Text

LER 90-016-01:on 900411,motor Control Ctr Relay 28/29-5 Setpoint Drift Occurred,Resulting in Analyzed Plant Condition.Caused by Utilizing Wrong Relay & Inadequate Review of GE Svc Info Ltr.Relay replaced.W/900613 Ltr
	ML20043H952
Person / Time
Site:	Quad Cities
Issue date:	06/13/1990
From:	Bax R, Michael Brown; COMMONWEALTH EDISON CO.
To:	; NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
	LER-90-016, LER-90-16, RLB-90-140, NUDOCS 9006270124
	Download: ML20043H952 (6)
	v • d • e

s

.. j i

V. 1 CommonweaRh Edison

' Quad Cities Nuclear Power Station !

22710 206 Avenue North - 1

_ oorcova, Illinois 61242 Telephone 309/654 2241 J RLB-90-140 l

1 June 13,'1990 l

U. S. Nuclear Regulatory Commission j Document Control Desk -

Washington, DC 20555 )

i l!

Reference:

Quad Cities Nuclear Power Station

. Docket Number 50-265, OPR-30, Unit Two :

l

Enclosed is Special Report 04-02-90-016, Revision 01, for Quad Cities Nuclear 1

Power Station. 1 This report is submitted in accordance with the requirements of the Code of l

' Federal Regulations, Title 10, Part 50,46: The licesnsee shall report the l nature of any significant change to or error' discovered in the limiting i emergency core cooling system (ECCS) analysis anc its estimated effect.

Respectfully, COMMONWEALTH EDISON COMPANY I

QUAD CITIES NUCLEAR POWER STATION

~R..L..Bax.

. Station Manager .

.RLB/MJB/jt 1.

Enclosure 1

,' cc: R. Stols T. Taylor ,

l INPO Records Center l

NRC Region III 1

1 2802H pyg6270124 900613 '

g ADOCK 05000265 PDC / t ,

e ~

~ DEVIATION ICvEsTIGATION REPORT (DIR). Form Rev 2.0 fp Facility Name PAGE

_ Quad Citiet Unit Two 1 lD[.1 0 1 5 Title Plant Found in an Unanalyzed Condition Due To Motor control Center 28/29-$ Relav 5etooint Drif t.

EVENT DATE DIR NUME'ER REPORT DATE ;

MONTH DAY YEAB, .STA UNIT YEAR y//

SEQUENTIAL NUPSER y/

/

REvls!ON NUMBER tiQ!Lill _. D AY YEAR 2

((

/

POWER /

01 4 11 2 91 0 01 4 Di 2 91 0 -

0 l 11 6 -

0l1 01 6 11 3 91 0 01 01 0 f CONTACT FOR THIS DIR NAME TELEPHONE NUMBER AREA CODE M. Brown. Reaulatory Assurance Ext. 3102 3 1019 6i E l4 l- 1212 14 l 1 COMPLETE ONE LINE FOR EACH COMPONEN URE DESCRIBED IN THIS REPORT CAUsE SYSTEM COMPONENT MANUFAC- REPORTABLE CAUsE SYSTEM COMPONENT MANUFAC. REPORTABLE TURER 70 NPRDS TURER TO NPRDS X Ele f i 61 2 Gl01Bl2 Y l l l I l f l I I I I I I I / I I I I I ;

$UPPLEMENTAL REPORT EXPECTED '

t1QfLIll DAY YEAR p

sVBM!ss!ON 1 YES fif ves. comolete EXPECTED SUBMIS$10N DATE) l N0 l TEXT Energy Industry Identification system (E!!s) codes are identified in the text as (xxl l PLANT AND SYSTEM IDENTIFICATION:

General Electric - Bolling Water Reactor - 2511 MHt rated core thermal power.

EVENT IDENTIFICATION: Plant Found In An Unanalyzed Condition Due to Motor Control Center 28/29-5 Relay Setpoint Drift.

A. CONDITIONS PRIOR TO EVENT:

Unit: Two Event Date: April 12, 1990 Event Time: 1545

. Reactor Mode: 2 Mode Name: REFUEL Power Level: 0%

1 REFUEL Mode (2) - In this position interlocks are established so that one control rod i only may be withdrawn when flux ampilflers are set at the proper sensitivity level and the refueling crane is not over the reactor. Also, the trip from the turbine control valves, turbine stop valves, main steam isolation valves, and condenser vacuum are l bypassed. If the refueling crane is over the reactor, all rods must be fully inserted and none can be withdrawn.

B. DESCRIPTION OF EVENT:

On April'll, 1990, Unit 2 was in a Refuel Outage. At 1820, QTS 170-12 MCC 18/19-5(28/29-5) Auto-Transfer Logic Operability SLrveillance, was completed for Motor Control Center (MCC) [MCC) 28/29-5. Part of this surveillance includes timing the transfer from Bus [BU) 29 to Bus 28 during a simulated loss of off-site power (LOOP) and failure of the Unit 2 Diesel Generator (DG) [DG). The transfer was timed at 38.99 seconds.

2800H

o DEVIATION INVESTIGATION REPOR7 TEXT CONTINUATION-

'gi Form Rev 2.0 FACILTTYiAME DIR NUMBER PAGE SEQUENTIAL -

REVIs10N ;

STA UNIT YEAR NUMBER NUMBER Duad cities Unit Two 01 4 Di 2 91 0 ""

0 l1 16 -

01 1 2 0F 0 l$

TEXT EnergyIndustryIdentificationsystem(E!!s)codesareidentifiedinthetextas(Xx) l

)

The acceptance criteria fo1 the time delay was 20 1 5 seconds. This criteria was provided by the Nuclear Engineering Department (NED) in a November 7, 1989 letter.

The letter was written in response to a discrepancy noted while testing Modification 4-1-88-6. This modification was installed so that MCC 18/19-5 would i transfer, if required, upon t. loss of DC power. During th?. modification testing 1 the transfer was timed at 25 50conds. The modification approval letter stated that the transfer time should have be n approximately 15 seconds. NED provided a letter stating that the relay (RLY)[62) setpoint and acceptance criteria was 20 + 5 l seconds. l l

On April 12, 1990, NED was asked to evaluate the 38.99 second time delay transfer. I At 1545 hours0.0179 days 0.429 hours 0.00255 weeks 5.878725e-4 months , after conversations with Nuclear fuel Services (NFS) and General Electric (GE), it was suspected that the time delay was outside the design basis ,

for a loss of coolant accident (LOCA) analysis. The NRC was notified at 1836 hours0.0213 days 0.51 hours 0.00304 weeks 6.98598e-4 months '

using the Emergency Notification System (ENS) phone in accordance with 10CFR50.72 (b)(2)(l). It was later determined that this event did not fall under the 1 10CFR50.72 notification requirement because it was not outside the design basis; it '

was an unanalyzed' condition which did not significantly compromise plant cafety.

On April 25, 1990, this event was determined reportable in accordance with

'10CFR50.46 as a Special Report.

Electrical Maintenance Department (EMD) personnel '!

removed the relay [RLY) (62), reset it to its nominal 20 second setting, and returned it to service. On April 14, 1990, at 1023 hours0.0118 days 0.284 hours 0.00169 weeks 3.892515e-4 months , QTS 170-12 was performed again on MCC 28/29-5 with an acceptable transfer time delay of 21.3 seconds.

C.- APPARENT CAUSE OF~ EVENT:

This event is being reported in accordance with 10CFR50.46: The licensee shall report the nature of any significant change or error discovered in the limiting Emergency Core Cooling System'(ECCS) analysis and its estimated effect. A significant change or error is one which results in a calculated peak fuel cladding ,

temperature different by more than 50'F from the temperature calculated for the limiting transient using the last acceptable model, or.is a cumulation of changes and errors such that the sum of the absolute magnitudes of the respective temperature changes is. greater than 50'F. The licensee shall provide this report '

within 30 days and include a proposed schedule for providing a reanalysis or taking other action-as may be needed to show compliance with 50.46 requirements.

This report will also serve as the final report for Potentially Significant Event (PSE) 90-07.

The cause of this event was setpoint drift. The cause of the time delay drift was utilizing a CR2820 relay in this application. GE SIL 230, Rev. 2, identified that these relays have a tendency to increase the time delay after long periods in a de-energized condition. GE recommended replacing these relays with an Agastat relay.

L 2800H

4

=

DEVIATION INVESTIGATION REPORT TEXT CONTINVATION

, , Form Rev 2.0 FAc!LITY_NAME DIR NUMBER P/LOL SEQUENTIAL REV!s!ON STA UNIT YEAR NUMBER NUMBER Guad Citiet unit two 01 4 01 2 91 0 ~

0l 1 16 ""

0 l 1 1 0F 01E TEXT . Energy-Industry Identification system (E!!s) codes are ice 91f ted in the text as (XX)

A contributory cause was the fact that-the LOCA analysis had changed without adeauate review.- When the LOCA analysis was changed from NEDO 24146A to SAFER /GESTR, the limiting. condition changed from the failure of the RHR injection-valves to failure of DC control power (battery failure) or a DG. With the RHR injection valve failure, MCC 18/19-5(28/29-5) feed transfer time did not need to be considered. With the failure of a DG as one of the limiting conditions, the transfer time became a significant part of the analysis. The SAFER /GESTR analysis became effective on June 23, 1988 for Unit 2.

There was an inadequate review of SIL 230. The relays-were not identified when other CR2820 relays were replaced under modification M4-1(2)-84-33.

i D. SAFETY ANALYSIS OF EVENT:

This relay being set at 38.99 seconds allowed operation in an unanalyzed condition. ;

The SAFER /GESTR analysis assumes 58 seconds for delay in LPCI injection which 1 l includes the following' l

1. 13 seconds for the Diesel Generator to come Up to speed and load on to the bus.

2. 45 seconds-for the Reactor Recirculation (Recirc) (RR) (AD] pump (P) discharge valve (V) to close. The Recirc pump discharge valve is required to go closed on the loop which the Low Pressure Coolant Injection System-(LPCI) (BO) is injecting into to ensure that the flow is into the vessel (VSL][RPV),.through the jet pump nozzles. Flow could otherwise be diverted into-the vessel-annulus via the suction line and back'out the broken recirculation 10o0.

The following times are actJal stroke times for the Recirc pump discharge valves:

1. 1-202-5A 23 seconds ,

-2. 1-202-5B 26 seconds l

3. 2-202-5A 23 seconds

4. 2-202-5B 24 seconds Using the actual stroke times rather than the assumed times, Unit I has a LPCI l Injection time of 51 seconds which includes 25 seconds for the relay setting and 26 !

seconds for the slowest Recirc pump discharge valve. Unit 2 had a LPCI injection time of 63 seconds which included 39 seconds for the as-found relay setting and 24 seconds-for the slowest Recirc pump discharge valve.

1 l

2800H*

-0 .; DEV!ATION INVESTIGATION REPORT TEXT c0NT2NUA710N

~

Form Rev 2.0 8

DIR NUMBER PACE '

FAC,ILITY NAME' SEQUENTIAL REVISION STA UNIT YEAR NLMBER NUMBER o Quad Cities Unit Two 01 4 01 2 91 0 -

0i 1 16' -"

0l 1 4 0F 015 TEXT Energy Industry Identification system (E!!s) codes are identified in the text as (XX)

GE'did an evaluation to confirm a 75 second LPCI injection time would not allow the- l 1 l peak-cladding temperature to exceed 2200'F. The following is a table in 'F that L

compares the analyzed peak cladding temperature (PCT) to the PCT if a 75 second :

I time delay is assumed.

58 sec. 75 sec. LIMIT PCT PCT PCT o NOMINAL 828 980 l-APP K 1377 1660 l UPPER BOUND 1275 1430 1600 LICENSING 1383 1670 2200 l ,

L Calculations show a probability of a simultaneoug LOCA, LOOP, and failure of the unit diesel to supply the bus is less than 2x10-' events per year.

.Before' June 22, 1988, the limiting condition for the LOCA analysis was failure of the Residual Heat Removal-(RHR) [BO) injection valves. The time delay for MCC 18/19-5(28/29-5) was not part of the LOCA analysis before June 22, 1988. l-l' The consequences of this event-are minimal because with a LPCI injection time of 75 seconds, the PCT ^1s well below the PCT limits. In this event, a LPCI injection time of 63 seconds was found, which would have provided an additional margin to the

! ' PCT evaluated.

E. CORRECT 7VE ACTIONS:

l The Unit.2 relay time was adjusted and at 1023 hours0.0118 days 0.284 hours 0.00169 weeks 3.892515e-4 months on April 14, 1990, QTS 170-12 l

was. performed. During.this surveillance the relay was timed at 21.3 seconds.

The Unit 2 relay was replaced on May 1, 1990 with an Agastat relay. QTS 170-12 was repeated and the Agastat relay timed at 21 seconds.

A Justification for Interim Operation was wr1tten for Unit 1 on-April 12, 1990.

The basis for the Justification for~ Interim Operation was that the Recirc pump discharge valves are assumed to close in 45 seconds for the SAFER /GESTR analysis.

Actual closure time is 26 seconds for the slowest valve. The LPCI injection time on Unit 1 is therefore 51 seconds which is less than the 58 seconds assumed in the analysis.

The time delay relays for MCC 18/19-5 will be replaced with an Agastat relay under work request Q83923 (NTS 2652009001601).

NFS has committed to formulating a plan to prevent recurrence of the inadequate review of the analysis (NTS 2652009001602).

QTS 170-12'will be revised so that the Recirc pump discharge valve stroke time is considered. The Rectre pump discharge valve stroke time and the time delay relay setpoint will be added so that the analyzed time, 58 seconds, is not exceeded l

'(NTS 2652009001603).

2800H

i . _ _ _

jf-] t Y- DEVIATION INVEsTIGA710N REPORT TEXT CONTINUATION.

. si Form Rev 2.0 FACILITYhAME- DIR NUMBER PACE SEQUENTIAL REVIs!0N STA UNIT YEAR NUMBER NUMBER And Cities Unit Two 01 4 0} 2 91 0 -

0l1 16 -

0 1 1 5 0F 0lE M: ; TEXT. Energy Industry Identification system (E!!s) codes are identified in the text as (XX)

A' search will be made for other CR 2820 relays that were overlooked during Modification 4-1(2)-84-33 (NTS 2652009001604).

S:tbstantial improvements have been made in the review of SIls since SIL 230 was ,

issued. The programmatic improvements in place now will prevent this event from occurring in the future.

10 CFR 50.45 requires formulating a plan after finding a greater than 50*F error in-the LOCA aralysis. In this case the corrective action was to reset the time delay-to within 1he 20 15 second tolerance and to cha',ge the surveillance procedure to i verify _ that the actual LPCI injection time is bounded by that assumed in the 1

-SAFER /GESTER LOCA analysts -

F .' PREVIOUS EVENTS:

The drift problem on the CR 2820 relays had previously been identified as a. station

' and industry problem. Modification M4-1(2)-84-33 had replaced other CR 2820 relays.- '

Based or, the corrective actions completed and to be completed, no further action is

~ deemed necessary.

G. COMPONENT FAILURE DATA: !

The failed relay was a General Electric relay, part number CR2820B127AA2, Manufacturers number 36080A.

1 L

L r

L l

2800H-

ML20043H952

Text

Reference:

Navigation menu

Search