ML20207Q052
| ML20207Q052 | |
| Person / Time | |
|---|---|
| Site: | Seabrook  |
| Issue date: | 01/14/1987 |
| From: | NRC |
| To: | |
| Shared Package | |
| ML19306D588 | List:
|
| References | |
| NUDOCS 8701210380 | |
| Download: ML20207Q052 (174) | |
Text
-
& u o. .
J O Y G 5 A' L
UN11ED STATES NUCLEAR REGULATORY COMMISSION
~
IN THE MATTER OF: DOCKET NO: 50-4h MEETING OF NRC WITH BROCKHAVEN NATIONAL -
LAB RE: SEABROOK EPZ t.OCATION: BETHESDA, MARYLAND PAGES: 1 - 143 ,
DATE: MEDNESDAY, JANUARY 14, 1937 AG-FEDERAL REPO , ERS INC.
~ '
Clfici.11 Reconers 444 North Cacitol Street N Washington, D.C. 20001 (202)347-3700 S 9clL 2\. '.a B ?O
.une.wter covt.ucz
- g. 4. . L ' -
CR29472.0 .
DAV/ajg- '
1 g 1 ~ UNITED STATES OF_ AMERICA 2 NUCLEAR REGULATORY COMMISSION
-3 MEETING OF 4 NRC WITH BROOKHAVEN NATIONAL LAB 5
- SEABROOK EPZ 6
Nuclear Regulatory Commission
-Room P-110 7 Phillips Building 7920 Norfolk Avenue 8 'Bethesda, Maryland i
9 Wednesday, January 14, 1987 10 l The meeting convened at 10:00 a.m., Victor Nerses, 71 l g Nuclear Regulatory Commission, presiding.
) 13 14 ii i
15 1
, 16 '
17 18 19 20 1
- 21 22 !
l 23 ;
l
, 24 l
,)
u 28 ;.
1 l
i ACE-FEDERAL REPORTERS, INC.
W . : bl b
4720 01'01' 2
~
- DAVbw 1 PROCEEDINGS 2- MR. NERSES
- Good morning.
3 My name is Victor Nerses, the Licensing Project-4 Manager for Seabrook. ,
5 I first of all want to p'oint out to you that this 6 meeting is being transcribed. There are a- few things that I 7 think you ough.t to know or make sure that we can get in the 8 record to reflect that it is. complete and accurate.
9 When a person speaks, they should identify 10 themselves, so the transcriber can get the information from 11 that person. When somebody has a question to ask or i 12 anything of that nature, please identify yourselves. ..
3 I think what he has made is a log of basically 1/ 13 i-14 the people at.the table. If there are going to be other 15 pa'rticipants, we ought to let the transcriber know who those 16 people will be.
'. 17 Before we start in, why don't we just go around
' 18 and 1ntroduce ourselves.
I 19 I already did. Vic Nerses, Licensing Project 20 l Manager for Seabrook.
l 21 ! (Introductions made.)
- 22 l MR. NERSES: Okay. >
, 1 23 ! I am passing around an attendance sheet. I would 24 appreciate it if people will sign it, put their names and I)'
25 some of the information about their af filiation. This I
- ACE-FEDERAL REPORTERS, INC.
.. .m. . ..
~ ~ ~ ' '- " ^-
- . ~ .. .
4 - . go .
^ -
4720 01 02 3 DAV6w 1 meeting is ' intended to have 'an opportunity for the Applicant 2 to discuss the Brookhaven National Laboratory Report. Prior 3 to the time we get into the discussion, perhaps there may be 4 some comments that need to be made. We will open it up to
~5 the comments, and we will continue on.
1 6 Are there any comments at this point in time that-t 7 . people need ' or want to make?
?
f 8 (No response.)
9 MR. NERSES: The agenda specifically, I am sure, t
10 will be identified when the Applicant starts their D
11 presentation. Prior to that time -- woll, I think we can go 12 ahead now. .. .
A' V- 13 MR. DE VICENTIS: I want to say a few comments I
- 14 before we get going. .
15 My name is John De Vincentis, New Hampshire 16 Yankee.
- 17 My understanding of today's sessions is for us to ,
18 present and discuss our comments resulting from our review 19 of the Brookhaven National Lab Report. I would like to 20 ; resolve as many of those comments today as possible. We 21l understand that since you haven' t s.een our comments to your
[ 22 report, it may require additional follow-up meetings.
23 We are encouraged by the recults of our review of l
the Brookhaven Report. We do believe that Brookhaven 24 l
'i.0 i 25 conducted a thorough and accurate review of our submittal.
i ACE-FEDERAL REPORTERS, INC.
, 201 347 3700 _ . NationwideCoversee mitMMA
p *,
l .
i.
I.
'4720 01 03 4 D
bIDAVbw l' Full discussion of our comments will highlight this point.
2 We have key members of our staf f with us that 3 assisted us.in preparing our studies which were reviewed by 4 Brookhaven, and they are available here to go into whatever 5 depth is required. -
6 our comments are outlined in six general areas.
7 The first on the introduction portion of the report, rather 8 general comments, that discuss their further validation of 9 the PRA studies required.
10 We will also discuss dif ferences in. the 11 application of the risk assessments and sensitivity
~
..
- 12' studies. . ,
13 . We will discuss interfacing system LOCAs, the
- 14 amplification of a discussion on check valve failure 15 frequencies. We ae prepared to discuss the shutdown events, 16 even though at the time of the issuance of the Brookhaven 17 Report, they were not comple ted. We do have :omments on the 18 information they did submit in the report, and we are 19 prepared to go in'to detail in discussing the steam generator tube ruptures' potential and the evaluation of direct 20 l 21 ! containment heating and, obviously, the containment status.
22 ) The format that we are contemplating for this l
23 i session is that Jim Moody will lead of f with a summary 1
24 ; introduction on just highlights, how we got to where we are i
25 today. Then we will have a lead individual or two sort of
. ACE-FEDERAL REPORTERS, INC.
. -- - - nmn . >-~c-- ~
4720 01 04 5 DAVbw 1 go through our comments, a page at a time, highlighting the 2 areas that are fruitful for discussion.
3 I assume that Vince and I will sort of moderate 4 the degree of discussion. If it does appear to be 5 nonfruitful, we will identify those areas for potential 6 future meetings, if necessary.
7 So if no one has any questions, I will tu r e. i t 8 over to Jim Moody.
9 MR. NOVAK: Let me make just one point.
. 10 This is Tom Novak.
11 I think it would be important, as your speakers 12 discuss the Brookhaven Report, to try, as succinctly as they-
'N I
. . l
/ 13 can, to. identify their understanding of what the Brookhaven i
14 Report is saying with regard to a specific subject. Then I 15 would ask Trevor Pratt and Charlie Hofmayer, af ter that 16 discussion, to see if, in their view, that is a reasonable 17 representation of their views and what the report is 18 suggesting.
19 I am not asking them, and I don't think they are 20 l prepared to provide rebuttal to the ccmments, I think tha t i
21 will take, as you say, more study, but I think it would be 22 important if we have to go through this discussion, to a t 23 l leas t have agreement, yes, that's wha t the report is 24 saying. We are not quite sure we understand what the report 8 25 is saying. So as we go through it, if we can, le t 's try to ACE-FEDERAL REPORTERS, INC.
202 347-3700 NationwMe Coverage 800 336 4 646
. i.
4720 01 05 6 m
h DAVbw 1 make room for that kind of a comment'. And to the extent yu 2 can, I think that will help us, and we won't have to go back 3 and ask ourselves at a latter time whether or not that was a 4 correct understanding.
5, That would be my only suggestion.
6 MR. MOODY: I am Jim Moody from New Hampslire 7 Yankee.
8 I may reiterate a little bit of what John said.
9 The intent here is to provide our comments on the draf t 10 Brookhaven Report.
11 I am going to go through fairly briefly and 12 summarize the major areas of disagreement we have with the 13 report.. Then we would like to come back and revisit it, and i 14 at that point, we might focus on what you suggested.
15 l MR. NERSES: Jim, do you have any handouts by any i
chance?
16 l 17 MR. MOODY: Somebody does. You might want to 18 hold the handouts, or at least the major handouts. I don't 19 care.
20 , MR. NERSES: Since you say you've got a slide 21 ! noting the differences and so forth, it might be worthwhile '
22 at this point in time.
23 MR. MOODY: Again, we recognize you are just 24 seeing these comments, cnd there will probably be 3 ,
25 an additional need for meetings. And as usual, we are ACE-FEDERAL REPORTEILS, INC.
202 347-3700 Naticawide Coverage M 33 M
<>4(9
+
IMAGE EVALUATION ((gj [D Q+
%g \f +k f \a? M/ k
$/ TEST TARGET (MT-3) f fj 4
[s g#
%% s,,
v&+q,,
1.0 if 2 M -
1 lf 9 E s 1.1
!. ~ E l=1.8 1.25 1.4 1.6 150mm #
4 6" #
4
% ++
4Iff;/ A7ff- .- N+g;;g+f 4 g,,,,7/ y
_ u
4> @ o O IMAGE EVALUATION ((//g 0[4rQ,8
\k 1@
[ %k/ %k/ TEST TARGET (MT-3) jf 4
.x\\\gf /gg\NYj o[,,tg M @p 4%'4 4
t 1.0 if m M l29 HE s !.:_ gpa 1.I L U8 I.25 1.4 1.6 150mm
- 4 6" #
4
'I 4*/; 9Y
.~
/f4 k 6
/ [\
m ;
%fff i& y a _ _ _ _ - - _ - - _ - - _ _ _ - - - - - - - -
+ 44r '
%$$ \tI/ < hdp IMAGE EVALUATION
/// /
q)ff \; f//
TEST TARGET (MT-3) ? dp
+#+ %s %
" 1.0 !f m m "ljHE ii E u
in He
'l.25 1.4 1.6
=
150mm
- 6" #
4 4 %p ,, / S4+%
y -
<));p(4%
4;;gy*//; y
~
3 n.
4720 01 06 ,
7 Asb DAVbw 1 prepared to meet at any time and do whatever is necessary 2 to resolve Seabrook.
3 (Slide.)
4 The first thing I would like ' to do is just review 5 some key events with regard to the Seabrook Station 6 probabilistic safety assessment. It was completed in 7 December 1983. It was a full scope Level 3 PRA submitted to 8 the NRC in February 1984 for their information. Two major 9 reviews took place af ter that. We had Lawrence Livermore 10 look at the plant model in the core melt part of the 11 analysis. They generated a large docu. ment of comments, and 12 we provided a detailed response to that about a year later.
~h
/ 13 The bottom line of .the Lawrence Livermore 14 conclusion and the NRC's was tha t there was no significant 15 safety issues identified. In fact, they felt that the core 16 melt frequency might be a little conservative. Our detailed 17 assessment also concluded the same thing; however, we did 18 obtain a lot of insights from that review, as far as 19 ! explaining the PRA results and the analysis.
20 j Some areas of the review that Lawrence Livermore i
21! made identified conservatisms, mainly, interfacing LOCA and I
22 l the seismic fragility of components.
They identified major 23 conservatisms in some other areas that affect early I
24 l releases. They had very positive comments.
LV 25 Another review was by Brookhaven, not the ACE-FEDERAL REPORTERS, INC.
3 2-347-3700 Nationwide Coverage 800-336-6646
- s s.
4720 01 07 8 N
D DAVbw 1 existing one but an earlier one, where they looked at the 2 methodology in the containmet model, and now they are 3 looking in even greater detail at containment analysis.
4 Their report was published in February of '86, 5 and it was pretty mt .h positive also. In July of this year, 6 we submitted an upda te that included the risk management and 7 planning study, an update of the original SSPSA. It also 8 included a sensitivity study from those results, using 9 WASH-1400 methodology source terms.
10 .We also submitted fragility inputs and the
. 11 se,ismic capacity of components, although that hadn' t been ,
12 quantitatively factored into our ,results.
13 Shortly thereaf ter,. we. -asked the NRC to review
. 14 these documents. Shortly thereaf ter, the review was 15 initiated by Brookhaven National Lab. There's been a number 16 of meetings. Two ACRS meetings in late September and early 17 october, and we are anticipating additional meetings in 18 February and March'. -
19 The draf t report was issued in December by 20 l Brookhaven, and of course, we filed that petition about ten l
I ~
21 days later.
22 And that gets us to today.
23 The comments on the draft report.
i 24 l (Slide.)
'h 25 As John indicated, our commen'ts are grouped into ACE-FEDERAL REPORTERS, INC.
' 202-347-3700 Nationwide coverage 800 336 6646 _ . .
)
L4720 01 08 9 DAVbw I six major sections, as shown here.
2 The first section providing general comments.
3 That focuses on the purpose and the summary of the draf t
^
i 4 report. Then the next five sections addressing more 5 technical areas, areas that Brookhaven has identified on 6 early releases.
7 I am going to go, through each one of these fairly 8 i briefly. Then we will go back and revisit them in more 9 de ta il . .
10 (Slide.)
11 I will identify what I think the major things are 12 that need to be revisited. The general comments that are ,
13 aimed at' the pref ace and summary up front, mos t of them -- t i
14 if I could find one word to explain the major disagreement
- 15 here, it is that it is one-sided. As I go through this, I
. 16 will particularly explain while we feel it is one-sided.
17 And the implication is that our sensitivity study is being 18 reviewed. In fact, it is obvious that the original PRA has 19 been reviewed, as you go through the Brookhaven Report, as 20 well as the update.
o
, 21 , The other thing that isn.' t explained properly, we 1
l 4 22; feel, is the role between the PRA and the sensitivity ,
l
- 23 l utudy.
- 24 i We have taken the PRA result and done a
- h 25 sensitivity study, using WASH-14 30 methodology source ACE-FEDERAL REPORTERS, INC.
. _ 1 . -m_ s- ece _
. . . . . . . . . .-.~ ^
7h720 01.09 10 :
er i 1 DAVbw 1 terms, and that actually jacked up the risk that we we,re 2 calculating by a couple orders of magnitude.
3 Now on top of that, we are doing additional 4 sensitivity studies, and we are continuing to stack 5 conservatisms. I think that needs to be explained 6 properly. And we have been guilty of doing the same thing 7 in some of our responses.
8 Along those same lines, it appears as though the 9 review has attempted to find ways to make the results look 10 higher. There was no attempt to identify conservatisms.
11 We think there are conservatisms in there, as well. The
- . 12 focus is.early risk profile. There are some cautious.
13 statements early -in the report about verificati'on',
14 completeness. We think'there is a lot.of more positive 15 things that can be stated about the PRA. WASH-1400 was 16 completed in 1975.
17 We have 11 years of PRA experience. The 18 knowledge since then, we've probably spent billions of 19 dollars in severe accident research that has to do with i
20 PRAs. The Seabrook PRA was built upon the design of the 21 Indian Point PRAs, done by the same contractor, who was also 22 reviewed in great detail. And during these reviews by NRC 23 and the contractors, this was factored into the Seabrook PRA.
() 24 And there's only so many ways in which you can have an early 25
~
ACE-FEDERAL REPORTERS, INC.
_ _ , _ . . _ _ M 347 3700 . _- . -_-_,
_ .Naticewide Covenne MVkttMMA
- ~
11
)720 01 10 1 DAVbw 1 release.
2 We think the information and the knowledge are 3 there. It is a matter of bringing it together and 4 explaining it. We think we've demonstrated that we are more 5 complete, as far as accident sequences are concerned than 6 WASH-1400. WASH-1400 didn't model reactor coolant pump 7 seal, LOCAs , earthquakes, which are dominating. The' latent
-8 health calculations, even when they are small at Seabrook, 9 aren't explicitly included in WASH-1400. As a matter of 10 fact, many of these issues we are going to talk about aren't
. 11 included in WASH-1400.
'T. 12 - So given the fact we feel it was one-sided, an 13 attempt to see how conservative can we make things or how 1'4 high can we make the results look, the envelope, the upper 15 bound, we feel the EPZ technical justification is not 16 diminished.
17 (Slide.) .
18 s Interfacing systems LOCA. The major review here 19 i is by Brookhaven.. We have some additional comments and input 20 to some of their review, but I think the major disagreement 21 we have, if you want to talk about this, is this calculation 22 of check valve failure rates. We will come back to that, 23 but basically, we think that while I think the difference is 24
{f- about a factor of 8, our value ccmpares very closely with 25 actual reactor experience.
26 (Slide.)
ACE-FEDERAL REPORTERS, INC.
_ _ . . . _ . _ . - . ~ -
- l l
12 4720 01 11 El 64 DAVbw 1 The shutdown and cooling events. Again, we have 2 some input to the Brookhaven review. I am not sure there is 3 anything major here, any disagreement. They question the 4 credit for the two RHR drop lines at Seabrook and provide 5 additional information to justify why we took credit for 6 it. Basically, we have procedures that move power from 7 those MOVs, so we can have a spurious event close an 8 isolation valve. We have also done a more realistic 9 analysis of the shutdown cooling events, as far as source 10 terms are concerned.
11 We will discuss that la ter . i
. I 12 (Slide.) i 13 Steam generator tube failure. We have sta ted i
14 right along that steam generator tube f ailure is very 15 unlikely. As a result of the Brookhaven review, NRC went 16 back and looked at it in great detail again on a 17 Seabrook-specific basis. This information was provided. I 18 am not sure that any plant has studied this event on a 19 I plant-specific basis the way we have. We show great l
20 ; margins. We s till think it is a very unlikely event. We 21 got some comments on the Brookhaven. write-up on this, and we l
22
- have some comments about the conservatisms and their i
23 l probabilistic assessment.
I 24 ! Again, this is an issue tha t is n ' t in WASH-1400, h l 25 j as well as the shutdown events of the previous one. I also l
ACE-FEDERAL REPORTERS, INC.
4720 01 12 13 G
Al DAVbw 1 believe it is a generic issue that won't be resolved 2 completely until research is complete. It is made up of a 3 series. You have to have the frequency which you have for l 4 the event to begin with, high pressure core melt and dry
! 5 steam generators and then frequency that the operator 6 doesn't pressurize and the frequency that the steam 7 generator tubes fail. l 8 There are a lot of things you can do to make 9 these numbers become small, to make this a. nonissue on a 10 plant-specific basis, b'ut I don' t think it is appropriate to 4 l
11 do that until research is completed here. We think- this l
. - i 12 number is very small. It 1s not important. Other people is l 9: l
.~
13 not so sure that there isn' t a ~ higher upper bound number.
l 14 .Until this number, we have some confidence in that number.
15 I am not sure it is appropriate to make a conclusion.
16 .'
17 18 19
. 20 1 21 .
a 22 23 4 24
!S 25 1 -
,4 .
. ACE-FEDERAL REPORTERS, INC.
. __ w.w.m .._. _ - - - -u - - - . - . - . .
--- ~~~-
2 .
l l
l4720-02-01 . 14 3
,w - DAV/bc 1 1 Slide.)
1 2 ,
Direct containment heating is similar to the 3 steam generator tube failure issue. Again, it's a generic .
i j 4 issue. The issue is based on generic experiments. I 5 believe they were done at Sandia.
. 6 I don't believe they represent Seabrook specific 7 configuration,s, especially above the cavity. And there's 8 been work done in that area. Some simple experiments by 9 Tausky. They've been discussed previously.
10 We believe the analysis is very conservative.
11 And, also, we take in the Brookhaven comments on containment
. 12 failure pressures, and reanalyzed this. We'll have Fred 13 Torri present this a little later. This I think is another 14 key issue we need to discuss.
t 15 (Slide.)
16 Containment structure. The Brookhaven review J
17 agreed with the yield stress calculations that everyone else 18 has performed in the past. They recommended that the 1 1
1 19 percent strain be used as a median capability of the 20 containmen't.
21 We believe that's very conservative. But it 22 actually doesn' t matter. The containment seems to be very 23 i strong, and its capacity is so good that even when we used 1 24 percent strain, it doesn't af fect our conclusions.
M '
i) 1 25 l i
There's a real strong link and a lot of agreement l
J ACE-FEDERAL REPORTERS, INC.
._ - ,-- . . . _ 2E147 1700 _. _._ _. . , . , , , , , - - -Nasionwide
__._ Coverase ._ 000 3%dM6
- . - ~ - _ . - . _ . _ - , _ . . ~
4720 02 02 -
15 Sb us DAV/bc 1 between Brookhaven in their review and what we've done.
2 That's probably in this area, even though it is very 3 conservative. ,
4 (Slide.)
5 In summary, while the first two -- I think that's B_
6 a typo -- that I think are k.ind of one-sided right now, and 7 the inclination is used properly in decision-making will 8 take a balanced look at the uncertainties, that would be.
9 consistent with WASH-1400 and NUREG 0396.
10 We want to talk about the appropriate use of 11 check valve data, obviously these two issues. And I just
'12 said the containment is very strong at 'this point.
ST Ll 13 I'd like to 'g'o back and start on our general
^
14 comments. I think maybe Carl will help us walk through --
15 Carl Fleming will help us walk through the first couple of 16 pages of our writeup.
17 Any questions?
18 (No response.)
19 MR. FLEMING: Does anybody want to make any 20 ! comments or questions before we go on to the next area?
21 MR. NOVAK: This is Tom.Novak, again.
I think it's f air to say don' t assume that we 22l 23 accept your summary as fact. What we want to do now is go 24 j back through each one of these and then make our comments.
25 I think you've provided a summary and we don' t ACE-FEDERAL REPORTERS, INC.
ll02 347 3*00 Nasionwide Coverase Arn1%Me
I 4720 02 03 16 D
Ek DAV/bc 1 hvae at this point in time a kind of an opposite summary, 2 just for the record.
3 So why don' t we go into the specifics. Let's 4 go back into the agenda and hit these, and then use this 5 approach I've suggested, prov,ide the details of what you 6 believe the Brookhaven report is suggesting, and see if we 7 can get a reading from our side whether or not that's a fair 8 representation of what the report is saying.
9 First, factually, is that what the report said?
10 Why don' t we try that? -
11 MR. MOODY: I suggest we start right with the 12 general comments. I guess the first overall f act, we felt 13 that it was kind of one-sided .and di,dn' t adequately describe 14 the full scope of the review that took place.
15 MR. ROSSI: You mean it didn' t describe the full 16 scope of the Brookhaven review? Is that what you're saying?
17 i It didn't describe all that you think Brookhaven really 18 reviewed? Is that what you mean? Or it didn't fully 19 describe everyth'ing that Public Service of New Hampshire had i
20 done?
21 MR. FLEMING: To amplify. od that with regard to 22 the one-sided characteristic, I think what we intend to mean 23 ' by that was the f act that the look that Brookhaven seemed to 24 look for ways that could possibly drive the results higher D . 25 l than they had calculated in our studies.
. . ACE-FEDERAL REPORTERS, INC.
m.m.s. ~.. - . c _ .. -
- s-
.j ,
4720 02 04 .
17 DAV/bc 1 It was not clear that they were trying to take a 2 position on where the true risk curves really lie. They 3 were only looki,ng for ways in which we might possibly get 4 them higher.
5 It's almost as though it was being done as a 6 process of negotiation. We started out with our asking
- i . 7 price, you know, and that almost set the stage of what 8 direction they were coing to go in, rather than trying to 9 characterize the level of conservatism that we may be 10 starting off with on the results to baseline how f ar it's-
~ ,
11 necessary to go further by adding additional conservatisms.
12* I guess the best way to express this is the 13 ,.Brookhaven report seeming to becoming to a profile 14 revelation, that they were able to do something with our 15 results.that made the curves appear higher.
16 . It shouldn' t be surprising that that result is 17 obtained. If the true character of the sensitivity study is 18 understood, our sensitivity study was arrived at by trying 19 to map what was,,,done in the NUREG 0396 and WASH-1400 as 20 closely as possible.
21
- That is, we used the' median accident frequencies, 22 best estimate assumptions on everything in the study that ,
23 had been addressed with uncertainties in the original PRA.
24 And we froze in on one issue and one issue along: source 3 25 terms, holding everything fixed at best estimate
.g.. t .ACE-FEDERAL REPORTERS, INC. .
720 02 05 18 1 DAV/bc 1 assumptions. We varied the source term methodology to be 2 able to see the isolated effects of conservative source term 3 methodology on the overall PRA.
4 It shouldn't be surprising that if one adds 5 additional conservative assumptions on top of that,'that 6 you're going to get higher curves.
7 So, statements in the Brookhaven report, like we 8 are not able to validate how the PLG report draws the risk 9 curves, really can't be attached with any significant 10 meaning because it just follows that if you add conservative
. 11 assumptions on top of the assumptions that we had, you 12 should get higher results.
13 So that was one p'roblem we had with the 14 Brookhaven report leading to better characterize what wo 15 did. We did a risk assessment study in the original PRA.
16 We updated it in PLG 0-342, and then we did a sensitivity 17' study on source terms. .
18 And that was the only purpose of the PLG. If 19 that is understood properly, then the meaning of going and 20 adding those sensitivity studies on top of that can be ,
21 better understood in the decision-making process.
22 The one-sided thing, the one-sided comment has to l
23 do with , you '<now, in all the sensitivity cases that woro
}) 24 looked at, very little is said in there about how much 25 , conservatism currently exists in PLG, version five.
ACE-FEDERAL REPORTERS, INC.
202 347 1700 Nationwide covemee en1tMMA
4720 02 06 . 19 DAV/bc 1 It seems to only look for ways in which one can 2 combine assumptions to make things higher. They didn' t 3 appear to look for ways in which the results might be even 4 more robust than they in fact were. ,
^
5 . So that's kind of an amplification on a couple of 6 those major points.
7 MR. NOVAK: Trevor, I think it would be 8 worthwhile if you could summarize perhaps what you set out 9 to do for us. And if there's any way you can reflect on the 10 Public Service of New Hampshire comments.
11 MR. PRATT: Surely. I think what you're talking 12 about is.mostly editorial, if you don't mind me saying so.
13 ,
I think this was a deaf t report which we submitted to the 14 NRC rather quickly.
15 We all recognize that a summary needs a lot of 16 work to explain these things in more detail. It should 17 reflect some of the words that are in the text.
18 The introduction, as well, needs expanding. I t' 19 was our intent to do that, to put into perspective all of 20 the various elements that went into it. That's one area 21 that is ' missing that we recognized,, and that will be 22 corrected.
'23 I don't think in any of the presentations that I 24 personally made to the NRC staf f, I in any way did not tell 25 them of the building conservatisms in your calculations with ACE-FEDERAL REPORTERS INC.
W.
. ,n" . , . *,;. a.- ..a.
_m
- 2ax.w.nco _ _
4720 02 07 20 DAV/bc 1 regard to using WASH-1400.
2 We made a presentation very recently where I did 3 point this out. I think that's an important point and was 4 certainly' pointed out,in the report. However, I think 5 that's part of the ground rules that we were asked to work 6 with.
7 In other words, the idea was to try to take the 8 median numbers of our accident sequence probabilities and 9 containment f ailure probabilities and to couple that with i
10 the types ' o f source terms one would have calculated in 11 WASH-1400 analysis that we actually applied at design at ,
12 Zion and Limerick, and in the hearings. We did all.of that 13 work. .
14 Again, if that's not clear, I can certainly spell 15 that out. One of the points we mado earlier this week, 16 l major contributors in terms of source te rms , the long, i
17 ' drawn-out source terms, using them throughout the code in 18 ' WASH-1400 are obviously conservative -- very conservative.
19 So it's very clearly conservatism and we can 20 point that out.
21I With regard to whether or not we were trying to 22 make things worse, obviously, we're not going to look for 23 ) things to make things better in the sense that, you know, 24 you're already coming in with a submittal that is a 1
25 ' reduction in results.
1 ACE-FEDERAL REPORTERS, INC.
~)
e 4720 02 08 21 O
d t DAV/bc 1 The question is how robust is that calculation?
2 In the sensitivity studies, it was simply an 3 indication,that, given the conservatisms of the WASH-1400 4 source terms, how robust that calculation was to 5 uncertainties in some of these.
6 Now it's recognized very clearly. And I tried to 7 get this across in several meetings, tha t in 396, the 8 contribu tors to the risk codes, they were all coming from 9i structural failure of the containment as a result of steam 10 spikes, hydrogen burns, et cetera.
11' So what that curve does, it reflects the 12 perceived f ailure modos of the Surry containment building.
13 Wha t we've now como up with is an improvement. .
I 14 We've now found the containments are stronger than they 15 wore, and Seabrook is a particularly strong containment.
16 What that then tends to do is push down those 17 curves into regions, which we didn' t pay much attention to.
18 In that sonso, one does feel a little bit unfair 19 in focusing in those areas on Seabrook. We are now focusing 20 on areas that were, if you like, not well identified, not 21 well analyzed, not well undoestood.
22 i So tha t, in that senso, what we're doing is 23 , moving into areas that woro not well thought about and 24 ! because of that, we're having to do sensitivity studies that 9 25 reflect our best judgment.
. ACEyEDERAL KEPORTERSt,1NC.
4720 02 09
- 22 DAV/bc 1 Our best judgment in all of these cases could be 2 yours, but it could be different. So we have to put ranges 3 ,on it.
4 Now, in the final report, we'll' attempt to 5 characterize that process a little better than it probably 6 is, but that certainly isn' t the intent of wha t we tried to 7 do.
8, The intent is to try to focus the NRC staf f so 9 that it's able to make decisions in this area. They have to 10 be concerned about these particular types of events. As you 11 say, there are only so many ways that you can lose 12 ' containment. And those are the areas that we have to focus 13 in on.
- 14 And you have to decide on what is lef t. But 15 that's all that study is designed to do. There are cases 16 where, for example, we did make conservative assumptions and 17 found they had no impact at all. And we said so.
18 Now I can bring that forward. They're buried 19 deep in chapter whatever, which I intend to do if that will 20 help to focus there, if we decide tha t that particular issue 21 is not important. -
22 In fact, it's interesting to note that all these 23 things that we originally set out to do on review, the idea 24 was to focus in those areas that we considered important, 9 25 which was containment structural strength, the interface of
. ... . ACE-FEDERAL REPORTERS, INC. ..
4720 02 10 -
23 M
4 DAV/bc 1 system LOCA, and so on.
2 They were found to be very robust. It was those 3 areas that we really didn' t focus on very well that we have 4 a problem. Steam generator tube rupture. Steam generator 5 isolation areas, which, as you say, were not addressed in 6 WASH-1400.
7 So we're breaking ground.
8 MR. MOODY: I'm not clear when we can address 9 them. ,
i 10 MR. PRATT: No, that's another problem. yes. But 11 I think that's what happened. You build a big, strong 12 containment building and all these other little things, how 13 significant they are, is a question more of policy, I think, 14 than of technology.
15 MR. FLEMING: I think we would agree that these 16 kinds of comments that we're bringing out here on the 17 summary, we agree are primarily of an editorial type nature 18 because as we get into the detailed, technical issues, I 19 think that we'll find that the area of disagreement is 20 rather focused and specific areas and issues. , And we can 21 discuss that technically with regard tio describing what has 22 been done.
23 One, this is a kind of minor point, but it's 24 something you might want to consider, we have agreed as a D ,
25 ground rule to try to make our case on the s'ensitivity
.;. .. .. . .. . ...s. ACE-FEDERAL REPORTERS INC.
^m ber x= -
w
3 o ,
4720 02 11 -
24 DAV bc 1 study based on the WASH-1400 source term
- methodology.
'2 In describing that in your report, it may be 3 misunderstood in the way it's written up right now that you 4 only review this little thin 0465 document.
5 As a matter of fact, we had to review all the 6 documents that we've discussed out here today, at least the 7 relevant portions of those. And then we're kind of 8 accepting the ground rule that we're going to use the source 9 term.
10' I think, if that comment's resolved, I think the 11 summary would better reflect the fact that it was utilized.
12 MR. PRATT: Absolutely. I intend to make that
- a. .
13 point strongly. It's very evident that the final document, 14 once it's reviewed, is basically on the source terms.
15 Most of the review concentrated on the previous
.16 document. ,
17 MR. FLEMING: Now, the other thing I wanted to 18 amplify on, the Brookhaven report ack,nowledges at this 19 point, and we recognize that and respect that, but I think 20 it needs to be emphasized more, that is, when one takes a 2L li'st of sensitivity analyses and somehow combines them together, as noted in the Brookhaven report, there isn't a 22l 23 real scientific basis for combining those.
24 We just wanted to reiterate that when one B 25 combines' three, four, five or six sensitivity errors and 1
ACE-FEDERAL REPORTERS, INC.
. . . . - . ~ . , . _ n .-. , -. -
4720 02 12 25 d)i DAV/bc 1 adds them up, in effect, you think that as far as the 2 decision-making process is concerned, there's a great 3 tendency to provide a distorted level of significance as to 4 what the underlying sensitivities are.
5 Individual sensitivity analyses seem to be 6 meaningful. You can see the effect of this issue by 7 comparing two dif ferent curves, say. But, if one takes five 8 or six sensitivity cases and adds them up and somehow draws 9 an envelop that's supposed to account for all of them, I 10 think what that's supposed to mean to the decision-maker is 11 rather obscure.
12 It has a tendency to distort or come up with r" l 13 I misleading conclusions about what the significance of these' 14 issues are when you, in fact, try to add th'em up.
15 In the PRA results, we make an attempt to do 16 this, to combine these uncertainties with a probabalistic 17 quantification. And that tends to be a rather controversial 18 area. Just how one does that, and so forth.
19 , - At least there's a way in which we can somehow I
20 i put some weight on the chance that all these stacked 21 conservative assumptions might turn out to be on the 22 unfavorable or favorable side of what you think the central 23 ' estimate should be based on.
24
- / 25 l l
. ACE-FEDERAL REPORTERS, INC.
4720 03 01 -
26 DAVbw 1 Many of the specific issues we get' into, it's a 2 little bit of a dichotomy here, because we are using 465 to 3 focus our decisionmaking with specific issues that we 4 acknowledge to be very uncertain. For example, the check
, 5 valve failure. We find ourselves arguing of where the check 6 valve failure frequency is. When one goes back to the 7 original PRA, one can see that we have an uncertainty ,
8 distribution that you can drive a Mack truck through. So we 9 acknowledge that this failure frequency is highly uncertain 10 and variable.
11 . So how one deals with uncertainty in sensitivity 12 takes on a dif forent character when you are talking about 13 the risk' asssessment re'sults, .which we characterize in '0432, 14 and tho' sensitivity results. That has to be understood 15 before a good decision can be made on it.
16 MR. NERSES: I guess all that says to me is that 17 you are suggesting that you have a proposed methodology of 18 treating uncertainties, which maybe perhaps is controversial 19 as versus the way Brookhaven handled the situation.
- 20 MR. FLEMING: Right; that's correct.
21 MR. NERSES: Is that the message? And 22 therefore, for those who need to make a decision on that, 23 that's significant?
24 MR. FLEMING: That is right. That is one thing I B ,
25 am saying. The other thing I'm trying to say is that the c.wi..,. x .I:<
... c. -ACE-FEDERAL --
REPORTERS .x-INC. .
4720 03 02 -
27 DAVbw' I decisionmaker, when he looks at the sensitivity study ought 2 to recognize that that is a sensitivity study versus a risk j 3 assessment, in which some approach is being taken to combine 4 all the uncertainties or sensitivities.
5 MR. MOODY: Which is what WASH-1400 attempted to 6 do.
7 MR. FLEMING: Right. They used a metholodogy --
8 we have a slightly dif ferent methodology, but the attempt to 9 put all the uncertainties into the risk curves at one time, 10 the original Seabrook PRA, this was more apparent, because 11 when one presented risk curves, we didn' t draw just a single 12 curve. We had a complete f amily of curves, and a genera' tion 13 1 of multiple curves was' the expression of uncertainty.
14 What occurred in the individual curves was a 15 expression of what the risk would be, given certain l
16 assumptions about what the resolution of that uncertainty 17 could be, and if you look at, for example, the early health 18 ef fect risk curves we received with the PRA, the way you 19 l would draw those risk curves, we indicated would be several 20 orders of magnitude up and down the scale from where the 21 central estimate was drawn. That is the result of our 22 methodology for combining uncertainties. It includos the 23 ! accident freqt:ency uncertainties, the source term 24 uncertainties, the uncertainties of some of the modeling h 25 assumptions, such as consequence modeling, and so forth.
. . . ... ACE-FEDERAL REPORTERS, INC. ...
..w.n _ ~ .. .- ... .... :
4720 03 03 ,
28 DAVbw 1 And it is very important to understand, we went 2 through 0465. We froze in on best estimate assumptions ,on 3 everything including meeting accident frequencies,- best 4 estimate assumptions on all the modeling parameters. We 5 varied one parameter source terms to see what the singular 6 ef fect of that issue was.
7 So as we go into the Brookhaven report, then we 8 find ourselves superimposing on that sensitivity case, 9 additional sensitivities, and we have to ask ourselves each 10 time we do that, you know, shat is that curve supposed to 11 represent? The combined effect? The source terms and
~
12 sensitivities? And say, shut down cooling events.
13 So I think if that is understood then, our 14 comment will be resolved.
15 MR. PRATT: I think we were asking whether you 16 had a way of dealing with it. I, don't believe you would 17 have dealt with it in the way of folding it into the 18 original assessment and then giving an assessed confidence, 19 as you had in the original PRA, and then reflecting, the 20 median of that distribution.
21 I think in our report, what we did there was --
22 and it's something I felt uncomfortable with, that is why I 23 wrote the words in there, when I added them all up and put 24 them in there. We were concerned about doing that, and you 9 25 can add things that are mutually exclusive, so that one has
.. :w ...u. , 9, ,.. ; ... . ..... ACE-FEDERAL REPORTNRS, INC.
==x-- - - - - . .
-n- . . ., .x- - -
1 .* .
4720 03'04 -
29 C:)6 DAVbw 1 to be very careful about doing that type of risk study.
2 .
I think what you are suggesting is that it would 3 be even more useful for the decisionmaker to look at the 4 individual sensitivities as each exists and make decisions ,
5 about that particular item rather than adding the whole 6 thing .together and saying this is going to drive.
7 You might say, okay, this item looks like it is 8 somewhat uncertain and should be fixed or dealt with, 9 whereas another item isn' t, but you add them all together.
10 Now I think, in presenting information to the 11 Staf f, one of the suggestions that was made was in the 12 summary that you display each of tho contributions rather 13' than showing the ef fect of the various items. Go through 14 each of the points, point by point, and explain the 15 significance of it.
16 MR. NOVAK: Let me make one point, and then we 17 will move on, just so the record is clear.
18 I think at this point, we should recognize that 19 Staff is knowl,ydgeable of what Brookhaven is doing, and your 20 comment -- I don' t know whether you really intended it to 21 mean that, but just so it is not misunderstood. We 22 understand what Brookhaven is doing, so in terms of the 23 decisionmakers, I think we will understand the work. We 24 have had a number of meetings we have coordinated what we
.hh
(
25 have asked Brockhaven to do for us. They are serving as, 1
l
. . i..; ... - . p. ; . . . , n.... ACE-FEDERAL REPORTERS, INC.
.=-_: x -
- , . ~ --- ----
--- - - - - - -.x------------ --
53720 03 05 30 ML/
2 DAVbw 1 principally, our technical, I guess , performers , if you 2 will. We have a comparable technical knowledge on the Staff 3 to do the work, but we've contracted with them to perform 4 these particular analyses for us. So I think from that 5 point of view, I think the Staff dees understand wh'at 6 Brookhaven is doing and will be able to make proper 7 judgments on your work, as well as what their review of your 8 work has been.
~
9 Why don't we move on?
10 MR. MOODY: Okay.
11 . Interfacing LOCA. .
"> 12 MR. FLEMING: Maybe I will just start out by U ^
13 giving you a brie'f summary of what we thought the Brookhaven 14 Report said about the interfacing LOCA.
15 Basically, what we saw in the report, they 16 acknowledged that their review had not been completed, so 17 maybe we haven't seen the complete presentation of their 18 comments on the interfacing LOCA, but by and large, it was a a
19 rather narrow focused issue that has a significant impact on l
20 ' the way the numbers are calculated that we apparently don't.
21 agree with. And that is how one calculates the frequency of 22 check valve ruptures.
23 In that regard, I think the principal area of y) 24 difference is the way in which Brookhaven had independently 25 calculated the check valve frequency and the way we had ,
.. ACE-FEDERAL REPORTERS, INC.
, .~
4720 03 06 31 m
J.2 DAVbw 1 originally calculated ours, comes to the point of how one 2 es,timates the number of exposed check valves for calculating 3 the number of. hours of service of check valve experience to 4 divide into the number of reported failures. In addition, 5 Brookhaven had done a wider search for check valve leak 6 events and was able to expand the treatment of f ailures to 7 cover a number of failures that were not included in our 8 analysis, which was a result of their independent check of 9 the LERs and a longer period of reactor years of 10 ! experience.
11 We consider that to be an improvement in the data ,
12 base with regard to the characterization of f ailures. But I d/ 1'3 then what Brookhaven has done is, they've taken the 14 remaining details of the PLG analysis and recalculated the 15 l ripple ef fects on through the event trees with the dif ferent 16 numbers of failures and different numbers of successes for
'1 17 check valve failures.
18 I guess, to summarize, our comment here, we do' n' t 19 agree that the Brookhaven f ailure rate provides a reasonable I
- 20 indication of the frequency of major disk valve ruptures.
It seem's like what they have done is come out with a better 21 22' failure rate for accumulator leak events and then by 1
23 : adopting our approach of developing a frequency magnitude 24 , relationship between frequency of leak events versus leak 5d 25 size, have como up with an unreasonably large value of the l
1
. 1 ACE-FEDERAL REPORTERS, INC.
202 147.)?00 Nationwide Coverset 8ak11MMA
4720.03 07 . 32 DAVbw 1 frequency of check valve ruptures. .
2 (Slide.)
3 Let me summarize what we have done in the 4 original PLG work.
5 'We estimated a total number of check valve 6 failures and a total number of check valve hours of 7 experience that produced those and developed a curve of 8 frequency versus magnitude, that if we take slices through 9 our curve, we had a range of possible values and uncertainty 10 distribution, a frequency of check valve leak events around 11 150 gallons per minute or more leakage than based on a curve 12 that we came up with. 'By fitting the data to a curve, we 13 extrapolated a number to come up with a frequency of a major 14 rupture characterized as 1800 gallons per minute leakage, 15 which would shif t us down the frequency scale, but a 16 comparable very large degree of uncertainty. 1800 gallons 17 per minute is a key value, because that is the capacity of 18 the ItHR relief valves. And leak events that exceed that '
19 would challenge the RHR system with reactor coolant system j 20 l pressure.
, 21 Assuming successful operation of the relief 22 valves and the way in which the PLG event tree analysis is 23 based, this is kind of a key number, which you would start 24l to ask questions about the type of piping integrity 'and have D 25 I interfacing scenarios develop at that point.
. ACE-FEDERAL REPORTERS, INC.
, 201 347 1700 . . _Nerionw6de coversee . AnMwM4A
~
- -+ ..
]4720'0308 . , , 33 DAVbw 1 What Brookhaven has done is, they've reestimated
, 2* the data base. They've added more failures and discounted'a 3 number of hours that we had taken credit for and 4 calculated. They shif ted their scale up. correspondingly by
, 5 a factor of about 8. They had m,aybe twice as many failures 6 in a factor of 4 less operating time. So you multiply these i -
3
=7 by a factor of 8. And then they used our same type of i
8 approach to extrapolate from frequency versus leak size. So 9 what they've basically done is, they've shif ted up scale the
, 10 curve. And then when one gets down to 1800 gallons per 11 minute, one is roughly again at about a fa'ctor of 8 greater
- 12 frequency. ,,
. ~ 13 .
Now in the PLG 0465.results, we are .using ,,
14 medians. We neglect the range of uncertainty, which we have 15 folded into the RMAP study. We just fix on the median, so 16 what basically we end up with is disagreement of roughly a 17 factor of 8 and the frequency of 1800 per minute leakage.
18 '
MR. ROSSI: Could you say something about what 19 was the of facts on the ultimaty conclusion of this?
-t
- 20 MR. FLEMING
- The effect was not great, in that 21 even using Brookhaven values, they were dble to show that ,
22 with a one-mile evacuation that the case still holds. The f
i 23 numbers that one calculates for this event come up rather 24 dramatically. There is a factor of 8 difference in this 25 f ailure rate, but that failure rate gets squared in the
.' ACE-FEDERAL. REPORTERS, INC.
' 203 347 3700 _ , , _ Nasionwide Cowrese
- 800 33H646
4720 03 09 . 34 52)
Lis .DAVbw 1 initial frequency calculations. So it proliferates. It 2 goes up almost two orders of magnitude in the final accident 3 frequency rates," but they were still able to show that t
I 4 with one-mile evacuation, the safety goals would be met, and 5 it did not have an appreciable impact on the goals versus 6 distance curves, because of the low frequency.
7 MR. ROSSI: Trevor, do you agree with that?
8 MR. PRATT: Absolu tely.
9 MR. ROSSI: So it wasn' t important in the 10 ultimate conclusion? - -
11 MR. PRATT: In fact, it really didn't have an 12 very. great 'ef fect at all on this with no evacuation, and it 13 only had'some effects when it required an evacuation,,very 14 small. But you already beat the goal without evacuation.
15 So you are talking about even smaller values in the terms of 16 dose versus distance.
17 MR. FLEMING: I think the fact that we consider 18 the Brookhaven analysis to be conservative, I think adds'a 19 little robustness to our case.
20 l MR. PRATT: I was just going to add that. I 21 think' what we are saying is that yo.u have a couple of orders 22 of magnitude in there to play .with.
23 l MR. FLEMING: But just to clarify our position on 24 this, so that the people who are going to be reading the 25 I report get our input into what they might characterize as y i
- 1. . ..:
. .. ~. ; . .?
ACE-FEDERAL
,. ,,., . REPORTERS, g.,,,_. c_ , INC. . .-
..'. l I
35
}7200310 1 DAVbw 1 a realistic approach to this type of analysis, what we try 2 to do is check all these curves against what we think is the 3 most relevant experience data of all. That is the fact that 4 we have not yet experienced a major disc rupture in 5 interfacing check valves in roughly 500 reactor years of 6 experience.
7 We took the Brookhaven comment to cause us to
]
8 really question ourselves what valves we should really be 9 counting in statistics. We went back and tried to estimate 10 what we define as interfacing check valves. We define an i
11 interf, acing check valve as any check valve that is sitting 12 against the reactor coolant system pressure or one check C) ,
13 valve removed from such a valve'. And'we estimated that 14 roughly 80 percent of the check valves that we had included 15 in our data base would fall into that category, wh'ich is 16 roughly 24, 25 valves per PWR plant. Basically, all the l- .
, 17 places where the RER system interfaces with the hot legs, 1
~18 the cold legs, the. charging pumps and the safety injection l
il 19 pumps coming into the system. There's a couple two or three 20 check valves in each one of those lines, lots of lines.
l:
L 21 So we are talking about roughly 25 check valves 22 per plant that fall into this category. We had assumed 23 .something like 30 valves per plant in our original analysis,
() 24 so we went ahead and knocked our experience down by that
.. - . - . . , . _ , z. _
25 factor, roughly 80 percent, but then added in the additional l .
....- . , . . ACE-FEDERAL REPORTERS, INC. .-
l ' . 2 : c _ r + m s -> 7->>oo._ 2: 8 -+o ce .
= oo>>+46 :-+
4720 03 11 . 36 9OEt DAVbw 1 experience that's accumulated unt'il the end of 1986. And if 2 you assume that we had a disk valve rupture and equate tha t 3 with 1800 gallons per minute, say, in that experience base, 4 the maximum likelihood or the point estimate you would get ,
5 by dividing one failure and that amount of experience, would 6 fall about right here_. But as a matter of fact, we haven't 7 experienced such a rupture, so this rupture has to be 8 repeated or viewed a something like an upper bound. And 9 depending on whether you go to classical statistics or 10 Bayesian statistics, there are different confidence levels 11 for that event. .
12 That turns out t'o be roughly equal to the mean 4
1.3 value iniour original submitta1, and it matches up pretty 14 closely to the median value at Brookhaven. So we thought 15 that this kind of evidence supports our view that the 16 Brookhaven characterization was conservative. And that is
. 17 basically the only real ma.tter, I think, that was discussed 18 in great detail on the interf acing system, although 4
19 Brookhaven then goes on to run their numbers through the 4
L 20 analysis.
h -
21 I would think we agree that our conclusions are 22 still robust.
- 23 That is really the gist of our response on 24 interfacing.
', ; 25 '.. ". MR. NOVAK: ,
Let me ask a question.
I think when 1 . . .
, .,.gu. 4,.;d,
.. 7. .=.y:a x' =N:! ACE-FEDERAL REPORTERS, INC.
.n..,_,.,__ . _ . , . _ -
- ac
- - e-
'4720 03 12 . . 37 DAVbw I we talk about the interfacing, I think the comments of 2 Trevor Pratt stand. - And I think they are important.
3 Let me just make sure. I think I aske.d this 4 question some months ago, When you talk about the 1800 5 gallons challenging the capacity of the ' leak valves inside 6 the containment, is that a number that you stand behind for' 7 the fluid condition that is present during normal operation?
8 MR. FLEMING: Yes.
9 MR. ROSSI: It is two-phased flow?
10 MR. FLEMING: Yes.
11 ,
MR. NOVAK: Now one other question.
12 In looking at other plants, would you call .this a '
13 larger than normal release capacity for a pressurized water 14 reactor design with reactor coolant volume that is 15, occurring? The 100,000 gallons?
16 MR. FLEMING: I believe it is somewhat larger L
l 17 than average, although I don' t think that is the essence as i
18 to why we are getting different results than we used to U
19 .get. I think the essence is that other analyses have just j 20 assumed, they haven't really calculated failure rates versus
~
r 21 leak size. They just call the even.t a disk valve rupture,
?
22 and by definition, the disk valve rupture creates a high l
i 23 pressure challenge to the direct coolant system. So I think 24 this is the first time people have tried to calculate the h
. .. .r. . . . . . . .
1 4{.
25 -
relationship between check valve frequency and rupture.
. . -: % ~:
'A>
g:. - m.my n.
d e :a. :... s c.; ACE-FEDERAL REPORTERS, INC. . : .A f.
_ y.. pp.w., y me -
.x, 73w._
c.. -;'
1 l
L 4720-03 13 38 M
Ct DAVbw 1 Another thing that needs to be pointed out is ta t 2 all the data that Brookhaven collected, which was a larger 3 data base than PLG collected, all those events were problems
. 4 with 0-rings, improper seating and other kinds of events 5- thati did not point towards structural problems with the
, 6 valve disk. The only way you could get-an 1800 gallon per j 7 minute leak through a check valve is to have the disk i.
j 8 rupture, and the way in which we have come up with the check 9 valve failure rate is by trying to infer, extrapolate the 10 experience from a few hundred gallons a minute of leakage, 11 becaus 0-rings went bad, or maybe a little bit of boron 12 plated out and interfered 'with the . proper seating of the 13 valve. -
14 We are trying to get reverse flow to initially 15 test it to be leak free. The check valve that hasa to be 16 put in that position before the plant goes into operation.
L l- 17 Test procedures are in place to verify that. That somehow
- 18 has to rupture.
1 4 19 li li 20 -
a L; .
1 21 .
[ 22 23 24 h . -. ..
.:.;& 25
. - ~ . ,
- ' ^ '
~'
.+._ ,, g _1.c ~'
....g3. . . . . _
~ yv -
t' ,.
^
? . . , .
-s_ _
e.A . v: .4 ACE-FEDERAL REPORTERS, INC. . s L. - . 7mm . - e t.2 w.m _ awc_ ; ; -
- :fre ._ _
.4720 04 01- -
39 i 9.
4 . i 544 DAV/bc 1 The other thing I wanted to point out.here and I l l
2 forgot is that back in the original Seabrook PRA, we used a i 3 different approach to estimate what we called gross 4 disrupture.
5 The approach consisted again of taking check 6 valve data on interfacing check valves and apply this, the 7 first stage of a two-stage Bayesian updating methodology 8 that we use' to calculate uncertainties in our f ailure rates.
9 It's interesting to note the approach that we .
10 came up with in the RMAP study is pretty much bracketed by 11 what we came up with in the PRA, using a different
. 12 methodology. -
+s
) 13 However, I'will note tha t, in this methodology, 14 the data was so weak that this distribution was heavily 15 influenced by the assumed prior distribution.
'16 So it reflected a lot of subjective judgment.
17 But different analysts, a different approach. But it lent 18 some additional confirmation to the fact that we think that 19 our check valve failure rates were reasonable.
20 I guess the next area is the treatment of 21 shutdown events. -IE I might sort of summarize what's 22 happened here, the original Seabrook PRA, the original and 23 the update of the Seabrook PRA did not include an explicit 24 contribution from events that might have occurred during
- -?25 shutdown modes four through six. _ . '4
~
. , y.'
. ,.m_.,m.. .m m.:.
. c . i.a..
..z...m._.. e.w: ,.. _
ACE-FEDERAL REPO.RTERS, INC. 4.
. ..~.J_7; .
e -
~, .
I 4720 04 02 -
40 dh h2 DAV/bc 1 Generally speaking, the PRA results can be said 2 to apply ,to modes one, two and three -- mostly one. But, 3 in certain respects, it also accounted for modes two and 4 three.
5 'There had been a study done by PLG on the Zion 6 plant to extend the Zion FLG to include a consideration of 7 shutdown cooling events. And in response to a request for 8 information we received from the NRC, we adapted that s tudy 9 and provided some additional PRA analysis to bound the 10 effects of shutdown cooling events.
11 . What Brookhaven has done, it has indicated that 12 they were incomplete in their review of that particular body
')'
13 of info' rmation but went an'd included our curves 'to indicate 14 poss.ible sensitivities to include the shutdown cooling 15 events in the conservative bounding analysis. .
16 What we have done in our response that we're 17 going to leave with you today is that "w e have gone back and
, 18 taken one element of ~ the conservatism out of the previous 19 analysis, which was associated with directly assigning 20 accident sequences that occurred at shutdown.
21 . We had assigned those directly to release
. 22 categories. It had been calculated for accidents that 23 occurred from initial power operation. Namely, we used
~
24 release category S-6 to characterize scenarios where you
~
a decay' heat problem with the equipment hatch l _ 25 might hav 1 . , , . . ,.
L .F .' ...
N )). % st a. A. . . . . . . .
.FEDE5AL REPORTERS, INC.
- e
_ ,, . _ . . . m. _ . - . . . . . . _ _ _ , . . . . _ _
l 4720 04 03 41 DAV/bc 1 open.
2 We did not take any credit for radioactive decay 3 or the slower. evolution of the scenario that took place 4 during shutdown. We still have not done enough work with 5 something that we would consider realistic as f ar as 6 consequences for such an event.
7 But we have been able to do something that's 8 rather easy to implement. And that is, we have delayed any 9 release times in the bracket model by 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />, which we 10 estimate would be roughly shorter than 95 percent of the 11 possible scenarios. that would be consistent with our
. 12 analysis. Sort of like the 95 th percentile and the time of 13 curves 'that we might see in some of these events, based on 14 our assessment of the frequencies.
15 And we calculated the risk curves and that has 16 resulted in a noticeable reduction, although it's not an
. 17 order of magnitude reduction. But it brings in the dose 18 versus distance curves to this limited degree -to where we l 19 have recompute.d those and presented those in our submittal.
n 20 And to the extent that Brookhaven may wish to 21 continue their approach of using those curves to indicate l-22 that sensitivity, they may wish to use the more useful l 23 curves.
l 24 All the more recent curves do is delay it for 48 9 ,: . :25 ' hours. All' that's really being done there is taking credit l .
C % .<, - L' .
CE-FEDERAL REPORTERS, INC. #
-. . ._. = . . . _
4720 04 04 -
42 R
j'hll l
DAV/bc 1 for radioactive decay and source term. We're still not j 2 taking credit for the slower evolution of the scenarios.
3 MR. PRATT: That's what I was going to ask. I
~
4 was wondering why it didn' t go to zero.. You're releasing at 5 the same time, you're just delaying. ,
6 Okay.
7 . MR. FLEMING: We're locking down the core 8 inventory due to radioactive decay. I'm sure, if we look
, 9 into the source term area, you'll find the thing was of f-10 scale. We've at least gotten to the point where we have 11 just one incremental.
12 MR.' MOODY: Something else, Trevor. We should 13 tell yo'u about the original curves we provided. In going.
14 back and doing this, we found that the realist'ic case tha t 15 was plotted really should have been of f-scale. It was up on 16 the scale and it was about an order of magnitude lower.
. 17 MR. FLEMING: 200 rem curve. The 50 rem curve 18 was on scale, but the 200 rem was off. There was a problem 19 in the way in which the curve was drawn.
20 MR. PRATT: One of the things that did come out, 21 and this is something we have to resolve in our report as
.22 well as' yours to' make sure it's consistent, is the use of 23 medians in means.
't 24 We got very confused. At Brookhaven a lot of h,. suv'
. . J'i_ % . 65 - J. . . . , s, .... .
~. .a. .
- t; 25 time we were using the mean values and some of the responses d' r.
' * ^
.u..
' ~
. .u. -
, ' ,. - . a-u.
.' ?;
- :n l , , ;. .
sqhspsa AssgACE-FEDERAL REFORTERS,INC. -
M .I
i . _. _ . _ . _ _ _ ._ _ _...
4720 04 05 43
?b Rot DAV/bc 1 were coming in at the median. I think we should clearly use 2 the median of the dose for the distance curve, and the mean
. 3 to be consistent with the safety goal.
4 MR. FLEMING: We acknowledged that we have 5 ,
contributed to some of that.
6 MR. PRATT: Okay. Just if I could recount then, 7 what you did was you delayed the inventory as a function of 8 the 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> to account 'for decay. But you kept the same 9 release numbers.
10 MR. FLEMING: We transla ted them. We translated 11 the release times to 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />. And what that does --
12 MR. PRATT: But you still, the 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br /> 13 integrat' ion plume --
^
- 14 MR. FLEMING: That's right, we just transla ted 15 the analysis out by 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />. All that does physically is 16 radioactive decay the source term, if you're doing an
. 17 evacuation scenario.
18 It also gives you more time' for evacuation. But, 19 in dose versus distance, that's not meaningful.
20 Other comments on the shutdown events.
21l What we had done was we .did the analysis in two 22 parts. We checked' the design results to the extent they i
23 l were applicable to Seabrook and we reviewed the shutdown l
24' procedures. -
&- :25I l~r .. . _.
~
~
We found there 'was a high degree of similarity in
~
\.
. 1 e
g.
- .a-
. - . ; 7 ;
~
- d.. ,..
.. i.i. .. .d._ni.h. ACE _ FEDERAL REPORTERS,
. ,~ _ , , , _ , , . _ . . . . . _ _ . , _ , , , INC.
.i.,m=.. 2 .
.=,
4720 04 06 44 DAV/bc 1 the way in which the plant was taken down to cold shutdown.
2 And the maintenance procedures had a high degree of 3 similarity.
' ' But the chief difference, there are two chief 4
5 differences that we noted that we would be hesitant to just 6 use design results .at f ace value.
7 One-was the fact that Seabrook had two RHR
, 8 suction paths. Zion just has one. I think that's an 9 advantage for Seabrook.
10 The other area was that the support system 11 interfaces were a lot different at Seabrock and Zion.
12 So what we de :ided to do was look at the Zion 13 NSAC. I'm not,sure what the number is. The NSAC report for 14 the Zion shutdown cooling events.
15 We took a look at those results and figured out a 16 way to account for the two drop lines by tackling the
. 17 reduction factor.. And that reduction factor made use of our-18 insi~ght that in order to have the kinds of events that wire I 19 of interest at Zion, where Movs could spuriously hold shut, j 20 in a sense, that was the only way to get suction to the RHR j 21 system at Zion. .
~
d 22- That woul'd give you a temporary loss of RHR. And 23 you wonder why the scenario is of interest there.
. 24 We took credit for two drop lines and we ih
~.. .- 25
.e . . , 2. :. . -; . ' . . ..
'.cali:ulated a reduction factor which said that the only way ~ .
- ... . s ..~ '
j
. . .. E '. i. -
,.- :. ~ -~^i g.0:.p .. ..:.- H
~-
- m :~ . . _. ,
dew .nn
- .,n:.1.w_w 9..L..
,. n.mgn6. __ ACE-FED,ER,.AL REPORTERS, INC. ,
4m.&. e.
- ~
4720 04 07 , . 45 C DAV/bc l- such events.could cause a problem would be if some kind of a 2 common cause failure would occur, so you h.ad multiple valve 3 failures or you had an unfortuitous combination- of doing 4 maintenance on one train and having the valves close on the 5 - one that was operating.
6 So we had to take it into account. The 7 Brookhaven review of that question, whether that reduction 8 factor would have been justified, and made a point about t
9 there could be some single failures or instrument 10 indications that could cause
- h f act MOVs in two drop lines .
11 to go shut simul'taneously.
12 -
And the information we're providing today, we're
.' 13 providing information about the procedures for RER cooling 14 in Seabrook which take advantage of the fact that there are 15 two drop lines. And they take steps to remove power and tag 16 out one MOV on different channels in each train such that l . 17 you really still have to. postulate some kind of a multiple
' ~
18 ~ failure.
19 This is information that had been included in our
,.s..
20 original submittal, so I think we'll provide additional 21 data, which you may want to take a .look at on that 22 particular issue. , f But it also appeared that the Brookhaven review 23 l 24 wasn' t completed on these events. And we may anticipate
-T ..
h..i_,.;y.,S.25 some more comments, I'm not sure.
. Those seem to be the B. .;f ~:i G 39.k.;_. ,',p:.' ~ ~. ,
- .4.
? y. .
.2... ,4
$':- } }.3
[.
- w. [. $$w$
s; J z, ~
.:, $:. .. .hCE-FEbENEREhbRTERS INCe /
G
.fs. r.~ w.~ '
'G* . O.*.':-*T ' 1.K+?c*: n . m +*~~s
.+4 ' i"-
- s. ---*Xi:'* 201.wl.tman ~ *
. . h'Nanamunda commune. * ' ' - r ' ana. tis.Asa *'Y'~'"#
a i 4720 04 08 46 DAV/bc 1 major comments.
2 And as we had indicated in our analysis, even by 3 conservatively including these events and the treatment of 4 these categories, as has been discussed earlier, it didn' t 5 really challenge the robustnecs of our conclusions about a 6 shortened evacuation zone.
7 I think the additional information we provided 8 here today will strengthen that case even further.
9' MR.. NOVAK: Just one small point. Perhaps you IO can keep the response brief.
11 The 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />, typically, it doesn' t take a power 12 . plant 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> to go from normal operation to shutdown 13 ' cooling ~.
14 Explain to me where you're getting the 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />.
15 Do you plan to do something special?
16 ,
Explain one more time the 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />.
. 17 MR. FLEMING:_ You can in fact get the plant into
~ ~
18 cold shutdown in something like four to six hours, going as 19 f ast as possible. Typically, we would expect it would take 20 ' 12 to 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />, in a typical scenario of cold shutdown.
21 But, when we did our probability of loss of RHR 22 cooling analysis, we used to figure out how much time the 23 plant could possibly be on RER cooling, we used data from 24 the Zion plant based on a large number of refueling outages
~ 25 and maintenance outages that they had.
.S - h m ,. .
.. ACE-FEDERAL m,m REPORTERS, INC.
s , .-. ~_ _ , _
i j
4 4720 04 09 -
. 47 DAV/bc 1 As a matter of f act, the use of that data kind of 2 pointed toward a very long period of time; roughly, 30 3 percent of the time, the. Zion units were in cold shutdown 4 because of the operating experience that they had.
~
5 If'you look at the data on the individual events, 6 we'found that roughly -- well, the average duration of an 7 outage was something l'ike 700 and some odd hours. So if you 8 ask yourself the question:
9 So we had to calculate the probability that you 10 would have a loss of decay heat removal over an average 11 outage of 700 and some odd hours, and given that you have >
12 such an event, you want to ask the question, w. . ell, what is 13 the time' of occurrence of the . loss of heat output?
14 If you go through that calculation using the 15 simple constant failure rate models that we have used, you 16 find that that distribution is roughly uniform over that 17 time. Given you have that, the probability is dis tributed
~
18 roughly equally over time.
19 So, on the average time of occurrence based on 20 the Zion data we would use, it would be three in 400 hours0.00463 days <br />0.111 hours <br />6.613757e-4 weeks <br />1.522e-4 months <br />.
21 We picked 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> because it would. be roughly, you know, 95 22 percent of the possible scenarios based on time of l
23 occurrence would occur at longer times.
24 We thought that was a conservative way to account
, s) 25 for the delay between when you shut the plant down and when ACE-FEDERAL REPORTERS, INC. 1 l 202.1471*no : r WM ' fTAm
, . i 4720 04 10 48
$$ DAV/bc 1 you actually 'got into the initiation of the heat-up.
2 Obviously, you could postulate scenarios.
3 Well, you could initiate the heat-up as early as 4 four to six hours. But, probabalistically, rather than do 5 the specimen case -- , .
6 -
MR. NOVAK: I think your answer is assume an 7 accident 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> af ter the beginning of the shutdown.
8 MR. FLEMING: That's my answer.
~
9 MR. MOODY: Then the best estimate, we would 10 probably use 400 hours0.00463 days <br />0.111 hours <br />6.613757e-4 weeks <br />1.522e-4 months <br />.
11 MR. ROSSI: Do you have any comments on that,
. ' 12 Trevor?
13 MR'. PRATT: No. I think that's probably- one of 14 the more significant comments. I think it's useful to also 15 have the information, to have you're working with average 16 numbers and you want to keep it on a much longer time. And 17 saying that everything beyond that is at a 95 percent 18 confidence level that you're moving up to, it's another one 19 of these areas where you start off with your median 20 estimated and you build in your 95 percent confidence.
21 MR. FLEMING: We ' re making a trade'-of f be tween 22 picking a time and putting the thing to bed, or arguing over 23 ' about what time we're going to do that.
24 MR. NEWBERRY: Isn't the data and information you i
fh ~
25 used to come up with that -- -
- e. : ACE-FEDERhL REPORTERS, INC. . e..
. . . . . . . ~ ._.. . _ . . . _ . _ . . . _ . . . . .. .
l 4720~04 11 . 49 C.b. DAV/bc 1 MR. FLEMING: It's all there. All the details 2 are in there.
3 MR. LONG: One comment, or question on that.
4 If the probability goes off the bottom of the 5 curve, then I don't think what I'm about to say has any 6 significance. But it strikes me that you could break out 7 the shutdown events into a couple of categories.
8 One is the evolution-specific. type of event where 9 you're draining to the mid-loop, or something like that, and 10 you're very near that point in- time. And you have some sort 11 of problem. And you lose RHR cooling.
12 - Those things might be attached to a type of 13 , release category tha't would be. occurring a short time af ter 14 shutdown, where you would use a decay heat for if it looked 15 like 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> or something else around the time you picked 16 now, but you'd only be assigning the probability of the 17 events that really occur associated with those evolutions.
18 And then, later on, you would take another 19 category where events occur randomly during a period of 20 shutdown and assign that on an average type of release 21 tategory. It looks like 400 hours0.00463 days <br />0.111 hours <br />6.613757e-4 weeks <br />1.522e-4 months <br /> ,or something like tha t 22 because I think that laat category won't have any effect, 23 , and the first category will have a lower probability.
i 24i So there's still a problem. That may be an 25 approach to trying to figure out a reasonable presentation ACE-FEDERAL REPORTERS, INC.
. .,,, , . . - - m _._
4720 04 12 . 50
~
DAV/bc 1 of risk of shutdown.
2 I agree with you that we have over-characterized 3 that.
4 MR. FLEMING: We have tried to sort out the 5 difference between those kinds of situations and event tree 6 calculations in tracing down. I'd have to go back and 7 check, but I believe that we've , looked at what's dominating 8 the probability of these accident scenarios is actually the 9 long-term, because that's where you have the most exposure ,
10 put into the calculation of failures.
11 The _ time intervals associat,ed with having the 12 plant down to mid-point and so forth ' are very short. And 13 that knocks down .the probabilities quite a bit.
14 But if one were to come up with .a realistic 15 estimate of the time for that kind of scenario. Where its 16 probability could be calculated, it probably wouldn' t be a - -
17 ' whole lot dif ferent than the 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br /> but, in that case, it 18 would be more' of a realistic estimate than a conservative 19 estimate.
20 MR. LONG: But it would be a long probability 21 as'sociated with it.
22 MR. FLEMING: Tha ts righ t. Yes, that's true .
l 23 l MR. MOODY: We realize we could do more work 24 here.
25 MR. LONG: I'm only suggesting that if it still
. ~
ACE-FEDERAL REPORTERS, TNC.
.e . - - -
4720 04 13 51 D
2 6 DAV/bc 1 looks like a problem, that may be a step that needs to be 2 taken.
3 MR. NERSES:. Maybe, at this time, it's 4 appropriate just to take a 10-minute break, and we'll start 5 again at 11:30.
6 (Recess.) -
7 8
9l-10 11 12' 13 -
14
~
15 16 17 18 19 j 20 i i
21; 22' -
23 24 i 3
l 25 1 l
l l..
~
l ACE-FEDERAL REPORTERS, INC.
5
. I l
720 05 01 52 1 DAVbw 1 MR. NERSES: Let's get started. Jim, can we 2 continue on?
3 MR. MOODY: I think the next issue -- well, I 4 don't want to call it an issue. The next subject that we 5 want to discuss is steam generator" tube failures, and Bob 6 Renry will talk a little bit about the subject, and then 7 Fred Torri will discuss some of our comments.
8 (Slide.)
9 MR. HENRY: I didn't bring any of the handouts 10 with me. These were Vugraphs that you've all seen before.
11 What I thought I would do is, just to make sure we are all 12 on the same wavelength, I will describe the issue'very 13 briefly and summarize very quickly what we have done and 14 what we have supplied as part of the Seabrook case and also 15 tell you how we briefed the comments you've sent back to 16 us. We should keep that phrase pretty crisp.
17 The issue itself is one of, given a severe 18 accident, we've lost inventory in the primary system. And 19 obviously, this is a schematic representation. With high 20 temperatures in the core region and natural circulatory 21 flows between the core and the steam generators, which could 22 l lead to temperatures high enough in the steam generator 23 ! tubes, such that they might fail.
The reason for looking at this is obvious.
f) 24 l 1
25 Because of the relief valves on the secondary side, if they
~
ACE-FEDEIML REPO_RTERS, INC.
, . i 720 05 02 53 1 DAVbw 1 were sufficiently high to fail, then fission products could 2 be released and bypass the containment.
3 To address this, we used the MAAP code developed 4 as part of the IDCOR program. Seabrook being a member of 5 the IDCOR program, it has it at their disposal to use, and 6 it provides us with an integrated system response model that 7 allows us to address this specific issue on a 8 sequence-specific and even react'or-specific basis. There is 9 nothing substantially different about the Seabrook four-loop 10 Westingho.use plant as opposed to other Westinghouse 11 four-loop plants, but we did take into account their actual 12 heat sinks, et cetera, and we did take into account the 13 Seabrook operating procedure.
14 I will get into that in a second.
15 (Slide.)
i 16 The issue-itself requires that you have models.
17 Again, this is just a schematic representation. It was ,
18 described under degraded core condition. ' Natural-19 circulation flows in the reactor vessel which is shown here 20 and natural circulation flows at the top of the vessel, the e 21 upper plenum, inner plenum of the steam generator and a ,
22 counter current flow that can be set up in the pipes 23 demonstrated by experiments that have been done at 24
) Westinghouse. And then an interesting kind of natural 25 circulation flow in the steam generators, where some CE-FEDERAL REPORTERS, INC.
.720 05 03 54 g5 1 CAVbw 1 tubes chcose to circulate in one direction. Others 2 circulate the other direction. So you have a gaseous 3 natural circulation flow path between the inlet to the steam 4 generator and the outlet flow. Excuse me. The hot plug and 5 the cold plug is the easiest way to say it. And those 6 models were developed as part of the IDCOR program, as part 7 of the 'DCOR NRC issue resolution phase.
8 As I said, experiments have been done to scale 9 set up at Westinghouse. .
10 (Slide.)
11 These were all part of what we discussed. We 12 take these benchmarks very seriously, because this is the 13 way that you can get some handle on whether or not you've 14 actually described the proper physical processes.
15 I won't dwell on these, but I just want to 1ct 16 you know that the kl.nds of measurements will be and the 17 scaled experiments which will be done with water and,SF-6, 18- basically, measure the power in the simulating core, which 19 is in the vessel off the page here. It will measure this 1
20 flow in the hot leg. It will measure temperatures on the 21 hot side of the plenum, the cold side and the returning flow 22 coming back from the cold side, which sees additional i
23 cooling as it comes back from the secondary side fluid.
24 I am going through this pretty quickly, but I
))) ,
25 just want you to have an idea. .
cACE-FEDERAL REPORTERS, INC.
720 05 04 55 1 DAVbw 1 (Slide.)
2 When we looked at these, we certainly considered 3 any time you develop a model, it has to be compared 4 with the experiments. As I say, one of the things they 5 measured was the flow rate,' and we tried two different kinds 6 of simulants. The simulant in one case being water and the 7 other case being SF-6. The water case used a power source 8 of 28 kilowatts. The experimental measured flow was .54 9 kilograms per second. The predicted was .5. For SF-6, the 10 experimental was .016 and the model .017. So in that case, 11 comparisons were quite good with the only experiments that 12 'have been done.
13 (Slide.)
14 As I said, they also measured temperatures. The 15 hot temperature, remember, on the hot side of the steam 16 generator and the temperature on the cold side and the cold 17 . returning fluid. And to just go through this quickly, when 18 they measure things like where they measured the 2.43 19 1 kilowatts power input -- excuse me. The steam generator, 20 the hot side was a 30 degree C. The cold side of the 21 generator was at 19, and the returning fluid was coming back 22 at about 10 degrees C.
23 I am just going through the numbers to give you 24 some feel for the comparisons, and these are the predicted l
25 values. Assuming that you have this number of tubes, which 26 is actually transmitting fluid out from the hot side of the
. .-. _ . ACE-FEDERAL REPORTERS, INC. .
l
. . . 1
~* , e 720 05 05 ,
56 1 DAvbw- 1 steam generator and just taking 12 as an example, the power 2 measurement was 2.3. 2.6 is your predicted. The hot side 3 of the temperature. ,30 degrees C. Predicted 29. l'9 versus 4 21, 10 or 11 versus 11. ,
5 so where there is information.available, we have 6 taken the model which was used for Seabrook specifically and 7 compared it against the information, and we find very good-8 agreement with that.
9 (Slide.)
10 We also went to reactor-specific cases to took at
,11 .the types of sequence-specific, operator-specific 12 uncertainty cases that should be considered when thinking of 13 this potential for bypassing the containment. Obviously, a 14 very serious issue. You want to make sure that you've 15 described processes to see how much margin you actually 16 .have. We looked at no operator actions. We have a base 17 case,. We look at the influence of pump seal LOCAs, for 18 instance, that could influence the natural circulation
. 19 flow. We went to operator action cases we had with 20 turbine-driven AFW with failure of the AltW and manual 21 depressurization of the primary system when core temperature 22 reaches 1200 degrees F, as it did in the procedures.
23 We also looked at uncertainty cases. We had a
)
24 high core melting temperature, which just means try to push i
L . 25 the temperature up high to keep this core open and keep it
, -. - - - : ACE-FEDERAL REPORTERS, INC. .~-
720 05 06 57 1 DAVbw I available for circulatory flow for a long period of time.
2 We would use low fgiction factors for the upper core natural 3 circulation to t'ry and see if we can make this flow larger 4 , and what would be the influence.
5 Low steam generator flows, which is another way 6 of forcing the temperatures to be higher than the 7 generators, because it is pushing more energy into the 8 generators, but we can't get rid of them, forced the 9 secondary side relief valve to be stuck open. This is one 10 of the things that came up at the meeting we had at 11 Brookhaven in October, I thipk it was. We went back and ran 12 this case, and lastly, no matter how degraded and decayed, ,
13 the core would never block. So it always looked like it had 14 three standing rods, which is the worst thing we could do, 15 but even then, we didn't find that you would ?.est the steam 16 generator tube integrity.
17 (Slide.) -
18 In all those cases, this is what the inlet gas 19 plenum temperature turned out to be. We went through, as I
. 20 said,-base case. I will give you some feel for this in a 21 second.
22 These are all in Kelvin.
The approach for 23 potential failure of the steam generator tube, you should j) 24 approach a temperature of about 1100 degrees Kelvin. This
! 25 represents the inlet gas plenum temperature, the tube
- + 7 .- .5.
. .. . ACE-FEDERAL REPORTERS, INC.
o
- 720 05 07 58 1 DAVbw 1 temperature. It's even lower than this. But this is a very 2 convenient way of showing all these sequences side by side.
3 Base case, 860 degrees Kelvin, and what is actually coming 4 in at the bottom of the steam generator. The seal LOCA 5 makes maybe a 5 degree C' difference on it. Manual l 6 depressurization, obviously, helps, because you pressurize t
7 the primary side. There's no driving force. You also, as a l 8 result of depressurizing it, interrupt the natural 9 circulation. Well, let me say it differently. You 10 decrease the influence of natural circulation, so that the
. 11 generators don't get as warm. -
12 ,
Th,e more severe high eutectic. temperature, you t
13 might force it up to about 880 degrees. The point is, 14 you've only increased it about 20 degrees Kelvin as opposed 15 to something like 1100 degrees Kelvin.
16 Low steam generator flow is about 980. Stuck 17 open secondary side hardly has any influence on it, because 18 the transient helps cool the tubes for a period of time.
19 And then as I said, last,'if you force the core, it never
'20 blocks. No matter how degraded it gets, it still looks like 21 there is 4000 rods still standing there. You can look at 22 that and still choose not to call it an uncertainty but a 23 sensitivity case, but force that up to about a 1040 degrees g 24 Kelvin.
25 Ma. aOssi, what are ehe other teems you assumed
. . - u . ~ ..- - . : < ACE-FEDERAL REP _ORTERSo INCe
720 05 08 59 1 DAVbw 1 about core blockage? I'm not sure I understand the last 2 case.
3 MR. HENRY:
I am not sure what your question is.
4 These ar'e uncertainty analyses off of base case. We've gone 5 through each one of these individually. For this last one 6 ir says, no matter how hot the core ever gets, it never does 7 anything.
8 MR. ROSSI: So all of the others do assume core 9 blockage?
10 MR. HENRY: All the others allow the core to 11 de form. .
12 .
MR. ROSSI: And you do a calculation of the 13 deformation of the core, except for the last one?
14 MR. HENRY: Yes. For instance, the difference 15 between this one and this one is, as opposed to something 16 like 2500 degrees Kelvin, that will certainly move and
. 17 deform in block. You forced it to be 3000 Kelvin. It will 18 still deform, but it took a lot longer to get there. And 19 this one, you just never let anything ever change.
20 So this one you consider uncertainty. I see no 21 way that this is uncertain.
22 ' MR. LONGr Bob, you said the temperature you 23 have to reach in the steam generator tube to challenge is
{} 24 about 1100 K'elvin? Is that still true if the steam c -
25 generator is depressurized?
s . . .
ACE-FEDERAL REPORTERSo INC. .
720 05 09 60 l' DAVbw 1 MR. HENRY: It is slightly difference, if the 2 steam generator is depressurized. I don' t have the exact .
3 figure here. 1050, or something. I am going to come back 4 to that in one second.
5 (Slide.)
6 The conclusions themselves. I showed you 7 temperatures, which were the gas inlet plenum temperatures.
8 The tube temperature is colder than the inlet temperatures.
9 We find the base case that the steam generator temperature 10 is about 750 Kelvin, several hundred degrees C away from the
. 11 point where you would question the integrity. And the 12 worst case, which is the one you asked, is.about 850. Tube 13 integrity also indicates that temperatures above 1100 Kelvin 14 are required to fail. It is about 1050 or so, 1000, I 15 can't remember the exact number, if you depressurize the 16 secondary side, and you have not done anything to the 17 primary side, and there is no feedwater also.
18 So in all those cases, we couldn't find a case 19 which got us to the point where we think that this is a 20 realistic probability here. We can't find a credible way of 21 getting there.
h 22 We have looked at some which I mentioned, but we 23 also looked at this, but we choose to build our case right
{)- 24 here, because we think, technically, that is, indeed, the l 25 point.
- 5. C r .
x . . . ACE-FEDERAL REPORTERS, INC. -
720 05 10 61 1 DAVbw 1 By our reading of your comments, it is our 2 interpretation that you don' t necessarily disagree with the 3 conclusions that we got out od reading the comments, Trevor 4 and Warren, were that this is a new analysis that you 5 haven't had a chance'to get involved in the depths of it.
6 There are some questions that come up, some details about 7 how mixing occurs in the plenum and steam generator, et 8 cetera, but we didn't find anything that says -- two 9 things . We didn't find anything that makes us feel that 10 you've disagreed with the analysis or that there was another 11 analysis that came out with a different conclusion.
12 We didn' t find 'either one of those.. And that's
) _
13 more out of the report, but also ccming down from 14 discussions we've had in the pas't.
15 So I guess what we would like to come away with l 16 today, if there were actually a point or which you need to t
17 have a very comfortable feeling for this, is some way of 18 assessing the analysis itself. We would like to define that l
19 , path and get about doing it, so we can bring that part to l 20 you.
21 That is all I have.
22 MR. LYON: Warren Lyon.
23 . We have, as you pointed out, some difficulties l
j])
24 with_certain aspects of the analyses which'we need to get 25 together with you and discuss and pursue and. reach a
...e. 1, g .. - s .. . CE-FEDERAL REPORTERS, INC. . . . J. . ..
$A720 05 11 . 62 1 DAVbw 1 resolution.
2 We gave a few questions. I think you are. aware 3 of those from what you've just said.
-4 'The seat-of-my-pants judgment says that this is a 5 nonproblem. My judgment additionally says I cannot prove it 6 t a reasonable satisfaction with the information that I 7 presently have in my possession, but we will be able to 8 prove it down the road a bit, given the kinds of interfacing-9 I think we nee'd, plus a little additional information from 10 the Westinghouse experiments, which is forthcoming.
11 That is my judgment as of today.
12 MR. HENRY: Our general comments on the
) ,
13 Westinghouse experiments should definitely continue to 14' finish their test program. They've already looked at 15 substantial variations. We can't find anything that looks ,
,. 16 like there is a major uncertainty or a major difference, I 17 should say, but I think that information is going to be made i 18 available.
19 We would like to make sure,'if there are
, 20 questions in the studies to be reviewed in some manner, then
- l. 21 let's define the process and the schedule and get about and L
l 22 bring it to a resolution.
l 23 .MR. LYON: I would add that there's really two
{[ 24 ways,to approach this issue.. One is the resolution one i.- w -; ,_ M . .
25 you've just outlined. .The other one, which has been
. ~ , .. .
~. ' -
lp c.,
- n. N .. .h ,5. DACE-FEDERAL REPORTERS, INC. .y
+
63
)720 05 12 1 DAVbw 1 suggested in a number of quarters, is to effectively get rid-2 .of it by assuring that the RCS is depressurized, if you ever 3 get into the situation, and on that basis, I believe the 4 problem could be shown to go away, with what we know 5 today.
6 MR. HENRY: I think Jim made that point up here.
7 MR'. MOODY: It's kind of a chicken-and-egg type 8 thing. I am not sure it's even chicken and egg. It is 9 difficult when you have sort of an unresolved safety issue 10 that is somewhere off in research, to try to resolve it, to 11 put in operating procedures to define ways to reduce the .
12 frequency or even having dry steam generators, which is 13 another approach. Those are things that can be done, but I 14 think I would do them as a last resort. Until it's 15 resolved, until it becomes a real issue, it's as much of a 16 nonissue as it is a real issue.
17 MR. LYON: I've avoided recommending specifically 18 the depressurization, I think, because, specifically, we l 19 ha0en't looked at the negative aspects of the 1
20 , depressurization as a means of avoiding this issue, 21 whichever one it turns out to be.
22 I will add one other observation. I think 23 Seabrook, in my experience, has gone further in looking into fj) 24 this. than any other specific plant that I've been associated l
25 with, investigating or critiquing.
+dMh4.b N,.e J-;. ACE-FEDERAL REPORTERS, INC. m -
i
. 720 05.13 64 1 ~DAVbw 1 MR. NERSES: Okay. Oh, I'm sorry.
2 MR. MOODY: I guess the only thing I am going to 3 reiterate it again. I think the NRC has made decisions in 4 the past on' resolution of unresolved safety issues research 5 ' efforts. I think this one falls in that category.
6 Okay, Fred Torri.
7 8
9 10 11 ,
q-rg 12 -
13 14 15 16
. l'7 18 19 20 -
21 -
22 .
23 ,
g .
24
~
~25 -m.
.s . _
> 2b;4:sgw .g.s; ?ssACE-FEDERAL REPORTERS, INC.
....I. u ,
- -. ; ~ -- .- - . . .- --
isi .* .
720-06 01~ . 65 1 DAV/bc 1 MR. TORRI: I think what we are saying is that.we i 2 believe that this part of our problem can be made to go 3' away based on the analysis that Bob Henry has just done for j 4 Seabrook.
5 And I think, if the difficulty is that we ha'ven't
, 6 been able to provide you with an analysis here, that you
~
7 have a little bit of a difficulty to perform an independent 8 verification analysis with the NRC tools.-
9 I think it takes getting together and reviewing 10 whatever detail necessary, the basis for our analysis
. 11 models. I think that will help to resolve this part of the 12 issue. , ,
13 'I 'think we should move forward with that. What
~
14 I'm going to say is that, given that we haven't quite 15 discounted that, what are our comments on these parts of the 16 Brookhaven assessment?
17 (Slide. )
18 It's our understanding that the probabalistic 19 assessment of the induced steam generator tube ruptures in 20 the Brookhaven Lab report -- namely, the conditional 21 probability of failure to depressurize the steam generators 22 and the conditional probability of induced steam generator 23 tube rupture, given no depreciation -- is based on generic l) ,
24 . information th'at is produced or was produced under some
.. .b.~. . m- , .<
25 other NRC-sponsored research program. - -
. . o. .
MJe-muin. ~-9! ACE-FEDERAL REPORTERS, INC. . .+p;$= a
720 06 02' 66 1- DAV/bc- 1 Therefore, it was probably intended to say that 2 it's based on generic data. We don't know in detail what 3 that is, but it was not clear to us what consideration of
, - 4 the Seabrook specific analyses and discussions that we have 5 provided as part of our submittal, what influence those 6 Seabrook specific issues and analyses have had on the 7 probabalistic assessment'of those two probabilities.
4 8 Now', in this context, in the context of the 9 information that we did provide, it seems to us that the .2 10 conditional probability for failures depressurized seems 11 high for the following reason: ,
M 12 Even though there are no plant-specific M -
13 procedures to'do'that, the FAI analysis has shown that there 14 are a substantial number of hours available before it's too 15 late to take this action.
16 We believe that in those hours core exit 17 temperatures get up to very high temperatures. And the F 18 Seabrook emergency staff would be staffed and people in this "
a ..
19 , room would be on the phone with people on the Seabrook n .
20 staff. '
p . .
.. 21 .And the advice from the senior accident n .
o 22 perspective would definitely be available. It would take
,: 23 some sort of very serious miscommunication or
!g , .a..
24 misunderstanding of wha't's , going on at the plant in order
.. ...,. +
. .A . . , v. . . . . , :: . , :, c , i.
f ' 25 for th'is not'to' happen.[ '* , '. .'
, f[.,~ * * ' '
. ..- ~: . ~ . . " .:p ;,
A +
_ . , .h. na id x ,.... MJa!A6 Fs- EDERAE REPORTERS, INC.
- .i
e ,
720 06 03 67 1 DAV/bc 1 So we believe that, in this context, given the 2 time that's there, we think it must be much less likely that 3 the primary system would not be depressurized and that this 4 number would have to reflect some rather short period of
=
5 time available.-
6 Maybe the kinds of times that used to be 7 predicted by March for these kinds of accidents as opposed 8 to the longer times in order to heat up the steam generator 9 tubes, you have to have these longer times because the heat 10 isn't going where it's staying in the core anymore, where it 11 w
.as in the old analysis which gave us the short times.
12 So maybe there is a readjustment necessary.
)
13 The second point we'd like to make is that it was 14 not clear to us -- oh, secondly, this number applies, as we 15 understand, only to depressurization prior to occurrence of 16 the failure of the first tube if that l's postulated.
17 At that point, depressurization, which now must
,18 be a very clear message to the plant people, would still be 19 effective in two ways, depending on how additional tubes 20 fail. Depressurization at this point can prevent the 21 failure of additional tubes by preventing the further 1
22 heat-up of the heat transported into the steam generators.
23 Or, if it doesn't do that, at least it can jh) 24 decrease the source term that would result from a. steam s . . g . . .- ,
1 25 generator tube rupture event without depressurization.
$4sA Jes.1> DACE-FEDERAL REPbRTERS,INC. ..M .
- - . - - . - . - . . ~ ~ -- - -- --
.,7...- . . - -
i.
720 06 04 68
'1 DAv/bc 1 So there is sort of a second aspect to the 2 operator action for depressurization that so far I think has 3 not even been considered in your or-in our discussions.
- 4 But I think they are real in terms of either 5 prevent}.ng failure of' additional' tubes or in terms of 6 reducing the source term. -
7 Now that sort of leads me into this other aspect, 8 namely, the conditional probability for an induced steam 9 generator tube rupture to occur, given no depressurization.
10 First of all, it was not clear to us in reviewing
. 11 the Brookhaven' report what role small steam generator tube 12 rupture versus large steam generator tube ruptures played.
13 We interpreted the range that was given, .01 to 14 .3 - .01 applying to small, single steam generator tube 15 ruptures, where .3 might apply to -- or the other way 16 around, .3 applying to the conditional probability for a 17 single steam generator tube rupture occurring, and .01 for 18 an' event that would involve a large number of tubes.
, 19 ' "/,It's not clear to us that that -- or it was very 20 clearly stated in the report. That is the interpretation-21 that Brookhaven had in mind. We have some evidence that 22 might be the case.
23 The .3 value for the conditional probability of a g , 24 conditional steam generator tube rupture, we interpret as
' '. / 1 25 meaning that there is a 30 percent chance that the results
. .- ~, ,
l&..a-sm.4 c.., .q.:a ACE-FEDERAL REPORTERSo INC.
~.
w
_...y_._ __. . _ _ . _ . . . _ . . _ _ _ . _ - _ _ . _ . . . . _ . . _ _
720'06'05 69 1 DAV/bc 1 do not happen as discussed, are substantially in error by 2 several hundred degrees.
~
3 And I hope that if we have a chance to review 4 that analysis with you more and to review the basis and the 5 analysis methods for that, that your confidence that that 6 analysis is in error, substantial error, would improve.
7 And, therefore, this number correspondingly should 8 decrease.
~
9 Again, we think-this is based on the evidence we 10 have that says this is sort of a generic earlier NRC
.11 assessment without the benefit of the information you have 12 now. And it's not clear to us what impact that the analysis 13 we have submitted is going to have on these two numbers, 14 except we think they're not reflected in there.
15 We think it does make sense to distinguish
[ 16 between small and large-induced steam generator tube
, 17 ruptures because they would represent.significantly 18 different source terms.
I' 19 While the source term that has been used so far
. 20 in assigning all these sequences to the S-1 release category 21 might be considered appropriate for a case which fails many 4
22 or a large number of tubes and would release, of course, l 23 rapidly for situations involving a single failure of a tube, p
i . .
24 .. w e think that a very different source term,would apply in
. . :n. ' . ' .u. .:. -
.- .m m:. .
. '25 the context of the source terms that we have defined and
, , .o ..
h.L.p:n.w..d ..uf:ijij. iACE-FEDERAL REPORTERS, INC.
. r .- ,
l 720 06 06 70 2 DAV/bc 1 used in the study. We think that the S-2-W release category 2 would most appropriately reflect that condition; although 3 it's not being calculated specifically for that event 4 se quence , it's the one that we have used that would seem to 5 be most closely representing the release that one might 6 expect for a single tube failure.
7 With those comments, I think it's clear that we 8 would like to proceed with reviewing in greater detail the 9 analysis that Bob has done. We'd like to review that with 10 Bob and hope that that then will have an impact on these 11 generator humbers as we see them.
12 MR. MOODY: Thank you, Fred.
13 MR. BARRETT: This is Richard Barrett. I have a 14 couple of questions. The first question has to do with the 15 distinction between the large versus the small-induced steam 16 generator tube ruptures.
. 17' Is there any analysis that.has been provided that 18 would allow us to determine the circumstances under which 19 you might get one versus the other of those two scenarios?
20 MR. TORRI: Not to my knowledge,
~
except some very
, 21 preliminary thinking of what might cause one or the other.
22 If the failure mode of additional tubes has some sort of 23 domino effect resulting as a consequence of the failure of
.24
{} u..
the first tube, then the actions I described and
' 25 depressurization of the area of the first tube because of
.y
=... M_ . _. . , . _ ACE-FEDERAL REPORTERS, INC.
. ~. . Y- .
720 06 07 71 1 DAV/bc 1 the signals one receives probably wouldn't be affected.
2 But if these tubes behave independently, and if l
3 they fail because different tubes fail at slightly different i 1
4 ' temperature levels, the stress is built up then because of 1
~ "
5 the very slow heat-up rate that Bob has shown.
6 I think there's substan'tial time between the 7 failure of individual tubes. That's what I meant was the 8 basis for saying that, in that case, actions to depressurize 9 might still have a benefit at that stage in terms of the 10 effect on source terms, or in terms of affecting whether 11 . it's a large or a small, in terms of affecting source CN 12 terms.
59 '
13 MR. BARRETT: Thank you.
14 MR. MOODY: I guess if there's one thing that we 15 should be able to agree on here, it's the conservative 16 calculation which takes a .2 and a .3, assumes with a
. 17 probability of 1 that this is an S-1 release category,, the 18 whole sensitivity study appears to have a very conservative i .
19 upper bou.nd.
l 20 ,
Would you agree with that, Trevor?
21 . . MR PRATT: Yes and no. I mean, in my opinion, I 22 , think that's the case. We have of course diverse opinions l .
. 23 and we've had views done of the Brookhaven study which think
. f. hat that perhaps may not be.
(h .
24
.. m .. . . . . . , . . ,.;,.: . ..,. . . , .
.n .
l / .g % But, no comments on the source term.
' ~
25 ,
So, yes, M. iu}c..u. . sy .y.h. .;:r:sd...dCE-FEDEdAL s _ _ ..
REPORTERS, INC.
WAb
72 0~7200608 .
1 DAV/bc 1 it's certainly my judgment that that was our intent.
2 Rich, did you have two questions?
3 MR. BARRETT: Yes, I did.
4 MR. PRATT: Did you get out your second one?
5 MR. BARRETT: Not yet. I can wait for the second 6 if you like.
- 7 MR. PRATT: I can make a couple of comments.
t 8 One, I want to add to it, I think, in terms of the 9 conclusion that Fred made with regard to the .2 being 10 generic data and not Seabrook-specific, he's absolutely
. , 11 correct. There's no question about that. We did not take 12 . specific Sea. brook information into account in coming up with 13 that number.
14 .We have a group at Brookhaven whose been involved 15 in this review that looks at operator actions, and we'll 16 have them look at this and give us his opinion.
17 MR. ROSSI: Does the .2 assume the procedures are 18 available to tell you when to do it?
19 - As I understand, when you were saying . 2, it does 20 not assume a sequence.
21 MR. PRATT:. It's judgment on the ability of the 22 operator to'do that in the absence of a specific sequence.
.23 .;. MR. ROSSI: So the .2 is a number that he would
{p . 24 not depressurize in ,the absence of any procedures?.
And you
. t.: n e. - 1.r..+ 3 . . . :: - ..:~ .<z..-~._,.
.- '25 .would leave it to a management decision to decide that 1
. . .n it. ..
. . .. i ..L . - .. . . .
..d.,. .i_&d_s... m.. A..-.~
G$$,..1CE-FEDERAL REPORTERS, INC.
, 2+
720 06 09 ,
73 1 DAV/bc 1 during the course of a severe accident. Why would you ever 2 not have preplanned procedures?
3 MR. WALSH: Harry Walsh. I'm the Operations
- 4 Manager.
t 5 Every procedure we have is dedicated'to recovery, 6 not to locating the degradation. We'reenteringnewhround 7 here. We're entering into grounds now where we're telling 8 the operator: Forget trying to save the core. Affirm at 9 some place that it's not going to be hazardous.
10 I think we're into areas where misconceptions by
. 11 operators could put us into degradations of cores that could gc l' 2 have been saved. -
13 I think that's why we should keep it out of the 14 operator's hands , into management's hands. The time scales 15 that we're talking about that this could occur is well into
. 16 the time scales established by our regulations that we have 17 to be there. -
- 18 MR. TORRI
- I'think that's why the time 19 information provided by the FAI analysis is as important as l 20 the abolute temperature information that was provided,
.. 21 because it provides, I think, a solid basis for judging how
" ' 2 2 much time might be available and emergency preparedness
- : . . ' I.[ 23 ' action outside of the procedures by the operating staf f l .
(h..c.. . :;;.qp:.L"
. -2:..'b- 4 ( 2.4, w
.,,ould.be effective. . , - a.. .-, -
, - : .7.: . n't ':+ y - ,
~ . - T :::: ~: - .
..~J::f;O:28
' ~ ~
"f_'di:And this gets at the whole basis of the whole,
, ..,9... .
M.- e~ .
N we ex.a.v, . .' Je.:. < n r~ . w .~..m.
CE-FEDERAL REPORTERS, INC.
%N ~-
720 06 10 ,
74 1 DAV/bc 1 the root thinking of why these emergency centers have been 2 set up, what types of situations they' re supposed to be able 3 to influence.
4 MR. PRATT: I can't speak for the group 5 involved. It's in a different division from my division.
6 But I can confirm that the .2 did not come from him. I 7 threw it in at the last minute in an attempt to give a 8 perspective on this thing. That's where it came from. It 9 only came from the work that's going out in research.
10 Now, in terms of the .3, again, this .3 came from 11 a memorandum that was written sometime ago. And, again, as, g 12 to what.the ranges of uncertainty might'be based on the C;/
13 phenomena involved, as you say, at that time, the author of
- 14. the memo felt more comfortable with the lower range than 15 with the upper.
16 So there was already a degree of confidence 17 ,
expressed. So the more you analyze, the more comfortable 18 one might feel.
19 The only additional reference as part of this 20 generic work that is going on in research was that there 21 were also some subjective judgments given there as to teh 22 conditional probability of small LOCAs.
. 23 J Given the high pressure sequence, they were
,h ';'
. 24 enveloped by this.. So we
. ... ,t. w;v ow just went with the envelope and we
__ 1 '25 conservatively put the' envelope at .2.
$.iw.es. .
- ze?XcE-FEDERIL REPORTERS, INC. .<Ma4
- y - . .- . .-. - - . . . - - - . - . . - - - - - .
l 720 06 11 75 1 DAV/bc 1 MR. TORRI: Is our interpretation correct that 2 the .3 you think would apply given the higher number would 3 apply to la'rge leak induced steam generator tube ruptures?
,4 .Or is the lower number one that would apply to t
5 small? .
l 6 MR. LYON: Excuse me. Warren Lyon. Let me u
7 interrupt for a minute here because I think some of the 8 background on these things is leading us astray.
9 The .01 to .3 probability that we're bantering 10 around, the first time I encountered that was two or three 11 . years ago, when I plucked those numbers out of thin air and i, ..
12 stuck them into an NRC memorandum. ,
13 I think that's what you're referring to, is where 14 you got that. Right, Trevor?
15 ,
. MR. PRATT: I was going to put it nicely. If you 16 want to characterize it that way...
- i j .
17 (Laughter.)
- l 18 MR LYON: I think we ought to go back and look 19 at that for a moment. That came about in somewhat the 20 following rationale. I ran into a situation where it was
, . ~21 .necessary to form some kind of a judgment.
~
22 .' 'Now I asked myself, okay, do I think it's going
,.,123 to happen? The answer was to myself no.
h .
' .'.: .... ;. 2 4 z.?y ~ ;; ~
,... _ t;Then I' asked myself: . Well, suppose it does
, w ~ :3 ' = ::- 'o " . ~ > ~. , ' . eti.y ,:;c.: ry +' 1.:
n~ a.
l.~'
,-//*g25' happen? How confident 'ani I'in"that my judgment is correct? ~
. . ' N.. . ;;, .
-.,,::.:;. . ;,;[ ,; ; R . _
. , . r .w.
- , xq . - ..
.w , _ : .w . m . .
- u. .
. c. .: - . . .
. . z.. - . . . . e-es,sA,,44.x .ewsEACE-FEDERAL REPORTERS, INC. ,..: d k, . ~ .
o .* . ,
1 720 06.12 76 1 DAV/bc 1 And I said: Well, I might be wrong. I'm 2 probably right. I'll kind of put a bound, so to speak, on 3 my guess that it's low at .01. But just to, cover myself, 4 I'll go to the .3.
5 But, factually speaking, if I'm right in my 6 assessment, it doesn't happen. And if I'm wrong, it 7 happens.
8 9
10
. .. 11 -
12 .
13 14 15 16 .
17 18 19 20 -
21 .. . . .; ,
- e .
22 ..
.-- 23 . -
Q. ;...mw...... . . . . ...~..c.....
. ., . 2 4
. .. . w: . . . . ... . .
.. : n. .u . . . . :. . . . . .
~
- . 25 .0.~ S
. > ~ . .J ' ' ' '
- i. .
, . . i . .
, . . . .x .
. . ,.o. . ~ ; s... ,. . .. .
c.ces,ee, m.... =. a..k.dk. A. CE-FEDERAL REPORT.ERS
. . .. f INC.
o . 4... -
. .~
720 07 01 77 1 DAV/bc 1 Hey, folks, this isn't a probability yet. We're 2 treating it as such, and we do this kind of thing all the 3 time. One guy that I'm working with pointed out very 4 carefully: You really shouldn't do this.
5 So, recognize wh'ere this number came from.
6 The second comment is the small versus large.
7 Clearly, from what I've just said, there's no correlation 8 between those two.
9 My own judgment is something of the following. I 10 don't think there's such a thing as small. I don't have any 11 . calculations to go on. But I think we're missing the boat
.12 .if we think there is such a thing as small. ,
13 The reasoning is .Something as follows:
14 If a tube lets go 'because it's over temperature, 15 the flow rates through the tube at the time of f ailure are 16 relatively small.
17 -
At the time that tube begins to leak 18 significantly or, if you wish, break, the flow rate will 19 become quite large. That is. going to have a tendency to 20 lead high temperature fluid from hotter regions into that 21 location. ,
22 '- I believe that no matter what kind of a leak you
?
- 23 ' ' start with, if it's of any significance at all, it will very
? ~. . 24 rapidly propigate to a' full tube gone. ,
O. .j ,. :
-;;.j7's..:.;h- .- .
L,::;i,my.gi :, .
- - . iE 25 .3X'T At the same time, this very hot fluid will be
... <; .. . ,,..:.;. ' r .c. , ., ,: C .. , ,
. w:g.v .
O .
- . r, . . a. .
s . .Ac. is..FEDE.RAL REPO.RT. ERS, INC.
. ~ _ , , , . . .
. .. . .. . ..24 ...
720 07 02 .
78 1 DAV/bc 1 spurting out that opening. And it's going to be spraying on 2 to adjacent tubes.
. 3 Those tubes are going to get heated up very 4 rapidly. And my hunch is you're now in a horse race as to 5 how many tubes.you take out before the reactor' coolant 6 system pressure has diminished to the point that more don't 7 break.
8 My hunch tells me that: Yo'u break one, you're 9 going to break a batch of them.
10 So I assigned, given no real analysis or 11 analytical knowledge, I assigned almost a zero probability r3 12 to only, breaking one. And I.put it all in the big tube us -
13 break basket, until I can soo something else that tells me 14 I'm all wet.
15 MR. PRATT: If I could just add a little bit 16 more. To get more, elegance back into this than just 17 Warron's gut feeling, we did mention in the report that 18 there was additional surveying and export opinion done as 19 part of the generic research.
20 And the numbers that Warren came up with were 21 relatively consistent with the assumptions of the experts.
22 So it wasn't just -- alboit Warron did a good job of coming 23 up with the range ho did. .
24 .MR. LYON: Shoor luck.
[h ,.;_~. .
r-
. , .. s: m .. ;. . ,
,~. . 25 .MR. PRATT There woro other opinions in thoro, As ,k !..,.$ . . :.. ,. i:ACE $FED 5RIL REP NTERS, INC. . . -a .
- 7. .
, . I 720 07 03 .
79 1 DAV/bc 1 too. We can get the numbers exactly, but I recall it was 2 something like .2 rather than the .3. .
3 The .2 going to the large and the .1 going to the 4 small.' If I recall.' Somebody can correct me if I'm wrong 5 in my numbers.
6 So we had more than the one individual's 7 feelings; how much more went into the other individual 8 feelings, I don' t know.
9 MR. LYON: I'll raise a flag, Trev'er, on this 10 kind of thing. I've seen it happen to me before on other 11 things where.I made a guess, and a couple of years later,.it 12 was fed,back to me as authoritative and everyone believing 13 it. .
14 Sometimes, these' things have a way of 15 propigating, especially when you're working where you really 16 don' t have data.
17 -
tbdIknow',insomeinstances,wherethose 18 numbers that I just bluntly guessed d'id influence people who
'$.9 were working on the NUREG 1150 kinds of things, to a greater 20 or lesser degree.
l 21 .
Now, whether that influenced all the experts, I 22 have no way of judging.
23 . MR. TORRI But I think the important point we're i .
g ,, ,n, ,. ,
24 , making,is that there is an analysis that has been able to .
, . .. . x ; .t- . . . . -sa. .. c .
. .. o ~..x.:t.s .a . c . ~... .
l M 'i ' '. 2 5 address this issue in a consistent manner.
And we feel that
- 'l , ... ..
, ,. c f, , :. .t s.
J'.... .' ? , ; , .. .
' 4. 50Ms . w . .. . $. 5- M
..... ..m . XCE-FEDERAL REP RTE. RS, INC.
. . _ , , , , , _ .o v. ._-. .
+ . .
720 07 04 -
80 .
1 HDAV/bc 1 that analysis should -- that your assessment of that 2 analysis should find its way into these probabilities.
3 MR. FLEMING: I want to amplify on 'a point here.
. 4 This business of what this probability associated with the 5 operator actions means and whether it assumes procedures or 6 no procedures, it may be useful to try to break this 7 scenario down into a couple of different scenarios that take 8 into account not only the accident scenario, but the i
9 scenario 'that's going to unfold between now and when 10 Seabrook goes into operation, about what will be the outcome 11 or resolution of this issue.
[g. 12 ,. It seems like there',s two or three possible us .
l 13 things I can think of. Either on one extreme, we will i
14 convince ourselves that this isn't a problem and cannot 15 justify putting in procedures for this event, and 16 essentially argue that the last term is essentially zero; or 17 on the near side of the loop, we may convince ourselves this 18 is in fact a problem and put in procedures and operator 19 training, and so forth, to convinc'e ourselves that the
- 20 second factor is acceptably low. .
21 .
. Or there may be a middle ground where we may 22 decide that there is sufficient uncertainty that this issue i 23, is not going to pa away, to get"it out of the way in getting 24, an operating license for Seabrook, we're going to go ahead
- ljg,
.,. , =: . : . /. - : ;; . c2 .. . -e : .: - . :- ~ n. ;~ w-25 anyway and put in procedures, and,make the frequency o'f the
, , . .- .A ;- .. :~.", ,
e r , e *
- l *
.,r- ,_- . * ,se - . *
- ~ - - 's.
Asihds $5NA$ FEDERAL REPORTERS,INC. ,, d.hkk.]
720 07 05 . 81 2 DAV/bc 1 operator action go down.
2 None of those scenarios are very well-represented 3 by collapsing them into a single sensitivity case and 4 multiplying by .3 because somehow we have to take a position 5 on what we think the outcome of these exchanges and reviews 6 .are going to take us.
~
7 So it may be better for the decision-makers if 8 they are confronted with a sequence of sensitivity cases 9 with different assumptions about what's going to be the 10 outcome; rather than wrestle with what does this .3
. 11 probability really mean. .
12 MR. PRATT: I think that's an excellent 13 suggestion. Again, I'm giving conclusions here.
14 In terms of resolving the last item there, I 15 think, given my knowledge of some of the experts out there, 16 it might be kind of difficult, you know, in spite of the 17 analysis, I already have feedback from some.
18 A lot of people believe, given these natural
~
19 circulation conditions, that failure of the primary system
, ,20 prior to the core debris dropping into the bottom of the
~
21 vessel is almost a certainty.
'22 _, In that regard, getting on to the next issue,
~
. - 23 that's very good because it ' eliminates the problem of direct
[j l24 ... ;heating, and so on. So, from that point of view, it's good m . n. .:.. =. ; y x .. . . e
~
m.m. , ._ .a .
'25 ,
news. ,~
. Jg,[ .. J, . ' ,
. s .- ,
^;.
2 .. =
.g@'t
.. - . .,. . I. _V.
..'.'T _::.._' n . Ab-FEDskAL REPORTERS, INC.
)
720 07 06 82 1 DAV/bc 1 And they believe from a chronological point of 2 view that that's going to happen. But if they then 3 introduce that as a way out of direct heating, then they're 4 not going to worry about where the better location will be.
5 You may have trouble convincing them. So 6 there' sone perspective that I don' t necessarily share. I 7 don't know that there's people out there that don't --
8 MR. LYON: Let me say it again. I'm not sure it 9 came across clearly earlier. Clearly, one way of making 10 this problem go away and making the direct heating problem, 11 to a good degree.go away -- although perhaps not totally --
g' 12 is depressurize. -
a .
. 13 I said earlier that's been suggested. If I were 14 involved in that kind of a process, I want to hear very 15 strongly from somebody like Larry over here:
16 Hey, what do you get yourself into? And where
. 17 are the problems? And what are the risks -- before jumping 18 ,on to that kind of a bandwagon.
19 And I think I'm hearing the same thing from you 20 folks.
21 MR. FLEMING: Exactly.
22 MR. LYON: The same is true for any other kind of
,23 fix for some of these pseudo problems.
, .., 2 4 .
,,7 , ' .;,MR. FLEMING: Absolutely.
h ..';;.. . ,n , ;. ; .
, ;.. . . v n , . ~ ., -
-. 8 . . .
~'
4 -f , . ./J' '25 '
.. 'i MR. HENRY: , Trevor, just one thing to keep the s
. - c., . , .. .. .
J.=O:s.a. n .ls s. A.. =% A.CE-FEDERAL REPORTERS, INC.
.. ... . .. . . c ... _ .. . . . . . .
, + ~ . . .M>.
l.
720 07'07 ,
83 1 DAV/bc 1. modeling in perspective, some of the people that you may be 2 talking about have some questions about the steam generator 3 tubes. .They haven',t looked at this part of the problem.
4 And 'it should b' e continually pointed out to them 5 that there are three circulation paths -- one in vessel, one
. 6 to get you to the generator and the other how the generator 7 circulates.
8 Most of the people that may have these questions, 9 the ones that I know, have only looked at circulation in 10 vessel. As a result, they can calculate high temperatures.
11 What you quickly find out when'you add the other q -
12 ones on. is the. result of calculating high temperatures in 13 , the'veissel is you've created another path that starts 14 distributing the energy.
15 And if you try to distribute it to the 16 generators, that's a huge energy sink, this is not -- well,
. 17 there's a lot of details in the analysis. You can get to 18 this conclusion with some fairly straightforward hand ,
19 calculations that says:
4 20 If you only keep it in vessel, sure, it will get j . .21 hot. But if you let it start moving out through the hot i . -
22 legs into the generators, it doesn't get particularly
~
I l
. N3 hot. " And the generators don' t get particularly hot either.
h4 .. ;. So it's always worth emphasizing the fact that ,
h- (.>.:..n M
{ ,
.7 ,
m.; b, V r .:-< . 3.-
% 1, 4 .r.4 4-;'. .i ....
- d. ')
' there's three pieces to'that pie and those people should
!'". . b .' C25
., ?- ,c
- .. . u .- - m .) u . .j
- 7.;.. .O. . - .
... 2-
.v . .;;.v.. ....; -.
..t...
- , . . . .. .. . . . .p. -
..e.. .
./, .m.Q.s w-
.db;wA.s, ggMh;.. ACE-FEDE.RAL REPORTERS
, f, INC. ; .Wyd5.. .
720 07 08 '
84 1 DAV/bc 1 look at all three pieces.
2 MR. NERSES: Jim, I'm trying to figure out how 3 much more time you think that we will need. I'm trying to l 4 figure out whether or not we ought to take a break at 12:30
~
5 and come back.
6 MR. MOODY: We probably should. What time is it L .
7 now?
8 MR. NERSES: 12:25.
l 9 MR. MOODY: Why don't we take a break. ,
10 MR. NERSES: How much time do you think we have?
,, 11 MR. MOODY:
Maybe a couple more houra, maximum.
l.
12 .MR. NERSES:
Fine. .
13 MR. LYON: Rich had another question. I only
- 14 heard the first one.
15 .
MR. NERSES: I'm sorry.
16 MR. BARRETT I would like to squeeze one t 17 question in before you take the break.
18 MR. NERSES: Go right ahead.
19 'MR. BARRETT I'd like to focus on this question L 20 of the .2 probability of failure to depressurize.
! 21
).It seems to me that, in general, these high 22 pressure sequences that we're dealing with here that give 1
23 ,
you this potential for team generator tube rupture are l . s . .
~
lg.
,3 24
- , . g . . >
, almost exclusively transient sequences, as opposed to
. , s . . .u . .,. .. .,. , . ; .. .c . .u, , . y. ,,3.2 ,. m, .s ,. . .. .
og _ .
3,25 small' or~large LOCAs. ', '
. ' ' [, f,',. * *:; .c. , ,1
. . n . '.
- l. .. . . .. . v. . ;< ._ . ,:. :. g. ;,:. .- ..
l .
. ..c.....
.gg.3,. . . . . . .. .
Mb.;mamut i!...dSNI_CE-FEDE_R4C l21..,d!NU REPORT,ERS , 2 ...
_ .. - wINC.= ..u.+aA.li.,I .-
. /
o .
720 07 09 85 1 DAV/bc 1 In general, you'd expect a significantly degraded 2 status of a lot of the plant systems. Particularly in a lot 3 of scenarios there are electrical systems -- AC and DC 4 systems -- and in many scenarios, the numatic systems that 5 might be'rel'ied on.
6 The discussion so far on the probability of 7 failure to depressurizo has focused on the failuro of the 8 operator. But is there a significant probability that even 9 if the operator tries to doprossurize, that the systems that 10 are required, the instrumentation and control systems, the 11 air systems,. the DC power or whatever kind of power is 12 required to open the valves, that those will be availablo?
13 Especially, for instanco, in a station blackout scenario, 14 which, in many PRAs, has turned out to be very important.
15 MR. FLEMING: We have not done a definitivo 16 ansessment. In general"'there would be contributions from 17 the operator an,d contributions from the hardware. We would 18 take a careful look at each, doing an assessment along the 19 lines that we've already dono for the operat'or accidents i
20 that we have analyzed.
21 . But I can comment, at the frequency lovel of the 22 first term, the frequency of the high prosauro core molt
. i.'23 'sconarios is ono of station blackout scenarios, as you noted
') ,. ,24 The principal hardwaro that would be nooded to
.i.:i,6 +*:
g ,t.~.
thora.; .sy ;.. ;
.. . nv.,y;, ;a. x. :.y . .
^ , , ,, I .' 2?nupport 5 the operator action to depressurizo would be DC -
. < ' .L .. .
....r..,... ,
s@hdh bshdACE-FEbE.Rk F1PdRTERS,INC. 3... 1
, _y . . ,
- t. . f 720 07 10 ,
86 1 DAv/bc 1 power.
2 The conditional probability of unavailability of 3 .DC power in the station blackout scenario, I believe that a ,
. 4 careful assessment would very likely conclude tha hardware 5 contributions would be small compared to the kind of numbera 6 w"e would probably assign for operator actions, which could 7 be in the range of -- I'm just speculating now -- 10 to the 8 minus 2 to 10 to the minus 3 for the operator action.
, 9 The contributions of the hardware wou'id normally 10 be expected to make a small contribution compared to the 10 11 to the minus 3. .
12 MR. ROSSI: Do you need the backup system for 13 doing dep'ressurization? .
14 MR. FLEMING: All we need is DC power.
15 MR. ROSSI: That's the way it's done now?
16 MR. FLEMING: Yes.
17 Mel, how rapidly do you depressurize? The 18 solarold opens the valve?
19 " MR. WALSN: It releases the valve, allows the 20 pressure underneath that.
< 21 ,
. MR. BARRETT: Air pressure?
21 .
MR. WALSH: System pressure.
J23. ,
MR. MOODY: You did remind me of an interesting
....3 24
[ point. This is Jim Moody again.
g:.i;"",92,,T.
._ . .% M'.,? .. .. . ;; O . . '; ... .... '
.: .v .
U . , .1 ,' ,'?25~ ; 3. . .' The value that we're'using here -- and it's a
.u : .m '
- s. - *
.... . . . c. c .
y$ a weax.p!
. 4 ,. ACE-FEDERAL REPORTERS INC.
.c-
. .3:.W...k.sh.m.c.u.
.. .-- - +. w. . ..r mM.: u
720 07 11 .
87 !
l1 DAV/bc 1 value that we gave in response to an NRC question -- is not 2 only a mean value, it's conservative.
3 It wouldn't be too difficult to reduce the 4 frequency in that number, even by looking more closely at 5 recovery that hasn' t been taken credit for so f ar, as well 6 as minor changes.
7 So this' conservatism is built across the i- 8 frequency of this event. And almost any one of them can
'9 make a difference. '
10 MR. FLEMING: The conservatism he's referring to
. 11 is the fact that we did not go through carefully, through a g
w 12 scenario by scenario analysis, and identify and leave out 13 those that would not, in fact, meet high pressure 14 performance.
l 15 We need to look at some hot damage states that wo 16 saw which tend to be dominating by those kinds of 17 scenarios. But we did not very carefully go through and
-18 look at recovery actions, and so forth, to come up with the 19 estimate.
. 20 , MR..LYON: Bob, sometime ago, you roughed out how i 21 .long it took to depressurize Seabro'ok with one valve. Do
( . . . . , ,
' 22 you recall what that was to get from full pressure down to l , ,. . ...
s, ,
- . 23 no more than 100 or 200 psi?
.. g. .. :;.. ;
h ..c . . . : .24ys.. O.s:.p'. R.:. ex..).e.[d,; . u L. n :4'I..
, :MR.
With a single PORV, I think you're *
- >,n t..HENRY . . .. . . -
. ' . ' . ' ,'. ;.25 depressurized in 15-20 minutes. .
- u. . . .
.., p, . ... - y ;. , . .,=, ,
. . . . . .. n . ,.; :. . . . .. .. . . :. ,. ;. .. ..
Q6N55$
a nw%.m 3.~ ,u.d.%nMOICE-FEDERAL m~
REPORT. ERS, INC. r.m.uk
.. . d
.,,.~. .
720 07 12 88 1 DAV/bc 1 MR. LYON: And you've got two and they're ,
2 safety-related.
3 MR. FLEMING: That's correct.
Okay.
4 .
MR. NERSES:
5 MR. TORRI: ThInt was an interesting question 6 though. .If we say there is that much time left to do it, 7 you want to be sure you don' t wait too long, be .' ore DC power
- 8. goes out.
9 MR. FLEMING: In the Seabrook PRA, we did 10 quantify a time distribution on the, availability of DC 11 . power. And there',s a fairly high confidence of having the
. 12 DC power available as long as 10-14 hours into station
- 13 ' blackout. It is a realistic asseAsment to take into account
. 14 load shedding and so forth. :
15 . MR. TORRI: That's sort of the time frame I 16 remembered in your analysis, isn't it, Bob?
17 MR. HENRYt I'm sorry.
18 MR. TORRI: , Ten to 14 hours1.62037e-4 days <br />0.00389 hours <br />2.314815e-5 weeks <br />5.327e-6 months <br /> of availability of DC i ,
19 power, the times you calculated to get to where the 20 temperatures were leveling off was. that long or longer?
21 ,
- ,. . .In my recollection, it was longer.
22
~
MR. HENRY: You don't have,that long that
. 23 auxiliary feedwater hasn't been available.
c... 24 , MR. MOODY: You're talking three or four hours.
g' db s,/', ..',f /25n .
. , , , ,,,n. ..:.
n .: 3. . w .. : . s . .. s .x . . . .: .w. . .; . . >+.
. MR. TORRI: For tubes to ' heat up. '.
$n . , L. :- 'e. .; : . ~ .c .
<: q-
?!: * . ' . 'f
. .. .:.:.q .*- . -
' c' : %;, .
MbdNe,. NeMiiAch-FEDERAL REPORTERS t INC. ig.$ 1 :
r
'~^720 07 13 89 x /
1 DAV/bc 1 MR. HENRY: To uncover the core, to degrade the 2 core, and for tubes to heat up, that's a period of two to 3 three hours depending upon the sequence-specific aspects.
4 If you have DC power that long, the principal thing you will 5 do is run your aux feed if you can.
6 That's about all. You have to have water.
7 'MR. WALSH: Turbine-driven.
8
- MR. NERSES: Okay, 1:30.
9 (Whereupon, at 12: 30 p.m., the meeting recessed, 10 ,to reconvene at 1:30 p.m., ,
this same day.)
11 . .
h ,-
13 14 15 16 .
17 . ..
18 -
19 .
20 .- -
21 . . ,. . .. .,s....,..
. , . ,, i. . ,
22 . ,.
.=
..i.,.......
', 2 3
. 1 1.',
. ..=
..y . .' 3 ..'
- r a
.c , ; /. . ,
- M , 7 a .. ., . , ;
s>
l...f a'~,
<>'. 4.'..<+.' '..-
3* '**k 1.~ m.r r; <. .- . . :. ..
9.;), .~.. ~ '..', :?".. .. : " . 2 4 ;
~ * * . ~ -
. . . .. . *,p.'y.t:,;'{.:.v',;.np.J:
.. .% . . . :. . ' , r y : .* . ... * ...- *
- ..: : u.b;>, .'
.. . q . ; ,c : , c.
. .f.f,,....... ,s .. ,,s,,_.
'25 . ,
( s ,
, , . ,; 1
.<.y , r;,
, . p, . y ; ., ,.. .a . ; . . , .n ,
,f: ', . r,; @;,.7 .!:J..;A
,,... . i. .; ,. s -
'j 4:
, ,. . . . : . . ," .t , . ,, ,. .
ENGMig'6.: 3dg;d.= . rfACE-FEDEEREPORTERS n gj,(sh. INC
=. .
"3720 08 01 90
/ .
v 1 DAVbw 1 AFTERNOON SESSION 2 (1:30 p.m.)
3 MR. NERSES: Let's get this meeting-going.
4 .. Jim? -
5 MR. MOODY: Are we ready?
6 The next subject is direct containment heating.
7 Bob Henry is going to start off again, and then Fred Torri 8 will finish up. .
9 MR. HENRY: As Joe said this morning, and most of 10 you have heard b'efore in previous meetings, we don't really 11 see where direct. containment heating is an issue that should 12 influence Seabrook. The df.s,cussions we' e had in the past, 13 several of the NRC Staff'and consultants who visited 14, Seabrook, more or less said things that make us think they ,
15 share that conclusion.
~
16 Let me just, in terms of the issue, remind you, 17 for containment to be sensitive to the issue of direct 18 containment heating, it means that tens of percent of the
> ~; ..
19 core, like half the core, has to be discharged directly when
. : . a., , . . 7
. . . , 20 .the reactor vessel fails at an elevated pressure, and all
.. 1 . : ,
., y .,; . -
...i21I" that material, thatISO tons of material, for example, must
, ,y .: . -
- 1. -g ; y .jj:q.3. y y. . , , ,
. .' 7 . 2 2 .be finely: broken down.g.g to. sizes which
., -.f,.J..; . are submillimeter and
?;
_ 7, .;. y. ; ., -:. .
. 23 . move throughout the containment volume, so that the energy
,;' 7 - f t .,...
7.y n. . -
jv..g,. ?-T::.DM
. C 24 is tiransferred directly from the. debris to the air inside
/ I,52. 4@N?-VitWsMyp9M16@$.ef.'(:@Wr.W9, - .
M.\7-l l*h:CJ, i. .jy 25 the containment, and then they.also . include some energy w* 7. ,. : *
..n...
,p.F,
,',ij'....f d' 1 . .
- :; . , .w .v '
r . . j~j. . . . -(..r
'S , # - ,,
.. ...~ v e ,
, g l-t.
4
.4
- 6 A p.. Jr,, ,e.
1;... .e' 4 .i- .y,.,.
.s L
- /y .3 , ' .*,'* 'jd, K .%4{?. .2f.3,*.gli,g.tgsp*
.emp.
t
, ~ * .*-
' w . *. . . 's;> y..gy ye . a. . z :. ;, . . . . . .-% e . w .' r v.-* a ;,;. ..>. v .O. w~ . . - . - .>.,
~
.' . - . . - . _ . ~ . - /. . . -
. 3d,Thblai .
~720 08 02 91 J -
1 DAvbw 1 that results from further oxidation of unreacted metals. I 2 The key issues are that that material.must find its way 3 directly into the containment atmosphere and the issues that 4 we feel are so specific,for Seabrook are the shape of"the
- 5 reactor cavity plus the instruments table region, which can 6 collect any material which is trying to come out of the 7 cavity region and deposit it in the lower compartment in 8 close proximity to the biological shield and the rest of the 9 instrument tunnel, and also the very tortuous paths and 10 tremendous amounts of structure that this material would se
. 11 ,
if it were trying to, move from thel weer compartment inside g 12 the missile shield up above the operating deck into the 13 upper regions of the containment.
14 IDCOR has done some experiments to demonstrate 15 what is influential for Seabrook type of containments.
16 These experiments are applicable to Seabrook, and they show 17 that given the structure in the near vicinity of the ,
.; , 18 ' instrument tunnel and given the three-dimensional character
' "19 of the flow path, if you will, that there is virtually no 1- .
[ 20 poter.tial for directly transmitting molten debris up into
.- 7 ,- , . :. _
' . ;,h$- :,, o y. 21' (the containment atmosphere. ' -
~ '
. ; m f -
-C. n . .
Nic.-1'-,i.~.' . < _-[., ,;.2j2'.
. I brought some Vugraphs which relate to those
. 7 _ ;. ;. .; . - -
. GM.~ 1?i ;-t 23 '; experiments. We went through them at the last meeting.
_;"' q [ ~f.:4;%-
3 :_ - ,;..- __.y :.. ..
k 24 _iUnless anyone has a burning need to see them, I won't bother.
hd x .A-dc .w. 5.: &&'MM h &@. :: 5&K. .-' w ~ ~ =&::+.12-1:t.n:x-P +s= . . -
!....Q,ips.3.w.Ql.
ur .+Oi m.25- i$$i
-- ~
yy If you want to'see the . experiments again, we
' .-gg+;[:f -A ..g ..pp,.
. showing them.. j p..,, g ; _ ;.ep. .,: _,_3. _,.
-M-K%, v. ' . n. m. . , .-;: .W , . . ,: U.: : N.; .: 7 '::-8.:a n.,w.' ..: -
- .-- f.
--. e.. w y ' ' > :<
gly.,a.... w ' : ::, .1 ,.: . .
> - . = * - ~ +. - -
. :w : -
M$LM .m 55@M2.
+.w.n=@,dd5 A,,,_ ACE-FEDERAL - . . . .
REPO.R,TERS,
.. = ~-
INC.
_ . . . ~ . . - .. 25 wu--
i 92 g7200803'
,1 .DAVbw 1 can walk through them again, but we did discuss them at the 2 last meeting.
3 Give those experiments, given th'e experiments
~
~
4 that have been done at Argonne, which were EPRI-funded and 5 part of the IDCOR effort, and 'even looking at the Sandia 6 . experiments, we see no conflict in all the experiments.
- j. ~7 They all make sense. Argonne has don.e the experiments with 8 the appropriate structure. IDCOR has done them with the 9 appropriate structure. The on'es at Sandia have never been 10 done with a structure that is representative of Seabrook.
11 all they've really done is show the material can leak o,ut of .
12 the cavity.
13 There is no disagreement about that.-
. 14 'So finally, to summarize. O' u r logical path is, 15 we agree the debris can come out of the cavity. Since the j 16 experiments have been done with the appropriate structures l _
j 1,7 they show, those struc ures are tremendously effective in 18 preventing direct containment heating. Those that have been 1! - .
0 '
19 ' don's with. the right' materials, such as Argonne, have been 20 e done.in a very staged, logical process.with IDCOR to show i; .. , .n.... .w * ;s,
- l__ -
L -
. .. .'21 that all these experiments with Sandia through Argonne make l;. . .: .
. - ; . . ;3. =,gy. _ y L'
~
c 22 fundamental sense. ' .N. w- :p.: 1.. 3 v .- .
y _ _- .- ; .e.=_.- g. s. -
, ? . M.. "e23 . , P jAnd.the final conclusion, then, if you can't get
.. .s .
24 4 kind of material directly into the _ atmosphere, then there is
!gg , j. ~.W, Yn~ .jMG? 25 4:4% c .: 9;.lAp.'nG ~
i i;&ys&ki;QQ:: yy.i'ff^p . ; _
- c. '.:.3 y: .l:; "e ~
.- u ;jy.
no issue for direct containment heating. .- @
- &l L.
- a, .
- .. - w.w w .. r: *^
'Qi a..-Q.
^* '
..n.
- ,,L s . s .) ,t.
l, ,
.. n. .. m,,
"s.*}.s'.,,, s ..
r
~
r
.'.5.
-.,ry 1--- rp .w......c....
Q', q'J}Q
- qil,s i.*& . *, ~. s, g'.,
~ *. ,*Q_
' . { g; .,
. 3~.
,G.jf
_ n w . - _0 _CE-F5 EPbNh$RS,INC.
~
5EACR ---
...;a-$aY n
t 93 g7200804 ,
1 DAvbw 1 MR. NOONAN: Bob, if you don't mind, I would just 1 i
2 like you b'riefly to walk through the slides real quickly.
~
3' MR. HENRY: I will be happy to.
4 -
.. (Slide.)
5 What we are trying to do -- and this says " Zion,"
6 but from a fundamental point of view, there is no difference 7 between Zion and Seabrook. These are IDCOR-funded 8 experiments and part of the IDCOR documentation.
9 The one I am going to walk through is a series of 10 experiments to try to build from something like Sandia has
.11 done to what the real reactor system is. When it says lip, 12 it means a representation of the reactor cavity, plus the .
13 lip, we will get to it in a second. I'll go tlirough that.
14 The 2-dimensi 7.al representation of the cavity plus the
. 15 lower compartment steam generator experiment with a reduced 16 area. I will define what that means. And finally get to 17 the 3-D, which is the actual meaningful test, because it 18 represents the changes, both in flow path configuration, 19 flow path direction and in local velocities, which we count
, 20 on.to keep all this material airborne.
l
,s -
e .
L21 .
r.
,.;.,.So without doing a very logical set of
- c. . 1. -. - ~ - -37 7
'22 -
experiments, it is hard. to decide what controls are
- .s ,,. , ..- ~
' .,4 23 ..e s s e n tial . That's what'these are" set up to'be. ^
So I've got
.W. - -
, z .- -
.' J '
. 24 -
h) . -
D'.N
~
. , . N.J.jkM.N N ..7 to'come ^ back and walk you through this.E$lh
-:~. AS.The key thing,' the eason I want this up here is,
- 2. ~
TM
'. ... s:-f.'25
- . : g.- .4..,.-: ; ;c - .=m -+ ; ; ~:
m _. . ; .:-m .ac ;a- w .; c.5. ,
. . , c. ,
. . : . /. = . .. ... . . s;. . c.
.+. -
. .,. . n :.' O--m., ,;.-- ..
. , := .
-;a- ' ?: : ? ' ~- .. -: W,::.
._m-, _~~ -s -.
.s dMW , . . . . d,.,d. n ~ .~ -W2ssA
.. . -a. f; ._ g4. n.m._CE-FED. ER,, AL.. REPO_RT. .ERS, INC. Aw O@#
720 08 05 94 1 DAVbw 1 this is a mass of material which remained in the test 2 section in percent. So we used Woods metal to represent 3 heavy, molten material matters coming out of the reactor
'4 vessel failure.
5 This number right here, which is 6 percent, tell's 6 us that 6 percent remained in the test section for that 7 case, which tells us 94 percent got dispersed.
8 When you finally get down to this, you see 97 9 percent remains in the test section. The reason why this 10 dispersed 94 percent and this retained 94 percent is
. .. 11 directly related to the configuration, which is directly 12 related to the configuration of.Seabrook.
~
13 So I will come back to this as we work through 14 it.
15 (Slide.)
16 The first experiment was one that represents a
'17 reactor cavity shown here in two dimens' ions. This is the
-18 end-on view. Woods metal supplied at about 300 psi was 19 discharged in a molten state in the cavity, followed by gas
'.,.-20 blowdown, which simulates the blowdown of the primary
,.w
- O?.M.21 system.. And you can Icok at this as being basically the
_'/gy ..*yk.s
'~
.. .p: ;. . . . , . .
~
l' 34
, l-99 22 same thing that was done in the Sandia experiments, where
' ,- n . 9.; - p. ., . .: .
~
l . , . - -ax:.q;O i23 .
they used iron thermite to represent the melt in experiments
., ., n . ,.
l ..
..;; that are approximately 50. kilograms. And they have a -
h
' .;7d .e@Q.~24; *c MiKEMWhis=S.T:M.w&K.d;W=-5 4% .
^'l-
- F?MFif.'jR25 representation of the Zion cavity which, for this
. . *,.: . . - m-it - : .v
-f
.. =:.: W w'y:;.
- , .. s. -
zQ % v-7.,3y.. ', _- ._
- ..;.g.u;;;ea
. ;, :; _ ,[ t;. %
.- . <- h, .,y ~~ ' , a,s z c ~:: x . . . ;.:. 9 m _., .z.
- 3.-. ' :
- .y. :.s;.- . .t.,; _ , , - .~ y - -'~.<;s :<
. M e v.c *w r w s .
b A'n' w s w. .2 L _, _ . . - m.... .m ._ ,. . _ _ .,m.
-, _ . . . . . . - - - - - _ _ . _ . _ . . . _ _ _ . . _ _ . . _ . . . _ _ ._ .. . _ ._-. i 1,.
9 20 08 06 ,
95
.1 'DAVbw 1 ,
fundamental issue --
2 Yes, Rich?
3 MR. BARRETT: Rich Barrett.
4 , ,
could you give me an idea of the scale?
5 MR. HENRY: "This is about'a 1 percent scale l: 6 slide, I to 2 percent of the linear scale of the actual t
l';
7 reactor system. So this cavity is roughly 6 inches long, as q- 8 the example. And we are going to keep the scale the same as 9 we walk through these. What that allows you to do, then, is 10 to see just how controlled the blowdown process is, 11 particularly, when you represent the regions on the side of
. 12 the vessel. We can think of it as something very similar to 13 the. Sandia experiments , which are like one-tenth scale.
'14 They use iron thermite and discharge that into the reactor 15 cavity with follow-on gas , CO2 and nitrogen.
16 What they find is all the material is swept out.
j, . 17 -
There is nothing in the reactor cavity that is going to give a . -
. 18 'the old material back.
q . -
l--. .; .: -
[
~
19 - ' "(Slide.)
.) . '
l3
. ; ~+ g , -
1, , ." .' 2 0 . c.,,,ib :.When we do this experiment, we find exactly the '
F-
] : y ?. : _.ycz? ?. **:?'- : - - -
1m + -
- ,.6.21 isame w , thing., .This.. : is the one that you saw. It says 6
. e.c. ;-...
.e - . .;e. .,. . .-ge n: , - . . .-
. u. -
!M "
7 22' percent stayed in. In this,.it says the test section. If
, :. . ; .. p ;. ,;; ,.:_ -
t-
. . e 63, 23'. .
..we,.,w..alk through this, the test section basically means the
..y..
- i. .
, . 97 - ,c ,:.
.. .~.c 24 ' whole picture'we've been showing you. , .
~ 5'
.. DIN d $h h E IfhdlM.M d O$5TM NRM 8 0-51stMn@ 3 0 $ s? i k lQ f 4f.yM25 a$$#/[.i$2fInf this. one ,~'it'.'. is ,just the reactor cavity.
So s a27 g&.kp.d: :. i v .* . :s
'e '* ::.:...:.
- .. .h? ga.~.=. :6 . :t;. :' %
';i. ,.n.9lN .,' ?5%x';f. . . ... a2. ~:'NHr;G ..q: :V ~#:: L':'; -On
. ~.. ::G
?. i.%;fr.?5.g w.M , . ;. .;..p.? 7 - .~.ct:..t.:..
- '-% . r.y :"i;T:c:'::ijW*:"
.7._ '-
.W
.. . c .
Jiggg=2jj.x M
E! f . N%CE-FEDERAL ~
REPORTERS, INC. -: - - . . - . . m#b
720 08 07 96 1 DA h I where to go? You go to 18 percent. This is sitting ikn a 2 big box. It will catch all the material. And 76 percent is i
3 on the vertical walls. The vertical wall that really caught 4 it was the vertical wall that was right outside here. And 5 18 percent was on the top cover in this box.
6 So if.that's all'you had, you would say, direct 7 containment heating would be an issue, but that, obviously, 8 is not all what you have.
9 The next thing you have is steam generator 10 representation in the testing sequence.
11 (Slide.) .
, 12 . And again, it's stris:tly done in two dimensions.
13 So you recognize this from what we just looked at. Now we 14 build this onto.it. This is the end-on view. Again, you 15 have Wods metal supplied at about 300 psi at 1.2 percent 16 scale. You get nothing that's held down here, as it comes 17 up now, and it goes through 'this vent, it looks like a 18 separator, but this area and this area and this area are 19 not greatly different. So the velocity you have coming out 20' - of here, out p.
of here, through here and through here,' is also
. 21 'not greatily different. So if that velocity is high enough
.; c ' .. _
9 : m . . . . .. .
~
J : .'22 to entrain and aerosolize out and sweep out, the material
.23 here is high enough to keep it airborne here. It will keep O . 24 . it' airborne there. It will keep it airborne there.
M ..: _[...: h 50.kS N .S~ N 1 C ?.N fs / 0 ^ O
~ ; 'T
. .':Of f*9.6 @ M m ~ -
- 5N
% 5'25 .,,_y . 4U.A So,when you look at the results, then, you find
. ;;;r 5:-:., .
,' .r ;
- b -w:. .-- * ; :g= ~::::n:ay .. .-
1,% '
i s y ..
..-..~.... .-....
w,.
c,.
.g . . . .
.a .:,:_ . :x. ..
y.. .
,.a. , . . . 7. '~,..
x -, .' 7 -f 7 .
- - - # ., y .
n- ,.
,.a c
-::u.
MM i%w
% eSnam 4 [5Q .$ dab-:naA A- ,CE-FEDERAL
.m_,_.-~..,_ REPORTERS, INC. -+.4 % y
~
720 08 08' 97 1 DAVbw 1 that more material is kept within the test section, and the 2 test section means the cavity, plus.the steam generator 3 compartment. Most of this is up now around the steam 4 generator, because it does separate out, when it starts 5 impinging upon some structures, but you still have about 40 6 percent, 50 percent, excuse me, that actually got out of the 7 test section.
8 So if that is all you had, you would say, direct 9 containment heating has potentia'l. But we still haven't 10 represented the actual configuration, because we forced this
. 11 to be 2-dimensional in character.
12 This one is basically the same geometry as this, U
13 except that this ' flow area --
14 (Slide.)
15 -- to the upper compartment is made smaller, so
, 16 we reduce these velocities in here somewhat, and we allow 17 more material, therefore, to be deposited. So even with the 18 fact that it is 2-dimens'ional, these two experiments show 19 you that structure makes a difference, but it still doesn't 20 look like an overwhelming difference. .
21 The most fundamental experiment to do is one that 22 says it is not a 2-dimensional system, it's 3-dimensional.
. 23 (Slide.)
(]) 24 Thisiswhatthe3pimensionalcharacterization.
~
25 looks like. Here is the reactor' cavity again to the same m :2 .L %@. ACE-FEDERAL REPORTERS, INC. ~
c.'~'
. 'r?:p.? % ' M?.M71700 Narinnwuie Coverase 300 336-6666
e >
'720 08 09 98 1 DAVbw 1 scale. You recognize the lip over here, which was the seal 2! tape before. Before we just had the steam generator here.
3l Now we've represented this in 3-D.
l 4' Here is the plan view of it. ,
5 (Slide.)
6' You are looking down. There are four steam i
7! i generators. The instrument tunnel is coming up through 8, here. This is a representation of the seal table region.
9 Again, this is for Zion. Seabrook actually even has a more 10 l restrictive seal table here, so the conclusions I am about i .
11 to come to will be even more robust for Seabrook. They were 12 ver.y robust for Zion. .
13 This represents the biological shield around the 3
14 ! reactor vessel and the simulation of the vessel with the 15 Woods metal is on this side here. These two regions 16 represent those open areas between the steam generators and 17 l Zion. They are wide open. You just put some screen in here 18 and represent the grating that is in there in the actual 19 ' case.
20 So new we did ' exactly the same test. The only 21 difference is, this is 3-dimensional. When the flow comes 22 out of here, it can go this way and sees a huge area of 23 flow, therefore, low velocities. If it sees local 24 structures, it can be deposited, if the velocities are too ,
i i 25 low to entrain it and to low to keep it airborne. !
1 I
ACE-FEDERAL REPORTERS, INC.
m ,,. m ~ ~ . , m,_
-. - . - . . ~ . - - - . - - - - - . . - - - - . . -
o o' I
i s i i
-*720 08 10 ! 99 1 DAVbw 1, (Slide.)
2 ,, So the debris pushed out of here comes this way.
l 3i It flows around. It can go up and out this way. It can i
4 also go out the simulated door, which is on the other side 5 of this machine. .
6i (Slide.)
7; And the results show us, when we make this I
8l 3-dimensional representation, over 97 percent is kept in the 9 test section. 2.3 percent was some fluid material that ran 10 i around and ran out the simulator door. . 2 percent actually 11 made it up and out as airborne material.
12 (Slide.)
13 So let'me review that for you. What that means 14 I is, we've got about 2 percent. When this came out, it 15 l essentially fell in a region like so. We did not see fine i
16 1 particulates uniformly dispersed all over, as you might 17 expect. If it had been discharged as a very find aerosol, 18 ; it might have accumulated as a puddle here. We found some 19 material which was reasonably small along these walls. In 20 fact, it would.never reentrain. We found material that 21 part of this actually flowed out of the simulated door over 22 here. .2 percent actually made it up and out of here. .2 23 percent has no relevance to containment response.
l 24 , So what these experiments show us is that if '
I 25 you just represent the reactor cavity and the seal table, ,
! I i
ACE _ FEDERAL REPORTERS, INC.
+ ** .
{ 3 f
l
'720:08_11 .
-100 i
1 DAVbw 1- . of. course, you will get things that look like they can be
-2.' dispersed, but if you want to do it in-2-dimensional-3, geometry, you will still see some dispersion. Although you 4 begin to see the influence of the geometry itself, what you 5 really need to represent is the 3-dimensic;nal flow path and l l
6 the structures, which is what was done in the smaller scale I -
Argonne experiments with uranium dioxide, Zr02, stainless 7l I
8; steel melt.
9- We see no. conflict between the Sandia results and 10 these and the Argonne experiments. We can put them all l .
11 i together, and it basically says it is not an issue for the 12 i Seabrook system. .
13 . MR. NOVAK: I am sure we.have some questions, but 1
14 3 I don't know how to lead them off.
15 l MR. HENRY: I should also say, Tom, excuse me, 16 I that it was our impression when we read the report and the i
17 i discussions we had in the past, that, again, there is no 18 I specific disagreement with our conclusions. Technically,
- 19 they came more from the standpoint of what else was being
. 20 done in things like 1150, but there were not specifically 21 any concerns.
i 22 l MR. PRATT: I think, again, the comment is i
23 ' true. What we were reflecting in the report was Zion's i 24 specific calculations that were done as part of NUREG .
l l 25 11.50. That was based upon the expert input to those i
l l ACE-FEDERAL REPORTERS, INC.
l
,, -.- p l
l *
'720 08 12 i -
101 1 DAVbw I assessments, to those that were done, I guess , about a year 2, ago.last summer. .
I 3i That's when the start-up took place. The 4 presentation that Bob gave here was given to the NRC Staff, ;
5 I guess, last fall.
6l MR. ROSSI: This was not done for the experts. 'I l
7" mean, the experts who did the 11.50 were unaware of this?
8 MR.-PRATT: They had quantified those estimates 9 before they saw this" presentation. So what happened is, as 10 part of the issue that was brought up by the experts, these i
11 j experiments were done, given to the NRC Staff to help 12 y resolve the particular issue. What it has resulted in is a i
13 ] redirection of experimental programs to take this effect*
14 into account.
15 f In particular, we had an experimental program not 16 in our division of Brookhaven looking at this phenomenon.
17 , 'These were very useful experiments that Bob did, and they
] 18 helped the NRC Staff redirect that effort to specifically
, 19 j these issues. Again, I don't think there is a technical 20 disagreement that this has the potential,to help resolve 21 this issue.
22 Whether or not all of the facts are in on all of 23 these experimental programs at this point is not certain, 24 but the quantification of the determinative entries in NUREG i
25 11.50, in my opinion, does not reflect these experiments.
ACE-FEDERAL REPORTERS, INC.
e s ; ,
'. \
- l
\
i l
'720 08 13 102 1 DAVbw 1- Someone can correct me, if that is not the case.
2, Another point on the event tree quantification, 3i the range of uncertainties that you get for the potential 4' for containment failure, the interesting point is, if you 5 remove direct heating as an issue, again, the ranges stay 6 large, which means that we have not just direct heating to 7k take head on. In terms of the results of NUREG 11.50, one 8' has to look at the assumptions made with regard to hydrogen i
9 ,! combustion, combining hydrogen combustion with steam spikes, 10 looking at pessimistic assumptions with regard to 11 containment performance and seeing how those tails of those 12 . distributions overlap, t
13 i i
14!
1 15 16 17 18 19 ,
-l 20 g ,
21 !
l I
22 23 24 25 ,
i l
ACE-FEDERAL REPOR1ERS, INC. l n,,-, s.......c. . . . . . , , , , < < ~
i l
'720 09 01 .l 103 l
1 DAV/bc 1 MR. HENRY: We haven't had the advantage that you 2; have to see exactly what has been done to 1150. The i
3l presentations tha't I have seen on some of these things, 4 they've asked a lot of questions about them. We don't know 5 how that's gotten factored back in.
6 I guess one other sidelight of this that we 7 probably should mention is I'm going to recommend a path on 8l this to the staff.
l 9, I made the same type of presentation last week to 10 the NRC special review committee on the uncertainties of I
11 , NUREG 0956.
12 ' I essentially said we'.ve gotten to the point-13 ' -
where we've got to fish or cut bait. It's either an issue 14 or it's not an issue. And the Sandia people have a system 15 that they've run they feel heightens their concern that this l .
16 ' could be an issue.
i 17 i ,
So my recommendation to them is to take this 18 facility and do this kind of experiment. This is a facility 19 , that can do maybe'100 kilogram scale experiments.
1 20 And I said let's stop all the speculation and 21 let's do an experiment on the biggest scale that we have, 22 and lets do the stuff that, from our viewpoint , from the 23 industry point of view, looks like it is more effective.
24 About all I got from the Sandia staff was comment .
25 that this should not be done because this scale is way too l
l ACE-FEDERAL REPORTERS, INC.
j l
. , . . . _ . _ . _ _ _ _ _ _ . . . . _ . _- _ . _ _ _ ._ . _. l i
l
- 720 09 02 104 ,
1 DAV/bc 1 small.
2 And I said that's crazy, that's the only scale i
3 we've got. It's fine. If the tests you have now say it's 4 an issue, let's find out if it is. I got the distinct 5 feeling that they weren't going to do this right off the l
6' bat, to find out.
i 7 ' If that is the case, I know of nothing that says 8' that this is an issue for any containment of the large dry I
9 type like Seabrook. In fact, if you piece everything 10 together, it basically says it's not an issue.
11 , If they're not going to do this experiment that 12 allows us to sort that out,, then my re' commendation is if 13 l that's the. way the research programs are going to get run, 14 I let's take what we have. Let's make the conclusions that 15 it's not an issue for Seabrook.
16 And if it ever shows up by whatever experiments 17 in the future, the NRC always has the flexibility of coming 18 back and raising that issue, as they do at any point in time 19 with any other issue.
1 20 I see no justification that says it is an issue. ,
21 MR. BAGCHI: I see that you and Trevor were 22 talking past each other, almost. From my standpoint, he was 23 talking about are tails meshing in so.the containment being 24 subjected to higher pressure? ;
I 25 MR. HENRY: Just say the last sentence. I didn't l ACE-FEDERAL REPORTERS, INC. j
m .-
- l
. 1 l i
720 09 03 ,
105 1
1 DAV/bc 1 quite understand it. I don't think we're talking past each 2, other.
i 3l MR. BAGCHI: Just the direct containment heating 4 by itself is not a challenge to the containment. I don't ,
5 think Trevor has a disagreement with that.
6; But, when you put together all of those things --
i the steam spike, the hydrogen burn, and some of the other 7 ll 8! situations that you come up with at higher pressures --
9! that's what you have to try and resolve.
10 And I was going to ask the utility to come back l
11 with some assessment of what they feel that ought to be. .
12 : MR. MOODY: That's in the PRA that was published 13 ., in Decem'ber 1983.
i 14 MR. BAGCHI: I realize that. However, all of 15 this work has been done. I would think that you would want to convince Trevor and his people that it is or it isn' t a 16 f 17 i problem, and try to give a perspective as to the magnitude 18 of the problem. ,
19 MR. MOODY: We included hydrogen burns, steam 20 spikes. F, red Torri is the one here who performed that 21 analysis.
22 MR. TORRI: If you cannot get.a substantial 23 fraction of the debris finely disbursed in the containment 24 atmosphere, then you're back at the pressure rise phenomena 25 that have always been considered -- steam blowdown versus ,
! ACE-FEDERAL REPORTERS, INC.
i
_ . . , ~ . . .. . - - - . . ~ - - - . . - -~ -
~*720 09 04 l 106 ,
1 DAV/bc 1l spike, hydrogen burn, and so on.
2l It's only when you combine that together with a 3 large fraction -- as Bob says, 50 percent or more --
i 4 disbursed in the containment, that you could then predict 5 positive factors that lead to high pressures. l 6l What Bob is saying is you can't get that material I
7I up in the atmosphere.
I 8 MR. BAGCHI: We heard that two or three months 9, ago. Bob made a presentation just in this very room, as I
\
10 ; recall -- 3D tests and so on. But that didn't convince the i
11 ; NRC staff.
~
j ~
. 12 MR. HENRY: I'm not trying to talk past anybody.
}
13 J I think we want to' break this up into the honest to God 14 issues and separate them out, what has to be done to resolve 15 j steam spikes, hydrogen burn, direct containment heating.
16 I think, as Jim said, we'have gone through it in 17 l the past,. And Seabrook has a very strong containment.
18 Steam spike is not a question. Hydrogen burn really isn't 19 either.
20 ' And we'd be happy to make a separate 21 presentation. But I think the key thing is this is one of 22 the ones that if I were to go out and look at what the NRC 23 and the NRC consultants had put into the open literature, 24 this is the one that people would pick up and would i
25 l superimpose that upon Seabrook. l 4 . ,l i i ACE-FEDERAL REPORTERS, INC.
. ,_ . . _ _ -~ ;.. _ _ _ . . ._ . . . _ . _ . _ _ - _ . _
- e. j
'720 09 05 , 107 i . ;
1 DAV/bc 1 We think that's a big mistake. We think that one !
2 can be resolved right here and now. If we can' t, we'd like 3 to identify a path that says: What do we have to do to do 4 it? Let's get going. ; )
5 I tried recommending that on the generic scale to 6 Sandia. And I really didn't get the kind of reception, I l 7 3 think, that they really should have brought it. l j
8 j MR. PRATT: Could I ask one question? And I'm
,9l probably showing my ignorance here, and I should know. l l
What is the situation with Ginsberg? Ginsberg is 10 j
. 11 the principal investigator at Brookhaven. If he was
.2 i , supposed to redirect this whole program to'look at these 13 geometries, now this is with similar experiments, I assume.
14 ! he's, continuing.
I 15 MR. HENRY: He's continuing. I didn't have a L
16 ' chance to see his presentation at the time , so I don' t know 17 i where we stand on doing something like this -- whether he's 18 . doing a verification of ic or working at it, or whatever.
19 MR. PRATT: But your point was, at Sandia, the 20 people wanted larger scale experiments. And in order to 21 address the scale question, they should look at this.
22 MR. HENRY: One could create an entire test i
l 23 , matrix that should be done. But, because of all the other 24 work that has been done and has been shown to be influential l.
25 , in this task, they ought to address that right up front. ;
i
! i i
ACE-FEDERAL REPORTERS, _____
LNC.
1
o .
l ,
! I d720 09 06
. l 108 ;
1 DAV/bc 1 They ought to address it right up front so that 2' we have a feel for it and the Commissio<. has a feel for it.
i 3l MR. PRATT: ,
The additional point that I was i
4 making, and I tried to make that in the report, perhaps 5 again, editorially, we should bring the assumptions to the 6j front of the summary.
i But the conditional probability that you came up 7l 8 with as far as Zion-specific calculations, in the text, we 9 did mention that there was a difference between the yield 10 capacity of Zion versus Seabrook.
11 So,one would expect a benefit there.. The point 12 that I was making originally was that because of the way 13 you generated the containment event trees in the NUREG 1150 14 I work, and Fred Torri was on the review panel on the event 15 trees along with myself, what one did was to establish a 16 fairly wide range of possible failure distributions for the i
17 containment.
18 For example, one would pick the mean of the yield
- 19 and calculate the five different distributions and have 20 people vote on where they thought the actual failure might 21 be.
22 What it tended to do was spread out over a wide 23 range. And perhaps , as you might expect , that affected the 1
24 , failure distributions.
25 In addition, one would then also couple in fairly
. l ACE-FEDERAL REPORTERS, INC.
--,., , . ~ ... _ , , . . . . .
l 1 l l l
- 720 09 07 !
109 l l
-1 DAV/bc 1 conservative assumptions with regard to coupling together 2; steam spikes, hydrogen burns, and so en, and have people l
3j vote on them.
4 And what we were finding when we looked at some i
5; of the results was that you could re" move direct heating as !
i 6 an issue, just take it right out of the containment event 7! trees.
8 In some of these other reactors which had lower 9l pressure capabilities, you still find significant i
10 l sensitivities in relation to that type of -- well, l
11 - conventionally, it's a lousy choice of words. But the steam I
12 spike, hydrogen burn mode of failure was still sensitive.
13 .l. That was the point I was trying to make earlier.
l Whether that's true at Seabrook with the higher pressure, we 14 l 15 l haven't determined at this point.
i 16 MR. BAGCHI: But that's a very important issue, 17 and that's what I was trying to bring out.
18 MR. HENRY: I think that's also something that 19 we're going to address as soon as I'm done. The only thing 20 I wanted to leave with you -- I was not trying to talk past 21 anyone -- I think these other issues can be addressed.
22 MR. BAGCHI: I know you're not. I wanted to make 23 sure that you do have an understanding of where we're coming 24 ' from. '
25 MR. HENRY: The one thing I did want to leave is 3 I l l
l ACE-FEDEI6\L REPORTERS, INC.
m-i ,
'720 09 408 110 l l
'I ~DAV/bc 11 that I' don't.believe in this. But what I heard-was that the 2; people at Sandia, if-they did this, they wouldn't trust this-l 3 experiment. I think that's just absolutely crazy.
i 4 MR. PRATT: They wouldn't trust your experiment.
5 MR. HENRY: No, they wouldn't trust their own i
6 experiment having a one-tenth scale. If there's a real 7 problem, it will demonstrate itself.
I 8l MR. PRATT: Does that imply then that they 9 wouldn't trust yours with the smaller gemometries?
10 (Laughter.)
11 l 1 MR. NOVAK: Before you leave, are these l
12 ! discussions written down in a. summary? Is this difference i
13_ ,
of opinion spelled out in any document?
i 14 ! MR. HENRY: This difference of opinion on what 15 experiments to do? Certainly, m'y recommendation of what 16 , experiments they do is written down. I don't know how the 17 l committee is going to keep their own records, outside of ,
18 l Herb Couts is chairman of the overall committee.
i 19 ' MR. NOVAK: I assume you've read the draft of 20 NUREG 115,0.
21 MR. HENRY: I have seen the work that's going to 22 be used as source terms.
23 MR. NOVAK: So, in fact, it's not a public 24 document. I wasn't sure.
l 25 MR. BAGCHI: No, it's not.
1 :
~
ACE-FEDERAL REPORTERS, INC._ ._ ---
o .
I I
f720 09 09 i 111 l
1 DAV/bc l' ' MR. HENRY: We've had presentations at various 2 meetings like this -- the Mach I owners , the BWR owners 3 group, IDCOR -- as it related to things. And it related to 4 questions on how it looks like things are going to be done. ,
i 5 MR. NOVAK: One last question. Is there any i
6i question of the repeatability of your experiment? That is, 7 were these single experiments or something? You just did 8 serially one at a time over any of those actually repeated?
9 Just see if there is repeatability.
I 10 l MR. HENRY: These two are close enough that you 11 ' can even look at those. It's the.same geometry, it's just I
12 ' reducing .the area. They're sufficiently close to tell you 13 , that they are repeatable. We have'done these. We didn't 14 I get a chance to do this one a second time. We could indeed 15 ; do that if that's an issue.
i 16 We did this one twice by itself. We did this one 17 ! in addition to that. We also did this on a larger scale, 18 twice the scale. It's still not a very big scale from the 1 i
19 percent to 2 percent models.
20 The key thing here, Tom, is that this progression 21 influences the geometry. This is exactly what you see from 22 the Argonne experiments, which were done with realistic 23 materials on a small scale. ,
t i i 24 Argonne also did woods / metal expe~riments like 25 i this and found the same thing. This progression shows how l
l . ,
! ACE-FEDERAL REPORTERS, INC.
- 720 09 10 112 1 DAV/bc 11 the model can be built, and the logic of the overall 2+ process.
I 3' MR. NOVAK: Is there an incremental pressure that 4 one could add, say, for every 10 percent of the core that 5 you assume is -- what are the normal pressure levels that 6l !
one would calculate?
7! Supposing it's not 10 percent of the core, it's 8! roughly half the core? Is there any rough rule of thumb?
l 9' MR. HENRY: Just roughly, 10 percent is going to l
10 l be worth something like one to one and a half atmospheres.
11 MR. NOVAK: Do we disagree on that? Is that just 12 simply a . thermodynamic calculation that we' re making?
13I MR. HENRY: It's a chemical reaction. It's an 14 l equilibrium calculation. You just bring everything to 15 : equilibrium - thermal, mechanical, chemical.
16 ] MR. NOVAK: So your answer is perhaps 10 percent i
17 ' on the order of 1.5 atmospheres would be the pressure?
18 MR. ROSSI: I think you're also saying that there 19 would be wide agreement on that calculation.
20 MR. HENRY: It's an equilibrium calculation.
21 !
Tom, that's not how much came out of the vessel, 22 that's how much came out of the vessel and directly 23 ! exchanged its heat with containment atmosphere.
i i l !
24 ' MR. PRATTI: That's why, in the days of Indian Point, when people tested then and we didn't know about 25 l ACE _ FEDERAL REPORTERS, INC.
l l -
'*720 09 11 i 113 l
1 DAV/bc 1' direct heating, these thermodynamic calculations, we didn't 2; have any problem at all.
i 3 When you're faced with these type of thing's, the 4 bounding thermodynamic calculations aren' t acceptable , so t
5 you've got to put some physical reasoning into it. And 6 therein lies the problem.
7! In terms of availability of NUREG 1150, NUREG i
8; 1150 is a very small document. It's standing on top cf 9, many, many feet of contractor reports.
I 10 It's my understanding that several of those l
11 l contractor reports were available. In addition to the ones 12 that were published when,the source terms came out, we sent 13 ) in parts.of those NUREGs as drafts to the NRC staff.
14 I So I know NUREG 1150 itself is not available.
15 ! But I thought that some of the contractor reports that went 16 l , into it were. I' don't know. That was my understanding.
i 17 l ,
MR. MOODY: Okay. Fred Torri.
18 i MR. TORRRI: I 'm Fred Torri . Notwithstanding the 19 arguments that say that this phenomenon is of the magnitude 20 that it can create a problem in a large, dry PWR 21 - generically, like Bob just went through, we did take some 1
22 time to ask ourselves: ' '
23 ' What does the Brookhaven report assessment of i
24 , direct heating and containment integrity or containment l
25 l fa~ilure pressure capacity, what does that combined 1
ACE-FEDERAL REPORTERS, INC. __
o .
I I i
~
- '720 09 12 l 114 l DAV/bc 1 assessment of Brookhaven mean in terms of containment 2 failure for Seabrook?
I 3l It appears that we've had a misunderstanding that 4l
.I'm glad is being cleared up. We had made an earlier j 1
5 submittal to you, that Brookhaven reference that I 6 discussed, that adopted the conditional probability of 7l containment failure due to direct heating as .01, which was 8! orally discussed in the previous meeting we've had.
9; And I believe it was this room. We. interpreted 10 , that as meaning that the pressure transient estimate for Zion due to direct heating was used together with the SSPSA, 11 f j . .
12 i Seabrook-specific containment failure pressure distribution, 13 which is in the report, the one that distinguished between 14 ! type B and type T failures, to arrive at the .01.
15 i So that the .01 was a Seabrook-specific i
16 ! containment capacity combined with the generic large dry PWR 17 Zion-pressurize estimate for direct heating.
And our submittal to you earlier in writing as an 18 f 19 l answer to your question meant that it was built en that 20 assumption. Reading the report, I've confirmed it with 21 Trevor. We understand that the total .01 number is a Zion-22 specific number.
23 In other words, it's Zion-specific estimates of 24 l the pr. essure resulting from direct heating and other i
25 i phenomena compared to design containment pressure, not the l
ACE-FEDERAL REPORTERS, INC. -
J
- 720 09 13 115 1 DAV/bc 1 Seabrook containment pressure.
2' 3
4 ,
i 5 !
6 7i 8 l; i
9i 10 1
,11 j. .
\-
12 ! -
j .
13 i 14 !;
15 !
16 l 17 l 18 i
19 20 21 22 23 24 '
25 ,
I ACE-FEDERAL REPORTERS, INC.
~ _ _ _ .
i I
- 720 10 01 i 116 ;
1 DAV/bc 1 (Slide.)
2 So what we felt, we felt it was necessary to make l
3i this assessment in the context of the Seabrook containment 4 capacity because that was, by and large, the largest I
5 difference between Seabrook and Zion. i 6 I think we make much the same assumptions as to 7 how much debris is involved in the direct heating
-l 8: phenomena. For Zion containment, the volumes are very 9l similar.
I 10 , And if one makes the postulate that the structures don't have an impact, which is the whole argument 11 f i 12 j . Bob was.trying to convince you -- that .they d'o make a 13 significant impact -- if you don't account for that, then 14 'I you get the same answers on containment in terms of how much i
15 l pressure increase you get.
I 16 l What is different is the pressure at which these i
17 l two containments fail, or at least what the analysis says 18 that the pressures are when they fail.
19 In order to be most constructive and helpful to 20 you, we felt that we wanted to do this reassessment in the -
f 21 context of your assessment of the Seabrook containment 22 capacity, not ours.
23 j We did that because that brings together all of i
24 ' the ends that your. analysis of the Brookhaven report had 25 into something that we feel could have been do_ne by you if ;
l ACE-FEDERAL REPORTERS, INC.
2 .
I
'720 10 02 117 !
1 DAV/bc 1 you had taken that extra step, or by Brookhaven.
2l Now, to do this right, one would take the 3 Brookhaven information on containment failure modes, if 4 there are' pressure levels for failure, and the leak areas f 5 resulting from these. failures, and one would generate a leak 6l I area similar to the one that we had in the SSPSA, which 7 gives you the probability of failure as opposed to the 8l ;
pressure for different failure types -- the large leak type 9 and the medium leak type.
1 0 lI We didn't have time to do that. We did an i
11 l approximate estimate of what the Brookhaven implied 12 distribution was . And we made this estimate based on the 13 ' following information:
14 The Brookhaven report or containment capacity i
15 ! seemed to agree with us that sort of a lower bound for 16 containment failure was indicated by the yield capacity of 17 1, the Seabrook containment.
i 18 l In fact, Brookhaven's estimate of ideal capacity i
- 19 is exactly the same as SMA's estimate in the SSPSA. It's 20 157 psig.,
21 Brookhaven estimated that for the two different 22 fai' lure modes that were closest, that were the lowest 23 pressure failure modes, that the hoop failure mode ,
- i. 24 represented was characterized according to BNL with a medium 25 l failure pressure of 175 psig.
~
l .
\
, l ACE-FEDERAL REPORTERS, LNC.
l I
'720 10 03 l 118 2 DAV/bc 1 That corresponds to the 1 percent strain 2; assumptions. That's saying that we feel there's a 50 .
31 percent probability if the containment fails under a 1 4 percent strain condition. And there's an upper bound of 216 5 psi, which was our 50 percent, our medium assessment, what 6i we could get if we went to failure strains with best 7l estimate material properties.
We don't agree that the 175 represents a 50 8f 9 i percent probability of containment failure. But, for this i
10 l purpose, we have accepted that because we wanted to use your 11 l numbers.
l 12 : The other failure mode,that was indicated by 13 , Brookhaven is the penetration X-8, which is a high energy i
14 ' feedwater penetration.
15 l Brookhaven's estimate is that the medium pressure 1
16 l capacity of this failure mode is 167 psig. They acknowledge 17 , that would be a Ty'pe B failure, not a large leak area but an 18 l intermediate leak area.
19 The only other statement that needed to be 20 interpreted from the Brookhaven information was that the 21 ; basement wall sheer failure mode, while the analysis by 22 Brookhaven of the finite element analysis indicated that 23 , that failure mode would not occur at a pressu're lower than 24 150 psig -- in other words, the lower bound -- it could 25 ' occur at some pressure higher than that.- .
! i ACE-FEDERAL REPORTERS, LNC.
t
.. ?
'720fl0.04 - -
119 l
+
' l .DAV/bc- 1i 'It's our assessment that the uncertainties, if 2 -l you accept this 157 psig as a lower bound for the sheer 3 failure mode,othat the uncertainties of that failure mode 4 are relatively larger than th'e uncertainties of these two f
! 5 failure modes. !
6 And we felt if one accepted a lower bound for t-7 that failure mode of 157, that the medium pressure for the 8 . sheer failure mode would be higher than 167 and, therefore, 9l would not control the median pressure at which the I .
10 containment would fail.
11 . Based on this line of argument, we have then
- 12. ' constructed a distribution for containment. failure which we 13- feel is based on-the following:
1 14 We have accepted 167. For the purpose of this 15 analysis,167 psig as the 50 percentile median failure 16 pressure for the containment. We have estimated the
$ 17 uncertainty associated with that failure pressure from the 4
18 . information given here, which gives us the widest 19 uncertainty; namely, the median of 175 a'nd the lower bound 20 ' which we interpreted as a 5 percent, probability of 157.
j 21 Note that the two are inconsistent. If we use i 22 this as the median, we should use the difference between 157 i
23 , and 167 as the indication of uncertainty. ,
l 24 l' We have taken the larger one to make sure that 25 the tails, the lower pressure tails resulted from the hoop
__ __ __CE-_[EDERAL REPORTERS tINC. _ _ _ _ _ _ _ _
. . . .m
'*720 10 05 - - 120 l
1 DAV/bc 1h distribution and from the sheer distribution and were l
2j covered by the wider uncertainty which we obtained, if we 1
~
3j take this as an indication of the uncertainty.
4 As a result, the distribution we obtained this
-5 way we compared with the Zion containment.
6 (Slide.)
7 And the result of this comparison is shown here.
8l The curve-shown here is design containment failure 9' probability distribution. And this curve is the 10 ; distribution we obtained just the way I described: A median 11 failure pressure of 167 psig, this is in psia. You have to 12 i add 15 percent and the median of ,182 here, and a standard 13 devia' tion of .066, which is derived from 175 psig median and 14 157 psig 50th percentile.
15 j First, let's look at what happens when we take 16 the .01 conditional probability of containment probability 17 derived for Seabrook. And we go with the design curve.
And that tells us that the corresponding pressure 18 l 19 ' is l'46 psia, including one psig.
20 ,
At that pressure -- now that must be a 21 representative pressure of what was calculated for Zion in 22 the Zion Containment, in order to get .01 containment 23 ; failure probability -- taking that same pressure, because '
24 'I the volumes are the same, for Seabrook, we find that the 25 l corresponding containment failure probability for i
ACE-FEDERAL REPORTERS, INC.
_ _ t
i -
~
- 720 10 06 121 1 .DAV/bc 1! Seabrook using the BNL assessment of the containment 2l capac'ity at Seabrook 13 1.5 times 10 to the minus 4.
I 3! This result, I believe, is conservative for two 4 reasons. One, I explained how we constructed this l 5 distribution with a wider uncertainty than what a strict 6 interpretation.of the data would give us, by taking 7 the information in the report that has the widest 8 uncertainty information and combining it with the lowest 9l median failure pressure.
10 But, in addition, if we really believe that the 11 yield is a lower bound .for the f ailure pressure, then-one 12 ; could argue that, in fact, at the yield pressur.e, the -
13 containment failure probabilities will drop off sharply.
. 14 !I It's like a cut-off. There's a discontinuity..
15 l And if we do that, we find that yield at 172 psia, that this 16 ' curve would actually be cut off vertically right here and ,
17 you'd never get even close to this pressure. ;
I 18 ' For these two reasons, we think that even taking 19 directly the Brookhaven information with the interpretatin 20 1 I described, we find that the early containment failure 21 probabilities, which were used in the SSPSA, which are still 22 embodied in all our maps in the WASH-1400 sensitivity i
23 ' studies, 1 times 10 to the minus 4 is very consistent with l l 24 l what we derive for that direct heating failure probability, *
~
25 l Using data entirely out of the Brookhaven review 1
ACE-FEDERAL REPORTERS, INC.
. __v . . _ _ _ _ . _ _ _
l l
'720 10 07 122 ,
i !
1 DAV/bc 1' report, we get either 1.5 minus 4 if there's no cut-off, or ;
2: if we use a yield cut-off for the failure pressure for this I
3' containment, we actually get zero.
4 There is zero probability based on this ,
t 5 interpretation that the containment failure is due to direct 6j heating if we accept yield as a cut-off.
7i This is I think certainly a reasonable argument.
8 There should be discontinuity in the containment failure 9 ,
distribution at yield. But, I think the strength of our 10 ! argument is that, even without a cut-off, we are very 1 -
. 11 compatible with deriving a containment failure probability
-12 i that is very compatible with.what has actually been used for 13 early containment failure probabilities in the SSPSA and all 14 l the submittals we've made to you.
15 l This analysis will be or is documented as part of i
16 l this submittal, which will be made available to you. We I
17 ; hope that the final report that Brookhaven may produce will 18 take advantage of that information.
19 MR. NOVAK: Do you have any comments?
20 MR. HOFMAYER: Charlie Hofmayer, from 21 l Brookhaven. Two questions.
22 On X-8, yu chose that, as you said, a Type B 23 failure?
i :
24 MR. TORRI: No. We didn' t make a distinction 25 l here between Type A, Type B and Type C. We made strictly I
~
ACE-FEDERAL REPORTERS, INC.
- 1, -
I
'720.10 08. 123 i l
'l . DAV/bc 1~ a probabalistic argument.
2 What you said is, if I recall correctly, that ,
3' X-8, the best estimate failure mode, was a Type B. But'the i
4 upper bound failure mode was a Type C. We did not make that i 5 distinction. In fact, we have not made any assumptions >
-6l about whether these are Type B or C. We assume they're all i
7' Type C.
81 MR. HOFMAYER: What about the X-267 l
9l MR. TORRI: X-26, your conclusion was that .the 10 best estimate failure mode was Type A, that the upper bound 11 failure mode was Type B, not Type C. A Type B failure mode
, 12 ; would not be represented by the release types we're 13 , discussing here. That would go into S-6 release category.
14 I Therefore, any contribution from that failure 15 l mode would not contribute to the frequency of S-1, which is 16 the one that's at issue here.
17 So we are not accounting for that penetration in 18 I this analysis because it would go into a different release l
l 19 category.
20 MR. ROSSI: That doesn' t buy you a whole heck of 21 a lot in terms of dose versus distance.
I 22 MR. MOODY: It's significantly different between 4 :
S-6 and S-1. But I think the main point here is the j 23 l ;
- 24 frequency. He has not addressed the other conservatism, i
25 which I hope is in our write-up. All of this is going to i
ACE-FEDERAL REPORTERS, INC. _ -_
l
'720 10 09 l 124 1 DAV/bc 1' ,
S-1. What he's done, as well as what Brookhaven's done, 2l which is another conservatism.
3 I think we've got 16 conservatisms here stacked ,
4 up.
i 5 MR. LONG: The reason I brought that up was I got 6 the impression, Charlie's question was going to the question 7f of being controlled by the probablity of X-26. In other 8 words, that penetration would be a potential Type D failure.
And the probability of that might be more 9l ,
10 : important to the frequency than your gross failure.
l 11 l MR. TORRI: We have not looked at that. One of
. 12 the reasons I cite, in order to do it correctly, one should 13 really regenerate those Type B and Type C probability curves.
14 with your data. For that penetration, that becomes 15 l important, you know, to be the Type B, not the Type C.
16 What we want to do 'here is address the frequency 17 of S-1, which was where all of the direct heating f ailures 18 were being assigned to it.
19 And I think, for that, this assessmen't here is
- 20 correct. What you're raising as a question is:
21 If you distinguish between Type B and Type C, is 22 there a contribution from the upper bound failure mode of 23 , X-26 that would show in release category S-6 that might be 24 - an important contribution.
25 l We have no answer to that. We have only looked l
l 4
ACE-FEDERAL REPORTERS, INC.
- 720 10 10 j 125 1 DAV/bc 1' a t S -1.
2i MR. FLEMING: There's an additional 3 conservatism. You kind of alluded to it. But I think 4 Brookhaven agreed in another area that our conservative ,
5 assumption in the piping fragility area, that it was 4 6l conservative to assign a 1 percent chance of failure at 7 yield stresses.
8 This analysis, we're essentially saying a 5 9' percent chance of failure at yield stresses. I think that, 10 in itself, is another sort of latent conservatism. There's 11 no evidence that materials fail.at yield pressures.
,12 l There may be,some uncertainty about what the 13 correct. yield value is. Maybe that's a better 14 i interpretation of what the 5 percent means, but there is no 15 i evidence that materials fail at yield. That's a 16 conservative assumptions.
i 17 l MR. TORRI: I think we'd certainly be prepared to 18 ! address X-26. At this point, I really concentrated on S-1 19 because thats the way direct heating came into the 20 picture. And we asked ourselves what is our assessment of 21 that as to X-1.
22 MR. MOODY: There isn't that much difference 23 between the two questions. j 24 , MR. TORRI: I'd have to look at what it is.
25 l MR. NOVAK: Before you leave, is it fair to say i
i ACE-FEDERAL REPORTERS, INC.
_,.1,m m__ _,_.
c .-
l -
1 l
l
'720 10 11 j 126 l l
1 CAV/bc 1 that what has been suggested as the Brookhaven understanding 2j that they have fairly characterized our statement or your 3i statements in the report?
i 4 First of all, let me just make -- in other words, ,
5 you made some statements regarding what your understanding 6l is of the Brookhaven analysis that you put forward. And you 7 used certain values that they have calculated.
i 8j Let me just ask first, were they correctly used I
9l or reasonably close to being used? ,
MR. HOFMAYER: Yes. I think we have to 10 l 11 understand a little bit better how he actually created those 12 i curves. Okay, particularly how you treated the sheer i
13 value. I think I'd like to understand that better and 14 reserve judgment on that.
I 15 l MR. NOVAK: For the purposes of my question, 16 i assuming that you're in agreement then, could you, without 17 ! going through the whole presentation, could you kind of 18 summarize for me .what you think you captured by going back 19 ' and doing that?
20 j I just didn' t appreciate the comparison of the 21 ! design. I think it would be worthwhile to say it now, i
22' without using any real slides. But, just sort of sum it 23 j up. .
I l
24 MR. PRATT: Just before you answer that, could l .
25 you put up the distribution curve for me, please?
~
8CEfEDERAL REPORTERS, INC.
s .*
l l l
l
- 720 10 12 127 1 DAV/bc 1 j MR. TORRI: The real meaning of this is, if you 2l substitute this value, which is now a Seabrook-specific I
3 value, so to speak, at least Seabrook-specific with respect i
4 to containment failure pressure, in face of this value which j 5 Brookhaven has used in its repo'rt, you find, no matter what i
6I you assume about direct heating, it does not impact, does 7 not have any impact.
I 8i It's a contribution due to direct heating which
^ 9! Brookhaven h'as shown as appearing in the lowest decade of l
10 l its curves. It disappears entirely. It's shifted down.
. 11 .
1 12 i 13 l 14 15 i 16 17 i
18 i
19 20 i 21 22 ,
23 l
24 i
. 1 25 j l
l ACE-FEDERAL REPORTERSdNC.
1
.. .- .. .~. . - - _ - . ..-. - ~ - - - - . . - -
'720 11 01 128 ,
- 2. DAVbw 1; MR. ROSSI: Let me try to say it and see if I say 2l it right.
3 What you do is, you take the direct heating 4 pressure, as calculated in the design, as Brookhaven I
5 calculated it in the design, the direct heating pres'sure 6 spike that Brookhaven calculated in the design. You assume 7 that pressure spike is the same for Seabrook. Now what you 8 did is, you took the Brookhaven calculations on the 9l containment for things like yield and that kind of thing.
l You took their calculations, the Brookhaven one, and you 10 l 11 . developed the failure probability curve for Seabrook 12 I containment now from the Brookhaven numbers, and you l
13 I combined the. Zion pressure spike with that failure 14 probability curve for Seabrook, and when you do that, you 15 j show direct heating is unimportant.
I 16 j Is that what you did?
17 MR. TORRI: In principal, yes, but I think I have ,
i 18 ' to qualify that. We don't know what the Zion calculated 19 ' pressure spikes are. All we know is that whatever it was, 20 it resulted in a conditional probability of containment ,
21 failure of .01. I 22 MR. ROSSI: So you didn't actually take the 23 pressure spike, you inferred it? j i
24 l MR. TORRI: We essentially inferred what that ~
25 l pressure was from looking at where the failure probability 1
\ .
l ACE-FEDERAL REPORTERS, INC. .
i
'720 11 02 ! 129 I
1 DAVbw 1i was, at what point the design pressure capability reaches 2 .01.
3 MR. NOVAK: Isn't that basically saying, if 4 you're dealing a small number, it doesn't matter if the 5 ' pressure spike through direct heating that has been f 6l calculated by Brookhaven was a small number, say, under 15 7 psi, it probably doesn't matter. So until you know the 8i magnitude of the number.
t 9; MR. TORRI: We inferred that number was something l
10 on the order of 14 6 psia, 131 PSIG.
11 MR. PRATT: Where.did you get the distribution 12 design?
13 3 MR. TORRI: Out of'the Zion PRA.
14 ! MR. PRATT: That's not what we used from NUREG 15 l 11.50.
l .
16 l MR. TORRI: What did you use?
17 MR. PRATT: Okay. We're going through the same.
process, so we can compare notes, because one of the things 18 l t
19 we did want to do in the final report, of course, is not
. 20 just to leave an ambiguous statement that we did this for g; 21 Zion and it is stronger or it might be less. So we are ;
I 22 trying to flesh all of this out. We could perhaps trade 23 some of that information back to you. You reviewed the l ,
- l. 24 : containment of entries and the quantification and the way l
25 they did things, and a lot of the problems with the
~
ACE _ FEDERAL REPORTERS, INC.
m.....m.~..... _ . , , ~
. . _ . ~ _ . _ . _. . _ _ _ _ .
i :
'*720 11 03 130 :
1 DAVbw 1; methodology is, it is very inscrutable. It is very 2l difficult to extract that information. We have a guy who is 3 doing that for us. It is going to be a major problem, I 4 think, with the document, but what they do there, if you 5 recall, when you reviewed Zion is, they looked at the mean 6 value, which is 134 PSIG, and they built the distributions 7; around that. So it follows very much the same pattern that l
, 8l was done for Surry, where they took the mean whatever it is 9i for Surry, 100 and whatever, then added up the 10 distributions.
. 11 So i.t is a compass distribution with weights 12' ! attached to it.
i
'13 I MR. TORRI: You did the same thing for Zion.
14 MR. PRATT: We did the.same thing for Zion, just 15 extrapolated it from the mean that was Surry-specific, and 16 we made it Zion-specific, based on the 134.
17 i . So we may not get exactly this number.
I'll have i
,, 18 ; to go back and check again.
19 ' MR. ROSSI: Nonetheless, you are dcing a similar 20 kind of thing, and that would be reflected on your final 21
- report. ;
1 !
22 ' MR. PRATT: Absolutely. I would actually prefer i
23 to go further than that, as well. I am going through the 24 l process for my own benefit of trying to understand some of 25 ' the rationale that's in your 11.50, some of these pressure ACE-FEDERAL REPORTERS, INC.
i
...,,_u. . s- . -- - . - - - . - - . - - -
. i 720 11 04 ,
131 j 7
1 'DAVbw 1! spikes. So that we're trying to look. What you are doing 2 is, you're getting to the end of maybe, there's 18 questions 3 in the containment of entry in coming up with the number for 4 which there have been multipie. additions and components.
5 And what we are trying to do is fish out of that information l 6 something that makes some physical sense to me personally.
7 If I am going to have to say anything meaningful 8lj about these results, I have to know where you're coming from i
9 and what the physical rationale behind-it is. So what we 10 have to do is to close that gap.
11 That is why it is ambiguous in the report. You .
did not have that picture available to you in the report at 12 l 13 ! 'the time, but I think it is an essential element that should 14 !, .be in there, and what we will be attempting to do is 15 ! precisely what Fred is doing to get what the median pressure
. I 16 ! load is. l l !
17 ! ,
. MR. TORRI: Something like this? ;
i 18 MR. PRATT: I meant actually to calculate what i
19 the median pressure load is from all this stuff, understand i 20 , what it's coming from, from a physical point of view, so ;
I . ;
21 l that we understand exactly the assumptions that have gone j i
i 22 into that, and then go into a distribution of this nature.
23 ! If Charlie likes that distribution, we will use l
.' 24 ' that one and come out with what the conditional probability 25 is that's Seabrook-specific.
ACE-FEDERAL REPORTERS, INC.
- c. -. --- - - - -
. :. g .. . - . - - -
,. :. u
-+
i
. r
. *720.11-05 132 1 ~ DANbw 1 MR. MOODY: This is all conditional on the fact 2 that someone observes that this can actually happen.
3 (baughter.)
,4 ,
MR. TORRI: There is, actually. I think there is .
I 5 merit, though, to thinking a little bit about whether, you 6 know, should it represent some sort of discontinuity on this 7 curve, because it is a physical condition,'where you are 8 much more confident that nothing is going to happen, and 9 from there on at yield, somehow, to say it's across the 10 yield pressure. I think that curve really should go much .
I 11 more like th',is. .
12 l MR. HOFMAYER: . Fred, even in your original.
l 13 study, you didn't propose that.
14 l MR. TORRIr We had a cut-off based on yield. The l
15 ! yield pressure, which we did the cut-off, the yield pressure l
16 I probabilistically. We looked at the distribution, the I
17 probability distribution of the yield. We cut it off
- probabilistically, according to the distribution.
18 l I i i 19 MR. BAGCHI: It is entirely possible that you 20 might be able to convince a lot of people that instead of l .
21 ! general yields, you could cut it off at some other point :
l l 22 i first or something like that. You could come up with some 23 of this, and it would undoubtedly be convincing to a lot of
.x - 24 people.
25 MR. TORRI: We did not cut off at a single ACE-FEDERAL REPORTERS, INC.
~
l o .
i
~*720 11 06 '
133 l1 DAVbw 1 i
pressure. We progressively cut off as we used more of the 2i yield distribution.
1 i l 3l MR. PRATT: What I am trying to eyeball here is a i
4 comparison with the original curve that you presented, where l i
5 the dominant contributor to the total failure, when it was i
6I wet, was really coming from the B. It looks really similar.
7 MR. TORRI: This might look somewhat like our B 81 distribution.
MR. PRATT I am just trying to look at it.
9l i It 10 looks very similar.
. 11 i -
Let me ask you another question, more generally.
.12 One of the things that's always disturbed me when I look at 13 i this type of distribution, where we had the gross failure i
14 i way below and the benign leakage type failure being the more l
15 dominant one at the lower pressures, I was always disturbed 16 j by putting a very short duration pressure pulse, which might I
17 l have relatively low work energy to what is available to do 18 'I i
damage, and saying, okay, that is going to create a failure, 19 , which is a leakage type failure which, in my opinion, needs 20 ; a lot of deformation, a lot of movement to open up.
1 .
21 I What is your feeling on there? Do you feel it is i 1 22 appropriate to apply pressure duration, say, in this range, 23 if we believe the distribution here, and assume that these
, 24 leakage type failures would really occur? They may not.
25 MR. TORRI: I think the pressure rise times i
ACE-FEDERAL REPORTERS,jNC.
S l l
'720 11 07 134 1 DAVbw 1l here are still very short or very long, compared to acoustic I
2 times.
3 MR. PRATT: So'it would still apply.
4 MR. TORRI: If you know the peak pressure. l I
5 The reason we in the Seabrook study treated only 6 containment failures a's Type C failures, is because we 7 argued we didn't want to bet on being able to predict the i
8l pressure spike, if the reactor is leaking. And that is what 9 you have to do, in order to get a split fraction between 10 Type B and Type C. But I think the pressurization times are 11 still long.
.L2 ; MR. PRATT: Thank you. .
13 ' MR. LYON: Warren Lyon. A quickie.
14 If I look at this two-decade change in the i
15 l failure probability, and if I accept that kind of thing, 16 , would I then be reasonable in concluding that I could 17 l completely forget about early containment ruptures, and the i
18 ' only thing I would have to be concerned about in the short 19 term would be something that bypasses containment.
20 , MR. MOODY: Or that there would be a low 21 probability, in fact.
l 22 MR. LYON: When I say " completely forget," I 23 implicitly was trying to say 10 to the minus 4 or smaller.
i
, 24 i MR. TORRI: It would not contribute.
25 MR. LYON: So I can just forget about i
ACE-FEDERAL REPORTERS, INC.
l
'720 11 08 135 l
l 1 DAVbw 1 containment failure for the early releases. All I have to do is look at, can I bypass, and that really is where it 2l 3' lies, and then we've got the longer term thing to also 4 address.
5 MR. LONG: You still have the penetration failures to be treated under S6 rather than Sl. 56, you've 6lI 7 still got that peak to look at.
8 MR. MOODY: The questions are pretty close.
9 MR. LO'NG: You've got to go through and argue the 10 ' distribution of the failure probabilities.
11 l MR. TORRI: Yes. You might have the 1 times 10 1
12 to the minus 4 in the SSPSA was really. lef t over from the 13 steam explosions, based on the steam explosions. You add i
the two, and you get 2.5 times 10 to the minus 4.
'14 I
15 ; MR. LYON: Probabilitywise, if I'm close to a I
16 l factor of 10, I thought I'm pretty close.
17 MR. TORRI: I thought you answered it in that
! 18 I context. If the containment is isolated intact at the I i
!! 19 ' beginning, this says you can forget about early containment i
20 failures.
I 21 l MR. BAGCHI Fred, I wonder if this was something 22 you did on the back of the envelope or this is something you 23 could leave with us, something that we can use?
j 24 MR. MOODY: It is in the submittal.
l 25 l MR. TORRI: I think I would like to add X6, X1 to
- ACE _ FEDERAL REPORTERS, INC.
s....._..~. _ , , , , , , ,
t _ ,- ._ _ _ _ _ . _ _ .m. m_c______.__.
- a. .. .
l
, *720;11 09. .l 136-
.11 ;DAVbw 1 l ~ 6.
2 MR. BAGCHI What I am really asking is,'if-you 3 .want to rethink it and' resubmit it again, we would like. to 4' use it. -
9 5 MR. TORRI: Okay. I have no reason to rethink !
6! what I did here, other than what I would do. The argument
~
7 about the discontinuity somewhere around here was not.in the 8 documentation. We did that in the SSPSA, and there is no i -
9' reason why we need to do it hbre, but I have no reason to 10 I redo this curve, but the arguments I have made with respect 1
11 l on the two accounts, and in my opinion are still valid, and l ,
12 l to some extent, it depends on how exactly you interpret the
! i 13 i shear figure. You haven't said what you think about the i .
14 l shear failure. You said it is not a problem up to yield. I i
15 ' interpreted you as saying that you don't quite know what you 16 get into above yield.
17 As part of the independent review of the SMA 18 , calculations that you asked for, Seabrook or Public Service 19 I has done some additional analyses, making more conservative
. 20 ; assumptions that SMA has done on a shear failure and has
. .l .. ,
21 ' found an answer, and that will support our argument that I 22 , made here that the median fail pressure of the shear failure 23 will not dominante the assessment. That is also documented
. 24 ! as part of that.
. 25 ; I believe you found 127 psi as the shear wall
- ACE-FEDERAL REPORTERS, INC.
l !
I
- 720 11 10 -
137 .
1 DAVbw 1l failure.
2 MR. MOODY: Okay, Fred. Okay, Fred.
3 MR. NERSES: Can we go on, Jim? How much more 4 have we got left?
5 MR. MOODY: There was the containment structural j 6 review, but I am not sure we have any major disagreements 7 i that we want to make a big deal about.
I 8' Let me just summarize. We've stated that the 1 I
9 percent strain is the median pressure capability. It is 10 very conservative. We've looked at their comments on 11 leakage, penetration failures. Fred has used that in ,
12 j looking at direct heating.
13 ! I think the bottom line is that the containment 14 , is so good at 1 percent strain, and there is so much margin, 15 ' that these issues are not causing us any problems. So there 16 i is no sense in arguing whether f ailure pressure, you know, l
17 , is 215 pounds or 175.
18 > Do you want to add anything to that, Fred?
19 > (No response.)
20 MR. MOODY: Does anybody disagree with that?
21 , (No response.)
22 , (S lide. )
23 ! MR. MOODY: Maybe we should try to summarize. I .
, I l
24 l just started from the bottom, and I think we just summarized 25 l the containment structural part of it. There's no b,gE-FEDERAL REPORTERS, INC. ..
- ~ _ . - . . _ _ . _ . . . _ _ _.m ._ _.- -. -.
'720:11 11 .
138-1 DAVbw 'disag reement . I will try to start at the top.
2 We don't believe that there's any additional 3 validation necessary at all. We think there's enough i-
-4 information. .There's been plenty-of reviews. There's only ,
1 5 so many ways you can fail a containment. We've discussed 'I 6 that already.this morning.
l .
!- 7 If there is any disagreements on that comment or 8, not, we can discuss them.
i 9' (No response.) ,
10 MR.. MOODY: We've mentioned the stacking of 11 conservatisms in such studies, and I think we heard 4
12 agreements that the final report is going to reflect the 13 , fact that this has been done and require a more balanced I
14 l look at this.
15 ' We talked about the valve failure data, and I 16 think Brookhaven's probably going to look at our comments
- i 17 on that. There was agreement regardless of frequency. It 18 doesn't really change the conclusions. The steam generator 19
' tube rupture. I think we agreed that it hasn' t gone away.
- 20 We believe that it's not an issue, and if it becomes an 21 issue, there are a number of ways of making it ;
i 22 insignificant, whether it's operator action, reducing !
23 l frequency, dry steam generators. It eventually will become ,
l 24 a nonissue.
l 25 Direct containment heating. I don't know. Did j ACE-FEDERAL REPORTERS t INC.
7 ._ - .
l l
- 720 11 12 139 1 DAVbw 1l we agree to anything on direct containment heating?
2 I think we agreed that even if you look at the 1.
3 percent strain and the lower failure pressures and replot 4 the pressure curve, it is still a low probability event, 5 even if we assumed direct containment, which we don't think 6j it can.
I 7i MR. TORRI: To the degree that our earlier 8f submittal was based on a misinterpretation of your 9 information, I think I earlier commented, it probably should 10 l be recognized.
l
. 11 MR. FLEMING: The other. thing I heard agreement 12 l on was a Seabrook-specific analysis was more powerful than 13 the Zion analysis.
14 MR. NERSES: I think you heard more than that. I 15 ' think what we would do is, and we are in the process of 1.
16 l doing is, we're looking at the containment of entry 17 l quantification with regard to Zion and coming up with the 18 median pressure pulses associated with the phenomenon and 19 finding out whele they are coming from and then compare them 20 against the distribution, and that depends on Charlie.
21 So what you would see in the report would be a 22 very clear discussion of where yours is coming from, what i
23 l makes up those loads and how they fit in. That is what we
,- 24 would look for.
~
l 25 MR. MOODY: I guess before we go from here, we l ACE-FEDERAL REPORTERS, INC.
m.,..,, _ , , . . . .
- m. _ ..... ~ .....
. . . . . . . . . ..e.or.. . - . = - - * - = = . - - ~ * - - **
e .
I
'720 11 13 i 140 l
1 DAVbw 1j are prepared to supply whatever you need.
2j MR. NERSES: I' suspect we will probably be going i
3' over some of the things, what's been presented here, and 4 we're also going to be looking at some documentation, and I i
5 suspect we've heard everything that we expected to hear.
6, We've covered all the items that we expected to cover. I I
I 7 expect that we are going to be back together at some time, l
8{ perhaps as early as next week, to get into further' detail.
9l I don't know if there is anybody else that has i
10 any further comments that they want to make, but I guess 11 ; that's where we are at, and I guess we can adjourn this 12 meeting. ,
13 I MR. FEIGENBAUM: Ted Feigenbaum, New Hampshire 14 ! Yankee, and I would just like to ask the audience if anybody i
15 l feels that there is any major areas of disagreement here 16 that we need to identify right now. People that have to get 17 together and when.
18 ' ,
MR. LONG: I think we've got some big areas that ,
19 ' there are differences down in writing. I don't kn'ow that we 20 know how we want to try to attack them right now, but I 21 suggest, unless somebody proposes a specific meeting on a 22 specific topic at this point, we'd discuss this and have a 23 telephone call to see if any particular group of people 24 , would need to get together.
25 MR. NEWBERRY: I think there may be a couple of I
ACE-FEDERAL REPORTERS, INC.
m,,, , -.~.m., m ,,_
__.2 . . . . - _
4 .
s *
- 720 11 14 141 1 DAVbw 1l areas where there is disagreement.
i 2I Now how we resolve those things depends on.what 3 Brookhaven is ultimately going to say in their report. I 4 don't know.- In some areas, shutdown events, I can see where 5 some of this information is going to help a great deal in 6 understanding.what they have submitted and came out in some 7 areas of the report.
l 8 I think we just need to look at the information 9 and assess it. -
l
~ 10 l MR. NOVAK: Let me make one statement. I think 11 in this case, it.is very helpful that we have a transcript.
12 We will make.the transcript available to Brookhaven and to
~
13'i Public Service of'New Hampshire and the Staff.
14 ! I would expect that Brookhaven's going to sit 15 ' down and study this transcript, use the material, understand i .
16 l what you've said, and then complete their review.
1 17 l I would want them to work on getting their report out. I would emphasize th,e need for us to close down on 18 l i ;
19 " this and disagreeing in areas, if that's the case. That 20 l doesn't bother me. I would prefer to say, Brookhaven has
. t 21 ; understood your comments of their earlier draf t. They 22 recognize they had some work to do when they turned it over to us. ;
23 l 24 l I think it's a compliment to them to put that out l
25 at such a period of time where they recognize there would be ACE-FEDERAL REPORTERS, INC.
_ _ _ _- - - ~
. i
'720 11 15 i 142 l
1 DAvbw 1] certain things brought up that reflect the need to go back 2l i and do some more work. It was never intended to be really a 3- polished draft. It was a good early cut at where they were, I
4 and at one time, we asked them to provide a draft. ;
i 5 My emphasis now will be to look at the material I
6 you've provided and then to close in on a report to us as 1
quickly as they can.
7l 8 And that is what we will be asking them to do.
9' That is a, bout where I would see us coming out.
10 ; MR. NOONAN: I think I would like to make one i
- 11. ! additional comment. We are treating this somewhat like an .
12 SER. If,we have questions, we will propose these questions 13 - to the Licensee. We will probably handle those in public 14 ' meetings. We will sit down and have a meeting sometime in 15 the near future, within the next six to eight weeks. We do I .
16 ; plan to try to go in front of the Advisory Committee on i
17 + Reactor Safeguards. I think then we will probably want a
, i j 18 f meeting prior to that, to sit down and have another i
19 i discussion about it.
20 I just want to be sure that you. understand, as we 21 go through this. If we have any questions, we will get on 22 the phone to the proper people in the Subcommittee, and then 23 , we will sit down and discuss it.
24 : ,
MR. NERSES: Okay.
- 25 l Let's adjourn.
i ACE-FEDhRAL REPORTERS, INC.
o '*
i
, '720 11 16 l 143 .
j l- DAVbw 1, I (Whereupon, at 2: 50 p.m. , the hearing was 2l adjourned.)
I .
3' ,
4 .
I 5
6li I
7 8l
. 9 ,- -
10 11 l . ,
, 12 l .
13 ,: ,
14 15 !
,\
16 j !
- - I 17 i .
I ! .
l, 18 I -
1 19 l
. 20 l
\ .
i .
j 21 ! i 22 23 ,
- i I
24 25 l ACE-FEDERAL 3EPORTERS, INC.
. a. , ..i, -. . . . - - . . . - -- . ~ ~ ----
e., . . .
CERTIFICATE OF OFFICIAI. ' REPORTER 4
This is to certify that the attached proceedings before the UNITED STATES NUCLEAR REGULATORY COMMISSION in the '
inatter of: .
NAME OF PROCEEDING: MEETING OF NRC WITH BROOKHAVEN NATIONAL ,
LAB RE: SEABROOK EPZ I
DOCKET NO.: 50-443 -
. PLACE: BETHESDA, MARYLAND .
.. DATE: ,
WEDNESDAY, JANUARY 14, 1987 were held as herein appears, and that this,is the original transcript thereof for the file of the United States Nuclear Regulatory Connaission.
t
/
(siet)
- V/
/ /
l.
(TYPED) N Y DAVID L.HOFFMAN l
Official Reporter RY,".7e'eVM#ffNRA 2"c-P
& f I .
l e
______._____m_.__.__.___ _ _ _ - _ _ - _ . -
- ^- --- - -
SSPSA KEY EVENTS l
~
4 ORIGINAL SSPSA SUBMITTED 2/84 e LLNL REVIEW OF SSPSA PLANT MODEL 4/85 -
e BNL REVIEW OF CONTAINMENT MODEL 2/86 !
e SSPSA UPDATE SUBMITTED 7/21/86 e MEETINGS - IN BETH$SDA
~
8/6/86;8/27/86;9/23/86;11/12/86 )
- AT BNL !
8/14/86;10/16/86;10/17/8.6
- AT SEABROOK 9/8/86;9/9/86;10/15/86 l
- 4. ACRS - SUBCOMMITTEE 9/26/86 -
FULL COMMITTEE 10/10/86 ,
e DRAFT BNL F]EPORT ISSUED 12/8/86 e APPLICANTS PETITION . FILED 12/18/86 :
i i
i .
! . -j j s ,
1 .
COMMENTS ON BNL DRAFT REPORT e
GENERAL COMMENT
S e INTERFACING SYSTEMS LOCA (,
- e SHUTDOWN COOLING EVENTS l'
{ e STEAM GENERATOR TUBE FAILURE e
DIRECT CONTAINMENT HEATING e CONTAINMENT STRUCTURE
[ .
I. -
t l
I 4
. i:
i :
P e
GENERAL COMMENT
S 1 e
SUMMARY
DOES NOT REFLECT FULL SCOPE OF .
REVIEW e ROLE OF RMEPS AND SENSITIVITY STUDY NOT EXPLAINED '
. O CONSERVATISMS SHOULD BE IDENTIFIED e EXPLAIN FOCUS IS EARLY RISK PROFILE e SSPSA MORE COMPLETE THAN WASH-1400 e BNL REVIEW HAS NOT DIMINISHED TECHNICAL
- JUSTIFICATION FOR A 1 MILE EPZ ,
l .
- i i
INTERFACING SVSTEMS LOCA i i
e CALCULATION OF CHECK VALVE FAILORE RATES l e TREATMENT OF OPERATOR ACTIONS
!. O RHR SYSTEM INTEGRITY .
i i
i . ;
i :
i ,
3 !
l i
l
- 3 i
i i
I
.j i ,
i SHUTDOWN COOLING EVENTS l e CREDIT FOR TWO RHR SUCTION PATHS .
e REASSESSMENT OF CONSEQUENCES i
i l -
t I '
- N ,
l !
(
I .
I j .
i i
l l
- c. , ,
j t
i STEAM GENERATOR TUB.E FAILURE e EXPERIMENTS VS MODELS -
O CONSERVATIVE PROBABILISTIC ASSESSMENT , ,
i e
I -
e ,
a
. e s-Y
J t
DIRECT CONTAINMENT HEATING O GENERIC VS PLANT SPECIFIC TREATMENT .
I e BNL ANALYSIS VERY CONSERVATIVE i
i-e REANALYSIS USING BNL ASSESSMENT OF l CONTAINMENT PRESSURE CAPACITY -
l
- i s
I l
l .
j' .
e
- 4
-l k l
l CONTAINMENT STRUCTURE -
e AGREEMENT OF YlELD STRESS e CONSERVATIVE 1% STRAIN MODEL i e ROBUSTNESS OF PSA CONCLUSIONS l
\.
l' -
i . I' l
i 1 I 4 i l l 6
[
Silft1ARY !
e AbDITIONALVALIDATIONISUDINECESSAPY . .
e STACKING 0F CONSERVATISMS IS INAPPROPRIATE .
I o APPROPRIATE USE OF CllECK. VALVE FAlLllRE DATA t
! e SGTR IS NOT A CREDIBl.E PATil FOR BYPASSIt!G C0llTAlllMENT i ,
e DIRECT C0tlTAlllMENT llEATitlG IS HDI APPLICABLE TO SEABROOK e - -
~
a CONTAINMEllT STRllCTURE IS EXTREMELY STRONG EVEN WITil BNL CONSERVATISMS .
j'
< 1 f
3 . .
3 t . .
-i I. l
-6 -5 -4 -
-2 10 10 10 10 10
.I I I I I 5 th % 95 th g med y ge
[ RMEPS (> 1S0 gpm)
BNL (> 150 gpm)
O RECIPROCAL OF INTERFACING CHECK VALVE EXPERIENCE IN PWRS WITHOUT FAILURES RMEPS (> 1800 gpm) k -
BNL (> 1800 gppi)
. SSPSA GROSS DISK RUPTURE I
i: i I i
-5 -4 -
-2 -
10 10 10 10 10 FREQUENCY ( events per reactor year j ,
i
- FIGURE 2-1 .
I COMPARISON OF CHECK YALVE FAILURE FREQU$NCY UNCERTAINTY DISTRIBUTION
. C-
l
- l l ,
l MAAP ANALYSES TO ASSESS THE POTENTIAL FOR TEMPERATURE-INDUCED STEAM' GENERATOR FAILURES
- 1. No Ooerator Action -
I a, Base Case, b, . Seal LOCAs, 2, Ocerator Action
- a. With turbine-driven AFW (not shown-sequence does not progress),
b, .
With failure of AFW and manual depr'essuri-
- zation of primary system when core temp. era-tures reach 1200*F,
- 3. Uncertainty Analyses of Base Case
- a. High' core melting temperature (3000*K),
- b. High core-upper plenum flow (low friction
- factors), .
) c. Low steam generator flow (low fraction of
) tubes carrying flow away from inlet plenum),
- d. Stuck-open secondary relief.
L e. No core blockage, .
E. -
6 6
=
~
i L -
HOT i LEG I I I. .- TUBES i
COLD LEG STEAM \ @
l gHOT LEG STEAM n( l
! GENERATOR -
SHELL i
GENERATOR SHELL PRESSURIZER- %
fg
\ l I
/ \ l
- I s/
HOT LEG LCOLD t i' I / TUBES REACTOR I LEG p l ,
00ME TUBES l
/ I HOT I
) I LE6 [N HOT LEG (lj g
{.
I
- r enR T PLENUM'
\
- l s I }
! COLD _
- LEO ----;
j ' ' '
- j
(( ) b CORE l COLD LEG N00E h COLD ( / !
! LEG -
g' l k< : >J INTERMEDIATE INTERMEDIATE - : J -
I LEG LEG 3*UNBROMEN" LOOPS ' ' I"8ROMEN" LOOP -
(NCDALIZATION sat.CE A8 UNBROKEta LOGP)
I i
' figure 7-3 Application of PWR primary system nodalization to a Westinghouse 4 loop design. l es 9
I
l
. 3.3kO/a . :
. # I i ' l~
i CONDITIONS AT THE TIME OF. PEAK '
l I .
STEAM GENERATOR TEMPERATURE IN -
i BASE CASE TRA.B t <
l T=700KT
.1m I I .
i e %
. 850K 640K l 5 .
760K
/
C l'
j 1160K ,
2.4kO/a7 j t ;
i o i r
- 1800K '
i I
(peak)
I -
l l 18kO/T ,
l l
i l
- ^ --
1-----------
_ _ _ _ _ - _ _ 7 _____ - , - --- _... =- .
j .
3.3ko/s
- 1 i .
i .
- CONDITIONS AT THE TIME OF PEAK STEAM GENERAT'OR TEMPERATURE IN
- BASE CASE Th8 B
, T=700K p '
l 1m i ,
850K 640K ,
j 1
- t l
j I [760K -
!, 1160K .2.4kO/s7 _
i j .., .
t e.
j n-_
i 1800K .
1 (peak) ;
. t .
i . 1 18kg/
i
.g a-~~. . - - - ~ ~ ~ . . ._ _n ..-+-._-.--.--.s-~ . .
. l
. . . D Waa e (TOTAL FLOW) [
L e'OUT" TUBE ,=
h
- , "BACK' TUBE .
4 ISEC = -
WHL
~>
/ T" CO TUP c
TH TC r
. WHL l s '"5 Hot leg and stecm generator natural circulation flow,
.-. Y
. O
'?rt~
A ' _ - _ _ : . _ __ ___ _ _ _ - .._-.: . _. _,.-__.:--.-. -
3 Wa 3 e (TOTAL FLOW) [
'OUT" TUBE ,e "BACK"' TUBE 4
~
.g . T33c =
_- i l
Wst _
/ T" CO UP "
' WHL TH TC -
Hot leg and stecm generator natural circulation flow,
. p
- , . , - - - - - , - - - - - - - - _ , -. ,,_. _ -_ - ,___-..._ .m.,_. ,. . . , _ . , , . . , , , . _ . _ _ . _ _ _ . , ._-_m.._,...__
-~ - -
- ( .
Wsa e '
(TOTAL FLOW) [. .
'OUT" TUBE .
A .
'BACK' TUBE
- g ..
.. . . \,
Tsec e -
~
, = . .
( ..
> ~~
Wst . 7 .Tsov
- 7 .
TUP c
- WHL -
TH Tc
%s ..
[ .
. s Hot leg and steem generator natural circulation flow.
s
< am.
~
. Fa' -
..M" h . .
COMPARISON OF COR.E-UPPER PLENUM FLOWRATES. .
IN THE WESTINGHOUSE EXPERIMENTS 70 '
MAAP,MODEL (IDCOR 85-2) ,
l ..
. .s.-
Experimental Model.
28 kw water tes't . 54 . 50 ', .
I.
.9 kw SF6 test ~ ,
.',.016 -
.017 e . .
I - . V e .. . ..
e . ,
,1 .--- .
. t .
.. . 1
- - s
/ ,. ,
s q e t
- e y
. :. - l l~* r,
=e
_ . - _ - - - , . , __m . . , .
- ,--.____w.-__.,,,,_-.-___m,, __,_.-.,,_yg,, ,., ._ . , . m., . _ ,,, _ , .,_ ,_,,_ _ _,__.- ,,,,
. . - ~
[. - -
, a
. COMPARIS0N OF HOT LEG-STEAM GENERATOR NATURAL -
I>
~
lCIRCULATIONMODELTOEXPERIMENTALDATA
. Experimental Model Prediction
. Quantity Measurement ng = 6* n g = 12 n g = 24
~~
0 (kW) 2.43 -
. 2,0 2.6 s 2.9
,, Tgg ( C)
- 30 30.7 29,2 28.4
- 19 24,2 Tc ( C) 21,7 18,8-Tco ( C) + 10-11 9.4 11,2 '12,8 !
- n g is the number of tubes carrying the out flow. -
I s
I O
N ] '
t
o ..
l PEAK SG INLET PLENUM GAS TEMPERATURE .
), AT MAXIMUM AP (*K) .
- l. Base case 860 2._,,' Seal LOCA 865
'3. Manual depressurization .
680
- 4. High eutectic temperature 880 .
- 5. High upper pl'enum flow ,
870 930 .
- 6. . Low SG flow Stuck secondary relief ~
- 7. 850 ,
- 8. 'Ho blockage . . 1040
~
L.
)
Q
-. - . . _ - __.-__,m *. - . . - . . . , , . . . - , _ _ . . . _ _,_,,,_I _
4 . .
s .
[ . CONCLUSIONS
- 1. Base case peak steam generator tube temperature is 750*K and ~1s only 850 K in worst-case uncertainty analysis.
- 2. Tube integrity analysis indicates that. temperatures of eb Jt 1100 K are required to fall tubes. Thus considerable margirr exfsts. - '
l'.
6 l
m O D
.. L; ~ '" - .
.._.s w .... . . . _ . .. . . . . . . . .. .. .
. . -' . \ i t
, . i ZION CAVITY SIMULATION EXPERIMENTS ,
Tesi: Floor of Unner Surface of Vert col Sect:.on Catch Volume Catch Volume WaLis Lip 6,07 ~0 17.8 76.]
. Steam Generator Geometry 818.0 7.38 2 11 . 11 17 l i Reduction Area 76,6 ~0 16.5 ~7
.2.37' 0 3D 9 7 . 11 O 208 l *0ut of doorgiy next to seal table wall. .
Fl j .
l i
i .
,, SEAL
! TABLE . o V//
j TRANSPARENT i
PLATE .
SUPPLY ;I TUBE :- ~
.I
,j SS _ c.'
. pj d -
N l s 1
- q. , .
,, . _ . , _ . . O. . ,
i l . . . . ~ . . . .. . - - - - - ----- - . - - -- - -
- '~-
DOOR l
i O .
KEYWAY l SCREENIN OPENING 1
/ TOP SURFACE l ;' ;
i ,
- l'!
! 'l .
SUPPLY !
TUBE .' '
f
+
l,l,'
.f D0O.
l l,, l "
L. I :;
. BIOLOGICAL . :l'
' . SHIELD :ll SEAL.
l TABLE l i LIP - .
, l -
O+ -
[ . -
l l STEAM GENERATOR .
I L
l ,
l
[
~
e 1 .
e f .. .
l-: -
9 *
- O % = 4
<.3 I I .
1- I l i i I
. i l 1 l '
I' ! '
I i i . I l
' 1 1 ;
- I i .
l
- SEAL i l
i TABLE ..' -
i I i kj i lgl'(j
,. i l I
.l~ DOOR I lSTEAM i r" i i
. BIOLOGICAL SHIELD i
' l8 l l l GENERATOR l
[ DOOR 1 l i i i i Ie l l l l fI7 l.;
SUPPLY TUBE ,
- e e
9 O
l
- e w
e l
._~*- .-_--..-,_.._L.- '
4 m---- L w ._-a a. _ aa e M6-W le '+*@t. -.6 'O.4h-e. .. ge.- e +,e6 Mey&Me- eae=e me g h ,
e 9
I
/A -
//
= _
{x
.EAL \ -
TABLE \\
TRANSPARENT \
PLATE -
SUPPLY TUBE z:;j ,,
T,,i;j!$5!!
i h 1 $!
// / l l
4
. e.
9//26~C/
//d~A / / #d .
/7f/37' d pA./J/.#d~f 1 7 X E A / C 7 #'
OF~ .ffA./A 00A dpA) 7'4/AJH6x37' Joss sx) 2x/& a474: .
/s/c/a~A .Zs n f .: /S~9* PS/ C I
/fSid/#A./ X-2; N00P : / 7 s~ fJ/gr MED/A /V / 6 9 PS/G V (O/fM/6~~ _
&/00 W - S ~6 ' M / A d o K b IUN 80/278d/#dX/2 F&&//ff- d, O/'
7 .
3,/
fd~N8$0&/d C 0AJ?~d/Al/7dR/ Pdfd//f4-
-y
/fX /0 (/Ve Cf/roPP)
O C W'ITH Cute P P )
l S$PSA ffSEZ1 / x /g W f
584 l .. .
s * .
. -- _. - -_ ._ ~ .- - -- ..
e l
.'..,e....,.,'..n..
~ .-
,. ,. . s.. s
,g .
io'
_. ...;.* = n= s + ' *-
.t i;: . :. e n S-D:.iihtjWsgr w ::m?.Q.f=.ys..
'.?*
.w .
2 G/V p.. .. u.sn.. . .W.,m.m.~yrg-~'.fi
..m.
m.,w2 .,
a :'
.w . . .
( 1 .,
. .. ?.-g.,ri.wr r. .. -p rs- .' '
. . '.?: w..
c.);: y. 2.~...,.
s . , s .~;. J i
c:.'.%<+.J,,-y
' Op, /
m . :..ap. y , . . 5- ..
s
.::=r w v..n. .
t r.:. s.mm.:9-s
/
.z. . u.s ...,...2 w:.v
. s.,-- . v. /
/
- i .. : 1' .
1
..r
.. . : . 10' ,.
-5 I 3~ i / >
.~ ,, 2 ~
i PSEA BROOK l _g Q..: ,' ..
I
/ (BA.MD 0/V 8NL DA A x.4-= x. :.to-,- . -.
.5 I
J y ; r
}
f
).
t
. u 2 > ] .'
I,.,:.
\p o
f f m . .
.,1,_. . . . . . '
1 .,.- - .
\.
p , , ~ .. ...,. _.
. .v.~. -
_. -c:n10 .w_v.3
,.2
,r
- ..s em :-s u .r -
.,; s y
.p.-. .,4.tr.. . . r. a. -- 5' . s t e
p r 1
- m. ,- .
. . . t-r'
/ - !A -q
- j ,. ~' f
. ? [' f , [ X $
,e /
l l j .....,,,,,,,,iii
- 2. -
130 140 150 160 170 18e 2
. . .# z_ - .
~-
PRESSURE (PSIA)
&% ..q P
%f. L =&/ :@ # M n u d 843 & 6 W B #l /14 M .
1;.s . ;.- f- .
d' __I
~
f $ (A TC. bf Q.n ?i,.~ L. . . .
. - ^{ 1.aae r pt }..r ;Jv - % i- . -
- ^;:x: . - -
. .m . _ . -'
8...-...
,@ s .[E*.
- 1, .' . # j'y
..e-1.e .
- ;3
. - *I**
e
, 3
~ W },4
. ';?:dba i '6*(WY b_$$. ^]'l { :.n.
,[. . :-l: *; _ ., _ __ }fl.llQ.k.. . # .L ,=.s-
~
'_h
_3 .a a be,a +4.r .--B -a e L -ma A- m 2 -
<As --e-r+ r -- -s, a,,w.-
._,._. ._-.__..-e-. -- - - - - - - - - ~~ ~ " ~ ~ ' ~
. 'x
-s e e 4
es.
/HBE A WWMAE ;
\
~ ffGfA/N/S7/C ASKECEMfA)7 SATEA I
\
- s F e m es/c ss rs
/
ds/)c/sppAr/oAs GF E&~ASA00K
, .S~/EC/F/G DAYA #A)C&EAK e
[frM ' $72# AW' P%"K/ ' of g&Ljs _/JMF l
0* S SA ESC /K ('N/EW AJo AWMD#X/5^A77om /r x 4 / c w j xryg;v r N/75^ NM7~ T4C/F/c ,9LAJg pgyg ffekts //5 7/A/ C w .r r . [#4tc .gxe .
1 fez &~hcr dd7 Y W o & / er.
s,
.o
- M
' e 'ew
._e b
g,- 'O, -
- . . . . . . . _ _ - . _ . - .. . . . _ . - . - _ . _ _ . . . - . . . ~ . . . .
t
, .J ,..
. W
'p f George S. Thomas q .%. ; I Vice Prod $elW Nuclear Producnon January 20, 1987 Pubuc Service of New Hompshire NYN-87-002 i New Hrmpshire Yankee Division f
United States Nuclear Regulatory Commission
.- Washington, DC 20555 Attention: Mr. Victor Nerses, Project Manager
. PWR Project Directorate No. 5 Division of PWR Licensing-A Esferences: *
(a) Facility Operating License NPF-56, Construction Permit CPPR-136, Docket No.s 50-443 and 50-444 (b) USNRC Letter, dated December 8,1986, "Trans-mission of Brookhaven National Laboratory Draf t Report of The Seabrook Emergency Planning .
Sensitivity Study", V. Nerses to R. J. Harrison
Subject:
Comments on Draf t Report
Dear Sir:
Enclosed please' find New Hampshire Yankees- (NHY's) comments * *co'n-cerning the Brookhaven National Laboratory (BNL) praf t Report A-3852
[ Reference (b)].
These comments are substantially the same on those provided to the Staff and BNL at the January 14, 1987, meeting held in Bethesda
' to discuss this draf t report. The only changes of substance (i.e.,
non-typographical corrections) are contained ~in the responses to items 1.1, 1.3, 2.7, 3.5, 4.0 and 5.0. NHY modified these comments as a result of the discussions at this meeting.
If the Staff requires any further clarification of these comments, J, , do not hesitate to contact the NHY Bethesda Licensing Office (Mr. R. E.
Sweeney) at (301) 656-6100.
Very truly yours,,
e ch S. Thomas M
Enclosure cc: Atomic Safety and Licensing Board Service List (Offsite EP)
T16.L2, 4.n 4G,:y P.O. Box 300 . Seabrook. NH 03874 . Telephone (603) 474-9574
.,.-t,e i
ASLB 57rvice Line (Off-Sits EP) e .
LHelen Hoyt, Chairperson .
Carol S. Sneider, Esquire Atomic Safety and Licensing Board Panel Assistant Attorney General U.S. Nuclear Regulatory Commission Department of the Attorney General East West Towers Building One Ashburton Place, 19th Floor 4350 East West Highway Boston, MA 02108 Bethesda, MD 20814 - .
Senator Gordon J. Humphrey*
Dr. Esseth A. Luebke U.S. Senate Atcaic Safety-and Licen. sing Board Panel Washington, DC 20510 U.S. Nuclear Regulatory Commission (ATTN: Tom Burack)
East West Towers Building 4350 East West Highway Richard A. Hampe,-Esq.
Bethesda, MD 20814 Hampe and McNicholas 35 Pleasant Street Dr. Jerry Harbour Concord, NH 03301 Atomic' Safety and Licensing Board Panel U.S. Nuclear Regulatory Commission Thomas F. Powers, III East West Towers Building Town Manager 4350 East Wast Highway Town of Exeter Bethesda, MD 20814 10 Front Street Exeter, NH 03833 Diane Curran, Esquire Harmon & Weiss Brentwood Board of Selectmen 2001 S. Street, N.W. RFD Dalton Road Suite 430 Brentwood, NH 03833_ -
, _ Washington, D.C.- 20009 Peter J. Mathews,, Mayor
~ ~~'
Sherwin E. Turk, Esq.-
- City' Hall ,
Office of the Executive Legal Director Newburyport, MA 01950 U.S. Nuclear Regulatory Commission
~ Tenth Floor 7735 Old Georgetown Road Calvin A. Canney Bethesda, MD 20814 City Manager City Hall Robert A. Backus, Esquire 126 Daniel Street 116 Lowell Street Portsmouth, NH 03801 P.O. Box 516 Manchester, NH 03105 Stephen E. Merrill, Esquire Attorney Geners' r Philip Ahrens, Esquire George Dana 48$bes Esquire h Assistant Attorne.i General Assistant A* r renc General Y Department of The Attorney General Office of sie e, s/ney General L Statehouse Station #6 25 Capitol Street
- 1. Augusta, ME 04333 Concord, NH 03301-6397 Mrs. Sandra Gavutis Mr. J. P. Nadeau Chairman, Board of Selectmen Selectmen's Office RFD 1 - Box 1154 10 Central Road Kennsington, NH 03827 Rye, NH 03870
- Letter of Transmittal Only
.- _ .~. _ . - . . . . . , - - , . . _ . _ . . . . _ . _ _ _ . . - , _ _ _ . _ _ . . _ ~
, .1 ' . ' .
ASLB Sarvico Liat (Off-Sits EP) *
(Continued)
.Mr. Angie Machiros Chairman of the Board of Selectmen Town of Newbury Ne wbury, MA 01950
'Mr. Willias S. Lord Board of Selectmen Town Hall - Friend Street Amesbury, MA _01913 Senator Gordon J. Humphrey*
1 Pillsbury Street Concord, NH 03301 (ATTN: Herb Boynton)
H. Joseph Flynn', Esquire Office of General Counsel Federal Emergency Management Agency 500 C Street, SW Washington, DC 20472 .
Paul McEachern, Esquire Matthew T. Brock, Esquire Shaines & McEachern -
25 Maplewood Avenue -
P.O. Box 360
- Portsmouth, NH 03801 .
Gary W. Holmes, Esq.
Holmes & Elis 47 Winnacunnet Road i .Hampton, NH 03842 Mr. Ed Thomas '
FEMA Region I ,
442 John W. McCormack PO & Courthouse Boston, MA 02109
. Robert Carrigg Town Office ~
Atlantic Avenue North Hampton, NH 03862 Judith H. Mizner*
Silvergate, Gertner, Baker, Fine, Good & Mizner 88 Broad Street Boston, MA 02110
- Letter of Transmittal Only
me.eus M% y ow 4 -w ^ me 4 . vM eme 9- 4.-- 4 oW9-t e e
- 4 ..
.t 6
O e
0 9
ENCLOSURE TO NYN-87-002 COMMENTS TO BNL ,
DRAFT REPORT A-3852 -
SEABROOK STATION 4
e 6
9 .46 1
i .
g 'ke Our Comments on tha BNL Dr:ft Rsport A-3852 ara summarizsd in this raport.
The comments are organized by the following major sections: '
- 1. General Comments
- 2. Interfacing Systems LOCA
- 3. Shutdown Events
- 4. - Steam Generator Tube Rupture
- 5. Direct Heating
- 6. Containment Structure 1.0
GENERAL COMMENT
S The general comments in this section are aimed primarily at the BNL PREFACE and
SUMMARY
but also apply to the whole report, in general.
Table 1-1 calculates the Safety Goal for Seabrook Station using BNL's stacked conservative sensitivities. As shown the Safety Goal is met with a 1 mile evacuation model.
1.1 Inconclusiveness of BNL report The pref ace and summary tend to understate BNLs actual review.
These sections should be written to be more consistent with the actual contents of the BNL Report to more accurately reflect the scope of the review ef fort.
The implication is that only PLG-0465 was reviewed. As demon-strated throughout the BNL report, PLG-0432 and the relevant parts of the original SSPSA wer,e reviewed and evaluated, -
PLG-0465.is nothing more than a sensitivity st' dyu showing the impact on results and conclusion using the conservative WASH-1400 methodology source terms. BNL used the PLG-0465 source terms and used PLG-0465 for sensitivity studies (risk comparisons).
The report leaves the impression that additional validation may i be necessary. We believe that adequate confidence for decision
[ making can be achieved by emphasizing review of those areas that affect early releases. They are the following:
- 1. Containment Bypass such as interfacing LOCAs
[ 2. External Events with Containment Damage l
L 3. Containment Isolation System
- 4. Containment Structural Integrity-WASH-1400 plus the NRC's 11 years of PRA experience and know-l 1 edge and the Seabrook evaluation of these areas should provide l sufficient confidence in our results. In addition, the following is considered important with regards to confidence and margins in the Seabrook results:
i 1
[ 1. . Thera cra cequences in tha SSPSA thst .iupact. ccely rolocacs that were not- explicity includ d in WASH-1400. This includss Airplane Crashes, Turbine Missile, and Earthquakes which doe inste public risk, fires, floods 'and reactor coolant pump seal LOCAS.
- 2. The WASH-1400 PWR results strictly apply only'to Surry at a so-called composite site. The Seabrook PSA results in PLG-0300, PLG-0432 and PLG-0465 are specific to the Seabrook plant, contain-ment and site. It should be clear upon comparison of these studies that WASH-1400 results are not applicable to Seabrook and the Seabrook study more accurately reflects the risks and accident n frequencies.
- 3. A one-mile EPZ can be justified even when a conservative one percent strain criteria for containment failure (requested by BNL) and conservative WASH-1400 methodology source term are used in a stacked sensitivity case. Although single sensitivity studies are useful, continual stacking of these upper bounds is not an appropriate way to address uncertainties for decision making purposes. Also, the 10 mile EPZ is based on best estimates (50% confidence level / WASH-1400). Even af ter stacking all the sensitivity cases, the BNL results still justify 2 miles when compared with the NUREG-0396 results. We believe that this goes well beyond what is needed to establish the robustness of the conclusions of the sensitivity study.
- 4. This is the second review by BNL of the Original SSPSA Containment Analysis. Also, the LLNL review of the' plant model identified.no
- significant safety issues and their comments on areas that offact early release were either positiye or identified conservatisms in
~
our model. These areas' include: .
o ' Aircraft Crashes (LLNL Page 4.3-2) o Turbine Missiles (LLNL Page 4.8-2) o Containment Isolation (LLNL Page 3.4-96) o V Sequence (LLNL Page 3.9-4 through 3.9-6 and Page 3-21) o Seismic (LLNL Page 4.1-74)
The overall SSPSA methodology is similar to other PRAs reviewed by the NRC (Zion, Indian Point). These reviews were f actored into the original SSPSA.
- 5. There are a number of potential conservatisms in PLG-0432 that should -be recognized.
S6 Enservative fragilities No credit for recovery af ter earthquake V double disc treated like CV Test frequency Factor of 2 in frequency model Probability of pipe break me as S7 plus no operator action Unit 2 Turbine Missile effects on Unit 1 S2 E credit for recovery af ter earthq'uake 2
. ~n_~,. .-. --. _ . - - - - - .~
w ' *'. - -
i
- 0: . ..
3 1.2 Characterization of PLG-0465 as a Risk Assessment ,
One issue that needs to be addressed in the final draft of the BNL review report, A-3852 is the difference between a risk ,
H assessment and a sensitivity study. PLG-0432 is a risk assessment and PLG-0465 is a sensitivity study. The former attempts to quantify all significant contributors to risk and to cuantify all identifiable sources of uncertainty. In addition, the calculations in a risk assessment. are performed very carefully to distinguish between two different kinds of probability: the relative-frequency of a random variable type and a subjective assessment ,
of a state of knowledge or expert opinion type. The-best risk assessment currently available for Seabrook is presented in PLG-0300 and updated in PLG-0432.
By contrast, PLG-0465 is a sensitivity study designed to invest-igate the singular importance of one and only one aspect of the uncertainty. That is the uncertainty in the estimation of radioactive release source terms. The curves in PLG-0465 were developed, using best estimate assumptions on all other parameters and median accident frequencies. This was done to avoid the " stacking" of uncertainties and to provide the highest degree of comparability with the results of NUREG-0396 and WASH-1400. The NUREG-0396 curves were based on median accident frequencies from WASH-1400 and risk curves that were generated using the best estimate or central estimate assumptions on
. -source terms and consequence model parameters. Even with this approach, the direct comparisons between PLG-0465 and NUREG-0396 are conservative because of a more complete treatment of dependent events and component fai. lure data.in PlG-0465. For examplhj NUREG-0396 does not include the seismic events, fires, floods, other external events, common cause failure, reactor-coolant pump seal LOCAs and other events that were included in PLG-0465.
Starting from the proper characterization of PLG-0465 as being based on conservative source terms and best estimate assumptions on everything else, it should not be at all suprising that the curvou will increase as additional conservative assumptions are added. The BNL Report summary states that "...the risk estimates quoted from PLG-0465 at the beginning of this summary do appear to be influenced by the various sensitivity studies performed at BNL". They are influenced, because they are generated by stacking new conservative assumptions (e.g. that there is a chance of early containment failure due to direct heating or thtt there is a chance of thermal failure of steam generator tubes during high pressure melts etc.) on top of a conservative source term methodology and fixing everything ,else.
3
o .
A final comment we have with these sensitivity studies is '
the mixing together of the two dif ferent types of probabilty.
The curves in NUREG-0396, WASH-1400 and PLG-0465 only include the relative frequency type of probabilty. All the uncertainity types of probabilty is isolated from the risk curves, i.e. the curves are computed as conditional on a set of assumptions regarding all uncertain paramenters. By contrast, the A-3852 curves reflect both. For example, the curves for early containment f ailure include expert opinion type probabilities that uncertain phenomena could result in early containment f ailure. How one then compares these curves is not clear and their usefulness in decision making is questioned.
In the final report of A-3852, the proper role of PLG-0465 and the BNL results as sensitivity analysis should be made clear.
1.3 Nature of the BNL Review Since the publication of the Reactor Safety Study, it has been generally accepted that risk assessments should be based on the most realistic set of assumptions that can be technically justified.
This is in contrast with the traditional approach employed in deter-ministic safety analyses (e.g., the Chapter 15 FSAR analyses) which are based on a set of " stacked" (i .e. , concurrent) conse rvative assump-tions. An important reason for basing PRA results on realistic assumptions is to avoid misleading conclusions ab,out risk levels and the principal contributors to risk. .
The use of conservative assumptions .in deterministic safety analyses is a way to deal with uncertainities in the models and data used for the calculations. An unsatisf actory characteristic of this approach is that the degree of overall conservatism in the final results is not quantified, nor easily controlled.
The overall degree of conservatism depends not only on the degree of conservatism associated with each assumption, which is seldom quantified anyway, but also on the number of such assumptions that the particular calculation happens to depend on. For example, if a radiation dose calculation, Z, happens to depend on say, 4 variables that are uncertain, i.e., Z = Z(A, B, C, D) and Z is calculated using conservative values of each variable such that each conservative value has a probability of .99 of being on the conservative side of the true value, then the calculated Z value will have a .99999999 probabili ty of being on the conservative side of the true value. For this reason we find the approach of combining all sensitivity curves in Figure S.7 to establish an upper envelope risk curves to be lacking in technical meaning and an inappr'opriate basis for decision making for emergency planning. Such stacking of conservative assumptions was not performed in the PRA analyses that provided the foundation for the decision to establish a 10 mile EPZ for all plants.
4
m . - . . ._-._ __ . _ . . .. _ _ ._ __ . __ _
- p. _ Thara cra two typreachco us:d in PRAs to address uncarteintics: ,
. The approach of propagating uncertainty probability distributions through the calculations to establish uncertainty probability i distributions on 'the results, and the approach of performing 4
sensitivity calculations. . The former approach is a systematic way to combine all the uncertainties into the final results ,
with the probabilistic " weight" of each point in the range of.
final results. This approach was followed in PLG-0432. The ;
latter approach is only useful to examine specific issues such as the case with PLG-0465. There is no scientific basis for combining or adding two or more sensitivity cases. In fact, 4
this tends to provide misleading indications of the significance of any and all underlying issues.
The curves used as ' decision criteria in PLG-0432 and reproduced as Figures S-2 through S-6 in the BNL report were taken from the Reactor Safety Study and NUREG-0396. Although the curves from these j studies have ranges of uncertainty associated with and.even quantified
. for them .the curves themselves are based on median accident frequencies j and best estimate assumptions on everything else, including what was then the best estimate assumptions on such factors as source terms,
{ containment strength and the progression of interf acing LOCA sequences. .
It is very important to understand that WASE1400 was based on what
- were then viewed as realistic assumptions on source terms and everything else and this was the basis for decision making, i.e., the decision to
', set the EPZ to 10 miles for all plants.
With this background, it is important to properly characterize y what was done by the BNL review and to, contrast their approach to what was done in the Peer Review group put together by NHY to raview PLG-0432 and PLG-0465. The BNL review can be accurately described as a qualitative critique of the two reports and a collection of sensitivity calculations used to explore the significance of selected issues. The BNL review appears to look at only those assumptions that could potentially drive the risk results higher. There is no evidence that identified the many conservatisms in the "best estimate" assumptions on NUREG-0465, which if found would support the bottom line conclusion that these studies provide a better justification for a 1-mile EPZ at Seabrook Station, than NUREG-0396 provided for 10 miles at all U.S. sities. One such conservatism is that the PLG-0465 results include many risk contributions not included in NUREG-0396, including:
l o reactor coolant pump seal LOCAs o seismic events and other external events o fires, floods and other internal plant hazards o common cause failures due to design, constuction errors, operator errors or environmental stresses
- o support system.f aults such as loss of component cooling It is quite clear from the SSPSA results that omission of the above contrib-utors would have resulted in significant reductions in the PLG-0465 risk levels. An other factor that must be taken into account is the fact that 5
o .
' all the issues addressed in the BNL sensitivity curves; such as how to estimate the frequency of the V-sequence, the concern
'regarding possible thermal f ailure of SG tubes during high pressure scenarios, the issue of direct containment heating, and the question of the importance of shutdown loss of cooling events were not addressed in WASH-1400 nor NUREG-0396.
A final observation is the decision by BNL to use the PLG CRACIT results instead of the corresponding BNL CRAC2 and MACCS results in the presentation of their sensitivity results. In addition, they added additional conservatism in the assignment of accident sequences to release categories (S1 vs S6). The only reason we can determine for this approach is that for a consistent treatment of multiple source terms BNL consistently found lower risk curves than those presented in PLG-0465. Apparently, BNL put together the worst possible combination of the PLG and BNL results to see how high
- i the risk curves could be driven up. Even with all of this, the NRC saf ety goals are still met at between 1 and 2 miles. BNL apparently is not taking a position on where the risk curves should be, or on which side of the PLG curves they lie. This is in sharp contrast with the NHY sponsored peer review team which took a rather strong position that not only are the PLG-0465 curves conservative but so are the PLG-0432 "best estimate" curves.
i e
1 I'
6 I -
. SOCALC .
PLO-0465 (MEDIAN)
EVAC =0 EVAC =! EVAC =2 RC FREQ 1.5 M1 POP ACUTE SO ACUTE S0 ME 50 S1 1.50E-09 6.47E+03 9.26E+00 2.15E-12 9.26E+00 2.15E-12 9.26E+00 2.15E-12 S2 7.50E-M 6.47E+03 1.21E+02 1.40E-07 8.33E+00 9.65E-09 0.00E+00 0.00E+00 S6 1.50E-08 6.47E+03 3.85E+02 8.92E-10 1.93E+02 4.46E-10 2.41E+00 5.60E-12 1.41E-07 1.01E-08 7.74E- 12 PLO-0465 (MEAN)
. EVE =0 EVAC =1 EVAC-2 RC FREQ l.5 MI POP ACUTE SO ACUTE SO ACUTE SO S1 4.00E-09 6.47E+03 9.26E+00 5.72E-12 9.26E+00 5.72E-12 9.26E+00 5.72E-12 S2 2.10E-05 6.47E+03 1.21E+02 3.93E-07 8.33E+00 2.70E-08 0.00E+00 0.00E+00 56 6.50E-07 6.47E+03 3.85E+02 3.87E-06 1.93E+02 1.93E-08 2.41E+00 2.42E-10 4.31E-07 4.64E-08 2.48E- 10 PLO-0465 +BNL(MEAN TO PESSIMISTIC)
EVAC =0 EVAC = 1 EVAC =2 RC FREQ 1.5 M1 POP ACUTE SG MUTE SO ACUTE SO S1 5.20E-06 6.47E+03 9.26E+00 7.44E-09 9.26E+00 7.44E-09 9.26E+00 7.44E-09 S2 2.10E-05 6.47E+03. 1.21E+02 3.93E-07 8.33E+0,0 2.70E-08 0.00E+00,0.00E+00 56 5.60E-06 6.47E+03 3.85E+02 3.33E-07 1.93E+02 1.67E-07 2.41E+00 .2.09E-09 7.33E-07 2.01E-07 9.53E-09 Sl= 4.0E-9(PLO-0465) 2.4E-6(BNL SG TUBE PESSIMISTIC) 1.4E-7(BNL Y SEQUENCE) 2.7E-6(BNL DIRECT HEATlHO-2.7E-4*.01)
=5.2E-6 S6 = 6.5E-7(PLO-0465) 5.0E-6(RAI 21 SHUTDOWN EVENTS-CONSERYATIVE)
=5.6E-6 F
TABLE 1-1 SAFETY 004. SPREAD SHEET l
I
\
. . ~ . . - . . _ . _ _ . . _ . .. _
2.0 INTERFACING SYSTEMS LOCA -
BNL has a number of comments on ' the PLG-0432 treatment of check v'alve f ailure frequencies. These comments are presented in BNL
~
Section 2.1 followed by a BNL reanalysis of V-sequences based on their reassessment of check valve failure frequencies. Our responses are provided in the following subsections.
2.1 NPE Data Base BNL states that " PLG selected a particular subset of those events listed in the NPE data base, namely, events involving check valves at the RCS-ECCS interf ace".
This is not true. As stated on p. 3-17, paragraph 1 of PLG-0432:
"...only those (valves) associated with PWR, ECCS and RCS were considered the most relevant for the valves consid,ered here, which are initially seated and testable". The data search that was performed in NPE included all the valves in these systems, not just those at the RCS-ECCS interface. Of the 21 leakage events reported in Table 3-8 of RMEPS (accounting for multiple events identified in several reports), the following types of failure events were indicated.
l Valve Type No. of Events Accumulator Check Valves 17 ECCS/RCS Interf ace Valves ,
4 ,,
Total 21 It may appear on the surf ace that only accumulator and ECCS/RCS interface valves were included in the data search. This is not the case. All reported events in the ECCS and RCS were included however in the search. The fact that other types of valves did not produce events in the final list of screened events is in itself statistical evidence that should be accounted for in estimating the failure rates.
It should be noted that the events of interest in the analysis are those that produce in excess of 1800 gpa of leakage through ECCS/RCS interf ace valves. In order to achieve such a leak rate, it is necessary to postulate a major rupture of the valve disc. There has never been such a failure occur in U.S. reactor operating exper-ience and we are not aware of such an event in any non-nuclear energy systems. Af ter reviewing the experience with all reported check valves events on all systems in NPE we decided to focus our data analysis on all normally seated check valves in PWR ECCS systems including accumulator check valves, those at the interf ace between the ECCS and RCS and those that separate the ECCS system from the RWST, containment spray and containment sump. In cases where inter-8
- - . -- ~. - . - . . - -- w . .
- - w aza . . .u.---.-. -- . -- - - - - -- !
, , , e j .
)*
- t ,. . .
facing check valves are numbered as belonging to a different systek (e.g. , ' containment spray) they were excluded. All these valves are normally seated and while they have different loading conditions, they
. have correspondingly different design criteria. Except for variations
.in pressure differentials.across the-valve discs and internal pressures, the environmental conditions of ~all PWR ECCS check valves are very similar (e.g. . temperature, boric acid, testing requirements) . Since p all the closed valves in-the ECCS are designed to the ASME code and are safety grade there are inherent margins of safety for the structural integrity of.each valve to remain intact in its normal and even severe abnormal environments. Therefore, we are not talking about events in which the valve d'iscs rupture due to excessive loads, we're talking about spontaneous ruptures of the material due to unforseen and undetected
. failure mechanisms (e.g. design flaws, material flaws, corrosion, fatigue, etc.) which reduce the capacity of the valve discs .co survive their normal environment. Many believe that such failures are essentially
, impossible. In view of these considerations, we maintain that the full PWR ECCS check valve population is an appropriate data base for this analy' sis. However, ' under no circumstances do we think it's appro-priate to limit the data analysis to accumulator check valves. If any subset of the EECS check valves should be used, it would be the ones being analyzed: the ECCS/RCS interface valves.
In the BNL analysis described on page 2-7 other valve types appear to be included in the BNL count of failure events but exposure time is limited to the data on accumulator check valves. The BNL survey identified about twice as many leak events (35 vs 17) and the exposure time is being limited to accumulator check valves only, which was about one forth of that used in PLC-0432 (2.3 x 107 hrs vs. 1.1 x 108 hrs). All other aspects of the PLG analysis were held fixed.. As a result BNL derived a failure frequency vs leak size curve roughly 8 times greater than analyzed in PLG-0432. In our opinion, BNL has come up with a more appropriate small leak frequency for accumulator valves, but has not contributed to a greater understanding of the frequency of interfacing check valve disc ruptures beyond that associated with providing a more complete survey of check valve leak events. The BNL analysis fails to take into account a very crucial part of our evidence about interfacing valve ruptures-the fact that we have never experienced more than relatively minor problems with 0-rings and imperfect valve disc seating on any ECCS valves, including
/ the interfacing check valves in nearly 500 reactor-years of PWR experience. This is a hard statistical fact that must be accounted for in any valid assessment of interfacing LOCA frequency.
To examine whether the BNL results for accumulator check valves provide reasonable results for interfacing check valves, consider the event of a valve disc rupture. Neither PLG nor BNL has identified any valve disc ruptures. The most severe event identified was one accumulator check valve leak of about 200 gpm due to 0-ring (synthetic rubber material soaked in weak boric acid) deterioration, not a structural problem with the stainless steel. Hence, there has been zero valve disc ruptures in PWR experience
. thus far. On this point, the PLG and BNL analysis is in agreement. What is at issue is how to estimate the exposure time.
i 9
l Th3 PLG cstincto of cxposura ti 3 includsd PWR cxpsrianca thru Novcubst 30, 1984, which was about 424 reactor-years and 1.1 x 108 component hours-for all ECCS check valves. BNL extended the data survej through December 31, 1985 for a total of 451 reactor years. There has been roughly an additional 50 reactor-years of PWR experience through the end of 1986 for a total of 500 reactor-years without any known valve disc ruptures, and certainly no interfacing check valve disk ruptures.. The average number of ECCS check valves per plant in the PLG analysis of exposure data is about 30. Hence, using all the ECCS valves, the total valve exposure time thru December 31, 1986 is:
(500 PWR years) x (30. valves per plant plant) (8766 hrs / year) = 1.3 x 108 hrs.
The above exposure time includes accumulator check valves, ECCS/RCS interface check valves and other check valves such as pump discharge check valves.
To get an idea what fraction of the ECCS check valves are interface check valves (i.e. those designed to isolate the reactor coolant system from the ECCS but excluding the accumulator check valves), consider the distribution of ECCS check valves at Seabrook shown on Figure 4-8 in PLG-0432. This figure includes the following ECCS check valves.
Number of Location check valves
- 1. RHR/RCS cold leg interface 8
- 2. RHR/RCS hot leg interface 4
- 3. SI/RCS cold leg interface 4 ,
- 4. SI/RCS hot leg interface 4
- 5. CHRG/RCS cold leg interface 6
- 6. Accumulator cold leg interface, . 4 ..-
- 7. ECCS pump discharge 6* *
. Total 36
- Two RHR pump discharge check valves are not shown in Figure 4-8 of PLG-0432. The discharge check valves on the charging and 2 SI pumps, are, however, shown in the
. Figure.
The above count does not include check valves shown in Figure 4-8 located in lines between the ECCS pumps suction paths and RWST and containment sump and indicated as belonging to the containment spray system. Of the above 36 check valves, 26 (items 1 thru 5) are ECCS/RCS interface valves ,
4 are accumulator valves and 6 are pump discharge valves. There are 12 of the interface check valves whose discharge side communicates directly with the reactor coolant system. All but 1 of the remaining interface check valves and the accumulator check valves have one normally closed check valve between it and the reactor coolant system. If the first check valve is leaking the second check valve experiences the RCS pressure load. Hence, the Seabrook ECCS check valve configuration has a total of l -
1 i
10 L
. ~
. - . - . . - . - ~ . - . _ _ . - . . . . -
'" l'2 + 14 - 1 + 4 = 29 ch:ck valvss that are in tha sc=a ecnfiguration es -
the interfacing LOCA check valves being analyzed . namely communica' ting with the RCS or in series next to one that does . Therefore a fraction of-29/36 = .80 of the ECCS valves at Seabrook are directly applicable to the situation being modeled in the interfacing LOCA analysis. This fraction j should be representative of PWRs in general. Hence, thru 12/31/86 we l have experienced (.80)(1.3 x 108 hrs) = 1 x 108 hrs of interfacing check l valve experience without experiencing a disc rupture. This neglects to l consider the roughly 3 x 10' hrs of experience associated with ECCS valves ,
' that had been included in the valve count of PLC 0432. While arrived at in a different way, the 1 x108 hrs used in PLC 0432 seems to be well sup-ported in view of additional experience that has accumulated since 11/30/84 and in consideration of some ECCS valves that possibly could be excluded as not interfacing with the ECS.
If one conservatively equates a disc rupture with a leak in excess of 1800 gym, we can compare the abcre experience with the PLG-0432 and BNL analyses of check valve failure frequencies. If we had in fact experienced a check valve leak or rupture having a size of 1800gpe, or nine times larger than the largest observed, and followed the same methodology as in the PLG and BNL analyses but with the above exposure time, then the median failure frequency at 1800gpa would have been estimated as (1/1 x 108 hrs) * (8766 hrs / year) = 8.8 x 10-5/ year. The fact of the matter is we did not observe such a leak or rupture, and therefore 8.8 x 10-5/ year is not a median but an upperbound. This result tends to support the analysis in PLG-0432 and strongly indicates that the accumulator valve only approach of BNL overestimates this frequency. . The median of the BNL analysis without experiencing a rupture is greater than this analysis would indicate even if a 1800 gpa rupture is assumed. .
-l It needs to be emphasized that the check valve failure rate estimates in PLG-0432 were assessed as having high degree of uncertainty. At 1800 gpm, the PLG-0432 estimates of failure frequency ranges from 1.4 x 10-6 at thq 5th percentile to 2.8 x 10-4 at the 95th percentile. The PLG-0432 upper bound is more than 3 times higher than the upper bound estimate obtained directly from the above analysis of exposure hours. The assessment of uncertainity in PLG-0432 took into account the sparcity of the data and the extra uncertainity associated with extrapolating the observed udta to larger leak sizes. In view of this, we do not feel it is valid for BNL to simply scale up this distribution using the same legnormal range factors,
. after reassessing the central tendency.of the distribution.
Another perspective on the reasonableness of the PLG-0432 check valve failure frequency estimate can be gained upon comparison of the 1800 gpm values of the failure frequency with the failure rate used in the SSPSA V-sequence analysis. That failure rate, in turn was originally developed in the Indian Point PSA and is compared with the PLC-04'32 and BNL failure races for 150 and 1800 gpa in Figure 2-1. The SSPSA distribution was developed using a different methodology and interfacing ECCS check valve data at 46 plants from start of commercial operation thru 12/31/82.
That analysis observed no disc ruptures in 1.15 x 107 valve hrs of oper-11
I ', I ._ _ _ . - . . . ~ _ __. _. ._
t s.
'.?-
ation. The methodology that was used to generate the SSPSA distribution was to break up the data into separate parts by plant as shown in Table 2-
- 1. Then, a plant-to plant variability distribution was developed using the first stage of the 2-stage Bayesean update approach described.in Reference (1). Because of the limited amount of data and zero f ailures, the SSPSA results are heavily influenced by the initial prior assumed in the application of Bayes Theores. In this methodology a unifora distribution is assumed over a grid made up of assumed possible discreet values of4and r, the logarithmic mean and standard deviation of the lognormal distribution. The latter distribution is used to characterine plant-to plant variability in the f ailure frequency. While the SSPSA analysis was only weakly influenced by the amount of evidence available, it is of interest to note it's excellent agreement with the PLG-0432 re-suits at 1800 gps. The SSPSA distribution brackets both ends of the PLG-0432 distribution.
A final check on the reasonableness of the PLG and BNL analyses is ob-tained by using the BNL results at 1800gpa in the SSPSA V-Sequence method-ology. In the SSPSA, no credit was taken for piping integrity or operator actions to prevent selt_o,e isolate the leak. If we use the BNL results at ,
1800gpa, recompute the y_I, and VS, initiator models and use the SSPSA plant model assumption, a mean core melt frequency, Ref (1), resulting from interf acing LOCA on the order of 2 x 10-5 per reactor-year is obtained.
This is comparable to the frequency of the dominant core seit sequences currently assessed for Seabrook. This is not a reasonable result, even without taking credit for piping integrity and; operator actions. ..
In suasary, we have the following response:
o BNL has provided a more complete survey of check valve leakage events for accumulator check valves.
o The BNL analysis of failure frequency using aceuaulator valve data provides unreasonably high values at 1800 gpa or greater leak rates for interf acing check valves.
o The BNL analysis does not take into account the most relevant evidence of all the absence of disc ruptures in check valves that interf ace between the ECCS and RCS in 500 reactor years of PWR experience.
o The PLC-0432 uncertainty distribution at 1800gpa is in ex-cellent agreement with the SSPSA disc rupture distribution that was developed using a different methodology.
o The BNL results at 1800 spa would produce unreasonable results for core damage frequency if no credit were taken for piping integrity or operator recovery actions.
Reference (1) S. Kaplan, "On a ' Two Stage' Bayesian Procedure for Determining Failure Rates from Experiental Data, " IEEE ' Transactions.
on Power Apparatus and Systems, Vol. PAS-102 No.1 January 1983.
12
- . . _ , , _ . . _ . . . . ~ .- . - ~ . . .
1 .- s -
4 i
.o Th3 FLC-0432 recults et 1800 gpa including the uncertainty j
distributions still represent the best statement of the possible !
range of values of check valve ruptures that freely account for -
applicable statistical evidence.
2.2 Check valve $pul'ation
'BNL comments that " to estimate the total number of check valve hours, .we used - the. total population of check valves at o the interfaces. This resulted in substantial overestimation of check valve hours".
It is correct that the total ECCS check valve population was used. However, as noted in Section 2.1 above most of these
. valves would be classified as interfacing check valves when defined as those between the RCS and ECCS, or one check valve removed from these valves. At Seabrook, 80% of the ECCS valves are " interfacing". As noted in Section 2 1 above the PLG failure rate distribution at 1800 spa is in excellent agreement with the observation of no disc ruptures in 500 L reactor years of PWR experience with about 108 hrs of interfacing valve experience.
2.3 Estimating Leak Rates
- .BNL states that "when estimating le.ak rate for accumulator 1
check valves from acctmulator inleakages, it must be recognized that the reduced leak races relate to two check valves in series, rather than leakage through a
- single check valve".;
It is agreed that there is, in principle, an element of uncertainty in estimation of leak rates associated with the ,
, . possibility that there may be multiple leakage between two f
series check valves. If this is in fact the case, as noted
- in the comment, what would be observed would be the limiting (smaller) of the leakages. However, unless the rate of leakage is very high, it would be unrealistic to expect this probles
, to impact the results in a statistically significant way. For
, independent events, the probability of observing mulitple check i valve leakages is vanishingly small. Even considering common cause failures, even a very high value of the beta factor of
- f .10 would lead to a high confidence that any particular leakage is coming from a single valve.
- l. It also should be noted that the leak rates in PLG-0432 were assigned very conservatively. We b'elieve this conservatism
, more than offsets the effects of rare multip1'e valve
- leakage.
l 13 l
L .
I
- 4,*
It should be noted that each individual check. valve'in the ECCS, . ~
particularly- the accumulator and interfacing valves must be periodically and independently tested. - Hence if there were multiple leakages, both would eventually. be detected.
2.4 Leakage Frequencies
'BNL states that " the leak failure frequencies versus leak rate curve presented in the study (reproduced in -Figure 2.1) is only a first approx-inacion for a more precise leak failure frequency versus relative leak rate curve. In particular, this curve pooled data involving a variety of check valve sizes. A more sophisticated treatment would require knowledge of'the size population of check valves at the interfacing pathways". -
As noted in 2.1 above most of the ECCS valves are interfacing valves.
2.5 Linear Extrapolation BNL states that "the largest leak rate in Figure 2.1 is of the order of 200 gpa, whereas the arena of interest ranges to 65,000 gpa. The " linear" extrapolation to higher races, is not necessarily j ustified . If the shape of the distribution is Pareto, the linear extrapolation is in order. However, if it follows s' Rayleigh distri-bution, the extrapolation is not correct (but conservative). Seabrook specific consideration (valve sizes, designs) are not.made in the analysis". .
- The results of the V-sequence analyses in PLG-0465 do not depend at all on-values of the curve beyond 1800 gpa. The plant respons'e to val'e v ruptures at a greater than 1800 gpa was assumed to be equivalent, i.e., the RHR system pressurizes to 2250 psia. Hence values on the curve beyond 1800 gpa'have no impact. Overall we have extrapolated the curve based on 2 decades of leak sizes by about I decade.
It is also important to note that in RMEPS, the analysis did not hinge
! .on a point value of valve failure frequency. Very large ranges of un-certainty were assigned including a specific allowance for the added un-certainty associated with extrapolation. -
We also disagree with the comment that design specific features of Seabrook were ignored in the analysis. There are three ways in which specific features can be incorporated into a PRA type system analysis. These include:
o use of design and procedure specific model of the system failure modes via the boolean equation, block diagram, or fault tree o use of plant specific failure data o screening of generic event data for applicabilty to the plant in light of plant specific features.
14
, 3** ,
Tha first end third cf th:co wara incorporstsd into tha Sasbreck analysis. The second is impossible for a new plant. . Hence' ve fi nd )
no justification for this comment.
2.6 Leak' Tests BNL states that "the initiator models implicity assume that the leak test of the valves " discover" all failures and valves behave as new af ter each test. The study does not describe the relevant test processes and the expected "real" efficiency of these test".
The initiator models are state of the art models and are consistent '
with the way which testing was treated in WASE-1400 and NUREG-0396.
Also, the addition of the d term tends to pickup previously unde-tected failures. The Seabrook test procedures were reviewed very carefully to justify credit for testing. Even so, the testing intervals are conservatively assigned (one year for MOV's,1 1/2 years for CV's) in view of the fact that procedures call for testing of the interf acing valves each time the plant goes to cold shutdown.
It is very conservative to assume Seabrook will only go to cold shutdown once a year.
2.7 Common cause BNL states that "the report does not consider common cause f ailures.
Such failures indeed happen due to boron deposition, improper maintenance
. such as installation of improper components (gaskets, seats, or valve disks) which may f ail almost immediately or at a later time".
It is .true' that the initiator models.did not include a term explicitly identified as accounting for common cause f ailures. However, 'such a term is not really necessary for the type of component configuration encountered here. In the term 23d, the) represents a random failure of the first valve which can be associated with a normally operating component for selection of the appropriate model. The 3 d term represents the demand failure of the second valve which can be likened with a standby component. Anytime you put together a combination of an operating and standby component, any f ailures of both components will occur at the same time because of the way the system is designed, regardless of whether the failures are due to in-dependent or common causes. Hence it is not necessary to introduce a separate term for ecamon cause failures. It is only necessary that the value selected for3 d accounts for both types of causes. We believe that the value selected for dd in PLG-0465 not only accounts for common causes but does so in a conservative manner. Also, the value configuration of the interfacing valves must be tested to confirm initial leak free seat-ing prior to each plant startup. Hance, boron deposition while it may occur is not applicable to the initiation of subs,equent leakage or rupture. Also it is not clear how gasket and seat problems could create a disc rupture.
2.8 Plant Walkdown
{ BNL identified the following concerns:
i
- 1. Ability of RER pump leakage to be detected in the Control Room -
concern lies with vault compartmentation design with equipment vault sump not receiving leakage promptly thereby delaying level detection input in the Control Room.
15
. 4 , '. -
- 2. Ability of RHR pump relief discharge into the PRT to be distinguishable in the Control Room from the pressurizer relief and. safety valve relief discharge - concern with the latter relief and safety valve discharge tailpipe temperature.
1' The connecting door from the RHR pump room to the CBS pump room is not necessary to meet any SB design requirements and can be removed. This l will facilitate the flow of any water leakage from the RHR pump in the RHR pump room to the CBS pump room 'and thus the level indication.
Existi'ng instrumentation provides adequate capability to distinguish RFR
^
relief valve discharge from either PORV or pressurizer code safety dis-charge to the pressurizer relief tank during a V-sequence event.
The pressurizer PORVs and safety valves have redundant monitoring available, 1.e.; the tailpipe temperature monitoring, and the tailpipe acoustic monitoring. Pressurizer discharge to the PRT is characterized by high RCS pressure, position indication of the PORV's and PORV block valves, high tailpipe discharge temperature and increasing PRT level, pressure, and temperature.
2.9 Operator Actions The authors of PLG-04:2 anc the V-sequence operator actions analysis are unfamiliar with the TEEM approach to assessment o,f human reliability. We note that the reference provided for TEEM is dated some nine months after PLG-0432 was published. We find it inapp,ropriate"to criticize the PLG-0432 analysis for its failure to anticipate a subsequently published contribution to PRA methodology by BNL. With regard to the quantification of 3 different operator actions using the same number, we disagree that the numbers must be necessarily different. It was simply judged that the most appropriate value for each actior. in NUREG-1278 corresponded to the same value and that the key difference was the hardware contribution.
With regard to the need for revising ECA 1.2, Training has been incorp-orated. Changing procedure may require additional studies to ensure that optimal procedure guidance is available for the operators.
2.10 Independent Check of IDCOR Result PLG performed an independent calculation to confirm the IDCOR result that peak pressure in the RHR system would be limited to RCS pressure. This
{ is provided in Enclosure 1.
All conclusions that were reached regarding the limiting failure 0 modes were backed up by stress calculations most of which were not presented in the reports reviewed by BNL. Additi6nal stress calcul-
[
ations were performed to show that 50% of the HX material would L have to be removed to result in stress approaching yield at 2250 l psia. In addition the potential for unfilled RHR piping was subse-quently addressed, quantified and found to be insignificant. Do c u-mentation to support the above except for the dynamic pressure calculation was submitted to NRC/BNL shortly before the issurance of the BNL draft report as part of RAI 75.
16 ,
. - . - . - , , - - -._ L -
-- ._:w.. _ - . - _ - _ _ - - . . . - . _ - ,
3 -.- .,.
y'
- - f.11 ' Trartrant of Psel Scrubbing ,
' Assignment of the traditional unscrubbed V-sequence to release category S1W is more conservative than WASH-1400, where the V-sequence was assigned to release category PWR-2. This conservatism was not significant at the frequencies predicted in PLG-0465, but it is worth pointing out in the context of the increased frequency estimated by BNL.
BNL has concluded that source term mitigation by pool scrubbing in the V-sequence is justified at least at the level credited by WASF-1400 (DF=100) for sequences where the break location is under water. BNL raises the question of pool, subcooling and on the basis of our submittal accepts the conclusion that the pool is subcooled. Additional confirmation for the pool subcooling argument can be provided independent of the analysis submitted .
earlier. The release of radionuclides and hot gases into the pool'in the RHR vault occurs very slowly. Both the stirring action by the gases and the slow release assure that the bulk fluid is isothermal at 212 degrees F, with some cooler temperatures at the walls. If temperatures higher than 212 degrees would exist, free convection mixing would render the pool is o-
! thermal even in the absence of a stirring effect.
Given an isothermal pool temperature of 212 degreer, the fluid below the surface is subcooled due to the hydrostatic pressure of the fluid. Based on PLG-0432, the RHR pump seals would be submerged under 21 feet of water before the release of radionuclides into the vault begins. The table below gives the subcooling between the pump seal and the surface for a range of pool depths. ,
Pool Depth Above Pool Subcooling (deg F)
Seals (feet) At SeaF Elevation P. col Average 21 24.6 11.6 15 18.4 8.8 10 12.8 6.2 5 6.8 3.4 0 0 0 This effect was neglected in the analysis submitted previously and it provides a more significant degree of subcooling particularly near the RHR pump seals where the most significant pool scrubbing occurs.
17
']
,L ...
, l l l Plant No. of Exposure Plant No. of Exposure Occurrences Time (Hours) Occurrences Time (Hours) 1 0 2.2 (5) 24 0 2.5 (5) 2 0 -
6.0 (4) 25 0 4.1 (5) 3 0 1.5 (5) 26 0 3.8 (5) 4 0 2.4 (5) 27 0 "-
2.9 (5) 5 0 1.8 (5) 28 0 2.7 (5) 6 0 4.2 (5) 29 0 1.8 (5) 7 0 . 2.3 (5) 30 0 2.9 (5) 8 0- 1.6 (5) 31 0 2.1 (5) 9 0 9.8 (4) 32 0 2.9 (4) 10 0 1.5 (5) 33 0 3.5 (5) 11 0 3.1 (4) 34 0 3.2 (4) 12 0 2.9 (5) 35 0 1.0 (4)
.13 0 2.5 (5) 36 0 1.8 (5) 14 0 7.5 (5) 37 0 3.3 (5)
!$ 0 2.9 (5) 38 0 3.2 (5) 16 0 2.7 (5) 39 0 1.6 (5) 17 0 . 1.3 (4) '
40 0 2.8 (5) 18 0 1.9 (5) *
. 41 0 .. 3.9 (5) 19 0 1.8 (5) 42 0 -
3.9 (5) 20 0 6.1 (4) 43 0 5.9 (5) 21 0 2.6 (5) 44 0 4.7 (5) 22 0 2.3 (5) 45 0 4.3 (5) 23 0 2.3 (5) 46 0 3.0 (5)
Toca! 0 1.1 (7)
Table 2-1 Plant Specific Osca Used in IPPSA and SSPSA Check Valve Rupture Frequency Analysi s .
18
-0 -4 ~
-2 10 10-5 10 10 10 I I I I I 5 th 8 th g median 3 mean{ 95 RMEPS (> l50 gpm)
BNL (> 150 gpm)
RECIPROCAL OF INTERFACING CHECK VALVE EXPERIENCE IN PWRS WITHOUT FAILURES RMEPS (> 1800 gpm)
BNL (> 1800 gpm) 55PSA*GR055 DISK RUPTURE I i I 4 .
f 10
~
10
-5 10
~ ~ ~
10 10 FREQUENCY ( events per reactor year )
l l
FIGURE 2-1 l COMPARISON OF CHECK YALVE FAILURE l FREQUENCY UNCERTAINTY DISTRIBUTION I
19
. - _ . - . . , - . - - , . - "_,,_,_.,.,_.,_,,,,,,,,,.,.n.,_
.. i3.0 SHUTDOWN EVENTS )
BNL provides comments ona ' ccidents during shutdown and refu'eling conditions in Section 2.2 of their report. They note that their
' review of .the information provided on this subject had not been completed. NHY responses to the comments provided thus far are pre-sented below.
3.1 Auto ' Closure Logic 'will ' Isolate 'Both Suction Paths '
(Single Pressure Transmitter)
BNL commented that "....another difference between Zion and Seabrook is that Zion his a single drop line and Seabrook has double drop li nes . This is not expected to be a significant difference because the auto closure logic at Seabrook will isolate both suction lines when a spurious signal is generated, i.e., a single pressure trans-mitter provides input to the interlock logic of the two inner isola-tion valves, and a separate pressure transmitter ,provides input to the interlock of both isolation valves...."
Per procedure, once the drop leg isolation valves are open, one valve in each line, associated with a DIFFERENT protection channel is tagged in the open position. This prevents a single failure of one protection channel instrument, interlock or a spurious action of the auto-close system from isolating both trains.
3.2 Credit for Two'RHR'Suetion Paths BNL made the following comment on the credit that was taken for Seabrook having two RHR suction paths "....this reduced the f requency of loss of RHR 'by a multiplicative factor of 0.145 and the do,r'e damage frequency by approximately a factor of 2. From the information available to BNL, it is not clear that this reduction is justified...."
When the RHR System is aligned for RCS overpressurization protection, two of the four (one on each drop leg) drop leg valves are open and power removed. This precludes a single f ailure from isolating both RHR suction lines.
When the RHR System is aligned for normal DHR, all of the drop leg isolation valves associated with the operating RRR train are open j and power removed.
b Therefore, no matter what f ailure is postulated, there will always
- be one RHR suction safety valve available.
The procedures for removing power f rom 2 of the 4 drop line MOV's preclude the spurious closure of valves in both lines at the same time due to the kind of spurious events that led to the experienced loss of DHR events. Therefore either mulitple independent or common j' cause failures must be postulated as was assumed in the derivation
, of the Seabrook reduction factor in the response to RAI 21. With res pe ct to the point raised by BNL that there would normally be one l
20
-- , v . _ . _ _ _ _ _ . _ _ . . . , , , ,
s -o
- 1 eperating RHR train and the other_in standby or.asintenance, the unfor-tuitous event in which a single drop line is isolated and the other train happens to be in maintenance was' explicitly accounted for in the derivation cf the reduction factors. - If a single drop line becomes isolated in the i
- normally running RHR train, there would be a high chance of recovery to No realign the available standby train and/or reopen the drop ~1ine.
credit was taken for RHR recovery in the Seabrook analysis of sequences
- borrowed from NSAC-84. However, if the drop line -on the operating _ RHR l
train were to close, there would be 3 different alarms in the control rton as noted.in our response to RAI 40. Because the reduction factor is cuch larger than typical non-recovery factors this scenario does not
' challenge the credit taken for 2 drop lines. Therefore, we maintain our pssition that the reduction factor is j ustified.
As noted by BNL, operator failure to restore the standby train was not explicitly included in the Seabrook analysis. To include such errors in the model, the correction factors presented in the response to RAI 21
- can be re-expressed as follows.
1 For spurious valve closure, the correction factor becomes.
R j3 MOV + (1 -pMOV) (.5) [0RHRM + (1 - ORHRM) O}
where gMOV=HOVcommoncausefailurefractionQRHRH = RHR tr shutdown OR = frequency of operator failure to restore RFR cooling The correction factor for errors in inverter swit.ching become's .
.5 [QRHRH + (I ~ QRHRM) ORI Table 2.7 in the BNL report lists 130 events involving loss of DHR, all which were terafnated by successful operator actionIntherefore the generic comparison to Seabrook frequency of non-recovery must be less than 10-2 most plants probably do not have the same alarm The capability NSAC-84 plant to alert does not.
i the operators to a loss i
Therefore a value of 10~gf for event RHROR pump suct on.is very conservative for Seabrook.
i But even if we assume this value, the spurious valve closure correction factor would only increase from .072 to .076 and the inverter switching Hence the effect is insignificant, even when s factor from .031 to .035.
- conservative recovery factor is used.
j 3.3 Loss of DHR BNL states that dominant causes of loss of DFR are; i
- ' 1. Spurious closure of RHR suction valves and,
- 2. Inadequate vessel level (RHR cavitation) i Addressing the first concern, let us look at one such scenerio for Seabrook, loss of DHR due to a failure high of a wide range RCS pressure transmitter. (Note: this is for example purposes only, the loss l
l could-occur for any reason listed in the BNL report, and the result l
would be the same).
)
' 21 ,
- - . . . . - - - - . . . - - - . . - - - - . .-- - =
- * ' 1So LTOP prettetica system ct. S:cbecek is cuch loco d2 pendant 'en humen .
_ j
, error, (due to automatic arming), and -the backup protection is afforded us again by the use of dual drop legs for the hot leg to RHR cooling system lines. Let us further review a " typical" overpressurization
. event. Since this type of event has it's highest probability of happening when the RCS is " water solid", we must make some basic assump-tions:.
Assumptions:
1 (1) Per operating procedure, the RCS isn't taken solid until Mode 5,
} (<180 degrees RCS).
(2) Both trains of RHR are aligned, one is in service. This includes, per procedure, that one of the two RHR suction valves in EACH RHR drop. leg is tagged open, power removed (the tagged valves are on opposite ELECTRICAL / PROTECTION trains) . This.is done to protect the system from an instrument failure, or single human error, isolating BOTH RHR suctions (and consequently, both suction relief valves) from the RCS.
(3) All equipment required by technical specifications to be made inoper-j able, has been made inoperable per it's applicable procedure.
Let us assume, at this point, with the RCS at <180 degrees F and water solid, the operating RER loop fails, due to a wide range RCS .
pressure transmitter failing high.
SEQUENCE OF EVENTS: .
- 1. Inadvertent closure signal occurs.
- 2. Operating loop RHR suction valve closes, non-operating loop suction valve doesn't close because it is tagged open per procedure (see initial condition #2 above).
- 3. Operator receives VAS alarms on low RER flow, low-low RNR flow and
- overhead alarm for low RER flow.
- 4. RCS pressure increases rapidly due to loss of letdown with continued j charging.
- 5. LTOP automatically actuates at 541 psig, this is independent of operator action, LTOP at Seabrook AUTOMATICALLY arms based on auction-eered RCS temperature and wide range pressure. .
- 6. If pressure reaches 450 psig, the non-operating RHR loop's suction
[ safety valve will open. For pressure to get to this point, some malfunction of LTOP would have to take place. This is due to only l
f 22 i
P
ONE charging _ pump allowed operable in' this stodo. Th3 RHR sofpty valvos lif t at 450 psig, _ passing over 900 gpa. The runout, of one centrifugal charging pump is approximately 550 spa. This 550 gpa is measured thru the safety injection flowpath, the operable charging pump is aligned thru the _ NORMAL charging path with it's flow ~ control valve in MANUAL. Although there is no procedural requirement for the flow control valve (FCV-121) to be in manual, it is highly unlikely it will be in automatic. This is because the FCV-121 has an electrical stop built into it that ensures RCP seal flow is provided in the event of a high pressurizer level at OPERATING pressure. This' stop position, with the RCS depressurized, would provide far to much charging flow to be balanced by letdown.
All of this results in substantially less charging flow than runout, and in any case, less:than the capacity of ONE RHR suction relief valve .
Regarding the second dominant sequence, the Seabrook operating staf f have provided the operator with procedure guidance for inadequate vessel level. ~The dominant sequence for Zion (local operator leaves manual vessel drain valve open during drain down operations) is less likely here because the drain down is:-
- 1. Monitored in the Control Room on RVLIS and refueling level indications.
- 2. Capable of being termi,nated from the Control Room without the utilization of a field operator.
As evidenced by BNLs report, out of the 130 industry events, all have been terminated by. timely operator action. Although some of the recovery actions have lasted longer than an hour (number not given by BNL),
- - the preceding statement of a loss of DHR 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> after a reactor trip while partially _ drained' causing core u'ncovery is ' difficult to underst'and. Altho ugh it is within the RHR/RCS system capabilities, and within technical specifi-cations to be in Mode 5 in 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br />, it is highly unlikely that the vessel could be drained down that fast with the reactor head still installed.
The reactor head, per technical specifications, cannot be removed for 100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> af ter last critically (i.e. ,100 hours0.00116 days <br />0.0278 hours <br />1.653439e-4 weeks <br />3.805e-5 months <br /> from trip) . Even if it were possible to d~ rain down the RCS under these conditions with the head in place, why would the core uncovet? Assuming a " typical" loss of DER, the
. RNR suction valves e, losing, there would be no mass loss, in fact, the RCS mass would increase due to the loss of RHR letdown and continued charging from the operating charging pump.
3.4 Relief Valves Ineffective /2 PORV's or Both Relief BNL states "....in addition to the PORV's the relist valves in the RPR system may be unavailable to relieve pressure. Each RPR suction line has a relief valve with 900 gpa capacity at 450 psig. However, these relief valves may be made ineffective if the RHR suction valves close automatically when the setpoint of 660 psig is reached, as was the case in the Turkey Point 4 Events. Actually, the Seabrook technical specifications only require either both PORV's of both RHR relief valves to be available".
Seabrook features reduce the frequency of occurence of this LTOP event at Seabrook.
23
.-m_.._ __ . m ~ - - - - -- --- --
.1 , ,.
- The RHR suction isolation valves are opened and power removed 'to the " Cross Train" RHR suction isolation valves during the time' that the RHR system is
[ in operation. This pecvides assurance that the failure of a LTOPS pressure transmitter will not isolate both trains of the RHR system thus blocking i
access to the RHR suction relief valves.
The autoclosure setpoint for the Seabrook Station RHR suction isolation valves is 660 psig. This is above the RHR suction relief valve set n
pressure and the total accumulation during relief valve lif t. The Turkey Point Event overpressure source was the charging pumps. The Seabrook Station RHR suction relief valves are sized to pass the flow from two L
S operating charging' pumps at the relief valve set pressure. Two charging pumps at the Seabrook Station are inoperable by Technical Specifications.
Thus,' there is added assurance for Seabrook that the RER system will not q be isolatedthe terminate and that the RHR suction relief valve will be available to transient.
I li 3.5 Applicability of LOCA's A
BNL commented that "NSAC-52 reviewed operation experience within 5 calendar years up to the end of 1981, and identified 10 loss of coolant events at PWRs. They were caused by the follwing causes:
p 1.
I Inadvertent manual initiation of RRRs supplied containment spray.
- 2. Inadvertent loss of inventory to the containment building sump i
and/or automatic initiation of recirculation mode of low pressure
,i safety injection.
M 3. Inadverent loss of inventory via the RHRs relief valves. *
- 4. Inadverent loss of inventory via aispositioned crossconnect i; or drain valves.
[ 5. RHRs valve packing gland removal during plant pressurization, p
1 dislodging the valve packing gland.
I
- 6. Gross valve packing leak".
t With regard to item 1, the Seabrook Station design is not the RHR supplied containment spray type. Multiple mispositioned valve alignments (motor and manual valves) would have to occur for these sequences at Seabrook. This would occur only by misaligning the refueling canal drain line in the discharge of the RHR pumps to the succion of the containment building spray pumps and subsequently to the containment building spray healer.
i O
24
Wme~ -e-.
. u - a m ow.~.. . ; ~. :
0'
- F:r ites 2,1:s3 of inv:ntary to tha conceinannt building sump cs a -
result of sump isolation valve mispositioning, testing, or automatic initiation of the recirculation mode of low pressure safety injection
- would require the failure'of a check valve to close. Thus the check valve in the Seabrook' Station RRR pump suction line from the sump pre-L vents. loss of inventory to the' sump if the sump isolation valve is inadvertently opened during RHRs operation.
The last .four are applicable to the Seabrook Station system designs.
- 3.6 LTOP/ Turkey Point Events BNL commented that "....The NSAC-84 analysis of low temperature overpress-urization may be too optimistic. Events such as those Turkey Point 4
. indicate that the frequency with which a rapid pressurization occurs with
'the RER system isolated and the PORV's unavailable is higher than 10-3 per year. The operators have only a few minutes to respond to the event" The IE notice referenced in the BNL Report, IE notice 82-17, Overpressur-iration 'of Reactor Coolant Systes, describes two avunts at Turkey Point and an inoperable LTOP system at North Anna. The North Anna event is not applicable to Seabrook since Seabrook uses solenoid PORVs rather than air with backup nitrogen accumulators.
The two Turkey Point Events involve failures in the operable LTOPS while the redundant circuit was out of service for cali' ation. In the first event a pressure transmitter isolation valve was round closed. In the second event a summator in the actuation circuitry failed, furthermore the summator was not included in the instrument channel surveillance. In both occurences the letdown via the RHR system was automatically isolated'by automatic closure of the RHR suction isolation" valve. Both events dere tensinated within two minutes by manual operator action after the RCS pressure exceeded the allowable values in the plant technical specifi-cations.
Two additional Seabrook features can reduce the frequency of occurence of this LTOP event for Seabrook.
The RNR-suction isolation valves are opened and power removed during the time that the RNR system is in operation. This provides assurance that the failure of a LTOP pressure transmitter will not isolate both trains of the RHR system thus blocking access to the RFR suction relief valves.
The autoclosure setpoint for the Seabrook Station RER suction isolation valves is 660 psig. This is above the RHR suction relief valves set pressure and the total accumulation during relief valve lif t. The Turkey Point Event overpressure source was the charging pumps The Seabrook Station RHR suction relief valves are sized to pass the flow from two operating charging pump at the relief valve set pressure. Only one is allowed to be operable for this mode of operation. Thus there is added assurance for Seabrook that the RHR system will not be isolated and that the RHR suction relief valve will be available to terminate the transient.
25
3.7_ A sessment of Consequences -
In our response to RAI 21, a conservative approach was followed to char-acterize consequences of loss 'of shutdown cooling events by assigning the resultant core damage sequences directly to release categories for power operation events. If a core damage scenario were to develop during shut-down, the scenario would evolve more slowly for a given configuration due to a lower value of af ter-heat and the source term would be reduced by r'adioactive decay over the period between the time of power reduction and shutdown to the time o,f release which could be days to weeks. The accident probability calculations in the Seabrook assessment used Zion data for estimating the frequency and length of outages. The length of the outages corresponds to the possible range of times to initiation of a core danage scenario as measured from the time of plant shutdown. The Zion data
, consisted of 12 refueling outages of average duration of 1,992 hrs. and i 49 maintenance outages of average duration of 488 hours0.00565 days <br />0.136 hours <br />8.068783e-4 weeks <br />1.85684e-4 months <br />.
Given a system f ails'over a period of time, the conditional distribution of the time of failure is roughly uniform. Hence the mean time of scenario initiation would be 12
- 1992 + 49
- 488 = 392 hrs.
(12 + 49)
- 2 If it is argued that a shorter time should be used then the accident pro . ,
abilities would be correspondingly lower because of a shorter time the RHR would be."at risk".to' fail. ,
Since the response to RAI 21 was prepared some additional CRACIT runs intre performed for release categories S2 and 56 with the release times delayed to simulate the delay to onset of the loss of RRR scenarios. The approach taken was to simply translate the release times for all release puffs by 48 hours5.555556e-4 days <br />0.0133 hours <br />7.936508e-5 weeks <br />1.8264e-5 months <br />. Based on the Zion data, about 94% of the scenarios would have longer delay times. By performing this translation, the radioactive decay of the source term is accounted for, but ' the slower evolution of
~
release due to lower af ter heat is not. Hence, these delayed release calculations are still viewed as conservative a,1though they are somewhat g
. less conservative than provided in the RAI 21 response.
The revised CRACIT runs were used to provide an enhanced assessment of the contribution of shutdown loss of cooling events to 200 rem and 50 rem dose vs. distance curves. The new curves are presented in Figure 3-1 and 3-2 for the realistic and conservative treatment of equipment hatch closure. These curves supercede Figures 14 and 15 in response RAI
- 21. Also, Figures 14 and 15 in response RAI 21 were plotted conservatively
- .too high. For example the 200 rem curve in Figure 14 should have been
'offscale.
26
- = _ _ _ _ _ _ _ - - - -
1 1
3 * * * * * * **I .
. . ....al a u, .
NUREQ.03gg -
N . = = = = = SENETlWIY STUOY -
QM MM M ATPewtM
......... swumo Lg ,'
(ze m -
j$ aFFr=4) j 0.1 ,
= .
2 -
~'% .
=d - \ _
$j \ \ so nnu gg s \ -
wo N \
5w \
it!$ 0 01 7 i I --
w .
h \A% \ .
- lW \g .
!- g I
\.
l .
u8 i
- i t
--q ,,) zw am O.Co l I Nh, ,,el . i e . . . .,i . , , , . . , ,
i 10 100 1 C00 CISTANCE(MILES)
Figure 3-1 COMPARISON OF 200 REM AND 50 REM 00SE VERSUS OlSTANCE CURVES WITH CCNTRIBUTIONS FRCM SHUTDOWN EVENTS WifA Tid #4 /s(egges
& % 9 9 firs, 27 7
-I.. .
, .. _ . . . . . . - . . . . - = - - - -
I . . . .
_" - NUMEG 0390
SENstrM7YSTUOY
~
~ FCM EWNTS ATPOWER ee eemesee '
SHUTDOWN M g> WITH NO CREDff f ,
- FOR ECulptesNT ~
HATCH RSCOVEfff N
j 0.1 .-
j,< .= .
.~'%
= .- \.
\ \~;r.. a SoREM
% N\
~
j5 m
gg 0.01 -
I SO W --
k ."500 ~
i< .
=h 4ml l g. . p \q. .
u8 .
is
- i. I-\
- l. 200 AEM 0.00 t I I' ' ' ' ' ' 'I ' ' ' '
I 10 100 ' ' ' .'c00 1
OISTANCE t M11.EB Figure 3-2 COMPARISON OF 200 REM AND 50 REM 00SE WASUS DISTANCE CURWS FOR CONSERVATiW EQUIPMENTHATCH Aa80 ASSUMPT10N ShastManabt YM44SCS OF NO del? CRE0liFO as Mrs. 1, -
28 21 .
. . . . - . _ _ . . . . _ ~ , . . _ . _ _ . _ _ _ _ _
~
' 4 .0 STEAM GENERATOR TUBE RUPTURE ,
BNL's comments on the steam generator tube rupture scenario is pro-vided in BNL Section 2.3 and an enclosure to the BNL Report.
1 Our detailed plant specific evaluation of steam generator tube creep failure (see RAI 47 resp)nse) indicates that this failure mode is
. very unlikely. There is no comperable evaluation that justifies the existence of this f ailure mode at Seabrook Station. In fact we are not aware of any analyses that shows tube f ailure occurs. Rega rdless ,
this is apparently a generic research issue and BNL evaluation conservatively assesses the f requency of this event without including probabilistic1y the Seabrook Analysis. If BNL uses their best '
estimate analysis consistent with WASE-1400 this issue becomes insignifi cant . Also, this issue will eventually be resolved as insignificant because of the many ways to reduce its f requency.
9 The frequency of this event is made up of 3 components:
F = fg
- fo
- fsc where :
F = the annual f requency of steam generator tube f ailure p- fE = the annual frequency of a high pressure core melt 1 with dry steam generators f o= the conditional frequency of operator failure to
, depressurized primary system given fE, fsc = the conditional probability that the Steam Generator
- tubes f ail given fg and fo The following is BNL's Pessimistic calculation of F. -
- F = 2.4 x 10-6 = 4 x 10 -5 (f e) * .2 (f o) * .3 (f sg)
If NRC accepts our evidence on Seabrook fSG, would be very small and the frequency would become very small (insignificant) .
It should be noted that even if fsc was to remain pessimistic, either fE Of f ocould be reduced significantly with plant enhancements.
Therefore, regardless of the final resolution, this issue will become an insignificant event.
Our additional comments are provided in Sections 4.1 and 4.2 below.
4 .1 Fluid Dynamics Experiments vs. Models l1
! Several comments suggest that the assumption of complete mixing in the fj steam generator inlet plenum is inappropriate. While dye observations in li the Westinghouse experiments suggested somewhat incomplete mixing, the i definitive temperature measurements indicate a high degree of mixing.
I Good mixing and moderate temperatures are expected near the tubesheet 29 i
"^
~
- .-. ,.,,: ~
,,_ . ._ . . _ . -- - --- - ~ ~ ' ~~
s.
,' . due to ths satrainmant cf coldar fluid by tha hbt jet af ter it exits the top of the hot leg. In view of the large margins which exist in the calculations (about 7000F- in the best-estimate case), an assumption of even very poor mixing would not change the conclusions with respect to steam generator tube integrity. .
It is suggested that the Westinghouse scale model data is preliminary and does not offer a good test of the model. While it is true that high pressure sulfur hexafluoride experiments have not yet been conducted, data spanning a wide range in densities, viscosities, and coefficients of volumetric expansion have been completed. The model predictions agree well with this data. Given the ranges of parameters tested, no new phenomena would be anticipated in'the high pressure experiments which would change this agreement.,
The reviewer commented that one or more steam generators could be depressur-ized, which would increase the stresst,s on the tubes and thus accelerate creep rupture. It should be noted that operator actions to depressurize the steam generators would occur only in the presence of adequate steam generator level, in which case failure of the steam generator tubes is not an issue. A MAAP case was run with the assumption that one of the steam generator relief valves stuck open. As shown on Figure 4-15 of the FAI report (RAI 47) , this negligibly altered steam generator inlet plenum gas temperatures compared to the base case. Based on the analysis in Appendix C (RAI 47) , some degradation in the secondary side heat transfer coefficient at the hottest portion of the tube surface would be expected at low pressure.
However, radiation (which was relatively small compared to convection at high pressures and was therefore neglected), would partially make up for this degradation. In any event, even the primary si't gas temperatures (let alone the tube temperatures) in this scena~i,o r were well below the levels required to threaten tube integrity.
The reviewer commented that the MAAP code, which was used for the analyses has not been verified and documentation was not available. As part of the IODOR program, an earlier version of MAAP was checked line-by-line by an outside reviewer and all comments were resolved. In addition, the MAAP code has been extensively discussed with the NRC and its contractors over several years as L . pa rt of the issue resolution process. Finally, MAAP has been benchmarked against a body of detailed code calculations, experiments, integral experiments and industrial experience including TMI-2 and LOFT FP-2.
L i'
l 30 i'
- ~.w,.-.- ._ .. . . . . ..n- , ._
4.2 Probability Assessments
.BNL assigns a probability of 0.2 to f ailure of the operator to depressurize the RCS prior to the occurrance of an induced steam generator tube rupture (ISGTR). In view of the long times predicted for tube heatup this seems to be an unreasonably conservative assessment.
The conditional probability of a small leak ISGTR given no repressurization of 0.3 appears very high in view of our analyses. This would mean there is c 30% chance that the results of the FAI analysis is in error by several hundred degrees centigrade, which we consider very unlikely. A more reasonable interpretation of the FAI analysis would assign a confidence level in the range of 0.95 to 0.99 for the conclusions to be substantially correct. The corresponding probability of an ISGTR then would be in the range of 0.01 to 0.05. This would also coincide more with'our assessment of multiple tube ruptures. The probability that several tubes fail given that one does fail could be fairly high if no depressurization occurs ef ter .the first tube has f ailed. In this case we would accept the value cf 0.01 for the probability of a large leak ISGTR given no depressurization.
However, given that the operator has not depressurized the RCS before the first tube f ails, we believe it must be more likely that he initiates depressurization after the first tube has failed. Whether depressurization
.at this time can prevent the f ailure of additional tubes will largely depend on the f ailure modes for the subsequent. tubes. If these tubes fail because of a domino effect then it is unlikely that depressurization would prevent further f ailures but it would still reduce the source term.
If each tube fails independently then depressurization would be likely to prevent further tube f ailures because of the slok tube heatup rate.* *0per-ctor action to fill the steam generator and flood the tube break soon' af ter core uncovery would also mitigate the source term substantially by a pool scrubbing ef fect similar to the V sequence. BNL assigns both the small 'eak ISGTR and the large leak ISGTR to release category S1W. This is a very conservative assignment at least for the small leak ISGTR, and it does not recognize the fact that the release characteristics of these two '
cases are significantly different. If this were not the case there would be no reason to distinguish the two. Release category S2W would best represent the release characteristics of a small ISGTR.
d 31
__.m _
1
e
- ~-
- 5.0 DIRECT HEATING ,
\
BNL's comments on the potential for early containment failure are
. provided in BNL Section 4 with references to BNL Section 3 regarding failure modes and their probability.
It should be pointed out that experiments performed by FAI conclude that this phenomena is not applicable to Seabrook when the configuration above the cavity is properly accounted.
The details of the BNL calculations for direct heating are not known. However, there appears to have been a misunderstanding with res pe ct to the interpretation of the conditional containment failure probability due to direct heating. We understood that the direct ,
heating peak pressures calculated for Zion were used by BNL to determine the conditional probability of containment failure for Seabrook from the Seabrook specific probability distribution for the containment f ailure pressure derived for the SSPSA (Figure 4-7 of PLG-0432). In other words we understood the conditional f ailure probabilty of 0.01 to be a Seabrook specific value but based on pressure transient estimates for the Zion plant. It appears from
. the BNL draf t report that the conditional f ailure probabilty of 0.01 is entirely a Zion specific value including the lower assessed '
strength of the Zion containment. The response provided previously was based on the Seabrook specific interpretation and it appears to
'have been very conservative, because it did not account for the most significant difference, namely the higher strength of the Seabrook containment which is supported by the BNL structural review.
The conservatism ~ in the assessment of direct heating ef fects is illustrated below in an analysis which 'is entirely based on the BNL assessment of the Seabrook containment pressure capacity. The most consistent approach to utilize the BNL assessment would be to regenerate the Seabrook specific containment f ailure probability distribution (Figure 4-7 in PLG-0432) using the BNL input. This was not possible within the time frame available. Instead, the following simplified approach was followed.
Under wet containment conditions, the lowest median f ailure pressure indicated by BNL for a type B or C failure is 167 psig for penetration X-8. BNL indicated that shear failure at the base could be an important f ailure mode but the finite element analysis indicated that this would not occur before the general state of yield is reached at 157 psig.
Note that on page 4-6 of the BNL report a value of 140 psig is reported for the shear failure mode which appears to be in error. Based on our previous analyses and BNL conclusion that failure will not occur below 157 psig (a lower bound) we assume that the median value for this failure mode would be higher than 167 psig. Therefore, the median failure pressure for type B or C failures is controlled by penetr& tion X-8 at 167 psig.
A statement of the uncertainties in the failure pressure can be derived l' from BNL's assessment of the hoop failure mode. BNL indicated a median failure presure of 175 psig with a lower bound at the general yield state which corresponds to a pressure of 157 psig. Interpreting this lower bound as the 5th percentile and assuming a lognormal distribution 32 i
. . . . - .__ _ ._ _ m . _ . _
s.
yields a standard deviation of 0.066. The overall co.ntainment f' allure pressure distribution based on the BNL assessment is then approximated by a lognormal distribution with a median value of 167 psig and a standard-deviation of 0.066. PLG believes that this is a conse rvative assessment for the lower tails of the distribution which are important in assessing the direct heating ef fects, because the same lower bound of 157 psig would apply to the failure mode of penetration X-8. Because of this lower median failure pressure, this combination would have resulted in a narrower distribution for the failure pressure and it would consequently have yielded an even lower containment failure probability due to direct heating than what is estimated below. Even though this would have been a more logical and consistent distribution, the more conservative repre-sentation was chosen to be sure that the larger uncertainties for the hoop failure and the shear failure mode are included.
The resulting distribution was plotted against the containment failure pro-bability distribution for the Zion containment taken from Figure 2.5.1-2 in the Zion PRA, which is raproduced in the attached Figure 5-1. The pressure where the Zion curve has a probabilty of 0.01 is 146 psia (131 psig). At this pressure the failure probability for the Seabrook contain-ment according to our interpretation of the BRL assessment is 0.00015, or a factor of 67 lower. This value compares very well with the value for early containment failure of 0.0001 which was used for release category S1 in the original SSPSA and which is still embodied in the analyses for the EPZ study (PLG-0432 and PLG-0465). We conclude from this approximate analysis that the direct heating effects are not changing any of the results or conclusions of the RMEPS study or the EPZ sensitivity study even when the BNL assessment of the Seabrook containment capacity and the NUREG-1150 data for direct heating are combined and 3ssigned a probability oT,'l.0.
This would indicate that these conclusions are reached even on the basis of a very conservative combination of all aspects of the analysis and assumptions raised by BNL.
33
. . . . _ _ ~ +- + - _. .,,, ,_
n (101 t
1 5 1 2 --! g /g .-----
108 .-
Ii C , , , , , , ,
.I s i ,
/
i i f
2 r
f
/
> . /
- 10-1 -
e .
< s ,
8 i f-l g I y 2 frSEA BROo/c *
~ ! / (BMED 0/V
- 0 ,0 '
BNL DATA) 2
~] l
( :l i
' /
f 2
/ /
10-3 .
/
S / j'
/ J
- / -
2 l X f,[X /0
...,,,,,,,,,,i 10 4 jl 180 140 150 160 170 130 PRESSURE (PSIA)
~
' j ' d~ / / / / $ zc g)g/g- 7 gi
~
Of'S/]/A'//fN/ $4]k9.') 64i fA2 DAj'n 0/d'G~C'/ HE^f/74f i
- Fu3 u.re. 5-I 34
_ _.__.s-_-.,-..._-..-_.-_. _ _ . ~ .. .,...,._..____-_..__m__...__________._____..__._.__
, _ . _ ~ __
-. - - -..a.- a . . _ . . . . ,.
4.*. .
- i .
. 6.0 CONTAINMENT STRUCTURE
- 6 .1 Independent Review .
An independent check of the containment strength calculations has .been completed. This review also includes additional documentation regarding assumptions made in the SMA analysis. The review is documented in. cal-culation number SBC-198.
6.2 Shear Failure at Wall-Basemat Intersection BNL investigated f,ailures at the above location and confirmed that " such a failure is not expected to occur for pressures .up to 157 psig". However, BNL states that "a shear f ailure mode at the base may develop at a pressure slightly above 157 psig". The basis for this statement is not providad in the report. l l
We have concluded that the above f ailure mode will not occur until internal pressure exceedes 300 psig, well above the 216 psig critical f ailure mode.
Our independent review, utilizing more conservative assumptions has concluded that shear f ailure in the concrete could occur at approximat ely 3 197 psig but that subsequent deformation would be limited by hoop and meridional reinforcement preventing liner tearing . Thus, this failure mode, while possibly initiating below 216 psig is not critical for cato-strophic failure.
6.3 Application 'of One Percent Strain as the Median Failure Pressure We disagree with the BNL judgment that one percent rebar strain,Jepre-seats the median ultimate capacity of t'he Seabrook containment. .
The median u11 mate capacity calculat ed by SMA (216 psig) corresponds to an average rebar strain of about 4.7% (S u = 109 ksi). These ultimate values are based on actual certified material test reports for the rebar material. A rebar strain of 1% represents loading conditions which pro-duce stress levels well below the median ultimate strength. In fact, 1% strain in the highly ductile (34% elongation) rebar material is typically located on the steeply ascending portion in the plastic region of the material stress-strain curve, well below the as-tested median ultimate strength of the Seabrook rebar material.
S e, d
I 35
, - . _ . . -: z =.z.-..:-..-. _ - _ _ _ - _ - - - - - . . - . - --
~ ~~ ~~
7Het.ostmE T
^
, , , hlCKARD, LOWE ANo GARRICK. lNC.
- *J .
7, ..
MEMORANDUM Date: August 22, 1986 Ref:
File: PSNH 1016.08 Copies: DRButtemer JWRead TO: K. N. Fleming /D C. B y DJWakefield FROM: M. V. Frank SHOCK WAVE E FtCT F LLOWING HYPOTHETICAL, CATASTROPHIC FAILURE SUBJECT OF ISOLATION VALVES SEPARATINGHIGH AND LOW PRESSURE PIPING IN THE RHR SYSTEM INTRODUCTION AND
SUMMARY
Reference 1 contains a characterization of the nature of the hydrodynamic shock effect following a postulated sudden catastrophic rupture of high to low pressure isolation valves in piping. Two situations were evaluated,'one with piping filled with water and one with voids in the water. The scenarios were evaluated in connection with an interfacing LOCA postulated to be initiated by a sudden, complete rupture of the discs in two series motor-operated valves in the RHR hot leg suction line, or an interfacing LOCA postulated to be initiated by a sudden, complete rupture of the discs of two series check valves in one of the RHR cold leg discharge lines. The situations were evaluated to determine if they could generate pressures that would seriously threaten the integrity of RHR system piping. The evaluation in Reference 1 appears to conclude that pressures higher than the reactior coolant system pressure at the time of the valve failure cannot be generated. Therefore, other failure mechanisms, such as RHR pump seal fa~ilure, are far more 1f kely. ,
I was intrigued enough by the argument to attempt an independent evaluation. My evaluation appears to verify the above conclusion.
Because of the similar design of the relevant plant features, this conclusion holds for both the Seabrook Station and the Diablo Canyon Station.
TECHNICAL EVALUATION Consider the situation shown in Figure 1. The pipe to the right of the valve (downstream) is initially filled with water of density pC.
temperature CT , and pressureCP . The volume to the left of the valve (upstream) is considered to be much larger than the downstream volume and is nearly filled with water of density pH, temperature T H, and pressure PH . Before the valve disappears, the following conditions apply:
TH>TC l pH < PC
~
Pg > PC V = 0; overall fluid velocity is zero both upstream and downstream.
'***~^-w ' * * ,o,.m~ .m_, e..._, _ , .
/ ,-' T0: Karl N. Fleming, Dennis C. Bley
,, , FROM: M. V. Frank -
SUBJECT:
SHOCK WAVE EFFECT FOLLOWING HYPOTHETICAL, CATASTROPHIC FAILURE OF ISOLATION VALVES SEPARATING HIGH AND LOW PRESSURE PIPING IN THE RHR SYSTEM DATE: August 22, 1986 PAGE: 2 FIGURE 1 CONCEPTUALIZATION OF SITUATION WATER g LEVEL. g RAREFACTION . COMPRESSION WAVE PROPAGATION j WAVE PROPAGATION TC'PC,PC UPSTREAM '
OOWNSTREAM SIDE 90' ELBOW SIDE POSTULATE THAT PH VALVE DISK
, INSTANTLY RUPTURES TH
~Tg > TC SYSTEM F1LLED WITH~ * * '
- WATER UP TO WATER -
!NITIAL Ph < PC LEVEL INITIAL STATE CONDITIONS PH>PC ATER RE GI EN THERE IS NO IN THE DIAGRAM
. FLOW; V = 0 .
O e
e 14129082286
_A______ _ _. , , - . , . _ , _ . _ _ _ . .
.' ._ y* . Karl N. Fleming, Dennis C. Bley TO:- .
FROM: M. Y. Frank' .
SUBJECT:
SHOCK WAVE EFFECT.F0LLOWING HYPOTHETICAL, CATASTROPHIC FAILURE .
OF ISOLATION VALVES SEPARATING HIGH AND LOW PRESSURE PIPING IN THE RHR SYSTEM DATE: August 22, 1986
-PAGE: 3 The sudden rupture of the valve causes a sudden pressure change, which manifests itself.as a pressure wave. The wave propagates at sonic velocity. A discontinuity in both pressure and fluid particle velocity exists-across the wave front. In fact, for the situation depicted in Figure 1, there are two wave fronts. A rarefaction wave propagates upstream, and a compression wave propagates downstream.
Reference 2 provides the following derivation of the velocity, pressure, and density relationship across the wave front itself.
The pressure and velocity changes across an acoustic wave in one dimensional flow can be derived from the momentum equation h+uh=-h (1) where u is the fluid velocity, P the pressure, x the. distance, and ge is the gravitational constant that is needed for consistent units when P ,
is expressed in pounds-force and p is expressed in pounds-mass. Since the velocity change with respect to digtance is relatively smal.i.in a
~
straight pipe, the Bu/3x term is negligible, and Equation (1) becomes au 9c 3P p ax (2) at The pressure wave travels at the velocity of sound a; therefore, it will travel .a distance (at) in time t. Let e = x t at (3) where the negative sign is required for a wave moving in the direction opposite to u. Substituting Equation (3) into Equation (2), one obtains i
(4) 2ah=-
which yields by integration .
P2-Pg " (5) 9Cu 2 - "1 The plus and minus signs result from the two opposing wave fronts. The terms P2-P1 and u2 - ul refer to the discontinuity of pressure and velocity across each wave front.
1412P082286
l y' TO:- Karl N. Fleming, 02nnis C. Bley M. V. Frank j ., , FROM: .
~' SUBJECTi . SHOCK WAVE EFFECT FOLLOWING HYPOTHETICAL, CATASTROPHIC FAILURE' 0F ISOLATION VALVES SEPARATING HIGH AND LOW PRESSURE' PIPING ~
IN THE RHR SYSTEM DATE: August 22, 1986 PAGE: 4 i
If AP = PH-PC is the total pressure drop across both the rarefaction and compression waves, then Equation (5) implies that ,
l AP = APg + APh- (6) where APH and APC are the pressure drops across the rarefaction and compression waves, respectively, i We note that, for a rigid system, total fluid momentum is conserved.
Mathematically, we can express this concept as AP 9C au (7) a *AH H+P!A"c!
C and pgauH-ACl4uCl=0 (8) 1s the velocity discontinuity across the rarefaction wave where and lauC augl is the absolute value of the velocity discontinuity (in the opposite direction)'across the compress.fon vave. Substitution of-Equation (8) into Equation (7) results in
- 9C AP .
au H" pat g
N and lAuc ! " p ea . (10)
Equations (6), (9), and (10) define an intermediate pressure, Pw, at one boundary of each wave. Thus, APH = Pg - Pw (11)
APC = Pw - PC Subtraction of Equation (10) from Equation (9) and substitution of Equation (11) yields PH+PC P y= 2 (12) 1412P082286
l T0: Karl N. Fleming, Dennis C'. Bley ,
.A -
FROM: M. V. Frank-
SUBJECT:
SHOCK WAVE EFFECT FULLOWING= HYPOTHETICAL, CATASTROPHIC FAILURE OF ISOLATION VALVES SEPARATING HIGH AND LOW PRESSURE PIPING .
IN THE RHR SYSTEM !
OATE: August 22, 1986 PAGE:. 5 ,
~
and
! APH = APC*
The maximum possible pressure applied to a pipe, due to the dynamic shock effect, would occur at the first bend. This effect is max.imized if the first bend is 90', as shown in Figure 1. We further maximize the effect if we postulate that the fluid pushed by the wave front must decelerate from sonic velocity to zero, thereby giving all of its momentum to the pipe wall before it reflects in a different direction.
The' total pressure, PMAX on the pipe is the sum of the dynamic pressure plus the maximum static pressure that could coexist concurrently with the dynamic pressure. In the situation of interest in which a compression wave propagates downstream to a 90* elbow in the RHR system, the maximum static pressure that could exist behind the wave front is Pw. This is expressed mathematically as PMAX " .A P0YNAMIC + PSTATIC For the compression wave, .
lAuCIAa C P
MAX "
+P g (13) gC Substitution of Equations (10) and (12) into Equation (13) yields P
MAX =
+P g ,
i T = PH-PC + 'N g+ PC Simplifying yields PMAX = Pg (14)
B Therefore, the maximum pressure cannot exceed the original upstream pressure, which, in this case, is the reactor coolant system pressure.
The foregoing applies to a system completely filled with water.
l Reference 1 points out the possibility of formation of an air pocket due 1412P082286 l
e *== *-
- . .eeo --++==am.- -
}
~
'fg
- i j2 , TO: Karl N. Fleming, Dennis C. Bley -
a FROM: M. V. Frank -
SUBJECT:
SHOCK WAVE EFFECT-FOLLOWING HYPOTHETICAL, CATASTROPHIC FAILURE OF ISOLATION VALVES SEPARATING HIGH AND LOW PRESSURE PIPING IN THE RHR SYSTEM DATE: August 22, 1986 PAGE: 6 to dissolution of air near the valve. There are two bounding situations with respect to dynamic shock effects that include two components of fluid (e.g., air and water), at least one of which is compressible (Reference 3). .
In the situation in which flow of the cir and water is stratified, with respect to each other, the velocity of the front lies between the sonic e velocities of either component. Since the sonic velocity in air is lower than in water, the factor, a, in Equation (13) would, in effect, be reduced relative to the all water situation.
In the situation in which the flow of air and water may be thought of as
, homogeneously mixed, the effective density of the mixture is lower. This would, in effect, reduce the factor pC in Equation (13) relative to the all water situation.
In either bounding situation, the maximum pressure applied to downstream piping generated by.a situation that includes an air pocket is less than the all water situation. In other words, air pockets or voids, such as '
described .in Refe.rence 1, Appendix B, cushion the shock, which.ts a conclusion we would reach intuitively as well. .
REFERENCES
- 1. Fauske and Associates, " Evaluations of Containment Bypass and Failure to Isolate Sequences for the IDCOR Reference Plants," FAI/84-9, Appendices A and 8, July 1984.
- 2. Tong, L. S., and J. Weisman, " Thermal Analysis of Pressurized Water Reactors," Transactions of the American Nuclear Society Annual Meeting, pp. 137-138, 1970.
f
- 3. Graham, B. W., One-Dimensional Two-Phase Flow, Chapter 6, McGraw-Hill, 1969.
r e
1412P082286 i
C _ _
. . . _ _ _ . . . . _ _ _ _ . . . _ _ _ . _ _ . _ _ _ _ _ - . _ ~ . . _ . . - . . _
INTERNAL DOCUMENTS 11/4/86 Memorandum for the NRC Commissioners from C. Kamerer on "Markey Field Hearing on Seabrook Emergency Planning" (undated) S. Lorg's Comments on PNL Draft Report on Seabrook Emergency Planning Sensitivity Study" 12/12/86 W. Lyon's note provided to S. Newberry on "Coments Pertinent to PNL Report Technical Evaluation of the EPZ Sensitivity Study for SeabrooY,' Draft Technical Report A-3852, December 5, 1c86" 12/31/87 Memorandum for V. Noonan from V. Benaroya on " Amendment for BNL Report -
A Review of the Seabrook PSA: Containment Modes and Radiolooical Source ~
Tems' PRA; NUREG/CR-4540" 1/6/87 (Draft Working Document) Memorandum for V. Noonan from V. Benaraya on "Coments Pertinent to RNL Report Technical Evaluation of the EPZ Sensitivity Study for Seabrook,' Draft Technical Report A-3852, Decenber 5, 1986" !
1/7/87 Note to R. Volmer, E. Jordan and T. Novak from E. Reis on "Seabrook Exemption Recuest" (undated) Memorandum for D. Matthews and V. Noonan from R. Perlis on "Seabrook EPZ Exemption Request" -
(undated) Memorandum for J. Scinto from R. Perlis on "Seabrook Waiver Schedule" 1/9/87 Note to E. F,ossi, L. Soffer, V. Nerses and D. Matthews from E. Reis on "Seabrook Petition for an Exemption" 1/12/87 " Review Coments on Seabrook Station Steam f;enerator Tube Response i During Severe Accidents" by T. Theofanous, UCSB l 1/13/R7 Note for S. Newberry from W. Lyon on "Seabrook Station Risk and Emergency Planning Perspective" l
l i
Markey/NRR 02/03/87
_ .- . _ . _ .. _ _ _-. _._.__._ _.. - ... _ . _ . . _ = _ _ _ _ , , -
- DP %
ENCLOSURE TO ANSWER NO. 8 INTERNAL DOCUMENTS O
0 x
v, , ,,, - - , , . - . - - - - - - - - - - - ,-- , ----- - ...-- ---
, en p4* W e4 4 --% '
ENCLOSURE TO ANSWER NO. 8 PERSONAL NOTES
$