ML20150C459

Describes Basic Aspects of 871221 Meeting W/Util in Bethesda,Md Re Ssi/Deconvolution Issues Re Audit Finding 1 & Provides Recommendations for Resolving Associated Issues

ML20150C459
Person / Time
Site:	Satsop
Issue date:	01/13/1988
From:	Philippacopoulo BROOKHAVEN NATIONAL LABORATORY
To:	Pichumani R NRC
Shared Package
ML20150C424	List: ML20150C420 ML20150C459
References
NUDOCS 8803180259
Download: ML20150C459 (3)

Text

~

y' r

1. ) e s y y:.dscg

" 6 BnT.V jUUd_ BROOKHAVEN NATIONAL LABORA ORY h{ ASSOCIATED UNIVERSITIES, INC.

Upton. Long Is:and. New York 11973 (516) 282s 2115 Deportrnent of Nuc! ear Energy F TS 656, January 13, 1988 Dr. R. Pichumani MS P-1114 U.S. Nuclear Regulatory Commission Washington D.C. 20555 Dear Dr. Pichumani In my letter to you dated October 15, 1987, I had proposed for a meeting with Washington Public Power Supply System (WPPSS) in order to discuss the SSI/ Deconvolution issues related to the "Audit Finding No. 1". As you know, such a meeting took place in Bethesda, MD on December 21, 1987. The participants from Brookhaven National Lab were myaelf and Dr. Costantino. The purpose of this letter is to describe basic aspects of this meeting and also to give recottmendations for resolving the WNP-3 SSI/ Deconvolution issues.

During the meeting WPPSS described the procedures used to check the effect of differences in seismic loads between the half-space analysis (no deconvolution) and the finite element analysis (with deconvolution) on the design of structures and components. In particular, the following items were checked: a) category 1 structures, b) piping and c) equipment. According to WPPSS, the accelerations associated with major concrete structures are bounded by the original seismic design basis. It appears however, that this conclusion has reached by comparison at the ZPA level. This may not be an acceptable procedure in that the lower frequencies associated with the SSI are not properly accounted for in this approach. Furthermore, the utility pointed out that the results of the half-space analysis produced an increase in accelerations for the steel containment vessel. This increase in seismic load was found to be acceptable on the basis of a review which was performed by Chicago Bridge and Iron Co. The utility was requested to present a complete review in order to address the impact of the half-space analysis on the seismic qualification of the steel containment vessel. In terms of piping, l WPPSS found that piping and pipe support stresses remained within code  ;

a?lowablew using the half-space analysis spectra. Finally, with regards to the impact of the half-space results on equi.pment qualification, WPPSS pointed out that all requirements are still met. Specifically, when comparing the i half-space output to the equipment specifications provided by the various vendors it was found that excessive conservatism is available to accommodate changes it. g's due to half-space analysis.

9903100259 G80310 PDR ADOCK 05000S00 g PDR

T l

A major portion of the meeting was concentrated on details of the half-space approach employed by the utility for WNP-3. The results of this analysis are questionable particularly because the method used to develop foundation impedances assnciated with embedment effects was incomplete. BNL had requested that the utility compare their impedance functions which were developed on the basis of various approximations to standard solutions typically used in SSI evaluations. The utility responded that such solutions are not applicable to this particular problem. During the meeting, however, available procedures were presented which can be used to resolve this issue.

Specifically, it was suggested that the following two items be addressed by WPPSS a) order of magnitude of impedances used in half-space analysis reasonably correlate with available solutions and b) the importance of the f requency dependency of these impedance functions should be addressed for the particular foundation configuration of WNP-3.

Based on the discussions during the meeting and our review of WNP-3 SSI/ Deconvolution issues carried out thus far, it is recommended that WPPSS should be requested tot a) Provf.o justification that ZPA alone represents an acceptable upper bound for the structural parameters of interest of category I reinforced concrete structures.

b) Provide detailed calculation to show that the stresses developed at the steel containment vessel from the half-space analysis are within acceptable limits.

c) Provide justification that the frequency independent foundation impedances are adequate for the frequency range of interest at WNP-3 site.

d) Provide correlation of sidewall impedances used in the half-space analysis with those obtained from available analytical solutions for rigid type foundations.

4 l

In my opinion, at this stage of the WNP-3 review of SSI/ Deconvolution issues, we do not have sufficient technical basis for accepting tha w'PPSS's half-space model as a valid one. Furhermore, it is recommended that after review and acceptance of the half-space model, compar. 2ns should be made between the results of the half-space analysis and the corresponding results from the finite element / boundaries analysis. The two results should satisfy the "enveloping" requirements of Alternate 1 (SRP 3.7.2 under revision), which is more appropriate when the input is a broad-band type spectrum similiar to that used in the WNP-3 SSI analysis. If the requirements of Alternate 1 are not met, then a site-specific type analysis should be considered by WPPSS in order to resolve the WNP-3 SSI/ Deconvolution issues.

Regards, W'U$ lD.

A. Philippacopoulos AP:db cc: C.J. Costantino M. Reich G. Bagchi L. Reiter B. Grenier I

I l

i