ML20113B729

Special Rept 85-02:on 850226,invalid Failure of Diesel Generator a Occurred.Caused by Operating Error.Event Not Considered Valid Test or Failure.Silicon Controlled Rectifiers Replaced

ML20113B729
Person / Time
Site:	Callaway
Issue date:	03/28/1985
From:	Miltenberger S UNION ELECTRIC CO.
To:	James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
85-02, 85-2, ULNRC-1072, NUDOCS 8504110364
Download: ML20113B729 (4)

Text

~'

,k n

• UNION EtscTaic COMPANY CALLAWAY PLANT MMLING ADDRESS:

P. O. BOX 820 FULTON. MO. SS2 51 March 28, 1985 Mr. James G. Keppler Regional Administrator Office of Inspection & Enforcement U.S. Nuclear Regulatory Commission Region III 799 Roosevelt Road Glen Ellyn, IL 60137 ULNRC-1072

Dear Mr. Keppler:

DOCKET NUMBER 50-483 CALLAWAY PLANT UNIT 1 FACILITY OPERATING LICENSE NPF-30 SPECIAL REPORT 85-02 DIESEL GENERATOR INVALID FAILURE The enclosed Special Report is submitted pursuant to Technical Specifications 4.8.1.1.3 and 6.9.2 concerning an invalid failure of a diesel generator.

n .

S. E. Miltenberger V Manager, Callaway Plant.

WRC/WRR/JWK/drs

Enclosure cc

Distribution attached 8504110364 850328 APR 2 M l PDR ADOCK 05000483 n S PDR ll TC#h

cc distribution for ULNRC-1072 American Nuclear-Insurers c/o Dottie Sherman, Library The Exchange Suite 245 270 Farmington Avenue Farmington, CT 06032 Records Center Institute of Nuclear Power Operations Suite 1500 1100 circle 75 Parkway Atlanta, GA 30339

.NRC Resident Inspector Missouri Public Service Commission D. F. Schnell J. F. McLaughlin J. E. Davis (Z40SR)

D. W. Capone /R. P. Wendling F. D. Field R. L. Powers A. C. Passwater/D. E. Shafer/D. J. Walker G. A. Hughes

-W. R. Robinson (QA Record)

W. R. Campbell G. L. Randolph J. M. Price R. A. McAleenan L. K. Robertson (470)(NSRB)

Merlin Williams, Wolf Creek SEM Chrono 3456-0021.6 3456-0260 Z40ULNRC G56.37 N. Date

SPECIAL REPORT 85-02 DIESEL GENERATOR INVALID FAILURE on 2/26/85 at 0231 CST an invalid failure of diesel generator (D/G)

"A", NE01, occurred while verifying operability of the D/G after an

, outage. The unsuccessful start can definitely be attributed to operating error and is therefore not considered a valid test or failure.

Starts of the D/G's have been tracked since the completion of Preoperational Testing on 5/11/84. To date there have been seventeen valid tests of D/G "A" and twelve valid tests of D/G "B." D/G "A" has experienced one valid failure (Special Report 84-02) and two previous invalid failures (Special Report 85-01). No failures of D/G "B" have occurred.

At approximately 0500 CST on 2/25/85, D/G "A" was taken out-of-service for a maintenance outage. As part of the outage a defective set of Silicon Controlled Rectifiers (SCR's) was replaced at 2000 on 2/25/85 under Work Request (WR) 004899. The defective set of SCR's had no effect on the operability of the D/G as a redundant set of SCR's is available. The D/G had been operating on the redundant set of SCR's since 4/20/84. Temporary Modification (TM) 84-E-167 had lifted the leads to the SCR selector switch to prevent the defective set from being energized. The temporary modification was cleared and the leads reconnected as part of WR 004899.

At approximately 0230 on 2/26/85, D/G "A" was started per Operations Surveillance Procedure OSP-NE-00002, Standby D/G Periodic Tests, to verify operability of the D/G after the outage. The voltage only came up to approximately 2500 volts (required voltage is 4000 volts) and the operator noted voltage regulator control was not available. The D/G was immediately secured by the operator. After discu'ssion with the system engineer, the original SCR bank was put back in service and the D/G started satisfactorily.

On 3/12/85 an investigation into the failure revealed that a second temporary modification (TM 84-E-500) had lifted the leads to the pulse transformers associated with the defective set of -SCR's. The temporary modification was to be cleared under WR 004899 but had not been identified in the work package. With the leads lifted, the voltage regulator was inoperable and.the generator failed to attain the required voltage. As a conservative measure, the D/G was placed on a fourteen day testing frequency until TM 84-E-500 could be cleared and verified as the cause of the failure.

On 3/21/85 WR 043589 landed the leads at the pulse transformers and cleared TM 84-E-500. The D/G was then started satisfactorily using the set of SCR's installed on 2/25/85. The satisfactory start after clearing TM 84-E-500 verified the failure of the D/G to be a result of a personnel error (failure to clear TM 84-E-500) during the replacement of the defective set of SCR's.

I

SPECIAL REPORT 85-02 Page 2 Administrative Procedure APA-ZZ-00320, Initiating and Processing Work Requests, requires the temporary modification number to be

. identified on the WR written to close the temporary modification. TM 84-E-167 was identified on WR 004899 but TM 84-E-500 failed to be identified.- To prevent recurrence, Administrative Procedure

.APA-ZZ-00380 Temporary System Modifications, will be revised to require that the temporary modification number be entered on the closure document, i.e. WR, prior to the installation of the temporary modification. Expected completion date of the revision is 5/15/85.

As' stated in section C.2.e.(2) of Regulatory Guide 1.108, Revision 1, August 1977, unsuccessful start and load attempts that can definitely be attributed to operating error should not be considered valid tests or failures. The failure on 2/26/85 is attributed to operating error in that a personnel oversight resulted in the temporary modification not being cleared. Therefore the testing interval need not be decreased to less than or equal to fourteen days as required by Regulatory Guide 1.108 and Technical Specification (T/S) Table 4.8-1.

The performance of D/G surveillance testing has been restored to at least once per 31 days. This is in conformance with the schedule of Regulatory Position C.2.d of Regulatory Guide 1.108 and T/S Table 4.8-1 which require the test interval to be not more than 31 days if the number of failures in the last 100 valid tests is one or zero, t

i

.J i

, . - . . - - ,-.