ML20195D491

Ro:On 880929,difference of 10 F Between Temp Indications for Primary Loop a & Loop B Observed.Caused by Small Amounts of Resistance Introduced Into Compensation Loop Leads. Connection for Compensation Loop Will Be Double Checked

ML20195D491
Person / Time
Site:	University of Missouri-Columbia
Issue date:	10/28/1988
From:	Alger D, Mckibben J MISSOURI, UNIV. OF, COLUMBIA, MO
To:	Office of Nuclear Reactor Regulation
References
NUDOCS 8811070028
Download: ML20195D491 (3)

Text

. ,

Research Reactor Facility UNIVERSITY OF MISSOURI g,,,a ,ch p,,,

October 23, 1983 T$e"pN$If$[I$

Director of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Mail Station P1-137 Washington, D.C. 20555

REFERENCE:

Docket 50-185, University of Missouri Research Reactor License R-103

SUBJECT:

Report as required by Technical Specification 6.1.h(2) concerning reactor operation with one of three reactor coolant temperature scrams not within Technical Specification limits DESCRIPil0N:

On Septenber 29,1938 at 2030, when coolant temperatures were approaching steady state values as part of a reactor startup to 10 MW, the duty operator observed a difference of 10*F between the tenperature indicati:ns for primary loop A (930A) and loop B (9808). These indications display the tenperature of the primary inlet water which is measured in the outlet p1pe of each of the parallel primary heat exchangers (Hx 503A and Hx 503B) before the two lines join together to form the reactor inlet pipe. The operator scrammed the reactor to investigate the cause of the greater than usual temperature difference.

Investigation by electronics technicians showed that the 9303 temperature indication was reading approxin.ately 10*F below actual temperature. The scram setpoint for the trip of this unit was set at 147.8'F (as verified by Co1pliance Check CP-8B carlier on September 29,1938). With this channel indicating 10*F low, a scram from this channel would require an actual temper 0ture of 157.8'F to be initiated which is not in accordance with Technical Specification 3.3.a which specifies a reactor trip setpoint of 155'F maximum for each reactor inlet temperature device.

ANALYSIS:

Technical Specification 3.3.a calls for three reactor temperature scrams as part of the Reactor Safety System, two reactor inlet temperiture scrams and one reactor outlet temperature scram. The two reactor inlet ta perature scrams are derived from trip units associated ..lth temperature elenents 930A and 930B. The temperature ele.nents for both 930A and 9308 are Roseacunt Model 442A, Type RG. This type incorporates a compensation loop to reduce the error introduced by a variation in the resistance of the resistance sensor lead wire due to changes in ambient temperature (see Figure 1).

981107002G 8G102S 6

%g* y DR ADOCK 0500 16p WFE im (Q, 4 COLUMBIA KANSAS CITY ROLLA ST. LOUIS M a%4i coyuq inc u on

Dircctor of Nuclear Reactor Regulation Page 2 October 23,19S3 r-----, no

.m...,,.sa e os.., ~_

, ,.m . . , _ , , o ,. . . m o m ,,

e,,s.........

s"'n,,ay,v,,o

, g.

__ b .,. l;~n

,. h -- w ~, ,, - ___

.g n I

w, % ,, >

os ii : . o. ... nu .n -

__n_

" 't,'

• • o i co . $oo.

L -- _ _ _J Figure 1 -

During the course of maintenance day activities on September 29, Compliance Check CP-8B, a calibration and scram setpoint check for 930A and 930B nad been performed and completed at 1355. This compliance check includes disconnecting the sensin; element (RTD) from the transmitter and connecting a resistance decade box to the transmitter to sinulate the RTD. After reconnecting the RTD, a final verification that the temperature elements are operational is performed by comparing their indicated temperatures to that of thermocouples in adjacent sensor wells. During the verification step, the 930A RTD indicated a temperature 1.1*F higher than its adjacent thermocouple and the 9303 RTD indicated a temperature 1.7'F lower than its adjacent the rmocoupl e . Both thermocouples were indicating 100.9'F. Tne procedure requires the associated RTD e tnermocouple to be within 2'F.

At 2006 on September 29, 1938, the reactor was placed in autonatic control at a 10 N power level following a startup from the scheduled shutdown for maintenance. At 2030 when coolant temperatures were approaching steady state values, the duty operator nanually scrartned tne reactor when he observed a 10*F difference between the temperature indications provided by 980A and 930B. A few degrees difference in the temperature indicated by 930A and 930B is not unconon, since tne temperature indications are derived at points in two separate cooling legs (Loop A and 1.oop B), out 10*F exceeded the maximum expected difference.

Tne investigation into the temperature diff erence between 0A and 93E included a calibration check of 930S af ter a pnysical check of a , connections that had been disconnected and reconnected earlier as part of Coapliance Check CP-3B. No oovious problems were detected. However, when the 930B leads were reconnected after the calibration check the 930A and 930B indications were in close agreement.

Since no conclusive reason was found for tne lower than actual tempera- '

ture indicated by 9305, a bench test setup was constructed by the electronics tecnnicians using a spare RTD and associated trip indicator. Tne tests were designed to determine if increased resistance, due to loose lugs or oxidation, at any RTD connection coulo cause a reduced temperature indication. As expected, int.reased resistance of the temperature sensor leads caused an increased temperature indication. However, an increase in the resistance of tne leads to the colpensatior, loop of less than 1 ohm could cause a reduced temperature indication of 10'F or more.

Director of Nuclear Reactor Regulation Page 3 October 23, 1938 The results of this bench test showed that the introduction of very small amounts of resistance into the compensMion loop leads would produce a reduced and relatively constant offset in temperature indication across the indicating range. It can be surmised that this increased resistance was due to a higher resistance connection such as oxidation of the connection legs.

The temperature indication, with increased resistance of the coaponsation loop, was still responsive over its entire range but with indicating lower than actual tenperature. With 930B indicating 10'F low, a scram on its channel would have occurred at 157.B'F instead of the 155'F maximum specified in Technical Specification 3.3.a. The reactor safety system was still capable of perforcing its safety function if a high reactor coolant teaperature condition had occurred, since both the redundant reactor inlet high tenperature scram (930A) and the reactor high outlet temperature scram (9013) were operable (Hazards Suamary Report, Addendum 4, Appendix A.3.3.6).

CORRECTIVE ACTION:

Imediate corrective action was taken by the duty operator when he scrarned the reactor to investigate the discrepancy between 930A and 9303 te.nperature indications.

The results of the investigation by the electronics technicians indicates that the continuity of the connection for the coapensation loop is imperative, in order to not have a nonconservative temperature indication (a lower than actual temperature). When reconnecting the RTO after future co1pliance checks, the connections for the compensation loop will be double checked to ensure a good lor resistance connection is made.

Sincerel , g<

1 R du V C' J. C. M Kibbe Reactor Manager Endorienent-Reviewed and Approved 9.hD. ii. Alger Assistant Director, Nuclear Technology and Operations xc: Reactor Advisory Co.uittee Reactor Safety Subco nittee A

lt*y s .m mag tytt: 11011 Ci 8184WI n *( C3.

,,t/gse154Mtit'. alt,1.IlQ gpqtg totJ til%P'.91 NIAI' 'I