ML20236W078

RO 19:on 980720,determined That Reactor Was Operating W/ Inoperable Alarm Circuit on Bridge Radiation Monitor.Caused by Spurious Alarms Due to Temp & Humidity Sensitivity. Maint Procedures CP-216 & CP-219 Were Modified

ML20236W078
Person / Time
Site:	University of Michigan
Issue date:	07/30/1998
From:	Simpson P MICHIGAN, UNIV. OF, MEDICAL SCHOOL, ANN ARBOR, MI
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 9808050062
Download: ML20236W078 (6)

Text

- _ _ _ _ _ _ _. . _ _ _ .

The University cf Michig:n Michig:n M:m:rirl Phoenix Project j Office of the Director i 2301 Bonisteel Boulevard Ann Arbor, Michigan 48109-2100 {

Docket 50-2 License R-28 United States Nuclear Regulatory Commission Document Control Desk l!

Washington, D.C. 20555 l

1 Re: Reportable Occurrence No.19 - Reactor Operation with In-Operable Alarm Circuit on the Bridge Radiation Monitor. q i

July 30,1998 l l

Description ofCircumstances k At approximately 15:45 on July 20 it was brought to the acting reactor manager's attention that the 0000 - 0800 shift crew had disabled the local alarm circuit in the Ford Nuclear Reactor's bridge radiation monitor at approximately 01:00 without authorization. The monitor had been giving spurious alarms due to temperatum and humidity sensitivity. The reactor was shut down at 15:48 and a source check of the bridge monitor was performed. i Results of this check showed that the remote radiation recorder alarm for this system which is located in the control room was also not operating. A further check showed that the radiation recorder alarm for the bridge monitor was incorrectly programmed. The reactor was kept shut down pending further investigation. It should be noted however, that the operators on shift had no reason to believe the radiation recorder alarm for the bridge radiation monitor was not functiona; when they disabled the local alarm.

Discussions with the reactor operators resulted in finding that the bridge monitor had previously given spurious alarms on Friday morning, June 26. They were authorized at that time to operate the reactor with the bridge radiation monitor local alarm tumed off from late that morning until the shutdown at 15:47 later on Friday afternoon . That is why they l thought they could disable the local alarm on July 20. We now believe the reactor may have been operated without a functioning bridge radiation alarm during this period also.

The bridge radiation monitor local alarm was restored to operation prior to the reactor start up on the following Tuesday, June 30.

The bridge radiation monitor is Requimd Safety Related Equipment as defined in the Technical Specifications and is required to be functioning during reactor operation with an alarm setpoint not to exceed 50 mr/hr . The normal setpoint is 40 mr/hr. The reactor was

{ 1 operated for approximately 15 hours1.736111e-4 days <br />0.00417 hours <br />2.480159e-5 weeks <br />5.7075e-6 months <br /> Monday, July 20 and very likely was operated for approximately six hours on Friday, June 26 without an alarm function on the bridge radiation momtor.

j The radiation measurement capability of the bridge radiation monitor was not disabled at I any time during these periods. The bridge monitor radiation levels are checked by the operators and recorded on the Operational Checklist every two hours when the reactor is 1 operating. A review of the chart record for the two periods when the alarm function was defeated showed normal radiation levels for the bridge radiation monitor during the intervening time between Operational Checks.

k s

DO eda O2

. 2 Time When Full Compliance was Restored Full compliance was restored on Tuesday, July 21,1998.

Immediate actions taken The enclosed figure is a block diagram of the bridge radiation monitor and associated cintuitry. The gamma radiation measurement circuit in the monitor outputs to a local meter, an intemal alarm circuit (local alarm) and the radiation recorder. The radiation recorder accepts inputs from the various Tech. Spec. required radiation monitors that are located throughout PMUFNR and gives the control room operator the ability to remotely observe the radiation monitors. In addition, the radiation recorder also has alarms that can be separately programmed for each input. These alarms indicate at the recorder and

. supplement the local alarms. The radiation recorder also has three alarm output relays, two l

of which are programmed to actuate under various circumstances. Output relay 1 is connected to actuate the High Reading Radiation Recorder (HRRR) alarm on the control panel. Output relay 2 is connected to the Stack Alarm on the control panel. All of the radiation monitor channels have radiation recorder alarms programmed. These alarms are

prograrraned to actuate recorder output relay I giving the HRRR alarm in addition to the

recorder alarm and local alann. The PML stack 2 gaseous ac'ivity t monitor (Stack GAD) is l also programmed to actuate recorder output relay 2 giving the stack alarm.

The calibration of the bridge radiation monitor was checked on Tuesday, July 21. It was l during this calibration check that the spurious alarms were isolated to the local alarm circuit in the monitor. The monitor was removed from service and replaced with a calibrated fully functional monitor unit.

The radiation recorder alarm programming was checked after the discovery that the alarm l was not functioning on July 20. The check showed that the bridge monitor channel alarm programming was incorrect and that the recorder alarm and HRRR alarm were not active.

The correct programming was restored. Between July 21 and July 27 all radiation monitoring instmments were given source and functional tests to verify their alarm capability.

j A memo (Attachment 1) was routed to all FNR/PML staff reminding them that absolutely l no alarms are to be bypassed or alarm set points modified without the explicit approval of j reactor management. The memo also reminded them of the need to properly document any L changes that are made.

Calibration and Maintenance Procedures CP-216 and CP-219, the calibration procedures for the facility's area radiation monitors, have been modified to include a function test of the HRRR reactor console alarm.

l Root Causes and Long Tenn Actions to Prevent a Recurrence Two items have been identified as root causes. First, an inadequate implementation of procedure CP-308, Equipment Out of Operation. An immediate change was made to CP-308 instmeting operators to promptly inform the On-call Supervisor or reactor management of faulty or improperly operating equipment. A further, more broad modification will be made to CP-308, after due considerat,on,i that will provide reactor management a

. 3 inechanism to track actions taken to repair or replace faulty equipment in a timely manner.

A memo (Attachment 2) has been routed to the staff regardm, g implementation of CP-308.

The second root cause is an incomplete understanding of the importance of assuring that all equipment, set points, alarms, and surveillance requirements stipulated in the FNR License and Technical Specifications are met and the absolute requirement to obtain authorization prior taking any potentially compromising action in regard to the same. A closed book training exercise (test) on the License and Technical Specifications will be administered to remind all Operator and Senior Licensed personnel of the Tech. Spec. requirements and the need for compliance.

Sincerely, e

Philip A. impson , Assistant Reactor Manager for John C. Lee, Interim Director Michigan Memorial- Phoenix Project I

cc: Theodore Michaels, USNRC Project Manager Thomas Burdick, USNRC Region III i

l l

L--____-_-____-_-_-_____

. 1 Bridge Radiation Alarm system Radiation Recorderin Control Room (0-50 mv output signall m

Measurement CifCult -

4 Altfm

• 88eys 1 l Y pmenaamm ans M - > analog and . out #1

oe, eigtat campey m

~

alarm reCorner alarm ouuMt red hght 4

D W ey Bridge Monitor (Ludlum 300)

Contros Room GM tube detector conow Room coneo6e Annunciator Alarm &

High Readmg Raseton Reconsor

Michig n Mem:ri:1 - Phoenix Project Phoenix Memorial Laboratory Ford Nuclear Reactor 2301 Bonisteel Boulevard Ann Arbor, Michigan 48109-2100 Tel. 734-764-6220 Memo to: Licensed FNR Staff From: Bernard Ducamp, Phil Simpson Date: 28 July 1998

Subject:

Bypassing of Alarms and Logbook Entries / Documentation Absolutely none of the alarms or alarm set points that are required by the Technical Specifications are to be bypassed or changed without explicit approval from the On-call Supervisor or reactor management unless allowed by Technical Specifications or approved ,

facility procedures.

All such changes MUST be appropriately documented either in the Reactor Logbook, procedure data sheet, or Equipment Out of Serv;ce Log.

I I

l i

- - - - - - _ - - - - - - - - - - - - - - - - - - - - ~ - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -- a I

Michig:n M:m: rill - Phoenix Project Phoenix MemorialLaboratory Ford Nuclear Reactor 2301 Bonisteel Boulevard i Ann Arbor, Michigan 48109-2100 L Tel. 734-764-6220 l

Memo to: Licensed FNR Staff From: Bernard Ducamp, Phil Simpson l

l Date: 28 July 1998

Subject:

CP-308 Equipment Out of Service An entry MUST be made on the Equipment Out of Service Log prior to performing any non-routine maintenance, trouble-shooting, or repair of reactor systems described in the License, Technical Specifications, Systems Descriptions, or Safety Analysis Report.

Reportable Occurrence 19 would not have happened if CP-308 had been properly implemented.

I i

i

_ _ - _ _ - _ - _ _ _ _ _ _ _ _ _ _ - - _ _ _ _ _ - - _ _ - - - _ - - - - - - _ - - _