ML20012D160

LER 90-003-00:on 900215,seven Control Rods Withdrawn During Reactor Startup While Rod Minimizer Was Bypassed.Caused by Personnel Error.Withdrawal Stopped & Keylock Switch Placed in Normal position.W/900314 Ltr

ML20012D160
Person / Time
Site:	Oyster Creek
Issue date:	03/14/1990
From:	Cervenka P, Fitzpatrick E GENERAL PUBLIC UTILITIES CORP.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
LER-90-003, LER-90-3, NUDOCS 9003260548
Download: ML20012D160 (4)

Text

. .

OPU Nuoleer Corporot6en Ng f Post Office Box 388 Route 9 South Forked River.New Jersey 08731-03B8 609 971-4000 Writer's Direct Dial Number:

March 14,1990 U.S. Nuclear Regulatory Commission ATTN: Document Control Desk Washington, DC 20555 Dear Sir Subjects Oyster Creek Nuclear Generatiag Station Docket No. 50-219 Licensee Event Report This letter forwards one (1) copy of Licensee Event Report (LER) No. 90-03.

rhrul&yourf '

'.hitzpatrick President & Director Oyster Creek EEF BDe dmd

(/ docs /dmd/0705A 1)

Enclosures cci Mr. William T. Russell, Administrator Region I U.S. Nuclear Regulatory Commission 475 Allendale Road King of Prussia, PA 19406 Mr. Alexander W. Dromerick U.S. Nuclear Regulatory Commission Washington, DC 20555 NRC Resident Inspector Oyster Creek Nuclear Generating Station Forked River, NJ 08731 34 IS 9003260p4890h3

{DR ADOCK 05 0219 s

[gh PDC y / (f

\\\ \

GPU Nuclear Corporation is a subsidiary of the General Public Utilities Corporation 1 ,

Ma

u o muc6e.a monaron, emaseem

=

• • 3 a ovee one no . .

. . LICENSEE EVENT REPO~.T (LE2)

. .a uev,. nam ni esensi muemen m - - =

Ovst er Creek. Unit 1 0 Il 10 t o I O I 21119 1 l0Fl0 13 T'T68 ** Control Rods Withdrawn During S tartup with the Kod Worth Minimizer Bypassed Due to Personnel Error evoert mee m 6 n museen m monom, wee ivi or=en eaciurm wwo6veo et estHitu h gav vtaa vtan "OM 6 , 4$ woerta par vtaa saputy mansu Waef numetaiol 01:101010 1 i l

~ -

Ol2 Il5 90 910 0l 0l3 0l0 0l 3 1l4 9l0

, nu 0 151 0 1 00 1 1 I

,,,,,,,,,, t== nosome se suewereo euneua=, to vs.: meouine =,e o, is cea , <en . .= w

"*"*8 m mem se an. m.rewenwi rs.rien g _

m aswnno unwm _

m.reweim _

ts.rtw ne, i t m .mnim _

une.,e, un.,en.m>

_ _.v e.,y,,, y,,,,

y _

manwnna L u n menn anwen.mnai aeans m innw een wenn es.temen.eini

. ,n , i een.ien., eenwen.,

tecetese co8vtaet Po86 T=le Lln fiel

=a.: ,etuao= =uu en Paul Cervenka 610 to 417 1 1 l- t 41 R i o h eens teve o=e une eon enca coneone=, saituae oeseniseo a t== apont nei Cauol SY$teW COM* owe =t gf Cause tvrtow cosie 8 owe =t g[ 'kE .

I I I I I I I I t i I I I t l I I I I l l l i 1 1 1 I I

' N '4 ouwse.onvatme oataweeveon. ,,

wowta oav ,pn sueueio=

vee rif v . esemen fJrPfCTt0 guesseesON Ontil =0 l l l ass,aae,ru . . ,., _ . . .,. .- . , o e.

On February 15, 1990, at approximately 2010 hours0.0233 days <br />0.558 hours <br />0.00332 weeks <br />7.64805e-4 months <br />, seven (7) control rods were withdrawn during a Reactor startup while the Rod Worth Minimiter was by-passed. The cause of this occurrence is attributed to personnel error on the part of the Group Shift Supervisor and the control Room operators for failing to observe rod worth minimiter indications. A contributing cause to this occurrence was a procedure deficiency. There were no procedural or administrative controls to govern the position of the keylock switch during startup. Upon discovering that the rod worth minimizer was bypassed, the reactor startup was halted. The Rod Worth Minimiter was unbypassed, the rod pattern was verified to be correct and the startup was resumed. Long term corrective action will consist of incorporating the incident report as Required Reading for all Operations personnel. Procedural changes will be made that will specifically require that the RWM keylock switch be verified to be in the

" normal" position prior to each startup.

I -**" u. _ . .. .

UCENSEE EVENT REPORT (LER) TEXT C;NTINUATION amoven ome =o vie-o*

o cuan ames pastuTV eseMB su pecstt gutett e past e ten muassen es

,,,, p..a..

" " " . 6 -"

Oyster Creek, Unit 1 ' "

0 l5 l0 [0 l0 !2 l1l 9 91 0 -

O l0 l 3 -

01 0 O!2 0' 013 vare u == . =sa .me==,ase a menw em DATE OF OCCURRENCE The date of occurrence was February 15, 1990, at approximately 2024 hours0.0234 days <br />0.562 hours <br />0.00335 weeks <br />7.70132e-4 months <br />.

IDENTIFICATION OF OCCURREliQI The first seven (7) control rods were withdrawn during a Reactor startup, while the Rod Worth Minimiter by-passed.

This condition is prohibited by Technical Specifications and is reportable as defined in 10CFR50.73(a)(2)-(i)(B).

CONDITIONS PRIOR TO OCCURRENCE The Reactor had been shut down to repair a Reactor Recirculation Pump seal.

The pump was repaired, the procritical checks were completed, and a Reactor startup was about to occur.

DESCRIPTION OF OCCURRENCE On February 15, 1990, in preparation for a Reactor startup, all of the necessary procedural precritical checkoffs were completed to permit a Reactor startup. Included in these checkof f s were steps to confirm that the Rod Worth Minimiter was loaded with an approved withdrawal sequence, initialized, and operable. The Rod Worth Minimiter (RWM) (EIIS Code IV) normal / bypass switch (Component HS) had been placed in bypass during the previous power operation, as required. The Croup Shift Supervisor, when completing the checkoff for Rod Worth Minimiter (RWM) operability, did not observe that the normal /by-pass swit chw' as still in the by-pass position. In bypass the RWM is inoperable.

At approximately 2010 hours0.0233 days <br />0.558 hours <br />0.00332 weeks <br />7.64805e-4 months <br />, a Reactor startup commenced. Two Control Room operators at the main control panel placed the mode switch to startup, and began withdrawing control rode (Component ROD) in accordance with the approved programmed sequence., Neither of the Control Room operatore noticed that the RWM keylock switch was in the "by-pass" position, nor did they notice that on the RWM full core display screen the "RWM By-Pass" function block was in an abnormal state (highlighted with a red background). During rod withdrawal, a third Control Room operator (CRO) on shift questioned why all of the control rods associated with the sub-group weren't highlighted with a magenta outline as they usually are. The CRO presenting the question and the other CRO at the main panel evaluated the indicators. It was then discovered that the RWM keylock switch was in the "by-pass" position. The Reactor startup was secured at 2024 hours0.0234 days <br />0.562 hours <br />0.00335 weeks <br />7.70132e-4 months <br />.

APPARENT CAUSE OF OCCURRENCE The cause of this occurrence is attributed to personnel error on the part of the Group Shif t Supervisor and the control Room operators for failing to observe and utilize RWM indications.

g ... -

=== me. u . ucu.a ..

UCENSEE EVENT REPORT (LER) TEXT CONTINUATION *** oven ome o vio-o*

. . o inne sm

,aoui, ass o, -eacnei nummen e a. .. is

....  ?'tg.*6 "l'3?;

Oyster Creek, Unit 1 016 l0 l0 l0 l 2!1 l9 91 0 Ol0 l 3 -

OI O O!1 0' Ol4 venu _ . - e w s. nn A contributing cause to this occurrence was a procedure deficiency. A modification replaced the RWM during the 12R Outage. Due to the state of the art design of the new RWM (computer driven), the existing system lineup requirements in the operating procedure appeared to be no longer necessary and were deleted. The operating procedure became an informational instruction guide for the operator interface but contained no guidelines for establishing, documenting or determining operability of the RWM. The " approach to criticality" procedure checklist, which is completed prior to Reactor startup, was never revised and still referred to the operating procedure and the old system lineup requirements. New procedures were written for configuration control to ensure that onif approved sequences could be loaded into the RWM.

The evolution that loads the approved soquence requires that the normal /by-pass switch be placed in the "by-pass" position and then utilizes a menu driven computer program to actually load the sequence. There were no procedural or administrative controls to govern the position of the keylock switch during the loading evolution.

ANALYSIS OF OCCURRENCE AND SAFETY ASSESSMENT The. Rod Worth Minimiser provides automatic supervision and, thus, conformance to the specified control rod withdrawal sequence. The design basis of the Rod Worth Minimiter is to serve as a backup to procedural controls, to limit control rod worth so that in the event of a control rod drop, the reactivity addition rate would not cause damage to the primary coolant system due to a pressure excursion, or result in significant fuel damage. During this event, the second CRO at the main control panel was independently verifying control rod selection and movement in^ accordance with the designated control rod sequence. The seven (7) control rods that were withdrawn while the RWM was by-passed were withdrawn in accordance with the designated sequence.

Therefore, safety significance of this event is considered minimal.

CORRECTIVE ACTION Immediate corrective action consisted of stopping all control rod withdrawal, placing the RWM keylock switch to the " normal" position, and verifying that the correct sequence was followed for the oeven (7) rods withdrawn. The resctor startup was resumed at 2135 hours0.0247 days <br />0.593 hours <br />0.00353 weeks <br />8.123675e-4 months <br />.

Long term corrective action will consist of incorporating the incident report as Required Reading for appropriate Operations personnel. Procedural changes will be made that will specifically require that the RWM keylock switch be verified in the " normal" position prior to each startup. In addition, changes will La made that will define the requirements for operability.

SIMILAR EVENTS 79-06 Failure to Use Rod Worth Minimizer g.o.. -

- __-_---_-_ _-_ ________-_- _ __-__