ML20209H692

Forwards Comments on Preliminary Accident Sequence Precursor Analysis Provided in NRC Re Operational Condition Reported in LER 269/1998-04

ML20209H692
Person / Time
Site:	Oconee
Issue date:	07/15/1999
From:	Mccollum W DUKE POWER CO.
To:	NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM)
References
NUDOCS 9907210044
Download: ML20209H692 (5)

Text

. - - -

Iv- 'j. , . x.

(*).

. # ~3 Duke Power Ik ,' ,LM ,,

Oconce Nuclear Site m ,, %,, 7800 Rochester liighway l

Seneca, SC 29672 W. R. McCollum, Jr. (864) 885-3107 omct Vice Pmident (864) 885-3564 m C j; ,

.m July 15, 1999  !

U.S. Nuclear Regulatory Commission  !

' Document Control Desk l Washington, DC 20555 )

\

Subject:

Duke Power Company ]

Oconee Nuclear Station, Units 1, 2, and 3  !

Docket Nos. 50-269, -270, and -287 Review of Preliminary-Accident Sequence Precursor i Analysis In a letter dated June 10, 1999, the NRC Staff provided a j copy of.the preliminary Accident Sequence Precursor analysis of an operational condition that was reported in Licensee Event Report (LER) No. 269/1998-04. This copy was provided for our review and comment on the preliminary analysis, whii:h was prepared by a NRC contractor at Oak Ridge National Laboratory. Duke Energy Corporation appreciates this opport. unity and has reviewed the preliminary analysis as requested. Our comments are included as an attachment to j this letter. l If there are any questions or further information is needed you.may contact R. P. Todd at (864) 885-3418 or H. D. Brewer at ' (704) 382-7409.

l

\

3 Very truly.yours, W. R. McCollum, J

' Site Vice Presi 'nt

, ta \

s O Attachments 9907210044 99071 9

DR A00CK 05 ,

L

ye , .o

1. --[ l.Ul1S. NuclanriRagu btory Commission 1, July:. 15',bl999-Page~2;

~

ci ( .

- gj t ,- ,

fxc .((w/attch) : 'Mr.- Luis A.~Reyes  ;

Administrator, Region'II 7 U.S. Nuclear Regulatory Commission

..S.  : ~p -611Forsyth Street, S. W., Suite 23T85 NWfi '

' Atlanta,.GA '30303

-s Mr. D. E. ' La' Barge :

g y.. }' -

U.S? Nuclear Regulatory Commission-Office of Nuclear Reactor Regulation Washington', D.C. 20555

-g'.

^'

x'c(w/okhtch): Mr. M. A.

Scott-U

. Senior NRC Resident' Inspector

y. ci Oconee Nuclear Station '

, y ri

'i.

g 1

p /.., '

3

>i,.,

I I

)

<1 1

i

}i I

)

i i

s p

I j.

., . l

'. i

r- ,

. 1

)

Attachment 1 The.ORNL precursor evaluation is thorough and well thought l out. Overall, the values selected for the various parameters I are reasonable. There is, however, one modeling assumption that Duke finds to be too pessimistic. Duke believes the non-recovery probabilities assigned to LPA-REC are too high. It is recognized that insufficient information is available on the i recovery being considered for quantification by any generally accepted technique. As a result, there is a natural tendency  ;

to assign conservative non-recovery probabilities. The considerations that Duke believes provide a basis for less pessimistic assumptions regarding the potential for recovery are identified below. The timing estimates provided are based on RELAP and MAAP analyses of a large hot leg break with all engineered safeguards available.

The ORNL precursor-evaluation assured a non-recovery probability of 1 for the large LOCA cases. Such an assumption implies complete certainty that the action will fail. A failure probability of 1 seems pessimistic for the following reasons:

1. Based on the RELAP and MAAP simulations, it is estimated that at least 30 minutes are available following the loss of injection before core damage would begin. While there would certainly be some concern over the cause of failure of the previously operating LPI pumps, it is also clear that inaction will lead to core damage (the inadequate core cooling section of the EP will be entered following core uncovery). Given that aligning the C LPI pump requires only a few minutes, at least 30 minutes are expected to be available for evaluation of the situation.

2. Significant core damage is not expected for more than an hour following the initiation of the LOCA. The TSC. or at

~

least some of the TSC staff, should be available to assist in the evaluation.

l

3. These timing estimates assume a very large break of the horizontal portion of the hot ..eg. Breaks at locations higher than the horizontal run of the hot leg will provide significantly more liquid inventory in the RCS at the time that injection is lost and extend the time to significant core heatup. Breaks at the small end of the large LOCA range may also afford additional time.

I

e Attachment 1 (Continued)

For these reasons, we feel that a value of 1 for the non-recovery probability reflects a degree of certainty that is not appropriate. A value of 0.5 for the non-recovery probability would be indicative of complete uncertainty in the recovery potential, reflecting neither a pessimistic nor

! cptimistic bias in the assumed recovery potential. A value of 0.5 is a more appropriate (but possibly conservative) selection for the non-recovery probability.

The ORNL precursor evaluation assumed a non-recovery probability of 0.5 for the medium LOCA cases. Such an assumption implies complete uncertainty in the success potential. A more optimistic view for the recovery potential I is appropriate for the following reasons:

l l 1. For the medium LOCAs, approximately 90 minutes is required to deplete the BWST inventory. The time to significant core heatup following the loss of injection should be longer than for the large LOCA case discussed previously. The time available for evaluation in suf ficient to establish a reasonable understar, r of the l nature of the events.

l 2. The TSC is expected to be available prior to the loss of injection. Because the TSC will be in place during the important stages of the event, their evaluation is more i likely to be rapid and correct. The availability of the TSC is expected to aid the control room in deter:nining the appropriate action.

3. Considerations of break size and location that are not the most limiting would also contribute to a higher likelihood of success.

The availability of the TSC to monitor the accident and assist in the diagnosis and decision making is expected to provide reasonable reliability in arriving at an appropriate course of action. Success under these conditions is likely and a non-recovery probability of 0.1 is judged to be a more appropriate value. The medium LOCA situation is judged to be similar to the small LOCA situation because of the TSC availability.

The discussion presented above provides suggested revisions to the non-recovery probabilities assumed for LPR-REC. While even 2

\_

fl l 1

I Attachment 1 (Continued) lower values than suggested might be appropriate, it is judged that the suggested values do not contain a significant bias in either a pessimistic or optimistic direction.

l l

l 1

l l

l l

l l

l

[

l i

i 3