ML20155G390
| ML20155G390 | |
| Person / Time | |
|---|---|
| Issue date: | 11/05/1998 |
| From: | NRC (Affiliation Not Assigned) |
| To: | |
| Shared Package | |
| ML20155G292 | List: |
| References | |
| NUDOCS 9811090042 | |
| Download: ML20155G390 (4) | |
Text
- -. -. _ - - -...---. - - -.-
p.
a.
W D
ENCLOSURE 1 M.b SAFETY EVALUATION BY THE OFFICE OF NUCLEAR REACTOR REGULATION F
REVIEW OF BWR OWNERS GROUP REPORT GENE-770-06-2 f
ON JUSTlFICATION FOR EXTENDING SURVElLLANCE TEST INTERVALS AND F
ALLOWED OUT-OF-SERVICE TIMES FOR BWR REACTOR CORE ISOLATION COOLING SYSTEM INSTRUMENTATION p.
1.0
SUMMARY
The staff has reviewed the General Electric Company (GE) Topical Report E
GENE-770-06-2, " Addendum to Bases for Changes to Surveillance Test Intervals
't and Allowed Out-of-Service Times for Selected Instrumentation Technical Specifications" (Ref.1), issued by the BWR Owners Group to support the proposed extensions of R C actuation instrumentation surveillance test intervals (ST!s) and allowed out-of-service times (A0Ts) for test and repair.
In particular, bases are provided for extending surveillance test intervals (STis) from 31 days to 92 days. Additional justifications are provided for extending 1) allowed P,.
out-of-service times (A0Ts) for surveillance test from 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> and
- 2) allowed out-of-service times (A0Ts) for repair from I hour to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
2.0 BACKGROUND
'I The staff, in its SER of December 9,1988 (Ref. 2), accepted the analytic methods presented in the NEDC-30936 Part 1 topical report (Ref. 3). These analyses have demonstrated the acceptability of a general methodology for considering Technical Specification (TS) changes for the ECCS actuation.
v
'I instrumentation in boiling water reactors (BWRs).
In the Part I report, the change in the water injection function unavailability was calculated by changing STI h
and A0T requirements one compunent at a time. The change was sufficiently small that the average unavailability with respect to time was ignored.
n ff In the Part 2 report, however, changes to the STis and A0Ts for all ECCS actuation instrumentation were proposed.
It was proposed to chenge all STis from 31 to 92 days, ali surveillance test A0Ts from 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />, and i
l all repair A0Ts from one hour to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />. The same changes are proposed for RCIC actuation instrumentation in GENE-770-06-2, the topical report under consideration herein.
h Because the RCIC and ECCS systems and their respective actuation instrumentation are so similar, the NRC staff has judged that the RCIC actuation instrumentation and ECCS (HPC1/HPCS) actuation instrtmentation modified STI and A0T risk impacts may be compared and evaluated according to similar criteria.
These are:
t a) A 4 percent or less increase in water injection function unavailability M
is acceptable, and p
b) where criterion (a) is not met, an increase of no more than 1.0E-6 per year in water injection function f ailure frequency is acceptable.
u i
The analyses presented in GENE-770-06-2 are based on the loss of Offsite Power l
(LOSP) initiating event, which contributes 40% to 90% of the overall core damage l
y frequency.
9811090042 981105 PDR TOPRP EMVQENE
^
l C
PM l
~
" ' ~ =
O 2
3.0 APPROACH The analyses performed required an evaluation of the average unavailability of the water injection function with respect to time. The approach used by GE was the determination of the average unavailability 'of the system over the STI of irterest, not including the impacts of any test or repair activities. Analyzed component unavailabilities are combinations of demand failure probabilities and average time dependent failure probabilities. Water injection function unavailability increases due to test and repair A0T increases were found in Reference 4 to have values of about 0.5%, which is very small, and were therefore not recalculated.
4.0 HRC ACTION The staff engaged the services of EG&G Idaho (WTO) to review the approach used and analyses performed in the GE report. This review was performed to determine j
the adequacy of the methods used to establish the technical bases for the modifications of STis and A0Ts for BWR RCIC actuation instrumentation.
THE GE/BWROG methodology first requires data on initiating event frequency, component rellability data including time-dependent and time-independent failure rates, system descriptions, and success criteria. Next, fault trees are generated for relevant systems, with emphasis on the modeling of the ECCS and RCIC actuation instrumentation. Then these fault trees are assembled to form the water injection The functional fault tree, with ECCS and RCIC unavailabilities as the top event.
water injection function failure fault tree is then evaluated using fault tree I
quantification codes.
A four percent water injection function unavailability increase was used as the acceptance criterion, in addition, where the four percent increase limit cannot be met, the guideline is augmented by limiting the impact of STI and A07 changes to an absolute increase of 1.0E-6/yr in water injection f ailure frequency. These l
r criteria are acceptable to the staff.
each of the product line groups were performed to show Therefore, analyses f or the specific effect of the proposed changes to the RCIC actuation instrumentation STis on the water injection function unavailability using the models and input l
data developed and documented in Refs. 3 and 4 For each product line group, l
the water injection function unavailability was first calculated with both RCIC and ECCS actuation instrumentation STis set at 31 days. Second, the unavailability i
was calculated with both RCIC and ECCS actuation instrumentation STis set at Third, the unavailability was calculated with a 31 day STI for RCIC 92 days.
j The actuation instrumentation and a 92 day ECCS actuation instrumentation ST1.
first calculation g Nes the base case unavailability, the second calculation i
gives the gross change in unavailability, and the third calculation gives the l
change in unavailability due only to the change in the ECCS actuation instrumen-tation ST1. The change in unavailability due only to the change in RCIC actuation instrumentation STI is obtained by subtracting the result of the third calculation from the result of the second. Corresponding sets of calculations for increases in RCIC actuation instrumentation test and repair A0Ts were not made because from Ref. 4 the increase in f ailure frequency for ECCS + RCIC when both RCIC and ECCS actuation instrumentation A0Ts were increa were all in the range of 0.5% which is negligible.
3 D
~
a
_..m___.m.
.m
~
3-In order for a licensee to use the generic analysis as justification for RCIC actuation instrumentation STI and A0T changes, the licensee should provide verification that either the appropriate BWR product line ger.eric model or one of the envelope cases for the appropriate product line provides an accurate or conservative representation of its plant as discussed in the enclosed EG&G Technical Evaluation Report (TER).
5.0 CONCLUSION
S On the basis of its review of the EG&G Idaho (WTO) TER, the staff endorses the finding therein that the methods used and results obtained in GENE-770-06-2 were verified. The staff therefore finds thr analyses presented in this topical report acceptable to support the proposed extensions of surveillance test intervals for RCIC actuation instrumentation from 31 days to 92 days.
Ihe TER concludes that additional analyses may be necessary for aarly-model plants for which the actuation instrumentation may not be seismically and environmentally qualified.
Such qualification is required under 10 CFR 50.49, in accordance with the licensed design basis. Therefore, the staff concludes that additional analyses are not necessary, beyond that required for a determination by the licensee that the i
generic aglyses contained in this topical report are applicable to the licensed design bcsi:.
i The GE/dWROG analyses provide adequate conservative bases for extending the RCIC actuation instrumentation allowed outage time for test from 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> to 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br />, and for repair from one hour to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
i
- iable 1 lists plant-specific conditions that each licensee or applicant must meet to make any proposed STI or A0T changes fully acceptable.
Table 2 summarizes the approved changes.
1
6.0 REFERENCES
1.
W. P. Sullivan, et al., " Addendum to Bases for Changes to Surveillance Test Intervals and Allowed Out-of-Service Times for Selected Instrumentation Technical Specifications," General Electric Company, GENE-770-06-2, Febrtfory 1991.
2.
Safety Evaluation by the Office of Nuclear Reactor Regulation - " Review of BWR Owners Group Report NELC-30936P-(Part 1) on Justification for Extending On-Line Test intervals and Allowable Out-of-Service Times for BWR Emergency Core Cooling System instrumentation," December 9,1988.
3.
D. B. Atcheson, et al., *BWR Owners Group Technical Specification improvement Methodology (With Ocmonstratioe for BWR ECCS Actuation Instrumentation) Part 1," General Elactric Company, NEDC-30936P, November 1985.
4.
D. B. Atcheson, et al., "BWR Owners. Group Technical Specification Improvement Methodology (with Demonstration for BWR ECCS Actuation Instrumentation) Part 2," General Electric Company, NEDC-30936P, June 1987, t
o
+
b
,f'
&~_',f-m--
- q g.' ;' y y yg,
',v.
- T w 5 g - ~ n ( ow q pgg-g - g sw, y
. ~. - -.
':%?";
l
's,.:
i s' N 4
l N
TABLE 1 CONDITIONS TO CLOSE OUT P1 ANTS For plant specific application of the TS changes for RCIC actuation instrumentation that are proposed, the licensee must:
(1) Confirm the applicability of the generic analyses to the plant.
t c.
Confirm that any increase in instrument drift due to the extended STIs (2) is properly accounted for in the setpoint calculation methodology.
(For s.
je Jn R. F. Janecek, dated April 27, 1988.;
l t
J TABLE 2 l
rr APPROVED STI AND A0T CHANGES s
(
r.
?
STI 31 days - 92 days (monthly - quarterly) l
((
TEST A0T 2 hours2.314815e-5 days <br />5.555556e-4 hours <br />3.306878e-6 weeks <br />7.61e-7 months <br /> - 6 hours6.944444e-5 days <br />0.00167 hours <br />9.920635e-6 weeks <br />2.283e-6 months <br /> T,.,
REPAIR A0T 1 hour1.157407e-5 days <br />2.777778e-4 hours <br />1.653439e-6 weeks <br />3.805e-7 months <br />
- 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> l
?
i<
I kr
.l
'f-i tV t
N l
,c
.s
'h s
't
- s.
- n
,3;
.