1CAN098311, Summarizes Nrc/Util 830906 Conference Call Re Proposed Installation of New Condensate Storage Tank,Resulting from Review of NUREG-0737,Item II.E.1.1, Auxiliary Feedwater Evaluation. Previous Proposed Concept Should Be Approved

Summarizes Nrc/Util 830906 Conference Call Re Proposed Installation of New Condensate Storage Tank,Resulting from Review of NUREG-0737,Item II.E.1.1, Auxiliary Feedwater Evaluation. Previous Proposed Concept Should Be Approved

ML20080K642
Person / Time
Site:	Arkansas Nuclear
Issue date:	09/20/1983
From:	John Marshall ARKANSAS POWER & LIGHT CO.
To:	Stolz J Office of Nuclear Reactor Regulation
References
RTR-NUREG-0737, RTR-NUREG-737, TASK-2.E.1.1, TASK-TM 1CAN098311, 1CAN98311, NUDOCS 8309290308
Download: ML20080K642 (4)

Text

_ _ _ _ . - - - _ _ - _ - -. -

i ARKANSAS POWER & LIGHT COMPANY POST OFFICE BOX 551 LITTLE ROCK, ARKANSAS 72203 [501) 371-4000 September 20, 1983 1CAN098311 Director of Nuclear Reactor Regulation ATTN: Mr. J. F. Stolz, Chief Operating Reactors Branch #4 Division of Licensing U. S. Nuclear Regulatory Coramission Washington, DC 20555

SUBJECT:

Arkansas Nuclear One - Unit 1 Docket No. 50-313 License No. DPR-51 ANO-1 EFW Safety Evaluation NUREG 0737 Item II.E.1.2 Gentlemen:

On September 6,1983, a conference call between members of your staff and AP&L was held to discuss the potential installation of a new condensate storage tank for ANO-1. This potential modification was proposed by AP&L as a result of NRC recommendations for modifications to improve the reliability of the EFW system. These recommendations were in addition to modifications previously performed by AP&L to improve the reliability of the EFW system.

The previous modifications, discussed in our [[letter::1CAN078308, Provides Addl Info Re NUREG-0737,Item II.E.1.1 Concerining Emergency Feedwater Sys (EFW) Evaluation,Per 820618 Request. Extensive Upgrade of EFW Sys in Progress.Upgrade of EFW Suction Initial Source Under Consideration|letter dated July 29, 1983]], (1CAN078308), were initiated based upon the AP&L review of the reliability required by NUREG 0737 Item II.E.1.1, " Auxiliary Feedwater Evaluation."

The main topic of the conference call was the need for making the propor9d condensate tank tornado protected as well as seismically qualified. AP&L proposes to make the tank seismically qualified, however, we believe that the reliability of the EFW system will not be significantly increased by making the tank tornado protected. During the conference call your staff requested that we submit the evaluation referenced in our [[letter::1CAN078308, Provides Addl Info Re NUREG-0737,Item II.E.1.1 Concerining Emergency Feedwater Sys (EFW) Evaluation,Per 820618 Request. Extensive Upgrade of EFW Sys in Progress.Upgrade of EFW Suction Initial Source Under Consideration|July 29, 1983, letter]]. The requested evaluation is attached and describes the methodology used to evaluate the effect of CST reliability on overall EFW reliability.

The probability of a tornado strike on the condensate tank was determined using NUREG CR-2944, " Tornado Damage Risk Assessment." The probability for a tornado strike of a force suf ficient to damage or destroy the condensate tank was calculated. For a structure 150 feet wide this probability was

~5x104 This number is conservative due to the calculational methods used, 8309290308 830920 A PDR ADOCK 05000313 P

W PDR MEMEiER MIDOLE SOUTH UTILITIES SYSTEM gg

Mr. J. F. Stolz S:pt mb:r 20, 1983 and in addition the condensate tank will measure approximately 38 feet in diameter vs. the 150 feet assumed in the NUREG. When this probability is compared to the values on the attached graph, it is seen that a number of this low order of magnitude can have no significant effect on the overall reliability of the EFW system due to the relationship between the condensate tank and EFW system reliability.

Our conclusions on the relative contribution of the concensate tank to the reliability of the EFW system are further supported by other material such as the paper presented by Mr. Frank H. Rowsome at the IEEE/ASME/ASCE Joint Power Generation Conference, Los Angeles, California in September 1977.

This paper was entitled, " Fault Tree Analysis of an Auxiliary Feedwater System."

The regulatory basis for your recommendations relative to the ANO-1 EFW system stem from NUREG 0737 Item II.E.1.1. This item required licensees to evaluate EFW systems to determine if modifications were necessary to improve system reliability. Therefore it is clear that only those modifications which would result in increases in EFW reliability are warranted. Although the evaluation submitted by AP&L concluded that additional modifications were not needed, AP&L has submitted a conceptual design for a new seismically qualified CST in order to satisfy NRC concerns expressed in your letter dated June 18, 1982, (1CNA068202). Imposing additional requirements simply to comply with current staff positions, the Standard Review Plan, etc. would constitute an unreasonable backfit and result in unwarranted expenditures. Since our evaluation clearly shows that additional modifications to protect the CST from postulated tornado damage would not serve to further improve EFW reliability, we request that upon consideration of this letter, AP&L's prev'ously proposed conceptual design be approved, jVery truly yours, John R. Marshall Manager, Licensing JRM:DH:sl Attachment l

l

i CONDENSATE STORAGE TANK SENSITIVITY ANALYSIS The Emergency Feedwater System (EFWS) at Arkansas Nuclear One - Unit 1 (ANO-1) was modeled as part of an Electric Power Research Institute (CPRI) project, " Development of Modularized GO Reliability / Availability System Performance Models." The objective for AP&L's involvement in the project was to develop an in-house capability for the GO methcdology. The model has some limitations but is of sufficient detail to allow evaluation of the effect of changes in Condensate Tank Reliebility on the EFW system reliability.

A sensitivity analysis was performed by running the model for various CST reliabilities between zero and one to determine the impact on EFWS reliability.

The model output suggests that improvements in CST reliability beyond 0.9999 do not significantly improve EFWS reliability. Seismically qualified tanks are generally above 0.999999 in reliability.* The attached graph shows model output for the EFWS reliability versus CST reliability. CST reliability is plotted from zero (0.0) to 0.9 and 0.999 to 0.999999999.

In using NUREG CR-2944 to determine the probability of a tornado strike on the condensate tank, certain assumptions were made. These assumptions included the determination that a tornado had to have an intensity of F2 or higher to affect the condensate storage tank. A tornado with an intensity of F2 has wind speeds of 113 to 157 mph. Some of the effects of an F2 tornado include: roof torn off frame houses leaving strong upright walls standing, trailer houses demolished, and light-object missiles generated.

Additionally, the strike probabilities were based on a structure 150 feet wide rather than the actual 38-foot diameter of the condensate tank. The use of this assumption should preclude the possibility of a tornado that does not hit the condensate tank but comes close enough to cause sufficient damage with missiles to impair the operability of the condensate tank. The probability of a tornado strike using those assumptions is ~5x10-8 When this number is compared to the relationship between conoensate tank reliability and EFWS reliability, it is seen that due to the order of magnitude of this number it has no effect on EFWS reliability.

4 Model References The GO methodology is adequate for modeling the EFWS at ANO-1.

This is demonstrated by a comparison of the basic model results with previous model results. The unavailability of the EFWS as represented by the GO model compares favorably with an earlier EFWS unavailability represented by a B&W model (4.29x10-5 vs. 5.4x10-s). The B&W unavailability figure is published in their ANO-1 EFW System Upgrade Reliability Analysis report.

• In the July 1982 Interim Report, Sequoyah Nuclear Power Plant Availability and Risk Assessment (Draft); Pickard, Lowe, and Garrick, Inc. use a value of 0.999999 for a non-seismically qualified tank.

0.999958-0.999957' - -

. m v

m v

m v o

>, 0.999956' -

a

9*

E 2

a: 0.9998 '-

m a

b 0.9996 '-

0.9994 4

0.9992 , . , , w , , , , , , ,

u . . . . . . .

0.2.4.6.8 g g g g g g g e e e e e e. e e e e e e @ e e e e e e e e e e e e e e e e e e e e e e e e

CST Reliability '

_ , . . . . . .