ML20082P449

Special Rept:On 950320,reactor Startup Occurred W/Source Range Monitor in Degraded Condition.Shutdown Taken by Console Operator

ML20082P449
Person / Time
Site:	University of Missouri-Columbia
Issue date:	04/19/1995
From:	Mckibben J, Meyer W MISSOURI, UNIV. OF, COLUMBIA, MO
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM), Office of Nuclear Reactor Regulation
References
NUDOCS 9504260269
Download: ML20082P449 (3)

Text

-

j \..

Research Reactor Center II i Research Park Columbia, Missouri 65211 UNIVERSITY OF MISSOURI-COLUMBIA w%3g2 i April 19,1995

' Director of Nuclear Reactor Regulation A'ITN: Document Control Desk '

Mail Station PI-37 U.S. Nuclear Regulatory Commission Washington,DC 20555

REFERENCE:

Docket No. 50-I86 University of Missouri Reseamh Reactor License R-103 l l

SUBJECT:

Report as required by Technical Specification 6.1.h.(2) regarding reactor startup with the  !

source range monitor in a degraded condition INTRODUCTION On March 20,1995, a reactor startup was commenced following a normally scheduled maintenance day.

The prerequisite Full Power Reactor Startup Checksheet had been completed, which includes a response i test of the fission chamber detector for the source range channel to verify it is operable as required by Technical Specification 3.4.a.

4 The startup commenced at 2025, after taking the initial set of Nuclear Data and recording source range count rate to be used in the 1/M determination of critical rod height. Initial count rate was 900 eps and Estimated Critical Position (ECP) was 16.9 inches. The stanup procedure (SOP 1I.1.1) requires stopping rod bank withdrrval in 5 inch increments until the rod bank is 2 mehes below ECP to take Nuclear Data.

At 2 inches below ECP rods are pulled individually to establish criticality.

The stanup had proceeded to the point where the rods were withdrawn 10 inches. At this point, the startup was discontinued because the senior operator in charge of the stanup saw no significant change in source range counts from those measured initially at zero inches. A review of reactor stanups for the past year indicates that, at 10 inches, source range count rate has typically increased 30 to 50% The reactor was shutdown by manual rod run-in at 2112 to investigate the lack of expected response.

Technical Specification 3.4.a. requires the source range nuclear instmment to be operable only during reactor stanup. The inadequate response of the source range channel, noted by the reactor operator before tenninating the startup, indicates that the operability of the source range channel was questionable for the 47 minutes the startup had progressed, even though it was response tested prior to startup as required by i the Full Power Startup Checksheet and met the conditions of T.S. 3.4.e. which states:

I "The reactor shall not be staned up unless the Source Range is indicating a neutron count rate of at least I cps and the Wide Range monitor is indicating a power level above I watt or the Souxe Range monitor is indicatinr o neutron test source of at least 2 cps and is verified just prior to stanup by a neutron t st source < e movement of the monitor that the channel is responding to neutrons."

E COLUMBIA KANSAS CITY ROLLA ST. LOUIS t

9504260269 950419 PDR

/, i 0 ADDCK 05000186 l S PDR

g. <e i

leuct to the Director of Nuclear Reactor Regulation April 19,1995 i i Page two *' l DESCRIPTION  !

The source range nuclear instrument detector is a one inch diameter fission chamber with eight inches of sensitive length. The fission chamber is mounted in a water tight container and is located in the pool  ;

outside the reflector region. It is connected to a mechanical drive system with 6 feet of stroke which i allows vertical positioning of the detector prior to stanup. When not in use, the fission chamber is  !

withdrawn into a boral thermal neutmn shield to prevent excessive bumout of the chamber when the j reactor is at power.  !

Technical Specification 3.4.a. mquires the source range nuclear instrument to be operable only during reactor startup. When the reactor stanup was commenced on March 20,1995, the source range monitor  ;

had been response checked as specified in T.S. 3.4.e. and by Step 11 of the Full Power Startup ,

Checksheet by driving the detector in towards the core and verifying increased count rate, and set to -

indicate greater than I count per second. The source range indicated 900 rps and the wide range monitor '

was reading 750 watts, so the initial conditions of T.S. 3.4.e. were met.

The source range channel appeared to be operable before reactor startup, but was determined not to be  !

responding adequately 47 mmutes into the stanup. The reactor was shutdown and the stanup channel was i again res pon.ce tested by driving the movable fission chamber over the length of its travel to verify an increased response when closer to the sour - (beryllium reflector). The source range counts increased as  !

the fission chamber was driven in, but die ecrease as expected when it was withdrawn. The count l rate, compared to data from several pre' artups, was considerably higher than nomial at various  !

fission chamber heights.

Electronics technicians were called in to troubleshoot the difference in response. The source range drawer and preamplifier were checked and were detemlined to be operating normally. The fission chamber detector and cables were removed for inspection and the cabling near the detector was brittle. The cabling from the fission chamber to the preamplifier was replaced and soume range counts returned to the range l logged for several previous startups.  ;

A short form precritical checksheet was performed, the soume range monitor was response checked, and  !

the reactor was started up without funher incident.

A N A LYSIS The discontinued stanup had followed Standard Operating Procedure and, when commenced, met Technical Specifications 3.4.a. and 3.4.e., within the limits of the historical method used to verify operability. This method had simply been to verify count rate increase on the source range channel as the fission chamber is inserted.

In retrospect, this method (which had served us well for 27 years) will detect if the source range channel is not responding, but provides no guidance for determining whether the source range is over responding.

The main concem with over response is that gamma or noise background could obscure the neutron contribution to the count rate such that suberitical multiplication would not be observed.

l

j : (, -l l

1 leuct to the Director of Nuc! car Reactor Regulation.

l- April 19,1995 g 'l Page threc*

t Review of previous stanup records indicate that for normal stanups after maintenance day (12 to 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> after shutdown), source range count rate is 100 cps or less for fission chamber height higher than 20 inches from full insened. The source range detector is usually placed in the 20 to 25 inch range for startups after maintenance day. The count rate indicated for the March 20 stanup was 900 cps, which  ;

exceeded typical count rates and may have been caused by high noise backgmund due to detector cable j insulation breakdown. Electronics technicians checked the source range discriminator circuits and j eliminated them as a cause of the higher than normal response.  ;

The source range channel is required by Technical Specification to be opemble only during reactor stanup. ,

It is not part of the reactor safety system, but provides improved monitoring oflow neutron flux levels at stanup to ensure that suberitical multiplication and criticality can be observed during a stanup. Safety System Protection from a Startup or Continuous Rod Withdrawal Accident (Hazards Summary Report, Addendum 5, pp. 25-29) is provided by the period scrams associated with the two intermediate range l instruments, which were operable.

During the March 20 stanup, the wide range monitor, which also monitors reactor power level down to source range level, was operable and indicating 750 watts. This indication is well above the I watt threshold stated in T.S. 3.4.e. and was capable ofindicating criticality if the stanup had proceeded further than 10 inches of rod withdrawal (6.9 inches below ECP).  ;

CORRECTIVE ACTION j The immediate corrective action was a reactor shutdown taken by the console operator when source range  ;

count rate did not change as expected with the rods withdrawn to 10 inches.

Review of the technique for response checking the source range channel indicated that no limit had ever been established for considering the source range operability degraded when higher than normal response _

is indicated. This has been corrected by a Standing Order to Reactor Operations staff providing a range of  ;

expected values for source range counts based on time after shutdown. Count rates higher or lower than the expected range require investigation by electronics technicians before stanup can commence. This evaluation will be in addition to the current response check of the soume range channel and will be added to the Full Power Stanup Checksheet at the next revision of Standard Operating Procedures.

l Sincerely, ENDORSEMENT:

Reviewed a d Approved i  % A j Walt A. Meyer Jr. J. Charles McKibben Reactor Manager Associaie Director xc: Mr. Alexander Adams, Jr., USNRC Regional Administrator, NRC, Region III Dr. John P. McCormick, Interim Vice Provost for Research and Graduate School Dean, UMC - ,

Reactor Advisory Committee Reactor Safety Subcommittee [

• emesiseoruo .

Commission Em 18,1997