ML20003H940

Advises That Mods Needed Re TMI Action Item II.F.1.6, Containment Hydrogen Monitoring,To Achieve Acceptable Levels of Reliability Based on NUREG-0737 Criteria.Three solenoid- Operated Valves Will Be Added Inside Containment

ML20003H940
Person / Time
Site:	Trojan File:Portland General Electric icon.png
Issue date:	05/04/1981
From:	Withers B PORTLAND GENERAL ELECTRIC CO.
To:	Eisenhut D Office of Nuclear Reactor Regulation
References
RTR-NUREG-0737, RTR-NUREG-737, TASK-2.F.1, TASK-TM TAC-47784, NUDOCS 8105080214
Download: ML20003H940 (2)

Text

,,

A+ Q w.a.

s 4

b,q uUt i

as % PcrtlandGeneralElectricCctrpany y

d MAY 0 7 WBM H..

co* * "

c, May 4, 1981 Trojan Nuclear Plant Docket 50-344 Licensee NPF-1 l

Mr. Darrel G. Eisenhut, Director Division of Licensing Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission Washington, DC 20555

Dear Mr. EisSnhut:

By letter dated January 2, 1981, PGE identified compliance with the NRC requirements of NUREG-0737. Following the January 2, 1981 submittal, the detailed design review continued to ensure compliance with the NRC requirements. In regard to Action Item II.F.1.6, Containment Hydrogen Monitoring, the design review of our existing Containment Hydrogen Analysis System to date has revealed that modifications need to be made in order to achieve acceptable levels of reliability for the Trojan Post-Accident Containment Hydrogen Analysis System based upon the cri-teria in NUREG-0707.

The modifications will consist of the addition of three solenoid-operated valves inside Containment. The design review has identified that conden-sation could accumulate in the low point of the Containment Hydrogen f

Analysis Train B sample line and may prevent its proper operation. A solenoid-operated valve will be added to the low point in the Train B l

sample line to allow periodic draining of this condensation. Although the Hydrogen Analysis System Train A sample line does not have a potential problem with regard to H2 monitoring, a solenoid-operated valve will be added to remove condensation for representative iodine sampling for the Containment atmosphere.

The Containment atmosphere sample return lines currently have a common isolation valve for both Trains A and B with an alternate flow which discharges to the Containment purge exhaust system. A solenoid-operated valve will be added to the Train A return line to provide redundancy back into Containment. All three new valves will close upon receipt of a Containment isolation signal. Installation of the valves inside Contain-ment will be completed during the scheduled refueling outage in 1981 and M

placed in service by January 1, 1982

.S Trojan's original design assumed about 8 days for hydrogen concentrations to reach 4 percent inside Containment. On that basis, intermittent

!D operation of hydrogen monitor several hours after a LOCA was considered appropriate. Since monitoring may now be necessary shortly after a LOCA, PGE is preparing an operating instruction to assure that condensation l

'2: 5 W 3 amen kee*. man Ctg a7204 810s08 eMy

o r

Fcrind General Eechic Ccn ceny Mr. Darrel G. Eisenhut May 4, 1981 Page two rates in the hydrogen monitor Train B are within acceptable levels by specifying a sampling frequency. This instruc tion will be Laplemented prior to the resumption of power operation for Cycle 4.

The instruction will be revised to allow continuous monitoring af ter the drain valves are operational in Janaary 1982.

Additionally, a review of the Trojan Containment Hydrogen Analysis System disclosed one area that conflicts with the NUREG-0737 requirements for redur.dancy. The power supplies to all Containment isolation valves are channelized by providing Train A power and Channel A CIS signal to the isolation valves inside Containment, and Train B power and Channel B CIS signal to the outside isolation valves. This design is based on accepted NRC design criteria for redundancy in Containment isolation.

Because of this desigs criteria for Containment isolation, the Trojan Hydrogen Analysis System does not entirely meet the design requirements of NUREG-0737 for redundancy. A singla Channel A or B f ailure could isolate both Containment hydrogen analysis loops; however, a single failure of this type could be corrected without major difficulty follow-ing an accident. It is our belief that Containment isolation, due to its importance, should take design precedence over the Containment Hydrogen Analysis System redundancy.

In order to further improve the reliability of the Containment Hydrogen Analysis System, additional system improvements are being evaluated. It is, however, our belief that the aforementioned modifications will provide an acceptable level of reliability for Containment hydrogen monitoring.

Sincerely,

1-I Bart D. Withers Vice President Nuclear c:

Mr. Lynn Frank, Director State of Oregon Department of Energy Mr. Robert A. Clark Division of Licensing U. S. Nuclear Regulatory Commission

- -