ML19338F115

Forwards Comments & Questions on Kwu Rept R141/141/79/E, Application of Susquehanna Test Measurement Results to Overall Loading of Suppression Chamber of Shoreham Plant by Depressurization Process, Dtd 791023

ML19338F115
Person / Time
Site:	Susquehanna, Shoreham  File:Long Island Lighting Company icon.png
Issue date:	10/03/1980
From:	Economos C BROOKHAVEN NATIONAL LABORATORY
To:	Su N Office of Nuclear Reactor Regulation
References
NUDOCS 8010070538
Download: ML19338F115 (2)

Text

k.

_j -

BROOKHAVEN NATIONAL LABORATORY ASSOCIATED UNIVERSITIES, INC.

Ucton, New YCrk 11973 Cecortment of Nucrect Energy (516) 345-2594 October 3,1980 Mr. Nelson Su Office of Nuclear Reactor Regulation U. S. Nuclear Regulatory Commission 7920 Norfolk Avenue Bethesda, Maryland 20014

Dear Nelson:

Enclosed is a "hard copy" of BNL's coments and questions on the KTG T-quencher load specification report which we telecopied to you earlier.

I trust you will find them helpful Best regards,

, pw..);

Constantino Economos CE:emm Enclosure

cc

W. Y. Kato, 3B2 G. Maise XOo3 s

//f 8010070$3V g

e INTEMAl.WDHl%GHAFT

• BNL Comments and Questions on KWU Report R141/141/79/E, " Application of the SSES-Test Measurement Results to the Overall Loading of the Suppression Chamber of the Shoreham Plant by Depressurization Processes" Dated October 23, 1979 by C. Economos C. C. Lin October 2, 1980 General Remarks - The report describes the development of an SRV air clear-ing load definition to be used for the Shoreham (SNPS) plant. This definition is considered to be representative of the actual behaviour of the T-quencher de-vice that will be used for load mitigation. The load specification bears con- -

siderable similarity to that developed for the Susquehanna (SSES) plant but dif- .

fers in some essential features. The similarities include: selection of a dis-crete set of pressure signatures from among a large library observed under test conditions to define the dynamic loads on the pool boundaries; application of multipliers to the observed amplitudes for conservatism and/or extrapolation to

" worst" or design conditions; application of frequency multipliers to account for differences between tested and prototype conditions and data variability; spatial attenuation of the pressure amplitude in the circumferential direction.

Differences include: use of " weighted" pressure traces for SNPS rather than actual traces as for SSES; use of different traces and frequency multipliers for SNPS for different load cases rather than a single " envelope" as for SSES; slight modification of vertical pressure attenuation to account for difference in quencher elevation above the basemat.

Comments l

1. The " weighting" procedure used to develop a single pressure trace for each SRV actuation is acceptable.

2. The pressure traces selected for load definition are acceptable on the basis of the ARS comparisons. We have qualified our acceptance here because we have some difficulty witri the use of an ARS which interprets the pressure his-tory as an acceleration of the support of an SDOF. We would have preferred to see the comparison made with ARS's that use the pressure as a force applied to the oscillating mass. We suspect that the results would not be qualitatively different but we are uncertain. It would be helpful if the SNPS-KWU people l could comment on this concern.

3. In developing pressure and frequency multipliers there does not appear to be any account taken for possible differences in SNPS line volumes from those tested during the KWU-SSES tests. This deficiency is particularly serious for the ADS load case since simulated ADS tests were only performed with the long l

- f

/

discharge line. A final decision as to the acceptability of the multipliers will require clarification of this issue.

4. In our judgement, the regression analysis used to determine pressure multipliers for " worst" cases has not been properly executed. We are performing our own regression analysis and intend to come up with alternate values for these multipliers. Preliminary results indicate that the multipliers will be greater than unity.

5. We have grave reservations concerning the acceptability of the fre-quency multipliers which are proposed. First, for the all valve case, the canbination of multipliers and pressure traces (Section 6.4.2) lead to a gap in frequency range running from about 7 to 8.5 Hz. This specification would not even bound some of the traces observed in the tests when transferred to the SNPS plant (Test 25.1, for example). Second, the ADS range which is proposed (6 to 6.6 Hz) strikes us as much too narrow considering all of the uncertainties in-volved (See also the concern discussed.in Comment 3). Third, too much uncer-tainty exists in our understanding of the SRV phenomenon to have confidence that the frequency shifts from test cell to plant all valve will be exactly as the analysis predicts. In fact, it is quite likely that the multiple bubbles will not oscillate in phase in which case power will be exhibited at every frequency lying between the two extremes corresponding to all valve and single valve.

Fourth, a number of internal inconsistencies have been detected within the re-port which cast further doubt on the correctness of frequency multipliers. This final aspect is addressed by means of the questions listed below.

Question 1 - Using Figure 6 of the report we deduce a multiplier of 1.32 to convert an 8 Hz bubble in the KWU test stand to Shoreham 1 quencher. From Fig-ure 7 we deduce a valve of 1.22 for the same conversion. In Section 6.2.2 it is stated that this multiplier is 1.26. Provide clarification indicating the source of this inconsistency and what the correct value is.

Question 2 - The situation addressed in Question 1 is a specific example of a universal inconsistency which we detect between Figures 5 and 6, on the one hand, and Figure 7. If we make the appropriate cross-plots from Figure 5 and 6, our version of Figure 7 would generally indicate higher multipliers than those shown. Provide clarification indicating the source of this inconsistency and which of the figures is correct.

l l

l

-% w