ML20082U091

Special Rept 91-003:on 910819,invalid Failure of EDG B. Caused by Unit Parallel Relay.Hermetically Sealed Relay Being Evaluated as Means to Prevent Similar Failures of Relay

ML20082U091
Person / Time
Site:	Wolf Creek
Issue date:	09/16/1991
From:	Bailey J WOLF CREEK NUCLEAR OPERATING CORP.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
91-003, 91-3, NO-91-0242, NO-91-242, NUDOCS 9109190287
Download: ML20082U091 (3)

Text

1 4 e W@ NUCLEAR LF CREEK OPERATING COR John A. Gahy Vr., Presioemt optirettorm September 16. 1991 NO 91-0242

'U. S. Nuclear Regulatory Consission ATTN Document Control Desk Mail Station F1-137 Washington, D. C. 2055$

Subject:

Docket No. 50-482: Special keport 91 13 the attached Special Report is being submitted in accordance with Tcchnical r Specification 4.8.1.1.3 concerning an invalid failure of Emergency Diesci Generator 'B".

Very truly yours, m W a.du /

.)ohn A. Bailey Vice President Operations JAB /jra Attachment cc: L. L. Gundrwn (NRC), w/a

- A. T. Howell (NRC), wla R. D. Martin (NRC), via W. D. Reckley (NRC), w/a l

4 1 l'd ('V {,,} *_g t i..

PO Box 411 Burlington, KS 66839 i Phone (316) 364-8831 9 I @ ! 00 2: ,' ; [ c;q, g , [,

ADO $v O Ss;*,;,,;, kg; - An Eauai Optetunity Employer M F HCVET FOS PDR

///

Attachment to NO 91-0242 l

Page 1 of 2 SPICIAL IMmf 91-003 invalid 1%ilmu of herTFMCV DINGl CiYWratDr "H."

his report describes an invalid failure of hmgency Diesel Generator (EDG)

"B" which occurred on August 19, 1991. Wis report is being subnitted in accordance with 'Ibchnical Specifications 4.8.1.1.3 and 6.9.2.

IMEIMFPION QF EVNIS On August 19,1991, at 0051 hours5.902778e-4 days <br />0.0142 hours <br />8.43254e-5 weeks <br />1.94055e-5 months <br />, EDG "B" was started for the perfonmnce of its nonthly surveillance test, to noet the nylirments of Technical Specification 4.8.1.1.2. The test nquires the EDG to be paralleled to the offsite electrical grid and run fully loaded (6.2 m ) for at least one hour. Shortly after the EDG was fully loaded, operators noticed the generator load was treMing upward. Nonelly a " droop" circuit prevents the generator frun assuming additional load when paralleled to the grid. With the generator load slowly increasing, operators took nunual control and unintained the load nmr 6.2 W for the ruminder of the test. The EDG was unloaded and shutd wn at 0230 hours0.00266 days <br />0.0639 hours <br />3.80291e-4 weeks <br />8.7515e-5 months <br />. The droop circuitry is used only when the generator is paralleled to the grid and is not used in the mergency operating node. W erefore, the EDG remained operable throughout this event.

During the test, the generator reached a muinum output of 6.8 W. Wis is well below its 30 minute rating of 7.4 W. Even if operators had not taken manual control of the generator load, various protective trips (such as generator overcurrent) would have stopptrl the EDG before any danage would o.: cur.

ROCrP CRISE NO OmRICPIVE JCPICNS Troubleshooting following the test detennined the Unit Parallel Relay (UPR) had failed. The UPR allcws the electronic governor to operate in droop modo parallel operation. One contact on the relay had intonnittent resistance readings, which prevented the governor fran properly performing its droop function. The EDG was taken out of service at 0814 hours0.00942 days <br />0.226 hours <br />0.00135 weeks <br />3.09727e-4 months <br /> on August 23, 1991, to allow for replacment of the UPR. We relay was replaced and at 1159 hours0.0134 days <br />0.322 hours <br />0.00192 weeks <br />4.409995e-4 months <br /> the EDG was started to verify operation of the droop circuitrv.

The EDG ran fully loadcd for one hour and the droop circuitry operated properly. At 1359 hours0.0157 days <br />0.378 hours <br />0.00225 weeks <br />5.170995e-4 months <br />, the EDG was declared back in service, resulting in the EDG teing unavailable for 5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> and 15 minutes.

Initial investigation indicates the high resistance readings were caused by foreign naterial on the relay contact. W e contacts have a very low voltage across them (1.5 volts), causing than to be very sensitive to any contaminates. We use of a henretically sealed relay is being evaluated as a means to prevent similar failures of this relay. mis evaluation will be cmpleted by 12-31-91.

- - Attachment to NO 91-0242

.Poge 2 of 2

', PhlTUtX CIAh'91FICATIEN Although the test of the EDG was cmpleted and it verified the EDG was available to operate in an mergency, nanual load control is not nonally required as part of this test. For this reason, this event is considertd reportable under 7bchnical Specifications 4.8.1.1.3 and 6.9.2.

This evmt is considered to be an invalid test and failum in Laccordance with Regulatory Position C.2.e(2) of Rcqulatory Guide 1.108, Revision 1, since the UFR and the droop function am not n,quind and do not operato in the anargency operating nrde. The event does not affect the current surveillance test interval of 31 days as required by Table 4.8-1 of 7bchnicsl Specification 3.8.1.1. Through August 23, 1991, them have 1xon four valid failures of EDG "B" in the last 100 valid tests. Prun the tino of the last valid failure of EEG "B" on Decamber 9,1989, EDG "B" has tudergone 29 successful valid tests. 2he last invalid failure of EDG "B" occurred on April 23, 1990 (Special Report 90-002).

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