ML20056C054
| ML20056C054 | |
| Person / Time | |
|---|---|
| Issue date: | 08/24/1992 |
| From: | Siess C Advisory Committee on Reactor Safeguards |
| To: | Igne E Advisory Committee on Reactor Safeguards |
| References | |
| ACRS-CT-2039, NUDOCS 9209160291 | |
| Download: ML20056C054 (3) | |
Text
}
0ho1039 CHE' STER P.
SIESS 401 BURWASH AVE, APT. 132 SAVOY.
ILLINOIS 61874 TELEPHONE: (217) 352-0989 24 August 1992 TO:
E.
G.
Igne. Senior Staff Engineer, ACRS FROM:
C.
P.
Siess, Consultant, ACRS
SUBJECT:
REVIEW OF DRAFT COMMISSION PAPER, " DESIGN CERTIFICATION AND LICENSING POLICY ISSUES PERTAINING TO PASSIVE AND EVOLUTIONARY ADVANCED LIGHT WATER REACTOR DESIGNS' As requested in your memorandum dated 19 August 1992. I have reviewed the NRC Staff's discussion and proposed position for Issues B and C in Enclosure 1 to this Paper: " Policy Issues Analysis and Recommendations."
My comments follow:
B.
Analyses of External Events Beyond the Design Basis Let me note first that the only external events considered in this item are:
Seismic Fires Internal Floods Tornadoes and Extreme Winds.
There are many other external events considered in the design of a nuclear power plant, such as: hurricanes, storm surges, tsunamis, river flooding, local flooding, volcanism, transportation accidents, gas cloud explosions, etc.
All of these must be considered in design and are a part of the " design basis", but only the ones listed above have been singled out for consideration "beyond the design basis."
[l Why?
A partial answer is provided by the Staff in the third paragraph of Section B [I will refer to it as B3] in which they point out that " fire, internal floods, and seismic events can be important contributors to core damage."
They point out further that "the g
y estimates of core damage frequency for the fire and seismic events h
continue to include considerable uncertainty."
This suggests that U
margin studies are needed to evaluate the sensitivity to uncertainty.
y hh:
Even if we argue that the frequency of internal floods is subject to uncertainty, we account for only three of the four items listed h
above.
I don't believe that we have ever worried very much about the
.g margins provided against uncertainty in the design for tornadoes and R
extreme winds; we have included ample margins in the design, and have Q
E been comfortable with that, just as we have for hurricanes, tsunamis, A
y etc.
G 1
I think that the Staff might be asked to explain why their discussion h
and position goes beyond seismic and fires.
0p o
With this introduction, I will go on to comments on the Staff's iI O proposed position and discussions leading up to it.
For convenience and,.I hope, clarity, I will discuss these more or less separately L
for each type of event.
gggg C g
But first I must point out that the Staff's proposed position will require that the pRA required by 10 CFR 52.47 for a Standard Design Certification must include external as well as internal events. Which events is not stated but can be inferred from what follows.
Seismic Events The design certification application will include a " modified margins analysis to determine the vulnerability to seismic events" (84).
The modification envisaged is to use insights from an internal event pRA to support the margins analyses.
This procedure is elaborated on in paragraph B4, which is specific in some places and rather fuzzy in others.
The result would be a plant HCLpF (High Confidence of Low probability of Failure).
If this value is 'less than about twice the design ground motion zero period acceleration, the designer should perform a more detailed evaluation to find the vulnerability against which to strengthen protection."
This requirement seems excessive in view of all the conservatisms built in to the process, plus the fact that not many sites will have a ZpA of 0.3 g,
which the standard designs must meet.
Note, however, that the details in paragraph B4 are not included in the proposed position in paragraph B10.
I believe that what is proposed, as best I understand it, is a reasonable approach and one that can be done.
Certainly, the results from a pRA can be useful in applying the margins approach to a plant that has not yet been built.
And, of course, using the margins approach avoids the impossible task of evaluating the uncertainties in the seismic hazard.
As a minor item, I object to the Staff's use of parentheses to equate
" robustness" to ' adequacy" in the first line of B4.
These are not the same thing.
Adequacy is what we provide in the design.
Robustness is the ability to be adequate in spite of our mistakes or underestimates.
All of the above will be submitted and reviewed at the Design Certification stage.
At the COL stage, it is not clear what further review will be made of the seismic adequacy of the plant.
I am not sure that any is needed, but note that paragraphs B8 and B9 refer to walkdowns which presumably would consider seismic adequacy.
I believe that some perspective on the approach to seismic events in comparison with other external events is provided by understanding that the Staff does not consider a seismic event to be
" site-specific" in the sense that river flooding would be.
The reason for this is that the plant is designed for 0.3 g, and the site selection process will be used to insure that the seismic potential at the site is within the design basis envelope.
Fires As near as I can tell, fires will be treated in exactly the same manner as seismic events.
Nowhere in the proposed position or in thc discussions which precede it is there a reference to " fire" separate from " seismic".
I must admit t h '.t I don't know how the " margin methods" apply to fires; the more detailed discussion in paragraph B4 is limited to seismic.
I believe that the Staff should be asked to explain this to the committee, and also to explain whether the walkdown referred to at the COL stage will cover fires, and if so hnw.
. Internal Floods In paragraph B3, fire, internal floods, and seismic events are mentioned as being important contributors to core damage.
They mention the uncertainties in analyses for fire and seismic events but do not include internal floods in this category.
At the end of that paragraph, they say that "The designer could use traditional probabilistic techniques to study internal floods."
These will be evaluated by the Staff at the design certification stage (B5).
No mention is made of review of internal floods at the COL stage, since they are not site-dependent.
Whether a walkdown will include such events is not clear.
Further, as indicated earlier, if there are no large uncertainties in the design bases for internal floods, it is not clear why they are included in the "beyond the design basis" category.
Tornadoes and Extreme Winds These are site-specific events and the Staff proposes that they may be designed for by enveloping their effects using bounding analyses.
The wording in paragraphs B10 and B5 about "using bounding analyses to show that the hazard is insignificant" is rather strange.
In paragraph B5, they go on to say that " Bounding analyses of a site-specific event should either (1) demonstrate that the frequency of occurrence is sufficiently low that it would not significantly contribute to risk at the site or (2) demonstrate that the design would be robust even if the external event occurred."
I find the first criterion very unusual in that I don't know how any one could demonstrate that the probability of a tornado is so low that I don' t have to consider it in design.
This criterion would be alright for say seiche, for a plant not on a lake, but it makes no sense for tornadoes or extreme winds; we always design for them.
The second criterion is of course valid; it is what we have always done.
C.
Elimination of Operating Basis Earthauake from Seismic Design Although the Staff has not yet formally proposed a position on this issue, it is clear from the discussions presented in this item that they will propose eliminating the OBE from the design loading for structures systems and components.
This position has previously been considered in a general form by the ACRS and has been approved in principle.
The more extensive studies carried out by the Staff, and reported in this submittal confirm the acceptability of this proposal.
In addition, the Staff has done a thorough job of seeking out those instances where the OBE loadings have indirectly to control design features not necessarily or entirely related to the OBE itself.
These include consideration of fatigue loading of piping and effects of repeated loading at moderate seismic levels in the design of piping systems and in the seismic qualification of mechanical and electrical equipment.
For these cases, the Staff has proposed design criteria not involving the OBE but providing comparable margins.
Final criteria for these cases will not be available until further studies and some research are completed, but the interim criteria seem to be well thought out and quite acceptable.
I am satisfied with what the Staff has done, and can think of no additional issues that need to be raised.
-