ML20011F440

LER 90-001-00:on 900123,fire Alarm Received for Fire Zone in Cable Spreading Room,Resulting in Operability of Deluge Sys. Continuous Fire Watch Not Established.Caused by Inadequate Procedural Guidance.Detectors disabled.W/900222 Ltr

ML20011F440
Person / Time
Site:	Oyster Creek
Issue date:	02/22/1990
From:	Fitzpatrick E, Godknecht M GENERAL PUBLIC UTILITIES CORP.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-90-001-01, LER-90-1-1, NUDOCS 9003060035
Download: ML20011F440 (5)

Text

m

.[ N $

I OPU Nuolent Corporation Ng 7 Post Office Dox 388 Route 9 South i

i Forked River,New Jersey 087310388  ;

609 971 4000 Writer's Direct Dial Number: j February 22, 1990 l

U.S. Nuclear Regulatory Commission ,

ATTN Document Control Desk W:shington, DC 20555 i

Dear Sir ,

t

Subject:

Oyster Creek Nuclear Generating Station Docket No. 50-219 i Licensee Event Report i

This letter forwards one (1) copy of Licensee Event Report (LER)

No.90-001.

Very truly yours, h

E. E. Fit rick Vice President & Director Oyster Creek

EEF MH $c (0705A

01)

-Enclosures l cci Mr. Williare,T. Russell, Adminisktrator Region 1 U.S.' Nuclear Regulatory Commission 475 Allendale Road King of Prussia, PA 19406 Mr. Alexander W. Dromerick U.S. Nuclear Regulatory Comminpion Washington, DC 20555 NRC Resident Inspector

. Oyster Creek Nuclear Generating Station Forked River, NJ 08731

/ ' \

9003060033 900222

[DR ADOCK 05000219

)

Pnn dh GPU Nuclear Corporation is a subsidiary of the General Public Utilities Corporation l

=" NuC6c.2...vLATon, --

Nee. . . u "N AspetOyto Onse es0. StNtes

. UCENSEE EVENT REPORT (LER) eAciufv i.4ent m oocalT Nousen m ====

Oyster Creek, Unit 1 0 l 5 t o 10 l 0 l 211 19 1 loFl 014

" ' ' ' ' ' Failure to Set Continuous Fire Watch Due to inadequate Procedural Guidarce and Personnel Error 4vlett DATOISI Lth stuneDem 101 AGPOft? DATB (Fi OTMtm P ACILifits serv 0Lvt0101 Dochti Nuessthist MONTH DAY vtan vtAm *8(% '4 , ",e upgyu pay ygan .assutymiawns 0 151010101 .I 1 0l1 2l3 90 9l0 0 l0 l1 Op 0l2 2 l2 9 l0 o,s,o,o,o, , ,

,,n ,,,,, TMis escom? is suomiTTID PURSUANT TO TMt al0ulAIMINTS OF 10 C9 A O (Case e*w or mom of 84. 8WMw'atl HH j Moot m N ,,,,,,, ,,,,,,, ,,,,,,,,,,,.i nitini i

~ ~^ ~

~

n esswonn mmi.nu m.nemm nstw i Te? l i n esswmm

. wnniii.

] wwwm

.nw n,

] m.nwme.m

.nwmi nAi

{ gggg A,

_ 2 _

~

nauwnnw -

unwmm ~

ee.nwan.mne n esswn n.: 1 so.n o me.m m.nemm Special Report LittNBit CONT ACT WOm TMig Lth H35 Aat A coes l M. 00dknecht 6 l0 p 971 1 1-i 1 41 11819 C0esPLtti ON8 LING #0m t Acte SOnePONINT f AILunt Ot0Chitt0 IN THIS R9 PORT 112) e m CAUSE SYSTtv COMPONINT

"(N($C. ypy,Y,a CAvlt syrTIM COM*0NINT "(${$C- 0 mk" @

S$h I I I i 1 I 1 i i e i  ! t i .

1 I I l l l I I I I i 1 l 1 M

$UPPLEMtttTAL htp0RT E XPICTt0 (14 MONTH DAY YlAR SUSWiS8 EON neteno svea,rw0N oens

~) vts tier . ~"] NO l l l A,4T a Act m.,. ,,x . . . ,.. .w ,,, ,, ,, ,-,,,. , H .,

on January 23, 1990, a trouble alarm was received for a fire zone in the cable Spreading Room. Initial investigation could not determine tne cause of the trouble alarm, but failure of the detector subsystem was not suspected. Investigation on February 8, 1990, found that all detectors

! for the subsystem were disabled and therefore the deluge system for this zone would not have automatically initiated, making it inoperable. A continuous fire watch is required to be set when an automatic deluge system is inoperable, but this watch was not established until February 8th. The failure to set the fire watch was caused by inadequate procedural guidance. The failure to take prompt action to evaluate and correct the problem is attributed to personnel error. This event is

considered to have minimal safety significance because the three other fire detection subsystems in this room would have alarmed in the event of a fire. Operators investigating the alarm would still have initiated the affected deluge system manually. Applicable procedures will be revised as needed and a critique will be conducted to evaluate the human performance problems. In addition, this LER satisfies Technical Specification 3.12.A.2.b and 3.12.C.3 requirements that a Special Report be submitted if the instrument / system is not restored to operable status within 14 days.

ye,, -

= .. _ ....m. v. ~ 1 Y UCENSEE EVENT REPORT (LER) TEXT CONTINUATION

<movse ous =o vio-om  !

, im.e ames  ;

nomev name m occmet =uusen m t.. mue,en m mes e  ;

ns. V "%l,2;' ^

"*;.t?: i Ovster Cresk U2.it t i vert n a . =u se m aea mm n

Date of occurrence The condition described within this report occurred between January 23,

• 1990 and February 8, 1990.

Identification of occurrence 1 on January 23, 1990, a trouble alarm (EIIC-FRA) was received for a fire zone in the cable Sprea, ding Room. A maintenance request was initiated to [

investigate the trouble alarm and an hourly. fire watch was set.  ;

Subsequent testing performed on February 8, 1990, found that all the detectors (EIIS-KP-DET) for an entire fire zone detection subsystem were ,

disabled and that the automatic fire suppression system (EIIS-Kp) for this i area would not have automatica.11y initiated, making it inoperable.

Technical Specifications require a continuous fire watch be stationed when '

an automatic deluge system is inoperable, but this watch was not set until February 8th. The automatic deluge system was inoperable for 16 days without a continuous fire watch being stationed.

This event is considered reportable under 10CFR50.73(a)(2)(1)(B). In addition, this LER satisfies Technical Specification 3.12.A.2.b and  !

3.12.C.3 requirements that a Special Report be submitted if the instrument / system is not restored to operable status within 14 days.

Condit'idne Prior to Cecurrence i

The, reactor was at 66.7% power, with n generator load of approximately 412 megawatts electric. '

pagglintion of occurrence on January 23, 1990, at 2015 hours0.0233 days <br />0.56 hours <br />0.00333 weeks <br />7.667075e-4 months <br />, a trouble alarm was received in the Control Room for one of the two fire zones in the Cable Spreading Room.

Investigation by the Operations Department could not determine the cause of the trouble alarm because the only indication on the local fire panel (EIIS-KP-PL) was a panel trouble alarm. There was no indication that any detector problem was present and none was suspected; however, the immediate action for an out-of-service fire detection system was performed. A maintenance request was submitted to investigate and correct the cause of the alarm, and an hourly fire watch was set. Subsequent investigation of the problem was not conducted until February 8, 1990.

Testing performed on this date determined that all of the detectors for the affected detection subsystem were inoperable due to a loss of the high voltage power supply (EIIS-KP-JX) to all detectors in the subsystem. The loss of an entire subsystem of detectors disableQ the automatic function ,

of the associated, deluge system making it inoperable.

gn.= 3.=

a

4 OfAC ,orm JBS. U S 8.UCLE.3 E 41ULf.fonY Comme. Neon

?

N .

UCENSEE EVENT REPORT (LER) TEXT CONTINUATION ***aoveo oue o mo-o*

.......m

. un m ooc. . . ..m 1,,,,.,,, ,.., m ma "t!,0. To*J." ,

= Oyster- Creek, Unit l' . .

0 l5 l0 l0 lo l 2 l1 p 9l0 --

0l0l1 --

0l 0 0l3 0F 0l4 nm s . .=== me n- mu w im q Plant Technical Specifications require that a continuous fire watch be  !

~

stationed whenever the deluge system for this area is rendered inoperable; however, this watch was not stationed until after the investigation on ,

February 8, 1990. Investigation into this event also shows that for an eight hour period on January 28, 1990, the hourly fire watch tour was not made.

Annarent cause of occurrence The failure to initially station a continuous fire watch was apparently caused by inadequate procedural guidance. Discussion with various operators revealed that they were unaware that a panel trouble alarm could be indicative of a complete failure of all detectors monitored by that panel. The procedures relating to the Fire Protection System also do not discuss the potential loss of all detectors in a subsystem when the trouble alarm is roceived, nor are they specific in defir.ing how a loss of  ;

an entire subsystem of detectors could impact the automatic suppression system associated with that zone. By. procedure, the Supervisor did initiate the request to investigate and correct the cause of the trouble  ;

alarm and instituted an hourly fire watch to monitor the area. Applicable procedures do not give direction to evaluate the operability of the automatic suppression system based on the condition of the detectors in the-zone, q Personnel error contributed.to the event and to the failure to take prompt.

action to correct the-proble,m with the fire protection system in that the rc ential-significance of the trouble alarm was not recognized. Past experience with trouble alarms on the fire protection panels indicate that they have not resulted from a failure of all the detectors in an entire fire zone subsystem.

Analysis of Occurrence and Safety Sianificance g The Fire Protection System for the cable Spreading Room is made up of two zonou each containing two cross-zone detection subsystems, and an q: automatic' deluge system. When any detector in one of the zones is activated, an alarm is received locally at the Cable Spreading Room and remotely in the Control Room. When a detector in the other subsystem for the affected zone activates, the deluge system for that zone will automatically initiate. The failure of either detection subsystem in a zone would prevent the automatic initiation of that zone's deluge system.

The deluge systems can also be manually initiated.

9 IoNU MO

^

o . .

ac * % === vi =vem unumoav co==.o

Y UCENSEE EVENT REPORT (LER) TEXT CONTINUATION awaoven ous =o mo-oio.

. exe.aes emme  ;

MUTV M He 00CR47 wu=Sem os LOR NUMSGA ISI - Paes ut

'**a " $'X ." -

l"J.?.*f '

Oyster Creek, Unit 1 o l5 l0 l0 lo l 2l1 l 9 910 -- O p l1 --

0 l0 0[4 oF 0 [4  ;

Tenina . s mac = me wnn This event is considered to have minimal safety significance. The second ,

detection subsystem for the affected zone remained operable with no problems during the entire period associated with this event, and would

~have actuated an alarm in the Control Room if a fire had started in this -

room. Additionally, because this is an enclosed room, the two detection  ;

cubsystems in the other fire zone would also have alarmed and would have initiated the deluge system associated with that zone. Operators

• investigating fire alarms'for this room would still be able to initiate

• the deluge system in the, failed zone manually. Although a continuous fire watch was not stationed in this room, an hourly fire tour was established r in the Cable Spreading Room during the period of the event with the exception of an eight hour interval on January 28, 1990. Analysis also indicates that if this event happened in any other mode no significant safety concern would be raised.

(' Corrective Action

\:

l -Short Term Upon discovery that all of the detectors for an entire fire zone detector j subsystem were disabled, a continuous fire watch was stationed and I immediate maintenance was initiated to repair the system. System p operability was restored on February 9, 1990. ' Guidance will be provided L to the operators regarding the interrelation between the fire detection I and automatic suppression syht' ems.

l i Lona Term l

1. Applicable procedures will be reviewed and revised as necessary.

2. A critique of this incident is being conducted and the corrective actions to resolve human performance problems will be determined and

' instituted as a result of that critique. ,

i Similar Events h 87-024 " Failure to Post Fire Watch for a Non-Functional Fire Barrier due to Personnel Error in Falling to Follow Procedure" w

$=M