ML20127A959
Text
r 77-
..n a
~
- c n /. ~
y.
- v j.
OCT08 W3 9
MEMORANDUM FOR:
J. Hulman, Chief y
Accident Evaluation Branch
.[
FROM:
A. Thadani, Chief-Q Reliability and Risk Assessment ~ Branch f
SUBJECT:
SEQUENCE FREQUENCY INPUT TO LGS DES d
Ref. 1: Memorandum from 8. Sheron to J. Hulman, Subj: " Set of Release
.J -
Categories for Limerick OES," dated August 15, 1983.
Ref. 2: Memorandum from J. Knight to F. Rowsome, Subj: " Review of Seismic Elements in the Limerick Generating Station Severe Accident Risk Assessment (LGS-SARA)", September 23, 1983 Per your request to me upon receipt of Reference 1, enclosed (as Enclosure 1) is a list of the sequence classes and their estimated frequency that we consider more representative.of Limerick for the DES than those from WASH-1400. The point estimates reflect mean values to be used in carefully constructed comparisons rather than as estimates of the absolute values.
The internal event sequences and' frequency estimates of Enclosure 1 are the same as those contained in BNL's review [NUREG/CR-3028] of the LGS PRA subnitted by PECo in March 1981.
The seismic and fire sequences and frequency estimates were obtained from the Limerick Severe Accident Risk Assessment Report submitted by PECo in April 1983. Although the staff's review of the LGS sequences and their frequencies is in progress, we are nearing completion i
and regard the values identified in Enclosure 1 to be representative of
'~
Limerick. Additionally, we do not expect the sequences or values to change significantly in the near future. We will keep you advised of any development.
l Althougn the above values are more representative of Limerick than the WASH-L400 values used in other OES caculations, we believe that the use of a single value, such as the point estimate, to characterize a sequence frequency could be a misleading simplification in the light of large uncertainties. Any oiscussion l
or use of sequence frequencies should take into a: count these uncertainties.
We have provided Enclosure 2 for insight into the I.:rge uncertainties around estimated frequencies such as those associated with earthquakes. The range reflects PECo's "95%" and "5%" limits. We believe that the high and low values of the estimated frequencies should not be characterized as 95% and 5%
p h
E
Contact:
E. Chelliah, RRAB 28338 f510IBOObh u
y-tmce)
- w e>
ons)
{
- otu m ne am *xu ma OFFICI AL R ECORD COPY
.,u~a.m.m,- -
s
+.
v.:
~.
El'.
OCT 06 G83 61
};
2 li 1
limits, rather only as a representative range of the seismic sequences
~3.
frequencies which incorporate a large part, buc not necessarily all, of the
- i uncertainties.
This range may be used in the LGS DES as a representative
!.i.!.
range for the environmental impact from earthquakes.
It is preferred to the
~
.j use of any single number.
A more extended discussion regarding the proper i.' } -
use of earthquake related frequencies can be found in the September 23, 1983
]*
memorandum (Reference 2) from J. P. Knight (DE) to F. Rowsome (DST.
g.
q.
.d
~.
Ashok Thadant, Chief Reliability and Risk Assessment Branch
Enclosures:
As stated cc:
T. Speis R. Mattson F. Rowsome A. Schwencer B. Sheron S. Acharya J. Repenthal J. Meyer R. Martin J. Knight R. Vollmer Distribution Central File RRAB r/f A. Thadani E. Chelltah F. Coffman E. Chellfah CHRON c c l.... A.
jf nA TA' U
......R....B..:. 0..S..T 3 7 R A.8..:.0..S..T........
.G..S. 8.. 0............
.G... M.....,....
.RR. A..S..;QS R
.c,r,co
=~~>
... ECD,gl,,],1,,ah,b,f,C,q f fm,an,,,,,,,(R,,,,,,, e. r...........Rd 10n'..
.AIhadan.......
on.,....W.o.L83......10./.013.........Wpza.3.........w
.aa.......101.41.u...........................
ac roau sts oo.m wcu cuo OFFICIAL RECOR D' COPY esc.c..wi m.
- h 9
N
.s.
5 ENCLOSURE 1
.dl
- n
'4, ESTIMATED FREQUENCY FOR SEQUENCE CLASSES 5[1 -([
REPRESENTATIVE OF LGS FOR DES 4
jf:
jl ESTIMATED F R E Q U E N C Y [RY]-1 Random Total Sequence Internal Events (2)
External Events (3)
Vessel ( ) Frequency Class (1) Transients LOCAs Fire Seismic Failure By Class-(4) t*;
(5)
CLASS I 9.5E-5 7.6E-8 4.6E-6 (7)
NA 1.E-4
~
CLASS II 4.1E-6 8.5E-8 (6)
(7)
NA 4.2E-6 CLASS III 3.3E-6 3.3E-8 (6)
(7)
NA 4.1E-6 CLASS IV 3.2E-7 5.2E-9 (6)
(7)
NA 4.3E-6 CLASS 15 NA NA NA (7)
NA 1.2E-6 (6)
CLASS S NA NA NA (7) 2.7E-8 4.4E-7 TOTAL 1.03E-4 2.'0E-7
- 4. 6 E-6 (7) 2.7E-8 1.1E-4 a
FOOTNOTE: (1) For sequence class description, refer to Table 12-4 of LGS SARA (2) Frequency estimates contained in NUREG/CR-3028 l
(3) Frequency estimates contained in LGS SARA (4) Total frequencies include the point estimates for seismic
-events from LGS SARA (5) Includes BNL's conservative estimate for Loss of Offsite Power (6) PECo has not identified any sequences (7) See enclosure 2 for seismic sequence frequencies and the Memorandum from J. P. Knight to F. Rowsome, September 23, 1983, (Reference 2) for a discussion of their proper use
1-3 J
~
- }/
- e ENCLOSURE 2 i
4 1
SEISMIC SEQUENCE FREQUENCY ESTIMATES FOR LGS DES CONTAINED IN LGS SARA
- f. j.
g, SEQUENCE FREQUENCY RANGE
[RY3-1 S.
CLA35 (1)
Low Median High POINT ESTIMATE f
Clas.s I 1.3E-9 1.7E-7 1.7E-5 3.2E-6 Class II (2)
(2)
(2)
- 5. 0 E-8.
Class III 2.7E-12 1.8E-8 4.9E-6 8.2E-7 Class IV 2.9E-13 2.1E-9 6.7E-7 1.1E-7 Class 15 8.0E-14 7.6E-7 7.0E-6 1.2E-6 Class S 1.9E-21 3.2E-11 2.5E-6 4.1E-7 Total Seismic Sequence Frequency 5.8E-6 FOOTNOTE:
(1)
Class I Sequences:
T E W 3 3 Class II Sequences:
T E W 3 3 Class III Sequences:
T3 (RPV) Rg E H
T E C C 3 S g 2 Class IV Sequences:
T E C C 3 3 g 2 Class 15 Sequences:
T R 3 g T E C S B g T E W S 3 Class 5 Sequences:
T3 (RPV) Rg T R H
3 B E
T3 (RPV) Rg E H
(2)
Estimate is not Available.
- -