ML17173A941
| ML17173A941 | |
| Person / Time | |
|---|---|
| Site: | Dresden  |
| Issue date: | 05/22/1979 |
| From: | Ziemann D Office of Nuclear Reactor Regulation |
| To: | Reed C COMMONWEALTH EDISON CO. |
| References | |
| TASK-03-10.C, TASK-3-10.C, TASK-RR NUDOCS 7906280263 | |
| Download: ML17173A941 (5) | |
Text
,_
MAY 2 2 1979
~k;~:~b-~~
Loe a 1 -PDR ORB#2 ROG Docket No. 50-237 NRR ROG VStello DE'.isenhut BGrimes RVollmer TCarter WRussell ATTORNEY, OELD OI&E(3)
DLZiemann P0 1 Connor HSmith JRBuchanan TERA ACRS(l6)
DDavi s
- Mr. Cordell Reed Assistant Vice President Commonwealth Edison Company Post Office Box 767 Chicago, Illinois 60690
Dear Mr. Reed:
RE:
TOPIC 'III-10.C :.. DRESDEN NUCLEAR PmJER STATION, mJIT NO. 2 Enclosed i~ a copy of our revised safety assess~lent of Topic III-10.C, Surveillance ReqL~irements on BHR Recirculation Pumps., This revision includes consideration of the comments received on the assessment issued b_y our letter dated August 17, 1978.
Your letter dated October 18, 1978,.provided comments on the assessment.
This*revision completes our assessment of Topic III-10.C which wnl be used as input to the integrated revie\\1 of the Dresden Unit No. 2 Plant.
If there are any errors in the facts of this revised assessment, please supply corrected information within 30 days of the date you*
receive this letter.
If_no response is received 1r1ithin that time;
_we wil 1 as_surne. that you have no furthE;:ir comments or corrections.
Enclosure:
Revised Assessment for To.pie III-10._C cc \\*J/encl osure:
Sincerely, Original Signed*by:
Dennis L. Ziemann Dennis L. Ziemann~ Chief Operating Rec?ctors Branch #2 Diyision* of Operating Reactors
- See next page
'f.9 0 6'2 8 0 ~:.6!'3.
- ::::. ;. ~~:i!:~: ::..
3*§? ~~-~-~*:*~*... *.:*.* '.::~i.; ::::~~-::::. : ::: : ::::: :: : ::: : : : : :: : : : : : : : : : :: : : : :: : : :: : : : : : :::. :
- , :::*. :'\\*. ::::::*. :::
DATii..
... ?!;J~~.?............~(~':!.!.~.............. ~!.~~--"""" """"""""""""'"""'..................... _
..... ;......... ~........,.,
, me POBH 31s <lM6> NRCM 02.co
'!t U.S. GGV*R'N~*NT PNINTING Ol"l"ICll: t 871 - &*I.. 7e*
\\
Mr. Cordell Reed cc Isham, Lincoln & Beale Counselors at Law One First National Plaza, 42nd Floor Chicago, Illinois "60603 Mr. B. 8. Stephenson Plant Superintendent Dresden Nuclear Power Station Rural Route #1 Morris, Illinois 60450 Anthony Z. Reisman Natural Resources Defense Council 917 15th Street, N. W.
Washington, D. C.
20005 U. S_. Nucle.ar Regulatory Commiss_ion ATTN:
Jimmy L. Barker P. O. Box 706 Morris, Illinois 60450 Susan N. Sekuler Assistant Attorney General Environmental Control Division 188 W. Randolph Street Suite 2315 Chicago, Illinois 60601 Morris Public Library.
604 Liberty Street
- Morris, I 11 i noi s 60451 K M C Inc.
ATTN:
Mr. Jack McEwen 1747 Pennsylvania Avenue, N. W.
Suite 1050 Washington, D. C.
20006 May 22, 1979
- SYSTEMATIC EVALUATION PROGRAM
- Topic III-10.C. Surveillance Requirements on BWR Recirculation Pump Discharge Valves Plant: Dresden Unit No. 2 DISCUSSION In July 1976, all BWR facilities whic~ had completed the Low Pressure Coolant Injection (LPCI) system modification, to remove the LPCI loop selection logic, were sent letters requesting that they apply for a license amendment to incorporate technical specification surveillance requirements on recirculation pump discharge and bypass valves. lhe recirculation pump discharge and bypass valves are required, for these plants, to close upon initiation of LPCI.
The closure of these valves is necessary to prevent the loss of cooling water by reverse flow through the pump or its bypass line and out the break. The failure of the recirculation pump discharge or bypass valve to close can adversely affect core cooling in a manner similar to the failure of a LPCI valve to open.
EVALUATION This topic applies to LPCI modified BWR facilities. Dresden Unit No. 2 retains LPCI loop selection logic and.therefore is not subject to the requirement of this topic. Furthermore, since the.unmodified LPCI is susceptible to single failures that can eliminate all LPCI flow, no credit is given fa~ any LPCI flows in the Dresden Unit No. 2 ECCS analysis. The staff's safety evaluation for Dresden Unit No. 2 core reload 2 describes the design basis event as the complete severance of the recirculation suction line assuming-a failure of the LPCI injection valve (safety eval-uation transmitted to licensee May 23, 1976).
An assumed failure of the LPCI va 1 ve prevents any LPCI fl ow from.entering the core. *An NRC Safety Evaluation dated December 27, 1974, for Dresden Unit No. 2 discusses the acceptability of the ECCS model used for the above assumptions.
The
. May 23, 1976.evaluation concluded that, with appropriate technical specification changes, Dresden Unit No. 2 met the performance requirements of 10 CFR 50.46 (Acceptance Criteria and Emergency Core Cooling Systems for Light Water Nuclear Power Reactors).
The LPCI logic network is designed to direct LPCI flow to the intact recirculation loop in the event of a loss of coolant accident (LOCA).
The logic network also was designed to close the suction and discharge valves of.the intact loop to prevent LPCI flow from bypassiryg the core and flowing out the break.
The staff review of Topic III-10.C indicates that since the LPCI loop selection logic has not been modified at Dresden
- Unit No. 2, the primary concern is not applicable as discussed above.
However, a different requirement does apply to Dresden Unit No. 2.
The staff has required that all BWR-3 1s perform a modification to ensure that the recirculation line suction valves remain open when LPCI is initiated on a LOCA signal.
Motor-operated valves are placed on BWR recirculation suction and discharge lines. Following a loss-of-coolant accident (LOCA), if either of these valves on the unbroken recirculation line closes and if the low pressure coolant injection system (LPCI) supplies ECCS water to that loop, then the LPCI water will flow through the jet pump nozzles into the lower plenum where it will contribute to core reflooding. If neither of the valves closes, the LPCI water could flow backwards through the unbroken loops' recirculation pump, around the downcomer, and out the break, thereby not contributing to core relfooding.
To provide redundancy, BWR/3 ECCS designs incorporated automatic closure of both the suction and discharge valves (on the unbroken loop only) upon receipt of a LOCA signal.
However, assumed single failure of the loop selection logic system can result in selection of the wrong loop as the broken loop. This would cause the following two events.
- 1)
All LPCI flow from both LPCI systems would be directed to the broken loop and would be lost out the break. This effect has been considered in BWR/3 ECCS-LOCA analyses; as a result, no credit is assumed for LPCI flow.
- 2)
Both the suction valve and the discharge valve on the broken recircu-lation line would close. If the break location were between those two valves, the break would be isolated from the reactor vessel. Although this could be advantageous under certain conditions, under other con-ditions it could introduce undesirable effects which have not been adequately considered in previously performed ECCS-LOCA calculations.
That is, for a certain range of break sizes, it is possible that core uncovery could occur with vessel pressure above the LPCI pump shutoff he~d. If break isolation were to occur at that time, LPCI flow could be delayed and/or reduced, resulting in a later core reflooding and a higher PCT.
With respect to Item 2 above, compensating effects exist that partially or wholly compensate for the above andesirable effects. The High Pressure Coolant Injection (HPCI) and the Automatic Depressurization System (ADS) would complete depressurization to the point where LPCI could function.
Although such LPCI operation would be delayed, credit can be assumed for the full complement of ECCS equipment since the required single failure has already been assumed (loop selection logic failure, selection of the wrong loop).
Preliminary calculations indicate that the above described compensating effects would sesult in PCT's for the worst size isolatable break that are below 2200 F. _However, a fully approved model meeting all require-.
ments of Appendix K to 10 CFR 50~46 does not exist which is capable of calculating a postulated break that becomes isolated. Also, the pre-liminary calculations were not performed for all sizes of BWR/3's.
Consequently, it.is not possible to categorically state that 10 CFR 50.46 requirements are met for all isolatable breaks for all BWR 3's.
Therefore, General Electric Company recommended, and we require, that the automatic closure feature on the suction valve be disabled. This makes break isolation a non-credible event which does not require analysis:
Two independent failures are necessary, i.e., closure of the discharge valve in the broken loop (requiring loop selection logic failure), and.
closure of the suction valve in the same loop (for example, by operator error).
No credit has been assumed for closure of the suction valve in any safety analyses other than ECCS-LOCA analyses.
For ECCS-LOCA analyses, closure of the suction valve provided a backup function for closure of the discharge valve on the unbroken loop.
With.
the recommended modification (suction valve closure disconnected), single failure to close of the discharge valve on the unbroken loop will now cause failure of the LPCI system.
However, this LPCI failure has already b~en taken into account by the ECCS-LOCA analyses on all BWR/3 plants.
No credit is assumed for LPCI operation on BWR/3 plants, since single failure of the loop selection logic can cause complete failure of LPCI.
Stated another way, the recommended change merely creates another potential path to a failure that is already accounted for in the ECCS-LOCA analyses, that is failure of LPCI; however, the recommended change precludes pos-sibility of an event which has not been accounted for in the analyses, i.e., break isola"tion.
By letter dated May 5, 1978, Commonwealth Edison informed the NRC that on April 21, 1978, the breakers for the recirculation system suction valves were racked out with the valves in the open position for their affected facilities (including Dresden Unit No. 2).
As stated above, elimination of the automatic closure feature on BWR/3 plants is a desirable change since it eliminates the potential for an event which invo)ves unreviewed safety concerns. The change does not create any new unreviewed safety concerns.
We, therefore, find acceptable Dresden 2 operation with disabled suction valve automatic closure following a LOCA signal.
On the basis of our review, we conclude that Topic III-10.C is acceptably resolved for Dresden Unit No. 2, and no further action is required.