ML20205R352
| ML20205R352 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 03/25/1987 |
| From: | NRC COMMISSION (OCM) |
| To: | |
| Shared Package | |
| ML20205R346 | List: |
| References | |
| NUDOCS 8704060323 | |
| Download: ML20205R352 (107) | |
Text
,
\\
~
UN11ED STATES
] NUCLEAR REGULATORY COMMISSION IN THE MATTER OF:
DOCKET NO: 50-443 MEETING WITH NRC AND MASSACHUSETTS OFFICE OF ATTORNEY GENERAL RE:
BROOKHAVEN REPORT (SEABROOK)
LOCATION:
BETHESDA, MARYLAND PAGES:
1 - 105 DATE:
WEDNESDAY, MARCH 25, 1987 ACE-FEDERAL REPORTERS, INC.
8704060323 870331 PDR ADOCK 05000443 Official Reporters PDR 444 North CapitolStreet Washington, D.C. 20001 (202) 347-3700 NATIONWIDE COVERACE
p
- CR30313.0 1
COX/cjg 1
UNITED STATES OF AMERICA
~'
2 NUCLEAR REGULATORY COMMISSION MEETING WITH NRC AND 3
MASSACHUSETTS OFFICE OF ATTORNEY GENERAL 4
RE:
BROOKHAVEN REPORT (SEABROOK)
DOCKET NO. 50-443-6 7
Nuclear Regulatory Commission Room P-110 8
7920 Norfolk Avenue Bethesda,-Maryland 9
Wednesday, March 25, 1987 The meeting convened at 8:05 a.m.,
V. Nerses presiding.
11 PRESENT:
12 R.
SWEENEY NHY-Bethesda Office V. NERSES NRC/NRR/ PAD-5 13 G. THOMPSON Institute for Resource and g,
Security Studies (Mass. A-G) 14 TIM M.
LEE NRC/RES/AEB RALPH MEYER NRC/RES/AEB 15 TREVOR PRATT BNL CHARLES HOFMAYER BNL RICHARD J.
BARRETT NRC/NRR/RRAB 16 FRANK J. CONGEL NRC/NRR 17 l T. M. NOVAK NRC/PLW-A STEVE LONG NRC/PLW-A CHARLES E.
ROSSI NRC/PLW-A 18 SCOTT NEWBERRY NRC/PLW-A FALK KANTOR NRC/IE/EPB 19 DAVID J.
PEEPLES New Hampshire Yankee ALFRED TORRI PLG 20 WILLIAM J.
DALEY, JR.
New Hampshire Yankee EDWIN J.
REIS NRC/OGC-21 G.
BAGCHI NRC/NRR/PWR-A/EB SUNIL WEERAKKODY NVS Corporation DONALD J.
PERROTTI NRC/IE/EPB 22 T.
C. FEIGENBAUM NWY 23 l J. R.
KYTE BMC, Inc.
I R. A.
HAMPE N.H.
C.-D.
A.
ELAINE I. CHAN NRC/OGC 24 JAMES T.
HAN NRC/RES WARREN C.
LYON NRC/NRR/DPLA/ FOB
(_,
25 J. A. MURPHY NRC/RES/DRSS K. N.
FLEMING Pickard,_Lowe & Garrick, Inc.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 4 646
30313.0
-cox 2
y 1
PROCEEDINGS 2
MR. NERSES:
Welcome, gentlemen.
This meeting is 3
for the Massachusetts Office of the Attorney General to meet 4
with the NUREG and Brookhaven people to discuss the 5
Brookhaven National Laboratory final report.
We established 6
this meeting, and we have all introduced ourselves.
I would 7
like to turn around and provide the floor for comments before 8
we begin.
Tom, would you want~to start off in saying a f'ew 9
words.
10 MR. NOVAK:
Well, yes.
Let me just make a few 11 comments.
This meeting is for the purpose, as I understand 12 it, for the State of Massachusetts and its consultants to
(-
13 provide comments to us regarding the Brookhaven report'on the 14 emergency planning sensitivity study and so forth submitted 15 by the Public Service of New Hampshire regarding the Seabrook 16 docket.
We have tried to assemble people that we feel would 17 be most responsive to your comments, can understand them, may 18 ask some clarification remarks.
We don't have an agenda for 19 the meeting.
That would have been helpful, but as we go 20 through it, we will just try to provide comments as best we 21 can.
Dr. Pratt and Dr. Hofmayer were the principal 22 investigators, I guess, for the Brookhaven work.
They will 23 be speaking for Brookhaven.
24 For anything regarding staff review, we will try 25 to answer those questions as best we can.
I think, as-I l
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage -
800-336-6646
30313.0 3
ccx V
1 understand the purpose of the meeting, and I would like to j
2 get a little clarification of that at this time, it's 3
primarily for you people to provide comments based on your 4
review of the -- first, I think the preliminary draft report, 5
which we sent you some time back', a~nd now the final report 6
from Brookhaven.
I guess I would first like to understand if 7
that is reasonably correct', or, if not, why don't we try to 8
best understand what we would want to accomplish by this 9
morning's meeting.
10 MR. THOMPSON:
I think we have two purposes.
One 11 is to give some comments and the second is to ask some 12 questions.
I guess the third is to respond to questions.
\\,
13 MR. NOVAK:
Fine.
14 MR. THOMPSON:
Maybe I should turn to my 15 comments.
Steve and I would like to make just a few minutes 16 of introductory comments and then just go to a back and forth 17 of questioning.
Is that okay?
18 MR. NOVAK:
Okay.
19 MR. THOMPSON:
Firstly, it seems to us that the a
20 application to reduce the EPZ at Seabrook is premature in 21 that it doesn't take account of the NUREG-1150 process which 22 is proceeding in parallel.
We believe that NUREG-1150 should 23 go through the review process and be issued in final form 24 prior to any major regulatory action at any individual 25 plant.
It seems to us wasteful to go through the identical
(-
ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8%33N
l 30313.0 4
cox ss 1
-- essentially identical process for one plant, when there is i
2 a generic activity going on.
3 Related to this, it has seemed to us, throughout I
4 the NUREG's investigation of this application, that there has 5
been very limited reference to the work reported in 6
That's been rectified somewhat in the latest i,
7 version of Brookhaven's review, which we have, but not 8
entirely.
It's also been done belatedly.
t 9
Which brings me to the question of outstanding 10 issues and outstanding uncertainties as represented in 1150.
11 As you might imagine, myself and Steve and the 12 Commonwealth of Massachusetts may wish to make comments in b-13 the future on 1150.
I don't want to prejudge those.
14 However, the report has identified a large number of-15 uncertain issues which bear upon EPZ reduction, issues such 16 as thermal-introduced steam generated tube-rupture and direct 17 containment heating.
These have a very clear. bearing on the 18 issue of EPZ determination, and it's quite clear from 1150 19 that there is a great deal of residual uncertainty.
20 So in that sense also, the EPZ reduction 21 application is premature.
22 Finally, it seems to us that the licensee's 23 material doesn't even satisfy the requirements that they have 24 set out for ther.selves.
They set out to demonstrate, by I
25 comparison with NUREG-0396, that an EPZ reduction was l
l l
/\\CE FEDERAL REPORTERS, INC.
)
202-347-3700 Nationwide Coverage 800-336-6646
' k,
\\
. ~ -
~ ~ ~ ~ ~ ~
._3.03lh t -
c,x 5
m-I reasonable.
2 However, that document points out that the 3
objective, or it stipulates an objective for emergency 4
response as achieving dose savings where dose savings are 5
above PAGs.
This has not been properly accounted for by the 6
licensee in this material or by the Brookhaven review.
There 7
seems to be an excessive focus on the 200-rem days, the 8
life-threatening rem days.
Even on the initial materials 9
submitted by the licensee, it's very questionable that they 10 have made their case by comparison with PAGs.
11 These are general comments.
We have a lot more 12 detailed technical points.
Steve would like to make some of
(-
13 those.
14 MR. SHOLLY:
I am sure you know, both of us 15 submitted affidavits when this issue came up before the 16 licensing board.
I don't want to try to belabor the points 17 we made, but just to refresh memories a bit, I think the 18 issue of possible accidents, releases reellting from sabotage 19 and terrorism needs to be addressed.
Obviously not a topic 20 that one wants to do a great deal with in an open public 21 forum, but by the same token, one can cannot ignore it 22 either.
23 The Event V interfacing loca analysis looked 24 through what Brookhaven did in their most recent report.
It 25 seems to me to be headed in the right direction generally, L
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800436646
I f-30313.0 Cox 6
\\s I
although we have a question about what happens to the l
2 hydrogen that gets released into the RHR vault area in the 3
auxiliary building.
I think it would be worthwhile to take a 4
look at what happens to that hydrogen, whether the auxiliary i
5 building maintains its structural integrity or not.
6 The -- with the focus on induced steam generator 7
tube rupture, there seems to have been forgotten that there 8
are accident sequences initiated by steam-generated tube 9
rupture.
Those sesk to have dropped below the' surface of the
't 10 sea of concerns here.
I would like to try to pop it back up 11 to the top somewhere so it gets addressed.
12 Along the lines of induced tube rupture, I think
\\.-
13 two areas of concern, one is what fraction of those sequences l
14 where induced tube rupture might be of concern is l
15 depressurization going to be possible.
In other words, there 16 must be some fraction of that class of accident sequences 17 where the pourers will not have power, won't be able to be i
18 opened.
I think that ought to be quantified.
19 secondly, the issue of depressurization, it would 20 be worthwhile to see some thermohydraulics calculations, see 21 just how far down in pressure-you can go.
22 Some of the Sandia direct heating experiments did 23 indicate that a fair amount of direct heating and melt 24 ejection can go on with pressures as low as 2-to 400 PSI 25 range.
That would require a rather substantial
/\\CE FEDERAL REPORTERS, lNC.
202 347-3700 Nationwide Coverage 800-336-6646
30313.0 cox 7
s-1 depressurization capability to get you down to that point.
I 2
think that ought to be looked at a little more.
1 3
Seismic sequences seem to have disappeared.
We 4
would do well to remember that seismic sequences accounted 5
for about 13 percent or so of the core melt trequency, and 6
about half of that was from seismic events in the.7 to 1.0 G 7
range.
That is important not only for accidents initiated at 8
power but also for shutdown accidents; external events during 9
shutdown seems to have been missed.
That would likely be a 10 substantial problem in the event of a very large seismic 11 event during shutdown.
12 Also in seismic, we raised the issue of whether (s
13 aftershocks occurring while the containment is pressurized by 14 an ongoing accident might reduce the abilities of containment 15 and might contribute possibly to an early failure.
Operator 16 response needs to.tne looked at, been concern expressed before 17 that operator performance under a high seismic load 18 conditions might be degrading due to stress, and also the 19 issue of relay chatter probably ought to be looked at in this 20 case.
21 Deinerting burns is something to ine looked at; a 22 fair fraction of core melt is from blackout, power is not 23 going to be lost forever.
Power will be recovered at various 24 times while an accident is in progress.
It might be possible 25 to do some bounding cases and see whether or not loads are L
ace FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage
- 800-336-6M6
30313.0 cox 8
L l
1 generated that could fall after containment; recovery of 2
sprays, steam explosions, it would be worthwhile to look at 3
the effect of heating of the upper head and bolts to see 4
whether or not a substantial amount of the energy that is now 5
being depicted as required to fail the vessel head goes away 6
and then is available to act as kinetic energy impacting on 7
the containment.
8 Shutdown accidents I have already mentioned, 9
seismic.
I think fires also ought to be looked at.
Gordon 10 mentioned the NUREG-1150 work and the various documents that 11 back that up.
12 I think it's important, if similar work is not l
(-
13 going to be done, either at Brookhaven or in the staff, that 14 these Zion results ought to be looked at for some 15 perspective, if nothing else.
16 Therefore you have a best estimate range of 1 to 4 17 percent somewhere for early containment failure.
That is to 18 be contrasted with the rather lower values put forward by PLG 19 and PSNH.
20 Finally, I think Gordon mentioned briefly the 21 focus on the 200-rem dose distance curves.
I think one ought 22 not to forget that holding down population dose, in terms of 23 person rem, was also a goal in the emergency planning r.ules 24 established.
And the thyroid doses were also a consideration 25 there.
So the dose distance relationships for thyroid dose V
ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30313.0 cox 9
ss 1
and the impact of possible emergency response option on 2
overall population dose ought to be looked at.
3 MR. THOMPSON:
I would just like to add a couple 4
of points to that.
Firstly, the Brookhaven review seems to 5
ne to really skate over the containment failure issue.
It
~
6 doesn't talk about the model test that's going to be done 7
this year.
The important' issue is whether an axiometric 8
finite analysis is adequate or whether there can be 9
departures from symmetry that are significant.
Another issue 10 is inhalation dose, which NUREG-1150 comments on at some 11 length, and the related issue is plume rise.
Apparently the 12 max curve incorporates criteria leading to less plume rise
(.-
13 than it gives us to assume.
14 MR. NERSES:
Okay.
15 MR. SHOLLY:
I might add one thing further.
One 16 thing we are interested in accomplishing this morning is 17 finding a little bit more out about what the staff is doing.
18 We have Brookhaven's report.
We also have Warren Lyon's memo 19 on induced tube ruptures and a few other menos on containment 20 leakage and stuff.
I an interested in finding out what the i
21 staff is doing in addition to those efforts.
22 MR. THOMPSON:
At the basic level, we are curious i
23 as to what the course of this proceeding is likely to be.
I i
24 don't know whether anyone can give us any guidance on that.
25 MR. NERSES:
Help me out on that.
L l
l i
l ACE. FEDERAL REPORTERS, INC.
I 202-347-3700 Nationwide Coverage 800-336-6646
30313.0 cox 10 1
MR. ROSSI:
Do you mean the review of the 2
petition?
3 MR. THOMPSON:
Right.
4 MR. NOVAK:
We have counsel here.
Perhaps they 5
will comment later.
~
6 Is that kind of a synopsis of general areas?
7 MR. SHOLLY:
Yes.
8 MR. NOVAK:
I guess what we are trying to do, or 9
what I would hope we would do, through this meeting, is to 10 clearly have a good understanding.
If there are any 11 particular questions about the Brookhaven effort,- I would 12 hope that we could have some discussions to understand that.
k/
13 It would seem to me'that you have kind of gone through it 14 all.
I think I scratched down some notes.
j 15 Probably what we have to do, and I would wonder if 16 it wouldn't be useful, is to try to go through it kind of by 17 subject.
That is, pick a specific concern, discuss it, and i
18 then say, okay, if both sides or all parties understand what 19 work has been done and the questions that are still 20 remaining, perhaps we could proceed on to the next subject.
21 That might be the best way to do it.
22 Otherwise, we are just going to continue to i
23 discuss it.
But we_ won't know if we have gotten where we 24 want to try to get to during the hours that we have available 25 this morning.
\\s.
ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
\\
30313.0 cox 11 A.-
1 Let me ask, I guess, our consultants from 2
Brookhaven.
How would you like to try to go through this.
3 Would it be useful to kind of take it one step at a time and 4
work through it by phenomena, by subject, by review area?
I 5
would assume that might be an useful way to go.
6 MR. THOMPSON:
Before we begin, I have one 7
question, which is the 1150 process.
8 As I understand it, the expectation is to get a 9
final version out around January or February of next year.
10 Is that the case?
11 MR. NOVAK:
We can ask our friends from research 12 to comment on that point.
There is an " official," quote,
(./
13 standing.
We might ask Joe Murphy from research today.
Why 14 don't you answer that question right now, Joe.
15 MR. MURPHY:
The current schedule is the beginning 16 of January.
17 MR. NOVAK:
Let's make sure we understand exactly 18 what you mean by that.
The NUREG-1150 report is out?
, 19 MR. MURPHY:
The NUREG-1150 report is out.
There 20 is a draft for comment.
The comment period expires, I 21
.believe the 25th of August, 21st of August.
I don't remember 22 exactly where.
Our schedule now assumes we will have a final 23 report that takes into account the public comments the 31st 24 of January in '88.
Our ability _to meet that schedule 25 obviously depends on what the nature and the character of the L
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-334 6646
l-
/
~ _.. _.
~
p f
~ ~ ~~ q 30313.0
~
[
cox 12 ks 1
public comment is.
That we won't know until the middle of 2
August.
]
3 MR. NOVAK:
Let me ask a clarifying question, 4
Joe.
There was some discussion of a peer review to be 5
sponsored by this Agency.
,j 6
Do you have any idea of the time, will it be in t-7 concert?
8 MR. MURPHY:
We are hoping to have that peer j
9 review in concert with the public comments.
I 10 MR. NOVAK As opposed to having the peer review I
11 focusing after all public comments have been received?
12 MR. MURPHY:
Correct.
Hope to do it in parallel.
\\-
13 MR. LONG:
Can I ask what the significance is of i
14 the final report on 1150?
i 15 MR. MURPHY:
Pardon?
j 16 MR. LONG:
What is the significance of having the 17 final 1150 report available?
I 18 MR. ROSSI:
Won't it r.eflect the public comments, 19 I think?
20 MR. LONG:
As far as 'M9 use by the' Agency or 21 whatever, what is the staten O '; i* as a final report.- It's a l
22 research report at that point?
23 MR. MURPHY:
It's a research report.
l-l 24 MR. THOMPSON:
May I ask about the peer review I
l 25 process, what that will consist of?
L ace-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-M,46
'r
~
-30313.0 cox 13 V
1 MR. MURPHY:
There are three elements of the peer 2
review process.
One is the public comment period.
We have 3
so far mailed out over 1000 copies to various and sundry 4
people for comments.
The second part of it deals with a 5
committee that is part of the uncertainty work that has been 6
going on at the research program at~Brookhaven.
A committee 7
has been formed to look at the uncertainty characterization 8
and methodology used in NUREG-1150.
9 We will be having a seminar at the Crown Plaza 10 Hotel in Rockville on the 21st and 22nd of April.
This is 11 not part of the peer review, but rather it's a seminar to 12 explain the methodology that was used in the report to help Nos 13 people get through it the first time.
14 In parallel at that meeting, the same time period, 15 the Brookhaven Uncertainty Committee will be meeting at the i
16 same hotel to conduct their first meeting and review of the 17 chronology.
18 In addition to that, we haven't weighed all the 19 plans yet, but we hope for a broader peer review.that will be 20 initiated soon that would consist of a committee of 10 to 20 21 people drawn broadly from the nuclear industry, the national 22 lab, public interest groups, that would conduct a broad peer 23 review, again with the goal of having that committee report 24 its work in the same time frame as the end of the public 25 comment period, which is the middle to the end of August.
L ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30313.0 cox 14 1-MR. THOMPSON:
Now, there was, in January, a 2
meeting of expert review groups, which covered some of the 3
same ground.
4 MR. MURPHY:
The meeting in January, Ralph perhaps 5
can address this better than I, covered four general topic 6
areas of phenomology that affect the uncertainty 7
characterizations that are in 1150.
This fifth meeting is 8
part of that, if you will, the Brookhaven meeting I referred 9
to on uncertainties, is part of that effort, in that they 10 will not be revisiting the work done in the four earlier 11 conmittees, but relying on it, and looking more at the 12 general uncertainty methodology.
The report of those
\\.-
13 committees will be available to the peer review committee, 14 the larger committee that I mentioned.
15 MR. MEYER:
I would like to comment on the 16 February review groups, because there is an important 17 distinction in what their purpose was and what your interest 18 might be.
19 The four groups that met in January met 20 specifically for the purpose of commenting on the adequacy of 21 the research plans for pursuing those areas of uncertainty 22 that had been identified in NUREG-9506.
Those subjects were 23 thoroughly aired in that document and its review process a 24 year ago.
So the follow-up was not to revisit technical 25 conclusions, but to look.at the adequacy of NUREG's research A-ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-33MM6
J 30313.0
'cox 15 ss 1
plans.
J
'2 MR. ROSSI:' I wonder if it would be worthwhile to 3
say something about what we intend to do in terms of our
'4 review from here on in.
5 MR. NOVAK:
We can.
I think that would sum up.
I 6
Then we can go'back to some of the technical areas. ~Why 7
don't we do that.
8 MR. ROSSI:
Okay.
Let~me just say a few words 9
about the relation of the Seabrook application-to i
10 NUREG-1150.
I see NUREG-1150 as a research product that is i
11 addressing a number of issues in a generic way.
The 12 application on Seabrook is a plant-specific application.
For b-13 a plant-specific application, you can look at design details 14 on that plant, and you can look at the phenomena and how they 3
15 might occur on that particular plant.
You can see whether 16 they are or are not problems in terms of the' unknowns and 17 uncertainties for the particular issues of concern.
18 There is no question that direct heating and i
19 thermal-induced steam generator tube rupture will have to be 20 addressed on Seabrook on a. plant-specific basis.
)
21 The uncertainties will have to-be.n order to=make i
I 22 a decision on the petition, whichever way the decision comes i
23 out, they will have to be as well defined as possible for 24 Seabrook.
The other issues that you mentioned, all of those 25 will, indeed, have to be looked at on the Seabrook
\\_
/\\CE FEDERAL REPORTERS, INC.
202-347-3700 2 Nationwide Coveiage.
800-334 6646 i
30313.0 cox 16 L
-1
. plant-specific basis.
2 Now, please appreciate that what has been done up 3
till this point in time is a preliminary look at the 4
application by Brookhaven, and they looked at some particular 5
key areas that we asked them to look at.
They have issued.
i 6
their report, and they have identified >various problems that 7
need to be looked at in more detail.- Direct heating, as I 8
identified, steam-generated tube rupture, as identified.
The 9
interfacing locus, it's probably been addressed more perhaps, 10 there's more known about those at this point in time.
They 11 have been addressed in the Brookhaven report and they will be 12 addressed in what comes in the future.
Accidents at shutdown
\\-
13 have been identified in the Brookhaven report as needing more I
14 work.
i 15 Now, where we go from here, the way we -- we 16 haven't defined our total review program at this point in l
t 17 time.
What we are doing now is working on defining a 18 thorough review program or a thorough review in the area that 19 a thorough review is appropriate of the application.
We 20 would, anticipate that that will involve contractors again, 21 such as Brookhaven, since Brookhaven has done a lot of work 22 on this, they will probably be involved.
23 All the contractors that will bo involved have 24 not, as yet, been identified.
25 But between~now and November, we open doing
/\\CEJFEDERAL REPORTERS, lNC.
202-347-3700
' Nationwide Coverage 800-336-6646
30313.0 cox 17 I
s-1 additional reviews in the areas of uncertainty, as we see
{
things where th'ey are right now as being preliminary, 2
that we 3
have identified areas that need further work, and we see more 4
interactions with the applicant, questions going to the 5
applicant, maybe additional walkdowns, whatever is necessary, 6
to do the review that is required in order to make a final 7
decision on the petition.
8 MR. THOMPSON:
You mentioned a November date.
9 That's your planned completion on that date?
10 MR. ROSSI:
Yes.
November is our planned 11 completion date at this time, that's correct, isn't it?
12 MR. HERSES:
That's what we have identified in the
\\s 13 affidavit to the board.
I also should add, Ernie, that a lot 14 of the details of this will probably be discussed this 15 afternoon.
I suspect that you folks will be available this 16 afternoon to meet the applicant?
17 MR. THOMPSON:
I will not.
i 18 MR. SHOLLY:
I can be here.
19 MR. NERSES:
It would be well for you to be there 20 so you can have an opportunity to listen to the details of 21 some of those things.
22 MR. ROSSI:
Yes.
The phenomena that are important 23 to this petition'that have been identified in Seabrook 1150 24 will clearly have to be addressed on Seabrook.
25 But addressing it on a plant-specific basis for L
ACE-FEDERAL REPonTens, INC.
202-347-3700 Nationwide Coverage 800 336-6646
m 30313.0~
cox 18 i
As 1
this petition involves some things that perhaps are not 2
available for a more generic one.
3 One of those things is that.you can look very hard 4
at the details of that plant.
You can look very hard where 5
there are uncertainties or even problems that are clearly 6
identified,'you'can-look very hard at that plant on-the 7
changes that can be made to the plant design, changes _that 8
can be made to procedures.
There are a number of things that i
9 can be done on Seabrook that would not be appropriate at this 10 point in time for a generic review.
Clearly this applicant 11 is interested in coming _out with a favorable outcome on the 12 petition, so presumably they are going to work very hard with N-13 the staff and our consultants to do analyses, procedure i
14 improvements, whatever else it turns out that it takes in i
15 order to tighten down on the uncertainties with their goal of 16 moving towards a favorable conclusion on the petition.
17 From the NUREG's viewpoint, we are going to do an 18 objective review, we will interact with the licensee, we will 19 interact with people like you to identify where the problems 20 are, and however it comes out will be based on our objective 21 review.
We don't know whether it will come out favorable or 22 unfavorable.
23 At this point in time, we would certainly not be 24 in a position to hypothesize on which way it will come out.
25 Scott, do you have anything more to say about the L
t
[
/\\CE. FEDERAL REPORTERS, lNC.
202-347 3700 Nationwide Coverage 800-336-(M6 t
~, ~,, -.. _. -,
30313.0 19 cox-1 review process?
You are more involved in the planning of 2
that than I am.
3 MR. NEWBERRY:
No, I think Ernie covered them 4
pretty well.
There clearly are classes'of things that we 5
think need to be looked at further.
Those things that were 6
identified in the Brookhaven review, and some things -- some 7
that you suggest we think we need to understand better to be 8
able to make a decision on the petition.
That's what we are 9
going to be talking about this afternoon and in the next few 10 months.
11 MR. THOMPSON:
When you discuss changes in plant 12 design and operating procedures, will that discussion be 13 confined to suggestions from the licensee or are you going to 14 raise some possibilities?
15 MR. ROSSI:
That would not be confined to just 16 suggestions from the licensee.
It would include anything 17 that the staff or our consultants could identify as being 18 areas that would be necessary in order to either eliminate a 19 problem, if it's clearly a problem, or in some cases to 20 reduce the uncertainties, j
21 MR. THOMPSON:
So it would presumably include 22 depressurization?
23 MR. ROSSI:
It would include depressurization as a 24 possibility.
25 Now, again, you know, staff's function is to
%s
/\\CEJFEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800.UMM6
30313.0 ccx 20 (s
1 identify problems and identify criteria that we think would 2
be appropriate in addressing that, but we would generally 3
provide flexibility of approach to the applicant.
I mean, 4
there obviously are alternatives, we think there will be 5
alternatives in a lot of these areas.
There may be choices 6
between design changes and procedures, and we would maintain 7
flexibility in those areas as long as whatever was finally 8
done was satisfactory to us.
9 MR. NERSES:
Okay.
Thank you, Ernie.
Tom, at 10 this point, do we want to go back and focus on the 11 Brookhaven?
We certainly would appreciate having their 12 comments on that subject.
(s 13 MR. NOVAK I think, in listening, I think it 14 would be good if we would go back to Steve and Gordon to let 15 you pick out a particular subject, and let's bring it back 16 up.
Then we would ask Brookhaven and any of the staff 17 personnel to comment.
18 I think if we run back through the issues as we 19 reflect on them, we will probably miss a few and not cover 20 the points in enough detail.
I think what we have heard is a 21 good summary thus far.
22 Now I would like to kind of pick a particular 23 area, based on your comments, based on your review of the 24 Brookhaven report, and see if, in fact, we agree with your 25 conclusions as a fair reading of what the material in the L
ace. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage RM)-33MM6 y
y
30313.0 21 cox L
1 report would suggest or so forth.
I think-that would be most 2
helpful from my point of view.
i 4
4 3
MR. THOMPSON:
Would it be appropriate to' start i
i 4
with direct heating?
i 5
MR. NOVAK Your choice, sir.
That would be 6
- fine, 7
I guess what I will like to do is why don't you i
a 8
give your characterization of whatever Brookhaven suggests in 9
the report, whatever shortcomings, whatever material they did 10 discuss in the report dealing with direct heating or any 11 subject.
Then I would turn to Trevor or Charlie as 12 appropriate and say, is that a fair. reading of the report, 13 and if so, good, then let's get that on the record and any 14 supportive comments, I think we would turn it around and make 15 it more intelligent to people who will read the transcript.
16 MR. SHOLLY:
I think I will preface this by saying 17 I got Brookhaven's report Friday and have obviously not been 18 up 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> a day reading it in detail.
I have been through i
19 it and I think I have a fair understanding of where things 20 are headed, but it's entirely possible we will misconstrue 1
21 something or misunderstand something at this point, because 4
j 22 we have only had the report a few' days.
1 23 In the area of tirect heating, I think what 24 Brookhaven has done'is identify the issue and contrast what 25 we have from the applicant in terms of plant-specific l
/\\CE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30313.0 cox 22 1
analysis, and contrast that with the NUREG-1150 work and say 2
there is a big difference here, the uncertainties, how much 3
material is going to be ejected, how much of that will be 4
involved in direct heating, will have a great deal of bearing 5
on what the loads will look.like at Seabrook.
But Brookhaven
'6 hasn't done a Seabrook analysis of the issue, and that 7
remains a ripe topic for future work.
8 MR. ROSSI:
I think we would all agree with that.
9 Brookhaven is shaking their head.
I will clearly state that 10 in the area of direct heating and I believe steam generator 11 tube rupture also, the problem has been identified.
It has 1
12 not been resolved.
Is that a fair statement?
i l
As 13 MR. PRATT:
Yes, it is.
14 MR. THOMPSON:
You will be applying the state of 15 the art method to Seabrook analyses?
16 MR. PRATT:
Again, that hasn't been determined by 17 the NUREG staff as to what the appropriate follow-up would be 18 and to do it.
Nothing has been established.
19 MR. THOMPSON:
Would that include a detailed 20 examination of the reactor cavity and the obstructions in the 21 pathways?
22 MR. PRATT:
Yes.
Those are a limit that goes into 23 one's judgment, the ability of a particular cavity 24 configuration to eject core debris into the containment 25 atmosphere and direct heating.
i L
/\\CE FEDERAL REPORTERS, INC.
202 347-37(X)
Nationwide Coverage MW)-336-%86 l
~
30313.0 23 cox I
4 5
.1 MR. THOMPSON: -Might it be possible that the 9
i 2
ongoing test at Sandia might be modified.in. order to account 3
for some Seabrook specific ~ configuration?
i I
4 MR. PRATT:
That.I --
i 5
MR. ROSSI:
At this point in time, again, as we go 6
through our review, things may change, because we will do our 7
review taking into account everything that we learn as we go j
8 along.
9 At this particular point in time we don't have an 10 intent -- and Tom, you can correct me-if you think I am wrong 11
-- of tailoring the NUREG.research program in a specific 12 direction for the application.
(s 13 However, we would want to reserve the opportunity 4
i 14 to do that, if it appeared possible, feasible and necessary 1
15 to do it.
16 At this point in time, we haven't made any plans 17 to do that.
18 Again, our direct heating will have to be 19 addressed.
One possibility is to prove that it can occur or 20 that it -- that you narrow down the uncertainties on_Seabrook 4
21 or the phenomena.
Another possibility is you use procedures 22 to depressurize and do it in such a way that the phenomena 23 does not need to be addressed.
That's another approach.
24 Either of those approaches could be successful and are i
1 25 available to the applicant.
(-
ACE-FEDERAL REPORTERS, INC.
i 202 347-37(X)
Nationwide Coverage 8(Xh33Mi646 L....
1 30313.0 24 cox v
1 They can pick either one, as far as I am 2
concerned.
3 MR. THOMPSON:
On the time frames, the review is 4
to be completed in November, but the direct heating research 5
program will extend to a longer time, so that at some point 6
in your review you are going to have to reach closure on the 7
new research information you accept.
8 MR. ROSSI:
We would have to reach closure on 9
Seabrook specifically; that would not necessarily mean that 10 the generic problem would have to be resolved.
11 Again, let me go back to the statement that there 12 are a couple of things.
Seabrook could make a design change
\\-
13 or identify one to eliminate the phenomena, or to demonstrate 14 to the satisfaction of experts on direct heating that the 15 phenomena, because of the design that already exists at 16 Seabrook, won't occur on Seabrook.
That's a possibility.
17 The other thing that they can do, again, is they 18 can reduce pressure, put in procedures, maybe design changes 19 to do it, whatever is necessary to make the case so that that 20 will eliminate it.
They have those options available to them 21 and neither of those options, in my opinion, require 22 resolution of the problem in a generic sense.
23 MR. THOMPSON:
All right.
What is the current 24 schedule for the --
25 MR. ROSSI:
Do you have something to add to that?
%a ace FEDERAL REPORTERS, INC.
202-347-370)
Nationwide Coverage MG)-3364M6
30313.0 cox.
25 1
MR. MEYER:
Yes.
It's kind of a general o
2 statement, but I think this is the right time to make it.
3 Direct containment heating, along with seven other topics, 4
was identified in NUREG-0956 as an area of major 5
uncertainty.
6 Now, I an one of the principal authors of that 7
report.
I can tell you that~in NUREG-0956 we intended to 8
point out major advances in source term technology, compared 9
with technology that existed at the time of 1975, our TID j
10 14844 in 1962, which documents are the basis for the current i
{
11 regulatory practices that we are operating on.
I 12 In the process of trying to demonstrate the major
- (s 13 advances in our technology, we candidly pointed out areas t
14 where we still had big uncertainties.
l 15 We did not intend that those c.reas be raised as 16 impediments to any further movement, since we were trying to l
17 put forward in NUREG-0956 a positive statement about a 18 product, a new technology.
19 So to me, as a researcher involved in identifying 1
l 20 this issue, it is somewhat uncomfor. table.to see the issue now 21 raised as an impediment that must be absolutely resolved l
j 22 before one moves away from these 1962, 25-year old 23 technologies.
24 MR. THOMPSON:
Could I make a response in a j
25 general way.
The issue here is relaxation of emergency l
4
/\\CEJFEDERAL REPORTERS, INC.
202 347 370(1 Nationwide Coverage N10 346646
. -,... _ ~
30313.0' ccx 26 (s
i response regulations.
We see the emergency response as an 2
independent means of protecting public health and safety, i
3 totally independent of plant operation.
The regulations are 4
currently based on reactor safety study, vintage source term 5
knowledge.
6 The reason why we give great attention to issues 7
like direct heating is that there are grounds for arguing on 8
the present knowledge about those issues, that the emergency 9
response regulation should not be relaxed.
I think it is 10 perfectly proper to point to uncertainties in phenomena, as 11 justification for not relaxing emergency planning 12 regulations.
As 13 I agree that advances have been made in 14 understanding, over the years, that it's very clear that 15 major uncertainties continue, and those have to be addressed 16 in the course of the hearing.
17 MR. MEYER:
You don't see a need to show a tighter 18 coupling over the concern in this particular case, but just 19 to identify an area where the technology has not moved far 20 enough to understand it with precision.
I mean, to me it 21 would seem that it would be more appropriate to tightly 22 couple the concern with a specific application instead of 23 just raising it because science hasn't progressed real far 24 down that pathway.
25 MR. THOMPSON:
I think we accept that if it can be L
ace FEDERAL REPORTERS, INC.
202 347 37m Nationwide Cmcrage a n336W 46
30313.0 ccx 27
\\-
1 demonstrated that depressurization will always occur, or will 2
occur with high confidence, I think Steve and I would readily 3
agree that the direct heating issue then is irrelevant.
4 MR. MEYERI We have seven more here that would f-5 continue to work on it.
6 MR. THOMPSON:
It's not our exclusive interest 7
either.
It just happens to be the one we started with this.
8 morning.
9 MR. ROSSI:
Well, you know, in summary, the issue 10
-- we don't have the answers or the conclusions on these 11 issues today.
The work that has been done up till now has 12 been done with the review of identifying things that have to
\\s 13 be looked at in the future.
It's unlikely that all of those 14 have been identified, and now we are going to basically roll 15 up our sleeves and start going through the tougher, more 16 detailed review; and the other issue that clearly has to be 17 addressed is the purpose of emergency planning period.
I 18 mean, that's one that will be addressed.
19 MR. THOMPSON:
As part of the Seabrook?
20 MR. ROSSI:
Yes.
It clearly has to be addressed 21 as part of the petition.
22 MR. REIS:
The question is, in the structure of 23 the regulations, will the purposes of emergency planning and-24 protection of the public health and safety be as well served 25 by a smaller EPZ as the one generally indicated in the Y
/\\CEJFEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage IKXb336-6646
30313.0 28 cox y
1 regulation.
So that has to be looked at as well.
The 2
applicants have indicated, in some ways, that they are 3
looking at it, and they had hoped that their petition covered 4
this.
We indicated at the last meeting with'the applicants 5
that there were some substantial questions left on that t
6 specific issue.
7 MR. THOMPSON:
Now, as I understand it, the 8
licensee is just asking for the zone to shrink.
Other ideas 9
that have been put forward include a smaller evacuation zone 10 with perhaps a larger sheltering zone..
11 Will those kind of possibilities be addressed in i
12 your review?
i N-13 MR. REIS:
It depends, because at that point you 14 get to the emergency plan itself.
Of course there has to be 15 a range of protective actions to take care of the unlikely 16 event of a major accident.
What that range of protective 17 action is will have to be looked at.
It may be that 18 something in there can be done.
That hasn't been focused on 19 directly.
20 But if you do shrink the zone, you would be 21 looking at -- and we are looking at how far you have to
)
1 22 evacuate and when you would sh'lter, how you would choose e
23 your protective action.
24 MR. THOMPSON:
I think I am correct in saying that 25 various analysts have raised the possibility of a smaller s-4 ACE.FeoERAL REPORTERS, INC.
202.147 3700 Nationwide Coverage mx1-33MM6
30313.0 cox 29 V
1 sheltering zone and a larger -- I am sorry, smaller 2
evacuations than the present and that the sheltering might 3
extend beyond 10 miles.
4 MR. ROSSI:
Ed, isn't it --
5 MR. REIS:
The regulation -- go ahead.
6 MR. ROSSI:
I was going to say that it would be-my 7
opinion, that as we go through the review, that the applicant 8
would have the flexibility of making modifications to the 9
specific details in this current petition.
10 MR. REIS:
That's true.
11 MR. ROSSI:
That he can do that based on 12 interactions with the staff and what the applicant perceives
\\,
13 as our review finding.
14 So I think that your basic question is, well, is 15 there some way of including alternatives in this effort, and 16 I think the answer has to be yes, as I see it, the applicant 17 would have the flexibility of modifying the petition based on 18 what our review shows.
19 MR. REIS:
Certainly.-
20 MR. NOVAK This is Novak again.
I would like to 21 get back to the specific subject and see if we can understand 22 where we are and then summarize and see if we can agree.
23 The point of direct heating, very simply, has been 24 a phenomena that the staff has recognized for several years.
25 I think NUREO-1150 helps focus on it and provides some V
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 8m33M646
30313.0 cox 30 v
1 information related to the uncertainties that one or a 2
collegial body of people may suggest that phenomena 3
provides.
4 Clearly, the point that we are talking about, is 5
the likelihood of early containment failure.
6 Now, we will speak to that directly, to the 7
utility.
We will establish that the staff considers that 8
direct heating provides enough uncertainty, based on what his 9
submittal has given us, with regard to direct heating, that 10 we are not convinced that the likelihood of early containment 11 failure is as small a value or as low as value as he is 12 predicting.
What he does with that statement is up to him, ks 13 really.
If you have gone back and read the record, I think 14 you are familiar with his general philosophy on direct 15 heating.
16 Based on the information he has provided, we do 17 not accept his rationale.
That is, that there is a simple 18 argument that the boundary conditions, the geometry of the 19 cavity and other parts, clearly make this phenomena not a 20 challenge to the Seabrook facility.
Based on the information 21 in the docket and in the waiver, we don't agree at this point 22 in time.
23 He may choose a number of avenues to try to 24 convince us that the probability is as low as they are.
- Now, 25 that's his choice.
%s ace FEDERAI, REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-33MM6
30313.0 cox 31 us 1
With regard to your comment on the 1150 review, it 2
would be my cpinion that we use 1150 as a source of 3
information.
I don't expect to tie it directly to any 4
licensing application.
That is, the generic review, the 5
final issue of the report, I don't believe, and I don't think 6
the commission has directed the staff, to use it as a 7
regulatory product at this stage.
That is still some time 8
off.
9 But, however, we are using what we consider to be '
10 understanding a phenomena that has been described in that 11 report and other reports to help guide us in understanding 12 risks to plant operations.
So from that point of view, I 13 think it is a source of information.
14 But I would not, at this point in time, want to 15 conclude that that process must be complete before any 16 regulatory action on any act, in any case, could go forward.
17 I don't think it's unique to Seabrook.
18 MR. SHOLLY:
Tom, I think obviously the focus of 19 the concern is on early containment failure.
But there are 20 two aspects that I wouldn't like to see get lost in the 21 cracks.
It may be possible, for example, to show that for 22 Seabrook's configuration procedures, design, what have you, 23 that the maximum pressurize one might expect doesn't get you 24 to the containment failure pressure.
If that can be 25 demonstrated to everyone's satisfaction, fine.
It wouldn't ace. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage A(U-336-6M6
30313.0 32 cx As 1
make the issue go away completely, though, because there are 2
two aspects to it.
One is possible impacts of melt ejection 3
and production of the aerosols on continued operation of 4
containment safeguards equipment, whether or not sprays will 5
be able to keep operating, if they are indeed available at 6
that point.
Perhaps you will get a large quantity of 7
various-sized aerosol materials into'the containment and foul 8
the spray pumps, and they will quit operating.
That's a 9
possibility.
Perhaps the containment fan cooler function 10 will be degraded, or it will fail entirely, also a 11 possibility that would impact your consideration of later 12 containment failure.
\\--
13 The other aspect is the impact of melt ejection 14 and direct heating phenomena on the sorts of radionucleides 15 you would see in the containment atmosphere over time; 16 chemical forms, in particular, oxidizing the typically 17 refractor stuff.
There has been some indication in the 18 Sandia test that you are getting some preferential oxidation 19 of some forms.
Molybdenum is one, I think, that has been 20 tossed around.
Those sorts of things ought not to be ignored 21 in all of this.
Although the key issue is early containment 22 failure.
23 MR. NOVAK:
Any comments that you could make on 24 those specific issues?
25 MR. PRATT Yes.
My/ focus is certainly on early f
v I
ace-FEDERAL RijPORTERS, INC.
l 202-347 3700 Nationwide { overage mxk336 H46
30313.0 ccx 33
\\-
1 containment failure.
I think the two other items that one 2
can bring up are of much less importance, particularly for 3
seabrook.
I think you can survive the phenomena created for 4
vessel failure, if you can survive that; then whether the 5
sprays operate or not, you would be talking about very late 6
containment, I believe, in those circumstances.
So I don't 7
consider operation of equipment in containment that crucial 8
for this particular case.
9 Again, the form of the radionucleides I think is 10 important, but important again if we have containment 11 failure.
I think what we see is if we have early containment 12 failure with direct heating, we see a releasing of the I
13 refractor products in the theme of the work that Sandia does 14 and certainly that range.
Again, I think these are issues --
15 they are issues really in focus of the containment failure.
16 MR. MEYER:
If you were to have a late containment 17 failure and core concrete interaction, which is usually 18 thought to be kind of incompatible with the high-pressure 19 ejection, either you plop it out and let it interact with the 20 concrete or you spray it out.
If you had the core concrete 21 interaction, you could have a sustained source of aerosol 22 generation, which would give the respective radionucleides 23 that I think we are predicting fairly accurately.
It's true 24 that the high-pressure melt ejection might produce a 25 different spectrum of radionucleides, because it would tend V
ace.FEDERAt. REPORTERS, INC.
202 347 3700 Nationwide coverage s k336-M 46
30313.0 34 cox j
(-
1 to be a mechanical dispersion process that wouldn't be 2
chemically selected.
Therefore, you can get more of the 3
refractor materials.
4 But that happens at the time of vessel failure, 5
and the natural aerosol deposition processes would tend to 6
bring that out of the atmosphere pretty effectively for a 7
late containment failure.
8 MR. PRATT:
I think it is just important to 9
emphasize if we do survive this phenomenon as a failure, we 10 are talking about a very long time in this facility, we would 11 agree with the calculations on the pressurization.
So there 12 is a lot of time for these processes to occur.
I think way
\\--
13 beyond what we would expect for continued release of the 14 fission product during core concrete interactions.
15 MR. SHOLLY:
The other action that comes into 16 play, of course, if containment failure is delayed to very 17 long time, his emergency response becomes both much less of a 18 problem and less easier to implement it if you have a plan or 19 not.
If you have 24 or 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> there will be very few 20 options where you can get people out of the way.
21 MR. THOMPSON:
One point about direct heating that 22 might or might not be an issue, that has crossed my mind, is 23 the potential deposition of core material on penetrations and 24 the heating of those penetrations.
25 MR. BAGCHI:
What kinds of penetrations would you
%s Ace FEonRAL REPORTERS, INc.
202 347 3700 Nationwide Cmerage 8(0 31MM6
30313.0 ccx 35 V
1 think would be vulnerable to deposition of fission products 2
in terms of release outside the containment?
3 MR. THOMPSON:
Well, if I would look at this 4
issue, I would start by looking at the potential trajectors 5
of projected melt and the arrangement of pipe penetrations 6
and equipment hatches and the like, and see if it's feasible 7
for material to conglomerate around those and then see where 8
that thermal-induced breach would lead to.
As I say, this is i
9 not something I have investigated.
If you have, you can put 10 the concern to rest.
That's fine.
11 MR. BAGCHI It was a very simple question.
I 12 wanted to understand whether or not your emphasis is on
(.-
13 sealed materials or any other type of material.
14
, MR. THOMPSON:
If you are ejecting many tons of j
15 material, then accumulation of a few tens of kilograms around 16 the pipe penetration could give an elevation of the 17 structural temperature around that.
18 As I say, it might turn out that this concern 19 would go away when you looked at it.
j 20 MR. PRATT:
We have recently gotten word from 21 Sandia on a calculation that they did with the contain code, 22 to explore the concern about heating of the containment walls 23 for any possible indications of reduced strength.
The 24 calculations, that I believe we reported in the weekly 4
25 report, one of our internal blurbs, showed that the Ace FEDERAL REPORTERS, INC.
i 202-347 3700 Nationwide Coverage 800 3364M6
30313.0 ccx 36 I
temperature rise was really insignificant, and that Sandia's 2
opinion, after doing these calculations, was that the DCH 3
phenomenon was principally a pressure threat to the 4
containment, not temperature.
S' MR. THOMPSON:
This is uniform distribution?
6 MR. LEE:
Let me expand on that.
What we want to 7
see -- that we are looking at those, and those concerns, in 8
addition to the containment and the aerosol generation.
9 The issue you are talking about is, of course, 10 over the past year, and we are initiating now for partial 11 containment line abrasion.
It's a direct part of the 12 material with the containment liner, including containment 13 penetration if it's in the area where it could be sprayed 14 with the material.
15 Another thing is what Rob just indicated.
We are 16 also looking at the possibility of not only the pressure of 17 the containment, mainly the containment, but the elevated 18 temperature could also weaken the containment material and 19 could possibly produce some stress in addition to, say, the 20 membrane stress caused by the pressurization with 21 containment.
So we are looking to those issues, and all I 22 can say at this moment, all of your concerns raised so far, 23 are within our program.
24 MR. SHOLLY:
Ralph, you mention the temperature 25 issue.
Is it that the temperature rise doesn't go high w
J
/\\CE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6M6
30313.0 cox 37
\\-
I enough to be of concern, or that it decays away relatively 2
quickly, or --
i 3
MR. MEYER:
I think it has something to do with 4
the time scale.
We did have a letter from Sandia.
The 5
bottom line was that they had looked at the temperature i
6 excursion and found it wasn't very big.
7 MR. LEE:
The key point we have to remember, in 8
the direct containment heating, all heat pressure rises and 9
temperature rises are quick, but they also decay very i
10 quickly.
j 11 So for us -- regarding the pressure, if the 12 initial part of the pressure does not fail the containment, j
(,
13 most likely the direct containment heating will not cause the 14 failure of the containment within the next few hours or even i
15 days, because all of the pressure effect of the direct 16 containment heating is within 30 seconds to 1 minute after i
17 the melt ejection.
l 18 What happens to this temperature, although the I
?
19 temperature decays -- well, the temperature decay goes hand 20 in hand with the pressure decontamination.
The finding of i
l 21 the Sandia calculations that the peak temperature goes up i
i 22 above a point where the containment liner may be weakened or I
23 even melted if that temperature can be sustained.
24 But, that temperature will not remain at that for 25 a significant period of time.
It decays very fast down to i
\\s l
I 202 147 3700 Nt nwi ov r g 336-6M6 i
30313.0 ccx 38
(-
1 the -- I think, if I remember the number correctly, somewhere 2
.around the 500 degree Fahrenheit or something, within 3
something like a 10, 15 minutes.
At'that temperature, 4
everybody reading that report will agree that the strength of 5
the material, the effect of the temperature on the strength 6
of the material, does not seem to be at issue.
7
.The only issue which may be remaining is the 8
thermal stress and the stress concentration.
d 9
MR. SHOLLY:
Have any estimates been made yet of 10 the pressure sometime after the melt ejection; assuming the 11 containment doesn't fall, what does the pressure decay away 12 to?
%./
13 MR. LEE:
Right now, the calculation we have 14 conducted, up to about 10 minutes only.
After 10 minutes I 15 am reluctant to say the number, because I don't remember that 16 query.
But you can go into NUREG-1150 because we have 17 provided that temperature, I mean the pressure history curve, 18 for one calculation.
19 MR. SHOLLY:
I will take a look.
I obviously 20 haven't had a chance to wade through that entire volume.
i 21 MR. NOVAK:
I would like to go back, at least 4
22 again to two points, I think, on direct heating.
I don't I
23 know if it was Gordon or Steve, but you seemed to make a 24 reference to some pressure level for which you believe that 25 the phenomena, as we might understand it today, still 4
V
/\\CE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 8(xk34(M6
. - ~
e<
30313.0 cox 39
)
L 1
represents a risk.
I think you used something between 2-and 2
400 pounds per square inch.
That would be reactant cooler 3
system pressure.
4 MR. SHOLLY:
I have seen some indications that at 5
pressures that low they have gotten a fair amount of indirect 6
heating.
I haven't seen it liberally quantified or 7
anything.
4 8
MR. NOVAK That's why I brought it up, so we 9
could understand and have something on the record as to what 10 specific experiments are available for scrutiny, both by the 11 ones that we have sponsored that are available to look at, 12 which gives a better understanding, or at least permits the 13 scientific community to review what that information might 14 suggest in terms of a pressure sensitivity.
15 MR. SHOLLY:
I think it was one of the Hicks 16 experiments, but I don't recall which one.
j 17 MR. NOVAK:
Why don't we ask our research --
18 MR. LEE:
If HIPS is what he is referring to, is 19 what Steve is alluding to, my best recollection is that the 20 lowest pressure we have conducted a test with, about 480 21 PSIG.
200 PSIG, you are referring to, most likely comes from 22 either of the tests -- I think it's in one of the tests.
23 MR. SHOLLY:
So 4807 24 MR. LEE:
fes.
480 is the lowest pressure we have l
25 tested.
!. y ACEfFEDERAL REPOkTERS, INC.
202 347 3700 Nationwide Coverage 800 336-fM6 L
77_
JJ313.0 40 cox l
1 MR. NOVAK:
Just for my own information, is there 2
a document that you can refer to for the record so that 3
people can review that data specifically?
4 MR. LEE:
Yes.
5 MR. NOVAK:
You might want to add it, if somebody 6
will look it up.
7 MR. LEE:
I think it's been discussed in 8
NUREG-1150 but we also have a Sandia publication.
9 okay, it's NUREG CR-4512.
10 MR. NOVAK The date of that report?
11 MR. LEE:
I think it was published in October of 12
'86.
I do have several extra copies in my office if anybody
\\--
13 would like to have it.
14 MR. SHOLLY:
Is that the one on the preliminary l
15 calculations, direct heating that Sandia did?
16 MR. LEE:
No.
No.
This is on the HIPS -- four 17 tests.
Okay, two SPIT tests, system pressurization, SPIT --
18 MR. MEYER:
- SPIT, S-P-I-T.
I 19 MR. LEE:
Two SPIT tests, 18 and 19, and two area 20 HIPS test, HIPS 2-C and 5-C.
This NUREG report I am 21 referring to is a test report of those four tests.
22 MR. SHOLLY:
Thank you.
23 MR. NOVAK:
I had one more comment.
I would like 24 to address it to the research people.
Is there a specific l
25 document or reference we can refer to which identifies the ss i
{
/\\CE. FEDERAL REPORTERS, INC.
l 202 347 3700 Nationwide Coverage 800-3346M6
?
l
30313.0 ccx 41 (s
1 proposed test matrix that will be performed regarding direct 2
heating?
3 MR. LEE:
Yes, I think there is a paper we have 4
provided to the Brookhaven --
5 MR. MEYER:
Probably'the best document to find 6
this now is in the SECY paper, SECY 86369.
The plan is laid 7
out along with milestones and schedules and budgets.
8 MR. NOVAK:
It wouldn't give test matrices, would 9
it?
10 MR. MEYER:
Yes, it's a fairly large SECY paper.
11 I believe there is a discussion of the 11 tests planned.
12 MR. ROSSI:
Is that public yet?
(,
13 MR. MEYER:
The SECY paper?
Oh, yes.
Last 14 December.
This plan may also be in several other papers, we 15 are looking to see if it's in Appendix J of NUREG-1150.
The 16 test matrix might be there in there as well.
17 MR. SHOLLY:
Ralph, I just got quite a volume on 18 FOIA requests that lays it out in painful detail.
19 MR. THOMPSON:
How far forward does this research 20 planning extend?
21 MR. LEE:
Okay.
The SPIT test and the HIPS test 22 ue have talked about have been completed.
Right now we are 23 going to so-called DCH test, direct containment heating 24 test.
These tests as we have indicated are currently 25 comprised df 11 tests and we have completed two tests.
\\s l
Ace-FEDERAL REPORTERS, INC.
2 C 347-3700 Nationwide Coverage 800-336-6M6
30313.0 cox 42 L
1 MR. THOMPSON:
When is the estimated coupletion of 2
all 11?
3 MR. LEE:
Right now, the rate of the tests we are k
conducting more or less are subject to the constraints on our 5
completing one test every three months.
We may be able to 6
speed it up a little bit with additional funding, but it's 7
not conceivable to be able to conduct a test more often than 8
once every two months or once every month and a half, mainly 9
because of the preparation period for the test.
10 MR. THOMPSON:
Does the supporting program at 11 Brookhaven, which is both, if I can remember, analytic and 12 empirical?
\\-
13 MR. LEE:
Yes.
The supporting program at 14 Brookhaven is both analytical and experimental.
On the 15 analytical side, we are concentrating on directing skat'ing 16 load, which would enable us to extrapolate the result of the 17 small-scale test to a larger facility.
18 On the experimental side, we are donducting about 19 1/40 linear scale test, using water and wood metal, to try to 20 identify phenomena important for the ability to transport.
21 MR. MEYER:
If you are interested in the ongoing 22 and future research-program in this area, then keep in mind 23 these review panels that you referenced earlier.
Brookhaven 24 will issue a report in April on those reviews, and we will 25 follow that with a NUREG staff report, hopefully about 30 V
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6M6 4
l i
1 30313.0 cox 43
(-
1 days later, that shows a revised research plan.
We don't 2
know yet the extent of the revision.
We hope that we have 1
anticipated things and that the revisions wil] be minor.
4 Nevertheless, this subject received direct 5
attention in those reviews, and there may be some changes.
6 There's a report number on the Brookhaven report.
I think 7
it's NUREG CR-4083.
We do not have a staff NUREG number yet 8
for the other document, but it will have a revision of the 9
research plan that was in the SECY paper that I mentioned a 10 little earlier.
11 MR. NOVAK:
I think I would like to see if we can 12 summarize.
If we have kind of covered the direct heating
(,
13 subject, then I think -- I would propose we take a few minute 14 break here to give the reporter a chance to exercise her 15 fingers and take a 10-minute break and move on to the next 16 subject.
17 (Recess. )
18 MR. NOVAK:
Can we get ba'ck on the record now.
We 19 are going to continue now, and, as I understand,.Gordon, you 20 would like to discuss containment response or containment 21 design.
I forget just exactly --
22 MR. THOMPSON:
Containment performance.
23 MR. NOVAK:
Fine.
Why don't we just pick up that 24 subject next and discuss it.
25 MR. THOMPSON:
I guess my first question is, is it v
/kCE FEDERAL REPORTERS, lNC.
202 347 3700 Nationwide Coverage 800-336-6646 j
l
p 30313.0 c5x~~~'
44 1
contemplated.that the final element analysis will be 2
repeated, as it was not.. carried to the point of numeric l
l 3
instability, whatever that represents.
I wonder if that is 4
going to be done.
5 MR. HOFMAYER:
A matter of time.
I guess we I
6 carried it to the extent that we have gone.
Potentially you l
7 could go higher, but we had to keep the steps small.
At that 8
point, we got it to a point where we felt, where, in this t
9 particular containment, the shear failure did not govern over 10 the hoop failures.
Now you have a race over how much 11 yielding you are going on, a large phenomena.
You could 12 carry it higher.
w-13 MR. THOMPSON:
Shear failure didn't dominate over 14 yielding in hoop direction but it might dominate over failure 15 in the hoop direction.
16 MR. HOFMAYER:
I am not sure I understand that.
17 MR. THOMPSON:
Well, my recollection is that shear 18 failure could occur at some point above 165 PSIG, based on 19 these calculations.
20 MR. HOFMAYER:
We are not for sure that that would 21 happen.
There is still a disagreement with the applicant.
22 They contend it's much higher.. I can't.say that I have 23 proven one way or the other you are getting significant' l
24 concrete -- high concrete stresses, high cracking down i
25 there.
We are saying, well, within the limitations of the s-ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 33MM6 l
c-
~
r 30313.0 CEN' 45 4
W 1
code that we are working with right now, you know, we felt 2
that, you know, it wouldn't be clear.
You could be debating 3
that, I guess, for a while.
4 MR. THOMPSON:
Now, if I understand correctly, the 5
footing for the enclosure of the building was not taken in i
6
,this calculation; is that correct?
7 MR..HOFMAYER:
No.
i l-8 MR. THOMPSON:- Given that footing is not complete 1
9 but only covers about 3/4 of the circumference, might the f
i 10 including of that footing have some effect on the structure?
l 11 MR. HOFMAYER:
No.
I don't think we b'und that l
o 12 analysis by any assumption that we made.
Basically, we held-l
' y, 13 the base.
I believe the other alternative might be you might 14 have some additional motion or something, then the question 15 would be, would that increase the shear or decrease it.-
16 MR. THOMPSON:
But the holding -- am I correct in j
i 17 saying that it's assumed that that footing is what is one of a
i 18 the things that restrains the base?
l 1
l 19 MR. HOFMAYER:
That's correct.
j 20 MR. THOMPSON:
That. footing is not complete.
l 21 There are gaps?
22 MR. HOFMAYER:
I think there is a 30 degree gap.
23 MR. THOMPSON:
At that point the base is not 24 restrained so there could be a double curvature of the points 25 where the footing stops?
sa i
l j
/\\CE. FEDERAL REPORTERS, lNC.
202 347-3700 Nationwide Coverage 800-336-6M6
n 30313.0 cox 46 s
3, I
La t
1 MR. HOFMAYER:
I wouldn't know for sure.
You t
a 2
could postulate that, I guess.
But the question would be i
3 would that have a big effect on the overall analysis, and our j i
4 judgment is that it wouldn't.
5 MR. THOMPSON:
Is this code being used to forecast !
1 6
the performance of the Sandia model?
l 7
MR. HOFMAYER:
Yes, it was.
~
1 8
MR. THOMPSON:
I see our research people are not
- [
9 here, but I am curious as to how this prediction process is 10 going to be done.
I understand that some 10 or 12 11 predictions are going to be made for the model.
I am curious l
12 as to the protocols for that prediction.
i' s-13 MR. ROSSI:
I think these two fellows will be back 14 shortly.
15 MR. NOVAK:
No, I think they have gone.
' 16 MR. ROSSI:
They have left their stuff here --
17 MR. NOVAK:
Let's leave the question and see if we 18 can get Ralph or Tim to respond to it.
19 MR. THOMPSON:
Then the question is how to handle 20 potential problems like corrosion or rebar corrosion over the 21 years of operation.
l22 MR. NOVAK Let me see ? if f-I understand.
First of 23 all, there was a question as to whether or not a specif'ic 24 analytical code, that I guess, Charlie, you were using, as to 25 whether it would be benchmark, in terms of how well it
(-
ACE FEDERAL REPORTERS, INC.
p 202-347-3700 Nationwide Coverage 800 336-f686
i 4
.sm.
4 t
~ 30313,0 c5[
.47 v
1 predicts containment response as part of subsequent tests to 2
be performed?
I am not sure that these two people are the 3
most knowledgeable people.
Joe, can you speak to that at 4
all?
5 MR. MURPHY:
No, I don't know the details of that 9
6 either.
7 MR. HOFMAYER:
I can answer the question.
The 8
code was used to make a prediction on the Sandia model.
- Now, 9
I did not do that.
People within our division did do that.
10 You asked me the process that they went through.
I really 11 can't answer that question.
There were a number of 12 predictions, I guess, made.
N.,
13 MR. THOMPSON:
I am just curious-as to whether 14 those predictions have been published, and, if so, where.
15 MR. HOFMAYER:
I think part of the problem, I-16 guess, or not problem, but a concern was that I think 17 research wanted people to make the predictions independent of 18 one another, so, therefore, they had the point that this is 19 being done right now.
They.are not interested in having 20 everyone compare notes; they would like to have the analysis, 21 do the test, and then compare.
22 But you would have to talk to them.
23 MR. THOMPSON:
I think it would be nice to have'a 24 protocol in which everyone did the analysis independently.
25 Then they are all published and then the test was done.
L ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3E6646
303 3.0 3
cox 48 l
i N-
~
1 MR. NOVAK:
Do you know if.it's following a.
2 standard problem format?' Historically, where research has, 3
and the NUREG' staff has set out to identify a specific test 4
to be performed, where a format was used which they refer to 5
as the standard problem, where all of the information was 6
provided and then people provided'their analyses and the 7
results were kept fron public disclosure until~after the test ;
8 was formed, and then the NUREG then'would issue the results 9
and show the comparisons from all people who participated.
10 Is this what we are talking about today?'
11 MR. HOFMAYER:
I assume that is the case, butLI 12 really would hate to answer, because it is only conjecture at 13 this point.
i 14 MR. NOVAK:
What is the. code'we are talking ~
15 about?
16 MR. HOFMAYER:
Brookhaven used the NFAP. code which i
17 is the same code that is referenced in our report.'
4 4
18 MR. SHOLLY Do you know offhand who at Brookhaven 1
19 did the prediction for the test?
Who did it?
20 MR. HOFMAYER:
Basically the analysis was done by 21 Jose Perez, who also worked on this project;-and Carl 22 Constantino was actually the principal analyst involved with 23 overall prediction processes.
24 MR. THOMPSON:
You are saying in your. judgment 25 three-dimensional effects would not=be significant.- I don't ib f
/\\CEJFEDERAL REPORTERS, lNC.
202-347-3700 Nationwide Coverage -
800-336-Mi46 1
4
.m-..+
w-r r
r rw-
-y er, e--
=
--m-e
+
-- -, +,
30313'.0 cox' 9
L
~
1 know whether that is true or not.
But let's suppose you are 2
in a threshold situation where you think the direct heating 3
pressure may reach some level,-and you think the containment 4
capability is some other level which is very close to it, S
then if the three-dimensional effect is a 5 or 10 percent 6
effect, then it can be significant.
7 If the challenge-and the capability are far_ apart, 8
then the small change is not significant'.
I just~ wonder 9
whether you have any feeling for the level of significance 10 three-dimensional effects might show.
11 MR. HOFMAYER:
It's hard to visualize-some of i
12 them.
We brought this question up-when the-applicant first i\\-
13 made their submittal when they were proposing very high 14 strains with the containment.- Possibly at those very high 15 defamations, our thinking was something you could consider.
16 On further thought of the whole. process, we-felt more 17 comfortable, and that is why we took_the. position-of the 1 18 percent strain as more of a median incentive value; and the-19 applicant has never agreed with that position.
But we feel 20 that there is a lot of uncertainty, as you get beyond that 21 point, and we feel more comfortable and it's a more 22 defensible position to go with something a little bit more in 23 line with what people have postulated for the failure.
24 MR. THOMPSON:
The Sandia test of concrete 25 containment model, that will not have a ring _ guard i %-
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646
7 30313.0 50 cdx
(_-
I surrounding it; is that correct?
2 MR. HOFMAYER:
I don't believe so.
3 MR. THOMPSON:
Will it be back-filled or sitting 4
on the ground?
t.
5 MR. HOFMAYER:
I really can't speak.
I would only 6
be giving you my brief understanding that.
I have not 7
studied the Sandia model or the setup there.
I.aun f amiliar -
8 with the fact that the test will take place, what the general 9
configuration and details; and I think perhaps research could' 10 give you more information on that.
I believe it's.on soil, 11 that particular model.
12 MR. THOMPSON:
Obviously we need to pursue that
\\s 13 elsewhere.
14 MR. HOFMAYER:
The base condition, because of 15 slightly different base conditions, may be slightly somewhat 16 more.
17 MR. THOMPSON:
The next point I meant was 18 potential corrosion of rebar over years of service,-whether 19 that might be significant.
20 MR. HOFMAYER:
The concrete is protecting the 21
.rebar.
That is basically what it is doing.
You are not 22 really in any kind of a -- even,-you are not in.any-kind of a 23 corrosive atmosphere to that extent.
So concrete has 24 performed admirably over the years.. There's.not a lot of 25 evidence of concrete structures failing due to corrosion or-
%s ace-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30313.0 cox 51.
(-
1 it would be, substantially be degraded, especially the 2
structures are very heavily reenforced, and they have a 3
substantial concrete cover.
So you have a protected 4
corrosive atmosphere.
5 MR. THOMPSON:
The next question is voids in the 6
concrete, what effect that might have?
7 MR. HOFMAYER:
You know, you are subjecting this 8
containment to a tensile environment, and basically the steel 9
is carrying the load.
10 I just can't envision if there were voids, and 11 there's no evidence that there were, maybe there is a 12 potential void, but it certainly would not be very large or A-13 substantial effect on a structure.
I can't visualize that 14 having a significant impact on anything.
15 MR. THOMPSON:
Even at the wall base junction it 16 would not be significant there?
17 MR. HOFMAYER:
Again, you are carrying the forces 18 of the steel.
You would have to have a void completely all 19 the way around or something to have some big effect.
20 MR. THOMPSON:
That's about it for containment.
21 MR. NOVAK:
Okay.
22 MR. ROSSI:
I just invited Tom back to the 23 meeting, because we were talking about the containment.
I 24 don't know whether he can answer it or provide anything more 25 or not, but Tom is coming back down.
s-ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
v 30313.0
'ccx 52 N-1 MR. NOVAK:
Do you have another particular subject 2
area you want to take up next?
3 MR. SHOLLY:
I would like to talk about the 4
seismic sequences.
I understand and I have seen the PS&H 5
submittal revising some of the -- I guess it's the fragility 6
estimates.
But I have not seen any quantified format as to 7
what it does to the seismic core melt sequence.
I would be 8
interested in knowing if the staff has any views on what 9
those changes in fragility-are going to do to the seismic 10 core melt frequency.
11 But working with that, aside for the moment of 12 what exists in the original PRA updates, there is a fair
%s 13 fraction of the core melt frequency, as I said, something 14 over 13 percent comes from seismic events and about half of.
15 that from very large ground acceleration values.
An area 16 that had been raised, both early on in the seismic safety 17 margins research program by structural mechanics associates, 18 and others at various times, is-whether an aftershock 19 occurring within 10- to 12-hour time frame, let's say whether 20 the containment is going to be pressurized,.perhaps to 80 or 21 100 PSI, whether a fairly substantial aftershock could 22 combine with that load to give you containment failure.
23 Then the other -- I think the other significant 24 aspect of this_is the impact of seismic events on shutdown 25 accidents.
I wonder whether -- I know Brookhaven is doing N-ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6646
4' 30313.0 ccx 53
(-
1 some work on reviewing NSAC-84.
I wonder whether the 2
incidence of seismic shutdown has come up during that 3
review.
4 MR. NOVAK:
Can we take those questions one at a 5
time.
I would first speak to the risks we perceive from 6
aftershock phenomena.
I think we did address that to some 7
degree.
I thought I read something regarding the likelihood 8
of aftershocks.
9 MR. ROSSI:
Did Brookhaven?
10 MR. HOFMAYER:
We did not.
11 MR. BAGCHI:
Aftershock we did not address in.that 12 way.
However, we do need to understand what your concerns N.,
13 are.
Based on my limited review of containment capability 14 due to major earthquakes, I feel that the fragility is very 15 high.
So high, that they are not likely to be affected by 16 the core melt type of earthquake that was raised at 17 Seabrook.
18 MR. ROSSI:
Their specific question was an 19 aftershock that occurred after containment was at high 20 pressure.
21 MR. BAGCHI:
If aftershock breached the 22 containment integrity, which would then lead to an accident, 23 so that it would be available for further shaking in the 24 aftershock, then we should not have a concern with 25 aftershock.
We need to think it through very carefully.
L ace FEDERAL REPORTERS, INC.
202 347 37U)
Nationwide Coverage 800-336-6646
30313.0 ccx 54 A-1 That's the point I like to make sure.
That's all.
2 MR. NOVAK: -For the record, what I recall reading 3
is a response staff developed to a Congressional inquiry 4
regarding earthquake phenomenon and potential aftershocks.
I 5
think there is a response that we did provide on the record.
6 MR. BAGCHI:
That response was along the same 7
kinds of lines that I just discussed.
8 MR. SHOLLY:
Would that have been relatively 9
recent, this year?
10 MR. NOVAK:
Yes.
I think the questions were 11 provided to Congressman Markey around March of this year.
12 We have another question on the table dealing with 13 concerns of earthquakes during a shutdown mode.
14 MR. ROSSI:
Let me just say something about that.
15 The accidents during shutdown is one of the things we have on 16 our list that has to be addressed more thoroughly.
So that 17 would, in my opinion, include the seismic effects on this.
18 Scot, you may want to add something more to it.
19 MR. NEWBERRY:
Do you recall, Trevor, what NSAC-84 20 did at the utility if it included external events?
21 MR. PRATT I don't think so.
22 MR. SHOLLY:
I don't think so.
I wondered if the
.23 issue came up in the review you are doing.
24 MR. ROSSI:
We have identified the general area of 25 accident shutdown as something that has to be addressed in s-ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
30313.0 cox 55 As 1
more detail.
2 MR. LONG:
External initiators have been 3
identified as part of that concern.
4 MR. BARRETT:
Is there some specific concern about 5
seismic accidents at shutdown as opposed to accidents in 6
operation?
7 MR. SHOLLY:
I have~two questions.
I haven't 8
given it a great deal of thought, but it's one that jumped up 9
immediately, because it's one that you look at when you have 10 accidents at full power.
Sorts of things you might have, 11 rather long-term loss of AC power, to the point where if you 12 had AC power, operators might be able to do some recovery 13 actions and get you out of the sequence.
Where if you don't 14 have the power, you can sit by, watch it happen, and not do a 15 whole lot else.
16 MR. BARRETT:
This is the matter of making sure 17 that our review of shutdown accidents is complete.
18 MR. LONG:
There was one other part to your 19 question.
That was the incorporation of the seismic 20 fragility update into a PRA.
The company hasn't produced an i
21 updated PRA with that material included, but they did do a 22 review and response and provided information in response to 23 one of our requests.
I don't remember offhand the number, 24 but we can get that material.
25 They did provide a -- I guess I should look to the
\\we ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6M6
30313.0 cox 56 s.
I company, but it was, what I recall, an estimate of what they 2
thought the effect would be when they did it formally.
3 MR. SHOLLY:
I didn't recall seeing anything.
4 That's why I asked.
5 MR. LONG:
They also identified a relay chatter 6
phenomena that they thought needed to be addressed.
You have 7
hit upon both of those topics, so they are on the table.
8 MR. BAGCHI:
I do want to point out about relay 9
chatter, that came up in the morning, there is an unresolved 10 safety issue which is now resolved, of which a letter has 11 gone out, 8702 or something like that, qualification of 12 equipment in operating effects or, earthquake effects.
Relay
\\s 13 chatter has been looked at fairly carefully, and all of these 14 can be incorporated in our findings for Seabrook.
You might 15 want to take a look at that.
16 MR. ROSSI:
That generic letter, I think, is 17 mostly directed at the older plants; is it not?
18 MR. BAGCHI:
That's correct.
19 MR. ROSSI:
The newer ones handle the seismic more 20 thoroughly and so would not come under the generic letter.
21 MR. BAGCHI:
Thank you.
22 MR. ROSSI:
Seabrook would not come under thap.
23 MR. BAGCHI:
Yes.
24 MR. ROSSI:
But the general question of relay 25 chatter would need to be addressed in some way.
ACE-F'EDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
i 30313.0 cox 57
\\s 1
MR. SHOLLY:
Is there a NUREG report or technical 2
report?
3 MR. BAGCHI:
The whole packet has gone out with 4
the NUREG reports that incorporate our comments along this 5
line.
6 MR. SHOLLY:
I will look at that.
7 MR. BAGCHI:
USI 846.
8 MR. NOVAK:
Just to be sure I understand, when you 9
talked about earthquake challenges during shutdown, is there 10 any particular plant configuration that was a concern, that 11 is, you may be in a refueling operation, for example.
Is it 12 just the general concern that-clearly shutdown provides s-13 different plant configurations which should be looked at in 14 terms of potential risk if an earthquake were to occur during 15 those modes of operation?
I think that's what I understood 16 your question to be, as opposed to anything that.you have 17 looked'at, and is there a particular vulnerability that you 18 see?
19 MR. SHOLLY:
Nothing specific.
It's an area we 20 are going to do some looking at in the next couple of 21 months.
22 MR. NOVAK:
When we are-finished with this 23 particular subject, we can.go back and see if we have a 24 better response regarding the specific tests to be sponsored 25 by research and what we might be able to add with regard to V
/\\CE. FEDERAL REPORTERS, lNC.
202 347 3700 Nationwide Coverage 800-336-6646 l
t
30313.0 cox 58 L
1 the analytical verification.
There was a ques, tion that came 2
up a few minutes ago, and I think it went to the point of can i
3 we add any more information regarding how the Agency is-4 sponsoring any form of standard problem with regard to 5
containment codes used --
~
6 MR. HOFMAYER:
Well, the question was the 7
pre-prediction process for the Sandia model test.
We, for 8
example, have had a pre-prediction of that.
I think the 9
question was what was the protocol of making that type of 10 information available.
There have been a number of I
~
11 pre-predictives, I believe.
They would like to know a little 12 bit more about that process, I think, in the question.
\\--
13 MR. THOMPSON:
Right.
~
14 MR. BAGCHI:
We can refer you to the project 15 manager.
However, all of this information is available 16 publicly.
These are blind predictions.
These are intended 17 to be the best analytical method that offers a correlation 18 with the test result.
After the test, there is going to.be a i
19 detailed analysis of the methodologies and a-determination 20 will be made which one does.the prediction best.
21 MR. THOMPSON:
I understood it's not normal 22 practice in the standard problems to publish the predictions 23 prior to the test; that's what somebody said.
24 MR. BAGCHI:
I am not sure I understood that.
25 MR. THOMPSON:
A comment was made while you were
.L ACE. FEDERAL REPORTERS, INC.
1 202 347 3700 Nationwide Coverage 80 4 336-6646
~__
30313.0 cox
.59
\\_-
1 out of the room that the normal procedure in the standard.
2 problem, for-instance, at Loft, is to secure a varidty of.,.
3 lined predictions, but not to publish-those predictions until 4
after the test is completed.
An alternative, of course,;is c 5
to publish them before the-test. '
~
6 MR. BAGCHI:
Predictors might not be published but 7
I think the results should be available.
I have to check 8
back and tell you whether or not it is.
I couldn't tell 9
you.
I would imagine that the predictions would be 10 available.
The DNL prediction is already available to us; 11 Frank's prediction, I think, has come in; a few other,s.
12 Personally, I do not know of any reason why the predict 1,ons..,
(_-
13 would not be made available.
14 MR. ROSSI:
I am not sure we have adopted the,same s
15 philosophy on code comparisons.across the board that-were 16 used on Loft.
Maybe somebody from-research'can comment on-u 17 that.
Do you know?
1
,g.,a 18 MR. NOVAK I am,tryi.ng to refer -.it doesn't 19 really matter if the information is available, that.is.the,.
20 important thing.
,.37 21 MR. THOMPSON:
The project manager can announce I
22 this definitively.
Who would that be?
23 MR. BAGCHI:
Dr. Costello.
o
.,4 24 MR. THOMPSON:
Another-question about the' test is i
25 the containment back-filled or is it sitting on the ground ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33M6m 1
30313.0 cox 60 1
service, the Sandia model?
2 MR. BAGCHI:
You are going to have lots of 3
differences between the model tested and Seabrook 4
containment.
5 MR. NOVAK:
But that wasn't the question.
6 MR. BAGCHI:
I know.
7 MR. NOVAK:
The question is, do you know the 8
specific geometry of the test?
9 MR. BAGCHI:
There is no specific circumference 10 surrounding the model in Sandia.
11 MR. THOMPSON:
I think we are done with seismic.
12 Maybe we could deal with induced tube rupture.
13 MR. NOVAK:
Fine.
14 MR. BARRETT:
Did you say " induced tube rupture"?
15 MR. THOMPSON:
Yes.
16 MR. NOVAK Why don't we go ahead --
17 MR. THOMPSON:
Induced or exacerbated tube rupture 18 because preexisting small leaks might be a precursor to 19 rupture, as identified in 1150.
20 MR. NOVAK:
Is it fair to ask, can you summarize l
21 what you believe 1150 has said about rupture events?
22 MR. MURPHY:
Ralph may be able --
23 MR. MEYER:
Sorry.
24 MR. NOVAK:
We will try.
That's the best we can 25 do with the people we have.
As we mentioned earlier, we did k-ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
4
-4 30313.0 cox 61 j
1 not have a specific agenda and it was difficult to estimate 2
or plan on who should be here.
3 Warren, do you have any particular recollections 4
of'what 1150 said about --
5 MR. ROSSI:
Trevor, I think --
{
We tried to 6
MR. PRATT:
I guess I will volunteer.
7 use what we understood to be a NUREG-1150 in terms of trying 8
to do a sensitivity study to give us a feeling as to how 9
important this might be.
So'the idea was to go in and look 10 at the exLert voting that had been done for Surry and to use 11 the numbers that they had suggested.
I believe we tried to l
12 represent those numbers as faithfully as we could.
13 There were several experts that were polled, and, 14 in general, the feeling was, there were two questions asked:
i 15 One was, what was the probabilities of the operator to i
16 depressurize might be, and failure to depressurize of the 17 failure rate was given at, I think,
.2, with a.8 success,
)
18 and that we put into the calculation a range of induced 19 failure which ranged from a low value up to the maximum of 20
.3, I believe, given that depressurization did not occur.
21 This was again based upon a repolling-of the experts that i
22 were taken by the team that had input to that particular 23 aspect of NUREG-1150 to give their opinions of what their 24 feelings might be for this issue.
25 We asked some people who were familiar with-i
(-
4 ACE-FEDERAL REPORTERS,'INC.
202-347 3700 Nationwide Coverage 800 336-6646 l
.~
. ~ -,. -...
30313.0 cox 62 1
operator actions; we had another group and division in mind 2
that deals with operator actions and human response.
His 3
personal opinion was he thought the.8 might be a little bit 4
optimistic.
If we were looking at a range, we should think 5
in terms of a 50/50 split as being the upper range of that 6
range, with a lower value being something else.
The final 7
sensitivity study did have an uncertainty on that operator 8
action, which went from.2 to
.5, simply because if we looked 9
at the particular subset of experts that wer polled to look 10 at that one issue, this is something that was a difficult 11 thing to call at this point, and saw no reason to suggest 12 that.8 was any better than.5 at this point.
A lot more 13 work would have to be done to come up with a number for that 14 particular procedure.
15 MR. THOMPSON:
On the question of conditional 16 probability of failure of the tubes, in the event no 17 depressurization occurs, was tube degradation specifically 18 included in that?
19 MR. PRATT:
In the sense that any of these things 20 were included in the voting process.
One of the things, I 21 wasn't personally involved in the actual discussions that 22 went in to come up with those changes, but all of those l
23 issues were talked about.
Again, it boils down to subjective 24 judgment on the part of the experts as to what the range 25 might be.
So the idea was not to necessarily peak out an
/%-
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30313.0 63 cox
(
1 extreme vote, but to try to sample what the general feelings 2
of the experts were.
3 MR. THOMPSON:
I have a question that is perhaps 4
best directed to research, which is, has there been any 5
attempt to estimate the extent of degradation of tubes of 6
operating reactors tabulated against years of operation, 7
including the likelihood of small leaks?
Has that been 8
tackled?
9 MR. NOVAK:
I don't think it's so much a question 10 of research as it is a regulatory consideration. RWe do have 11 specific criteria which require plugging of steam 12 generators.
This is based on the clear understanding of 13 phenomena that challenged steam generator tube integrity.
I 14 don't know if we have the specific experts here this morning, 15 but, clearly, I would consider this to be a rather 16 well-behaved phenomena in terms of our understanding.
I 17 think our criteria for tube plugging are conservative, and I 18 think the fact -- generally speaking, industry is moving 1
19 toward processes to recover from potential failure 20 mechanisms, so as to provide a recoverable tube; so there are 21 efforts ongoing to improve the material characteristics of 22 the steam generator tube size to permit continued operation.
23 Ernie, do you have any additional comments?
24 MR. ROSSI:
No.
I don't have anything to add.
25 There have been various studies on integrity but I am not
/
i ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
~
,2
~
30313.0 cox.
64 1
familiar with them.
2 MR. NOVAK:
I was thinking of some of the 3
Westinghouse work going on that involved shot binning.
4 MR. ROSSI:
A new plant like Seabrook will benefit 5
from the latest technology of taking care of steam generator 6
tubes and what to do when you do have leaks and minimizing 7
leaks and that kind of thing.
That's clear, they will have 8
whatever the up-to-date technology would be.
9 MR. ROSSI:
I think Warren Lyon indicated that he 10 may have some comments on some of your questions on steam 11 generator tube integrity and induced rupture.
12 MR. LYON:
Let me back up a little on the operator 13 aspects of it.
The things that Trevor was referring to are 14 really Surry or more generic in nature.
As you will see this 15 afternoon, I am getting ahead of myself just a little bit, t
16 the considerations that go into the operator are very plant 17 specific.
One needs to start asking what kind of procedures 18 exist, what kind of training exists, what pressure does one 19 need to reach.
How does one assure that one stays below that 20 pressure once it's reached, what criterion are involved in 21 flagging and when to initiate that process.
22 These are all plant-specific kinds of things which 23 one would investigate, as a part of going into do we want to 24 depressurize.
Then, of course, there is the negative side.
25 What does it cost you if you decide to depressurize.
None of r
\\'
\\
ACE. FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646 r
7 r
w-r
t i
30313.0 65 cox I
this work has really been discussed in detail between the 2
licensee and the staff yet, but it is on our agenda.
3 I think, when you were getting into the 4
discussions on the tubes, your,real question is what is the 5
real word of tube condition and how is this being factored 6
into the considerations.
As you can tell from reading what I i
7 have already written, some aspects of that are -- have 8
already been flagged.
What are the real conditions on the 9
steam generator secondary side versus on the primary side, 10 for example, is specifically spelled out in what we have 11 written.
12 The licensee has started to look into some areas I
13 that we have discussed on what are the real conditions of the 14 tubes.
Again, we have got a lot of work to do there that we 15 haven't completed.
So we are kind of in an interim position, 16 but we are not forgetting those things.
17 Does that help any?
18 By the way, there is one thing that you asked way
{
19 back early that you didn't bring up yet, and that is have we 20 forgotten that accidents initiate with steam generator tube 21 rupture?
The answer is, no, we haven't.
The old' original 22 PRA stuff effectively said steam generator tube rupture is a 23 real minor contributor to core melt.
24 Now, our opinion was, hey, if we have got this 25 real minor contributor, and the kind of number I think of, if i
r
(-
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
^ ~~ ~
L
-L
'~
~
~
~
30313.0 cox 66
- s I
1 you believe a number of that size is of the_ order of 10.to 4
2 the minus 8, which we have got something like that, that's-3 3
the one I remember.
I would.have~to check it to be sure.
'If-4 you have got something of that order of magnitude, and we are 5
concerned with an initiator that is more in the 10 to the 6
minus 5 frequency in asking where does that go with regard to i=
7 steam generator tube' rupture, you look'at the biggie first, 8
and you know that that one is or is not a problem, and then 9
you can decide on the other-one.
10 In regard to the point'of the condition of'the 11 tuces and the preexisting leaks, the licensee and we have 12 discussed that some, as you will see.
We may.get into it
(
13 this afternoon.
We intend to'go further in that'line as 14
- well, i
15 MR. SHOLLY:
If I might say something, the 10 to 1
l 16 the minus 6 number for steam generator tube ruptures, as I 1
f 17 recall, is the fraction of a steam generator tube ruptures, 18 early failure, early large release.
The core melt frequency 19 for steam generator tube ruptures, I think, is up in the few j
20 times 10 to the minus 6 ballpark.
I haven't looked at the l
i 21 affidavit and report and whatever that I submitted.
i 22 MR. LYON:
You may be right on that number.
f 23 MR. SHOLLY:
I think there is a problem with 24 binning there, that you will get something on the order of 25 What Battelle calculated with TMLUSTDR source-term.
At the
(_
ACE FEDERAL REPORTERS, INC.
l 202 347 3700
- Nationwide Coverage 800-336-6646
~"*
T,
L..
- ,'. ~
~ ' ' ~ ^ ~
30313.0 67 cox 1
very least, somethinglof that magnitude, if you get a core 2
melt, even if a second valves are cycling, you will get some 3
release.
That needs to be looked at.
A few times 10 to the 4
minus 6 is a small percentage contributor to core melt 5
frequency.
But if the containment holds, if high pressure 6
melt ejection and direct containment heating goes away, that 7
might be a real bell ringer.
8 MR. LYON:
That's one we would want to take a look 9
at.
If that goes by and starts to rear its ugly head, we go 10 far enough, have we gone far enough ahead to look at it.
All 11 right.
12 MR. THOMPSON:
A question arises of propagation of
(
13 failure.
I believe you have stated that you anticipate that 14 rupture of one tube is likely to trigger rupture of 15 surrounding tubes.
16 A similar question is what preexisting hole size 17 do you need to propagate into failure of that tube by 18 altering the flow patterns and increasing the local 19 temperatures?
20 MR. LYON:
Absolutely correct.
21 MR. THOMPSON:
Do you believe it's possible to 22 calculate a range of hole sizes, preexisting hole sizes, that 23 could cross a stability threshold and promote propagation of 24 failure?
25 MR. LYON:
There are two aspects of what you raise
((-
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30313.0 cox 68 1
that I see.
One is, is the flow perturbation due to a leak 2
sufficient to cause a concentration of flow, if you will, 3
that gets that spot hot and makes it propagate.' We have 4
discussed that some.
5 MR. THOMPSON:
Right.
6 MR. LYON:
I lost my second one.
7 MR. THOMPSON:
The other point I mentioned, which 8
you have identified in your memoranda, is the propagation of 9
failure from one tube to the other.
10 I would also mention --
11 MR. LYON:
Well, they are, of course, the 12 situation is essentially can you, because you have a leak, 13 cause -- I know what the second one was, it wasn't that.
Can 14 you, because you have a leak, cause localized heating in 15 adjoining tubes and cause them to let go?
Of course, that's 16 not going to happen unless you get sufficient heating to'have 17 an initial impact at the leak itself.
If the fluid reaching 18 that initial leak spot is not going to be superhot, then-you 19 are not going to propagate.
20 That brings up really the second point that I was 21 going to raise that I forgot, and that is, is the flow rate 22 in that area sufficient to perturb the circulation occurring 23 throughout the reactant coolant system, such that you get 24
~ high-temperature fluid moving toward that location?
Of 25 course, if that happens, then things will begin to b
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
- -m W pdw Mwemes-m eq.%.
- e S
e e e$e
+.m m
m e
m e a g
g=-
e p=
1,,
30313.0 cox.
69
(~
1 propagate.
2 MR. SHOLLY:
As opposed to going through the-o 3
pourer or coolant pump seals or something like that?
4 MR. LYON:
Yes.
e 5
MR. THOMPSON:
It seems'to me in principle you_can-6 perform an analysis where you prov'ide a minimum hole size 7
which will perturb the flow to that point.
Then if you look 8
at experience with steam generators, you can see if that hole 9
size is likely in tubes.
10 MR. LYON:
No.
The licensee has performed 11 calculations in this area.
They have also looked at. flow 12 rates which are associated with reactor coolant pump seal
(
13 response, and have effectively said, hey, this isn't going to 14 cause significant flow perturbations insofar as our results 15 are concerned.
We are still looking at those areas.
16 MR. THOMPSON:
I asked you a. question' earlier, 17 which was is there any systematic documentation of the state 18 of steam generator tubes in operating 1 reactors.
As I 19 understand the response, it was that there are criteria used-J 20 for plugging or for expansion of. tubes.
But it wouldn't be 21 correct to assume that every tube is degraded to just above 22' that threshold.
That would be an' upper limit of 23 degradation.
24 MR. LYON:
Let me' answer that one this way.
25 Before we can get our review complete, we will have a ka
/\\CEJFEDERAL, REPORTERS, lNC.
202-347-3700 Nationwide Coverage 800-336-6646 i
n._; -_____
y_,
.~ ;
30313.0 cox 70 7 -
l 1
reasonable understanding, as best we can calculate it, of the 2
kinds of holes one would have to have to seriously perturb 3
the results.. One could then take those kinds of things, one 4
could take the strength of the tubes, assuming various states 5
of degradation, and how that influences the results, and then 6
crosscheck against the criteria, and if you want to take it 7
further, the probability of whether you are within or.outside 8
of those criteria.
1 9
But we will have developed an understanding of the 10 influence of those kinds of things.
11 MR. SHOLLY:
There is one other aspect to this 12 whole issue of induced tube ruptures.
Most of the 13 discussion, and I think most of the writeup you did, focuses 14 on thermal-induced ruptures.
There was some work done in 15 NUREG-844, I have heard-varying opinions about how 16 conservative or how realistic that work was, that suggests 17 that anticipated transient without scram atlas sequences and 18 without main steam break sequences.
At the time you are 19 operating, you are in a period of enhanced vulnerability for 20 tube failures during those events.
21 Those are events where, for instance, a loss of 22 feed water, where you have nothing on the secondary side, 23 more than operating pressure on the primary side, with some 24 enhanced vulnerability for some period of time you are 25 operating, to producing tube failures.
I don't know whether i,
\\_
i ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 80 4 336-6646
[
]
30313.0 cox 71
(
1 that is being brought into this whole discussion of reduced 2
tube failures at some point.
3 MR. LYON:
I' haven't specifically looked into that 4
area on the induced.
You are correct.
What I have been 5
specifically stating personally, what is' going on'under the 6
reactor coolant systems and damage, and what is being done to 7
the pressure boundary.
I haven't been pursuing the atlas and 8
what kind of pressure differentials you can get across the 9
tubes.
10 MR. THOMPSON:
Let's imagine your tubes are i
11 degraded if you have a loss of feed water atlas with degraded 12 tubes; you might well spring some holes which might then 13 propagate after the core melt has occurred.
14 So it's a potentially significant train of 15 events.
16 MR. NOVAK:
I think the question is as you develop 17 the scenario, what is the probability of that scenario.
18 MR. SHOLLY:
Atlas didn't come out horribly low, 19 10 to the minus 6 range somewhere.
20 MR. NOVAK:
If you connect it now with a failure i
21 to recover in any sense, in other words, then you go to a l
22 core melt, that adds another number that you have to look at j
23 the equation.
I think you have to look at the challenge rate 24 of that type of scenario to then decide.
Is it reasonably 25 likely.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
.,. ' ~
': ~ ~
^ '
., :T
30313.0 cox 72
.1 The other thing, I am not-sure.
Let me go back 2
and ask Warren.
There was a question, if you had a 3
preexisting leak in a steam generator tube, you' ould do a 4
calculation to see what size it would be to be in order to 5
influence a flow pattern so as to cause other tubes to 6
overheat.
Was your question that you would then look to see 7
if the technical specifications.would permit operation with a 8
leak of that size?
In other words, is that the way you were 9
going to attack the problem?
10 MR. LYON:
We already know what kind of leak rates i
11 the tech specs would permit.
It's only a few GPM at most, 12 and you start shutting down.
I 13 MR. NOVAK:
It's not a few GPM.
It's a fraction 14 of a GPM.
15 MR. LYON:
That depends on where it is.
In the 16 tubes, it's a relatively restricted number, yes.
We are 17 talking not very much.
18 MR. NOVAK:
Right.
)
19 MR. LYON:
Then one can look at the flow rates 20 which occur within the steam generator, under these kinds of' l
21 conditions, the severe accident conditions.
22 One can, in the calculation methodology, include a 23 range of leak rates and see if that does anything to those 24 flow rates and does anything to the thermal responses, and 25 one can begin to generate a picture of the influence of leak i
l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 33MM6
'~~~~ ~T' 1:1 l ".~ ~ ~ ~ _ l'f, - iT.f ' '
30313.0 i
cox 73
,n 1
rate on what happens to the tubes during a severe accident,.
-2 if anything.
Now, one of the points that I~made in my report-3 e
4 was before one enters into all of this, one needs to 5-understand the weaknesses of the analysis techniques that are 6
being applied, so that you understand exactly what it is you 7
are propagating.
8 So that'you then understand weaknesses in the 9
conclusions.. We haven't really completed that process.
10 We have questions in regard to the basic modeling, 11 which, in my opinion, need to.be addressed before we can then 1
12 say, okay, now we are really ready to look at the influence 13 of preexisting leaks.
14 MR. NOVAK:
I wasn't trying to use your time.
I 15 thought perhaps my comment might clarify a point of i
16 discussion.
Why don't we just move on.
17 MR. THOMPSON:
It did.
18 MR. SHOLLY:
The other aspect though, preexisting 19 leaks is another area that needs to be looked at.
I realize 20 tech specs require plugging at, what, 40 percent degradation, 21 which is also not unheard of to get degradation considerably 22 beyond that, 70, 80 percent.
If it does happen.
There is 1
23 some probability that that is going to-happen.
If you get a i
4 24 very severe wall thinning or cracks at that depth, probably l
4 25 needs to be looked at.
You have a tube that is much weaker
(-
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 6646 J.T.!!.~.T '
' ' ZT "217 " ~TEFl-I Z fl.,.-
~1
~ ~. '
i
~
30313.0 cox 74 1
than it was.
This is one of.the things that came up in 2
NUREG-884 with atlas and steam breaks.
You have cracking or 3
thinning to a point where you have only 20 percent or 50 4
percent of the original tube thickness available, and we are s
5 much more vulnerable to tube rupture than you would be with 6
tech spec limits.
7 MR. THOMPSON:
Now, the next question on tube 8
rupture is by-product heating:
The focus in your memorandum 9
is not on fission heating but on convective circulation.
10 Even if you have a hoop seal in the pump inlet, you may 11 nevertheless be passing some fission products and perhaps j
12 getting selective in those tubes, particularly in the case of c'
\\
13 a pump seal loca.
I was wondering if you would have any 14 position on that?
15 MR. LYON:
Definitely.
As you probably know, I 16 have been concerned about the influence of migrating heat 17 sources for several years.
The licensee calculation scheme, 18 as I understand it, includes those.
My understanding is they 19 don't believe there will be enough migration of fission 20 products into the tube region so that that has a significant 21 influence.
I haven't gone into the modeling in detail there, 22 so I have not personally reviewed it, but it's not 23 forgotten.
My opinion is that near the core you can get 24 substantial migrations in substantial local heating.
I 25 haven't addressed, in detail, as I said by review, as to how b.
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646
30313.0 cox 75
[
1 much of that migration can get out into the steam generator 2
regions.
3 MR. SHOLLY:
There is some suggestion in some of 4
the source term calculations that Battelle has done with 5
Westinghouse plans, that fair fractions of fission product 6
inventory, of cesium and iron, can wind up somewhere between 7
the hot steam leg and the steam generator.
Where, exactly, 8
that is coming out, I don't know.
I talked to Rich Denning a 9
little bit about this.
I am not sure that their modeling 10 that finally noted that they can tell you, but at least the 11 suggestion of that work, there is a potential there.
Whether 12 any of the more detailed thermohydraulic codes or any of the 13 more advanced core modeling codes are going to be of any help 14 there, I don't know.
15 MR. LYON:
Your point is well taken.
Our 16 calculation capability methodology in these areas is still 17 under development.
We have to include these kinds of -
18 considerations in any considerations we reached.
Just how 19 well are we going to be able to understand what is going on 20 is the basic question.
We have to include, as you started to 21 allude to, the perturbations or changes which can occur if 22 you are getting net flow through the steam generators, as 23 might occur if you had a pump seal loca or you had a loca 24 initiator in the vicinity of the pump, which would have the 25 potential of perhaps being a little larger.
As you will find i
(_'
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coterage 800-336-6646
30313.0 cox 76 I~
1 out this afternoon, we will be getting into some of the 2
things that potentially could cause that to happen, which one-3 would need to look into if the RCS is going to be at high 4
pressure.
5 MR. THOMPSON:
The thing with the loca initiators 6
is you.possibly also have feed water source.
7 MR. LYON:
That's correct.
When I said loca 8
initiator, the one I was referring to is in conjunction with 9
loss of secondary water.
10 MR. SHOLLY:
Station blackout?
11 MR. LYON:
Principally.
If you have water on the 12 second side, these are basically no, never minds, as far as 13 overheating the tubes.
14 MR. THOMPSON:
If there is a net transfer of 15 fission products through the tubes, the deposition might be 16 highly more uniform so that a small quantity of fissio.n 17 products might be significant, j
18 MR. LYON:
You don't need half of the core up in 19 the steam generators to get yourself in trouble, you are 20 right.
A localized area would be the one to look for.
~
21 MR. THOMPSON:
Stuck open, secondary side relief a
22 valves.
If the secondary side is drying out a recycling of 23 the relief valves, what does this say about the reliability 24 of stuck open valves.
25 Also, if there is some leakage from the primary
(
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80 4 336-6646
- EENe e
m a
g eh a
+=Wm
$i a*
4
- 4at e p
w
- g
30313.0 cox 77 i
i side to the secondary side vented through the relief valves, 2
what insult does that give to the valve and-what effect does 3
that have on the probability of sticking open.
Since the 4
Battelle analysis of source term with a tube rupture assumed 5
that the valves would close, but those valves are suffering 6
an insult greater than is intended.
7 MR. LYON:
I am no expert on relief valve 8
behavior.
When they are passing things for which they are 9
designed are chattering or bouncing around, you would want to 10 talk with someone else.
The way I have been visualizing the l
11 valves under core damage conditions is -- I think the word 12
" insult" is a very good one.
I haven't been using that 13 word.
I will have to remember that.
14 But if one is talking about passing fluid at 1000 15 PSI or so, at temperatures that have been already been 16 significantly high that we have caused failures in the tubes, 17 so we are talking 1000, 1500 Fahrenheit general range, I 18 personally would have difficulty believing that those valves 19 could be relied upon to close.
That is the basic assumption 20 that I have been following in the pursuit of these things.
21 Assume they are going to fall and see where it 22 leads you.
Then one could go back if need be and address 23 that.
24 MR. BAGCHI:
The staff has the ability to review 25 the safety of safety valves, pressurized relief valves.
b ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646 2..
i l
30313.0 cox 78 1
The relief valve that you have on the secondary 2
side are not dissimilar.
Tests do indicate that the relief 3
valves do not traditionally indicate a problem with 4
receding.
5 MR. THOMPSON:
Even after repeated cycling?
6 MR. BAGCHI:
Chattering, more likely, of cycling.
7 Chattering does damage the seal, so 1* there is a separate 8
and distinctive event after that, there may be some 9
problems.
10 MR. THOMPSON:
Often in these scenarios you have a 1
11 period of steam-generated dryout preceding core melt, so you 12 have had some period of minutes of cycling or chattering or 13 whatever.
Then, if you get a tube failure, you subject these 14 already damaged valves to further insult.
15 MR.' BAGCHI:
I don't think it's a receding 16 problem.
The problem you have there is a leakage problem.
17 It's not a stuck open valve.
They are spring-loaded.
It 18 would be a matter of degree.
19 MR. THOMPSON:
Could I have the citation on the 20 pressurizer valves?
21 MR. BAGCHI:
TMI action item II.D.1, one can find 22 that in NUREG-0737.
23 MR. SHOLLY:
Are you referring to the EPRI tests?
24 MR. BAGCHI:
Yes.
25 MR. SHOLLY:
It would be interesting to look at k
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 336-6646
i 30313.0 cox 79
('
1 the conditions of those tests as opposed to what you did in 2
your --
3 MR. BAGCHI: -We are not talking about 1500 PSI 4
fluid -- 1500 degree Fahrenheit fluid.
t
~
5 MR. SHOLLY:
Passing hydrogen in your. fission 6
products as well,-perhaps?
7 MR. BAGCHI:
The point was raised about a stuck 8
open valve.
I don't think it will be stuck open.
There 9
would be a degree of leakage.
It's just a personal opinion.
10 MR. LYON:
There is experience that says on 11 occasion they will fail open.
Davis-Besse, of about three 12 years ago, was an example of that in the field, but something 13 like that is a bit of a fluke and is not a failure mechanism 14 which would be induced by a severe accident condition.
15 MR. SHOLLY:
It's a problem that Babcock & Wilcox 16 plants seem to have.
I would like to come back to the Event 17 V question I had a little earlier.
Obviously the principal 18 focus of concern there is the paths the fission products 19 take, and whether or not the failure point is underwater, you 20 are going to get fission product scrubbing or whatever.
But 21 regardless of whether or not that point is underwater or not, j
22 you are going to have hydrogen release into the RHR pump 23 vault and presumably later, somewhere else into the auxiliary 24 building.
25 I wonder if Brookhaven or if the staff has looked h
ACE. FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-33 & 66 4
30313.0 cox 80 s
-1 at that particular matter and come to any conclusions as.to 2
what happens.
3 MR. PRATT:
I can't speak for the staff.
I know 4
as part of our evaluation, specifically for Seabrook, we i
5 didn't include that consideration.
6 MR. SHOLLY:
Have you looked at it for any other 7
plants?
8 MR. PRATT:
For my divisional part, no, I don't 9
believe so.
10 MR. SHOLLY:
That you know of?
11 MR. PRATT:
That I know of.
12 11R. LYON:
May I ask what your concern is there i
13 basically?
14 MR. SHOLLY:
For Event V scenarios, where the 15 release point is not submerged, there have been analyses 16 done, N Core has done some work, that suggests there would be 17 a lot of continuation of that source term in the auxiliary 18 building.
I am wondering if those analyses are based on the 19 nornal structural integrity and residence time of gases and 20 fission products as they pass through a tortuous path, or 21 whether there is a potential there for a large hydrogen 22 deflagration to blow the thing apart basically and open up 23 the large hole.
24 MR. LYON:
Your concern is not with the submerged 25 releas-s, which our information says is the majority of them, k.
I 1
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800436-6646
30313.0 1
cox 81 i
l 1
it's the potential for a type break which exists above the 2
water down at the bottom of the vaults.
3 Trevor, can you address the assumptions that were 4
made on the Event V releases above the -- that did not have 5
scrubbing and what kind of binnings you put that into?
6 MR. PRATT:
Makes no difference.
7 MR. LYON:
Makes no difference as far as your 8
calculations are concerned?
9 MR. PRATT:
For those sequences.
10 MR. SHOLLY:
I understand what you have done 11 there.
It's a conservative binning project.
I am trying to 12 look at a point if this thing progresses much more, we might 13 get in a shooting match between the source ters package and 14 MACCS.
I am wondering where this issue will impact.
15 MR. PRATT:
This issue is being focused in very 16 well, I think, in the comparisons between the MACCS and the N 17 Core package, specifically BWR, mach 1 and mach 2.
There you 18 have the primary containment completely surrounded by the 19 reactor building.
The question is how much retention does 20 one take for fission products in the reactor building between 21 the two codes.
If you look at the calculations with MACCS, 22 you will find very significant retention, very high 23 contamination.
24 If you look at the calculations with the source 25 term code package, you will find much less.
There is a
(
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
30313.0 cox 82
(~
1 reason for the number of differences; one is the size of the 2
hole in containment and the throughput of the fission 3
products in the reactor building.
Another one i the 4
possibility of hydrogen burns which is modeling in the source 5
term code package, which produce large pressure pulses which 6
tend to sweep the fission pulses out.
Whereas in the MACCS 7
calculations you gets a more gradual burning of the hydrogen j
8 in the flame and much less gradual expulsion of gases.
9 Again, we are talking about maybe a factor of 4 and 10 decontamination prepared with much larger numbers.
11 You should remember, though, that those 12 calculations come at the end of calculations where you take 13 into account retention, which of course the source term 14 calculations that we are using in S-1-W now does not 15 calculate that.
This is a thing of very great importance, I 16 think, to longer-term analyses where you are dealing with the 17 revaporization of the iodine, which would probably be 18 retained in the primary system in the source term 19 calculations; its reemission, which in the N Core analysis, a 20 large quantity of that cesium iodine is released.
21 Again, most of it is retained in the reactor 22 building in their' calculations.
23 on the source term package code calculations, that 24 element is not in there, we do not calculate this point of 25 revaporization.
However, it was taken into account in the
/
v ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 6646
_i_-_-.--- -
n y
i yy_
30313.0 cox 83 i
1 uncertainty ranges in NUREG-1150 by polling the expert 2
opinion and calculating that factor.
-So you see that effect 3
in your ranges.
Yet I am not sure of the applicability to 4
what we are talking about today, because the binning that we 5
do have now, I think, is fairly conservative.
6 MR. SHOLLY:
On that basis, I will let it go.
But 7
I understand what you mean.
8 MR. THOMPSON:
I have some questions on 9
consequence calculations.
I noticed that Brookhaven has used 10
- MACCS, M-A-C-C-S.
4 11 But they don't appear to have used it for 12 inhalation; am I correct on that?
i 13 MR. PRATT:
I should have brought my expert.
14 MR. LONG:
I may be able to say something.
When 15 you say we haven't used it for inhalation.
16 MR. THOMPSON:
200-gram dose in the Brookhaven 17 report, which apparently is red marrow dose.
18 MR. LONG:
That code has a hazard function with it 19 that I think has a standard setup, Trevor, and the guy who 20 ran the code did provide in there a probability of death to 21 an individual standing on the plume center line.
I believe 22 that has a standard setup for all of the doses to the various 23 organs.
So I believe that includes the lung dose.
We can 24 check on that.
What he did for the 200-rem curve and so on 25 was to go on and create a hazard function that was a step (m.
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646 1
' ~ ~ ~
30313'.0 cox 84 e
i function at 200 rem, so that you would be able to count the 2
probability of exceeding 200 rem.
He used the red marrow 3
dose because that is a reasonable approximation'of the 4
geometry of the whole body.
5 Sc what he was doing there was to try to match the 6
0396 curves.
7 MR. THOMPSON:
NUREG-1150 has identified lung 8
doses of potentially significant issue.
9 MR. LONG:
Right.
10 MR. THOMPSON:
It seems to me, therefore,'it ought 11 to be considered in the Seabrook context.
12 MR. LONG:
What I am saying is I think one of the
/
13 vehicles for considering it would be to look at the 14 alternative curve in the Brookhaven report.
That's the 15 individual's probability of death.
We can check, but I 16 believe that the normal sweep of doses that MACCS calculates i
i 17 is included in that particular curve and somewhat different 4
18 from the 200-rem curve.
19 MR. THOMPSON:
Another question about MACCS, l
20 NUREG-1150 identifies plume liftoff behavior as being treated 21 differently in MACCS than the other two.
22 I was wondering if any differences were identified 23 between MACCS and CRACIT, C-R-A-C-I-T.
24 MR. PRATT:
If you look at the calculation that 25 was presented in his section, you can see the difference in
' k ace-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide coverase 800336646
30313.0 cox 85
(-.
1 the probability, OEN, on the calculation.
That's largely due 2
to the fact that the S-1-W has a high energy release and showsthelowerprobabilityofthefatalitiesc1bsein.
3 1
4 MR. THOMPSON:
Right.
I don't remember seeing in 5
your report a specific discussion of the role of plume 6
liftoff in causing these differences.
7 MR. PRATT I think in the latest report, you may 8
not have seen it in a draft report, but I think in the latest 9
one we did have a qualifier in the latest one that we were 10 concerned about this effect for the S-1-W release.
One of 11 the differences between our draft report and final report is 12 in our draft report we were dealing with total numbers of 13 fatalities and trying to infer how that would impact our 14 comparisons with the safety guard.
15 We then went in the final report to looking at the 16 region within one mile of the site.
We found very little 17 effect.
i 18 Now, if I can refer you to table -- the table 19 appears all over the report, but the first place it appears 20 is table 2, you can see for the S-1-W release, these are j
21 calculations that were provided by the applicant, they are 22 not our calculations.
We used the effects to look at changes 23 in frequency.
This is on page -- this is Roman numeral, what 24 is that, within one mile there is 796 -- this is XXDR-30.
25 MR. SHOLLY:
Figure S-4 shows you the same thing k_
l ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6646
~L :
1 A
i 30313.0 cox 86 i.
(~
j 1
graphically with less distance.
2 MR. PRATT Yes.-
That was'the point I was 3
making.
4 The question that we had, the qualifier-that we 5
have now put into the report, is we run MACCS for this i
6 release, we would expect to see a high fatality risk closer 7
in.
Therefore we would expect to see more impact-on our j
8 uncertainty calculations or sensitivity study calculations j
9 than we have seen by using the calculations that were
)
10 provided on the applicant.
We did not have this information l
i 11 because we did not have population distribution, s o TMt i
12 couldn't get a table.
Our decision was made to use.the 13 applicant submittal for CRACIT to show.the effect of our I
14 sensitivity studies in the section 6, to do some comparisons i
i 15 and try to identify what differences might apply if we had 16 used MACCS consistently throughout.
17 MR. THOMPSON:
Just going back to table S-2 that 18 you have said is an analysis of the applicant, this is for i.
19 some average population, does not include beach populations?
20 MR. PRATT I think that would be better addressed 21 to the applicant.
22 MR. THOMPSON:
The next question is whether you 23 feel any of these codes, CRACIT, MACCS or anything else, 24 provides an accurate estimate of doses at a coastal site such i
i 25 as Seabrook?
(
1 a
l ACE-FEDERAL REPORTERS, INC.
]
202 347-3700 Nationwide Coverase 800-3 4 6646 L
- 22:J-.
- ,.- - - L *.. L'::2: ~ ' :_T: 2.-
J
)
30313.0 cox 87 1
MR. PRATT:
Again, CRACIT, as you know, is a 2
different model than the one we were using.
It does have a 3
plume meandering, modeling in there, which can take into 4
account the terrain.
The way we run MACCS, although the 5
intent is in the final version to have that capability, the 6
version we were using did not.
So we are still talking about 7
one direction of movement once you start the calculation.
8 You can have multiple puff releases, but nevertheless the 9
wind direction stays the way it is and so on.
So the option 10 is not there yet.
We are not as sophisticated as CRACIT in 11 its calculations.
So I would tend to think the CRACIT 12 calculations for center line dose are pretty conservative.
13 You can get variations depending on weather.
14 MR. SHOLLY:
Do I understand you to say that the 15 version of MACCS that you use for calculations would, in the 16 case of a multiple puff release where you might get 4, 6, 19 17 hours1.967593e-4 days <br />0.00472 hours <br />2.810847e-5 weeks <br />6.4685e-6 months <br />, would run that out through one plume, one compass 18 sector, or would it move it around if the wind direction 19 changes?
20 MR. PRATT:
I would have to defer that and give 21 you an answer in writing.
I could surmise --
l 22 MR. SHOLLY:
It's important.
23 MR. PRATT:
I appreciate it's important, but it's 24 not the full capability of MACCS that we had hoped to be 25 available to us at this time.
(s ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33M646
30313.0 cox 88 1
MR. LONG:
I will take a hazard and subject to 2
check with Art Kunkle, but we did'a sensitivity study, which 3
I think you saw in the draft report, lookingatbhatcauseda 4
plume to be narrower or wider.
There is at least a plume 5
spreading factor that is included with time.
6 MR. SHOLLY:
Same thing as CRACIT and CRACIT 2, 7
MR. LONG:
Right.
I think they did, but we will 8
have to check and see if they did in this run.
We were using l
9 the actual data bank to run MACCS.
10 MR. PRATT The main question that you are 11 bringing up, in the CRACIT code, you were allowed to put in 12 road networks, and you can look at a plume and where the 13 plume moved.
My understanding that capability does not exist 14 in the version of MACCS as we used it.
15 MR. LONG:
You had indicated a concern about 16 whether or not the models accurately represented a coastal 17 environment.
What is your concern for accuracy, are there --
18 MR. THOMPSON:
It's an phenomenological concern.
19 Fumigation is a known phenomenon for coastal sites.
I still 20 believe it's within the FSAR.
But there is also a phenomenon 21 that has been observed of a helical circulation traveling i
22 along the coast.
I believe that is not accounted for in 23 CRACIT or otherwise by the applicant.
I wonder whether 24 anyone thinks that is potentially significant?
25 MR. LONG:
I don't know of any model that accounts JACE FEDERAL REPORTERS, lNC.
202 347 3700 Nationwjde Coverage 800 336 6646
30313.0 cox 89
_m 1
for that as a dose prediction.
You have to account for the 2
additional dispersion as a function of equidistant travel.
3 You have to get a different direction of the plant as well.
4 I don't know exactly who will be available to evaluate that.
5 In the scheme of things, I don't see where that is 6
necessarily a large factor, unless you would like to argue a 7
population distribution with respect to the wind direction or 8
something.
9 MR. THOMPSON:
If there is a mechanism causing 10 selective concentration along the coastline, and I don't 11 claim right now that that is true or otherwise.
That, if 12 true, is relevant to the fact that the population in summer 13 is similarly distributed along the coastline.
14 MR. LONG:
My point is that the helical pattern 15 comes from having differential heated surfaces and a layer in 16 which you have a land series phenomenon above that is 17 traveling with a continental air mass flow.
Plume is 18 released in one area, travels to the other, lifts, goes up, 19 sees a countervailing wind direction higher on, cools, comes 20 back down and continues.
If the wind patterns are not 21 directly 180 degrees offset, you have a pattern, at the two 22 different elevations.
You have a fair large traveling path 23 from the rising as well as the sinking aspect.
In terms of 24 calculating a point on the ground from the plant to the first 25 time this comes down and the second time it comes down, the k
ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-M46
30313.0 90 cox
,-s 1
actual trajectory is a lot different in terms of its 2
dispersion characteristics, than you would model in_the gas 3
plume case.
4 I guess what I an asking for is do you really have 5
a reason to believe that the direction along the coast may be 6
so significant that it would overcome the additional 7
dispersion?
8 MR. THOMPSON:
In a limited review of the 9
literature on this, I have encountered a discussion of air 10 pollution being selectively concentrated in cities or towns 11 north of Chicago, presumably due to this effect.
In certain 12 conditions, the air concentrations of pollutants were much 13 higher over a period of hours-than you would have predicted 14 from gas dispersion.
15 MR. LONG:
You are talking about area sources 16 there, I believe, right?
17 MR. THOMPSON:
Right.
18 MR. LONG:
Do you have a reference?
19 MR. THOMPSON:
I can get it to you.
i 20 MR. LONG:
Okay.
It's hard to envision how you 21 would apply that to this case.
That's the reason I am 22 looking for something further.
23 MR. SHOLLY:
One other aspect of the consequence 24 modeling end of this.
I am interested in finding out if you 25 have gotten from Brookhaven, from MACCS rems, person rem ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336 6646
30313.0 91 cox 1
comparison for different assumptions, for instance, no 2
protective actions, one-mile, two-mile, even a 10-mile case.
3 MR. LONG:
I haven't gotten them.
Wi have to run 4
them.
5 MR. PRATT:
We don't have a population.
6 MR. SHOLLY:
Of course.
Would there be a problem 7
-- I don't know what the requirements are for MACCS 8
population distribution.
Couldn't you set some data from 9
10 MR. PRATT No.
It's just that we didn't do it 11 for this particular piece of work.
We focused on the dose i
12 calculation and went.with that risk measure.
There is
}
13 nothing wrong with -- there is no difficulty.
1 14 MR. THOMPSON:
Anything like that that needs to be 15 done for Seabrook needs to discriminate seasonally because of 16 the summer beach population.
17 MR. LONG:
The two tools that are available have 18 different strengths and weaknesses.
The CRACIT code is much 19 better at looking at evacuation phenomena.
The MACCS code 20 has a more recent dose response function to it.
It would be 21 nice if somewhere somebody would put these two together for 22 us, but I don't think we are going to see that during the 23 course of this review.
The reason for looking at CRACIT was 24 really to look at the evacuation assumptions.
We have to plan how we will evaluate the other factors for the tools we 25 k-ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646
l 30313.0 cox 92 s
1 have available.
We haven't completed mapping that out yet.
2 MR. SHOLLY:
Emergency response does not have an 3
overall impact on population dose in the CRACIT; I am 4
wondering if any differences emerged in the lung models, if S
that starts to accumulate a large population dose, within a 6
10 mile area or not.
I sort of expect it doesn't, but I 7
would like to have that suspicion confirmed.
8 MR. LONG:
Your suspicion is the reason we haven't 9
looked at it with high priority, really.
10 MR. THOMPSON:
During the staff's review, there is 11 going to be a number of further consequence calculations.
12 MR. ROSSI:
I guess the consequence calculations I
13 the staff would be doing would be directed to do at what we 14 need to do to evaluation the petition.
If it didn't bear on 15 evaluation of the petition, we wouldn't feel we would need to I
16 do it, so I suppose it bears on the evaluation of the 17 petition.
Is that a fair way to state it from what you 18 know?
19 MR. LONG:
Trevor, not having gotten a contract 20 from us yet, is not in a good position to answer the 21 question.
22 MR. ROSSI:
Well, from the aspect he has been 23 working on, the evaluation process of the petition.
I just 24 want to make this statement, the work we do in reviewing the 25 Seabrook submittal will be directed towards what is required i(
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-33 & 6646
30313.0 cox 93
(,
1 to view the petition and what comparisons you need to do 2
between Seabrook and whatever the basis for the' -- what was 3
done in NUREG-0396 on the basis of 10 miles.
In listening to 4
all of your conversation, I get a little bit concerned that.
5 we wouldn't want to use new knowledge in only the areas which 6
would tend to make things worse without also taking account 7
of new knowledge and other areas which mainly are an 8
improvement, because you could get caught into a situation 9
where you go and use things that have been found in the last 10 several years that preferentially could make things worse, or 11 also include a way to make things work better.
You wouldn't 12 want to do either one of those.
13 MR. SHOLLY:
I.have a general feeling on that.
If 14 there is something that will make things better, the 15 applicant will direct his energies toward pointing that out, 16 and to the extent our efforts are limited by budget, it will 17 tend to focus in on those areas, might make things worse, and 18 you being in the middle will sort that out.
19 MR. ROSSI:
From what you do that might be 20 appropriate, but from the NUREG standpoint, we want to do it 21 as objectively as we can, 22 MR. SHOLLY:
I understand.
23 MR. THOMPSON:
Documentation of MACCS.
Does 24 anyone have any idea when that will be published?
25 MR. MURPHY:
I think it's due out in about two s
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336 6646
30313.0 cox 94 1
months.
I can verify that date for you.
2 MR. SHOLLY:
Dan Alpert is working on that or 3
someone else?
4 MR. MURPHY:
Dan has moved to solar power.
e 5
MR. SHOLLY:
Three people have left, what is going 6
on now?
7 MR. MURPHY:
The reason it has been delayed'is 8
because the principal investigator is now working on solar 9
cells.
10 MR. THOMPSON:
I am not sure we have discussed the 11 sabotage question, other than Steve mentioning it earlier.
12 Just for the record of the transcript, I would like to say 13 that our position is that we believe there are viable s
14 sabotage scenarios which can lead to large early release, but 15 we don't think it's appropriate to pursue those publicly, and 16 that some means of closed procedures has to be found to deal 17 with them, but we do think they are very much pertinent to 18 the application.
19 MR. NOVAK:
I think, for the record, we will 20 review that comment and see what can be done to coordinate 21 it.
It would seem to me that this.would probably be a 22 meeting where we would have the applicant, the State of 23 Massachusetts, representative, and the staff together, to try 24 to consolidate what can be said and basically have it on one 25 record.
We will probably have to work something out in that ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationmde Coverage 800 336-6646
30313.0 95
,,-cox l
i 1
area.
2 MR. SHOLLY:
I-think the remainder of the issues, 3
there is not a whole lot that can be said, other than it 4
would be a good idea to check it specifically for Seabrook.
5 There is sort of mixed evidence.
Whether or not you can get i
6 a pressure that will fall containment or not.
I don't know 7
what we can do here other than beat.it to death in a minute I
8 or two.
I think we have hit the principal areas that are j
9 driving what the profile of risk is going to look-like for 10 the plant.
I hope you have a little bit better idea of where 11 we see those issues lying and how things are going to sort 12 out over the next couple of months.
7 t
13 MR. BARRETT:
I would like to ask you a question 14 about RCS depressurization.
I got the impression, while you j
15 were discussing direct containment heating, while you were 16 discussing the induced steam generator tube rupture, that 17 this is an option that you generally favor; is.that a correct l
18 impression?
19 MR. SHOLLY:
My only personal opinion now is it is 1
20 something worth looking at.
I am not persuaded yet that 21 simply using the polars is going to get you down to a low
.i 22 enough pressure that ejection and heat containment is going
~23 to go away.
If someone has some thermohydraulic calculations l
24 that shows that fairly conclusively, I would like to see 25 them, because that will make our job simpler over the next k_
I ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 33H646 L :T ::' '
f,J.~
30313.0
)
cox 96 l'
i 1
couple of months as well as everybody else.
But my 2
impression is the polars aren't going to provide you enough 3
depressurization to get you there.
Maybe one wants to 4
consider a backfit of some sort, perhaps, that would get you 5
there on depressurization.
It's a touch and go subject, God i
j 6
knows.
It's one there are ups and downs to, and something i
7 that would have to be looked at very, very carefully before 4
4 8
you would ever want to consider it.
It's sos.ething that is 1
9 clear enough conceptually that it is worth raising.
10 MR. THOMPSON:
Then, of course, you have to think 11 about steam explosion.
It's quite clear that there is no 12 analytic basis for describing a conditional probability to
)
13 steam explosion, one is relying on expert judgment,-the
~
14 range of that expert judgment for the steam explosion review 15 group ranging up to conditional probabilities of 10 to the l
16 minus 2.
If that's the best you can do, an important public-17 policy issue is whether you want to weaken emergency i
18 planning, emergency response regulations, where a phenomenon i
19 of this importance is not well defined.
i.
20 MR. BARRETT:
Are you talking about steam i
21 explosions, the so-called alpha failure mode?
l 22 MR. THOMFSON:
Right.
There is a further i
23 complication in that if the sequence starts out at high l
24 pressure, and there is some heating of.the upper vessel 25 region, the kinetic energy needed to generate -- to turn the k
ACE FEDERAL REPORTERS, INC.-
l 202-347-3700 Nationwide Coversse 800 3 % 6646
= wome ese N pve me,.>-,
e+
g e e e' * *
.s *= w %+ e.
e
--*i-=*-- +. - -
- * = =
,a e
4-
bS A
-4' e
n m
30313.0 cox 97 1
vessel head into a missile may be reduced.
2 So you have to be careful with depressurization thatyouarenottradingoffonesetofphenomenkfor 3
4 another.
i-
- 5 MC. BARRETT:
I am missing something here ---
6 MR. THOMPSON:
If steam explosion is more likely 7
at the low pressure than at high pressure, then by 8
depressurizing you are putting yourself into that regime 9
where steam explosion is more likely.
10 In choosing whether or not to weaken the emergency j
11 response regulations, you are relying upon expert, judgment as 12 to the conditional probability of the alpha failure modes.
13 That's an important policy issue, and many members of the
)
14 public might decide that they wanted to preserve emergency 15 response regulations absent any firm analytic basis for 16 probability.
17 MR. LONG:
What is the argument for getting a 18 higher probability of steam explosion at low RCS pressure?
19 MR. THOMPSON:
Well, again, a judgment of thel j
20 experts is that triggering is less likely at higher
~
21 pressure.
I' don't have a personal opinion on that, but it's 22 the consensus of the experts.
23 MR. LONG:
Basically a vaporization difference or 24 what?
25 MR. THOMPSON:
I presume.
(c ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 900 336 6646
.m..,,._..,._
-.7. -
30313.0 98 cox 1
MR. BARRETT:
I think that's very complex, it's 2
more complex than that.
3 MR. LONG:
What I was looking for is something 4
that they were -- is this an accepted?
5 MR. SHOLLY:
Just the example of a risk 6
trade-off.
If you do pressurize and can get rid of high 7
pressure melt ejection, you want to make sure you are not 8
trading it for something equally nasty or worse.
9 MR. PRATT:
Again, I haven't been following the 10 latest things from Marshall Berman's program in Sandia on 11 when explosions happen and when they don't.
I know he was 12 having trouble in the earlier days in triggering explosions 13 at the higher pressure.
It's easier to do it at lower.
But 14 given his high ingenuity, I believe he was able to do it at 13 higher explosions, and it's only a matter of time.
16 MR. LONG:
But there is material available?
17 MR. PRATT:
To have a lower probability than a 18 higher --
19 MR. I,0NG:
We do have some empirical material 20 available?
21 MR. PRATT:
Yes.
This program has been going on 22 for years, many, many years.
23 MR. BARRETT:
Let me get back to my original 24 question.
Go ahead.
25 MR. SHOLLY:
One other aspect to it, assuming for i_-
ace-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33M646
.____7______.__,.
30313.0 99 cox-(--
s 1
a moment the high pressure melt ejection is a big problem and 2
provides you with a containment load that can fail the 1
3 containment with some moderately large probability, so that 4
it dominates the early failure mode, if you start looking at 5
how to deal with that in a mitigative sense, depressurizing-6 the reactor coolant system would seem to be an easier way to 7
go, and perhaps even a cheaper way to go, than trying 8
something like a large high-volume filter vent.
I can see 9
where if you are trying to lop off the upper end of a 10 pressure spike on high pressure melt ejection, you would need 11 a big containment area to do that.
To qualify for seismic 12 events, you are starting to look at a lot of dollars, a 13 research program and some considerable lead time for 14 construction.
15 Whereas if you can add a valve or two, and some 16 procedures in training to accomplish the same objective, that 17 would be a much more viable way to go.
18 As I say, it's strictly conceptual at this point.
19 A lot of competing factors to consider.
20 MR. BARRETT:
I would like to raise one more 21 question.
That is, I don't exactly recall, early on in the 22 meeting, we had some discussion about the application of the 23 state of the art methodology.
I think what you meant by that 24 was the NUREG-1150 methodology on a plant-specific basis to 25 Seabrook, with regard to direct containment heating.
I think 4
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80 4 336-6646
30313.0 cox 100 1
there was a reference here to the Zion, Zion results being i 2
to 4 percent of containment failure probability.
I would 3
like to get'back to'that.
Is this something that you believe 4
is a productive avenue to pursue, or was I perha,ps misreading S
your intention there?
6 MR. SHOLLY:
This is out of the Brookhaven, 7
NUREG-4551, volume 5, February '87 draft, where the --
8 looking at the probability distribution for early failure, 9
the median came out around 1 percent, mean came out at 4 10 percent.
Depending on your preference for best estimate, you 11 are in that range.
It seems from my reading of it so far to 12 be driven by direct containment heating.
l 13 Without getting into a great deal of discussion j
14 about whether it's a good thing to try and do or not, if one 15 just takes that result and slaps it on Seabrook for the 16 moment, that's much higher than the early failure probability 17 predicted by PSMH's studies.
That being so, and the plants 18 not being hugely different in terms of the direct heating 19 susceptability, and forgetting for a moment the cavity 20 business, it suggests that perhaps it's worth going the extra 21 mile and doing some plant-specific look at it.
22 Perhaps it's the case that the extra 15 or 20 4
23 minutes of pressure before you get to the median failure I
24 pressure buys you a much lower number.
I don't know.
But 25 it's certainly worth looking at becauseuit has the potential k_.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
.wr 30313.0 101 cox 1
to be the driver as far as early probability failure goes.
I 2
would rather see it done on a plant-specific basis and see 3
that it was 10 to the minus 3, here it was over'51 percent 4
and probably Seabrook's in the middle.
That's real nice, 5
pretty obvious, but I don't know how far that gets you in 6
trying to decide whether or not the petition gets granted or 7
not.
8 MR. THOMPSON:
I have a comment about the 9
uncertainty in NUREG-1150.
Polling of experts, my own 10 inclination is to the Marshall Berman school of uncertainty, 11 which is rather dubious about this polling approach, and 12 believes that one should use, for example, the technique he 13 has used for steam explosion, which is to not describe 14 probability distribution functions where you don't have clear 15 empirical basis for them, and to use, for example, the low, 16 middle and upper third or range of each parameter, as a guide 17 to determine what are the important factors.
He has written 18 this very eloquently in various places.
19 MR. SHOLLY:
I must say I saw some of the earlier 20 drafts on what Sandia did with Surry, optimistic, 21 pessimistic, central.
I thought that was real instructive in 22 how the reactor safety community as a whole is reflected by 23 those choices.
Whether it got transferred to this, where you 24 get distribution of pressure, those are assigned percentages, 25 you start mushing this together in source term bins and-
-s ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 1
,4
30313.0 102 cox 1
1 source term clusters, it starts to lose the physical ground 2
and gets a heck of a lot more ephemeral.
Just a comment.
3 MR. MURPHY:
A problem we had, with the 4
optimistic, pessimistic, central approach, at one time we 5
wound up taking 29 things we believe to be conservative, 6
multiplying them with the other, and having no idea what that 7
meant in terms of a statistical standpoint.
On the other 8
hand,-29 optimistic things, and central was not defined as a 9
best estimate, which was a very difficult comment to get 10 across to people.
Central was described as something between 11 optimistic and pessimistic.
Basically at the point where 12 both the optimist and pessimist were unhappy with their 13 answer.
14 For instance, the central estimates were the 15 source term code package, we know were not in the source term 16 code package, that we believe are real.
So ycu couldn't call 17 it a best estimate.
18 Marshall has written eloquently on the subject, 19 but he also participated in the process.
The answer, I 20 think, is that expert judgment never substitutes for good 21 science.
But in the absence of the detailed models, you 22 can't ignore what you know, and you try to express the best 23 you can.
24 As an example, if I took my best estimate of 25 direct containment, the pressure associated with direct V
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
30313.0 103 cox 1
containment heating is surry.
My best estimate of the 2
containment response of Surry, I would say, the containment 3
can take the pressure.
The modes of both of them do not 4
intersect if I take the two.
I would say direct containment 5
heating is no problem at Surry if I took my best estimate.
6 But granted they are subjective in how they intersect, we r
7 come to the conclusion that direct containment heating is a 8
problem that I cannot discount at Surry, based on present 9
knowledge.
I have to do some more work.
10 Looked at in that vein, I think the subjective 11 judgment is quite valid.
12 MR. THOMPSON:
Where you can run into problems, 13 though, is the way you are dealing with intersection of tails 14 of distributions.
You are using that -- the inferences you 15 draw from that, to determine policy as in present 16 application, by, perhaps, claiming that the probability of a 17 person at a given distance acquiring a given dose.
I think 18 you are going beyond the scientific basis, 19 MR. MURPHY:
The decisionmaker has to understand o
20 very clearly the subject activity in the process, and the 21 nature of both sets of tails, where they com*e from, what they 22 are derived from.
Provided that understanding is there, I 23 don't have a trouble with the decisionmaker using that.
24 information as long as he understands what he is using.
25 MR. LONG:
In a way, we are being more v
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80 4 336-6646
+ -
30313:0 cox 104 1
conservative, in a sense that we in the past have typically 2
been using median or.means as comparisons.
Here you are takingmediansorinthiscasemeansthatleadtbone 3
4 conclusion and saying, but the credibility of that conclusion 5
is somewhat tainted, because there are other opinions out 6
there.
We are trying to sample those in some quantitative 7
way.
If you consider a credibility distribution as opposed 1
8 to a probability distribution, it's still something that we 9
would like to include in the decisionmaking process.
10 MR. SHOLLY:
That's an interesting way of looking 11 at it.
I will have to remember that.
12 MR. THOMPSON:
I would agree for certain purposes 13 it's an useful process, but I would urge caution in using it 14 for policy decisions, i
15 MR. LONG:
I think we understand your caveat.
I 16 was trying to point out as a matter of moving in the 17 direction you are advocating, to overlap.those tables in the 18 first place.
19 MR. NERSES:
Any more comments or questions?
20 MR. NOVAK:
I think it's been a very informative
~
21 morning.
I certainly want to thank you for being as candid i
22 as you can with regard to your concerns.
I think we have had 23 as open a discussion as we could, based on the material we 24 had to discuss.
I would say that with regard to any 25 subsequent meetings, we will certainly have discussions j
~
i J
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-33(4 646
-~
f w
w --,
- y
-w y
y Tr v--Y
--'T
U-30313.0 CoX 105 1
beforehand to look at the need for it, the subject matter and 2
the timing, so that that can be worked out.
3 Thank you for coming, andwewillad3burnthe 4
meeting at this time, thank you.
5 (Whereupon, at 11:38 a.m., the meeting was 6
concluded.)
e 7
8 9
10 11 12 13 i
14 15 16 17 18 19 20 21 22 23 24 25 J
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
= ~
r p
CERTIFICATE OF OFFICIAL REPORTER This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION 1n the matter of:
NAME OF PROCEEDING:
MEETING WITH NRC AND MASSACHUSETTS OFFICE OF ATTORNEY GENERAL RE:
BROOKHAVEN REPORT (SEABROOK)
DOCKET NO.:
50-443 PLACE:
BETHESDA, MARYLAND 1
(
DATE:
WEDNESDAS, MARCH-25, 1987 were held as herein appears, and that this is the original transcript thereof for the file of the. United States Nuclear Regulatory Commission.
i
( sigt) //pI (TYPED)
[
[
WENDY S. COX Official Reporter Yeh'o'N S N M 5Ndn NC*
..- -..-.~..-,. - - -.
,_q, g
g
-,,--.--y
-m C
y