ML20235H785
| ML20235H785 | |
| Person / Time | |
|---|---|
| Issue date: | 07/09/1987 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1600, NUDOCS 8707150263 | |
| Download: ML20235H785 (311) | |
Text
ACRSFIloo gRG NAL UNITED STATES NUCLEAR REGULATORY COMMISSION 1 IN THE MATTER OF: DOCKET NO:
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 327TH GENERAL MEETING O-V-
LOCATION: WASHINGTON, D. C. PAGES: 1 - 175 DATE: THURSDAY, JULY 9, 1987
, u
[ {p ,, -[j E D r' -
,{
~ ' -
,,",i -{i (
30 l'\j 0g nv
~ " ' " "
ACE-FEDERAL REPORTERS, INC.
'N Official Reporters 444 North Capitol Street Washington, D.C. 20001 b707150;'63
^'"" B70707 "o" (2 2)3 #37 i-1600 NATIONWIDE COVERAGE
- ( f PUBLIC NOTICE BY THE UNITED STATES NUCLEAR' REGULATORY COMMISSIONERS' i ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 1
THURSDAY, JULY 9, 1987 The. contents-of this stenographic-transcript of1the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards (ACRS), as_ reported herein, is an uncorrected record of the discussions recorded at.the meeting held on the above date.
No member of the ACRS Staff and no participant at
() this meeting accepts.any responsibility for errors or inaccuracies of statement or data contained in this transcript.
O O
i CR31630.0 COX/sjg 1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR SAFEGUARDS )
3
- *
- 1 j
4 I
327TH GENERAL MEETING 2 5
6 Nuclear Regulatory Commission 7
Room 1046 1717 H Street, N.W.
Washington, D. C.
9 Thursday, July 9, 1987 10 The 327th General Meeting convened at 9:35 a.m.
11 ACRS MEMBERS PRESENT:
12 13 DR. WILLIAM KERR, Chairman, presiding 14 DR. FORREST J. REMICK 15 DR. HAROLD W. LEWIS 16 DR. J. CARSON MARK 17 MR. CARLYLE MICHELSON !
18 DR. DADE W. MOELLER j 19 DR. DAVID OKRENT 20 MR. GLENN A. REED 21 DR. PAUL G. SHEWMON 22 DR. CHESTER P. SIESS I
MR. DAVID A. WARD 23 7,
MR. CHARLES J. WYLIE V
25 ACE FEDERAL REPO RTERS, INC.
202-347 37(K) Nationwide Cosert ye 8(G336446
31630.0 !
2 I cox 1 PROCEEDINGS 2 (Discussion off the record.)
3 CHAIRMAN KERR: Mr. Murphy, we have a schedule. j l
4 As you know, it is such that we should be finished by 10:30.
5 I know that you are not in control of that entirely. We hope 6 you would keep that in mind.
7 MR. MURPHY: I will try to keep with the 10:30 8 decdline. In the process of doing that, I will skip several 9 of the slides that are in your package.
10 Let me start off by saying good morning. You are 11 right, Dr. Kerr, I did enjoy the discussion. I think it 12 actually is quite helpful for us, as well. The objectives of 13 the study in broad scope were to provide a greater 14 understanding of the frequency risks and uncertainties due to 15 severe core damage accidents based on the assessment of 16 internally initiated accident sequences at five reference 17 plants that have different plant and containment designs.
18 Finally, secondly, they assess the usefulness of 19 the methods we use and the information contained in 20 evaluating and providing insights to decisionmakers on a 21 variety of plant-specific and generic issues and to try to 22 help focus the resources of the NRC in a more effective 23 manner.
24 Of course, finally, in terms of the draft report,
("))
25 publish it as a draft for comment to obtain comments from the ACE FEDERAL.. REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-3364 616
31630.0 .
cox 3 1 industry at large, or the public at large, f
2 In terms of methodology, we used an abbreviated !
3 analysis of the frequency of accident sequences. We employed 4 the insights we gained from previous PRAs to alter the depth 5 of the analysis in a variety of places. The second comments 6 in all these things referred to the internal peer review we 7 gave, and I won't address that unless you want me to.
8 I will talk more about the depth of the analysis 9 of the accident frequency characterization.
I 10 Extremely detailed containment event trees.
11 Proceed on to an analysis of severe accident phenomenology i
12 and sourcr. terms using state of the art tools. Consequence 13 analyses employing the improved modeling and what we believe 14 are the latest health effects models, which have been subject 15 to a lot of peer review.
16 Finally, a risk estimation with comprehensive 17 uncertainty analysis.
18 The emphasis of that uncertainty analysis is in 19 the accident phenomenology and source term areas.
20 The uncertainty analysis in the accident frequency 21 analysis is more or less the standard PRA approach, 22 propagation of data uncertainties thrcugh the reliability 23 models. We did sensitivity studies to investigate different 24 modeling assumptions. This latter part has been done in some 25 PRAs, in general has been omitted.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0 cox 4 1 In terms of the uncertainties associated with the k
2 risk analysis, one of the major criticisms of the old reactor 3 safety study was that the uncertainties were not adequately 4 addressed. In fact, the uncertainties associated with 5 accident phenomenology, while they were addressed in the 6 appendices, when you went to the figures that showed the 7 CCDFs for early fatalities, the overall risk figures, they 8 weren't included at all. That element that was written in 9 chapter -- Appendices 7 and 8 in WASH-1400, never made it to 10 the conclusion section. That information was not included at 11 all.
12 Our uncertainties are large, and that's because O 13 our understanding of severe accident phenomena is not 14 complete. Our knowledge base has increased considerably from 15 where we were in the days of WASH-1400.
16 In the last 10 years, we learned an awful lot.
17 But one of the things we learned is what we don't know. In 18 other words, we come to appreciate that some things are 19 things we should worry about that we were aware weren't 20 existing at the time of the safety study in '75.
21 This fact that our knowledge base has improved 22 substantially, but we know that the models, the detailed 23 computer models we have now leave out the phenomena we
() 24 believe are important, and are not yet ready to calculate 25 them in detail, makes it difficult in trying to express the Acn FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800 336-M46
31630.0 cox 5 l
1 uncertainties.
O 2 For this reason, we relied on the subjective 3 judgment of experts, based on what the extant knowledge was.
4 We asked them to' define reasonable ranges of potentially 5 important' issues. Then we polled them in an Adelphi-like 6 process to obtain weights roughly akin to a degree of belief.
7 as to what they felt on each of these issues. .
8 These then were propagated through the tree in a 9 statistics : type f ashion. I think it's important to mention, i
10 though, that subjective judgments by experts can never 11 substitute for good science, and that.we.need to push on 12 forward either to resolve the areas that appear to be most O 13 important with experimental validation, or we may have to 14 make regulatory decisions to include margins to encompass the 15 uncertainties.in the areas where we feel it is necessary to 16 take regulatory action. I don't mean that across the board, 17 but just where the item is high enough to draw and determine 18 that action is needed.
19 The method for risk uncertainty analyses is to 20 select potentially important risk factors, frequency, 21 containment performance, structural loads as well as 22 structural strength and the source term issues.
23 Judgmentally select discrete weights and values ;
() 24 for parameters. Then develop a reasonable range within which 25 the value of the risk would lie using a statistical sampling ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 3364M6
31630.0 cox 6 g 1 _ technique.
2 MR. EBERSOLE: When you select the discrete values 3 and weights, to what degree are these standardized? To what 4 extent are you merely reproducing WASH-1400 in the context of 5 the data that went into it? You don't use new stuff, do 6 you? I am trying to say if you were put in two different 7 ball parks, had to generate your own statistics, would you 8 come out with the same answers?
9 MR. MURPHY: I guess I don't really understand the 10 question.
11 MR. EBERSOLE: You drop in.the input, the 12 statistics, from some common set of sources. Doesn't this 0 13 make your product have a commonality that it ought not to 14 have?
15 MR. MURPHY: I don't think so. Let me show you 16 what we did. I can you will see it's quite different from 17 what was done in the WASH-1400.
18 Take a look first at an issue that wasn't j 19 considered at all in WASH-1400. This is direct heating 20 pressure. We called together our panel of experts and i i
21 provided them with analyses that would support delta P 22 associated with direct containment heating for a station 23 blackout sequence that ranged from 10 psi up to as high as )
() 24 130 psi. Each one of these had a different bases for the 25 calculation, but, in general, the basis was the percent of ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage Mn33M646
l 31630.0 (
cox 7 l
1 the cote involved in the high pressure melt ejection. The 2 experts were polled to what their degree of belief was in 3 each of these different models. That's represented by the 4 ranges shown here. In other words, one expert, the highest 5 expert, felt that a delta P of 45 psi had a weight of about 6 70 percent. One other expert below us felt it was about 15 7 percent, and the average was in the 38 percent range.
8 MR. MOELLER: Excuse me. What is the weight, the 9 weighting factor?
10 MR. MURPHY: Really an expression of the degree of 11 the belief of the analysts in the various end products. In 12 other words, we don't have a good calculation on method for p
\- 13 calculating what happens in direct containment heating. So 14 we proposed a series of different assumptions, of different 15 models. The analysts almost, in every case, said there is 16 nowhere where I am 100 percent sure this is the right model 17 and zero percent, have zero confidence in the next model. So 18 this is a degree of belief type weighting. It would be 19 easier to understand if you had the individual weights, one 20 by one. These are presented in appendices B and C of the 21 NUREG CR 4551 reports and a brief discussion of each weight, 22 where it came from, what the experts' views were.
I 23 But, in any event, the range was like so. And for
]
() 24 our calculations, as we went forward, we used the average of 25 the expert opinion, which is represented by the X on this ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(Kb33MM6
31630.0 cox 8
,' 1 box.
I
\_-)
2 The next Vugraph, I think, illustrates the 3 processes maybe a little bit better. Same problem, expressed 4 somewhat differently, now in terms of absolute pressure, with 5 a combined -- if we take the information that was on the last 6 side, convert it to absolute pressure, the direct containment 7 heating loads and the expert's view of what the pressure rise i
8 associated with them, is reflected by the dotted bars here.
9 We asked similar questions, what the strength of 10 the containment was and got this result. Where the low bound 11 here was the analysis that was done for WASH-1400. The upper 12 bound was based on the fact that that is four times design 0 13 pressure of a place where the steel vessel failed where they 14 tested it at Sandia. The one given the highest weight was 15 the one that had been done by Stone & Webster. Some of our 16 experts felt that the capacity of the containment might be 17 higher than what Stone & Webster calculated, so credit was 18 given for that.
19 The point I want to make, if we had gone with the 20 best estimate calculation and said this is the -- our best 21 estimate of what the pressure rise was associated with direct 22 containment heating, and this is the strength of the 23 containment, you come to the conclusion that direct ,
() 24 containment heating isn't important from a reactor safety 25 standpoint. I will show you, as we go on, we certainly ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Cmerage mig-336-6646 I
L 31630.0 cox 9
,, 1 didn't come to that opinion in this analysis.
\_)
2 MR. EBERSOLE: You were way out in a containment 3 challenge here. What about the thing that caused the 4 containment challenge? If you look at the plant in one way, 5 big heat source, bunch of pipes and valves, isn't it, with a 6 lot of electrical junk that makes it run. We know now that 7 valves are not reliable like WASH-1400 thought they were.
8 They don't operate under duress. Their statistics are 9 entirely different. We are finding that out from another 10 program. So did you use new data on the valves --
11 CHAIRMAN KERR: Jesse, if you are changing the 12 subject. ,
13 MR. EBERSOLE: I am not changing it. He is out 14 here on containment challenge --
15 CHAIRMAN KERR: I think Mr. Shewmon wants to ask a 16 question about containment challenge.
i 17 MR. EBERSOLE: Okay. +
18 MR. SCHWINK: I didn't understand your last 19 statement, you have the most probably DCH load, which is 20 about 70 pounds there.
21 MR. MURPHY: Yes.
22 MR. SCHWINK: Containment failure is twice that 23 load. Why is it that this says that there is a high
() 24 probability of DCH failing containment, or did I 25 misunderstand you?
ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage Wn33MM6
131630.0 cox >
10 1 MR. MURPHY: Nowhere does it say that.
O. 2 MR. SCHWINK: I thought that was a conclusion you 3 reached for your statistics.
4 MR MURPHY: That's not a conclusion I reached 5 either here or in the study. Let me explain what is driving 15 the bounds of our analysis. What I said was if'I did~a most 7 probable -- just used a most probable value, the motive of 8 distribution,.I would come to the conclusion that direct 9 containment heating is totally unimportant. What'I am saying 10 is because of the uncertainties of a distribution, this 11 intersection of these two tails is what drives the-tip of my 12 uncertainty bounds, on terms of containment failure under O 13 load.
- 14. This is important because, if further research, 15 either tells us that containment is stronger, so that this 16 curve can shift that way, or says that the loads are slightly 17 smaller, so this can shift this way, I am talking about the 18 enter section that tails, and slight shifts, either one, can 19 dramatically change the results that I will show you later.
20 CHAIRMAN KERR: Is this further research going to 21 take the form of a different group of experts?
22 MR. MURPHY: It potentially could take the form of 23 a different group of experts. One of the criticisms we have
() 24 in our elicitation process, and one of the things we are 25 trying to improve, is the way information is provided to the ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage Mn336&l6
L 31630.0 cox 11 1 experts and how the process is done.
2 CHAIRMAN KERR: If you said further research --
3 MR. MURPHY: No, what I am talking about basically 4 here is the -- in terms of direct containment heating, 5 there's a lot of uncertainty, but there's a lot of research 6 in progress. Tests in progress and further analysis in 7 progress. Various people have various feelings on direct 8 containment heating, where it is going to go. I have my own 9 personal views. But I would rather wait and see what the 10 results of the tests for further calculations are. Some 11 calculations, I have seen recently, some containment 12 calculations, suggest to me that, for at least the Surry 13 plant, I can expect to see the shift this way.
14 MR. MICHELSON: When your experts looked at the 15 containment, were they only looking at the containment 16 shell?
17 MR. MURPHY: Basically.
18 MR. MICHELSON: Not necessarily in penetrations, 19 special design valves which are not necessarily designed to 20 the same standards as containment and so forth?
21 MR. MURPHY: That's correct.
22 MR. MICHELSON: Those were not included?
23 MR. MURPHY: Those were not included. It's the
() 24 leakage from the containment atmosphere to the shell that 25 appears to be the most important.
ACE-FEDERAL REPORTERS, INC.
202 347-37(X) Nationwide Coserage 800-33MM6
31630.0 cox 12
_s 1 MR. MICHELSON: The seats of those large
(') 2 ventilation valves won't withstand the kinds of pressures you 3 are talking about here.
4 MR. EBERSOLE: This doesn't contain the MARK-I 5 flap, does it, about the direct heating in the middle, et 6 cetera?
7 MR. MURPHY: No, I will get there in a minute or 8 two.
9 CHAIRMAN KERR: We have a choice here, to have 10 some rather broad discussion on how 1150 is going to be used, 11 and we can spent our time on details of how it was carried 12 out, if you would like, but we will miss the opportunity to O 13 get more general comments, so keep that in mind.
14 MR. MURPHY: In terms of the scope of analysis on 15 the accident frequency calculation, this was a comparison of .
16 several NRC sponsored studies, as well as the study that was 17 done by IDCOR as part of their work to resolve and bring the 18 severe accident issue to closer.
19 RSMAP, reactor safety methodology application 20 plan, study of four plants done by NRC in '78 time frame.
21 IREP in '80s; RMIPE, which is on LaSalle, which will be 22 eventually incorporated into 1150, very detailed state of the 23 art level 3 PRA.
() 24 I think the bottom line is in areas we have done 25 more than in the IREP studies, in other studies we have done ACE FEDERAL REPORTERS, INC.
202 347-37(X) Nationwide Coverage 8(W)-336-(M6
31630.0 l' cox 13 1 somewhat less. I think the bottom line is the technique we O 2 heta used gives us a product roughly as good as we got out of 3 the IREP studies. Much more efficient, in terms of the cost 4 to the Agency, it's somewhat less than half of what an IREP 5 study costs.
l 6 MR. MARK: I found it a little bit difficult to !
7 read from that set of inscrutable symbols, the general 8 conclusion you have reached. The danger is large, small, 9 middle, what?
10 MR. MURPHY: What I mean here in terms of data, 11 NUREG-1150 looked at in-plant data more than it did 12 WASH-1400, RSMAP or IREP, but more than RMIPE.
.O 13 MR. MARK: Is there a way to summarize from your 14 conclusion what you have spread out there.
15 MR. MURPHY: I guess it would be more or less by l l
16 adding up all these things and integrating together. You say i
17 we have done a little more into the field of data analysis, j i
18 our initiating event was about the same. The human factors 1
19 analysis was roughly akin to what's been done on the I 20 WASH-1400. On the IREP studies, better than what was done on 21 RSMAP. Not what would be done in a current state.
22 MR. MARK: What is RMIPE?
23 MR. MURPHY: Risk management program, by NRC.
h
' 24 This we hope to have into the final NUREG-1150. PRA is still 25 in progress. Most detailed PRAs have been done in the scope Ace-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-3364 446
31630.0 cox 14 1 of analysis.
7 2 MR. LEWIS: You didn't mean what you said about 3 adding up the columns, did you?
4 MR. MURPHY: Well, these, of course, are totally 5 qualitative to start with.
6 MR. LEWIS: That's correct. So adding those makes 7 no sense.
8 MR. MURPHY: You get some idea of the weight by 9 combining the length. What I am saying integrating it all i'
10 together, in my judgment, is that the IREP and the 1150 are 11 roughly of the same character.
I 12 MR. LEWIS: You can't integrate all together
'- You didn't mean -- I don't want to 13 qualitative judgments.
14 pick on you.
15 MR. MURPHY: In terms of the way I am going to 16 present the results of the front end study, we used a 17 technique for display of information that was not the best.
18 I am not going to talk about any of the numbers on this, but 19 just to explain what we did, so you can understand the 20 results.
21 We did our analysis in typical PRA fashion, 22 propagated our uncertainty through it and got the result.
23 This is where most PRAs have stopped in terms of severe core
() 24 damage frequency. Took your best shot at it. You propagated 25 data uncertainties until we got an answer. The statistics of Acn-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6M6
)
31630,0 cox 15 1 the thing gave you a 5 top 95 bounds.
J 2 In doing this, the analyst makes a number of 3 modeling assumptions, modeling assumptions which are usually 4 not considered at all in terms of how he goes forward with 5 the analysis, but in the process of making those, his best 6 judgment in here is because we do sensitivity studies then on 7 those which expressed -- the analyst thought he was most i
8 shaky on his assumption on. Typically, there are a dozen or 9 so of those.
10 The results I am going to show, they each can be 11 changed and recalculated. The results are going to be -- I 12 will show you, show a box. The box expresses the range in 13 which the mean fell over all these sensitivity studies. The 14 base case will be -- as I indicated, will be a box, and all 15 the box indicates that from the sensitivity studies that have 16 been done, the lowest the mean got was here, the highest the 17 mean got was here.
18 MR. SIESS: In the mean sensitivity study, you 19 have arbitrary function, that's correct, you go from 0 to l 20 100?
21 MR. MURPHY: No, it was arbitrary, but it was done 22 in the sense we asked the answer list, did you have two or 23 three assumptions that you made in this area, that you are !
() 24 uncertain about. What model did you use for handling 25 dependent failures? We had some discussion yesterday on the ACE FEDERAt. REPORTERS, INC.
202-347-3700 Nationwide Coserage 8(K)-336-6646
t 31630.0 cox 16 l
1 way we handled the beta factors. That was one of the things l 2 that was treated as a sensitivity study. We reduced the beta j
3 factors from the generic beta factors that EPRI had 4 developed, i 1
5 MR. SIESS: What would be arbitrary? l 6 MR. MURPHY: If we used the data that EPRI had 7 without modification. That was one of the changes that was 8 made.
9 MR. SIESS: In deciding what to change, you didn't 10 say I will take zero and 1 and see what the effect is, g l
11 MR. MURPHY: In general, no. Although one of the 12 uncertainty studies on common cause was to assume the beta 0 13 factor was zero.
14 MR. SIESS: That's arbitrary?
15 MR. MURPHY: That's arbitrary. Others, dependent 16 on success and failure definitions, should you assume the 17 HPCSS pump will fail, not good data.
18 MR. SIESS: What were your assumptions, yes and 19 no?
20 MR. MURPHY: On that one, it's a yes or no type 21 decision.
22 MR. SIESS: No, only because you make it one.
23 MR. MURPHY: We found that Peach Bottom, the
() 24 diesel generator was probably the most reliable we had ever 25 seen, when we looked at the plant-specific date and when we ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Ccurage FM3M 6646 ,
l 31630.0 '
cox 17 1 got to the analysis of Grand Gulf, in particular, we look at r]c 4
2 the effect on that plant if the failure rate were as low as 3 indicated by the Peach Bottom analysis.
4 MR. SIESS: Nobody ever looks at that.
5 MR. MURPHY: It was an arbitrary decision as to 6 what went into the sensitivity studies.
7 With that explanation, the demands on the 8 distribution also, in terms of the final figure, reflect the 9 highest of the sensitivity studies on the upward bound and 10 the lowest.on the lower. It's not to imply these happen at 11 the same time. They are associated with different i
12 sensitivity studies. This is a display technique. Many d
O 13 people misinterpret this to say that the uncertainty range is i
14 that wide. I think that's a misunderstanding of what we are 15 saying.
1 16 MR. SIESS: What are you saying?
17 MR. MURPHY: What I just said. Over the 18 sensitivity studies, you have this type of range from the 19 highest to lowest. I think you get more information to look i 20 at this type of picture, how it varies from issue to issue 21 and what is driving it. I am concerned it might put it in 22 the one compact figure, that there is so much information 23 compressed in there, nobody is getting the message I am O 24 trying to give them. That message comes more if I put 10 or 25 11 of these across the page, and you can see how, over any I
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6646 5
31630.0 cox 18 1 sensitivity issue, it affects the answer.
O 2 But apologizing for that factor of it already, 3 these are the results we got in terms of the frequency of 4 core damage, the horizontal line is the mean of the 5 distribution for each of the plants for the base case.
6 This is the mean analysis. In general, the 7 numbers were fairly low, particularly in terms of the 8 boilers. For several of these plants, we think we note 9 something that again brings up, I guess, the nonrandomness of 10 the sample. These plants have had previous PRAs, and there 11 appears to be evidence that the utility learned from the fact 12 that they had had a previous PRA, because we went back to O 13 rebaseline the study from the earlier looks, we found the 14 numbers tend to go down.
15 MR. SIESS: How come Zion is clean shaven?
16 MR. MURPHY: The reason for Zion being clean 17 shaven, these were done -- all the others were done using the 18 same methodology. Zion was based on a different study. It's 19 difficult to break out the uncertainties associated with the 20 data and the technique they used.
21 CHAIRMAN KERR: Joe, would you say that with l
22 respect to Peach Bottom and Surry, you found anything 23 different from the WASH-1400?
() 24 MR. MURPHY: I think we found on those two plants 25 in particular, virtually everything had been identified as a ace FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cmerage 8(A336 6646
31630.0 cox 19 1 significant contributor to risk in WASH-1400 had been v
2 modified. Utility had learned lessons from WASH-1400 and 3 fixed the system. We found what was driving che analysis was 4 quite different than before.
5 CHAIRMAN KERR: What about the core melt 6 frequencies using whatever, new approaches, new betas?
7 MR. MURPHY: Core melt frequency has, in general, 8 dropped. In Surry, we had a couple things that were added 9 that weren't in WASH-1400.
10 I think one of real lessons is we went through 11 this rebaselining effort. There was -- particularly on these 12 two plants. There was real indication that the utilities had O 13 learned from the earlier PRAs and had modified their plants 14 to reflect that.
15 CHAIRMAN KERR: By "less," is there an order of 16 magnitude between the results?
17 MR. MURPHY: Yes. As I recall, Peach Bottom was 18 -- they were either two, three and four times of 10 to the 19 minus 5 in the reactor safety, as I recall. These are still 20 in the same ballpark, but down somewhat.
21 MR. MICHELSON: Did the previous Peach Bottom 22 study include external events? j
\
23 MR. MURPHY: Not in any comprehensive way.
() 24 MR. MICHELSON: Still does not, of course?
25 MR. MURPHY: That's correct. In terms of what we ;
l ACE. FEDERAL REPORTERS, }NC.
202-347-37tK) Nationwide Coserage 8(Kk336-6M6 l
(
31630.0 cox 20
-1 found influencing the accident frequencies, we found that 0: 2 station blackout was a major _ contributor to the core damage E 3 at every plant,.almost totally dominant test boilers.
4 CHAIRMAN KERR: Maybe you will talk about it 5 later, but if one goes beyond core melt tofrisk, I think that 6 the new source term code package results would be likely to 7 have the greatest influence and wouldn't have much influence 8 on core melt frequency.
9 MR. MURPHY:
I will show you one figure in a 10' minute. I think the big difference 1:s in the source term-11 rather than in the accident frequency. Although the accident 12 frequency has changed.
O 13 CHAIRMAN KERR:- Not because of source term?
14 MR. MURPHY: No, I would guess that Surry, which 15 happened to be'the one I was most fam'iliar with, that the' 16 design of the plant that was analyzed in WASH-1400 was 17 presumably-the design as of 1972. If you look at that plant 18 today, I would be surprised if the utility hasn't put close 19 to $100 million in modification of the plant. It isn't the 20 same reactor we analyzed in 1972. What our analysis shows is 21 that it's a better one. The lessons that were learned from 22 the PRA have been taken and effectuated in the plant.
23 Station blackout is important. LOCAs resulting from loss of
() 24 cooling of the reaccor coolant pump seals are deminant 25 contributors. This is an area we are having continuing ACE FEDERAL REPORTERS, INC.
202-347-3XO Nationwide Coverage 8(n-33M646
31630.0 cox 21
, 1 discussion with the utilities about in terms of trying to 2 factor in the latest data on the effect of loss of pump seal <
3 cooling onto the LOCAs. That conclusion potentially could 4 change if we could get better data-upon which to base our 5 model.
6 ATWS was found to be a smaller contributor, 7 primarily because we gave credit to improved training and 8 design modifications from the ATWS rule.
9 Design of support systems was found to be 10 important and common cause failures continue to be 11 important. Yes.
12 MR. SCHWINK: There was a fair amount of 13 discussion about insights. In the draft, as I see it, there 14 is nothing labeled " insights," there is nothing labeled 16 " conclusions
- yet, nothing of that sort. Would you say that 16 this is what are the main ir. sights of the study?
17 MR. MURPHY: Main insights of the portion of the 18 study dealing with accident frequency characterization.
19 MR. SCHWINK: Thank you.
20 MR. MURPHY: Actually, one of the difficulties in 21 the way of expressing it, 1150, almost in its entirety, is a 22 conclusionary document. The main report essentially are 23 insights, in many ways. There is a hierarchy of report
() 24 structure. You have the detailed NUREG CR 4550 and 4551 25 reports, where for a given plant, if you put them together, ACE. FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coserage 8(n336446
-31630.0 cox 22 1 they are about that thick. Those are the basic technical 7
2 documents.
3 MR. SCHWINK: I would suggest that 500 pages of 4 conclusions is no conclusions at all, given the limitations.
5 Human mind, or could you grasp it to believe that?
i I
6 MR. MURPHY: I will agree that we need to express them better. What I am saying is I wouldn't say that volume j 7
f l
8 1 of 1150 has no conclusions. I would say basically it's all i
9 conclusions. l 1
10 MR. SCHWINK: I suggest that you look at the table j 11 of contents or any part of it you want to show me, and I 12 would be delighted to see a conclusions section. Mine fails hr ,
13 to have one.
i 14 MR. MURPHY: I agree with that.
15l MR. SCHWINK: Hopefully the last one will.
16 MR. MURPHY: I agree with that, too.
17 MR. SCHWINK: Good.
1 18 MR. MURPHY: No, your criticiem is valid and we 19 recognize it.
20 We talk about a little problem we have in the 21 display of information. I will use as an example the 22 probability of early containment failure at Surry.
23 Unfortunately, the graphical technique we use to present the i
O 24 results has led to misleading information. We developed our 25 detailed containment event trees. I went through the issues ACE.FEDER.AL REPORTERS, INC.
202 347 3700 Nationwide CoseraFe 800-33MM6
31630.0 cox 23 j 1 and direct containment heating was one of the most important O 2 of them. Then we went through the analysis to get the 3 conditional probability of early containment failure. We 4 presented the results using bar graphs where we drew a 5 horizontal line to represent or actually a plus sign, to 6 represent where the point of.the Monte Carlo run resulted.
7' So you could get a feel for what the distribution was within 8 the band. I think this technique didn't work at all. We 9 need to find a better graphical technique. The reason I say 10 that is if I plot this data directly, you see quite a 11 different trend.
12 What this tells me is that in terms of the 13 probability of early containment failure, the preponderance 14 of the samples we ran said that the probability of failure is 15 low, it's relatively low, less than 10 rcent. But there is 16 a residual tail that does go out, of very skewed 17 distribution, of a long tail that is associated with direct 18 containment heating, primarily. The points in here that are 19 not direct containment heating, that are on the tail, are 20 basically associated with a combined hydrogen burn and steam 21 spike. These suggest to us where we need to do further 22 analysis and research. This is why I was making the point of '
23 the intersection of the tails before. Because if those tails
() 24 pull apart a little bit, a number of these points will go 25 away.
ACE FEDERAL REPORTERS, INC, 202-347-3700 Nationwide Coserage 800-336-(446
31630.0 cox 24 1 MR. MARK: I am a little puzzled by the scale that O- 2 is shown there. I see that the tail is less than 10 percent ,
3 of 3 percent of something, but I am not quite sure of what.
1 4 MR. MURPHY: What is plotted along this axis is 5 the probability of early containment failure.
6 MR. MARK: Against the probability of failure at 7 all?
8 MR. MURPHY: No, number of samples. i 9 MR. MARK: Probability in the event that I am 10 sitting here and the plant is there, or probability of what?
11 MR. MURPHY: Number of samples of the Monte Carlo 12 runs that were run.
n
(_) 13 MR. MARK: In event of core melt, at Wisconsin or 14 what?
15 MR. MURPHY: Core melt, after initial core melt.
16 MR. MARK: 3 percent?
17 MR. MURPHY: 3 percent is here.
18 MR. MARK: That's 0.
19 MR. MURPHY: No, 20, 40, 60, 80.
20 CMAIRMAN KERR: It's a high pressure core melt --
21 MR. MURPHY: This is the probability of early l 22 containment failure. This is the number of Monte Carlo 1
23 samples that were run.
() 24 CHAIRMAN KERR: He was asking for the conditions 25 on which each failure occurred. In the case of with DCH, l
Ace-FEDERAL REPORTERS, INC.
202-347-37(K) Nationwide Coserage 8(Kk3364M6 l
31630.0 cox 25 i
. 1 it's more than core melt, it's a high pressure core melt that
(_s) 2 is necessary to achieve this.
3 MR. MURPHY: Probability expressed here is 4 conditional on core melt. Each sequence is weighted by its 5 . frequency.
6 The point I am trying to make is the display tool 7 we used didn't display this.
8 CHAIRMAN KERR: What was the probability of high 9 pressure core melt given core melt?
l 10 MR. MURPHY: That depends on the accident 11 sequence. For some accident sequence that probability is 12 zero, for others it's very high.
O 13 CHAIRMP.N KERR: You didn't get an overall 14 probability as you did --
15 MR. MURPHY: In terms of what is here, calculated, 16 total core melt frequency.
17 CHAIRMAN KERR: Given the core melt frequency, 18 what is the likelihood that high pressure core melt would 19 occur? Or, if you don't remember?
20 MR. MURPHY: I don't remember. In one of the 21 slides in your Vugraph, it's basically the parallel of 22 contribution through the core melt frequency.
23 The result of all the plants that we looked at are
() 24 expressed here. Again, recognize that this display technique 25 distorts matters, in that it does not show the way we had ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(K)-336 6646 .
1
--___-_-_______--__-O
31630.0 cox 26
- 1 hoped it would, what the distribution within the boxes, in i t
%.)/
2 the terms of Surry, both bars on Surry, Zion and Sequoyah, 3 these things are peaked much more sharply on the bottom than )
l 4 it appears in the figure. In the case of Peach Bottom, you 5 will see it's essentially bimodal. The question there 6 develops mostly around the question -- the attack of the dry 7 well wall. We have that expressed twice for Peach Bottom 8 with and without liner meltthrough. This is the molten core 9 flowing across the base mat and coming in contact with the 10 dry well wall and melting a hole through, that being your 11 mode for early containment failure.
12 MR. MICHELSON: This is all based on the O 13 assumption that the failure results from the shell failure 14 and not from penetration failure, because you said earlier 15 penetrations weren't included in your study?
16 MR. MURPHY: That's true.
17 MR. MICHELSON: You have to somehow verify the 18 penetrations or even stronger?
19 MR. MURPHY: The penetrations were looked at in 20 this study. I don't believe there are any penetrations down 21 at the level where you would have the melt.
22 MR. MICHELSON: Where I am talking about melt, 23 that's true. But I am talking about pressure, that, of
() 24 course, is different.
25 MR. MURPHY: Sequences where this tends to happen ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 8(x133t>4646
i 31630.0 cox- 27-1 are sequences that don't necessarily have a high pressure,
.O. 2 but the flow of the melt tends to-cause the problem.
3 CHAIRMAN KERR: In the case of the Peach Bottom
.4 with low line of meltthrough, this is the one on the right; 5 isn't it?
6 MR. MURPHY: In terms of no liner, we have these 7 results in our final' report, these will change. . 'They.will 8 change on the basis of the BWR owner's group has now 9 performed a very detailed analysis of the structural strength 10 of the Peach Bottom containment. There were flaws, we knew, 11 in the study that was done by Ames Lab that we relied 12 primarily on.
-O 11 3 CHAIRMAN KERR: Changed downward?
14 MR. MURPHY: Yes.
15 CHAIRMAN KERR: What about Grand Gulf? Did it 16 change in that?
17 MR MURPHY: Not that I am aware of. The accident 18 frequency may change. We have had a lot of discussions with 19 the utility at Grand Gulf. They have made some modifications 20 to both the plant and the procedures for coping with station 21 blackout. As.you recall, station blackout is by far the 22 dominant accident sequence of that plant as we have analyzed 23 it. The frequency of that is going to go down, so in terms
-( ) 24 of risk there will be a change, in terms of conditional 25 frequency there may not be much. j ACE. FEDERAL REPORTERS, INC.
202-347-3XO Nationwide Coverage 800-336-6646
.J
m ._ _ _ .
31630.0 cox 28 1 CHAIRMAN KERR: So I could conclude from that, or 2 at least I am, that the likelihood of early containment 3 failure, given core melt, I guess, is bigger for Grand Gulf 4 than it is for Peach Bottom. It may be significantly bigger 5 for the new work on --
6 MR. MURPHY: As you said it, I think that's a true 7 conclusion, but there is another insight that has come out of 8 this work that says, essentially, from a risk standpoint on a 9 MARK-III containment, if the suppression pool is available, 10 you don't get a very much different answer with or without 11 the containment. The suppression pool is so effective on 12 that plant in reducing the risk.
O 13 CHAIRMAN KERR: It seems if the effect is 14 containment, then one should describe whatever it is you are 15 describing as containment failure. If, really, containment 16 is in the prevention phase.
17 MR. MURPHY: There's a problem with the whole 18 definition of what containment pressure means. We took it as 19 a failure of the shell. With the suppression pool available, 20 you get a substantial amount of risk reduction.
21 CHAIRMAN KERR: So you can have the structure i
22 fail, water will stay put --
23 MR. MURPHY: Yes. Difference is in Peach Bottom,
~
() 24 if you have the liner meltthrough, you bypass the suppression 25 pool. We haven't been able to find anything about by passing ACE. FEDERAL REPORTERS, INC.
{
202-347-3700 Nationwide Coserage 800-336-6646
31630.0 cox 29 7
1 suppression pool with anything other than minimal frequency.
2 In terms of the source term calculations, it was 3 necessary for us to consider a number of parameters that were 4 not in the source term code package. The source term code 5 package, as I am sure you recall, is discussed at length in 6 NUREG-0956; and among the things discussed there are eight 7 different features of the code. Where there are phenomena 8 not considered, not incorporated in the code.
9 CHAIRMAN KERR: Would it then be a valid statement 10 to say that the source term code package as published was not 11 used in the study?
12 MR. MURPHY: No,' it was used as a base upon which 13 to proceed.
14 CHAIRMAN KERR: Not that source term code package 15 that was used, a modified one was used?
16 MR. HURPHY: Having spent almost $1 million in 17 running the source term code package, I don't want to say it 18 wasn't used.
i 19 CHAIRMAN KERR: What are you telling me? I l
20 thought you were telling me it had to be modified.
21 MR. MURPHY: Let me say what was done. We made 22 the source term code package runs on a number of accident 23 sequences, typically six or seven, per given plant. We found
() 24 it was necessary, however -- it's an expensive code to run in 25 its present form. It also has the difficulties that things ;
)
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 8(n336-6646 j l
31630.0 cox 30 1 like the revolatilization of refission products are played
(')
v 2 out, on cool surfaces and that revolatilization is included 3 t'.p to time of vessel breach, but after cime of vessel breach, 4 it's not included.
5 CHAIRMAN KERR: I am not trying to be critical, I 6 am trying to find out what was used.
7 MR. MURPHY: We started off by running the code 8 package. Then we looked and we said there are certain 9 phenomenon not modeled in the core package. For those 10 phenomena, we need to rely on again the subjective judgment 11 of the experts by using a simple equation.
12 CHAIRMAN KERR: Modified code package modified by
/~ \
(_) 13 expert judgment is what you used. Am I mistaken?
I 14 MR. MURPHY: No. I think that's a fair 15 characterization.
l 16 CHAIRMAN KERR: There are people who might even 17 want to try to understand how this work was done. It seems l
18 to me they need to know whether -- l 19 MR. MURPHY: Here are the results from one 1
1 20 sequence. This was TLB prime station blackout at Surry. The 21 numbers referred to the various radionuclides groups. The 22 figure is a little bit askew in the way it's grouped in that 23 group 1 stops here and group 2 stops here, et cetera.
() 24 But for the group 2, this is the group that was 25 used in reactor safety. Later in time, we developed the ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
-31630.0-cox. 31 1 March / CORRAL code.
- O 2 What we used for the study is represented by this 3 .bar. There is a distribution in the bar that I haven't 4 shown. The reason it's higher than the' source term code 5 package is that there are phenomena that are real that are
-6 not in the source term package _that we had to add to 7 represent what the source term was. We start off with the 8 source term code package result. Then, effectively, went 9 back using a simple metric tool weighing heavily cn1 the 10 judgment to reflect these factors, such as the reevolution of 11 iodine that was deposited on the hot -- cooler services of 12 the reactor vessel --
O
+ 13 CHAIRMAN KERR If you were doing a state of the 14 art PRA you would not use the source term code without 15 modification?
16 MR. MURPHY: I would do what I had just done in l 17 1150. If you use the source term code package alone, there 18 are times where you are going to underpredict. There are 19 also times, there are several of these figures in the 20 report. There are times when you could overpredict. You 21 will find parts of some of the sequences that are shown in 22 1150, where the indication or where the source term code 23 package is above our uncertainty range, 24 What I am saying is the phenomena that aren't in f])
25 the code package yet are such that I think you have to modify ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33'# 6646
1 131630.0 c'o x 32 l 1 that if you want to give a realistic appraisal of the source l O. 2 term.
l 3 MR. MOELLER: You explained it, but it's probably /
4 nonunderstandable to me. Where is the data for krypton and 5 xenon? In other words, group l'is not on the graph?
6 MR. MURPHY: Group 12 is not on the graph. !
7 MR. MOELLER: Thank you.
8 MR. MURPHY: A summary of what we spent most of 9 yesterday talking about in terms of extrapolation of results 10 from one plant, from~these five plants that we have looked at 11 to others. We feel the methodology is transferable. . The 12 insights'regarding the important features are useful in
)O I 13 focusing analyst's attention. You can't make the direct 14 translation, however, with the quantitative results.
15 I would like to finish up, since I have got five 16 minutes, with the comparison to the safety goals that we are 17 talking about. This is early fatality safety goal, 18 quantitative design objective. We compare the results of the 19 five plants we have analyzed, too. The goal was here, and 20 all the plants are clearly within the goal. The goal, as the 21 Commission has spoken of it, is in term of the mean, and the 22 mean of these distributions tends to be roughly in the 23 middle, sometimes skewed slightly up, sometimes skewed O 24 119 at1r aowe.
25 The point being that in terms of early fatalities, ace FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cmerage 80fk33MI>46
l l
31630.0 cox 33 i
'l all the plants are within the goal, or the entire range of O 2 uncertainty, we calculated,.even though it is very wide.
3 CHAIRMAN KERR: Would you hazard a guess as to 4 changes as they occur, after one takes into account external 5 issues, or are you reluctant to do that? You might not have 6 any basis for it.
7 MR. MURPHY: Main thing I.would expect to happen 8 with external initiators, I could be wrong, this could be 9 totally speculation, I would expect the accident frequency of 10 Peach Bottom to come up, it's very low right now. I would 11 expect the external events may bring it up somewhat. I would 12 .not expect them to bring up a tremendous amount, because,-our O 13 people who have started the work on the Peach Bottom analysis 14 are very impressed with the design details to be given to 15 that plant in terms of seismic and fire, the two they have 16 started looking at right now. I don't have results yet, but i
17 the overall impression, I was told, was that the plant looked 18 like it was designed for an earthquake that was far bigger 19 than.what it was designed for.
l 20 MR. WARD: What was that, it looked like it was
-21 designed for earthquake, it looked like it was designed far 22 greater than what it was designed for.
- 23 MR. MURPHY
- I am sorry, I guess I didn't say that
() 24 very well. I apologize. i 25 What our analysts told us, when they went to Peach ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33M646
31630.0 cox 34 l
1 Bottom and looked at the design of the plant, they found that Oc 2 the anchorages, the detail that was given to seismic matters, 3 they felt was more typical of a plant in California than what 4 they would expect to see in a relatively low seismic region 5 in the East Coast.
6 For that reason, we may not shift very much when 7 we add seir-'ic considerations to the Peach Bottom plant in 8 doing it. Again, that's a matter of detail. I wouldn't 9 hesitate a guess right now. I would guess that they are 10 going to have the general trend of pushing these up. I would 11 find it very hard to believe that they would push it up to 12 there.
O 13 MR. MICHELSON: You said something about fire 14 also.
15 MR. MURPHY: They had the general impression that 16 as they were looking for problems in the fire analysis --
17 this is in its infancy.
18 MR. MICHELSON: What does that mean, the fact they 19 haven't found any?
20 MR. MURPHY: Just what it says.
1 21 MR. MICHELSON: Are you inferring that the risk 22 would be lower than you had thought? )
23 MR. MURPHY: I am saying that they are 10 to 20 0 24 percent in the analysis. What they are telling me now they 4
I 25 have not found any analysis, the analysis is continuing and I ace FEDERAL REPORTERS, INC. l 202-347 3700 Nationwide Coverage 800-336-6646 1
l i
1 31630.0 l cox 35 l
l 1 won't draw any conclusions from that. j
)
/"}
~
2 MR. MICHELSON: Peach Bottom didn't do a fire PRA j i
3 'yet did they?
4 MR. MURPHY: We are just in the process of doing 5 it. I 6 MR. MICHELSON: They have no basis to compare it l 7 since that's the first one.
I 8 MR. MURPHY: They are 10 to 20 percent into the 9 analysis, they are impressed with what they see so far. We 10 haven't taken it to conclusion.
11 MR. MICHELSON: I am trying to understand what you 12 are saying.
O k- 13 MR. EBERSOLE: I can't help but ask you, when I 14 see Sequoyah up there, will it be any different when 15 Mr. White gats through with it down at TVA, or is that the 16 way it was?
17 CHAIRMAN KERR: I wish you could have helped.
18 MR. MURPHY: This is the way it was at the time we 19 looked at the plagrc which- Nas roughly two years ago. We have 20 had a lot of discussions with people at TVA. There are 21 portions of the analysis that we are going to be modifying.
22 CHAIRMAN KERR: It's still below the safety goal, 23 isn't it? r
(} 24 MR. MURPHY: Yes. It's still below the safety 25 goal. I think that's an important thing to emphasize. I ACE Fzonic.L REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
g.
31630.0 cox- 36 1 .would, expect,'then, on the results of our discussions, things. .
.O lL .are going to drop. We are still trying to get more detail.
3 101. EBERSOLE: The enormous: cost' of this long 4 shutdown ought to be measured in some context, but I.can't
.see that it's measurable.
~
5 6 MR. MURPHY: .In terms of latent fatality safety 7 goal, here all the plants are well.below the safety goal by 8 orders of magnitude.
.9 CHAIRMAN KERR: Joe, I think I understood you to 10 say yesterday, and I wanted to verify that, that much of the
'll early containment fatality seems to be from the assumption 12 that 5 percent of the people actually'did not evacuate?.
~O 13 MR. MURPHY: Yes. That's a key point. I should 14 have made it. Thank you.
15 MR. MOELLER: You could almost ignore the latent 16 safety goal?
17 MR. MURPHY: The latent safety goal, in everything 18 we have looked at, is far below -- if you meet the early
~
19 goal, you meet.the latent goal. I think that's clear. In 20 terms of the number of early fatalities, we are predicting, 21 for most accident sequences, almost all the late'-- the early 22 fatallties come in that portion of the population, which we 23 assume does not evacuate. And it's based on the assumption
-() 24 that 5 percent of the people do not evacuate.
25 MR. REMICK: Do you assume that they shelter?
ACE-FEDERAL ' REPORTERS, INC.
202-347-3700 Nationwide Coverage 80(k336446
7 31630.0 cox 37 1 MR. MURPHY: We assume that they shelter, is that O 2 correct, mark? Show -- the final one --
3 MR. MARK: Excuse me, shelter in what?
4 MR. MURPHY: Homes.
5 MR. MARK: Mobile homes or brick built houses with 6 basements or homes they dig in the ground, what?
7 MR. MURPHY: Basically normal activity. They go 8 inside their homes.
9 MR. MARK: Homes inside.
10 CHAIRMAN KERR: You just_ assume some shielding ;
11 factor, don't you?
12 MR. MURPHY: Yes.
O k- 13 CHAIRMAN KERR: Do you know why offhand? That may 14 be too much detail. That's okay.
15 MR. MARK: Forget it.
16 MR. MURPHY: I think one of the key things coming 17 out of the study is that the number of early fatalities is 18 very affected, not surprisingly, by the people who don't 19 move.
20 Secondly, if you look at the information, we 21 develop some information there for use that shows how the l
22 information in 1150 can be used in an emergency preparedness 23 arena.
() 24 The message there is that if you are going to 25 evacuate, it's far more effective if you evacuate before the ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-3364M6
31630'.0 cox 38 f
]
1 release of the plume, than if you evacuate into the plume.
O 2 'That sounds like common sense, but that's not necessarily 3 ' reflected in'the emergency plans that exist today.
4 CHAIRMAN KERR: But it's not very common.
1 5 MR'. MOELLER: As I listened, I wondered if the i 6 statements you have made are compatible with what our ,
7 subcommittee heard, one of the ACRS subcommittees heard a 8 couple weeks ago. The report of the studies from Sandia and 9 from the New York Power Authority and from the studies of 10 Frank Rowsome, which,show that the early fatalities, as well 11 as the latent fatalities, are little affected by whether you 12 evacuate immediately or whether you take shelter and later O 13 relocate the population after the cloud has passed by. Your 14 numbers are showing that if you could repeat what you said, 15 that the early fatalities driven by the fact that you assume 16 5 percent of the population does not evacuate properly, that 17 would sound contrary to these other data which our 18 subcommittee heard.
19 MR. MURPHY: That information is in chapter 10 of 20 the report. I have a Vugraph here, but I have the wrong 21 one.
22 CHAIRMAN KERR: We are now three minutes over.
l 23 MR. MURPHY: The only one I had left was the f
O 24 comparison with the large release, which is not a safety 25 goal, but something that has been remanded to us for ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3364M6
a a
=31630.0 cox 39 1 discussion, and the committee has commented on it, and I will~
2 skip that'one and that concludes'my presentation.
3 CHAIRMAN KERR: Further questions. Mr. Remick.
4 MR. REMICK: Just a comment, Joe. I was not aware )
5 'of.the assumption of 5' percent not evacuating. It seems 6 reasonable, but it would seem to me to be helpful if you had
.7 at'least one example to show with and without that 8 assumption, to see what a difference'it made.
9 MR. MURPHY: .Yes, I think so. There are, again, 10 in chapter 10, there's one place where we had what we called 11 a "very effective emergency action," which I think was down 12 to, was it 1 percent, Mark? 1 percent didn't leave, then we l - ()
l 13 increased the evacuation' speed. The decrease in the number 14 of fatalities was on an order'of an order of magnitude or 15 more. Very substantial.
16 CHAIRMAN KERR: Further-questions.
17 MR. MURPHY: It's kind of buried in the middle of 18 the report.
19 CHAIRMAN.KERR:' If not, thank you very much, 20 Mr. Murphy.
21 15-minute break.
22 (Recess.)
23 CHAIRMAN KERR: Next item on the agenda is
() 24 integrated safety assessment program.
25 MR. WARD: Thank you, Mr. Chairman.
ACE-FEDERAL REPORTERS, INC, 202-347-3700 Nationwide Coverage MO-33MM6
31630.0 cox 40 1 Most of you probably remember something called the O 2 SEP, systematic evaluation program, I think those letters 3 stand for, which was completed two or three years ago. That 4 was a program which was called sort of a cooperative program 5 between the NRC Staff, NRR Staff, and licensees who operated 6 the 10 oldest plants, or at least plants that had been 7 licensed in the early days of regulatory activities. This 8 program was intended to assess those plants against what was 9 called kind of a modern set of regulations and decide what 10 upgrades, if any, were needed in any of the particular plants 11 to bring them into some sort of essential conformance with 12 the more modern regulatory requirements.
O 13 The program was unique in it did not require 14 literal conformance with each and every regulation, but would 15 accept what I might call risk-based arguments, for example, 16 in making cooperative judgments between the NRR Staff and 17 Licensee Staff about interpretations of meeting regulations 18 in detail, or whether certain regulations had to be met or 19 not.
20 The program was, I think in most quarters, 21 including here at the ACRS, regarded as rather successful 22 one, because there was a perception that both the NRC and 23 licensee got a lot of value out of the investment in upgrades
() 24 that were made, and one of the reasons for that is that NRC's 25 procedures, and the practices under this program, permitted .
l ACE FEDERAL REPORTERS, INC. j 202-347-3700 Nationwide Coverage MO-336-6646
i 31630.0 cox 41
- . 1 the licensees of those 10 plants to take sort of an
%-)
2 integrated approach to upgrades. That is, rather than having 3 to fix up item X by the end of July to deal with issue X and 4 Y, and so on, it was permitted to look at a whole draft of 5 issues and perhaps develop a fix here or here that would deal 6 with several of the issues at the same time and was permitted 7 a schedule to accommodate that kind of thing, in addition to 8 making what I might call risk-based arguments about the 9 necessity or priority to assigning certain fixes.
10 Today we are going to be considered another 11 program called ISAP, integrated safety assessment program, 12 which is kind of son of SEP, might look at it that way. It's r
13 a program that started out with a certain enthusiasm, I guess 14 back in '84. There seemed to be the intent at that time, 15 some endorsement from ACRS, about applying this sort of 16 approach on a broader scale, perhaps to all plants.
17 About the essential features of the SEP program to 18 be applied. Chet is not here to keep me straight, but as I 19 recall the SEP program dealt with the changes in regulation 20 in the TMI action plan, but not the whole set of generic 21 issues that are out there. The ISAP program was intended to 22 deal with all outstanding issues at a given plant, including 23 all of the USIs or generic issues that reached resolution of
() 24 the Staff and will deliver some new requirements or 25 suggestions for licensees to undertake. I am not sure what ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80u-336 6646
31630.0 cox 42 1 happened to the broader idea of the ISAP program. Staff (v_) 2 seemed to go -- or bulk of the Staff seemed to go on to some 3 different things. But there remained a core of an ISAP 4 program that was scaled down to what's been called a " pilot 5 program." This pilot program is a cooperative effort between 6 Northeast Utilities and the Staff and an ISAP group and the 7 Staff to apply this sort of process to a whole list of 8 outstanding issues which existed for these plants to date.
9 Northeast Utilities is sort of a unique outfit, I think.
10 It operates four nuclear plants, Haddam Neck and 11 Millstone 1, 2 and 3. They are all different and the fact 12 that they are different, I think, has led the utility, maybe,
) 13 into having somewhat different interests about how to 14 approach this sort of thing than other utilities might have.
15 At any rate, they seem to be quite enthusiastic 16 about the program, and I should say is the program, the pilot 17 program, applies just to their oldest plants, Haddam Neck and 18 Millstone 1. They have been working on it for a long time.
19 Part 1 -- essential part, not all of the program, but one 20 essential part of the program is to have the ability to make i
21 for what I call " risk-based evaluations" of things. Of I
22 course, having a plant-specific PRA is a big help in doing 23 that. All of the SEP plants did not have plant-specific
(]') 24 PRAs, surrogates and borrowed techniques, so forth, to make 25 the risk assessments that they did.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 804 336- % 16
1 31630.0 cox 43 1 In this pilot ISAP program, Northeast actually has (b
2 made, so far, at least, level 1 PRAs for each of its four 3 units. So it has that as the tool as an important part of 4 the program. I think that the small part of the Staff, the 5 ISAP group that has participated in this, many of the people 6 in that group also participated in the SEP program. The 7 small part of the Staff thinks it's an effective program and 8 a good way to deal efficiently with the multiple issues that 9 it sees that utilities at various plants have to resolve.
10 The utility seems to be very enthusiastic about 11 the program; and, as I understand it, they would like to 12 continue on and have the Staff give them the okay and agree 13 to cooperate with them in extending the program to their 14 other two plants and also to future generic or other issues, 15 other regulatory safety issues that come up in the future. I 16 think most of the members of the subcommittee agree with the 17 utility and the Staff in this area. We think the pilot 18 program has been carried out very well, and we think it's a 19 useful model for extending more broadly. Whether it's really 20 practical to apply it at some time in the future to all 21 plants or not, I don't know. Paul.
22 MR. SCHWINK: Why are we hearing about this ;
23 today?
(} 24 MR. WARD: Okay, i 25 MR. SCHWINK: Everything that you say is true, and 1
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-M46
31630.0 j cox 44 '
1 it was true a couple of years ago, although they have gone i
( farther with it now. Is it coming to some conclusion, is 2
3 there talk of phasing it out, is the Staff considering l I
4 seriously opening this door again so that others could come
~5 in, if they want, of utilities? l l
6 MR. WARD: Yes. The first phase of it, the pilot 7 phase, is about to be finished. The Staff has written -- a 8 draft of the assessment that Northeast has done on the first 9 plant. It's a draft form. So to some extent, we will be 10 reviewing that. But I guess the main reason we are hearing 11 it today, and what I would like to see the committee write a 12 letter on, is to look at this process, whether or not we O 13 think this pilot program has, indeed, been successful and 14 whether we think that the process should be extended, perhaps 15 that Northeast would perhaps offer it in some way to other 16 utilities. I think that's what's going on today.
17 MR. SCHWINK: Fine, thank you.
18 MR. WARD: Unless any of the other subcommittee 19 members would like to comment, we will go now to the agenda.
I 20 MR. EBERSOLE: I would just like to make an 21 observation, Dave. You can't tell what you are looking at 22 when you just listen to the name of a program. ISAP, in my 23 view, has produced, at least these two plants is not what I
() 24 would call an ISAP, it rides critically on the presence of 25 the SEP program. So ISAP, remember that Dave said it's a ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Courage 800-336 6646
31630.0 cox 45 1 study of the residual problems, having already considered (3
2 those, SEP and IREP, if you looked at a plant that had never I
I 3 been attended to by SEP and IREP, then IREP would be the 4 whole show. But as it is presented here, it has a front end 5 adder to it which is SEP and IREP. So this is only the back 6 end of a process that includes those earlier investigations.
7 Did anybody say that I said that wrongly?
8 .
i 9 MR. THOMAS: Cecil Thomas from the Staff. Jesse, 10 the details of the program has continued. The details i
11 obviously have to be worked out. We would envision going 12 back to a plant that had not been in an SEP, that wanted to 13 participate in ISAP, we would envision going back and picking 14 up the same types of issues that were considered in the SEP 15 program.
16 MR. EBERSOLE: Then it would be a true ISAP?
17 MR. THOMAS: Yes, sir.
18 MR. WARD: Any other comments. First on the 19 agenda, we will turn to the Staff. Mike Boyle is going to
]
20 present.
21 MR. BOYLE: My name is Mike Boyle. I am the 22 integrated safety assessment manager from Millstone 1. Today 23 I am going to discuss the history, scope -- I was going to 24 discuss, but Mr. Ward took care of 99 percent of my l
25 presentation for me.
ACE FEDERAL IlEPORTERS, INC.
202-347-37tK) Nationwide Coserage 80(k336-6646 j
31630.0 cox 46 1 MR. WARD: Sorry.
O 2 MR. BOYLE: Besides the history and scope, review 3 the methodology for ISAP, give an example of the lessons j 4 learned and conclusions that we found in conducting ISAP for {
5 both Millstone 1 and Haddam Neck. I am also going to touch 6 on the recommendations that we feel for the future course of 7 action for ISAP.
8 As is previously stated, NRC initiated the 9 systematic evaluation program, SEP, in 1977. Its purpose was 10 to review some of the older operating power plants against 11 regulations and requirements that had evolved since the plant 12 had received its operating license.
O 13 Phase 1 of SEP was the compilation of those issues 14 that were going to be evaluated in SEP. Mr. Ward stated that l
15 TMI items were evaluated, they were not -- anything that was 16 being handled generically for all the plants was handled out 17 of SEP. We only look at those issues that were handled 1
18 generically and the issues to be compiled for were compiled 19 before the TMI accident.
20 Phase 2 of SEP was a pilot , review of 11 plants.
21 Review included a deterministic review of safety issues that 22 were identified in phase 1.
23 Plus there was a very limited PRA conducted of O 24 individual issues when they were amenable to that. The Staff 25 conducted an operating experience review from the time the ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 84336-6M6
31630.0 cox 47 I plant was licensed until the end of SEP.
.O 2 Phase 2 identified significant findings relative 3 to the safety of reactors, and it also identified useful 4 evaluation techniques for all reactors.
5 As originally envisioned, phase 3 of SEP was the 6 application of the lessons learned in phase 2 to all the rest .
i 7 of the reactors in the country.
8 As a development of the TMI. action plan, the 9 interim reliability evaluation program, IREP, was initiated-10 by the NRC. Its purpose.was to perform a limited number of 11 plant-specific PRAs to supplement the risk, reliability and-12 evaluation findings in WASH-1400. It was also to define O 13 methods to conduct additional plant-specific PRAs to enable 14 consistent comparable results between all the plants.
15 One of the significant findings from both of those 16 programs is that issues related to safety of operating 17 nuclear power plants can be more effectively and efficiently 18 addressed in an integrated, plant-specific review. The Staff 1
19 took the good points of both IREP and the SEP program, that 20 being the deterministic review of safety issues in SEP, and 21 the operating experience review of the plants in SEP, 22 together with a full level 1, at least a level 1 PRA that was 23 recommended from the IREP program, merge those two together
() 24 into'one program called "ISAP." This was initiated with a 25 Commission policy statement issued in Federal Register in ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
1 31630.0 l cox 48 g- 1 November of 1984.
's / .
2 Subsequent to that, for a number of reasons, one 3 of which was budgetary, the NRC recast ISAP into a two-plant 4 pilot program. One of the reasons for that was to get a 5 quick result to be able to find recommendations for 6 conducting ISAP for all plants. It was then decided that the 7 two plants should be plants that alzeady had participated in 8 SEP in order to do deterministic reviews which would take 9 quite a number of years to complete. It wouldn't have to be 10 done again. It could get through the pilot program rather 11 quickly.
12 Northeast Utilities volunteered Millstone Unit 1 13 and Haddam Neck to participate in ISAP. It should be noted 14 that Millstone Unit 1 also participated in IREP.
15 ISAP was to include integrated assessments of 16 operating reactors on a plant-specific basis. It was to 17 evaluate all current licensing actions that were on the 18 books. All licensee initiated plant improvement projects and 19 selected unresolved generic and safety issues. Those 20 selected were from NUREG-9033, which is the prioritization of 21 generic issues. We only selected those which had a high ,
22 priority in that NUREG. We did this to establish 23 implementation schedules for all of the items for both
() 24 plants.
25 MR. EBERSOLE: If you were to change that word in ACE. FEDERAL REPORTERS, INC.
202-347-3700 Natioriwide Coverage R&3364M6
l 31630.0 cox 49 1
1 the first line, "are" to "would be" wouldn't you go back and O 2 include all the elements of SEP and IREP? 1 1
3 MR. BOYLE: It was. It was a foundation of ISAP. j 4 MR. EBERSOLE: That would be more than residual 5 safety, that includes past solved safety issues.
6 MR. BOYLE: Yes. The basis for making many of our l 7 decisions was the fact that we had an IREP and SEP foundation l 8 for the plants. I 9 In addition to the integrated schedule that 10 resulted, the procedures for maintaining and updating the 11 integrated schedule were also to be developed. Major 12 elements of ISAP were the review of lessons learned from l
' 13 SEP. And, as we have stated, this was not performed in the 14 pilot program since the plant had already completed SEP. The l
15 performance of a plant-specific risk assessment, the analysis 16 of operating experience --
17 MR. MICHELSON: Which operating experience do you 18 mean?
19 MR. BOYLE: Staff itself performed an operating 20 experience review for the Staff.
21 MR. MICHELSON: Plant-specific operating 22 experience?
23 MR. BOYLE: Yes.
24 MR. MICHELSON: Not sister plant experience or 25 industry?
ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage tun 33M646
l 31630.0 cox 50 1 MR. BOYLE: No. J 2 MR. MICHELSON: Thank you. ,
)
3 MR. BOYLE: We analyzed all the data generated )
I 4 from the individual plants. I I
5 MR. MICHELSON: Individual plants that were under j l
6 study?
7 MR. BOYLE: Yes. ,
i 8 ,
MR. MICHELSON: Thank you. ;
9 MR. BOYLE: This differed from the SEP and we i 10 asked the people performing the review to not only give us ;
11 numbers, but try to explain the numbers and give indications 12 of trends of plants, if things were getting better, getting O 13 worse in different areas.
14 The licensee was then asked to perform an l 15 integrated assessment of all of these items. The integrated l
l l 16 assessment proposal was supplied to the Staff. We reviewed l 17 it and issued our findings in the draft report, NUREG-1184, 18 which has been supplied to you.
19 We asked then the ACRS, licensee and a peer group 20 to review that draft NUREG report and give us comments, f-21 In addition, the licensee was asked to supply a 22 proposed integrated implementation schedule.
23 The comments and a final schedule will be made
() 24 part of the final report.
25 MR. EBERSOLE: A previous subject was 1150. How ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage mL336-6646
l 31630.0 cox 51 1 do you see any --
2 MR. BOYLE: Later.
3 MR. EBERSOLE: All right.
4 MR. BOYLE: I will go over that later.
5 MR. EBERSOLE: Okay.
6 MR. BOYLE: This is getting a little more detail 7 on how ISAP is actually conducted. The Staff set up an 8 integrated assessment team for each of the plants. The team 9 was made up of an integrated assessment manager, project 10 manager, project director for the plant, resident inspector, 11 a reliability and risk analyst and whatever specialist we 12 needed on call, such as human factors or other engineering
[}
L 13 disciplines.
14 The integrated assessment team and the licensee 15 then independently went out and made a compilation of all 16 issues that they thought should be included in an ISAP 17 review.
18 The two groups then met, and between themselves, 19 hammered out which issues should be included, which not 20 included. Those completed were to be inc3uded in ISAP.
21 Regulations, orders with a due date or a completion date in 22 the near term, projects that were almost complete and would 23 benefit at all from ISAP.
() 24 The two groups defined the scope and review 25 criteria and the justification for continued operation for ACE FEDERAL REPORTERS, INC.
, 202-347-3700 Nationwide Coverage Ro-33M646
31630.0 cox 52 1 both of those plants. The ISAP review is divided into three O 2 parallel phases. The first was the topic evaluations. There 3 were not any new topic evaluations in the pilot program 4 because the SEP lessons learned had already been completed 5 for the plants. However, other items that were still
{
6 licensing actions still on the books for which evaluations 7 were due for a licensee, if a licensee sent in a review, we 8 then evaluated that in ISAP.
9 The licensee was also asked where it was amenable 10 to supply probabilistic analyses for the individual topics.
11 At the same time, the licensee was performing an overall 12 probabilistic safety assessment, and the Staff performed O 13 their operating experience review.
14 From the PSA, and the operating experience review, 15 the licensee and the Staff identified additional issues which 16 were not obvious when ISAP started, which should be included 17 in ISAP 4.
18 MR. MICHELSON: Let me interrupt for a moment, on 19 operating experience, why did you not want to look at other 20 operating plant experience as it may look to this plant?
21 MR. BOYLE: I can't really answer that. I don't 22 know. We thought at the time that the specific plant l
23 experience was probably more important.
O 24 xa x cn8' son: "evde = re 1 9 rte =t, b=t i
25 certainly limited sample.
I ACE FEDERAL REPORTERS. INC. I 202 347-37W) Nationwide Coserage h(0 33M646
31630.0 cox 53 1 MR. BOYLE: Although both plants have been 2 operating for quite some time. l I
3 MR. EBERSOLE: This is to say there is no generic 4 component in ISAP.
5 MR. BOYLE: We tried to make a plant-specific 6 review instead of a generic review.
7 MR. EBERSOLE: Again, it's not integrated, because 8 it contains none of the generic problems.
9 MR. BOYLE: It does contain generic problems 10 evaluated on a plant-specific basis.
11 MR. WARD: A lot of general experience or generic 12 experience is in the NRC issues; ISAP has to deal with O 13 those.
14 MR. BOYLE: I&E bulletins, notations, generically, 15 whatever tied them into the plants, we did look at those 16 things.
17 MR. EBERSOLE: Okay.
I 18 MR. BOYLE: I should also note from the operating 19 review, if we reviewed something of high safety significance, j 20 we required immediate action be taken to rectify that.
I 21 Otherwise, the issue that was identified would be evaluated 22 in ISAP with all the rest of the issues.
I 23 MR. EBERSOLE: So what you said, if you found l
() 24 something that was really bad, you fixed it?
25 MR. BOYLE: Right away. j l
/\CE FEDERAL REPORTERS, INC.
{
202-347-3700 Nationwide Coverage 800-3364M6 j I
i
31630.0 cox 54 MR. EBERSOLE: Then it disappeared. I think o
1 i
2 that's interesting. In other words, you brushed out what was 3 the most important things, aga.fnst the old set of rules that 4 built the plant. That sounds funny to me. I think those 5 ought to go on the record at the front end.
6 MR. BOYLE: They are on the record. They are in 7 the report.
8 MR. EBERSOLE: That's part of the ISAP product.
9 You are telling me you threw it away. )
l 10 MR. BOYLE: We did not throw them away. They are
]
1 11 in the report. They are identified. In fact, I have some 12 examples later for you.
13 MR. MICHELSON: On the operating experience, one -
l 14 more question. Did you look to see how the utility had used i 15 the SERs and SOERs issued by INPO and incorporated them into i 16 their design or operation?
17 MR. BOYLE: Yes. People who did the review went 18 from Oak Ridge. They went through every piece of paper they 19 could find from anybody on plant experience, including INPO.
20 MR. MICHELSON: That's a little different 21 statement you made a minute ago when you said you looked only 22 at specific plant operating experience.
23 MR. BOYLE: We tried to cut it only on the plant, O 24 though.
25 MR. MICHELSON: I think I am hearing two different ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Cmerage 8(n336#46
l l
31630.0 !
cox 55 l 1 answers, but go ahead.
2 MR. BOYLE . Staff, at this point, issued their 3 Staff integrated report which is being evaluated by the ACRS 4 peer review which is similar to the SEP. Final assessment t
report is due to be issued this month or next month, in the l 5 1 1
6 near future.
7 ISAP not only identified issues requiring
\
8 immediate rectification, but also identified areas that the 9 Staff had thought to be important or that the licensee had 10 thought to be important, and found out that these areas were 11 not as important as previously thought.
12 For Millstone 1, the PSA identified that 64 O 13 percent of the total calculated core melt frequency was due 14 to failure to maintain adequate long-term cooling. Here the 15 licensee identified this to us, made some immediate 16 corrective actions and initiated a long-term decay heat 17 removal study.
18 This is all on the record. For Haddam Neck, loss 19 of motor control in the switch gear room would cause a loss 20 of function. Licensees also started to take immediate 21 corrective action. i 22 MR. EBERSOLE: Largely, that was AC power failure 23 or were there other contributions to loss of that?
() 24 MR. BOYLE: There is. I think when we get into 25 licensees' presentation, they will go into this in more ACE-FEDERAL REPORTERS, }NC. I 202 474700 Nationwide Coserate 800-336-6M6 l
1
'I l 31630.0 cox 56 1 detail.
k 2 MR. MICHELSON: Looking at the PRAs that you i
3 looked at for these plants, were these broad scope PRAs, 1 4 including external events?
5 MR. BOYLE: No, the original PSA was not external 6 events. However, they have supplied an internal flooding and 7 fire analysis.
8 MR. MICHELSON: Not in the sense of a PRA with a 9 risk estimate on it?
10 MR. BOYLE: I believe see. It's going to be 11 discussed later in their presentation.
12 MR. MICHELSON: Thank you.
O 13 MR. BOYLE: For Millstone 1, the licensee was 14 asked to address degraded grid voltage protection, as were 15 all the rest of the licensees. They proposed a fix for 16 that. After they had completed their PRA, they saw that the 17 fix that was being proposed and the Staff accepted was 18 actually going to degrade safety instead of increase safety.
19 The same thing for Haddam Neck, their demineralized water 20 storage tank fix was going to add more problems than it was I
21 going to solve.
22 MR. REMICK: Millstone 1, I forget. Is that a 23 plant with emergency condenser?
() 24 MR. BOYLE: Isolation condenser.
25 MR. REMICK: What is the policy that that is a ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 800-336 4 :6
F l
l 31630.0 l CoX 57 1 high percentage of the risk? Why, in the case of Millstone
{
l
() 2 1, with the isolation condenser, that's kind of a passive 1
3 decay heat removal system, why was the risk so high with that 4 system at present, what is the problem, the valve not j 5 opening, one valve that is closed not opening.
6 MR. ATEFI Failure of IC did not, major portion, j
7 it was the long-term decay heat removal problem, which I 8 guess the licensee will discuss, that was calculation that 9 that system might not work, so IC was not the preliminary 10 problem at issue.
11 MR. REMICK: I guess maybe it will come up with 12 Northeast Utilities' presentation, if the isolation condenser
) 13 works, what is your problem in decay heat removal? That I 14 don't understand. If I recall, the isolation condenser has 15 fire water that can be put into the shell side -- I thought 16 that was a pretty good system. Assuming one valve kept open 17 and the other closed.
18 MR. EBERSOLE: If you depressurize, like BWR, l 19 toward the secondary system, needs a driving head temperature 20 to let to low atmospheric temperature, gives you pressure 21 cooling if you have a depressurized climate. You lose the 22 thermal driving force if you depressurize.
23 MR. REMICK: Not if you have heat there. You have
() 24 got to have AC generator steam.
25 MR. MICHELSON: Maybe we will find out later. I ACE FEDERAL. REPORTERS, INC.
202-347-3700 Nationwide Coverage f40-336-6646
31630.0 cox 58 j 1 don't understand it either.
O 2 MR. WARD: I don't know how much detail. Rick, 3 could you answer Dr. Remick's question?
4 MR. KACICH: Richard Kacich from Northeast ]
5 Utilities. Your basic premise is correct. If the isolation 6 condenser works, there's no problem at all. This scenario i I
7 deals with the fact that isolation condenser is inoperable j 8 for whatever reason. Then as he mentions, NTSH concern and a 9 problem that might be a problem in the long term, it's a 10 long-term decay heat removal. l 11 MR. REMICK: What is the problem that the 12 isolation condenser might not work? Is the valve not O 13 opening?
14 MR. CAMP: That's the most likely single --
15 MR. REMICK: If I recall, you keep three out of 16 four valves open, one is closed.
17 MR. KACICH: That's correct.
18 MR. BOYLE: I would like to go on to the 19 conclusions that we found in the ISAP pilot program.. We 20 found that the performance and operating review of the Staff 21 gives Staff and licensee better understanding of plants' 22 capabilities and characteristics. The integrated assessment 23 has potential to identify common elements in many different
() 24 issues, or at least some issues, and deal with or propose a 25 single resolution to take care of those. Integrated ACE FEDERAL REPORTERS, INC.
202-347-37(0 Nationwide Coserage 8(0-336-(M6
7 31630.0 j cox 59
,_ 1 assessments also provides an opportunity for the Staff and V
2 licensee to address pending licensee requirements on a 3 plant-specific basis.
4 ISAP also provides a formal process to evaluate 5 the prioritization of all proposed actions. This is as pro 6 -- this is as opposed to generic letter 8320, integrated 7 living schedule program, which doesn't give the Staff or 8 licensee a formal process for developing an integrated 9 schedule, a program, but not a process by which to view the 10 program.
11 The pilot program has demonstrated the potential 12 benefits to licensees, the public and the NRC of integrated
/~'s
\~ 13 assessments, using plant-specific PRAs and operating 14 experience reviews. Currently,. the pilot program is 15 scheduled to be completed by the end of the year, hopefully 16 by the end of the fiscal year.
17 Purpose of the pilot program was to develop 18 methodology and recommendations to the Commission in how to 19 proceed with ISAP for all plants, not just two plants. Staff 20 is currently preparing a Commission paper that is going to 21 detail the ISAP pilot program experience and make 22 recommendations to the Commission on how we should proceed 23 with ISAP.
() 24 The preliminary recommendations that we are making -
25 to the Commission is, indeed, we like ISAP very much. We ACE FEDERAL REPORTERS, INC, 202-347 3700 Nationwide Cmerage 800-3364 646
__________________..______________.__j
31630.0 cox 60 ;
l 1 think that the utility that is participating in ISAP thinks l 7._
(_/ it's a very good program. We would like to continue the 2 j i
3 program. We would like it to be made available to all !
l 4 licensees. We would also like to incorporate parts or whole )
i 1
5 other' programs that are related to it. And as was just )
6 brought up a few minutes ago, like severe accidents. Also 7 the integrated living schedule program. The Commission paper '
8 is currently with the NRR office director, and it is expected ]
l 9 to be issued to the Commission by the end of the month. J I
10 MR. MICHELSON: Will that paper require a full I l
11 scope PRA on the part of a participant?
12 MR. BOYLE: A current thought right now is if a rm
() 13 full scope PRA was required of a utility for another reason,
+
14 such as severe accident program, that it might be a good idea 15 to include an ISAP-type process along with that.
16 MR. MICHELSON: You are not requiring a full scope 17 PRA in order to do an ISAP?
18 MR. BOYLE: If an ISAP were to be done, it would 19 be necessary to have a full scope PRA in order to make good 20 conclusions.
21 MR. MICHELSON: You are saying it is a part of the 22 program?
23 MR. BOYLE: It is a part of ISAP as it is
() 24 presented now. That's why we would like to incorporate other 25 programs and have risk assessment in them. That's the end of ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverare m)-3364M6 l _ ..
31630.0 cox 61 1 my presentation.
2 MR. MICHELSON: Any other questions? Chet.
3 MR. SIESS: Haddam Neck and Millstone 1, they had 4 been looked at in relation to current regulatory criteria?
5 MR. BOYLE: That's correct.
6 MR. SIESS: Essentially, those discrepancies had 7 been disposed of. If there were a new ISAP program involving 8 a plant which had not been through something like SEP, would 9 they have to look at the extent to which they complied with 10 current criteria or would the ISAP simply deal with the 11 requirements that were in force? )
l 12 MR. BOYLE: As ISAP is currently formulated, the j 13 Commission's policy statement, the SEP lessons learned are an 14 integral part of the program. A utility not having been 15 through SEP would have to look at those lessons learned.
16 MR. SIESS: Something like that.
17 MR. BOYLE: Lessons learned are subset of SEP l
l 18 topics evaluated.
l l 19 MR. SIESS: It would be no requirement. Backfits 20 would have to be in both.
21 MR. BOYLE: In all likelihood.
22 MR. SIESS: As they were in SEP.
23 MR. BOYLE: It should be noted that the licensees
() 24 volunteered to make those changes.
25 MR. SIESS: The threat was there, but I am not ACE-FEDERAL REPORTERS, INC, 202 347-3700 Nationwide Cmerage 8(0-336 4 46
l 1
31630.0 cox 62 1 even sure it was even made.
O, 2 MR. BOYLE: That's right. )
3 MR. SIESS: That's a part of the policy statement 4 for the pilot ISAP?
5 MR. BOYLE: Yes.
6 MR. SIESS: Whether it would be a part of it for 7 the extended ISAP would depend on what the Commission says.
8 MR. BOYLE: That's correct.
9 MR. SIESS: Do you think it would make your job 10 easier or more difficult?
11 MR. BOYLE: Personally, I think it would make the l
12 job last much longer. For many plants, it's not altogether O 13 certain how valuable the SEP lessons learned are. If the 14 plant was just recently licensed, it would make no 15 difference.
16 MR. SIESS: Do you think it would make the plant 17 safer?
18 MR. BOYLE: Depends.
19 MR. SIESS: SEP plants, there were relatively few 20 things that had to be done.
21 MR. BOYLE: Relatively few major items that had 22 been done, that's correct.
23 MR. SIESS: Of course, would be moving forward in
() 24 time. If you had your druthers, would you rather have the 25 ISAP just deal with requirements, in effect?
ACE FEDERAL REPORTERS, INC.
202-347-3X10 Nationwide Coverage 800-3364M6
31630.0 cox- 63 1- MR. BOYLE: Speaking for myself, I think the SEP ]
.O 2 lessons learned might be better evaluated outside of ISAP.
u
)
3 CHAIRMAN KERR: What is the relationship or what-i 4 is the relationship between ISAP and IPEM?
5 MR. BOYLE: Unfortunately, I don't know what an 6 IPEM is and I don't know anybody else that does right now. I 7 know there are quite a number of recommendations for~what an 8 IPEM might be. If ISAP and severe accidents were somehow i 9 merged, I would think the IPEM would be a full level 1, at l 10 least a full level 1 PRA, and not some small little smart 11 PRA.
12 CHAIRMAN KERR: Has any thought been given to a
'O 13 possible merging of the two, other than over coffee?
14 MR. BOYLE: A lot of thought has been given to not I
15 only severe accidents but other programs that have some 16 relationship to the ISAP process.
17 CHAIRMAN KERR: Thank you.
18 .MR . SIESS: Does the utility -- I will save the 19 question.
20 MR. WARD: Any other questions or comments for 21 Staff? Thank you very much.
22 We will turn next to the Northeast Utilities. I l
~
23 understand that Mr. Blasioli will be the spokesman.
O 24 xa 8t^s ot = xv #e e 1 veu1 81e to11- em 25 representing Northeast Utilities. On behalf of NU I would )l ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6M6
i 31630.0 cox 64 1 like to thank the ACRS committee for inviting us down here to O- 2 give our perspectives on ISAP. We did participate in the 3 ISAP subcommittee meeting on Tuesday. We believe we had a 4 very constructive dialogue. What I will try to do is 5 summarize the key points from that meeting.
6 CHAIRMAN KERR: After thanking the ACRS for 7 hauling you down here, shall we believe anything else you 8 say?
9 MR. BLASIOLI: Okay. Some of the other 10 participants we have are Richard Kacich, manager, generation 11 facilities, licensing; Jack Quinn, supervisor, Millstone Unit 12 1, licensing; myself, supervisor, generation facilities, O~#
13 licensing; John Bickel, supervisor, probabilistic risk 14 assessment; Mitch Lederman, engineer, generation facilities 15 licensing.
16 On Tuesday, Ed Mroczka was down here who was our 17 senior vice president, nuclear engineering operations.
18 I wanted to briefly go over some ISAP background.
19 Mike Boyle discussed most of this, so I will try to minimize 20 my discussion.
21 Very initial ISAP proposal by ourselves was made 22 in September '83. The Commission policy statement on 23 systematic safety evaluation of operating nuclear power
() 24 plants came out in November of '84. There was a final 25 Commission approval of the ISAP pilot program for Millstone ACE-FEDERAL REPORTERS, INC.
202-347-37(K) Nationwide Cmerage 800-336 & 46
i 31630.0 cox 65 1 Unit 1 and Millstone Unit 2 in May of '85 that specifically O 2 identified those two plants as the pilot program plants. We 3 submitted our ISAP final report on Unit 1 on July 31 of 1986, 4 and we did similar submittal for Haddam Neck on December of 5 '86. Draft NUREG, which you are all familiar with, came out 6 in April, '87. We are currently respond'ing, working with the 7 Staff to respond to that NUREG with the anticipation of a 8 final NUREG issued shortly thereafter.
9 ISAP objectives. When we started the program --
10 first, we want to try to improve safety in the most efficient 11 manner considering all of the constraints we have to work 12 with. We want to responsibly disposition a large backlog of
( 13 projects, that includes NRC projects, SEP, TMI, INPO 14 commitments, we want to try to establish a more effective 15 interface with the NRC, including a more enhanced 16 communication with them, to try to find a better way of 17 resolving licensing issues.
18 One of the objectives was to support cost 19 containment initiatives, improve resource planning, 20 facilitate ALARA initiatives, to facilitate an integrated 21 project planning on behalf of a multiunit facility, keeping 22 in mind we have four nuclear power plants, we want to try to {
23 resolve issues on an integrated basis.
O 4 ca^'ax^" xena= Let e eex v " eboet the riret !
l 25 bullet. Do you have a feeling that safety could be improved l l
ACE-FEDERAL REPORTERS, INC.
202-347-3hW) Nationwide Coserage 8(n336-6646 l
J
I 31630.0 l cox 66 l
1 or whether it needed to be or were you told by somebody? I j
2 am interested in the significance of that statement., j i
3 MR. BLASIOLI: I think in the past when a new NRC 4 requirement came out, it was assumed to increase safety. 'We j 5 wanted to get a better handle on how much each of the new 6 requirements does increase safety, prioritize them into which ]
7 ones have the biggest benefit first and address the ones that
(
I 8 have less benefit.
9 CHAIRMAN KERR: You didn't have an overriding 10 concern that your plants were unsafe? i l
11 MR. BLASIOLI: Definitely not.
12 CHAIRMAN KERR: Given a list of requirements, you O 13 wanted to try to understand their contribution to safety.
14 MR. BLASIOLI: Yes. We definitely had a l
15 distinction in our mind that there was difference that had to 16 be associated with the NRC requirements that were on the l 17 books at that time.
18 CHAIRMAN KERR: Thank you.
19 MR. BLASIOLI: Bottom line of the objectives and 20 why we were interested in ISAP, we are a public service 21 organization, we are held accountable for prudent decisions 22 for expenditures. We are looking for, and we think we found, 23 a process to make better decisions, contain costs, but to do
() 24 so without compromising plant safety.
25 MR. MOELLER: On the ALARA initiatives, you are
/(CE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coseraye 800-336-6646
31630.0 cox 67 1 thinking of ALARA in terms of all changes, et cetera, that 2 you might make at the plant. You are not thinking of just 3 exclusive on rad protection?
l 4 MR. BLASIOLI: We are zeroing in primarily on the 5 occupational dose.
i 6 MR. MOELLER: You are, thank you. I thought you 7 might be thinking ALARA -- it's similar to your first 8 bullet. Doing things in the most efficient manner. You 9 supposedly reduce occupational exposures in an efficient and 10 effective manner.
11 MR. BLASIOLI: Real briefly, I would like to get 12 into some of the ISAP methodology and the processes. When we
(*
~
13 first initiated and started working with the NRC on ISAP, we 14 had several meetings with them, a fair amount of 15 correspondence, and ultimately on January 31, 1985, the NRC 16 sent us a letter for both Millstone 1 and Haddam Neck which 17 outlined the scope of ISAP, what topics were going to be 18 included in ISAP, the kinds of topics that are included and 19 were included were NU initiated projects and various safety 20 concerns.
21 I would like to take the opportunity to address 22 the question that came up during Mike Boyle's presentation.
23 We have a nuclear safety engineering branch in our
() 24 organization which does look at the various new safety 25 concerns, LERs, we would be making recommendations to the ACE-FEDERAL REPORTERS, INC.
202-347 37(K) Nationwide Coserage 8(Xb3364646
i 31630.0 cox 68 1 plant management, those would be included in the program.
O 2 NRC initiated pr6jects, unresolved SEP issues, unresolved TMI i
3 issues, generic issues, just unresolved safety issues, 4 period.
5 MR. MICHELSON: Which unresolved safety issues are 6 you talking'about? Those that have actually been through the 7 full resolution process or those in the process yet?
8 MR. BLASIOLI: They would be several. They would 9 be those that had been all the way through the process, we 10 were trying to implement through a plant-specific basis as 11 well as those --
12 MR. MICHELSON: You also considered those in O 13 process to see how they might apply?
14 MR. BLASIOLI: Yes, that's correct. To quickly 15 run through the analytical ranking methodology, or ARM.
16 Each project is scored by four attributes, public 17 safety attribute, industrial safety, economic performance 18 attribute, which looks at the affects the plant modification 19 has on plant availability, plant efficiency, electrical 20 output put core thermal output, lastly personnel productivity 21 attribute, which looks at the effect the modification has on 22 people, equipment, overall environment. I 23 As we ranked each one of these attributes, we
() 24 would convert them to equivalent units, we would weight them 25 accordingly. We would sum them up for one integrated score.
ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 8(Kb336446
31630.0 cox= 69-1 MR. MICHELSON: Is this a subjective process?-
O 2 MR. BLASIOLI:- It includes both. subjective and 3 objective, to varying degrees, depending on the different-4 attribute.
5 Once we came up with a ranking.for particular 6 projects, the way we would then come up with an overall 7 implementation schedule would be to factor in,the various 8 factors that are listed on this slide. First one is external 9 impacts. What is the-source of the particular requirement.or 10 project that we are trying to initiate? Is it NRC,.is it 11 INPO, is it the state regulatory body? Whatever-it is, we 12 would evaluate and we would consider the impacts of not doing 5# 13 a particular project based on its origin.
14 The Arm rankings, obviously, have a very 15 significant factor in determining what gets done and what 16 doesn't get done. We would do the highest ranked ones first, 17 but based on these other factors, some of the low and medium )
i 18 ones would be put in there ahead of the higher onen.
19 ALARA goals, occupational exposure, I talked about 20 before. We do have corporate goals averaged over a 21 three-year period for how much man-rem exposure we want to 22 have on our plants. We factored that into what gets done at 23 a particular time.
() 24 MR. MOELLER: How do those compare with the INPO 25 goals for the utility?
ACE. FEDERAL REPORTERS, INC, j 202-347-3700 Nationwide Coverage 800-336-6M6
31630.0 l cox 70 We, in fact, use those INPO goals j 1 MR. BLASIOLI:
O -2 to come up with ours. I believe a general statement would be 1
3 that ours are generally lower than the average INPO goals. l l
4 MR. MOELLER: Thank you.
5 MR. BLASIOLI: The capital and expense budget, we 6 are concerned we properly maintain cost without compromising 7 safety. Equipment and material availability, lead times.
8 Something you may need, even though it's the number 1 or 9 number 2 ranked project, if it takes you 18 months to get a 10 particular piece of equipment, you have to factor that into 11 your schedule.
12 Personnel resources, always a factor. Outage O 13 duration. What we are hoping to do is get more objective for 14 future outage durations to base the duration on the actual 15 refueling that has to occur and work the projects around that 16 schedule. In the past what has happened is we take what we 17 have to do when we do an outage duration based on what has to 18 be done, not on the length of refueling.
19 Regional power needs. That is something very 20 recent. As a matter of fact, it's very important to us right 21 now. We have Millstone 1 shut down, we have Connecticut 22 Yankee, Haddam Neck coming down in a couple of weeks, Pilgrim 23 is shut down, Vermont Yankee is going to be coming down end O 24 of the first week in August, 25 When we started Millstone 1 outage and get back on ACE-FEDERAL REPORTERS, INC, 202-347 3700 Nationwide Coserage MXb336 4 46
-31630.0 cox 71-1 to prevent possible brownouts in Connecticut makes a big difference to us in scheduling our outages.
2 3 End product is unit implementation schedule.
4 There is a significant difference in our mind between ISAP 5 process and so-called " integrated living schedules." What we 6 do, based upon the Arm rankings and these other factors, we 7- have a threshold to draw particular topics from implementing 8 the plant at any point in time. So there is a concept that 9 we have built into the ISAP program, whereby we would propose 10 to the NRC, if it's an NRC requirement, or INPO, if it's 11 INPO, that there would be a particular topic that we would 12 not, in fact, implement.
13 I think we also believe that our process is more 14 sophisticated than a normal living schedule.
15 I would briefly go over PRA and summarize some of 16 the uses of PRA at Northeast Utilities and summarize the 17 overall status of where we are in our living PRA program.
18 One of the things, although it's not listed on 19 this particular slide, that we use PRA for, is in the public 20 safety attribute of ISAP. That has a major influence on the 21 overall outcome of a particular project. Plant design change 22 request. That includes both hardware and emergency operating 23 room procedures. We look at it to make sure we are
() 24 increasing safety and we are not decreasing it.
I 25 Design optimization. That's a front end use of ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(G346M6
31630.0 cox 72 1 FRA --
( .
2 CHAIRMAN KERR: Is it always clear where you are 3 or are not increasing safety. Are there close calls are or )
I' 4 is it generally straightforward?
5 MR. BLASIOLI: John, correct me if I am wrong, !
l 6 utilizing the PRA aspect of it, it would come out fairly l
7 clean in terms of quantifying whether it's breaking even, or 1
8 if you are going to be doing better. They are nodding in 9 agreement.
10 MR. MICHELSON: Are the PRAs that you are using 11 full scope PRAs in making these determinations?
12 MR. BLASIOLI: My next slides will get into that.
( 13 MR. MICHELSON: Okay.
14 MR. CAMP: Let me see if I can help you with 15 that. In most instances it's not a close call.
l
\
l 16 Deterministic criteria, in general, good engineering, serves 17 to make sure the design changes get you in the right l
18 direction. Occasionally we have found that when you do 19 things by the books, in terms of meeting applicable codes and 20 standards and design criteria, there's some inadvertent 21 things that did expect to happen. This tool enables us to 22 find them before we implement them.
23 CHAIRMAN KERR: Thank you.
() 24 MR. BLASIOLI: Design optimization, that's a front 25 end use of PRA. We used that to assure that the ACE-FEDERAL REPORTERS, INC.
202-347 37(K) Nationwide Coserage m)-336 4 16
31630.0 cox 73 1 unavailability of a particular design change is as low as 2 possible.
l 3 Tech spec changes. Very similar to the 50.59 ;
4 safety evaluation item listed up there. PRA lends itself 5 nicely to 50.59 evaluations. Prioritizing operator training i i
6 requirements. That's a pilot program for Millstone 3 right l
7 now. That's where we are looking at the dominant sequences 8 in the PRA to make sure that we are properly training the 9 operators.
10 CHAIRMAN KERR: Have you ever looked at the 11 question of whether one should do testing during the time the 12 reactor is operating as contrasted with trying to schedule
/'~T 13 all testing when the plant is down?
14 MR. BLASIOLI: I'm not real sure. John.
15 MR. BICKEL: John Bickel, Northeast Utilities. We 16 have look at that, there are some activities in that area.
17 There very clearly are what people call "high risk 18 surveillance." In other words, surveillance with which to 19 carry out and put the plant in a state that you are maybe one 20 relay, changing state, from a spurious scram. Doing those 21 tests and refueling obviously is a better idea.
22 However, you get into a point of test or tech spec 23 required. You would have a tremendous number of tech specs
() 24 that you would have to change in the overall effect of all of 25 that. It's a nontrivial task to try to evaluate where the ACE. FEDERAL REPORTERS, INC.
q 202-347 3W) Nationwide Coverage MKL336 %46
31630.0 cox 74 1 best balance is. What we have attempted to do is go after O 2 specific ones where we felt that the chance after a spurious 3 scram was great enough that it warranted backing off in 4 frequency. We have submitted individual types of tech spec )
5 changes'in those limited areas.
6 CHAIRMAN KERR: Seems to me at least once a week, 7 an example of somebody doing a test bumping into a cabinet, j 8 hitting a lead-in and causing a spurious scram, that is for 9 other population of plants rather than your own. But it j 10 seemed to me somebody should look into this. At least you 11 have looked into it some.
12 MR. BLASIOLI: Prioritizing QA inspection
, f~
13 resources. Again, we would use PRA, it's a pilot program, we 14 will be utilizing PRA to help determine what systems would 15 require or should require the -- our concentration of QA 16 resources.
17 Emergency procedures optimization. As we have, in 18 fact, found in different problems, as we have been going 19 through the PRA development, we are looking to our EOPs to f
20 see if we can't better optimize the way they are written to 21 help solve some of the particular problems that have arisen.
22 PRA program development. I would like to skip to 23 the 1981 bullet just to shorten the presentation a little
() 24 bit.
25 MR. WARD: We have got plenty of time. I mean, ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage NOM 36-6646
31630.0 )
cox 75 1 unless you are going to spend a spectacular amount of time on 2 the last chart. But if there is an essential point you i
3 wanted'to make -- !
I i
4 MR. BLASIOLI: There really isn't anything 5 significant for the first couple. I would like to talk about j i
6 1981, for the PRA task force. We formed a task force, and we <
7 came up with some recommendations that said we wanted to do i 8 PRAs for all four of our plants. We initially had decided to 9 do them based on the age of the plants, doing the oldest ones 10 first. As it turned out, in the process of getting the 11 operating license on unit 3, as I think you said on Tuesday, 12 we were made an offer we couldn't refuse. We end up doing 13 Millstone 3's PRA first. That was completed in 1983 as a 14 PRA.
15 In 1985, we subsequently started a Millstone level 16 1 PRA which was, in fact, completed in 1985. Similarly, in 17 1986, we completed a level 1 PRA for Haddam Neck, Unit 1 fire 18 PRA and Unit 1 fire PRA. In 1987, we had completed Millstone 19 Unit 1 flooding PRA and Haddam Neck PRA. We had initiated 20 Unit 2 level 1 PRA and Millstone unit number 1 level 3 PRA.
21 MR. MICHELSON: Clarify for me, what do you mean 22 level 3?
23 MR. BLASIOLI: That would include containment 24 consequences.
25 MR. MICHELSON: Full scope?
Ace FEDERAL REPORTERS, INC.
202-347 37(X) Nationwide Coverage 8(K)-336446
31630.0 cox 76 7
1 MR. BLASIOLI: Yes.
k 2 MR. AICHELSON: I thought full scope by definition 3 included fire. It was an external event.
4 MR. WARD: Level 3 means it goes to health 5 effects. Initially these are internal event PRAs. ,
6 MR. MICHELSON: Even a level 3 may not include 7 external events at all. That's what I want to make sure.
8 Which ones of these are -- Millstone 3 PRA does, in fact, 9 include external events. Millstone Haddam Neck's only 10 partially do in terms of the fire and the flooding PRAs.
11 MR. WARD: Millstone PRA included seismic -- I am 12 confused about that. Did the big PRA include all external O 13 events?
14 MR. BLASIOLI: Yes.
15 MR. CAMD: One point of clarification on this i
16 whole thing. We made the judgments about where to start.
17 The place to start was with the level 1 internal events.
18 Millstone 3 was a little bit of an anomaly in that we had to 1
19 do the whole thing, external events, all the way to level 3, l l
20 as part of a licensing process. That's what we did. Then it l 21 was a measure of taking the resources that we had and doing i 22 it in the logical sequence, so that we started with the older 23 plants, internal events, level 1. Now we are building on
() 24 that to include fires and floods. Long-term game plan is 25 level 3 PRAs and across the board.
ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 8(n33 MMS
31630.0 cox 77
-s 1 MR. MICHELSON: Including all external points.
U 2 MR. CAMP: Only exception is seismic, PRA 3 Millstone 3 includes seismic, we have not a firm commitment 4 yet to include seismic on the other units. Again, that's a 5 measure of where you want to devote the resources first.
6 MR. WARD: Now, something about that that bothers 7 me, Millstone 1, as I understand it, the isolation, as 8 Dr. Remick pointed out, the isolation condenser is a valuable 9 thing to have. In Millstone 1 PRA shows that. You said that 10 you -- although you haven't done a PRA for Millstone 1, you 11 did, under the SEP program, beef up the isolation condenser 12~ to SSE requirements, I guess.
13 But as I understand it, the later BWRs, GE BWRs, 14 did not incorporate the isolation condensers of the style 15 that is in Millstone 1. I may be wrong about this, but the i
16 impression I had is a primary reason for that is there is 17 something big and heavy that's up real high. It's difficult 18 to build a -- to be seismically resistant. I guess I have to 19 question whether a seismic PRA for Millstone -- if that's the 20 case, I have to question whether seismic PRA for Millstone 1 21 really is unimportant or relatively unimportant as you say it 22 is.
23 MR. CAMP: I will take a stab at trying to answer
() 24 that. But I think there was a clear recognition of that 25 particular fact in the course of doing the SEP. Reflecting ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(Xh33M646
i 31630.0 cox 78 1 on all of the lessons learned in SEP, that seismic topic 3-6 7_
2 occupied the most resources for all of the plants that 3 participated, including Millstone, and resulted in the 4 expenditure of the most dollars in terms of hardware fixes 5 that were fully ~ implemented.
6 While the SEP process did not involve a seismic 7 PRA in the sense that we have been talking here, like was 8 done in Millstone 3, it was a rather significant effort, with 9 the tools we had at that point in time, to take a look at 10 what upgrades passed the thresholds that were established in 11 SEP about what is worth fixing in the grand scheme of things, 12 alluding to your remarks earlier about the aspects of the SEP O 13 process. Our decision on the seismic PRA to date reflects 14 the studies and efforts that have been put into it thus far.
15 Even though there wasn't a PRA back then, 16 everybody recognized that a large component that high up in 17 the air was a vulnerability from a seismic standpoint.
18 That's why when you look at it today, the supports on it are 19 unbelievable. I am not sure we could do much more with it 20 anyway. It's kind of a pragmatic thing at this point in 21 terms of where you are going to get the most bank for your 22 buck, or in this case, where you are going to get the most 23 mileage out of your PRA resources. Current decisions reflect
() 24 our judgments on those points to date.
25 MR. WARD: Specifically, do you think that your --
Ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coterage MXk33MM6 ,
)
l l 31630,0 l' cox 79 1 the isolation condenser would survive an earthquake two or O 2 three times the SSE? I mean, as I understand it, most 3 seismic PRAs showed that the risk in seismic comes at rare 4 events that are in the range of two or three times the SSE.
l 5 So what do you know about your isolation condenser? ,
l 6 MR. BICKEL: John Bickel for Northeast Utilities. ]
I; 7 I don't think we could claim that we know the exact !
8 fragility, the actual capacity of the thing. There's stuff 9 that exists. I think the thing that Rick is trying to allude 10 to is a pragmatic approach. We try to use PRA as an ;
i 11 engineering tool 'n making what I would call some very j 12 concrete decisions. In the seismic arena you have, I think l
13 it differs very much from internal flooding and it differs 14 very much from fire analysis, and that the level of agreement 15 among what I call knowledgeable experts and hazards is more 16 than an order of magnitude apart. If I am to do a seismic 17 PRA, and I am going to commit resources to it, we don't know 18 how to use the thing, because we don't know what the numbers 19 mean. There is a considerable amount of disagreement about 20 the usefulness of it.
21 I am basically sinking a lot of money and time 22 into something that I cannot use to help support 23 decisionmaking on the units that I have.
24 In five years, if there was enough agreement that 25 I could weigh the risks of a seismic earthquake with some ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8 6 336-6646
31630.0 cox 80 J degree of certainty versus the risks from a stuck open relief 2 valve and some failures of decay heat removal systems, and I 3 could trade off one for one, I would then have a way of using 4 the decisionmaking process. Right now it doesn't fit in.
5 It's a problem. If like five years from now there was a 6 great level of concurrence on how you equate seismic risk 7 versus internal risks, we might have a different -- maybe a 8 slightly different viewpoint. .Right now, I think what we are 9 taking is a prudent, go slow approach, reevaluate it in a few 10 years.
11 MR. WARD: It seems to me there is something 12 unique at Millstone 1. If.I understand it, you depend on the O 13 isolation condenser as an important part of your system. It 14 cuts off a lot of event trees or whatever you call them.
15 On the other hand, that isolation condeneer 16 apparently is particularly vulnerable to an earthquake. So 17 it would seem you wouldn't have to do a -- because of its i 18 importance in that vulnerability, it might be useful simply 19 to look at the fragility of that particular system without 20 doing a full-blown seismic.
21 MR. SCHWINK: They have said they analyzed it, 22 they beefed it-up. The SSE for many people is considered to 23 be enough. That itself is extremely improbable, though you
() 24 don't know the probability with certainty and that keeps l
25 bugging Dave every so often.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
~ - - - -- ---_
31630.0 cox 81 1 MR. WARD: Well, typically, we want to talk of a
() 2 safety goal, we would talk about core melt likelihood of 3 having 10 to the minus 4 or less. I don't know what they 4 guess is the return period for their SSE, but it's probably l
5 less than 10,000 years.
6 MR. SCHWINK: At that time that means some of 7 their supports might even get out of the elastic range, which j
8 they probably designed it for.
9 MR. WARD: That's why they designed it.
10 MR. SCHWINK: I have never bumped into a 11 mechanical engineer who was willing to go to a plastic 12 design. They probably found a unique set but I would bet a 13 fair amount of money that the design was for the elastic 14 range.
15 MR. WARD: General Electric has not used this 16 design of isolation condenser in later plants.
17 MR. SCHWINK: Part of it may have been cost as 18 well as seismic arrangements. They have been cutting costs a 19 lot.
20 MR. EBERSOLE: I think we mentioned to a certain 21 extent, that in ways like BWR, if you don't have driving --
22 what do you do if you have temperature driving gauges to get 23 to the atmosphere, like BWR, you have to fall back on 24 support.
(]}
25 Now, the advantage, and I think I have mentioned
/\CEJPEDERAL REPORTERS, INC.
202-347-37(o Nationwide Coverage 8(n 336-6646
_ _i
31630.0 cox 82 1 this, there is a better way, the primary vessel, that it 2 breathe through the suppression, atmosphere is a predamaged 3 clean stream, these folks haven't looked at that, they said, 4 but they would; am I right? That would, I am sure, greatly 5 enhance the chance to make this Millstone 1 look a hell of a 6 lot better. I believe you said you did get some benefit out
'7 of it, maybe 50/50 heat flow. You get something out of it, 8 but not much.
9 MR. BICKEL: That is correct.
10 MR. SCHWINK: When you are finished with that, I 11 have a different question.
12 MR. WARD: I am finished.
13 MR. SCHWINK: As I recall, after looking at what 14 you could do against salt sprays and so on, you committed to 15 take Unit 1 off the line next time there was a hurricane 16 coming up the coast, or severe storm, because you couldn't 17 avoid a sudden outage. Is that true of the whole site, all 18 three plants?
19 MR. BLASIOLI: We have procedures in place for all 20 three plants that would require us to take it off line if a 21 tornado, hurricane or particular external event was coming j 22 towards.
1 I
23 MR. SCHWINK: That external event is something
() 24 that will blow seawater over your switchyard. Is that
{
25 basically it?
ACE FEDERAL REPORTERS, }NC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0 cox 83
_s 1 MR. BLASIOLI: Even if we are expecting large s
2 losses of power lines, we need that as well. If it's going 3 to do damage to the site, we expect to lose off-site power 4 into the site, yes. It could be all of those.
5 MR. EBERSOLE: I don't think you mentioned in the 6 full committee hearing anything about your great gas turbine, 7 your airplane engine. Does it only serve one unit?
8 MR. BLASIOLI: Yes, Millstone 1.
9 MR. EBERSOLE: Others you ride on diesels?
10 MR. BLASIOLI: Yes.
11 MR. EBERSOLE: Do you have unit interties on l
12 electrical systems?
13 MR. BLASIO!I This cutage, we are in fact putting 14 in a tie between Millstone 2 and Millstone 1?
15 MR. EBERSOLE: Will it be subterranean? I 16 MR. BLASIOLI: I don't know. Do you know?
17 MR. QUINN: I believe most of it is internal to 18 the building. The switch gear is out in the yard and it does 19 run shortly into the building.
20 MR. EBERSOLE: Fine.
21 MR. BLASIOLI: Future activities for our PRA l
22 program. This is a five-year plan that you are seeing in 23 front of you. We do intend to complete Millstone 2, level 1,
() 24 PSS, during 1988. That's the only level 1 PRA that we don't 25 have currently completed. Schedule update for the existing ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 8(0 336-(M6
F 31630.0 cox 84 1 PRAs, Millstone Unit 3, revision 1 by August of '87. Haddam O 2 Neck,-revision 1 by December of '87, Millstone.
3 2 by February of 1988. We have one on Millstone 1 4 that was December of 1987. .We intend to upgrade all our'PRAs 5 to level 3. That will help us to implement the severe 6 accident policy at the same time and safety. goals policy of 7 the NRC.
8 I would like to briefly summarize some of the-9 experiences that we have had with'ISAP in the' pilot program 10 for Unit 1 and Haddam Neck. Some of the insights that we 11 have obtained through the implementation of ISAP.
12 As we were telling the subcommittee on Tuesday,
( 13 there are significant risk contributors that we have 14 identified. We have evaluated them and appropriate 15 modifications have been proposed. We have used the ISAP 16 process to help us decide how important some of the 17 particular identified risk contributors were and, in fact, 18 what some of the proposed modifications needed to be 19 prioritized for. In the case of Unit 1, the items we list I 20 here, long-term cooling issue, which, as we talked about, I a
21 think was talked about this morning by Mike, was a major V
l 22 contributor to core melt frequency. We have, in fact, been 23 able to revise some procedures and we have reduced the
(} 24 contribution to core melt frequency some degree, and there 25 are additional studies that are going on to allow us to ACE-FEDERAL REPORTERS, INC.
202-347-370() Nationwide Coserage MK336446
1 31630.0 l cox 85 1 1 further reduce the CMF.
73 k-] 2 Degraded grid voltage issues. There's a case i
]
i 3 where we were in the process of implementing a design, and in 4 fact we uncovered through the PSS and ISAP process, the fact 5 that that particular design would increase the likelihood of 6 having a station blackout event. We then went back to the 7 drawing board. We have since redesigned those particular 8 mods.
9 MR. MICHELSON: How degraded a voltage are you 10 talking about?
11 MR. BLASIOLI: Different thresholds, 90 percent, 12 85 percent, 70 percent. Different subpoints.
O 13 MR. MICHELSON: Did you study these by PRA 14 techniques?
15 MR. BLASIOLI: We studied the modifications that 16 we were going to be making to the plant to resolve the 17 issue. In doing so we were making things more complicated.
18 MR. MICHELSON: " Degraded" doesn't mean lowered 19 voltage?
20 MR. BLASIOLI: Yes, it does.
21 MR. MICHELSON: How do you know the effects of 22 lower voltage? How do you know the effects of 70 percent 23 voltage on your equipment? Do you have some test results or
() 24 something that you use for all of this?
25 MR. BLASIOLI: We have test results that would ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336-6M6
31630.0 cox 86 1 tell us below what value we no longer have qualified
() 2 equipment or we could be potentially damaging that
/
j l 3 equipment.
4 MR. MICHELSON: This was not a systems interaction 5 question but damage to equipment question?
6 MR. BLASIOLI: If you let the voltage get too low 7 you could in fact degrade safety-related equipment.
8 MR. MICHELSON: You didn't look at it from the 9 systems interaction as to how the equipment may malfunction 10 under 70 or 100 percent.
l 11 MR. CAMP: Maybe I can help a little bit. This l
12 was a generic NRC issue we were implementing on a O 13 plant-specific basis. I think the thrust of the generic 14 issue was there was a recognition where the grid would be 15 available but at a voltage level where you could be damaging 16 ehe aquipment and your protective systems wouldn't disconnect 17 you soon enough to go onto the on-site power supplies. So, 18 in the course of doing that, we engineered a design that l 19 again, by the normal deterministic criteria, everything 20 looked fine.
21 Then we used the TSS methodology to help look at 22 it and found it wasn't a substantial move in the wrong 23 direction but was a move in the wrong direction so we don't
() 24 want to do it.
25 MR. EBERSOLE: With respect to Millstone 1, I ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8 % 336-(646
-31630.0 cox 87 I
1 think we went astray yesterday on one matter. We were asking O 2 you about suppression bypass. We learned a few things, one l
3 of which was you said that you thought the issue was focused 4 in the possible sticking of the vacuum relief valves, not in )
5 the breaking of pipes, carbon steel pipes at the roots, doing 6 low pressure, rumbling, laboratory discharge, when you get 7 down to 100 pounds or so.
8 I think that was -- that went astray for this 9 reason. You have put tail pipes, like all.the-BWRs, on your 10 safeties. You convey the discharge through the dry well now, 11 and there is no atmospheric scrubout. So your vacuum relief 12 valves:will never work. They don't have to do anything. You 13 have to ask whether those, contrary to the German design, are 14 good enough in single pipe configurations. They use double 15 walls. Of course, MARK-III uses better than double walls, I-16 think they are buried in concrete. Yours hang down 17 pendulously and better not break off because Brookhaven says 18 you won't last long if they do.
19 MR. WARD: Do you want a response, Jesse?
l 20 MR. EBERSOLE: I just want to have a disagreement, l
21 if there's one there. Anybody disagree?
22 MR. BLASIOLI: No disagreement.
23 CHhiRMAN KERR: What is the issue? I couldn't O 24 ==aer te=a it-25 MR. EBERSOLE: Insufficient bypass. If you breach ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0 cox 88 1 the suppression bypass process, you are gone.
.C--.
2 MR. SIESS: Talking about failure of the lines 3 from the relief valves.
4 MR. BICKEL: Jesse, could we add one possible 5 thing.
6 MR. EBERSOLE: Yes.
1 7 MR. BICKEL: One of the points that I think ought 8 to be recognized is that -- I think we touched on it, I 9 think, very briefly the other day, was that first of all, the 10 likelihood of a -- what you are talking about is a scenario 11 where the containment is very quickly overpressurized by 12 steam and that you basically bypassed the pressure
' O 13 suppression capability. There are, in our plant-specific 14 emergency operating procedures, there is a very direct 15 procedure addressing exactly what to do on that. We !
16 generally believe that the procedure would work. That 17 procedure directs the operator. If it becomes apparent that 18 the pressure is increasing, he has got this situation. He is 19 instructed to open up all available SRVs, open the main 20 condenser. We have one horrendously large condenser in that 21 plant. In that action, you essentially divert all the steam 22 to places where the pressure will be controlled. If you 23 execute that procedure, you can -- you do avoid the bursting
() 24 of the containment. If that occurs, that's a bad event. We 25 recognize that. What we are saying the other days, those
/\CE-FEDERAL REPORTERS, INC. ,
202 347-3700 Nationwide Coverage 800-33M446
[
31630.0 cox 89
_ 1 types of scenarios are likely to cause, as a result of a LOCA V
2 event, coupled with a failure of the vacuum, break your 3 valve, versus a lifting of a relief valve coupled with 4 failure of the pipe discharge pipe above the water line.
5 MR. EBERSOLE: In order to say that, you have to 6 say a LOCA is more likely than failure of a steel pipe in a 7 wet water space.
i 8 MR. BICKEL: The LOCA we were talking about is a 9 pipe that is normally pressurized. LOCA you would be talking 10 about would be one in the normally low pressure end of the 11 discharge pipe.
12 MR. EBERSOLE: Synthetic vibration.
13 MR. MICHELSON: How long would it take to 14 implement your procedures? You have to do some kind of 15 gerryrigging on the valves or that sort of tuing, don't you?
16 MR. BICKEL: All they have to do to reopen the 17 valves is an equalizing valve.
18 MR. MICHELSON: You have an accident from high 19 containment pressure. You have to look at your MSIRs on an 20 accident. This is an accident. There's pressure, that makes 21 it an accident. It takes a few minutes to figure out, unless 22 you have a quick flip switch or something. Brookhaven says.
23 MR. BICKEL: Agree, but, still, if you have one of
() 24 these tail pipes busted, one of six, there are five 25 remaining.
ACE FEDERAL REPORTERS, INC.
202-347-37(K) Nationwide Coverage 8(KL3364M6
'31630.0 cox- 90 1 MR. EBERSOLE: I certainly endorse this, and I O 2 would like to reverse the idea of lock-in. I'would rather 3 blow the turbine all away.
4 MR. MICHELSON: Did you look at the Brookhaven
.5 calculations? Did you agree with them?
6 MR. BICKEL: I am not familiar with the Brookhaven 7 calculations.
8 MR. MICHELSON: You have made your own though?
9 MR. BICKEL: We are working on our own.
10 MR. MICHELSON: I was surprised at how fast the 11 containment was pressurizing.
12 MR. BICKEL: It wouldn't surprise me if the folks O 13 from Brookhaven came up with a rate that fast.
14 MR. EBERSOLE: You mentioned spray. Do you have 15 that?
16 MR. BICKEL: We do but their limitations on 17 operations are such -- they are very restrictive.
18 MR. EBERSOLE: Let me tell you what happened when 19 you looked at this in generic issues.
20 MR. QUINN: Jack Quinn from Northeast Utilities.
21 Part of the procedures also, if you get -- before you get two
'22 pounds in the torus, in the dry well, one of the procedures 23 is to start the 2-inch line, and stoking the valves, before
() 24 you get to that.
25 MR. MICHELSON: The calculations show you blow the ACE FEDERAL REPORTERS, INC, 202-347-3700 Nationwide Coverage 800 336-6646
31630.0 Cox 91 1 containment about 10 minutes. This is three times design, rh .
2 whatever they use, three or four times design. -It's unreal.
3 MR. QUINN: Another mitigating factor, lines are 4 seismically designed and the penetration where they come 5 through the down cover, poke through that area, is well 6 braced with a wagon wheel type design to prevent those from 7 shaking at that point. But whether that exceeds this 8 vibration, I don't know. That isn't just a single point of 9 contact.
10 MR. WARD: Paul, let's go ahead, please.
11 MR. BLASIOLI: Next item I wanted to quickly 12 discuss, we used to have liquid chlorine tank cars at 13 Millstone. We have since replaced those with a sodium 14 'hypochlorite system due to the high ranking that that 15 particular project received in ISAP. The Haddam Neck plant, 16 one thing that came out of PSS was loss of MCC-5. That's not 17 a redundant motor control center. The significance from a 18 core melt frequency point of view was evident when we l 19 completed the PSS. We are working towards ways of resolving 1
20 that issue, in fact, some plant modifications that will be 21 going partially into place in '87 as well as '89 outage will 22 go a long way towards reinvolving that issue.
23 MR. MICHELSON: Nonredundant motor control 24 center?
25 MR. BLASIOLI: Yes.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 l
31630.0 cox 92
- 1 MR. MICHELSON: Safety-related one?
(v)
~
2 MR. BLASIOLI: Yes. It has redundant power 3 supplies to it, but there are single failures. It was part 4 of the regulatory basis design of the plants and approved 5 from day one.
6 MR. MICHELSON: Apparently you didn't have to have 7 redundancy in those. days.
8 MR. BLASIOLI: This was back in mid '60s.
9 MR. MICHELSON: I thought we always had 10 redundancy.
11 MR. BLASIOLI: There was an exemption.
12 MR. MOELLER: When you mentioned chlorine as a 0 13 significant contributor, is that through the control room?
14 MR. BLASIOLI: Off-site consequence as well as 15 on-site consequence. Yes, it's both.
16 MR. MICHELSON: As I recall, Haddam Neck also has 17 some interesting fire protection problems in the control 18 room. Are you going to talk about those at any time in part 19 of this?
20 MR. BLASIOLI: I hadn't planned on talking about ,
l 21 it. I 22 MR. CAMP: We had no specific plans to talk about 23 it. I can tell you that the biggest single long-term
() 24 solution involves the installation of another switch gear 25 building which will be finished in 1990.
ace FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-336 6646
_ - - _ _ _ - - - - - _ _ _ _ _ - _ _ _ _ _ _ _ ~
31630.0 cox 93 1 MR. MICHELSON: You are in the process of 2 designing and constructing a separate building?
3 MR. CAMP: That's correct.
4 MR. MICHELSON: Right now it still exists?
5 MR. CAMP: Significant modifications in the whole 6 Appendix R area. This will be the last element of wrapping 7 it up.
8 MR. BLASIOLI: Another area that was discovered 9 through the ISAP and PSS process was the small and medium 10 break LOCA issues at Haddam Neck. We discovered there were-11 some very small breaks in the Loop 2 lake; those will be 12 partially corrected this outage and completed in the 1989 O 13 outage.
14 Demineralized storage tank issue, modifications 15 are about to be implemented. We met all the standards that 16 we typically would need to meet. When we processed that 17 through the PSS and ISAP we found we were increasing the 18 unavailability of main feed and auxiliary feedwater. In 19 fact, we have since decided not to implement that particular 20 change.
21 We have currently Haddam Neck PCB oil-filled 22 transformers. We will be replacing those transformers during 23 the upcoming outage.
O 24 xa x c"8' son: er rer1ec1 9," whet a v= ee=7 25 Are these inside of vital areas?
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0 J' cox 94
,_, 1 MR. BLASIOLI: Yes, they are.
u 2 MR. MICHELSON: Did the NRC say they had to be 3 replaced, or did you just decide to? What is the hazard, 4 safety hazard?
5 MR. BLASIOLI: Personnel hazard as well as a fire 6 hazard.
7 MR. CAMP: These transformers are located in the 8 switch gear room. They could be contributing to a fire in 9 that area which would have very significant consequences.
10 MR. MICHELSON: You can use fire-resistant oils 11 besides PCBs?
12 MR. CAMP: We have just opted for the course of 13 action, the dry type transformer.
14 MR. MICHELSON: You are not the only plant in the 15 country that has this particular problem. I can see that's 16 the right direction to go, but I was curious that's your 17 rationale.
18 MR. CAMP: Goodly amount of it dealt with the 19 probabilistic treatment we got from the PSS, where it was the 20 single most important risk contributor from a fire.
I' 21 MR. MICHELSON: Fire or explosion that you are 22 worried about. The transformer literally blew up?
23 MR. BICKEL: Like a disbursive fire, if you have a i l
() 24 face to face fault, the oil overheats.
25 MR. MICHELSON: Thing blows apart.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0 cox 95 1 MR. BICKEL: But the main thing, we were O 2 concerned, not the explosive business, but the fact that it 3 was going to disburse potentially flaming oil and other 4 switch gear.
5 MR. MICHELSON: Oil is supposed to distinguish- --
6 flash point is so high it's supposed to distinguish by the 7 time it lays down.
8 MR.'BICKEL: We had talked with over own fire 9 protection organization. I guess from the NFPA. They have l 10 very' carefully pointed out to us, PCB oil, the right 11 temperatures, it will ignite and it will burn. It has to be 12 a little bit hotter.
( 13 MR. MICHELSON: I don't agree with your decision 14 at all. It's interesting because some other people around 15 the country have similar potential.
16 MR. BLASIOLI: We obtained it by doing the fire 17 risk and internal flooding analysis.
18 Some additional insights. One of the key things 19 that we discovered as we were going through ISAP is the 20 ability to integrate various resolutions from different.
21 selected topics, whether they be NRC requirements or NU 22 initiated topics. Good examples of those for Millstone Unit 23 1, are the following. Containment vent and purge issues. We
() 24 had four or five issues, again, combination of our own 25 initiated ones, as well as NRC requirements. We were able to ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
i
-31630.0 cox 96 1 come up with an optimum resolution whereby we are only G
.O 2 implementing two fixes, instead of five. We believe that we 3 have, in fact, resolved all five of the issues.
4 MR. MICHELSON: Are you going to continue to purge 5 and vent the containment during normal operation?
6 MR. BLASIOLI: The fix we are putting in place in 7 this outage now, we already have a dry well compressor system 8 which has already reduced the amount of purging we have to 9 do. We are tapping into that system now and we are putting 10 in a torus pump, whereby we can take suction from the torus, 11 go into the dry well and TAP back in. Once we solved the 12 oxygen issue that was the 1-pound dry well and torus. We 13 were able to meet that in tech spec modification. You have 14 eliminated the need to vent and purge which helped resolve 15 some of the other issues.
16 MR. MARK: That last item, I don't quite see the 17 connection.
18 MR. BLASIOLI: Tornado missile / electrical 19 backfeed?
20 MR. MARK: Yes.
21 MR. BLASIOLI: I can skip the others --
22 MR. MARK: If you are going to, go ahead.
23 MR. BLASIOLI: That was an SEP topic, when we
() 24 evaluated in ISAP, we determined it to be relatively high 25 value, which was to assure a tornado missile electrical ACE FEDERAL ' REPORTERS, INC.
202-347-3700 Nationwide Coserage RKb336-M>46
. ~ - -
31630.0 cox 97 1 supply to the condenser system. We did not at that time take O 2 credit for electrical backfeed between Millstone 2 and Unit 2 3 based upon Appendix R fire protection reasons. We have now 4 gone back, reevaluated the ISAP project for the SEP topic, 5 and have shown that we have significantly reduced the 6 significance of that and no longer are planning on having a l 7 missile-protected system.
8 MR. MARK: Do I have it clear then there was some 9 possibility of a tornado missile interfering with the 10 electrical interconnection?
11 MR. BLASIOLI: Yes.
12 MR. MARK: That you have paid attention to?
O 13 MR. BLASIOLI: Yes. The only one I want to go 14 over on this slide would be the BWR water level 15 instrumentation /SPDS. Instrumentation we currently had the 16 in the plant, there was the potential for the operator not 17 being able to recognize reference light flashing. We had 18 originally scored the vessel water level instrumentation 19 relatively high. But what we were able to do was by virtue 20 of the SPDS design, provide information to the operator so he 21 could immediately recognize reference light flashing. We 22 have since decided that the significance, again, of the 23 vessel water level instrumentation was not as high as it was )
() 24 in the past.
25 MR. MICHELSON: What does he do if he recognizes I
ACE FEDERAL REPORTERS, INC. I 202-347-37(X) Nationwide Coserage 800-3364M6 I
'31630.0
- cox 98
~
1- the' flashing?
OL 2 MR. BLASIOLI: I believe the directive is to flood 3 the vessel'up.
4 MR. MICHELSON: Not everybody has a flashing 5 indicator either. Is that a prevalent thing in the industry 6 'now?
7 101. BLASIOLI: What was the question?
-l 8 MR. MICHELSON: The flashing on the' level 9 ' instrument.
10 MR. BICKEL: John Bickel for Northeast. All it 11 really.is, pressure temperature correlation. It tells you if
.12 you are at saturation.
O 13 MR. MICHELSON: Runs through a computer and
's 1
14 signals an alarm.
15 MR. BICKEL: My understanding is it puts a little 16 tag on things and starts to roll or flash.
17 MR. MICHELSON: I wonder if other people are doing 18 that.
19 MR. EBERSOLE: Is that the end of that?
20 MR. BLASIOLI: That slide.
21 MR. WARD: Better move on, Paul.
22 MR. EBERSOLE: I have one important point. I 23 don't think we mentioned this funny regime that Diablo Canyon ;
.() 24 is in trouble with now, the modes, didn't we go into that, 4, 25 5 and 6 where we find what amounts to an astonishing ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
31630.0 cox 99 1 potential for core melt and the shutdown mode, because O 2 everything is informal, jury-rigging, untested, 3 unformalized. It's a big mess. None of the investigative 4 search processes, they have looked hard and organized more 5 into this, it's coming down the pike now. There is a big 6 letter coming out.
7 MR. BLASIOLI: We are aware there are activities 8 going on. We would consider looking at whatever the NRC 9 issues or INPO or the industry issues.
10 MR. EBERSOLE: Letter on the way out, hung up, 11 perhaps, by CRGR but perhaps CRGR is better than it once 12 was.
O 13 MR. BLASIOLI: I believe Jack Quinn from Unit 1 l
14 engineering has an answer.
15 MR. WARD: Would you come to the microphone, 16 please. Even though Mr. Ebersole doesn't, you should. l 17 MR. QUINN: Just about the comments you made about 18 tubing or water level, we do not use any type of indication 19 during that period. We used installed, it's recalibrates for 20 an open vessel or normal lower level emergency.
21 MR. EBERSOLE: That's one of the better aspects of 22 a boiler, BWR. You have a bunch of BWRs.
23 MR. WARD: He is asking about your other units.
"a ou ""=
O 4 e orry- 1 aoa't *= -
25 MR. WARD: Paul, we have about 10 more minutes.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(XL336 (646
31630.0 cox 100 1 MR. BLASIOLI: Okay. I will go through this ;
() 2 quickly then. As far as benefits go, we believe that ISAP 3 has offered safety improved in most efficient manner, better 4 integration and review of plant-specific issues, allowing us 5 to review and evaluate generic topics on a plant-specific j 6 basis, and we have established, we believe, a more effective 7 regulatory interface, which I will be getting a little bit 8 more details in a second. We have improved utility and NRC
)
9 resource management which I think is critical to both us and j i
10 the NRC. We believe ISAP represents a win win situation for )
11 both NRC and Northeast Utilities. ,
I 12 As far as NRC and NU having interface, we have a O 13 framework for enhanced communication between NRC and 14 Northeast Utilities. Movement towards a regulatory interface 15 where utility and NRC priorities are converging. We have 16 established a forum for better resolution of technical safety 17 issues, the program is ongoing and improving at all times.
18 I would like to capitalize on something Dr. Siess 19 said on Tuesday, the ISAP process is not only progress but 20 philosophy. I think that's the key between ourself with the 21 interface with the NRC. It's a mutually -- philosophy that 22 we both mutually agree.
23 Future NU ISAP activities and conclusions.
() 24 Something that's ongoing right now, as a matter of fact, is 25 the institutionalization of the ISAP process in-house. We ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0 cox 101 1 wanted to become a way of life, normal way of doing business
'O 2 within NU for all f0ur of our plants. We intend to further 3 develop NU's living PRA program. We intend to integrate 4 resource planning for all four of our nuclear plants. We 5 want to expend resources in the most efficient manner across 6 the board for all four plants.
7 . Conclusions, I guess the bottom line is we would 8 like to continue to expand ISAP, continue Haddam Neck and 9 Millstone 1 using methodology in-house and expanding that l 10 same methodology to Millstone 2 and Millstone 3. We would l 11 request the NRC to continue support for implementation of 12 ISAP. That concludes my planned presentation.
13 MR. WARD: Okay. Carlyle.
14 MR. MICHELSON: Would you clarify again, you had a 15 full scope PRA for Millstone 1; is that right?
17 MR. MICHELSON: Full scope, external events. What 18 confuses me a little bit, I thought I heard from the Staff 19 they expected to have a full scope PRA.
20 MR. BOYLE: No, we do not.
21 MR. MICHELSON: What kind of PRA do you require?
22 MR. BOYLE: We would like the best PRA we can 23 get.
() 24 MR. MICHELSON: If you don't use a full scope PRA 25 approach, you don't begin to address flooding, fires, ace. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3364M6
~
1 31630.0 i cox 102 J l external events other than earthquake and so forth. You
() 2 don't get into the true system interaction.
I 3 CHAIRMAN KERR: I think that normally means, level q 4 1, you go up to core melt. Doesn't have anything to do with 5 initiating event.
6 MR. MICHELSON: That's right. That's why I asked, 7 did they include external events. The answer was no.
8 MR. SIESS: You are not asking the right 9 question.
10 MR. MICHELSON: What should I ask?
11 MR. SIESS: Do you want to include external 12 events.
O 13 CHAIRMAN KERR: Yes. That's the question to ask.
14 MR. MICHELSON: The answer is no, it's not 15 required under the ISAP program to recognize external 16 events.
17 MR. BOYLE: In those cases we use subjective means ,
)
)
18 to evaluate. We don't have a PRA. We recognize it's there. i 19 And in the prioritization of the issues we understand these )
I 20 things and have to make engineering judgment. j 21 MR. MICHELSON: I would like to see how you
\
22 understand these things without doing a PRA. l
)
23 MR. SIESS: A lot of these requirements are not l
() 24 related at all to seismic or flood. j 25 MR. MICHELSON: System interaction is very much, j ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage MKk33M646 I
31630.0 cox- 103 1 of course, an external event problem.
2 MR. SIESS: That's only one generic issue.
3 MR. MICHELSON: Just one of an examplo --
4 MR. SIESS: The object isn't the PRA. Object is 5 USI, generic issue and requirements of prioritizing. The PRA 6 is only one, except towards the ISAP.
7 MR. MICHELSON: System interaction happens to be 8 an issue that is called a generic issue. It's well 9 understood what the problems are of systems interaction, PRAs 10 claim to do it on internal events, but not on external events 1
11 as I understood the PRA people.
12 MR. BOYLE: I think you should recognize that the O 13 ISAP Commission paper is not yet before the Commission.
14 MR. MICHELSON: We don't know what you require 15 yet. Thank you. That takes care of it.
16 MR. WARD: Paul, thank you very much. Let me -- I 17 appreciate your presentation. Let me just take a couple 18 minutes and turn to the committee, now. I have two questions 19 or two points I would like you to think about or comment on.
20 First, I propose that we do write a letter on this topic at i
21 this meeting on the -- I think it would be useful to tell the I I
22 Commissioners what we think about this process, so I propose l 23 we write a letter and I would like to have your input now and 1
() 24 if anybody has something to say about that.
25 The second thing is, I would like you to consider ace-FEDERAL REPORTERS, INC.
-4 202-347-3700 Nationwide Coserap MO-33MM6
w 1
31630.0 cox 104 1 what the task of the committee should be, or this ISAP
( k' subcommittee should be.
2 When the SEP process was going on, 3 we had a subcommittee, and, as a matter of fact, it went full 4, committee, and also reviewed the SEP report for each plant, i 5 and I think probably commented to the Commission for each 6 plant that went through the SEP process. We haven't done 7 that here. The subcommittee didn't review this first -- what 8 was the one we had, Millstone 1 ISAP in any detail. We don't ;
9 have any schedule right now to review the next one, Haddam 10 Neck 1 in any detail. I guess I would like to hear from the 11 committee what its charge is. Do you think we should?
12 MR. SIESS: We don't review the implementation of 13 USIs and GIs on plants now, the only difference between the 14 way they are being implemented now and the way they would be 15 implemented for ISAP differences, we learn from generic items 16 subcommittee that the Staff didn't even know a lot of them 17 hadn't been implemented. I think under ISAP you darn well 18 know where they stand on the list and what is being done 19 about them. Under ISAP, some of the items can go away 20 because of the integrated solution. That makes it different 21 than the enes tossed out there and everybody do everything.
22 I would think from time to time we might want to look at 23 where there were tough decisions about going away, whether
() 24 the Staff wants to bring them in. But I don't see how we can 25 look at generic implementation of USIs and PRAs on every ACE FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage MKK3E6M6
'31630.0 cox 105 1 plant, f3 V 2 MR. EBERSOLE: We had an enormous tabulation on 3 what's been done?
4 MR. SIESS: This was the Simms report, which I 5 mentioned. Staff is now trying to find out the status of 6 implementation of generic issues and USIs, and putting it on !
7 the computer. First thing they found out was a lot of them 8 hadn't been implemented. They didn't know it. They hadn't S asked. They didn't say that was good or bad, now they are 10 trying to track them. The system will work better without i
11 our attention.
12 MR. EBERSOLE: Wasn't it Harold Denton who was
(
13 enthusiastic about the ISAP?
14 MR. SIESS: Everybody has been. ,
I didn't know 15 anybody that didn't like ISAP.
16 MR. EBERSOLE: I like ISAP. I think it's the only 17 way to go, until we can get standard plants.
18 MR. SIESS: Oh, yes, sure.
19 MR. SIESS: I mean, rubber stamp, daisy. Cookie 20 cutter.
21 MR. SIESS: No. They all have different 22 utilities, i
23 CHAIRMAN KERR: Are you fellows telling Dave Ward j l
() 24 what the secretary is supposed to do. l 25 MR. SIESS: I thought I did.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coveraye 800-336 4 646 I
31630.0 cox 106 1 MR. WARD: I think Chet did. Is there any --
2 okay. I think I get what Chet said about my second point.
3 What about the first point? I am going to draft the letter.
4 Give me some input if you want to.
5 MR. EBERSOLE: Let's go with ISAP.
6 MR. REED: I am glad the word standardization was 7 mentioned. Why did we get into ISAP anyway? Well, we get 8 into it because at time zero, when the Model T forward was 9 built, designs are not mature, you come along, immaturities 10 of designs, come along the trail, backfits, incidents, all 11 these kinds of things. Eventually, ISAP should go away, if 12 we ever got --
G
/ 13 MR. SIESS: I don't think so.
14 MR. REED: You mean we never will reach maturity.
15 MR. SIESS: I think different utilities will have 16 different plant betterment programs they want to factor into 17 their maintenance system.
18 MR. SIESS: If the Commission thinks utilities 19 will standardize operations at all their plants, you are 20 dreaming.
21 MR. EBERSOLE: Pan Am --
22 MR. SIESS: Delta might be different tuan Pan Am.
23 CHAIRMAN KERR: Just a minimum standard, which is
() 24 what the standard plan is.
25 MR. EBERSOLE: Minimum?
ace FEDERA1. REPORTERS, INC.
202-347-3700 Nationwide Coverage 8m3364M6
31630.0 cox 107 1 . CHAIRMAN KERR: -
'O 2 MR. WARD: Gentlemen, it's your meeting, Bill. '. I 3 am not going to take responsibilities for this. .
4 MR. SIESS: State of the plants have nothing to do 5 with this.
6 MR. REED: I would like to complete my statement.
7 My concern is, if you continue on trail of ISAP and i
8 organization support, and I get the feeling that Northeast j 9 Utilities has a very large, I will call them desk engineering 10 group, that you begin to decouple, lots of desk engineers 11 getting more remote from operation, shuffling these figures, 12 grinding out these cranks, coming out with a figure that might be quite unrealistic and decoupled from the real scene, 13 14 valve or pack. I think there's been a tendency on the part 15 of Northeast Utilities to allow this kind of organization to 16 grow, grow and grow. I suspect you will find they will have 17 a very large organization in nuclear activities. I am not so 18 sure if the other companies might want to, like the Swiss 19 only had 200 people involved in their nuclear plants and they 20 run one of the best plants in the world.
21 MR. LEWIS: How do you know that?
22 MR. REED: Asselstine said that from his visit. I 23 wonder if you are building empires and people from scheme O 24 efter scheme.
25 MR. SIESS: You would rather continue as we do ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage MKk336-6646
31630.0 cox 108 1 now?
O 2 MR. REED: Staff ought to close the circuit on 3 things that are' deficient. If you think ISAP closes the 4 circuit, we can shed some of the other investigations, teams, 5 cats, dogs, whatever, fine.
6 MR. SIESS: I don't know what yon are talking 7 about.
8 MR. REED: I don't like to continue to build.
9 MR. SIESS: This deals with how the utility meets 10 the hundreds of requirements that have been tossed at it.
11 They are out there. Whether the plants, ISAP. I don't care 12 whether it's Northeast Utilities or Podunk Utilities, they t
13 have two plants that are in ISAP and two that aren't. And 14 the two plants that are not, they have gone the old way, they 15 have done them.
16 MR. REED: You are saying we will not eliminate 17 the old way if we go to ISAP.
18 MR. EBERSOLE: Glen, there was an implication what 19 you said that Northeast Utilities doesn't have the shift 20 engineers, shift supervisors, coupled into this effort.
21 CHAIRMAN KERR: Gentlemen, I want to point out 22 that we begin our meeting after lunch at a quarter of 2:00.
23 We can continue to discuss these interesting topics as long 24 as you like, but they are going to cut into the lunch hour.
(])
25 MR. WARD: Well, that ends the subcommittee ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0 cox 109 1 report.
O 2 MR. MARK: I would like to ask Glen whether he 3 would rather hear from Northeast running the plant or 4 Davis-Besse.
6 MR. SIESS: Neither one has anything to do with.
7 ISAP.
8 CHAIRMAN KERR: Our lunch break is beginning.
9 (Whereupon, at 12:45 p.m., the meeting was 10 adjourned, to reconvene at 1:45 p.m. this same day.)
11 12 13 14 15 16 17 18 19 20 1
21 l
'I 22 23 O 24 25 ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(Kb3364446
{
31630.0 cox 110 1 AFTERNOON SESSION (1:45 p.m.)
CHAIRMAN KERR: Mr. McPherson is here. He will 2
3 introduce the next topic and speaker.
4 DR. MC PHERSON: Thank you, Mr. Kerr. My name is 5 Don.McPherson from the Department of Energy. I am with the 6 Light Water Reactor Safety and Technology Office. Within 7 that office, we have a series of severe accident technology 8 programs of which TMI-2 is one of them.
9 We have, as our objective, the transport of the 10 core and the disposal of the core, and furthermore within the 11 accident evaluation program, we are responsible for 12 understanding what happened during the accident, applying O 13 that understanding to the resolution of severe accident, j 14 source term issues; and in transferring the information we 15 learned from this program to a technical field and to the 16 public. As a result of this objective, I am especially 17 pleased to be here to provide the results of our accident 18 evaluation program to you, the ACRS, and to make it a matter 19 of public record.
20 We have two representatives of our contractor, 21 EG&G, here with us today. I wish to introduce these two 22 people. Phil Grant is the manager of the entire TMI-2 23 program for DOE, and he will be discussing the status of the
() 24 defueling and the means that are being used for that work.
25 Jim Broughton is the manager of the accident evaluation ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-3364M6
31630.0 cox 111 7, 1 program, that is, one portion of the overall program. He
(.J I 2 will be telling you of the status of the core, the 3 information that is being learned from the. samples taken from 4 the core, and the scenario which we now believe the accident 5 went through.
6 I will introduce to you Phil Grant.
7 MR. GRANT: Thank you, Don. As Don indicated, I 8 am planning to focus on the defueling activities in the first 9 part of the presentation. In order to do this, I have 10 proposed to produce a snapshot of what the end state 11 condition of the core is and where we are at in the defueling 12 process. This snapshot is based on all the data we have 13 today, both visual examinations, both probing, sample 14 acquisition, and also core bore sampling and probing. This 15 best describes what we believe today was the end state of the 16 reactor after the accident.
17 The first region, I think both of you are 18 familiar, as I briefed you about a year ago concernir.y those 19 conditions, the green area here is what we know of as the 20 " void area." This originally represented about 316 cubic 21 feet of area. It was, of course, created when the oxidized 22 coolant was joined by the other coolant. This core data was 23 measured with topography data and we measured the terrain
() 24 from that data. The blue terrain represented the loose 25 debris. This represented 35 tons of material. Again, this ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 80433M646
~
r ,
31630.0 cox 112 j l
1 was quantified not only as part of the fueling, but also on l O 2 part of the mechanical probing of depths of the various )
l 3 locations in the core. The red outlined area represents what 4 we previously called the " molten mass" material. The 5 sequence of how these layers were formed will be discussed by j 6 Jim when we get into the accident scenario and the sequence 7 events. I will not refer to that now. Anyway, this previous j 8 molten mass of material had a crust on it of material and had 9 a ceramic center core. This represented approximately 28 to l 10 30 tons of material. The yellow area shown here, the 11 standing stub assemblies below that molten mass, originally 12 the core had 177 stub assemblies.
13 Of course, a lot of that was destroyed when the 14 oxidized zirc was collapsed as some dissolved part of the 15 molten mass. That region of standing rods represent about 60 16 tons of the core. Then we have the lower head region, 17 including the lower port assembly and the fuel that accessed 18 through that lower head. We are estimating that is in the 19 order of 20 to 25 tons. Then some of the newer data, I will 20 show actual photographs, slides of some damage done to the 21 core, but we have done some probing in the core former 22 areas. We are estimating as much as five to seven tons of 23 fuel debris in that location.
MR. MARK: Please, you have mentioned all these
(]) 24 25 numbers in terms of tons. What is the total number of tons ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 800-33MM6
l 31630.0 cox 113 -
1 of the maAerial that is involved here?
2 MR. GRANT: Total tonnage is about 293,000 pounds 3 or slight,1y under 150 tons of material.
4 MR. MARK: When you say 15 tons, you are talking 5 about 10 percent.
6 CHAiRMIN KERR: Actual fuel is what, 120, metric.
7 MR. GRANT: Yes. I wanted to paint you a graphic 8 -- the only other thing I wanted t; mention to you, based on 9 some'end core probing through the instrument lines, we have 10 concluded, again, this is very preliminary, we have about 8 11 inches of metallic nonseismometer material below the metallic 12 material. Jim will go through that --
0 13 CHAIRMAN KERR: Go that again. You have 8 14 inches --
15 MR. GRANT: Metallic material that is nonsource 16 material.
17 MR. REED: On the wall, this is over 4 inches?
18 MR. GRANT: 5 inches.
19 MR. SCHWINK: Control rods is a fair part of 20 that?
21 MR. GRANT: You are getting ahead. Let me paint l
22 the picture and we will tell you how we got there.
23 MR. SCHWINK: Before you leave this, when that
() 24 melting occurred, is there any agreement on where the steam 25 bubble or the extent of the steam bubble in that picture you ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0 CoX 114 1 have there?
O 2 MR. GRANT: You are referring to the original 3 bubble after the accident, when we had the collapse and how i l
i' 4 that related relative to the core?
5 MR. SCHWINK: I am referring to the steam bubble 6 that you think was present when that fuel melted and came 7 running down.
8 MR. GRANT: Jim, are we going to get into any of 9 that?
10 MR. BROUGHTON: I will cover that.
11 MR. SCHWINK: All right.
12 MR. GRANT: Let me show you where we are at with O 13 regard to defueling in the various regions, what is lef t and 14 some of the complexity in the defueling of those regions.
15 Region 1 shown here is the area that we completed, that 16 defueling has been completed to date. It represents 17 approximately 114,000 pounds of material out of a total of 18 about 293,000 based upon our current best estimates of total 19 mass of material. This region 1 defueling included the loose 20 debris area that I referred to earlier, and also about 95 21 percent of the molten mass of material that had resolidified 22 as part of the accident sequence.
23 Region 2 here is what we referred to as a midcore
() 24 region, a couple of things still remaining there. I alluded 25 to earlier, we used the core bore operation to kind of Swiss ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 8(Xb336W46
31630.0 cox 115 1 cheese this hardened resolidified area. The accessing O 2 diameter in the core is about eight feet. There were some i 3 outer regions that couldn't be Swiss cheesed or rubblized by 4 the core bore. That's what we refer to as the doughnut, j 5 outer ring. There are also some rock-like materials, some of 6 which weigh over 1000 pounds in the core. I will show you 7 what those look like, and other loose material in this region 8 2.
9 Region 3, which is the stub assembly, or what we 10 call the lower core region of the original core. As I 11 mentioned, there are 177 stub assemblies. To date, GPU'has 12 defueled 39 of those stub assemblies. While I have got it, O 13 why don't I show you what that means.
14 Here is a top view, looking down on the core, of 15 the 177 assemblies. The ones shown in blue are the stub 16 assemblies that have been physically removed, loaded into 17 fuel cans and removed from the core proper area. The i 18 doughnut that I alluded to earlier in the core bore, major 4
19 part of it, seems to be up in the northern quadrant and what 20 we know as the southwest quadrant.
21 The off-green colors here are actually the core 22 locations that we took strata samples that Jim will get into, 23 and the significance of those analyses. The other region
() 24 here shown was a 19 by 19 array, where we Swiss cheesed all 25 the way down into the -- about 12 inches above the lower grid
/\CE-FEDERAL REPORTERS, INC. I 202-347-3700 Nationwide Cmerage 800-336-6646
31630.0 l cox 116 l 1 support plate, all the way down to the lower spacer grid to
\
2 demonstrate that the core bore could be used as a way of 3 rubblizing fuel intact assemblies as well.
4 MR. MARK: Please, how does one assess the 5 temperature distribution.across that diagram you have got.
6 Is it high in the middle, high in the red spots, blue spots 7 or what?
8 MR. GRANT: You can't assess it from this. This l
9 is physical location, i
10 MR. MARK: I know you can't. I am asking you to !
11 tell me how it goes.
12 MR. GRANT: We will get into that in the accident O 13 investigation, what we have hypothesized as a sequence of the 14 various events of the core.
15 MR. MARK: Very good.
16 MR. EBERSOLE: If you could review for me the 17 original basic logic of why you are doing all this stuff. Is 18 it to study the degradation of the fuel under these 19 conditions? Is there any faint hope that you will run the 20 plant again? Why didn't you store it in place?
21 CRAIRMAN KERR: Mr. Ebersole, Mr. McPherson did 22 explain that early on.
23 MR. EBERSOLE: I missed that.
() 24 CHAIRMAN KERR: He told us why DOE was doing it.
25 MR. EBERSOLE: It must have been a short Acn FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 8 6 336 # 46
i 31630.0 l cox 117 i p- 1 statement. What was it?
%.J l 2 MR. WARD: It was earlier. !
i 3 MR. SIESS: We don't have time for a long one. l 1
4 DR. MC PHERSON: It is our objective, of DOE, to 5 understand what happened in the accident.
I 6 MR. EBERSOLE: That's enough.
7 DR. MC PHERSON: To apply that information to the 8 nuclear industry.
9 MR. EBERSOLE: There's no hope you can regenerate 10 this one?
11 DR. MC PHERSON: It is not DOE's hope.
12 MR. GRANT: Let me go back briefly to some of the O 13 other regions that we need to focus on in defueling, those 14 being regions 4, 5 and 6. Technically, they probably present 15 the most complex part of defueling, the remainder of the TMI 16 core. The one, I will start on region 6 and work from top to 17 bottom, is outside the core former areas. I have some 18 photographs of damage done to the core formers. But based on 19 current probing down through the flow holes, we concluded 20 there could be seven tons down there. So defueling will 21 require removing the baffle plates and dislodge that 22 material, and whether that is interfused with other 23 structural steel, including the barrel, is unknown.
() 24 A new problem will be going through the lower 25 course, 13 plates, largest of about 13 1/2 inches in ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0' cox 118 1 diameter, weight about 20 tons. Access in that flow
- O 2 distributor and forging plate, on the order of squares of 3.5 3 by'3.5. inches. -So gaining access to the lower head and' 4 defueling some of the regions that we visually have seen fuel 5 debris -- and I will show you on some of those slides --
6 will be a very complex operation.
7 Then once you access through, of course, removing
- 8. the remaining amount of material. There are some approaches 9 that GPU is taking, including accessing the lower head 10 through the annulus and using vacuuming systems,'and what we '
11 call a " passive vacuuming system." Because of some of the 12 data we have here, we are not sure we are going to find much O 13 of that material being vacuumed. There's a large unknown 14 there. -Jim will get into some of the data as to what we 15 found in the lower head.
16 MR. REED: I guess you are saying you can't pull j 17 the barrel.
18 MR. GRANT: Right. The plan is not to pull the 19 barrel, the plan is not to remove physically -- the plan was j 20 not to remove what we call the " core assembly attachment."
21 There is a plan that calls for pulling out the core assembly 22 in pieces, but we view it not as a viable way of defueling 23 the system.
() 24 MR. EBERSOLE: Is it welded in place?
25 MR. GRANT: Part of it is probably welded because ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
31630.0 cox 119 1 of the accident and the molten flow of material. The way tON./ :i 2 it's assembled, it's assembled in a reverse order. You can't i 3 take it apart just like you install it.
4 Anyway, we are looking at different ways of 5 accessing, ranging from very simplistic vacuuming systems to 6- cutting and enlarging hole access holes using the core bore 1
equipment I referred to earlier or an underwater cutter and 1 7
8 actually cutting out pieces of the core. Once you get down 9 there, you have 52 end -~9 guide tubes and end core 10 instrument strings which basically present a forest'of trees 11 down there, and the tubes are 2-1/2 inches in diameter. They 12 also preclude a strict access to removing a lot of the fuel O 13 in the lower head.
14 The point I wanted to make about regions 4, S, 6, 15 they offer a very complex, technically challenging obstacle ,
16 to defuel that.
17 That's kind of a snapshot of where we are at in 18 defueling and some of the things coming down the road. The 19 last Vugraph I am going to show you here, before I show you 20 some slides, is kind of the progress rate to date. What we I
21 plotted here is the weight of the core as a functional 22 calendar of time, and the upper curve represents the pounds 23 of core debris loaded over time, and as of actually the 7th, O 4 we had 113,000 pounds. We have also overplotted the shipping 25 rate. I am not going to present too much on our shipping Ace FEDERAL ' REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
31630.0 cox 120 1 program today, but basically we have two NRC license casks, 2 type-B casks, designed for shipping fuel debris. We 3 basically can ship seven canisters, fuel canisters, per 4 cask. Typically a fuel canister holds between 1100 and 1500 5 pounds. Some of the more recent ones are lighter loaded 6 because of packing efficiency. So we have shipped about 7 88,000 pounds of core debris or roughly 78 percent of the 8 fuel. We have made this in 12 shipments from the island in 9 double cask shipments.
10 MR. MARK: You mentioned two casks. Is that 11 because of a limitation on number of casks or restriction on 12 NRC licenses?
13 MR. GRANT: No restriction on NRC licenses. They 14 issue one license for the cask. It's up to the manufacturer 15 to build casks according to that certification for it.
16 There's no restriction on that. The reason two casks were 17 chosen initially gets back to looking at the maximum estimate 18 of cool core defueling progress, and we also looked at truck 19 cask shipments, rail cask shipments and ended up with rail 20 cask. We sized it accordingly based on the amount of fuel 21 that we anticipated on defueling. Also it was sized on what 22 type of fuel canister would be replaced into the reactor 23 vessel with existing fueling design and efficiently optimize
() 24 fueling operations. There's two casks today.
25 It brings up another point. GPU is in the process l
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3364M6
31630.0 cox 121 n 1 of leasing a third cask. They are doing this with the b 2 anticipation that the stub assembly removals that I referred 3 to, and some of the other defueling operations, the 4 performance will improve. Therefore, they have concluded 5 that a third cask will better optimize the rate of removal of 6 core material from the TMI site.
7 The only other point I wanted to make here, you 8 see some flat areas. I think most of you are aware there are 9 biological growth problems, loss of turbidity. Those 10 impacted on the fueling progress. We also, when we were 11 Swiss cheesing the core and doing a core strata sampling 12 periods, there were some flat periods. There's also been k 13 some technical problems in removing the material that is 14 created as well. It's a combination of problems as well as 15 technical difficulties.
16 Other bottom line here, based on the progress rate 17 today, it's obvious defueling activities will go well out 18 into 1988.
19 MR. MARK: Couple of questions. You mentioned a 20 biological effects which interfered with things. What is the 21 general radiation level in which those biological things 22 prospered, 10 to the 6th rads, 10 to the 9th or what?
23 MR. GRANT: Depending on proximity as to whether l
O 4 1t ee ree1 aedrie or structere1 etee1- weao heve 111e= or 25 dose rate within the various core elevations. We took a look ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 804336-6M6
I 31630.0 cox 122 1 at that --
^ %')
2 MR. MARK: Were those things you mentioned rates 3 or total exposures. f 4 MR. GRANT: I am sorry, total rates, hour for 5 hour5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br />, 10 to the 4th, 10 to the 5th.
l 6 The point I guess you are getting at is biogrowth ;
7 and the impacts, The reason we believe that these things 8 survived is that the ecology and the evolution of them is very short-lived, therefore, rad effects, and also the fact !
9 10 that you have absorption of dose with the water, but, yes, we 11 have actually identified bacteria, algae and some euglena 12 that were within the coolant systems.
(}-
\ 13 MR. MARK: And in radiation fields of the sort you 14 mentioned?
15 MR. GRANT: Yes.
16 MR. MARK: That was one question. Another was, 17 will it come up in later discussion whether the distribution 18 of stuff in this only example we have of a meltdown, is i
19 likely or thought to be likely to be similar to the accident 20 progression in any light water reactor, that is purple stuff, 21 zone 5, and your slide before last, will that also happen in 22 a PWR when it is melting down and contributing to direct )
i 23 containment heating and the imaginary scenarios of that
() 24 sort. The third question is an estimate of the amount of 25 hydrogen actually formed. i ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 8043364646
31630.0 cox 123 1 MR. GRANT: We will be getting into some of the 2 generic applications. Jim will be touching on it. This is, 3 by the way, a PWR. B&W 177 reactor, TMI-2. 4 i
4 MR. MARK: You say it's a pressurized? l l
5 MR. GRANT: Pressurized water reactor. 1 I
6 MR. MARK: That crud in the bottom, which has 7 nothing to do with active fuel, is --
i 8 MR. GRANT: We speculate it's control rod 9 material. Jim will get into that sequence, where it has
., 10 generic applications, both in source terms and other areas, 11 we will touch on briefly.
12 MR. MARK: That is going to answer my questions, 13 thank you.
14 MR. GRANT: Now I will show you some of the 15 equipment we are using.
16 MR. MOELLER: Excuse me. You said we, you are 17 with DOE. The plant staff, I presume.
18 MR. GRANT: I should have explained that. We are 19 integrated with the GPU staff. We are involved in their 20 defueling program. I have what I call a " technical 21 integration office" that is set up right at TMI. They are in 22 there administering, controlling the funding, the DOE 23 contributions and insuring that we are linking progress with l
O 24 defueling, with funding administrations. We are also 1 25 involved -- we are accepting these canisters, so I have a ACE. FEDERAL REPORTERS, INC.
202-N7-3700 Nationwide Coverage 800-3364M6
1 31630.0 cox 124 l
1 staff there that is doing quality assurance, receipt i O 2 inspections on the cans. We are also accepting for shipment 3 the materials on-site. DOE is the shipper of record. We 4 represent DOE as well. We are integrated there as well as 5 back at Idaho, 6 The next sequence of slides, I am going to walk 7 you through some of the defueling operations and the equipment that is being utilized. This is a top shot looking down at what we know as the defueling plant form. There are l
- J ' two jib cranes, one-ton jib cranes. This is the open slot, Il 11 where the operators access the --
12 MR. SCHWINK: How far are they above the core O' 13 midplane at that point?
14 MR. GRANT: Their position relative to the loose I 15 debris was about 24 to 26 feet. For accessing the lower head 16 they will be over 40 feet above the actual defueling or fuel 17 debris area. This platform does rotate. Depending on the 18 quadrant of the core, they are in the process of defueling, 19 they operate accordingly. There is a hydraulic system that 20 doesn't show here.
21 MR. MARK: How far are they above water 22 elevation?
1 23 MR. GRANT: Water elevation relative to this is j
l
() 24 about three feet below the platform. Here is another angle l 25 looking in, they have TV cameras into the coolant. They have ACE FEDERAL REPORTERS, INC.
i 202-347-3700 Nationwide Coserage 800-336-(M6
31630.0 ;
cox 125 1 access of observing that as they are doing the operations.
2 They also use extensive hydraulically operated tooling. I 3 will get into showing some of that, too, so the whole 4 hydraulic power station is also built and set in the 5 background over here.
6 One cf the major tools that originally was an R&D 7 tool was what we call a " core bore" system. This was 8 basically a drilling system that was using conventional 9 drilling technology for underwater offshore drilling, and it 10 was applied in such a way to try to retrieve strata samples 11 from the core, including that previously molten mass of 12 material and also accessing down through the stub assembly O
\/ 13 areas down to the lower head. Jim will get into some of the 14 data on that.
15 The unit is also now being considered, as I 16 mentioned, for enlarging holes, up to 6.5 inch diameter holes 17 in that lower core assembly, going through what we might call 18 an " air lift" system and try to vacuum the material out of ,
19 that. There is also consideration for using the core bore 20 for actually drilling out the corners of the core former i
21 walls such as we can use it as a way of dismantling that core !
22 former plate. As I mentioned earlier, the original diameter 23 access of a core bore was roughly 4-foot radius, 8-foot
() 24 diameter, subsequently been modified to allow accessing of i
25 all regions of the core including to the core former wall !
ACE FEDERAL REPORTERS, INC.
202-347-37(X) Nationwide Coverare 8(Kh3364M6 I
1 31630.0 i cox 126 1 areas.
O 2 This also shows the existing platform where you 3 saw the operators. This is what we called the "defueling 4 carousel." It holds five fuel cans and basically rotates and 5 allows them to bring a fuel can into a load location where 6 they can put fuel debris. You also have three elevations in i
7 which this carousel can position the canisters such that you 8 are lowering the canisters down in the areas you defuel.
9 Next one, please. Here is an actual photograph of 10 the core bore equipment installed, in the operating, this is 11 the main term motor, down into the core region. This is one 12 of the Chris drills. This was used for actually retrieving 13 strata samples from the core. Rouj5ly 3.5 inches in 14 diameter. It's what we call a " ceramic cutting tungsten 15 carbide bit." Hollow shaft fo,r Swiss cheesing the core, we 16 used a solid face bit to actually rubblize that material.
17 I ought to also mention this equipment was 18 developed out at EG&G, fully tested, procedures developed. !
19 The TMI-GPU crew went out and trained on it before it was 20 shipped to the TMI plant.
21 I am going to show you about four or five tools 22 that are currently being used with regard to harvesting 23 fuel. Can you raise that by hand a little. This is what we
() 24 call our " cutting tool." It's approximately 4-foot in 25 length, basically a nipper. We use this to cut through some ACE FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Cmerage 80(L 336-6646
31630.0
. cox 127 1 of the fuel pins, stub assembly areas, to harvest some of O 2 that material. This is actuated by a hydraulic system. We 3 believe this hydraulic system, because of leakages in some of 4 the other connecters, was one of the ultimate sources for the 5 biological growth in the reactor coolant system.
6 MR. MICHELSON: How big is it?
7 MR. GRANT: 4-foot.
8 MR. MICHELSON: No, diameter.
9 MR. GRANT: This is roughly 2-foot.
10 MR. MARK: One substitute square is behind it.
11 MR. GRANT: Yes. Very massive piece of-12 equipment. Next one, please.
O 13 This is what we call the " clam digger," again, a 14 hydraulically operated. This is a lower part of the clam 15 mouth. There's an upper part up here. It's also 16 approximately 4-foot. This tool was used as defueling part 17 of the loose rubble debris. This is what we call the " spade 18 bucket." It has a hard faced -- you can see the teeth here.
19 There was a compacted layer of coal after it was 20 rubblized. This was one of the major units for removing it.
21 It's basically dropped in the bucket, it collapses, you bring 22 it over and load it into the fuel can.
23 MR. SCHWINK: Width of that is a foot or what? I
() 24 MR. GRANT: Almost 2 feet, 18 inches. We have a 25 small version of this and a large one. This is the large ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide CoseraFe 8(Kk336 6646
31630.0 cox 128 1 one. This is a -- what we call a three-point gripper, again, O 2 hydraulically operated tools. Most of these tools you are 3 seeing now were developed by Westinghouse to support the 4 defueling operations.
5 Basically, this tool is used to grip individual 6 rods or other loose debris that you might want to access.
7 This is what we call the " cam lock fuel assembly 8 unit," and basically it's hard to show this, because we are 9 off the screen. But there's a square 8.5 by 8.5 opening here 10 where you lower this down over a partially intact stub 11 assembly that has the fuel stub pins sticking up. Basically 12 those stubs come through, hydraulically driven. This is a O 13 wedge that compresses against it. It grips it. You 14 physically pull this up with the assembly. It was used on 15 some of the peripheral assemblies that were almost totally 16 intact with some undercutting, but we are not using it for 17 the existing stub assemblies, because, in those, we have to 18 access and actually grab the lower end fittings to retrieve 19 and remove those.
20 These are photographs in the core.
21 The next few slides are slides of actually looking 22 down into the core area. This is after the Swiss cheesing of 23 the core, and there's an area that is hard to see here. But
() 24 basically it goes down to partial stub assembly sticking up.
25 Out in the outer section is part of what we call the ACE FEDERAL REPORTERS, INC.
, 202 347-3700 Nationwide Coterage 800-336 # 46
31630.0 cox 129 1 " doughnut." There are some loose pins here that fell out.
O 2 MR. MOELLER: Those long rods are fuel rods?
3 MR. GRANT: Yes, roughly 3/8 in diameter, fuel 4 pins.
5 This is a similar shot looking more outboard, and 6 these are actually fuel pins extending up.
7 I have a two minute video of what we are looking )
8 at. Water clarity was good. I shouldn't say more about 9 these slides, l
10 MR. EBERSOLE: Did the RCCs preferentially i 1
l 11 disappear, the cluster control rods? Did they precede the 12 fuel on down? l 13 MR. GRANT: We will get into that. But the 14 sequence of events looking at the burnable poison rods and 15 control rod material, Jim will get into that in the accident 16 sequence, based on which collapsed first and the flow of l 17 molten material.
1 18 This is, again, a terrible photograph. The core 19 former walls, which form the original core area, are 20 basically plates, 3/4-inch stainless plate and they form 21 corners in some locations where the actual assemblies fit 22 down. A point of interest on this one was we found a melt 23 through on the core former wall, this core former wall would Q 24 run down, of course the plate would extend down to the lower 25 grid support plate. In this case you can see kind of a clean ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 804 334 6646
31630.0 cox 130 1 cut. The photograph doesn't show it that well, but there was O 2 actual fuel debris and pins sticking through this area. This 3 was our first indication of potential major degradation of 4 core former wall and a potential pathway for material to flow 5 down into the outside of the core former walls.
6 MR. MICHELSON: Core formor wall, how does that 7 relate to what we call the shroud?
8 MR. GRANT: Same.
9 MR. MICHELSON: Same thing?
10 MR. GRANT: Yes.
11 MR. MICHELSON: Thank you.
12 MR. GRANT: Here is a shot of a lower end fitting 13 -- the fuel assembly would sit in this position. The fuel 14 pins would be down into what we call the A crate, lower end 15 fitting session. This was one of the assemblies that was I l
16 physically pulled up out over the core and laid over on the
~
17 bed for eventual cutting and loading into fuel cans. You can 10 see in some instances the physical removal, when they try to 19 grapple these from the top, they actually pull fuel pins out l
20 of it and obviously made a mess which requires extensive 21 cleanup. But this is the lower portion of that.
22 I Here is looking down after a stub assembly had 23 been removed, looking down on the pads, the assembly
() 24 saddles. This is part of what we call the lower grid support 25 plate and the pads. In the center of this region, which is ACE. FEDERAL REPORTERS, INC.
} 202-347-3700 Nationwide Cmcrape 8( 4 336 6646 )
l
31630.0 i cox 131 1 hard to sec, this was one of the instrument fuel assemblies, i O 2 This kind of coaxial cable is what we call part of the 3 instrument end core. I will show you some shots of the end 4 cores from the lower head and the guide tubes that kind of 5 direct those end core lines.
6 This is a shot taken down in the lower head, 7 looking back up through that elliptical flow distributor that 8 sits on the end. That flow distributor is approximately two 9 inches in thickness and the actual hole was about six inches 10 in diameter. This is a nodule of the molten material that 11 was oozing or flowing through that opening in the flow 12 distributor. We haven't been able to sample any of this, but 13 we have sampled the material, some loose material down in the i
14 lower head. You will hear from Jim that we identified most 15 of this as ceramic material, rather than a metallic ceramic.
16 This is another shot that was taken earlier down 17 in the lower head inspection. The reason I wanted to show it 18 to you was it's obviously proof positive that there was some f 19 molten flow of material down in the lower head region, versus 20 there were some earlier thought that a lot of this material 21 as it flowed down below the water level, as it quenched, it i 22 broke off and went to the lower head.
23 So the indication of some molten mass of flow 24 material down there.
({}
25 This is a shot looking at one of the end core f l
ACE FEDERAL REPORTERS, INC. j 202-347-3700 Nationwide Coverage 804336MM
I' 31630.0 cox 132
,. 1 guide tubes. To give you a perspective, that is down. This 2 is what we call the guide tube, roughly 4.5 inches in
)
) 3 diameter. Inside it we have what we call the instrument end 4 core nozzle. This nozzle penetrates through the reactor 5 vessel walls, in the lower head. This was a -- one of the 6 shots looking down at that unit and looking at some of the 7 fragmented material that was laying around. No damage to 8 this one. The reason I am leading you into this, we will 9 show one that we found that was damaged.
10 Next one, please. Here is another shot of the end 1 11 core nozzle. The distance between the lower head and the end 12 core guide tube is about eight inches. Here is one of the 13 nozzle penetrations through the head, basically intact and no
't 14 apparent physical damage to it.
15 Here is one --
16 MR. SCHWINK: Instrument tube nozzle?
17 MR. GRANT: Yes. Here is one where we have 18 identified damage, and to kind of lay this out, this area 19 here shown in white was the original part of the guide tube.
20 Again, they are 4.5 inches in diameter. The upper section of 1
21 the guiae tube is over an inch and a half in thickness, and i 22 it tapers down and the thickness at the lower end of this 23 guide tube is actually about 2/10 of an inch. Here in the
() 24 center is the end core nozzle. As you can see, basically the 25 nozzle has been removed, and there's some nodules of melted ACE. FEDERAL REPORTERS. INC.
202-347-37(K) Nationwide Coverage Ek3364M6
31630.0 CoX 133 1 material on top of it. These are one of the samples we are 2 anticipating as we mine down into the core that will preserve 3 that and do the necessary chemical structural analysis on the 4 materials.
5 That's basically all the photos I was planning to 6 show you relative to defueling. I have a transparency here 7 showing what one of the end core nozzles looks like.
8 This shows basically the nozzle, the instrument 9 string within the nozzle, the guide support tube that I was 10 telling you about. This is the bottom of the distributor and 11 of coursa this is the reactor vessel, lower head.
12 MR. SCHWINK: You made a point that this was not O 13 attacked by molten material, if I understood you correctly.
14 But I had thought that there were some others there they 15 thought they had found, or you had found molten material 16 inside of one; is that incorrect?
17 MR. GRANT: I don't think I said this wasn't j 18 attacked, we found some that were. There were about four end 19 cores that had been damaged. This one, which was guide tube 20 45, it's the R-7 location, was the worst of the four that we 21 found thus far. We haven't inspected all of them, because we 22 basically have been accessing these coming down the outer 23 annulus. We could only accees those peripheral end cores.
() 24 So, yes, this one is damaged by, obviously, some heat source 25 and the molten materials. Here is another graphic of what ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-336-6646
31630.0 l cox 134 1 you saw in the photograph, and the areas that we have seen, a O 2 melt through of the guide tube.
3 That's' kind of all I had planned from a defueling 4 report on the reactor vessel, what we have identified thus 5 far. Jim ~will walk you'through our accident evaluation.
'6 program, kind of what the objectives are, where we.are at in 7 accomplishing those objectives. I will end with the accident ~
8 scenario and some of the significant.results to date.
9 101. BROUGHTON: While I am getting my slides out, 10 Dr. Kerr,'do you want me to try to end at 3:15, which would 11 shorten my talk considerably, or do you want me to --
12 CRAIRMAN KERR: Research team --
't ,
13 MR. BROUGHTON: I will shoot for 3:15.
- 14. Dr. McPherson explained to you the relationship 15 that the TMI accident evaluation program is sponsored by 16 DOE. The primary goal of the program is to provide evidence 17 for more realistic source terms. In this afternoon's talk, I 18 will very briefly touch on --
19 CHAIRMAN KERR: Excuse me, is a microphone i
20 available?
21 MR. BROUGHTON: I will speak up. Thank you.
22 I will briefly touch on defueling and examination 23 results which are currently in progress. These results will
() 24 be more thoroughly documented in reports to be published 25 later this year and next. I will then get into the accident Ace FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
31630.0 cox 135 1 scenario itself, basically a qualitative description of the 2 scenario as we envisioned that it occurred, and then 3 summarize for you the impact of the program as we see it on 4 some of the more pertinent technical issues that the NRC is 5 facing.
6 MR. EBERSOLE: Are you going to identify the 7 things that would stop the accident before it got so bad?
8 MR. BROUGHTON: No, sir, but the accident was 9 terminated by the addition of water in the core.
10 MR. EBERSOLE: I meant long before that. 1 11 CHAIRMAN KERR: Jesse, why don't we let him do his 12 presentation. He said he had a lot to cover.
C'- 13 MR. BROUGHTON: This is an outline of this 14 afternoon's presentation. I will briefly again put up the i
15 objectives. I will go over this very rapidly, the 16 information that is required from the TMI-2 research to meet 17 our objectives. There is a series of photographs in your 18 handout that summarizes some of the examination results. !
i 19 I am going to skip over this in the interest of 20 time, and go directly basically into the accident scenario 21 itself. I believe that based on some of the questions I have 22 heard this afternoon, I think the committee is basically 23 interested in that, and the potential impact from this
() 24 research on some of the more pertinent technical issues.
25 I will conclude this afternoon's presentation with ACE FEDERAL REPORTERS, INC, 202 347-3700 Nationwide Coverage 800-336 4 36
31630.0 cox 136 I
I a summary of the program status.
O. 2 As Dr. McPherson said, the primary objective of 3 the TMI program, accident evaluation program is to understand 4 what happened during the accident and then to apply that 5 understanding towards the resolution of some o2 the technical 6 issues, and finally then to transfer the results of the 7 program to the government and to the nuclear industry.
8 This slide and the next slide summarizes the basic 9 information required from -- for the TMI-2 research. It's 10 available to us from three primary sources. First is 11 defueling observations and then evaluation of on-line 12 instrumentation which was operational during the accident
'- Now, that instrumentation, of course, was not 13 itself.
14 designed to provide data under these severe conditions, but 15 in actual fact it did provide considerable information to 16 us. ;
17 Finally, this information from evaluation of 18 samples which have been acquired from the reactor itself 19 during the defueline process. Phil projected this slide. It 20 does show the condition of the core after the accident, at 21 least as we see it. We have taken samples from the debris 22 bed and from the lower plenum. Those samples are currently 23 being -- examination on those samples are currently in
() 24 progress.
25 Last summer we performed a drilling operation ACE FEDERAL REPoaTEns, INC.
.i 202-347-37(K) Nationwide Coserage 8(K)-33MM6
31630.0 cox 137.
I through this hard crust, that provided us samples of material 2 down to the distributor. In two occasions we had attempted 3 to take samples of debris on the lower head itself. But 4 because of the friability of this material, we were unable to 5 acquire a sample at that time.
6 This is a photograph of debris from itself. You 4
7 can see fuel pellets, cladding. We estimated that about 30 8 percent of this material was previously molten.
9 This is a photograph of one of the 11 particles we ,
10 have acquired from the lower plenum. We have sectioned and 11 examined these particles and find that they are basically 12 homogeneous in nature, primarily ceramic with some metallic, !
13 and peak temperature reached 30 to 300 Kelvin, depending on 14 the zirconium content and oxide material. There is extensive 15 porosity in the material, and the fission -- measurement of 16 fission products across these sections show that the fission 17 concentrations are basically uniform.
18 MR. MOELLER: You say ceramic material, you mean 19 its fuel?
20 MR. BROUGHTON: Basically UO2, ZrO2 compound.
21 MR. MOELLER: Basically resolidified?
22 MR. BROUGHTON: Yes. I have summarized here some 23 of the significant fission product resulting from the debris
() 24 itself within the core. We find there has been significant 25 retention of iodine and cesium in the high temperature core ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-33M646
31630.0 cox 138 1 debris. Significant cesium to about 20 percent, iodine about O 2 20 percent in the upper debris, about 2 to 5 percent in the j l
3 lower plenum debria. Also we find that there's been j 4 significant releasm of ruthenium from the fuel but retained l 5 in the metallic debris. ;
l I
6 As I said, we did a core boring operation. Last j 1
7 summer we acquired 10 core bores from the core, their 8 locations are shown here. There was a pattern to the badness )
l' 9 that one first sees on this slide. We have an east-west I
10 sampling line, a southwest to northeast sampling side and 11 then two north-south sampling sides. We concentrated on the 12 east side and attempted to acquire a sample directly through 13 the region where we believe the core material, molten 14 material, had relocated into the lower plenum. It appears 15 that region is out here in the southeast corner.
16 These are montage photographs of four of the core i 17 bores, G-8 and G-12 are near the core periphery. G-8 and K-9 18 are basically in the center of the core on the bottom of the 19 core itself, the rod stubs extend up to about two feet to i 20 four feet in length. We had anticipated acquiring solid i 21 columns or cores of material from the consolidated region.
22 But because of the fragile nature of that ceramic material, l
23 in the central region, it broke up in the core boring O 24 operation. We basically acquired chunks which were l
25 sandwiched between solid plugs from the bottom supporting ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 80(b33MM6
31630.'0 cox 139 ,
1 1 crust and from the top supporting crust.
O 2 The twisted nature of these fuel pins was caused 3 by the core boring operation itself. Basically these pins 4 'show very little, if any, damage at all as a result.
[
5 Accident. That is below the previously molten material.
6 The core bore examinations are currently.in 7 -progress. We are doing physical, metallographic and 8 radiochemical examinations. The metallograpnic examinations.
l 9 on the metallograph have been completed. We have samples at ,
1 10 the test area north. The results of these examinations'will 1
11 be written up and distributed to the NRC in early to mid j 12 1988.
13 Summarized here are the density measurements which 14 have been made on the rocks and the cores from the core bores-15 and also shown are some reference densities for zirconium, )
16 zirconium oxide and uranium oxide.
17 These are some closeup photographs of one of.the I
18 large plugs,.which was taken. This is the large plug at the j
)
19 bottom of the K-9 core bore. This is the supporting crust. J 20 This is a cross section, photograph of the cross ,
21 section that shows the fuel rod remnants surrounded by )
22 previously molten metallic structure. You can see that this j i I 23 previously molten material has flowed into the cracks. If I 1
0 24 you look closely, you can see that the zircoid cladding has 25 basically been dissolved away from these fuel pellets. In ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 8(0-3364M6 j
31630.0 cox 140 1 this region right here is the remnants of -- what sua believe O- 2 is the remnants of another or third fuel column here.
3 We sectioned that, or I have got four photographs, 4 photomicrographs in the handout from region 1, here, you can 5 see where the metallic material has flowed directly into the 6 -- one of the pellet cracks or interfaces between pellets.
7 We are focusing our examinations at this point in time across 8 these reaction' interfaces that we see.
9 This is the remnants of some cladding with the 10 fuel on the bottom of the photograph and the ceramic mixture 11 on the top. We are looking at these reaction interfaces to 12 learn more about the materials interactions that occurred 13 during the accident.
14 In region 3, which is out in the previously molten 15 metallic that surrounded the fuel columns, we found an 16 interaction zone from basically a metallic ceramic mixture 17 going to basically all metallic mixture. We are looking at 18 the composition and interactions that have occurred. We have 19 found that there is significant silver in this metallic 20 structure, as well as some iron and nickel. But at present 21 we have not found a significant amount of metallic 22 zircalloy. It appears that the zircalloy in this region has 23 been completely oxidized.
() 24 MR. SCHWINK: Region 3 is down at the bottom.
25 Region 3 is basically in the center of the photograph between ACE-FEDERAL REPORTERS, INC. ;
202-347-3700 Nationwide Coverage 8(n336-(M6 i
_ _ ____ ~ _ . _
i 31630.0 cox 141 1 the two fuel columns.
O (J 2 MR. SCHWINK: How many feet from the bottom head 3 is it?
4 MR. BROUGHTON: Regions 1, 2, 3 and 4 that I am 5 very rapidly walking through here are regions from that cross 6 section of the bottom supporting plug. That plug, we 7 believe, was taken at about two feet from the bottom of the 8 core.
9 MR. SCHWINK: Thank you.
10 MR. MARK: Are you in a position to estimate the 11 fraction of the total zirconium that got oxidized?
12 MR. BROUGHTON: I don't have that number on the i
V 13 top of my head, sir, but I think it was about 55 -- about 50 14 percent of the zirconium.
15 MR. SCHWINK: That's the kind of number that has 16 often been --
17 MR. BROUGHTON: It appears right now, with the 18 examination results that we have in hand, that most of the 19 zircalloy in the high temperatures became oxidized. But our i
20 examination of these structures are not complete so I j 21 couldn't say that all the zircalloy was oxidized.
1 22 MR. MARK: Your answer is perfect. This is also i 23 the kind of number that is assumed for the generation of O 4 avar ee#.
25 MR. BROUGHTON: That's basically consistent with ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 804336-M46
31630.0 cox 142 1 the total value of total hydrogen production, I believe.
7 s.
t)
That's about 400 to 450 kilograms, but I am not sure of the 2
3 number.
4 Region 4 is down at the bottom of the photograph 5 of the cross section, and it is in that region where we j
\
6 believe there is the remnants of a fuel column, and what I 7 have labeled here as a ceramic phase, we think, is the 8 remnants of a column of fuel surrounded by the metallic 9 phase, which is in the process of dissolving or liquefying 10 the UO2.
11 I wanted to show this slide and the next slide to 12 show, to illustrate the behavior of the products as we are O 13 seeing them from the TMI-2 accident. This is a photograph of 14 the cross section of that large plug at the bottom of the K-9 15 assembly. Shows the three fuel pellets, in this case, 16 surrounded by the metallic, previously molten metallic 17 structure. This is a groove of this cross section, and you l 18 can see, by the brightness here, that most of the fission 19 productivity is associated still with the fuel columns 20 themselves, even though the zircalloy cladding has been 21 dissolved and become part of the metallic structure down 22 here.
23 This is a section of the K-9, previously molten
() 24 metallics, previously molten ceramic in this case. The 25 radiograph shows most of the fission activity now is ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 8m336 6646
31630.0 cox 143
_, 1 associated with the previously molten metallic. We have 2 identified this activity as being associated with ruthenium 3 in the metallic, with much less activity down here in the 4 ceramic structure, indicating that with time, at temperature, 5 the ruthenium is migrating from the ceramic material to the 6 metallic when it's available. That's not really surprising, 7 based upon some more recent results. But the early separate 8 effects experiments did not show this occurring.
9 MR. MARK: Is there a vertical direction on that 10 slide, some of this on top?
11 MR. BROUGHTON: In terms of direction, this is at 12 the top of the consolidated region of previously molten 13 material. The previous slide was a particle or a large rock 14 from the bottom crust of previously molten material. In that 15 crust the fuel rods had not been melted and in vertical 16 orientation. Here we have got a composite from the very 17 top. So the two are separated by about four feet in the core
)
1 18 itself, this being at the top, the other being at the l 19 bottom.
20 MR. BROUGHTON: This is one of the rocks from the 21 0-9 position, which was out near the periphery of the core.
l 22 It looks, for all intents and purposes of this photograph, to 23 be a previously ceramic rock. However, when it was sectioned
() 24 -- it appears that the machine is having some ' fficulties.
25 On the previous rock, when we sectioned it, we found that it ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 8(XF336-6646
i 31630.0 l cox 144 )
f 1 was metallic and there were very few small pores but only a
(-) couple of large voids or pores in the rock.
2 l 3 This is an example, another example of a composite 4 structure. If you look closely you can see the remnants of a 5 fuel rod here and cladding here. It was sectioned in this 6 fashion. This is a cross section -- this is the photograph l 7 of the cross section of that particle. You can see the 8 previously molten material surrounding the fuel rod. This is 9 basically the fuel pellet. The molten material has flowed 10 into the crack and the pellet here and on the top here, you 11 can see the disintegration of the cladding. Showing some 12 photomicrographs here, the interaction of the molten material
() 13 inside the fuel crack, the interaction surfaces and over here 14 a closeup of the disintegration of the cladding by the molten 1 15 material. This is a photograph of an all-ceramic particle.
16 It looks very much like a lump of coal with substantial 17 amount of porosity.
18 This is a cross section of that particle, You can 19 see there's a substantial amount of layered porosity. We 20 cannot yet explain that porosity. The photomicrographs of 21 the voids or pores are showing an interaction layer on the 22 inside surface of those voids.
23 We acquired several control rods in a core boring
() 24 operation. This is a cross section of the tip or melted tip 25 of one of those control rods and on the lower left here is a ACE-FEDERAI. REPORTERS, INC.
202 347-3700 Nationwide Coserare KO-336 6646
31630.0 cox 145 1 typical microstructure for a fabricated control rod alloy and b 2 on the right here is the structure of this melted, previously l
3 molten control material. j l
4 MR. EBERSOLE: May I ask a question, since I guess l 1
5 this is the control rod part. If I could hypothesize a 6 slightly different sequence of events, if the control rods )
7 were to have melted prematurely before the core and sumped on 8 down, in essence, lost the top 1/2 or 2/3 of the core, I 9 think it's thought that the operators are trying to do 10 anything and everything they can to throw water back in the 11 core. If they were successful, they would throw water --
l 12 MR. BROUGHTON: I believe the boration in the
'( water is sufficient to maintain a subcritical configuration.
13 14 MR. EBERSOLE: Without rods?
15 MR. BROUGHTON: Without rods.
16 MR. MARK: That piece you pointed at a little 17 whila ago was fuel, that was uranium oxide at a density of 18 about 11. It withstood that state?
19 MR BROUGHTON: The fuel that is still contained 20 within the cladding itself still has a density of about 10.9 21 grams per cc, in that neighborhood.
)
22 MR. MARK: Right. The other material.
I 23 MR. BROUGHTON: Previously molten ceramic has a
{
O 24 much 1 wer density primarily because f the p r sity in the 25 structure. I think theoretical densities I showed you ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6M6
31630.0 cox 146
_ 1 earlier -- I forget what ZrO2 is. Depending on zirconium V content, the density is down what?
2 3 MR. MARK: I am wondering about the layering that 4 might ultimately be thought to take place in the bottom of i
5 the vessel, The fuel wculd be at the bottom or these other I I
6 things would be there just es well?
7 MR. BROUGHTON: If you will let me address to that 8 when I get to that point in the accident scenario, it would 9 be a little simpler to see. It's a good question. I have a 10 hypothetical answer for you --
11 MR. MARK: Please proceed.
12 MR. BROUGHTON: -- with the data from the core 13 boring operation. We had visual data of what the inside 14 surface of the holes looked like. From the contour maps we 15 have constructed vertical cross sections for each of the fuel 16 assembly columns shown here. In this slide, I have got the 17 vertical cross section for column G, basically, in the center i 18 of the core. This is the voided region, at the top of the i
19 core, below that was debris bed. Then there was this solid 20 surface that was intangible to us until we did the core 21 boring operation. The blue lines represent the two core 22 bores that were taken at G-8 and G-12 during the core boring 23 operation. This enclosed region here is the solid
() 24 consolidated region of previously molten material. That 25 solid region is composed of two basic structures, a central ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 804 336-6646
31630.0 cox 147 1 -- what we have dubbed a " homogeneous region," homogeneous in r"s O 2 the sense that we believe it was all molten, but it's 3 composed of previously molten ceramic encasing, previously 4 molten metallic inclusions or veins of material. We saw that 5 all through this structure. Then that region, homogeneous 6 region, was encased in an agglomerate or by an agglomerate 7 which was composed of rod stubs or rod remnants with 8 previously molten material flowing in and solidifying in the 9 material.
10 The bottom, in this region here, the fuel 11 fragments have a vertical -- basically a vertical 12 orientation. Whereas out here on the sides, the fuel 13 fragments had been disrupted and knocked around and we found 14 a rather random orientation.
15 But beneath that, where the rod stubs -- we 16 believe the minimum liquid level was around here at about 17 2-foot and we see very little, if any, damage to the rod 18 stubs in the bottom of the core.
19 Summarized here on this table the estimates of 20 core volumes and masses for each of the regions of the core, 21 the void volume, upper core debris, molten zone within the 22 core itself. Also the molten debris in the lower plenum, 23 standing rods. Basically we have concluded that at least 35
() 24 percent of the core was molten during the accident. About 20 25 to 25 percent of the core was fragmented, oxidized and ACE. FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage MXb336-6646 1
31630.0 j cox 148 i
. 1 fragmented. That was in the debris bed itself on top of the
(> 2 previously molten region. About 40 to 45 percent of the core 4
3 was unaffected by the accident. That's basically rod stubs 4 or rod remnants at the bottom of the core.
5 MR. MARK: What about the radiodistribution. You 6 said at the bottom of the core?
7 MR. BROUGHTON: Yes, sir.
8 MR. MARK: Edges, too?
i 9 MR. BROUGHTON: Some of those rod' stubs at the 1 10 edges extended all the way to the top of the core. Most of 11 them, of course, were at the bottom. You will see that in a 12 couple of slides. There were some rod stubs that extended to O 13 the top of the core. On the east side of the core, where the 14 core former wall was exposed, that is where the damage 15 extended all the way out to the core former wall itself.
16 I have summarized here the results of our fission 17 product research in terms of distribution of those fission 18 products and the inventory itself. We anticipate that there 19 has been significant retention of noble gases in undamaged 20 fuel assemblies. Iodine and cesium was dissolved in the 21 coolant water, consistent with cesium, iodine and cesium --
22 CsOH, ruthenium and antimony associated with metallic 23 materials, but essentially remain completely within RPV.
() 24 Lanthanides remain.
25 We have done considerable work to develop accident ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3364>46 I
31630.0
'cox 149 1 scenario, it's based on the~results of the examinations that
~
2 I have summarized for you, as well as evaluation of on-line 3 instrumentation and also the results from some of the more 4 independent accident experiments, such as the.PVF and LOFT 5 experiments.
6 This is a plot of reactor system pressure as a 7 function of time after reactor scram, going from 0 to 300 8 . minutes, does not extend out for the full 15.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> until 9 -forced cooling was reestablished. For the first.100' minutes, 10 it was primarily a thermal' hydraulic event. The coolant' 11 pumps were on, cooling was maintained within the core, but 12 the system was losing coolant inventory through the open 13 block valve.
14 At 100 minutes, the.B-2 pump, I think it was the 15 .B-2 pump, final pump was turned off, and the core began to 16 uncover. The block valve was still open, liquid levels 17 dropped, temperatures increased in the core. For the next --
18 until 174 minutes, when the B-2 pump was turned on. The B-2 f 19 pump was left on for about 15 minutes, but we believe it only 20 pumped water for 5 to 10 seconds, and we have estimated that 21 it pumped about 1000 cubic feet of water. Enough to 22 significantly refill the reactor vessel, but we also believe i
23 that, due to steam binding, there was significant bypass of ]
24 water around and inte one of the other cold legs. 15 minutes
(])
25 later that pump was turned off. But at 200 minutes the HPIS i e i ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 1
31630.0.
cox 150 1 was turned on at high flow rate. It remained on for 17 O. 2 minutes and at that time, we believe, pumped sufficient water 3 into the system to. totally reflood the core.
4 However, it was not until 224 minutes when.we 5 believe the molten material relocated from the core proper ;
6 into the' lower plenum.
7 As evidenced by a 300 psi pressure spike in the <
8 reactor system pressure, as well as response of-the SPND, 9' .self-powered neutron detectors and thermocouplers located in 10 the core and finally source range detectors which are out on 11 .the northeast and southwest sides of the core. We evaluated 12 those source range monitors. We believe they tell us about
'O 13 20 tons of material relocated from the core into the lower i
14 plenum in about one minute or less at this time period here.
15- I will now follow through with a series of 16 drawings of the condition of the core at different times, as 17 we currently see it. Between 100 and 174 minutes, the water 18 level dropped in the core and it began to heat up from the 19 top. Of course, the highest temperature, the higher 20 temperatures were in the central region of the core. Again, 21 the oxidized cladding was melted, liquified fuel and began to 22 flow down. At about 150 or 160 minutes, the minimum water 23 level was reached just above the second grid space there is
() 24 shown here at about 20 to 24 inches, and that stopped the 25 progression of this molten material downward in the core. A i 1
ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336 4 46 !
l J
31630.0 cox 151 l
1 stable crust was formed by the cooling of the water, and we
( )
2 get the beginnings of this crucible forming with partially 3 molten solid plug of material above it.
4 By 174 minutes, 72 to 74 minutes, we believe that 5 this configuration was contained, where the top part of the 6 core was essentially completely oxidized, metallic zircalloy 7 that remained had flowed down, dissolved fuel with it, formed 8 a consolidated region of partially molten material in the 9 central part of the core as shown here. The funnel shape of 10 this material was caused by the redirection of steam flow 11 around and the increased cooling with time, as the blockage 12 spread radially outward.
k# 13 We got a little bit ahead of ourselves here.
14 Let's go back one. During the oxidation of the zircalloy in 15 the top half of the core, there was substantial heat transfer 16 from the core to this upper core support plate. We have seen 17 damage to the structures, and that was -- but we see it in a 18 bimodal nature as shown on the next slide, where we see the 19 damage confined to the central region, top half of the core, 20 north half of the core. Another area here in the south half 21 of the core.
22 We also can see that some of this stainless steel ,
23 is foaming, indicating that oxidation was occurring near the 1
() 24 melting point. Yes, Dr. Shewmon.
25 MR. SHEWMON: I wanted to get to the temperature.
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646
l 31630.0 j cox 152 1 You are basically' burning the chrome out with the steam?
O., I 2 MR. BROUGHTON: Burning the chrome out. Also 3 oxidizing.
4 MR. SCHWINK: That's what I meant by." burning,"
5 thank you.
'6 MR. BROUGHTON: Some of it was oxidizing, l
7 oxidizing very vigorously, very foamy, looks not unlike a.
8 head of cauliflower. I am sure you have se'en photographs'of
-i 9 it.
'10 Other areas, noted here with the yellow, down in 11 here, here, the stainless shows evidence of melting. It has 12 dripped and'run, but no oxidation, indicating in these.
O- 13 regions that the gases that were rising off the core were 14 basically nonoxidizing, probably hydrogen. There are a few 15 isolated regions of melting in those structures outside of 16 the two primary zones, but in reality these other areas are 17 undamaged and when you look at them in the videotapes, they
.18 show very little, if any, discoloration or damage at all.
19 The gradients from undamaged to damaged are very, very 20 sharp. In fact, I have got a photograph of one assembly that 21 shows melting on one side of the sleeve that goes around the 22 top of the upper end box, melting on one side, and 23 essentially no damage or discoloration at all on the other 24 side of the assembly, eight inches away.
' t] _
25 After the pump was turned on and the injection of ace-FEDERAL . REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3364M6
l
, 31630.0 cox 153 j i
1 water, we believe that the water level rose to about midcore !
2 elevation. The heat transfer resulted in the thermal shock 3 and mechanical shock resulted in fragmentation of those 4 oxidized rod pellets or rod stubs at the top of the core, but
'S little effective cooling of this solid or consolidated region 6 here in the central part of the core: At this time we -
7 believe there had been no relocation in the lower part of the 8 core.
9 Shown here is a planed view of the core with the 10 instrument locations with solid dots, between 100 and 175 11 minutes all the thermocouple went off scale. They indicated 12 the temperatures had exceeded 700 degrees Fahrenheit, where O 13 there were no measurements of temperatures made, higher than 14 that, because the system wouldn't handle the output.
15 However, at 175 minutes, when the B-2 pump was 16 turned on, some of the thermocouple, those denoted in pink 17 -- this is consistent with the region I have shown of 18 previously consolidated molten material that we find today.
19 This leads us to believe that this solid region was formed by )
20 175 minutes when the water was injected by the B-2 pump, it :
21 cooled these thermocouple or the new junctions that had 22 formed, but not the central regions because those junctions 23 were never exposed or in a position to be cooled.
() 24 MR. MARK: Did they come through their position to 25 the bottom head?
ace FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coserage 8(KL3364M6
31630.0 cox 154 1 MR. BROUGHTON: All the thermocouple, and SPNDs, 2 came up through the bottom head, up through the core, with --
3 there were, I believe, seven SPNDs spaced, evenly spaced, 4 through the core, and the thermocouple junction was six 5 inches above the upper core support plate.
6 During the accident, because of the high 7 temperatures, those leads were destroyed. But new 8 thermocouple junctions were formed at the -- at a lower 9 elevation. We see this in both the LOFT and PVS test. Those 10 will provide an accurate measure of temperature. You just 11 don't know where it is.
12 Yes, Dr. Shewmon.
O 13 MR. SCHWINK: Was that 175 minutes probably also 14 the maximum temperature for the outlet?
15 MR. BROUGHTON: I think it probably was. But we 16 are completing some analysis for our standard problem, so I 17 can't say for sure.
18 MR. SCHWINK: Enough.
19 MR. MICHELSON: These thermocouple junctions that I 20 are forming had to be around the bottom crust of the molten 21 zone, weren't they?
22 MR. BROUGHTON: Yes. That's consistent. We, EG&G 23 and GPU, have made measurements using the no balance
() 24 technique, of where these new junctions exist today. Around 25 the periphery, the core of those junctions extend 2 to 4 feet ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3364 446
31630.0 cox 155 1 up into the core, consistent with this elevation here. Where
,]
2 GPU has removed assemblies at this point in time, they are 3 finding, when they go back and look, that the instrument lead 4 that was in that assembly is now down here on the support 5 plate and it's 2 to 4 feet long. The measurements we made, 6 what we are seeing during defueling are consistent with the 7 fact that those thermocouple junctions are in here. In the 8 central region of the core, those thermocouple junctions are 9 down here.
10 MR. MICHELSON: They are real, also, but down in 11 the bottom.
12 MR. BROUGHTON: I will get to the reason for 13 that. Between 175 and 224 minutes, the central region of 14 this consolidated mass continued to heat up. It became 15 completely molten during that time period, even though the 16 water level in the core may have been up near the top of the 17 core. It's because there was insufficient surface area to 18 remove the decay heat and prevent it from melting before 19 stable thermal conditions were established.
20 We have looked at the instrument response of 224 21 minutes. The pressure increased 300 psi. The source range 22 monitors show an increase of a factor of 2 at this point.
23 The cold leg temperature increased to 224 minutes indicated 24 superheated steam went into the cold legs at this time. The f])
25 SPND alarm pattern shows any pattern.
ACE-FEDERAL REPORTERS, INC.
202-347-37(X) Nationwide Coverage 8(X)-33(>6646
i 31630.0 ,
cox 156 l l
1 This is the vertical cross section for row 6. The i i
U~. reason for showing this is the top surface of the 2
3 consolidated mass dips over here on the east side. We had i 4 quite a number of data points to give us an accurate 5 description of the surface. We know'this dip is real. We 6 had 10 data points from which to develop this lower surface.
7 We had to project it out beyond where we could get our core 8 bore so we could get core bores at column 0 but not out at P f 9 or R. Projecting that surface, we had anticipated finding 1 I
10 the solid region or fuel rods extending after this elevation 11 here. Basically there is an anomaly here and what we 12 believed it represents is the flow point for the material, 13 the molten material out over the top of this consolidated 14 region and down into the lower plenum.
15 Shown here is what we believe the core looked like 16 at about 226 minutes following, just after the relocation.
17 What we think happened is this upper crust was 18 thin and unstable. In fact, it may have represented a melt 19 front from 200 to 226 minutes, whereby the molten region was 20 gradually progressing up into the debris bed itself.
21 It was unstable and thin and we believe it 22 cracked, allowing some of the material, 20 tons of material 23 to flow out into the lower plenum as shown here.
O 24 xR. MIcustson: Where was the water at that point, 25 there it is. I didn't see.
ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coserage 8(n 336 6646
31630.0 cox 157 1 MR. BROUGHTON: Up here. At least above the O 2 midplane elevation. Yes, Dr. Shewmon.
3 MR. SCHWINK: That metal that got there first that 4 you are supposed to tell us about.
5 MR. BROUGHTON: I believe,.and there is not a 6 consensus on this, amongst my colleagues at EG&G. I believe 7 that the metal that we now find in the lower plenum, this 8 region right here, or what we think is there, there is some 9 uncertainty in these measurements, what GPU did was they 10 inserted a very small gamma detector in the inside tube of 11 one of the instrument strings, and if the position of that 12 gamma detector was correct, then there appears to be a layer O 13 of nonradioactive material or material resting on the lower 34 head. That would be silver, we would anticipate, or metallic 15 zircalloy or steel.
16 MR. MARK: Like, what, 2 or 3 inches?
17 MR. BROUGHTON: As deep as B inches in the center 18 region, at the very center of the lower head. I said earlier 19 that we have found veins, large veins of previously molten 20 metallic in this region right here. What I believe happened 21 was that when this break occurred, that metallic material, as 22 well as the ceramic, flowed out and down into the lower i- 23 plenum. As it cooled, the metallic material settled out on
() 24 the bottom, with the ceramic material being on the top. It 25 turns out the density of the metallics is on the order of 7 ACE FEDERAL REPORTERS, INC.
202-347-37(0 Nationwide Coserage 800-3364646
31630.0 cox 158 1 to 8 grams per cc, the density of -- I think the density --
O 2 MR. SCHWINK: Silver is higher than that.
3 MR. BROUGHTON: Yes, 8, in excess of 8 grams per 4 cc. I forget the exact number. The density of this 5 previously molten ceramic material is 7 to 7.5 grams cc.
6 MR. MARK: The bottom stuff would have solidified 7 anyway. I 8 MR. BROUGHTON: As Phil pointed out, there is a
. 9 break in the core former wall. We believe that the material 10 flowed into the core former wall. It flowed all the way 11 around. Now we find what appear to be four breaks in the 12 core former wall on the west side of the core where defueling O 13 of rod stubs is in progress. Preliminary evaluation leads 14 one to believe that the break is the result of material in 15 the bypass region melting through and possibly flowing back 16 into the rod stubs. I believe that the primary flow is down 17 through the core proper itself because of the high resistance 18 to flow that exists in this bypass region, and the very rapid 19 nature of the relocation as indicated by the source term 20 monitors. When we defuel completely, we will have the 21 physical evidence that will tell us which scenario is true.
22 MR. MARK: Approximately when would you expect --
23 MR. BROUGHTON: When will we have that I
() 24 information?
25 MR. MARK: Roughly.
ACE. FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserare 8(XF3364M6
31630.0, cox 159 1 MR. BROUGHTON: Within the next year.
i%
V 2 Now, very quickly, I would like to just summarize, 1 3 where we believe the major impact from this research will be v
4 on the NRC issues. I summarized here the major areas of 5 uncertainty as identified in 0956, looking at 1150, Appendix 6 1150, the wording is different. I believe there is a great 7 deal of similarity between this and Appendix J of ,
1 8 NUREG-1150. I have X'd the areas where I believe the TMI-2 9 research will have significant impact. The one area where !
10 there might be confusion in your mind, when we X'd high 11 pressure melt ejection, what I,had in mind there was the
- 12 failure of the lower head itself prior to injection of the
( ,
's 13 material out into the containment vessel. In reading 1150, I 14 see that that event is included up here in core melt \
15
, progression.
sf 16 I summarize on the left column here the primary or 17 significant results from my research to date and specific 18 impact that that would have on those issues. Aboutp5 19 percent of the core has melted or at least that much, It 20 certainly indicates the core melting is possible during a 1
21 severe accident with only limited damage to the upper 22 plenum. We have examined two lead rods from the upper s'
23 plenom, one from H-8 and E-9 and find no melting.
t
() 24 Temperatures on the bottom of those lend rods reached only 25 1200 degrees in the center, about 1000 degrees on the ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6M6
31630.0
-cox 160 1 periphery. Only 700 degrees at the top, indicating that in
/~}
k- 2 general the upper plenum was not damaged, even though there 3 was significant melting. I believe that TMI potentially 4 could help address the basic issues or uncertainties i
> 5 associated with natural convection, heat transfer from the 6 core to those upper structures during the course of an 7 accident.
8 The melt debris in the lower plenum was eventually 9 cooled without failure of the lower head. Water, and the 10 impact here is that water in the reactor vessel, continued 11 ejection from the high pressure ejection system, did 12 tirminate the accident and did prevent failure of the lower 13 head even though this material was resting on the lower head 14 for a number of hours after it initially flowed down there.
15 There was no energetic steam explosion. This is 16 consistent with our current belief of no steam explosion, at
'7
, least at high system pressure.
18 CHAIRMAN KERR: How would you characterize it? I 19 notice you did get a high increase in pressure.
20 MR. BROUGHTON: That occurred over a 20 to 30 21 second time period, when you look at the raw data coming off 22 of the pressure transducer.
23 CHAIRMAN KERR: Do you know what the time response
{} 24 of -,. transducer is?
25 MR. BROUGHTON: les. It would have made a more l
ACE. FEDERAL REPORTERS, INC. I 2F 347-37(X) Nationwide Coverage 8(K)-336-6646 1
r 31630.0 cox 161 1 rapid rise had we had an explosion in the sense of a step 2 change in pressure across the system or across at away from.
3 MR. MICHELSON: That step was caused, you think, 4 when the molten material flowed down the shroud area and to 5 the bottom of the head?
6 MR. BROUGHTON: Yes. In discussions with Dennis 7 Berman at Sandia, he was surprised there was not more 8 disbursement of the molten material within the core as it 9 flowed down, which is what is needed in order to get the 10 surface to volume ratios in the rapid heat transfer. He was 11 very surprised that we have only found what appears to be a 12 relatively narrow column or channel region where the material b'-- 13 flowed into the core, into the lower plenum. The experiments 14 that they are doing at Sandia show, when they power this 15 material into pools of water, it is disbursed out and broken 16 up. We didn't see that here.
17 MR. MICHELSON: Did they power that in in a 18 configuration consisting a lot of fine rods?
19 MR. BROUGHTON: They did not.
20 MR. MICHELSON: Rods were still intact in the 21 area?
22 MR. BROUGHTON: Yes. There is a great deal of 23 difference between what they have done experimentally and
() 24 what we have done here. That's why I believe with a better 25 understanding of what happened at TMI, I think that we can ACE-17EDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800 33MM6
, 31630.0 ;
cox 162 l I
1 have an impact or help close this issue or reduce this l O 2 uncertainty in the area of steam explosions.
i 3 CHAIRMAN KERR: But will you ever be able to 4 predict that?
5 MR. BROUGHTON: I don't know.
l 6 MR. SCHWINK: What about the Idaho?
{
7 MR. BROUGHTON: Right now research is at Sandia, ]
8 not Idaho.
9 MR. SCHWINK: He is talking about research and 10 modeling.
11 MR. BROUGHTON: Retention of iodine and cesium in 12 the debris was significant. What this indicates in terms of
,A
'%) 13 the issues is that iodine and cesium would be present in that 14 molten material when it is ejected out of the core and into 15 the containment. For direct release to the containment by a 16 molten core concrete interaction.
17 Finally, on this slide, ruthenium appears to be 18 retained in metallic debris. Ruthenium is released from the 19 fuel but needs to be accounted for in our models.
20 MR. MICHELSON: Was there a drain line in the 21 bottom of the --
22 CHAIRMAN KERR: Mr. Michelson, if you could get 23 closer to the microphone.
24 MR. MICHELSON: Was there a drain line in the 25 bottom of the reactor vessel?
ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-33MM6
31630.0 cox 163 1 MR. BROUGHTON: No.
2 MR. MICHELSON: I thought all vessels had drain 3 lines --
4 MR. REED: I believe that went out with the S1W 5 experience in about 1953.
6 MR. MICHELSON: On boilers, they do.
7 MR. REED: I am talking pressurized water.
8 MR. MICHELSON: Okay, then there was no drain 9 line, 10 MR. WYLIE: You may be right -- maybe not.
11 MR. MICHELSON: Well, yes.
12 MR. BROUGHTON: To the best of my knowledge, there t
13 is no drain line. The only penetrations I am aware of are 14 the penetrations of the instrument leads and there is a 15 single weld of the instrument penetration nozzle to the lower 16 head on the inside surface of the lower head.
17 MR. MICHELSON: No way to blow down the bottom of 18 the vessel but clean it up once in a while then.
19 MR. BROUGHTON: To my knowledge, no.
20 We now have seen molten core materials have caused 21 limited melting of instrument penetration nozzle and core L
22 former wall. This could provide a more realistic 23 understanding of core damage progression and eventual failure
() 24 of the reactor pressure vessel structures. We are in a 25 process of doing some fairly extensive analysis of the ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3364M6
31630.0 cox 164
,, 1 -thermal response of the lower head. These leave possible U 2 thermal and-material interactions between molten core 3 materials and lower head itself. If those exist and we 4 looked at them, it would help enhance our understanding of 5 the processes which control damage to the lower head during a 6 severe accident.
7 MR. EBERSOLE: At this point, could you say that 8 it is correct to do what I think everybody thinks they will 9 do, that if you get into this state, nevertheless, you will 10 always try to throw water into the core, no matter what?
11 MR. BROUGHTON: I believe so.
12 MR. REED: On that basis, since it's not a e
13 characteristic core, and it was the first core, how do you ;
1 14 see this whole pattern of events unfolding if you had what 15 would be a more likely core configuration in aging?
16 CHAIRMAN KERR: Your Honor, I object. That's 17 conjecture on the part of the witness.
18 MR. BROUGHTON: It would, of course, result in the 19 greater decay heat, maybe a more rapid initial cleanup. It 20 would have some impact on the initial release of fission 21 products because the boundaries would be saturated or open to 22 the void volume of the fuel itself. !
23 So release of fission products might be greater or
() 24 more rapid.
25 MR. REED: Are you saying maybe the core would ACE-FEDERAL REPORTERS, INC.
l 202 347-37(X) Nationwide Coverage M4336 6646 1
l 31630.0 .
cox 165 i
, 1 melt quicker and the rods wouldn't? i 2 MR. BROUGHTON: I am saying the initial heatup 3 might be more rapid until you get to the very rapid oxidation 4 of the zircalloy. At that point in time,'.:hether it's low l
5 burnup or high burnup core, heat release, by the oxidization I 6 of the zircalloy, rapid oxidation is much, much greater than 7 decay heat. The release of fission products in a low burnup 8 fuel and high burnup fuel are controlled by different 9 processes. Low burnup of fuel is basically fusion controlled 10 until the UO2 structure is broken down by melting or .
1 11 liquefaction. Whereas high burnup fuel have concentrated on 12 grain boundaries. At high enough burnup, the pores on the A_
'\~ 13 grain boundaries are interconnected so the release can be 14 more rapid. You can have a greater initial release. Once i 15 the fuel was melted, you get these solid structures formed, I 16 don't see where burnup is going to have a significant impact 17 on the sequence of events during the accident.
18 Even though this accident doesn't neatly fit into 19 any of the categories we have established for severe 20 accidents, it was at high pressure. There was water 21 available. In many respects it's similar to a TML3 primer 22 and ST2D sequence. This was described as a bastardized 23 sequence of events. The operators responded to what they saw
() 24 to mitigate the transient they were in. Hopefully they would 25 prevent this from occurring.
ace.FEDERA'L REPORTERS, INC.
202-347-3700 Nationwide Cuerage 80(k336-6646
~31630.O cox- 166 ]
1 MR. EBERSOLE: Let me ask a question. With O 2 respect to the heatup of the core on a theoretical-basis, if 3 fission products went off to someplace, away from-the core, 4 like the noble gases, et cetera, what percent of the heat 5 generation runs off and is no longer in the core? . .
6 MR. BROUGHTON: It depends when during the If it's during the oxidation process, !
7 accident that occurs.
d 8 oxidation of zircalloy is about a decay factor of 10 greater 9 than decay heat, so it doesn't have an impact.
10 Once you reach this configuration, it appears that 11 the fission products that are in this solid consolidated 12 region are pretty much sealed there, because in this case we O 13 .had water around it forming a cool surface; it's going to be. l
- 14. difficult for those fission products --
15 MR. EBERSOLE: What percent is that of the 16 original?
17 MR. BROUGHTON: We haven't completed that today.
l 18 We are seeing 20 percent of the cesium and iodine up here, j 19 nearly all of the low and fission products. Krypton and 20 xenon has been released from this material but it's retained 21 down here. There's not a clear answer to your question.
22 MR. EBERSOLE: Complex mix.
23 MR. BROUGHTON: Complex answer and complex mix of i
24 conditions. l 25 I would like to now status the program and start ACE-FEIMRAL llEPORTERS, INC.
202-347-3700 Nat;&vit Coverage 800-3364M5
31630.0 cox 167 1 with that using this slide, to show where we have sampled and 2 are in the process of doing examination.
3 The green indicates those structures that we 4 believe have been adequately sampled for the purposes of the 5 existing goals of the program. The lead rod, centrol rod 6 lead screws, dry screws from upper plenum, two of them. We 7 have samples of structures from the upper cores, samples from 8 debris bed, samples from the solid consolidated region, the 9 rod stubs themselves. We will have samples of the lower 10 plenum debris, and get more samples of the lower plenum 11 debris, and we will require at least one sample of the melted i
12 instrument structure down here that we have seen.
I rs l k-) 13 On this slide and the next slide, I have 14 summarized the remaining research. We have the complete 15 examinations of those structures and complete accident 16 scenario after we get the information and publish those 17 results.
18 We are -- we have basically completed the 19 qualification, evaluation qualification of the on-line 20 instrumentation and we are also in the process of working 21 with the CS&I countries in establishing a program for the 22 TMI. We are completing the demonstration calculation for 23 that exercise. We are starting the transition of' using the research results to determine'some of the unresolved severe
(} 24 25 accident and source term technical issues. Over the next ACE FF.DERAL REPORTERS, INC. {
202-347-3700 Nationwide Coverage MO-336-6646
p -
31630.0 cox 168 1 year and a half we will complete this work.
()-
2 In conclusion, I believe that the TMI-2 accident 3 was a mitigated severe accident. That is, that there was 4 fuel melting.
5 The postulated scenario for the accident is 6 basically consistent with the results from our sample 7 examinations and the on-line instrumentation as well as the 8 results from other independent smaller integral effects and 9 separate effects.
10 Finally, I believe that the results from the TMI-2 11 research will have a significant impact on resolving existing 12 uncertainties in the identified technical issues.
) 13 MR. EBERSOLE: What do you say about the fact that 14 it was mitigated by unorthodox and not planned for methods?
15 I remember they were fearful it was a reality if they went 16 into the RHR injection mode, seals would leak, they would 17 transport stuff out into the aux building. What did they 18 do? They evaporated into the condensor.
19 MR. BROUGHTON: They were responding to what they 20 believed to be a liquid full system as indicated by Ji'uid 21 level in the pressurizer. As such they were in a form of 22 feed and bleed operation. HPIS was turned on periodically.
23 It wasn't until after 15.5 hours5.787037e-5 days <br />0.00139 hours <br />8.267196e-6 weeks <br />1.9025e-6 months <br /> into the accident
)
{} 24 that someone realized or recognized the situation that they 25 were in and ordered that the block valve be closed, the high ACE FEDERAL. REPORTERS, INC.
202 347 3700 Nationwide Cmerage 800-33M646
31630.0 cox 169 1 pressure injection system turned on.
2 I think what it demonstrates is with water in the 3 system, a relatively small amount of water in the system, j
4 there is still a significant margin of safety to protect the i
1 5 integrity of the reactor or the reactor coolant system.
6 MR. EBERSOLE: They took it out onto the
-?
7 condensers, not onto the emergency system, in view of c.
8 flaws in the seals and other things?
9 MR. BROUGHTON: I can't comment on that. I don't 10 know. But I know that large consolidated region, it took a 11 long time for that to cool and the limited water in the 12 system eventually cooled it.
13 MR. EBERSOLE: They stood away, didn't they, from 14 putting the circulating system out into the equipment area.
15 Which would have been --
16 MR. BROUGHTON: I think the answer to your 17 question is yes, but I am not that familiar with it --
18 MR. EBERSOLE: I thought that was a recognition of l 19 the shortfalls of the system.
20 MR. BROUGHTON: I can't comment, I don't know.
21 MR. MICHELSON: Would you comment on why you 22 didn't burn through the instrument tubes at the bottom of the 23 vessel when the molten mass burned through the bottom.
() 24 CHAIRMAN KERR: Would you repeat the question.
25 MR. BROUGHTON: Why didn't the material burn ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 4 46
l 31630.0
.cox 170 '
I through the instrumentation tubes or head immediately after-t3
\/ 2 it burned through to the lower plenum? I believe the answer 3 to that is, first off, we are in the process of looking at 1
4 this in detail with the available analytical tools. But I l 5 .believe the answer is tied up with the thermal capacitance of 6 the head,.very large structure, 5.5 inches thick, solid 7 steel, has significant thermal capacities. The lower plenum 8 itself, the reactor vessel was half full of water. That ]
l 9 acted as a significant heat sink.
10 Also, when the material flowed down into the lower 11 plenum, it did fragment to some extent. That permitted it to 12 be cooled. I think the combination.of the cooling of the ;
(} 13 water and the large thermal capacitance of the lower head 14 protected the lower head. But whether or not there is damage ;
15 to the lower head, we know now that there is damage to some :
16 of the other structures.
17 MR. MICHELSON: When we try to transfer this 18 knowledge to other plants like boiling water reactors, I 19 think we ought to take a real close look at that draining.
20 Boilers used to have and I think still do have a drain line i
21 in the bottom of the vessel. It's not a real heavy gauge i
22 pipe compared with the thickness of the vessel, because, 23 it's, I think, somewhere, two to four inches in diameter.
24 But it looks like an ideal burnout point if you get a molten
.[ }
25 mass entering that drain line. Drain valve is a fair ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-3364M6
......_._....J
31630.0 cox 171 1 distance from the bottom of the vessel. If you start pouring O,
2 molten metal down into the drain line you may burn out the 3 drain line.
i 4 MR. EBERSOLE: It will get into the rods --
5 MR. MICHELSON: No, the rod drives are much 6 stronger than drain lines.
7 CHAIRMAN KERR: I think we ought to see how much 8 information we can get from our speaker before we conduct our 9 own discussion, in light of the schedule. Mr. Mark.
10 MR. MARK: The progress of damage through the core 11 was on a rather long time scale in this accident. In recent 12 discussions, there's been a lot of hangup, complication, on O 13 attempting to calculate the core damage progression. I 14 suspect, mostly, in situations where the time scale would be 15 greatly compressed, or thought to be possibly greatly 16 compressed.
17 To what extent do you think the lessons that you 18 have just been showing us so nicely can be transferred 1 19 perhaps on a different time scale to attempts to calculate 20 and discuss a picture, core melt progression and possible 21 severe accidents?
22 MR. BROUGHTON: First I would like to say that I 23 believe that the 2.aitial progression of damage to the core
() 24 during the TMI-2 accident was very typical, from 100 to 175 25 minutes was very typical of what would probably occur during ace-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646
-31630.0 cox 172 1 'most of the -- an S2D or TMLD prime accident. Beyond that j O, >
2 pump the B-2 pump was turned on, first off. What we'see is '
)
3 consistent with what we have learned from the PBS severe 4 field damage experiments'from what we saw in the LOFT I
5 experiment, and I believe I think what I see in the NRU tests 6 .that-the NRC has conducted. But those tests and experiments 7 basically terminate'at that point in time. They are 8 terminated before the damage or the rods are completely j
\
9 consumed. That's where we stood with TMI. Water was kept in l 10 the core.
11 We have put together a standard problem from which 12 others can benchmark their analysis capabilities. We have O 13 provided that CS&I to most of the countries in Europe, Japan, 14 Canada and the NRC, as operating through the INEL; Sandia and 15 Los Alamos are running a program; Battelle Columbus is 16 running the 0772 codes. I think when March -- the 17 development on March is complete, they will probably include 18 that.
19 MR. MARK: I thought March had been --
20 MR. BROUGHTON: Basically, I believe that the TMI 21 will provide us a significant and important benchmark for 22 these analysis tools.
23 MR. MARK: Will that not have a good chance, then, O 24 e = err 1=9 the vre e=t re or vo eid111 tie 1=
25 attempting to calculate the core melt progression picture? !
ace-FFDERAL REPORTERS, INC.
202-347 3700 Nationwide Coserage 800-3364546
31630.0 cox 173 1 MR. BROUGHTON: I would hope so. It's my opinion I
th 2 that it will.
3 MR. SCHWINK: He is not as optimistic as you are, I 4 but somewhat.
5 MR. BROUGHTON: Another way for me to state this 6 is I believe those codes should be.able to reasonably predict 7 the TMI-2 accident with reasonable accuracy, or correctness, 8 if we are going to have confidence in them.
9 MR. MARK: I don't want to predict the TMI-2 10 accident, I want to get rid of the direct containment heating 11 morass.
12 MR. BROUGHTON: That's really outside the scope of (3
V 13 the TMI-2 accident except from the standpoint that the damage 14 that might exist in that lower head, understanding it, would 15 help us understand the conditions or the processes that 16 control the meltthrough, the complete meltthrough.
17 CHAIRMAN KERR: You said earlier that this would 18 help us understand something about natural convection heat 19 transfer. That has a bearing on the DCH problem.
20 F.R . BROUGHTON: Yes, sir, it does. It's the 21 combination of the natural convection and the heatup of the 22 lower head and competing effects for failure of the reactor 23 coolant system.
O 24 MR. sCHWINK: MetrROG people say in the scenarios 25 they have worked with so far, even after the molten material ACE FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
31630.0 cox 174 1 comes down it resolidifies. That's not inconsistent with 2 what you found there.
3 MR. BROUGHTON: It's not inconsistent. That i
4 appears that that is exactly what happened. For the d 5 remelting to occur, water to the reactor vessel has to be 6 completely cut off or essentially cut off.
7 CHAIRMAN KERR: Further questions or comments.
8 MR. EBERSOLE: I think it was not within your 4 1
9 scope to look at the -- what I said earlier, the coolant of l l
10 crack arrestors, which would have stopped progression to this 11 disaster. For instance, like real knowledge and inventory of l
12 water, didn't have it. The recognition of super heat, they O 13 didn't have it. The information transfer from the 14 engineering sector to the operating sector.
15 CHAIRMAN KERR: I think that's a comment not a l 16 question.
l 17 MR. EBERSOLE: Do you find this, in your review, l
1 l 18 which has been extensive, to be about the right view; it 19 shouldn't have happened?
l 20 MR. BROUGHTON: I don't know what should have 21 happened. I have looked at the accident from a different 22 perspective. Not one yet from the perspective of how best to 23 mitigate or apply this understanding to develop the correct
() 24 procedures for mitigation. But I do believe that we are 25 looking at a mitigated severe accident and there's ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Cmerage 800-3364M6
l )
31630.0 j cox 175 1 information there that could be help.
- \
2 MR. EBERSOLE: I am looking at the prevent end of 3 it.
1' 4 MR. BROUGHTON: Yes. How do you prevent it from 5 occurring and how do you mitigate it?
6 CHAIRMAN KERR: Mr. Mark.
7 MR. MARK: Mr. Chairman, I would like you to ask 8 you to please thank our last two presenters for a I 9 fascinating, absolutely marvelous presentation.
10 CHAIRMAN KERR: I will thank our last two 11 presenters for an absolutely fascinating, productive and 12 informative session. i i
13 We will take a 10-minute break. i 14 (Whereupon, at 4:50 p.m., the meeting was 15 concluded.)
16 17 18 19 20 21 22 23 0 24 25 ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646 f
CERTIFICATE OF OFFICIAL REPORTER O
This is- to certify that the attached proceedings before the UNITED STATES- NUCLEAR REGULATORY- COMMISSION in the matter of:
NAME OF PROCEEDING: ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 327TH GENERAL liEETING DOCKET NO.:
PLACE: HASHINGTON, D. C.
DATE:. THURGDAY, JULY 9, 1987 l
were held as'herein appears,.and that this is the original transcript thereof for the file of the United States Nuclear Regulatory Commission.
(sigt [
(TYPED)
HENDY S. CO Official Reporter ACE-FEDERAL REPORTERS, INC.
Reporter's Affiliation O
g NRR STAFF PRESENTATION TO THE ACRS I
SUBJECT:
Integrated Safety Assessment Program 1
DATE: July 9,1987 Q PRESENTER: Michael L. Boyle PREShNTER'S TITLE / BRANCH /DIV:
Project Manager / Integrated Safety Assessment Project Directorate / Division of Reactor Projects III/IV/V and Special Projects PRESENTER'S NRC TEL. NO.: 49-27636 l
l l
1 l
O l
1
O V
HISTORY OF ISAP e NRC INITIATED THE SYSTEMATIC EVALUATION PROGRAM (SEP) IN 1977 TO REVIEW OPERATING PLANTS AGAINST REGULATORY REQUIREMENTS THAT HAD EVOLVED SINCE THE MAJORITY OF REACTORS HAD BEEN LICENSED.
e PHASE I 0F SEP DEFINED A SPECIFIC SET OF SAFETY ISSUES TO BE REVIEWED e PHASE II 0F SEP WAS A PILOT REVIEW OF ELEVEN PLANTS; REVIEW INCLUDED:
- DETERMINISTIC REVIEW OF SAFETY ISSUES PRA 0F INDIVIDUAL ISSUES O -
OPERATING EXPERIENCE REVIEW
( ,/
e PHASE II 0F SEP IDENTIFIED SIGNIFICANT FINDINGS RELATIVE TO THE SAFETY AND EVALUATION TECHNIQUES FOR ALL OPERATING REACTORS.
e PHASE IIIOF SEP WAS TIE PLANNED APPLICATION OF THE PHASE II FINDINGS TO ALL OPERATING REACTORS.
1 i
j 1
l O
l Q)
HISTORY OF ISAP e THE INTERIM RELIABILITY EVALUATION PROGRAM (IREP) WAS INITIATED BY TIE NRC FROM THE TMI-2 LESSONS LEARNED (NUREG-0660) e PURPOSE OF IREP WAS TO:
PERFORM PLANT-SPECIFIC PRAs FOR SEVERAL PLANTS TO SUPPLEMENT THE RISK-RELIABILITY FINDINGS IN WASH-1400 TO DEFINE METHODS TO CONDUCT PLANT-SPECIFIC PRAs FOR CONSISTENT, COMPARABLE RESULTS e TIE RESULTS OF IREP WERE TO BE APPLIED TO ALL PLANTS IN THE NATIONAL RELIABILITY EVALUATION PROGRAM (h1EP)
O 1
l 1
i
(^')
t/
HISTORY OF ISAP e ONE OF THE SIGNIFICANT FINDINGS FROM BOTH SEP AND IREP IS THAT ISSUES RELATED TO SAFETY OF OPERATING NUCLEAR POWER PLANTS CAN BE MORE EFFECTIVELY AND EFFICIENTLY IMPLEMENTED IN AN INTEGRATED, PLANT-SPECIFIC REVIEW e NRC MERGED THE DETERMINISTIC REVIEWS OF SEP PHASE III AND THE PLANT-SPECIFIC RISK ANALYSIS OF NREP INTO A SINGLE PROGRAM, THE INTEGRATED SAFETY ASSESSMENT PROGRAM (ISAP, 49 FR 45II2) e ISAP WAS MODIFIED IN 1985 INTO A TWO-PLANT PILOT PROGRAM THAT WAS TO INCLUDE PLANTS ALREADY REVIEWED IN SEP.
e NU VOLUNTEERED MILLSTONE UNIT I AND HADDAM NECK TO PARTICIPATE IN ISAP O)
L i
l O
l l
l 4
i
)
O I
SCOPE OF ISAP INTEGRATED ASSESSMENTS OF OPERATING NUCLEAR POWER REACTORS ARE CONDUCTED ON A PLANT-SPECIFIC BASIS TO EVALUATE ALL LICENSING ACTIONS, LICENSEE INITIATED PLANT IMPROVEMENTS AND SELECTED UNRESOLVED GENERIC / SAFETY ISSUES TO ESTABLISH IMPLEMENTATION SCEDULES FOR EACH ITEM. IN ADDITION, PROCEDURES ARE DEVELOPED TO ALLOW PERIODIC UPDATING OF THE IMPLEMENTATION SCHEDULES.
MAJOR ELEMENTS OF ISAP ARE:
REVIEW 0F LESSONS LEARNED FROM SEP PERFORMANCE OF PLANT-SPECIFIC RISK ASSESSMENT ANALYSIS OF OPERATING EXPERIENCE INTEGRATED ASSESSMENT OF TOPICS INTEGRATED IMPLEMENTATION SCHEDULE O
)
i
, j b
ISAP REVIEW e ISAP CONDUCTED BY AN INTEGRATED ASSESSMENT TEAM (IAT) e SCREENING REVIEW BY BOTH IAT AND LICENSEE TO DEVELOP TOPIC DEFINITIONS, SCOPE, REVIEW CRITERIA AND JUSTIFICATION FOR CONTINUES OPERATION.
e ISAP EVALUATION IS DIVIDED INTO THREE PARALLEL PHASES TOPIC EVALUATIONS (DETERMINISTIC AND PROBABILISTIC)
PERFORMANCE OF PROBABILISTIC SAFETY ASSESSMENT (PSA)
EVALUATION OF PLANT OPERATING EXPERIENCE e LICENSEE PERFORMS AN INTEGRATED ASSESSMENT OF TOPICS TO DETERMINE WHICH WLRRANT CORRECTIVE ACTION BASED ON PERCEIVED SAFETY SIGNIFICANCE AND TO DEVELOP COST-EFFECTIVE ACTIONS TO RESOLVE q
v MULTIPLE ISSUES, WHERE PRACTICAL.
e STAFF ISSUES DRAFT INTEGRATED SAFETY ASSESSMENT REPORT e DRAFT ISAR REVIEWED BY LICENSEE, PEER REVIEW GROUP AND ACRS e FINAL ISAR ISSUED: INTEGRATED IMPLEMENTATION SCHEDULE AND METHODOLOGY TO MAINTAIN SCHEDULE IS FORMALIZED.
C) v
I 4
U EXAMPLES OF ISAP FINDINGS 1
MILLSTONE I 64% OF TOTAL CALCULATED CORE MELT FREQUENCY WAS DUE TO FAILURE TO MAINTAIN ADEQUATE LONG-TERM DECAY HEAT REMOVAL CAPABILITY HADDAM NECK
- LOSS OF MCC-5, IN THE SWITCHGEAR ROOM WOULD CAUSE A LOSS OF FUNCTION OF CRITICAL EQUIPMENT AND PREVENT SAFE SHUTDOWN ISAP ALSO IDENTIFIED AREAS THAT WHILE DEVELOPED TO INCREASE PLANT SAFETY OR AVAILABILITY, ACTUALLY INCREASED RISK.
- MILLSTONE 1, DEGRADED GRID VOLTAGE PROTECTION
- HADDAM NECK, NITROGEN BLANKET FOR THE DEMINERALIZED WATER O STORAGE TANK. i O
p i
e
_ ,e\
r
(_)
ISAP PILOT PROGRAM CONCLUSIONS I
e PERFORMANCE OF PSA BY TE LICENSEE, AND REVIEW OF TE PSA AND OPERATING EXPERIENCE BY TE STAFF HAS LED TO A BETTER UNDER-STANDING OF TE PLANT'S OPERATING CHARACTERISTICS AND CAPABILITIES BY BOTH TE L7CENSEE AND TE STAFF.
e INTEGRATED ASSESSMENT HAS TE POTENTIAL TO IDENTIFY COMMON ELEMENTS IN SEPARATE REVIEWS AND PROPOSE A SINGLE ACTION TO RESOLVE THEM.
e INTEGRATED ASSESSMENT PROVIDES AN OPPORTUNITY FOR TE STAFF AND LICENSEE TO ADDRESS PENDING REQUIREMENTS ON A PLANT-SPECIFIC BASIS.
o ISAP PROVIDES A-FORMAL PROCESS TO EVALUATE THE PRIORITIZATION OF ALL PROPOSED ACTIONS.
o PILOT PROGRAM HAS DEMONSTRATED TE POTENTIAL BENEFITS TO LICENSEES, THE PUBLIC AND THE NRC OF INTEGRATED ASSESSMENTS g USING PLANT SPECIFIC PSAs AND OPERATING EXPERIENCE REVIEWS.
V) e PILOT PROGRAM IS CURRENTLY SCHEDULED TO BE COMPLETED BY THE OF 1987.
1 o
r; tg ,
.Q-
.V '
FUTURE OF ISAP e STAFF IS CURRENTLY PREPARING A COMMISSION PAPER DETAILING:
ISAP PILOT PROGRAM EXPERIENCE RECOMMENDATIONS FOR CONTINUATION OF ISAP e PRELIMINARY RECOMMENDATIONS CONTINUE ISAP ISAP BE MADE AVAILABLE TO ALL LICENSEES POSSIBLE INCORPORATION OF OTHER RELATED PROGRAMS (e.g., SEVERE ACCIDENTS, INTEGRATED LIVING SCHEDULES)
LO e COMMISSION PAPER IS CURRENTLY UNDER NRR UPPER MANAGEMENT REVIEW l
e- EXPECT PAPER TO BE' ISSUED BY END OF JULY 1987.
I l-l
~
s 9
t n
e n m i o
u s c i s
o m D m o
e0 C c5 n1 r y
e1 r -
)
t f
, 'i / .t oa f
eGa ~
o ..
l u
gi g R eErR (Df,fW% e U R ks N r a
i e
R l c
r u t
o N c S.
a U e
R
~
6
nformation =ow in \UREG ~50 e Core Damage Frequency Estimation (Chapter 3)
Definition of Plant y Damage States Containment Analysis (Chapter 4)
Definition of Containment Failure Bins Source Term Analysis (Chapter 5)
Frequencies of C '
pro il tjes Release Characteristics Plant Damage of Containment States of Containment Failure Bins p Failure B,ns i
Offsite Consequence Analysis a
W (Chapter 6)
Consequences of p u Containment Failure Bins Risk Estimation (Chapter 7) i Risk-Reduction Potential (Chapter 8)
Plant Specific Results Plant Specific Extrapolation of Results PRA Data Base (Chapter 9) and insights Generic insights y Regulatory Use (Chapter 10)
Key Uncertainties
& Limitations u Planned improvements (Chapter 11) e rrL /
2
5.
n n
p n q o
ts c ss i
s hius keo
,t s nf ege hi t s nf om r ci i
n m
digne ids n ri t go o
,c n ng diCn yce a n p i
t a
cn am ddaer i nl y r e
egs e s ed ei ho u a eia vn s v c u rif2loi) a t d i
s q sf emst t spc(edgu n o
fr aaa n a d c d o dpne f
oehep esl ha t
n s aR o r
, r s gort nctnn egn sa t f e i nc o eer mna i r
el e v e
i v de n r d qf e u t l
e apt t cd o0 t
c aeeef husumNa t l au n b5 q e
t vssd i
f a a11 j sr eat ov iyo e; r e-b e sbn ,e e v s erarhe ot n hG t E O do nt sda tih s) e (1 vtf aon a nR u e nc n: nl oU ru act s edp ahnal l
f ui no usm ge t sN t
a srd t g edsece seerrrv nla e
ri ee tess i
t ui t e n gtn wandt e unkno e
i a a es ce mif aioia t t hgmdef me a pi lnpn t irf e oh er r t e nn c ct soo cd i
deayecm si aee ng icel l t i
t r i n
v ai r o nlc an nn r u u r rei a eaiemsio s
s mc r m r e el i
t ba pd n e ef n t a odv e a op e
ont t neo ofn sti o ehn Tp or Taairc Tit
- q lllllllll [
4 D
R R
o d y r s b el ne e ca a e sht r s d
urn t r gp y npo e l oe n e a de ei w i i
dne e st x l
ei v n a
crS ce i
v de ds one y alt d e ny t _
e ana r ab mtx i n
a r o s yd de et ,
e ge e t v s C e l o w vo r e _
e eR sin t
r o
ni e ot rd c
n d t e ve p u y
f otu A
, n e mr mtec e v
i g y sQ e o ,
ge v nl s i
j o csl nb s nua t eo i yu n l
eon n ho os e Ood uir qvet e e m
pt t tr l
p s, h e
o e r n n n mle r h fr pi i a e a ed p
t e y t d e- s o m
f o mb i
nin ch o M srd o os ct - em s
c cn af ys is f e ysw o es o p t h
t p dces re l
ac i w
l t e et u' ne ahiu va o l
i i ngvo a aff n aiers tW ef et r s sg e e o di n rG i
t do gw ch a te gi n st f nlt yy on e ea m an iiyt s a l ei t
s su ve i
i ue i t
s v yl eo au l
mrs ee i
s ys r qh e t w e _
rl ns bpan r v l amr s se k t i e n eiv s bmf o xn nr e At ep ot i AeoC EU Claer R O - * * *
- l ll lll l t
s h n t n
go ui h e r
s o r s e h fa f i
s y . t f
i d
a l
n sA e A i eR t y
tn P a g
t n ida r i t
i a
t r a s e
t r ed v e cn n c
n na ut is U
a n s t ono e y t i c ai st n
e ds ep u
q f
l e dm i
u u e od t s r n o ss F
t i om y a n t ayt t ig e vn d gl i i i i t l ab i
c ies A
c pa ndo ol r e i
e Pr Sm e
o t
A e.r O te k
r f
o ed e ce t s t i
n i nnr ne ag t u
l y f oe sa ye t
xn i
t s nd e e
ge t l e ar b ualu l
t l s a nl ae ne u ip st c i
s q dm ib t
n o ol b rdi an y e n at da ei l d ao t c t en vasl t a a s br se so as evu s n rt n me d eo uo ba l s ta nt A e s dn sm ,e i
y s ns ui dl t sr s t
r n oin eeias t e ei l re e t
n d r e s n a vts o en u pif s pmi xic r O i a d ue rt xes e r ee t a am pu edi yp edc n r ce o mb f t bxyu e
c t
oy n i y
otona s eros nd be sd t t t rt s n et u eh g p ht a u ndr e reo ne aa eht U up l
r eS r eS mip mi gm k wyt l an t e
sy gqium ge e o i
s se sd at bef de ur y d;r u e c R ef ei j l J c rn o e ia tS ic t e a e ea
,l e n n n c gS e vgt i
edo i
ta o r ia a dr eo r
o idn viet ic rt t e r r l f e
t cee e
t csdnei ec ca ee cv wc oa t
r jwo l t e
jdeg ne e bop bon r a
UR ne Us KR ne h unf uo S gi sm T Sko O .
k s
r i s
e r t f e r o m o ee s ar f
l uu e a s aiq s t vn
. p hs y ge n hc d -
e s l i .
a a e n y n a wla et A s dn mg -
y r na hi n -
o ad cl t t ip _
n c sn hm i
a f a ea w a t
r t l u a ns e n a e at vad i
ha l
c c i t
c n t r n e wit U o t e p a el rb ei ts k m m r ca ga i
s i o i sl i nt as R k s
f r dav r g
e t en c~a i
r r p l o y s eg bs i
f l l
t nm l
ei n au n
d i a yer e ss oe o t n cmt y u sl i h
t emneien l l
ar s a y el e t e oh qnu u t ac r e nt e t
r M pt pi lk t gr eoo ee o c
ei nFCS mm ga l
el d l
ei- v - - dr vu eo Sd r ua Jp Dw
) ll !
2 1
D R
R O
0 '
4 1
- h ~
0 I
2 1
l a t _
X " 0 i
c n e
I 0 )i s
t 1 s t m (
p is t ce t
aeu e r r c 0 t
n e
i n I 8
Ss DI *
. m e
rI s f e O o oru r
c n
f g es i n 0 e t d s I 6 r ul i
ue u p p t r s nm i
nP s e
gg .
. x -
0P r
ia t
an '
I 4
rS Mi a t e e p H x I 0 2
E -
_ X 0
0 8 6 4 2 0 1
0 O 0 0 0 Byd E=$o3 O _
l
~
0 0
- 2 e
g sn da 0 aR o
6 1
L e r
gu ns ~
)
i a
is t
s ae _
p _
er 0 2 (e HP 1 r
u s
t e s cr e el u r
} P ri ia t n
DF f t 0 m e
on e r
8 ni n e u a s
t n -
om sn d s
e r C o
i r a i aP o
L e at r u
pn H 0 o
l t
n Ci a 4 DF mC o
l P
a i i x
_ ~
Cd n y r i 1 r _
u a S _
O 6 5 4 3 2 1 0 0 O o 0 _
)
f f oE'.2O,_i
Information F.ow in NUREG-1150 e
~Co re:De~m a g'elFre q u e nc y~ Es tinta tio'n x hw J._ 't )(Cliahter13)SM"M--45 Definition of Plant y Damage States Containment Analysis (Chapter 4)
)
Definition of !
Containment Failure Bins Source Term Analysis (Chapter 5)
Con i a Frequencies of po Release Characteristics e
Plant Damage of Containment States of Containment Failure B,ns p i Failure Bins Offsite Consequence Analysis (Chapter 6) h Consequences of p p Containment Failure Bins Risk Estimation (Chapter 7)
Risk-Reduction Potential (Chapter 8)
Plant Specific Results Plant Specific Extrapolation of Results PRA Data Base (Chapter 9) and insights Generic insights 9 l
Regulatory Use (Chapter 10)
Key Uncertainties
& Limitations n Planned improvements 9 (Chapter 11)
___.___._.__a
i O
/ - = -
- E -
- - m
/ E l E E I E E E l l
i s ae E = E E E I E I =
t ed Dut i
9/o = = - - - = - = =
f S o r 9/ E - - E = E E = -
l eh vt e
/ E .
E E E E _
E E E E O
LeO 0t o n o s 5 i t m 1 d a e 1
e r
m r l s t
s y
o o Ga Ep f
n t s sr t
r n t m
e S
t e
s I
n o o s r u o
Rm f o
e v
t c C S y p C a
Uo y E F
a y
/
n o
l a
p u n NC c a
r i g
n a
n r
e v
i t
a i c
r S
r o
m u
c a t
a m o u t c e m t i c t e ht c a i t
n u e c l o A D I H R A E O C O
3 asis "or :le "30x anc W1is<:ers" Curves e 1E-2 3 Key: U 3 pper 95%
_- Mean
-- Low 5%
1E-3-- ,
- a Study 2 Bound 5 o
(5.3E-4) 2 &
c
,g ?
_, Study 2 Mean :s B o____________ -
o 1E > (1.5E-4) m e
@ 1'0 E -
m : c' b D
= E Y
h u-s.
o Base Case a
c e
o E1 -
$< 2.9 E- 5 0 - - - - - - - - - e - - - - - - - - - - - - - -
Q Mean *$ 'E S3 -
m --
(2.9E-5) - E E - o y 1E-5-- 5 0 -- g
. : g <>___________ . .
g B
. Study 3 Mean m u -
m (7.4E-6)
E 3.7E -
e- ..
1E-6--
o E E _ __ __ _ _ _ __ _ _ _ _. Study 3 Bound gg (1.7E-7) 1E gn -
O .5 g, eo=
bb
==[ _
1 33*$
II$0o o
OOI o n. O I I I I I Base Case 1 23
~ Sensitivity Studies :
Collective Result g
(!- -( _l.,. "e
]
O d
_ ~ nl f
~
I au GrG e
g as mt aa n -
I h m co at et DP l
P Bo ee r
o c n Cer h a
y e e. . _
_ I o
u rf q
ee_.
vR.
S e O e
Sf o. .
f os. e n ni I
o o c i
Z s n.
ie r
aq. u per mF o
Ii 1i I
y r
r u
C S
- _ ~_ _
- - _ ;: =: _
3 4- 5 6 7 E
1 E
1 E
1 E
1 E
1 e
M$ 'S8y BQ c58L L
O f 5 lu 0 G 1 l
a 9
e 1
d 9 - - - - - x p n i g a r
9 c G 2 i n
r a *s p mtn t m
o e- l i
y r
a a B t
o 6 1 2 0
1 a s -
DlP h c
8 - - - 1 -
2 x s e
a c e
r e P e 8 e n
oc h a
4 t
o n
Cne y 0 o 6 3 9 1 1 1 er u - 5 - 3 x a or t e S q
e 0 1 c y
f n 4 e
$e oR i Z
n o 2 1
- 8 1 - -
9 7
0 1
x f q
e r
u t r 5 e
bfuo 1 5
g a
m i y 0 tr s r
r 8 4 0 8 4 6 1 a u 3 2 2 - x d ne S 6 2
e r
o oi c ,
c Cn y c ,
t o
l e ne ou n
s au pq iq t
ue r
e w t g e t r
ok ui s
br iF o n a br i
c er r t e ng oa t
u o
l' t
e A
C O
e n
o pe Drr m) a1 i
ro t t n
i nF Cm t a k
c a
i f
s s u
L g
mta e r (y o
o s co r
r nD e
f lB O B n CW oCcy l t P .
ce r r n fo o f ic a
f g S on ee n au pb eo o A f e u r i ctr i
Ws i
PC it s s a so s r
C e T o oo ev sl q e nn o
t t
S L L L oO t n A LC SF r i
ro I
Pc
~
lL
e e
g n
g r a
s i em e r s
e wa od i
c . o nd pe e t n c ad r r
a Co e e es hn Dc s )
h u
q - v e r r t a rg r
o a s et o o ds t y
e hu t ne ne u o r
_ o tb u n ur i
l acn b u F t i bi Ce i
r q r f r ot n iar r t e t
n _ o t n tS Au n S go l o
odW
_ t ,f e .
u nc gi.n c n ceT d b i r o r vA
_ l .
_ i r otn ey o i Z )s i
c eo co na
._ t (l c _ n l re ph sa t j a (t o l a n A -
_ c r
f om i
mim mt me eg r m e)ta r
e a )s l
pR e
s o g -
o sf o t n
j a sd a orf s )r y e arh awl W l i
c oer e esg st sy g (a P n m l t a e o n (
e a ma bu n i r
o w r uis r n uqooo l u
s os oat l pg l e ot t i rl f a t c u pn i
l f
i t
g e d e s ul i f(aSc arl e n n ber u n s soo eynne t ep i
o i t p us sr s
ke oes f oc ue ng e ed um l
cg f! n at e .
aa su s do n t c tape o y r g ny ui l r
u _
bm l
ep at t e wsac ie c nbmst r
t a n a sn t sn n eoe oCoiaet _
e -
i od Aa Sei ri f
dpu mDCDB F t e ar Clo i
Wlr d emq m t o Sc Oo Lc Ta o Aem ho er C
o- - - -
Tcf e
,lllll lll lllIlLl 1
9 I
information Fow in NUREG "50 e Core Damage Frequency Estimation ,
(Chapter 3) I Definition of Plant y Damage States
'[CoMainrKehEAnhlysI5d?3
{J @W(Chapter $);._uw - (d.
Definition of Containment Failure Bins Source Term Analysis (Chapter 5)
Frequencies of Release Characteristics Plant Damage Po il t es of Containment of Containment States p Failure Bins Failure Bins Offsite Consequence a Analysis W (Chapter 6)
Consequences of l p u Containment Failure Bins Risk Estimation (Chapter 7)
Risk-Reduction Potential (Chapter B)
Plant Specific Results Plant Specific Extrapolation of Results PRA Data Base (Chapter 9) and insights Generic insights y Regulatory Use (Chapter 10)
Key Uncertairities
& Limitations o
u Planned improvements (Chapter 11)
e r
l i
u a
F t
n e
m % ++ %+ = !
- != g.
i n t at cg t
n e r n n a i Da i
t ol e CP N oH l
y yr r r au ES
- 4 I8 * + t n s isU l: y=
y t t i
l i
c eg r
b in a Di ta b ht e o iH r W P
0 0 0 8
0 7
0 0 5
0 0 0 2
0 o 0 9 6 4 3 1 1
0 0 0 0 0 0 0 0 0 l lllll ll
i C
1 O l g )
O o o
o e
8 o i 2 - w '
53 d LL. b
- =
- s ;
c 9 m i G) D
)
E
( o o o a- ,
.E -
- o. e {
%e o Oy O
EC oe o s
< $ 1 00- og E
>,>, E
-u Q oe u u
@3 4-o dg es ,
1 W@ I O
EO O I 00 U
! O O o -
>, 6 O lii
- S
.t j y 49 W
=
.o o <
4< oo- N.
CO O O Qd
.O <o O o o u og Q,,
< IO o 4O l l l I l o
o "e m N co o O o - '-
. seld weg Jo JegwnN l
l l
i j
O nformation Fow in NUREG-750 {
Core Damage Frequency Estimation (Chapter 3)
Definition of Plant ir Damage States Containment Analysis (Chapter 4)
Definition of Contai'n ment Failure Bins a >
Frequencies of C" '
pro il t es Release Characteristics Plant Damage of Containment States of Containment Failure Bins p Failure Bins Q
V Offsite Consequence Analysis (Chapter 6)
Consequences of p 1 r Containment Failure Bins Risk Estimation (Chapter 7)
Risk-Reduction Potential (Chapter 8)
Plant Specific Results Plant Specific Extrapolation of Results PRA Data Base
! (Chapter 9) and insights l
Generic Insigh p Regulatory Use (Chapter 10)
( Key Uncertainties I
f & Limitations u Planned improvements (Chapter 11)
!:FL
ili 4llj a
f h l u
r g eu G e no d r
i r Lh t 2*3 .
n
- a u ot l G
r r h Ne l
i a eg M F i n uo m L r t ht h +
++ +2_j.i j.i t
o n t t il t
o e WeM B m h c
i n +* !
! " ;i
- i 2 Z +++i_. j_i a
t a P e
n h a
o y C p1 , j
=ia " o u
y q l
e r S a
E n
- + t cg + t g: i o
y t e ri n Z
t i
c eg i
Da t
l i
inr e b .
Di ta N oH a ht e * ++ .+ + . ll - g*
b iH y o
r W r r
u P : + { * ;
4 + A ; :: H:=: = 1 __ ! !
S 0 0 0 0 0 0 0 0 0 0 0 0 9 8 7 6 5 4 3 2 1 1
0' 0 0 0 0 0 0 0 0 .
o s
Ay "
ld c s
dteu c s
iC i
m, oS m,F -
Myt ViD F / i s -
DC l
mi n i
/N t asl ra s i
VO C l ea i
bt N s idt a
1
' r O*Fe :
rt cd e -
o e e C )i r FRut o f Fdic ei c c *Oa ci rn i SCe EFF di cei f ef pc uU o V - .l dise e ce- p S e :
F* (1 ia epes Ns c-R I*iec i
h m s- ns i
ct r o
OC p CC at e e nuc i t
rr o F
F+S+ yaqe fl ep ef i c = iPSS .
d s i u
o md ae B a t 7 M-m r s *
- aU P _
o _
I
4 O_
7 i
o +
r g a s n
e c
x
- n i t
m u
d i
i t
e n
c 6 o a s S + pt up n
o ,
t e o oy r s e
smd t
l r ur x G r
k m,ui s ul en,e m ;i m ni na s
ur ueh p
u _
ei - - d niuiht Rat nds! r n 5o i
l t cec e! aa u r Fn + uxict brla G olyl a n 1234567 f o o e i d rP i
X d a
l c
na y R e g
u n O oEr s r k
a 4iod i
rt,u + y c a a R auS po O d
t u
P e1
)
x S Ld n mkc y RAoi C (B oa _
f t
e R Om0 i
3 ClB 4 a S Ce1 r 5 o r H
/T1 -
n x oC eG o t cR cE r
i t
a e AuRoU t
a g
+ R MS N xo+
i 2
S x g5 ~! : !-
'i g5 -
O 1 2 3 4- 6 E E 1
1 E
1 E E E 1 1 1 a8o.E _$ gee.$*
O
9 G E
_ i f
B T
A S T U S O C E N K O I N S S F D A I T U C R N N T A I S D F L N O P L E A U R R R F U C E E E T I T S H S U F C T T U F I A L O C R L
l E R E A S O R O P :
E T A F S Cl R -
E S N T R F L E O N O O E k I A N
O A
R E
U T
A T
A D
L P
E Z
I 9
I F E N Y S T S F U L A N O H E L A T F T G H O R N U I T P T A A B H A T E R Y R S E T T L O A R A X L P A M E A M D I R I N S T E A T S N G L U
E _
E N h -
G I 0 S O D 1 E T L R l I' i A h Y G E E L S Y E T Y T T G R T : C h O A T E A L S A R L T T T F 0
T H S I M k C. G Y T -
i I L h L E -
T S A A; E h -
E N l i l S i F I A U U ~0 -
9 -
i In"ormation =ow in NUREG ~50 0 Core Damage Frequency Estimation (Chapter 3)
Definition of Plant l y Damage States Containment Analysis (Chapter 4)
Definition of Containment Failure Bins Source Term Analysis (Chapter 5)
Frequencies of C '
pro il t es Release Characteristics Plant Damage of Containment States of Containment Fa ture Bins y Failure Bins Offsite Consequence Analysis g
(Chapter 6)
Consequences of p n Containment Failure Bins MWQf%*$RisktEs'ti&ia tioirif?'WT S w .,::V$NMes(Ch'ip t'er7)5$#f##$,94%)
Risk-Reduction Potential (Chapter 8)
Plant Specific Results
- Plant Specific Extrapolation of Results PRA Data Base (Chapter 9) and insights Generic Insights p Regulatory Use (Chapter 10)
Key Uncertainties ,
& Limitations u Planned improvements g
(Chapter 11)
!rL
9
_ qo l
a u
d i
i v
d I
n e d
. g nl f a
r e
__ " _ , E = m g = . =$ = = . e_
au ei l
G rG A
vM y 1 t
i ti on hm l
kt h . co a si -~-- - O ="= = E = 8 =__ at Pe o t
t a iW R B F h a
y y o
l rl _
.- " , nll
,i mm " =- u aa i
q e
Eo _ S h G oo iy t n o g
_Z= Ea5 5= E==E =z i .
Wte i
Z n i
t f a na-oS s
_ yC r
)
H H t
e n
i r - _ = " - "" ==B g m= ___= _
So rD u e m
a p (
N i n
a
- t n
m -
yr )H o o _
- - a=s= 5 ==~m _3. 5~ - .
-=_ urC C C S (D t
c
- l a
o i
D e
r
- G
=
H
- N r_ : =5 : r::
C i : - _ - _ - D 6 7 8 9 0
- - - - 1
_ E E E E -
1 1 1 1 E
_u em mea a*a
.- E ~E 2Bw $22.m5 3 1
n 4
o
. n a
l a
u d
i i
v d
I n
e s d nl f y
ge ai r
l 5 h =a .1 = - au rG eM G t v i
l A0 t
a ton 1
i hmco a kt h - = a -B."-
- at Pe o t
F iW R
si B
t n h a
tela y
o ao _m _" u q
LG S e
h y t
it n
- Wfea _ E'E = . E="m r i Z
o i
t g
n n a oS s
)
H H e
i yC r
t n
r ~- rD e a
p
- e So u
m n
N
( i a
m t n
o yr )H o C
C l a
-3= 5 _==a= - rC u
S (D tce o r G i D
H C
=
_ E: - - :: - _ =5 3 ~ r=:_- r:: - D 5 6 7 8 9 0
- - - - - 1 E E E E E -
1 1 1 1 1 E mj moC #._m >a$
~E, m 2_. it'5gc-:
1 lI l!l l ,Il l
m_
. =&
N O h i
c h
Weo r t s h
aMe i Trt i ol a d f nl 1 t
_=s = E a= ag 5 E_e__=# au a rG ,
enF G _
e g
si asy elt r l
r l
eua RsE hm a eR e _;=
5g 5 5="_8Me= e " at co L e t P o g _
r a B dl e a L h
soG.
o i a
y o
K_ == E g li == E dE u p y q e
ot S Pf r e.
O h
- a. .
i n
o tS
_ - - - E~ ;5n
. _ i Z g i
i n
e Ws-t a
- ) e H H na.. yC oe s l
, - =E_E===E=_
So u
r rD t
n e
ie. N m n rR a
(
i a
t p -
yr )H o n m -" M e m" mg=ii = i
- m- urC C o S (D tc e
C -
.v il i
e r
t a D tao nG - =
e - H T
- : - r3: - E: I
- 5- - 7:: ~ C 5 6 7 8 9 D 4 - - - - -
E E E E 1
E E 1
1 1 1 1
O ,g e >Bmo5ec '- c $'5 0 #.2mo.2'
O-g O O LO p
", W j
'E i l
W C C b
1 (t e
W O m (0
) 'r c~
/ [g_.
C)
O m O
(0
- _u
- C J O w w D O m w c D D o C o w Q D -
w 0] T>ctfr c > C e '
(a UJ O o cc 0"
G o c- Oou e to O Q g fl O >
c :
p& W 0 g $
E w a >s ^ m O_
O X X W O O
C JJ LL nn O
O l . . . .
t
e U
L 0
N M OS YA D S TROE ER E I
E F GT T A_ T I
A ONT S ROI G U L I
T8 7
PI ME U9 D T TMF : 1 E
T N A OA T
Y B S Y T ,
e A E N C S AL R SSRO MEY R E HJ U
G E SEO T ERS C T T R
N SPI A O I S V E N A D R A
- Iillf Ii\c
e S
T E I
T A S E
S E
I L J S P I
U I I T T C TA I L
1 I
L K
A 0 NS F I
I C O C S I
A N A N A N I
F F R O T
P N I
N C I
T I
O O N
O I
T A C I U I S R I
G T E T A
I L E T
RIN A
R E
N O R E
I B E N
T A G AT N E
S N E
B R,G P EE G L
L I
G O R EN I R,G M P E S S
E M R ,T N N TS I
E N R,G O GS OI N S G I
AN SR VI I
N R N GE O
I E NC e S M I
I R NE E R S I EL LC VS I
I AC V E E N RS -
ML R I
P E EE N U N TA EI C PS A U 7 - I PG SL US M H UN -
I -
C L S9 A O L D E
N A1 K N I
S E
K L E
HL Y M I
U Q
A L C I S B B L T U D .
. . L R M A H E RJ A K N H O H C C L
U H C T
N I R
A A P
O I J J M e
t O
E H D T N ]
A S NG F 1 L ]
E I
O)
N I
O . A L A }
T 2 T ) O N N I
L 1 A) 06 N I F I F
1 "
I 1 R4 1 T P .
T T 5E8 I P OR U 4P9 5 N A A RO1 S NO D T S
8 U I S T EP N A F
9 FR E 1 I
T I
U E 4 OEB Y N .
N N R U
H
( C E O O A T O N M NIO E S T S )5 N T T L EN R (
T P NE R O ETV L L L 8
9 I OM G MAO VM AI 1 N K TS)
N U C K Y E L(UN Y E E S S7 T O RA L N )6 L S9 E8 O C B A AS R OM N T VT P 8 I
MS1 A L A S EN P F( OT M 9 A1 " A L B S YYA L AM T N S) D ,YI 4 T R O) C TE P N A A L6 DR P P3 I O R L L8 E A B 1 8 E P LFR F( A I
A O89 OAE I
G P M 19 H R 1 SO M1 - A C S P PSWS R KC FO YL OC FE I
PR NCO I
MP G S R A B E OIT P MT N E LU L E EDN RE I
I LE SMA CI L M T T (T T T(
T NA B L P A R T RD I
AT P
MEE TL D I
MO R I MOFGE I
T E OYUM SC AP D BP BP A EU I
NS( CSN NAS A UE UE R TS I
I FI H SR SR DINS I e e e
- e e
_ e
~
e R A E S N F j N O A A G d M G N 0 O E S I T L C E N N K A V N E C F I
T A I
C A R A L I
B E I P S F F E T T T E
I E G N N C V T R I
C I
T G S E J
I T S A R N N E O L V O R I
C M N N E M
N I T P E O
/
N A J
B E
H I
T Y
T I
N I
A A
L I
T D
E e P T I I T S L T N O N O I
T N E I A P U O O A R TY P I Y S C R R GI E
A T I
D V T U A E L TI T S E I T S O L I
F Y L C O S A N U I
A B E C E E ET S I S F R T TI N E ST F T E A A V NC E R V T T U-O OE E O O I
L I
I R P J R P R I L T I
P SO P P C CL E R O U A A U M
I RP M S MI F F M e e e e e e
- e_
4 e
S E
S U S L D
E C
O -
R P _
D PN AA e S Y I
G O
L O
D O
3 H
f g
p T
O O
E M
a Y
g 4
e e
S S S N E ]
N R U !
l R E S 0 E C S I
C N S Y N O E T O C U E C Y S F Y T S A T E I
S P E F N D A F A A E S A S L P V I
S / L F / S S N O S T E O S O T C U I E .
C E S T R E E J S C N P J O I A U O O R P I /
R P E M S C P D S T E S D E D D U E T E E S T A V V S I
A I T L L C
O I
T I N
O I I
S S R N
l I
- E E E
- C R R N U R N N E
_ N N U U G e e e
- e ll 1l l ll!flll
m e
)
M S R E
(
A U T
B E E Y I R T T G i U U O i B B I
L A I R R E
O i
R T I
l U I Y D U A A O O E B Y F I E ~
GH Y T R C T OT I
U i N V B B l I LE A A T D
I R C OM E Y M R U e
I D R T D O
1 E
O G O A F F O N C Y A R R HI S T S E P TK S E L P L EN I F A E C E T
MAR C S N N
I M
N N
E C O L J I O O O L S N S A R B R O R C P U E C E I P P E P T H Y C - - - -
L A E
A e N
A e
e -
T E N L A L j U P l
[
D L E R Q E A S U H
T U X I
T _
GT OI N N N OA FR O UV DA E LT G E
- I S M ONE N EN G
N TBAS A ELE D E L
E o
I D I P
U SAC LI P OM UG W E D N OS C NI R R UN N M HE E _
I L E E L T CPE T HD L C O XAT OO SI ER _
T E P CN OA S/E E TWU I
N FP GS A MIM SA F S E M N LD I
/
AN T L U P S D ON H I I E NO T AN LKOAN GL E EN DL A T C
T OT I
NNA AM O G NE N US A
T A SE L R
I AC R T MAPU R TG R U G
I UF OXRLAQEUE P E D E LI TEAACEPOR T AD C - - - - - - - - DH N VO A NC I
lllIll. ;L
T e-N _
E _
M U L _
S S 0 _
S E E I S T I
S L A I T
K U S T T I
RAS e C A I E T H S
I T
L R I
B O A N
- B O -
R
. P o-
S E
I S
T I
T L N I E -
T M U E N R S T O I
U E N S I T Q C O A
I S A E R T E E Z R U A H S G l
M G O Z T N N S l T
I S T E M R E A P I
N R I
T U H O I
P O Q C S A R
N O
N E N Y N O -
C T I S
O N O R T T I I
T E T O R C E R
A E E G N A U U A T P U A N O C Q L A S D A T F E A R I I N E R H A I R E) I
) C) -
P C Z C F V PM AM OM l E E E OA QAR RA F N M P C Y R PR O G S G G GG T L N T YG I
N I
E O NO N
S P A E O A L F ZR I I CO _
ZR NR O D C L A T P T P I I EP S
T N I I I
T N E R T R T I I GT N G H V 9 RO A S C R 5 OO OO I
A L E E U
. I RI L I RI L EL MIP Z
I P D T S 0
5 P (P P (P E(
L I
e
- e e e e e e T
U (lll 1 .
Y -
D U D D T E E S DD D T T V L
A E E E TT EE T D L L LE E ED TE M D LE E D T
0 N O PP T MA E M MM ML ET P E TIA O OI D E OO OP L I C CI T MES R CC P N A AI N AA CM M A R I
P A T A A O O R P RA O E P R L
L H
E RR RC PP P A A1 N G C P G P E 3 1 N3 E R Y T 1 R I AN L L L P R L DI DL V T R C E E E VV EE P EO OE E O E D I EE VR E I EVO REL O VE D P DC L L L FI FL F L L F
E 1 C 3 1 1 2 1 9 T T T 1 MR0 OD
.A .
OO .N O.NA OO .NA O A 0 E N N E NN A L N L NN L N R 4 L 1 L TA CITTPTPTTP T L
G I I I I E N S RNNKNKNNKN I I OSHBUI OUUCUCUUCU E E R EE EF EE N E E E E N E A NNIL KNN N N NN N P WAOMSOOMOMOOMO TICT E ATTATATTAT A FR S E TSSDSDSSDS R AR L L RA L L L L D L D L L D LL P RUI HRI I AI LAI L L I AI DHMTPMMHMHMMHM 5 6 8 9 1 3 5 6 7 7 7 7 7 8 8 8 8 8 9 9 9 9 9 9 9 9 9 1 1 1 1 1 1 1 1 1 e * * * *
- e *
- e
- f l l)ll
8 8 ) )
1 2 9 0 0 1 )
1 .
- 0 V V j S
V R E E d S
P E ( (R 0 2
R 7
(
8 7
8 8 O
9 91 N s8 1 S A 9 R Y T T R 1 E E I N P T B R A N
E I
S G U MU T T U M N I
E N G E R E E V N I
TU C B 3 T M I
I D E T T C S X
E V S A A L F
3 T 1 E Y T
- L .
N O .
L C S E I M OON A O I
L Y R S L N O E ET P T T C U H TI N KN I s P I L
T T A T O U F A U CU D E R N P F O P E E P E S N N D -
N UN L L OMO I
L C A O DT E AT A C O L L DS S
I T E A G E U L AI D L L D E Y L DI A P R T M
E MH M R E E H - - - G V F O C P E A C S U S S e
- e e e e
- Ill il
e U
P L 0 _
I L T H
I TI UW TS SE e AC EN HE I T R R E OP NX E
e
D E
T A j U
L A S d H V E U
0 E
G ,D S ES U DE UI S ES ON FOP I S T S N RO I T O SI EK E I
HT N R S TA E P E
U UN M D S SA E S DT I S S ST C I
A E EN SN RO SI EM A E L C AE GP S
Y NE OI T U E O L
A I
M T A S S S T L RR N AE U C BI EI S Y KTR OS A
- TL RI F I
UE S AS E K BP TD S S G AE E
T
.G L R NT I T E B TO ER I
R S
I C E ON O O A SMAF U SG T P NLVL I
L SUWSN E N KT SSI HA SA T O OI DCH P I I 5 D D A E YD SLO GS I I I
RR N CR O K - E E R YAO T P U M G P CCMZ I
I EC T L NL S NO E DY Dl E l
D AAF N AR N ER EH CP NM NA MFARLEL N KL P O A SA I
I T TD M A U AO L NI N FA S GRI I
I D S L F R E RR N D LL NGD DS I
GN AM -
H EE I
I OEO AOMEI L TRT N SA MLDS HLSDO OFI I
e e e e L l
e
/
N O
S C E PI I
T b D A D L U E E S O S T E 0
N S T E N F I
I R U E K A D E S M C E H S U A T T T I R B B AO 1 T T .
OP L E F OE
.N S N
L A
I L
T S I
F OH T TD E R T
H H I
N N V G N GI UAE L C E
I G O L E E L o
S U I T NI NGR E E NO I
U D L
OR T T U A
/
E L R L
O E SP I A H S V L T W S T E L L N L S N R O I I S ME E M O D E :
MS O I
E RS S N SE T T S T I D I
A E L E A V A D R U U L T P N R S N D G S S DR A E S E R
M A OWP CBST O T I X
N -
I E - - -
e e
~
e
T N
E S M j E E R G d E U F N S O E A 0 N S C N I
N A A A C O M M I I F F T R E T I A E C N C U T E E L N R I
P I U C S A Y O I
- V S S F F T E T I
E T E N D L I R I
F T A N T C S
L AS U R E P T /
N O Y N M
F WN R D E O EE B N I
VM O N I
W E E T A D E R R A Y E
I I L V CU U T V E Q I
O R I
F G L E E I T
R D I
CR R P E U E E M
I T P"C V D A SI I E Y R - R T V T G TE C O E NN E R F E T AE F P A N LG F M S I P" E I
- e e e
^
llllll l
N E L E R L E E A H
W W F T O j E E B C N d O
N AG I T
0 O
I F
RI N U T EG L A TR O E C NE S I I V E C NS UE YN R G A RO N MT L I
I F OC A V R ML I T E C O E OI T A R I R
C U L N P T UA H M N D T GS C I I E S E E E D
- C C N E A RI T T N AH AIR O R A -
R HT N SO F G _
/
NR DI R N E O R P U M I O
U RN A G N OD W C R OS R
N F N ON F E O KA T D U S R F T N A S I OF N A S S M WA I
E E A E T MY D Y T R M S VLE T VE I
I G
AC OI OF O RR T RA R F N MU PS P e e e e llll! '
4 S '
N SO '
EI I
S TU I
L L T C I
U N O
TC S _
AD EN
A S
I '_
I
T S S E A j I
T E S I H d L T S I R 0 E T C U O O T N R S R P A O S P E F E A H G
I S T T
I I
R N I
V E O N I H N N T T F A S L
C F O P T A O T E A N
- N N CL E O E M RP R I T MA U R U A P R O A T Z O G L O S E U
l l E R E L F A V P RC _
N E DU O E DA E N T S T S ER R I
TH RI E L S - U V TI T N N I I F L I
N U I
e e e
D N E A U d N
P X
I T
U 0
E N _
D S OP N T C A AI N CS I E U R F U N N R O S S I
T E A TN N N L SO O E TI OC U A I
S CU Q T U ON E N
- L T R R E C S U SM E O EE N T F T P I I L
O I L L L
I L
M C I T A I
TI U
O UR T T S TO S F AM AT E A E R HR HO TG TP RO RP OR OU NP NS e e
i
,. 1 -- pa af c
- oq'o
~
"g UNITED STATES 8' n NUCLEAR REGULATORY COMMISSION 5 .$# ' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS O WASHINGTON, D. C. 20555 C % , ,',#,.# July 8, 1987 MEMORANDUM T0: D. Ward, Chairman Integrated Safety Assessment Program Subcommittee FROM: Richard Savio, Senior Staff Engineer c~I c-b /( m. b
SUBJECT:
ACRS MEETING 0F THE INTEGRATED SAFETY ASSESSMENT PROGRAM SUBCOMMITTEE, JULY 7, 1987, WASHINGTON, D.C.
A Working Copy / Summary of the subject meeting minutes is attached. I would appreciate your review and corrections as soon as possible.
O Copies are being sent to ACRS Members and Consultant who attended the meeting for information or review.
Attachment:
As Stated cc: ACRS Members P. Davis R. F. Fraley M. W. Libarkin G. R. Quittschreiber R. K. Major T. G. McCreless O
1
)
WORKING COPY
.O DATE ISSUED: 7/8/87 ACRS SUBCOMMITTEE MEETING MINUTES OF THE.
INTEGRATEDSAFETYASSESSMENTPROGRAM(ISAP)
JULY 7, 1987 WASHINGTON, D.C.
The ACRS Subcommittee on ISAP met on July 7, 1987 at 1717 H Street, N.W., Washington, D.C. The purpose of this meeting was to review the effectiveness of the ISAP process and to use the Millstone 1 and Haddam Neck ISAPs to understand how the process functions. ACRS action is scheduled for the July 9-11, 1987 ACRS meeting. The objective of these discussions will be to develop at least interim coments on the ISAP process and the usefulness of extending this process to other plants.
The Subcommittee met with representatives of the NRC Staff and represen-tatives of Northeast Nuclear Energy Company (NNEC0). The Subcommittee O discuss 4ons se9en et 9:00 e.m. end were conciuded et enout S:SO p.m.
All of the discussions were held in open session.
The principle attendees at this meeting were:
ACRS D. Ward, Subcommittee Chairman J. Ebersole, Member C. P. Siess, Member P. Davis, Consultant ,
R. Savio, ACRS Staff NRC C. Thomas, NRR/ISAP M. Boyle, NRR/ISAP A. Wang, NRR/ISAP B. Alefi, SAIC D. Gallagher, SAIC O
ISAP Meeting Minutes July 7, 1987 O
NNECO E. Mroczka R. Facfora R. Kacich J. Quinn P. Blasidli J. Beckel M. Lederman Highlights
- 1. The NRC Staff discussed the history of the development of the ISAP program. The ISAP concept evolved from the Systematic Evaluation Program (SEP) and the Interim Reliability Evaluation Program. The SEP was started in 1977 to review operating plants against regu-latory requirements that had evolved since the particular reactor had been licensed. An initial program involving 10 plants and comparing these plants against the SRP was completed. The notation of IREP was developed from the TMI Action Plan and was intended to involve the performance of a relatively simple PRA which was to be used to identify risk outliers. The experience with these two programs indicated the resolution safety issues would be more effectively achieved if they were addressed in an integrated, risk-based, and plant-specific review. The ISAP program was initiated in late 1984 as a ns.5. hod for integrating ongoing plant safety improvements and licensing actions. The intent was t develop a procedure for combining, scheduling, and doing a risk-based evaluation of changes in plant procedures and equipment.
A plant-specific PRA and a review of plant operating experiment were to be part of the ISAP. The ISAP procedures involved an initial screening to sort out those topics which could be reasonably included in an ISAP-type procedure. The sources of topics for the ISAP were general licensing actions, USI/ Generic Issues, licensee-initiated plant improvements projects and topics which arose from a review of the plant-specific
ISAP Meeting Minutes July 7, 1987 PRA and a general review of plant operating experience. The November 15, 1984 Commission Policy Statement on ISAP which de-scribes this procedure is included as Attachment A. A pilot program involving NNEC0's, Millstone 1 and Haddam Neck plants.
NNEC0 has completed the ISAP evaluation for these two plants. The NRC Staff has issued their draft evaluation (ISAR) on Millstone 1 and expects to issue their draft evaluation for Haddam Neck in August 1987. A list of the ISAP pilot program milestone is given on pages 1-3 of Attachment B. The NNECO and NRC work rescheduled in the of discovery new safety issues (derived from the PRA in-sights) and a ranking of all the safety issues. The NNECO and NRC rankings are displayed on pages 4-7 of Attachment B.
- 2. The NRC Staff is preparing a SECY paper describing the lessons-learned in the pilot ISAP and giving recommendations for the future use of ISAP. A draft paper has been prepared and is currently undergoing management review. It is very likely that the NRC Staff will recommend that ISAP be made available in the future to Licens ees who wish to participate.
- 3. There was some discussion as to how the Millstone 1 and Haddam Neck ISAPs could be compared to the use of the Integrated Living Sched-ule (ILS), and the use of the SEP process. The SEP-type comparison of operating plants is included in the ISAP process (Millstone 1 and Haddam Neck had already been through the SEP process). The ILS allows a prioritization of safety issues but not the combining of, or dropping of safety issues.
- 4. The Millstone 1 and Haddam Neck ISAPs were Level I PRAs with a limited treatment of external events and without any consideration q of seismic risk. There was discussion of the adequacy of this b approach. It was suggested that the consideration of seismic
l 4
15AP Meeting Minutes July 7, 1987 events, a better treatment of external events, and upgrading the j PRA's to a Level 3 (i.e., consideration of both core melt frequency J and containment performance) were necessary for a complete evaluation of the ISAP issues. l
- 5. NNECO discussed their experience with the ISAP process. NNEC0 plans to extend the program to Millstone 3 and Millstone 2. They believe that the process provides an effective method for resolving safety issues in a cost-effective fashion.
- 6. NNECO discussed the scope, limitations, and principle insights of the Millstone 1 PRA. The analysis was a Level 1 PRA (i.e., ,
consideration of core melt) with plant specific data being used for component reliability and transient frequency to the extent that such information is available. Fires and internal floods were included in the models. The estimated core melt frequency in the original estimates (1985) was 8x10-4/ year (meanvalue). With the implementation of some plant modifications, the current core melt frequency is estimated to be 5x10~4/ year (mean value). Loss of long term decay heat removal and loss of AC power were the dominant contributors (64% and 12%, respectively, in the 1985 estimates and ;
42% and 19% in the 1987 estimates).
- 7. NNEC0 ranking of safety issues was done using their Analytical of the impact on Ranking Methodology (ARM). This procedure involves a consideration of public safety, personnel safety, economics, and personnel productivity.
- 8. Program costs for the ISAP were in NNEC0's opinion reasonable. The PRAs(whichareusedforpurposesotherthanISAP) cost $580Kfor Millstone 1 and $1.4M for Haddam Neck. ISAP costs were about $500K
, per plant. NNECO believes that the ISAP process is an efficient method for interacting with the NRC.
ISAP Meeting Minutes July 7, 1987 O
- 9. There was some discussion of the scope of-the PRA and possible sources of modeling inadequacies. The issues discussed were:
- complex systems interactions where, were not induced in the models.
- modeling of human error.
- the performance of fire dampers and the potential for the propagation of high temperature environments from area to area via the HVAC ducting.
- valve performance under accident loads and accident con-ditions.
O
- performance of the SDV system in Millstone 1.
- 10. The use of PRA in NNECO operations was discussed. PRA is used to facilitate the resolution of safety issues and as input into ongoing design activities. PRA is also being used to prioritize QA/QC inspections. PRA insights into the effectiveness of the NRC QA/QC requirements (Appendix B) was discussed. The limited work done by NNEC0 has not shown safety performance differences between Appendix B and non-Appendix B hardware.
NOTE: Additional meeting details can be obtained from a transcript of this meeting available in the NRC Public Document Room, 1717 H Street, N.W., Washington, D.C., or can be purchased from ACE-Federal Reporters, 444 North Capitol Street, Wash-ington, D.C. 20001,(202)347-3700.
O
/~f' f f ~ W
. 45112 Ted:ral R;gister / Vol. 49. No. 222 / Tnursday. Novernber 15, 1964 / Rul:s and Reguhtions NUCt. EAR REGULAYORY !
COMWSSION i 10 CFM Part 50 Commiss3on Policy Statement on the ,
Systematic Safety Evolastion of I Opefuting Nuclear Power Reactors l acewev: Nuclear ReTaletory l Commission.
Acroc Notice of Commission Poljey Statement.
suenanasry: nts Policy Stateinent describes a pilot program for which the Commission has developed the regulatory poEcies and practices to conduct integrated assessments for aparating aucisar power isactors. This i program is called tha lategrsted Safety l Assessment Program (ISAP) and will t i
addreas signiEcant regulatory requirements which have evolved since the plant was originally limaneed sed pending hcans&ng actions which have ,
evolved from a vanary r.f aher sources. l An integrated assessment will be conducted on a plaal-s part of a trial program,pecdc basis, to evaluate au as,
.licensms issoes on a given facility and to establieb schedules for any necessary plant i...y.m..la.ts. In addition.
procedures have been established to g) a!!aw for a periodic updatint of the nroulting implementation schedules for new licensing issues that arise in the future.
POct PuertWest cospodessaTWhee COedTAcTt Dennis M.CrutchSeld. Assistant Director for Safety Assesement. Division of Licensing. U.S. Nuc!est Regula tory i Commissien. Washington.12.C. 2D555.
Telephone (301) 492-7422.
super.seeeertany swooms atum: In 1sn.
the Nuclear Reguls tery Commission initiated the Systemene Ers!catien Progrem (SEF;. Pheee ! of SEP deficed a specfic set of safety treues tiepics) to be renewed for operenrrgmrcient power plants. Phase II of SEP wesi pr"let revrew of these tepics for eleven of the oldest domestic opeTstir:g netters.
Results have evehd f em SEP crer the last two yeers and idennSed sp.rEcent experience relative to the safety and eveinstien technigdes fer operattrrs plants.
In taso. Congress enaclad Pub. L $6-255 (the NRC Authorization Bi!1 for Fiscal Year 1990k Section 120 el Pub. L 96-295 required that the NRC develop a program lor the syuemstic asfety enlumtiens of operst::E resc: rs. De
( ) program proposal would have extended SEP to en ceulostice which required licensees to cec: pere the.r plar:t design to the acceptance crueriale ti.c
, A-1 6
a .. . . .. . . _ _ . ._
Federal Regi: tee / Vol. 49. No. 222 / Thursday. November 15. 'l984 / Rules and Regulations 45113 ndard Review Plan (NUREG-os00).* Historica!!y. licensing issues have ne only exceptions ml! be issues for p t pregam was not implemented fo? been evaluated generically. and which de NRC Staff explicitly
\ . . rating reactors; the Commission guidelines for any necessary cor ectve determines that prompt acton is determmed, and the Congress speed, actions have been applied urufer . to recited to protect the heald and safety that the scope of the program was too su plants. While this approach has cf de public. Such actor.s include the broad to efficiently evaluate the safety provided an effective means to ensure short. term response to buUetias issued of operating reactors. Congress revolution of these !ssues the generic bythe OfSce of h.specti n and subsequently specified in later implementation has not given sufficient E.nforcecent.
Authorization Bills that funds should not attention to plant specific Supe of Evalushn be spent to icplement that program. 'claractaristca which have a direct However, those activities were useful in bearing on the appropriateness cf the ne scope efIEAP is intended to be as that they foessed attention on the needs corrective action and the relative comprehenske as practcal.
importance of the inue in relation to an Consequently,it wdl consist of end difficulties associated with the deterministic. probabuistic, and systematic safety evaluaton of overst! plan for any necemary plant operating reactors as they relate to a improvements. b some cases, operating exp" ence evaluations, which consideration of plant. specific , will serve to identdy specSc issues to constantly changing technology and charseteristics have identi5ed be addressed in an integra:ed increasing scope of regulatory alternauve corrective actions which assesa=en t.
requirements.
Provide an equivalant or greater ne deter =inistic review areas. or Following the T U-g accident. the LSAP topics. wdl be derived on a plant-NRC developed the TM1 Action Plan meseurs of safety, often at less cost to the licensea. e/ecdc basas dur.ng a scnening review (NUREC-0660)* from the safety lessons learned.Two aspects of the DG Acton Cocaequantly, the NRC had deve*oped md the Ucesee at de beginning of de the regulatory procedures and attecdant prog s=. ne issues to be cens2dered Plan are particularly significant to the an p) a set of SEP Tepics fer which de avsluation of the safety of cpersting policies to conduct intersted NRC Staf hes found signi5 cant plants:(1)it identiSed a large number of aneuments for operating power 6Humces been current hcensing correctve actions to be implemented by reactors.This approach is called de csteria and typical design entena In operating plants and (2)it initiated the blegrated Safety Assessmect Program existence when operstng plants were Interim Reliability Evaluation Program (ISAP). b order to ecsure the Lcensed W a3 pening bcens=g (IREP)in which plant. spec $c effectvecess of this program. It will be sedoca for de plant. beludsg =uM-probabilistic nsk assessment (FRA) started on a trial basis and the plants to be reviewed have been se'lected by the plant actons30 Acton Man dies were to be performed by the requan=ents and plant.epecSc O ff for severn! operating reactors to NRC Staff from these licensees who v .pplement the risk reliability in6cated an intenst to voluntarJy hcecepg adeu W de unnsohd geene issues fu which nscluuen on a experience Imm the Reactor Safety participate in such a program.
E' P Study (WASH-1600). % licensing Based on the results of dis tnal d Jp ti "r isp d by actions resulting from nn have program, the NRC wn! decide, in about a the licensee increased the scope of outstanding year, whether or how this program be extended to ede opuntng g Q~,g.,g gpg published en March 14.1983. (48 FR la y, expen a tu at m a o 3. 10- 2),inc! cates that the quantitative 1 REP indicates that there art plant. goals and design objectves ml! not be specific strengda and woimses from Implementation Schedu! e iability point of view, t:Mt warrant To previde a stable environ =ent to e u ti n p o wb.e e e
" n. beyend the cenduct LSAP. the Commission bes interpreted as regir.::g dat Ucensees or detum s cahy based tuues. authonzed the staff to suspend specific applicants pe-ferm a probabdistic One of the most signi5 cant existing !=pletcentation scheddle scalys:s; however, the Com=:ssion j conclusions drawn from SEP and IREP is requirements for the plants to be conteLes to believe that pecbabiliste that issues related to safety of operating reviewed. Each affected Ucensee mU be analyses provide a valuable adjunct to nuclear power plants can be more exp,eted to propose and justfy deferal the determinisce regulatory effectively and efficiently implemented for spec 1Sc implementation require =ents sed enhance e .gineering in an interated, plant specific review. requirements that warrant funier jud;=ents,if dey are proper!y In addition. the expenence from SEP has evaluation.The associated performed and appbed. Cor.sequently, syrved to focus on the set of current implementation requirements ted other the Com=ission beheves that a plant-heensmg enteria which should be ufety issues will be evaluated specfic prebsodistic safety essess=ent evaluated for opersting plants and collectively in an lategrated anessmeet. (PSA) should be perfor=ed in espenence from IREP has served to ne staHis only authorized to defer conjuncuen mth ! SAP.ne plant.
defin the methods to conduct a plant
- substantive regulatory and other speciSc PSA mil provide a basis for s ici e probabilistic safety analysis so requirements to the extent a!! awed by cost /beceSt evaluations for the t at onsistent. comparable results de Comnduion's procedural deterministcaUy. based issues and will cou e obtamed which would enhanc* regdationa.nca. the staff wiU use de dso idendy potental strer. gds and an tntegrated plant safety anessment- provisions in to CFR 50.12 to grant any wednesses in d plant desy:n and
'"*E N"*' cperstien which s ou!d be censidered in Corw mr tw pwh.d by a::n tznan. tmple =enta ten an interated es. ssment. l 6 adition any new .
O*J = bmers i. % Negauon,5,me,.
Dms.ea of Temat Leon.uuen ud requinments which evolve late in de An cperatng expenen:e essaation wdl be cend.c:ed in paraUei wid de pm comt trs weu, a,*wy course of or followtng ! SAP win be tepic evaluaticas and plant specific s deferred for the plant; nvolved sed EEC,.YyQQQ",Q".*d Dwaeat of c cun,= 5:an pwt mani n d Incorporated in an implementacan PSA.This eva!uation mll be used to ' j j
5e-asMa. vA ann. schedule updata, se desenbed below. Identtfy issues related to sig .Jicant A-2 /
(o
1 i
I c!!; Tedsral Ee,.ister / Vol. 49. No. U.2 / Tharsday. November IS 1984 / Ruhs and Ragnahtjons 1
trends. t vent precursors. plast appropdatrnegs of cc=ec.*N. e ac11ons l
- ar.agemer.! and cperstor. and proposed by the Lcanaev for each of the
. maintenance practcek 13 aditoc. the identLad Isauta [s a dra.ft report. The h
c;. erat.as exper.emca evakaOc:
- 13 draft rer .: wiU be issund for puhuc cen =est .ad peer review.Should the j pr:mde a davene perspectee fer the '
- teraled asusa=ect.The evakatic: NRC Sia!! cd Lcansas dis.apee on the m2 cc-sast of as analp.s ar.d corrective acten for smy issue, that categer-:30cn of repo table evecls a d m.aner wiU be resolved is occarda ce forced plast shutdowns e.ed an e 6 Co==isaica's precadures for eval a L co of oversllIm'** bd ta.
P' * *^ Fouovens :esolutien of any cc==ents Evalus don W on the d:sft report, the NRC Staff will The 15AP Tore eyelusbons and sequest 6st the licensee establish and pla:t. specific PSA wdl be evnducted in lutify i=plecientation scheddes for each of the corteedve actions and any par Del The licensee wtD unceUy per'er deter ::::see enalven (cr the ongoing amelyses that =ay be necevaary plant speede set of ISAP Teptes by to estabi.sh appropriate cerreenve compar=g the as-bui!t deQn cf de accons. The NRC Staff wiU bdge the facihty to the ertret Ocewng crite -a. adequacy of the preposed icduary codes and sesaderda, ci other t=p},=eeration schedules besed on the apprepnate acceptanca emena and also technical evakation of the issues prere.s nai p,w~mu for esch issue preeented in the d ah report sud issue based cs t.t:e PSA.The etaff w,U review g,,4.g,,g,7 ,3 % 1 ggst ,, pert, the iicresee s analvoce and inue safety ev alcate: rep = ts'=*ich ide-tfy Liceca60s Acnoe and Schedde Updates
'E O d *"'"*** h
- b **;#7"** ne fi=aj report wtu seres as the basis entens and ary anecdant safare inues a=d doc ==entsten fer a license M hdd be c.%.cMMb a=asd=ect t=cntperating and intepa3ed aseess=c=t. Schedules for the for=ali=i=a de i=ple=uitatics licecsee's asalpes a:d ataIf rvaluacen wt!! be estab!; shad d=. a the sc recing scheddes. The license a.=end=ent wiU also estabfish procedures to periodica!!y l l review to echarca an e25cest use cf resou re s.
updata the i=ple=ectatics schedule.
. A PSA wedd be cce.d::teci by the Acy cew i=ple=entatics licensee in ac:ordarca with as NRC require =ents that.arise dunng or Procedures Guide (NUREC/ cps-:s:4 foUowing an LSAP review wdl be The Procedures Guide desc=bes deferred. enept far these issues for apprepnate and c=tsistect gnethed.s for which1ha NRC Staff detennices that (1)it: tater defin:trert (2) e pplicatics of prompt acti:D is regtdred to ensure the data. (3) succeseMadure cntena. (4) bealth and safety or ce==cn defense a:alys.s (:) quahty centr:1. and (6) and sect.rity of the public.
docu=e=tatco and preeectaton of ne defured i=ple=entade resulta. Is adit =n. the NRC Staff will require =ects will be evaluated ide ufy methoda by wh2ch ths PSA coDectvely as part of an should adhese unreeolved ge=em issues; i.e. aalery issues for wh:ca i=pte=ertatics oc! edde updata. The accepta:ce enten.a do sos yet exat. update evelcations wtndd be cenducted Dunes the acreer.ca review, culesteces Periodically, but not =cre than at Eve-wiu be estabiisbed to c>onitor the yearbtriesla.The evalcation would p=ptu cf the PSA and to ensure follow the same genersj course as an apprepnate isteracten between the. 1 SAP review and would c=nsider a NRC Sta!f and the lica.caee. The licer.>ee revised PSA. which has been epda ted to will be expected to use the PSA to reflect ent ectve actens and plant idectfy sp*e coct .bu::n to risk Improve =entsas they are coepleted.
that shouh! be spec 5:aUy considered in The revtsed i=ple=entation schedule the intepated eesess=est. For the t .a! will similarly be incorporated and prepa=. the extent and sture of plant for=al =ed tw a new license specific probabdisce a alyses wiU b' amendment.'
established oc a case-by. case hasis.
The issues rused in ite ISAP Tcpic Dated at Washiry:m. Dr. thre 9th dey of f evaluatie s and the PSA. a:d the November tm.
crerat=g experier.ca evaluatics wiU he j ) censidered ce!!ec vely is as intecated For the E: lear Ret 'dio y C:==.wien.
assess =erL Cec;.sacts cn correct.ve Sam W 1. Cl.A aetic:s wiU be based on q.uC.itat:ve see,,wy af p, cyn,c .u.a essess=ents of the vaLa and i= pact of each acuen.The NRC 5taff wiu present gru o.m.n tw w w its concluions regarding the need Ier or = - caos ==-ei-ao A-3 7
pWW b
-.. l O
v ISAP PILOT PROGPAM MILESTONES March 23, 1984 SECY 84-133 described a four plant ISAP pilot program in lieu of SEP Phase III and the National Reliability Evaluation Program.
November 15, 1984 ISAP policy statement was published in the Federal Register (49 FR 45112).
May 6, 1985 SECY 85-160 described a revised ISAP pilot program for two plants: Millstone Unit 1 and Haddam Neck.
May 17, 1985 Northeast Utilities (NU) submitted a proposed set of " topics" for both plants; the proposed scope was a revision and update to two previous proposals (June and December 1983).
June 1985 Meetings held with NU to discuss scope of topics, pending licensing actions, generic issues, and plant improvements.
July 10, 1985 Millstone 1ProbabilisticSafetyAnalysis(PSA) submitted; imediate corrective action (long-term cooling) and topic-related issues were defined in the forwarding letter.
July 31, 1985 Staff evaluation identified 80 topics for Millstone 1 and 70 topics for Haddam Neck, as well as projects that were to be completed independent of ISAP. Documentation requirements were defined for deterministic, probabilistic and plant improvement summaries.
August 13, 1985 to NU submitted individual topic safety analyses for Millstone 1 February 4, 1986 together with applicable topic probabilistic sumaries.
August 19, 1985 NRC issues draft Millstone 1 operating experience report for comment.
September 23, 1985 Draft SAIC PSA evaluation report for Millstone 1 issued to NU for factual corrections and coments.
October 3, 1985 NU submitted comments on ISAP operating experience review analysis. l October 21, 1985 NU submitted coments on draft SAIC Millstone 1 PSA evaluation report.
November 20, 1985 Commission Memorandum and Order issued granting a schedular extension to 10 CFR 50.49 for the environmental qualification of eleven valve operators to be resolved under ISAP for Millstone 1 but not later than August 30, 1987.
l 4
l l
O January 3, 1986 Final SAIC PSA evaluation report issued to NU for use in the Millstone 1 integrated assessment.
February 14, 1986 NU submitted a proposed ISAP schedule for Haddam Neck.
February 19, 1986 Comission briefing on ISAP status.
March 3, 1986 Safety evaluation for Millstone 1 issued which describes the results of the topic reviews: (1)specificissuestobe addressed in the integrated assessment, (2) resolved topics, and (3) three new topics and the related issues resulting from the PSA and operating experience review.
March 26, 1986 NU submitted the supplemental fire analysis for the Millstone 1 PSA.
March 31, 1986 Haddam Neck PSA submitted by NU; imediate corrective actions (small break LOCA) and topic related issues were defined in the forwarding letter. i April 22, 1986 SECY 86-121 presented first annual ISAP progress report.
June 13, 1986 to NU submitted individual topic safety analyses for Haddam Neck November 18, 1986 together with applicable topic probabilistic sumaries.
July 3, 1986 Draft Haddam Neck operating experience report issued to NU i for coment.
July 31, 1986 NU submitted their proposed resolution of the Millstone 1 integrated assessment issues together with a priority ranking for each issue.
August 19, 1986 NU submitted coments on draft Haddam Neck operating
. experience report.
September 30, 1986 Staff issued final Haddam Neck operating experience report.
November 16, 1986 Draft SAIC PSA evaluation report for Haddam Neck issued to NU for factual corrections and coment.
December 12, 1986 NU submitted their proposed resolution of the Haddam Neck integrated assessment issues together with a priority ranking for each issue.
January 14, 1987 NU submits coments on draft SAIC Haddam Neck PSA evaluation report.
April 14, 1987 Staff issued draft Millstone 1 Integrated Safety Assessment h Report (NUREG-1184) for coment.
Final SAIC PSA evaluation report issued to NU for use in the May 27, 1987 Haddam Neck integrated assessment.
k O May 26, 1987 Coments on Millstone 1 draft ISAP received from peer review group. Comments from NU due at end of month.
May 1987 Staff issues second ISAP progress report and recommendations for future ISAP Actions.
- July 1987 Staff to present ISAP experience and recommendations to ACRS; ACRS to issue letter report.
- July 14, 1987 Staff to issue draft Haddam Neck ISAR to NU and peer review group.
- July 30, 1987 Staff to issue final Millstone 1 ISAR and start process for Millstone I license amendment to incorporate integrated schedule plan.
- August 14, 1987 Coments on Haddam Neck draft ISAR due.
- September 1987 Staff to present Haddam Neck ISAP to ACRS.
- 0ctober 1987 Staff to issue final Haddam Neck ISAR and start process for Haddam Neck license amendment to incorporate integrated schedule plan.
O
- Current completion schedule.
O )
+
' O k 8 8 5 C n - 4 0 8 1 3 2 534 7 4 7 9 6 2 9 05 3 0 4 E a - 1 1 2 1 1 2 3 2 2 1 1 1 1 2 M R -
O N
+
+
e e e e s e e e s e e e e e e e e e e s e s e n u v C n R a N R -
t ig ig g ig g ig g ig gg ig ig ig ig gdgi h h h h h h h h h h h h h h h h u H H H H Hi H Hi H H H H H H H Hii dPd e i ii d d ic i vvui iv i ni iududvdvdiv dn dn d i i i e de d ed e e e e e e e e e e e e e e e e e e e nd i i i i i i i i nd vd nd ud ud i di di
+
M M M M M M M M M M M M M M M M M M M M M M M
+ +
e - 6 6 2 58 0 2 1 3 3 6 5 3 6 9 1 6 M r - 6 1 5 1 0 1 5 06 5 0 0 01 5 9 0 R o -
A t - 7 1 2 1 0 1 4 80 2 0 0 0 1 0 3 0 S -
+
+
d e 0 0 00 o t e a u 2 o3 7 6 8 2 3 4 2 0 0 o o o o 2 4 8 3 8 9 C7 C C C CC C C 0 r
g a l 9 t 4 1 0 5 1 3 0 9 6 7 t t t t 6 8 1 4 3 8 N 9 N N N N N N N 7 8 0 0 5 6 3 4 8 4 8 2 5 0 3 - 8 eV - 3 4 2 1 9 2 23 1 4 4 4 4 2 t - - - - -
n -
I +
+
l e
n .
nd 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 o o 0 5 0 0 0 0 0 00 0 0 0 0 0 5 0 0 0 0 s r 1 2 r P 0 0 1 1 0 0 2 1 0 2 0 0 0 0 0 0 2 e - - - - - - -
P
+
D C
E +
Wlen y n t 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 0 0 o e 0 0 2 0 0 0 0 00 0 0 6 0 000 0 0 0 sf r a 0 0 0 0 0 0 3 00 0 0 2 0 0 03 0 0 0 eS -
P -
+
e+
t t i n e a c m 0 0 0 0 0 0 0 1 0 0 0 0 0 0 0 0 0 0 0 n r 0 0 0 0 0 0 000 05 1 0 0 00 0 0 5 o o cf 0 0 0 0 0 0 1 00 0 0 0 0 2 0 0 0 0 0 E r 1 -
O e P +
+ ,
t y i t 0 5 0 0 5 0 0 00 5 0 0 0 0 50 0 5 0 l e 0 2 6 2 1 0 4 20 7 0 5 5 02 0 0 2 0 b f u a - 0 52 1 0 0 3 60 3 1 0 1 0 5 0 1 0 1 P S - -
1 1 -
+ +
1 2 22 2
+
n e 59 6 6 6 7 0 5 60 1 1 1 1 1 1 1 1 1 7 7 78 2 2 2 2 23 29 99 991 1 1 1 1 1 1 1 2222 4 5 5 55 1 23 4 6 6 8 9 1 3 4 5 1 0 1 2 5 R i A t 3 3 3 3 3 3 3 3 33 1 1 1 1 4 3 3 3 3 333 3 3 3333 33 333 333 3 3 S t !.
I e - 3 3 3 3 3 3 33 33 4 4 4 4 4 3 3 3 3 3 3 3 3 3 33333 3 333 3 33 3 3 3 S
+ 3 3 33 3
+
1 P c 5 1 7 9 234 8 21 A i S p 5 9 6 6 6 7 8 3 4 8 2 8 9 9 9 9 1 2 3 4 6 6 8 9 1 3 4 5 1 0 1 2 3 I o 0 0 1 1 1 1 1 2 22 0 0 0 0 0 01 1 1 1 1 1 1 1 222 24 55 5 0 T
+
1 l 1 1 1 1 2 2 27 1 1 1 1 1 1 1 1 1 1 1 1 l 1 1 1 1 I I 1 1 1 2
+ .
k n -
s o a i t
t
. i n v s s s o r o e t n i p P r f i
t o
a t e al r
o r l e
v n o
l i
u a
i D
r c s pi a wF t i
i e s A V t o p n
f i l v na g n t a l f u r o i
d a r n i o n sk e e p o V T e i e s e ck s t M / h e t r e p s r a a n e w
d st t s
e a o u e u n s i s d uB M eS l et a r l t r e y p /n de /t n a o t t e y u oi t P o y s
- cl a a r nS t s n s i l cb u o r e _
- i F n r R B o l aI o y n et a e o t C u s r e e e n n i y e r .M t I s v a n r r s r l e t y et p ed y o o t a v e i ni t A u P t a l
c a e rp t sO g n s y yS r r a a s u et d ni et i t i
t c
cl e t pt nl l e pL u e n ei eiD l t s l e o gb vt o rl i
l a
d e
c e eS iniae ace g o nt M
aF l p
E S S o o e eS t n e oi rF T C o a e ad d r o at r O
nt t ct t u gj rD e t nl yd i rt L c o e u r t r e e vR _
sid n o oS i s g n F n n P iR M t P l a s rl _
rl e el M v y nt a r a ol r yb rf c o p n o a r e el a rSi uF gt i I e eW r e aH a eil e u s n rV o et v e _
l nf o yt H aHf o e l
o oS r ot l o yi n r e e e l t i eV u t v s yi t H t d o e a o r t r r o e t ei wd C n ef u od s t e o r t n P e o t a t nt s sP e r r n eW M t s n y S f t oi oR e p l C D a t ot , wC a r n i s s a R N DP e s s aP r f C i m t B l o nl t o l p e e y i r d ol J oS de na ra y B o ol o i a - uR e e nP eB S V D S M a - - - - i R e C nt eC T e 2 a e u e r g P c x n t r p cd g o R c
r 7 / Vl eF e e e nl l t y n o o c 7 7 7 7 cl s s i gn e g aT e n e ni e/ s r r y s i e e e a fn r i dat y9 99 9. A o edr A 9 1 ut T ol l S ed it e r a t r i V n o gt g ad a ut e ip I 1 t l t l nt s gt w w v mf n e1 1 1 1 t t eli r S el d r r ol l e c o-T -
C i b n p e sf n a yt ei a r f U S n a o e p n f ot r e r rt o a6 6 SS o oU p o yT n a r e el s a a r s nR p od ut ui e g o oif o
+ F R UCC P S I OLS D W P P S T S R R R R P C B A C H A I F BE D F I F S P
s
'v<
+
' Uk e e o e e C n 8 7 9 5 6 0 2 254 4 t 9 6 0 6 5 2 1 8 3 2 B E a 1 1 1 4 4 - 3 3 1 3 3 2 3 2 3 3 2 O
N R N
+
+
e e e e e s e e s s s e e s s C n k n u v v v u n u lu i ui ui v i i i i i i i iiiiv uw wu w w w w w w s w w w w s w w u w w w w w w w R a de de d d d d d d d d d d d d d o o o o o o o o o o o o oo o o o o o o n o oo e e e e e e e e e e e e eL t L L Lt L LLt L L L LL L L LLLLLLL N R M M M M M M M M M M M M M M M
+
- + 2 3 0 6 2 7 0 1 9 e 3 4 1 1 1 3 0 0 0 M r 2 4 2 0 0 7 0 0 0 50 0 0 0 1 00 0 R o A t 0 0 0 0 0 1 0 0 0 000 0 0 0 0 0 0 S - - 2
+
+
d 0 0 e
t e a u C 2 1 9 C 9 9 C 6 6 3 0 4 5 o o9 8 0 0 6 50 4 9 0 C C Ct C CC r l N 1 7 0 N 3 N - - 4 2 3 t t 8 9 7 4 5 N N N N N N N g a 9 1 5 5 1 3 8 eV 3 6 1 4 4 t - -
I n +
+
l e
n . 0 0 0 0 0 0 0 0 0 0 0 0 0 0 nd 0 0 0 0 0 0 0 o o 0 0 5 0 0 0 0 0 0 0 0 0 0 0 0 0 050 5 0 s r 2 0 0 202 0 1 00 0 0 r P 3 1 0 0 1 1 1 0 0 e - - - - - -
- P +
O C
E 4 N l N e n y n t 0 0 0 0 0 0 0 0 0 0 0 00 0 0 0 0 0 0 0 0 o e sf 5 4 0 0 0 55 0 0 7 0 0 0 0 0 5 5 0, 0 2 0 r a 3 4 1 0 1 0 0 0 0 1 0 0 00 00 0 0 0 0 0 eS - - -
P -
+
e+
t c i n e a 0 0 0 0 0 0 0 0 0 0 c m 00 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 1 0 00 5 0 0 0 0 0 0 n r 5 6 5 o
cf o 3 5 0 0 0 0 0 1 0 0 0 00 0 0 0 0 0 0 0 0 O
E r e P +
+
c y i t 5 5 5 0 0 0 0 0 0 0 0 0 50 0 0 0 0 1 4 0 l e 2 7 0 1 0 0 0 0 0 1 1 0 2 0 0 0 1 0 0 1 0 b f u a 1 1 0 0 0 0 0 1 0 0 0 1 0 0 1 0 0 0 0 0 0 P S -
+ +
1 2 2
+ 1 1 1 3 n 7 8 0 2 7 9 2 2 4 5 05 6 6 6 6 6 6 6 2 6 1 4 23 683 93 03 34 o 5 5 6 6 6 6 7 8 8 82 22 1 4 1 3 56 7 99 1 1 1 1 1 1 1 1 1 2 R i 333 33 A t S c 3 3 3 3 3 3 3 3 3 3 1 1 1 4 4 3 3 3 3 3 3 3 333 3 3 3 3 33 33 3 I e 3 3 33 3 3 3 33 3 4 4 4 3 33 3 33 3 3 3 3333 3 3 3333 3333333 S
+ 33 3
+ 0 2 P c 1 6 3 4 5 6 9 1 1 A i S p 5 6 8 0 5 7 0 0 2 3 1 3 5 6 7 9 9 0 56 6 6 6 6 6 6 2 6 1 4 6 8 9 0 I
T o 0 0 0 1 1 1 2 3 3 3 0 0 0 0 0 0 0 1 1 1 1 1 1 1 1 1 2 23 33 3, 34 2 2 2 2 2 2 2 2 22 1 1 1 1 l 1 1 1 1 1 1 1 1 1 1 1 1 1 I 1 1 1 1 1
+
+
s . .
n s g r r t
o n n n e e o i o ip o t t n n t i i r t I I g P at i n
c a P a n n i c t r ri o o f i d n o ot i i i f e e d d s n n e t t d i t e u c od a u e eT b u n M o l l r V V i d e M e et / / n s r e t s t R v s s n
o ot n e
- V t R s n eiA y e n s n i i s
- O s y o u g LI e o t t y e y S a r d
M i t n S t i
s sL o di s te r e e s e r u
i t
a aS c c r
u ot Do i y t u r t N c I nf t a s s n a iip l t T e t n a a t i S nf n p c a A j tn ra peF S w a t W n o e c f f i si n a e at nB n e W i o l i i ii r s sT aF y t nl e e - gi s et pt f i l ei c t f r g s oP S u p g n iVS A s o v ei ct n t i s s l n q e n i g s - u e ed c i a sE ed a a r6 P r t
s yi iS t t n e si r n yd r n a nR r i a v r l t o l l e3 e uil n o oh o uS F r t a l oM nCCt - n l
e ip n uVa id Fo rP t s Dl a at i n iatn Est g sl e e e e e r o t 3 en O h e
t e et v1 l P ol 5 t b s aE r u el et t i h i a2 at eT s e t e o e n e u n 6 u T it l
t sC o o e a a k p r iP BI t i e
e oA e r a n u n
g el e c s r e n s r n o r i eP h o e s uf D ai el s V
S p s
S nS e u
e t P d S ol a s h l n/
t P gi t t a8 l t l n t n n e t i a e eR L ah o p p3 n s t g sii e e - B t t d y s o a i o t i t t e a e -- at R eS t cVn t 0 e r n et R c s o ga eC r t et u a oR I R ei cl sP C e w p a eiL t e si r r i I u u1 b qi q2r r LP g l
a n n n r v o t w y st i C l E E d e t W n ei8 e u ot l R yM ei et O R - - y a r a r r y a - - 5 eA nD i r oV w4 t sC aB 2 r0 o T n ei e st l S ,syl yLo n o e n t n r n r c mb eS ot eii ui el o9 9 eU o y t l 7 7 n p w cd nD R u r S c c1 aD ni
,C B A e r22 4 e v d t n i g c g n cl eV S C r aW i a g e aT at e r g s r / gt s o n i
p o e a e et r d r w g6 U V mC t r r st t a s t 1 1 rR ld f i b eA od r t r I e eV re ece se eeeC a iai pd l t
T o y o x r o 1 W CSI rBt d nt y r i C a e ei n s n t i n p s e o a o n e oS6 nS o o u a i S nl t t t S s a o
+ HCE DT 4 R M P R S OE SCBC I S C R R E F C M C W F M E E I iI T ER B
4 O t e e C n 0 1 3 7 9 67 8 l 1 1 I E a 2 3 2 .3 2 3 4 MR O I 4
4 hc K
w -
C n R a k
w w w w w w w o o o o o o o o -
s l- w wwwwwwww o o o o o o o o o N R L LL L L LL L L L L L LL L L L L 4 -
. 4 9 e 9 2 5 8 50 4 2 2 M r 1 0 0 2 1 0 0 0 R o 2 A t 0 0 0 1 30 0 0 4 S
4 d
4 -
e t e a u C C 9 2 8 9 9 50 6 1 2 C C rl g a N N 3 3 9 4 0 5 6 3 N N 2 0 2 9 eV 1 t
n I 4 4
l e
n . .
nd 0 0 0 0 0 00 0 0 0 _
o o 0 5 0 0 0 50 0 05 _
s r rP 1 0 0 2 1 00 1 0 0 e - - -
P
.l 4
4 e
n y nt 0 0 0 0 0 00 0 0 0 o e 0 5 0 2 3 1 0 5 0 5 sf r a 0 1 0 1 3 50 0 0 1 eS P
4 e4 it nc e a c m 0 0 2 0 0 80 0 0 0 n r 0 03 0 000 0 0 0 o o cf 0 0 0 0 1 00 0 1 0 E r O e P 4 4
e e e e e e e e e e e e e e e e c y t t t t t t t t t t t t t t d t d t d i t 0 5 5 0 0 00 0 0 5 e e e e e e e e e e e e e e e e e e e l e 1 0 0 1 000 0 0 0 l l l l l l l l l l l l l l p p p p p g p p p p p p p p p p p p p pl pl p b f u a P S 0 0 0 0 0 00 0 0 0 m
o o m oo oo oo oo ee oe oo oeo oo eo ee er oo mr oo o r o C CC C CC C C CC C C C CDCDC D
. 4 2 2 4 1 1 n 3 7 8 3 4 6 1 3 4 8 3 4 7 9 1 0 7 8 9 0 2 3 5 7 2 4 5 6 9 9 5 6 1 o 4 4 4 55 5 6 6 66 7 7 7 7 8 22 2 4 22 2 23 3 3 3 34 4 4 4 4 5 6 6 7 R i A t 3 3 3 3 3 3 3 3 33 3 3 3 3 3 1 4 3 3 33 3 3 3 3 3 3 3 3 3 33 33 3 3 S c I.
I e 3 3 3 3 3 3 3 3 33 3 3 3 3 34 4 3 3 33 3 3 3 333 3 3 33 33333 3 S
4 3 3 _
4 P c A i S p 3 7 8 1 2 4 9 1 26 1 2 5 7 9 4 0 7 8 9 0 2 3 5 7 2 4 5 6 9 7 3 4 9 -
I o 4 4 4 0 0 0 0 1 1 1 2 2 2 2 2 0 2 22 2 3 3 3 3 3 4 4 4 4 4 01 1 1 T _
1 1 1 22 2 2 2222 2 2 22 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 22 2 4
4 s s n s i e e g
o s e i e g g s r s d e y a n t r n n e 7 t l r a a u a i e 7 a a g h md h t t 5 e n p n C r eC s s 0 e
8 r A U o g o c e)
S y s 6
l B i nS f oST 3 d c A r n t i T n rT 4 a sE n 0 n d e e a t i P l - r o nR a 4 5 o e et z s g g a3 o l e oU m0 -
e i l es a y i e n d gn n8 t c
a i N r 1 P t l r T i n ni o lt gr t o - L n s a i H S u e e s t ait i L t st B a
e e o af c o f 1 - o a c sl r I i e oe F r g r s ck t e a s e c( R a r i e n P o e a c t r l e1 e n et 1 t i t t, e y c r C o nT a
t eT n aT u3 n nP y
f n i o i vt n R s0y a r o p l n e D eF 7 s o D e d i pt l W oM el r e c . t e t e ot e a a eni dS f t D e e t n st n n e s l n eM a e rV et n o ei p r n m ut n a e n a ed t d o at t S t a n a ip l F n i n ii nt n s a ol s n n ed r i y n et o t .t o u s M P y s s h s e I a e v a e s a y o eM t y Pl y o e i S i m S r t M t Sl c et s p T p y d a st e n4 R nS 2 V e t e eS e n u p e n r
- 7. ol w noSnR v e cl O
nP a n t d t i3 t i e rI e s anR r ot yv o u s a - p s a r rt u p g r e r o p y S D l
eo Hi r vt t e uiSB e
M s s P eM 1 i oM e0 r oi 8 r P t 5 t d t f S r nt w uid o s e y o ona w c o l l ef Ot wi s t ei a at l r4 e - t T - t t oI e n i oF r l ct r rl n o e2 r s2 sL 2 d i E r yd l e ee nD e u e op sc eP - P t oBt o eb C l r n eh r n v e r ol i F f n st e s2 t l P s6 P R B e o s c e a e l et t e s o o s oC d m r s e2 e o - o o1 - - d e t i s ot R t e n ct u e i er o ei s n r el l. P 3 1 e g c nt e p a n
- 3. t i
T s e a nM a g sVP d R T t e oP ot r a eF e H n i yR l Sel e n n t i e r r el l P s s r a e si t syt 3e 1 1 s - 2 e 2. 3 342 f5 t a iI ct r e a t si r aV yW o n n H i e et t c d T eel ol A s o eS n 3 n1 e e e e e e n a s i
p r a t e a t t r e e et n I s o e s p s e n eA s f t entee r t t ivi bul Sy et eI wh r d t t V t w r l r c n U o - a o aP r ig ptgee8uo rapWe0 t nf C y u rh n U Oy t t aC et t p t ta S S aCs y eV e oeu veC s e e e e L y st a d r a T
4 W C SL B H US RR 4 C D S F St C R MI I R I I I I T M A S D R S T E D
9
. 'D Bj 5 I=, .
- 4. l Ea b
,! g .%b i!
- +
+
%2
=>
l_T
+
i_-
12.
I83 I[e B
Wzc: i.
8.
W +
N 82 8h .
oil a. +
U W. m
=t *
- 1 23 .mm. . III
.8I9 a-l,9~ - -
- 3 ., - - .
er
- bd W
. ~~
+
N i e i
~
C L's
'" : b
=.: A * )
u Cft g.
?,-t"3 1V .b m
0 k
D )b E H 0
0 0,
000 5,f/ s) 0l 000 000 0, 5, 0, 00 00 0, 0, 000o 000*
- 0. 0, 0, -,0 0 0
0 0
0 0
0 0,
O T 4 1 8$ 780 94 1 00o 4 1 3 s A
MI"T 1 1
421 1 21 ' 1 9 2 it e
I T i v
T N it S
E A c
a U g Q i n
N le f
u e
O I
d er u
T t f
u U )
e >
ir g
n u
B I
b a
lm)e*a d
s n
ul o R )
a) u b c a am- yt T e a r
Aer Vu u a n cw r
r e
S I 7 d
R s
o dA l
l eds ir ll ed oai NVa r l((i l lo w
O o
D 8
/
8
/
7 la i
DiR eDn d r e r
iaa=
laee=
i r
r r t t aa i A t
T o
e r
e L f o t a
e ie t
s de unw i t
eMMs a ee t
c a
A s OUe a Mss'a n o t
a M n( ( t o l e oss B e r
doo e e sLLooo i
r I
l a s !
eed t g o gii e al e o R
i N
r e o elbtt l
a Miar ut t R S t
O t a k s l g Rmr T c de Fnn*=
rddi ee r C R
d e
E c eeelo e O_
t I et doiie c M o n r sse Aia T
P R i
n ossR Sd f
i M ahccccn et m !
e e p
T d r I
CAA Cl os i
e Hl o - - o s x R
C l
e e
etuia gnm rbbe s r esoo rnt ssi AAw t F s e L A
e A euu n e eo ooe s r
S u a oe LDR wSSFt t i wRL wM e
r V T E
D D f
e ))) o L )a b )c ) o L )a b ) o)PP" L a b )c o
) )
C o
E x
O T
l i
a r
M abc t
e a
m E E Q
la n
o R Z 1 2 3 4 5 6 7 it i
d d
O A C N t
e o -
D W ,
E g3 T j A N M 0 I
T S
E g g g -
0 g- 3 g1 e
L;j :
O I
0
] - -
k -
i
ESTIMATED END STATE CORE CONDITIONS SUMMER 1987 L -
o mtTITFI'!q s s 5 s e 5 e ::: _
E E pAAAn g ._
sf_
w {
l [r s
L '
1 i 0 : 0c0 0 0 0
0 !
,- [k ,
)
l h:ogl0kO[0f.!
af$ 1,la}
o 1 et:e 1
, 1 ag%=ca >:
ao 3
l y$:65$' : 1 )
ll @
)}lJ FQ:gg:
j:. mgp 3
t IN h
,'w.
l O uuu G
Y l A l B j C l Dl El Fl Gl H l K j L l Ml Nl 0l Pl R l
$iiII:l4:N$$$$$f 44
- ::: vM::.: ..
'3 IE -
l+ VJ iiii -...
'2 p p ::::
33 y nock 10 y
S
+ P + O x -z -5
, p + z 7 y y' ,
6 5
+
' t 4
P P M+
iii! M&Wli L+#s iti:
2
!!!!:!!!VA L6iRMH4 !!!
t!!!!!!!!4!!! !!!!
w Th aterialin the P4 to R8 region
< 30" is presumed to be resolidified crust
@ Q 61" to 80- because of the melting damage to the
@ 30" to 40" @ 81" to 100-baffle plates in this area.
instrumented Location
[ 41" to 50" @ 101" to 127,,
g oonui g eossibie oonui o core oriiiee tocation P r,ig;g d CORE STATUS AS OF 7/5/87 g
b
.I k .$h$$b$bbE!EEEEEEEE o. o ll' !!! lli li! lll lli lll lli lll lll jll l[jflllf[llfl'jjflllflj;fl9lf;;gf gggflg;f g;;f g;; g, ,
t t
E C l 3
z e
t i r
ll ,
2 . E
$g b$
o i
h E
a !!
W di ~
- 8. i 8i LD h 5 A'! 1 D' LLJ 03 d
- t l .
- LIJ g
- -
E E O O ;i Y'
/ V/
/o
/
~ !-
!: !l! I si
!si !! H G l I ?j n iE ? S i- g I
, l' ~
- i l-
?! jl e
- i 9
p" e E n p ;t si
!" #i g .~7 e e ;i lw*I o =
s = ".
!! 5 a"
3 lgl
. E is!!.! 0 0 F.3 0 -
f E
E E
fe
-E e
- Illi $-
Di
%g V'
o N' E NI 3" L e"
"d
" El g IIIlIllIll Ill It!lIll lli Ill lli lli Ill ill 111 lll lij (([ lll [l[ lll lll [jj lll lll lll lll gjj lll lll ggg lll o
EEEEEES20 9 o o no R**C o o o o* *o 1o2 oCto= o2 o e eo o eom o r~
- mou a -o n n n n n n n n n 10001 X 's ell 03Ov07 "13nd I
5; L V S' I
0 '
M NIN - "
0 E E 1 E , '
A R ?
f OW O .
T T EC 5 B NI
' D A
L C
'6 1
. /
3
=
k
- 2 a - 6 1
1 3
, 7 "
4 1 )
- - 7 6 R /
8 - _
1
/ 3 -
R
- (
3 1
- 5 1
n - 4 O
- N M#
E 2
- D I
3 U G
A["
r.
1
- 8
/
7 R E
6 V t - O 5 C v
1 -
/
- N 3 5 -
I 2 2 c 1 A6 I -
D 2
/
1 A 4
l
=
f i
U 1
Q h
I
\
\
/
\
.a
- 8, k 1
. ~
.o $
y.$ 4 -
gT / ( _
\ qi s \- t
j ;
3,: , 3 --
~ "
I 9 I 9
- = =
j I +
+ 1
= . = f l
- a 1 9
9 -
- 9 p ,
{,
=
[D 84g bi =
l s
- = I* l, l s r Q
- n , n n mm n , ,, n, n ru n rary n i n ,.n -
'u=
44Uii111..ii1..Ii11.y 7 .
I q
/
l
\
e m
a r _
g _
o r -
P n n : o 7 y t 2 t i o b h d
g u
8 9
e 1 a t n o r
I e 9 u s B.
Ml a P e
r
. e n -
T v M.
J J u ,.
E t
n a e a a
d e
),
i c
c A
ol hag y annr doiro
/iet t
ae ar Nnio gb na EL
- I
O 2
0 5 _
4 2
0 T _
S _
0 _
3 4
h P c
r a s e l t
e s u e s r e
r e _
2- r v _
n e I
o s M i t
a n
o T
n _
a i
m m m t o
e a o a8 a$g .
i 1
r d _
f x i r j;+
e O i n d e
r d a
n e
2t s!
I e
- n l s n c u t e i
u a s Ms. .
u i t
v q e g t Ts i m
O j b
c e
r n
o i
l n
e u d n
e f
t oa l c
s o i t f i
c cn i u
a e c ah-pc t _
d m d a a s _
n r t d mte t-s i
r _
a o n e i
l t i
i o
f n a t an m s m c a i t e a u a
r i
c i
f l
u nd r l
c i
g n teic g t i
s n s o o a g o oc r o r
P B i
S P Pa P C o o o a a a o _
O _
3 0
5 4
- O 2
0 T
S 0
3 t e 4 n r e s td P e ve nn d eu ea mc i
c s s c s ni t s a fi r s
ol a e e e g n nc v om i
v i r
u i oin th go d
t d uc o c d l
oe s t t -
j e e n em r
mic ab r
l b e oe r
gu p op O p a
t t ge nc r
pd n
O h ir f a m t a
du no o y a
r h
w t
as sr t t y u sl l
sd g d r
e n s ula edn o
r t n
a da nt un rno r
i rt i
P s r e f e aa -
e l yd se nr d pic pc n a ut l
ce n
Tni n r
U Aa o o o O_ -
g 4 0-5 4
2 0
t -
n T S
h e 0 3
c r
s n
m u
4 P
a i o t r
s e t c n s a s i
ne r o n o
l ba y
d i
oR t a
r i
i t
me t e d eh a2 -
p o
n.
o s se r
mI c a wl , no o rM o
d.
n a
y r
a t
r ol o .i t
fT n d pV a r
n o n u
pR u
u g
I m i t
o seh i f
i cor a r
u b et r
n o
sf g d od c a i f n a cna e B de n o
c i a
o t s e
r o
c r i t er i
m n gu t u e i
at c n e
q t s
t n mur r r
e y l a at Ds C u
R S P
- e o d ~
o ;
- e ;
, l' -
5 0- O 5
4 2 -
0 -
) T -
d s m S
0 a s r 3 e i r
e s d f o 4 P
du t n o -
en l
r -
a o a rt i
m i
t n c c o i m
in u e a s e
qo r o
r e
i o
p a _
e c f c io t n n i e d c
R (h o sada t l b
a n
a _
nc n i i n
o r x s iaor i i
t s
t e
al o
b r
u t c
t u O a e o p mia r d d s ,e n o me m o
st ea a r p
rR r o
l o c t um r t
n o
f d af r o n i s
I n2 -
n a e pt n o
c i f
s _
I e me i
cM t a e t xe f
o d e
sT a
t s k t t c
e i
n a
ad t Bmo d E
n en Pa f
f E R e
r o o e o f
O
, I
1 4
End State Core Configuration 1987 9
- c; C ; c h -
ME5 35 5 5 5 33 3 5 5 , ,
q 5 5,b m ,
l -
l
,_, -c -
l_.._!.,_e_ q
_,- )' Upper grid damage I M 28 inlet
- ~ ~
// A1 inlet A [ '
0 ~:0 0
= : 0: d 0L -
0
- Pt 0' ,
0 ;;-0 -0 :o e :. 0 -
w$r') I t 5 B outlet -
o lpw# p ore C void
)'I l' i l l5
\ Q af ) .: ,
[h / -l.::'2N. :: b .- .-
Upper debris 2 4*/ '
, ;E.i..Ii0.;'..':'?
'c', ,i bed
{
4
.fl[.3.f'f.j:$?."'I.E:(%
.. 5 r..
e>' '
' .I.f': ' .ye,')f[*.b.k.l }
- Crust (agglomerate)
.lt[ ti; ; l.l:',l.lf;? h. ' '
-Previous molten
! lh; ' .*..*:'#.
material g I ),% U "' :
's 2 gNj N ,
hj _e - _-9 _,
- , G=y 9 2 o*o Lovie plenum Previously . ! i!: ,
molten material 4.... .........
Failed instrument g y"-
structure Uuu g
~,~* - 7 1246 f
4 m
T% i,.'y. O
-s e ,1
~ -
a
. g1 *
=-
-I p-a -
, 5' l' '
.s e; s A*
d, ,
- "?
%' s+= pg:
v
- l , ,, s.
7 e . : =-
${g h .
- '(
7 g',7 - - ['
- <, . ,. ;f' ' 's
.S$ m- .
- h~- Y7$
e-_
'la yh . - ,
,;.9 h'. b T*-
r- S h. > '
3 u,ma,j N
-3 7Qj
. 4. ~
F' my. h fs~
- s. +4 r g%"a:,
x.
-- M[.[;
u, .
e pf
O 6 0
5 4
2 0
T e S 0
b 3 d o 4 P
n t f a d t
n od si c e a s ee gi x tl na l
t a i c
i f r et e m nm a i i
n b rl m g m t
s g e
l i de ad e s D aw o
r f n d i t
rt g a e b r s bu b ne l b i h
x m
e n a s m i
duf i r e gs d 5 pio io nn l ah cto b
e s e d pt t d c n a
i i
b t s ,b f e a O Un f
i ic ha xr e te e-cl ea o
e r
i p
c 1
n oma e i
sn i i i pre u t m e ul l t a a w oa ea r r yx r
ei r ne f um e p
e cs b t
e aE et a g
f n oet m e
n e e s s om t t n e m r ed t n dl e o sm k K a
l o hg muo l
i c
t m e0 r he a oply p0 yt a _
p u io s n msu e2 2 l s e ur f S i st i
a o ci o g
at r u ou i t o
r bid ser v eo vc % .
ex i vb ea r r 0 Do I t p Aa Pf 9 a = o =
- O
4 4
O
)G M a ., , m b, , y 3 ,,
l
., n4 g g d,G:' ' 4 o
& & -lT;,4 ~ e-
..,- ' . ,NT I* .. %l, A*f ' :;%,. "
':Q , j
\l,2.A ' il6"D. '
f gij .
p::y;,.f
- -
- 14 , .
an
%,7 rif;J. ; *}
' ,,; b *,
. /!L * " .
hi/N/,$ 4-e R ..J,, +n*
s, . . 5, m, .rw- . ;Q
$d y
.ss . -
+
g?ey..'h-
- O
}#1 Sw .
l;$$g d, . J hh! )
gn r i. .
I JQ(! 4. ,
w p*I er:'.4 ; .,
? 11,,,I .
$~.
p ) F -- 5..
R$n rs..ys-a u ,r,..
[h., [d ,. < I. ..a );! 4 5
%j[. *h, ,
-~ : . : . ,q
- .w<pg.,
((? U; , -c g ,['1 Y 4^{ j';gh ;;,,\. ^;. , 'y:lYf";.' b fj g i fff . . . .
sth
":ks. '-
2 j'
y .,' ?, , , 0 s.<N / Y. ,
. , ~3. =; .a% ; -
, ; . . ,g
- R,47. '.J _..; ' '
V N, 1'
,a y: . . J L';4 Q...ll g;f.j 2
ilsy';y?WR"c ]C s Ml 'j,:
7
'gl;,(. p , , y,( ,,
u
' % ;*% ,,l
' ~
UQQ s'd$l%'MNNy[:,.4.f T u
O
7 0
5 4
e 2 i
s 0 T
S w s 0 K r e e 3 4 _
l P 0 h t
p 0 o m 8
m )
2 2 - a s
u O n s s s n s )
Z, r e e e s o
l e n U w h n
g r
c Pot (
r s t
e u a i
ou b o n "
r a 2or s t o e n i
Oo i
r e r
t a
Up b u r wi ( e t t
n La om ie cr mv y
m m
d f
o f c
a r
e c
n -
x ard 0 e) t o -
f en 0 r ug n c oE ca 2 t n ait a s ce t c
o i
ys es t rl ee f n i i u
bu nl d
- r i r o 1 pm ge o to e m2 r ab n m ir su p o eO t t me r e ag e o f r
t ka (U ic b
ia n
o mD hr f i p s pm e r eK x eht s u ao a p0 i
f h s s i S s ee e e0 g1 l ew m r
o l
l cb c a3 r
pd i
tr n i
t r ed vn me a ca f
i n
Pc aa P a
Aa Sl U
- a e *
- 8 0
5 4 _
2 _
0 _
T _
S y .
0 es 3 4
P hu t o n nv i
e u
i o i sr e f t Cp ms n 7 3 d or e -
t e '
da n f t r u c -
d e n s ai er u _
R r st -
Ib a s 9
2 e e c t 1 d l -
c f ore s ei r
l l
a u l mte d n oa um o cire o
i i
t et n r nr a en P e tu m h t n i
n t
e a r r e uo ri er t i
o t po n
amc t n n
at e
s cen ce i
s i f
i t et ir f
i nht nhl F ggo gi Shm i i Sw i
a a
O 9 0
5 4
2 8 0 2 0
e 8 2 8 6 0 0 3 T n i s g n 2 3 3 8 1 1 1 S r - - - 0 a - - - - 3 o"b id e R 0 6 8 1
5 9 0 6
0 9
4 P
t e D 1 3 7 )
Ug nn r /
t ea i t
e p e
p e g
i C
p r e
er U a r
e 2 2 1 2
8 2
5 5
3 9
0 9
4 1
(
i s b m
Ry r v 1 +
s y u o A l
a n
n a.
t tn ce uv L
., a f
o e
n 4 N g di e 0 k l5q,y r E a ofo g 6 f . 2 ! ,s. $ p eva r t mn 0 2 0 ; i Pn e u a n R 1
2 1
9 1
- b MkD, ll.
O spe l ee sj O e 0 9 3 4 0 5 6 '
g a j: thys s nrce Pr l
l 4 7 0 r}
na (g' on i
ope e
, .,,1 aniar t
et eee s o r g
w a 2 6 5 7 0 5 4 o n ai tc r
c cobr sl i
s L e v
1 0 1
8 .
1 1
l e mnp t a F A -
C o
hiu t )g e
n i
wnaom d
e d e
p v4]
4.
hr i
t wed uc 0 e4 -
r i l
c e
gli5 u u "
C s b S R u r E
u C
dac( r e e na ul l s n o 9 1
7 s S 4 4 r
av o paim a i d
2 1
3 2
1 ' 8 5
4 md ds a
e R a o CanRo
- f M
- O .
u
es r
n _
o o _
Bi t e a rc oo CL pt
+ ol e lot u
" o R B
P e e O
. e N M
L e e K H
Y
. e G F
E i
e e D -
C i
B
_ p o A lot le
" t
+ Au 5 4 3 2, 1 1
0 9 8 7 6 5 4 3 2 1 1 1 1 1 JlI -
l
O a
1 3
j, gi> l o
3l g.
- jl ; y
{j .
'{ 3]
p; ~ .'
4
- $ lg y g i
.. gs hg1 4a-1 , k g,m M 's -l h q!{ 1[i i.j i
e u#o
}f f
1l 1
1 5
4 2
0-T S
0 3
4 P
s n
i o
t a
i n i c l a
h c m p i a a m .
i r e g
x l a
c o h c
E l
o i
s a t
i y e d e
h a M
r P R o o o o B
e r
o C
- ll l ll
o 2 1
5 4
2 0 .
T S
y 0 3
t 4 i P s 9 -
o r 1 o
p )% t o
s n
(
5 i
e e p 5 -
t i O s
n e e)
S am 2 8 5 6 D r e/ c 7 7
8 7
5 6
6 5
9 0
o f o
v A (g 1 y
r 8 a e3
)
3 9
4 7
9 m gm o o n
a /c t t m Rg 3 8 u (
4 5
7 6
S -
s s k
c e 2 2 o
r o O O R C Z r
Z r
U -
o -
- Tm n Q,,?lly wSpp, .
/O'
. (. , _
[<^
[. 5 .0 9k. " , , < -~ %V # ,a...; '
i ,t",%; . '
y.
' %,N hf;.gj-[_~ 4*~
, 47%. '
,J I
.@', . , .es-9
- t .91
' x- ._.w, , - ^ ,, .
( 7 1 i .3 pc ' '-
g,_, , --
s -
Jr,s., + ,.y _ s ,
'y..
l b 4
.4. .. y(<;me
'}
hi
-; ak s
'.9 <3 r.
= '?? ?? ' '
,3 g v 4% .2
.*g
% . .' .,,. \ f WlYf}.npR*
4
+~
- n ?> s:, f '.
- y'
_>. _ % ,y. 'V
' Y"$9 .
pp .
I,,
.i '
9 ' , If ,,
k - _ %
9.,
Al~
n,.
, , . .. . y _ ~
fj
?.+
l $$
h f
~
r c .;.t 1 - r. >
f p ,
y ;el* .'
_%+ , ,
.C E ' ~ t
- $ te k '
t .. .#
s f
- O
- ?-
,,f" u
O
. :4 7 y , , ,r Q
, , _;;g , ? * : }
. hl A%d'.:
g ? Q ).-
r- *%
_qa -'#, / * . 1.
c
- l .b.. ,
+
.~ ?? ~,,
_? ,.r ( , o g
u .
,s _
~ : -y
, ;,. / ,g.
44,4 7
,.' s
.Nwh
,6 it .
.M;,:
'F*}
.,* .64
- ;; <";; [%p;!Q
~
Qc. i ~ U. ,
.s ,
"'s.,, -
y ., ' ,{ e i.
., O.. ,'
A ++
ll' , ,
g,u;w/S*n},
g*a 3.'
s 7 <
spyI,R:7'[:* ,,; t. , ,
,. c sfw < . ;f
,k l5f
yc n
gp.
Y
O 3 8
2 4 4 1
n n n 7 o o i
o i g
e i
g e
t R R
l u
o s '-
i s =-
I. r
~- ,,
D t'
Y~ _., - A.
- f ';
g -
. :s :
i n i
- j. '
'n 0
d ,f ,
d a
,f.f 4 x.
.g, a
_'"Y A
~
l ' .
C g r
, ,M ~
d n p , ". ,
O a u I,
n --
f z
o 3 .
}; L"' .
i . c~
t
- c .
a r
mNf u t
e ~ .
I n
l i
a r
t e
a M _
e r
1 n
3 n
o i o
g i o
C R e
R g
e O
4 8
4 e
1 7
- 7
"%*~
f*^
1 t
l e < b M
=
l 7 e '
i 8
/
t l
l p
e l# -
5 p
i~
,D -
- '~
e p
' f. ~
Y?
j "' .-
o I g t
i n 7
,' - [ y9 F.
h' i
w R'/~,k'.g}- .
e l
o -
A .
) .
e f L '- '
' ' mg
< . ,?
c _ 0 0
C {* k i
l - 2
.' [.
j,T l i *: 1 t
a Mf ; k ~
e }/4 i l
M ,e$
e' r l u; c e 1 t f a u x r F e
n i
m t n
i o i l
l c
a i
l t
e g t e
l e
e M /
t le p
R l e
P e
l l ll ll l
O 5 8
4 1
7
. - . 3 k .
n
=1 6
'w M n l Yd g r 7
_8 i
o '
a *
?.
t -
s c -
a r
s p t
e - ; _
~
y.
i n [;3 y .
.. t l
g.
e u
j '; ' ,;-
f r
=: .j i
/ '
g O n
- f i
d 4 9 h.
f '.
- .p m
- i n3
- 0 d ,.
0 2
l a _
,Q+( = t.~
l C
iRg f s X .'
l 2
a e
n r u
g n
l e
i o t i
x i
d d
F u
g m l a
e i c
m C
R a r
e C
O
6 8
4 O
1 7
t n t ! A. ',. , , <
1
.% il- E; :. P ;a:lf;j p. [m. * -
lW'1
.f>
,\ :?4 0 4
e ?r ih y l*- c%s?O?::,O'fC 6 -
i -j f - M d =:I".9 - & 4 N h .
7 a ..
. A u' l
b, . " e 8
! i g;J;- -
- t. .
r , -
Yi.i, g' g ,
'y y s,. ' . ' hm s4 l @h , 3- .
,k.-
, a.
i a g yt , -
s x .p~* ?E, A q h.
e 9
3 mp n .
0 0
i o l 2
g e l R
f e
c f
a r
t e
I n
e
- ill 1 Il] ll!
1 ll o 7 8
4 1
7 s
e s j a ,'; }, i
~
3 0
6 h
p ME
' y . * \c - [i.- . . .M
_' 7 8
c 3-n . -
5 i -
g; :.. -
3 .
m g j ' *.
is e
a r .a
+
a j
e 4
- w. .
? ,
c s
M -
- g 4 d
T
~ 3 m n
w
'4M, a ^f i
c ,
9 l ,.
~- + ' I.
o l
r:
a c. a.
p.
L. c t - -
4 1
m e -
yP -
y.
4 h
0 0
m _
2,2 i o.
f i o
O-
'3 y '
l e
r il t
u ,e e s
s s .
a ;
i x h a
p h
p h a
p . '
c c M i l
l c i m
i l
l a
- t a a r t e e e 4 M C M n -
i o
g e
R O -
Il
l Beta Gamma Autoradi6 graph of Particle K9-P1
_ \
~
. \
I ,'-
f 4..
3
. " i s
O -
O L_______ . . . . . . . . . . .
Beta Gamma O Autoradiograph of Particle K9-P2 e
.t
.s h : .
.w p
G.
\
~'
s.
W a
- t bwj'%..f*
)
O r i p3.
J ! x;A,
.ny (% [nrtu,dey?.
.A u
<'~
. y
- 4 *
- j
- M,,cW i g'6A l
y
. k' _
,' e
. * ~
m
, + k,
~.~'4 -
'> e&< <a; *;
4 j~ A
. Y '4
- d. %~, '
~
i C
f _ ;fg '
- ~~
r-4 '
~ 1 ,t
-v 9,4, _
g
?"
i X t
p f ._ '
i
~- '
O_
9 -
0 -
I
% m 4
c k ,
~ 5
/
g
- 09 1 y 0
, <. y% 2 72 _
t
\wp - ~
\% \
^\
~
s ~~
~
O
{\,
i ,
- /, ;I/l! l ,II,I '
1l!
O - >
- x. ro - (. 7 1
2 8
4 1;.g 3- 1
^
~$ 7. A 7
M
.. ~ CLz:
g~ - i g3,$
8
.= s.p'wY
.,7 bg
-. eur
, , 4 'g t _
e e,%,sg ^ .,J *.? , $;* > 'eg .9 cu~
t
?
4g
- 1 r
,iu g:
t f .
, ~ p* . s s
r..
k +,'h c r~ j' i
'L ' .? i t
r N
r*y}g A 4 g J.~Q e d
D n
e
$ 4 '::~. w< q! s y '
,r mp k%;sA ~ i3. + Q+.- ;. [00
, s l
e * . p1, f' ,.. :- l' 1
c l
/ .
i t ) sk;,, pi ' ? . ~ . ^- . -
- l rB a -
P1 O ~c P-i l
l 9
a0 t ( - j; f
j4 3 9
e ,c -
O k n'% mT g, i Q s;n *-M7 ' s, M - '
8 e r
[5N: V. <
o
+ : n p q PA>CmgO q a5 e g
y A e;a @n} phak p ,4
. r l
~
- ji t
i
=
,:V ' $ xw an ppe
,L:k
.$i
. p!
y t r
a
[(nyi'
= m
~ p?
W i c
h*2 l l
a
- ~We4 ?l t
e 82+WQ..
~www g,.3 y;.
M
~ %c$
O
9 9
, a s :.m k?$flfk,.e 30 :,lnme(x: a
- j. m 7 '
R' b; !s Y! :I f
$ $s ? , g% ,., '5l;jlf:'lG ',' ., ~$3
, , i,+ .
> e
,;, y 3. ' 1.,. $, s.1;.l*! e-
. , q' : . ; Q;9-[ ,, g y,., ,' .
..,. : p .3'g -
' ?$;
fw+~~. h j 5i t Y W k .h b. l l ., ? s. - .
a
. G. k h h.. I I [ v
,Y n, . u .mi[,**
4 ,~ ,,^ Q*. ,, ,
. ;v p y. %.
D,w knEYdi y s .% :, s .,r/s:I :: 6 , ;
w *' *mr > ..
- w. u . , * ;? .g,.g
%i -:, y;; r Ap.,; ts ~ .
~ - 4v -v .
7 ,4
~- up . w, , . . ,,, .s.. . - - ,.
,, + , , +
- # a
,.-w , ',. e p.
~
s ' , i .] ,.et' =
,+ .} ., fi k .
9 $
V ,9 4
'%:b
+-
,s-
+,
,a g"s ..#
V
'~
' *g i
.)
s V $
- a. -
m Z
P
//jj k a., ,
// ,
v
~
- // k.N4^^
a
/y 54. '
i 1
f llI/h f)D '
r
- l O
4 7 7
8 4 4- 1 7
M 7
8 l
e i
c -
t r
a P s g c i
c 0 9
1 l
e u i d
n i
l l
a M F d t
e m 7 l a m a) r ;
?
8 C n e
eD - q t
l o
C1 1 6:
m s
O
& P
, u i
o v
5 $, e i
l l
cN( s P
r c
t a i l
l a
e M t
e M
d 3
0 e d "3' 4 4-i x f" i M 7
N~' 8 M f~
S' 7
~
M:
f$
O
0 e
O oo ~ t c4
@. ~O c4 s co C4 r @@
f .
.e* y h
.~
4 e do Q ::y l sc
- h::,'
?
';!'ir
! ^
- v. . ji,,, m 7; { ; ,ll,1._.
'/
jf -
' }I?f%y:,,, , yd*
e
//[f f, f g
- m;,
(/[jf ,e s
/ $,
- 3. ' 2
$ %l%l s EMh x ' %a f' f A,..fsp %
Wq 4 g
g%j%jg y
r g W
~'
ny$??kiy .
'f j
.i + m . , e- \
w.
lllll l
o 5 5
5 4
5 4
M
- e r
o 8
7 4
1 7 p 7 8
d n
u o
}
r a
e n
o z
n o
mi p c t
2 a 9 0 r 1- 0 2 t e
M 1 n
7 I 8 i l
e ? c c ;
i m
.iE i
t ) '
{-
a r
l rB e
~
P6 a - -
g1 _ f.
,s ; f c
o O c-P .% ,a ' '-
3 Ug n o
i i8 .t.L e t
c mG a(
f, j ._
,V-e s a s r l
i r se e ot r a C Cm 0 6
4 -
e nM7 r
o 8 p d
n u
o r
a e
n o
n z
.o n i
s u i o
l c mt p c
,i n 0 a
r i
c 0 1 t e
l l
n O t M
a e l I
' ll\l' ll lll I
s f
p
+
p #
~m x b
$g 37 e
,y -
4 p
/j. ,
- w i
p
+
y- k A =p Q
mY((h ,3 l l .f:g[:{. k b nl h up (:m ~ S d e
\ ?
/@ ,geyk, E/ _. f. 4 e, s #f}[,-
v '
p^
- k. -
ap t
y$
l&g p'
/
fs
> 2a p, 4 ,#
8 4 #
/
4 Y, ;,
- j#
p ,,
L- i
. ; ,y L;'i;u(@$ . @
N q9y -,
'M 'T
5 _
s m n
o\s i o 9 4
~
t 8
a 1
g -
e 3
]
r 4 l
R g 2
~ P e T l
O s S I
el 0 N t at as 2 G
l M
ei n
- r 4 l
e P l
L K
ic m mev m o e n
l H l l
al g g a G r r t
~
l e g e
- _ -_ - _. F
- o. M m 2_ E _
l E
C A l.
l u -
I I
D C M ]P/ E s o
I
- I B
- _. . l A
C _5R 1 1'9 }_6 k 32T_
1 n
i o a t -
o c, {.,
l
- - i '
I.:
1 e
- l 2
j d
S , ,
R o 3 d le 4 e
s ,
r g lon i F
p g 5 o w
s od d cd n R e
so id o 6 ly il B .
oc q n
b s
r i r
t S 7 me d
3 C ,i i
V o
b8 D
e r
G s ' -
8 s 9d A
s e
e o p
s i
p d 0R r U o 1 le o
v r i
le 1
1 u
F u
2 ,; ; f C
i 1
f 2
1
- o i G /rp- m 3 j t o
1 i
g$ 4 t
d d t o I n s 1
2 i B i
, 5
- _ ' ,
- : i : ' , i : \ * . 3Y
- - - - _= - 5- 1
- 0 0 0 00 90 807 0 0 605 0 04 03 0 1 2 1 4 3 1
- 1 1 1 mu 2GE
) 8E Y3o3{'0 ggo j $B< gzg?w
L O
a N
' m = N O k o 3$ *
' W - - :
h_
8 F3=
g e 2, e
40 -
W g 2 r
$g$
E C
C i
O 9 g ,
D Wm s
' M N O N C .fm j ,u9 3 i
6 W e e
~
E E
s u O
c N N c - o E m^ m m
e u 5fli d=c i 0
O V E
.o e .
c u e
g .o I C e o "O E c o e
- a B E o o - e F Ti
- e o a "
- o o LIJ a > 5 c. 5c a stz 2: $
- C t.
o ct o, o 80 0 > a 2 G S O
1 l l
O 3
2 n 1 o
3 2
- 0-i t
d s T S
e t u nt ic n
e 0
0 3
b i
se t p i cim n P
r cx l l hi r t
t ue - aie -
s d s r t
ewpx s i oi r
e mye l a
D t i ps a l r b hee te g t n ot u w t o r i t tl ea s an c iss niv wlpcl se mid a a dml d n u i sd f or l
oh eoa o l l
ea d s o e. t a cm m ut s o
r ad gn cb nH i
cnsli a
f ne id r
P e a dsO i
oamic s
sm hn t
aer ot l
s y iu d be eC l wtn -
n oi l v y yd n rf a o nb de l
od nl er m sn oat e a nr l
f i
s oes i sa mian t i iu ac n dI
~
np i
s iao s mC s
i t
n sie eid c t
es r
F t n e u a std er m nu l
ih e u de al _
ri st afn _
f t f e i nt nc o e rd cw t auu b ni so ec t e dn m s - be d ng n at a y aa ul iaV iy nl l r cm s niPd aa a i f
ia nd in ei s n e ht t
r eRlu hc n
ti s
m g do u ae ho a a in S oc Rmht s Lb
- u l
m a = * -
u S
O
9 e 1 M M M l 4 C 0 N N N i
N 2-1 3
2 0-7 u 5 D M D 5 8 R_
0 N N N 0 O
S S
P - _
t b 3 1 3 D 7 cu S 0 0 0 N 0 d r 2 1 S 2 1 0 3 o) r s 0
2 3 P (% C 4 1 5 5 5 nn oo I 1
2 2 2 M 5 2
i i N
st e se subs l 7
4 M 1 M 8 4
9 iboa N N F ir NG et r s t
n i
e d oD m m s ed e n re C- e s
a t
s y
i s
e r
ut sc ae e et x b g
s g n o
r o md e E n t
n a
i d
n i t
c t t oo i
d l o
l a e NN i
l is u n d uk o b i
m i
s ==
bn c y t a
o MD r a r r t u ot t o a n NN t
c dn c i
l i
x o l a
a a a u c t e e e o R R A D T e_ -
O
) _
g - 6 - 8 0 5 C - 2 - 1 0 1
2-1
( 3 u 2 w 0-9 T u S R_
- 3 -
1 9 0 -
- 1
> 0 3
P h - - 8 9 t S 5 9 c - -
0 -
u )
+
d S r -
1 2
4 7 9 ~
o) 1
( _
r s - 7 C
5 3 4 P (% - - -
n n ioo ! -
- 5 -
3 0
5 7
sit e se O
su
=
l i bs - I - I 2 Fb oa - - 5 NG i
l r ets oner% )
l b
e a
a s i i o(
t 2 3 9 6 l
i a
~
sDacCss 4 2 1 1 v a
e FoM r f a de t
V- s s
l e
ey wt eo s n n d e a s s I
b s d m
m em) E*l v- oa e
p s
i b
r t
e u n i nrt f a EmSa d e a l e dn (d 1 o
r d d i
p e t o e [u =- -
o l
t c
r e s r
e ciro' t
en c a p n w je-ot t p o o r e x i
n U C L Pre O
o 8
1 5
4 O
r 2 0
d a T e S c 0 3
4 P
y _
i o d V r s a is y
la n n a
e g in c r e
e S n ig n _
t E :
C G n _
e O d
i c
c n t s A eta t
az i
o t neu eds i
nt dc er l
t ri a nc 2 set ta dcd eah p
erc
- e na de r e a I Er a i" a l
nv s I e sre M
h c
O T -
O-
O 7 1
5 4
2 o
- d _ _ T-e s n o e s p n 4 .
) o i o P -
7 t 1
2- e a l
v e ro c _
0 a e
r 0
2
(
k v
c ol Cr e -
n o d _
u o l e )
fB s 4 =
s S I _~ l o
c 2
2 l: p s = P H e)
( du et e t ds _
imv a 0 rge
0 r _' a9 v g 2 3 ee P
tt n
kc (1 o )
4 :
Do r
c
)
S m
m u
e l B 7 lr t e
Lr i s)3 1 m n9
(
u n
e t
a1 r -
t4 p17 :
e r
op i
(
m s cut O y m(
u i:i' l
ie a a e p s t ih m m S
i B i n T mm u ) _
r p0 0 = 0 -
o 0 i;i' 0l 0 t =
n (1 (
1 1 a
t ln o _
c o o t
n) a C l ad ole e oo co R f- c oe
- r so sc o(
L
_ _ =
0 5 g 5 O 1 g 2
y 3 e 5 $ 8 n_ E e E $
O lll!Ifll
F Hypothesized Core Damage Configuration V at 150-160 Minutes
- =fE"h*5" W
!IE g g g g E E E 5 5 E_l_ _ _ _
yc ~
(
28 inlet -
~ ~ ~
f~
A1 inlet Q- >' s b \d l
- = =0
- 0: 0 : 0 _ 0
! 0 0 ,
B outlet 0
[0l0:0f0l0:0{,s' sF , ,o i -
k (S(1(f27
-Oxidized intact 7
I h j .' S 'e$1!)]$9 rods
, g g ]
eIIe ;
i fm echI 3
-High temperature f ;
,.Jt.
)
2 ,
l ;ri
~
yQ .
.!L . .
-Solidified crust 2 L f" I,L lil}h.- f56 near liquid level d!! V.;.r-MJM9 -Approximate { tnypr-Q a liquid level
- h. ll q -mg 4 r m El"!
*O e =o =Os=qw g
C 9 O +~Tuu ' m.
..- l
l 4 Hypothesized Core Damage Configuration at 173 Minutes ej J=lis^e e A-kg s 5 5 5 5 5 =
, 4 L hb NOM D ,
.c c_ _.
1 j Upper grid I f damage ( 28 inlet '
~ ~ '
~
- A )
A1 inlet { A
=
0 0; 0 : 0 =d j0 iO!
- 0 ! !- !$dl 3Is. j I B outlet 0 :o= Partial y 6 ! 0 ,
1 .
$0
[f ' ? I L,'fh[M.},L -High temperature h 2 ." jy y
'4 )
, gM3 x fuei rod remnants j L ,y! RfM J
) dlfj , i l, :
[ j dl v- h m ' 1 3 ! Partially molten ZrO2/UO2 N Solidified ZrO2/UO2 2 -.mc: :. {.s. j between fuel rods, g q,,q l.e., supporting ] N i l 1.' .. NDN-I 1 i
) N crust 4 1
2
.]S /pdD r 9-9 -
i Approximate liquid water r ggg 2 level ) f3: s%w, a o
=&&h>-
I s., i EUU r.nu
... O
O 1
~
5O
** D V'
z = @
**.I T @
a, w
\ @
L _ TF G ? ,. , h @ K: O _
@ \
O .D r-I
.J ~
my .
,w M. @
g l @ ( @ u o> 1 E r- i
, F8u '
O l
Hypothesized Core Damage Configuration (175-180 Minutes) i
- %'"EE x 1_ 2 2 _5 5_ 5 g $
g g qq l 55 EM - q -
~'
p-Upper grid
- damage
,e A y
!f %
!/
~ '
~ '
l , l A1 inlet [0 / s" // / L 8
=0
= 0 =
0 : 0 J: 7" d B outlet _ E = , f g, -Fragmented g
,.' , I fuel rods
'S j ,-[ ,, -Possible upper
?h '
3
/ crust 3 j qs, 4 h,, --Partially
- ,, i
( 2 O [ [ ? L,"gx molten
~
g:fT[tj gg ZrO2/UO2 between Approximate d - .i i ' L 8 L' fuel rods liquid level - - c 4
- 4. .. > =. -%
l [ 2 j 1 1 ., i ( fi
) K Solidified ZrO2IUO 2 between fuel rods
> i @MN(if== q
-e-e 9p __ _-
I 4 a$$k7' o w C 6 4
i O_ d 8 r* o 1
"c' 5 Y=
4 2 o 0 Nd"t T S 0 3 b
*o2" *' *
"4 4
P 4' la p f 'fe v r m e P u - 3$ / t i n r R r hN}9 4 Nk/ u o P t f e gNx9DSQhNN t n o O C A )s gt e hNkNN$RNx\Qb' \NNxNp N nun $R\\N%EN M i l oi b%yNQN\NQ N h NhNxY L oM DhkNxNtNx h%\NNd K C l e71 5 _\. $9\\Nk6N \N\xNQS
- H O p( G ut on
%4NQ\\NQhxNhNx@ F ce /
MNkNxN+Xx\hQ\N$\t os k\[NNk6 NhN NxNQM' i E mn k rar D eT ibRQNx@h$ [kh\h\ h T
$, $ jb C
e r
% B A
o C-I n 5 4 3 2 1 0 9 8 7 6 5 4 3 2 l - 1 1 1 2 1 1 O 1 l l l , l! l ll !
Hypothesized Core Damage Configuration (224 Minutes) 9 a ggg - g
= = = s_ s
~
sss5
@ N 5 -
n- NE _, q l j
,c -
Upper grid ; damage ( /th Y ~
, \
2Binlet
~
~
// A1 inlet
\k -
s M/ U l j'
= : 0 : 0 : 0 =
d/ d
; 0 ;&
;' 0 l0:0 & 0 ,
t 0 0 ; 0 $ j B outlet , oyo:g'@o s g o , y oid V i j - Upper debris a h "
,,.3,.
3 , '. ,. bed h Approximate - (j < ' $.,;g liquid level- .
-Upper crust
? ,
'.v.7,seg
i t w : 1 2 "t 4lj.g' Molten metal .
' i{,g::.
~
and ceramic - -'
~
.g. .
2 h ; h T l l N
\ olidified S
r[ Eff?.c b ZrO2IUO2 f 4dnl nnM=h s < T;,, between
, rting 4
u
'h s9e_
n, luu 9 7.im P 0
e 4 ) i s 2 p e 2 s 0 a t e a 0 3 r c n r e ( i n e t s e r o e t a s n s a t r u p o i e n t pt e o a a r c r e t a su en i n m e p d . e g m RiM r u n e m r t s a t a n s r g a e e e e p r c l
- D r
m S u d o N r u C o C P S t R S s . . I n ' 2 3 4
. O ..
o - en 8 r o oit 0 2 m gCar - m
\ ig 5 =.
l 4
, M l R 2 0
_$ l P O
,( l N t e T S
l M a r 0 3 c e l l 4 P 6 l l K im m o H a l G r lg l e g - w l F E C A o l o l e @ R-t l e I A n f' g,,f h f - 2_1 i o _s i t c ,O e - - j _ _ - - _ - ~ s R S i d o P i r O s i i e u s f g N n m o Mu i n d e id lo r i b n L C a C 6 t ly i is r - ' ;? - S Kb d b K , m i io e He s e r i V d r e p fg n y GA s o p s d id i U d F o d oi R o r C i i lc d e r l f f e E le p D F u
- p io C
i
- o m e
s i p i k , k ,; d i t B d o
'ds 3' n D k
2 i
"ir g
_= A
- - - - - - - - = -
o 0 4 3 2 1 0 0 0 0 1 2 epa ok 5 1 1 1 0 90 80 706 05 04 0 1 g $o g$ j SB< gg 3 20 0 d
O Hypothesized Core Damage Configuration y (226 Minutes) MW
'g3 355 5 5 2 s 5 5 5
= -
' 4- a L
c e
.r \ _
rm x
' N (
L i
. : = 0: 0 0 : 0 0 d
- 0
~
~
10 0 0 0 = 0 f0 , Enlarged S :?:S- k
,( / void
.: 0 3: -i-3%,.t 24 . J - , t t
~
O . l ;;g ; ,f",s . c' 0:(iig/-/ l0, -upper bed debris
,i Approximate water level -
ii,f'*#N},f ib N upper crust Crust failure 2 4%J near southeast y [ i.
]
core _/ :. . . j > periphery jf ,7 .
) ...
)., . .,
[
.: , .: : t j e {
Probable ' 7 1 -
~ ' '
h
"' [ Q
~
reloca ../ k ... ;
~
Lower plenum route . debris I aj.
~
kbf, - .. J$l.......%' D"
.,. /4' l Falled instrument b '~ht S'M' ' I N,Y,, n a. 7..
~
structure ..v m .w- , u IIou' 7 1245 l l l
g 3 0 3 4 2 n 0 2i o T S t 0 I a x xxx x 2 4 Mm To r t r o P y f n p s t I n n t cut ar a i d d o r n t pa r e ne o sa c6 n5 n e i s sl e g Fr e i U 09 o r d f - y n O oG h n i t o sE d o sn a z aR Sn a i t ci o i r C o eU t y Rn o jc et n ea oma p ArN i n a is ni n se l t set n ri t r v esoe uf r r t r or nmibl o e c ig t aor i o ema cc t j a n l up set oiu orure cmd U c e o l npsc M f i r t oxsn nhpe r o cleiee o c ama mpcgoe t r a l n e r r
- o r
tuer ne aher rdisni A aoetie g o yd s N C gS H C H l oiF O a *
- o * ***
ee rt r 2 2 5 4
- e oa )
I 2 yd cw P H m a s 0 T i r g y e S l nne d e bn s o amo l t 0 e s er i r a u ee 3 hms t u n oe nr du e d pl o e 4 P r r i i f t u wne o n o f o ss on no rt pt c elpe cei eu t c f s l c ee nmtc bd c brn i se e d i r ina dr o a n ep l t ir sps a nta e lup p ope pur ( t bme t e s oi nn t o n ge p len se nt e ce hosis I mt c itnhn a t i sd ei c es r y r us r po e e t r t l ese ufid sl pleo e s v c ch be c f n o d m__m et u r r a o hg s n mim da oc n l i t t e t ih l o r egc ch ue- ort a r o aur cmt at ed r
- w. t t
nn a olee wrt ro r f f o ed p as e scc ses t s sd e tr i nt r n a eo ti ss Cer n at n e ur na tdo a dn dte ler o i l ei ios t u s i ctep dmu nis ip t Wt arm e np ox Ce l If a o r o l m ucn Raa cd e e I l R t s a n m w a e m u a e e r s i n i c r o n e t s ol c ar i d e f c lp c i ne t i n i f rd i sa a n g o ee wlo t e g Cm t e s % r dn r t 5 loo e e i l c n ne S u s e 3 y i n y sl l e o I at lo b oi s _ l a t r R e i n f m b t a r b tu s o dt s r e im de en a no nn oa a n ad e c xd n v wis i t si c pi or e e e eo ni f pll a pl t t s rl ep er tbn i at pe loa eeg u e hx Am Mw Te Rds i Rm e
,l f f lll{ l i1 Iu
,l 3
O 2 5 4
- 2 f l o o r
0 d er e t nd T S n r ol u oa ceh 0 3 a 4 caf i s P f od er s e s g n a n sw eo n i dno cl i o ni as t s o r e pht - t t c s r res eo at ht a egr e t . p vc d d e nrt ou g m ra ep i i upc cer t gt u n a a sr am t , s sa s ed bm i l amV eaP dg n n I RdR Ul i ~ o Ol gi a t inn d ete l l e l an ae z i r eh e z f uo v a on f o t et l a ore eP h l s gn n ol l mmw d no D i i ai t t a n el rl e eraw a ed t a mter atwn l t n a o n e e s n i t a mde em pr mbl r a e v r eit r omtn f o hn e sir e c e s ci e er l t eia ot t e O b ndmo e e uc bcm l i a R l t sr o utd a s n sr ser e ot o Mcian Picn ( O 1 l
i Reactor Sampling and Examination Status l x.,),
# I/ / / A Adequately sampled and examined E~E 2
=>
's ;5DE 5 5 5 5 5 55;
= g = = 'Ju-y 5 = m i Control rod J4 f . ._ ,; _ , _ u _ ;]_ . . .c- ~
leadscrews
> _:7 /
( / / ' Upper end fittings
..- /
/ and control rod i
/ / ,/ spiders a / / ,/ ' . %
.e y / / -
/
/-
/
//
i ' / >'3 / (' d L ,
} 0 fD Ejk o /
40 0 0 , i'j)E*, 0 0 k . 23 b k
%/g\:\y\}k)Ar " ""
0f- 1 Peripheral .ff d j.l
, ,. i fuel rods %
N- j,ly,j .,r',./ / f,h+. '* Previously
- Relocation ,) ' g, . .
< .,f , -
Jg
)x ,,.
- ,,,,, molten core materials gg j route -
I . p ueF rod M .* .. v ., - j
;#@ll .'folf, 2
i g- u-5 Lower plenum 11 _ q debris s 3...:!:'
- 4 g _ _=oggenoTb o /
hd{$y o Molten core .. f . materials and \ ' ligF k \ .- /fj. .d'. structures interactions - %h~ ~ . m.,,W..j& e (), I . g, - Blocked instrument String 7,g
o _ 4 2 5 4 2 l 0 i a T S r t e y t 0 3 4 . a a n P n mm w d e i u e n n m h a p l e nr a eo o pct sm t y i u l r l e w f er o v m d reehiz nee t n e e hacr oswl ri t sr u v d o rg l o blo r e ombs es i n i a o r tdnedu e t c r a n er n omen t r p u et r o sP e c uwareds
,nt epn e d
o r c o s p _ Rg n n se er pat el but u rl a n e nr pr t e ol p el git ip e cuGt opoio pd a . t ame n rSagaspci ri ain r ns i nawzs ot r nnils cf _ i nx i ms o ei mnkef an ahs _ iEa ael x rdief nt o noer cc gastf eb eiei t l m e o ntcms bt e hl u l e e ba auimr npsp e p l pr e s etrb aevdmbm r r i l d R m odC o tn set n D e S ociou c c C o a pC n a rI i S - - - - A- - e
- _
o
5 0 m 2 n0 r e n 5 i o3 el tb o 4 2 0 t a i t T ao ei l a S _ i ct t ca r v r e 0 3 4
) f4 i
l a71 r o ua o n e P nd u p sht g h e qm i et r r dw n i t u do r l ap o n e i n nf ner a e gt wit a n i f t nib l or r o h n n o dy es dp c o i oi t n a a r ds io hn ys rc t a cl u m c c p dr a a( u dl c e r m u us at nt em s a ea r c t c ee r t n ad nn er an un e ve e x oa t ao tr o se i Rg dt i sit se ss gr o a a ea dr t e si c dt h n gl r e nP ns an ba ai f l e vo ir oo e r i e o o (t l ign sm ta q u ss t c pt c an i ce a ee r e v gd eu mi t rd e dat nu u s n cmp o ao e a ei s r xe s et ese td n i a RxE iptsl psel h t oa l c l adn r uei o smumi y n l b i cn) t r s ahaaos a oiou l i nCmCP pctNCF mtd ea A- - - i a o (ll1,ll i
{ll l a
,v 6 2
5 4 2 t 0 n s T e e t S d i s s iec t ngu 0 3 i c t n 4 P c nio iad a a et da n art mo r e i eap r ci ecd e cm vn n a u eo v e aa h gos r i s 2xe - in csf l l i d e I Mlpe witstd o s t a T mn hi cr exde na n g e ao i o i i t m hst t i a a gt as e nl n s r o mtn sieo evri o l u a s f o f o r e m rl 2s5l o6o s c a i r su I e9 r px n w al t r t Mn 0etor o t n nus esn ToGmp C d e cei s r e etE hcRen as i c dti hn t aUt p a c f omNsr l ei- ,t a t a wno s innd 2 l t ion u n t t ndia s sem u aese l I t M ot o s cie sr ei f s T pif s rf i ir nngat e end hon e ge or ee l h T Tca hid Tsipr a o o O
,llf !l llI!l llll1f}}