ML20244D633

Forwards SER Providing Risk Perspective of Util Proposal to Increase Allotted Outage Time from 3 to 7 Days for safety- Related Sys,Per Review of Util PRA (WCAP-10526).Review Results Listed

ML20244D633
Person / Time
Site:	Byron
Issue date:	01/15/1986
From:	Speis T Office of Nuclear Reactor Regulation
To:	Thompson H Office of Nuclear Reactor Regulation
Shared Package
ML20236A787	List: ML20236A785 ML20244D633
References
TAC-M57242, NUDOCS 8601230336
Download: ML20244D633 (4)

Text

_ _ _ _ _ - _ _ _

, h 'ghan

1. 'o

~g UNITED STATES

- ! "' o NUCLEAR REGULATORY COMMISSION

h. $ WASHINGTON, D. C. 20555

%.,,,# JAN 151985 l MEMORANDUM FOR: Hugh L. Thompson, Jr., Director Division of PWR Licensing-A, NRR FROM: Themis P. Speis, Director Division of Safety Review and Oversight

SUBJECT:

SAFETY EVALUATION REPORT RELATED TO THE LC0 RELAXATION '

PROGRAM FOR THE BYF,0N GENERATING STATION (TAC NO. M57242)

We have completed our review of the probabilistic risk analysis

. (WCAP-10526) submitted by Commonwealth Edison in support of their proposal

to increase the Allowed Outage Times (A0Ts) from 3 days to 7 days for nine safety related systems in the Byron plant. This Safety Evaluation Report provides our risk perspective of the proposal, and is submitted as supporting y information for decision making by NRC. In brief, the results of our review

! are:

I Flagged a significant plant vulnerability in connection

! with Byron Unit 1 operating with its two pump essential i

service water (ESW) system without established cross-tie 4

capability for ESW backup from Unit 2; interim resolution 1 of this problem, pending completion of Unit 2, was obtained j by Commonwealth Edison's commitment of December 6, 1986 to j make at least one ESW pump from Unit 2 available for Unit 1 i prior to the operational restart of Unit 1; and i Support a decision to permit the proposed increase in A0T for six safety related systems in the Byron plant.

! Support denial of the proposed increase in A0T for the essential service water system, diesel generators, and the auxiliary

, feedwater system.

The Applicant's submittal was reviewed by the Reliability and Risk Assessment Branch with technical assistance by the Brookhaven National Laboratory (BNL). The Applicant's risk analysis utilized various PRA l! models and data developed for the Zion Probabilistic Safety Study, the j, rationale for this being the general similarity in the design and operation of the Zion and Byron plants.

l,

Contact:

A. Spano, RSIB Ext. 28302

a. . r, , i JM 151986. i I

]

i e

Our review approach involved the following considerations:

(a) _ Single Unit Operation: The review was performed in the context of the existing operational status at Byron, i.e., operation of Unit  ;

1 only, with no established capability for inter-unit sharing of i the essential water system. This is of consequence in regard to the l assessment of risk posed by the Byron 1 two pump ESWS, which, in. '

the absence of the potential backup capability that could be pro-vided by the Unit 2 ESW pumps, can fail with a frequency of'about 9x10 4 per year. Failure of this vital support system fails important safety functions, including the cooling of the reactor coolant pump (RCP) seals, which can lead (with an assumed probability of 0.5) to a non recoverable seal LOCA. Thus, the Loss-of-ESW initiating event by itself is estimated to-provide a direct contri-bution to core melt frequency of about 5x10 4/ year. In this connection, i the large difference in Byron 1 core melt frequency results obtained . ,

between BNL and the Applicant's study is principally due to the fact j i

that this accident initiator was not considered in WCAP-10526. I Estimates of the core melt frequency when a Unit 2 ESW pump is avail-able to Unit 1 are presented later.

(b) Risk Index The Applicant's study presented the impact of the proposed change in A0Ts in terms of changes in offsite health risks.

/

The staff analysis used core melt frequency as a risk measure, be- '

cause it is more directly related to changes in A0T, and because, in general, the core melt frequency is the limiting constraint to the proposed numerical guidelines of the NRC safety goal.

(c) System-Specific A0T Risk Impact: The Applicant's results apply to the case where the A0T change is applied to all nine proposed systems at the same time. This does not yield information on the relative risk impact posed by each of the various systems. The BNL analysis examined the effect of an A0T change on individual systems as well as on the proposed group of nine systems.

(d) Sensitivity Approach: The Applicant's A0T analysis was based on  !

Bayesian estimated values of the frequency and duration of maintenance events. The relevant data base, both generic and

• plant specific, is of meager proportions. In the BNL analysis, uncertainties related to frequency and duration were addressed using a sensitivity approach in which the core melt frequency was linearly related to the mean maintenance duration time (Tm) for the systems of interest. The analysis covered the Tm range from 19 hours2.199074e-4 days <br />0.00528 hours <br />3.141534e-5 weeks <br />7.2295e-6 months <br /> (estimated to be the mean maintenance duration time for {

the current 3-day A0T technical specification) to the full 7-day {

period of 168 hours0.00194 days <br />0.0467 hours <br />2.777778e-4 weeks <br />6.3924e-5 months <br />.

i i

ZrrI7 ~ ' ~ ~ ~ ~ ~ ~ ~

l ;, , w;

% ! 5 49 l 4-l The interim BNL results indicated the Byron 1 baseline core melt frequency for the existing 3-day A0T condition to be about 1x10 3/ year, the main con-tribution to this arising from the vulnerability posed by the Byron two pump ,

ESWS with no backup. Discussion of these results in meetings with the l

4 Applicant at NRC on December 3 and 6, 1985 focused on the question of the i expected reduction in core melt frequency that could be obtained if:

1 (a) revisions were made in three calculational factors affecting the i i

calculated annual frequency for loss of the Unit 1 ESW system and (b) one of l 1

the two Unit-2 ESW pumps were to be made available as additional emergency i l- backup to the two pump system in Unit 1.

j. The net result of revising the three factors under (a) above, was an P approximate one-third reduction in the initiating event frequency for loss '

of ESW in Byron 1 (from about 1.5x10 3/ year to about 9.5x10 4/ year; see Appendix 1, Enclosure), with the net overall core melt frequency for Byron 1 j with 2 ESW pumps reduced from about 1x10 3/ year to about 7x10 4/ year. (These -

l estimates do not include a possible contribution due to common cause failures l between the ESW pumps). ,

t 1 This estimated core melt frequency of 7x10 4/ year for the Byron 1 two pump ESW is high. However, with the availability of a Unit-2 ESW pump (as committed to

.l by Commonwealth Ejison on December 6,1985) the initiating event frequency for loss of ESW is reduced by a factor of about 25, while the Byron core melt frequency point-estimate is reduced to nearly 2x10 4/ year.

1 As discussed more fully in the Enclosure, th9 core melt frequency results obtained with the mean maintenance duration time increased to a full 7 days

indicate the Byron 1 risk impact for the two-ESW pump configuration to be j unacceptably high for the A0T increase applied to the ESWS, appreciably high for the DG and AFW system, and of little significance for the other six proposed  ;

systems. While this conclusion clearly holds for Tm=168 hours, we believe 1);

the conclusion also holds for reduced values of Tm comparable with Zion plant experience.

1 j For the 3 pump ESW configuration, we believe the risk impact associated with increasing the A0T for the ESWS would still be too high. As indicated in the Enclosure, the primary reason for this is attributed to the sensitivity

of the loss of ESW initiating event frequency to maintenance. Accordingly, a ,l

relatively high rate of increase in core melt frequency is estimated to be  !

obtained for the 3 pump case, amounting to approximately 1x10 6/ year for each l

additional hour of maintenance.

)' In conclusion, the results of the staff's probabilistic risk evaluation of

, the Byron A0T proposal would (a) support relaxation of the 3-day A0T for the t1 following systems: chargingpumps,safetyinjectionpumps,RHRpumps, fan  !

coolers, containment spray pumps, and component cooling water pumps; (b) not

! support relaxation of the 3-day A0T for the following systems: essential l water system, diesel generators, and auxiliary feedwater pumps.

i l

i

4-Finally, it is important to point out that the risk perspective obtained in our evaluation is dependent, in particular, on the assumptions of-frequency and maintenance outage times provided in the Applicant's submittal. The technical specifications. set forth for Byron 1 provide, in general, no control on the. frequency of maintenance events, and, therefore, on the cum'lative u outage' times for the various systems. At the least, reasonable prudency would appear to warrant self-monitoring' actions by the utility to obtain trending.information on the frequency of maintenance' events, especially for the risk important ESW and DG support' systems, and the two pump AFW system.

~

no . _ _ ,

.hemis P. Spels, Director t .

Division of Safety Review and Oversight ,

Enclosure:

As stated cc: H. Denton R. Bernero

. .F. Miraglia j J. Knight .

V. Noonan j C t01shan % j E. Butcher'

H 1

j

. _ . - - - - _ _ _ _ _