ML20236S171

Ro:On 980615,large & Rapid down-shift on Three Different Meter Indications from One Ni Drawer Was Discovered.Caused by Failed Preamplifier in Wide Range Amplifier Subject Ni Drawer.Replaced Cabling,Detector & Preamplifier

ML20236S171
Person / Time
Site:	University of Missouri-Columbia
Issue date:	07/15/1998
From:	Mckibben J MISSOURI, UNIV. OF, COLUMBIA, MO
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 9807240171
Download: ML20236S171 (3)

Text

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

r t

+

Research Reactor Center 1

Research Park l II, Columbia, Missouri 65211 Telephone (573) 882-4211 l

FAX [5731882=3443 UNIVERSITY OF MISSOURI-COLUMBIA July 15,1998 Document Control Desk U.S. Nuclear Regulatory Commission Washington, DC 20555 i

REFERENCE:

Docket No. 50-186 University ofMissouri Research Reactor License R-103

SUBJECT:

Report as required by Technical Specification 6.1.h.(2)

Description At 1838 on June 15,1998, approximately eleven minutes after arriving at 10 MW following a normal startup, the duty shift operator noticed a large and rapid down-shift on the three different meter indications from one nuclear instrument (NI) drawer. This NI drawer provides indication for  !

channels #1 (source range indication), #2 (intermediate range indication), and #4 (power range indication). This was accompanied by source range low count rate and channel #4 downscale (less )

than 95%) annunciations. Within seconds the duty operator manually scrammed the reactor.

Immediately after the scram, the meter indications for these instruments displayed normal shutdown readings, with the exception of channel #1 period indication which was off-scale low (or negative).

An electronics technician determined that the preamplifier in the wide range amplifier for this NI drawer had failed. The preamplifier was replaced. After the NI drawer and associated indications and trips were tested satisfactorily, a normal reactor startup was commenced. With the reactor at 5 MW in manual control, the duty operator again noted a large and rapid down-shift on the intermediate range and power range meter indications for this NI drawer-intermediate range channel #2 and power range channel #4. The reactor was immediately shutdown by manual scram.

After extensive troubleshooting, the electronics technicians isolated the problem to the detector and cabling, which were replaced. They again replaced the preamplifier in the wide range amplifier.

This NI drawer and associated indications and trips were tested satisfactorily and no further problems of this type have recurred.

With the degraded signal to this NI drawer causing the indications to behave erratically, the associated short period rod run-in and scram (channel #2) and high power rod run-in and scram (channel #4) trips were deemed inoperable, which is a deviation from Technical Specifications 3.4.c \O and 3.3.a.

4 i

I t

g <: ,

~us i

9007240171 TeO715 I' PDR ADOCK 05000186 S PDR i.

______-_________________--________a

n- 1 9.

Letter to Document Control Desk July 15,1998

- Page two Ana1xaia

= The problems occurred with a Gamma-Metrics nuclear instrumentation system. This system has the neutron finx detected by a fission chamber and the resulting signal being processed by a model

. 900301 amplifier before being sent to a model 900302 neutron flux monitor. The neutron flux monitor provides source range, intermediate range and power range indications.

Technical Speci5 cations require three independent high power acrams and rod run-ins and two independent shdtt period scrams and rod run-ins trips. The high power trips are provided by three different power range channels. The short period trips are provided by two_ different intermediate range channels. The Gamma-Metrics NI drawer provides one intermediate range channel and one -

power range channel and associated trips. The safety function of the intermediate range is to .

assure protection of the fuel elements from a continuous startup rod withdrawal accident (Hazard Summary Report, Addendum 5, Section 5). The safety function of the power range is to assure that a safety limit, f der the most severe anticipated transient,is not exceeded (HSR, Addendum 4, .

Appendix H). Each intermediate range channel provides a rod run-in at a 10 see (minimum) period ;

' and a reactor scram at an 8 sec (minimum) period. The power range channel provides a rod run-in at 115% (maximum) power and a reactor scram at 120% (maximum) power. When the Gamma.

-Metrics instrument failed, the reactor was shutdown within seconds. All other nuclear instrumen-

' tation channels-intermediate range channel #3 and power range channels #5 and #6--were operational and would have provided the required scrams and rod run-ins.

, The reactor safety system is dependent on the short period trip protection during reactor startup, .

and the high power trip protection to assure a safety limit is not exceeded. Additionally, the power range channels have a downscale alarm (less than 95%) to warn the operator of a potential nuclear instrumentation problem (e.g., loss of signal or erratic rignal). With the intermediate range and

, power range channels coming from the same NI drawer, the power range downscale annunciation also provides a warning to the operator of a problem with the intermediate range instrumentation.

A previous indication problem with this Gamma-Metrics equipment was apparent by erratic startup counts, which was corrected by replacing the detector and cabling. This most recent failure was .

caused by the detector cable degrading (lower insulation resistance), which in-turn caused the pre- . j amplifier to fail. The degradation is believed to be caused by gamma irradiation. The company .

)

technical service representative verified the cable degradation as the cause of the pre-amplifier i

' failure by using an insulation testing instrument -AEMC Megohmmeter, Model 1100N with a full

. scale reading of 1,000 gigohms. Our electronic technicians had earlier checked the cable and detector resistance but were not able to detect the degradation because our instrument was a model 21 J Megger with a full scale reading of 1,000 megohms.: An AEMC Megohmmeter Model 1100N -l instrumentation has been procured and is available for future troubleshooting. q

. The company technical representative also indicated that the degradation of the cable insulation j would cause the pre-amplifier failure by burning out the R2 resistor in the pre-amplifier. This -

resistor serves as a current limiting resistor to protect the pre-amplifier. The R2 resistor in the two replaced pre-amplifiers was found to be failed open. This knowledge will help in determining and correcting cable insulation degradation problems in the future.

~ The trip settings for the reactor scram and rod run in trip settings for Channels #3, #5 and #6 were

verified to be correct after both times that channels #2 and/or #4 may have been inoperable. These redundant NI channel functions would have caused a rod run-in and/or scra'm at the desired set

! points.' 'Iherefore, the safety system would have pedormed its intended function. It should be r noted that the rod run-ins are not part of the safety system and only serve to introduce a shim blade insertion on a reactor transient before the safety system is activated.

V=

_ __ _2___:__ _ _ _ _ _ _ _ _ . ._.-__= =- - - - - ~ - -- -- -"

s ei . - 6

' Le'tter to k)ocument Control Desk July 15,1998 Page three Corrective Actions At two different times when the duty operator observed a large and rapid down-shift in NI meter -

indication, the reactor was shutdown immediately. After the first shutdown, the pre-amplifier was replaced and the indications and trips from the associated NI drawer were tested and operability verified. After the second shutdown, the detector, cabling and pre-amplifier were replaced, tested, and its operability was verified.

To decrease gamma irradiation damage rate, the possibility of raising vertically, moving further away from the core, and/or shielding the cable / detector connection are being evaluated. A new insulation testing instrument with higher sensitivity was obtained to facilitate earlier detection and-troubleshooting ofinsulation degradation problems.

Sincerely,-

/

J. Charles McKibben

. Associate Director xc: Mr. Alexander Adams, Jr., USNRC Mr. Tom Burdick,NRC Region III Dr. Brady Deaton Dr. Edward Deutsch Reactor Advisory Committee Reactor Safety Subcommittee i

1

- (/14/l4 d CHRISTINE M.ERRANTE l Notary Public -Notary Seal - t STA'IEOFMISSOURI BooneCounty l My commiulon Expires: April 14,1999 ,

i f

d s; i L' -

O .j