ML20127C563

FOIA Request for List of Validated Computer Programs for Processing of Leak Rate Data & Listing of Volumetrics Faulty Program

ML20127C563
Person / Time
Site:	Zion, LaSalle, 05000000
Issue date:	03/08/1985
From:	Reytblatt Z WARREN WILSON COLLEGE, SWANNANOA, NC
To:	Harold Denton Office of Nuclear Reactor Regulation
References
FOIA-85-233 NUDOCS 8503150278
Download: ML20127C563 (2)

Text

-- - - - - - - - - - - - -

.i  %

v.

(WarrenWilson College ,.. i . .... . s. C. -

March 8, 1985 Mr. Harold R. Denton, Director EREEDOM OF INFORMATION Office of Nuclear Reactor Regulation JCI,REQUESI.,

U. S. Nuclear Regulatory Commission Washington, DC 20555 g, APPEAL P a. id d_3 _ gs

~-

1 In the Matter of i

COMMONWEALTH EDISON COMPANY, ON THE EMERGENCY RELIEF VOLUMETRICS, Inc.

(Zion Station, LaSalle County DECISION Station)  ;

AND ALL LIGHT-WATER REACTORS '

USING VOLUMETRICS SOFTWARE INTRODUCTION

{

On January 9, 1985, the Petitioner filed an cmergency relief petition.with the Office of the Executive Director for Operations seeking an immediate termination of use of the Volumetrics computer program for processing leak rate test raw data  !

until the sofware is debugged and revalidated, the petition being supported by the evidence from the Zion leak rate test showing that the program does not perform addition and/or division correctly which consequently may lead to underestimating abnormally high leak rates to the degree that such leak rates would appear to be within normal limits. Such misrepresentation according to Appendix J to 10 CFR constitutes a danger to public safety and requires your immediate action.

As stated in the NRC letter of February 20, 1985, to the Petitioner (Exh. B), the Petition has been considered by the NRC, and was fcund to.be providing an unsufficient basis for emergency relief on the part of the NRC. The NRC Decision does not dispute the computer output supporting the Petition. The NRC Decision has not supported its Decision by any essential calculations, discussion or references.

BACKGROUND Several fraudulent Zion tests employed the Volumetrics software. Many other tests which have not yet been found deficient, employed the same software. Regulations governing the safety related software provide for debugging and revalidating such software when mistakes in calculations have occurred.

DISCUSSION

1. Contents of the NRC Decision ,

The Decision (a) acknowledges receipt and concideration of the charger (b) con ras that the charge is new; and (c) introduces a concept of an " underlying subject" which, it is claimed, has been previously raised by the Petitioner and considered by the staff. Further ((d)), the Decision informs on the NRC satisfaction with some different calculations, and (e) the Decision states that the charge is not new.

. .. vA (704) 298-3325 e u .- . - e.- a, man n unoin dMn ann / . .A su e- I

4 -

2. Discussion of the Decision (a) , (b). The Petitioner ' agrees with the NRC statements. ~

(c) The statement is irrelevant to the fact that the software does not handle properly arithmetic operations.

. (d) Rules governing usage of safety related computer software prohibit such usage after just one occurrence of an incorrect result has been disclosed, regardless of satisfaction or dissatisfaction of the NRC with other results. It is quite common that a faulty computer program yields the correct results most of the times, and the incorrect results - only in a few times. Such programs are not allowed for safety related calculations.

(e) The Decision contradicts its own findings (see (b)). However, even the assumption that the Petition did not raise "new technical issues" is not a basis for a denial if it is shown, as it has undeniably been, that there are immediate safety concerns. It appears that the NRC Decision is deliberately confusing the Petitioner on the good reasons for emergency relief actions.

CONCLUSIONS

1. The NRC is in violation of 10 p. 2 not responding to the concrete piece of evidence - failure of the sottware in question to perform arithmetic operations of addition and/or division.

2. The NRC arguments are either selfcontradictory or irrelevant.

3. The NRC failed to show why the software proven to produce an arithmetically wrong result at least once, should not be debugged and revalidated.

REQUESTS

1. Immediately revoke validating documents of the software in question, order the debugging and revalidation.

2. I am enclosing an FOIA request for the list of (validated) computer programs for processing of leak rate data, and for the listing cf the.Volumetrics faulty program. Your cooperation is kindly sought in acecrdance with 10 CFR, Part I, Subpart A, Paragraph 9.4.

3. In accordance with 10 CFR, Part I, Subpart A, Paragraph 9.4 and applicable provisions of the FOIA, 552 USC and amendments thereef, I request that all"the records related to revalidation of the volumetrics program in the future, be made e available to me for inspection at the time of their submittal to the NRC or shortly thereafter and prior to finilizing the validation.

I shall expect to receive your response in 20 (twenty) working days.

Sincerely,.

(

),

ZinovyV.Re7tblatt, Chair Department of Mathematics and Co puter Science

{

i n,