ML19347C203
| ML19347C203 | |
| Person / Time | |
|---|---|
| Issue date: | 08/12/1980 |
| From: | Plesset M Advisory Committee on Reactor Safeguards |
| To: | Dircks W NRC OFFICE OF THE EXECUTIVE DIRECTOR FOR OPERATIONS (EDO) |
| Shared Package | |
| ML19347C199 | List: |
| References | |
| ACRS-GENERAL, NUDOCS 8010170024 | |
| Download: ML19347C203 (2) | |
Text
. _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
/~'s
(,,j'
> e
'g UNITED STATES E..,J/ '$
NUCLEAR REGULATORY COMMISSION P'
N./
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS
'o,s, w /
- wassincros. o. c. 20sss
...../
August 12, 1980 Mr. William J. Dircks, Acting Executive Director for Operations U.S. Nuclear Regulatory Commission Washington, DC 20555
SUBJECT:
CASCADING FAILURES IN NUCLEAR PLANTS
Dear Mr. Dircks:
The Committee has, in the recent past, had occasion to address several matters which may have had accident or failure " cascades" as a common element.
The Committee has been increasingly concerned that this aspect has not always had the early and vigerous attention that it warrants when it has arisen either : the course of a licensing review or as the result of analysis of operating events.
In general, a cascade may be visualized as a series of failures each occurring as a consequence of some previous event which gave rise to a set of conditions (environmental, electrical, mechanical, etc.) not origi-nally considered in the design.
As a rule, the later failures in such sequences tend not to be considered in the same depth, if at all, as those occurring earliest, ever though they may be causally related.
Examples are:
The potential environmental effects of operation of the PORV and block valves cperated in a bleed-feed mode, on equipment in containment, and in particular on the power and control circuitry associated with those same valves; The potential for BWR containment overpressurization due to failure of a safety / relief valve discharge line as a result of sustained ficw at resonance conditions through a stuck-open relief valve (the initial failure);
The consequences of an instrument line failure if this small break LOCA can also lead to a loss of control or safety function. Generally, instrument line failure analyses stop with evaluation of the resulting SBLOCA; 8 810170 o A
. i.
Mr. William J. Dircks August 12, 1980 The effects on containment electrical penetrations of " post-accident" conditions within containment which could cause overcurrent or short-circuit tonditions of non-IE equipment therein.
This should be con-siderad in light of the nonqualified nature of much of the equipment currently used to clear such faults; and The consequences of failure of an HPCI steam supply line, outboard of the isolation valves in that line, particularly in light of the probable effects of such a break on these valves and their controls.
In NUREG-0372, " Review of Licensee Event Reports (1976-1978)," the Com-mittee observed in a discussion of Systems Interaction,
" Redundancy and defense in depth are widely used in essential reactor systems to assure their availability.
Implicit in such usage is the assumption that a high degree of independer.ce exists between the redundant elements (or the various echelons of defense in depth).
Occasionally an LER disclosas an unintentional or previously unrecognized interdependence between such elements.... Because of the potentially serious impli-cations of such situations, more attention nreds to be directed to seeking them out."
The Committee believes that many considerations of failure consequences are often either unrealistically narrow or are not investigated at all if they appear to have obvious primary effects which are supported by a gen-eral analysis. The instrument line failure is a good example.
The Committee recognizes that this is a complex subject and that any deci-sion on an approach to requiring the consideration of cascades as the rule rather than the exception will require a great deal of study and interpre-tation.
We expect, therefore, to have our Subconnittee on Safety Philos-ophy, Technology and Criteria begin in the near future to schedule discus-sions for consideration of these matters.
Tne Committee would appreciate your identifying for us appropriate points of contact within the NRC Staff for this purpose.
Sincerely, Milton S. Plesset Chairman 1