ML20086T757

RO 50-20/1995-4:on 950720,operated W/One Shim Blade Fully Inserted.Caused by Failure of Licensed Operators to Follow Written Procedure for Investigating Mismatch.Failed Proximity Switches Replaced

ML20086T757
Person / Time
Site:	MIT Nuclear Research Reactor
Issue date:	07/28/1995
From:	Bernard J, Lau E, Newton T MASSACHUSETTS INSTITUTE OF TECHNOLOGY, CAMBRIDGE
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
RTR-REGGD-01.016, RTR-REGGD-1.016 50-20-1995-4, NUDOCS 9508030147
Download: ML20086T757 (5)

Text

-

I

• 2  % i NUCL EAR REACTOR LABORATORY AN INTERDEPARTMENTAL CENTER OF b,[WW[ flk MASSACHUSETTS INSTITUTE OF TECHNOLOGY O. K. HARLING 138 Albany Street, Cambridge, Mass. 02139 4296 J. A. BERNARD, JR.

Director Telefax No. (617)253 7300 Director of Reactor Operatons Telex No. 9214734AIT-CAM Tel. No. (617) 25'i 4211/4202 July 28,1995 U.S. Nuclear Regulatory Commission Washington, D.C. 20555 Attn.: Document Control Desk

Subject:

Reportable Occurrence 50-20/1995-4, Operation with One Shim Blade Fully Inserted Gentlemen:

The Massachusetts Institute of Technology hereby submits this ten-day report of an occurrence at the MIT Research Reactor (MITR) in accordance with paragraph 7.13.2(d) of the Technical Specifications. An initial report was made by telephone to Region I (Mr. Stephen W. Holmes) on 07/20/95.

The format and content of this report are based on Regulatory Guide 1.16, Revision 1.

l 1. Report No.: 50-20/1995-4 l 2a. Renon Date: 28 July 1995 2b. Date of Occurmnce: 19 July 1995

3. Frilily: MIT Nuclear Reactor Laboratory 138 Albany Street Cambridge,MA 02139 l

l l

l b .$ li n i .,  %

l 9508030147 950728 PDR i

S ADOCK 05000020 g I

g PDR

' Reportable Occurrence 50 20/1995 4 Page 2 0f 5 -

4. Identification of Occurrence:

Operation of the reactor without all blades within 2.0 inches of the operating position, MITR Technical Specification No. 3.11.2c, and failure to adhere to written procedures for operation of systems involving nuclear safety, MITR Technical Specification No. 7.8.1.

5. Conditions Prior to Occurrence:

The reactor had been prepared for startup in accordance with the " Full Power Startup" checklists (PM 3.1.1.1 and 3.1.1.2). The " blade in" indications -

for two shim blades (#3 and #4) were out-of-commission because of malfunctioning proximity switches. These switches are installed in the reactor core tank and are actuated by permanent magnets attached to the blade assemblies. All other indication of blade position is obtained from the servo-mechanism that drives the blade. An in-core experiment, the Boiling Coolant Chemistry Loop (BCCL) had been installed the previous day. Two licensed senior reactor operators (SROs)  !

were in charge of the stanup. One served as the console operator-in-charge and the other as the supervisor. In addition, two individuals who were in training were present. One of these individuals performed the actual blade manipulations under .

the direct supervision of one of the licensed SROs.

6. Descrintion of Occurrence:

Reactor startup began at 1017 with the estimated critical position (ECP)  ;

calculated to be 7.90 inches. This position was attained at about 1115. (&ig: The approach to the ECP was made in very small increments because this startup was performed by a trainee. As a result, individual blade reactivity contributions could .

not be distinguished. If larger incremental movements of the blades are made, as is y normally the case, then the dynamic period caused by such movements is readily observable.) Withdrawal of the blades continued until the shim bank reached 8.40 inches at about 1120. The startup was then halted in accordance with the existing MITR procedure PM 2.3.1, " Normal Reactor Startup," because the ECP had been exceeded by 0.50 inches. Step 11 of this procedure requires that the shim bank be l lowered by 1.0" and states " sample reactivities should be checked and each blade should be driven inward in sequence while monitoring the neutron level. This will confirm if each blade is in fact coupled to its magnet drive." This action was not done. Instead, after some discussion between the two licensed SROs, it was i decided that the reactivity difference was due to the BCCL contribution not having ,

been properly estimated. Instead ofinserting the blades one inch as per procedure, I the blades were again withdrawn and the reactor was taken critical at a shim bank height of 9.60 inches. This was 1.70 inches above the ECP. The additional reactivity was 1.44 beta. An experiment was then run at low power for about one hour. Power was then increased and the reactor was brought to 4.8 MW at 1359.

At this time it was noticed by the licensed supervisor that the AT across the core was reading higher than would be expected at 4.8 MW. A heat balance calculation was then perfonned and it was determined that once equilibrium thermal conditions ,

were attained (~24 hours) the thermal power would be 5.04 MWt. Reactor power t was immediately lowered to 4.5 MW. Operation continued at 4.5 MW until 1720 when the operator-in-charge performed a reshim and noticed that movement of  ;

blade #4 caused no reactivity effect. That is, its movement did not cause a period.  ;

S 4

i

. " Reportable Occurrence 50-20/1995-4

. Page 3 of 5 The reactor was then manually scrammed. (Note: A shift change had occurred i 1500-1600.)

7. Descriotion of Apparent Cause of Occurrence:

The main cause of this occurrence was the failure of the licensed opeters to follow the written procedure for investigating a mismatch between the calculated and observed ECPs. This is step #11 of the MITR procedure PM 2.3.1, " Normal Reactor Startup." There were, however, several contributing factors, as listed below.

Because of an electrical interaction between the magnet current circuitry and one of the period channels, magnet currents are routinely lowered to about 60 mA compared with the normal setpoint of 80 mA. This is done in order to reduce noise on the period channels prior to startup. In this case, the currents on magnet #4 and

1. 6 were reduced to 60 mA. The drop-off current for each blade is tested before each startup with typical values being 20-25 mA. Accordingly, it was felt that 60 mA would be more than adequate for blade withdrawal. Testing subsequent to this occurrence determined that the minimum current necessary to reliably pick up -

blade #4 is 64 mA. It is about 30-35 mA for the other blades. Because the " blade -

in" indication on blade #4 was out-of-commission, there was no immediate indication that blade #4 was not attached to its magnet. (Note: The " blade in" indication is generated from proximity switches installed in the reactor core tank.

Given the hostile environment in which these switches operate,it is not unusual for them to fail). t Uncertainty in the reactivity worth of the BCCL was the reason that the ECP was thought to be in error by the licensed operators. This reactivity had been measured on several occasions to be -407 m . In November of 1993, the in-core section of the BCCL was modified (lead replaced with aluminum as per SR#-0-93-10) and the reactivity was remcasured to be +200 m . This +200 m value had been the subject of some skepticism and because this was the initial ,

startup after the BCCL was reinstalled for the current fiscal year, the ECP error had been attributed by the operators to this uncertainty.

It was subsequently noted that blade #4 had not been picked up during a startup on July 11, 1995. This problem was properly diagnosed and corrected prior to the reactor's being brought critical. This occurrence was not communicated to all of the reactor operators, including the two licensed SROs who conducted the startup on July 19,1995.

8. Analysis of Occurrence:

Following the reactor scram at 1720, analyses wer'e done to determine evidence of fuel damage. These included core purge and primary coolant samples, visual fuel inspection, and linear flux strip chart examination for high frequency noise (boiling). No evidence of fuel damage was found. In addition, blade #4 was inspected for mechanical binding. None was found.

A power density and thermal-hydraulic analysis was also done to determine )

if any limits on the core power distribution had been violated (MITR Technical

" Reportable Occurrence 50-20/1995-4

.- Page 4 of 5 Specification No. 2.1) during operation with blade #4 full in. This analysis ,

showed that there were no violations and that the core was operated below incipient 1 boiling at alllocations.

9. Conective Action:

Temporary corrective actions subsequent to the occurrence included the daily analysis of primary coolant water for evidence of fuel element failure and that performance of all reactor startups with either the Director of Reactor Operations or the Assistant Superintendent for Reactor Operations present in the control room.

Also, the failed proximity switches were replaced. However, another has since failed.

A meeting of the MIT Reactor Safeguanis Committee was convened on July 24,1995, to discuss this occurrence. The Committee's recommendations were as follows:

1) Development of a method for better communications among all operators, especially regarding abnormal situations such as that which occurred on 07/11/95.

2) Management action to ensure that all personnel follow facility procedures.

3) Determination of the minimum current required to pick up each blade. This was done on July 25,1995.

4) Development of a checklist to guide operators when the reactor is not critical within 0.5" of the ECP.

5) Development of a special procedure for verification that a blade is connected to its magnet when the " blade in" indication is out-of-commission.

6) Management action regarding resolution of electronic equipment problems.

i 1

l

Reportable Occurrence 50-20/19954 Page 5 of 5

'10. Failure Data:

ROR #50-20/1980-2 involved the binding of a shim blade during r'. actor -

startup.

Since ,

Q Thomas H. Newton, Jr., PE Edward S. Lau, NE Asst. Superintendent for Engineering Asst. Superintendent for Operations MIT Research Reactor MIT Research Reactor f 4. d ohn A. Bernard, Ph.D.

Director of Reactor Operations MIT Research Reactor JAB /gw cc: MITRSC USNRC - Senior Project Manager, NRR/ONDD USNRC - Region I - Project Scientist, Effluents Radiation Protection Section (ERPS)

FRSSB/DRSS l

4