ML20138E111

Safety Evaluation Re Leak Test Parameters for Valve MO-8812. Changing Test Medium for Valve from Gas to Water & 33.6 Psi Differential Pressure Acceptable

ML20138E111
Person / Time
Site:	Trojan File:Portland General Electric icon.png
Issue date:	10/10/1985
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20138E091	List: ML20138E085 ML20138E111
References
TAC-56813, NUDOCS 8510240578
Download: ML20138E111 (3)

Text

. _ .

f TC e

UNITED STATES E ~ )g( j'[g p, NUCLEAR REGULATORY COMMISSION C. -[ WASHINGTON, D. C. 20555

'%d ,$

t SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REGULATION LEAK TEST PARAMETERS FOR VALVE MD-8812 -

PORTLAND GENERAL ELECTRIC COMPANY ,.

TROJAN NUCLEAR PLANT DOCKET NO. 50-344 I. INTRODUCTION As part of the application for an operating license for the Trojan Nuclear Plant (the facility), the Portland General Electric Company (the licensee) comitted to submit a permanent test program for valve y M0-8812 which was acceptable to the NRC and would verify the capability

of the valve to meet applicable safety requirements. A test program for this. valve, designed to meet these objectives, was submitted to the NRC by letter dated August 29, 1978. This proposed program was approved c -

by the NRC in a letter dated January 15, 1979.

_. By letter dated January 15, 1985, the licensee requested the test conditions be changed. The requested changes consisted of: (1) the leak test medium reduced frombe 60 changed from gas psid (pressure to water,)and difference (2) the to 22.5 psid. The test pressure be staff's review of this matter indicated the proposed pressure difference of 22.5 psid might not be conservative. The licensee was therefore requested to

. re-examine this request and revise the proposed test pressure if the re-review indicated such revision was needed. By letter dated July 12, e' 1985, the licensee responded by increasing the proposed test pressure

~

from 22.5 psid to 33.6 psid.

II. EVALUATION Valve M0-8812 is normally open and is installed in a 14-inch diameter line. It is the principal valve in the supply line from the Refueling Water Storage Tank (RWST) to the suction header of the Residual Heat Removal (RHR) pumps. The valve is located outside containment and is normally separated from the containment atmosphere by at least two containment isolation valves. In the event of a loss of Coolant Accident (LOCA) of sufficient magnitude, borated water initially would be drawn from the RWST and delivered via this valve and the RHR pumps to the ReactorCoolantSystem(RCS). When all of the borated water provided for this purpose has been delivered to the RCS, plant response to the accident enters the recirculation phase. In this phase, valve MO-8812 is closed and containment isolation valves (M0-2069A&B and M0-8811AAB) between the containment sump and the suction header of the RHR pumps are opened. In this configuration, water spilled from the RCS as a result of the LOCA can be drawn from the sump by the RHR pumps, cooled by the RHR heat exchangers, and returned to the RCS to continue the cooling process. - *.

Ohft 0 0 0 44 p l'DR

Because of the possible presence of high levels of radioact;lve materials in the containment sump following a LOCA and the exposure of490-8812 to .

these materials during the recirculation phase of a LOCA, leikage from M0-8812 is a safety coacern. This leakage may occur either b'y stem leakage or by backflow to the RWST and thence to the atmosphere. It is to reduce the probability of leakage via these pathways that periodic leak testing of M0-8812 is needed.

As noted above, the licensee previously proposed and the staff approved a leakage test procedure for M0-8812 involving use of a gas as the test medium and a test differential pressure of 60 psid. Also as noted, the licensee now proposes the test medium be changed to water and the test

, pressure be reduced to 33.6 psig. The licensee's basis for requesting a

] change in the test medium from gas to water is that the medium considered

. in the FSAR accident analysis was water (Section 15.6.5.6.1, page' 15.6-35 to -37). Thus, by changing to this test medium, the test can be

-l. _ compared directly to the assumptions used in the FSAR (3000 cc/hr for a

- total of eight valves). On the basis that use of a test medium si,ailar

~~

to that which would be present during the postulated accident reduces the uncertainty of the measurement and thus provides greater assurance acceptance limits are met, the staff concludes the proposed change in test medium is acceptable.

Regarding the proposed change in test pressure, the present value of 60 psi, corresponds to the design pressure of the containment building. This value, therefore, is conservative because it envelopes the worst case w- accident. In support of a lower test pressure (33.6 psi), however, the licensee notes the containment pressure at the time of initiation of the recirculation phase is substantially lower than the maximum containment sressure, and thus, the 60 psi value presently used is unnecessarily 11gh. For example, referring to Figure 6.2-91 of the FSAR, the licensee notes the maximum containment pressure for the Design Basis Accident (DBA) occurs 116 seconds after the beginning of the accident. At this time, the containment pressure ranges from 57.3 to 59.9 psig, depending upon the assumptions used for the specific calculation. By contrast.

., based on the minimum water inventory permitted for the RWST, the licensee has calculated the earliest time the recirculation phase could begin.

The licensee finds this to be 647 seconds after the beginning of the accident. The licensee notes that by this time, the FSAR figure shows the containment pressure would have declined to 39 psig. The licensee has adjusted this value to account for elevation differences between the water levels in the RWST and the containment, plus an allowance for uncertainty in reading the figure. With these adjustments, the licensee arrives at a requested test pressure of 33.6 psid (Note: Although the licensee's submittal of July 12, 1985 defines the test pressiare as 33.6 psig, discussions with a licensee representative confirmed attest pressureof33.6pa,idwasacceptabletothelicensee).  ;,

4

w. . i

l ,

l The staff has reviewed the licensee's calculations and adjuptments.

! Although the staff identified some differences between its estimates and those of the licensee, the final value for a conservativt test pressure was the same. Based on these considerations, the staff concludes the proposed revised test pressure for M0-8812 of 33.6 psid is acceptable. The staff points out, however, this value is acceptable only if the differential nature of this value is observed. That is,

! the value of 33.6 psid is to be the difference between the pressure

! measured on the upstream face of the valve and that measured on the

! downstream face.

( . III. CONCLUSIONS

'N ~

Based on the more accurate measurements that will be obtained using water

. as a test. medium, the staff ' concludes that changing the test medium for l

valve MO-8812 from gas to water is acceptable.

c .- . Based on the licensee's analysis and the staff's review of the analysis, the staff concludes that 33.6 psi differential pressure across the valve s is a suitably conservative test pressure for valve M0-8812 and is, there-fore, acceptable.

Principal Contributor:

G. Zwetzig. Region V I 98 l

I

.I

• L