ML20212R145
| ML20212R145 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 03/25/1987 |
| From: | NRC COMMISSION (OCM) |
| To: | |
| Shared Package | |
| ML20212R050 | List: |
| References | |
| NUDOCS 8704240309 | |
| Download: ML20212R145 (86) | |
Text
.
ORGEE mi1EDSTATES NUCLEAR REGULATORY COMMISSION IN THE MATTER OF: DOCKET NO: 50-443 MEETING WITH NRC AND LICENSEE (PUBLIC SERVICE COMPANY OF NEW HAMPSHIRE)
. AND REPRESENTATIVES OF BROOKHAVEN LABORATORIES RE: BROOKHAVEN REPORT (SEABROOK)
I l
(
LOCATION: BETHESDA, MARYLAND PAGES: 1 - 84 DATE: WEDNESDAY, MARCH 25, 1987
, , ACE-FEDERAL REPORTERS, INC.
( OfficialReporters 444 North Capitol Street Washington, D.C. 20001 (202)347-3700 8704240309 870406 PDR NATIONWIDE COVERAGE I' ADOCK 05000443 PDR
CR3'O313.0 COX/sjg 1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 MEETING WITH NRC AND LICENSEE 4 (PUBLIC SERVICE COMPANY OF NEW HAMPSHIRE)
AND REPRESENTATIVES OF BROOKHAVEN LABORATORY 5
RE: BROOKHAVEN REPORT (SEABROOK)
DOCKET NO. 50-443 7
8 Nuclear Regulatory Commission Phillips Building 9 Room P-110 7920 Norfolk Avenue 10 Bethesda, Maryland I
Wednesday, March 25, 1987 12 The meeting convened at 1:15 p.m.
14 15 16 17 18 19 20 l
. 21 l 22 1 23 24 l
( 25
)
l ACE-FEDERAL REPORTERS, INC.
3 2-347 3700 Nationwide Coverage 800 33H646
.4.%- 4 - a+- + - . .w -
a a_
-30312.0 cox 2 V
1 112S1E2IEgS 2 MR. NERSES: I don't have a gavel to call this 3 meeting to order. I think you can hear me. What I would 4 like to do is get our meeting started. Today's meeting is to 5 have an opportunity to discuss with the Seabrook people the 6 publication of the Brookhaven National Laboratory report. We
- 7 are here to listen to any comments to that and also provide
. 8 you an opportunity of our review in our comments, where we 9 stand, and some of the problems we feel or areas of concern 10 that we feel need to be explored.
11 Why don't I just start this out with any 12 comments. Tom, do you want to start off.
(- 13 MR. NOVAK: Let me just make one brief comment. )
14 This is Tom Novak with the staff. On the 18th of March, we 15 did meet with Public Service Company of New Hampshire and 16 other utility owners of the Seabrook station. The sheting --
17 there was a transcript of the meeting, it is available to the 18 public.
19 During that meeting, we did point out that we 20 would use this meeting as an opportunity to express, as 21 candidly as we can, the concerns and the work remaining with 22 regard to the waiver request filed by Public Service of New i
23 Hampshire.
24 We have asked a number of people from the staff, 25 as well as the consultants, to articulate as best they can
(
/\CE-FEDERAL REPORTERS, INC.
! 202-347 3700 Nationwide Coverage 800-336-6646
l 30312.0 cox 3 1 the results of our work of the waiver requests.
2 So we expect to cover that in this afternoon's 3 agenda. I don't have a specific agenda. I think the 4 suggestion that we first try to see if, in fact, the utility l
5 has any specific comments that they have prepared, 6 presentations regarding the review, we might entertain those
. 7 first, and then come back and articulate in more detail the 8 residual concerns that are expressed in the Brookhaven 9 report.
10 MR. NERSES: Tom, before we start, I think we have 11 made one round, but I think it would be fruitful for us just 12 to introduce ourselves again.
(s 13 MR. NOVAK Why don't we do it very quickly. With 14 the size of this audience, we are going to waste a lot of 15 time in informalities that aren't necessary.
16 MR. NERSES: I understand that. There are a lot 17 of new faces. We don't have to have it on the record.
,18 (Discussion off the record.)
19 MR. NERSES: Thank you very much. Why don't we
. 20 proceed, then.
21 MR. MOODY: Do you want us to start?
22 MR. NERSES: Do you hat c some comments?
23 MR. FEIGENBAUM: I have some opening remarks. I 24 want to thank everybody first of all for meeting with us here 25 today. As Tom mentioned, last week some of the chief
(
ace-FEDERAL REPORTERS, INC.
l 202 347 3700 Nationwide Coverage 800-336-6646
f 30312.0 i cCx 4 L
1 executive officers of the joint owners involved in Seabrook 2 met with the upper management of the NUREG. It was agreed 3 that at that time that a drawn out long review process of 4 this petition was really in nobody's benefit. At that time, 5 at that meeting, the NUREG pledged to apply whatever 6 resources were necessary to move ahead to -- hopefully
. 7 expeditiously on a review of the petition. Seeing that 8 Seabrook is ready to operate and is complete and that the
, 9 petition review is essentially the critical path for 10 operation.
11 We would like to bring the spirit of cooperation 12 that was voiced at that meeting to this meeting, as we 4
A- 13 believe the petitioner view right now is at a very critical 14 juncture. As you know, the process started about nine months 15 ago when we submitted our technical studies, and Brookhaven 4
16 has performed an excellent service for everybody's benefit in 17 putting to rest a number of issues and culling out from the i
18 various documents a number of key issues that have to be 19 resol.ed, remaining issues.
. 20 We would like to understand today, from the NUREG, 21 which of those issues that you deem to be important and which 22 ones have to be addressed expeditiously. We plan to do that, 23 and we pledge to provide the resources to move ahead on these 24 rigorously and as quickly as possible.
25 For the record, I would like to just mention that s
Ace FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage an346646
30312.0 cox 5 I we do strongly believe that some of the issues raised by 2 Brookhaven are not important safety issues for Seabrook, and 3 this is well documented in our request for additional 4
information in our supplementary submissions to the NUREG, 5 and in comments to the draft report.
6 However, we also realize that we have been not
. 7 been completely successful in convincing the NUREG staff that 8
these issues are not important for Seabrook, and, theretore, 9 we intend today to face those issues head-on in possibly a 10 different manner than we have in the past.
11 I think it's to everybody's benefit, if we try to 12 keep on track on the remaining issues. I believe that some i ss 13 form of management controls of this process from here on out 14 will be necessary, because of the large number of consultants 15 and people involved, and, of course, of the importance to the 16 Seabrook licensing process.
17 Today we are here to specifically address the 18 NUREG and BEL concerns that still remain. We realize there I'
19 i
are other emergency-related issues involve in the planning of 20
. the review. But we would like to hold that for a separate 21 meeting. We feel the agenda is full enough today, just 22 talking about the technical issues specifically, that we 23 might want to hold off for a few weeks in having a separa.te 24 meeting on the emergency planning issues.
25 That's really all I have to say at this point,
, \.
I ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage mn.336-6M6 l
L____________________________._______
l l
30312.0 -
CoX 6 1 s-1 Vic. We are kind of flexible.on format for the meeting. I 2 am sure it will all come out at one point or another.
3 So we would suggest that we would like to hear j 4 perhaps from your contractors and yourself what issues are of 5 key importance to you, then we could address them one by one J
i 6 or after you have completed your presentation.
. 7 MR. NERSES: Just to get the record clear, we had 8 first thought of having'you make a presentation, then we 9 subsequently had a discussion about that. You felt that we j 10 could probably forego that for an opportunity to get right l 11 into where our concerns are and get into that. I think, 1
- 12 Ernie, you are prepared to get into that aspect of it right
- b. 13 off the bat.
i 14 MR. ROSSI: Yes. Let me run down a quick summary, 15 I guess, of what issues we see that have to be addressed on 16 our review of the petition. Clearly, one of the major issues i
17 is going to be to identify and analyze an approach that will 18 result in significantly improving the confidence that high 19 pressure core melt sequences will not result in larger pieces
. 20 induced by either steam generator tube rupture or by causing 21 containment failure by direct heating. The results of the 22 approach that is identified will need to be incorporated into 23 the PRA.
24 As the staff sees it at this point in time, there 25 are a couple of alternatives, and there may be others that u
ACE FEDERAL REPORTERS, lNC.
- j , 1 202 347 3700 Nationwide Coverage 800-336We I.
m _A - - - . - - . _ . - . - . . _ me. -s e i
30312.0 cox 7 c;x 1 the applicant may want to identify. One alternative is to do 2 analyses and experiments to demonstrate that the phenomena of steam generator tube rupture, induced by high temperatures 3
4 and direct heating, are not significant for the Seabrook 5 specific design.
6 Now, that approach would need to address
. 7 particular areas of concern to the NUREG experts and 8 consultants, and we believe that that would -- it would be 9 necessary to convince the NUREG staff and consultant experts 10 that are most knowledgeable about those phenomena that have 11 been involved in NUREG-1150 that the issue has been addressed
[
12 on Seabrook in a plant-specific way.
%- 13
- We would also note that this could very well take j 14 substantial time and effort, and that it has a lot of 1
15 difficulties to do that within the time frame that you are 16 asking for our review of the petition.
17 The second approach would be a mitigation approach 18 that would clearly preclude the phenomenon. As an example 19 would be depressurization of the reactor coolant system in
! , 20 order to eliminate the possibility of a high pressure core 21 melt sequences. i 22 To go that route, some of the things that you 23 would need to address would be to determine and justify the i
24 reactor coolant system pressure reduction that would be 25 necessary in order to preclude either direct containment s
/\CE FEDERAL REPORTERS, lNC.
! 1 202447 3700 Nationwide Coverage 800 3466e
i30312.0 scox B N
1 heating or induced steam generator tube rupture, establish 2 and justify that the capability exists at Seabrook for doing 3 the pressure reduction, and that would require thermal 4 hydraulic analyses, and it would also require consideration 5 of the equipment survival ability that is needed to do the 6 depressurization, and deal with whatever else may be going on
, 7 at the same time in the environment that would exist at the 8 time you would have to do the depressurization. It would 9 involve establishing a clear criteria that would be used for 10 the operators to initiate the depressurization, and it would 11 -- we would need procedures and implementation of the 12 procedures, including appropriate training to do that s- 13 depressurization.
14 The goal would be able to demonstrate a high 15 probability of successfully being able to depressurize the 16 reactor coolant system under the accident conditions in a 17 time frame that would have a high probability of avoiding the 18 high pressure melt sequences.
{
19 That approach also would need to look at the
, 20 potential negative aspects of depressurizing. Negative 1 21 aspects would include things that might be made worse by 22 doing the depressurization, or additional accident sequences 23 that that depressurization might lead to.
24 That is a fairly significant issue that we see 25 needs to be addressed and may very well be one of the more
\_
ace FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage fm336-6646 '
u- _. - - _ .__.
30312.0 cox 9 w
1 difficult ones.
2 Other items that have been identified by the 3 staff --
4 MR. ROSS: I talked to Novak this morning. He 5 asked it we would give a little bit of elaboration on this 6 point. Perhaps this would be a good time for us to do that.
- 7 Would that be all right? ,
8 MR. ROSSI: You can do it now or I can run through 9 the other six items. Why don't I do that. The others go 10 pretty fast, I think.
11 MR. ROSS: All right.
12 MR. ROSSI: As identified by Brookhaven, the s, 13 applicant needs to complete a Seabrook-specific PRA for 14 accidents during shutdown that includes etternal initiating 15 events and appropriate source terms, usir.g Wash 1400 16 methodology; that study would need to be incorporated into 17 the risk scudy.
18 For that, you should consider the issues that have 19 been raised by Brookhaven in its review of NSAC o4. The
. 20 staff's calculations of the whole body and thyroid dose 21 versus distance curves for the design basis accidents 22 generally agree with the Public Service of New Hampshire 23 results, as long as the meteorological dispersions modeled by 24 the same methods as used in NUREG-0396.
25 When the Seabrook-specific meteorological data is s
ace-FEDERAL REPORTERS, INC.
202J47 3700 Nationwide Coverage 800 4 36-6646
t
-30312.0 cox 10 q
.1 used by the staff, however, which is the standard manner for 2 design basis accident analyses, the staff results indicate .
o 3 that the upper protective action guide levels for thyroid I 4 dose would be exceeded out beyond the EPZ zone that you are 5 requesting.
6 One possibility is to -- well, first of all, I
, 7 think there is an issue as to what methodology is used, 8 whether we use the NUREG-0396 for this particular review, or 9 whether Seabrook meteorological data is used. That, in my 10 mind, is still an issue that we need to discuss.
11 Other approaches are to consider additional 12 measures to reduce the design basis accident source term to a
%- 13 point that would mee', at least the upper thyroid PAG level, 14 at the proposed EPZ radius using the Seabrook meteorological 15 data.
16 One example, or a possibility that's been
- 17 discussed by the staff, is looking at what can be done to 18 reduce the type C valve leakage rates, which account for much l
19 of the source term.
20 This, I guess, is not going to be a subject of 21 today's meeting, but clearly it needs to be addressed at some 22 point in time. That is providing information on the basis 23 for the statement in the petition that emergency planning for 24 the reduced EPZ radius would provide a suitable basis for ad 25 hoc protective actions significantly beyond the reduced EPZ !
u ace. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-(M6 l
w
,30312.0 cox 11 L
1 radius.
2 MR. FEIGENBAUM: We would be prepared to discuss 3 that today. '
4 MR. ROSSI: You are prepared to discuss that 9
5 today. I thought that was one of the items you were going to 6 put off till later.
, , 7 The last two items are simply to continue our ,
8 review, and we will need applicant's support and the staff 9 contractor reviews of containment bypass sequences.
i At some 10 point in time we will give you RAIs on that issue. Also we f
11 need to relook and complete the review of the PRA plant 12 systems model. Again, we would expect that that would be i i
s, 13 handled in meetings with RAIs like other reviews are 14 handled.
15 Now, those are really what we have as the bigger i 16 issues. There are probably a number of smaller ones that are 17 going to come up as we do the review. We may eliminate some 18 items as we do the review. You may find some other issues as 19 we go along.
20 Now, I guess, maybe we ought to --
21 MR. MOODY: One clarification. The last two 22 containment bypass and the last two systems analysis, taose '
23 are two separate ones?
24 MR. ROSSI: Yes. We haven't divided this as two 25 separate ones. The containment bypass sequences were looked s.
ace. FEDERAL REPORTERS. INC.
202 347 3700 Nationwide Cmcrage Hu)J LWe
30312.0
.cox 12 w
I at in some detail by Brookhaven. I think there are still a 2
few loose ends that need to be closed out there.
3 The last item really is the staff in going to need 4
to address the original PRA for the plant, the front end.
5 That's the front end review, and we will need whatever 6
support we need in answering the questions that we come up
. 7 with as we do that review.
8 MR. MOODY: Thank you.
9 MR. ROSSI: Again, the direct heating and steam 10 generating tube rupture is one of the -- while it looks like 11 it may be a very difficult issue, so now, I think, would be a 12 good time to let the research people --
N. 13 MR. NOVAK: Let me make one statement. In earlier 14 meetings we heard from Public Service of New Hampshire 15 b specifically, Bob Henry had discussed his views, which are 16 reflective of the utility's views over concerns on direct 17 heating.
It would be my opinion that Brookhaven, and 18 supported by the staff, continue to believe that the generic 19 concerns that I will ask Benny Ross and his people to address
. 20 continue to provide us with a certain degree of uncertainty 21 that is still on the table, so to speak.
22 So what t I would like to do is have the office of '
23 research, in a manner, summarize the generic understanding of 24 things such as direct heating phenomena. Then leave it to 25 the utility specifically to address it on a plant specific b-ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage alo 336 66M
{N */
w 30312.0 cox 13 1 basis. I just want to make that separation. I would draw 2 the research concerns to be generic and then require the 3 utility to respond to them as appropriate with the Public 4 Service of New Hampshire. Benny, thank you for showing up.
5 MR. ROSS: You might want to dim the lights 6 halfway if it works. I am going to show some pictures.
7 Then, I think, Tim, you have some slides?
8 MR. LBE: I have some readoffs.
9 MR. ROSS: This is just a cartoon on how the high 10 pressure melt sequence starts from the bottom of the vessel.
11 There is really not a national consensus on whether its 12 independent, like Tom was mentioning generic, or is there s, 13 some feature of a cavity in one design that would promote 14 direct heating and another design that would be a target.
15 Nonetheless, it is one of the high pressure melt 16 ejection issues. It's shown here as melt expulsion and 17 general heating of the containment.
18 We built this facility at Sandia, and it's 40 feet 19 tall -- there are dimensions here in meters, but I can't 20 figure out meters, so we will have to use feet. The object 21 here is to inject up to 80 kilograms of hot metal.
. 22 Parametric way, I think total is 15 test --
23 MR. LEE: 11.
24 MR. ROSS: 11 at least for now, and different 25 parameters, and try..to do some modeling work on exactly.how s_
l ace. FEDERAL REPORTERS, INC.
- o:.m-mn Nanonmde Cmcrage so 336 (M6 l l l
l
C _ a ,
30312.0 cox 14 s, i j 1 efficient heating hot metal is in a containment atmosphere.
l 2 One of the things we have done generically, this
- 3 is for Surry, we have done it for other plants, this would be -
i ,
l.
4 something you would probably have to do if you took the first [
2
- 5 option that Ernie mentioned. That is, if you want to try to
\
6 guess direct containment heating for your facility, instead l i
j -
7 of making it go away. Expert curves, some people put quotes 8 on " expert" and some people don't, were constructed, and this 9 chart shows, in the empty rectangles, degrees of relief of j
j 10 what the load would be on the containment, and then the cross !
l '
11 rectangles are the response of the containment to load. I I
i
, 12 Whenever you see an empty rectangle to the right of a shaded s,
t
! 13 rectangle, the analyst would assume there is some probability i j
14 that the containment would fail for this event.
i 15 i
MR. NERSES: Are you able to hear him back there !
16 okay? '
- 17 MR. NOVAKt Yes.
t 18 MR. ROSS: One outcome could be that you decide 1 19 you want to go down the same trail that we did and construct !
I
- i. 20 a plausible histogram for direct heating and other loads for !
i
) '
21 different sequences, and overlay with it what you think the
- 22 containment response *would be. And then these would somehow t
l j 23 work together to give some probabilities of early containment I 24 phase.
I
{ 25 If you decide to do that, then in the interest of
. N.
5 l
1
- ACE. FEDERAL RuronTans, INC.
j i 202 347 37(x) Nanonwide Coverage N00-334%46 4
..,,,-----,-,,.x-
.g--,---..m,-,,,n_ ,- ,- -- n,, -n-
_ . _ - , _,g,w.,
, ,e,n m,sesm4,-,_,,,,,a vm8 w m e-e-
t i
l 30312.0 cox 15 I L.
l L
1 science, research would be glad to work with you because we 2 just learn that much more. We always like to get more 3 plans. We ca'n tell you who did this work, how it was done, 4 it's about 18 months old or so. We have learned some more.
5 You might get a different-looking histogram than we got for 6 your plant.
i 7
When you take this approach, you can calculate, 8
through a sampling program, what the probability is for early 9 containment failure. Some of the times, when you sample 10 these degrees, I believe you will get a high containment 11 f
structure response and a low heating load, and you say no 12 failure. You do a number of samples, then you get sort of a s, 13 scatter diagram that looks like this. This is just a number 14 count of the number of times that we would sample early l 15 1
containment failure with and without direct containment 16 heating.
17 It looks a little better if you look at an l 10
(
interval, you can come across it at the 50 percent level with 19 no direct containment heating and get an immediate 20 probability of something less than 10 percent for Surry.
21 If you did your p) ant sort of the way we did ours, 22 you can construct curves like this.
23 The implications are clear, if you have a low l
24 containment -- conditional containment failure, then you are 25 going to get a low probability, it may be acceptably low, I
~~
l Aca FnounAL REconTuns. INC.
202 347 3700 Nationwide Coverage mn DMM6 i
30312.0 ccx 16 i
I don't know, of getting a high early radiation exposure 2 off-site.
3 The curve is less optimistic if you throw in 4 direct containment heating. . Instead of 8 percent, you see 5 numbers on l5 percent or so at the median level.
6 The same data, just looked at it another way, on a 7 linear scale, probability, the little Xs are samples, with no 8 direct heating to the right, most of the little Xs are down 9 at the bottom. This is the same line, just new line.
10 The next question is what do we know about 11 containment response, we finished a 1/6 scale at Sandia.
12 These are some late construction photogrsphs, over 1000 s, 13 sensors put en it. Then test containment would be 14 pressurized to failure sometime within the next 30 days or 15 so. We have done a lot of pretest predictions. We hope to 16 validate and get higher confidence in the models used to 17 calculate the capabilities pressures for containment. This 18 is some other photographs of the same structure. It was 19 built to code, has a liner. Right now we are probably 20 running T tests, if they haven't already been done.
21 I assume you have something like this from your 22 PRA. Certainly, when you start looking at the residual risk, !
l 23 you are going to need to know what are the major 24 contributors, because some of these contributors are direct 25 containment heating sequences and some aren't. I assume you s
ACE. FEDERAL REPORTERS, INC, 202 347 3700 Nationwide Coverage M0lk3 M 6646
m.
30312.0 cox 17
's I have this type of information if you decide to go the first 2 way that Ernie mentioned. That is, to try to see whether 3 this is a risk-significant event for your plant.
4 Why don't you turn on the lights, if you would, 5 now.
6 Now, the other side, the latter option that Ernie 7 mentioned, if you decide to try to make it go away by 8 modifying the hardware and procedures, again, we have done a 9 lot of work in the last five years on the venting capability i 10 through your PORV. If there is some question about 11 validating models or getting a warm feeling, that you can get 12 that many kilograms per second through a PORV, we will work
%, 13 with NRR to review your model, because we have done a lot of 14 work. I think in! hat area we wouldn't have to do anything 15 else. We have got our research already done.
16 As far as timing, I am going to make a guess. If 17 you tried to do this sort of thing, I am going to guess it 18 would take at least six months. What do you think, Joe?
19 MR. MURPHY: That sounds reasonable.
- 20 MR. ROSS: To crank it through, try to estimate 21 the loads, and get a good idea of what the loads would be and 22 the containment response, and generate enough technical 23 support that you would feel comfortable in defending it 24 anyway. It took us somewhat longer, but I think we at least 25 have gone through it once. We have a reasonable idea how to s..
ACE. FEDERAL REPORTERS, INC.
ll 2c347 3700 Nakmwide coverase an.3weM6
-. . - _ -. ._ = _ . . _ _ .-
30312.0 cox
- 18 y
1 do it. We are always trying to add more plants to our data 2 base, so we wo'uld be glad to add this butterfly under glass,,
3 if you would, so we could study it some more.
4 MR. FEIGENBAUM: Would you like us to respond?
5 MR. ROSS: There is a little bit more. Tim had 6 some pictures he wanted to show.
. 7 MR. LEE: Dr. Ross has discussed the implication 8 of the issue of heating, of the risk to the plant. I will 9 just continue the research program, direct contaminant 10 heating issue only.
11 But I didn't have too much time to prepare for 12 this presentation. I don't know how the material here could
(- 13 fit into everybody's needs, but I will try to supplement my 14 presentation by answering whatever questions you may have. I 15 am trying to give you the benefit of the latest information 16 concerning our understanding of the issue.
17 I will start with telling you about information 18 that is available at this moment and discuss what implication 19 they will have.
. 20 on the analytical side, we do have some 21 calculations on the so-called containment role in groups 22 under our program. This is accident specifications specified 23 by the HUREO office and asks many plants to do calculations 24 of what consequence on the containment pressurization, it we 25 assume the amount at the core will melt in the containment
\_ ,
ACE. FEDERAL REPORTERS, INC.
m2.H7 37a) Nathmwide Cmerage im316-6M4
~ _ _ - - - _ _ _ _ - - - - _ - _ _ _ _ _ _ _ _ _ - _ _ - _ - _ _ _ _ _ - _ _ _ - _ . _ - _ _ _ _ _
.A 3031200 cox 19 s
I under pressure.
2 Some of the calculation work was provided in the 3 presentation of NUREG-1150. I will refer you to NUREG-1.150 4 if you want to investigate the results of this calculation 5 further.
6 All I am trying to point out is this calculation
. 7 does not consider the rate, like the rate of the temperature 4 8 reaction, the rate of the heat transfer. In this
- 9 calculation, we assume that the heat reaction will be 10 completed instantaneously, that the heat will be transferred 11 through the containment atmosphere to equilibrium. Not 12 included in this calculation also are some mitigating
, s, 13 effects, like the structure, and also the effect of possibly i
. 14 -- possible effect of the quenching by water, which may be 15 presented in the cavity or on the floor of the containment.
16 So the result of this calculation is on the 4
17 conservative side.
< 18 We have a more recent calculation conducted and 4
19 included in the NUREG-1150, which contained nome summary 20 conclusions and calculated the sensitivity of the direct 21 containment heating.
22 This result is also presented in NUREG-1150, and I 23 will just show you a chart of how the results came out.
24 They tried to compare the maximum pressure 1
25 calculated and compared it with the assumed finer ultimate l
1 1
l Acn.Fuonnai. ReconTans, INC.
. 202 347 3700 Nationwide Coverage N043%6646
~
a ..
30312.0 cox 20 L
1 terms of the containment,' the highest _ pressure of the 2 containment we can afford with failure.
3 This shows you for the Surry containment, there 4 are many sequences which reaches that pressure or even 5 exceeds it. Ag'ain, if you need more detailed discussion of 6 this, we will get you a copy of NUREG-1150, and it will 7 provide you more information so that you can see more.
8 There are also some calculations made by the 9 University of Wisconsin concerning the rate of the dispersal 10 of the core debris from the reactor plant only. They did not
. 11 consider any additional structure which may present of the 12 exit of the heat and also the containment. But the_ question A- 13 of the structure didn't relate at all.
14 on the experimental side, we have completed an 15 experiment, SPIT, the first is SPIT-18, which is about 1/20 16 on the scale. We also completed the HIPS test, which is 1/10 17 scale, which is conducted within the containment.
18 As Dr. Ross showed you, we recently procured this, 19 the containment volume. We are starting a new series of
. 20 tests we call the DCH. We have completed two tests in this-21 facility, DCH-1 and DCH-2, but I don't have much information 22 on DCH-2.
23 On the containment pressurization, DCH-1, CWTI is l 24 another test, which is conducted under the sponsorship of 25 others. SPIT-18 and 19 is conducted with steel cell, which
\
ace. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coversee Mn3%.MW6
' "-**~
1 30312.0 cox 21
- s.
- I has only radiated pressure or something that is 3 PSIG. That 2 pressure was easily exceeded in the test.
3 Then, of course, the DCH-2. In all of those test, 4 significant pressurization of the volume, the containment 5 volume, was observed.
6 On the subject of the debris dispersion, those
- 7 tests results are written, but all except the CWTI tests are 1
8 conducted without containment water.
9 On the subject of the effect of the structure, the i
10 two tests -- this is two tests in the Sandia and on a smaller
} 11 acale, currently ongoing; there are several past which the 4
i 12 sub3 ect is the effect of the water.
%, 13 In looking at the test results, I would have a 14 portie.n that all of the tests conducted, we keep at the 1
15 Commission. By that I mean that none of those tests are 16 , coaducted with exactly the identical composition for the --
17 l are representative of the molten core material. In other 18 tests they use a so-called core melt which included some 19 steel, some material which is important and causes 20 oxidization of the metal that provides half of the energy, 21 which results in the pressurization of the containment.
22 Also, the scale of interest in that we know that 23 as of this moment, the the b: ale of the test facility. The 24 result is more biased against this, which means that the 25 larger the facility is, you will observe for the calculation, ace. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 80n 3%6M6 f
p - m...
30312.0 cox 22 L
4
, 1 more an effect of the contaminated heating. Again, I will 2
come back to that discussion later on.
3 I will go to next the mitigating effects. Also 4
discuss some of the test results which are on each of the 5 subjects.
First of course is the depressurization of the 6
reactor coolant system, because that's the problem of the
! 7
. high pressure ejection, which as we understand, should be 8
part of the program of direct containment heating. There cre 9
some programs in the NUREG office, and also with the 10 industries, to look into the possibility that some 11 i calculation may exist in the reactor system that could carry 12
[theheatfromthecoreintosomeotherpartofthereactor
(- 13 coolant system, and induce a failure of another part of the 14 RCS boundary before the core will melt through the bottom of 15 the reactor. .The depressurization of the reduced failure 16 could affect the system, and if there are any questions on 17 this subject, I will refer it to Jim Hofmayer, who's a 18 manager.
19 But there always is a concern that Whether the 20
. induced failure could depressurize a system in time to 21.
alleviate the problem, or whether it's proper to rely on the
, 22 failure of the safety system, or the safety system could have 23- 1 reviewed a consequence of an accident of this magnitude. So 24 the active system instead of relying on the induced failure i!S to depressurize the RCS.
L ace FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-3364M6
i.
3031200 '
cox 23 L
1 I can just say at this moment we just started to 2 look into this program, and there are many questions we did 3 answer before we can consider the system as a viable 4 alternative to eliminate this problem. Of course the 5 depressurization with the existing hardware and also the 6 impact of the depressurization, and the entire reactor system
. t 7 on other safety issues, such as steam explosion or how much 8 to measure.
9 Now, the second thing which could have some 10 mitigating effect on the DCH is the effect of the structure.
11 I believe I don't have the benefit of the readings of 12 material from the applicant, but I am guessing from the s, 13 presentation, and from previous representations, there are 14 some major mitigating circumstances. This might be the item, 15 ,
the effect of the DCH.
I 16 But in considering whether the structure can 17 reduce the amount of the melt particles which may participate 18 in heating the containment atmosphere, we have to consider 19 several things. First of all, we still don't have any
, 20 acceptable scaling law that can extrapolate the results of
! 21 the small scale test to ur. ,
22 As a matter of fact, we even have a problem of I i
23 quantifying the effect of the scale. By scaling, I mean also i 24 like a melt, probability of the melt at temperature and so 25 forth. -
I s- '
ace FEDE (AL REPORTERS, INC.
202-347-3700 t ationwide Coverage 800-336-6646 1
y _ _
i 30312.0 cox 24
'\-
1 But even just considering the geometry scale, we 2
all know that the larger the facility, the extent of the 3
containment heating is likely to increase, but we still 4 cannot quantify it exactly here. We are working in that 5 direction.
6 Another thing I would like to bring up, recently,
- , 7 in the past six, eight months, both Brookhaven and Sandia 8
started identifying the effects of melt containment into the 9
cavity, most likely could go through a rapid reaction with 10 the steam available, I guess the frequency -- steam will be 11 available in the containment atmosphere.
12 The objective will be containment atmosphere, or
\_. 13 even with the reactor cavity, with the high speed. We 14 measure the rea : tion with the steam, and the preliminary 15 calculations indicate that it does. Before the particles get 16 out far from the cavity, almost 100 percent could be 17 contained or oxidized.
18 The result of that submittal of steam reaction is 19 hydrogen, which is migrating throughout the containment.
20 Even with the intervening structure.
21 The calculation will show you the effect of this 22 on the container. The result is even more dramatic because 23 t of the volume of the container in Sequoyah is smaller.
24 Up until the time when people were seeing if steam 25 '
reaction could take place, it should be able to trap the
(- '
ace-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-33M446
30312.0 cox 25
.i \-
1 metal particles in the upper containment area. So that's 2 part of the heating. But if you contain the particles, 3 hydrogen can be generating in the lower department. Thus 4
hydrogen can go through the ice chamber very easily and reach 5 the upper containment or even burn in the ice chamber, and 6 the resulting pressure is enough to further contain it.
. 7 There is also another subject that we have on 8
mitigating effect, the effect of the water which may present 9 the reactor cavity or on the floor of the containment.
10 But as of this moment, the evidence is mixed. We 11 still cannot see even whether the pretence of the water could 12 help or not. CWI test will deposit particles down toward a
(, 13 pool of the water on the floor of the chamber in which_you 14 lay the containment. That's a configuration, but their tests 15 showed that, yes, the water - presented the water on the 16 floor is effective in quenching some of the particles iof the
! 17 pressurized -- peak pressure and Loating.
18 But, on the other hand, the Sandia test and the 19 HIPS test showed that water in the test cavity is pushed out 20 by a pressure spike.
But the high speed film taken during 21
.I this test showed that the water in the test cavity quenched 22 the water on the floor.
23 What we showed, when conducted with the water, the 24 melt produces a large pressure spike, and that pressure spike 25 is pushed out of the water from the cavity, so that the l
%- 1 1
14CEJFEDERAL REPORTERS, INC. !
202-347-3700 Nationwide Coverage 80 4 336-6646
)
1 30312.0 cox 26 f (_
1 particle does not mix uniformly with the water, and the 2
effect of quenching was not observed to be significant.
i 3 MR.'BARRETT Tim, could I ask a question about 4 structure, the effect of structure? I 5 MR. LEE: Yes.
6 MR. BARRETT: Do you have any tests that show the 7
(
effect of structure for a small scale?
8 MR. LEE: Right now, to the best of my knowledge, 9 we have some structure, which, again, I don't have, i
10 Brookhaven is initiating a test which would include some 11 structure.
12 But the main problem here is how you are going to
\_, 13 end up with the result.
The effect of the scale I am talking 14 about will be multiplied if you keep on adding to the 15 structure. I just mentioned that in the smaller scale 16 facility, it resulted in a bias against the DCH. That's 17 because of two reasons we know of right now. One is 16 increased variation in a smaller scale, in a small scale 19 facility.
20 So that increases the heat lost to the structure 21 and frees up the melt particle, so less melt particles are 22 ejected into the containment area.
23 The other and more important reason we know of 24 that the scale of the facility could extend to the DCH is the 25 length of the task over which the mere particle can fly L
, ace. FEDERAL REPORTERS INC.
202 347 3700 Nationwide Coverage 800 3 4 6646
e . __
30312.0 Cox
- 27
. N.
i 1
freely before it hits the outer structure, that length also p, 2 decreases in a small scale facility.
3 We show that chemical reaction and heat transfer 4
is more vigorous when the particles have freedom of 5 movement.
6 That reason is incorporated into the small scale
. 7 test and we still have a problem identifying it.
! 8 So the best advice I can give you is any test 9
result available so far can be used on the point of the 10 assessment or the tendency or may be used to assess what 11 solutions are needed for a number.
- 2 MR. ROSS:
Mel. Maybe we ought to kick it back to
\_, 13 i
the See. brook people who came this distance and give them 14 first choice on questions. You had some, I believe.
j 15 MR. FEIGENBAUM: No, I don't. We would like to 16 address this issue. We have a different approach to it at 17 1
this time, I think, than we have had in the past. Jim, do 18 you want to present that.
19 MR. MOODY: Yes. I have four of the six issues 20 t here, the other were DBA and ad hoc discussions which weren't 21 in this presentation.
22 MR. FEIGENBAUM:
Can you speak up.
23 MR. MOODY:
I think we will address four out of 24 the six issues that were identified earlier. The third and j 25 l fourth, DBA and something to do with ad hoc actions, . we can '
(- )
1 I
4
/\CE-FEDERAL REPORTERS, lNC.
202 347 3700 Nationwide Coverage fl00-336-6M4
_ - . _ _ . . . _ _ _ _ . _ ~ , _ . _ - - _ . _ . _ . _ . . _ . _ _ , - _ . ,
_ _ _ _ _ _ _ _ _ - _ . , _ _ _ . L
30312.0 cox 28 L
1 discuss later, but was not the intent of this presentation.
2 MR. ROSSI: I think you were not aware of the DBA
, 3 dose question until today, is that the case? I am not sure 4 it's in the Brookhaven report. I think that was a staff item i
, 5 that was done off that. I am not surprised you are not ready 6 to address it.
7 MR. MOODY: The other thing I will try to do is 8 concentrate on the specific issues. I am not going to talk 9 about all the comments we have on the final Hrookhaven 10 report. We will put that in the documented form probably in 11 a few weeks. We will submit some initial comments.
12 But just to summarize, we could see in the final 0
s, 13 report that many of our comments on the draft had been 14 addressed in some way, not always positive, but they were 15 considered. On the other hand, we saw some comments that 16 hadn't been addressed or may not have been addressed. We 17 think there is a good reason for -- loca was stated to be a 18 significant contribution, so we suspect that may have been 19 the reason that it wasn't -- our comments may not have been 20 addressed specifically.
21 We are still encouraged by the Brookhaven report.
22 We believe it supports our waiver request. When you look at 23 the best estimate considerations that Brookhaven considered, 24 it doesn't affect our results at all.
25 We picked up, I think you have the four issues I
./ i i
ace FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
l 7 .
1 30312.0 cox 29 s
I right here. We are planning to-look at shutdown events in 2 much greater detail. We have already initiated efforts on 3 that. I am going to talk about the induced steam generator
, 4 tube rupture and the direct containment heating or 4 5 containment loads issue is pretty much together.
'6 There was some indication that additional
{-
1 7 completeness is needed in the review, a search for early
- 8 release sequences, and we are, of course, ready to support on 9 those areas.
10 This is kind of a busy slide, and it gives the 11 Brookhaven areas of review. We believe we have tried to I 12 address where we think there is agreement and where there are Ag -
13 some open issues still remaining.
14 Many of these are insignificant.
, 15 As I indicated, facing loca, they concluded it had 16 no impact on the risk results. However, on the record, we 17 will give our comments again, or refer to our draft 18 comments. We believe the calculation of the initiating event 19 is not correct, and we had disagreemente with that.
. 20 MR. ROSSI: Could I just say one thing. I 21 i
personally think as you go through this and state what you 22 believe that Brookhaven included in their report, if 23 Brookhaven does not agree with it or have a comment, you
- 24 ought to make it right at the time, if you would, please.
25 MR. PRATT Absolutely.
.! \-
i
/\CE. FEDERAL REPORTERS, ]NC.
1 202 347 3700
' Nationwide Coverase Mn3M MA6 1
l j 30312.0 cox 30 i .
1 MR. FLEMING: Could you clarify for the -- what 2
you stated for the record was not correct was the Brookhaven 3 analysis.
, 4 I
MR. NEWBERRY: I understood that.
5 MR. ROSSI:
Which is on check valve failure f 6 rates. i 7
MR. MOODY:
We believe that the initiate frequency 8
that Brookhaven is using for their loca analysis is 9 incorrect. We will comment on that again.
10 On shutdown events, we said the conservative 11 impact had minor results on that, but Brookhaven had some '
12
{ reservations, thought more in-depth study was needed and we
- %. 13 agreed.
There were some interpretations of NSAC-84. I think 14 we have some agreements as to the impact of Seabrook-specific 15 features.
We can talk about these in a little more detail 16 later.
17 Induced steam generator tube rupture. There were 18 two quantifications, I will get to this in a minute. But the >
19 best estimate of the lower frequency calculation had no 20 impact.
The conservative one had some minor impacts. Again, 21 there's a lot of concern with uncertainty and need to resolve 22 this to higher competence levels expressed by Brookhaven. We 23 have done extensive analysis on induced steam generator 24 rupture events.
We have discussed that before. We don't 25 think it's an issue. There is a lot of margin on the tubes, A-ace. FEDERAL REPORTERS, INC.
202 347 3700 Nationwkle Coverage 2336-fM6 1 . . .
i
, 30312.0 l cox 31 u s-1 but it doesn't appear it will be fruitful to go down that 2 path. We will talk about that in a minute.
3 MR. LYON: On that one, I will hold until you talk 4
about it, but I am not sure we agree with all those things.
5 MR. MOODY: I was talking about things we don't 6 agree with, I think.
7 MR. LYON: I was looking back. You say points of 8 agreement, probably no impact, for example.
9 l MR. MOODY: Well, in the Brookhaven report, there 10 were words to that effect. It probably has no impact and is i 11 of such uncertainty in their mind.
- 12 MR. ROSSI
- Is that what you believe the N- 13 Brookhaven report says, Trevor?
14 MR. PRATT If he has seen the words, they must be 15 4
in there. I think the range of implication would show there 1G is a range of sensitivity studies.
11 MR. ROSSI: The upper range was important, median 18 was not, and lower was off the scale.
19 MR. MOODY: Yes, we used the better estimate
. 20 calculation, it had no effect.
21 MR. PRATT: That's right.
i .
22 MR. ROSSI: I think before you did recognize the 23 fact that the uncertainty band needs to be tightened down in 24 order to appropriately support use of best estimate.
- 25 MR. MOODY
- I believe that was Brookhaven's L
ace FEDERAt. REPORTERS, INC.
202 347 3700 Nationwide Coverage min-336-6646 j
~
30312.0 ccx 32 V 1 position.
2 MR. FEIGENBAUM: Jim, are you going to address 3 that in more detail,on a later side?
4 MR. MOODY: Yes. Contain CIS failure and leaks, 5 again, conclusion is no impact or minor impact on the 6 results. We agree. Since then, since we submitted this
- 7 information on our comments on the draft report, he have 8 traced down the four large openings that would be synonymous 9 with the category of S-6, 10 One of them doesn't apply. Two of them would 11 apply to an S-2 type of release. A fourth one appears it may 12 apply as an S-6. So from the work we have done since the
, A- 13 last time we have talked, the fact of the reduction and S-6 14 ue believe is necessary. We are not planning to talk about 15 that. We will submit that in our detailed comments.
- 16 By the way, the traceability of these events in 17 NUREG-4220 has not been easy. The four events we have been 18 able to find with nuclear power experience; and the smaller 1
19 has been reported in nuclear tech spec-type of events is very 20 difficult. We are not sure whether we can go much further or 21 whether it's even fruitful.
, 22 containment capacity, I think everybody calculates i 23 the yield stress, United Engineers, Structural Mechanics, now I
24 Brookhaven, have all calculated the yield stress within a 25 couple of pounds of each other, if not almost exact. I think 1
- ACE. FEDERAL REPORTERS, INC.
I 202 347 3700 Nationwide Coverage 800 336 6646
30312.0 ccx 33 1 we all agree with what the yield stress calculation 2 presents.
3 There is disagreement on failure criterion, 4 probabilities containment failure, beyond yield. I believe 5 Brookhaven his recommended a 50 percentile at the 1 percent 6 strain calculation. We believe that is very conservative.
. 7 Fe will comment on that again.
8 Lantainment loads. We agree that Seabrook 9 specific loads should be used, but I don't think we are going 10 to recommend that this option 1 be the solution. I will get 11 to that in a few minutes. We have done some analysis and 12 consideration of this on Seabrook specifically. We believe
(, 13 that the configuration at Seabrook makes this a non-issue, 14 l bat, again, we don't see this being resolved.
15 MR. LEE: Excuse me, could you tell us what level 16 the yields are --
17 MR. MOODY: Excuse me, I can't hear you.
18 MR. LEE: Could you tell us what pressure, the 19 yields are the load from the direct containment heating.
. 20 MR. MOODY: What --
4 21 MR. ROSSI: He is asking you what pressure did you 22 use in your study for direct containment heating. Is that
, 23 your question?
24 MR. LEE: Yes.
25 MR. MOODY (s
! /\CE. FEDERAL REPORTERS, INC.
, 202 347 3700 Nationwide Coverage anJWM46
. _ ~ . . _ , __ . - . - , - - . __ -- _ _ . - ._ , - - . _ - _ - . - - - - _ _ _ _
30312.0 ccx 34 1 MR. TORRI: Fred Torri. Our assessment on the 2 work that Bob Henry has done and work that heat core has 4
3 done, and specific review of the Seabrook configuratio.i, 4 leads us to the conclusion that the material will not be 5 disbursed in the containment atmosphere. Not a question of '
6 whether or not you will get to it. But whether or not the
- 7 question was disbursed in the containment atmosphere.
8 MR. ROSSIr Did you also treat the hydrogen 9 production effect that was discussed by Mr. Lee a few minutes 10 ago, even if it doesn't get dispersed into the upper part of 11 the container? Have you looked at that?
12 MR. TORRI: We considered hydrogen generation from s, 13h metal water reactions and hydrogen burns at the time of l'
14 venssi breach, up to and including 100 percent of the core 15 zirconium reactor. We concluded that the pressure 16 containment would not approach even the yield capacity of a 17 containment by some 50 or 60 PSI margin. Below the yield 18 capacity of a containment by about 50 or 60 PSI margin.
19 MR. LEE: This is including the material submitted
. 20 with the application?
21 MR. TORRI: The hydrogen burn study which we did 22 in the original is still the source.of reference available to 23 hydrogen burning. It's contained in a separate chapter in 24 the Seabrook PRA. It has not been resubmitted as a separate 25 document as part of the EPZ request, but is included as a Acn FconnAL ReponTens, INC.
202 347 3700 Nationwide Coverage 20 346M6
30312.0 cox 35 I reference to the SSPSA. The resulting containment 2 probabilities are of course still fully concluded in the EPZ 3 material and all the calculations we have done as a result of 4 the submittal.
. 5 MR. NOVAK: Excuse me, before we pass out. Let's 6 open this thing and give us a little more breathing room.
7 MR. LEE: Just one more comment. Is it possible 8 to get a copy of your calculation and an idea of the DCH 9 regarding Seabrook?
10 MR. TORRI: You are asking for a copy of our 11 documentation relating to DCH or hydrogen burning or both?
12 MR. ROSSI: Let me ask a question, did his
(, 13 description of the hydrogen generation agree with yours about it the phenomenar I think he said 100 percent zirconium water
- 5. 5 reaction.
16 MR. LEE: I have to reserve in my comment, up to, 17 up to what?
18 MR. TORRI: 100 percent.
19 MR. ROSSIt Zirconium water.
20 MR. LEE: I will have to reserve comment because I 21 haven't seen the report. But a couple things I would be 22 interested in. First, how they excluded it, all the things 23 from interacting with the containment. And secondly, if the 24 metal steam reaction is considered as indicated, up to 100 25 possible zirconium has been included, up to 100 percent of ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage En3%fM6
30312.0 cox 36 1 all of the zirconium in the core, and you consider that to be 2 a hazard then?
3 MR. TORRI: Yes.
4 MR. ROSSI: I think that the reason that you 5 assumed that the core melt did not get up into the upper part 6 of the containment was a small scale experiment; is that 7 correct?
8 MR. TORRI Continuing for the details, I would 9 like to refer to Bob to discuss that, but it was a 10 combination of the multidimensional effects observed in small 11 scale experiments combined with a review of the actual 12 I specific configuration of a Seabrook cavity, the instrument
%, 13 tunnel and the lower container compartment around the steam 14 generators, the open area, which the opening is adjacent to 15 the instrument tunnel exit.
16 MR. ROSSI: Let me ask a qdestion. You are 17 convinced that the containment failure criterion that is now 18 being assumed by Brookhaven is lower than what you think it 19 ought to be. You are convinced that that is the case.
20 Now, if your containment failure criterion were 21 the acceptable one, then with direct containment heating 22 loads, that you could get a general consensus would be 23 conservative. Where would that put you with respect to 24 direct containment heating problem? In other words, instead 25 of arguing amongst the experts about whether direct L
Ace.FeoERAL REPORTERS, INC.
202 347 3700 Nationwide Cmcrase 80ri.3%6646 l
E __
~
i ,
, 30312.0 cox 37 i s ,.,
I containment heating is or is not a phenomenon on your plant, ;
i . 2 maybe the easier approach is to address where does a 1
- 3 containment really fail and would that get you to a point 4 where there would be general agreement that would serve you j 5 in direct containment heating loads, so we could accept the 1
l 6 failure criterion or you have convinced us that the failure
. 7 criterion was right, where would that leave you?
8 MR. TORRI: You are asking whether a combination 9 of the NUREG-1150 direct containment heating loads, combined j 10 with our assessment of the containment capacity, would yield i
11 failure or not?
j 12 MR. ROSSI: Right.
I
- (, 13 MR. TORRI: We have submitted, as part of an RAI, 14 our interpretation of information received from Brookhaven on j 15 g the implications of direct containment heating loads on their ;
16f Zion analysis, translated back to Seabrook as best we could.
, 17- And did conclude that as a result the containment 1
! 18 probabilities would be no higher than what was derived as the 19 PRA, as the early containment failure probability.
! . 20 We have not been able today, in the NUREG-1150 l
j 21 material, been able to trace all the steps through it to be
- 22 able to give you an answer on that question on the ways of j 23 our own review of NUREG-1150. I am not sure that's an easy 24 job to do. We just haven't been able to trace all the steps 25 through all the combinations to find that answer.
1 %-
l I
L i
ACE FEDERAL REPORTERS, INC.
202 347 37(r Nationwide Coverage Roo)EAM6
. _ _ . .-. . . . __ -. . - . . - _ = . . - . . - ..
P 30312.0-cox 38
%w' 1 There is some information, some definitions that i
2 are not explicitly stated in the NUREG-1150 documentation 6
l 3 that have kept us from completing that.
Trevor, do you 'have any comment on the l
4 MR. ROSSI:
i j
5 question?
l 6 MR. PRATT I was trying to come up with some 1
description of how we presented the information in the final j 7 8 report.
j 9 After we talked last time, Fred developed what he j 10 considered to be a failure distribution for Seabrook that
! 11 kind of fitted the types of concerns that Charlie had.
12 1
So what we then went back to do was look at the
- ss 13 Zion-specific calculations we had done and reported in 1
! 14 NUREG-1150. If you recall the first report that we issued, 15 the draft report, took those calculations, and said, okay, i 16 j this was directly applied to Scabrook, this was the type of 17 answer we would get. i l
18
] So we then went back and tried to unravel from !
19 those calculations a median containment load. It's not a 20 trivial exercise as Fred indicated, because he was sampling a 4
21 large number of calculations, so we did it for a particular 22 )
I.
accident sequence and came up with a median load of about 135 23 PSIG. If I compared that, 135 PSIG, the distribution i
24 produced at the last meeting, then I come to the conclusion 25 written in the report, that it is a very low probability of N.
l, '
l i
/kCEJFEDERAL REvonTEns, INC, 2024474700 Nationwide Covernse f004M.%46 i
.~
l i
I 30312.0 i l
! ccx 39 i rs
- 1 containment failure based on that load.
2 However, the distribution of load one sees is very 3 wide.
1 4 MR. ROSSI: You are saying now the uncertainty in 5 the loads?
l 6 MR. PRATT: Yes.
l - 7 MR. ROSSI: My question was, if you accepted an 1
8 upper limit of that uncertaincy, you would indeed have the l 9 overlap even with the most favorable containment failure I 10 criterion. You would still have that.
11 MR. PRATT: But the question, what you really have 12 to do is propagate it through and sample it so you can get a i
- %, 13 sense of the sampling distribution, the density of the c
i 14 calculations, so you can get some confidence of where you i 15 are. i j
{ 16 Again, if I could refer you to the calculations
! 17 that were presented earlier from NUREG-1150. '
1 f 18 If I could refer you to the NUREG-1150, volume 1 ,
1 l 19 figure, ES-11, it does appear in several. cases. This is the 1 I . 20 Surry result that was presented to you earlier. We move to {
21 the Zion results and you see a clustering downward of the l 22 calculation. What I am doing, is if I look at that and 1
j 23 compare it against Seabrook, where I have a generally higher
- 24 yield than I do with Zion. My gut feeling is that this black i
t 25 area down here is very low'at Seabrook, but I don't know how I \- .
i i ACEJFEDERAL REPORTERS, INC.
t 202 347 3700 Nationwkle Coverspe_ mb3%%86 l
30312.0 ccx 40
. %, c 1 far the tail goes up, because I don't know how they overlap. ;
l 2 MR. ROSSI: That is the first option we talked .
I 3 about this afternoon, that the loads with the uncertainties i
! 4 are less than what the containments would take. l i
- 5 MR. PRATT: I think this does show the curve as i
i 6 well.
1 7 MR. MOODY: To add to that, Trevor, if you look at l
8 the NUREO containment yield pressure, they are both, as you l 9 said, Surry, Zion, Seabrook, is 20, 30 pounds higher. But if 1 10 you look at the construction of the Seabrook containment 11 versus the other two, the structural design is different.
l 12 Its capability beyond yield is better. Its typical design is 1
(' 13 better. We don't have a structural engineer here, but i 14 reenforced containment of Seabrook is much better.
15 I think I was down here on loads, sounds 16 familiar. The last two source terms and consequences, :
17 Brookhaven and we agree that the source terms are i
i 18 appropriate, there was reasonable comparisons between CRACIT, l
l 19 CRACIT-2 and MACCS. Some minor, I don't know if I call them i .
20 disagreements but issues, some comments by Brookhaven about 21 the conservative of S-1 source terms versus 1400 washing ;
1 l t
22 methodology and different calculations for S-1.
23 MR. PRATT I might add to that, in terms of the l a
24 S-1 release, that was just a capture. The main point of it l 25 was whether or not we thought the calculations were
\_-
i I
- /\CE. FEDERAL REPORTERS. INC.
4 202 347 3700 Nationwide Coverase 800 3bMW6
r_., d --4,- 4-e+ A -4i. _ -4__ s au _m .A_-. . , Aa4 11.-a- JL--+- - _ . - as A. .J_( --r .4 -_4 -4 --4-- * -+- - A N.
i
) I 30312.0 cox 41 ,
%s .
1 consistent with Wash 1400 methodology. The only question was 2 could they ever be exceeded, and we have this document they 3 might be.
4 MR. MOODY: We have assigned a probability for our 5 S-1-W, 1.0. If we took a NUREG-1150 type of source term, I 6 am not sure it would make much difference.
7 MR. BARRETT I want to clarify that point about ;
8 the source terms. Does the Brookhaven report actually say 9 that the source terms are conservative, that point of 10 agreement in the middle there, source terms are appropriate 11 and conservative. I thought the Brookhaven report simply 12 concluded they were a good representation of Wash 1400.
A., 13 MR. PRATT That they were appropriate to Wash 14 1400. For the source terms calculated, that in general Wash 15 1400 methods give you higher releases than the new method, 16 but if you are dealing with the early, you get high releases, 17 then we would calculate the Wash 1400. I think that's 18 consistent with the 956.
19 MR. BARRETT The source terms that are most I
20 important to this particular waiver request are the early 21 failure systems bypass.
22 MR. PRATT The S-1-B. '
23 MR. MOODY: It's very small. Doesn't impact the 24 results.
25 MR. FLEMING: I will have to disagree. I don't L
Acc. FEDERAL. RuronTens, INC.
2n2447 3700 NadonwWe Cmcrase mn)% M
o I 30312.0 cox 42 l
1 think that the ability to defend the waiver request hinges on 2 whether we agree what conservative source term is for S-1-D.
3 I want to say I don't think it's important to 4 resolve whether our 5-1-N source term is conservative or
! - 5 not. I think it's important to shiw that probability is 6 acceptable as smaller, i 7 MR. PRATT: Absolutely. I think the difference 1 !
8 between whether you are putting lithinium out, that's an l
9 unacceptably high release. You have to get ritt of it by i 10 probability. [
1 fl 11 MR. FLEMING: Exactly. ,
j i 12 MR. FRATT: That's why I relegated that to an i
. L 13 issue.
14 MR. MOODY: I am going to discuss the shutdown j 15 events first in a little more detail. First of all, the l 16 submittal we made earlier, RAI-121 was a bounding and 17 conservative analysis. We used NSAC-64 as a guide. It's the !
! 18 only in-depth analysis we are aware of that is available.
.1 19 Hasn't been included in the SSFSA, hasn't been included in
} 20 NUREG-1150, NUREG-0396 or any other FRA.
l 21 MR. FEIGENBAUM: Jim, could you speak up, please?
1 22 MN. MOODY: Do you want me to repeat that?
23 MR. FEIGENBAUM: No.
24 MR. MOODY: We are still confident that the
] 25 analysis was conservative, but we are following Brookhaven's L
l
! Acc.FeneRAt. REPORTERS. INC.
l l 202.H7.n00 NationwWe Covetase mO46m l
!. I
'r ,
30312.0 i cox 43 I
1 suggestion to look at it in greater depth. We are going to j i
2 follow the NSAC-84 outline and methodology. Our comments on l
) 3 that seem to be positive. l i
1 i 4 We are going to perform a more in-depth and
- l. 5 6
updated review of our permits in events, and we have done a review of events since '82 and '83 than was available in t
l 7 NSAC.
! 8 We are considering seismic events, fires and
) 9 floods, i '
10 MR. ROSSI: You are looking at spent fuel and
)
11 things that can occur during refueling and that sort of thing
]
l 12 also, spent fuel storage?
l i s, 13 MR. MOODY No. We are looking in the reactor i i
j 14 cavity. Not in the spent fuel.
1 15 MR. NEWBERRY: Have you thought of the spent fuel ;
i) l j 16 type events? i
! i
! 17 MR. MOODY We did. r
! i i
18 MR. ROSSIt I think when you look at accident !
I t f 19 shutdown analysis events, you have to look at the ones that l
{ 20 might be there and put them to bed. It would appear to me at I 1
21 22 some point in time the question of spent fuel storage is going to come up, and you might as well address an issue that
)
l lP 23 has been through this. If it's c'learly not an issue, you i j 24 might write down a reason why it's not an issue in enough ,
l 25 detail, and same thing, anything else that you run into. l i A. l r
! Acu.FennnAL RuronTuns, INC. !
l l 202.M M a n N aik m . We cov m se muswa !
i, !
30312.0 ccx 44 1 What you are really being asked to do is look at events that 2 are important here that have not been important on other 3 plants, where you are not looking for the low probabilities 4 of releases that you are looking for here. So you ought to 5 look, do your own look rather than just use the methodology 6 of something that's been done in the past. Do others agree 7 with that on the staff?
8 MR. NEWBERRY: Yes.
9 MR. MOODY I know we looked at it some time ago, 10 but we will revisit it.
11 Evaluating Seabrook-specific procedures with 12 regard to shutdown and specific design features. I am going
%, 13 to requentify the sequences, the data, include uncertainty, 14 revisit the site-specific source terms and consequences for 15 these events, and we will evaluate potential improvements and 16 procedures or instrumentation, whatever is necessary for this 17 event. We plan to complete this by June 1987.
18 Now, earlier, I indicated that our best estimate 19 analysis had no impact on our results. We are going to go 20 through this again anyway in greater detail, to insure that.
(
21 We also believe that there is a potential solution 22 to this in the short term, if necessary, if this time 23 schedule is too long, that could include restrictions on the 24 equipment hatch, if that type of review could be done much 25 quicker, so there are other shorter paths we believe need to ACE.FEDl!RAl. Riti>onTens, INC, j 202.H7-)W Nation *6de Cmersee makiwm*.
I
l l
l 30312.0 ccx 45 ,
1 be resolved. .
2 MR. NOVAK Don't take the slide off yet. Trevor, 3 do you think the areas of uncertainty are reasonably covered l 4 there. In other words, to the extent we can articulate in !
5 the areas for investigation during shutdown, we ought to deal 6 with it at this time. ,
7 MR. PRATT You think that's fine. i 8 MR. NEWBERRY: What does the first bullet mean, 9 Jim. Are you going to have a model of the Seabrook system?
10 MR. MOODY: We are going to use the similar !
11 methodology and approach that was used in NSAC-84. Of 12 course, we will consider the comment that was provided on
%, 13 N-84 as well as Brookhaven.
14 MR. BARRETT Are you going to look at the 15 pressure values of the RHR system?
16 MR. MOODY: Yes, overpressurization, draindowns, 17 loss of coolant. Did you want to add something?
18 MR. NOVAK No. I just want to make sure we have 19 got our comments on the table now. :
20 MR. FLEMING: I just wanted to clarity by using 21 the same methodology, what we mean by that is to follow the 22 approach of developing operator action based on seabrook
/ 23 rather than -- but extending the procedures out to release 24 dates, that was done in 21.
25 MR. FEICENBAUM: Jim, I imagine sometime before
(.
Acn.FeonnAL RuronTuns, INC, l 2G14M7m Nanon*We Contsee mwou M.46 ,
30312.0 cox 46 9
L_ !
1 June of '87 we will have a more detailed work plan for what 2 we are going to do in this area. Maybe it would be 3 beneficial if you went over that with the NUREG and their f 4 contractors when it's available so we can talk more in depth '
, 5 about any loose ends that might not be included there.
6 MR. MOODY I think we need that kind of input as 7 to what they feel is necessary and -- !
8 MR. FEIGENSAUM: What we are looking at today, is i 9 that kind of approach reasonable? l 10 MR. MOODY Again, I think we should explore other 11 ways to resolve it in a short term, if it's necessary.
12 MR. NEWBERRY: Will this include external event l 4., 13 initiators? i 14 MR. Mo0DY: Yes. Any other questions?
15 okay, next, I will discuss the unresolved 16 phenomenological issues, induced steam generator tube 17 rupture, and the containment loads, direct containment 16 heating loads, t
19 This is a very simple representation of what it 1 20 takes to have this type of an event, early release. We have ,
t 21 a high pressure core melt and dry steam generators. The 22 operators tail to depressurine and the probability is that 23 the tubes failed.
24 What I have here are the two calculations that 25 Brookhaven provided as sensitivity studies. ,
w I
Ace.FeneRAi. REPORTERS, INC, m.m.mm Nmkin*We Cmernor m m.) k M l I
t 30312.0 I
cox 47 1 Here we have the lower calculation, which has no l
2 impact on the results. Then their upper bound calculation 3 that has some impact on the results. We discussed this [
4 morning -- that is the next slide.
5 We provided some detailed analysis in RAI-47 on 6 steam generator tube failures. We found there was quite a 7 bit of margin for the tubes. We don't believe they are going 8 to fail. The map code was used that models the three 9 circulation paths, and as reasonable benchmarking with the 10 Westinghouse experiments.
11 Another option in this area is to pursue NUMEG 12 review of the MACCS code and verification or perform
(, 13 ,
additional analyses. We are beginning to back away from this I t' area. We are going to stay on the record as stating we don't
(;
believe this is a real issue. But it looks like a long-term 16 resolution. We are not prepared to put our eggs in that 17 basket.
18 on the front end, I think Brookhaven or the NONEG 19 recognized that this value that we provided was a
. 20 conservative frequency. It's a mean frequency. We have j 21 looked at it since and the mean value was actually 4 times 10 22 to the minus five. We haven't calculated a mean value, but 23 it's probably closer to 10 to the minus 5. We began, in a 24 preliminary way, to evaluate sequences, procedures, and 25 consider potential improvements in this area. We also began w
AcibFimenAI. Riti'ontains, INC.
n m.no Nemm.We Commee snamm
30312.0 ccx 48 l 1 to look at procedures in the tech support center, training, 1
2 simulator use on some or the sequences, evaluating PORV I 3 capability environmentally and additional analyses. We have 4 done some preliminary work in this area, and I guess we are 5 reading for the second option.
6 MR. ROSSI: Basically what you are saying is you l' 7 are going to go the route of procedures for criterion in 8 pressurizing, and state that can be done with high confidence 9 rather than pursue the one that says if you stay at high 10 pressure, there really isn't a significant probability.
11 MR. MOODY: We think this probability is very 12 small.
%, 13 MR. ROSSI: You are not going to try to use that 14 as your basis.
l 15 MR. MOODY We are not going to try to resolve 16 that.
17 MR. ROGGI When you do the depressurization, you 10 will, I assume, identify the pressure to which you have to go l 19 and demonstrate that you can, indeed, get to that pressure, 20 and that that pressure will solve the problem?
21 MR. MOODY Yes, that will be addressed.
l 22 MR. ROUSti Warren, do you have anything to add 23 about things they will be concerned about when they go that 24 route?
25 MR. LYON: I have been sitting here thinking about w
ACit Fitnitari. Riti>onTitus, INC.
20M4MM NanonwWe Cmerne *n m u.e
I n t
l
} 30312.0 '
cox 49 l .
I u l
1 some of this as you have been talking. As you know, Ernie, I l
2 have a presentation put together. I am thinking in light of 3
the flavor we are seeing with some of the other comments, I 4 should throw most of that away. There's one or two slides 5 that I think would be appropriate to discuss when we get to 6 it.
. 7 MR. MOODY: Okay.
6 MR. BARRETT: Jim, you mentioned that the high 9
pressure core melt frequency is -- that this might be an 10 upper estimate of it. Is that based on going back and 11 quantifying, or reexamining, or are you contemplating --
12 MR. MOODY When this was provided, we took a look
(, 13 at the plant sequence dates. We knew we had procedures that 14 didn't apply. We tried to bound it rather than go through a 15 great number of sequences. Since then, we have had an 16 opportunity to go back and look at it. It's really on the 17 order of 3 times 10 to the minus 5 rather than 4. This is 18 really a mean, not a median. Median will be less than that.
19 MR. BARRETT You are not contemplating any
. 20 changes to the plant to improve it?
21 MR. MOODY We are considering those things, but
!* 22 we are not leaning in that direction significantly.
l 23 1 In these sequences, many of those sequences, we 24 believe, have a potential for recovery that hasn't been taken l 25 credit for, so the number could be smaller.
w Acti.FlintinAi. Rr:PonTrns.INC, 202 M S m NMiamakle Omange muMAMM t_ _ _ _ _ _ _ _ _ _ _ - - _ - - - - -- - - - - - - - - - - - - - - - - - - - - - - - - - - - ~ ~ - ' ~ ~ ~
30312.0 50 cox 1 MR. NEWBERRY: How much of that is ATWS? Do you 2 recall?
3 MR. MOODY: It's a good contribution, maybe 50 4 percent.
l 5 MR. NEWBERRY: 50 percent?
6 MR. FLEMING: Not with dry steam generators. We 7 do have a significant frequency of ATWS.
8 MR. MOODY: On the next slide.
! 9 MR. FLEMING: On the order of 10 to the minus 5 or-10 grea,ter, but those have higher probability of steam generator 11 core.
12 MR. ROSSI: Are you working at all on reducing, I
\.- 13 am not sure it will even help,.in reducing the ATWS 14 probability?
15 MR. MOODY: We have looked at it in a preliminary 16 way, yes.
17 MR. ROSSI: I don't know what a diverse reactor 18 trip would do, for example.
f 19 MR. MOODY: The post-Salem, some of the post-Salem o 20 changes actually made may help us. -
21 MR. FLEMING: The logical proceus to lead us to 22 resolve the outstanding issues, we haven't getten to.the i
.23 point of reducing ATWS per se. That, in and of itself, is 24 l- going to directly attack these issues. We are focusing on 25 new stem generator tube ruptures or dry steam generators.
(-
/\CE FEDERAL REPORTERS, lNC.
202-347-3700 Nationwide Coverage 800-33MM6
f 4
30312.0 cox 51
's 1 For direct containment heating, all core melt sequences.
2 In that context ATWS itself is not of great 3 significance.
4 MR. NEWBERRY: The thrust of my question was to 5 see if the ATWI contribution was there. I know other studies 6 it was not, very low probable, as I recall. So that's why I 7 asked the question.
8 MR. MOODY: Early containment failure, we have 9 similar boxes in, I think 6 times 10 to the minus 10 is the 10 upper bound. We have similar things, looking at here, except 11 the frequency is the total core melt frequency, in essence.
12 So there were a great number of sequences.
As 13 MR. ROSSI: You have to be able to justify that 14 the pressure you depressurize to is a pressure which will !
15 eliminate the problem, if you go that route. Also, you have i
16 to go back in and look for sequences that maybe created this, 1 1
17 and may be made worse from the depressurization, so you need )
18 to do that too.
19 But I think the justification of the pressure 20 would be very important.
21 MR. MOODY: We are looking at the negative effects 22 of depressurizing?
23 MR. ROSSI: Yes.
24 MR. MOODY: Again on this one, I have pretty much 25 given up on pushing this path. We don't believe it's an ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
30312.0 1 cox 52 v
1 issue and yHe will continue to state that on the record. But 2 we see the containment capability beyond yield as something 3 that won't be resolved that quickly. The containment loads, 4 NUREG-1150 issues are just too long-term.
5 Again, we are focusing in onto pressurizing.
6 First of all, for most of what we have read about 7 these issues, there seems to be a lot of concurrence that 8 pressurization does solve most issues.- One thing we would 9 like to do, as soon as possible, is to continue to get.that 10 concurrence.
11 MR. ROSSI: Clearly, the NUREG will agree with the 12 depressurization number 2 under several caveats. One is that
(-- 13 it is demonstrated it clearly works. You know what the 14 pressure is you have to go through to do it, and you make a 15 good demonstration that it couldn't create new problems. It 16 may ne when people get'into understanding it better, if you 17 find that there are some new problems created from it or some 18 new phenomena that now you have to worry about under 19 depressurization, that will have to be looked at.
20 of course, the third thing.is that you have to 21 have criterion procedures that are deemed to be clear and 22 have a high degree of working. You know, I would hope that 23 the depressurization doesn't end up to be something that 24 until you study it looks good, but when you study it, you 25 find it creates a new phenomenon or new problem that gets you (s -
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646
m - ;,., , ;; ~ ,
30312.0 cox 53 1 back up to where you were to start with. If that happens, 2 obviously that wouldn't be a solution.
3 MR. MOODY: We have taken a preliminary look and 4 we have considered the types of things that you are 5 addressing. It looks promising.
6 I think in about a month we can put some kind of a 7 report together that addresses this.
8 MR. ROSSI: Your first bullet of obtaining NRC 9 concurrence with depressurization solution, it sounds like 10 it's worded in an oversimplified way. If there is a good 11 technical basis for doing it, that the experts on the staff 12 and our expert consultants will accept, and there is a' strong s- 13 assurance that it doesn't create new problems, then that will 14 be the criterion. But it's hard for us to have an up front 15 concurrence that that is the way you ought to go. We need to 16 look at the downside of the procedures of all of that too.
17 MR. MOODY: We are going to evaluate --
18 MR. TORRI: May I add something. What is meant 19 there really is that we would like to be confident that you
= 20 do consider their statement a successful path in resolving 21 the issue. If we are successful in providing the information 22 that constitutes resolution, as opposed to, for example, 23 having to do both this as well as other things, that might 24 address what you might call residual concerns.
! 25 MR. NOVAK: Let me speak to that.
's ACE-FEDERAL REPORTERS, INC.
- - 202 347-3700 Nationwide Coverage 800-336-6646 I
30312.0 .
cox - 58 L
1 MR. TORRI: We don't want to go down this path for 2 three months and then find out what you consider a viable 3 solution is something entirely different.
4 MR. PRATT: I can understand that. On the other 5 hand, we can't make any promises that this approach will, in a
6 fact, give the results that you want. I think it's an open
. 7 point. I think everyone I have talked to on the staff feels 8 that depressurization is an approach that should be looked 9 at. But it has to be looked at, really, very carefully.
10 I don't know that a month is enough. I am 11 surprised. I would really suggest you exercise a lot of 12 looking at how this would be managed, where the decision ss 13 process would come from, not that there is just a procedure 14 in a control room. That is not going to cut it. You know it 15 and we know it.
16 So you really have to -- if you are going to 17 i follow down any particular avenue, it's got to really have a 18 strong basis that would support it and can be scrutinized. I 19 think that's a major point, because I think what you have to 20 do is you will have to compare decisions such as-this to 21 other very difficult decisions, and whether or not people 22 will make them when they have to. I think that's just -- I 23 would leave that point a very strong point, that we are not 24 ruling out depressurization as a viable option. What we are 25 saying, though, is it has got to be a damn good approach
(-
ACE-FEDERAL REPORTERS, INC.
[ 202-347-3700 Nationwide Coverage 800 336-6646
30312.0 -
cox 55 s.
1 before it will sell.
2 MR. ROSSI: I guess no one in the room at this 3 time knows of a fatal flaw in that approach. If they do now 4 they ought to bring it up now. We may discover one if we 5 don't know it today.
6 MR. FLEMING: I know Jim will make the point in 7 the next slide, but I think it should be made now, it may not 8 be in everyone's mind up front now that this idea of 9 depressurizing the reactor coolant system is not a novel 10 idea. As a matter of fact, it's already in _the training and 11 procedures for the existing plant for a broad class of 12 accident sequences. It is not in there for all of them at 4, 13 this time.
14 So by introducing the opening of the PRV, we are i
15 not talking about a major innovation in the concept of the I 16 procedures.
17 MR. NOVAK: I disagree with you. At this last 18 meeting, your operations people had a contrary view, that 19 they were very hesitant. They were not effectively
- 20 suggesting that this was the way to go.
21 MR. MOODY: We were talking, I believe, about the 22 TMLB prime sequence. It.is~the exception for the functional 23 restoration guidelines.
24 MR. NOVAK: 'Depressurization is what I thought we 25 were talking about, very broadly.
%. l l
ACE FEDERAL. REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6M6 1
30312.0 cox 56 V
1 MR. FLEMING: In functional restoration guidelines 2 for loss of core cooling at Seabrook, for sequences with AC 3 power available, it explicitly said step 17, to open up the 4 PORVs after exhausting other opportunities. The concerns 5 that were raised and the caveat that is in existence right 6 now is significant loss of power.
7 MR. .LYON: Which is the major contributor to the 8 situation with the steam generator secondary dry core melt 9 and so forth; correct?
10 MR. FLEMING: Not the major but a major.
11 MR. LYON: What is the major?
12 MR. FLEMING: The second sequence, per sequence s- 13 happens to be a loss of 1 DC bus with a transient hold by I
14 failure of emergency feed water. That happens to be a little 15 bit more likely than this particular sequence. It is a major 16 but it is not the major.
17 MR. MOODY: That one also happens to have, we i
18 believe, more recovery in it that we haven't taken care of l l
19 yet. So we will get into that. That is the reason why our '
20 plan is going to address this one first. Then we are going 21 to go back and look in greater detail.
22 MR. BARRETT: Jim,-'would you tell me a little more 23 what you mean by addressing the TMLB sequence. -What are you l
24 going to address?
~
25 MR. MOODY: I think a more important concern about b
14CEJFEDERAL REP RTERS, lNC.
202-347-3700 Nationwide Coverage 800-336 6646
~
30312.0 cox 57 2
1 why we are concerned about the effects of negative l 2 depressurizing is we don't believe this is a real issue. We I i
I 3 are sort of in a risk management paradox here. We don't l
4 think it's an issue, and we are going to do something to 5 resolve it.
6 First of all, in BWRs, the RCS system is manually 7 operated, or it used to be for several years. So these types 8 of operations have been used in the past.
9 We intend to implement it in the tech support-10 center as a decision of procedures, and with training. We 11 think the effectiveness of this solution can be demonstrated 12 fairly easily. We don't need exotic codes, where there may
(, 13 be controversial concerns about the verification of codes.
14 MR. ROSSI: You do need to be able to justify the 15 pressure that you have to get below, and be able to do an 16 analyses to demonstrate that you have the equipment that can 17 get you down to that pressure. That much you will need to 18 do.
19 MR. MOODY: It may be optimistic, but we believe
- 20 it's fairly straightforward.
21 By implementing it in the tech support center, and 22 the whole solution, as its complexity, there is no:added 23 complexity to existing procedures, plant design, which is 24 important when you consider the negative aspects.
25 It offers a solution to both issues, we believe.
x ACE-FEDERAL REPORTERS, INC.
l 202 347-3700 Nationwide Coverage 800 33 M 646
1 4
l 1
30312.0 cox 58
\~
1 MR. NOVAK: Before you leave that slide, two quick 2 questions. One, do you have today a feeling for the earliest 3 time that a depressurization would be followed, in terms of 4 from the initiation of an event.- Have you backed up to say 5 just how soon in an accident environment must a commitment to 6 depressurization made?
7 MR. MOODY: We looked at TMLB prime, a couple of 8 hours at least, Fred?
9 MR. TORRI: Three hours.
10 MR. ROSSI: Does anybody from Brookhaven or the 11 staff have a comment on the soonest you might have to 12 depressurize for induced steam generator rupture or
%, 13 containment heating?
l 14 MR. PRATT: Three hours.
15 MR. ROSSI: Roughly three hours.
16 MR. BARRETT: Depends on the amount of fluid 17 lost.
18 MR. TORRI: If you have too big a loss through 19 pump seals, you don't have to open the course.
. 20 MR. FEIGENBAUM: Jim, that would be part of our 21 complete package that we would submit for review.
22 MR. MOODY: Yes.
23 MR. ROSSI: Let me ask a question. When you say 24 implemented as a technical support center decision, does that 25 mean that the operators would not do it without an ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
30312.0 cox 59 s
1 instruction from the technical support center?
2 MR. MOODY: For the TMLV prime, yes, that is 3 true.
4 MR. ROSSI: The operators would not have the
. 5 authority to make the decision themselves based on the 6 procedures it would take. A decision first from the
8 MR. WALSH: I am Larry Walsh. I made the comment 9 at the last meeting. I am the operations manager. What we 10 are trying to do here is put this decision out into the 11 technical support center so proper management can choose and 12 direct their activities to occur that an operator feels s, 13 differently about. He wants to save the core, and we are 14 doing something differently. That's why the decision was to i
15 go out to the TSC for that decisionmaking process. There is 16 one man in every shift that.is trained at the TSC, and he is 17 the duty superintendent.
la MR. ROSSI: You would always have a person on-site 19 that could make.that decision?
20 MR. WALSH: Yes.
21 MR. NOVAK: Shift technical advisor?
22 MR. WALSH: Yes. He is the senior plant member on 23 every watch.
24 MR. HAN: Jim Han from research. You open one or l
25 both. PRV, the second question is does ours correspond to l
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336 6M6
i i
30312.0 cox 60 L
1 the status of your core? What is the status of your core 2 after three hours?
3 MR. MOODY: I believe three hours was the 4 beginning of the recovery of the core for the TMLV prime.
5 MR. HAN: Our calculation shows two hours, two 6 hours water reaches the top of the core, TMLB but not for 7 Seabrook.
8 MR. MOODY: Surry has steam generators.
9 MR. TORRI: On the Seabrook steam generator it's 10 substantially larger than the Westinghouse plants.
11 MR. ROSSI: What kind of steam generators do you 12 have on Seabrook?
\s 13 MR. TORRI: Model F. I believe it's about 30 or 14 50 percent increase in secondary site of water inventory.
15 Translates directly into extended time as to core.
16 MR. NOVAK: Let me go back to one other question, 17 then you can certainly move on. You mentioned training with 18 simulators. I would expect that your simulator would not be I 19 modeled to follow through on this event.
i 20 MR. MOODY: That's true. We can't model core melt l
21 sequences. We were planning on looking at the front end. l 22 Some have these sequences on a simulator.
23 MR. FLEMING: I would think by the time we would 24 reach the point where we would recommend these course of 25 actions, that the simulator would still be useful.
(-
/kCE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
. . _ _ _ - __ -rm,- - - -~
30312.0 cox 61
\_
1 But to cover all the considerations about adverse 2 effects and all the different accident scenarios that could 3 lead to this kind of concern, simulator has proven in the 4 past to be extremely useful to shake out these kinds of 5 problems.
6 MR. NOVAK: Given I don't have to proceed beyond 7 it, but can the simulator be used as a viable tool to see if 8 in fact a decision is made early enough? In other words, can 9 you use the simulator in a manner to test the decision 10 process in terms of.being timely?
11 MR. MOODY: The timing?
12 MR. NOVAK: Yes.
A. 13 MR. MOODY: The front end part'of it you can.
14 MR. WALSH: Larry Walsh again. You may be able to 15 do that later on, but right now we can't, that is true. The 16 basic principle we were going to use for this is the 17 initiator. To get into the event that causes the need for 18 depressurization. If the simulator didn't carry us on far 19 enough, it would be a walk-through exercise beyond that, 20 range of simulation. l 21 MR. NOVAK: I know we are holding up, but this was 22 really the purpose of the meeting, to try to get as candid a 23 view we can on some of these.
24 I think the staff experience, if you look at 25 experience in terms of stress decisions, whatever you want to ACE FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage H00 336-6646
30312.0 I cox 62 w
I call them, the experience has been that people hold back from 2 making these decisions, and you have got to face that 3 recognition and deal with it. A lot less significant events, 4 people did not follow what they should have followed, 5 perhaps.
6 MR. ROSSI: Because they perceived it as making 7 things worse, that they would be able to recover in some 8 other way.
9 MR. LYON: This is getting into a point that 10 bothers me. Larry, why do you feel this is saying, hey, we 11 bought the farm, if you start depressurizing under a TMLB 12 prime condition?
%, 13 MR. WALSH: Why do I feel that we bought the farm 14 by doing this?.
15 MR. LYON: Yes.
16 MR. WALSH: In 22 years I have always been taught 17 to save the core, not to let it melt.
18 MR. LYON: Why do you feel it's burning the core?
19 The reason I pose this, I have got these four big tanks of 20 water sitting there. I can't get those into the core until I 21 depressurize, and it would seem to me that whether I have got O
22 TMLB prime or anything else, if I am in a situation with dry 23 steam generators, and I can't make up to the reactor coolant 24 system, it would seem to me I could buy more time by dumping 25 those tanks. That's why I wonder if you feel it would be 1
ace-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-334 6646
~
3 l 7 ,
30312.0 I cox 63 Nil I committing to having " bought the farm," so to speak.
2 MR. WALSH: I am considering procedures other than 3 TMLB prime will put those in there. My concern is not so 4 much trying to get the cumulators in, because I think the 5 operators would have thought of that on their own and put it
, 6 in there.
7 My concern is reducing that amount of heat / energy 8 transfer that is in the core, and listing those relief 9 valves, therefore maintaining some continuity of core and l 10 pressure in the vessel and the system, and it gives you more 11 time to recover from whatever it is that is preventing you 12 from adding water. As soon as you open that PORV, you are
(- 13 reducing that time scale in the logic of "I am going to have 14 it open to the atmosphere containment," which'is better for 15 the general public, but it's not necessarily better for the 16 core, because now you are reducing the time scale that you 17 have to react to the situation you are in.
18 MR. ROSSI: Your analyses are going to look at l
19 that. That's the kind of thing you would do analyses to look '
. 20 at, what the down side is, up side is, procedures ought to be l 21 and criterion. Presumably your analyses will look at that 22 kind of thing.
23 MR. FEIGENBAUM: I think what Warren's point is 24 when you depressurize this, what your options are is to open 25 up water into the pressure.
~
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33M486
30312.0 4' .cox 64 l- (..
1 MR. TORRI: I think Warren's point is an excellent 2 one, that we may not have thought about it, and the logic of 3 exhausting all opportunities for keeping the core cooled for 4 as long as possible -- I think he has made an excellent 5 point. At least I have not really thought of the -- I.
6 thought of the accumulators as a way to put water over the
' ~
7 debris after its melted through the vessel to quench'the 8 core. I think in PRA space, a lot of us have thought that '
9 way. We have not thought about getting the pressure down 10 before the core gets damaged beyond repair to bring the 11 accumulators in to refloat the core. We have not taken i 12 credit for those kinds of events in risk analysis.
i l (.) 13 MR. FLEMING: I think another comment that needs
! 14 to be pointed out, though, is the fact that the existing 15 procedures and existing training program and the whole 16 philosophy of the post-TMI lessons learned, did not have.that j 17 issue, these issues, mitigating.these issues as a focal 18 point. We are trying to come back and revisit. The 19 procedures were written without a high credibility with 20 respect to enhanced steam tube rupture.
) 21 In a station blackout sequence, all your options 22 to provide long-term decay heat removal at low pressure 23 depends on AC power. After you get to some point in time, 24 you have to depend on your RHR. When you have no AC you have 25 to rely on steam water pumps. We have to go back and revisit
(_-
i i
ace-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cover
- age 800-336-6M6 i
_ _ . - ~_ -.._ _- _ ., ___ ._,_ _ _ _ _. . . _ _ _ _ _ __
30312.0 cox 65 L
1 that and look for potential adverse effects.
2 MR. NERSES: Would it be fruitful at this point to 3 take a water break for 10 minutes and let our reporter have a 4 break here.
5 (Recess.)
6 MR. MOODY: I think this has been stated two or 7 three times, but let me just go through the slide. The 8 reason we are going to look at TMLB prime is it's a station 9 blackout sequence, and the existing functional restoration 10 guidelines do not handle station blackouts but they do handle 11 the known station blackout scenarios. It's also a dominant 12 contributor to both issues.
s- 13 The elements of the action plan, and somebody over 14 here I think was asking a question, and I hope this answers 15 it, develop a procedure for the tech support center, this has 16 been discussed earlier. I think, already, performing the 17 time analysis for the sequences, questions have been asked on 18 this example, we are going to form some preliminary PRA 19 analyses on the sequences. We have talked about checking it
. 20 out on the simulator, some of them. We will finalize the 21 procedures, the PRA evaluations, initial report. This is 22 kind of general, there are a lot more details to it.
23 MR. FEIGENBAUM: Jim. That would also include a 24 thermohydraulic analysis? l l
25 MR. MOODY: Yes. ;
l s_ l l
/\CE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage MXL3346646 l
4
-30312.0 cox 66 v
, 1 MR. FLEMING: Yes.
2 MR. NEWBERRY: Would the procedure be detailed 3 enough so that you talk about criteria or plant-specific 4 symptoms in this procedure?
5 MR. MOODY: I believe that's correct.
6 MR. WALSH: Yes, definitely.
7 MR. MOODY: Definitely. This was up here 8 previously, I think we should revisit this. We can skip the 9 first bullet.
10 But I think within about a month, we can submit a 11 report, and we have been working on this for a month 12 already. There was some comment about our ability to do 4., 13 something in a month. We have done quite a-bit of 14 preliminary evaluations on this already in the last month.
15 Also, I believe, a meeting within the next couple 16 of weeks, we should schedule a meeting, I would like to not 17 discuss this, but the shutdown events and develop a more 18 detailed plan for us evaluating these issues, submitting i 19 additional information to the NUREG and their review
. 20 process. , t 21 We will continue after that first month or so to 22 look at other sequences in more detail, and also resolve i 23 comments, questions from the NUREG in their contracts.
24 We believe that this process can be completed in !
i 25 the July time frame. We think it's a reasonable schedule.
1 I
/\CE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage H00-336-6646 l
T ~
l 30312.0 cox 67
\) . l 1 Anybody else?
2 MR. NEWBERRY: Jim, would you please step back.
, 3 Maybe you have already made the point on why you are focusing l 4 on TMLB prime for the high pressure sequences. Are there not 5 other sequences that make up the 4 times 10 to the minus S?
6 MR. MOODY: Yes. The reason we are focusing on 7 that one, the procedures the operators now have do not 8 address the station blackout sequence. So for that 9 particular' sequence, we have to go to the tech support center 10 to make the decision and implement opening the PORVs --
11 MR. NEWBERRY: Where you have dry steam 12 generators, you are saying that the emergency procedure i
\.- 13 guidelines already correct your pressurization?
f 14 MR. MOODY: Yes. We are still going to go back .
15 and look at that after we have done the TMLB prime.
I
! 16 The next to the last bullet, in addition to 1
17 resolving comments, we are going to complete the evaluation 18 and look at sequences that already go into the functional 19 restoration guidelines, look at the procedures, existing
. 20 procedures, and evaluate those. Any other comments, 21 questions?
22 The last slide addresses the fourth area, and 23 probably picks up the issues 5 and 6, containment bypass, the 24 supporting of continual review in those areas, and systems 25 analysis on the front end, which I suspect has to do with ss '
ACE-FEDERAL REPORTERS,'INC.
202-347 3700 Nationwide Coverage H00 33MM6 F --w --w- ,y-. .< _
g.. . + , . , ~ - _ . _ . - - - . - >.w- -.-. - , - - - , - .s, -----.-,-,-r.-
. -,,--,.,,-..v....
1 30312.0 cox 68
- v 1 missiles and other ways of having early containment 2 failures. I assume that would be the scope. We have looked 3 systematically a number of times, and we have provided some 4 different forms in earlier presentations, the ways in which 5 you can have an early large release. Of course, we have 6 already talked about structural capability and core 7 phenomena, interfacing bypasses, we have looked at that, and 8 we have looked at it a couple of times.
9 We have identified six paths where there are two 10 boundaries, two check valves and series, two MOVs, all of l
11 their paths are very small or they have'more than two 12 isolation boundaries. We have looked at hatches, purge
\.- 13 lines. I believe part of the purge line' consideration 14 probably hasn't been reviewed, containment isolation systems i 15 analysis. I believe this is still unfinished from the l 16 Brookhaven report. We have in the Seabrook PRA addressed 1
l 17 missiles, seismic events, and their impact on early releases, 18 and where we are prepared to support you in reviewing those 19 areas in the future. Questions?
I- 20 MR. FEIGENBAUM: Maybe you could explain to us 21 what you plan to do in those areas and how we might be able 22 to support you.
23 MR. MOODY: That question wasn't to me, was it?
24 MR. FEIGENBAUM: To the NUREG.
25 MR. NEWBERRY: Couldn't hear you.
\-
/\CE. FEDERAL REPORTERS, lNC.
202-347-3700 Nationwide Coverage MD 336-6646 I. - , . _ . _ _ _ _ , . _ _ _. _.
30312.0 cox 69 s
1 MR. FEIGENBAUM: In the Brookhaven report, .
2 Brookhaven indicates a more thorough systemic review of other 3 possible early release categories. We would like to have 4 some idea of what is going to be done in that area and what 5 might be done in order to support that.
6 MR. ROSSI: We are going to be going back and
. 7 looking at the SSPSA and looking to see if sequences have 8 been left out that ought to be in there. That's basically 9 what we are going to be doing.
10 We are going to have contractors do that work.
11 MR. FEIGENBAUM: You don't need anything from us 12 at this point?
%, 13 MR. ROSSI: No. But at the time that we start 14 that work, we will probably -- would probably be helpful to 15 have some meetings where we guide our contractors through .
16 that, because in the past we understand that was a problem, 17 how to follow it, that sort of thing. Any tutorials we might 18 want might be helpful at some point in time.
i 19 MR. FEIGENBAUM: We would certainly want to do
. 20 that.
21 MR. ROSSI: It would be the the kind of thing 22 where you give us guidance on what is in there, how it is l
23 laid out, that sort of thing, that make it easier for our 24 contractors and our staff to go through and do the review.
25 Do you have anything?
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646 e
30312.0 cox 70 y
1 MR. NEWBERRY: No.
2 MR. MOODY: There is a focus here on early release ,
3 type of sequences and how they were done. I assume that is 4 the focus.
5 MR. ROSSI: That would be the primary focus, yes.
6 MR. NEWBERRY: We are still talking about how best 7 to do that. If you have any ideas, I guess you would like to 8 hear that. One alternative is to do a top, down, inside and 9 out review of the PRA. In some cases that didn't seem to be
- 10 totally fruitful. There are certain areas we probably don't 11 need to spend a whole lot of time on.
12 Clearly we are interested in establishing a better
(.. 13 understanding of the likelihood of a large early release.
14 MR. FLEMING: If I might just throw out some 15 suggestions there, it's from all the studies done, the 16 sensitivity study and the discussion of this particular.
17 waiver request and so forth, it's -- I think it's been pretty 18 well demonstrated that in the release categories that we have 19 been dealing with with this PRA, the only two release 20 categories that have a significant contribution to 21 challenging our case, or associated with the S-1 and S-6 22 release categories, that pretty well limits you to a couple 23 of narrow areas for review, including the things identified 24 on that slide. The large equipment hatches, which is the 25 concern for shutdown cooling events, that those might be open v
ace-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-3 % 6646
30312.0 cox 71
\-
1 and tried hard to shut. The purge lines, which one aspect of 2 the PRA scenarios leading to the chance that the purge lines 3 might be open during the sequence and the chance that they 4 might fail, the specific set of scenarios and specific plant 5 damage states for the purge line sequences. Those are 6 dominated by seismic events.
- So if one focused on the F states, 3-F, 1-P and 7
8 7-F, for example, you could focus on a review of the seismic 9 analysis of those particular sequences.
10 The only other ways to get large holes in the 11 containment are missiles. We have look at aircraft and 12 turbine missiles, and those analyses are specifically s- 13 identified in the PRA. You could conceivably break the 14 containment structurally due to an earthquake. We have 15 discussed the seismic capacity of the containment due to 16 seismic loads. That's pretty well identified in the PRA.
17 As far as the interfacing loca is concerned, the 18 only large releases that contact with the reactor containment 19 entry and go outside the containment, either have three or
- 20 more normally closed checked valves or motor operative 21 valves, or they are in the analysis, and some review of the i a l 22 plant documentation can be done to confirm that. )
1 23 The steam generator tube rupture initiating event l 24 does not seem to be a viable candidate to get you into a 25 large release. The way we have analyzed it in the PRA, we v
ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage H00-336-66W.
m
30312.0 cox 72
! s_.
, 1- analyze a core melt frequency, which is dominated by problems i
2 in the support' systems following the initiating event, with 3 the high probability of isolation.
2 i 4 But given that there is a failure to isolate, l
I 5 which we calculate low probability, those are S-2 releases or 6 small holes. I don't know how fruitful a review of the steam
!* 7 generator tube rupture would be, even though you may not 8 agree with every particular detail of that analysis. 1 2
9 I think from the standpoint of S-1 and S-6, the j 10 fraction of the front end plant model that is really l 11 pertinent to those release categories, could be carved'out i
l 12 and identified.to help focus your review. I can't think of
) s, 13 any other conceivable way to get a large hole in the
- 14 containment that isn't characterized on that slide and 4
j 15 doesn't have either extensive resolution paths already l 16 identified. And also in the RAI information that was l
i 17 submitted, there was lots of additional' checks made, more- i l
l' l
18 documentation provided, all the penetrations went to the l l
j 19 containment.
1 .
. . 20 We took a conservative shot at NUREG-4220. I am 21 just really at a loss to know what else you will learn at'the.
- 22 front end model that would be pertinent to S-1 and S-6.
23 MR. NEWBERRY: You may be right.
24 MR. MOODY: I think NUREG-1150, which is now 2
25 available in draft form, has made some attempts to also look i
/\CEJFEDERAL REPORTERS, INC.
k 202 347 3700 Nationwide Cmerage mXb336-6M6 i
., . - . - , . . - - _ _ _ - . . ~ _ - . . - - , - . . - - - - -__ . .- _ - -- - -
i 30312.0 cox 73 s
1 at early large releases. People have been trying to do this 2 for the last 10 years, since Wash 1400.
3 I guess I will put the summary back up and 4 concentrate on the second two issues. We still feel that the 5 best estimate calculations that Brookhaven has identified, 6 even though they may not be complete analyses, supports our 7 conclusions, and notwithstanding that, we have plans and we 8 have a course of action to address the three major issues.
9 We just talked about the completeness search for early i
10 releases. We are prepared to start new that area, front end 11 review of whatever supplier.
12 MR. NOVAK: Do you have any concentrated special s, 13 effort in the review of the NUREG-1150 document as an 14 utility?
15 MR. MOODY: I wouldn't say a concentrated effort.
16 MR. NOVAK: Dedicated, 25 hours2.893519e-4 days <br />0.00694 hours <br />4.133598e-5 weeks <br />9.5125e-6 months <br /> a day?
a 17 MR. MOODY: We are reviewing it. Out of 18 interest --
1 l
19 MR. ROSSI: Do you have your consultants looking
- 20 at it? l 1
21 MR. MOODY: Yes. I suspect we will comment on
=
22 it. I am not sure I would consider it a top priority 23 effort.
24 MR. FEIGENBAUM: Why do you ask the question, 25 Tom?
(-
ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336-6646
A 30312.0 74 cox i
\s 1 MR. NOVAK: Well, I think it's germane to New i
2 Hampshire's view -- I would expect you would express your 3 views as to the credibility of certain phenomena as they 4 apply to large dry containments. You could speak 5 generically. Certainly I think you have got a lot at stake, I
6 and the emphasis that you place on your work has to somehow 7 be consistent with what the generic concerns that are 8 expressed in 1150 would suggest.
We have initiated review of it.
4 MR. MOODY: We 9
I 10 haven't made a decision as to how extensive we are going to 11 comment on it.
i l 12 We are going to comment on two areas we want to s,
o 13 comment on, I think that's the containment area and their 14 treatment of uncertainties.
15 MR. NOVAK Is your formal presentation over?
16 MR. MOODY: Yes. I think we addressed four out of 17 the six issues, the major issues that were identified. The 18 other two issues on design basis accidents and ad hoc 19 considerations, I think we would plan to address that in a 20 future meeting, unless you want to ask specific questions.
21 MR. NOVAK What I would want to do, it seems to 22 me that we need to get from you, we can look at the 23 transcript, but I would suggest that you formalize the l
. 24 efforts now that you are going to go through in as much 25 detail as you can to give us an appreciation of what you are l s-l l
Acs FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage WXb334 % M
30312.0 cox 75 1 going to do beyond what is in your application to date.
2 MR. MOODY: For these issues?
3 MR. NOVAK: Yes. Articulate them as best you can 4 so that we can have an opportunity to read them and 5 understand what additional avenues you are pursuing.
6 MR. FEIGENBAUM: Yes, we plan to do that. Along 7 that line, Tom, I don't mean to minimize the extent of the 8 effort. We certainly understand what is involved here in 9 resolving these severe accident phenomena. However, our 4
j 10 schedule, the way we have laid it out so far, we didn't show 11 you our detailed schedule, but we have put some thought into 12 it already. Unless we run into some severe technical s- 13 problems, it looks like we will be able to submit all of the 14 information that we plan to provide you to resolve these 15 issues, at least these four that we talked about today, by 16 sometime in the midsummer, July of 1987.
i 17 We would like to know whether you can support that 18 effort in reviewing these reports that we are going to 19 submit, and whether we possibly can lay out a schedule to see
. 20 where this all brings us, as far as a review of the 1
! 21 petition.
22 There was some indication provided to the board 23 that it would take until November of '87 to finish your 24 review, and maybe we should talk about that a little bit.
25 MR. NOVAK: We can. One comment, which is v
ace.FEDERAI, REPORTERS, INC, 202 347 3700 Nationwide Cmerage luo-13MM6
l 30312.0 cox 76
\_
1 probably an obvious one, the better job you do, the easier it 2 is for us. Our efforts will pretty much come from what you 3 can say conclusively on the record, and this is no surprise.
4 If you can make a very strong case for a given point of view
, 5 and it has technical substance, then I think we can 6 objectively look at it. If we agree, fine, and if not, we 7 can so articulate. I don't know of any way to commit to a 8 schedule until we know in a lot of detail just specifically 9 what your plans are. Then I think we can certainly work with 10 you on the schedule.
11 MR. FEIGENBAUM: Jim, the next step would be to 12 submit packages as far as describing how we plan to address i
%- 13 these issues. What kind of time frame are we talking about?
14 MR. MOODY: We are planning to submit, within a 1
3 15 couple of weeks, our comments on the final report. We would
- 16 hope to fold into it what you are requesting. I think we 17 would also suggest a meeting within a couple of weeks to look l
18 at in more detail our plan for addressing these issues, and l 19 possibly our plans for addressing them and reviewing them.
20 There may be some things that NUREG can begin to do already, 21 for example, like independently looking at blowdown 22 calculation for depressurization. I am not sure that NUREG 23 necessarily has to wait for us. But I think a meeting within 24 a couple of weeks to discuss in more detail our plans, as 25 well as detailed comments from a final report. There may be
%- l ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8m336-6646
).
30312.0 cox. 77 y
1 things, for example, on the shutdown events, that we could 2 submit as pieces of that package.
3 Right now, we are anticipating around the 1st of 4 June, for example, on completing that evaluation. It's not 5 clear that we ought to wait that long to submit information l .
6 or there may be some specific additional more detailed
- . 7 questions that Brookhaven or the NUREG has that we can 8 address.
9 MR. FEIGENBAUM: The way it would work is we would l 10 submit our plans for closing out these issues in a i 11 narrative-type format, and give the NUREG an opportunity to 12 look at it, and have a meeting to go over it in more ' detail, 4, 13 each individual plan?
4 14 MR. MOODY: Yes.
! 15 MR. FEIGENBAUM: Would that seem reasonable?
16 MR. NOVAK Certainly you have said that in 17 reality you are going to be setting the schedule. I think we I 18 can respond only at the pace at which you~are able to bring 19 these things to a reasonable closure. I will wait and see in 20 two weeks or three weeks for your initial response to the 21 Brookhaven response wherein you would discuss in narrative 22 form what you would intend to do.
23 Let me just, beforelwe'close, I would like to ask
' 24 both Charlie and Trevor to give me or give us their most i 25 candid views as to how you perceive this discussion. Did we 4 ss ACE-FEDERAL REPORTERS, INC.
! 202 347 3700 . Nationwide Cmerage ' 1400-336-6646
. l
30312.0 cox 78 v
i sharpen our differences to the point that you think they are i 2 articulated now that the utility can go forward? That's one 3 of the things we want to accomplish, is a clear articulation 4 of where our uncertainties are.
5 MR. l'OFM AYER : I think in containment area the 6 ultimate difference is what we consider the mean probability 7 of failure; I don't want to discuss that today extensively.
8 MR. BARRETT: I can't hear back here.
9 MR. HOFMAYER: I am saying the major difference is 10 where we consider the mean probabilities of failure of 11 containment. You have reiterated your point that the 12 proposal that you put forth originally is your position, and j
%- 13 we have said in the report that we take the position ot 1 14 percent strain, and I guess that difference still-exists.
15 MR. ROSSI: How important is that to the bottom 16 line conclusion, though?
17 MR. HOFMAYER: Doesn't seem to be at this point in 18 time.
19 MR. ROSSI: If the other things go away, then that
- 20 isn't an important issue anymore?
21 MR. HOFMAYER: Appears that way.
22 MR. BAGCHI: There isn't that much we can do 23 between now and then to emphasize or increase our confidence 24 in another number, for instance. We probably would benefit 1
25 from the test. I don't think we can do anything between now s-ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33MM6
30312.0 cox 79 w
1 and then to improve on that number.
2 MR. TORRI: I think what we are suggesting is that 3 we have made a commitment-to seriously look at all aspects of 4 depressurizing this coolant system at the right time and 5 investigating all aspects of that.
6 What we are_asking is that the NUREG consider to
. 7 put its priorities on that as well through work in parallel, 8 so that we can accomplish that in the most optimum schedule 9 that is possible.
10 In that vein, we would take, for example, 11 questions like how high does the pressure get, or what is the 12 containment failure pressure, become relatively important.
%- 13 We would think that we would suggest that -- we would hope 14 that your priorities of efforts would also go towards the 15 direction of seeing whether you can convince yourselves that 16 we can substantiate that the depressurization is the viable 17 solution.
18 Our preliminary analysis indicates that it is, and 19 we do not get into conflicts with that approach that either
, 20 are for you or for the operations unacceptable, as opposed to 21 going off in different directions now. We pursue this 22 approach, you continue to pursue'looking at impact of-23 'NUREG-1150 assumptions on Seabrook. Given that seabrook has 24 made this commitment, I think that's where we think the 25 priorities ought to be in an optimum allocation of.
s-ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage R00 3364M6
]l ?
}\
Y , --'
lJ cox 80 1 resources.
2 MR. NOVAK: I think it's fair for us to look at ;
3 the research work being supported by this Agency to see it we 4 can, in the relatively near term, establish criteria for j t
. 5 pressure, for example. l l
6 I think if we can talk with our experts and come i 7 to a consensus, we are saying, with reasonable confidence, we
~
8 believe that a pressure of this magnitude in the reactor 9 coolant system gives us the kind of assurance that should you 10 have a failure of the vessel, that the phenomena is much more 11 behaved and would carry with it the concerns that the high ,
12 pressure core melts, too. I think that's something we should
%- 13 do on our own, because we are still going to have to deal 14 with it. You have a responsibility to make that same quest, 15 but I can, I think we can do it independent of you through 16 our own research people to establish what we know, based on 17 the data that is available, plus what will come forward in 18 any near-term tests. ;
19 MR. TORRI: We think that's a very integral part l 20 of having to make this solution work, there is no question 21 about that. We think that trying to determine what the e
22 pressure on Seabrook would be, if a breach were 2250 pounds 23 in the primary system, is not directly relevant to the 24 situation we are trying to implement. Because of that, we 25 are saying maybe that is a secondary question relative to a v
ACE-FEDERAL REPORTERS, INC.
202-347-3%D Nationwide Coverage NKL33MM6
30312.0 cox 81 w
1 primary question that you have identified. There are other 2 primary questions that need to be resolved. How long does it 3 take to melt down, how big an area do you need, are the cores 4 large enough, do you need one or two. Do they work? Can it 5 be opened, can it be kept over. Is there enough heating 6 capacity to keep them open in case of station blackout.
- 7 MR. LYON: And what is the probability of opening 8 them in the first place, which is key to avoiding.the 9 difficulties that you see if you stay at high pressure.
10 MR. MOODY: I think the point, and I wanted to get 11 back to this, I don't think you meant this, but hopefully you 12 won't wait for the next two weeks to start thinking about
, s, 13 these. Hopefully people are thinking about them from the 14 questions and comments we had today. Hopefully you will 15 start down your independent path of pressurization and 16 shutdown of events as you already have.
17 MR. NOVAK: Let me clarify, I think Vic Stello l i
18 last week spoke for the Agency by saying we will continue to I i
19 give this a high priority consideration. It.will then .
.l
. 20 require support from all elements of this Agency, research as 21 well as the office of NRR. You did get that comment from 22 another office. So I think there is no question _about.it.
23 I would like to go back, while we have a few 24 minutes, to see if there is anything specific that either i 25 Trevor or Charlie would like to make at this time. If not, I ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmcrage *10-33MM6
30312.0 cox 82 s-1 fine, but if there is anything you would want to get on the 2 record at this point in time, please do so.
3 MR. PRATT: No. I think it's been a useful 4 meeting. I think it simply addressed the issues that were
- 5 brought out in our report. I think the suggested route of 6 solution is a reasonable one. I would like' Carl's suggestion 7 in terms of carving out the provisions of the PRA that would 8 lead us to concentrating on those areas. I would intend to 9 support that and go back through the thing and take a focused 10 approach, given the time frame, could be very fruitful again, 11 that gets back to our statement in the report about ,
12 confidence. We haven't done that, if we would, we would have s- 13 the confidence. I think that's fine.
14 My only concern in terms of the containment loads 15 and how they compare with the stuff that was done in 1150 is 16 that you go down the route of putting all your eggs into one l 17 basket, you run some risk, a little bit -- I think what I l i
18 showed you in NUREG-1150, in which there was a trend, and I 19 think by doing not too much calculations, one could show a 20 much smaller bar for Seabrook which I think would be useful 21 to you. It could be done very effectively, I think it would 22 be an useful exercise. I think research would be happy to do 23 that. That would give them a data point in their 24 calculations. I wouldn't necessarily dismiss that as a route i l 25 to go somewhere. You may not want to hang your hat on it but !
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage H00-33MM6
30312.0 cox 83 v
1 it, would be useful.
2 MR. TORRI: To do that, don't you need a fair 3 amount of detail on the front end of the PRA?
4 MR. PRATT: No, no. It's conditional --
o 5 MR. TORRI: Conditional on core melt?
6 MR. PRATT: Right.
7 MR. TORRI: You have to correctly represent all 8 the sequences that go into it?
9 MR. PRATT: Yes, indeed. We did the exercise for.
10 Zion using the information that is coming from the front 11 end. We are not suggesting that one would want to revisit 12 the front end. Indeed, if you have to talk about that -- we
%- 13 are talking abou't concentrating our efforts in those parts of 14 the front end that leads to this problem. Those would 15 identify the bypass sequences. 'You then look at those 16 accident sequences such as the high pressure cases, and that 17 number of 4 times 10 to the minus 5 that we look at, convince 18 yourself that that was the right number. .
l 1
19 For those particular damage dates, you run through 20 the entries that you have. It's time-consuming, it's 21 difficult. You have to be careful -- you reviewed the trees a
22 along with me. They are non-trivial trees. So if it's a 23 non-trivial exercise to put all of that stuff in, again, to 24 do it in the time frame. It's a suggestion that you may not.
25 want to eliminate, but that is a possibility.
v ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage Exk336-6646
r ---
]
-30312.0 cox 84 L
1 MR. TORRI: I guess what we are saying is the same 2 resources would be the ones who are also looking at how long 3 it would take to depressurize, what the right time frame is, 4 and our opinion would be that by looking at blowdown and 5 characterizing the requirements for that would be, in our 6 opinion, a higher priority, if it's competing with the same 7 limited resources that you have at your disposal.
8 MR. NOVAK: Let me make one final comment. We had 9 a meeting this morning with representatives from the Attorney 10 General's office in the State of Massachusetts. I think 11 there were a number of comments made that we will want to go 12 back and look at in more detail. I would ask you to do the s, 13 same thing. We will be looking at that. They have requested 14 a meeting at some time in the future. It may very well be we 15 will have a joint meeting. We will have to look at that. I 16 would encourage you to look at the comments that were 17 provided to us this morning. I would expect the transcript 18 to be ready in a day or two. When we do have one, we will 19 make it available to the Public Service of New Hampshire and 20 to the Attorney General's office of Massachusetts.
21 Are there any more comments from the staff or o 22 utility? I guess we have covered what we intended to. We 23 will just call it a day. Thank you very much.
24 (Whereupon, at 4:05 p.m., the hearing was 25 concluded.)
%s ACE. FEDERAL REPonTens, INC. I 202-347-3700 Nationwide Coverage 800-336 6646
CERTIFICATE OF OFFICIAL REPORTER
\s This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the matter of:
NAME OF PROCEEDING: MEETING WITH NRC AND LICENSEE e (PUBLIC SERVICE COMPANY OF NEW HAMPSHIRE) AND REPRESENTATIVES OF BROOKHAVEN LABORATORY
. RE: BROOKHAVEN REPORT '(SEABROOK)
! DOCKET NO.: 50-443 PLACE: BETHESDA, MARYLAND h
s, o DATE: WEDNESDAY, 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.
(sigt) g (TYPED) v WENDY S. COX d Official Reporter ACE-FEDERAL REPORTERS, INC. l Reporter's Affiliation l l
l i
e L