ML20058M742

Ro:On 930907,determined That Scram Setpoint for Low Flow Scram for HX 503B Leg Below LSSS of 1625 Gpm Required by TS 2.2.Caused by Defective Alarm Trip Unit.Defective Alarm Trip Unit Replaced W/Spare Unit

ML20058M742
Person / Time
Site:	University of Missouri-Columbia
Issue date:	09/27/1993
From:	Mckibben J, Meyer W MISSOURI, UNIV. OF, COLUMBIA, MO
To:	Office of Nuclear Reactor Regulation
References
NUDOCS 9310060374
Download: ML20058M742 (4)

Text

.

Research Reactor Facility Research Park Columbia, Missoun 65211 Telephone UNIVERSITY OF MISSOURI-COLUMBIA e

a September 27,1993 Director of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Mail Station P1-137 Washington, DC 20555

Reference:

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

Subject:

Report as required by Technical Specification 6.1.h(2) concerning reactor operation with the scram setpoint of one of five safety system channels that provide scrams for low Primary ,,

Coolant flow not within Technical Specification limits due to equipment failure  ;

Int roduction ,

On September 7,1993, while performing compliance testing of the Primary Coolant flow safety channel connected to flow transmitter PT-912E, the scram setpoint for the law flow scram for the heat exchanger 503B leg was below the Limiting Safety System Setpoint (LSSS) of 1625 gpm required by Technical Specifier tion 2.2. This Primary Coolant Flow Scram is one of four required by Technical Specification 3.3. An additional backup to these Primary Coolant low flow scrams is provided by the core difTerential pressure scram. The reactor safety system was capable of perfor ming its safety function if an actuallow flow condition had occurred while operating at full power, because the remaining four safety system channels that provide scrams for Low Primary Coolant Flow were operable.

These additional safety channels are: (1) Primary Coolant Flow from FT-912A;(2) Heat Exchanger Differential Pressure for each heat exchanger leg (DPS 928A and DPS 928B) and Core Differential Pressure (DPS-929).

Descriotion On September 7,1993, the reactor was shutdown for scheduled maintenance. Compliance testing (Compliance Procedure, CP-4B) of the Primary Coolant flow safety channel connected to flow transmitter, FT-912E, revealed that its scram trip setpoint was cccurring at slightly greater than 1400 gpm. This was below the 1625 gpm LSSS for this heat exchanger leg of the primary coolant system. The trip point for the scram from FT-912E is normally set for 1725 25 gpm. Two additional scram trip tests were performed and each scram trip verified the trip at slightly greater than 1400 gpm. The last compliance test (CP-4B) of the safety system channel for FT-912E before ,

the September 7 failure was completed on May 20,1993.

%gf M ^400% j Eb E COLUMBIA KANSAS CITY ROLLA ST. LOUIS /h. p)

% ,_m-~~ /f I dr g /WO 7G ,W , 9310060374 930927 9 ga ^*" **ge6 4

_ -~ - - - -- -- - -. . . - . ..

4 t

Letter to Director of Nuclear Reactor Regulation

- September 27,1993 Page two ,

The calibration test portion of CP-4B showed that flow transmitter FT-912E was operating i properly. The problem was determined to be in the alarm trip unit 920C (GE Model 560) which ,

initiates the scram trip for the safety system channel connected to flow transmitter FT-912E (see Fig.1).  ;

The electronic technicians replaced alarm trip unit 9200 with a spare. This trip unit is part of a dual alarm unit which also includes 920D, the trip unit for pool system low flow. The scram .i function for the pool system was tested to be operating at its specified setpoint before removing the dual alarm unit. After the spare dual alarm unit was installed, the compliance testing of the scram  ;

trip for each unit (920 C & D) was performed to validate each scram trip point was in specification.

Additionally, on September 7,1993, the compliance procedure (CP-4A) was performed for the-Primary Coolant Flow channel connected to IG912A (the safety system channel for primary coolant flow developed from the flow orifice in the opposite heat exchanger leg from PI'-912E). The  ;

compliance testing of the three other Primary Coolant Low Flow scram channels (DPS-928A, DPS-928B, and DPS-929) were last performed on June 1,1993. The compliance tests of these channels revealed their trip points to be within specifications.

l Annivsis

~'

The electronic technicians investigated the alarm trip unit and found that a capacitor had failed (the capacitor was in two pieces)in the amplifier of the alarm trip unit. This capacitor I

(capacitor "F"in Figure 2) prc.vides a positive feedback path for the amplifier which ensures positive '

latching of the amplifier circuit. Any change in input initiating a relay change (to alarm or reset)is magnified by this capacitor to prevent relay " hang-up" around the alarm point.

Electronics technicians were able to verify that the problem was a relay latching problem by a l series of bench tests where capacit. ors of various capacitance values were substituted and the relay action was always positive and at the preset trip point. When the unit was tested with no capacitor the rehy would trip, but at a lower value than the set point. The relay action was also very sluggish and accompanied by relay chattering when the trip setpoint was approached. Review of electronics related maintenance history revealed that this was the first failure of this type to occur with the GE Model 560 dual alarm trip unit.

Primary coolant flow rate, reactor power, pressurizer pressure, and primary coolant temperature are the set of measurable operating (process) variables used to develop the reactor safety limit curves for the MURR. The limiting safety system setpoints (LSSS) are the settings specified for each of these variables, so chosen that automatic protective action will correct the most severe abnormal situation anticipated before a safety limit is exceeded.

Hazards Summary Addendum 4, Appendix II, Bases for LSSS for Modes I and II operation provides postulated transients with three process variables at their LSSS and allow the fourth to depart from its IESS to show the safety margin inherent in the LSSS value. An example that shows the conservatism of the primary coolant flow ISSS, postulated Mode I operation with i pressurizer pressure reduced to the LSSS of 75 psia, reactor power and coolant inlet temperature raised to their LSSS of 12.5 MW and 155 F, respectively. The safety limit curves predict that the .

primary coolant flow rate could be reduced to approximately 2400 gpm before DNB would occur, implying a safety margin of 800 gpm be!ow the LSSS of 3200 gpm on coolant flow through the core.

Since 50 gpm of primary coolant flow is diverted to the cleanap system, the LSSS of 1625 gpm per loop corresponds to the actual core flow rate of 3200 gpm for 10 MW ( Mode I) operation. ,

m

3 2

Letter to Director of Nuclear Reactor Regulation

September 27,1993 Page three This example demonstrates that even under extreme conditions (as opposed to steady state nominal conditions of pressurizer pressure greater than 78 psia, reactor power less than 10 MW, and coolant inlet temperature 120 F)if flow had been reduced in one heat exchanger leg of the primary coolant system to 1400 gpm, the reactor would not have violated a safety limit. If a loss or '

reduction of primary coolant flow had occurred any of four additional safety system channels were operable to perform the safety function at primary coolant flow rates (or differential pressures equivalent to primary flow rates) greater than required by the primary flow LSSS. There was no period of full power operation with reduced flow since the last satisfactory test of this safety system.

channel scram on May 20,1993.

Corrective Action Immediate corrective action involved replacing the defective alarm trip unit with a spare unit -

and performing the compliance testing of the apare alarm trip unit for both primary coolant flow and pool coolant flow. Follow.up action to determine the cause of failure was performed by electronics '

technicians. The positive feedback capacitor in the trip unit amplifier was found to have failed -

open. As a precautionary measure, the feedback capacitor in each of the two Model 560 alarm trip units in use and the spare unit will be replaced by January 1994. The compliance testing of the

• pool flow portion of the Model 560's in use is scheduled for December 1993, so the capacitor replacements will be done in conjunction with these tests.

i S,i cerely, Walt A. Meyer, Jr.

Reactor Manager ,

ENDORSEMENT:

Reviewed and Approved sk/ y- s J. Charles McKibben ,

Associate Director t

Attachment:

Figs. I and 2 n 1 j -

xe w/ene: Regional Administrator, NRC, Region III Reactor Advisory Committee

> ~

[f './L W 4ti "<s=q q M;].g 4p' . e

-3 Reactor Safety Subcommittee castm N.traANTE 1 NOTARY PUBLIC STATI 0F MISSQJRI BOONE COUNTY MY COMMJssiaN Exp. APA. 14.1095 l

l l

i 1

4 .. .

-*' PRIMARY FLOWLOOP B Ha 6038

• RCW B.aen

s. . a I,

...-.3,,.

acw ,

a,T. rwa . .in og.

Rosemourit riOOM i14 NSTRUMEta CABNET

&U Q~:P I SQ ROOT oO G i e in POWER SUPPLY CONUITER O gp no e,ig EP he Sin ,, ,, , it le 1. 3 -

Nb 00@- 000 s i si 3 5 !

. . .l - . ~

ia i zoli i i i s aa 120V, euM2 .20V, SOHZ

.emsu.c ow . ,

y wo4,e -

Elf 0"rc t 6 6i6 6 2 O-is isjer is E r-P -

rmyOER iQ +

Ot SU '

epio 2

_.g "D" gq)AL qq "C" , sp en, essa see. essa $s

}

__4,

,,s,- - - .6-

, , , , , .. i. . . . . . . i

-98 000 000 000 l D II c- . s. _a e 'W //

g ; g-f r ? t - -

I

• i f f j j 1 an xu s =o [i N s20V 0+E #

1DN.ner ifbV. GONZ

.m a gu.s.a ,n's g? '/ .rus tems

.- i.m> ,

-t e . i s . .uma e . . o.i s so,.. e.

w (* T est PL U S) 7

,n X 8 '

l 3r , a

=

s- T ~"y >-

ergm.

[=_ _j zgJ ,

c i , dH g [ ,g

.u ..i . _.  ! - <

.. q c  :

l

'%, -CD--- 1

~

i 1

.i

,4

_s PhIy

...... r t. _ g _ . ,

~

_.J , .

n n,n.,

i

......i u(, , , . ,, 4, n ,.& l ,.

>>.. ..- , , , , . . . . .. e s.~ ..

.x,.

2<.

... .. n rm ana Figure 2. Model 560 Alarm Trip Unit