ML22230A157
| ML22230A157 | |
| Person / Time | |
|---|---|
| Issue date: | 05/17/1978 |
| From: | NRC/OCM |
| To: | |
| References | |
| Tran-M780517 | |
| Download: ML22230A157 (1) | |
Text
RETURN TO S
- """ "'"c,,o EGRET ARIAT RECORDS f
"1 Transcript of Proceedings
\\
l NUCLEAR REGULATORY COMMISSION
.,.1-,,
~o BRIEFING ON ANTICIPATED TRANSIENTS WITHOUT SCRAM (ATWS)
(open to Public Attendance)
May 17, 1978 Pages 1 -
38 Prepared by :
C. H. Brown Office of the Secretary
DISCLAii*\\i:R This is an unofficial transcript of a meeting of th2 United States Nuclear Regulatory Commission held on May 17, 1978 in the Commission 1s offices at 1717 H Street, N. W., Washington, D. C.
The meeting was open to public attendance and observation.
This transcript has not been revie1*1ed, corrected, or edited, and it may contain inaccuracies The transcript is intended solely for general informationa1 purposes~
As provided by 10 CFR 9.103, it is not part of the forma1 or informal record of decision of the matters discussed.
Expressions of opinion in this transcript do not necessarily reflect final determinations or beliefs.
No pleading or other paper may be filed with the Commission in any proceeding as the result of or addressed to any statement or arg'..1ment contained herein, except as the Com~ission may authorize.
1 2
- 3.
4 5
6*
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23
. 24 25 UNITED STATES Of AMERICA NUCLEAR REGULATORY CbMMIS.SION
.BRIEFING ON ANTICIPATED TRANSIENTS WITHOUT SCRAM(ATWS)
(Open to Public Attendance Corrimissioriers' Conference Room Room 1130 1717 H Sbreet, N.W.
Washington, D. C.
Wednesday, May 17, 1978 1
The Commission met pursuant to recess, at 11:00 a.m.,
Joseph Hendrie, Chairman, presiding.
PRESENT:
Chairman Hendrie Commissioner Gilinsky Commissioner Kennedy ALSO PRESENT:
J. Hoyle J. Kelley L. Gossick A. Kenneke
- s. Hanauer R. Mattson E. Case
- s. Levine
( ;.-
1
- 2.
3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
\\
P R O C E E D I N G S CHAIRMAN-'HENDRIE:
This is one of those,occasions of joy which come-s really too rarely to the Commission, 2
a briefing on Anticipated Transients Without SCRAM.
In this case, Mattson*,.-Hanauer,: I don't know whether to inciude you in the*.guilty partyi***L'ee, or not.
Why don't.you *sit bae;k a little bit MR. GOSSICK:
I intend to stay out of the iine of fire.
CHAIRMAN HENDRIE:
and I'll *ask these other fellows what they are up to and who is going to make the, presentation.
MR. GOSSICK:
Dr. Mattson will lead: *of,f followed by Steve.
MR. MATTSON:
We a~e here to give you a status report today on ATWS and ~ot to ask you for a decision.
We have sent some paper up to you,~including a NUREG report,. the number is 0460 and I will probably refer to it that way this morning, and a copy of some remarks that I gave to the public meeting on the 19th of April and to the ACRS Subcommittee on ATWS on the 20th of April.introducing the report and laying out a plan of action for bringing ATWS to a conclusion.
I'd like to summarize that briefly today, give you a flavor of where we are headed and where that will
l 2
. 3 4
5 6
7 i
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
\\
\\
\\
eventqally inv6lve you in the decision process, and give you a sumrn:ary overview. of some of the: techn.:j.cal things that are said in: the r,eport.
First, I,ought to highlight that the technical report, DSS,.Division of.Syste!Ils Safety, staff report.
- It is not yet a position. for.the Office of* Nuclear.Reactor
'I*. I',, !:** *.
Regulation.
- That report is a 'reevaluation 'of the. staff*'
position on ATWS.
It traces through* the history *of WQrk on this technical subjeet;:, commencing in the late 1960s* when the issue w~s identified.as a safety concern by the ACRS.
I In tracing that.history it marches by some rather important
~ilestones ~nd I will. list a couple:
One was the production in 1973, by.the Regulatory.
3 S'taff of the AEC of a document* known affictipnately as WASH-1270, which one can think of as the five+/-year-ago
.predecessor of this NUREG report that you have. in front of you today.
That>Jwas *followed in 1974 and 1975 by a series of reactor vendor rep9rts proposing analytical methods for iTWS and a seri~s bf *staff. status reports on those analytical methods.
Also, about the same time frame, the reactor safety study was published, *in 1975, *wASH-1400.
The.in 1976 there was a major contribution to the body of knowledge on ATWS by the Electric Power Research Institute, EPRI, surveying and
2
- 3.
,4 5
6 7
8 9
10 11 12
- 13 14
,15 16
- 17.
18 19 20 21 22 23 24 25 drawing conclusions from the* body of data on.the frequency of transients and, the. probability 'of failure* of SCRAM.
This reevaluation of the staff position on ATWS was directed by Ben Rusche, that is, ordered to commence by Ben Rusche the f6rme~ Direct6r of NRR about a year ago~
The
.tebhnical work *which formed the *original. basis for the report 4
was led by
- S.teve
- Hanauer drawing upon experts from -- throughout NRR and Research and a couple of consuitants.
Principatly, though, the experts were from the Division of Systems Safety.
Last*fall,.he turned it back to NRR, it was assigned to the Division of Systems Safety where we have managed the technical development, rev,iew and report drafting since.that
.time.
In our work over the last six months, *there'.have
- been major contributions from other sectors from the agency, I have mentioned three.
The Office of Research and the Probabilistic Assessment Branch has pulled a major ore in helping us in our application of probabilistic assessment
.techniques. *The Division of --.the Environmental Division of.NRR, Harold Denton's division, is in the main responsible for the value impact assessment which. we* think is one.of the more thorough going and complete of any yet done by the staff.
And the Division of Operating Reactors has supplied us with technical insight and considerations on the backfitting issue and will continue to do so in some major way in the course of
1 2
3 4
5 6
7, 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 5
the next several months.
The new report is intend~d to form the. technical basis for a final policy decision on ATWS.
We don't consider it yet to be the final technical word, for example,* changes are occurring t?day and will occur over the next several weeks in response to the process~s which is in motion to review the report, namely, the ACRS review and the Regulatory Requirements Review Committee review.
We would expect by the time the report reaches you for a decision, with a proposal for implementation, it will be sufficient for reaching a final decision on ATWS.
We believe that the technical issue has drug on about long enough.
I think the Chairman's informai remarks carry that same message.
We believe *that by and large the dialogu~
between government and industry is complete on this issu*e.
There is continuing discussion going on as we try to explain and make clear the words in our report so that people may more effectively comment on it.
The principal additions of this report to the state of knowledge on ATWS are several.
They include the use bf probabilistic.techniques, the contexting of the probability of:*severe consequences arising for ATWS, relative to the probability of severe consequences from other causes which are provided by the reactor safety study.
It is yet a another review of the available data on
1
- 2 3
4 5
6, 7
. *a 9
- 10.
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 these very I.wow probahili ty events and a* final contribution 6:15 va.lue :.impact 'assessment.
'
- In the media* we have been reading that it i's the industry.' s yiew that ;not much has changed in the years that have co~~* since the iast technical*report, with the qualifi~
cation'.:that these things* have 'been added, we* would agree 6
that the technicai' position in the *main has.remained.relatively stable.
That indicates.to us that it is time for a d~cision.
I woui~ like to burn to some slides.
Tom, are
~pu, badk there and' can give me Slide-No. l?
(Sl'ide)
Briefly the proces~ ~hat we have s~t in motion for reaching this final decision on ATWS *is described on*
that ~hart.
A public briefing has been held, the ACRS Subcommittee meeting has been held, the RRRC':s 'first meeting on the subject was yesterday, we hav~ another ACRS Subcommittee meeting on the 26th, we would expect several more Subcommittee meetings.before we reach the full ACRS for comments.in June or July.
We are here with the Commission breifing today, we would expect to get back to you-about August with.a Commission paper recommending a course of action.
One other miles tone we have to accomplish :bef cire *2 then is RRRC recommendations to the Director of NRR which he will take into consideration in formulating the Commission policy paper with his recommendation for action.
2 3
4 5
6 7
8 9
10 11 13 14 15 16 17 18 19 20 21 22 23 24 25 7
In looking ahead to.the likely content.of that
. recommendation,. it is my opinion at the moment and*. ~t remains r
unchanged.through,several weeks of public. discussion of this subject now, that rulemaking is the, most likely course of
- action td~finally resolve this i~sue.
lhere ~r~ pros a~d cons,
- both sides t9 that., I also believe.tha:t it is reasonable.* to expect tI:iat* such, rulem~king co.uld be a~c.omplished in the course of about two years, 'with i:qiplerrientation expected after that.
COMMISSIONER GILINSKY :,
Two-year rulemaking?
MR *. MATTSON:
We have a leg up, 'Commissioner.
It is
. a li:ttle shor.ter, than one would expect from ---
CHAI.RMAN HENDRIE:
Are you start led. at the shortness,
.the abruptness or.the lorigness qf it?
MR. MATTSON:
L.took it to be the fo'rriler, bu't perhaps it is the latter.
The'average is somewhat longer than two years, but we have in the rep0rt, the draft of a proposed rule.
COMMISSIONER GILINSKY:
The average for what?
rulemaking,?.
MR. MATTSON:
For rulemaking, from beginning to end.
COMMISSIONER GILINSKY:
For NRC rulemaking?
MR.. MATTSON:
- I don't believe Mr. Minogue is in the room, but I think his rule of thumb is in the order of three years.
1 2
3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 CHAIRMAN HENDRIE:
Does that include that petition that we finally dismissed that came in in.'68 or '70 or something like.that?
MR. MATTSON:
That would* include a number of things, including ACCS which I'm sure weights the data.
COMMISSIONER GILINSKY.:
You can get a CP license sooner than that.
(Laughter)
CHAIIDII..AN HENDRIE:
They are trying to fix that.
MR. MATTSON:
So I guess the principal thing that I would like the Commission to draw from that chart is that we will be back to you some time the end of the summer for more indepth and less status-oriented discussions.
That is, we will be looking for a decision.
I'd like, having dwelt on the procedural aspects
.of where we have been and where we think we are going, to turn quickly through an overview of what ATWS is and what we are saying about it in this report.
8' If we could move on to slide 2, I don't spend a lot of time here (Slide)
=- but just so we are all talking about*the same thing, ATWS is composed of two elements.
First an anticipated transient, events expected to occur one or more times during the life of a plant. Some examples are loss of feedwater,
1 2,
3, 4
5 6
7 8,
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 tur.bine trip, loss of off-site power and so forth.
The second.half of ATWS is the failure of*the shutdown system, the SCRAM system.
The failure,to function.
The.mystical common mode failure appears in that element of the ATWS formula.
In the next slide --
(Slide)
-- we show a.little bit more about t.he nature of these elements.
Why is there an ATWS pro?lem?
Well, we are concerned that the frequency of ATWS events, having severe consequences, by severe consequences we mean things ranging from public -- releases of radioactivity leading to
'~
doses in excess,6f~part~100 guidelines ~anging up to events of a more catastrorhic nature as described in the reactor safety study.
And the frequency of such events with severe consequences shown here s.implistically as the frequency of anticipated transients with potententiil for severe consequences, that is, not all transients, if there were no SCRAM would lead to the same consequences.
Times of probability fo SCRAM failure given the.antpcipated transient with the potential for severe consequences.
Now, the debate over the years which has raged on all three points.
First, what's an acceptable value for this 9
frequency of ATWS events with severe consequences? I'll spend
1 2
3 4
5
- 6 7
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 quite. a lot of time talking about that in.. a few minutes.
For shorthand, we say that's,our safety goal~.. How safe do*
- we -want* thepiant~*to be for the severe ATWS events?
Th*e s*ec.ond element, the mj,ddle cclumn,what
- value 0f the freq~~ncy of antiblpated trin~ienis with potential 10 fo~ severe conse_que~ces should be used. in this formu],a.. ' That Is an -*area, on-which' today, we feel there is 'much less controversy than).in the past.
The numbers used in the reactor safety
'study; the numbers used in the EPRI document, the numbers proposed. by the i*ndustry through* th~ *years, and the number proposed here are traceable to fairly good data at this point, given the operating experience with the number 'of plants in the* country. * 'Not a lot of controversy, over pumber.
What value the probability of failure of SCRAM should be used, relatiyely at least, much more qontroversy in that area.
And then, in the bottom line, what are we saying in this.report?
A safety goal of 10 to the 6th.
For a f~equency of ATWS events with severe consequences 10 to the minus 6 *~*.:.
COMMISSIONER GILINSKY:
Let me ask 0you about that.
Now, this is for the entire class of these events?
For each particular type of event?
I mean, it doesn't seem to me to be a very w.e.11 defined concept.
MR. MATTSON:
The types of events for which we are concerned are defined in the report.
1 2
_3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 COMMISSIONER GILINSKY:
And this is for_ the entire class?
MR. MATTSON:
For the entire class there'are some that are important.
For those that are important the pn:>babil.i, ty of yielding at ATWS with consequences.i,n excess of part 100, we propose should be shown to he less than 10 to the minus 6th for each reactor.
11 Now, the next slide will turn to it if I would just
-summarize the next two points on this one.* That is something I do want to go in to.
The value of 6 for the frequency of anticipated transients with potential for severe con5-equences comes from operating data.
And the probability of *scAAM failure comes from a variety of statistical and relia,bil.1.t:y assessments with several orders of'magnitude difference between the staff position and the historic industry position, principally owing to the weight given'by the staff to the potential for common mode failures.
But when you perform the multiplication up in the top line and compare the two numbers, you that.we-~.
have, and the number is more precisely stated in the report, at present an average probability of occurrance of an ATWS wit.h severe consequences for reactors operating or under design in the United States, of 2 times 10 to the minus 4th and a safety goal of 10 to the minus 6th.
And the propos~l is to bring the design capability of the facilities into conformanc
2 3
4, 5
6 7
8 9
10'.
11 12*
13 14 15 16 17 18 19
- 20.
21
- 22.
23 24 with the safety goal.
COMMISSIONER GILINSKY: Now, how does that relate to the numbers in t.he reactor. safety study?
MR. MATTSON: The reactor safety study looked at two specific reactors.
The Surry* Reactor. ~ndi,~~E:~~li-~~~§~_t_qm*--
Reacto~~. The Surry Reactor;*~*~r~~s~rize~ wate~ ieaot~r.
The reactor saf~ty* study and the st~£f in NUREG. 0460*.
12
- essentially, show the Surry reactor, a Westinghouse pressuriz~d water reac~or is capable of meeting the 10 to* t.he minus 6th goal t.oday.
I will get on a little bit later to saying wh~~
we think generically about tlle,various class(;;!S of facilities, but subject to confirming analysis and a little more detailed look at Surry, :~.it's 'fairly apparent to us at this_ poi~t that
'I the Surry reactor would meet the goal and. that -othei We~tinghouse reactors would.be similarly close.
COMMISSIONER GILINSKY: What about the other.* one?
MR. MATTSON:"*p~~~h-Bo*Etom_;reactor -- we say in 0460 that there is new information available irt some.of these areas since th~ reactor safety study.
The reactor safety
- study showed ATWS to be an important contributor to the overall risk of severe consequences portrayed by the. reactor safety study.
we*say.it is an even more _important contributor.
COMMISSIONER GILINSKY:
So does this \\_*-sw~p~}:hie:
.... --... -~-
-** -~-
1 2
3 4
5
.6 7
8 9
10 11 12 14 15 16 17 18 19 20 21 22 23 24 25 MB,.,MATTSON: Swamp is the wrong word; but we can bring down the overall risk -- the o.verall pr.obabili ty., o_f
- s*ev~r~ c~nsequenc~s.*. sig~ificantly by fixil)g. the 'ATWS problem on boiling water reactors,, including Peach Bottom.
. COMMTSSIONER GILINSKY.:,
- Well, let !Jle ask you, --
is your estimate. here different.than *the one in the. reactor safety study?
MR.. MATTSON:
Yes, slightly higher.
COMMISSIONER GILINSKY: Slightly higher?
MR.'MATTSON:
Yes.
MR. LEVINE:.. Let me say a word.
13 The problem is one of the fact that ther*e* are other ways of shutting down the reactor besides~:this* SCRAM'system which were* considered i:n: the reactor safety study and which are in fact considered in the ATWS report.
These -numbers that you see there represent an incomplete statem~nt in the total situation.
There is another fac~or of 10 6r 106 in probability reduction when you look at the. backup shutdown systems, which Roger, I'm sure will be getting.to in his discussion.
The numbers we had are about the same for this portion -- ATWs:*was defined as the anticipated transient and the failure of the normal SCRAM system.
Beyond tha.t there is an additional probability to get the reactor shut down which other additional things have to bear.
I 2
- 3.
4
,5 6
7
- s 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 14 COMMISSIONER GILINSKY: So why is there a problem?
MR. LEVINE:
Well, there is a question now of how close do you come -to the 10 of the minus 6, when you consider the other mechanisms.*
COMMISSIONER GILINSKY:.So you are saying just a little bit away or ---
MR. LEVINE:
The number 10 to the minus 4 is a factor, but 100 away from 10 of the minus 6 and the question COMMISSIONER GILINSKY:
Yes, but the question is MR. LEVINE:
-- how close do you_get there with these backup shutdown system.
MR. MATTSON:
The problem MR. HANAUER:
I guess we had better go into it.
The problem is more complicated.
In the first place, the Peach Bottom reactor has and had during the safety study evaluation some of the equipment which we are planning to require for all boiling water reactors.
The second point is that the Peach Bottom reactor has the CHAIRMAN HENDRIE:
Another way of saying that, Steve, is for the run of BWRs the ATWS situation is considerably.-
less cheerful than it is at Peach Bottom.
MR. HANAUER:
Well, there are two classes of BWRs.
Some have this additional equipment and some don't.
The Peach Bottom reactor was in the more favorable
1 2
3 4
..5
. 6 7
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 15 group.
COMMISSIONER GILINSKY:
So it was not a fortunate choice.
MR. MATTSON:
In this problem.
MR.. HANAUER:
Yes..
- The other point is *that whereas the reactor safety study made prpperly for safety studies an evaluation qf the chance that these additional systems would bring the plant to a satisfactory situ~tion, the look we have taken is in some respects more conservative ~han that in the safety study~ and we come out with the same -- I'm making this up as I go along we come out with the same functions ;--we:'.are. asking for more performance and reliability in these funcitions than was provided at P~ach Bott6m.
Therefore, we.would evaluate that Peach Bottaro needs faster equipment, to us, confirms and partly not confirms to us the safety status evaluation.
I didn't say that too well.
I have to warn you it's a fairly complicated subject which I will go into in whatever detail you ne~d.
COMMISSIONER GILINSKY:
Let me ask you this:
_suppose we haa not taken Peach *Bottom, but one of this.other class*
of reactors.
Would we then have come out with numbers which are significantly larger or the probability of these events/
which are significantly larger that we obtained on the reactor safety study?
1 2
3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 MR.. HANAUER.: 'Yes, sir.
COMMISSIONER GILINSKY:
By what, like a factor of 100 or something?
MR. HANAUER:
More like a factor of 5.or 10.
16 MR. LEVINE: By the way,, this information was given to the regulatory staff, even before. the study was done, that is, the fact that" not all plants had used this equipment and +/-hat made a significant difference in the results.
MR. HANAUER:
We have known this for a long time.
MR. MATTSON.:
Yes.
If we could go.to the next slide.
(Slide)
He we will talk a*little bit about how we get to the safety'objective that we have proposed.
The interest, of course,'is what's an acceptable objective for protecting the public from expbsure from radiation.
We hav~ proposed that COMMISSIONER GILINSKY:
Let me ask you again about this 10 to the minue 6th.
You may be getting to it, I see it at the bottom of your slides, but the::thing. that concerns me here is that it is not clear how many items you stuff into this package that it is suppose to be when you add it up, 10 to the minus 6th, there is a certain amount of arbitrariness?
MR. MATTSON:
Yes, that's what I was going to talk about.
l 2
3 4
6 7
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 17 COMMISSIONER GILINSKY:
You are going to talk about
'it?
MR. MATTSON:
Yes.'
COMMISSIONER GILINSKY: Very good..
MR. MATTSON: WASH-1270 set.a safety goal of 10 to the minus 7th.and contained a rationale that if the:ire. were l0*such causes of severe consequences -- potential causes of severe consequences of this nature, then a goal of 10 to the minus 7th for each of the 10 consequences would yield overall, a safety level of 10 to the minus 6th probability of severe consequences.
10 times 10 events at 10 to the minus 7th.
The reactor safety study came along after WASH-1270 and said that the core melt frequency, was somewhath,ighe:1; than ha~ been generally understood be;Eo:1;e, but desc~ibed a spectrum of consequencE!s which was much different than had generally been understood before.
The core melt frequency of 5 times.10 to the minus 5th was spoken to in the reactor safety study without catastrophic consequences of the sort that. have been held in mind prior to that time.
In light of that information, in using the same rationale that was in WASH 1270, the staff could not continue to support 10 to the minus 7th for one of 10 contributors, rather 10 to the minus 6th for one of 10 contributors for
1 2
3 4
5 6,
7 8
9 10*
11 12 13 14 15 16 17 18 19 20 21 22 23 24 25
- an overall probability of 5 times 10 to the minus 5th made much more sense.
Now,. the factor of 5 in there is an i:nteres~ing number
- als,o.
One might -argue that you could make the ATWS objec~i~e _5 times 10 to the minus 6th, but the 5 actually goes away when you take the ATWS cdntribution out of.the overall cor_e melt frequency and when you make some estimation for* improvements in safety for future reactors rela.tive to the estimation from the reactor safety study..
18 So the.number for core-melt frequency excluding ATWS is somewhere in the range one to 3 times 10 to.the minus 5th, we say in the report.
- COMMISSIONER GILINSKY:
So you are saying you will bring it down to the level of other* events?
MR. MATTSON:
Yes.
COMMISSIONER GILINSKY:
That's your objective?
MR. MATTSON:
Yes.
And ther.e are several other events of the order of 10 to the minus 6th or slightly higher which contribute in the main to that 1 to 3 times 10 to the minus 5th over~ll probability of core melt.
And this brings ATWS into line with. those, ~ut does not go as far as proposed in 1973 to go to 10 to the minus 7th.on the basis of the insights gained in the o~erall ris~
by the reactor safety study.
CHAIRMAN HENDRIE:
10 to the minus 7th, we bandied
1 2
3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 those numbers about and they didn't look too cheerfully in those days.
I !eally shouldn't have,picked that up, stuck in*the 1270.-- it was much too simplistic as*sort of a ~is~ussion ~nd it had and continues to have the
, ~~traordinary dis~dvantage that it is an absolutely undemonstr-
~ble and you are not a heck of a lot better off, but not
. notably'* so.
10.to the minus 6th is still something that you are *not going to demonstrate for plant experience.
MR..,LEVINE:
That factor 10 is a very important difference.,*.
CHAIRMAN HENDRIE:
But you are getting a lot close
- to the demonstrability of it..
COMMISSIONER GILINSKY:
Let me ask you, Roger.
Suppose* iomeone comes ilong and recalculates the reactor safety study and discovers that what you think is 10 to the minus 5 is really 10 to the minus 4.
You then suggest that the ATWS be knocked down in the order of magnitude?
MR. MATTSON:
The Commission would first have to gi~e me the same kind of implicit guidance that I had from the Commission today on the reactor safety study, that is, that it is generally sound and generally* describes an acceptable risk, because that is implicit in my use of the reactor safety study to change 10 to the minus 7th to 10 to the minus 6th.
COMMISSIONER GILINSKY: Where does that come from,
1 2
3 4
5 6
7 8
9 10' 11 12 13 14 15 16 17 18*
19 20 21 22 23 24 25 has 'the Commission ever said that?
MR. MATTSON: Well, the Atomic Energy, Commission said that in 1974 and the i~formation has never been or the guidance have never been overturned.
COMMISSIONER GILINSKY:
Well, T don't think there is guidance from this Commission on that.
20 MR. CASE:
Well, it is more implicit than explicit guidance.
The implicit guidance is that a well done study ---
. MR. HANAUER:
We didn't stop reactor licensing in view of that - -i~:f:qrmat,i.on,i,n -pa,rticular:.
COMMISSIONER GILINSKY: I am getting a curious set
- o:t: arguments.
MR. MATTSON:
Well~. it is also the corner stone that you must have, in order to do probabilistic licensing assessments at all-~-
COMMISSIONER _ GILINSKY:
I remembe,r testimony along the lines if they weren't safe we would have shut them down wouldn't.we have?
MR. HANAUER:
Yes.
COMMISSIONER GILINSKY:
I don't think that's a good argument.
I mean, you cari consider certain levels of safety to be a6ceptible or tolerable for a period of time, but_you can have quite a few goals.
You know, we just got through a session on safeguards in which we are tolerating a situation
l' 2
3 5
6 7
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 we would not in the long.'run consider respectihle.
- I'm not* saying there is ~nything. wr*ong. ne~essarily, or *that that should riot' be the guida.nce.
MR.* HANAUER:
There is no. notion *.that:... it is the last word either.. The~~ are-~ it is, first of.alli the 21 best information. available.
And the sec~nd place, it impelle.d*
neither the staff,rior the Commission to conclude that there was *anything seriously wrong with reactor safety.
So in the spirit of using the best available information, we have ha.sea. our.goal to some considerable
.extent, on the results.of the study..
If the study came out very different, if the next study or some recalculation came
.out very different, we would have to rethink this.
.MR. MATTSON:
It is.important to ~ecognize that the approach to developing licensing cr+/-teria and the requirements that we are proposing to put on the design of reactors do not flow in any direct quantitative fa~hion from the reactor safety study number or from the 10 to the minus 6 safety goal*.
What we have done is examine the general safety goals stated in probabilistic quantative terms in WASH-1270, said it too low, it needn '.t be that low and** move.it up in order of magnitude.
But*then we are still stuck, as I will describe later, with being unable to state in licensing
.criteria probabilistic methods, totally p~obab:i,listic methods for deciding how a system should be configured, what components
1 2
3 4
5 q
7 8
9 10
- 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 22 it sho~ld contain, what safety stress limits it might be designed for for pressure:*cqntaining components, what operator reliability can be relied upon,** et cetera, et cetera.
So we have fallen back to a det.erministic licensing approach which has in it an element of conservatism.
When you move from the probabilistic to* *the determin.istic, no matter how much you say you are doing it with realism as we say in this report, there is an element of conservatism which provides a hedge against changes in that. 5 times 10 of the 5th number.
The additional hedge we have looked at is in our review.of the reactor safety study,.through the years of its availability and in the context of ATWS,. we would conclude that the number will likely _move down and now up; that 5 ti.mes 10 of the minus 5th is high.
COMMISSIONER GILINSKY:
Well, I think that if indeed you are using or guiding yourselves by some implicit Commission approval, I am wondering about a document and I think that is something the Commission ought to take up.
MR. MATTSON:
I think that that's a very important point you raise, and I personally would *recommend that perhaps the time has come for that.
Any probabilistic assessment in reactor safety hinges upon the implicit understanding of the general acceptabil ty of the reactor safety study results.
That's the only benchmark
I
'\\,
I 1
2
.3 4
5 6
7
_,8 9
10 11 12 13 __
- 14 15 16 17 18 19 20 21 22 23 24 25 23 MR. HANAUER:
This is one of*our first forais
- hito a*n explicit *use of* numbers iri establishing safety goals and ther.efore,* it is riot_..,,,.surprising that new considerations of
.this sort_ s~ould come up._.
COMMI$SIONER G~LINSKY: *. Well, I think the Commission
. ought 'to gbr~-, guidance on:Utt.
- MR. MA'i'Y.rSON:., That is what we -wiB.. be looking for the Commission mo.st for, when we cornea back to you for a decision, is how safe is safe enough for ATWS?
Is it required to b~ fixed~ot~i~n~t~it required to be fixed?
If it is, to what deg1;ee?
And that's the question_ --.that',s the name of*
this game.
That'1 s* what 'this is all about.
MR. :HANAUER:*,That will be an issue during the*
rulemaking als6.
You don't have tp decide that in any final way in putting out a proposed rule for consideration.
MR. MATTSON:
Let*me briefly summarize the kinds of solutions to tl].e ATWS issue that have been proposed arid discussed in the past.
They are fairly straightforward.
You_ e_i ther decrease the frequency of anticipated transients which is impracticable to demonstrate maintenance of that throughout the' life_ of the plant and probably over::-,.:::.*:::'.~:*
constrains the operation of the plant for other reasons to the point that it is just not thecptacbical way to proceed.
Decreasing the probability of SCRAM failure,
1 2
3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 24 intellectually, intuitively and an acceptable way to proceed, but.you have to demonstrate such a low probability.of SCRAM failure in order to achieve the safety objective~- and.this is what the Chairman referred to earlier -- it becomes a number that is beyond the realm of technology to underwrite,.
numbers in the range of 10 to the minus 5th,. 10 to the minus 6th reliability.
The third.option is to decrease the severity of ATWS event~.
Said another way, provide mitigating ~ystems which assure *.that the probabiii ty of* severe co~sequences is sufficiently low.* That's our preferred approach.
It has*
been the preferred approach of the staff for some years~
The fourth option, some combination of the three, for example, if some cost effective improvement in rods and*
drives could be proposed in combination with some.lesser mitigating system than would accrue if you d.idn't decrease the probability of SCRAM failure, there might be fruit in that area.
No one has proposed it, so*we haven't examined it in detail, it doesn't seem cost effective.
CHAIRMAN HENDRIE:
The last one has the difficulty of involving the probability matters in *the regulatory position in a very direct way?
MR. MATTSON:
Yes, sir.
CHAIRMAN HENDRIE:
For all that you look at the reactor safety study here in the course of the staff's
1 2
3 4
5 6
.7 8
9 10 11 12 13 14 15 1,6 17 18 19 20 21 22 23 24 25 25 rec6nsideration for the 99th t!me of this problem.
You eventually come down to the foll,o'wing judgments:
(a)*
something ought to be done about ATWS, it is the staff's judgment, I take it, that it ain't good enough as it stands.
MR. MATTSON:
Yes, sir.
CHAIRMAN HENDRIE:
(b) you conclude that the best way to fix it is.to implement consequence limiting measures rather than trying, as we started out in '73 to improve SCRAM systems or equivalent.
Now, as in othei places where the staff has said well, here's an accident sequence or type of accident sequence that we think is a problem and ought to be fixed.
You now propose to class ATWS as a design basis accident.
MR. MATTSON:
Yes, sir..
CHAIRMAN HENDRIE:-- which means that the plant designs ought to be such that they can deal with that class of accident~ with consequences within the Part 100 ~uidelines.
In providing consequence limiting measures there is rio consideration of how probable the design basis accident is.
MR. :t>'.IATTSON: In the proposed licensing system, that we have, yes.
CHAIRMAN HENDRIE:
As in the classic approach, you just say it will be declared as a design basis accident. We know it is low probability, never mind, but nevertheless, we have declared it a design basis accident and that means you have
- 1.
1 2
3 4
5 6
7 8
9 10' 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 got to* have something in the plant *that takes care of it.
So that the approach here comes back finally, for all' the talk while we consider the place of.how this fits with the reactor safety study and those considerations of the staff, really does come back in the classical modes and in effect or dir~ctly saying,*we believe ATWS.should be a design basis accident.
If it is a design ba;is accident, why we don't look at the combined probability consequence spectrum and pick some acceptable combination of probability and consequence for the fix; we say design basis accident fix it.
MR. MATTSON:
Yes, sir.
With one exception --:-,
CHAIRMAN HEND~IE:
Well, No. 4 up here would come away from.that and say, no, what we would do wouid in fact be to do: a probability consequence analysis and then fix some acceptable place on that and say ~t has to be at least that good.
But that really does pull you into the risk analysis business.
MR. HANAUER:
There is something, however, that I would suggest needs to be said.
Chairman.
.Every word you said is exactly* right, Mr. Chairman.
CHAIRMAN HENDRIE:
That's the advantage of being (Laughter)
MR. HANAUER:
But I would point out what we already 26
1 2
.3 4
5
.'6 7
8 9
10, 11 12, 13 14 15 16 17 18 19 20 21 22 23 24 25 27
- know, and that ip that the selection of the design basis
. accidents, a'nd the. sel~ction of, -- i°:n: det.aiL.,:,w.nich: 1.asstimptdcons
'one shouid \\make and whi,ch values and parameters one uses in evaluating them are choices from a spectrum, and that these choices h~ve
- been made in the past. based on unq:uantifiecL engineering j{idgment and which *probabilities :and risks were somehow involved.
They were one of the ingredients even though we really didn't know how to do it in a quantitative
- _w~y.
The difference is that this time we have explicitly taken some numbers and sh.own, in a much less than optimum way how we have.used some quantitative
~isk and probability ideas in choosing the accident, the assumptions to be made in the evaluation, a~d in at least 6ne case, the ~alrtes of the parameters to be used in the evaluation.
And that's the difference.
MR. MATTSON:
The one exception to that was the same exception I had to your statement.
'(Slide):_.**
If I *could turn to the next slide, we wiil talk about briefly, the proposed licensing requirements. _ Quickly, they treat*radiological concerns, primary system integrity, fuel integrity -- you can read them as well as I can.
The exception that I want to talk about is the mitigat ing system desigri: and the draft of the rule as it currently
1 2
', 3 4
5
'6 7
8 9
10 11 12 13 14 1-5 16 17 18 19
, 20 21 22 23 24 25 28 exists in.the report proposes an, option on the reliability or availability of that'mitigating system~
It says you can proceed one of two ways,, Mr. Designer.*
The first way is the old way, you'.can make it safety grade and do all of the things to gold-plate that system that go
~long :with the definit;i.on of the word "safety grade".
That includes IEEE-279, that includes seismib qualification, environmental qualification, all of those things required pur~uant to our. regulation for safety grade equipment.
Or, you can show a reliability -- an unreliability for that system of 10 to the minus 3rd or better.
And that 10 to the minus 3rd system is good enough for. a mitigating
- ystem.
Now, that makes our licensing requirements a bit of a mix of the deterministic and of the probabilistic and it
- is fair to say at this point that mix is causing some controversy.
It caused some with the ACRS, causing some with the Ratchet Committee and ---
CHAIRMAN HENDRIE:
Well, let me tell you as a member in good standing of the American Society of Mechanical Engineers as well as other learner societies, I just make it Category I safety system and my deep intuition tells me I would save an awful lot of design in work and analysis just going down.the classic route.
MR. MATTSON:
That may very well be true.
1 2
3 4,
5, 6
- 7 8
9
. 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 29
. CHATIIRMAN HENDRIE:. 10 to the minus 3 or better for --
that's pretty good reliabili-i:y ~*
I wonder; if my car starts 9 9 9 timeS"
- out ::of every l_, 000 or.':what else t~at I use in the ordinary, life works 999 times.6ut of 1~000 or better.
MR. MATTSON:
We.,think 10 to the minus 3rd'..is achievable and can be demonstrated.
There is some controversy on that stibjept ---
.,MR. CASE:
,Well, in fact our judgment says safety grade does that, but as you point out, it is so.metimes difficult to prove that.
CHAIRMAN HENDRIE:
Yes.
It is going*to be shaky to get there.
MR. MATTSON:
As we go through the next slide ~hich briefly characterizes what we would be looking for in mitigating systems CHAIRMAN HENDRIE:
Before you go away from that, please, since we are having a briefing and don't have to face proposed rules, regulations or whatnot -- if in its wisdom the staff judges that a rule::,.and an appropriate guidance is the proper set of actions for the Commission to take, may this time we please h~ve the rule contain the essential elements, acceptance criteria in summary for~ and that all of the rest of that good* stuff be in regulatory guides
_: ihitead of regulations?
1 2,
3 4
- 5.
.6 7
8' '
9 10 11 12
- 13 14 15 16 17 18 19 20 21 22 23 24 25 MR.** MATTSON:
That is exactly the proposal*
The primary acc.eptance cri t.eria *. would be 'in the regulation and t,h'e. evaluation model~. would be in s*upporting regulatory guid~s~
CHAIRMAN HENDRIE~
I know he.has been dea'iing with Appendix K for a long time and MR~ MATTSON:
We have-learned the same iessoI)..
Okay, if I could go back.to the next slide..
(Slide) 30 The mitigating system we are talking about,* again,.
this just summarizes what I have already said. The classic safety grade requirements -- one exception there:
the expected natural phenomena we would be willing to speak to the OBE instead of the SSE in terms seismic qualification for this mitigating system.
Of course, able to withstp.nd the ATWS-environmental.
conditions inside a*containment, diverse, automatic, relative to othe;- sys terns.
Some credit for operator. action:
there is.
- a. magic number in the report aimed at 10 minutes, L *believe, with some support in ANSI standards work and a general movement in our safety work toward that 10-minute* criterion.
And then the bottom two line:s, either safety grade or reliability analysis with a demonstrated unreliability of 10 to the minus 3~d with 50 percent confidenc~.
CHAIRMAN HENDRIE: Well, at least you didn't hang up in the 95 percent confidence level.
There is some solace in
1 2
3
- 4 5
6 7
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 31 that.,
MR. MATTSON:
That's an important f~ature of what.
we are trying.. to do in the evaluation models is to say we want to use realistic evaluation models for a. design basis accident.
We are having some difficulty with people.believing _it. :--' :.,
esr~ci~lly sine~ there are some notabl~ exceptions t6 that
~ith moderator temperature *coefficient and use of service level C limits for the pressure vessel.
But, in the main, the initial conditions and the other parameters are treated realistically and we agree with that,. and the industry evaluation models that we have been reviewing over the last years.
If we could turn to the last slide.
(Slide)
At the bottom line, what do we think this safety goal and these requirements will lead to in design of plants?
As I said earlier, with Westinghouse, some plants we think will.make it and other plants will need circuitry changes, combustion engineering and,:Babcock ~
and Wilcox, the other two PWR manufacturers, the same thing plus some relief valve additions.
In the case of the boiling water r~actors, those two lines are a little bit difficult to understand, let me try it. It is recirc-pump trip for the boilers, plus a high capacity boron system, got to pause there a moment ---
1 2
. 3 4
,6 7
8 9
- 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 COMMISSIONER GILINSKY:
That's a new system.
MR. MATTSON: Yes.
General El~ctric has been in with thi staff informa,lly and has ora_lly pr~posed an. automat:Lc boron syst?m o:f: some higher capacity than as previously been discussed for ATWS con~iderc;1_tions.
We continue to put th~ lin~ there at the bottom 32 of the reliability, the high pressure make-up*system, the high pressure coolant injection system and t_he reacto:t. core isolation cooling system on current BWR'designs.
It is likely that if the capacity is su_fficient and the:*quality.of the high capacity boron system is sufficient,* that the. concern with the reliability of present high-press~re make-µp systems go away, that with the safety grade, high capacity automa,tic boron system and with the existing high pressure make-up capability, and with recirc-pump trip that the boiling water reactors can be found to meet the safety goal.
, That doesn't say I c~n do it tomorrow, there is a
- fair amount of technical work., evaluation model, things to be done, but it is our estimation of whe+e we would come out in the future, but that would be the case.
I think General Electric shares that estimation.
We need, at this juncture, in the development of the position and consideration of these things*by ACRS and the RRRC to hear from General Electric on that subject.
I am
1 2
3 4
5 6
,7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 informed that we will in the course* of* its presentation to the ACRS wh_ich is scheduled for like, * *Jul:y:.
The ACRS Subcommittee', 1.should have mentioned earlier, is going to provide two days. for an oppo,rtuni ty for industry 'to comment. the NUREG report., One day will be 33
,spent with two of the reactor vendors and the At?m.ic Industrial Forum, the other two vendors plus EPRT will make' presentations and respond to que*stions from the Subcommittee.
That completes my presentation,.if you. have further questions?
MR. KENNEKE:
May I ask one question.
As your thinking is now would you haye the safety goals numerically defined in the rules, or would that appear
- MR. MATTSON: It would not appear in the* rule at all.
It would be in the technical ba~is and be used in the, value impact assessment and would be used much the same way as* I have used it.today, to context what we are trying to achieve, but it would not appear in the regulation~.
It wohld be deterministic criteria matching the <;reneral framework of those regulations to that.
CHAIRMAN HENDRIE:
To the extent that ~e are trimming here on some sort of an overall safety map, we are trimming up a corner that has long been in controversy.
In effect, the ATWS approach of the staff has always been in the context, that
1 2
3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 as you build more plants, why it is inevitably true you get increased reasons to trim some of these things out.
34 What sort of a'gradation of requirements do you see reasonable and possible from very old plants to construction permits?
MR. MATTSON:
Th~re are some alternative~ available.
The a~tiactiveness of the p~os* and cons, if you will, of ~.those various alternatives,. I believe will be one of the primary subjects for consideration by the Ratchet Committee and by the ACRS.
The report does not go to that subject to any large degree.
When we start from where we are at today, I can tell*
you what I see we will be considering.
CHAIRMAN HENDRIE:
Well, rather than discuss it at length, let me I think it would be useful for the staff to consider that sort of approach,--
MR. MATTSON:
We are doing that.
CHAIRMAN HENDRIE:
-- to the extent that older machines that had been operating*for a long time don't have fully upgraded ATWS provisions, they indeed represent a small local risk element, I guess, but in the overall scheme of things it doesn't loom very large and against the costs.
What you'.have is typically, ver often, is a safety yield per dollar which rises as you come into latter plants, and what it means for some of the earlier plants is, yes, you have got some safety, but boy, it's very expensive to implement it.
And
1 4
5 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 I think* it is a fair proposi~ion.
C0MMI$SIONER GILINSKY:
Well, you ~pply that
' principle every time you pao~fit:.,\\:*~oo.*
'.I MR. MATTSON:
Yes.
CH.A;I:RMAN HENDRIE:
Yes, 'and what we are,talking a:bout here is. again, -- 'i~ fact, 'if people* don't or.der some plants, why the whole r~gulation here will be backfitted.
(Laughter)
MR. MATTSON: One way to come at this gradation and the alternatj.ives for gradation of requirements, gradation versus age of plant is to look at the contribution to the cumulative probability of core melt, that is, th_e total core melt frequency* from all causes for various levels of ATWS fix, if you will.
Just to be simple, if I said 10 to the minus 6th for all future plants and 10 to the minu* 5th for all past 35 plants, I can sum the contribution to the core melt frequency, the total core melt frequency.from let's say the hundred older plants versus the contribution at another safety.objective for a hundred new *plants and show the difference for no backfit versus backfit to va~ying levels~**
Again, that's the kind of thing that depends fairly strongly on using the reactor safety* study, and the numbers which can be derived on core. melt frequency from the reactor safety study.
That is using them in a relative risk mode,
r 1
2 3
4 5
6 7
8 9
10 11 12 13 14 15 16 17 18 19 20 21
- 22 23 24 25 that is, you are comparing increments and deriving changes in the overall risk from the incremental changes in the pieces* rather than saying that the reactor saf~ty study is an absolute measure of risk.
36 CHAIRMAN HENDRIE:
I. think in that context there is much less*question ~bout it.
Buppose it weren't labeled WASH-1400 to reactor safety study, but were a set of publications and ba6k-up
,file reports, and MIT reported nuclear technology, a journal arti~le or whatever, staff wouldn. 1_t hesitate to.use the -pr,ipcip es laid down there as well as some of the relative numbers.
MR. *MATTSON:
There is one other principle that is being brought to the backfit question, and that is the principl which underlies the systematic evaluation program.* The basic outline there is to look at the plant generally, at a number.
of areas, and then make a.decision overall on the safety of the plant, not piecemeal on 175 issues.
That same concept can be brought to _bear for the ATWS question for the older facilities, it is in the context of overall safety*and it is an important contributor.
CHAIRMAN HENDRIE:
I recommend that approach for for the older plants, because it would make little sense.
On a particular older plant, to fix ATWS and thereby make a 10th of a percent difference in the overall risk situation, that is, there may be a~.iirimher of the older plants in which
1 2
3 4
5 6
7 8
9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 37 ATWS doesn't represent the same proportion of the core melt risk as.it ~oes in the Surry, Peach Bottom tases as calcuiated in WASH-1400.
And ther~ it ~ak~s progressively less sense to fix ATWS on a sort of a blind-across-the-board basis, or*
another way of putting* it is just as I did before, that in those cases. your safety yield per dollar of in.vestiment is going way down.
COMMISSIONER GILINSKY:
There are several measures, you know, there is overall safety for all of the reactors, and then there is that particular reactor.
MR. CASE:
Yes, we are talking in that context.
COMMISSIONER GILINSKY:
And you have to look at several ways.
I think you also have to be cauctious about using this safety study. I want to repeat that point, assuming kind of an implicit Commission support for certain numbers on the basis of not having shut down reactors.
You could also argue that none of these reactors have the ATWS fix and we have not shut that down.
So why institute it now.
I think it is just an overcritical way of looking at it.
MR. CASE:
You must appreciate that the biggest criticism that industry made against our ATWS position of three or four years ago was WASH-1400 came out and you haven't taken this account into account in your deliberations.
We were forced to do this because -- and it was a fair question
l 2
- 3.
4*
- S 7
8 9
10 11 12, 13 14 15 16 17 18 19 20 21 22 23 24 25 that they ~aised.
. COMMISSIONER GILIN.SKY:
Well, you obviously want to take into account work that has been done,: but I, think you want to be careful about imputing to the Commission some guidance concerning overallr~sk levels.
I.think that is something the Commission really hasn't addressed.and something I think it should address.
CHAIRMAN HENDRIE:
Okay.
Well, thank goodness this was a briefing and not a decision.
I can simply say thank you very much.
38 (Whereupon, the meeting was concluded at 11:55 a.m.)
J