ML19351G349

Final Deficiency Rept Re Use of Defective Computer Program for Seismic Qualification of Wm Powell Co Valves.Erroneous Results of Program Had No Impact on Facility Valves.Seismic Suitability of Valves Has Been Demonstrated

ML19351G349
Person / Time
Site:	Zimmer
Issue date:	02/12/1981
From:	Borgmann E CINCINNATI GAS & ELECTRIC CO.
To:	James Keppler NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION III)
References
10CFR-050.55E, 10CFR-50.55E, M-25, QA-1397, NUDOCS 8102230563
Download: ML19351G349 (2)

Text

I f

\

.~

Yk "

THE CINCINNATI GAS & ELECTRIC CONIPANY crf CINCINN ATI. OMIC 452 0 February 12, 1981 QA-1397 EA BORGMANN stoce . ct *et s eco

,p f%

Jh U. S. Nuclear Regulatory Commission \

Region III 799 Roosevelt Road N h Qf')*/

g , y {, 3 -1 Glen Ellyn, Illinois 60137 67 FEB 2 01C81* M N

Attention: Mr. J. G. Keppler, Director & u.s. co*u m uncau u "c" g

%qh N RE: WM. H. ZIMMER NUCLEAR POWER STATTS UNIT I 10CFR50.55(e) M-25, SEISMIC QUALIFICATION, '<*/%b [M ,

POWELL VALVES, CONSTRUCTION PERMIT CPPR 88, DOCKET NO. 50-358, W.0. 57300-957, JOB E-5590 Gentlemen:

This letter is our final report on the subject 10CFR50.55(e) event, which related to the use of a defective computer program for seismic qualification of valves furnished by the Wm. Powell Company.

In 1976 Midwest Technical, Incorporated, was subcontracted by the Wm. Powell Company to develop a computer program for the purpose of calculating valve resonant frequencies used in conjunction with the seismic analysis. The computer program developed for this purpose was known as FREON. This program was princi-pally developed for use on Powell's contract with the Mississippi Power and Light Comoany. However, FREQN was also used on Powell's Seismic Report No. 71376 for Zimmer.

In mid 1971, Bechtel, MP&L's A/E for Grand Gulf, subcontracted a firm to reanalyze a Powell valve for BWR New Loads. During the course of this reanalysis it was discovered that the FREQN Program yielded erroneous results. The defect in the FREQN Program related to the number of elements used to model the valve in the analysis. Subsequent to the Bechtel work, Powell discovered that the formulas used as the equations of motion in FREQN were also errcneous in that they were based on displacement along the axis of the stem as opposed to dis-placement perpendicular to the stem.

Rather than correct the defects in FREQN, Powell elected to develop a new finite element program for seismic analysis. Seismic Report Number 71376 was revised using the new finite element program. This resulted in a revised resonant frequency of 48.5 Hz, which is beyond the cutoff frecuency of 33 Hz as stipulated in the S&L technical specification. In effect, the erroneous results of the FREQN Program had no impact on Zinner valves. FREQN calculated a resonant frequency greater than 33 Hz and the new finite element program also calculated a resonant frequency greater than 33 '1z. Specification requirements were met.

Gel $

830223o 3 3 5

5f

i n

Mr. J. G. Kepoler U. S. Nuclear Regulatory Comission Pegion III Februa ry 12, 1981 - QA-1397 Page 32 Powell has re-evaluated all seismic reports, including those not using the FREQN Program, and is in the process of updating all seismic analysis methodology to the current state-of-the-act and NRC seismic criteria. The re-evaluation of the seismic report is being routinely monitored and reviewed by Sargent & Lundy and the CG&E General Engineering Department.

The seismic suitability of the valves referenced in our reoort of November 20, 1980, nas been demonstrated, and it is our feeling that 50.55(e) M-25 can be closed.

Very truly yours, THE CINCINNATI GAS & ELECTRIC COMPANY By P E. A. BORGMANN SENIOR VICE PRESIDENT RLW:ec cc: NRC Resident Inspector Attn: F. T. Daniels NRC Office of Inspection & Enforcement Washington, D. C.-