ML20211B167

Special Rept on 990614 Re Test of Scram Set Point of Two of Five Low Primary Coolant Flow Safety Sys Channels.Caused by Trip Set Points Being Too Close to TS Limits.Compliance Procedure 23 Will Be Revised

ML20211B167
Person / Time
Site:	University of Missouri-Columbia
Issue date:	07/13/1999
From:	Mckibben J, Schoone A MISSOURI, UNIV. OF, COLUMBIA, MO
To:	NRC (Affiliation Not Assigned), NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 9908240237
Download: ML20211B167 (4)

Text

..

Research Reactor Center Research Park Columbia, MO 6521i University of Missouri-Columbia

,1,0v (373) ggygy 3, t ax (573) 882-6360

,, n .

July 13,19N Director of Nuclear Reactor Regulation ATTN: Document Control Desk Mail Station F1-37 U.S. Nuclear Regulatory Commission Washington, DC 20555

REFERENCE:

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

SUBJECT:

Report for information purposes concerning test of the scram set point of two of five low Primary Coolant Flow safety system channels INTRODUCTION On June 14,1999, Compliance Procedure 23 (CP-23), which tests three of the five primary coolant flow / differential pressure safety channels, was performed. In the first performance of the compliance test of the scram obtained from the differential pressure across primary heat exchanger 503A, the measured serem set point for the low flow scram was 2 gpm below the Technical Specification 3.3 limit of 1625 gpm. This Primary Coolant Flow Scram is one of four required by Technical Specification 3.3, two from each of the two primary heat exchanger flow paths. An additional backup to these Primary Coolant low flow sc ams is provided by the core differential pressure scram. The core differential pressure channel, DPS-929 was tested as part of the same test procedure and on the first test war, found to be 19 gpm below the 3200 gpm required by Technical Specification 3,3. Without making any adjustments, these two differential pressure channels were re-tested and provided scrams that were within the Technical Specification 3.3 limits. Even if these two channels had been inoperable, there would have been no failure of the safety system because the remaining three 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, the flow transmitter for the calibrated orifice plate in the 503A heat exchanger leg; (2) Primary Coolant Flow from FT-912E, the flow transmitter for the calibrated orifice plate in the 503B \I heat exchanger leg; (3) Heat Exchanger Differential Pressure for heat exchanger 503B leg )

(DPS-928B). O o @Ull 9908240237 990713 f ADOCK 0500 6

/

k . h

[ .

l l

. Director of Nuclear Reactor Regulation

, July 13,1999 Page two

• ' DESCRIPTION On June 14,1999, the reactor was shutdown for scheduled maintenance. CP-23 [to test the Primary Coolant Flow safety channel connected to differential pressure transmitters DPS-928A, DPS-928B and DPS-929, which measure the difTerential pressure across heat exchanger 503A, heat exchanger 503B, and the core portion of the primary coolant system located in the reactor pool respectively) was being performed. To correlate the differential pressure trip point for each transmitter to a flow rate, CP-23 is a dynamic test conducted with the primary system operating. The primary coolant pump outlet diaphragm valves (540A/B) are incrementally throttled to reduce flow. The scram is received from the respective differential pressure trip unit when the differential pressure reaches the trip set point due to decreasing flow. CP-23 tests are done at six-month intervals.

In performing CP-23 for each trip unit, the differential pressure and a corresponding flow rate are recorded after the scram is received. The corresponding primary coolant flow indication is derived from the FT-912A and FT-912E differential pressure measurement across the calibrated flow orifices in the primary coolant legs going to each heat exchanger.

FT-912A measures the flow rate through heat exchanger 503A and FT-912E measures the flow rate through heat exchanger 503B. For the 929-trip unit, the flow indication from FT-912A and FT-912E are combined.

The CP-23 procedure has a calibration test performed on the respective DPS-928A, DPS-928B or DPS-929 transmitter before completing the dynamic test. All three differential pressure transmitters were in calibration. A Moore FCA and a lioneywell Vutronik dual monitor switch or dual trip unit initiate the scram trip for the safety system channel connected to flow transmitter DPS-928A and DPS-929 respectively.

The first test of the scram trip set point for DPS-928A occurred at 1623 gpm with a 3.85 psi indicated pressure. This was below 1625 gpm, the primary coolant flow limit in Technical Specification 3.3. CP-23 requires the scram trip point limits for DPS-928A to be 1675

• 25 gpm. Prior to adjusting the trips, the scram test was conducted two more times and the tests showed the trips for DPS-928A to be 1646 and 1645 gpm with an indicated pressure of 3.9 psi both times. The previous compliance test performed on DPS-928A had been completed on  !

December 14,1998 with a scram set point of1692 gpm and an indicated pressure of 3.95 psi.

The DPS-928A trip unit also has a different trip set point to cause the 546A/B convective cooling valves to open, but there is no Technical Specification limit on this trip. The natural convection valves 546A and 546B trips functioned properly.

The first test of the scram trip set point for DPS-929 occurred at 3181 gpm with a 25-psi pressure indication. This is below Technical Specification 3.3 limit of 3200 gpm. The CP-23 trip set-point limits for the scram from DPS-929 are 3300 50 gpm. DPS-929 was re-tested once prior to adjusting the trips and the test indicated the trip at 3255 gpm and an indicated pressure of 25.5 psi. The previous compliance test performed on DPS-929 had been I completed on December 14,1998 with a scram set point of 3319 gpm and an indicated pressure of 26.0 psi.

The Electronics Technician adjusted the trip set point of the scram on the Moore FCA to the 1675

• 25 gpm range. After this adjustment, the DPS-928A scram test was performed again and the scram occurred at 1667 gpm with an indicated pressure of 4 psi. He also adjusted the trip set point of the scram on the Honeywell Vutronik monitor to 3300

• 50gpm range.

After the adjustment, the DPS-929 scram test was performed again and the scram occurred )

at 3291 gpm and an indicated pressure of 26.2 psi.

J

. Director of Nucl=r Re:ctor Regulation

. July 13,1999 Page three On' June 2'1 and July 12,1999, the scram set-point test portion of CP-23 was again performed for DPS-928A and DPS-929. In these tests the trip units performed within the procedure limits of1675 25 gpm and 3300 50 gpm respectively. On June 21st, the scram set point for DPS-928A was 1667 gpm and an indicated differential pressure of 4.0 psi. For DPS-929 the set point was 3291 gpm and an indicated differential pressure of 26.4 psi. On July 12th, the scram set point for DPS-928A was 1662 gpm and an indicated differential pressure cf 4.1 psi. For DPS-929 the set point was 3310 gpm and an indicated differential pressure of 26.5 pai. Additionally, on June 14,1999, the compliance procedure (CP-4A) was performed for the Primary Coolant Flow safety system channel connected to FT-912A (the safety system channels for primary coolant flow developed from the flow orifice in the 503A primary heat exchanger leg). The compliance tests of this channel revealed its trip point to be within specification.

ANALYSIS CP-23 is a dynamic test conducted with the primary system running and the scram coming from the differential pressure measurement across a heat exchanger in the case of DPS-928's and across the core for DPS-929. The primary coolant pump outlet diaphragm valves (540A/B),large isolation valves not designed well for throttling, are throttled to reduce the primary coolant flow. The scram is received from the respective DPS-928 or DPS-929 trip unit when the differential pressure reaches the trip set point due to decreasing flow. The corresponding flow rate is obtained from the flow indication derived from the differential pressure sensors across the calibrated flow orifice plates. The 540 valves are hard to adjust.

The flow is decreasing when the scram is received. The differential pressure and flow are obtained when flow has leveled off at a new lower flow rate. The values obtained are conservative; they are either at or below the actual values that cause the scram trip. If these valves are throttled too quickly or in too large of a step, a larger conservative error in readings can be created. This can explain why on subsequent tests that measurements above Technical Specification limits were obtained prior to adjusting the trips. The procedure does not address the need for smallincremental steps with a time delay between steps when approaching the trip set point.

While the scram trip set points were within Technical Specification limits, the set points on the Moore FCA switch for DPS-928A and on the Honeywell Vutronik dual monitor switch for DPS-929 appear to have drifled down prior to the test of June 14,1999. The two re-tests (June 21 and July 12,1999) have indicated no further significant drift.

Ifit were assumed that DPS-928A and DPS-929 were inoperable and a loss or reduction of primary coolant flow had occurred, each of the three additional operable safety system channels would perform the safety function at primary coolant flow rates (or differential pressures equivalent to primary flow rates) greater than required by Technical Specification 3.3. Therefore, there was no period of full power operation with flow below Technical Specifications

• 1imits since the last satisfactory test of this safety system channel scram on December 14,1998. l

)

7-

~

. Director of Nucirr Rcactor Regul: tion

, July 13,1999 Page four ROOT CAUSE There are three root causes:

)

e To determine the flow rate for the scram set points requires a dynamic test. The flow is decreasing when the scram occurs but the flow rate cannot be determined until the flow rate is not changing. It is difficult to make small flow changes in this system. This can cause the measured flow rate to be below the flow rate that caused the scram.

. The trip set points are too close to the Technical Specification limits. The potential l conservative error in the test and/or a drift in the set point can cause the measured set i point to be out of specifications, f

. The need for smallincremental steps with a time delay between steps when approaching l the trip set point has not been addressed in either the procedure or through training.

CORRECTIVE ACTION DPS-928A and DPS-929 were re-teeted to validate that they would actuate within Technical Specifications limits prior to readjusting the trip set points. The corrective action necessary to be within the CP-23 limits was the adjustment of the trip settings and subsequent re-testing to ensure the trips were consistently within the set-point limits. Follow-up action included testing the trip of the FT-912A Primary Coolant Flow channel to verify that it functioned at the set point. The DPS-928A Primary Coolant Flow channel and DPS-929 core differential pressure were re-tested on June 21" and July 126 and each time the trips were within the proper set-point limits of 1675 25 gpm and 3300 50 gpm. The DPS-928A Primary Coolant Flow channel and DPS-929 core differential pressure will be re-tested again during August to ensure that the channelis not drifting. CP-23 will be revised to set the trip set-point limits for DPS-928A and DPS-928B to 1725 25 gpm and for DPS-929 to 3400 50 gpm.

Additionally the procedure will be revised to stress the need for smallincremental steps with j a time delay between adjustments when approaching the scram set points to avoid getting i erroneously low values.

Sincerely, ENDORSEMENT:

Reviewed and Appro ed

/ &$t*PtA Anthony Schoone J. Charles McKibben Reactor Manager Associate Director xc: Mr. Alexander Adams,Jr., USNRC Mr. Tom Burdick,NRC Region III Subscribed and sworn to before me this Dr. Ed Deutsch 14th day of July, 1999.

Reactor Advisory Comnu.ttee Reactor Safety Subcommittee ba kt tle O Ph mh Michelle M. Blevins Notary Public My Commission Expires: 2/10/2003 MICHELLE M. B&W Nn'ary Public-Notary Seal

.. 'nTE OF MISSOURI N Mtv COUMY My G.a.. w. I .q ar Feb. 10,2003 J