ML20213F397

Safety Evaluation Supporting Util Responses Re Effects of Proposed Mods to Improve long-term Reliability of Auxiliary Feedwater Sys Per Recommentations GL-2 & GL-3

ML20213F397
Person / Time
Site:	Trojan File:Portland General Electric icon.png
Issue date:	11/10/1986
From:	NRC
To:
Shared Package
ML20213F384	List: ML20213F382 ML20213F397
References
TAC-60783, NUDOCS 8611140188
Download: ML20213F397 (2)

Text

.

g[gan%qI 'g UNITED STATES y ) p, NUCLEAR REGULATORY COMMISSION

-* WASHING TON, D, C. 20555 7,, j k..v / ..+

SAFETY EVALUATION AUXILIARY FEEDWATER SYSTEM NRC RECOMMENDATIONS GL-2 AND GL-4

. INTRODUCTION

By letter dated October 23, 1980, the staff forwarded its Safety Evaluation (SE) 1 of Portland General Electric's (PGE's) responses regarding requirements of the Auxiliary Feedwater (AFW) systems, which were identified in our letter to PGE dated October 3, 1979 and in NUREG-0611, " Generic Evaluation of Feedwater Transients and Small Break Loss-of-Coolant Accidents in Westinghouse-Designed Operating Plants." Among the responses evaluatcd were those related to long-Term Recommendations GL-2 and GL-4. Our SE concluded that PGE's responses to GL-2 and GL-4 were acceptable.

By letter dated November 25, 1985, PGE provided a detailed report of the Trojan AFW System design which included a discussion of proposed modifications that were intended to enhance system reliability. Responding to a staff request, PGE letter dated April 29, 1986 and an October 7, 1986 teleconference provided additional information regarding the affects of the system modifications as they related to GL-2 and GL-4.

DISCUSSION AND EVALUATION Long Term Recommendation GL-2 States:

" Licensees with plants in which all (primary and alternate) water supplies to the AFW systems pass through valves in a single flow path should install redundant parallel flow paths (piping and valves).

Licensees with plants in which the primary AFW system water supply passes through valves in a single flow path, but the alternate AFW system water supplies connect to the AFW system pump suction piping downstream of the above valve (s), should install redundant valves parallel to the above valve (s) or provide automatic opening of the valve (s) from the alternate water supply upon low pump suction pressure.

The licensee should propose Technical Specifications to incorporate appropriate periodic inspections to verify the valve position."

Our October 23, 1980 SE concluded that PGE's response to GL-2 was an acceptable alternative to the prescribed requirements. The staff determined that the in-sta11ation of low suction pressure alarms and safety-grade automatic trip of the AFW pumps on low suction pressure, the installation of control room position indication for the single locked-open primary AFW supply line suction valve, and the use of a new electric motor-driven AFW pump utilizing a separate flow path from the condensate storage tank (CST) to the pump suction, would collectively, be an acceptable alternate means of assuring AFW pump protection in the event of an inadvertantly closed valve on the primary AFW supply from the CST.

8611140188 861110 PDR ADOCK 05000344 P PDR

J PGE's November 25, 1985 letter proposes the replacement of the common suction valve with a flexible coupling. The flexible coupling and all non-Category I piping have been reanalyzed by the licensee and meets seismic Category I design requirements. In addition, an October 7, 1986 telephone conference confirmed that this " upgraded" piping has been analyzed for and meets Seismic Category II/I considerations and criteria, and will be incorporated into the ISI program. Also, a debris screen will be added to the CST outlet nozzle to prevent blockage. PGE also proposed to replace the safety related low suction pressure trip of the AFW pumps with a safety related low CST level trip. However, low AFW pump suction pressure would still be alarmed in the control room.

Based on the above, we conclude that the proposed modifications continue to provide an acceptable means of assuring AFW pump protection.

Long Term Recommendation GL-4 States:

" Licensees having plants with unprotected normal AFW system water supplies should evaluate the design of their AFW systems to determine if automatic protection of the pumps is necessary following a seismic event or a tornado.

The time available before pump danage, the alarms and indications available to the control room operator, and the time necessary for assessing the problem and taking action should be considered in determining whether operator action can be relied on to prevent pump damage. Consideration should be given to providing pump protection by means such as automatic switchover of the pump suctions to the alternate safetygrade source of water, automatic pump trips on low suction pressure, or upgrading the normal source of water to meet seismic Category I and tornado protection requirements."

Our October 23, 1980 SE concluded that PGE's response to GL-4 was acceptable since the concern raised in this recommendation was similar to that of Recommendation GL-2, namely, providing automatic protection for the AFW pumps in the event of loss of normal suction supply for what ever reason.

Other currently proposed modifications to the AFW system as stated in PGE's November 25, 1985 letter, in addition to those which were described under GL-2, include the installation of a tornado missile barrier around the above-grade suction piping near the CST, the addition of stiffeners to the CST wall in the vicinity of the suction nozzle, and judf rious placement of CST level transmitters and cabling. These modifications are intended to provide additional protection and availability of the AFW pump's normal water supply (i.e., the CST) and other components from damage during a seismic event or tornado.

Based on the above, we conclude that the proposed modifications continue to provide an acceptable means of assuring AFW pump protection.

CONCLUSION Based on our evaluation, the licensee's responses regarding the proposed modifi-cations are consistent with the recommendation of GL-2 and GL-4, and are therefore, '

acceptable.  !

l i

Principle Contributor:

T. Chan )

1