunger sealing o-ring. The leakage may have b..n compounded by the turning gear not being properly rected out.

The turning gear interlock valves were replaced, and the turning gear was verified to be properly positioned. The Undervoltage Sequencer Test was then successfully repeated. The maint nance procedure which governs manipulation of the turning gear is veing revised to include verification of proper interlock valve operation, ie. plungers ergaged and no leakage present.

J (3421z.VS) j 1

e 4

• DIESEL GENERATOR 1B VALID FAILURE SEPTOKBER 27, 1991 On September 27, 1991, the ID Diesel Generator was auto-started per ESF Actuation Start Signal Test, IBVS B.1.1.2.f-8. When the start was initiated, the 1B Diesel Generator reached rated speed within ten seconds as required.

The voltage indicated zero on one phase and off scale high on tne other two phases. Investigation determined that a fuse had blown on the secondary side of the potential transformer that providec power to the voltage control and meteritig circuit. The blown fuse, F-8, was replaced. During subsequent testing the diesel generator attained rated speed and voltage within ten seconds.

A possible contributing factor to the fuse loading was a strip chart 16; order that was coanected to the voltage metering circuit for the test which was being performed. Since the recorder could not be positively identified as a source for the fuse overload, the root cause remains indeterminate. The failure was classified as a valid failure. The blown fuse is considered an isolated event, and no further corrective action is planned.

(3421z.VS)

e g . .

DIESEL GENERATOR IB INVALID FAILURE OCTOBER 1, 1991 On October 1, 1991, during a maintenance run on the IB Diesel Generator, the fuel injection pump for cylinder 6R seised. This caused the associated cylinder to stop firing. The seized pump did not affect the diesel generator's ability to maintain rated load. The diesel generator was shut down, and the defective pump was replaced. Subsequent testing did not identify any more similar problems.

The pump which failed had been recently installed, with approximately one hour of run time prior to the failure. Previous failures of this type had been attributed to foreign material present in the pump internals. It could not be determined if the contamination was present during manufacturing, or was introduced during installation. Further investigation of this failure mode will be considered by the Cooper Bessemer Owners' Group.

(3421z.VS)

t DIESEL GENERATOR IB ItWALID FAILURE OCTOBER 2, 1991 On October 2, 1991, during a maintenance run of the 1B D8esel Generator, the engine was shut down due a large fuel oil leak which developed at the fuel injection pump for cylinder 21. . The cap screws which secure the pump head to the pump casing had come loose. An unsuccessful attempt was made to tighten the screws without stopping the engine. Since the leak was severe enough to constitute a potentia] fire hazard, the engine was shutdown. The pump was replaced, and subsequent testing did not identify any more problems.

The pump which failed had been recently replaced, and had approximately 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> of run time prior to the failure. This is considered to be an isolat.ed event, possibly caused by a manufacturing defect. No further corrective action is planned at this time.

The failure occurred during a maintenance run, and is clossified as invalid.

4 (3421z.VS)

6 DIESEL GENERATOR 1A INVALID FAltlJRES OCTOBER 16 AND 17, 1991 On October 16 and 17, 1991, during three maintenance runs of the 1A Diesel Generator, three separate fuel injection pumps seized. Each of the pumps seized within the first hour of operation after the pumps were installed. The Diesel Generator was running unloaded at the time of each of the failures.

After each failure, the Diesel Generator was shut down, and the pump was replaced. After these three initial failures, no similar problems were encountered.

Previous failures of this type had been attributed to foreign material present in the pump internals. It could not be determined if the contamination was present during manufacturing, or was introduced during installation. Further investigation of this failure mode will be considered by the Cooper Bessemer Owners' Group.

i (3421z.VS)

L_ _ __________________________________________________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - . _ ______ ..