ELV-02901, Special Rept 1-91-2:on 910522,operator at Diesel Generator Panel Observed Fluctuation in Field Voltage.On 910524,when Diesel Generator Tied to Grid,Field Voltage & Amps Spiked. Caused by Failure of Voltage Regulator.Regulator Replaced

Special Rept 1-91-2:on 910522,operator at Diesel Generator Panel Observed Fluctuation in Field Voltage.On 910524,when Diesel Generator Tied to Grid,Field Voltage & Amps Spiked. Caused by Failure of Voltage Regulator.Regulator Replaced

ML20081J681
Person / Time
Site:	Vogtle
Issue date:	06/17/1991
From:	Mccoy C GEORGIA POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
1-91-2, ELV-02901, ELV-2901, NUDOCS 9106240102
Download: ML20081J681 (4)

Text

. . . . . - - - - - - - - - - - - _ _

y* om e- o . . , s 4 ? i ., e",e Coa.i-

1.

• o I. < c e \q , q

- o, e, ,, 4 Is i i a  : bi n

C. K. McCoy ()(,'O(Qld s f()W(,T

, . . t ,- ,

- e, i

. .am -

June 17, 1991 ELV-02901 1019 Docket No. 50-424 U. S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, D. C. 20555 Gentlemen:

V0GTLE ELECTRIC GENERATING PLANT SPECIAL REPORT VAllD DIESEL GENERATOR FAILURE In accordance with the requirements of the Vogtle Electric Generating Plant Technical Specifications, Sections 4.8.1.1.3 and 6.8.2, Georgia Power Company hereby submits the enclosed Special Report concerning a valid diesel generator failure.

Sincerely, nyrl' CC. K. McCoy CKM/NJS/gmb

Enclosure:

Special Report 1-91-2 xc: Georaia Power Comoany Mr. W. B. Shipman Mr. P. D. Rushton Mr. M. Sheibani NORMS U. S. Nuclear Reaulatory Commissign Mr. S. D. Ebneter, Regional Administrator Mr. D. S. Hood, Licensing Project Manager, NRR Mr. B. R. Bonser, Senior Resident inspector, Vogtle mos m m nos17 1I pn o

noocx osooo.m PDR yL ,

l

V0GTLE ELECTRIC GENERATING PLANT - UlllT 1 TECliNICAL SPECIFICATION SPECIAL REPORT l-91-2 DIESEL GENERATOR IB VAllD FAILURE A. REQUIREMENT FOR REPORT This report is required in accordance with Technical Specification (TS) 4.0.1.1.3, which requires all diesel generator (DG) failures, valid or nonvalid, to be reported to the Commission in a Special Report pursuant to TS 6.8.2.

B. DESCRIPTION OF EVENT On 5-22-91 at 0851 CDT, DG IB was successfully started for a monthly surveillance test per TS requirements. After the DG was tied to the grid and operated at full load for approximately 45 minutes, an operator at the DG panel observed a fluctuation in field voltage. The fluctuation was observed to repeat, and the control room was informed. About 6 minutes after the initial fluctuction was observed at the DG panel, an operator in the control room observed DG 1B indicated reactive power to increase from 2800 kVARs to 4500 kVARs and output current increase from 800 amps to 1200 amps. A third operator immediately decreased DG load, and kVARs and amps decreased but then spiked again. The operator decreased DG load a second time, but reactive power then went ta negative 6000 kVARs. This prompted the operator, at 1016 CDT, to open the DG output breaker, and the DG was declared inoperable. After notifyirg the system engineer, the DG was shut down at 1030 CDT, and an investigation was initiated.

An event critique team was formed to lead the investigation of the event. A troubleshooting plan was developed, and after an oscilloscope and a recorder were connected to appropriate points within the DG control panel, a DG troubleshooting start was made at 1806 CDT on 5-22-91 in an attempt to obtain information concerning the cause of the spiking ')roblem. The DG was operated at rated voltage and frequency; however, the DG was not tied to the grid or loaded. The problem did not repeat, and at 1906 CDT DG 1B was shut down.

Due to a similar recent event t the Grand Gulf Nuclear Station and discussions which occurred between Southern Nuclear and Grand Gulf personnel following that event, a concern existed that the contacts of the rectifier bridge transfer switch may have become oxidized. Therefore, action was taken on 5-23-91 to replace the transfer switch for DG IB. Resistance readings were taken across the contacts of the removed switch and were found, in general, to slightly exceed the desirable limit. Following the replacement _ of the transfer switch, at 1702 CDT on 5-23-91 a second troubleshooting start was made for DG 18. At 1711 CDT, DG 1B was tied to the grid, and after operating at full load for approximately 7 minutes, the spiking problem was observed. An operator decreased DG load to 2000 kW, and kVARs and amps stabilized. At 1910 CDT, the DG output breaker v:as opened to swap to rectifier bridge number 2. The DG was retied to the grid and operated at full load for approximately 66 minutes. One spike in field voltage, kVARs, and amps was observed near the end of the run on rectifier bridge number 2 but no' decrease in DG load was required. At 2123 CDT, DG 18 was shut down.

1

On 5-24-91, action was taken to verify the torque of electrical connections at the exciter reactors and the power transformers. Also, the clearances on the rectifier bridge transfer switch were verified and the remote gate firing module for rectifier bridge number 1 was replaced. At 0521 CDT on 5-24-91, a third troubleshooting start was made. The DG was tied to the grid using rectifier bridge number 2 and was then operated at full load for approximately 65 minutes. During this run, field voltage, kVARs and amps l were twice observed to increase above normal readings. The first increase lasted for approximately 2 minutes, and the second increase lasted approximately 20 seconds. No reduction in DG load was necessary. At 0728 CDT, DG 1B was taken off the grid to transfer to rectifier bridge number 1. l The transfer was made but at 0743 CDT, due to an uncertainty regarding the observed SCR waveforms, DG 18 was shut down prior to being retied to the grid.

Further action was then taken on 5-24-91 to connect a recorder to additional points within the DG control panel, and a temporary modification was implemented to remove and jumper-out two relays which control auto-manual transfer of the DG voltage regulation. Following the completion of this action, at 1356 CDT on 5-24-91 a fourth troubleshooting start was made. The DG was tied to grid, and after operating at full load for approximately 30 minutes, the field voltage, kVARs, and amps were observed to spike. This happened repeatedly until at 1558 CDT, DG IB tripped due to a loss of generator field excitation.

A decision was made to replace the DG field voltage regulator and to connect a recorder to four additinnal points. After completing this action, two DG runs were made to complete adjustments for the voltage regulator. DG IB was then started at 0149 CDT on 5-25-91 for a fifth troubleshooting run. The DG was tied to the grid and ran at full load for 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> without the spiking problem being observed. The DG was shut down at 0438 CDT. Due to the success of this troubleshooting run, DG IB was restarted at 0509 CDT on 5-25-91 for an operability run. The operability run was also successful, and at 0810 CDT on 5-25-91 DG 1B was declared operable.

C. CAUSE OF EVENT The direct cause of the event was the failure of the field voltage regulator. While the rectifier bridge transfer switch and the remote gate firing module for rectifier bridge number 1 were replaced, and a temporary modification was implemented to delete the relays which provide for auto-manual transfer of the DG voltage regulation, subsequent troubleshooting following each of these actions found that the spiking problem was still present.

The root cause for the voltage regulator failure is indeterminate. However, the voltage regulator which failed will be sent to the vendor's (i.e., NEl Peebles) test facility for failure analysis.

2

l D. ANALYSIS OF EVENT l

DG IB was out of service for 69 hours7.986111e-4 days <br />0.0192 hours <br />1.140873e-4 weeks <br />2.62545e-5 months <br /> and 54 minutes. However, during the l time that DG IB was out of service, the redundant diesel generator (i.e., DG 1A) was continuously available for emergency service. The operability of DG 1A was demonstrated on 5-23-91 by performance of an operability run pur TS requirements. Also, the operability of required ac offsite sources was verified per TS requirements.

The failure of DG IB on 5-22-91 is classified as a valid failure per i Regulatory Guide 1.108, section C.2.e. The troubleshooting failures that occurred for DG IB (i.e., the 1702 CDT start on 5-23-91, the 0521 CDT start on 5-24-91, and the 1356 CDT start on 5-24-91) are classified as invalid failures per Regulatory Guide 1.108, section C.2.e. For DG IB, there has been 1 valid failure in the last 20 valid tests and 4 valid failures in the last 100 valid tests. Therefore, the test frequency for DG 1B remains at once per 31 days in accordance with TS table 4.8-1 requirements.

E. FOLLOWUP CORRECTIVE ACTIONS

1. Replacement relays for the relays which were removed under the temporary modification have been installed and the jumpers which were installed under the temporiry modification have been removed.

2. A representative of the voltage regulator vendor arrived onsite on June 11, 1991. Under the direction of the vendor representative, minor adjustments were made to DG 1B to further balance the SCR currents.

Additionally, the voltage regulator that failed will be sent to the vendor's test facility for failure analysis. Should a definitive cause be determined for the failure, a supplement to this Special Report will be submitted.

3