ML20206H338

Forwards Auxiliary Feedwater Sys Reliability Analysis, Vols 1-3,per 870213 Commitment.New Analysis Provides Means of Assessing Relative Improvement in Auxiliary Feedwater Sys Reliability Obtained Through Mods Implemented Since 850609

ML20206H338
Person / Time
Site:	Davis Besse
Issue date:	04/03/1987
From:	Shelton D TOLEDO EDISON CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
Shared Package
ML20206H340	List: ML20206H338
References
1367, NUDOCS 8704150331
Download: ML20206H338 (3)

Text

.. . m-

)

l TDLEDO EDISON Docket No. 50-346 DONALD C SHELTON Ves Premders-Nudear (419)249 2399 License No. NPF-3 Serial No. 1367 April 3, 1987 United States Nuclear Regulatory Commission Document Control Desk Washington, D. C. 20555 Gentiemen:-

On February 13, 1986, through Serial No. 1241, Toledo Edison committed to perform a Comprehensive Reliability Study of the Auxiliary Feedwater (AW) System.

Toledo Edison has since completed this detailed reliability study which is provided as Attachment I to this letter. This study, performed by Impe11 Corporation, evaluates the following three AW System configurations for Davis-Besse.

" Prior Configuration" -

This is the AW System configuration as it existed on June 9,1985.

This configuration is referred to in the attached study as the

" Existing Configuration."

"Two-Pump Configuration" -

This configuration incorporates the short-term modifications implemented prior to restart from the June 9, 1985 transient. This configuration includes only the two Turbine-Driven Safety Grade AW Pump Trains.

m 4 "Three-Pump-Configuration" -

This configuration includes the short-term modifications associated

]

og with the " Planned Two-Pump Configuration". In addition, this configur- l bo ation includes the new Motor-Driven Feedwater Pump (MDFP). The MDFP

• serves as a manually initiated backup (third train) to the two safety l l

grade AW pump trains. This is the currently installed configuration at Davis-Besse. .

OC

$ The unavailability of each of the above AW System configurations was Q

C n.c.

evaluated for each of the following conditions.

O g

Loss of Main Feedwater (LMW) with offsite power available.

THE TOLEDO EDISON COMPANY EDISON PLAZA 300 MADISON AVENUE TOLEDO, OHIO 43652

Dockst N5. 50-346 License No.;NPF-3

. Serial-No.'1367~

Page 2

- Loss of Main Feedwater coincident with a loss of offsite power (LOOP).

- Loss of Main Feedwater coincident with Station Blackout conditions,

~

the Loss of Offsite and Onsite AC Power (LOAC).

" Unavailability" is defined as the " probability per demand that the A W System will not perform its mission".

InNovember1985,ToledoEdisoncompletedarelikbilidyanalysisbasedon the predetermined assumptions and guidance provided in NUREG-0611. That analysis was provided in Appendix IV.C.2.3 of the Davis-Besse Course of-Action. This new analysis utilizes a much more detailed and plant specific approach-than the NUREG-0611 based analysis. Specifically, the guidance-provided in NUREG-0611 considered only intact steam generators.- Consequently,

! failures resulting in low pressure conditions.in the steam generators (e.g.:

! stuck open Main Steam Safety Valve) are not considered in-the NUREG-0611' study.

i NUREG-0611 considered all AFW support systems.(i.e. Service Water and. Electric-

. Power) to be available with a probability of 1.0. Additionally,' NUREG-0611 j specifies the failure events to be considered as well as the failure rate for i these events. The failure events specified in NUREG-0611 only address system initiation. Faults which can occur following initiation are not considered.

j This new analysis includes consideration of faulted as well as intact steam

generators, includes a detailed model of the Steam and Feedwater Rupture

Control System (SFRCS), utilizes component failure rates derived from plant specific data, considers additional human errors and considers potential failures in AW support systems as well as failures following system initiation. Therefore, the numerical results of these two studies are not comparable nor can the acceptance criteria specified in NUREG-0611--be applicable to the results of this study.

't This new analysis does provide a means of assessing the relative improve -

i ment in A W System reliability obtained through modifications implemented-since June 9, 1985. This relative improvement in AW System reliability -

is illustrated in the attat' led table. - This study concludes thatcimplemen-tation of the short-term modifications-and installation of the MDFP. improves

, AW System reliability by a factor of 20 over the AW System as it existed on June 9, 1985. The study also identifies areas for future consideration.

Toledo Edison expects that future modifications resulting from the long-term-items discussed in_the Course of Action document will further improve the reliability of the A W System.

Very tru y yours, j

DCS:SJS:p1f

+

Attachment cc: DB-1 NRC Resident Inspector E A'. B. Davis, Regional Administrator (2 copies)

,~,r -~g av, ,- p---.- s-, w,- - , , -, , , , -- e y-e.., ,-w-, g e- -, y

v - . . - - . _ .. . -. . - . .

.. .. : =

- Docket No. 50-346-

' License'No. NPF-3 Serial No. 1367 ,

j - Attachment

SUMMARY

OF'RESULTS-UNAVAILABILITY OF THE DAVIS-BESSE 1 AUXILIARY FEEDWATER SYSTEM (AFWS).

i Unavailability.Value of:the AFWS

~

Initiating ,,

t Event Prior Two-Pump 'Three-Pump Configuration Configuration. Configuration (presently.-~ installed)

Loss of Main Feedwater .062 .0068 ' 003' 1

(UiWF)

Loss of Offsite Power .071 .015 . 0092 (LOOP)

Loss of All AC .35 .27 .27 i

(LOAC) i i,

t.

= p'. ~ - - - ,4 ,1sy-#.-- & -yhwre9- -97a qgr w