ML20085N468
| ML20085N468 | |
| Person / Time | |
|---|---|
| Site: | Waterford  |
| Issue date: | 11/07/1991 |
| From: | Burski R ENTERGY OPERATIONS, INC. |
| To: | NRC OFFICE OF INFORMATION RESOURCES MANAGEMENT (IRM) |
| References | |
| W3F1-91-0649, W3F1-91-649, NUDOCS 9111150157 | |
| Download: ML20085N468 (3) | |
Text
--
V ENTERGY f"T" ~"u"+ """'""
- - -.. ~.. _
_.,_,_'677;_.
R. F. Burski j
W3F1-91-0649 A4.05 OA November 7, 1991 U.S.
Nuclear Regulatory Commission ATTN:
Document Control Desk Washington, D.C.
20555
Subject:
Waterford 3 SES Docket No. 50-382 License No. NPF-38 Idaho National Engineering Laboratory (INEL)
Draft Report on Intersystem Loss of Coolant Accident (ISLOCA) Risks Gentlemen:
Waterford 3 on September 10, 1991, received a letter dated August 15, 1991, from the NRC Project Manager requesting comments on the subject report within sixty days of the receipt of the NRC letter.
The purpose of this letter is to provide Waterford 3 comments on the subject report.
Overall, we believe the report is reasonably accurate and indicates that the ISLOCA risk at Waterford 3 is very small. -The report highlights the excellent administrative controls, training, and interlocks that prevent human errors from contributing significantly to the ISLOCA risk.
Specific and detailed comments are provided in Attachment one (1).
We would like to thank the NRC for the opportunity to review and comment on the draft report.
Please contact me or Robert J.
Murillo should there be any questions regarding these comments.
Very truly yours,
- 0
.fw RFB/RJM/ssf Attachment cc:
R.D.
Martin (NRC Region IV),
R.B.
- McGehee, N.T Reynolds, NRC Resident Inspectors Office I
-9111150157 911107 PDR ADOCK 05p00382 il P
POR 7
Attcchment On3 to W3F1-91-0649 Pago 1 of 2 4
Waterford 3 Comments on Idaho National Encineerina Laboratory (INEL) Draft Report on Intersystem Loss of Coolant Accident (ISLOCA) Riska General The SDC system does not include two high pressure rocirculation lines to the RCS hot legs.
The high pressure safety injection system does inclucc two injection lines to the hot legs as shown on Figure 3.3.
[Page 17)
In Figure 3.4 & B.11, air operated valve SI-106A/B is not shown between the RWSP and SI-107A/B.
Also, check valves SI-107, 1071, and 108A/B are drawn backwards (showing flow in the reverse direction *.
[Page 21, B-24]
Eecuence 1B - SDC ISLOCA Durina Startup l
This sequence relies on the Automatic Closure Interlock (ACI) to l
isolate low pressure piping as pressure increases during startup.
Waterford 3 has removed this interlock, but has installed alarms to alert the operator that the valves are not fully closed when pressure increases above the alarm setpoint.
This alarm provides equal (if not better) protection against an ISLOCA.
Thus, even through ACI has been removed, a value of 1.0E-03 is appropriate for ACIF in the event trees of Figure 4.1 and B.2.
[Page B-4]
The screening values for ACIF (1.0E-03) and human failures OFIP, FTD, and FTDON (1.0) given in the text are not the same as shown in the event tree in Figures 4.1 and B.2 (3.0E-03 and 0.5).
[Page D-7) gecuence 2 - LPSI Disct3rce ISLOCA No. credit for an alarm on high pressure between the upstream check valves (indicating a failed open check valve) was given because an annunciator card for a similar alarm on the high pressure hot leg injection line was pulled.
The alarm for the LPSI discharge line has typically been in service and would therefore, be available to alert the operator not to open the l
valve with a potentially failed open check valve.
Credit for this alarm would reduce the core damage frequency for ISLOCA below 1.0E-06.
[Paga 28]
l l
I I
Attechm nt One to W3F1-91-0649 Pago 2 of 2 v
The analysis assumes that the operator fails to isolate the leak given that the operator has detected and properly diagnosed.he LOCA.
This assumption does not account for an operator's first reaction which is to undo the wrong action particularly if there were an immediate and obvious indication of a problem (i.e.,
close the LPSI discharge valve that he has just opened).
An operator failure probability of 1.0 for FTI in Figures 4.2 and B.4 is thus overly conservative.
[Page B-12)
Secruence 5 - ISLOCA Duriac BJC Tha analysis of this sequence assumed that check valve SI-108A/B was failed open with a probabilit) of 1.0 based on the as-found condition of the valve.
Although this valve was not leak tight and would have allowed pressurization of the downstream low pressuro piping, the leak rate through the valve was not sufficient to result in a large ISLOCA.
The back flow rate through EI.
08 was estimatod, based on test data, to be about 15 gpm.
Since this is well within the capacity of one charging pump, the operator would have sufficient time to depressurize and prevent core damage.
Thus, the core damage probability is even lower because another failure (e.g. gross failure of BI-108) l would need to be included.
(Note that closing motor operated BI-IC6 also prevents draining the RWSP in the event of a break in the suction piping).
[Page 42, B-23]
I