ML20050C672
| ML20050C672 | |
| Person / Time | |
|---|---|
| Issue date: | 04/06/1982 |
| From: | Shewmon P Advisory Committee on Reactor Safeguards |
| To: | Palladino N NRC COMMISSION (OCM) |
| References | |
| ACRS-R-0971, ACRS-R-971, NUDOCS 8204090239 | |
| Download: ML20050C672 (3) | |
Text
-
/p Hog #'o
)
UNITED STATES hob Og7/
NUCLEAR REGULATORY COMMISSION
)
t 3,
r ADVlsORY COMMITTEE ON REACTOR SAFEGUARDS 0,.
WASHINGTON, D. C. 20555 j
%.a p
+...
+
April 6, 1982
't.
REcEtygg ',
Q 6; *APRg 198&c3
~
~
The Honorable Nunzio J. Palladino 4
Chainnan 9
U.S. Nuclear Regulatory Commission 8
y Washington, D.C.
20555
SUBJECT:
INSTRUMENTATION FOR MONITORING WATER LEVEL OR INVENTORY
Dear Dr. Palladino:
During its 264th meeting, April 1 and 2,1982, the Advisory Committee on Reactor Safeguards met with representatives of Babcock and Wilcox Company, Combustion Engineering, Inc.,
and Westinghouse Electric Corporation to discuss several proposed systems designed to indicate the approach to or the existence of inadequate core cooling (ICC).
The Committee also had the benefit of coments from the NRC Staff.
A Subcommittee meeting was held on March 31, 1982 to discuss the design features of these systems and their use in the management of reactor transients.
We are pleased to observe that the NRC Staff has developed an approach which will integrate the installation and use of ICC systems with that of other new systems which are being installed in response to other post-TMI-2 requirements.
We were told that the scheduling of installation and use of ICC monitoring systems is expected to be done on a plant-by-plant basis, and will take into account the commercial availability of these systems as well as the schedule for installation of other backfit items.
The NRC Staff has indicated that they believe that use of the ICC monitoring system should be introduced into operating and emergency procedures very carefully and only after appropriate operator training, including experience on simulators, if feasible. We support this approach. Both the use and the testing of these systems must take into account the probability they are likely to be most useful in emergency situations.
It is important that operators understand both the capabilities and the limitations of the sys-tems in order to use them with confidence when they are needed.
The NRC Staff has concluded that the proposed Westinghouse system and the proposed Combustion Engineering system are acceptable on a generic basis, subject to further exploration of a small number of unresolved issues.
The approach being taken by the Staff seems reasonable.
k 8204090239 G20406 L bl h PDR ACRS i
R-0971 PDR
Honorable Nunzio J. Palladino April 6, 1982 We agree with the following tentative conclusions of the NRC Staff:
1.
Core exit thermocouples and saturation margin monitors are not suffi-cient for an adequate ICC monitoring system for PWRs.
2.
Both the Westinghouse and Combustion Engineering vessel inventory monitoring systems correct identified deficiencies in present ICC monitoring instrumentation.
3.
A multi-step review process remains to be completed to assure careful phas'.q-in and full integration of inventory monitors.
We believe that the current approach of the NRC Staff to dealing with the ICC problem has sufficient merit that it should continue in the proposed di rection.
We plan to continue our review of this area as further develop-ments occur.
Additional comments by Members M. Bender and H. Lewis are presented below.
Sincerely,
\\.
P. Shewmon Chairman Additional Comments by ACRS Member M. Bender Concerning Reactor Vessel Level Indication System Although a great deal of valuable study has clarified the use and applica-tion of the inadequate core cooling monitoring system for PWRs, the feature intended to show reactor vessel coolant level has not been shown to have great operational value.
The proposed systems are not unambiguous in their response under all circumstances.
The Westinghouse RVLIS uses differential pressure to determine liquid level and measures differential pressures of 1 to 10 PSI against a background system pressure of 1500 to 2000 PSI.
It must correct for density and dynamic head.
The emergency operating procedures would need very thorough development to make RVLIS diagnostica11y useful.
It would have beer. of doubtful value in the Ginna event or the TMI-2 accident.
Honorable Nunzio J. Palladino April 6, 1982 The Combustion Engineering heated junction thermocouple system would be more effective under TMI-2 conditions and is less subject to ambiguity due to system operating conditions, but it, too, has some limitations.
The basic requirement is to provide guidance for operator action.
The urgent need indicated by both Ginna and TMI-2 circumstances is rapid primary system depressurization and reliable shutdown cooling.
I believe emphasis should be placed on being sure that such operator actions are unambiguously permissible regardless of liquid level indicating devices.
Additional Comments by ACRS Member H. Lewis Concerning " Water Level Indi-cators" I see no reason to repeat all the comments I have previously made on this subject.
In the interim, the Staff has commendably adopted a far more systematic and considered approach to this question, and that has miti-gated but not extinguished my concerns.
The remaining ones are:
1.
To change the name from " water level indicators," which they are not to " inventory monitors," which they are also not, does little good.
In the absence of dynamic effects, the Combustion Engineering system measures the mean void fraction in the upper plenum, no more and perhaps a bit less when dynamic effects are important.
The Westinghouse system measures differential pressure, and, in the absence of dynamic effects, this is more closely but not precisely related to pressure vessel inventory. That they each give some information is indisputable.
2.
Since the information they do provide depends upon many things such as pump status, flow problems and dynamic effects, etc., it is not clear to me that an operator dealing with an unfamiliar upset can know whether his upset is of such a nature that he can believe the instrument.
I do wish the Staff would decide whether it is better to know partial inven-tory (Westinghouse) or void appearance (Combustion Engineering).
This is scenario-dependent and I have not seen the issue clarified.