ML20027B352

From kanterella
Jump to navigation Jump to search
Fifth Interim Deficiency Rept Re Computer Program Verification Methods Lacking Adequate Visibility.Bechtel Method for Finalizing Brown & Root Computer Calculations Encl.Next Rept Will Be Submitted by 821222
ML20027B352
Person / Time
Site: South Texas  STP Nuclear Operating Company icon.png
Issue date: 09/13/1982
From: Oprea G
HOUSTON LIGHTING & POWER CO.
To: Jay Collins
NRC OFFICE OF INSPECTION & ENFORCEMENT (IE REGION IV)
References
10CFR-050.55E, 10CFR-50.55E, SFN:V-0530, SFN:V-530, ST-HL-AE-880, NUDOCS 8209200303
Download: ML20027B352 (5)


Text

.

The Light Company nosm ughdng& P-r m ux n00 n-s

.wxasmoi m3m8mn September 13, 1982 ST-HL-AE-880 File Number: G12.96 SFN: V-0530 Mr. John T. Collins

, /l

~

/ $7 Regional Administrator, Region IV Nuclear Regulatory Commission j ~

k 611 Ryan Plaza Dr., Suite 1000 Arlington, Texas 76012 u

Dear Mr. Collins:

J South Texas Project Units 1&2 Docket Nos. STN 50-498, STN 50-499 Fifth Interim Report Concerning Computer Program Verification On May 8, 1981, Houston Lighting & Power Company (HL&P), pursuant to 10 CFR 50.55(e), notified your office of an item concerning computer program (code) verification. The verification methods lacked adequate visibility to the user as to whether or not the program versions in use had been verified.

The information supplied in this report represents our Fifth Interim Report.

As previously discussed in our Third and Fourth Interim Reports, a consultant firm was engaged by Brown & Root, Inc. (B&R) to address the appropriateness of the modeling options, acceptability of code usage, and acceptability and adequacy of the methods selected for computer program verification. As a result of a review by Bechtel of B&R subcontract activities, Bechtel recommended that the consultant's work be terminated to avoid unrecessary duplication of Bechtel transition reviews of past B&R engineering work.

Based on the Bechtel recommendation and, additionally, in recognition that the consultant's scope of work was not dispositive of the computer program verification issue, HL&P directed B&R to terminate the consultant's work in December 1981 except for preparation of a report covering the work already accomplished. The consultant prepared and sub-mitted a report on the work completed prior to contract termination.

This consultant's report was transmitted to BPC for resolution of this item.

As part of Bechtel's efforts to resolve this issue, a method was devel-oped for use by the various Bechtel engineering disciplines during their review of B&R computer calculations. Details of the Bechtel method for finalizing B&R computer calculations are provided in Attachment I to this report.

} 6 )7 8209200303 820913 gDRADOCK 05000498 PDR

September 13, 1982 ST-HL-AE-880 File No.:

G12.96 SFN: V-0530 Page 2 l

The consultant's review of B&R's Nuclear Analysis Disciplines computer program verification reports (CPVR's) and calculations included 26 CPVR's and 147 calculations. However, there was not necessarily a one-to-one corre-lation of calculations and CPVR's (i.e. QAD CPVR's were reviewed, but no calculations using these codes were). Therefore, status of CPVR's and calculations is presented separately.

The consultant's report reviewed 147 B&R calculations performed by the B&R Nuclear Analysis Discipline that had utilized computer codes. Of that number 54 had been revised, not used in the design or had been superseded by other calculations primarily performed by NUS.

Ten calculations have been or are in the process of being superseded by Bechtel calculations either because the design has changed or inadequate computer program verification was found.

There are currently 43 calculations that have been identified as requiring change primarily for design criteria reasons.

Four calculations have been identified as desirsb!a to keep pending verification of the computer programs. Thirty-six calculations are still undergoing review.

Of the 26 CPVR's reviewed by the consultant, Bechtel is considering retaining 5 codes, not retaining 13 codes and has not completed its evaluation on the remaining 8 codes. The 13 codes are not being considered for retention because the calculations utilizing them have been redone by NUS, require rework because of design changes or the CPVR is inadequate.

For those codes not retained the calculations will be redone using a verified code.

For those codes retained the CPVR will either be made consistent with paragraph 5 of Bechtel Engineering Department Procedure (EDP) 4.36 to the satisfaction of the Engineering Group Supervisor (EGS) or the calculation verified to EDP 4.37 as is shown in the figure in Attachment 2.

B&R also identified 57 significant computer programs used by the remaining disciplines on the South Texas Project (STP). Of this number BPC is considering retaining 34 codes, not retaining 17 codes and has not completed its evaluation of the remaining 6 codes. Of those not being retained, 10 cases result from deficient CPVR's and 7 cases from a need to redo calculations for other reasons.

The next report concerning this item will be submitted to your office by December 22, 1982.

If you should have any questions concerning this item, please contact Mr. Michael E. Powell at (713)877-3281.

Very truly yours, ll

.W/

rp,

r.

7

/

Exe tive ce President MEP/mg Attachment

Houston Lighting & Power Company cc:

G. W. Oprea, Jr.

September 13, 1982 J. H. Goldberg ST-HL-AE-880 J. G. Dewease File Number: G12.96 J. D. Parsons SFN: V-0530 D. G. Barker Page 3

- C. G. Robertson R. A. Frazar J. W. Williams R. J. Maroni J: E. Geiger

'j H. A. Walker S. M. Dew J. T. Collins (NRC)

D. E. Sells (NRC)

W. M. Hill, Jr.

(NRC)

M. D. Schwarz Baker &Botts)

R. Gordon Gooch Baker & Botts)

J. R. Newman Lowenstein, Newman, Reis, & Axelrad)

STP RMS Director, Office of Inspection & Enforcement Nuclear Regulatory Commission Washington, D. C. 20555 G. W. Muench/R. L. Range Charles Bechhoefer, Esquire Central Power & Light Company Chairman, Atomic Safety & Licensing Board P. O. Box 2121 U. S. Nuclear Regulatory tommission Corpus Christi, Texas 78403 Washington, D. C.

20555 H. t. Peterson/G. Pokorny Dr. James C. Lamb. III City of Austin 313 Woodhaven Road P. O. Box 1088 Chapel Hill, North Carolina 27514 Austin, Texas 78767 J. B. Poston/A. vonRosenberg Mr. Ernest E. Hill City Public Service Board Lawrence Livermore Laboratory P. O. Box 1771 University of California San Antonio, Texas 78296 P. O. Box 808, L-46 Livermore, California 94550 Brian E. Berwick, Esquire William S. Jordan, III Assistant Attorney General Harmon & Weiss for the State of Texas 1725 I Street, N. W.

P. O. Box 12548 Suite 506 Capitol Station Washington, D. C.

20006 Austin, Texas 78711 Lanny Sinkin Citizens for Equitable Utilities, Inc.

Citizens Concerned About Nuclear Power c/o Ms. Peggy Buchorn 5106 Casa Oro Route 1, Box 1684 San Antonio, Texas 78233 Brazoria, Texas 77422 Jay Gutierrez, Esquire Hearing Attorney Office of the Executive Legal Director U. S. Nuclear Regulatory Commission Washington, D. C.

20555 Revision Date 08-23-82

=

s BECHTEL'S PROGRAM TO FINALIZE BROWN & ROOT, INC. COMPUTER CALCULATIONS Each Bechtel discipline will review the Brown & Root, Inc. (B&R) calculations and associated Computer Program Verification Reports (CPVR's),

turned over during the Transition Phase as part of the work package re-views, and develop their status. As a result of this effort along with his previous experience, the Bechtel Engineering Group Su'pervisor (EGS) will establish which calculations are required to document his design effort.

In order to minimize the recalculational effort the EGS will use B&R calculations previously transmitted in all cases where their appropriate-ness and correctness can be verified in line with the attached flow chart and as discussed below.

A.

If the calculation is not required as part of the design bases documentation it will be marked Preliminary and filed under Preliminary or Superseded Calculations.

B.

If the calculation is required as part of the design documentation and, in the judgement of the EGS, is appropriate and meets all other Industry Codes or Standards associated with the design, it must still meet certain standards of acceptability as delineated in Bechtel Engineering Department Procedures (EDP's) 4.36 and 4.37 before it can be marked Final.

1.

B&R calculations developed using codes with documented CPVR's consistent with B&R procedures may be marked Final if, in the judgement of the EGS, the verification adequately meets the requirements of Paragraph 5 of EDP 4.36.

This level of acceptance is to be used only in the case of relatively simple "one of a kind" solutions and where, in the judgement of the EGS, the reasonableness of the solution is self evident.

2.

B&R calculations not meeting the "one of a kind" criteria of Paragraph 1 or calculations which, while meeting B&R criteria for code verification, do not meet the requirements of EDP 4.36 but, in the judgement of the EGS, were made with the valid codes, must be verified before acceptance as final in accordance with EDP 4.37.

As a minimum this verification can be accomplished using limited reruns or bounding calculations of the design problems on the similar verified versions of BPC or contractor codes.

If, in the judgement of the EGS, the reruns provide reasonable and consistent results, he can accept as Final, single or groups of calculations encompassed by this verification.

3.

B&R calculations which, in the judgement of the EGS, do not meet the criteria of Paragraph 1 or 2 above, must be redone in accordance with EDP 4.36 or 4.37 before they can be accepted as Final by the EGS.

In addition to the above verification effort, from time to time, Disci-pline Chiefs, at their discretion, may require that calculations be given additional review prior to acceptance, and will so advise the EGS.

g vai s.38 fift

--g!

h gg(*

SA I

-g I

=

n a

R R

!!st

=.

==

o E

I i.:

85 5

=

e E

a a

r 84 E

a EM

[ms Et

5 3

eo

-es b$<=

"E ;**Eg hE!

no

,[:=e

=is

< w *e l

M g

E-

=

g u

g

m. I. _

slevi5 I5l!!al[izl s

.=5!55 n=

es--l ij!!l

~

8 E

~

3 s

2 a

n E

"II!E i=I;'$

!!3

!!E5 8

an 4

E.

h.

Eit 8 5