ML20209B597
Text
,
- p*Wuq%',
UNITED sTATEG h [ [3..,
NUCLEAR REGl.lLATORY COMMISSION g
j WASHahGTON. D. C. 2C555 0
0, '
a
%.~.,~.. * *r AUG5 1982 MEMORANDUM FOR: Darrell G. Eisenhut, Director Division of Licensing FROM:
Richard H. Vollmer, Director Division of Engineering
SUBJECT:
DIABLO CANYON DESIGN VERIFICATION PROGRAM-SCOPE OF PHASE I This is in response to your memo of July 26, 1982 concerning the scope of the Phase I Independent Design Verification Program (IDVP) for Diablo Canyon. The semi-monthly reports through No.18 and Interim technical reports Nos. 1, 2 and 3 have been reviewed by the DE staff.
In addition, we have benefit of the study of the annulus structure performed under our contract by Brookhaven National Laboratory.
The majority of the review work being performed under Teledyne is still at a inconclusive stage with the formal context of the IDVP as shown by the following summary of Errors and Open Items taken from the semi-monthly reports through No. 18.
CLASS A CLASS B CLASS C CATEGORY NUMBER CATEGORY NUMBER CATEGORY NUMBER Closed 1
Closed 1
Closed 15 Analyzing 2
IDVP 1
Analyzing i
Evaluation Require 1
TOTAL 2
Analysis 4
Modification Complete Waiting to 1
TOTAL 20 be worked TOTAL 5
At best, two Class A errors and one Class B error are at a stage where an unambiguous result could be audited. Similarly ITR No.1 is in essence a status report calling for further sampling and ITR No. 3 references the additional sampling called for a ITR No.1.
These somewhat meager results do not seem commensurate with the large amount of time and resources that have been devoted to the IDVP effort.
I believe that the reason for the IDVP becoming a protracted process is that there has not been a vigorous implementation program for the early findings that could have generic implications.
Instead we have seen a searching for explanations and solutions to " local" problems in a attempt to show low impact. While this approach is in keeping with the gyg4WOh}f D 0' yI l
)
1
, AUG 5 1992
~
6 overall philosophy of the IDVP it is remarkably inefficient when problems are discovered in the early stages of the design process as has occurred.
In general, I concur that some reassessment of the present Commission order is timely. The results of the IDVP to date have indicated considerable uncertainties in both the structural models presently employed to generate floor response spectra and in the methods used to process the raw spectra for use in piping and equipment evaluation.
Some of these uncertainties have arisen as a result of a finding that the structural models did not comport with as-built structures. This finding raises serious question as to the validity of the initial assumption that subcontract work done prior to June 1978 was a sufficient element for review.
Whereas I feel action on the part of the staff is clearly appropriate at this time, I feel that there is more to be gained by pressing for a vigorous implementation program by PGE/Bechtel rather than a modification of the Commission order that might be construed to preempt TES only 90 days after the Phase I program plan was officially approved.
I am concerned that such an action would be a severe blow to the IDVP concept and could throw excessive review responsibility back on the staff.
I I would prefer, as a first option, that we press for PGE/Bechtel to aggressively pursue the implications of the IDVP findings to date. Such a program would include:
(a) verification that all required structural models are a good characterization of the as-built structures, (b) verification that the methods employed for development of floor response spectra are acceptable, (c) verification that the floor response spectra curre'1tly on record are acceptable in light of the findings of steo (a) ard (b)
- above, (d) verification that piping models presently on record are acceptable characterizations of as-built systems.
I feel that any attempt to limit the scope of these verificatione through step (c) above (such as to systems required for low power) is neither practical nur efficient. Once the magnitude of variation
>~ AUG 5 1932 0
0 between currently recorded spectra and acceptable spectra is understood, it may be possible to set meaningful bounds on a low power restitution program.
N
.'ichard H
- ollmer, irector Division Engineer ng cc:
H. Denton E. Case R. DeYoung R. Mattson L. Chandler R. Jones J. Crews T. Bishop J. Knight F. Miraglia H. Schier11ng L