ML16134A473

Response of Babcock & Wilcox Co Plants Following Loss of Nonemergency AC Power

ML16134A473
Person / Time
Site:	Oconee
Issue date:	08/26/1996
From:	Raughley W NRC OFFICE FOR ANALYSIS & EVALUATION OF OPERATIONAL DATA (AEOD)
To:
Shared Package
ML16134A475	List: ML16134A473
References
TASK-*****, TASK-AE AEOD-*****, AEOD-T96-03, AEOD-T96-3, NUDOCS 9609040320
Download: ML16134A473 (11)

Text

AEOD/T96-03 TECHNICAL REVIEW REPORT RESPONSE OF BABCOCK & WILCOX COMPANY PLANTS FOLLOWING A LOSS OF NONEMERGENCY AC POWER Prepared by:

William S. Raughley Reactor Analysis Branch Safety Programs Division Office for Analysis and Evaluation of Operational Data 9609040320 960826 PDR ADOCK 05000269 I

S

_PDR

INTRODUCTION Management review of Office for Analysis and Evaluation of Operational Data (AEOD) and Office of Nuclear Reactor Regulation (NRR) reports (Refs. I and 2) on the Oconee emergency power system questioned whether Oconee had considered the worst case emergency electrical loading. The concern was that a complete loss of nonemergency ac power at Oconee could lead to reactor trips at the three units and potential overcooling transients. Such a scenario could lead to emergency core cooling system (ECCS) actuation at all units. Multiple ECCS actuations would result in exceeding the present emergency power system loads, and exacerbate voltage and frequency performance issues already identified by NRR and AEOD.

AEOD was requested to review the operating experience to characterize the response of Babcock & Wilcox Company (B&W) nuclear plants to a loss of nonemergency ac power and determine the implications to the Oconee emergency power system.

SUMMARY

A review of the operating experience at B&W nuclear plants from 1987 to the present found that complete or partial losses of nonemergency ac power did not result in any overcooling transients or full ECCS actuations. However, four out of five losses of nonemergency ac power resulted in initiation of automatic start of the emergency feedwater pump motors from the emergency power supply following a loss of feedwater or feedwater transient. At Oconee, start of the emergency feedwater pump motors is not considered in the electrical loss-of coolant accident/loss-of-offsite power (LOCA/LOOP) analysis of the transient voltage, current, and frequency at the safety equipment connected to the emergency power supply.

OPERATING EXPERIENCE The operating experience review used License Event Reports (LERs) for B&W plants from 1987 to the present. The review identified 18 events involving a complete or partial loss of nonemergency ac power to one or more emergency buses that resulted in loading of one or more emergency power supplies. Thirteen of these events occurred with the reactor shut down and 5 occurred with the reactor above zero power.

Twelve LER abstracts describing the 13 loss of nonemergency ac power events with the reactor shut down are provided in Attachment 1, "LER Abstracts-Loss of Nonemergency AC Electrical Power With the Reactor Shut Down."

The response of the B&W plants to the five loss of nonemergency ac power events with the reactor above zero power are summarized below.

1

LER No. 270/92-04 On October 19, 1992, Oconee Unit 2 experienced a loss of nonemergency ac power and a trip from 100 percent power. One emergency power supply automatically loaded with the required shut down loads. A feedwater transient resulted in the start of the motor driven emergency feedwater pumps (MDEFPs).

LER No. 287/96-001 On March 16, 1996, Oconee Unit 3 experienced a loss of nonemergency ac power and trip from 100 percent power. The unexpected load shed signal of several secondary system pumps resulted in a loss of feedwater, and Unit 3 tripped. One emergency power supply loaded with the required shutdown loads and the MDEFPs.

LER No. 302/89-023 On June 6, 1989, Crystal River Unit 3 was at 12 percent reactor power and experienced a loss of nonemergency ac power following the trip of the start-up transformer. A loss of main feedwater occurred. The emergency diesel generators loaded. The MDEFP failed to start automatically as expected. and was manually started.

LER No. 302/92-001 On March 27, 1992, Crystal River Unit 3 was at 98 percent reactor power and experienced a loss of nonemergency ac power. The emergency diesel generators started and loaded. The make-up pump was started manually after it failed to start automatically. The emergency feedwater pumps started on loss of voltage and closure of the emergency power supply output circuit breaker.

LER No. 346/87-011 On September 9, 1987, Davis Besse was at 100 percent reactor power. A feedwater flow transmitter failed, and a subsequent operator error resulted in a reactor trip.

Several anomalies occurred including the failure of a bus transfer. The emergency diesel generators started and loaded. Approximately 4 hours4.62963e-5 days <br />0.00111 hours <br />6.613757e-6 weeks <br />1.522e-6 months <br /> into the event, a turbine bypass valve failed open, allowing the main steam to dump directly to the condenser. At this time steam generator No. 2 level decreased rapidly and both auxiliary feedwater pumps started.

2

DISCUSSION Review of the events while shut down found the emergency power supply was used to restore power to required loads following each of the 13 loss of nonemergency ac power events. In 2 of the 13 cases, an engineered safeguards isolation signal(s) resulted in the operation of some motor-operated valves. In a third event, a switch was inadvertently bumped and resulted in full ECCS actuation; however, being defueled, many systems had been disabled and only motor-operated valves and dampers loaded. These events were not of interest for the purpose of this evaluation; with the reactor shut down, the emergency power supplies were judged to be lightly loaded from the engineered safeguards and other required loads.

Review of the five loss of nonemergency ac power events summarized above found no overcooling transients, or full ECCS actuations. In four events that occurred at Crystal River and Oconee, the loss of nonemergency ac power resulted in initiation of automatic start of the MDEFP following a loss of feedwater, or feedwater transient. In the fifth event at Davis Besse, a feedwater transient initiated the start of the emergency (auxiliary) feedwater pumps several hours after the start of the event.

Oconee was evaluated considering that a loss of nonemergency ac power could result in loading of the MDEFPs for all three units to a single emergency power supply. However, it does not appear that tests or analyses fully considered timing effects for a loss of nonemergency ac power, or a design basis LOOP/LOCA accident, where one emergency power supply accepts loads from three Oconee units. The MDEFP start can be delayed 30 seconds (Ref. 3) after a low-water level in either steam generator. Consideration of the MDEFP time delay, its tolerance, and the tolerance of other timed devices used in loading the emergency power supply may result in overlapping motor starts and lengthen the time of transient variations in voltage, current, and frequency at the equipment.

FINDINGS

1.

The operating experience shows that a loss of nonemergency ac power at a B&W plant does not result in full actuation of ECCS.

2.

A loss of nonemergency ac power at a B&W plant is likely to result in automatic start of a MDEFP following a loss of feedwater or feedwater transient. Oconee analyses and tests did not always consider MDEFP time delays or timing device tolerances that.affect the safety equipment voltage, current, and frequency when loading the emergency power supply.

3

REFERENCES

1.

U.S. Nuclear Regulatory Commission, Draft Special Study, AEOD/S96-XX, "Oconee Electrical System Design and Operation," June 1996.

2.

U.S. Nuclear Regulatory Commission, Draft Report "Oconee Emergency Electrical Power System and Other Related Matters," July 1996.

I.

Duke Power Company, Oconee Nuclear Station, "Final Safety Analysis Report,"

December 31, 1995, pps. 10--22.

4

LER Abstracts-Loss of Nonemergency AC Power With the Reactor Shut Down The following losses of nonemergency ac power occurred with the reactor shut down. The events are grouped by plant in ascending order of the event date'.

OCONEE UNIT 3

• • k*k*k**k*k**

DOCKET YEAR LER NUMBER REVISION DCS NUMBER2 NSIC' EVENT DATE 287 1987 002 0

8704130053 204108 03/05/87 POWER LEVEL -

000%. ON MARCH 5, 1987 AT 1515 WITH UNIT 3 IN A REFUELING OUTAGE, PERSONNEL BEGAN CUTTING ON THE RIGHT HAND SIDE OF PANELBOARD EB-1 IN ORDER TO INSTALL A NEW WATT-HOUR METER. CUTTING INDUCED VIBRATION CAUSED ACTUATION OF THE STARTUP TRANSFORMER (CT-3)

LOCKOUT RELAY THUS CAUSING LOSS OF ALL INCOMING AC POWER TO UNIT 3.

NONESSENTIAL ELECTRICAL EQUIPMENT WAS THEN LOAD SHED, AND BOTH KEOWEE HYDRO UNITS STARTED.

POWER WAS REESTABLISHED FROM KEOWEE WITHIN THE REQUIRED AMOUNT OF TIME. THE ROOT CAUSE OF THIS INCIDENT WAS THE FAILURE OF THE DESIGN REVIEW PROCESS FOR THIS MODIFICATION TO ADEQUATELY ADDRESS THE INTERACTIONS OF NEARBY EQUIPMENT DURING THE INSTALLATION PROCESS.

THERE WERE NO RADIOLOGICAL RELEASES AS A RESULT

'OF THIS EVENT.

AS SUCH, THE HEALTH AND SAFETY OF THE PUBLIC WERE NOT AFFECTED.

OCONEE UNIT 3 DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 287 1988 004 0

8811020343 211014 08/30/88 POWER LEVEL -

000%.

ON 08/30/88, AT 1628, WITH UNIT 3 IN A REFUELING OUTAGE WITH THE REACTOR DE-FUELED, A FALSE UNDER-VOLTAGE CONDITION WAS INPUT INTO THE MAIN FEEDER BUS MONITOR PANEL (MFBMP) #2.

AT THAT TIME MAIN FEEDER BUS (MFB)

1. 1 WAS OUT OF SERVICE.

THESE CONDITIONS LED TO AN INDICATED LOSS OF POWER TO BOTH MFBS, CAUSING A LOAD SHED TO OCCUR.

AS A RESULT OF THE LOAD SHED, POWER WAS LOST TO VARIOUS PLANT EQUIPMENT INCLUDING THE SPENT FUEL COOLING PUMPS, WHICH WERE COOLING THE OFF-LOADED CORE STORED IN THE SPENT FUEL POOL.

ALSO AN EMERGENCY START OF THE KEOWEE HYDRO GENERATOR UNITS WAS INITIATED AS A RESULT OF THE LOAD SHED.

THE LOAD SHED SIGNAL WAS INITIATED AS A RESULT OF THE LOAD SHED.

THE LOAD SHED SIGNAL WAS INITIATED BY A CONSTRUCTION AND MAINTENANCE (CMD)

CREW LIFTING ELECTRICAL LINKS ON THE MFBMP, CREATING A FALSE MFB UNDER VOLTAGE SIGNAL.

IMMEDIATE CORRECTIVE ACTIONS INCLUDED INVESTIGATING THE CAUSE OF THE LOAD SHED, RESTORING POWER TO THE MFBMP, SECURING THE KEOWEE HYDRO GENERATOR AND RESETTING THE LOAD SHED CIRCUITRY.

THE ROOT CAUSE OF THIS EVENT WAS THE FAILURE OF THE PROCEDURE, USED BY THE CMD CREW, TO ENSURE ADEQUATE PRECAUTIONS WERE The abstracts are from the Licensee Events Reports as submitted to the NRC.

2 DCS Number is the U.S. Nuclear Regulatory Commission's document control system accession number.

NSIC is the Oakridge National Research Laboratory's. Nuclear System Information Center accession number.

1 Attachment

GIVEN SO THAT THE LOAD SHED LOGIC WAS NOT INITIATED ON THE MFBMP WHEN THE CREW PERFORMED THEIR JOB TASK. A LOAD SHED SIGNAL WAS INITIATED WHEN THE ELECTRICAL LINKS WERE LIFTED ON MFBMP #2, CAUSING AN UNDER-VOLTAGE SIGNAL.

THREE MILE ISLAND UNIT 1 DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 289 1987 002 0

8704080083 204109.

03/02/87 POWER LEVEL -

000%. TMI-1 WAS IN REFUELING SHUTDOWN WITH THE "A" DECAY HEAT REMOVAL SYSTEM IN OPERATION. ELECTRICAL MAINTENANCE PERSONNEL HAD JUST COMPLETED MOLDED CASE CIRCUIT BREAKER TESTING IN ACCORDANCE WITH PLANT PROCEDURE 1420-BKR-L IN THE lA AUXILIARY TRANSFORMER CONTROL CABINET.

THE MOLDED CASE CIRCUIT BREAKERS PROVIDE PROTECTION FOR THE TRANSFORMER CONTROL POWER, COOLING FANS, AND OIL PUMPS.

WHILE PERFORMING AN ADDITIONAL PROCEDURE WHICH REQUIRES CLEANING OF THE CONTROL CABINETS, CHECKING FOR LOOSE CONNECTIONS, ETC., THE ELECTRICIAN NOTICED A RELAY COVER WHICH WAS NOT PROPERLY INSTALLED.

UPON ATTEMPTING TO RE-INSTALL THE RELAY COVER, THE RELAY OPERATED CAUSING THE 1A AUXILIARY TRANSFORMER AND THE 230KV 8 BUS TO TRIP AND LOCKOUT. THIS IN TURN CAUSED THE lB EMERGENCY DIESEL GENERATOR (EDG)

TO START AND LOAD.

THE AUTOMATIC START OF THE EDG IS REPORTABLE UNDER 10 CFR 50.73(A)(2)(IV).

THERE WAS NO SAFETY SIGNIFICANCE TO THE AUTO START OF THE EDG.

THE ROOT CAUSE OF THE EVENT WAS IMPROPERLY DRESSED WIRING ON THE FAULT PRESSURE RELAY WHICH DID NOT ALLOW A COMPLETE FIT OF THE RELAY COVER. MAINTENANCE OF THIS TYPE IS SCHEDULED DURING PERIODS WHEN THE EFFECT OF INADVERTENT OPERATION IS MINIMAL.

THE RELAY WIRING AND COVER INTERFERENCE WILL BE RESOLVED DURING THE 7R REFUELING OUTAGE.

CRYSTAL RIVER UNIT 3 DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 302 1987 025 0

8712240085 207506 10/16/87 POWER LEVEL -

000%. ON OCTOBER 16, 1987, CRYSTAL RIVER UNIT 3 WAS SHUT DOWN IN A REFUELING OUTAGE. AT 2119, PERSONNEL WORKING IN THE VICINITY OF THE UNIT STARTUP TRANSFORMER RAISED A METAL POLE AND MADE ELECTRICAL CONTACT WITH A 230 KV FEEDER INTERRUPTING THE PLANT OFFSITE POWER SUPPLY. THE FOLLOWING SIGNIFICANT EVENTS RESULTED:

THE ENGINEERED SAFEGUARDS SYSTEM ACTUATED, THE "B" DIESEL GENERATOR STARTED AND LOADED, NORMAL POWER WAS LOST TO THE SECURITY SYSTEMS, AND THE REACTOR BUILDING PURGE ISOLATED. ADDITIONALLY, POWER TO THE FOLLOWING WAS LOST:

ONE NEUTRON MONITORING CHANNEL, THE AUXILIARY BUILDING VENTILATION SYSTEM EXHAUST FANS, THE CONTROL BOARD ANNUNCIATOR AND EVENT RECORDER, AND THE EMERGENCY NOTIFICATION SYSTEM PHONE.

THIS EVENT WAS CAUSED BY ACCIDENTAL GROUNDING OF THE UNIT STARTUP TRANSFORMER 230 KV FEEDER.RESULTING IN INTERRUPTION OF THE OFFSITE POWER SUPPLY. ELECTRICAL DISTRIBUTION SYSTEM LINEUPS WERE RESTORED TO THEIR PRE-EVENT STATUS AND THE DAMAGED 230 KV FEEDER WAS REPAIRED. WORK ACTIVITIES IN THE VICINITY OF THE UNIT STARTUP TRANSFORMER HAVE BEEN DISCONTINUED.

2

CRYSTAL RIVER UNIT 3 DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 302 1989 025 0

8908010251 214918 06/29/89 POWER LEVEL -

000%. ON 6/29/89, CRYSTAL RIVER UNIT 3 WAS IN HOT SHUTDOWN (MODE 3), COOLING DOWN TO PERFORM MAINTENANCE ON ONE OF THE TWO EMERGENCY DIESEL GENERATORS (EDG).

AN ELECTRICAL STORM WAS IN PROGRESS IN THE AREA. AT 2015, THE PLANT SEPARATED FROM ITS NORMAL OFFSITE POWER SOURCE. THE OPERABLE EDG AUTOMATICALLY STARTED AND REENERGIZED THE "A" 4160V ENGINEERED SAFEGUARDS (ES) BUS.

TWO MINUTES AFTER SEPARATION FROM THE NORMAL OFFSITE POWER SOURCE, OPERATORS ALIGNED POWER FROM A SECOND OFFSITE SOURCE TO THE "B" 4160V ES BUS, AND ENERGIZED THE BUS. OPERATORS RESTORED THE NORMAL OFFSITE POWER SOURCE AT 2137.

THE REACTOR WAS COOLED BY NATURAL CIRCULATION FLOW FOR APPROX. TWO HOURS DURING THE EVENT. SEPARATION FROM THE NORMAL OFFSITE POWER SOURCE OCCURRED DUE TO AN ELECTRIC FAULT IN THE 230KV SWITCHYARD.

THERE WAS INSUFFICIENT DATA AVAILABLE TO DETERMINE THE EXACT CAUSE OF THE ELECTRIC FAULT. HOWEVER, IT IS BELIEVED THAT THE EVENT WAS CAUSED BY LIGHTNING.

CRYSTAL RIVER UNIT 3 DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 302 1991 010 0

9111260245 223427 10/20/91 POWER LEVEL -

000%. ON OCTOBER 20, 1991, CRYSTAL RIVER UNIT 3 (CR-3)

WAS IN MODE 5 (COLD SHUTDOWN) FOR A SCHEDULED MAINTENANCE OUTAGE. AT 1443 THE BREAKERS FOR THE OFFSITE POWER TRANSFORMER OPENED, DISCONNECTING THE ENGINEERED SAFEGUARDS (ES)

BUSSES FROM THE OFFSITE POWER SUPPLY.

DECAY HEAT COOLING WAS INTERRUPTED FOR LESS THAN A MINUTE WHILE THE EMERGENCY DIESEL GENERATOR (EDG) LOADED THE ES BUS AND OPERATORS RESTARTED THE DECAY HEAT PUMP (DHP).

UPON STARTING THE DHP, A PURIFICATION RELIEF VALVE LIFTED CAUSING A DROP IN PRESSURIZER LEVEL. OPERATORS QUICKLY IDENTIFIED THE SOURCE AND ISOLATED THE PURIFICATION SYSTEM. AT 1447, THE OPERATORS MANUALLY ENERGIZED THE REMAINING ES BUS VIA THE CR-3 STARTUP TRANSFORMER.

THIS EVENT WAS CAUSED BY A PREEXISTING WIRE INSTALLATION WHICH INADVERTENTLY APPLIED 115V AC TO THE CR-3 BATTERY BUS.

THE ERRONEOUS WIRE HAS BEEN REMOVED.

THE ASSOCIATED BREAKER RELAYS HAVE BEEN REPLACED WITH LESS SENSITIVE RELAYS TO MINIMIZE THE POSSIBILITY OF SIMILAR EVENTS IN THE FUTURE.

INFORMATION WILL BE PROVIDED TO ALL OPERATORS CONCERNING THIS EVENT.

THE ASSOCIATED EMERGENCY PROCEDURE (EP) WILL BE REVISED TO ADDRESS RESTARTING THE DHP.

CRYSTAL RIVER UNIT 3 DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 302 1993 002 2

9511140158 0

03/29/93 POWER LEVEL -

000%.

ON MARCH 29,

1993, FLORIDA POWER CORPORATION'S (FPC)

CRYSTAL RIVER UNIT 3 WAS IN MODE 5, COLD SHUTDOWN.

ALL ELECTRICAL POWER WAS BEING SUPPLIED VIA A BACKFEED FROM THE 500 KILOVOLT (KV)

SWITCHYARD.

AT 0151, THE 500KV OUTPUT BREAKERS OPENED, INTERRUPTING THE BACKFEED.

BOTH EMERGENCY DIESEL GENERATORS STARTED AND ENERGIZED THEIR RESPECTIVE ENGINEERED SAFEGUARDS BUSSES.

DURING THE LOSS OF ONE (500KV BACKFEED)

OF THE TWO AVAILABLE OFFSITE POWER

SOURCES, DECAY HEAT REMOVAL COOLING STOPPED AND WAS RESTARTED FIVE MINUTES LATER WHEN THE ALTERNATE SOURCE OF POWER (230KV)

REENERGIZED THE SAFETY RELATED BUSSES. DURING THE TIME DECAY HEAT REMOVAL WAS NOT 3

AVAILABLE THE 94 DEGREE FAHRENHEIT (F) REACTOR COOLANT TEMPERATURE INCREASED A MAXIMUM OF APPROXIMATELY 14 DEGREES F NEAR THE CENTER OF THE CORE, WHILE TEMPERATURES NEAR THE CORE BARREL INCREASED SUBSTANTIALLY LESS.

THE OFFSITE POWER FROM THE 230KV SWITCHYARD WAS AVAILABLE DURING THE ENTIRE EVOLUTION. SALT WATER ENTERED SWITCHYARD CABLE TRENCHES AS THE RESULT OF A WINTER STORM. SALT WATER IMMERSION OF A DEGRADED CABLE IS SUSPECTED TO HAVE CONTRIBUTED TO THE CABLE FAILURE, SUBSEQUENT ACTUATION OF RELAYS WHICH OPENED BREAKERS, AND LOSS OF ONE OF TWO AVAILABLE OFFSITE POWER SOURCES. THE DAMAGED CABLES WERE REPLACED AND OTHER CABLES WERE TESTED AND REPLACED AS NECESSARY.

.FPC HAS APPROVED A PROJECT FOR CONTROL AND POWER CABLE REPLACEMENT AND SEPARATION.

CRYSTAL RIVER UNIT 3 DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 302 1993 004 0

9305180308 0

04/08/93 POWER LEVEL -

000%. ON APRIL 8, 1993, CRYSTAL RIVER UNIT 3 (CR-3) WAS IN MODE 5 (COLD SHUTDOWN) FOR A PLANNED MAINTENANCE OUTAGE.

ELECTRICAL POWER WAS BEING SUPPLIED VIA BACKFEED FROM THE 500 KILOVOLT (KV) SWITCHYARD. ONE OF THE TWO CLASS 1E ELECTRICAL SYSTEMS WAS OUT OF SERVICE. AT 1803 THE 500KV BREAKERS OPENED, INTERRUPTING THE POWER BACKFEED. THE IN-SERVICE CLASS 1E 4160 VOLT (V) BUS DEENERGIZED.

THE "B" EMERGENCY DIESEL GENERATOR STARTED AUTOMATICALLY AND REENERGIZED THE BUS.

THIS REPORT IS SUBMITTED IN ACCORDANCE WITH 10CFR50.73(A)(2)(IV).

DECAY HEAT REMOVAL FLOW WAS STOPPED FOR APPROXIMATELY THREE MINUTES DURING THE TRANSITION OF LOADS TO THE EMERGENCY DIESEL GENERATOR. REACTOR COOLANT SYSTEM TEMPERATURE, INITIALLY 89 DEGREES FAHRENHEIT, INCREASED APPROXIMATELY 4.5 DEGREES FAHRENHEIT DURING THIS PERIOD.

THE EVENT OCCURRED WHEN PERSONNEL FOLLOWED AN INCORRECT SEQUENCE FOR REMOVING DC POWER TO THE 500KV BREAKER PROTECTIVE RELAYS AS A RESULT OF INADEQUATE WORK INSTRUCTIONS BEING USED TO REPLACE A BATTERY CELL IN ONE OF THE SWITCHYARD BATTERY BANKS. ADDITIONAL ADMINISTRATIVE CONTROLS HAVE BEEN ESTABLISHED TO REDUCE THE RISK OF INTERRUPTION OF OFFSITE POWER AND DECAY.HEAT REMOVAL. ADDITIONAL CORRECTIVE ACTIONS, BASED ON THE RESULTS OF A HUMAN PERFORMANCE ENHANCEMENT SYSTEM (HPES) ANALYSIS, WILL BE ADDRESSED.

RANCHO SECO At Atsk******* WWWWWWWWWWW~kt

• *WWWW*

• • • • e****

• * *g DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 312 1987 028 0

8706160676 204868 05/14/87 POWER LEVEL -

000%. DURING COLD SHUTDOWN CONDITIONS A HIGH VOLTAGE ON THE DISTRICT'S ELECTRIC POWER SWITCHYARD WAS SENSED BY THE EMERGENCY POWER SUPPLY SYSTEM AT RANCHO SECO. AS A RESULT OF THAT SIGNAL, THE BRUCE/GM DIESEL GENERATOR "B" AUTOMATICALLY STARTED, THE SEQUENCER UNLOADED AND RELOADED THE BUS AND THE DIESEL GENERATOR OPERATED SATISFACTORILY. AN UNPLANNED AUTOMATIC ACTUATION OF AN ENGINEERED SAFETY FEATURE IS REPORTABLE TO THE NRC ACCORDING TO 10 CFR PART 50.73(A)(2)(IV). PROPOSED TECHNICAL SPECIFICATION AMENDMENT 147, REVISION 2, PROVIDES ENHANCED CONTROL (ALARM ONLY INSTEAD OF TRIP) OF BUS VOLTAGES AND PREVENTS BUS TRIPS AND SPURIOUS DIESEL GENERATOR ACTUATION DUE TO OVER-VOLTAGE CONDITIONS.

4

ANO UNIT 1 (NOTE: THIS LER REPORTS TWO EVENTS.)

• • • • • • • w*********

w*t***

DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 313 1989 040 0

9001160051 216389 12/05/89 POWER LEVEL -

000%. ON 12/5/89 AT 0645 AND 12/6/89 AT 2205, WHILE THE PLANT WAS SHUTDOWN IN A MAINTENANCE OUTAGE, AUTOMATIC ACTUATIONS OF AN EMERGENCY DIESEL GENERATOR (EDG) OCCURRED AS A RESULT OF LOSS OF POWER TO A 480 VOLT (V) ENGINEERED SAFEGUARDS (ES) BUS.

PRIOR TO BOTH EVENTS, THE B5 AND B6 480V ES BUSSES WERE CROSSCONNECTED TO FACILITATE MAINTENANCE ACTIVITIES. THE DECEMBER 5 EVENT OCCURRED AS A RESULT OF A PERSONNEL ERROR WHICH OCCURRED WHILE OPERATORS WERE ATTEMPTING TO "SPLITOUT" THE B5 AND B6 BUSSES AND RETURN THE ES POWER DISTRIBUTION SYSTEM LINEUP TO NORMAL. THE ERROR RESULTED IN A LOSS OF POWER TO BUS B6 WHICH CAUSED THE OFFSITE FEEDER BREAKER FOR 4.16 KILOVOLT BUS A4 TO OPEN AND INITIATED A START OF THE 'B' EDG WHICH TIED ON TO THE A4 BUS.

THE DECEMBER 6 EVENT WAS ALSO THE RESULT OF A PERSONNEL ERROR.WHICH CAUSED A LOSS OF POWER TO 480V ES BUS B5.

THIS CONDITION CAUSED THE OFFSITE FEEDER BREAKER FOR A3 TO TRIP AND THE 'A' EDG TO START. THE MOMENTARY LOSS OF POWER TO A3 CAUSED THE OPERATING DECAY HEAT REMOVAL (DHR) PUMP TO TRIP.

DHR FLOW WAS LOST FOR APPROXIMATELY 9 MINUTES AND RESULTED IN A REACTOR COOLANT SYSTEM TEMPERATURE INCREASE OF 17 DEGREES.

MANAGEMENT BRIEFINGS WERE CONDUCTED FOR THE OPERATING CREWS PRIOR TO RESTART FROM THE OUTAGE COVERING THE LESSONS LEARNED FROM THESE EVENTS.

ANO UNIT 1 DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 313 1993 004 0

9311010183 0

10/04/93 POWER LEVEL -

000%.

ON OCTOBER 4, 1993, AT,1647 HOURS, ANO-1 WAS IN COLD SHUTDOWN CONDITIONS WHEN THE RED TRAIN 4160 VOLT ENGINEERED SAFEGUARDS (ES) BUS AND 480 VOLT LOAD CENTER WERE UNINTENTIONALLY DE-ENERGIZED.

IN PREPARATION FOR INTEGRATED ENGINEERED SAFEGUARDS ACTUATION SYSTEM (ESAS)

TESTING, THE TEST LEADS FOR THE RED 4160 VOLT ES BUS FEEDER BREAKER WERE INADVERTENTLY SHORTED WHICH COMPLETED THE TRIP CIRCUIT FOR THE FEEDER BREAKER. THIS RESULTED IN THE FEEDER BREAKER OPENING WHICH DE-ENERGIZED THE RED TRAIN 4160 VOLT ES BUS AND ASSOCIATED VITAL 480 VOLT ES LOAD CENTER. THE EMERGENCY DIESEL GENERATOR (EDG)

AUTOMATICALLY STARTED DUE TO THE UNDERVOLTAGE CONDITION OF THE 4160 VOLT BUS; HOWEVER, THE EDG OUTPUT BREAKER FAILED TO CLOSE.

THE EDG WAS SECURED.

PRIOR TO THE ESAS TESTING, THE GREEN TRAIN HAD BEEN ESTABLISHED AS THE PROTECTED TRAIN, AND THE GREEN TRAIN OF DECAY HEAT REMOVAL (DHR)

WAS AVAILABLE THROUGHOUT THE EVENT.

THE ROOT CAUSE OF THIS EVENT HAS BEEN ATTRIBUTED TO A PROCEDURAL DEFICIENCY IN THAT THE INSTRUCTIONS FOR THE SETUP OF THE ESAS TEST EQUIPMENT WERE INADEQUATE.

CORRECTIVE ACTIONS.INCLUDE PROCEDURE REVISION AND DISCUSSIONS OF LESSONS LEARNED WITH THE APPROPRIATE MAINTENANCE AND ENGINEERING PERSONNEL.

TROUBLE SHOOTING REVEALED THAT THE EDG OUTPUT BREAKER FAILED TO CLOSE BECAUSE OF A FAULTY LATCH CHECKING SWITCH.

THE OUTPUT BREAKER WAS REPAIRED AND RETURNED TO AN OPERABLE STATUS.

5

DAVIS BESSE DOCKET YEAR LER NUMBER REVISION DCS NUMBER NSIC EVENT DATE 346 1990 006 0

9005080090 218131 04/03/90 POWER LEVEL - 000%. ON 4/3/90, AT 0905 HOURS, WITH-THE REACTOR DEFUELED, THE STATION EXPERIENCED A SAFETY FEATURES ACTUATION SYSTEM (SFAS) LEVEL 1 THROUGH 5 ACTUATION. IT WAS INITIATED BY AN ACCIDENTAL BUMPING OF A BREAKER SWITCH. WHEN THE SWITCH OPENED, IT DE-ENERGIZED A BUS AND RESULTED IN THE LOSS OF POWER TO SFAS-CHANNELS 1 AND 3.

THIS CAUSED A FULL SFAS LOGIC ACTUATION. BEING DEFUELED, MOST ENGINEERED SAFETY FEATURE SYSTEMS HAD BEEN DISABLED TO PREVENT INADVERTENT ACTUATION. AS A RESULT OF THE AUTOMATIC OPENING OF ONE OF THE ISOLATION VALVES, THE BOUNDARY OF AN ONGOING HYDROTEST CHANGED, AND AN OPEN DRAIN LINE WAS EXPOSED TO 500 PSIG.

THE CLOTHING OF ONE WORKER WAS CONTAMINATED. OPERATIONS MANAGEMENT HELD MEETINGS WITH ALL CRAFT PERSONNEL TO DISCUSS THE IMPORTANCE OF BEING CAREFUL AROUND PLANT EQUIPMENT AND THE CONSEQUENCES OF THIS EVENT. THE NRC WAS NOTIFIED VIA THE ENS UNDER 10CFR50.72(B)(2)(II).

THIS IS BEING REPORTED AS AN LER UNDER 10CFR50.73 (A)(2)(IV) AS AN AUTOMATIC ACTUATION OF ESF EQUIPMENT.

6