ML20133G951
| ML20133G951 | |
| Person / Time | |
|---|---|
| Site: | 05000447 |
| Issue date: | 03/15/1985 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-2282, NUDOCS 8508090135 | |
| Download: ML20133G951 (7) | |
Text
6'd -Q AcR5 -ass:L.,
DATE ISSUED: 3/15/85 7
![
OM ONUS
/
7 Q&
MEETING MINUTES 43M FEBRUARY 14-15, 1985 MEETING OF THE COMBINED SUBCOMMITTEE ON GESSAR-II AND RELIABILITY AND PROBABILISTIC ASSESSMENT The ACRS Subcommittees on GESSAR-II and Reliability and Probabilistic at the Best Western Airport Park Assessment met on February 14-15, 1985 The purpose of this meeting was to continue the Hotel in Englewood, CA.
Subcommittee review of GESSAR-II for a Final Design Approval app The principle topic in these discussions was seismic to future plants.
The Subcommittees heard presentations from representatives of risk.
NRR, BNL, Jack R. Benjamin and Associates, and General Electric Com The meeting was begin at 8:30 am on February 14, adjourned at 6:30 pm February 14, reconvened at 8:30 am on February 15, and adjourned at Parts of the meeting were closed for the 12:00 am on February 15.
discussion of proprietary material. The principle attendees were:
D. Okrent, Subcommittee Chairman J. Ebersole, ACRS Member D. Ward, ACRS Member H. Etherington, ACRS Member C. P. Siess, ACRS Member R. Savio, ACRS Staff M. Bohn, ACRS Consultant D. Scaletti, NRC Staff M. Rubin, NRC Staff R. Bari, BNL K. Shiu, BNL J. Reed, Jack R. Benjamin & Associates 8508090135 850315 DE.31GNATED ORIGINALS PDR ACRS 2282 PDR IA D
m Certified By_
1 i.
. February 26, 1985 P
GESSAR-II/RPA Meeting Minutes J. Quirk, GE R. Villa, GE K. Hollzclaw, GE D. Hankin, GE D. Foreman, GE Highlights General Electric, as part of their FDA submittal for GESSAR-II, has 1.
The performed a PRA which includes an assessment of seismic risk.
seismic risk evaluation was performed as a " add-on" to the original The NRC PRA and was performed in response to a NRC Staff request.
Staff is having BNL perform the technical review of both the PRA and the seismic risk appendix.
GESSAR-II, as a standard plant design, does not have a specific 2.
site for which a seismic hazard (i.e., the estimated frequency of Soil occurrence for a spectrum of earthquakes) can be developed.
A and foundation conditions could also~ vary for different sites.
seismic risk evaluation, to be useful, would have to establish the level of risk associated with the plant and what the significant A suitable seismic vulnerabilities existed in the design.
" envelope" of seismic hazard functions and soil and foundation conditions would be required. The potential seismic risk contribution of those parts of the plant which were outside of the scope of the GESSAR-II design would need to be established and The intent would be to controlled via interface requirements.
provide assurance at this stage in the licensing that the seismic design would be adequate and well conceived and to identify improvements at a preconstruction staage.
General Electric described the methodology which was used in their 3,
A seismic hazard curve was developed with seismic risk evaluation.
the intent of encompassing most of the possible sites in the j
i l
'd
2 GESSAR-II/RPA Meeting Minutes February 26, 1985 s
Eastern United States. This hazard curve is shown on Figure 1 and compared to seismic hazard curves used in other industry PRA's.
Figure 2 contains a more complete collection of seismic hazard curves, including a number which were generated in recent LLNL evaluation. The curves generally represent best estimates accompanied by large uncertainties and are generally based on surveys of expert opinion.
4 General Electric has, based on expected performance, developed fragility requirements for critical structures and components in the GESSAR-II design and has established these values as specification requirements. BNL in their evaluation of GESSAR-II has concluded that the fragility of the GESSAR structures and components could in many cases be expected to be lower than what GE has established. The NRC Staff is proposing that the Applicant referencing GESSAR-Il either use the BNL estimates in their seismic evaluation or provide additional justification that the use of the GE estimates is appropriate. A comparison of the BNL and GE fragilities is given in Figure 3.
5.
BNL has nearly completed their review of the GE GESSAR-II seismic risk analysis. They have concluded that the GE analysis is less detailed than other seismic risk evaluations which have been performed and that additional work needs to be performed before the GE evaluation is useful in representing the core damage risk from a seismic event. Tae seismic hazard, fragility, and systems analysis areas were all found to be deficient in some ways. BNL performed their own analysis and concluded that the core damage frequency varies from 10-6/ reactor year to 10-3/ reactor for the range of fragility estimates and seismic hazard curves used to describe sites in the Eastern United States. This would leave earthquakes as a potentially significant contributor to the frequency of core melt for some sites. BNL identified two potential suppression pool 2
4 4
v GESSAR-II/RPA Meeting Minutes February 26, 1985 I
bypass sequences (collapse of part of the suppression pool wall and draining of the suppression pool through a failed RHR heat exhanger) which were not considered by GE. GE is currently reviewing the implications of BNL findings on potential suppression pool bypass.
'f 6.
The NRC Staff and BNL have evaluated the differencies between the GE and BNL results. GE has calculated the frequency of core melt from an earthquake (using the LESSAR-II hazard curve) at 6x10-7/ reactor year (point estimate). The BNL modifications of the GE system analysis would increase the core melt reactor year frequency to 6.3x10-6/ reactor year. Using the BNL fragilities would increase the core melt frequency by about a factor of 10, from 6.3x10-6/ reactor year to 6.7x10-5/ reactor year. The use of
" worst-case" fragility values and unfavorable siting conditions could increase this point estimate to approximately 10-3/ reactor year (which would exceed the proposed Coninission Safety Policy core melt guideline of 10-4/ reactor year for all contributors). The inclusion of relay chatter in the analysis accounted for much of the increase BNL's in estimates of core melt frequency.
7.
The significance of relay chatter was discussed to some extent.
Relay chatter has been treated either superfically or not at all in the seismic PRA's which have been performed to date. The degree to which relay chatter will affect the performance of the reactor is dependent on the configuration of the individual circuits.
Circuits which well " trip" or " latch" with a single relay contact accuation would have the most serious effect. Solid state relay may also malfunction in an earthquake.
NOTE: Additional meeting details can be obtained from a transcript of this meeting available in the NRC Public Document Room, 1717-H Street, N.W., Washington, D.C., or can be purchased from ACE-Federal Reporters, 444 North Capitol Street, Washington, 0.C., 20001 (202) 347-3700.
- r
P e3,
LIMERICK - -.-
INDIAN POINT D
ZION
- a I
OYSTER CREEK -
i g
\\
d
\\
i-s i
s s
\\\\\\
\\'
\\
\\'
'g
\\
GESSAR 0-5,
N s
5 N
- \\
OYSTER
\\
CREEK ~
LIMERICK g
c_
x
' 's s\\
210N s
INDIAN POINT
\\
- 30'7
,s 1
I I
I I
I I
i
)
i O
0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 3.0 EFFECTIVE PEAK GROURD ACCELERATION (g)
GESSAR II Seisuic Hazard Curve Compared to Hazard Curve Studies.
dd
.~,
4 i.
IE-i!E'1 -.
1 SEu AR II B Limerick-ERTEC 2 22cn-55 MAP 9 Lkerick-LLNL 3 Zion-D&M (no truncation) 10 Millstone PSS
~
4 Zion-D&M (truncated) 11 Millstone-LLNL
~
5 Indian Point-D&M (same as Limerick (no truncation)
LLNL) 3 IEM,e
~
6 Indian Point-VCC 12 Seabrook-D&M 7 Indian Point-D&M/WCC 13 La Crosse-LLNL
~
co..bined 14 Watts Bar-LLNL (not shown)
-=
3 1E-923 'i 7
~.
g
~
G 14 U-0 1
=
L-1E-925
~
id I
\\
3 i
(
.=
3 1E-SEE --
5 3
E 6
4 1E-C7 :-
3 3
E i
I I
t
- 1. 5 3,3
.1 1
r PGA (g)
Cep: : son of the GESSAA II test-cstie.:-c Figure 2-2.
h u t r t' cc:vs to the results cf other J
s t i c.i hutrd studies.
(
s'f I
Jock R. Senicmin & Associcries,Inc.
B Consuffing Engineers
e DELETION 4
, m,.,.. m.m e..
.s >x m i
i
.