ELV-03109, Special Rept:On 910821,valid Diesel Generator Failure of DG 2B Occurred,Troubleshooting Revealed K-1 Exciter Shutdown Relay Was Latched in Shutdown Position W/Reset.Caused by Component Failure

Special Rept:On 910821,valid Diesel Generator Failure of DG 2B Occurred,Troubleshooting Revealed K-1 Exciter Shutdown Relay Was Latched in Shutdown Position W/Reset.Caused by Component Failure

ML20083B693
Person / Time
Site:	Vogtle
Issue date:	09/20/1991
From:	Mccoy C GEORGIA POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
ELV-03109, ELV-3109, NUDOCS 9109250167
Download: ML20083B693 (3)

Text

. _ _ _ _ _ _ _ _ _ _ _

c,., 4 ,,3 n,.u., cynim 4 ) Li,e : a a fe Ot ' P# ' a cs ,

FeJ(W r ll u L' m

[.. yyx />nsn' r

, , ny , . a m o c.

m c.x.accuy Georg!albwer s ..v,, w w so .. , s ,, cr,,. , ,o September 20, 1991 ELV-03109 1120 Docket No. 50-425 U. S. Nuclear Regulatory Commission ATIN: Document Control Desk Washington, D. C. 20555 Gentlemen: .

V0G1LE ELECTRIC GENERATING PLANT SPECIAL REPORT VAllD DIESEL GENERAlofLFMLURE in accordance with the requirements of the Vogtle Electric Generatirig Plant Technical Specifications, sections 4.8.1.1.3 and 6.8.2, Georgia Power Company hereby submits the enclosed Special Report concerning a v.111d diesel generator failure.

Sincerely, lY C. K. McCoy j

CKM/NJS/gmb

Enclosure:

Special Report 2-91-2 l

xc: Georola Power Company Mr. W. B. Shipman Mr. M. Sheibani NORMS t

! L S. Nurlear Reag]atory Ccmmission

! Mr. S. D. Ebneter. Regional Administrator l Mr. D. S. Hood, licensirig Project Manager, NRR Mr. B. R. Bonser, Senior Resident inspector, Vogtie

, n ?

mm,w mno p)"

[DR ADOCK 05000425 I

,\

pon L

)

V0GTLE EtECTRIC GENERATING PLANT - UNIT 2 TECHNICAL SPEClflCATION SPECIAL REPORT 2-91-2 VAllD DIESEL GENERATOR FAltVRE A. ' REQUIREMENT f0R REPORT This report is required in accordance with Technical Specifications (TS) section 4.0.1.1.3, which requires all diesel generator (DG) f ailures, valid or nonvalid, be reported to the Commission in a Special Report pursuant to TS 6.8.2.

B. DESCRIPTION Of VAllD FAILURE Of DIESEL GENERATOR 2B On 8-21-91 at 0842 CDT, DG 28 was started for a monthly surveillance test per TS requirements. The DG start pushbutton was depressed and control room -

operators observed annunciators nor.nal for a DG start come in. However, both control room and local meters for DG 2B showed no indication of generator voltage or frequency. Also, a lack of generator field voltage and amperage was indicated by local meters located on the generator control panel, and after about a minute, an annunciator for DG 20 low voltage was received. An operator stationed at the DG reported that the DG had started and other than the lack of meter indication, appeared to be operating normally. Apparently, the D3 field had failed to flash. After discussing the observed indications with the system engineer, DG 28 was declard inoperable and was shut down at 0905 CDT.

Troubleshooting following the DG shutdown revealed that the K-1 exciter shutdown relay (i.e., a Telemecanique Catalog No. A143E 125 Vdc mechanical latching relay) was latched in the shutdown position with a reset signal present. A 0.25-ampere fuse was also found to be blown; however, this fuse was located in the 125-Vdc circuitry for the synchroniting relay and would ,

not have contributed tu the field failure to flash. The K-1 relay was removed, and further investigation determined that the mechanical latching mechanism for the relay was failing to retract completely on a reset signal.

A mechanical latching mechanism was obtained from a spare K-1 relay and was installed in place of the malfunctioning latching mechanism. The K-1 relay was then reinstalled, and after functional testing of the relay, DG 2B was started for a maintenance troubleshooting run at 0211 CDT on 8-22-91. All indications came up as expected, and the DG was paralleled to the grid and slightly loaded. No problem was observed, and DG 20 was shut down at 0223 CDT. At 0257 CDT, DG 2B was restarted for an operability run. The operability run was successfully completed, and at 0641 CDT on 8-22-91, DG 2B was declared operable.

! C. CAUSE OF EVENT The root cause of the event was a component failure of the K-1 relay. As described above, the mechanical latching mechanism for the K-1 relay malfunctioned and caused the K-1 relay to remain in the shutdown position.

Based on a review of industry experience and discussions with the relay manufacturer, this failure appears to have been an isolated occurrence.

1

A failure of a Telemecanique (formerly ITE) relay utilized as a K-1 shutdown relay for an emergency diesel generator was identified as having occurred at the Perry Nuclear Power Plant early during 1990; however, that failure involved a failure of the relay to latch.

The cause of the blown 0.25-ampere fuse is indeterminate; however, it is likely that the fuse blew while being tested.

D. ANAL.YSIS OF EVENT The function of the K-1 relay is to shut down the DG excitation system by shorting the secondary side of the power current transformers which supply current to the field. This relay is actuated on either a DG shutdown signal or when the DG is >1 aced in the maintenance mode. The control circuit is designed so that t1e K-1 relay will reset within approximately 2 minutes af ter actuation on a DG shutdown signal and immediately after the DG is taken out of the maintenance mode. Prior to the start of DG 28 on 8-21-91.

DG 2B was last shut down from an operating condition at 0200 CDT on 8-13-91 and was last taken out of the maintenance mode at 0810 CDT on 8-21-91 (i.e.,

after completion of a prestart cylinder moisture check). Therefore, the malfunction of the K-1 relay must have occurred on either 8-13-91 at 0200 CDT or on 8-21-91 at 0810 CDT.

The blown 0.25-ampere fuse would .10t have impacted the capability of DG 2B to start on an emerge.1cy start signal or to provide power to its associated emergency busses.

The elasped time that DG 2B was declared inoperable was 21 hours2.430556e-4 days <br />0.00583 hours <br />3.472222e-5 weeks <br />7.9905e-6 months <br /> and 59 minutes. While it is possible that DG 2B may have actually been inoperable from 8-13-91 until 8-22-91, it is noted that redundant DG 2A was continuously available for emergency service throughout this time. The operability of DG 2A was successfully demonstrated on 8-6-91, 8-13-91, 8-20-91, and 8-22-91. Also, the operability of required ac offsite sources was successfully demonstrated per TS requirements on 8-21-91 and on 8-22-91 after DG 2B was declared inoperable.

The failure of DG 2B on 8-21-01 is classified as a valid failure per Regulatory Guide 1.108, section C.2.e. For DG 2B, there have been 2 valid failures in the last 20 valid tests and 2 valid failures in 55 valid tests.

Therefore, the test frequency for DG 2B has been increased to once per 7 days in accordance with TS table 4.8-1 requirements.

E. CORRECTIVE ACTIONS The blown fuse was replaced, and the K-1 relay was repaired as noted above.

I l

_ _ . . _ . _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . . _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . _ _ _