Text

LER 97-004-00:on 970221,trip of High Pressure Safety Injection Pump Occurred While Filling Safety Injection Tank Resulting in TS Violation.Caused by Particle Lodged Between Surface of Indication disk.Y-phase Relay Was OOS
	ML18067A443
Person / Time
Site:	Palisades
Issue date:	03/24/1997
From:	Roberts W; CMS ENERGY CORP.
To:	;
Shared Package
ML18067A442	List: ML18067A441; ML18067A443;
References
	LER-97-004, LER-97-4, NUDOCS 9704010031
	Download: ML18067A443 (6)
	v • d • e

f' NRC FORM 366 U.S. NUCLEAR REGULATORY COMMISSION APPROVED BY OMB NO. 3150-0104 (4195) EXPIRES 4/30/98 ESTIMATED BURDEN PER RESPONSE TO COMPLY WITH THIS MANDATORY INFORMATION COLLECTION REQUEST: 50.0 HRS. REPORTED LESSONS LEARNED ARE INCORPORATED LICENSEE EVENT REPORT (LER) INTO THE LICENSING PROCESS AND FED BACK TO INDUSTRY. FORWARD COMMENTS REGARDING BURDEN ESTIMATE TO THE INFORMATION AND RECORDS MANAGEMENT BRANCH (T-6 F33), U.S. NUCLEAR REGULATORY COMMISSION, WASHINGTON, DC 20555-0001, AND TO THE PAPERWORK REDUCTION PROJECT (3150-0104, OFFICE OF

- (See reverse for required number of digits/characters for each block) MANAGEMENT AND BUDGET, WASHINGTON, DC 20503 FACILITYNAME(1) CONSUMERS POyYER COMPANY DOCKET NUMBER (2) Page (3)

PALISADES NUCLEAR PLANT 05000255 1 of 6 TITLE (4) TRIP OF HIGH PRESSURE SAFETY INJECTION PUMP WHILE FILLING SAFETY INJECTION TANK RESULTING IN TECHNICAL SPECIFICATION VIOLATION EVENT DATE (5) LER NUMBER (6) REPORT DATE (7) OTHER FACILITIES INVOLVED (8)

MONTH DAY YEAR YEAR I SEQUENTIAL NUMBER REVISION NUMBE.R MONTH DAY YEAR FACILITY NAME DOCKET NUMBER 05000 9~ FACILITY NAME DOCKET NUMBER 02 21 97 - 004 - 00 03 24 97 05000 OPERATING THIS REPORT IS SUBMITTED PURSUANT TO THE REQUIREMENTS OF 10 CFR§: (Check one or more) (11)

MODE (9) N 20.2201(b) 20.2203(a)(2)(v) x 50. 73(a)(2)(i) 50. 73(a)(2)(iii)

POWER 20.2203(a)(1) 20.2203(a)(3)(i) 50. 73(a)(2)(ii) 50.73(a)(2)(x)

I LEVEL (10)

I 99.~

I 20.2203(a)(2)(i) 20.2203(a)(2)(ii) 20.2203(a)(3)(ii)

20.2203(a)(4) 50.73(a)(2)(iii) 50.73(a)(2)(iv) 73.71 .

OTHER 20.2203(a)(2)(iii) 50.36(c)(1) 50.73(a)(2)(v) Specify in Abstract below or I I 20.2203(a)(2)(iv) 50.36(c)(2)

LICENSEE CONTACT FOR THIS LER (12) 50.73(a)(2)(vii) in NRC Form 366A NAME William L. Roberts, Licensing Engineer TELEPHONE NUMBER (Include Area Code)

(616) 764-2000 COMPLETE ONE LINE FOR EACH COMPONENT FAILURE.DESCRIBED IN THIS REPORT (13)

CAUSE SYSTEM COMPONENT MANUFACTURER REPORTABLE CAUSE SYSTEM COMPONENT MANUFACTURER REPORTABLE TONPRDS TONPRDS y ED RLY G080 YES SUPPLEMENTAL REPORT EXPECTED (14) MONTH DAY YEAR I YES If yes, COMPLETE EXPECTED COMPLETION DATE x I NO EXPECTED SUBMISSION DATE (15)

ABSTRACT (Limit to 1400 spaces, i.e., approximately 15 single-spaced typewritten lines) (16)

On Febn..iary 2.1. 1997, at 2004 hours0.0232 days 0.557 hours 0.00331 weeks 7.62522e-4 months , the plant was stable and operating at approximately 99.6%

power. While supporting performance of Technical Specification (TS) sampling for safety injection tank (SIT) boron concentration, High Pressure Safety Injection* Pump (HPSI) P-66A tripped due to a malfunctioning Y-phase time-overcurrent relay. This resulted in two emergency core cooling components (the safety injection tank and the HPSI pump) being inoperable, a condition prohibited by Technical Specifications (T$), and the plant entered Technical Specification 3.0.3. This event was very similar to the event that occurred on July 17, 1996 andreported in LER 96-010.

Immediate investigation by Operations led to the resetting of the tripped relay by hand. The safety injection tank was refilled *using P-66A, and TS 3.0.3 was exited. The plant was in the TS 3.0.3 action statement for approximately 32 minutes. The plant remained in a 24-hour action statement per TS 3.3.2 due to the inoperability of the HPSI pump, as the problem with the relay reset had not been resolved. The breaker relay was found to have a small metal particle lodged in its operating mechanism. The particle prevented proper reset of the breaker's Y-phase time-overcurrent relay induction disk after the P-66A motor start. After removal of the metal particle, the relay was tested and performed satisfactorily. The Y-phase time-overcurrent relay was cut out of service as permitted by plant design and the HPSI pump declared operable at 0043 hours4.976852e-4 days 0.0119 hours 7.109788e-5 weeks 1.63615e-5 months on February 22, 1997, exiting TS 3.3.2. Subsequently, this relay was replaced with an identical '

model on February 25, 1997, with no further problems identified 9704010031 970324 PDR ADOCK 05000255 S PDR

NRC FORM 366a U.S. NUCLEAR REGULATORY COMMISSION 4/95 LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION FACILITY NAME 11 \ DOCKETl2\

05000255 YEAR ILER NUMBER 16\

SEQUENTIAL NUMBER REVISION NUMBER PAGE 131 20F6 PALISADES NUCLEAR PLANT 97 - 004 - 00 TEXT (If more space is required, use additional copies of NRC Form 366A) (17)

EVENT DESCRIPTION On February 21, 1997, at 2004 hours0.0232 days 0.557 hours 0.00331 weeks 7.62522e-4 months , the plant was stable and operating at approximately 99.6%

power. Performance of monthly Technical Specification (TS) safety injection tank (SIT) boron concentration sampling was in progress for Safety Injection Tank T-82C after tanks T-82A and T-828 had been sampled and refilled. The plant had declared T-82C inoperable due to low level and pressure and entered a one-hour allowed outage time (AOT) per TS 3.3.2.a. to obtain the sample. At this time High Pressure Safety Injection Pump (HPSI) P-66A tripped when started to refill T-82C. The trip of *P-66A concurrent with T-82C's inoperability, was a condition prohibited by Technical Specifications which required that TS 3.0.3 be immediately entered.

Local inspection of P-66A's breaker (152-207) found the Y-phase time-overcurrent relay [SB; ISV]

target dropped. The Shift Supervisor found the relay's induction disk rotated to its trip position.

The disk was reset by hand. P-66A was then startE!d to refill T-82C while observing operation of the relay's induction disk. The relay;s induction disk did not properly return to its pre-start (reset) position with this pump start. Proper safety injection tank T-82C level was restored, T-82C declared operable at 2036 hours0.0236 days 0.566 hours 0.00337 weeks 7.74698e-4 months , and TS 3.0.3 was exited. No reduction in plant power level occurred during the time the plant was under TS. 3.0.3. With 'P-66A still inoperable, the plant remained in a 24-hour action statement per TS 3.3:2.

Electrical Maintenance personnel performed rneggar and phase resistance measurements of the P-66A motor circuit. Values measured were satisfactory and compared well to the same measurements r:nade in July 1996 during the previous event (LER 96-010): The System Engineer performed an operability determination recommending P-66A as operable with the Y-phase time-overcurrent relay cut out of service since the motor protection was still provided by other operable relays. The Y-phase time-overcurrent relay was cut out of service, P-66A was declared operable at 0043 hours4.976852e-4 days 0.0119 hours 7.109788e-5 weeks 1.63615e-5 months on February 22, 1997, and the 24-hour allowed outage time per TS 3.3.2 was exited.

CAUSE OF THE EVENT Apparent Cause:

A small metal particle lodged between the.top surface of the induction disk and the relay's permanent magnet prevented the induction disk from rotating back to its reset position. Sampling had been completed on two of the four safety injection tanks and the HPSI pump had been started twice to fill the tanks as part of the sampling process. The two previous starts had positioned the induction disk near its full travel such that when P-66A was started the third time to fill T-82C, the induction disk completed its travel and eaused the time-overcurrent relay to pick up - resulting in the trip of P-66A.

NRC FORM 366a U.S. NUCLEAR REGULATORY COMMISSION 4/95 LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION PAGE 13\

I FACILITY NAME 11 \ DOCKETl2\ LER NUMBER 16\

YEAR SEQUENTIAL REVISION NUMBER NUMBER 05000255 3 OF6 PALISADES NUCLEAR PLANT 97 - 004 00 TEXT (If more space is required, use additional copies of NRC Form 366A) (17)

Root Cause:

At present, no determination can be made concerning the origin of the metal particle. As a result, no definitive root cause conclusion can be made. Considering the relay housing design, gasketed front cover, and the limited time the cover was removed, it is unlikely that the metal particle was introduced into the relay on site. It is more likely that the particle was introduced during manufacture. - '

ANALYS.15 OF THE EVENT Palisades has four safety injection tanks, one for _each safety injection loop, mounted high in the containment. Each tank contains a volume of borated water under a nitrogen overpressure used as a motive force for injection into the Primary Coolant System (PCS). Periodic sampling of _the boric acid concentration in the tanks requires one tank at a time to be declared inoperable, as sampling often results in draining the tank below its TS limit. The tank is refilled using HPSI. pump P-66A. Tripping of the high pressure safety injection pump during refilling, with the SIT inoperable due to low level and pressure, results in a condition prohibited during plant oper?tion and requires entering TS 3.0.3.

Troubleshooting was pe'rformed on February 25, 1997, on the degraded 150/151-207 Y*phase time-overcurren~ relay while installed in th_e plant. The relay was tested for its .as found!' settings.

Testing of the time-overcurrent section of the relay found that the induction disk would rotate when test current was applied but would not reset when the test current was removed. Subsequent manual movement of the induction disk found that the disk would not reset if moved anywhere within its "as set" rotational range. Visual inspection found a metal particle lodged between the top face of the induction disk and the permanent magnet. The particle appeared flat and was measured to be approximately 1 mm in length at its largest dimension. .When within the field of the pE;irmanent magnet, the metal particle stood up such that it contacted the top face of the induction disk and the magnet-face. While the disk could be rotated by hand or an applied current (motor starting or test current) the disk return spring was not capable of overcoming the drag created by the lodged particle. Once the particle was removed, the time-overcurrent portion of the relay was tested and found to operate properly. The induction disk was no longer sticking and the disk fully returned to its reset position after all electrical or manual manipulations. Although removal of the metal particle restqred operation of the Y-phase time-overcurrent relay, the relay was replaced on February 25, 1997.

NRC FORM 366a U.S. NUCLEAR REGULATORY COMMISSION I

v 4195 LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION .

I FACILITY NAME 11\ DOCKETl2\ LER NUMBER 161 PAGE 131 YEAR SEQUENTIAL REVISION NUMBER NUMBER 05000255 4 OF6 PALISADES NUCLEAR PLANT 97 - 004 '00 TEXT (If more space is required, use additional copies of NRC_ Form 366A) (17)

Possible Source of Metal Particle in Relay The Y-phase overcurrent relay is contained in a semi-flush, panel mounted housing which is attached to the 152-207 breaker cubicle door. With the exception of the removable cover for the housing (which is gasketed), the housing has no openings through whiGh debris could enter. As such, the metal particle was either introduced into the relay when the cover was removed or it must have been inside the relay when rece.ived from the manufacturer.. Inspection of the cover's felt gasket found the gasket in good condition with no obvious intrusion paths for foreign material.

Inspection of the relay housing found no foreign material inside. The cover is known to have been removed on only two occasions since the subject relay was installed*, once during the initial relay installation and during relay troubleshooting performed on October 2, 1996, for a different problem with the same relay. When the cover is removed, the ,procedure generally_ related to maintenance :,

or inspection of relays, System Maintenance and Construction Services procedure PD-01 "Test Procedure for Nondirectional Time-overcurrent Relays", contains inslructions to. remove dust and foreign particles from the protective relay, specifically list!ng the rotating .disk.

Past 150/151-207 Y-Phase Relay Failures On ~uly 17, 1996, ~he. High Pressure Safety Injection Pump P-66A tripped while filling Safety Injection Tank T-82C. The 1996 event was nearly identical to the February 21, 1997 event. Initial inspectiori of the. 1996 event found the 152-207 Y-phase time-overcurrent relay's induction disk at its trip position, which indicated that the relay had been responsible for tripping the pump. Upon opening the breaker cubicle door;, the induction disk returned t6 its reset position. Subsequent troubleshooting could not repeat the event; therefore, the root cause of the event could not be determined. Although malfunction of the relay could not be repeated, the Y-phase time-overcurrent relay was conservatively replaced on July 19, 1996. Therefore, the relay that caused

. -the February 21, 199Tpt:imp trip had beeri installed only seven rn.ontns. -*- . - .. * . - . * - - - * *

As the February 21, 1997, and the July 17, 1996 event were nearly identical, the actions that were taken as a result of the previous event were reviewed to determine if the recommended actions to prevent recurrence were adequate.

Actions.to Prevent Recurrence frbm the July 17. 1996, Event

1.

Inspect similar relavs in the plant for similar indications.

All time-overcurrent relays on C, D, E, A, B, F, and G buses (in excess of 120 relays) were inspected to determine if a similar "sticking" problem existed in other similar relays. All of the relays inspected were found to be properly reset.

NRC FORM 366a U.S. NUCLEAR REGULATORY COMMISSION 4/95 LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION I

-**--FACILITY NAME 11\ DOCKET12\ LER NUMBER 16\ PAGE 13\

YEAR SEQUENTIAL REVISION NUMBER NUMBER 05000255 5 OF 6 PALISADES NUCLEAR PLANT 97 - 004 - 00 TEXT (If more space is required, use additional copies of NRC Form 366A) (17)

2. Review the stock balance of spare motor protection relavs for all safetv-related motors.

Procure necessary spare relays as considered appropriate by System Engineering.

New relays are on order but not yet on site.

3 Utilizing the Industry Experience Program. compare Palisades preventative maintenance.

testing. aging and replacement practices-regarding motor protection relays against the industry standard and revise if necessary.

A NETWORK request for information was submitted September .16, 1996. The maintenance performed and the three-year frequency schedule for Palisades' protective relays aligned well with that reported by the r~sponding plants. The data indicated that the industry calibration *interval was from 18 months or each refueling outage up to an interval of every five years. No changes to Palisades' protective relay maintenance/calibration frequenGies were deemed necessary.

An NPRDS query was performed for the subject relay model, GE 121AC66K20A. The query found only one occurrence of a relay's setpoint found out of specification.

4 Revise SOP 3 to permit the use of HPSI Pump P-668 to fill SITs at full PCS pressure if justified bv engineering analysis.

The procedure to allow this operation to take place is in the process of being revised. Had the procedure revision been completed, it would have provided a means to restore T-82C to service more quickly, but had no direct bearing on the cause of this event.

-- - -* . - ~. -

SAFETY SIGNIFICANCE When P-66A tripped, it rendered one redundant HPSI train inoperable and interrupted the refill of T-82C. Although T-82C was inoperable due to its level being outside of Technical Specification limits during sampling, it is likely that its safety function of delivering water to the PCS could still have been performed with the water that was available. Even if it could not have performed its safety function, the Loss of Coolant Accident (LOCA) analysis for Palisades assumes that only 3 of 4 SITs are available to inject their contents into the PCS .. Since T-82A, B, and D, were operable, the loss of T-82C would not have had any significant effect on the plant's ability to respond to a LOCA. The condition only existed for approximately 32 minutes.

This condition only existed for a short period of time since prior to the attempt to refill T-82C, two of the other safety injection tank samples had been obtained and the tanks refilled by using P-66A the same day.

-.1.

NRC FORM 366a U.S. NUCLEAR REGULATORY COMMISSION 4/95 LICENSEE EVENT REPORT (LER)

TEXT CONTINUATION I

FACILITY NAME 11\ DOCKET12\ LER NUMBER 16\ PAGE 13\

YEAR .SEQUENTIAL REVISION

. NUMBER NUMBER 05000255 60F6 PALISADES NUCLEAR PLANT 97 - 004 - 00 TEXT (If more space is required, use additional copies of NRC Form 366A) (17)

CORRECTIVE ACTION CORRECTIVE ACTION TAKEN AND RESULTS 'ACHIEVED The Y-phase time-overcurrent relay was cut out of service and the pump was returned to operation within the 24-hour LCO ..

The Y-phase time-overcurrent relay was r~placed on February 25, 1997 and returned ~o service. The replacement relay was thoroughly inspected and cleaned prior to installation

.and a relay housing inspection showed no signs of foreign material.

The actions taken in response to the previous similar pump trip event were reviewed to determine their status and possible a.ffect O[l mitigation of this event. It was determined that the identified deficiency with the P-66A p*ump Y-phase time-overcurrent relay that caused the.

February 21, 1997, pump trip is isolated to this relay. Previous investigations and '

inspections have shown that the identified deficiency with the particle restricting reset of the induction disk is not expected to occur with any of the other plant relays .. Similar relays have been*in operation since initial plant startup in 1971 and, with regular inspedions and cleaning, _have been shown. to be reliable.

ADDITIONAL CORRECTIVE ACTIONS PLANNED Nondestructive material analysis of the metal particle which was found to cause improper 150/151-207 Y-phase relay operation will be performed, if possible, to determine if partide*

origin is from the relay. The material's origin may be useful in identifying upgrading controls needed to prevent r~currence, e.g., assuring cleanliness during manufacture, accounting for

--*wea( of the discduring operation, orproviding Foreign Material* ExClusion durihg insfalfahon - - - ---

or testing ..

Since the replacement relay installed on February 25, 1997, is from the same lot as the relay which failed on February 21, 1997, results of the findings of the nondestructive material

analysis will be used to determine what additional actions, if any, are r.ieeded for the replacement relay.

ADDITIONAL INFORMATION A similar event occurred in 1996 and was reported under LER 96-010.

ML18067A443

Text

Navigation menu

Search