ML20244C350
| ML20244C350 | |
| Person / Time | |
|---|---|
| Issue date: | 05/10/1983 |
| From: | Ernst M NRC OFFICE OF NUCLEAR REGULATORY RESEARCH (RES) |
| To: | Thomas C Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML20234E460 | List:
|
| References | |
| FOIA-87-40 NUDOCS 8305270140 | |
| Download: ML20244C350 (3) | |
Text
e I o s l'NITID STATES f!!'CLI A4 REGULATORY COMMISSION Vi AbHINGT ON. D. C. 205%
,,,,, May 10,1983 met:0RANDUM FOR: C. O. Thomas, Chief Standardization and Special Projects Branch Division of Licensing Office of Nuclear Reactor Regulation FROM: Malcolm L. Ernst, Acting Director Division of Risk Analysis
- Office of Nuclear Regulatory Research
SUBJECT:
INPUT FOR OKRENT REQUEST ON SARRP BWR MARK III ANALYSES In response to the request of D. C. Sca11etti of your staff, we have prepared a brief discussion of the results to date of our Severe Accident Risk Reduction Program as they relate to BWR Mark III plants. We understand that this information was requested by Dr. Okrent during an ACRS subcommittee meeting on GESSAR, and hope that the enclosed material will be of use.
~~
Ma o m L. Ernst, Acting Director Division of Risk Analysis !
Office of Nuclear Regulatory Research l
Enclosure:
As Stated ,
cc: D. C. Sca11etti B30Cn0J# 3g.
. i l
I Preliminary Results from SARRP 4 on BWR Mark III Plants The.results obtained to date from the SARP,P program for BWR Mark III plants i are best described in the two conference papers provided as Attachments A l and B. While these papers are roughly six months old, nothing done since that time leads us to question the accuracy of their general results.
When reading these papers, we believe that it is very important to keep in mind a few points.
(1) The work is generall> ased on the Reactor Safety Study (RSS) and RSS Methodology Application' sProgram (RSSMAP) risk assessments, although I
some specific modifications have been made. The accuracy of Grand Gulf RSSMAP results, and the extrapolation of these results to other Mark III plants, may introduce significant uncertainties.
(2) In general, risk reduction calculations have been performed without consideration of competing risks, so that data shown in the attachments (e.g., Attachment A, Table III) generally are maximum potential benefits.
The exception to this general rule is the assessment of filtered and !
unfiltered vents, which makes use of the more detailed studies of these specific features. '
(3) It is not intended to use these results, as such, to support Commission "0/84" decisions on severe accidents. Between now and the end of 1983, these results will be refined using updated information on:
accident sequence likelihood from the Accident Sequence Evaluation Program; and accident consequences from numerous RES programs.
Considerable sensitivity studies will also be performed over the remainder of the year to investigate the impact of key assumptions, design varia
- tions, etc., and thus provide a much more robust technical basis for Comission decisions.
J Attachment B uses Sandia's filtered-vent containment system (FVCS) work to display the benefits and costs of severe accident " fixes" for a variety of benefit measures. For the FVCS cases for BWR Mark I and Mark III designs, the benefit measure includes the effects of competing risks; in these cases, a first evaluation of benefit uncertainties is also provided. It should be emphasized that the cost for the "BWR add-on" shown, for example, in Figure 1 represents only a vent. Since credit is being taken for radionuclides retention in the suppression pool, an unfiltered vent was determined to be an adequate j
{
l t
l i
._--__--____--______U
2 design for these plants. The cost displayed for a "PWR add-on" is for a comparable filtered vent; it is clear that the design cost is quite sensitive to the capability of the suppression pool to remain an effective filter. In addition, since the "BWR add-on" cost is in the range of 1 million dollars, the required plant outage to connect the vent to the containment could significantly increase the cost if the outage exceeds the original estimate.
i In addition to the SARRP and FVCS work discussed in the attachments, another Sandia program on the benefits and costs of alternate decay heat removal (ADHR) systems provides insights on severe accident prevention features for BWRs. Af ter consider.ing a number of possible ADHR systems, a detailed study was made for the most promising option for BWRs, an additional low-pressure make-up and pool cooling train. The final report on this work (NUREG/CR.2883, in printing) indicates that such a system reduces core melt frequency by factors to 3 to 6, using the Grand Gulf RSSMAP as a basis. The cost estimated for this train was roughly 30 to 40 million dollars.
L l
I
.