ML19350D009
| ML19350D009 | |
| Person / Time | |
|---|---|
| Issue date: | 12/05/1980 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-1785, NUDOCS 8104130074 | |
| Download: ML19350D009 (5) | |
Text
^4x^ t ) neun
&'l0N'&
DEC 5 n
\\ '. ~.
f.
gggg e
]CRS - 19 5 5 i
f,-
?
. 'e' ACRS Subcomittee on Extreme External Phenomena met on the af ternoon of
(
sN'OMo~ber8,1980inWashington,D.C. The purpose of the meeting was to discuss D
4\\ N #.p the status of the NRC Staff review of the seismic qualification of the auxiliary yV feedwater system (AFWS) in operating plants and the use of quantitative criteria
+[
in seismic evaluations. The principal attendees were as follows:
ATTENDEES:
AFFILIATION D. Okrent ACRS Member H. Etherington ACRS Member J. Ebersole ACRS Member J. C. Mark ACRS Member C. P. Siess ACr.S Member R. Savio ACh5 Staff R. Burns Consultant P. Davis Consultant J. Hickman Consultant W. Lipinski Consultant T. Novak NRC J. Knight NRC H. Levin NRC G. Holahan NRC T. Chang NRC
- 0. Garner NRC L. Shao NRC G. Bacghi NRC R. Borsum B&W C. Powell La C. Kamp E. Grecheck VEPCO D. Bridenbaugh MHB P. Pettit AIF R. Leyse EPRI RISK EVALUATION FOR OPERATING REACTORS - G. HOLAHAN, NRC lhe NRC Staff has identified 10 operating PWRs which do not have seismic qualified AFWS and has completed preliminary estimates of the risk associated with this feature of these plants. There is a body of thought within the NRC Staff which holds that AFWS would be the largest contributor to a core melt associated with seismic events and that in this regard presents a higher risk than the failure of the reactor to scram or loss-of-coolant accident. Mr.
Holahan summarized the Staff's approach in evaluating the risk associated with these 10 plants. He noted that conventional and nuclear power plants which have experienced seismic events have suffered little or no damage and that this 8104130OW
o EEP Oct 8, 1980 is viewed as a qualitative indication in that systems which are built utilizing ordinary engineering practices (not necessarily designed to seismic criteria) do, in fact, have a high degree of seismic resistance.
In the NRC Staff's study two seismic event probabilities were used. One was judged typical for Eastern sites and the other judged typical for Western sites. Equipment failure proba-bilities were characterized by using safety factors of 1 and 2 with the safety factor of 2 being used as a best estimate and the safety factor 1 as bounding value.
(A safety factor of 2 implies a 50% probability of failure of that system at twice the design earthquake). The analysis'did not take credit for redundancy in a, specific system but did account for diverse systems for the removal of decay heat. Of the ten plants in the study Yankee Rowe was the only plant in which the original design did not involve at least some degree of siesmic analysis for the AFWS. ' Yankee Rcwe is pretently shut down. for tur-bine repairs and is expected to remain so for an extended period of time.
A sumary of the probability / estimate for the loss of decay heat removal due to a seismic event for a typical Eastern U.S. plant designed to 0.29 is given on page 1 of Attachment 1.
A safety factor of 2 is used in these computations and a probability of 6x10-4 occurrences pet year was associated with the ground acceleration of 0.2 g.
The probability of loss of auxiliary feedwater and a back up decay heat removal system was also estimated with the same safety factor used for both systems.
A summary of the results obtained for all 10 plants is given on page 2 of.
The study associates the highest risk with San Onofre 1 (8x10-4 per reactor year for the complete loss of decay heat removal). The Subcomittee O
e
.k EEP Oct 8, 1980 noted that a typical (or regional average) Western earthquake was used in estimating the risk associated with San Onofre 1 whereas the plant is, in reality, located close to a fault on which large earthquakes may occur with This could imply a higher risk than has been estimated for greater frequency.
The annual societal risk associated with a typical (3x10-4 per
. San Onofre 1.
reactor year) probability of seismically induced core melt is summarized on page 3 of Attachment 1.
This societal risk is compared to that estimated for the WASH-1400 PWR and to the nonnuclear hazards associated with seismic events.
The risk associated with this plant was approximately a factor or 10 higher than what was associated with the WASH-1400 PWR and factor of 100 lower than what would be associated with the nonnuclear hazards from seismic events.
The Staff nas concluded that risks' associated with the operation of plants with AFWS which are not seismic qualified is not so high as to warrant immedi-ate plant shut down. It is the Staff's ir.tention to expand the scope of their preliminiary risk study. This work will include a more thorough study of all PWRs to quantize the degree of seismic resistance built into the auxiliary feedwater systems, the use of the available plant-specific ground acceleration information and the use of plant-specific design information in developing 1
safety factors. It is expected that this study will take several months and would confirm and refine the estimates obtained in the already completed preliminary study.
PLANS FOR FUTURE WORK - H. LEVIN, NRC Mr. Levin discussed the Staff's plans for expanding upon the preliminRry study described by Mr. Holahan. He indicated that the NRC Staff intended to make use of resources which had b'een developed in the SEP program. A licensee
f
. Oct 8, 1980
(
EEP questionnaire will be developed and circulated and the licensees responses evaluated. The plants will be categorized into groups and selected plants (an estimated 10-12) will be designated for site visits by NRC Staff review Detailed evaluations will be conducted for selected plants (an esti-teams.
mated 3-4) the recommendations and conclusions will be documented.
It is ex-pected that this will take about seven months. Additional risk assessment will j
be performed if required and would extend the program by about two months. Sig-
)
nificant results would be exoected 3-4 months after the start of the program.
SURVEY OF OPERATING PWRS - D. GARNER, NRC Mr. Garner summarized the results of the survey of all operating PWRs which have been made by the NRC Staff. This survey was of a preliminary nature and involved contacts with the plant owners to obtain their opinions as to the relative seismic l
qualification of the AFWS in their plants. Page 4 of Attachment 1 summarizes the results of this survey. The systems tabulated on this sumary with an "X" are systems which the licensee believes will not withstand the SSE. The absence of an "X" does not indicated, however, the systems are Category 1 or that the NRC Staff has reviewed the licensee's conclusion.
S GENERAL DISCUSSION Mr. Ebersole stressed the importance of a reliable feed and bleed system in removing decay heat. Systems of this type would provide a means for heat and,_
if of large enough capacity, provide the means for rapid blow down of the reactor so that low pressure heat removal systems and, if necessary, makeshift coolant makeup systems could be used.
i Oct 8, 1980
(
.EEP Dr. Okrent and Dr. Siess noted that the direction of the SSMRP program has been such that it has contributed very little to the understanding or resolving of the AFWS* issue. They noted that the SSMRP work could be contributing to resolution of this issue if the program had been oriented toward an earlier input into the licensing process and that this had been recomended prior to Subcomittee reviews.
The benefit which could be obtained from performing seismic evaluation and It eliminating obvious failure mechanisms was amphasized by the Subcommittee.
was suggested that this might be used as an immediate means for reducing risk associated with the operating plants. The Subcomittee indicated that it disagreed with the Staff's approach of resolving the auxiliary feedwater system They indicated that the risk levels estimated (if accurate) warrant issue.
considerable priority by the NRC. Evaluation by the affected utilities with*
recommendations for their resolution, improved quantification of the risk on a plant-specific basis; and the implementation of remedial measures as practical, when appropriate, was recomended.
The Subcommittee indicated that they believed that an ipportant probabilistic risk evaluation such as the one performed by the Staff for the seismic quali-fication of AFWS should be subjected to, peer review and a greater degree of internal quality control. The Subcomittee also noted that quantitative risk criteria are being used on a case-by-case basis by the NRC Staff and indicated that an effert should be made to develop guidance as to the level of NRC priorities / action which should be associated with given risk level.