ML19280B308
| ML19280B308 | |
| Person / Time | |
|---|---|
| Issue date: | 11/02/1981 |
| From: | Knight J Office of Nuclear Reactor Regulation |
| To: | Case E Office of Nuclear Reactor Regulation |
| Shared Package | |
| ML19280B307 | List: |
| References | |
| REF-10CFR9.7 SECY-81-631, NUDOCS 8112040054 | |
| Download: ML19280B308 (3) | |
Text
'
p ne,
u UNITED STATES a
NUCLEAR REGULATORY COMMISSION y4,er g
2.y i j
WASWNGTON, D. C. 20S55
$1 os a
K.% y NOV 2 1951 MEMORANDUM FOR:
Edson G. Case, Deputy Director Office of Nuclear Reactor Regulation FROM:
J. P. Knight, Assistant Director for Components and Structure Engineering Division of Engineering, NRR
SUBJECT:
CONTAINMENT INSTABILITY The CP/ML rule as presently drafted does not require consideration of instability (buckling) for either containment loadings due to 1) inadvertent inerting or 2) the 45 psi minimum internal pressure require-ment related to accidents releasing large amounts of hydrogen.
The design philosophy for each of these two off-nomal conditions is significantly different; because of this difference we believe that the exemption for instability consideration under the inadvertent inerting condition may limit the usefulness of the rule by presenting the opport.ity for technical challenges to future operation of plants choosing post accident inerting systems.
Inadvertent inerting is an event of relatively 'hbh likelihood during the lifetime of the plant.
ASME Code Service Liiiiit A stress criteria are therefore required in the rule to assure with high confidence that inadvertent inerting occurring at any time in the life of the plant, or several times for that matter, would not result in degradation of the containment structure. That is to say there would be no loss of the margins of natety that were present prior to the actuation of the inerting sy! tem. Under these conditions, no need for formal regulatory intervention after such an event should be expected and continued plant
- peration would be expected after minimal activity.
On the other hand, the 45 psi minimum pressure requirement in the rule is designed to set a " floor" for containment capability in the event of a serious accident that has a very low likelihood of occurrence. ASME Code Service Level C stress criteria are specified by the rule with the expectation that, although containment function will be preserved, there may be some damage after such an accident that would require repair before the original containment margins are restored.
Said another way, this is a one time event and should it occur, a hiatus would be expected during which inspection, ar.alyses, design and repair activities took place.
e112040054 811404 CF SUBJ g
Edson G. Case liOV 2 19 81 With the rule as currently written, in essence the Commission is reaching a judgment that explicit consideration of buckling need not be required in containment design for these two abnonnal occurrences as long as the specified stress limits are met.
Implicit in this judgment is a finding that faGures endangering the public health and safety due to buckling are very unlikely.
W h a finding, if based on a known quantity 1.e.,
a containment that has never been exposed to conditions other than to those for which it was completely designed in accordance with ASME rules, is reasonable since it is based on a knowledge of the margins that result from these design practices. On the other hand, if the containment has experienced several events that it was not completely designed for, such a judgment is more likely to be challenged.
Buckling is a cumulative phenomena. 'he rule requires that the con-tainment be capable of being tested at 1.15 times the calculated full inerting pressure. At least one cycle of unanalyzed loading is thus implicitly mandated.
Despite reasonable judgment that a large number of such cycles might be required before there was a safety concern, the basis for the implicit Commission finding begins to erode if we consider a containment that has gone through one or more inerting pressure cycles.
This would result in a containment that has undergone repeated loadings that were not completely considered in the design analysis.
Absent any specific analytical and/or experimental bases to the contrary, it seems logical that exposure one or more times to a loading that was not included in the design analyses would be enough to trigger legitimate requests for review. Once the issue is joined in various regulatory forums with a requirement for explicit demonstration of (1) lack of discernible damage and (2) margin to incipient damage, it would be difficult at best to rely only on judgment as to the probable lack of significant damage. At the least, it seems likely that the licensee and staff would be faced with post event inspection and analyses.
Such inspections and analyses are subject to numerous interpretations making further argument likely.
In addition, Commission-sponsored programs (NRR and RES) underway for several years are studying the ASME code rules and analytical methods presently available for use in design against buckling.
Preliminary results show that this subject is not without controversy and that conservatism may not be as great as expected in some aspects of design.
Based on these considerations, in my opinion prudent rule development would require that present ASME code requirements for buckling be met for all high likelihood events that might affect the containment, such as inadvertent inerting.
Based on recent staff discussions with the involved applicants, each of the current containment preliminary designs
M2 19 81 Edson G. Case affected by the CP/ML rule can accommodate this requirement.
Should the requirement be included in the rule, it would reduce the licensee's ability to optomize the containment design from an economic standpoint at the final design stage, with an estimated potential cost of 0.5 to 2.0 million dollars.
" /0 '/
-l
./( m.//1 r
J.
P. Knight,' Assistant Director for Components and Structures Engineering m
Division of Engineering, NRR