ML20042B937
| ML20042B937 | |
| Person / Time | |
|---|---|
| Site: | Maine Yankee |
| Issue date: | 03/19/1982 |
| From: | Garrity J Maine Yankee |
| To: | Spei T, Speis T Office of Nuclear Reactor Regulation |
| References | |
| JHG-82-50, MN-82-56, SHG-82-50, NUDOCS 8203260312 | |
| Download: ML20042B937 (2) | |
Text
.
f d
MAME&%.i B
HAllHEE A10mlCPOVERCOMPARS*
guaugrgjain"g$$n k
N (20D 623-3521 March 19, 1982
. ang MN-82-56 i
0 g
s U
9 thited States Nuclear Regulatory Commission E-C%q7 Washington, D. C. 20555 7
5 -<d e -
- qh-; J id Attention:
Mr. Themis P. Spels 5(;# D M
Assistant Director for Reactor Safety 8
Division of Systems Integration a g p ['
p
References:
(a) License No. DPR-36 (Docket No. 50-309)
(b) USNRC Letter to YAEC, dated January 26, 1982 (c) MYAPC Letter to USNRC, dated March 15, 1979 (WMY-79-18)
Subject:
Single Failure Assumptions in ECCS Analyses
Dear Sir:
Reference (b) reported that Westinghouse had discovered that for some of their plants the assumed single failure of a low pressure safety injection pump, previously believed to be the worst single failure for LOCA analysis, might not in fact be the most limiting.
It appears that this discovery is attributable to the way break ECC spillage is calculated to affect containment backpressure and steam binding retardation of refill /reflood.
This leads to higher calculated peak clad temperatures in the reflood phase of recovery from LOCA.
We have considered this concern and conclude that the hypothesis is more related to the calculational techniques and simplified model of containment response than to real phenomena.
We understand that this concern arises when exercising a model which assumes break spillage exits the break as a fine spray with a steam condensing efficiency of 100% (instantaneous thermodynamic equilibrium with containment atmosphere).
This is consistent with BTP CSB-6.1 guidelines and 10 CFR 50 Appendix K, and certainly provides a conservative estimate of the effect of containment backpressure on refill /reflood cooling performance.
Maine Yankee ECCS licensing calculations utilize the methodology outlined in Reference (c), and conform to BTP CSB-6.1.
In our methodology, break spillage is added to containment spray system flow.
Flow from one LPSI pump thus is calculated to produce a conservatively low containment back pressure.
With two LPSI pumps running, the break spillage would not be double that calculated with one pump running because system flaw resistance and pump flow vs. head characteristics combine to provide a somewhat lower flow.
In addition, we believe that the assumption of one pump flow break spillage represented as spray conservatively bounds the true effects of two pump flow gDk spillage on conbi *
'kpressure.
f&Ogoh y%
g P
f e
MAINE YANKEE ATOMIC POWER COMPANY U.S. fOclear Regalatory Commission March 19, 1982 Attention:
Mr. Themis P. Spels Page Two l
Further, experimental data (LOFT Test L2-3) has shown that peak cladding temperature occurs during the blowdown phase (not reflocd); that no sigli,'icant clad temperature excursions occur during reflood.
This seems to indicate that higher break flow does not materially affect the peak clad -
+
temperature.
Moreover, L2-3 and other testing has confirmed that actual peak clad temperatures are several hundred degrees lower than Appendix K calculations would indicate.
It thus appears that the low head safety injection flow question does not give rise to real safety effects, so there would not appear to be any impact on Maine Yankee operating limits.
We trust that the above information is satisfactory.
If there are any questions, please feel free to call.
Very truly yours, MAINE YANKEE ATOMIC POWER COMPANY
</
John H. Garrity, Senio Director Nuclear Engineering and Licensing JHG/bjp i
-..