ML20082F228

Forwards Justification for Not Implementing Mod Recommended in NUREG-0737,Item II.K.3.22 Re Rcic.Costs Associated W/Mod Cannot Be Justified Since Safety Benefit Can Only Be Classified as Marginal

ML20082F228
Person / Time
Site:	Browns Ferry
Issue date:	11/23/1983
From:	Mills L TENNESSEE VALLEY AUTHORITY
To:	Harold Denton Office of Nuclear Reactor Regulation
References
RTR-NUREG-0737, RTR-NUREG-737 NUDOCS 8311280479
Download: ML20082F228 (4)

Text

_ _ . . .

a

.+

C TENNESSEE VALLEY AUTHORITY.

CHATTANOOGA.. TENNESSEE 374ot 400 Chestnut Street Tower II November 23, 1983 Mr. Harold R. Denton, Director Office of Nuclear Reactor Regulation U.S. Nuclear Regulatory Commission Washington, D.C. 20555

Dear Mr. Denton:

In the Matter of the ) Docket Nos. 50-259 Tennessee Valley Authority ) 50-260 50-296 By letter from G. C. Lainas to H. G. Parris dated October 20, 1983, we received a request for plans and schedules for implementation of NUREG-0737, Item II.K.3.22 at the Browns Ferry Nuclear Plant. Item II.K.3 22 requires modification of the Reactor Core Isolation Cooling System (RCICS) to provide for automatic switchover of the suction from the condensate storage tank to the suppression pool. The information was requested pursuant to 10 CFR 50.54(f).

By my letter to you dated December 23, 1980, we submitted justification for not implementing the modification discussed in II.K.3.22. This was further confirmed in my letter to you dated April 22, 1983, our response to Generic Letter 83-02. Our basic position remains that the modification is of marginal safety benefit at best, and that an objective cost / benefit analysis would show the' modification to be unnecessary.

Enclosed is a detailed technical discussion of why we continue to believe that the modification is unnecessary. If you continue to disagree with our position after review of the additional information enclosed, please get in touch with us through the Browns Ferry Project Manager. We will be happy to meet with you at your convenience to further discuss our position.

Based on the enclosed information we believe that continued operation of Browns Ferry is justified and that the licenses for units 1, 2, and 3 should not be modified, suspended, or revoked.

Very truly yours, TENNESSEE VALLEY AUTHORITY l .

L. M. Mills, Manager Nuclear Licensing

! Subscribed agd sworn to before me this M 36 day of //cvcMef 1983

,- (OLl0 dt h

Notary Public f I' My Commission Expires /8 - d 8 Y l

. Enclosure cc: See page 2 8311280479 831123 An Equal Opportunity Employer PDR ADOCK 05000259 P PDR

s ,

Mr.. Harold R.-Denton November 23, 1983 c'c (Enclosure);

U.S. Nuclear Regulatory Commission

~ Region II ATTN: James P. O'Reilly, Regional Administrator 101 Marietta Street, Suite 2900

. Atlanta, Georgia : 30303 ,

Mr.'R. J. Clark Browns Ferry Project Manager U.S. Nuclear Regulatory Commission 7920 Norfolk Avenue Bethesda, Maryland 20814 a

m.we., -w-e w+ --ev ea- - - - --

. . ENCLOSURE

(

REFERENCE:

. LETTER FROM G. C. LAINAS TO H. G. PARRIS DATED OCTOBER 20, 1983)

JUSTIFICATION FOR NOT IMPLEMENTING THE MODIFICATION RECOMMENDED IN NUREG-0737, ITEM II.K.3.22 '.' c.

BRCWNS FERRY NUCLEAR PLANT RCIC is an auxiliary system designed primarily to provide relatively low flow (rated 600 gpm) vessel makeup during isolation events when normal feedwater supply is unavailable. During events that require significant, amounts of high-pressure coolant injection, the 5000-gpm safety-grade HPCI system.is relied upon. Failure of HPCI to operate during these events would result in operation of the ADS and LPCI systems; therefore, RCIC is not needed' or useful during . events requiring large amounts of high-pressure coolant injection.

While both HPCI and RCIC initiate at the same level setpoint, procedures exist to secure HPCI injection relatively early during events that require small amounts of high-pressure coolant makeup in order to prevent overfilling the vessel. Thus, the large amount of water in the 375,000-gallon capacity condensate storage tank is reserved primarily for the low-flow RCIC system (600 gpm) and CRD system (~100 gpm). At 344,000 gallons in the CST (low-level alarm), the tank would not reach the low CST level suction switchover setpoint until greater than eight hours into an event at full RCIC and CRD flows. At the technical specification minimum level of 135,000 gallons, it would not be reached until more than three hours into an event. These timeframes are more than sufficient to allow the operator to perform a manual switchover of the RCIC suct4on source under a situacion of relatively low stress and therefore good reliability.

It should be understood that the above condition of remaining on full 600-gpm RCIC flow for such an extended period of time is highly unlikely. The only required injection is that necessary to provide makeup for the inventory lost due to decay heat and pressure relief to the suppression pool. The 135,000 gallons are sufficient to provide for makeup for more than seven hours. With the normal water level of greater than 344,000 gallons, makeup is available for approximately 35 hours4.050926e-4 days <br />0.00972 hours <br />5.787037e-5 weeks <br />1.33175e-5 months <br />. These times are within the recovery mode of operation, and the operator will be under low stress conditions.

In addition to the above discussion, if only RCIC were running and the water level decreased in the CST suction supply, HPCI would realign to the suppression pool at the switchover setpoint. In the event of RCIC trip, there would be several hours before the water level decreased to the HPCI initiation point again. HPCI would then begin injecting to the vessel from the suppression pool if RCIC were not manually realigned to the suppression pool and restarted.

Further analysis of long-term events in NUREG/CR-2973, " Loss of Decay Heat Removal (DAR) Sequences at Brewns Ferry Unit 1 - Accident Sequence Analysis," and NUREG/CR-3719, "The Effect of Small Capacity, High-Pressure Injection Systems on TQUV Sequences at Browns Ferry Unit 1," show that after approximately four hours, RCIC is no longer needed and CRD flow will maintain sufficient injection to the vessel. In practice, RCIC would probably be tripped off or realigned to the CST-to-CST test mode for pressure control with CRD supplying makeup after several hours. This would therefore negate the necessity for a RCIC suction switchover.

. .w. c.. *. 4+ -

O We additionally believe that it is more prudent to rely on the suction switchover only for HPCI to prevent a potential common mode failure causing

. both high pressure systems to inadvertently and irreversibly realign from the available CST to'the suppression pool which could be at an elevated temperature. The desirability of protecting the condensate storage system as an external coolant source is clearly recommended in the NRC Severe .,,

Accident Sequence Analysis (SASA) program using Browns Ferry unit 1 as the model plant.

Finally,- the costs associated with performing this modification cannot be

. justified since any safety benefit derived can only be classifled as marginal if not nonexistent.

a N

=

0 4

l l

' 1 l

,w

. - . - - . - _ - - , - - - - c -

+e-- - - - - -