ML20247Q715
| ML20247Q715 | |
| Person / Time | |
|---|---|
| Issue date: | 05/23/1989 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1739, NUDOCS 8906070020 | |
| Download: ML20247Q715 (431) | |
Text
- - _ - - - _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ __ _ . _ _ _ _ _ _ _ _ _ _ _
\
. m b<m'lA_
. 3 O
UNITED STATES I NUCLEAR REGULATORY COMMISSION i ADVISORY COleCITTEE ON REACTOR SAFEGUARDS I
In the Matter of:
JOINT SUBCOMMITTEE ON THERMAL HYDRAULIC PHENOMENA AND CORE PERFORMANCE O .
5 1
, .)
Pages: 1 through 284
~
Place: Bethesda, Mary. O p g7- ~~
cate: May 23, 1989 .
- VI. Og-l,.E,4Wg ;
.- ----------------------------------------- ...------ ?O$8 Uf6 Of the C0ilWi ;
O HERITAGE REPORTING CORPORATION l OficialReporters j 1220 L Street, N.W., Suite 600 j e906070020 090523 Washington, D.C. 20005 PNV f$739 r (202) 628-4888 i
' i b
, 1
> ( UNITED-STATES NUCLEAR REGULATORY' COMMISSION- .!
!Sdc,'50T&I--
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS-In the Matter of:
JOINT SUBCOMMITTEE ON THERMAL HYDRAULIC PHENOMENA'AND CORE PERFORMANCE' Tuesday, May 23, 1989 Bethesda,.-Maryland
.The above-entitled c.atter came on for hearing, pursuant to notice, at 8:30 a.m.
PARTICIPANTS:
Mr. Ward, Chairman Mr. Catton Mr.-Kerr Mr. Lee Mr. Lipinski Mr. Plesset
, Mr. Schrock Mr. Sheron ,
1 Mr. Shotkin Mr. Wylie p p
O ** 9- e rei 9 career tie-(202) 628-4888
= _ _ - _ _ - _ _ _ _ _ .
D.
2' 1 PROCEEDINGS I
2 MR. WARD: The meeting will now come to order.
o 3 This is a meeting of the Advisory Committee in 4 Reactor Safeguards,. Joint Subcommittee on Thermal Hydraulic 5 Phenomena and Core Performance.
6- I'm David Ward, Chairman of Thermal Hydraulic 1 7 Phenomena Subcommittee. Bill Kerr is Chairman of the Core 1 8' Performance Subcommittee. He is expected to arrive later 9 this morning.
10 Other ACRS members in attendance are Mr. Wylie and 11 Mr. Catton. And we have ACRS consultant, Mr. Lipinski, Mr.
12 Lee, Mr. Plesset and Mr. Schrock.
11 3 The purpose of the meeting is to discuss: first,
{} 14 this morning, the NRC's research office, thermal hydraulic 15 research program plan as documented in NUREG-1252 and in a 16 proposed SECY paper; and second, the status of the ongoing 17 effort to address implications of core power oscillations 18 that occurred at LaSalle Unit 2 about a year ago.
19 Mr. Paul Boehnert is a cognizant ACRS staff member 20 for the meeting. Rules for participation in the meeting 21 were announced as part of the notice of the meeting, 22 published in the Federal Register on May 10, 1989.
23 A transcript is being kept and will be made 24 available as stated in that notice. Roquest that each 25 speaker identify herself or himself and speak with Heritage Reporting Corporation (202) 628-4888
3 (f 1 . sufficient clarity and volume so that she.or he can be 2 readily heard.
(
3 We're received no written comments nor requests to j
)
4 make oral statements for members of the public. l 5 First, a couple words to the Subcommittee on the 6 research program.
7- As you recall, about a year ago the Subcommittee, 8 and then the full Committee, reviewed fairly comprehensively 9 the plans for.the thermal hydraulic research program; and we 10 made some, again, fairly' comprehensive comments on that in 11 the ACRS letter of last June, I believe it was.
12 The Staff has responded to our comments, and I 13 think they disagree with a few of them and agree with
{} 14 others. That letter and the Staff's response and Mr.
15 Boehnert's summary of the responses is in the package of 16 material that was attached to the Meeting Summary that 17 Boehnert provided.
18 I'd suggest as we listen to the discussion this 19 morning of the plans for the continued thermal hydraulic 20 research program that you take a look at that if you haven't l 21 already and consider what you're hearing in light of the 22 review and comments that the ACRS made last year.
23 Do any of the other members have anything they 24 would like to say at this tima before we go to the agenda?
25 Any of the consultants?
Heritage Reporting Corporation
.h (202) 628-4888
4 m
() 1 Let's see, we'll be covering, of course, the BWR 2 core power oscillation this afternoon. And I know Mr.
3 Lipinski and Mr. Lee are here primarily for that, but we'll 4 appreciate your comments and input to what you hear about 5 the more general research program this morning, too.
6 Okay, Lou, are you going to start or Brian?
7 MR. SHERON: My name is Brian Sheron from the 8 Office of Research. I'll give you a brief introduction on 9 where we're going with the thermal hydraulic research 10 program, maybe a little perspective of why we're going 11 there.
12 Let me give you a brief history. This is probably 13 nothing new to anybody, maybe just for emphasis, that we've
.r's 14 been basically developing thermal hydraulic systems codes
\)
15 for at least 15 years, probably starting with the 16 promulgation of the ECCS rule. Even before that there were 17 codes, I guess, the flash code and stuff which were the 18 origins of RELAP and the like.
19 The code development progressed in, I guess, in 20 almost in order of our understanding of risk. Back in the 21 early 70's, I think, the Large Break LOCA was the principal 22 het topic in the regulator's agenda. As you know, there l 23 were extensive ECCS hearings, questions on whether the ECCS 24 systems would, in fact, provide protection against the 25 postulated double and the guillotine break.
Heritage Reporting Corporation
(-)> (202) 628-4888
5
() 1 I guess it was around 1974-75 that WASH-1400 came 2 out. And towards the latter half of the seventies, I think 3 people had started sinking in the concept of risk, that the 4 Large Break LOCA was in fact not dominating the risk to the 5 health and safety of the public, but it was other events.
6 Actually, even before TMI, back when I was in the 7 Reactor Systems Branch under Dr. Rustozi, we actually 8 started to put more emphasis on the small breaks. And TMI 9 came along, which was in essence a small break accident, and 10 that sort of reinforced the emphasis. And we spent a lot of 11 time and money on looking at the small problem.
12 One of the interesting things was that the codes, 13 although they were designed for large breaks, were in fact
{} 14 15 flexible enough so that they could be revised to accommodate the small break. Also, following TMI in the early '80's, 16 there were a number of events, transients. I remember 17 Crystal River dumped 20,000 gallons of water on the floor 18 through a stuck open, I guess not through stuck open put 19 through a PORV. There was a steam generator tube rupture, I 20 believe at GINNA, which caused the operators to trip the 21 pumps. And there was just a general emphasis on transients 22 all of a sudden, not your classical Chapter 14 SAR 23 transients, but what I would call them more prcbable, i 24 realistic transients, things that were likely to happen.
l 25 And we spent a fair amount of time studying those.
l
[
l Heritage Reporting Corporation l
O- (202) 628-4888 l
1 l
6'
( 1 We studied all of these in all of the thermal-hydraulic 2 facilities that were available to us, primarily in LOFT, in l 3 semiscale, in the FIST facility, TLTA, which was its-4 predecessor for the GE plants, European facilities, Japanese l 5 facilities in the 2D/3D program.
6 We tried to get very, very strong experimental 1
7 data base that'would cover all of these, and were pretty.
I 8 much through with the exception of a few specific items
- 9. which you'll-hear about later today, the B&W plants, the-10 steam generator performance. And we are now finishing up 11 the UPTF experiments over in Germany.
l 12 In conjunction with the experimental program, as 13 you know, we were also continually developing the computer 14 codes. These were primarily the TRAC code and the RELAP 5
}
15 code now.
16 We've gone through, and you've all heard this, I 17 guess, the scaling study that Dr. Zuber did, which was done 18 to support changing the ECCS rule.
I'll skip down a couple bullets here; but it, in 20 fact, did quantify or estimate, I would say, the uncertainty 21 for the Large Break LOCA. We're planning on continuing that 22 and quantifying the uncertainty on the Small Break LOCA to 23 demonstrate the applicability of this methodology.
24 Dut I think if you look at the codes today, and 25 you look at the degree of maturity, their ability to i.
i l-1
(] Heritage Reporting (202) 620-4888 Corporation
)
( y
'/ "% .
1 calculate these events, I think we've generally concluded
( ) ..
2 that they have reached an acceptable level of accuracy and 3- maturity.
4 And I have to temper that in the sense that we're 5 not saying that the codes are perfect, they can calculate 6 everything every time out to' five decimal places. What we 7 have to do is we have.to say,'given that we have limited 8 resources within the Office of Research, where can I get my 9 biggest payoff? The real question is, how much more money 10 should I spend refining these codes, versus looking at other 11 . areas of research interest like human factors'and severe 12 accidents and the like.
13 So you.have to remember that when I make this 14 statement, it's made in the context'of looking at an overall 15 research program in terms of how much more money should I 16 put in this area versus other areas. And given that we 17 have been developing these codes for 15 years, I guess as I.
j 18 say, I don't hear people screaming, saying these codes are 19 totally unacceptable and can't calculate everything. In j 20 fact I think they do a very good job.
21 So anyway, that's really what has brought us to l 22 where we are today.
23 I would also point out that -- remember, too, you 24 have to ask yourself how we use the codes. And basically 25 it's to predict stylized accidents. That's how we've used !
O rie 9- a ro ei 2 c ver tia=
(202) 628 4888 ,
8 b .1 them in the past; that's how we develop them. We had the 2 accident; we postulated.the double and the guillotine break, 3 and then we built these codes to predict that break. We 4 designed all of our experiments around that particular 5 accident.
6' What we've learned obviously is that in real life, 7 we don't usually see the stylized accidents. Sometimes we 8 do, but.for the most part, these are complex accidents.
9 It's a combination of equipment failures and human error.
10 If anybody said, "I could predict the probability that each 11 one of those different events and the sequence was going to 12 happen," what you do is you wind up with an event that's ten 13 to minus 37 or something.
14 I think there was one event; that was the Davis-
-{ )
15 Besse. I think we got a question from Congressman Markey 16 who was trying to make-a point. He said, "What is the 17 probability that the Davis-Besse event would have occurred?"
18 The objective obviously was, if you go through every error 19 that was made'along the way, or every failure, and you put a 20 probability number on it then you multiply them all i 21 together, you'd get a very, very small number. And then if 22 somebody wanted to make a point, they could say, "See, PRA's 23 are not very good because obviously your PRA methodology 24 would have said that this had a probability of 10 to the I
25 minus 16th, and it happened."
Heritage Reporting Corporation (202) 628-4888
~9
,- ~
i 1 . Well, in fact, I think that you would say it may-2 have had an individual probability of 10 to the minus 16th, 3 but there may have been 10 to the 14th ways that you could 4 have achieved that end result, which was a loss of all feed-5 water. All individual scenarios, each one having 10 to the 6 minus 16th probability, so that when you' add them all up, 7 you still get 10 to the minus 2 or 10 to the minus 3.
8 The point here is that when we ask ourselves, "How 9 well can we calculate the stylized accident?", I really 10 don't get excited about it because that's not telling me how 11 well I'm going to calculate what's going to happen in real l 12 life.
13 What's going to happen in real life is what you'll
{} 14 15 hear later; it's like the LaSalle event, something we had
.never predicted, something where it was not a design bases 16 accident and the like, but it happened.
17 So when we ask ourselves how well do these codes 18 have to do, we always have to keep in mind that, even though 19 they can't predict the stylized accident very well, the one 20 that we really worry about is probably one we haven't 21 thought of. And I guess my feeling is that once -- the way 22 we have confidence that these plants can perform acceptably, 23 is when we know that the codes can calculate reasonably well 24 .he stylized accidents.
25 We gain from that a confidence that, because we Beritage Reporting Corporation (202) 628-4888
'10
. l' understand the stylized accidents,;..we have'this warm feeling 2 that,there are other events,'we might.not be able to' predict 3 them, but we think that the plant's going to be able to 4 handle-them. Juni it's based on our- understanding that the 5 plant can handle the ones that we have challenged it with. L
')
6 MR. MARD: You know, Brian,'I think, the Committee 7 'came to maybe a slightly different-conclusion about the 8 ' status of codes and the need for the utility of further:
I 9 development last year. I'd characterize it more as --.it's. ;
i 10 not so much that the codes had reached a state of j 11- development or a state of accuracy that they.were good 12 Lenough for all the uses you had for them, but rather that it R
.) '
13 didn't.look like further development was going to make them 14 much better than they are. There wasn't real much potential
{}
- 15 for reaching a new plateau of completeness and accuracy and
. 1 16 so-forth in the codes. And so there wasn't much more point, 17 there wasn't any point in continuing to develop them. But ]
l 18 that sort of means that the uses to which you:put the codes, !
l
.J 19 you have to keep that in mind.
l 20 MR. SHERON: Yes.
21 MR. WARD: I don't know whether I'm saying -- I !
l z2 think that's a little bit different perspective than what -- j 23 MR. SHERON: Let me get on a little bit, because I 24 ara not advocating that we kind of close up shop and put the 25 chain and lock on the door and be done with it. That's not a
Beritage Reporting Corporation (202) 628-4888 L
b _ _ . . . _ _ _ . _ _ . _ . _ _ _ . _ . . _ _ _ . _ _ _ _ _ _ _ _ _ . _ _ . _
-4
11
() I what the program is heading towards. It's basically a 2 reduction in level, if that's the way to look at it.
3 I would point out, I agree with you 100 percent.
4 I came to the Office of Research having spent 11 years NRR.
5 I was told by Mr. Stello that one of the reasons they moved 6 me to Research was to try and bring, I guess, some 7 regulatory flavor, some thinking, to the office.
8 I've tried to do that in the sense, you know I 9 guess my feeling is that the main -- our job is to support 10 the main mission of the Agency which is to regulate 11 commercial nuclear power. I really don't look upon us as a 12 pure research agency that is supposed to be doing pure 13 research just because we should be investigating things for 14 their own sake.
(-)s 15 Our job is to do what is necessary for this Agency 16 to carry out its function. And I think anything beyond that 17 is really -- should be left maybe to other agencies.
18 I ask my staff for every program that is under my 19 direction. I ask them usually about three questions when I 20 get the proposed contract, for example, to go out, and that 21 is: "Who wants this data that you're going to get from this 22 program?" Second, is: "What are you going to do with this 23 information once you get it? Is it going on a NUREG report 24 and sit on a shelf? Does NRR want it? How is it going to 25 be used?"
l l
I' Heritage Reporting Corporation
\-} (202) 628-4888 L________.
12
) ll' And'if they don't have an answer to that, then I 2 usually don't sign a piece of paper, because I don't want to 3 do research just because - you know, if people don't think 4 ahead on what they're doing the research for,. what is it 5 they're trying to solve? What is the problem? And then I 6 also ask: "How likely is it that I'm going to get results 7 that I can even use?" These are tough questions that I ask-8 my staff. For those programs that pass basically that test,.
9 those are the ones that we'll go forward with.
10 And so I apply those not only like the thermal-11 hydraulic research, but to the other programs as well. And 12 this gets into the question of how likely are the results 13 going to change my perception of safety? I think that's the
{} 14 15 biggest one in thermal-hydraulics, and that is, is more code development likely to change my perception on how nuclear 16 plants perform or behave? Right now I guess I agree,-and 17 the answer is, probably not very likely that there's going 18 to be a big change.
19 MR. CATTON: At least not with the codes that you 20 have. I have a little different historical perspective than 21 you do, I think, Brian. Those codes were developed for fast 22 transients, mainly the Large Break LOCA, and their internals 23 are -- that's two fluid modeling and it's based on weak 24 constitutive relations. As a result, those codes are'not 25 going to change my perception about how nuclear power plants e ti=9 c re r ti -
O- rie 9-(202) 628-4888
13
(_). 1 operate.
2 You got the wrong sec of codes. It's been going 3 down the wrong path for -- well, not necessarily the wrong 4 path -- but the Large Break LOCA question has been 5 addressed, and I think that's the conclusion the CSAU came 6 to, not that the codes were any good, but through a 7 combination of tuning and adjusting and whatever, they are 8 able to address the problem that was their charter.
9 But they're not the codes you ought to be using 10 for transients or Small Break LOCA or anything else.
11 MR. SHERON: Are you basically saying that these 12 codes are no good for any of this, and we don't --
13 MR. CATTON: No, no. It's just they're not the 14 proper code. When you have fast transients, you write a
~}
15 code in a particular way. If you have a different class, 16 you do it differently.
17 MR. SHERON: But I don't understand what's 18 unacceptable about the codes that exist. They've been 19 verified against a spectrum of transients. They've been 20 shown their ability to calculate them against experimental 21 data. It's one thing tc say that, but it's another thing to 22 provide evidence that says that they're deficient. And I 23 guess that's what I'm asking for. Is there evidence that 24 they're deficient in their ability to calculate these 25 transients?
Heritage Reporting Corporation (202) 628-4888
14
,e w
(,) 1 MR. CATTON: Well, I get very nervous when I look 2 at the internals of these codes and find inconsistencies.
3 And I can cite a number of those if you want. Whenever one 4 of these incidents comes up, I can recollect Davis-Besse, 5 when calculations that were carried to Congress had to be 6 done by hand, because the codes couldn't be made to work 7 soon enough. They're big; they're bulky; they're not '
8 appropriate for the kd.nd of thing I think you have to deal 9 with in the future.
10 You need a different kind of code. You 'eed a 11 code like the one that they have at Brookhaven: simple, 12 clean, understandable.
13 MR. SHERON: And hard wired.
'} 14 MR. CATTON: Well, it doesn't have to be. That 15 just happened to be his choice.
16 MR. SHERON: Well, the minute you go away from 17 hard wiring, you're into the complexity again, and the time 18 consuming it takes to, set up a deck.
19 MR. CATTON: I think that anytime that you put 20 together a code that exceeds your understanding of a 21 phenomena that mak.es it up, you're in trouble. You have to 22 rationalize it every time you use it. There should be some 23 sort of consistency between your understanding of the 24 phenomena and what's built into the code. And with these 25 codes today, tnat's not the case.
() "
Heritage Reporting Corporation (202) 628-4888
15
'T 1 MR. SHERON: I apologize, but I don't share your (V
2 concern on this because I think that what you're advocating 3 is a whole new code development.
4 MR. CATTON: No, I'm not.
5 MR. SHERON: I'm not sure than what your - .maybe 6 I don't understand then what you're advocating in this.
7 MR. CATTON: Well, in the 1974-75, when this 8 program was started, there were admonitions from every 9 . direction that you need balance. One of the things that you.
10 need to do is to build on information that's Lmternal to l
11 these codes, and that was never done.
12 You have a code whose structure may be terrific, 13 whose numerics may be second to none, but it's filled with
-w 14 things that are weak. I guess that's what I'm proposing is 15 that this whole arena has just never been looked at. For 16 whatever reason, research shows not to do that kind of 17 thing. As a result, a lot of the internals of these codes 18 are just somebody's perception of what the physics are.
19 MR. SHERON: Wouldn't you think that that would 20 become obvious when one goes through the verification 21 process.
22 MR. CATTON: I would have hoped so, but it didn't 23 seem to do that. As a result, you wind up with things like, 24 you could pick -- I can give you an example. Some of the 25 SETF or CCTF runs where you can't get the word fraction
(')
k>
Heritage Reporting Corporation (202) 628-4888 l
)
16
~I) 1 right, you get the temperatures right. The reason is that 2 the heat transfer package got boxed around because the word 3 fraction was wrong, and the approach was never to go into 4 the internals and get a consistent set of correlations and 5 data. That was never done.
6 And there are other examples like this. Even when 7 you tune thinas like interfacial drive, factors of 10, and 8 going in and changing a physical primary factor of 10, 9 that's obscene in the outside world. Yet, the code seems to 10 predict the temperature right.
11 I can only come to one conclusion, and that it's ;
12 highly tuned to that particular system. That's okay if 13 that's the only system you're ever going to look at. But if
{} 14 you start talking about your advanced reactors, questions 15 come up like the steam generator tube rupture at North Anna.
16 Westinghouse uses potential flow to get the cross-flow 17 velocitics. I don't think anybody in this room would accept 18 that. Yet TRAC or RELAP 5 or any code you have can't really 19 address that question. You don't know what the damping is 20 that results from steam. I mean, there's conflicts when you 21 read the literature. Some isay that it enhances damping, 22 some say it doesn't. Maybe that's not a safety question.
23 If it's not, then you don't need to worry about it.
24 MR. SHERON: Well, I guess, you know -- I mean I'm 25 sure you can sit there and point to deficiencies in the
() Heritage Reporting Corporation (202) 628-4888
l 17 J f( l' codes all day; I'm sure'I could, too. Again,'it's a
- 2 ' question of, is there something that is causing us to '
'3 rethink and to believe'that perhaps there's a deficiency l
4 that's fatal in this code, that's unable.to-calculate i .5 anything new and different.
6 MR. CATTON: I would be-very concerned about 7 things that are too new and different for these codes.
8 I'll give.you another example. If you were 9 dealing with either one of those two codes, and this room 10 was one.of your nodes and it had fog in it, the next time 11 step, it~would be water on the floor. That's just because 12 of the way that these things are built. That's incorrect' 13' and everybody knows it. Well, as.long-as you never have to 14 deal with a problem like that, you're.probably'all right.
- 15. Do you know that that's what you're going to be doing? I 16 don't think you do. You want a predictive tool. I'm not 17 sure how predictive your tool really is.
18 I'm taking to much time I think.
19 MR. SHERON: Well, okay. I understand.your l2 'O ' concern. I guess I don't feel as troubled about it.
21 MR. WARD: Yes. I'm not sure I do. I mean, are 22 you proposing -- what are you proposing, that a different p 23 direction for code development will work?
l 24 MR. CATTON: No, no. In part, yes, I guess. But 25 first you have a set of codes out there you spent a lot of Heritage Reporting Corporation (202) 628-4888
o!
18 L1 money on. I was shocked when I added up a few of those
.2 numbers ~from those bar graphs. 'let, we have codes that are-
-3' basedLon constitute of' relations. Those constitute of-4 relations.were guessed in- 1975-76'when they put the codes 5 .together. A lot of them have been shown over'and over not' 6 to be right.
7.- Now,'that's'not expensive research to fix those, 8 .but at least he ought to pay attention to'it. That's where 9 :I'm coming from.
10 The second thing is the:need for codes in the 11' future is the Small Break LOCA, the trunsients,. things like 12 this. The codes that they have now are not the kind of code 13 they should write. These are quasi steady processes for the 14 most part. You don't use a code like TRAC of RELAP 5 or
. O-15 anything even close to it for those kinds of. circumstances.
16 MR. SHERON: But the fact that we are. I mean, 17 you know, obviously maybe if we were to go back and start 18 all over again from scratch, we would write a different code-19 to handle the slower-transients. Given that for whatever 20 reasons, the agency chose to use these codes to modify the 21 existing codes to handle these events.
22 MR. SHOTKIN: Brian.
23 MR. SHERON: Yes.
24- MR. SHOTKIN: At the risk of trying to be helpful, 25 let me say that later on, we'll be able to tell you that the Heritage Reporting Corporation
( (202) 628-4888
1 i
19
()~ 1 final versions, final planned versions of the codes, have 2 done pretty much what Professor Catton has mentioned. We've 3 'gone through most of the fundamental correlations and
-4 changed them. Now, I don't know whether Profcssor Catton 5 would agree with all the changes,.but at least we've had an 6 international group of experts come up with ideas of what 7 should be the correct correlations to put in. And they have 8 been pretty much put into the final versions of the codes.
9 MR. SHERON: Just to finish what I was saying,
'10 now, and that is that given that we've gone this far, and 11 for whatever reasons we have these codes, they're being used 12 for the small break in the transients. The question is, are 13 they unacceptable?
14 MR. CATTON: I can't answer that question, really.
)
15 But I recall being at semiscale and watching an experiment 16 where they had two phase counter current stratified flow.
17 Your codes can't handle that; yet, your codes are supposed 18 to deal with small breaks.
19 Now, I always hear the argument, but it really 20 doesn't matter, that the bubbles are traveling upstream and 21 the water is traveling downstream. That ain't the way the 22 physics look. How important is it? I'm not sure anybody in 23 this room can answer that question. Yet, we're going to use 24- that code with incorrect physics. It appears to give the 25 right answer. And it's not that its difficult. Counter O a rie 9- eorei 9 (202) 628-4888 cerect tio-L __ ______ ___ ___ . _ _ _ _ . _ _ _ _ _ _ _ _ _ _
20 current stratified flow is a trivial problem.
)- 1- It's not that-2 difficult to deal with. You could do it on a sheet of 3 paper. Yet the codes don't have it. I don't understand 4 why.
5 MR. SHERON:- Let me ask a question though.
6 MR. SHOTKIN: I don't think that's correct, 7 but I don't --
8 MR. CATTON: They may have changed something in 9 the near future.
10 MR. SHOTKIN: No. We had correlations in there 11 for years for that, but I don't know whether it's worth 12 arguing this.
13 MR. SHERON: I would just point out, though, that 14- I recognize that these may be valid criticisms of our codes.
15 Do you think this is equally true of the industry's codes?
16 I mean, we license nuclear plants on industry calculations, 17 not to NRC calculations. And apparently, I guess, the ACRS 18 has found these plants acceptable, and I wonder if this is 19 something that you feel is of paramount safety importance or 20 what. I'm trying to put it into perspective because I've 21 always had this feeling NRC has been held to a whole 22 different set of standards than the industry when it comes 23 to scientific research and the like.
24 MR. CATTON: Maybe you ought to be, because you're 25 spending tax payers' money.
(\ Heritage Reporting Corporation (202) 628-4888 !
i l
_ - . _ _ _ _ _ - -- - _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - - - - - _ - - _ _ _ - _ - _ _ - - - _ - - - - _ _ - - - _ _ _ _ _ _ - _ _ J
21-() 1 MR. SHERON: But-the industry is responsible for 2 .the health and safety of the public.
3 MR. CATTON: That's true.
4 MR. SCHROCK: Could I add a sentence. I'm pretty
- 5. much in agreement, what Ivan has said. We've been over this' 6 ground so many times. We've identified specific things that 7 are just clearly incorrect in the code. And they still are 8 there today, and still the result is claimed to be okay.
9 Put that together with the fact that the codes 10 were oversold in their early stages of development as being 11 first principal codes. Many people out there who still 12 think that they are first principal codes, they use them out 13 of context, buy them too, all kinds of things, face
{) 14 15 reactors, whatever. It's been a case of overse11.
Now you got something which seems to be, in the view of the 16 regulators, satisfactory for whatever needs you have. Those 17 needs have never been made very clear to me. Because 10 sometimes we hear the codes are used by the regulators; at 19 other times, we hear the plants are not licensed by these 20 codes; plants are licensed by industry codes. What is the 21 objective for the codes if it's not to give the regulators a l 22 mechanism for assessing the safety of the plants. I mean, 23 it comes out double talk to me, Brian.
24 NR. SHERON: No. I mean, we -- I wrote a user 25 need letter to the Office of Research when I was at NRR in
()- Heritage Reporting Corporation (202) 628-4888
22
(~3 1 1984, which very clearly spelled what the regulators uses or u) 2 objectives were with these codes. And it was to do three 3 things: One was to audit license -- and as a matter of 4 fact, I think I got it. It's on the next U graph, as a 5 matter of fact. Let me just skip to that real quick.
6 Okay, A, B, and C. One is that we have the 7 capability to analyze operating reactor events. Two is that 8 we should have the ability to address licensing issues when 9 they come up; i. e., it could be design basis accidents or 10 calculations or whatever. And the third is to be able to do 11 the front end of severe accident sequences; namely, the risk 12 dominant sequences like station blackout.
13 MR. SCHROCK: To me it sounds contradictory to the 14 words that you used only a few moments ago in which you said 15 that we don't license plants, to these codes we license them 16 to industry codes 17 MR. SHERON: I said we license them, we don't 18 evaluate them. We evaluate the plants to whatever happens 19 to be on the Commission's plate at the time.
20 From the standpoint of licensing today, a plant 21 has to do a safety analysis report, and they have to do a 22 Chapter 15 which has all the stylized accidents and 23 transients. That's all they got to submit to get a license.
24 Whether that's the right set, I don't know. But in the 25 Commission's wisdom, it still is.
(~) Heritage Reporting Corporation k/ (202) 628-4888
n.
23 1 .That's what we license a plant to. When the-
]}
2 Commission'makes'a decision, it says this plant is safe ,
3- enough to be licensed. It's based on the safety analysis 4 report that's submitted. It's not based on all these.other 5 jillions of calculations.that we may have done in the past 6- and the like.
7 If we have a specific issue that comes up, let me 8 give you a' clear example of one. You'll hear about this 9 'later, so I'm really taking away some of Lou's thunder here.
10 One of them was that Tom Murley, and you may have heard 11 about it recently, has a concern about the interfacing 12 system LOCA. He's worried about it. We've had a couple of 13 events'near misses, I guess, the BIBLIS event, Vogtel had 14' one. And so he's asked the Office of Research to support 15 and. help NRR in addressing this generic concern.
16 We put together a program which is multifaceted.
17 It involves doing PRA's; it involves human factors. One of 18 the pieces is doing an analysis of an interfacing system 19 LOCA with the thermal-hydraulic codes. What we're trying to:
20 do is figure out, to get an estimate of how long an operator 21 has to respond and to perhaps take corrective action in the 22 event that there is interfacing system LOCA. What that will 23 involve is doing basically a small break analysis with the 24 codes, TRAC or RELAP, whichever one, I don't know which one 25 Lou is going to choose, for a small break in a pipe that's
/, ~'j Heritage Reporting Corporation v (202) 628-4888 i
E__.______________.__ _ . _ _ .__ _ l
24
] [ 1 going to have a?very long L over D, maybe 60 feet away.from 2 the primary: system.
3 How accurate do I have to be? I don't know. I-
'4 don't think I'have to.be'very-uccurate. The question is how 5 long.does an operator.have? Is it two hours? Is it 15 6 minutes? If it's two hours, I don't care if I have plus or:
7 minus 15 or 20 minutes on either.end of that. All I know is-8' that the operator has hours to act as opposed to minutes.
9 It gives us a feel.
10 So this is the type of thing that we use the codes 11 for. I'm not licensing those plants with that analysis.
12 I'm doing an assessment. It's going to be one piece of 13 information that's going to go into a big pot of information 14 that Tom Murley's going to have in front'of him. He's going' 15 to have the results of inspections. And regional inspectors 16 that are going to go out to these plants and they're going 17 to.look at the procedures that the operators operate these I < 18 ' plants under, and they're going to decide whether or not the 19 procedures are deficient, whether there's a chance an i 20 operator can go out there and open up two valves and cause 21 an interfacing system LOCA. He's going to look at the 1
l 22 information that research provides. He's going to put it 23 all together and he's going to decide what he wants to do, 24 whether he wants to fix the plants or whether he thinks 25 they're okay.
l- Heritage Reporting Corporation (202) 628-4888 !
1 _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ -
1 25
(_) 1 So that's basically how these analysis, these 2 calculations that we use are applied in the regulatory 3 process. They are one piece of a puzzle. We don't make a 4 decision based on one calculation or whatever. If it got j 5 down to the point whether or not a plant should operate 6 based on one NRC calculation, I'll guarantee, the first 7 thing we would be doing would be going to that plant with a 8 show cause order saying, "You do the calculation. You tell 9 me why you think your plant should operate if it's hanging 10 on the edge of a cliff."
11 So these analysis that we do today with these 12 codes are basically, they're scoping analyses. They're to 13 give us a better feel for how these plants respond. They're 14 designed to help us understand how severe a problem that we
(' }
15 really have. Is it a bad problem? Are we on the edge of a 16 cliff? Is there a big uncertainty?
17 If there's a big uncertainty with regard to an 18 analysis that puts them on the edge of a cliff, Bob Jones is 19 over there from NRR; he could probably reiterate this. But 20 my guess is that, if we found some deficiency in our code 21 that was causing us real grief, I'm sure the industry codes 22 would have it, and he would be the first one to be writing a 23 letter out to them, dragging them in and making them explain 24 about that deficiency.
1 25 All these codes are applied in the regulatory A Heritage Reporting
('~j Corporation (202) 628-4888
H p
26 1 1-' : process. .But like'I-said,.we don't make licensing decisions'
.2 based solely on one calculation or-a number-of calculations.
i:
3 It's just one piece of a number of-pieces of a bigger 4' puzzle.
5 MR, SCHROCK: Would the plants be-licensed without 1 6 this kind of supplementary analysis? would you accept the 7 industry codes as a basis for licensing.iflyou didn't have 8- these tools?
9 MR. SHERON: I'm going to ask Bob Jones to answer 10: that,-because'that's his responsibility in NRR.
11 MR. JONES: This is Bob Jones from NRR. .I think 12- the simple question in today's environment is, _yes, we 13 would. We would not need our own codes to license these O =' 14 plants. But the. simple answer'is not totally correct.
15- That's based on the fact that we have had this research'over 16 the years. We have a better understanding of how these 17 plants perform. We do have the thermal-hydraulic code 18 improvements, the experimental data, the industry efforts.
19 So it's that growth over the 15 years that give us the 20 ability to do it without using our own codes.
21 Brian mentioned, how does NRR use codes? In'all 22 honesty today,.we don't uca them very much. In fact, I 23 can't remember the last time I've really done a calculation 24 except, I guess -- I have to take that back; we did one for 25 the Westinghouse large break LOCA model. In that arena what Heritage Reporting Corporation
-O (202) 628-4888 1
o 27 j ) 1 we basically did was compare their results to what we get 2 with our codes, and try to understand what are the 3 differences.- Do they make sense? Is it because of model 4- improvements? Is it something we don't know?
5 Looking for major problems or deficiencies, we 6 don't do many audit calculations any more. We used to, not 7 that much any more. Our resources are now more applied to 8 going out looking at plants, doing inspections, looking at 9 AITs, IITs, to try to go and fix mostly hardware problems at 10 the plants. Thermal-hydraulic calculations are not where 11 our primary emphasis is today.
12 MR. SHERON: As a matter of fact, any calculations 13 usually that NRR needs done, I think for the most part, they 14 would ask'us to conduct them.
.O 15 MR. JONES: Generally, that's correct. New rules,-
16 exploratory research,.those are generally done by Research 17-18 MR. SHERON: Primarily, Los Alamo and Idaho.
19 Sandia doesn't. I guess they might do some when they run 20 these combination codes like MELPROG TRAC or something, and 21 then Brookhaven which does the BWR analysis force.
22 Let me move on, if I could here. Let me give you 23 the overall objectives real quick. You probably -- it's 24 been up here awhile; you've seen it.
25 One is we want to bring the major development -- I Heritage Reporting Corporation
.O (202) 628-4888
28
() 1 say major thermal-hydraulic development programs. This 2 basically where we're sitting here and cranking out new 3 versions of the code on a very frequent schedule to a 4 successful conclusion. What we've done to accomplish this, 5 is we told both Idaho and Los Alamo that the next version of 6 their codes, which is RELAP 5 MOD 3 and TRAC PF 1 MOD 2, 7 will for their purposes be the final version, the production 8 version. We've told them to incorporate into these codes 9 everything they ever felt was needed. In other words, it 10 was, this is your last shot right now. Go under the 11 assumption that there won't be a new code unless there's a 12 reason for it.
13 So they had delayed their issuance of these
(~% 14 versions of the code until they could get in all of the
(/
15 improvements and the light modifications that they had 16 originally wanted -- were probably saving up for future 17 versions. You'll_ hear more about this later, about the 18 release date on these codes and the final assessment. But 19 basically the approach would be is that we would not 20 authorize the development of a new version of the code 21 unless the laboratory or whoever could come in and explain 22 to us and convince us that the deficiencies in the code, for 23 whatever reason, were sufficient enough to warrant the 24 development and reissuence of a new version.
25 The second overall objective would be to maintain
()
Heritage Reporting Corporation (202) 628-4888 1
l
i 29
( )~ l a capability to do calculations, to do the analysis, to 2 support NRR at some minimum level. And I'll talk a little 3 bit more about that in a minute.
4 And then last is to take these codes that we've 5 developed over the past 15 plus years, and apply them to 6 reactor issues. I used the analogy, we build the Cadillac 7 and there's no sense now that we've built it, keeping it in 8 a garage polished. Let's take it out and drive.
9 MR. CATTON: Now that you've mentioned the 10 Cadillac, I'll quote a good friend of both of ours: "These 11 codes are like fancy Cadillacs with Yamaha engines." That's 12 a quote, Novak Zuber. "The codes are like Cadillacs with 13 Yamaha engines."
r~~3 14 MR. SHERON: Well, Novak is in Dubrovnik,
(_/
15 Yugoslavia, so he can't defend himself.
16 MR. CATTON: I happen to agree.
17 MR. SHERON: No comment. You're my friend and 18 Novak's my friend, and I don't like to disagree with my 19 friends.
20 Let me go through the specific objectives of our 21 program now. I think first and foremost is we want to 22 maintain a capability within the Agency for thermal-23 hydraulic analysis of light water reactors.
24 And as I said before, this would include being i
25 able to analyze operating reactor events such as the Davis-Heritage Reporting Corporation
(~') (202) 628-4888 l
1 g
I 30
- )h 1~ Besse,'Ginna, and you'll hear.more about later, the LaSalle 2 Levent. .
3 Licensing issues, questions come up. A licensee 4 -may ask NRR,- "We'd'like to change our diesel generator start'
'5 . time,." or "We'd like to change this valve opening. time," or,
- 13. something. "We want to reduce the size of our PORV,".or we.
i 7 make.it bigger, whatever it could be.
8 .But there may be issues that come up, and it may-9 be.necessary for the Staff to have to perform independent 10 ' assessments.:to see if we could. independently reach the same 11' conclusion on'the acceptability of that, that_the licensee
'i.
1:2 obviously had already made. So we want to have codes 13- available to do those type of calculations. And as I said?
f q 14 before, this,. what I call ~ generic'research, which is
{
15: basically making sure we understand in sufficient detail'the 16 front end of.some of the severe core accident, damage-type 17- accidents,. the risk significant events.
18 This covers basically the.whole' spectrum of E 19 anything a reactor would be likely to see. I don't know .
!: :20 what else there is; so it's pretty. encompassing.
21 Secondly, in order to carry out this, we need to 1
22 maintain a cadre of experts, both contractor and in-house 'i L 23 people, that can achieve this capability. What this means, ]
24 really is, if I just tell the. laboratories, for example:
25 "Okay, we're done developing codes. It's been nice. We 1
Heritage Reporting Corporation (202) 628-4888 I. '
31
,n
() I want to keep a couple of you guys sitting around twiddling 2 your thumbs until we need you."
3 If they're good, they are not going to sit around.
4 They're going to be gone; they' re going to be working on 5 Star Wars on something else.
6 So in order to keep this cadre of experts, we've 7 already asked -- I've asked the management of the labs what 8 did they think it would take to keep a good group together 9 interested in doing this work with sufficient interesting 10 work; that they would be available and be called upon to do 11 calculations when we need it, but would have interesting 12 work to do. This in itself to me implies there is always a 13 continuing minimum level of development going on.
{} 14 15 These guys are not going to be sitting around waiting for the staff to call them up and say, "Well, we 16 need this calculation today. When you're done, just hang in 17 there and we'll call you in a couple months for another one 18 or something."
19 What it means is that they should have a program 20 in which they could be doing development and improvement, 21 this small cadre of experts, which would be their sort of 22 like their base research. But that, when and if there was 23 need for analysis, they would be right there and be ready to 24 do it.
1 25 Basically, I'm keeping a set of mechanics working I
I
([ ) Heritage Reporting Corporation (202) 628-4888 ;
1 l
~-_ _-_____________________- _ _ _ _ - _ _ l
32 c'^' 1 on that Cadillac constantly, keeping it tuned up, keeping it
(-)
2 running, taking it out on a test track, adding bells and 3 whistles as necessary.
4 That's what that means.
5 MR. CATTON: Brian, I think you and I disagree on 6 the focus, not on the statements. I would feel a lot better 7 if that cadre of experts was trying to make that radiator 8 work better in that Cadillac -- not changing the radiator, 9 but making it work better. And you can go through, you can 10 pick areas, the way they deal with interfacial drag and the 11 number of criteria.
12 Somebody ought to be thinking about, what can we 13 you do about that? Not changing the Cadillac, but maybe 14 changing its generator to make it work a little bit better,
()
f-
/
15 to make it closer to physical reality than it is now.
16 And the number of people in your cadre of experts 17 wouldn't be any different, but the focus would be different.
18 It wouldn't be necessarily on code development. I don't 19 think you need a 3D two-step approach to solving these 20 equations when the pieces of the equations are not right.
21 You need somebody looking at that piece of the equation to 22 fix it. Make your Cadillac run smoother. Right now it's l 23 only running on two cylinders.
24 MR. SHERON: To me that's something that, yes, 25 these people could be doing. What I'm saying is I'm not Heritage Reporting Corporation
(]
(- (202) 628-4888
1 i
33 ]
() 1 2
looking for --
HR. WARD: You haven't defined yet what they're '
3 doing. I haven't --
4 MR. SHERON: No , I haven't.
5 MR. CATTON: They keep using the word, " code 6 development", and when you say " code development" it 7 has --
8 MR. WARD: I've been waiting to hear what he's
-9 going to say they're doing.
10 MR. SHERON: I'm not.
11- MR. WARD: You're not going to say it. Is Lou 12 -going to say it?
13- MR. SHERON: Are you going to say it, Lou?
14 MR. SHOTKIN: When I get up, I'll-say something 15 about it.
16 MR. WARD: Okay. You've given sort of a concept 17 here that you keep a cadre available to do item number 1, by 18 having them work on something, and you haven't told us what 19 that's going to be. And Ivan's objecting to it already, I 20 think, immediately.
21 MR. SHERON: He is suggesting things that they can 22 work on.
23 MR. CATTON: -- tighten up the bolts, doing b 24 analysis, and, essentially, playing with the code.
25 MR. SHERON: And I'm saying that I can certainly 1
1 Heritage Reporting Corporation (202) 628-4888 L__________________
1 34 1 accept what Ivan is-suggesting as a fruitful area that they- 1
( ).
2 could be certainly working on. .This would be something we 1r 3 would -- you know, I very much welcome these types of; 4 comments.
5 MR. CATTON:- I also think it would do something 6 else for you. By doing that, they would know what the 7 numbers mean. And one of the problems I have with code L 8 developers as a rule, is that any number is okay as long as l-9 I produce one.
10 MR. WARD: Okay, maybe we better -- but I hope 11 we're. coming back to this in a more --
12 MR. SHERON: We'll come back and --
l 13 MR. WARD: It's a little squishy right now.
l 14 MR. SHERON: -- we'll agree with Ivan, pretty 15 much.
16 MR. WARD: Okay.
17 MR. CATTON: What do I do now?
18 MR. WARD: I don't know.. Let him go ahead.
19 MR. SHERON: A third specific objective here, 20 which is really, again, coupled with this, is to maintain-l 21 the code development and research at a level, minimum level 22 necessary: A, to ensure that the codes are acceptable for 23 advanced LWR analyses. Now, you'll hear a little bit more 24 about this. We have a program underway right now at Idaho 25 that is designed to look at the advanced LWR's and to Beritage Reporting Corporation (202) 628-4888 l
! 35
(}
1 basically try and point out to us areas where they think the l 2 codes may not be properly verified or acceptable.
3 Secondly, we want to make sure that we have a l j
1 4 program in place such that new information that might come 5 nr, in the next several years, or even tomorrow, doesn't 6 invalidate our current understanding of the code accuracy 7 here. Keeping in mind -- you know, the Europeans are still 8 running some of their facilities; although I think probably 9 within several years, they'll be winding down their 10 facilities as well. But there's always the potential that 11 we will be getting new information, new data. We want to be 12 aware of it as it comes in, and we want to understand it.
13 And we want to make sure that there's nothing in it that 7s 14 tells us that we got a big mistake or something in our
(
15 codes.
16 MR. WARD: It seems to me you don't want to so 17 much make sure there isn't anything in it, but make sure 18 that if there is something in it that invalidates the I 19 current understanding, that you recognize it and do 20 something about it.
21 MR. SHERON: Yes.
22 MR. WARD: Isn't that really what you're saying?
23 That's not what B says. B says somehow you're going to 24 defend the current accuracy in the face of any new 25 information that happens to come up.
(~NI Heritage Reporting Corporation As (202) 628-4888
4 36 1 MR..SHERON: Well, what I'm saying is ensure new 2 -information does not invalidate the current understanding of
- 3. 3 code accuracy. What that means, that in order to ensure
~4 that, I may have to fix the code in order to make sure that' 5 it's accurate. Maybe I've just worded this wrong. But'I 6 think'--
17 MR. WARD: Oh, okay. To keep codes as accurate as 8 you think they are now, you might have to make some changes.
9 MR.'SHERON: Right. That's what I mean by 10 " ensure."
11 MR. WARD: All right.
I 12 MR. SHERON: " Ensure" means possibly changing the 13 codes if I have to.
14 MR. WARD: All right.
%/
15 'MR. SHERON: And then the last one is " achieve (2) 16 above", which is, as I said before, I'm not going to 17 maintain a bunch of superstars if I don't given them l 18 interesting, challenging work to do.
19 And my last slide is: we want to establish and 20 maintain low-cost, experimental capability at Universities 21 through construction and operation of scaled loops 22 representing major U.S. reactor types.
l 23 Several years ago, I guess it was in '84 or so 1
L 24 when research had first proposed to build the University of 25 Maryland loop -- maybe it was '83, I can't remember -- but I Heritage Reporting Corporation O- (202) 628-4888 w___-__-__________--__-_- _ _ _ _ _ _ - _ _ _ _
l 37 (q j 1 think the original plan was that maybe because all of these 2 major facilities like semiscale and LOFT and FIST are all 3 shutting down, maybe we should build a loop at every 4 university of a repres'Atative U.S. reactor type.
5 And I think at the time, NRR, yours truly, at 6 least my feeling was that, let's build the University of 7 Maryland. Let's run it; let's assess it; let's understand 8 it, and let's see if this is really going to produce useful 9 information. Let's make sure that there aren't any scaling 10 distortions or something that just totally invalidates the 11 usefulness of the data.
12 We've done that. We've build the loop; we've run 13 it; we've got the data. The data from the loop has been
(~N 14 evaluated and assessed by a number of experts. The scaling d
15 rationale has been investigated, and the like. I believe 16 there's a report in preparation right now which will in fact 17 provide this verification of the usefulness of it. And 18 based on that, we have put in the out year budget, starting 19 next year, money so that we can build representative loops 20 of either like a Westinghouse, PWR, a GE boiling water 21 reactor, perhaps one of the advanced designs, the advanced 22 over the OR designs, at universities. These can be run and 23 maintained -- I don't know what a graduate student costs 24 these days; you would know that better than I would, but we 25 think we could probably get by for -- once they're built, we
() Heritage Reporting Corporation (202) 628-4888
38
() 1 can run them and get useful information to the tune of 100-2 150K a year.
3 So I think that if there's maybe three of four 4 loops around the country eventually built and running, we 5 can maintain a fairly good capability in the order of 6 perhaps 600-700K a year, which I think would be a pretty 7 useful investment, worthwhile.
8 This is something we're still -- we don't have a 9 lot of details on this yet. Lou 1 think will talk a little 10 bit more on it later. But right now, this is what our 11 planning is. We put the money in the budget out through ' 91 12 or ' 92, presuming that it would take two or three years to 13 build these at the tune of something like a million dollars 14 a loop.
15 And then the fifth here is: to retain involvement 16 in international thermal-hydraulic activities. I put a 17 caveat in here: provided the resource commitment is 18 minimized and there is substantial benefit to the NRC.
19 What that means is just because somebody has built 20 the loop, doesn't mean that we have to jump in, in a big 21 international cooperative program and run our codes and do 22 all these analysis and send 30 guys over there a year. The 23 Commission's having some heartburn right new on foreign 24 travels, so obviously before we get involved in any more 25 international commitments of major proportion, we're going
/) Heritage Reporting Corporation (202) 628-4888
39 (m
wJ 4
1 to have to make sure that, I think, that both of these are 2 satisfied.
3 But, again, the intention is not just to bow out 4 of the world arena in this area. We will maintain our 5 contacts overseas; we will maintain involvement in 6 international thermal-hydraulic activities if there is 7 substantial benefit to the NRC.
8 And lastly, which you'll hear more about later, 9 because Lou will be talking about this is: we want to expand 10 the applications research using these codes to 11 systematically assess reactor behavior, both for the 12 operating LWR's, and I think we want to start study on the 13 advanced LWR's .
") 14 These are going to be coming up from licensing (V
15 soon. I think it would be good if we had a research program 16 that kind of ran these plants through their paces. Let us 17 understand where there might be some weak points, or 18 whatever, and also understand that if there's any problems 19 with the codes, that we're going to have to fix and the 20 like.
21 So anyway, that is basically the objectives that 22 we've set forth for the future of thermal-hydraulic 23 research. And I would probably venture to say that since 24 the Commission has a five year plan and we're required to 25 plan out therefore for five years, this is at least our
() Heritage Reporting Corporation (202) 628-4888
1 40 j 1 planning for the next five years in this area.
t( ) '
2 So Lou will talk more about the financial 'l I
3 commitment, but I guess without the -- if you subtract out '!
i 4 the accident management work he's doing, we're talking money .
5 somewhere in the order of $6-7 million a year. So I think l 6 that's still a substantial amount of money being put into an 7 area that's been studied pretty extensively for 15 years.
8 For comparison, human factors is running somewhere 9 around, I guess, 9 or 10 million a year. And that's a 10 program that's high on the Commission's agenda. -We have 11 substantial-user needs not only from NRR but from NMMSS and 12 the like. So I think there's a program that we're trying to 13 start, get off the-ground, get new things going on it, and 14 we have a lot of user needs. And that's around 9 or 10 15 million, and I think PRE research -- this is life after 16 1150 -- will probably be around 6 or 7 million a year. And 17 then severe accident research is going to be up in the l 18 neighborhood of 20-23 million over the next couple years.
19 So that from my division, that gives you an idea 20 of where -- because next year I think my fiscal year l 21- budget's around 50 million. So this gives you a rough idea l'
l 22 that Lou's branch, if you add on the accident management, is 23 about roughly 20 to 25 percent of the division budget. And 24 I have four branches in my division. So he's basically an 25 average in there. So that will give you a dollar idea of
' (] Heritage Reporting Corporation (202) 628-4888 l
L _ _ _ _ _ _ _ _ _ _
41
) 1 where we are. He'll give you more details on that a little 2 bit later.
3 That's my presentation.
4 Yes, sir.
5 MR. LEE: Is there any plan for thermal-hydraulic 6 support for alternative advance reactor types like -- metal 7 cooled or high temperature gas core reactor and things like 8 that?
9 MR. SHERON: We have not been asked by Billy 10 Morris' division in research to perform any research, any 11 specific thermal-hydraulic research, analysis, whatever, on 12 the LMFBR's.
13 Right now, I guess, there's still a capability 14 where we're involved. There is still funding coming in from 15 foreign sources for the Summer code, but that's about it I 16 think right now.
17 Most of that work is being done out of Dr. Morris' 18 division under Tom King. And I guess that's sort of been 19 totally over in that division, and we haven't really gotten 20 involved, or we haven't been asked to gear up and do any 21 research in that area right now. So we haven't done 22 anything.
23 Well, I thank you very much.
24 MR. WARD: Thank you, Brian.
25 MR. SHERON: I apologize. I am going to have to
( Heritage Reporting Corporation (202) 628-4888
42 1 run. I have another meeting that I have to be at, at 10
{
2 o' clock, and Mr. Beckjord threatened me-with my job if I 3 wasn't there.-
- 4 MR . .SHOTKIN: The package that you have has 5 actually three presentations in it. . I'll give the first one .
6 'now,'and then the last two D graphs will come up as 7 appropriate in the agenda.
8 Right now I'll be talking about thermal-hydraulic 9 research plans and funding cver the-next five years.
10 'Let me start with the planning assumptions.
11 .They're based on what Brian just presented. What he 12 presented was also planning assumptions; I've just repeated 13 them in a slightly different way.
14 First, we're planning to complete most major 15 thermal-hydraulic programs by fiscal '92. What this means 16 is that we'll be completing them between now and fiscal ' 92, 17 and.later on I'll be telling you just which programs will be 18 completed in which year.
19 Most of these programs are coming to a natural 20 end, through international commitments where agreements run 21 out, or with the codes where we just had an extensive 22 discussion here on the code development. And I'll be 23 talking, and David also, Bessette, will be talking about 24 some. international programs that we're gearing up to help us 25 improve these codes, to get the best thinking throughout the
/"N Heritar ;'
ing Corporation
, u 9-4888
43
('T L) 1 world on what some of these -- the best constitutive 2 relations are.
3 And maybe I'd better stop here and say that the 4 codes that will be coming out the end of this year, the TRAC 5 MOD-2 and RELAP MOD-3, which are really the only two codes 6 we're working on now other than codes for the LaSalle 7 stability problem that you'll hear about later; I believe we 8 have taken the advice of the ACRS that we heard when you 9 reviewed the Q-A documents for the codes, or the models and 10 correlations documents. We've taken the findings of the 11 CSAU study. We've taken the advice from the international 12 code assessment program that has been using these codes for 13 the last several years. And I believe that we have made the 7- 14 best effort to change most of these constitutive relations
'O 15 that were identified as being suspect.
16 I believe that the 3D2 step in numerics is 17 finished or has been finished for several months. The only 18 new numerical thing I heard about, and I guess Dave can talk 19 about it later, is a Japanese suggestion on how to handle 20 the fuel rod behavior that actually did cut down the running 21 time of some of the codes.
22 But I think most of the work in the past year has 23 been on the constitutive relations and on the advice that we 24 did get from the ACRS review, from the CSAU study and from 25 ICAP.
(' Heritage Reporting Corporation (202) 628-4888
U.
44 1 MR. WARD: Lou, I don't want to drift off now on 2 something you're going to cover later, but how - you seemed 3 to agree a little while ago that ongoing work that will keep 4 this cadre of experts busy might be addressing, might be 5' .doing the sort of thing that Ivan was talking about.
6 MR. SHOTKIN: Absolutely.
7 MR. WARD: But what I thought I just heard you say 8 is you've already done that, or much of it.
9 MR. SHOTKIN: I think that we have done most of 10 it. Again, I think these codes are so complex that whatever 11 we do, we'are going to be bound to get some criticism that 12 we haven't done enough. And I'm assuming that that will 13 occur with these next versions that come out, and that we
(} 14 still will always find constitutive relations to change.
15 One of the things that we are doing to get ready 16 for that is looking at the advanced light-water reactors.
17 These are the 600 megawatt plants, and we're asking our 18 contractors, or we're asking Idaho to look at the modeling 19 within the code, look at these reactors and tell us are
- 20 there any basic modeling that looks suspect when we come to 21 analyze the phenomena that are going to occur in these 22 codes. And it might be the stratified flow models and the 23 horizontal pipe; we have correlations in there. They may 24 say that those aren't good enough for passive, natural 25 circulations system. And that would lead then to people
() Heritage Reporting Corporation (202) 628-4888
45 1 working on constitutive relations.
-(}
2 Brian said this;-let me repeat it. What we want 3 to do'is keep good people involved. And I~think we have to 4 think of interesting work to keep them on board so that when 5 we really need them in a priority --
6 MR. WARD: No, I understand that. But --
7 MR. SHOTKIN: There may be people who want to work 8 on constitutive relations. And, yes, I think we'll always 9 be able to find work'like that.
10 I think we'd also like to keep experts who are 11 ~ interested'in looking at the overall systems approach'to 12 reactors. How does the whole system behave? I think we 13 want to keep those people involved doing interesting work as 14 well. So-I think we're going to have a variety of. work that
~O 15 will be in our base-line activities. One part of that would 16 be looking at constitutive relations, but I don't think we'd 17 have everyone sitting there looking at constitutive 18 relations.
19 MR, WARD: I don't think I meant to imply that you 20 should. But I certainly would like to see something that I 21 could read that would tell me what you're doing with them.
22 And I hear you, yet every time I take a look at a code that 23 somebody's using, I found, well, gee, it still got that in i
24 it; it still has this that's wrong.
25 There's an inconsistency between what you're p Heritage Reporting Corporation (202) 628-4888
h s E 46 l
1 saying and what I observe. And I'm sure this research is 2 going on. But I would like to see a little more h 3' information.
4 MR. SHOTKIN: I think we owe it to this 5 Subcommittee or the ACRS, if they' re interested, when we do 6 come out with the next final versions of the codes, to come 7 back and have you review it. And we'll go up and list on 8 one side of the page all the constitutive models that the 9' ACRS said were wrong,~or the CSAU or ICAP, and then here's 10 the fixes that we put in; here's the ones that we didn't 11 touch and give the reasons why we didn't touch them. I 12 think we would be perfectly willing to do that.
13 MR. CATTON: I think that's too late, Lou. In my 14- view, if we're to have any input, we should see what your Q. 15 plans are now with respect to those constitutive relations, 16 and comment. Then if you say that what we're saying is 17 silly, you go about your business. But if you come after 18 it's all done --
19- MR. SHOTKIN: Okay. We could -- I don't know if ,
20 you have the details today, but you've already seen that, 21 the development plans for MOD RELAP, MOD-3 and TRAC MOD-2 2:2 after it was presented. I can't --
23 MR. CATTON: The plans that I saw didn't address a-24 number of these problems. They didn't address this business 25 of the heat flux splitting. They didn't address this Heritage Reporting Corporation O (202) 628-4888
47-b 1 business of the:entrainment and how it's handled. There 2 were a whole list of --
3 MR. SHOTKIN: Then we didn't address them, Ivan.
-1 4 Then we have to come back and we'll tell you, here's what we
~5 addressed, and here's what we didn't address. And, you 6 know, would we open up for criticism for'that, then we would 7 have to show you why we didn't address them and what we 8 planned to address when.
-9 MR. CATTON: But wouldn't it be better if we did 10 that at the front end, and then you could say, hey, we're 11 not going'to do that; rather than'say, gee, we didn't do 12- that.
13 MR. SHOTKIN: But I'd be glad to at any time. We 14 got to come back this afternoon and do it, but I think we've 15 ~done it already.
16 MR. CATTON: I'm not asking-you to plague this
- 17. whole group with that. I would like to see it..
18 MR. SHOTKIN: Be glad to any time.
19 MR. CATTON: Because my recollection is things 20 that were listed at the ICAP meeting were more just the l- 21- gloss that goes over the surface. Some of the things that l-22 came out in the CSAU process were not on any of those lists.
23 MR. SHOTKIN: Okay, why don't we promise to send 24 you the complete list of model improvements or constitutive 25 relation improvements that our contractors came up with, or O- rie e- eerei e career tie-(202) 628-4888
l 1 l
~
l 1
i 48
( 1 ICAP came up with, and then show you the ones that actually j l
2 got into the final version of the code. And it's not a 3 complete list; we didn't do everything.
l 4 MR. CATTON: Fair enough.
5 MR. SHOTKIN: I'll send it to Boehnert. The~ fact !
6- that-we presented that at our last management review meeting 7 a few months ago. And I think that these development plans 8_ have-been on the books for half a year now. We haven't 9 really changed then. But we'll be glad to send them to you.
10 You may find that you've already seen it.
11 MR. CATTON: What would be good, if you sent along 12 a copy of the CSAU lessons learned with it. I'm not sure 13 that everybody here has seen them.
{} 14 15 MR. SHOTKIN:
MR. WARD:
I'd be glad to do that, also.
What was that? I mean, what form was 16 that issued in?
17 MR. SHOTKIN: Well, right now, what has happened 18 is that we've been waiting to get Westinghouse comments on 19 that document, and it isn't published yet. But it's just 20 about ready for publication. But we have draft copies I'll 21 be glad to send.
22 In order to develop and maintain the expertise to 23 meet future agency needs, I guess the assumption here is 24 that there won't be future agency needs for this kind of 25 work. That's one of our assumptions. I guess that's number i
Heritage Reporting Corporation (202) 628-4888 4
W
49
- v[') ' 1 three.
2 If there's any question of where some work should 3 go if it isn't clear that it belongs at a particular place, 4 then we definitely put it into the technical support center.
5 MR. CATTON: Lou, is that where that work on 6 valves took place in that support center?
7 MR. SHOTKIN: Which valves?
8 MR. CATTON: Motor operated gate valves. There 9 was some testing done at Idaho.
10 MR. SHOTKIN: Did we do it?
11 MR. CATTON: Yes, you did.
12 MR. SHOTKIN: Okay.
13 MR. CATTON: I'm not sure where within Research it r3 14 was done. Yes, I just heard about it a couple weeks ago.
\~/
15 MR. SHOTKIN: I know Griffith is doing work for 16 us.
17 MR. CATTON: I'm not talking about Griffeth.
18 Somebody was looking at what happens to water operated 19 valves with full flow through them; whether they can open 20 and close them, the forces on the diaphragm.
21 MR. SHOTKIN: I can find out who did that, but we 22 didn't do that.
23 MR. CATTON: It seems to me that when you bring 24 these things together that experience with that sort of 25 thing is what's going to maintain your cadre sharp.
Heritage Reporting Corporation (s] (202) 628-4888 l
50-i' 1 MR. SHOTKIN: . Absolutely,_right.
2 MR. CATTON: .And I get.the-feeling that there's 3 walls between'the different areas. To me that work was 4 ' thermal-hydraulic research and belongs together with the 5 rest of this business.
6 'MR. SHOTKIN: That's one of the reasons for going 7 to Idaho is that there's a variety of branches and division,.
8- not just ours, that supports a variety of work at Idaho.
9 And that I think-is not true of other labs,~the variety of 10 work that goes on.at Idaho. And as you pointed out, here's 11 something that's very directly related to thermal-hydraulics 12 that evidently somebody else is funding at Idaho.
13 MR. WARD: Yes, Ivan. 'I suspect that work was 14 funded by NRR out of Bob Bear's (ph) shop bacause he was the-
~O- 15- one that was down here talking about that to the Committee.
16 MR. SHOTKIN: Bear is now in Research, but in 17 another division.
18 MR. CATTON: Oh, okay, yes. Sorry.
19 MR. SHOTKIN: I can go into.more detail later when-20 I go through the budget on this. But we are defining base-21 line activities for the technical support center. At the 22 beginning of the year, during the planning period, we don't 23 define priority issues for the technical support center.
24 This is now three years old. Every year most of our base-25 line work gets bumped by priority issues. And I'll go over
'n U
Heritage Reporting Corporation (202) 628-4888 i
/
p 51 X
j) :L what the priority issues were this past year and then what 2 got bumped, and then what our plans are for next' year.
- 3. Brian mentioned the small-scale integral: test 4 -loops. This is a new concept, a new approach. I think
'5 we've been pleasantly surprised with how things have gone at 6 : Maryland. I think more than the' fact that we're surprised, 7 I think the people at Maryland are enthusiastic about the 8 results they're getting from their loop. And I think that
- 9. that has. led us to say, " Hey, why won't we do'this somewhere 10 else?" Maybe it's not a bad idea. It's certainly a very 11 cost effective idea to set up these loops at universities.
12 And we certainly can't build another major loop at a DOE 13 lab, we just can't afford that.
14 MR. WARD: Are you going to talk anymore about 15 this? I mean, is this something on your agenda?
16 MR. SHOTKIN: I don't think so, but I could --
17 MR. WARD: I'll ask you a question now, then.
18 Brian mentioned -- apparently what you're talking about is-19 perhaps four more loops at four different universities.
20 MR. SHOTKIN: I don't think Brian gave us that 21 much money.
22 MR. CATTON: He said $600,000 or $700,000 a year. .
23 MR. WARD: Well, he was talking about -- that's 24 what I was trying to get at. He mentioned --
25 MR. SHOTKIN: I think we'll do it one at a time, Heritage Reporting Corporation (202) 628-4888
[. _
52
) 1 Dave. I think that's our approach is that we'll go out when 2 RFP next year and get the proposals in, and we'll 3L probably -- we'll pick one. Maybe we'll pick two if they 4 look good and go with that. But I think we want to not dive 5 in with.four or five right away, because we're really not 6 sure whether we want to have one for advance reactors right
- 7 now.
8 MR. WARD: Advance reactors. I mean, there's 9 really two rather different concepts.
10 MR. SHOTKIN: Well, the advance reactors, that's 11 another story. We talked with DOE and they're also going 12 through this process of making their choice out of 13 proposals. It's not clear where their funding -- whether 14 they're going to choose one or more of these designs to go 15 ahead with.
16 MR. WARD: Okay. Brian mentioned an initial cost 17 of a million dollars --
18 MR. SHOTKIN: That's what we're planning.
19 MR. WARD: -- operating, a running cost of 100,000 20 a year. But as I recall the University of Maryland facility 21 cost considerably more than that.
22 MR. SCHROCK: I was going to comment on this. I 23 think your budgetary planning would be very faulty if you 24 really thought you were going to get much in the way of a 25 sustained capability with an investment of 100 to 150K a
() Heritage Reporting Corporation (202) 628-4888
i M
-53 l' year. That's not much money-anymore.
2- MR? MARD: : .Even at a university.
3 MR. SCHROCK: Any place.
4 MR. SHOTKIN: I see Richard is not-volunteering 5 anything right now.- But I can get the' exact number of what
-6 Maryland costs. My remembrance is that it.was close to a-7 million: dollars.
8- .MR. LEE: -Richard Lee from Research.
9 I'think the initial phase of building that-
' 1 10 ofacility~ costs 750K, and then later, I think we added about, 11
~
for.-just the construction, is close to maybe 300. .So it's
- 12. approximatelyLabout a million dollars to build it.
13 MR. SHOTKIN: We're going,to ask for proposals to
- come in.
14 And if.somebody says they can't'run it for 100K,.
15 then we'll have to listen to is.
16 The other planning assumption'is that thermal-17 hydraulic related issues.are going to keep coming. And our 18 experience over the last-10 yearu or so is that we do get 19 new' issues coming up. The current issues that we're working-20 on, the BWR stability that you hear this afternoon related 21 to the LaSalle event. We're intimately involved in that 22 with NRR doing analyses.for them. A year later we had set' 23 up a TPG like we had with the CSAU, and I think that's 24 - working out quite well.
25- We, at the request of AEOD, we are looking at the h' ~ Heritage Reporting Corporation (202) 628-4888
54 1 NRC simulators at the Technical Training Center and f~')Y
\ .
2 developing input techs for our thermal-hydraulic codes.
3 We're going to, then, for the same plants as the simulators, 4 and then run.them and see how good the simulators are, 5 because AEOD is looking to' upgrade these simulators.- So 6 we'll do a base-line check of where they stand today. And 7 then when they upgrade them, we'll be able to check if there 8 are any improvements.
9 Brian mentioned the' interfacing systems LOCA.
10 That's a project that'NRR is interested in this. We're just 11 'about ready to get their user need letter, although we've 12 been working with them for the past month or so. And we 13 just decided to go ahead and start a program on this. It 14 will be primarily a PRA human factors program, but there 15 will be a sna11 thermal-hydraulic accident management 16 component to it. In fact, we'll probably put the final 17 quotation marks or the final parentheses around the work.
18 And then finally the OTSG behavior. You know 19 we've had this tag group, and you'll hear more about it 20 later, that finished their job, came up with their report.
21 And at the last PMG meeting that Richard Lee was at, we've 22 decided to go ahead with one of the proposals, a little more 23 than the funding we have, and we've asked them to go back 24 and reduce the scope. And then we told them how much money 25 we had, and we also told them to tell us what they can give O Heritage Reporting Corporation (202) 628-4888
55
( 1 us for the amount of money that we have.
2 So we'll be coming back with what we intend to do .
3 and how much it's going to cost. But I think we're going to 4 resolve this-issue which would really complete all of the 5 thermal-hydraulic issues that we know of associated with B&W 6 plants.
7 And then of course we'll continue to explore new 8 initiatives when we plan our research. This is just a 9 bureaucratic activity. The current in these' years are 10 accident management. We're not going to talk about that 11 today. We do owe the ACRS, I believe it's a different 12 subcommittee, a briefing of what we're doing on our accident 13 management program. And I think we're at the state now 14 where we will be able to come and talk about it.
)
15 We're starting up a program looking at advanced 16 ONR safety issues. And this is just'the 600-megawatt plant.
17 And as I say, DOE will choose before the end of this year 18 which plant or plants they're going to go with. We're just 19 trying.to get ahead of the issue a little bit looking at the 20 two designs that are on the table now that we know about.
21 Idaho is starting to look at code capabilities for these two 22 designs.
23 MR. WARD: I guess I didn't -- DOE is going to 24 pick either the PWR or the BWR design?
25 MR. SHOTKIN: Here's what I know --
Heritage Reporting Corporation (202) 628-4888
56
^j 1 MR. WARD: Okay.
v 2 MR. SHOTKIN: -- based on what we talked with DOE.
3 They have the two designs that we know about that this 4 Subcommittee has heard about, or we talked about, were part 5 of some first-phase of their long-term program. That phase i
6 ended. They are now getting ready for their second phase '
7 where they have asked for proposals to continue. They have, 8 I think, several proposals; I think it's more than two 9 proposals. And they will, depending on their funding and 10 how things look, they will choose anywhere between zero and 11 X number of these proposals for actual development. It's 12 not clear in their minds yet, because they were just talking 13 with us, how that's going to go.
~ 14 But meanwhile, we talk with them. We are looking
's)
15 at the one proposal that happens to be the Westinghouse AP-16 600 that has the most documentation associated with it, 17 that's actually come into NRC. So that's the one we're 18 looking at.
19 And then David Bessette is going to talk later 20 about the international code maintenance consortia. This 21 will be an extension of ICAP, post ICAP. They're 22 interested, these foreigners of 12 or 14 countries who are 23 involved, are interested in continuing to work with us on i
24 the computer codes, will come out the last versions the end 25 of this year, give it to ICAP. They'll have another two I i
. i (N
\-)
Heritage Reporting Corporation (202) 628-4888 '
l
F i
p 57 1 years to assess.these codes. And what we're talking about
~{
2 is after .that -- after fiscal ' 91, .they are interested in1 --
3 or we're proposing to them that they supply us with p~
4 funding, and that we jointly set up code maintenance 5 activities at our labs. So this'will help us-maintain the 6 expertise.
7 Now you can relax. This is just bookkeeping.
8 We've defined a new budget structure which.I think I'd-like
- 9. to just let you'know what we've done and why:we've done it.
10 We've given you the Commission paper which -- I'll 11 say later, but I'll let.you know now, we'd like your 12 comments on it. That's one of the reasons we're here. In 13 the' Commission paper and in our planning, we have structured 14' thermal-hydraulic research into two elements. These are
=O 15 code words, budget words. The two elements is called plant 16 performance, which you'll recognize. It used to be thermal-17 hydraulic research. And the second element is reactor 18 applications. And this is this new activity where we're 19 going to take the' completed research and apply it to 20 regulatory issues.
21 Under the old budget structure, we had.under there
.23. .the new budget structure. But on the PWR large LOCA, PWR 24 small LOCA, and other experimental programs, a lot of these
-25 activities are coming to a completion. And we lumped all Heritage Reporting Corporation O (202) 628-4888 i
___.___1_i_______ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ . _ _ _ . _ _ __
58 i, ) 1 that together into one activity that's called experiments 2 and analysis. And then the modeling stays the same.
3 Under reactor applications we have --
4 MR. KERR: Excuse me, I don't understand. There's 5 a column there labeled "new", and you said that all those 6 things had come apparently to an end.
7 MR. SHOTKIN: No, no. They are coming to an end.
8 They' re coming to an end by fiscal ' 92.
9 MR. KERR: Okay. So what is it that's new?
10 MR. SHOTKIN: This is just a budget structure.
11 This is a --
12 MR. KERR: Well, I see something that says 13 " experiments and analysis." Is that not what's new?
14 MR. SHOTKIN: No, the names.
b'
/' These are new names.
15 MR. KERR: What do the names reprcsent?
16 MR. SHOTKIN: This name represents these previous 17 programs. As I say, it's bookkeeping. Because I think 18 sometimes you like to look at the details of our budget.
19 MR. KERR: I don't question your nomenclature; I'm 20 trying to understand what it means.
21 MR. SHOTKIN: Okay.
22 MR. KERR: You can call it anything you want, if 23 you'll tell me what it means.
24 MR. SHOTKIN: Okay.
25 MR. KERR: It means something that's coming to an
(
') Heritage Reporting Corporation (202) 628-4888
59
() 1 end. Is that right?
2 MR. SHOTKIN: No.
3 MR. WARD: No. That was kind of a throw-away 4 comment.
5 MR. SHOTKIN: What we did, we looked at programs 6 that were coming to an end. We're getting ready for the 7 next five years. And we looked at programs that are coming 8 to an end. If I could be off the record, it would be 9 better. But sometimes just keeping these things going, you 10 got to write a lot of sentences of justification for every 11 one of these lines. And we thought, well, a lot of these 12 are coming to an end in the next five years. Why don't we 13 combine into one line-item, and it will just save us the f~T 14 work of having to justify it and have to write comments on
\_.)
15 it, every exercise we have, which are several times a year.
16 And it also helps in showing that we are bringing 17 this total program, total testing program outside of B&W 18 down to some minimum level which you'll see will come out to 19 be mainly the university test facilities in the future.
20 And under " reactor applications" we had three old 21 budget categories and we've come up with three completely 22 new ones. Containment balance of plant, we're bringing that 23 to a conclusion. We've worked with NRR to see if they 24 needed any work. At first they said, yes, they needed work.
25 And then when we really kept working with them, it turns out
() Heritage Reporting Corporation (202) 628-4888 I
1
60 1 they didn't need anything. So that's going to zero.-
.{ }
2 We have the technical support which really goes 3 over into -- we're calling it " operating reactors."
4 MR. KERR: Does that imply that the NRR thinks 1
5 that they now understand containment performance well l i
6 enough?
7 MR. SHOTKIN: No. It's that they just don't need 8 thermal-hydraulic analysis to help them in their decisions. 'I 9 MR. KERR: Okay. j l
10 MR. SHOTKIN: I shouldn't even say that. They i I
11 don't need thermal-hydraulic research. They have the codes; 12 they use them. And if they need anything, they pay for it. j l
13 There was some of the work that we were going to I 14 do was on taking the COVA (ph) code that they use and making I
15 it more. user friendly, cleaning it up a little for them. j l
16
~
They had one technical issue which was the effective or the 17 amount of super heat in a containment after a steam-line 18 break -- I think it was a steam-line break -- and the fact l l
19 that you could get stratification. The temperatures would I 20 look different than some of the codes we' re calculating. ;
21 And then there was a test program that we were 22 talking about. But after just a long period of time, they.
1 23 really weren't interested in getting into that. Or if they i l
24 were, they're doing it themselves. l l
25 MR. CATTON: What about things like whether or not !
Q Heritage Reporting Corporation V (202) 628-4888 ,
l L _- _- -_ _ - - _ - _ - _ - _ _.
61
() 1 the valves will close to isolate the containment. That's a-2 thermal-hydraulic question.
3 MR. SHOTKIN: That's the interfacing systems LOCA 4 problem.
5 MR. CATTON: No, the containment. isolation 6 valves -- not high pressure to low pressure.
7 MR. SHOTKIN: Okay.
8 MR. .CATTON: Those are big valves. From what I 9 understand they don't know whether or not they'll close.
10 Isn't that a' thermal-hydraulic question?
11 MR. SHOTKIN: . Well, it's partly thermal-
.12 ' hydraulics; I agree with you.
13' MR. CATTON: Loads are thermal-hydraulics.
14 MR. SHOTKIN: The way it's handled here, it's
{ j 15 being done in our Divisi,on of Engineering. That sort
_ of i
16 work. That doesn't mean that they don't pay thermal-17 hydraulic experts at the DOE labs to look at it; it's just L 18 that the focus is more on the engineering aspects of it'.
19 There's a division of engineering that's more in-20 structural behavior, materials behavior, and that's where 21 that work is being done. I was using the word for our 22 division of engineering -- that's where that work would get I j
23 done. !
j 24 MR. CATTON: As a result it turns out it's not 25 being done, i i
O aerieae- R eerein, cergeration (202) 628-4888 1-L___ _.______ _-__- - ---- - -
61-
) l' the valves will close to isolate the containment. That's a 2 thermal-hydraulic question.
3 MR. SHOTKIN: That's the interfacing systems LOCA 1
4 problem.
5 MR. CATTON:- No, the containment isolation.
6 valves -- not high pressure to low pressure.
7 MR. SHOTKIN: Okay.
8 MR. CATTON: Those are big valves. From what I 9 understand they don't know whether or not they'll. close.
10 Isn't-that a thermal-hydraulic question?
11 MR. SHOTKIN: .Well, it's partly thermal-l 12 hydraulics; I agree with you.
l 13 MR. CATTON: Loads are thermal-hydraulics.
1 14 MR.-SHOTKIN: 'The way it's handled here, it's H15 being done'in our Divisi,on of Engineering. That sort of 16 work. That doesn't mean that they don't pay thermal-17 hydraulic experts at the DOE labs to look at it; it's just 18 that the focus is more on the engineering aspects of it.
19 There's a division of engineering that's more in 20 structural behavior, materials behavior, and that's where 21 that work is being done. I was using the word for our 22 division of engineering -- that's where that work would get 23 done.
24 MR. CATTON: As a result it turns out it's not 25 being done.
() Heritage Reporting Corporation (202) 628-4888
( ,
-62 i'
1 MR. WARD: Well,, I guess I'm not sure they're 2_ right. I mean, it's an equipment problem, and this other 3- division of research deals with the equipment problems.- I 4~ don't understand what you two fellows are driving at 5 frankly.
6 MR.-CATTON: I'm not sure that we're driving at-7 the same thing. But it is my observation that you tend to 8 work hard at what you know best. As a result, people who 9 are working with these valves, if they are at all,_just 10 don't seem to address some of the thermal-hydraulic-11 questions that they should in order to establish what the 12 loads that they have to deal with. They are taking loads 13 that somebody gave them without question -- and doing a good 14 jol uith what-they have.
15 MR. SHOTKIN: And so we have three new elements 16 under reactor applications: the operating reactors, which 17 is primarily the tech-sepport center. We also included our 18 work on the simulator benchmarking.
19- The LWR system studies right now is really the 20 advanced light-water reactor system studies. Although we 21 understand that there might be an application for CANDU 22 reactors, so we may have to get involved in that one I'm-23 sure.
24 And then we have the vn'1ysis for operating 25 reactor events, and that's where we do the work right now on O rie e- eerei - cerrer eie-(202) 628-4888
_________--__z________ _
j
i P.
J li 63
.x . 1- th's LaSalle stability problem.
-2 MR. WARD: What is this application for a'CANDU 3 reactor?
4 MR. SHOTKIN: There was a note that came in that 5 Mr. .Stello thinks that there might be an application for 6 CANDU reactor. And NRC might start thinking about what we 7 have to do to get ready for that. That's about it
.8 MR. WARD: Yes.
9 MR. SHOTKIN:' We haven't done anything more than 10 that.
11 MR. WARD: It seems that could be rather major 12 undertaking.
13 MR. SHOTKIN: Yes, I don't.know. I'm sure the 14 Canadians have done a lot of work.
15 Now under -- just to complete the new budget 16 structures so that you could find where some of these 17 programs are that we talked about -- under B&W testing we 18 have MIST which we are finishing up. The final reports are 19 in. There is one more test that will get running in a month 20 or so. There's OTSG testing I told you about. We're going 21 to start up that program with industry, once we get the 22 details straightened out.
23' And then we're planning to do a small-break LOCA, 24 CSAU -- CSU is code scalability -- applicability and 25 uncertainty. And we intend to do that for B&W plants and
( Heritage Reporting Corporation (202) 628-4888
64 L 1 use the MIST data to do that.
2 Under experiments and analysis, that new activity i
.3 under the plant performance element, we have the 2D/3D 4 program which I'll talk about later; the ROSA-IV program; 5' BETHSY, and then these small. loops. Maryland is the'one-we 6 have now. We're thinking of having a few more.
7 Under modeling we have TRAC, RELAP. Actually we ;
8 have ROMONA (ph) and a few other smaller codes. We have !
l 9 ICAP, and then ICAP prime which is the follow-on program to ,j 10 ICAP which will be this code of improvement' program.
11 Under reactor applications, under operating I 12 reactors we have the technical support center, thermal-13 hydraulic technical-support center, this program on )
11 4 ' simulator fidelity. LWR systems studies right now is just l
.O 15 the advanced LWR's.
l 16 What we have under analysis for operating reactor 17 events.is'all the work we have going in BWR stability, and 18 you'll hear about that later today. And we have a small 19 project on reactivity accidents, which is related to the l
20 Chernobyl accident. And that's finishing up.
]
i 21 Now with that introduction, I hope you can j 22 remember everything. If you can't I'll go over it again.
23 Here's the funding that we have over the next five 24 years. For each one of these activities, again, we have two 25 elements. We have the plant performance elements. And l
Beritage Reporting Corporation O' (202) 628-4888
65
'l - here's_the total each fiscal year'for plant performance.
- 2 And you see what the plant is. The plant - .and it's 3 starting at 7500 in 1989. The plan is to bring this down in 4 a natural way as we close our program, down to a base level 5 which we achieve in about ' 93 of $3 million a year. And 6 that's the number that you see in the Commission paper. And-7 that $3 million would be split right about equally between 8- experiments and modeling. I'll go into more detail.
9 Under reactor applications, which is the second 10 major element, you see that we have a big increase between 11 ' 8 9 and ' 90. And then that comes down to a long-term 12 funding program. In the Commission paper I think we say 13 between 3 nnd 4 mil?, ion, and we're showing it at 3.75
{} -14 15 million-which is split about equally between operating reactors, advanced reactors. And then we have in the out 16 years a small amount put in for analysis for operating 17 reactor events. But what has happened is that every year we 18 switch money down from one of these other areas down to do 19 this. And that's why this year you see that over a million 20 dollars that is primarily for the BWR stability, that money 21 had to come from somewhere else. We didn't have that in 22 the budget at the beginning of the year.
23 Why don't I go through each one of these lines 24 unless there are any specific questions.
25 The B&W testing, as you see, we have that i
L
'( ) Heritage Reporting Corporation (202) 628-4888 l
L
I 66 r^N 1 completed in ' 92. Going down, most of the money is in the O
2 budget in '89. We hope to be able to spend that in '89; we 3 may have to, if we can get the facilities started, it may 4 mostly get spent in ' 90.
5 Under experiments and analysis what we will have 6 is things like 2D/3D going to an end, and then pick up the 7 university testing. So that the funding stays. It's about 8 2 million over the next couple years, and then gets down 9 about a million-and-a-half in the out years.
10 Under modeling, we'll have most of the work.
11 We'll finish the codes in '89. It goes down in ' 90. We put 12 in some money in '91 to take the results of the OTSG and 13 develop models in ' 91. The ACRS Subcommittee was worried 14 about our ability to use the results of the B&W testing and
'~
15 put models in the codes, and that's our plan to do that in 16 '91. And then we'll go down to this base-line level, about 17 a nillion-and-a-half, which would be primarily code 18 maintenance in those out years, and maintaining expertise.
19 MR. SCHROCK: Lou, could I ask you: Do you 20 approach these five year plans with the knowledge that 21 budget expectations for the agency are going to be this, the 22 amount for research for this? Therefore, what piece of it 23 can we apply to this technical problem? Assess the 24 technical problem that you have. Look at it realistically 25 in the sense of how much do we need in order to impress
(^g Heritage Reporting Corporation ,
() (202) 628-4888
- - _ _ - _ _ _ - - _ _ = _ _ _ _ _ _ - _ - - - . - - _ _ _ - _ _ _ _ _ - - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - . . . - - - - - -
f 3 67 c ..
1 these' issues.
2 . MR . SHOTKIN: It's.done both' ways.,'I think in the 3 ~ severe accident area,. we can identify today technical 4 -problems-that we want to be working.on over the next three
- 5. to five years. And.we can name them: we,can even' estimate 6 them.
~
7 In.a; program like this that'is a mature program 8 coming to an end, we're finishing up. .We can't identify-9 particular~ problems that'we'll be working on two years from 10 now. I mean, we could say maybe a few constitutive 11 relations. But we can say we want to maintain that 12 expertise, and we make an estimate of how much we1want to:
13 maintain. And that's these numbers. The out year's three j 14 million for plant performance, 3.75'million for reactor 15 applications. And that's based on looking at what we're 16 .doing.today, how many issues we -- what we see' coming'on in 17 the future. We see the advanced light-water reactors. coming 18 on. We see some accident management work coming on --
19 maybe. But we don't see particular research issues in this 20 area that are clearly defined today.
21 You see the OWR system studies. We are starting 22 that up with a small project this year, and then we have.put 23 in the budget a planning, this amount of money in the out 24 years.
25 Analysis for operating reactor events, we have O rie 9- eorti 9 corror ti -
(202) 628-4888
I 68
() 1 money in here. This looks at the work on the Chernobyl 2 implications and on the BWR stability primarily, and we put 3 in money for next years. We're going to finish up most of 4 that work next year, put in a little bit more, and then we 5 come down. But what we expect is that as new issues arise, 6 these numbers will have to have more money in there, and 7 we'll just take them from some other part of the budget from 8 the out years. It seems we can't really define what those 9 issues will be. We know they will come up with a certain 10 regularity.
11 And finally, the operating reactors, there's a --
12 I have to explain this increase from a million to about two-13 and-a-half-million between '89 and '90. About 500K, it 14 goes to the simulator fidelity in each year. So what this 15 says is that in '89 we have put in 500K in the technical 16 support center, and next year we plr to put in two million.
17 Well, how could that be?
18 What happens is that -- what happened this past 19 year, and I went through it in detail -- is we started the l 20 year with a certain number of base-line programs. And we l 21 thought we' d be doing a technical support center that came 1
L 22 out to about $1.7 million. These included synthesis 23 reports. We were going to build some input decks. It turns 24 out we didn't build any input decks. We were going to i
25 analyze generic issues. It turns out we didn't do any. We l
1
() Heritage Reporting Corporation (202) 628-4888 1
1
y 69 )
r3 1 were going to assist in industry audita. We didn't do any.
V 2 And then some new things came along that we had to ;
3 do. We started up accident management. So we switched some 4 of that money into looking at accident management.
5 We had to do some little more work on getting 6 ready for the OTSG testing. So we did that.
7 We had the BWR stability issue come along at the 8 beginning of the year. We hadn't budgeted for that, so we 9 took money from the technical support center and put that 10 into BWR stability.
11 So that explains why this number is so low, is 12 that some of it disappeared off of this view graph and ended 13 up in accident management, which we're not talking about
,. 14 today. And most of it ended up down here, looking at the
\)
15 BWR stability.
16 Next year we plan to write -- this is our planning 17 on the base-line for the tech-support center. We plan to 18 come out with synthesis reports. What we're planning right 19 now is to complete the work on natural circulation, look at 20 perhaps secondary feed and bleed that would be connected 21 with B&W testing program. And it's not clear yet how far 22 we're going to go on interfacing systems LOCA, but if that 23 'comes to be a -- require the need for a synthesis report, 24 we may want to do it on that.
25 We do plan to build input decks. A few years ago r^s Heritage Reporting Corporation
(_) (202) 628-4888
70
. v[') 1 we had a group meet and decide what should the agency be 2- doing in order to get ready for the future. The consensus 3 was that we ought to be building input decks. So we put 4 money to build input decks, one on the Vogtel reactor, and 5 one BWR to be determined.
6 And then the other base-line activity that we're 7 planning to do next that we couldn't do at all this year is.
8 look at code applicability as applied to either Westinghouse 9 RCE plants for small-break LOCA, and then either 10 Westinghouse of CE for loss of feed-water. This is what 11 we're going to plan to do, but as the year goes on, we'll be 12 asked to be involved in higher priority issues, and we 13 probably won't get to do a lot of this work. We'll just put f' % 34 it off until next year.
b 15 MR. KERR: How many input decks do you now have --
16 any?
17 MR. SHOTKIN: We have good input decks -- well, 18 varying degrees. I would say about a dozen really good 19 input decks; and of lesser quality, probably another dozen 20 MR. KERR: How many do you think you need?
21 MR. SHOTKIN: We would like to have one for each 22 representative type. And I think there are 23 23 representative types of reactors of the 120 that are there.
24 So if we have, I would say, about 20 good input decks, I 25 think that we could probably handle almost anything that Heritage
(~') Reporting Corporation (202) 628-4888
,- '71 1 comes up. )
(~)Y
\_
2 i l'
3 4
5 6
7 8 ,
9 10 11 12 13-S 14
( 15 16 17 18 19 20 21 22 23 24 25 Heritage Reporting Corporation O, (202) 628-4888
_ _ _ _ _ - - - - - - _ - - - - - - - - _ - - - . - . - - - - _ - - - - -- - .- - - - - - - - - -- ------_--J
72
(~)
L.'
1 MR. KERR: That is the basis then as something 2 that you will need them.
3 MR. SHOTKIN: This is maintaining expertise, 4 keeping good people busy doing interesting work, but 5 something that is useful to NRC. And we had several 6 meetings with NRR, AEOD, Research, various branches in 7 Research, and this is the consensus of what they would like 8 to see done as a base line activity. As I say, last year we 9 had money in there and we had to divert that money to do the 10 BWR stability work.
11 MR. WARD: Lou, I did not understand what you 12 meant by the code applicability for loss of feedwater and 13 for Westinghouse plants.
14 MR. SHOTKIN: You remember that we had this plan gs)
(_
, 15 about two years ago. This was when I think Gillespie was 16 the division director. And I think that we presented it to 17 the ACRS where we were going to look at each plant type and 18 for a range of transients and see whether the thermal 19 hydraulic codes that we have today or any time that we want 20 to use them are applicable to that geometry and that 21 transient. And I think that one of the CSAU was an overkill 22 for what we want. We did do a small study that came out 23 last year on applicability of --
24 MR. WARD: Well, wait. The CSAU that was done was 25 for a large break.
Heritage Reporting Corporation
(^}
\~ (202) 628-4888
(
t
73.
'l'. 'HRi'SHOTKIN: Right', in the Westinghouse. plant.
r< 2~ MR. WARD: Yes.-
-3 MR. SHOTKIN: And what we had going'on,.actually 4 if you remember historically, before we-diverted *ourJeffort' 5 .to look at.large break LOCA at Westinghouse,-we had an-
-6 effort to look-at loss of feedwater.in a B&W plant. Jmd 7 that report came out;1ast year, the applicability of'the 8 computer codes to look at a loss of feedwater transient in 9 the-B&W plant. ..And that is the sort of thing that we think.
10 that we ought to be doing.
11: I do not think that we have.ever done it'for a 12 combustion engineering plant. We have just gone under the 13 assumption.that Westinghouse is like CE because of the U-2 14 steam generator. And if we can do a two by four in a B&W, 15 then'we can do it'in a CE. And I think that we ought to -j i
- 16 probably do a more . <:areful look at small break LOCA:in a CE I l
17 or another Westinghouse plant and perhaps loss of feedwater.
18' .MR.-MARD: Apparently of these dozen input decks,.
19 you have enough to do this sort of work.
20 MR. SHOTKIN: Yes, we'have at Calvert Cliffs.- d 1
21- MR. WARD: Then what are;you going to compare it l 22 with? i 23 MR. SHOTKIN: What we are doing again similar.to 24 CSAU, we are going and looking at the code itself and 25 looking at the phenomena that we expect it to model for this Heritage Reporting Corporation l (202) 628-4888 l
)
d 1
74 1 geometry and this transient, and do the models and the
})
2 constituent relations that look like they apply to that, are 3 there any deficiencies that we can identify. We will do 4 some analyses, and apply it to any relevant data that 5 exists. There is data on U-2 steam generators. It might be 6 Westinghouse data, or we may have to'extrapolatova little.
7 But again this is a base line activity. This is a 8 maintaining expertise type activity.
9 And I will tell you, if the ACRS has any -- I mean 10 we heard Professor Catton give a suggestion of:looking at 11 constituent relations. If you have any other suggestions of 12 base line activities to maintain expertise, we would be very 13 glad to hear them.
14 MR. LEE: I have a question. I guess in base line 15 activities in general, but in particular in generating these 16 inputs for various power plants perhaps'that there seems to-17 be a lack of utilization perhaps of what seems to be a 18 parallel effort in the industry especially with the RETRAN 19 code. A lot of utilities have made a lot of investment and 20 to my understanding they use a RETRAN code. And they have 21 developed their own simulation capability with that 22 particular code. And I was wondering if you would comment 23 on that.
24 MR. SHOTKIN: Well, my comment is that it is 25 great. And I know that they have conferences and that they Heritage Reporting Corporation (202) 628-4888
E 75 g()' 1 do interesting work, and I think that it is very-useful to 2 the utilities. Is your question do we.have access to that 3 information, is that your question?
4 HR. LEE: Or are you planning to make use of their 5 -expertise in generatingthis input in perhaps a more 6 efficient manner for example?
7 MR. SHOTKIN: -Any time that we can get utilities 8 to help us, we take it. It is usually difficult to get a.
9 utility to help you until they have a problem. The LaSalle 10 staff was very helpful after they had their transient.
11 Davis-Besse was very helpful after they had their transient..
12 But typically NRC Research cannot go to a utility and say~I 13 would like to model your plant and expect very much help
. 14 from the utility staff.
15 In fact what we thought a few years ago was that 16 we talked to EPRI to see if we could get hold of the input.
17 decks that they had for RETRAN, and their answer was that it 18 does.not belong to them but that it belongs to the utilities 19 and we have to get permission from the utilities.
20 MR. LEE: Are you planning to pursue that path a 21 little bit further to see if you can make some comparison, a 22 comparison between RETRAN and RELAP-5 capabilities for 23 example?
24 MR. SHOTKIN: Oh, yes.
25 MR. LEE: Do you have plans to do that?
Heritage Reporting Corporation (202) 628-4888 1
_ - _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ - _ _ - _ _ .- _ _ _ __ - -___ _ _ D
76 1 lMR. SHOTKIN: Are you talking about comparing how 2 well.RETRAN does to our computer codes?
3 MR. LEE: Or certain power plants for example or' 4- certain trends.
5 MR. SHOTKIN: We-do not have.anything specific in 6 mind. That is more of something that the NRR might. request-
~ l We know 7 us to do. I think that we know what is in RETRAN.
8 the modeling and we know the kind of transients that they' 9 do. And we do not have any plans to do any specific bench 10 marking against RETRAN.
11 MR. LEE: It is your judgment that there really is 12 no need or that you do not anticipate much benefit coming 13- out of such a comparison or corroboration if you will?
14 MR. SHOTKIN: If we can get a corroboration and 15 partuof the corroboration was they would say how good is our 16 code, will you tell us, we will be glad to do that. But the 17 real world is that we just cannot go knock on the door.of a 18 utility and expect them to free up an engineer's time to 19 help us do research. The world does not work that way, 20 unless there have a problem.
21 Like thinking historically when we did our 22 pressurized thermal shock studies on the three utilities and 23 we were doing work for them, they were very cooperative. We 24 visited them and they gave us anything that we wanted, and L 25 they hired consultants to help us. When a utility has a 1
Heritage Reporting Corporation (202) 628-4888
i 77 !
. n.
( J- l' problem that generates NRC interest.like LaSalle, their 2 staff also, and'GE also was very helpful.
3 Why we chose Vogel is we felt if interfacing 4 systems LOCA comes up, there is a bigger problem, and having H
5 the event'at vogel maybe we will get Vogel's staff 6 interested in helping us. But just randomly going and 7 knocking on the door, they have better things to do with 8 their time.
9 MR. LEE: Let me pursue this once more. I am 10 somewhat surprised and concerned in some way that there has 11 not been as much corroboration between two groups of people 12 developing capabilities of a similar nature, not much of a 13 direct comparison between RELAP-5 and RETRAN capability for 14 example in the models.
15 MR. SHOTKIN: RETRAN is based on RELAP-4.
16 MR. LEE: But they have had a lot of modification 17 made since the days of RELAP-4.
18 MR. SHOTKIN: Yes.
19 MR. LEE: So right now they look quite a bit like 20 a different animal to me. Again these are parallel, and 21 somewhere along the NRC and the industry each has made a l- 22 tremendous amount of investment. And we would like to see a l'
23 little bit more corroboration if possible.
24 And does the NRC have access to the RETRAN code?
25 MR. SHOTKIN: Oh , yes, sure. The corroboration h Heritage Reporting Corporation (202) 628-4888
78
- (~) 1 would be more reviewing it, reviewing a submittal from V
2 industry. We have corroboration throughout the world and 3 throughout the United States with people who are not the 4 regulated industry primarily. When it comes to the 5 regulated industry, we do cooperate on the B&W testing 6 program. B&W is using RELAP-5 and we corroborate with them.
7 We are not using RETRAN. And on PTS, they did not use 8 RETRAN.
9 When we get into a cooperative program with 10 industry, they do not use RETRAN. So there is no natural 11 cooperative push between us and the utility industry on 12 RETRAN.
13 MR. LEE: I think that it is the utilities g- 14 primarily that use RETRAN and not the vendors. The vendors
~'
15 have their own codes more often than not. So I think that I 16 would like to see a little bit more of the utilities' 17 involvement that may be possible.
18 MR. SCHROCK: If I could proceed for just a 19 moment. There is a fundamental thermal hydraulics issue 20 that comes out of this that is brought up between RETRAN and 21 RELAP-5. That is that the EPRI people have embarked on this 22 path of claiming that we have constituent relations that are 23 flow regime independent, whereas the two NRC codes are 24 continuing with a pretty well accepted theme fifteen years 25 ago. And success in modeling these complex phenomena
('
Beritage Reporting (202) 628-4888 Corporation
l 4 79 f 1 depends upon having the clear picture at each stage through i
1 (f
2 the development'of the transient as to which flow regime or.
3 ' flow regimes the system is currently operating in, and 4 developing adequate constituent packages to describe the l 5 . behavior in those specific flow regimes.
6 Now we have EPRI doing something altogether 7 different, and this is a very fundamental question. It 8 would seem to me that it is of great importance to the NRC 9 to evaluate it on its merits as a funduaental issue.
10 Does it ever get discussed in NRC?
.11 MR. SHOTKIN: Mr. Jones, if you could help on 12 this. Because we certainly never get involved in research.
13 He is shaking his head no.
<N 14 MR. SCHROCK: Is that because you think that the g
15 idea is a poor one or do you think that it does net merit 16 the time and effort that is necessary to evaluate it?
.17 MR. SHOTKIN: I guess the question is are there 18 any licensing or regulatory issues where RETRAN is used in a 19 ' submittal. If they are just using RETRAN to make themselves 20 . comfortable about how their plan operates, that may be.one 21 issue. Mr. Jones is going to make a remark.
22 MR. JONES: This is Bob Jones, NRR. With respect 23 to RETRAN use in the industry in general, they are using it 24 mostly in the single phase mode for a transient analysis.
25 In most of the two phased type transients, they have not
() Heritage Reporting Corporation (202) 628-4888
80 l( ) 'l pursued it at'all. So from the re'gulatory-sense, we see no 2 reason to pursue it.
3- MR. LEE: I am surpriced to hear that. I thought 4 that the RETRAN has a full-blown two phased capability just 5 like RELAP-5 does.- And I would like to know why you.think 6 that the utilities do not want to use RETRAN~with two phased 7 flow of transients.
8 MR. JONES: You will have to ask the industry.
9 All I can say is I was not involved in the RETRAN review. I 10 know that RETRAN-2 was reviewed many years ago. I think.
11 that the review was complote in 1983. And one of the areas-12 of primary concern at that time was the two phase modeling.
13 There were some concerns about it. They were not sure what 14 they wanted to do. They had two different. options.at the 15 time. So we left it in their hands, and nobody has picked 16 up the ball to go further as far as I know.
17 And all of the utilities that have come in to use 18 RETRAN have restricted its use to basically normal 19 Chapter 15 transients and accidents, which means that you do.
20 not get into two phase conditions in general.
21 MR. SCHROCK: Well, certainly what I said about 22 them pursuing a path of constituent relations that are full l 23 regime independent is very well documented. There is just 24 lots of EPRI literature about this and lots of discussions 25 of it. I would kind of like to get back to the point that I
() Heritage Reporting Corporation (202) 628-4888 l !
_ _ _ _ _ _ _ - _ _ _ _ .- _ __ i
81
(~') 1 was trying to make and see if I cannot make some progress c
2 with it. And that is that I believe that NRC should be 3 interested in the fundamental aspects of what it is doing in 4 technical calculations, and not providing the answer that if 5 there is not regulatory need that we do not think about 6 fundamentals.
7 You have got great difficulty in constructing a 8 viable research program. You have always had that 9 difficulty. And I do not understand why it is that you 10 constantly have this sort of retort that we just heard in 11 relation to a fundamental issue. Here is one where the 12 industry seems to be going in one direction while NRC is 13 continuing in another direction. I do not agree with what 14 EPRI is doing in that in the technical sense at this stage, f-)
LJ 15 and I think that it is going to take a heck of a lot of 16 proof to demonstrate that it has merit, but I do not see how 17 it can be ignored by the regulator if that is what is 18 happening out there in a significant way in the industry.
19 I think that it has to be evaluated. It is clearly the kind 20 of thing that your cadre of experts ought to be addressing.
21 MR. JONES: When have an application which will go 22 into the two phase regime, we will go look at it. Research 23 has their experts there, the thermal hydraulic experts who 24 go out and look at the fundamental equations, and they are 25 responsible for the codes that are turned over to use for
/~) Heritage Reporting Corporation
- (202) 628-4888
a 82
,-m (8s ) 1 our use. And if it is going to be resolved anywhere, it 2 will be resolved there. But we are facing an era of 3 basically very limited resources and the best way to apply 4 it.
5 I think that that is a question that may be 6 suitable for this carry-on program in research to maintain a 7 cadre of experts, which I think that that was an excellent 8 suggestion made earlier that Brian has stated that they are 9 going to go back and think about. And I think that I would 10 add this as something to the list that we would consider.
11 MR. SCHROCK: Very good.
12 MR. CATTON: If your experts do the kinds of 13 things that he is suggesting, they wiL not be experts for 14 very long.
(3) -
(_
15 MR. JONES: They will not do it unless they are 16 paid and given the direction to do it. Realistically that 17 is what will happen. We are not telling them to ge do it.
18 And the RETRAN applications that we have got are very 19 limited. You have already heard our basic conclusions s
20 already on what we think the codes are. We tuink that they 21 are good enough to do the job that we need them to do at 22 this poir;t, and we have got a plan that will continue to go 23 forward and maintain these codes and look at the stuff as it 24 comes through and maintain the expertise. And hopefully 25 that is t-he type of useful work that they would be doing.
<3
()
Heritage Reporting (202) 628-4888 Corporation
I 83-
/'T : 1 MR. SHOTKIN: Let me make a suggestion that under V
2 code applicability that we add the possible task of looking 3' at-RETRAN. And I will go back and check with my management 4 and with NRR. It is not something that is a natural for us 5 to do, but you heard NRR, the representative here, express 6 at least some mild interest in it. And I will certainly go 7 back and try to get it into our program..
8 MR. CATTON: Maybe I missed a point, Virgil, but 9 were you not suggesting that they should interested in the.
10 fundamental issues on which their codes are based, and that 11 the EPRI thing and RETRAN was just an example?
12 MR. SCHROCK: Right.
13 MR. CATTON: I do not think that you are being 14 advised to go and in particular look at RETRAN. I think U-s 15 that the suggestion was the fundamental issues on which your 16 codes are based which should receive some attention.
17 MR. SHOTKIN: Okay. We will come back and tell 18 you what we are going to do and why. And.first we will see 19 if this gets management approval. It certainly is something 20 different than what we usually do, unless we are asked to do 21 it, to go and try to review an industry submittal before it 22 is submitted.
1 23 MR. KERR: At the risk of confusing things 24 further, as a listener it seems to me that what is being 25 suggested is that this is new work and a new direction that Heritage Reporting Corporation (202) 628-4888 1
L_1________-___-_
84.
(j' -1 is being taken by people who presumably are trying to _
'l 2 , describe the thermal hydraulics phenomena. And that'is you 3 are going to be aware of~what is going on and be experts, it 4 is something that you ought~to look at. Not that you are
- 5. trying to review RETRAN so that you approve or disapprove
.EF it, or license it or' write a topical report on it. But that 7- you understand the basis on which.these relationships are 8 derived and the applicability that they have, just as you 9 would want to understand the other. development in the 10 ' thermal hydraulic area.
11 MR. SHOTKIN: I agree and I have no problem. I 12 will use your words. There is no problem in looking at j '..
13 But as I say, I would first have to go back and discuss it rs 14 back home, and then we will get back to you, and we will say
' (-)
15 here is what we are-doing and why we are doing it, but it is. ,
16 not in our program right now.
17 Are there any more questions on this viewgraph?
18 (No response.)
19 MR. SHOTKIN: If not, I can go to my final 20 viewgraph that I have for this part of the presentation.
21 You are asked are there any programs affected by funding 22 availability. And this is the first one that we mentioned 23 in our Commission paper. And it says on new thermal 24 hydraulic test facilities for the 600 megawatt lightwater 25 reactors. Well, I should have said major test facilities.
h Heritage Reporting (202) 628-4888 Corporation
- U_.-__.__._..__.___-_....-
F x r
- 1. _
85-
. We might have a university test facility for the advanced.
L'- ' (f .3_); 1" 2 lightwater reactor.-
'3
. But what'I want to say'is that I know that we have'
.4 'come before this subcommittee and before the ACRS'with
~5 various' plans for continuing experimental 1' capability and a' 6 modular' test facility which was a big facility located at.a1 p 7 DOE lab, and we do not have that in our plans at all-p 8 anymore, not for the lightwater reactors or~for the advanced I h 9' light water. reactors. What we may have is'a_ university 10 facd,ity to look at this, q i
11 MR. WARD: You mentioned that there is going to-be 12 another found of DOE support for the 600 megawatt reactors.
13 Is that just going to be design development or 14' might that'be experimental work?
15 MR. SHOTKIN: My guess and it is a guess is that 16 it is just design' development. I think that we know that GE i 17 has already run tests, and I'think that Westinghouse has run 18 some small scale tests on theirs, and I do not know who1eise
.19 has submitted. But I believe that it is just design. j i
l20 MR. SCHROCK: I hear comments from these people to i
)
21- the effect that their aim is to have plants certified. l 22 MR. SHOTKIN: Yes, that is right.
23 MR. SCHROCK: How can that be done without 24 supporting experimental evidence. You cannot certify the j l
-25 plants on the basis of site alone. !
l
-0 rit e- eerei 9 cereer eie-(202) 628-4888 l
1 L
i 86 I
). 1 MR. SHOTKIN: Professor Schrock, that is just why 2 we are starting to look at this a little ahead of time.. We 3 do not.have a submittal yet from the industry, but we know 4 that the time scale is~ going to be tight. And we know that 5 if we do have to do any experimental testing that.we will 6 have to.make plans long enough before so that we will have 7; the results irt time for the review. And I agree with you 8 that that has been our whole motivation for starting this, 9 that there is going to be a tight time schedule. That when-10 the stuff comes in that there is only going to be a short 11 time for NRC to review it. And if we are going to do any l
12 review like we have done in the past, then we have to get 13' ready for it.
14 But we do not see right now, we have not put
.O '
15 anything in for any major test facility. What we have, if 16- we have'to, we could build a small scale university 17 facility. You know, I do not think that is that bad. I 18 would say that from what we know of these 600 megawatt 19 designs that we do not think that there is any major 20 facility needed. I think that we know a little bit about 21 them.
22 The second program. I will be frank, I had to 23 reach for these a little bit. You wanted something, and I 24 could say this one, but then I had to sort of reach for 25 these, and let me tell you a little bit about them. What I Heritage Reporting Corporation O. (202) 628-4888
87'
( ; 1 am trying to say is that I do not think that we are hurting 2 for funding in any area right now that I could identify. We 3 do not have any new software plan for accident management.
4 We will use existing-codes like RELAP/SCDAP. Right now the 5 . source term code package probably when it is ready MELCOR.
-6 We do not thin that we will use TRAC very much.
7 You would say well, maybe you need to develop a 8 new code, but we really do not see any justification'or any l 9 need for doing that, and we do not have any of- that funding 10 in our budget in case for some reason we decide that'the 11 codes do not run fast enough or we really need to do a lot 12 of work on accident management studies.
13 MR. KERR: Well, that probably means then that you 14 have decided, and maybe this is a valid decision, that you 15 do not.really need to know very much about accident 16 progression in vessel, because certainly existing codes do 17 not tell you anything about that. They do what you tell 18 them to do, but they do not really give you any insight 19 other than what you insert.
20 MR. SHOTKIN: Before I answer that, let me just 21 say one thing. MELCOR is designed I believe to be the fast 22 running accident management code. If there is any code that 23 is supposed to be doing it, it is MELCOR. And I can say 24 does MELCOR have good models in it for, as you say, in 25 vessel melt progression. And I do not care too much about
() Heritage Reporting Corporation (202) 628-4888
s 88 melt progression, but I care about adding water to the
( l' 2 degraded core,.what is the effect.
3 We have a research program under accident 4 management that.does look at the consequences of adding to 5 larger degraded cores. Right now it is a scoping analysis 6 program. We-have written a user need letter to the branch, 7 to Castansi's branch, saying for accident management that we 8 think that there is not'enough information here, could you.
9 please have a research program. And that-is one of the 10 issues that-is in the severe accident research plan, that 11 they will be looking at that, probably at some experimental 12 work. They will be looking at code uncertainty, how well 13 the models do, and what model improvements need to be made.
14 MR. KERR: I think that you are telling me.then.
15' that.this is not thermal hydraulics, but this is severe 16 accident.
17 MR. SHOTKIN: Yes, right. But under accident 18 management, the part of our work that I am not talking about 19 today is that we are responsible for identifying for that 20 .Castansi's branch what work has to be done for accident 21 management, and I think that we have already done it for the 22 stage of the accident that they talked about. We think that 23 it is very important. There is a little disagreement that 24 we have in-house. We-think that it is important for the
'25 positive effects, but certainly everybody agrees that it is s
O' Heritage Reporting Corporation (202) 628-4888
=_-_______:____ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
89
() 1 important to know the adverse effects of adding water such 2 as hydrogen generation, steam production, steam explosion, 3 the shattering of whatever existed. A11yway that is in the 4 research program.
5 MR. CATTON: Well, RELAP-5-SCDAP, as near as I can 6 tell, that was used to address the question of the in vessel 7 recirculation and where the failure would occur. Yet most 8 of its internals are highly speculative.
9 MR. SHOTKIN: Part of our severe accident research 10 program, and again this is in Costansi's branch --
11 MR. CATTON: If that falls into the other program.
12 MR. SHOTKIN: Yes, it is to look at just that. We 13 are aware of these deficiencies and uncertainties. And my 14
~
job is to say we want to use these codes and here is what we 0(N 15 want to use them for, and you tell us how good you think 16 that they are. And if anything more has to be done to make 17 them better, do it, because we want to use them.
18 MR. CATTON: Where is the boundary?
19 MR. SHOTKIN: The boundary is very clear, let me 20 see if I can explain it, between our branch and Costansi's 21 branch. They do research on experiments and co-development.
22 We use the results to apply it to accident management. So 23 we apply the research to accident management and they do the 24 research. Now when we look at accident management, we think 25 that we are also doing research on what procedures and e
')
q, Heritage Reporting Corporation (202) 628-4888
90 1 strategies are, but they look at experiments on phenomena 2' and code development.
3- MR. CATTON: Have you asked them to look at 4 RELAP-5-SCDAP?.
'5 MR. SHOTKIN: Yes. We have even identified the 6 one place were RELAP-SCDAP gets one answer and TRAC MELCOR-7 gets another answer.
8 MR. CATTON: I am not surprised at that.
9 MR. SHOTKIN: Which one do we believe and why.
10 .Yes, we have identified that already.
11 MR. CATTON: I listened to a presentation on 12 RELAP-5-SCDAP dealing with the question of where the primary 13 system would fail. And the person doing the calculation was 14 not aware that the heat transfer coefficients that he was 15 using might be_off by a factor of two or three, and that 16 that could change the location of the failure significantly, 17 and he thought that he had done a sensitivity analysis, but 18 he had not done really much of anything.
19 Now whose branch was that worked on, was that done 20 under you?
21 MR. SHOTKIN: If you are talking about the I
22 analysis of depressurization to minimize direct containment 23 heating?
24 MR. CATTON: No. About where the failure in the
, 25 primary system would occur, heating and overheating.
1 ,
i i
h Heritage Reporting Corporation (202) 628-4888 l
L_ -_ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . __ _____-____ _-_ __-______-______-_- - - .
-- . I
V J 91.
I\ 1 MR. SHOTKIN: We are not looking at that, so it
'V 2 must be Costansi's branch.
I 3 MR. CATTON: Okay.
4 MR. SHOTKIN: I would say that we look at 5 something that is directly related to an accident management 6- strategy or procedure. If it-is some general concept, then y 7 we may be looking at it.
1 8 MR. CATTON: This.certainly was an accident-9 management question I think.
10 MR. SHOTKIN: Okay. Well, I am sure that there 11 will be some overlap, but I think that we will work it out.
12 The work will get done whether we do it or whether someone 13 else does it. It will get done.
14 MR. WARD: Does that about wrap it up?
15 MR. SHOTKIN: Yes. I just had one other thing 16 that I sort of reached for as sort of a lead-in but let me 17 tell you about it. I say that there is no specific testing 18 for accident management. The Germans on the 2D-3D project 19 on the UPTF uave proposed a follow-on testing program for 20 UPTF on what they call accident management. And of course 21 this is before severe core damage, it is that kind of 22 accident management.
23 And we have looked at it and we are reviewing it.
24 We are going to be discussing it with them next week. -Right 25 now it looks like UPTF has so many deficiencies that we
( Heritage Reporting Corporation (202) 628-4888
92
() JL ' would.have to'know how they are going to run the tests and 2 what is the usefulness of the data. But the Germans I 3 believe will try to'get some international consortium 4 together to do further testing in UPTF, at least to. call it 5- accident management testing. But we do not have any.
6 specific funding in that to participate in that.
7 MR. SCHROCK: You do not have an item in the 8 budget?
9 MR. SHOTKIN: I do, but we are not talking about 10 it today. That is a third element. There is plant 11 performance,sreactor applications, and then accident 12 management. And accident management is a separate ACRS-13- subcommi' tee, and we. owe the ACRS a review of that program.
14 But the Commission paper that we are talking about today is 15 just on the first,two elcuents, and I am not talking 16 accident management today. I will answer questions, but we 17 did not plan anything on accident management today.
18 MR. SCHROCK: How does the budget for accident 19 management compare with this?
20 MR. SHOTKIN: It is $4 million a year.
21 MR. WARD: Okay. We have not taken a break yet, 22 so I think that we need one. Let's come back at 11:10.
- 23. (Whereupon, a recess was taken.)
24 25 )
O rie e- gerei-e cerrer eie-(202) 628-4888 1
1
1, 93
.([ 1 MR. WARD: Let's take a look at the agenda. We 2- want to finish by the scheduled 12:15. And if we back up,.I 3 would like to leave 15 minutes for III, the subcommittee 4 discussion.
5 And if we end up with this next topic C, that Mr.
6 Bessette is going to cover, we want to use 20 minutes. Then i
7 20 minutes for D and five minutes-9 piece for the next.two 8 items.
9 So David, you will finish up with this at 11:30.
10 MR. BESSETTE: Okay. By definition.
11 I have a lot of slides, so we ought to be able to 12 get through these quickly. ,
13 There are our three activities again -- B&W 14 testing, experiments and analysis, and modeling.
15 Under B&W testing, there are three projects.
16 There is the MIST integral facility; University of 17- Maryland Loop; and the future or proposed Once Through Steam 18 Generator program.
19 MIST. This pertains to MIST.
'20 There are 50 test on small break LOCA done under 21 Phase III. There were some additional tests funded by one 22 of the utilities.
23 And basically what this has done is applied a 24 small break LOCA data base comparable to what we had done 25 under semi-scale and other facilities for the Westinghouse O rie ee a verei 9 cereer tie-(202) 628-4888
l 94
(
) 1 and CE designs.
l 2 So this extends the small break LOCA data base to 3 include B&W.
4 Following this Phase III there was some desire for 5 some additional tests on transients. These are highlighted i
6 by some different transients that occurred in B&S plants.
7 And that was MIST Phase IV. And that included eight tests.
l 8 And the MIST contract ends this December.
1 9 There is one final test remaining to be done in 10 MIST. So it's 58 done, one to go.
11 The type of tests run in MIST: There was facility 12 mapping, effective boundary conditions, effective break 13 location, feed and bleed, steam generator tube rupture,
/'s 14 noncondensibles, effective reactor coolant pump operation.
(-)
15 Phase IV there was again sma31 break LOCA without 16 high pressure injection. All the Phase III tests included 17 normal HPI. Some more scaling tests, station blackout, 18 intermediate-size break and secondary side effects.
19 MR. J. LEE: I have one quick question for you.
20 In these, this series of tests, do the 21 participants perform pre-analysis of tests as well as post-22 test analyses? Or is it primarily a post-test analysis 23 that is performed?
24 MR. BESSETTE: I think it is some of both.
25 Richard?
e
( Heritage Reporting Corporation (202) 628-4888
95
) 1 101. R. LEE: Yes. - We did both.
2 MR. WARD: Keep going.
3 MR. BESSETTE: This is status reporting.
4 All the data reports are in for the Phase III.
5 The code calculation results are in preparation-and due in.
6' June of this year. So.that's next month.
'7 Post-test analysis. The RELAP5 calculations are-8 complete and the TRLC calculations are scheduled for the end 9 of this fiscal year.
10 Turn to University of Maryland Loop.
11 The University of Maryland was oriented toward 12 looking at different scaling approach, looking to'some of 13 the atypicalities of MIST.
14 We just had a meeting on May 9 to look at.the MIST 15 atypicalities and how University of Maryland addressed 16 those.
17 We had Peter Griffith and Ishi there.
18 The results were discussed from the University of 19 Maryland tests on natural circulation on different two phase 20 inventories. And it showed basically good agreement between 21 the Maryland results and the MIST results as far as when you 22- transition from one type of natural circulation to another 23 and from single phase, two phase and boiler condenser.
l24 MR. SCHROCK: You say that the scaling is okay in 25 both cases?
O aerie 9- eereine corger tion (202) 628-4888
96 1 MR. BESSETTE: We understand the scaling effects 2 well enough.
3 So this just goes into more detail on the meeting 4 at the University of Maryland.
5 There has been a report prepared by University of 6 Maryland on preparing the results from the two facilities.
7 And the content of this report is shown in this slide. That 8 is the outline.
9 And it addresses the four, basically the four 10 atypicalities being looked at. That is the reactor vessel 11 vent valve, downcomer effect of the pipe diameter and the 12 wall heat capacity.
13 And this for reference just gives the list of 14 experiments done at the University of Maryland.
O 15 MR. CATTON: How many TRAC calculations are going 16 to be done?
17 MR. BESSETTE: Altogether cn MIST and University 18 of Maryland?
19 MR. CATTON: Well, I remember at the outset the --
20 MR. WARD: Ivan, we have to keep going here.
21 MR. BESSETTE: It's something like a half a dozen. ,
22 MR. WARD: Okay. {
23 MR. BESSETTE: And there will be some counterpart 24 calculations done on the facilities. ]
i 25 MR. WARD: Okay. I think you answered it. That's !
Heritage Reporting Corporation (202) 628-4888 i
I
97
' 'T 1 cood.
~
(_)
2 MR. BESSETTE: Now, the third part of the B&W 3 testing is the OTSG.
4 We have done air-water tests at INEL and we have 5 also had the test advisory group, or technical advisory 6 group, to plan the larger OTSG program.
7 This was the air-water facility at INEL and also 8 they had a, that was about a 161 tube, I think it was, and 9 then they had a single tube steam water facility, to look at 10 these phenomena, the feedwater penetration, distribution, 11 spreading at a tube sheet, carryover and the final three 12 phenomena here were from the single tube looking at vapor 13 generation, condensation and heat transfer.
,_ 14 And from this there has been an aux. feedwater 15 model developed and that is being put into the codes right 16 now.
17 Now, this slide pertains to the technical advisory 18 group to plan the data needs, to look at data needs for OTSG 19 and to plan the facility design.
20 The conclusions, I think these have been presented 21 before.
22 For most phenomena, there is sufficient data where 23 data needs have been identified, for seven phenomena. Part 24 of the OTSG process was to do sensitivity studies on the 25 effect of the aux. feedwater flow distribution and spray
/" ' Heritage Reporting Corporation i
(202) 628-4888
l i
l 98 l 1
. r~~ '
1 condensation. l
(- l 2 This shows the seven phenomena identified, the 3 data required or data desirable.
4 Most of the, five of these seven phenomena 5 associated with say rapid depressurization events on the 6 secondary side or high steam flow events such as a steam 7 line break, and the idea is that such data ate needed in 8 order to do best estimate analysis.
9 There is a desire to have both a composite 10 facility, that is sometning like MB-2, in order to get say 11 integral effects on the steam generator secondary side.
12 There is also a desire to have separate effects, 13 experiments in order to do model development for the codes.
- 14 The most recent meeting was on March 17 where the 15 parties endorsed the TAG findings and reaffirmed the 16 commitments on the NRC side to spend up to $2 million and on 17 the B&W Owners Group side to spend up to $2 million for a 18 total of $4 million to build the facility.
19 There is also a need to develop facility cost 20 estimates. And those cost estimates are in progress right 21 now and are being developed by INEL and by B&W.
22 MR. SCHROCK: Where vill that facility be?
23 MR. BESSETTE: Chances are separate facility would 24 be located at INEL I think and the composite facility would 25 be at Alliance, is my guess.
r' Heritag4. Reporting Corporation
(-)/ (202) 628-4888
l l 99 i
(~}
\_/
l Now we turn to the next program element, 2 experiments and analysis.
1 3 That includes three projects. There is 2D/3D at 4 BETHSY which is basically a Westinghouse 3-loop facility.
i l 5 It is built and run by the French, ROSA-IV which is 6 basically a Westinghouse four-loop facility where four loops 7 have been combined into two.
8 This is the 2D/3D status.
9 The experiments in two Japanese facilities are 10 completed. In the German facility, UBTF, just 26 out of the 11 planned 30 tests have been run, and the last four tests will 12 be run this Summer. The last four tests deal with the B&W.
13 This is where the analysis stands. The SCTF Core o 14 I facility is complete. The thermal mixing tests on B&W, or 15 in UPTF is complete. This is a hot leg CCFL test in UPTF.
16 SCTF Core II, CCTF Core II, and so on.
17 Basically what all this is saying is that most of 18 the work is complete.
19 This is showing the things still in progress.
20 You see different results. UPTF contributed to 21 the PTS resolution. UPTF showed for reactor conditions that 22 in the reflux condensation mode we are far away from a CCFL 23 limit in the hot leg.
24 Also in the Japanese facility it showed that you 25 tend to get a uniform temperature profile across the core
/)
('
Heritage Reporting Corporation (202) 628-4888
<=
l 100'
( ) 1 during a reflood even if you have a skewed, core power 2 skewed; the cross flows cause additional cooling in the hot 3 portions of the core such that you get a fairly uniform 4 reflood and temperature distribution across the core. So 5 there is, whether you have a large peaking-factor in the 6 core does not really matter too much.
7 And the data has been-used for the code 8 assessment.
9 This shows the remaining work to be done. This is 10 what is in progress, is a final overall report summarizing 11 the ten-year 2D/3D program effort.
12 And this shows the final publication is scheduled 13 for February of 1991.
(~Y 14 This shows al reports have been -- the hierarchy o
%)
15 reports under 2D/3D, at the first level of detail you have 16 thousands of reports. Then you get into summarization and 17 overview and you get down to, the next level is 500, then 18 30, then this one final overall report.
19 This is the BETHSY facility, which is located i 20 Grenoble, France.
21 It is a 1 to 100 scale. So it is, in terms of 22 these facilities, a relatively large scale. It is about 23 half the size of ROSA-IV.
24 Our agreement with the French calls for us to 25 calculate four experiments and the French, using their own O erie 9-(202) 628-4888 rore1== cerrer eie-
101 1- code, would calculate the same four experiments. And we 2 would get together to compare results. .And'it was code to 3 code benchmark. You try and understand where one code does-4 well and the other one does not, the reasons for it.
5 Basically, the tests selected are counterpart 6 tests to ROSA-IV, or areas of the. code that we have known in 7 'the past to be deficient and we want to assess with the 8 latest version.
9 MR. WARD: What is the name of the French' code?
10 MR. BESSETTE: CATHARE.
11 MR. WARD: Okay.
12 MR. BESSETTE: We have received the data from the 13 first of these four experiments. We will be using RELAP5, 14 MOD 3 to calculate these experiments.
15 And this just shows when we do the other three 16 experiments.
17 This is the ROSA-IV program. This is the Japanese 18 facility.
19 This is 1 to 48 scale. These. steam generators are 20 about four times the size of BETHSY. There have been 41 21 experiments done so far. There have been a variety of small 22 breaks, different break sizes, break locations and break 23 orientations, that pretty much tell you everything you want l 24 .to know about small breaks.
25 We periodically go to the Japanese when we have an <
l l
Heritage Reporting Corporation (202) 628-4888 i
102 1'
1 issue:that'we'want to have addressed. We have no domestic 2 facilities ~and when something does'come up we normally.
b L 3 request the Japanese to run' experiments.
4 MR. WARD: Could you give us a quick example of-5 one that's gone there?
6 MR. BESSETTE: . Things like the loop seal behavior
? during a small break, the' core uncovery and recovery, 8- instrument tube line rupture.
9 MR. WARD: Okay. That's good.
'10 MR. BESSETTE: You know, also I guess the 11 depressurization to avoid direct containment heating, as.
12 showing up there.
13 I will skip these next few.
14 Just talking about these-issues of loop seal O 15 clearing, different phenomena that have been addressed in 16 ROSA-IV.
17 We are using the ROSA-IV results to do 18 developmental assessment on both RELAP and TRAC as part of 19 the co-development effort.
20 So all these programs are tied together. They are 21 all tied by means of the code.
22- And let's see. You have a bunch of slides that
=23 you can look through.
24 We finish up with the third of the three program 25 elements or activities,.is modeling. This includes three
(- Heritage Reporting Corporation (202) 628-4888
i 103 O 1 ereseet - rhere i ra^c, azt^> =a **e c^r vroer -
I 2 MR. J. LEE: Excuse me. Is there a difference 1
3 between TRAC PF-1 and TRAC PWR? '
4 MR.'BESSETTE: No. TRAC PWR just is convenient, 5 so you don't have to specify exactly which version of TRAC 6 you are dealing with.
7 This current TRAC-RELAP development effort began 8- in 1987. We obtained a consensus amongst the ICAP members, 9 those using the code,-and the general approach as to how to 10 resolve these deficiencies.
11 And there was a general feeling that the codes as 12 they stood in 1987 were not as good as they could or should-13 be.
( 14 Developments should be complete in June of this 15 year. There's a meeting on June 7 and 8 to finalize the 16 development of assessment effort.
17 Code release is due in October. We plan a two-18 year debug and assessment effort under ICAP during 1990 and 19 1991. And we have no further development plans, that is no 20 plans for TRAC-PF1/ MOD 3 or RELAP5/ MOD 4.
21 This shows the people contributing, or the 22 organizations contributing on TRAC. It is primarily United 23 Kingdom, Japan and we expect to get some work done by CRAY.
24 Let's see. The next few slides juts give the list 25 of models being worked on and the schedule for that work.
O = erie ee aegerei 9 cereer eie-(202) 628-4888
104
,,)
( 1 And you can see some of the things, the code did 2 not conserve momentum before, it should now. And that was 3 fairly important in transition regions between like the cold 4 leg and the downcomer.
5 You will be happy to know we finally got the old 6 A&S decay heat curve out. It's not in there in any form 7 now, I hope.
8 And this is a similar list for RELAP, it covers a 9 number of different problems in the code that are associated 10 with large break and small break.
11 And RELAP and TRAC are both on a similar time 12 schedule for release October 1.
13 And now this final slide is just to give the ICAP 14 status. ICAP is due to continue until 1991.
(~))
15 There is a general feeling amongst ICAP members 16 that there should be some follow-on program associated or 17 tied to the codes and to their use or application.
18 MR. WARD: Very good. That was a good summary.
19 Let's see. Lou, you have a short presentation, 20 then back to Dave.
21 MR. SHOTKIN: You want to know the programs being 22 completed and the justification.
23 MIST we are completing this year. We are 24 finishing that. The justification is just that we are 25 finishing it. We have done all the plan testing and
() Heritage Reporting Corporation (202) 628-4888
105 1 analyses.
2 The OTSG testing we have not started yet but we do 3 expect once we start.to finish about 1992. And that would 4 complete all the B&W testing that we know about or that we 5 plan for.
6 2D/3D is scheduled to, the testing will end fiscal 7- 1989. The final reports, well, David showed.1991. And the 8 note that the accident management testing proposal was being 9 evaluated which could extend our participation in UPTF 10 beyond 1990.
11 ROSA-IV and BETHSY, both of those will come --
12 well, ROSA-VI in 1992, I think BETHSY ends in 1991, and that 13 is just the end of the formal agreements.
14 We might want to extend some participation in ROSA 15 forward on a just as needed basis.
16 The codes, the three codes shown will be finished 17 this year. These are the last planned versions of the 18 codes. And then they will go to ICAP and ICAP will finish 19 in 1991 and they will complete assessment of these last code 20 versions.
21 And then as David said, that is a new code 22 improvement program that has been proposed to the ICAP L 23 members.
24 And that is the summary of what we are completing, 25 between now and 1992.
O Heritage Reporting (202) 628-4888 Corporation l
106 1 MR. CATTON: How much is NRC's participation in (f
2 ICAP costing?
3 MR, SHOTKIN: I believe it is about 600K a year, 4 something like that.
5 We have it roughly split equally between Idaho and l.
l 6 Los Alamos.
7 Idaho does RELAP and TRAC BWR and Los Alamos does l
8 TRAC PWR.
l l 9 MR. SHROCK: What is the total of the ICAP budget?
10 MR. SHOTKIN: We don't share the cost information 11 that the other countries are doing. But if you estimate
- 12. about 50K a calculation and 14 countries and each one is 13- doing about three a year, it's whatever 45 times 50K is is 14 like the total budget.
15 MR. WARD: Okay.
16 MR. BESSETTE: One more slide this time.
17 In dealing with the future, I think you have heard 18 this already, really it is just maintain the current codes 19 and that the RELAP5/ MOD 3 TRAC PF1/ MOD 2, the BWR version of 20 TRAC and the RAMONA code which is used for reactivity 21 transients and BWRs.
22 There is the desire to keep a window on what the 23 foreigners are doing, the foreign thermohydraulic research 24 programs. And the idea would be to do this via a program 25 like ICAP to give us access to foreign facilities and Heritage Reporting Corporation (202) 628-4888
f , ;i: M l'T' ,.
w 107 p )( 1 foreign expertise. .And we would do this with'a multilateral 2 program based.on TRAC and RELAP.
3 The'next element is' university-experiments and we L 4L wouldfexpect this would constitute about 25 percent or so of-5- the. total funding. So that's out of this $3 million a' year 6 spending plan that about $800,000 would go to universities.
L 7 And then there is the reactor applications program y
8 element. And that is looking at doing analysis of operating 9 events, analysis of advanced light water-reactors, 10 regulatory issue resolution such as this'long-term loop seal-11 behavior following a LOCA, things that just arise like that.
12 Synthesis of research results and development of 13- plant models, this things you've heard about. And the 14 simulator support.
}
15 MR. WARD: Now, the simulator support, that is for 16 the Chattanooga simulators. And does this include 17 development of new programs?
18 I guess what I am driving at, don't the simulator ^
19 . vendors, haven't they all developed new programs for the 20 simulation programs that are --
21 MR. BESSETTE: I think they have all improved 22 .their modeling. But it is still now as detailed as'what you 23 have in RELAP for example.
24 MR. WARD: Well, no, but it never will be, right?
25 MR. BESSETTE: So the idea is to do benchmarks for
\
.O rit e- ee ei e career tie- .l (202) 628-4888 ;
4
o l
L 108
-[} 'l particular transients to see how the simulator answer
-2 compares with the RELAP answer.
3 MR. WARD: Oh, that is what this program is going-4 to be?
! 5 MR. BESSETTE: Yes.
. 6 MR. WARD: This isn't developing all new software 7 for those, but using what is in there now and seeing how 8 good it is?
9 MR. BESSETTE: Yes.
10 MR. WARD: Okay.
11 MR. BESSETTE: so we get a better view as to what 12 limitations do simulators have, where can they be.used.
13 MR. WARD: Okay. I understand.
14 Where does your keep the expert cadre happy come 15 in? Is that the first one?
16 MR. BESSETTE: Yes.
17 MR. WARD: We talked about several specific items 18 and you were asking for suggestions of things that might be 19 done.
20 But is that all included in your first bullet up 21 there?
22 MR. BESSETTE: Yes, it would be.
23 MR. WARD: Ivan was next, I guess.
24 MR. CATTON: The codes that you are going to 25 maintain don't necessarily have the capability to deal with
/~ Heritage Reporting Corporation (202) 628-4888
109
'l' the multi-dimensional instabilities that you have in BWRs.
)
2 What are you going to do?
3 We can ask that same question this afternoon. But 4 you are putting up here what your future program is. And I 5 don't see a capability to really handle the BWR stability 6 problem and I don't see any mention of what you are' going to 7 do about it.
8 MR. BESSETTE: By multi-dimensional do you mean 9 the non-uniform?
10 MR. CATTON: The axi-symmetric or symmetric mode 2 11' type. instability.
12 MR..BESSETTE: Yes. Providing the code validation 13 goes well, I mean, RAMONA should have that capability.
14- MR. CATTON: RAMONA has poor thermal hydraulics O 15 and good 3D kinetics, at least as I understand it.
16 The other codes have poor kinetics and 17 questionable thermal hydraulics for the kind of problem you L 18 have to look at.
19 MR. BESSETTE; The program we have on-stability 20 right now in progress in looking at getting a better handle 21 as to how reliable these codes are.
22 MR. CATTON: But where -- and I understand what 23 you re doing now and we are going to hear about that.this 24 afternoon.
25 But sort of anticipating, where are the -- where Heritage Reporting Corporation p O (202) 628-4888 l
o 110
]
. f- s .
(,) 1 'does the money come from to do what you'have to do in order. '
2 to predict the BWR stability?' Or do you think-that 3 everything is already done?
4- MR. BESSETTE: I don't think --
5 MR. CATTON: I don't see anywhere to put it here.-
6 MR. BESSETTE: Of course it all has to be shoe-7 horned into this item.
8 MR. CATTON: I would hope it would not come under 9 a category. called " maintain." If it does, then I don't 10 understand the meaning of the word.
11 I just wanted to raise the issue. The rest of it 12 can wait until this afternoon.
13 MR. BESSETTE: I guess probably what we need to do.
14 is go through the current stability work we are doing to get-15 a better view as to what kind of shape these codes are in.
16 MR. WARD: Just quickly, does that include l
17 development of a new code?
18 MR. BESSETTE: No. These plans assume we don't 19 have.to do a lot of new work relating to stability as far as
.20 the codes go.
21 MR. WARD: I guess that's what we' re going to 22 hear, that it's not, the resolution of that problem isn't 23 dependent on developing a code that will deal with the whole 24 issue.
25 MR. BESSETTE: Yes. This is all predicated on the
( Heritage Reporting Corporation (202) 628-4888
111 l 1 idea that there aren't any real big problems with the codes.
v J-2 MR. WARD: Well, I'm not.sure. There's sort of a 3 problem with the codes in that it can't deal with this, issue 4 very well.
5 But it may be that you choose to deal with that 6 issue by some means other than a code, than a big systems 7 code. And I thought that is what you were really saying.
8 But maybe we will hear more about that this afternoon.
9 Do you see what I mean, the difference?
10 MR. CATTON: You can always design around the 11 problem.
12 MR. WARD: Well, you know, we did a lot of 13 designing before we had big system codes. They aren't 14 necessarily the answer to every issue that comes along.
('S
\m/
15 MR. CATTON: But somehow you have to delineate the 16 stability.
17 MR. WARD: That's right.
18 MR. CATTON: What they are doing is they are using 19 these codes and they are having troubles with these codes.
20 They take too damn long to run, or this reason or that 21 reason.
22 Yet, I don't see anything here that allows for 23 something they might have to do to them.
24 You could do it by test. You don't have to 25 necessarily use the codes. It might even be cheaper to do l l 1
() Heritage Reporting Corporation (202) 628-4888 t
}l -
1 112 r /'y 1 it by. test.
! y o 2 MR. WARD: Is there a simple answer to that 3 question?
4 MR. SHOTKIN: Yes. You will hear it later..
5' MR. WARD: Okay.
6 MR. SHOTKIN: But the stability boundaries, 7 Professor Catton, are being defined by Jose Marks Luger and 8 the LAPUR code.
9 MR. WARD: I missed that Lou. I'm sorry.
-10 MR. SHOTKIN: I know. It's the stability' )
11 boundaries are being defined by the LAPUR code, Jose Marks 12 Luger of Oak Ridge. We have a. joint program with NRR. He's 13 . funded by NRR.
14 We are taking the results of his stability work-
' () 15- and we are doing our analyses using three codes -- the l
16 Brookhaven Engineering Plan Analyzer, we are using TRAC BWR l 17 to look at that a well as the owners group and GE are using l 18 TRAC'BWR, their version, and we are using RAMONA.
19 And we meet periodically with the technical j 20 program group where we look at the results of all of these 1 1
21 codes and come to some general consensus of what are the ;
4 22 physical phenomena going on.
23 MR. KERR: Would his code have predicted the 1
24 LaSalle incident?
25 MR. SHOTKIN: I think that we have been able to Heritage Reporting Corporation O* (202) 628-4888 1
?
- _ _ _ _ _ _ _ _ _ - - . _ _ _ - _ _ - - - _ _ _ _ - - - - _ - _ - _ - - - - . . ~ - _ _ - . _ _ - __. _ .
li 113 f( -1 predict.the boundary of the LaSalle incident with LAPUR and 2 I have seen it with RAMONA.
- 3. So I think at le,ast with those-two.
'4 MR. WARD: Well, we are going to hear more about 5 that this afternoon, right?
I, 6 MR. BESSETTE: Yes. this is one case'where.we do l
7 -have a' simple, faat running code that has proved quite 8 valuable.
10 11 12 13 O
15.
16 17 18 19 20 21 22 23 24 25 Heritage Reporting Corporation (202) 628-4888
--u____._m__________ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ . _ _ _ . _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . - _ _ _ _ _ _ _ _ _ . - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _
1 ,,
l 114 1- MR. WARD: Okay. Virgil had a question. ,
2 MR. SCHROCK: I was going to ask_about these 3- resnits. It seems to me that that is the kind of activity 4 that would be highly suitable for experts that you are
.5 maintaining at the national labs. Who is doing that?
6 MR. BESSETTE: That is being done at IENO. The 7 reports so far that they have produced, there are two.
8 There was one on natural circulation and feed and bleed, one 9 on feed and bleed.
10 MR. WARD: That is not out, is it?
11 MR. SCHROCK: Feed and bleed is out.
12 MR. BESSETTE: Natural circulation was in draft 13 form.
14 MR. SHOTKIN: Natural circulation we did not like
-O' 15 and we are having them redo it.
16 MR. BESSETTE: But the idea is that you take 17 something specific like feed and bleed and you look at the 18 type of experiments that have been done and you look at the 19 code capabilities and make some general assessment of the 20 state of the art.
21 MR. SCHROCK: But in your thinking on the various 22 items, you do not associate that with maintenance of the 23 expertise in the labs.
24 MR. SHOTKIN: Yes, we do.
25 MR. BESSETTE: Yes, I think we do. I think that
,q Heritage Reporting Corporation
() (202) 628-4888
l 115 i
? I all of these really or most of them are maintenance of 2 expertise but not all, maybe half.
3 MR. WARD: Okay, all set. Now to the next topic.
4 MR. SHOTKIN: This is actually the name of the 5 topic that you put on the agenda. And we have given you the 6 Commission paper, the draft Commission paper. We are not 7 scheduled to go to the Commission with this Commission ~ paper 8 the week of July 17th. And if'you will look at the very 9 'end, we have a blank space that says comments, that the ACnS l
10 has reviewed this and comments are as follows. We have 11 given you a copy of the branch plan that came out last year l 12 that you can look at in conjunction with the Commission l-13 paper.
14 And I think that everything that we have talked
.O 15 about today is covered in the Commission paper, and we would 16 just like your comments on these long range plans for the 17 thermal hydraulic program. We have been asked to go to the 1
18 Commission with this and we have determined what the base 19 line programs are four or five years from now and how we are 20 going to get there, and we would just like your comments on 21 it.
22 MR. WARD: All right. You were just going to 23 spend five minutes or less going over your response to our 24 last letter.
25 MR. SHOTKIN: Sorry about that.
Heritage Reporting Corporation O (202) 628-4888
_ _ _ _ _ - _ _ _ _ _ _ _ ____________- i
l 1
116 rm 1 MR. WARD: We do have that in our package.
b 2 MR. SHOTKIN: I am sorry. I jumped up too soon.
3 I forgot about that one. We have responded to your letter.
4 We have responded to each of the comments.
5 Do you have any particular one that you would like 6 to discuss?
7 MR. WARD: Yes, that is a good way to do it. You 8 have got that package in front of you, gentlemen.
9 MR. CATTON: The steam generator tube rupture 10 comments by the ACRS came about because of I believe 11 North Anna. And the letter essentially says that you agree 12 that Westinghouse has done a fine job in defining it as 13 fluid elasticity. Maybe I did not read the right documents.
,. 14 But when I took a look at what Westinghouse had done, I
15 Westinghouse used a potential flow to calculate a velocity 16 and that is kind of silly.
17 It seems to me that there are some real research 18 questions. Now if they have no safety relevance, that is 19 one thing. But we are seeing more and more steam generator 20 tube ruptures. Yet the phenomena that cause them cannot be 21 described all that well.
22 MR. SHOTKIN: Actually I anticipated you asking 23 that and I did come prepared. I do not know whether you are 24 going to like the answer. We work closely with your 25 regulatory staff. And in general before we go off and look
(
s Heritage Reporting Corporation (202) 628-4888 l
l
117 1 at a problem and before we define it as a safety problem
}
2 worth looking at, we work with our regulatory staff to see 3 what their interest is.
4 Now in North Anna, there have been two bulletins 5 put out by NRC. The first one was 8802 that did address the 6 problem that you are talking about, the steam generator tube 7 rupture because of flow induced vibration. That was issued 8 in 8802. And what it said was this could be a problem in 9 your plant, please looking for denting in your tubes when 10 you go down, and when you find denting plug the tubes.
11 And I checked with Mr. Chang, the NRR person 12 responsible for this, and he said everyone responded, and 13 people did look and they did find tubes that were dented and
,- s 14 they did plug them.
( ')
15 What you are saying is interesting and is 16 something that we should be looking at. But 17 bureaucratically the way that it works is that this 18 mechanical engineering branch would ask Wayne Hodges' branch 19 within NRR to help them understand this flow induced 20 vibration in the Westinghouse document. And then if Wayne ,
21 wanted results or help in that, he would ask us. Neither 22 has taken place, because the bulletin went out and the NRR 23 was satisfied that if there is a safety issue that they have 24 headed it off with this bulletin.
25 Now since that happened, again I think at the same
(~') Heritage Reporting Corporation L> (202) 628-4888
118
() 1 ' place at North Anna, they had one of these tubes come out.
L .2. So they. issued another bulletin just this past week, 8901, 3 that says failure of Westinghouse steam generator tube 4 mechanical plugs. And there are about 23 plants, and they 5 are asking,them to look at their plugs,-and there is a 6 certain kind that they have to look at, and if they have 7 them to either replace or repair them.
8 We do have a research program going on to look at-9' the effects of a plug coming out. Now this is not the flow 10 induced vibration but it is connected, because to fix the 11 flow induced vibration they look for denting and put in a 12 plug. Now they are worried about the plugs coming out. And 13 actually Joe Hopenfeld of the Office of Research is 14 sponsoring a program to look at the effects of what happens
}
15 when a plug comes out. And if it comes out of one tube, how 16 will it interact with other tubes.
17 MR. CATTON: I think that you missed'the point 18 that I was trying to make, and maybe I did not read the 19 right documents. But the Westinghouse documents describes 20 the process that they went through. They changed the 21 recirculation ratio to reduce the velocity across the tubes, 22 thereby getting away from the question of fluid elastic 23 instabilities. Now if they knew what the velocities were 24 that caused them, then that would be a reasonable thing to 25 do.
Heritage Reporting Corporation (202) 628-4888 i
119 11 I maintain that they do not know what those .
b>)- 2 velocities were, and they do not know that they were in the 3 fluid elastic regime, because the method of computation was 4 not proper for that problem. And I thought gee, this is an.
5 opportunity for research to do something meaningful.
6 MR. SHOTKIN: We have looked at your review of the 7 Westinghouse document, and we do not have any problem with 8 your view. In fact, we think that you are probably correct.
9 The question is is this something that requires research to 10 look at, or can the regulatory staff come up with a 11 resolution. And the regulatory staff feels that they have 12 the resolution with these two bulletins, 8002 and 8901.
13 MR. CATTON: So they do not deal with the question 14 of changing the recirculation ratio at all?
15 MR. SHOTKIN: I would have to look at the 16 bulletin. I think that is true.
17 MR. CATTON: If I heard what you said, they are 18 just plugging tubes, or unplugging tubes, or changing plugs 19 in tubes.
20 MR. WARD: It does not have anything to do really 21 with the issue that you raised.
22 MR. CATTON: That is right. And now you have got 23 advanced reactors coming along with tube generators, and you 24 have no way of assessing whether or not they are going to 25 get around this problem. Your present codes cannot deal p
< >g Heritage Reporting Corporation (202) 628-4888
o 120
[ '1 with it.
2' MR.'SHOTKIN: That is correct. But in the 1
3' edvanced reactors, I think~that we'will have a chance to 4 look at this.
u 5 MR. CATTON: It is sort of like now is the time to 6 be-sure that you have some kind of a small' program going 7 that will give you the basic information that you need, and 8 I do not see it.
9 MR. WARD: How about giving us a paragraph for the
- 10. letter that we are-going to write on this, because we are 11 going to write a letter in June.
12 Are there any other questions on the letter'that 13 we wrote last year and the responses?
14 (No response.)
' O 15 MR. WARD: . We did have a comment about the loop 16 seal clearing problem and so forth. I guess that I was a 17 little confused. There seems to be a program in ROSA-4 to
.18 deal with that, is that right?
19 MR. SHOTKIN: Yes.
20 MR. MARD: And your response, is that what your 21 response said? Well, I will check. I do not want to take 22 the time. I thought I had it.
23 MR, SHOTKIN: We have got it. I think that answer 24 still reads. Do you want me to read the answer to you?
25 MR. WARD: Yes.
Heritage Reporting Corporation (202) 628-4888
1.l4 I- '
121 I 1 MR. SHOTKIN: "The issue of Vop seal clearing was 2 considered for large break LOCAs with the results showing 3 satisfactory long term cooling," and then we refer to a EG&G 4 report, EGG-TFM-07fi93. " Specific tests were run in 5 ROSA-4 at the NRC request to confirm this judgment. We'are 6 not aware of any other area where long term cooling is a 7 particular concern."
8 MR. WARD: But I thought I saw.in your present 9 program that there are still tests to be done in ROSA-4, or.
10 are they call completed, I mean what they have just 11 presented to us?
12 MR. SHCTKIN: I think that they are finished. I 13 think that he was just taking credit for it.
14 MR. WARD: All right. Okay, thank you.
15 Let'me see, what else do we have that you were 16 going to present?
17 MR. SHOTKIN: We are finished.
18 MR. WARD:- That was it, okay. That is great.
19 MR. CATTON: I have a question.
20 MR. WARD: Sure.
21 MR. CATTON: On the mist and the once through 22 steam generator studies, you guys were going to check into 23 the Oak Ridge work. And I guess that I supplied somebody 24 with the person's name and phone number. Did you do that?
25 MR. SHOTKIN: Yes.
Heritage Reporting Corporation (202) 628-4888
m 5 122 y MR. CATTON:
11 Did they do_you any good?
o a y f 2 MR. BESSETTE: I do not know the answer ~to_that, 3 but they.are being read right now. I am not sure that there i
4 is any conclusion yet.
5 MR. CATTON: Because I called the guy too, and he 6 told me'that they had done modeling and everything on 7 vertical tube operators'which are once through steam
- 8. generators.
9 MR. BESSETTE: They were sponsored by the Office 10 of Saline Water.
11 MR. CATTON: That does not make it bad. Our 12 department chair says that what we do is bad because it is 13 nuclear engineering. They did some good work.
14 MR. LEE: One quick question regarding your
("~)- 15 specific Recommendation 8, and is to do with RELAP-5 16 Mod 2, whether it is going to have boiling water reactor 17 capability.
18 Do I understand correctly from your response to 19 that recommendation that it will not have BWR 20- capability?
21 MR. BESSETTE: We are not doing any development 22 focused on BWR water, but one can do BWR analysis with 23 RELAP. I mean people use RELAP for BWR analysis. But our 24 development effort does not consider BWR specific issues.
25 MR. LEE: Like jet pump models and so on are not Heritage Reporting Corporation
\ (202) 628-4888
123
=1- there yet, so BWR analysis would be fairly limited with 2 RELAP-57 3 MR 'BESSETTE: It was not-part of the development 4 plan to make RELAP fully applicable in the BWRs. -
5 MR. LEE: I am just-trying to understand your -
6 response. . I am not criticizing or anything like that. So 7 RELAP-5 Mod 3 will still have a limited capability regarding 8 BWR analysis.
9 MR. BESSETTE: Yes. . Other than some general work-10' on the two phase, there is nothing directed at BWRs.
11 MR. CATTON: Will the results of the Exxon work.on 12 RELAP-5 be published?
13 MR. BESSETTE: Everything that they are.doing for 14 us is -- well, they are not actually doing too much modeling O 15 work that I know of.
16 MR. CATTON: I understand that. They were 17 essentially going to go through the same process that the 18 other group did on TRAC, were they not, on RELAP-57 19 MR. BESSETTE: Doing a CSAU, you mean?
20 MR. CATTON: Sort of.
21 MR. BESSETTE: Well, what I know of that they are 22 doing is they are doing calculations or assessments relating 23 to large break LOCA using the 2D-3D data, and from that 24 determining code performance. And that is helping to find 25 what models need to be worked on in RELAP. But they are not Heritage Reporting Corporation O (202) 628-4888
h.
f-124
( )' 1 really doing any model development.
2 MR. CATTON: I understand.that, but will the 3' results be available, or-are there any available yet? I 4 would like to see them if they are.
5 MR. BESSETTE: There is communication between the 6 people doing the work at the Union and EG&G.
b 7 MR. CATTON: Well, EG&G is doing it for them is-my 8 recollection. They put up the money.
9 MR. BESSETTE: Well, EG&G is doing the model 10 development. But in addition to that, the Union is doing 11 calculations on their own.
12 MR. WARD: Could you pursue this afterwards. I am
'13 not saying that it is not important, but I think maybe it is 14 lesser important. We have got ten minutes here. Paul, we
[}
15 have time at'the June meeting which is the week after next 16 for a review of this so that we can write a letter. We can 17 talk about what we want to have presented. This is June 8th 18 or 9th, two hours on the 8th or 9th.
19 What I would like to get in the next few minutes 20 are our consultants' comments. The ACRS is being asked to 21 comment essentially on the SECY paper that was attached. I 22 do not know if you have had a chance to read it, but it 23 covers the material-that we have heard here this morning.
24 The members of the full committee will have another crack at 25 this, but we will not have another crack at your comments O seritaee Revereine correratien (202) 628-4888
'} 1 2
unless we hear them now, so I will invite those now.
MR. PLESSET: I think that what I heard on their 3 plans for university laboratories that it is a little bit 4 naive and not well thought out. They have to think about 5 stability and endurance as well as the cost. I think that 6 they are a little bit off on those.
7 MR. WARD: What do you mean stability and 8 endurance, institutional you are talking about?.
9 MR. PLESSET: If they have some professor in mind, 10 is he going to be there for more than a year or two, and 11 that is mostly what I had in mind. And I think that they 12 are off on their finances too.
13 MR. WARD: What do you think about the idea of 14 trying to involve universities?
15 MR. PLESSET: That is a good idea. I think that-16 it is a very good idea. It could be very worthwhile, but it 17 is going to cost more than they are thinking. They need 18 more preparation than they are going to do.
19 MR. WYLIE: Would it not be more useful in the 20 long run to have that at one university instead of 21 scattering it out over four or five?
22 MR. PLESSET: Well, it might. They might be able 23 to do a better job and have more money. But I think that 24 they are trying to get some multiplicity and some diversity, 25 and I think that that diversity might be helpful. Not four O Heritage Reporting Corporation (202) 628-4888
126 1 or five but maybe two.
2 MR. WYLIE: I was thinking the ability to have 3 stability.
4 MR. PLESSET: If it were one with a big 5 investment, it would be more stable. That is a 6 problem.
7 MR. WARD: Virgil, if I can I look for some 8 written comments from you. That would be fine. And 9 anything additional from either of you that you think that 10 the full committee ought to consider in writing a letter, I 11 would appreciate very much getting comments from you. And 12 we can get that into the book so that the full committee and 13 read it.
14 Let me ask Walt and John, do you have any comments
\/ 15 that you would like to make at this time from what you have 16 heard this morning?
17 MR. LEE: A specific comment regarding the need ,
18 from my point of view to study natural circulation phenomena 19 in a close load situation, not just for the current 20 generation, not just the full B&W plant where the current 21 emphasis seems to be placed on, but also for the boiling 22 water reactors, and for advanced boiling reactors, and 23 for the metal cooled reactor systems somewhat down the road.
24 Let me try to go through these three types of 25 reactors quickly. For boiling water reactors, I thought l
f-) Heritage Reporting Corporation
(_/ (202) 628-4888 l
127.
E .(m,) I that one:of the. uncertainties that we have had regarding'the 2 anticipated' transient scram events is the expected power 3 ' level after the recirculation pumps are tripped, and that 4 has to be handled through careful understanding of natural 5 circulation phenomena.
6 Then for the boiling water reactor, at least one 7 of the designs that I have seen involves passive safety 8 features relying heavily again on natural circulation, heat ~
9 to heat removal. If not, just decayed removal even at some 10 other power operation.
11 And of course for metal cooled systems, the way 12 that I perceive the Argonne effort in their so-called safety 13 demonstration test, the tests relied heavily on the 14 experience in representing the amount of sodium involved in 15 the safety demonstration kind of test. HSo again we do'not 16 seem to have first principle modeling capability to
'17 represent natural circulation phenomena in general, and I 18 would like to see a little more effort. I understand that 19 there are some reports written and so on, but I understand 20 this to be our current status of understanding.
21 MR. WARD: All right. Thank you, John.
22 Are there any comments from any of the members?
23 (:No response.)
24 MR. WARD: As I said, we will be hearing more 25 about this at the June meeting. Okay. Let's break for
() Heritage Reporting (202) 628-4888 Corporation
- p. .
!?
n
[:. 1
'128 I 1 lunch then,fand come'back atL1:15.
l 2- .(Whereupon, at 12:15 p.m., the hearing was 3 : recessed, to reconvene at 1:15 P.m.,- this same day.)
c: .
1 .4-
- f. 5 1
6 7
-8 9
4 10 11 12 13
- 14
~ 15 .
16 17 18 19
' 20- I 21 22 23 24 25 h- Heritage Reporting Corporation (202) 628-4888
_______________--______ _ D
f 129 1 AEIEBHQQH EEERIQH (o
1 2 MR. WARD: We'll pick up again.
3 The first item this afternoon is Roman Number IV.
4 We are going to discuss the LaSalle Cure Power Oscillation 5 Event.
6 I understand Mr. Rausch is going to speak for the 7 BWR Owners Group.
8 MR. RAUSCH: Are we ready?
9 I'll basically just be giving an update on Owners 10 Group activities regarding the studies and instabilities, 11 BWR's instability.
12 This is the way we see the purpose of the meeting.
13 It's very similar to a meeting we had April 6th with the NRC 14 staff and some of the Research people. Update you on what t
-- 15 we are doing to resolve stability issues. Give you a status l
16 report on the development of long-term corrective actions, 17 which was really the main focus of our April 6th NRC 18 meeting. And can be the main focus of today, if you wish.
19 And then depending on interest, at the bottom here
'20 we've had some letter dialogue back and forth between the 21 Owners Group and the NRC and a_ clarification nature of the 22 major Bulletin Supplement 88-07 Supplement 1 that deals with 23 regional oscillations. And we've had discussions along that 24 line and we can go into those if you are interested.
l 25 And we have a disclaimer at the bottom. As an
(^g Heritage Reporting Corporation
(_) (202) 628-4888
130'
(. 1 Owners Group Stability Committee, the things we are saying 2 here' basically are pretty much our perspective and then, of-3 course, they can't represent a commitment on the part of any 4 individual utility.
5 MR. WARD: Well, is that because--is this caveat 6 because of your basic charter and relationship--
7 MR. RAUSCH: Yes.
8 MR. WARD: Or because of the preliminary nature of 9 the thing 10 MR. RAUSCH: Mostly basic charter. We are so far 11 along with the: reports and the work we've done that we don't 12 mind showing you what we've done.
13 It's just basic charter. Even when the Owners 14 Group comes up with a position, it may be a consensus 15 position, but there could be a few utilities that say "I 16 don't believe it" or "I'm not going to do what you say.
17 MR. WARD: All right.
18 MR. RAUSCH: So that's basic charter.
19 This is just a very crude agenda. A little bit of 20 time on introduction and overview. More time on results to 21 date. And really the long-term corrective action status.
22 I don't believe I need this much time. We will 23 fill the slot if you wish with questions, but I don't have 24 two hours' worth of presentation material.
25 MR. WARD: Okay. You might get a few questions O erie ee nevertiee (202) 628-4888 cereer tie-
F i
- 1 131 i
<- 1 here and there.
( -
2 MR. RAUSCH: Oh,-I suspect I might.
3 This is an overview of the Owners Group Stability
.4 Committee. Our major programs objecti Ls were really funded 3
5 in 1988. We called it Phase I Analysis, which was 6- investigating plant response to regional. instabilities. And 7 I'll update you on that.
8 MR. CATTON: Regional instabilities?
9 MR. RAUSCH: Yes. I heard some discussion about 10 Mode 2, or whatever, but basically core-wide instability is 11 ' where the entire reactor thermal hydrologically and 12 neutronically is going up and down in power.
13 MR. CATTON: In a one-dimensional sense?
14 .MR. RAUSCH: Axial dimensional sense. Yes.
' \-
~ 15 Radially they.are pretty much phase. So the entire reactor 16 throughout the core is going up and down in power at the 17 same time. It's a function of the density waves propagating 18 through each channel.
19 MR. CATTON: So when you say regional 20 instabilities, that could be an axi-symmetric mode, for 21 example? 1 22 MR. RAUSCH: Yes. I prefer to think of it as--I 23 usually describe it as the left half and the right half.
24 While the left half is at higher power, the right half is at 25 reduced power.
Heritage Reporting Corporation (202) 628-4888 l
l
132 D' 1 MR. CATTON: And then they swap?
[G 2 MR. RAUSCH: Exactly.
3 MR. CATTON: Or that ante-symmetric, okay.
4 Do you get axi-symmetric modes too where you have 5 the center going up and the edges going down?
6 MR. RAUSCH: It's never been observed to my 7 knowledge. And I also don't believe that enough code work 8 has been done to see if that's possible. It certainly--I 9 would say it's certainly theoretically possible but it may 10 be in practice impossible to make happen. I don't know.
11 And then that was our objective. To identify 12 appropriate remedial actions. And I'll step through a 13 couple of quick chronology bullets and you'll see a lot has g-) 14 been done in that area. We have identified actions. We
(/
15 have taken them of our own volition working with GE and they 16 are reinforced through Bulletin 88-07, Supplenient 1.
17 Phase IA was to be a simple study of large 18 amplitude oscillations using LaSalle type neutronics and the 19 similar base. It turned out to be a Peach Bottom base. And l
~
20 to see what happened to average power response during ATWS.
21 I can update you on those calculations as well 22 MR. CATTON: When you do your large amplitude 23 oscillations, is the initiator small in amplitude or large 24 in amplitude?
25 MR. RAUSCH: It was just a simple full 3D TRAC I
Heritage Reporting Corporation
()' (202) 628-4888 i l
1 i
j
________________d
133
()
1 calculation and there was no initiator.
2 MR. CATTON: You just used the noise and the code 3 to start it?
4 MR. RAUSCH: You don't really need anything. I 5 guess you would call it noise and the code, but it actually 6 started from this condition and oscillated on its own.
7 MR. CATTON: Considering the damping in a code-8 like track, that's pretty good.
9 MR. RAUSCH: I could show you what it looked like.
10 And then we had a host of post-Phase IA activities that 11 really started right about exactly the time we met last. I 12 believe that was, what, December 15th or thereabouts.
13 We did a formal review. Worked with EPRI, not 14 really formal, but a very detailed peer review for the work 15 we did since it was somewhat uncharted territory.
16 We've had an entire subcommittee devoted to 17 providing guidance on the interim corrective actions, and 18 how they relate to current operating and emergency 19 procedures.
20 We've had a subcommittee devoted to gathering 21 world-wide data on instabilities. And our major activity 22 has been associated with long-term solutions. I'll spend a 23 little time explaining what we've gone through to identify 24 viable long-term solutions.
25 MR. CATTON: And what version of TRAC did you use?
() Heritage Reporting Corporation (202) 628-4888
(
134 l m
.y 1 MR. RAUSCH: TRAC they call it.
2 MR. CATTON: Derivative of what? I don't think 3 the Reporter.got it.
4 MR. PFEFFERLEN: EG&Gs. The one that's in the 5 public domain.
'6 MR. CATTON: Remember when they were building 7 their advanced thermal hydraulics code. They made 8 modifications to the BWR, TRAC BWR, that was at EG&G for 9 their own purposes. How long did it take +;o make a run on 10 TRAC?
11 MR. RAUSCH: In terms of computer time?
12 MR. CATTON: Yes. >
13 MR. RAUSCH: I don't recall, Hank. It was over--
14 MR. PFEFFERLEN: I can't say. It was a long time.
15 MR. RAUSCH: Hours. Many hours. It's a function 16 of how long your time stepr are. How many seconds you run 17 the calculation out. I think the time steps were very 18 small.
19 MR. CATTON: You had to run in tenths of second, 20 didn't you, in order to get the limits?
21 MR. RAUSCH: Oh, yes. Right.
22 MR. CATTON: How many runs did you make?
23 MR. RAUSCH: Which calculation? The regional 24 oscillation or the ATWS assessment again?
25 MR. CATTON: Well, you had two kinds of Heritage Reporting Corporation (202) 628-4888
135 r~ 1 calculations.
2 MR. RAUSCH: Right.
3 MR. PFEFFERLEN: One is large amplitude 4 instability. And the second was ATWS. The ATWS you ran out 5 about 100 seconds today, if I remember right.
6 MR. RAUSCH: Right. And then there was a small 7 number of seconds run, maybe ten seconds or--
8 MR. PFEFFERLEN: Ten-second steps or so.
9 MR. RAUSCH: There's a small block made also at 10 reduced core height.
11 MR. CATTON: But how many of these runs did you 12 make?
13 MR. RAUSCH: For the Phase I calculation there 14 were six major runs. And for the Phase IA there was
' 15 really, I would say, 1 1/4 major runs.
16 MR. CATTON: 1 1/4?
17 MR. RAUSCH: Yes. The next one we looked for a 18 trend and didn't see it. And it wasn't really funded, so 19 they didn't see anything happening so they turned it off.
20 I have a lot of chronology slides. They are 21 available in your package and I won't go through them in 22 extreme detail but we found it useful in talking with the 23 staff and even amongst our own, inside our companies, to 24 point out the somewhat hectic pace we've been working under.
25 I go back to November because this is right about Heritage Reporting Corporation O (202) 628-4888
136
(): 1 the. time frame where we discovered that the margin, the CPR 2 margin, .available for' regional oscillations was not as large
'3 as we thought. That started a series of events.
4' We had a normally scheduled meeting with the NRC 5 staff. We updated them on the work we have done. .And 6 inside--less-than a week, we .iad already defined interim 7 corrective actions. We telecopied them to all the 8 utilities, and obtained commitments back on either 9 implementation or a schedule for implementation from 10 everyone inside of a week.
11 And we basically told the NRC that we had a lot of 12 areas that we wanted to work on simultaneously. Again, the 13 most important being long-term resolution issue. And.we 14 laid out our plans for that.
15 MR. CATTON: Did you ever figure out what was 16 missing from the earlier calculations that resulted in the 17 loss of margin?
18 MR. RAUSCH: There are two issues here. Are you 19 talking about decay ratios or CPR?
20 MR. CATTON: Well, there were some early 21 predictions of where the boundaries of the instability were.
22 And they were off by 20 or 30 percent or something.
23 MR. RAUSCH: There are two very distinct issues.
24 The first one is decay ratio, which is what I think you are 25 referring to. Decay ratio would be, as you could call it a Heritage Reporting Corporation (202) 628-4888 1
e 137
,m.
1 power flow map description of where you could be unstable.
- a()
2 Those calculations had a licensing basis 3 associated with them, if the plants wished to use them.- And 4 it was if you calculated decay ratio of .8 or less, you did 5 not have to have what was called stability tech specs.
6 It's turned out since then that the calculations, 7 we've shown that the model is not at fault. It's really the 6 level of detail has been used for all these years was not 9 sufficient to show, like under more extreme or different 10 power distribution conditions, your decay ratio can go up 11 substantially.
12 MR. CATTON: So you just didn't search the 13 spectrum properly?
(~g 14 MR. RAUSCH: I think you could ssy that's they way tv /
15 it was.
16 MR. CATTON: So it's not the fault of your ability 17 to calculate.
18 MR. RAUSCH: That's right. At least for in phase 19 oscillations. And, as a matter of fact, one of the first 20 questions asked by the AIT team, Larry Phillips was 21 involved, was to explain what happened to your decay ratio 22 calculations.
23 And GE went back and redid it under the actual 24 conditions as best as they could get them, which was very 25 accurate actually. They calculated 1.0 decay ratios, which
()
i I
Heritage Reporting Corporation (202) 628-4888 1
l
_ __._______ __ __J
l 138
(~3 l' is exactly what we have.
V 2 Now, that's quite a distinct issue from margin and 3 CPR. There is no real disagreement or issue' remaining for 4 critical power ratio margin during in-phase oscillations, 5 where the entire reactor is responding, simply because the 6 APRM system, which will trip the reactor, responds one to 7 one with the oscillations.
8 There is no cancellation effect. You get very 9 large instrument response and you will get a SCRAM and by 10 the time you SCRAM you haven't lost very much in terms of 11 critical power ratio.
12 MR. LEE: Are you trying to make a distinction 13 between what you call in-phase and regional oscillations?
14 ET. RAUSCH: Yes.
15 MR. LEE: You can theoretically consider a 16 situation where power oscillation would move up and down 17 without any distortion?
18 MR. RAUSCH: No. We've seen it in some European 19 tests. Very common tests. The NRC staff has seen it. If 20 you have the regional oscillation, axi-symmetric or whatever 21 you want to call it, where left and right halves are going, 22 as long as you not at very large anplitudes, it's quite 23 reasonable to postulate a cancellation,effect on the core 24 average power.
1 25 The core average power stays about constant. The
(~ Heritage Reporting Corporation j
()) (202) 628-4888 i i
_ _ _ _ _ . _ _ _ _ _ . _ _ _ _ _ _ _ _______________________________________.___Q
4 139
{
detectors on half the core, the detectors on the other half
(]) 1
- 2. the core. One is going up, one is going down.
3- Fortunately, as soon as the amplitude goes up, a 4 fair amount, maybe 50 to 100 percent of scale, you start 5 getting a non-linear effect. The rise of the power shape is 6 much steeper. The fall is steeper. And you have'a wide 7 trough. So you have a spiking effect.- And the positive 8 spike will not be cancelled out by a negative trough.
9 MR. LEE: This is a time domain problem you are 10 referring to.
11 MR. RAUSCH: Yes.
12 MR. LEE: But I thought you were trying to make a 13 distinction in the spatial oscillation essentially. When 14 you go to regional oscillation. And what I am trying to q
15 understand is-whether in your opinion you think it is 16 possible to have this time domain oscillation without 17 spatial distortion in power.
18 MR. RAUSCH: That's an awful--time domain 19 oscillation without spatial distortion in power. Yeah.
20 That's an in-phase oscillation--
21 MR. CATTON: Yes. And that's what he's referring 22 to.
23 MR. LEE: Yes, that's what I am asking you. Do 24 you think it is possible?
25 MR. RAUSCH: I'm not sure what you are
- Beritage Reporting Corporation (202) 628-4888
140
~
. 1 postulating.
2 MR. LEE: It's a simple question, Do you think 3 you can have the power level go up and down without any 4 distortion in spatial shape. That's what I'm asking you.
5' MR. RAUSCH: Oh, certainly. Yes. That's been 6 observed for a number'of years in tests.
7 MR. LEE: But then you qualify it as saying that 8 the spatial distortion is small enough so that you leave it 9 like that. That was the impression I got. So that's what 10 I'm trying to understand.
11 MR. RAUSCH: Yes, yes. And the tests support that 12 in terms of the entire reactor is acting in phase. If one 13 small part is perhaps going a little bit ahead of another 14 o
part or whatever, it's been insignificant.
15 If you correlate the local power readings to the 16 events, where they have been in phase, they are acting as if 17 the whole core were going up and down together. The 18 isolated parts of the reactor have been going up at the same 19 time as you would expect. And the left' half and the right 20 half are going at the same time.
21 There is a phase- you know, it is a density wave 22 phenomenon. It comes up from the bottom. Goea to the top.
23 I think it's roughly a half a second transit time for the 24 density wave to propagate, from core bottom to core top. So 25 you have an axial phase-like.
Heritage Reporting Corporation O- (202) 628-4888
141
~
.y r- 1 MR.' LEE: Which means that your spatial distortion 2 in power should be inherent.
3 MR. RAUSCH: Axially spatial, yes. But it's'not a 4 cancellation type. It's not phase-like enough where you are 5 really getting a major cancellation. You are not getting 6 100 percent of what one detector is doing, but you are 7 getting 70 or 80 percent.
8 MR. LEE: But that already implies that I thought 9 there was a spatial distortion in power distribution. Even 10 when you are talking about in phase oscillation.
11 MR.-RAUSCHe Well, averaged over time, there is 12 not.
13 MR. LEE: Pardon me?
15 given seco ,
16 MR. LEE: Well, I'll go into that later on.
17 MR. PFEFFERLEN: Is the question do you get some 18 mitigation of the signal because of this axial distortion, 19 and the answer is that was accounted for in the study that 20 was done in Phase I where we did take the phase of the power 21 distribution through the core as if you are looking at the 22 different levels of LPRM.
23 If that's your question, we did try to account for 24 that.
25 MR. LEE: I wasn't questioning whether you did
'p Heritage Reporting Corporation (202) 628-4888
'h_ __ ________--_._____________________________.-________._________-.____________m ._m_ _ _ _ _ _ _ __..___-__m - _ _ _ _ _ _ _ _ _ _ --.___-._._________._____-___________.___________._______-m. . - - _
142 1 account for it or not. I' d like to understand the physics
[^))
~~
2 behind it a little bit better. That's all I wanted to know.
3 Indeed, the spatial distortion cannot be separated from what 4 you'd like to call in phase oscillation whatsoever. It's 5 inherent in the oscillations.
6 MR. RAUSCH: Certainly.
7 MR. LIPINSKI: All your 64 in-core detectors were 8 behaving in synchronism.
9 MR. RAUSCH: 172.
10 MR. LIPINSKI: 172, okay. Now, what is their time 11 response? Are they filtered or do they give you rapid 12 response?
13 MR. RAUSCH: They are filtered in terms of inputs e-' 14 to process computer for a physics calculation. But they
(_)S 15 are--
16 MR. LIPINSKI: And what's the filter--
17 MR. RAUSCH: They aren't filtered significantly 18 for input safety systems. That was all evaluated in the 19 first round of the bulletin in June.
20 MR. LIPINSKI: Do you have any idea what that 21 filtering tank constant is?
22 MR. RAUSCH: Yes. It was a factor of 10. Too 23 quick to--to slow--to quick, whatever the right way is, to 24 be significant. It wasn't very--
25 MR. LIPINSKI: Rapid response compared to the
() Heritage Reporting Corporation (202) 628-4888
143 1 frequency of oscillation?
- 2. MR. RAUSCH: Right. -Yes, we asked those same 3 questions ourselves in the May time frame after the event.
4 So that turned out not to be of significant impact.
5 The AP arm instruments, the device itself, the 6 protection system, does not have a significant filter.
7 MR. LIPINSKI: Let me ask the question again.
8 MR. RAUSCH: Sure.
9 MR. LIPINSKI: Okay. That's to the safety system.
10 But what about the data that you recorded, was that 11 filtered?
12 MR. RAUSCH: What the operator sees, it depends on 13 which device you are looking at. Again, electronically 14 there's very small filters.
15 It turned out that there was some range of damping 16 available inherent in chart recorders, mechanically and 17 electrically induced, and we've made recommendations for 18 people to use the fastest response.
19 MR. LIPINSKI: Okay. But you recorded data and 20 from that data you drew a conclusion. What I am trying to 21 establish is the data that you saw that was recorded, was 22 that filtered or was it fast?
23 FR. RAUSCH: The data that we recorded at LaSalle 24 was not--only really the average power range response was 25 what was used in our study. And it was just used to qualify Heritage Reporting Corporation 0 (202) 628-4888
__ _ -____ _ = _ . . -- _ _ -
1:
b
,. . '1'44 1 'this axial phase-shift; phenomena, in terms of LPRM response
({
2 .versus. avid response.- And that was not-significantly l
l 3 filtered either. It was fast enough.
l
- l. 4 We are dealing.with a very slow phenomena. It's'a 5 .5. hertz.
6 MR. LIPINSKI:- Yes. But if you've got filters l 7 that are a couple of second hertz, then it's going to.
j 8 distort your data.
9 MR. RAUSCH: No. And our chart recorder' hooked up 10 to.what was called the GE Star Trek recorder.. It's capable 11 of millisecond response.
12 MR. LIPINSKI: Okay.
13 MR. RAUSCH: Well, I'll just go quickly through 14- some of these things. At the same time frame we happen to 15 have our Enforcement Conference talking about the adequacy 16- of our design control because of the low decay ratio 17 numbers.
18 We ended up with satisfactory--I think the way it 19 -ended up was admitting to the design problem in this case 20 but not really a design control problem. And we had no fine 21 because of our extraordinary corrective actions associated 22 with what the site had done and associated with the Owners 23 Group work.
H24 We ended up with another meeting with the NRC 25 regarding putting more--making the Bulletin a little Heritage Reporting Corporation
=O (202) 628-4888
l 145
[^) . 1 stronger than what we had in our interim corrective actions.
V 2 And we came to an agreement that on two pump trips, all 3 plants would be required to SCRAM. That was added to the 4 interim corrective actions that the Owners Group had come up 5 with with GE.
l l 6 We had our first big Stability Committee meeting 7 after the NRC meetings and that's when we formed the 8 organization I was discussing earlier associated with review 9 of results, long-term solution, interim safety actions, and 10 gathering stability data.
11 We had the meeting here. And it's in your 12 handout. I want to qualify these bullets. I should have 13 changed them a little bit.
,g
- 14 What I was referring to here was that ACRS didn't U
15 believe a statement was made that stability--operating 16 issues of stability were not a public health and safety
. 17 issue. A serious one. I think that statement was made back 18 in December. But there are still very many questions 19 outstanding on ATWS and emergency procedure guideline 20 issues.
21 It's partly why we are here again today.
22 The Bulletin Supplement was issued in late 23 December. We had another Stability Committee meeting 24 getting going with our subcommittees, responding to the NRC 25 for first round of clarifications for operating conflicts
[) Heritage Reporting Corporation (202) 628-4888
t 146 j
'l that we were' finding.
2 Whenever'you take something as complicated as 1
3' every time you' hit a certain power flow condition, you have '
l l
4 to do the following, there's bound to be some questions as .i 5 - to what takes priority. g 6 MR. WARD: Tom, on your item there where you say l
I 7 'the ACR doesn't believe stability--
8 MR. RAUSCH: Yes.
9 MR. WARD: As I recall, we didn't write a letter, i
10- _did-we? The only way the ACRS-- j
'11 DR. KERR: I was trying to recall. I don't-know.- j 12 MR. CATTON: No, we did not.
- 13. hR. WARD: The only time the ACRS ever says what l 14 it believes is in a letter.
O 15 MR. RAUSCH: Okay. .
I 16 MR. MARD: And we are kind of sensitive to people !
17 lifting comments that are made around the table as an ACRS !
18 collegial opinion. j l
19 MR. RAUSCH: This slide was really used in terms l l
20 of putting things in context for the Owners Group. And we_
21 lifted it-- l l
-22 MR. WARD: Well, that's even worse then. ;
l 23 MR. RAUSCH: Well, I think in terms of what we !
24- heard, we were just basically saying in terms of serious 25 public health and safety, we are agreeing that this is not p Heritage Reporting Corporation v (202) 628-4888 1
I
s j; 5
147 1 that kind of phenomena, 2 MR. WARD:. Well, the. committee may or may not 3 believe that, but it hasn't decided.
4 MR. RAUSCH: All right. I'11 be more careful in 5 the future.
6 Basically just a series of more meetings and it 7 takes.a while to get funding. This is a-major bureaucracy 8 dealing with the Owners Group.
9 A lot-of conference calls with the NRC. A letter 10' back commenting on the Bulletin Supplement.
11 Very, very many subcommittee meetings. -We had 12- four subcommittees going. Among,them was getting this 13 . experience, world-wide experience. We talked with both BB-fy 14 Atom and ANF/KWU.
D 15 Jose March-Leuba was kind enough to come to one of 16 our meetings and give us a little discussion of what work' 17 they were doing in stability.
18 And it just keeps going, but NRC sent a letter 19 back to us responding to our comments back in January.
, 20 Basically all I'm trying to show here-is it's been l
21 a busy group. And I guess the last key data on-there is we 22 do have a major meeting tomorrow, an annual meeting, of all
~23 the BWR' Owners Group executives where the funding, major 24' ' generic funding for the next phase of work for what we call 25 the long-term solution is taking place.
Seritage Reporting Corporation (202) 628-4888
- - _ - _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ . .-. A
i 148 1 So we are anxious to get through that hurdle
. f}-
)
q 2 tomorrow. 4 3 I might say at one time, I think back in April,~ we i
4 had already figured in utility time put into--a couple of 5 subcommittees ended up to about half a man year just inside 6 of a few months. So we were very busy.
7 The results to date. Here if we want to get more 8 into detail about what the Phase I calculation was, I can do 9 that now.
10 There's really an engineering scoping analysis to 11 evaluate thermal margin, critical power ratio margin, under 12 regional oscillation conditions.
13 The regional oscillation model was based on test gg 14 data, which we used to drive TRAC thermal hydraulics. In G
15 other words, a power distribution versus time and space was 16 developed and TRAC was driven with this distribution in a 17 series of flux contours and then the thermal hydraulic 18 response was observed and power ratios calculated and 19 neutron monitoring system response study.
20 MR. LIPINSKI: Now, was this a full three-21 dimensional representation?
22 MR. RAUSCH: Yes. But there was, what, eleven i 23 rings, Hank?
I
! 24 MR. PFEFFERLEN: Yes.
I 25 MR. RAUSCH: So there was essentially ten
() Heritage Reporting Corporation (202) 628-4888 l
l
149 I L 1 symmetrical--five on one side of the reactor and five 2 contours on the other side. Kind of in ring shapes and a !
3 hot channel in the center of one of the rings, studied in 4 detail.
5 With five and five, yvu can get a fairly good 6 contour.of a flux- you know, left and right. And it looks a 7 lot like what is called a first order or second order vessel 8 function. But in a real reactor, it would obviously be a 9 little more distributed than that.
10 MR. LEE: This is two dimensional radial 11 cylindrical RZ calculation, or is this regional--
12 MR. RAUSCH: It's 3-D, X, Y, Z.
13 MR. LEE: X,Y,Z?
14 MR. RAUSCH: But contours were drawn and 15 encompassed individual bundles, fuel bundles. And they 16 belonged either to one contour or the next contour. Each-17 ring had this discrete number of fuel bundles in it.
18 MR. LEE: How many total number of measures were 19 used, do you remember?
20 MR. RAUSCH: Well, in terms of neutronics, it was 21 full 764 times 25. 764 bundles, 25 axial.
22 In terms of the hydraulics, again, there was 23 eleven. Eleven channel types ir the reactor.
24 MR. LEE: And you did say something about 25 something driving something. I missed that.
Heritage Reporting Corporation (202) 628-4888
i 150 ,
l' NE. RAUSCH: Yes. In order to make this kind of 2 study and do sensitivities-on it, these flux contours,.these 3 power contours, were used to drive the TIUU2, the TRAC model.
4 The TRAC was not running free mode, oscillating on its own. -
5 The power being calculated in TRAC was being fed in.
=
6 MR. LEE: But there's no feedback on power 7 whatsoever.
8 1R. RAUSCH: It had to fit it, that's right. But i
9 'we were able then to carry the amplitude beyond what has 11 0 'been' observed. We maintained the. shape and raised the j 11 overall size of the oscillations. That's the only way we 12 were able to do that.
13 Now, to do that, you can't study all the 14' variables.
I )' 15 MR. LEE: To me it seems like a zero dimension J
16 neutronics calculation with three dimensional thermal 17 hydraulics, would you agree with that?
18 MR. RAUSCH: Yes, that's correct. I would say I
19 that would be correct. But the power feedback in every i 20 neutronic node, I mean there was an actual power in every 21 node in the reactor that determined the--for example, the 1 22 time dependent pressure of each--
23 MR. LEE: Yes, I understand. That's why I said 24 the three dimensional thermal hydraulics, but zero 25 dimensional neutronics.
Heritage Reporting Corporation O (202) 628-4888 1
l 1
l
_--_-______-________A
151 4
7[ 1 MR. RAUSCH: Yes.
(3-2 Our conclusions are easily stated but we found the 3 potential for safety limit violation was predicted not 4 predicated. Predicted under some conditions. Namely, very 5 high initial bundle powers. So logically you'd say the 6 higher the power density of the reactor, or the individual-7 plant was capable of having very high bundle powers, the 8 more the loss of margin, or the closer you could have come 9 to the safety limit during this~ postulated regional 10 oscillation.
11 Now, on the good side, we did find that even under 12 the very, very high assumed initial power conditions, that 13 the' average power range monitor system, APRM system, did 14 respond very well and at the 10 percent peak-to-peak level 15 which is being used now for operating procedures in I
16 conjunction with the Bulletin Supplement, there was still 17 margin.
18 DR. KERR: I didn't understand how you determined 19 the amplitude of the oscillations. Initially I got the 20 impression you got those from observations made in reactors.
21 MR. RAUSCH: Yes. A base case was really a k-orso 22 amplitude and then it was multiplied--the shape was 23' maintained. It was just simply multiplied by a constant. I 24 forget how high it was. 1.5 or something like that.
25 DR. KERR: Was there anything unusual about 1.5 as Heritage Reporting Corporation O (202) 628-4888
152 1 1
) I contrasted with 2 or 3/4 or--
f'~'/
u- 3 2 .MR. RAUSCH: The way the study worked, it really )
3 wasn't that sensitive to that. Wnat it was basically doing 4 was making a. calculation of critical power ratio change 5 versus instrument response or oscillation magnitude and 6 correlate the two so you can really extrapolate the results.
7 DR. KERR: So it doesn't make any difference what 8 the local oscillatory amplitude--
9 MR. RAUSCH: Oh, I'm sure it does. But that was 10 studied. In other words, that was studied. The oscillation 11 amplitude is one of the major variables we have, but the 12 carrying it to extreme. cases wasn't--in other words, having 13 very high contour versus a medium contour didn't 14 significantly change our plots.
15 If you have a high bundle power oscillation, you 16 have a higher impact on critical power ratio, whether the 17 contour was real high or the contour was average height. It 18 wasn't that sensitive.
19 MR. LEE: What's the maximum amplitude that you 20 have seen in this particular simulation?
21 MR. RAUSCH: What was that, Hank? Do you have the 22 report out? Do you have that report handy? Well, we' d have 23 to look in the report.
24 25 Heritage Reporting Corporation
(] ~
(202) 628-4888
153
~
\ 1 t/
2 MR. RAUSCH: It was based on a model that we give 3 to and got the APRM response taking into account the actual 4 base shift and the radial power distribution. What we were 5 looking at the model for was given that the plant operated 6 that way, our question was how would the neutron monitoring 7 system see this and how would the hydrology respond and 8 that's what we modeled and that is what the results were 9 based on.
10 They were not intended to look for the largest 11 amplitude and, say, given the case that we have observed, 12 how will the neutron system respond and how does the flow 13 respond.
f- 14 APRM is the average of 28--
(',3) 15 MR. CATTON: Those are not your APRM--I am talking 16 about the in-core sensors.
17 MR. RAUSCH: That's what I'm talking about. Each 18 in-core sensor, and we average them as you would taking into 19 account the various axial and radial distribution.
20 MR. CATTON: So you are looking at average values 21 rather than instantaneous in-core values?
22 MR. RAUSCH: Well, what I was saying, we were 23 trying to determine what kind of an ACRM response would come 24 out of this type of oscillation.
25 MR. CATTON: I think the information of interest f)
'd Heritage Reporting Corporation (202) 628-4888
154 1 is what's going on in the core in detail in terms of what
.{ }
'2 the fuel temperatures are.
3 MR. RAUSCH: We have that information.
4 MR. PFEFFERLEN: That's what the flow part of it 5 was intended to do.
6 MR. RAUSCH: At very, very high assumed initial 7 bundle powers, our highest cases, the highest power 8 distribution case, if you want to call it that, and the 9 highest oscillation magnitude case, flow reversal is 10 obtained near the end of the fuel assembly. I can't 11 remember if it was the top or the bottom.
12 You actually had a very, very short period of 13 time, you had the hydraulic effect of working towards 14 counter f] 7w, and that's, of course, where you have your 15 problem, because you have low flow conditions. It's very 16 transitory and all our engineering judgment is telling us 17 it's a re-wet phenomena and you won't really get a very high 18 temperature rise. And I think the cladding temperature 19 model we looked at it did show some temperature rise but not 20 enough that you were going to fail the fuel, but it's still 21 our limit of interest.
22 I need to make one perspective on all of this. I 23 think there's a very big misconception that anyone who gets 24 associated in these analyses to think that these are the 25 analyses to demonstrate acceptability of our current Heritage Reporting Corporation (202) 628-4888
i IL, r- l 155 l 1 syrtems. And they are not. They were intended to
}
2 demonstrate to show in an order of magnitude sense how much I 3 margin we had due to this hydraulic phenomena of a regional' 4 oscillation, both from an instrument standpoint and from the
)
5 detailed hydraulics going on in each bundle.
6 Now that we see we don't have a very large margin ;
7 for extreme cases, our answer is not to sit back and say.we 8 are going to make our existing system work. We are going to 9 sharpen our pencil until our existing system is shown to 10 work. It's quite the contrary. We are working on long-11 term solution options, all of which have automatic 12 protection features. Only for some plant classes are we 13 even considering still the use of the average power range 14 system that has the automatic feature.
15 For. example, a BWR 2 is probably still quite 16 appropriate because the BWR 2 instruments are quadrant 17 based. This quadrant right down here feeds one average 18 power range channel. .This quadrant feeds another. This one 19 another and another. And it's hard to imagine, especially-20 for a low power plant coming up with oscillation contours 21 where you get any significant cancellation. So this system 22 will respond very, very well.
23 But even still, you'll see when I get into our 24 long-term discussion phases, we are in the point of studying 25 the various options we have, all of which can mitigate, in
'[] Heritage Reporting Corporation (202) 628-4888 i
-l
156
['} 1 our opinion, prevent or mitigate any postulated mode of ss 2 oscillation.
3 So to let the cat out of the bag, for example, if 4 you had several of your detectors throughout the reactor 5 input to a safety system individually, any one of them shown 6 to have an oscillation characteristic, and you had your 7 reactor SCRAM on that, these questions go away.
8 And that's our approach. There are other 9 approaches like that. I'll discuss those in a few minutes.
10 So we aren't trying to sit back and say things are all 11 right. It's just a little tight. It's quite the contrary.
12 We are on a very aggressive long-term solution schedule.
13 So I guess the conclusion for now is that there rm 14 are still some substantial APRM indications and very 15 substantial local power range monitoring indications 16 available for the safety limit is approached. Those 17 readings are being used actively by all plants in this 18 country in their emergency procedures and in their 19 monitoring procedures when they are in this region.
20 They are also supposed to get out of the region if 21 they do get in there, but I can talk about that a little bit 22 if you like.
23 We do have very good instrument response. And 24 these calculations support interim corrective actions that 25 w3 ve developed. Actually they do go hand in hand.
Heritage Reporting Corporation
[^t (202) 628-4888
157
"\ ' Now, the Phase IA calculations are a little more--
(d 1 2 much easier to understand from a theoreticians point of view 3 since we are not driving anything. This is just a fully 4 coupled 3-D TRAC G calculation performed to evaluate the 5 impact of oscillation on average reactor power.
6 These were in-phase oscillations. There was not a 7 -level of detail put into the model to where you would expect 8 to get a regional oscillation. But here what was of 9 interest to us was the time dependent kinetics and thermal 10 hydraulics and looking for the non-linearities that were t
11 postulated. Well, we know there are non-linearities in all 12 the neutron feedback.
13 And we wanted to see if having large amplitude gx 14 oscillations would result in a power increase and thereby O 15 raise questions regarding ATWS containment heat load.
16 As we mentioned, full 3-D coupled neutronics and 17 thermal hydraulics TRAC G model. The results were we saw i
18 small magnitude oscillations. Natural circulation 19 conditions, all by itself. The calculation did it. Then 20 all control rods were withdrawn and a fairly reactive point 21 in cycle at LaSalle, which is the same time as the 22 conditions of the actual LaSalle ovent.
23 And we got larger amplitude oscillations. They .
24 went up to about 200 percent of rated, after they steadied 25 out a little bit. It took perhaps 20 seconds to build up to
() Heritage Reporting Corporation (202) 628-4888
158
(} 1 that magnitude. 30 second. A very artificial case where 2 natural circulation you withdraw a very large number of 3 control rods, but we needed to get large amplitude 4 oscillations.
5 Our conclusions are that the average core power 6 increased slightly. It war, less than 7 percent. And that 7 we required an inlet flow and sub-cooling increase, in other 8 words, a feedwater flow increase, to sustain this power 9 increase.
10 A normal ATWS event, the feedwater flow would be 11 controlled by procedure or automatically the feedwater pumps 12 would be tripped. You would have a steady source of water 13 injection. The reactor water level would drop. The amount 14 of recirculation flow would drop. The power level would
(-)3 15 drop.
16 We did try reducing the water level by removing 17 feedwater flow for a number of seconds. And we did not see 18 the oscillations go up. We saw them stay about the same or 19 perhaps even go down.
20 MR. CATTON: In your use of TRAC, did you do 21 anything to check to make sure that its internals were all 22 in proper working order for your use? You know, one of the 23 problems with TRAC at the outset was the fact that it 24 doesn't calculate void fraction very well. The fluid 25 interfacial drag is not quite right. There's lot of those
() Heritage Reporting Corporation (202) 628-4888
159
(~)
\_/
1 kinds of problems with it. And it's been tuned to the 2 large-break LOCA. And you are not looking at a large-break 3 LOCA. I'm not sure what's been done to remedy these things.
4 MR. RAUSCH: I'm not really qualified to answer 5 your question in detail. I do know that there's been--on 6 the physics side--there have been improvements. It's been 7 driven by panacea. And there are major improvements in the 8 physics, including some work in the void feedback back about 9 three or four years ago. That's in this model.
10 I really don't know enough about the TRAC void 11 modeling being used. Although I think the fact that it did 12 oscillate at the same conditions as the LaSalle event on its 13 own is a pretty good indicator that the thing was working
,f- . 14 right. And it was against the Peach Bottom mechanical base i
15 deck, I'd call it, or thermal hydraulic base deck. And 16 that's the same deck they've used to qualify against, for 17 example, Peach Bottom. Turbine trip test. So there's been 18 work done on that and Hank Pfefferlen will be talking a 19 little bit about what they've done to qualify TRAC.
20 DR. KERR: Did you say earlier that for the Phase 21 I work you did one complete run plus part of another? '
23 DR. KERR: I mean Phase IA, yes.
24 MR. RAUSCH: Yes, that's right. The part was the 25 water level reduction.
()
Heritage Reporting Corporation (202) 628-4888
160 1- DR. KERR: Okay. And.do you anticipate that had 2 you done another run you would have repeated the results of
~
3 this first one?
4 MR. RAUSCH: Yes. We are at the stage right now 5 where GE and EPRI were asked by some of our executives--an 6 Executive Oversight Committee to go back and study what kind 7 of calculations could be done to close out the issue, if 8 more is'needed.
9 DR. KERR: And should I conclude that given the 10 particular reactor model that the average power increase 11 would always be less than 7 percent?
12 MR. RAUSCH: I think Jose will.be able to address 13 some of that, but I think he'llLbe talking about some 14 calculations he's done. EPRI is going to start doing some-15 calculations in that area also but I don't know if they 16 intend to make them of a licensing grade or a conference 17 grade. But just for information.
18 We think our primary argument as to why the power 19 increase is not significant is really related to what ATWS 20 is. It's the water level in.a BWR if you constrain the 21 feedwater flow, you can strain the power. The water level 22 will drop to match.
23 That's the direction we are thinking. And you 24 can't really change that physics argument. It will work.
25: .It's just a matter of degree.
f.
l Beritage Reporting Corporation (202) 628-4888
I 1'
161' l
.f~ 1 And what was particularly interesting to us is if V '
- 2. you look at this. power increase in this scenario, if you 3 look at the fact that the sub-cooling increased due to the.
4 fact that the feedwater flow increased, it all started.from 5- the power increase. Take those conditions, average them 6 after the power increase, and input to a steady state
,7 physics calculation, you get the identical power number past
~
8 the TRAC case. Exactly the same.
9 DR. KERR: Well, number, one, since you limit 10 feedwater run, I don't know what you'd do if you did another 11 run, but you probably do. But, number two, I don't know 12 whether the TRAC is representing what happens in the actual 13 . cooler on not. Again, perhaps you do. Or have confidence 14 that you do.
15 MR. RAUSCH: Yeah. Like I said,-I think you have 16 to remember in an ATWS event, that this power level increase 17 is only ou interest until the standby liquid control is 18 turned on. So it's a short window of power increase.
19 DR KERR: Well, I don't think'anybody knows 20 exactly how long it takes for the standby liquid control 21 system to come into effect.
22 MR. RAUSCH: Right.
23 DR. KERR: So there are a number of uncertainties 24 associated with this it would seem to me.
25 MR. RAUSCH: I can't argue with that. There are a Heritage Reporting Corporation O (202) 628-4888
162 em 1 number of uncertainties.
(
2 MR. PFEFFERLEN: This was just a pure non-SCRAM 3 case. It did not follow ATWS.
4 MR. RAUSCH: This was just a study of large 5 magnitude oscillations just to see what happens to the power 6 level. That's all it was. And like I say, we are 7 considering doing more. One of the problem is you have to 8 very carefully choose what cases you do.
9 DR. KERR: I can certainly endorse that.
10 MR. LIPINSKI: In using TRAC, let's establish what 11 parameters you have as measurements from the core. As far 12 as I know, it's neutron flux only. You have no other 13 information.
14 MR. RAUSCn: In terms of output of TRAC 7 15 MR. LIPINSKI: Thermal hydraulics. No , no. I'm 16 talking about the actual core. What measurement data do you 17 have from an actual core? Neutron flux?
18 MR. RAUSCH: During what type of assessment are 19 you talking about? In terms of hydraulics?
20 MR. LIPINSKI: Well, I'm asking about the thermal 21 hydraulics. You don't have thermal couples in there. You 22 don't have individual flow meters. So you have no way to 23 measure the thermal hydraulics that's taking place within a 24 core.
25 MR. RAUSCH: Not directly. There are some Swedish l
Heritage Reporting Corporation
(~)/
(, (202) 628-4888
INTRODUCTORY STATEMENT BY THE JOINT THERMAL HYDRAULIC PHENOMENA AND CORE PERFORMANCE SUBCOMMITTEE - CHAIRMEN'S REPORT
' - MAY 23, 1989 The meeting will now come to order. This is meeting of the Advisory Comittee.on Reactor Safeguards Joint Subcommittee on Thermal Hydraulic Ph'enomena and Core Performance.
I am D. Ward, Chairman of Thermal Hydraulic Phenomena Subcommittee.
W. V. err, Chairman of the Core Performance Subcommittee is expected to arrive later this morning.
The ACRS Members in attendance are: I. Catton and C. Wylie.
Also in attendance are ACRS Consultants: J. Lee, W. Lipinski, M. Plesset, V. Schrock.
The purpose of this meeting is.to discuss: (1) the NRC-RES thermal hydraulic research program plan as documented in NUREG-1252, and a proposed SECY paper, and (2) the status of the ongoing effort to address h- the implications of the core power oscillation event at LaSalle Unit 2.
Mr. P. Boehnert'is the cognizant ACRS Staff Member for this meeting.
The rules for participation in today's meeting have been announced as part of the notice of this meeting previously published in the Federal Register on May 10, 1989.
A transcript of the meeting is being kept and will be made available as.
stated in the Federal Register Notice. It is requested that each speaker first identify himself or herself and speak with sufficient clarity and volume so that he or she can be readily heard.
We have received no written coments or requests to make oral statements from members of the public.
(Chairmens' Comments - if any)
We will proceed with the meeting, and I call upon Dr. Lou Shotkin of NRC Research to begin.
163
() 1 plants'that have eight instrumented channels on it. That's 2 data that might be available sometime.
3 MR. LIPINSKI: Yes. But when you were making your 4 earlier discussion on using flux to drive TRAC and telling 5 me what TRAC was doing, how do I compare the output of TRAC 6 to what's happening to a core if I don't have core 7 measurements.
8 MR. RAUSCH: That's the same issue you have with 9 the Oden or ready codes or ANS co-transient code and any 10 integrated system coda. Really what people do to qualify it 11 is they look at the integrated response.
12 MR. LIPINSKI: Yes. But you are driving it so you 13 had no integrated response.
14 MR. RAUSCH: Right. But if it works undriven in 15 an integrated fashion, then--
16 MR. WARD: Wait a minute. This Phase IA wasn't -
17 driven.
18 MR. RAUSCH: That's right. Phase IA was not.
19 MR. LIPINSKI: I backed up to the other one. In 20 terms of the fact that he doesn't have measurements--
21 MR. WARD: Okay. But the preliminary one was just 22 driven by the neutron power measurements.
23 MR. RAUSCH: The neutron power, right.
24 MR. WARD: And they weren't making any comparison 25 of thermal conditions, what they observed.
() Heritage Reporting Corporation (202) 628-4888
h p ,
164' a.
1 MR.:LIPINSKI: But he's drawing the' conclusion--
E ~2. MR. : WARD: Because he didn't have any observed.-
3 MR. LIPINSKI: Iknow. But he's drawing ~.a 4 conclusion. In other words,-he.has a flux shape.that he 5' puts into a code that gives him a thermal hydraulic effect, 6 but he has nothing to tell-him what was.actually happening 7 in the core in terms of thermal hydraulics to compare it to 8 -the code.
'9 MR. WARD: Well, I don't think he was doing--in 10 the first one,'he just concluded that, hey., there'areisome
-11 big--they run right up against margins or beyond margins--
'12: MR. RAUSCH: It's a problem that needs fixing, 13 exactly..
14 MR. WARD: --with this sort of thing. .'I think.
O.~ 15 that's all they got from that first one. They confirmed 16 they got a problem.
17 MR. RAUSCH: And this one--
18 MR. LIPINSKI: It's-kind of an expensive way to do-
'19 it, isn't it?
20 MR. PFEFFERLEN: I think it's safe to say that we-21 feel that TRAC does a pretty good job on the thermal
.22 hydraulics because of separate effects tests and things that 23- have been conducted in tests on the hydraulics part of it.
24 But you are absolutely right.- What remains to be 25 looked at is how does TRAC do a stability calculation and Heritage Reporting Corporation O (202) 628-4888
r p 165 b
('N ' 1. that is planned. But it is not under'our belts right now.
'2 We feel that pretty good that with the driven neutron, that 3 .the TRAC will do a good job.of looking at the hydraulics.
4 'MR. LIPINSKI: What's your confirmation of that?
5 My comment is though depending upon the uncertainties in 6 TRAC, your margins may be even worse than what you think 7 they are.
8 MR. PFEFFERLEN: That's possible. We are not 9 touting this as a mounding calculation. I think Tom 10 indicated that this was not the end. This was an 11 identification that the next step needs to be taken, and I 12 think I agree with you. You are absolutely right.
13 MR. LIPINSKI: But you are arguing that TRAC did a 14 good job. What's the basis for that argument?
15 MR. RAUSCH: The fact that without having to turn.
16 any screwdrivers it oscillated at the same type of 17 conditions as the LaSalle event on its own.
18 MR. LIPINSKI: But you forced the oscillation by 19 inputting the power.
21 MR. WARD: You are mixing up the preliminary one 22 and the IA.
23 MR. RAUSCH: The IA was a hands off calculation.
24 .The TRAC was allowed to free run. A couple 3-D. And it 25 did. It oscillated. It looked exactly like the LaSalle O Heritage Reporting (202) 628-4888 Corporation
166 g 1 event.
p 2 MR. CATTON: Amplitudes and all?
3 MR. RAUSCH: No , not exactly. But it's such a 4 fine line you wouldn't expect that. Just'a tiny difference 5 in decay ratio, for example. Just the slightest modeling 6 ' difference I think would impact the amplitude. But just 7 coming to a steady state of oscillations at the same 8 conditions as the LaSalle event I think is quite remarkable.
'9 MR. CATTON: I think so too. Probably 10 compensation errors.
11 MR. RAUSCH: Perhaps.
12- MR. LIPINSKI: No. There were probably some 13 places you could put in adjustment factors.
14 MR. CATTON: Well, I wouldn't go that far.
'O 15 MR. RAUSCH: Well, I can't emphasize enough that l
16 in terms of--at least in terms of what we would call 17 operating issues, the issues associated with margins during 18 normal operation.or postulated instability events is being 19 taken seriously as its own issue. We realize we have to put 20 a fix in ahead of the final answer as to what these 21 oscillations could look like which is a Research issue. I l 22 characterize it as a Research issue. That's really 23 associated with ATWS. You know, what could happen to the 24 oscillations, what could they look like.
l- 25 .As long as you have a fix in place that Heritage Reporting Corporation (202) 628-4888
167
-(): 1 encompasses hardware and procedures or whatever, tech specs, 2~ that shows that you will not have oscillations or the plant 3 will SCRAM at first indication, all those interesting
~
4 Research questions become only relevant to ATWS.
5 Well, the next three slides or so are just--they 6 are only barely relevant. They are just an indication- you 7 can look at them if you wish--of what some of the 8 subcommittees have done gathering world-wide data.
9 The bottom line in this subcommittee is that we 10 are obtaining a report from Stoller Corporation, or co-11 funding it with a foreign utility. And that plus all the
- 12. meetings we've had with the foreign fuel suppliers have 13 given us an awful lot of information, at least as to what's 14 available today.
15 DR. KERR: What have the European operators done, ;
I 16 if anything?
17 MR. RAUSCH: There's a wide range of things. Most i
18 of the European plants like to do testing to establish 1 19 actual margins during either start up or some of them do it 20 even monthly. I don't think everybody in this country is 21 quite as comfortable with watching the plant oscillate like 22 that. Even regional oscillations. They take a lot of I 23 measurements and then they can check by their decay ratio 24 codes how well they are doing.
25 But the biggest fix that's really out there, if I l
() Heritage Reporting (202) 628-4888 Corporation l
l l
l l
t _ __ _ _ _ _ _ _ __ _ _ _ _ _ . _ _ _ _ _ _ _ _ __ .__ _ _ _ _ _ __ _ _ _ _ _ _ _ _ . . _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _
a
168
/3 l' you.want to call it a fix, is the use of stability monitors.
d 2- They are more wide spread in Europe than they are in this 3 country. There are only one or two running right now in 4 this country. I think there may be more pending further 5 Owners Group work.
6 And the stability monitor is very useful under 7 controlled conditions. When the operator is moving control 8 rods or he's in an unstable area.
9 For a rapid entry into'a instability event, a 10 stability monitor needs some time to calculate it. It's.not 11 as useful. It may still play a role.
12 DR. KERR: Thank'you.
13 MR. WARD: Are the ABB and KVU designs essentially 14 similar in regard to this issue to the GE design?
~
15 MR. RAUSCH: Similar in terms of fixes?
16- MR. WARD: Yes. Are there reasons to expect the 17 problem might be different or might be similar? The fixes 18 might be similar?
19 MR. RAUSCH: Yes. There are considerable 20 differences. The KVU plants are more--they are very similar 21 to GE. They have licensing exchange agreements. The ABB 22 plants have much tighter orifices. And much stronger 23 recirculation flow capabilities with their internal pump l
24 plants. They design them later in the game and I think that 25 may have been one of the reasons they did that was to make Heritage Reporting Corporation
.O (202) 628-4888
169 1 .more inherent stability margin.
2 MR. WARD: I see.
~
3 MR. RAUSCH: So there are some. differences. But I 4~ .would say overall the KVU thought process is very'similar to 5 what the Owners Group is doing right now.
6 MR. WARD: Okay. I interrupted you I think.
'7 MR. CATTON: I just wanted to know what a 8 stability monitor was.
9 MR. RAUSCH: Oh, I'm sorry. It's essentially a 10 fast state-of-the-art PC or a micro computer connected to 11 any number'of local power range detectors. And an average 12 power range detector. And it acquires noise data over time 13 -and correlates it. Fancy words. ANF has one. It has an 14 auto correlation function and then there are some that do--
~s-15 I'll trip myself on these. They are essentially 16 transformers, but they can make on-line decay ratio 17 measurements with X amount of time delay.
18 DR. KERR: It's magic.
19 MR. LIPINSKI: It's not magic. It's state-of-the-20 art.
21 MR. RAUSCH: They have been proven I think that 22 they do work. I wouldn't anticipate a major problem 23- licensing them. I don't anticipate it. I don't know. I 24 haven't talked to the staff. But in terms of--as long as 25 the use matches their capabilities. They do have s Heritage Reporting Corporation
( (202) 628-4888
170 I
rs h 1 limitations.
U 2 MR. CATTON: If it's just a PC, why arer.'t they 3 used more here? They are pretty cheap.
4 MR. RAUSCH: A PC is reasonably cheap. It's not 5 overly expensive. It's an operator aide. We are seriously 6 considering it right now for use in LaSalle. It's just not 7 a--we look at it more as just exactly that. It's kind of an 8 aide. And right now under the Bulletin requirements, you 9 can't really operate anywhere near the problem area anyway.
10 MR. CATTON: That means you know where the problem 11 area is. It's well defined.
12 MR. RAUSCH: That's right. We've got a buffer 13 region in there, but that's right. We don't know exactly 14 what the problem area is. But the PC isn't the problem with l
()
'- 15 them. There's a large number of hook ups to all the i 16 detectors. And then if you want to have it available all 17 the time, you need two of them. And it's not fully tested 18 out in terms of--well, I guess it is in Europe. Its 19 usefulness is still questionable I guess. !
i 20 Corrective Action Review Committee. I talked a 21 little bit about this. Really they reviewed the bases for 22 interim corrective actions that ended up in Bulletin l 23 Supplement I.
l 24 We've clarified some of the things where there's 25 been conflicts in terms of what the operator should do under l
l r- Heritage Reporting Corporation
's (202) 628-4888 l
l
l-
'171 hL 1'- certain conditions. We've looked'at adverse operational
.O.' 2' consequences.. And developed some procedural guidance for 3 internal owners' use.
4 And we had a separate ~ subcommittee' working on 5 getting the Phase I and IA reports out. They coordinated-6- the-EPRI peer review which I think gave a lot of technical 7 credence to the work that was done.
8 A lot of the limitations that you gentlemen have 9 been pointing out, were pointed out by our committee or by 10 EPRI. A&1 we are all aware of the limitations of the Phase.
11 I and IA analyses.
12 -They coordinated the comments in the draft reports 13 and we obtained a special assessment of ANF fuel.
14 . Essentially the same--just the thermal hydraulic portion of D
Ass 15 the hot channel. We gave them the input conditions to 16 compare the 9 by 9 fuel type versus what the 8 by 8 fuel 17 type.that was-calculated, using different correlations. And.
18 there's no real surprises.
19 Reports are due to be out late this month in Phase 20 I and sometime in June for Phase IA and we will be 21 transmitting them both to the NRC and to ACRS>
22 The one with all the action now, all four 23 subcommittees have been very active early in the year. The 24 long-term solution subcommittee. Really their charge was to 25 identify viable permanent solutions to stability issue, not Heritage Reporting Corporation O (202) 628-4888
172 1 ATWS, the operating issue, which applied to all GE product 2 lines and all fuel types.
3 This Bulletin, meeting with the NRC, really should 4 come more'near the end here but we've been involved'with 5 detailed consultations with major fuel vendors, consulting 6 firms. We've developed a very wide spectrum of potential 7 solutions using guidance from the whole committee.
8 We've estimated' resources to establish feasibility 9 solutions and what workscopes it would take to further prove 10 that a given concept.could work.
11 We've estimated development and hardware costs 12 associated with each type of solution. And we developed a 13 plan for defining the final solutions.
14 And somewhere in here we met with the NRC last 15 month to make sure we had no major disconnects in the 16 direction we were going and that meeting went very well.
17 So really right now we're in the implementation 18 phase. We are kind of reorganizing again to get the right 19 people involved.
20 DR. KERR: Is it accurate to say that your 21 preliminary results lead you to believe that an ATWS 22 situation would produce a self loading oscillation and would 23 not cause difficulty?
24 MR. RAUSCH: Certainly. That's preliminary yet.
25 There are and there will be a large number of theoretical or Heritage Reporting Corporation O (202) 628-4888
- p. .
W ,
173
) 1 'research type questions. As long as you allow a reactor to 2 . free run and you postulate no SCRAM, you get some very 3 interesting physics.- And, you know, we have limited 4 calculations to date, so--
=5 DR. KERR: I understand. And the reactor, as i't's 6 going to operate not in the abstract situation.
7 MR. RAUSCH: And really the efforts I'll be 8 talking about from here until near the end is the operating.
9 domain. The ATWS issue is quite--we separate that quite 10 distinctly in our--
11 DR. KERR: Well, see, the ATWS issue is the only
~
12 one I'm concerned about because I don't believe--I mean the 13 reactor may cause--I mean the Type I oscillations may cause' 14 you some economic damage, but I can't believe that.that's 15 going to cause a serious accident. But the ATWS thing, it 16 seems to me, could with what wo know at this point.
17 MR. RAUSCH: Well, we haven't seen anything that 18 would indicate any real reason for damage or--
19 DR. KERR: The real reason for damage would be an 20 oscillation with a very large amplitude and I have not seen 21 that ruled out. Your preliminary results have convinced you 22 it's unlikely.
23 MR. RAUSCH: Yeah. Okay.
24 DR. KERR: I'm not convinced that one can get one, i 25 I just haven't seen anything that would convince me that one Heritage Reporting Corporation (202) 628-4888
174
^i 1 could not. And I think in ATWS it's fairly unlikely. So (Q
2 I'm not suggesting that--
3 MR. RAUSCH:- Well, we have--Hank has about three 4' cn four slides addressing what we've done to date, and some
- 5. of it is a rehash of what-our discussions already are, but 6 if you--I only have a couple more slides' associated with 7 long-term solution. If your interest isn't very detailed, 8 it won't take very ?.ong.
9 It's obviously of interest to the owners. Because
- 10. that's where our major effort is going right now.
11 MR. WARD: We are interested. Go ahead.
12 MR. RAUSCH: Okay. Those actions are more 13 conservative than SIL-380 Rev 1 which is the basic guidance 14 document in terms of how to recognize and respond to 15 instability.
16 There are additional requirements that the owners 17 came'up with. There was a SCRAM anytime there was a two-18 pump trip. There was greatly enhanced training and operator 19 awareness stability issues of instability.
20 An instability event is very unlikely with 21 recommendations in placs and the probability of a thermal 22 limits violations is, of course, even less likely than the 23 event itself.
24 We feel that the region of exclusion is larger 25 than the region of instability occurrences and--
/'] Heritage Reporting Corporation (J (202) 628-4888
K 175 1 DR. KERR: I don't understand what the 2 significance of that bullet is.
.- 3 MR. RAUSCH: This bullet?
4 DR. KERR: Yeah.
5 MR. RAUSCH: We've encompassed the--knowing 6 instability events. That's all it is. With'some 7 conservatism. In other words, the part of the power flow 8 map that's been X'd out or put into caution space. It 9 envelopes all known events. That's all it is.
10 DR. KERR: I thought there was some' significance 11 maybe to the larger, but what you are telling me is that you
- 12. know where the instabilities will occur.
13 MR. RAUSCH: We don't know in great detail. And 14 that's one of the things we are looking into.
~.O 15 MR. LIPINSKI: How do you know if you have a 16 thermal limit violation?
17 MR. RAUSCH: It's only based on this Phase I work.
18 MR. LIPINSKI: Okay.
19 MR. RAUSCH: Had we not done that Phase I work,'it 20 could be ignorance is bliss I guess. I feel better that we 21 now know there's a problem there.
22 The long-term solution objectives, I guess is 23 fairly obvious.
12 4 The close-out stability from an operational point 25 of view is an issue.
Reporting Corporation p) s Heritage (202) 628-4888
i s
176 r%' 1 While minimizing the impact on plant operation--we j
. (_)
2 don't want a trigger-happy SCRAM system.
3 Recognize plant differences and provide some 4 flexibility to optimize applications to'each plant.
5 This one page really has all the--the only l 6 detailed information we have in terms of what options we.are 7 pursuing right now. I can discuss it at whatever depth 8 you'd like.
9 Before I go through some of these, I'd like to 10 point out that every solution has an automatic action 11 associated with it.
12 Now, they are basically into three or four 13 different categories that we're looking at. I'll start with' 14 the bottom one. I won't call that a category. It's just on O 15 there for completeness and we never took it very seriously.
16 And that is investigate conservatism of the present safety 17 limit. As I talked about the fact that this is a re point 18 phenomena, almost by definition.
19 There's like to be some conservatism using the-20 MCPR limit as your limit of interest, but we realize that 21 that's fraught with problems to try to pursue that any 22 further.
23 So really the first major category of solutions is
- 24. really prevention.
25 We call it regional exclusion. It's basically 7' Heritage Reporting Corporation (202) 628-4888
177 1 taking the part of the power flow map and putting a SCRAM if
]}
2 you enter that part. I could try to give you a little 3 -better feel for that on a map.
4 For example, you could take perhaps a shape that 5 looks something like this. We don't really know. But by 6 further analysis, we establish a line where if the plant-7 fell below that line, it could_be a defunction of plant type-8 or different characteristics of some plants, of the 9 different product lines, both power density and 10 hydraulically.-
11 If they had an event where .they ran back in here, 12 into the instability region, the plant trips. It's that 13 simple.
14 Now, the problem associated with this one is it 15 cuts off part of the--porhaps a line may be way'over here.
16 We haven't done the analyses yet. We are getting ready to.
17 The further to the right this line is, we need 18 some confidence in this line. The more of an operational 19 problem it presents in getting start ups and perhaps pump-20 run backs or operator power reductions falling down into 21 this region.
22 So one of the alternatives we're looking at is the 23 second bullet on the slide I just took off. It's called 24 prevention by regional exclusion but with limited bypass.
25 For example, this line here you may allow the plant to
(~') Heritage Reporting Corporation v (202) 628-4888 i
178 1 operate under this region only in a controlled fashion if e-)
~
2 the stability monitor is on, for example. And you'd still 3 have a trip protection right below there. So if you did 4 have a fast-flow reduction, you'd still SCRAM anyway. And 5 that would have to be supported by a different type of 6 analysis than just a single region would be.
7 MR. WARD: What do you mean by controlled bypass?
8 MR. RAUSCH: Procedures and tech specs that would 1
9 say if your stability monitor is operable, maybe so many 10 hours1.157407e-4 days <br />0.00278 hours <br />1.653439e-5 weeks <br />3.805e-6 months <br />, if necessary, whatever. You would be allowed to 11 start up in this region and get out of it as long as you are 12 monitoring while you are in there and you still had a backup 13 automatic SCRAM protection.
14 MR. WARD: Administrative controls you mean?
5/ 15 MR. RAUSCH: Yes. You may have a key switch on or 16 something in terms of turning off one of the SCRAMS.
17 These solutions require some sophisticated decay 18 ratio analysis to establish those lines. We recognize the 19 need to even scope that out to see what it looks like before 20 we put a lot more money into it and we also recognize the 21 NRC's role in such a solution.
22 We are not talking about anything resembling decay 23 ratio calculations of the past vis-a-vis numbers that would 24 be conservatively based on using more limiting power 25 distributions, et cetera.
g- Heritage Reporting Corporation
(,)j (202) 628-4888
179 gg 1 MR. WARD: What sort of operational problems do R.) ~
2 you think that might present? I mean how many forced 3 shutdowns per year per plant?
4 MR. RAUSCH: It's really a function of the plant 5 types. There are a lot of the newer units, the 5s and 6s, 6 have single failure proof recirculation systems. In other 7 words, it takes two failures to trip both recirc pumps.
8 MR. WARD: Okay.
3 MR. RAUSCH: We are generally thinking that it's 10 likely for a high power density plant where this region is 11 going to encroach a desirable operating mode which is why we 12 talk about this limited bypass, but it really depends on how 13 high you let this line go.
14 There are some owners right now that are thinking q
A/ 15 they wouldn't mind letting it go quite far if that gave them 16 the simplified solution in terms of--this type of line is 17 rather easy to put in. It's real easy to design such a 18 system such a system, such a SCRAM system like that.
19 And with today's fuel types, there's no need to 20 operate way up here, high and to the left. They can kind of 21 come up like this.
22 But other plant types, where they aren't single 23 failure proof or they've had histories, or there are 24 concerns about operators for whatever reason, maybe he 25 looses feedwater heaters and he need to take power fT Beritage Reporting Corporation
(_) (202) 628-4888
L 180 7,-
> 4 ).
1 reductions, they wouldn't want to live with something like 2 this.
3 MR. WARD: So you think a typical plant.might have 4 one extra forced shutdown per year or ten?
5 MR. RAUSCH: I would hope that nobody would choose 6 a solution unless it had something less than one per year.
7 A lot less. That would certainly be Commonwealth Edison's 8 opinion.
9 MR. LEE: Then'if you were to perform some other 10 more conservative power flow map analysis you will end up 11 with a larger region that would you like to avoid.
12 MR. RAUSCH: That's right. We understand that.
13 The existing analyses, the ones that were in place at the.
14 time of the LaSalle event assume the normal power flow map.
15 And calculated decay ratios based on that and they were less 16 than .8. Now, we already know right now if we redo that 17 with much better assumptions and we are looking at using 18 different codes also. A code capable of calculating--giving 19 some indication of regional oscillation decay ratios.
20 Our line was already from LaSalle. We already 21 know it's not way over here. It's up here somewhere. 50 or 22 60 percent core flow is what we are guessing.
23 MR. LEE: This power map and the recommended 24 regions, ABCs, may not be the final areas at all.
25 MR. RAUSCH: Oh, certainly, that's correct. As a Heritage Reporting Corporation (202) 628-4888
)
i
______________________________-______O
181 l matter of fact, these interim A, B and C--this slide is from
(~}
J 2 the Bulletin Supplement. We probably--it depends on plant
- 3. type. Some plant types might be very conservative. Others 4 ones you may cross parts of this. We don't even know yet.
5 We don't think so, but we really don't know until we do the 6 calculations.
7 And it might depend on how you may look for, you 8 know, once in a hundred year power distribution, it's 9 particularly bad. We don't know those answers until we do 10 it.
11 I've kind of covered the first two bullets, 12 regional exclusion options. This one is basically available 13 for perhaps a BWR 2 and a 3. It's saying that with further 14 analysis perhaps we could show for a lower-power density l
15 plant that the APRM's average power range system will still 16 respond with plenty of margin and safety limit.
17 We are pretty confident that you could be shown 18 for a BWR 2 with a quadrant APRM system as I've discussed.
19 The problem with that is analytical. We are not sure what f
20 it would take to do that. We are leaving it open until it's 21 closed I guess is the way to look at it. We realize that we 22 need more than our Phase I analysis to support that 23 solution.
j 24 But there are a group of plants that I know I
25 roughly 40 to 50 percent of them, maybe more, that do have
/"% Heritage Reporting Corporation
(-) (202) 628-4888 L~ _ - - - - - - - - - _ _ _ _ _ _ - - - - _ _ _ _
g.
A
- 182-7
-? 1- 'it_already, an-automatic-SCRAM,. a flow bias-automatic SCRAM,,
'2 that doesn't have any time delay. That's' essentially this.
3 line--somewhere down this line here. It' shapes up like 4' this. Where.they do enter that' region, they do SCRAM.
.5: LaSalle had such a system but it had a time ~' filter..
6 on.it. Intentionally. It's called.a thermal _ power monitor.
7 And had it been more like'Dresden'and Quad Cities without'
- 8. that time delay it would have SCRAMed much quick'er.
9 The next category is probably the..most interesting 10 one and I've hinted at this already. This'would-be a 11 mitigation, in other words a SCRAM, based on LPRM response.
12- And there would be an automatic. action for a specified:LPRM' 13 signal. And the number'and distribution of detectors needed 14 in the' reactors would have'to be determined. And that's the 15 kind of' thing.we are~ going to be looking at is.how many.
16' would it take? How far away can it-be and still be 17 ' sensitive enough to catch it before there's any question of
- 18. thermal hydraulic limit, CPR violations?
19 There are qualified. micro processor based systems 20 available. One of them is called a NUMAC. It's a General 21 Electric system that can essentially be smart enough to 22 detect an oscillation very rapidly without tripping every 23 time there is a small power fluctuation like for a valve 24 closure. They could look for an up cycle and a down cycle
, 25 in some range of frequencies.
Heritage Reporting Corporation
.O (202) 628-4888
___-_m. --_.-__________m.m__.-_____ _ _ _ . _ _ . - _ _ _ _ _ _ --
i 183 r~st 1 So we are quite confident that this will work.
'J 2 It's just a matter of--it's a little more expensive hardware ;
i 3 option. And we need to look at it closer before we become )
i 4 comfortable with the plant trip- plant liability question.
5 But we feel there's probably some core flow again where the 6 system could be turned off perhaps above 60 percent core 7 flow. And it automatically enables itself when you come 8 down into the region of instability.
9 So this is the heaviest analytical. This one is 10 not too far behind but we think it could be better. We 11 think the methods can address it in terms of uncertainties 12 in the regione.
13 This one has the most questions analytically, the 14 higher the power density goes. And this one has the less
/ 15 questions analytically.
16 MR. LIPINSKI: Do you regard these additional 17 systems as being safety systems?
18 MR. RAUSCH: Yes.
19 MR. LIPINSKI: And they'll meet all the 20 requirements of safety systems?
21 MR. RAUSCH: Yes.
22 MR. LEE: Have you also considered the possibility 23 of making the system itself more stable?
24 MR. RAUSCH: Oh, yes. That's the direction our 25 future fuel design changes is always maintained or get l r- Heritage Reporting Corporation l (_) (202) 628-4888 1
l l
l
184 1
1 (w, 1 better instability. But unfortunately--when we brainstormed
%-) '
2 all the possible solutions, we had thirty some. And they j 1
3 didn't all break down into these five or six ct.tegories, or 4 four or five categories. There's all kinds of things you 5 can do. The fuel elements or the orifice on the bottom that 6 would make radical improvements. They range from 7 prohibitively expensive to taking seven years to change.
8 So unfortunately if you raise the single-phase 9 pressure drop, which is a very stable thing to do, you lose 10 recirculation flow because you are increasing the '
11 resistance.
12 And there is no real simple quick fix, although I 13 think evolutionary fuel design changes, for example, use of 14 more cold water inside the lattices, less voided area, can (D
k# 15 have measurable stability improvements and that's the 16 direction of BWR fuel designs.
17 MR. LEE: How about changing enrichment?
18 MR. RAUSCH: We don't think--it's really a power 19 distribution phenomena. You could radically change the 20 power distributions and provide enormous margin, but again 21 that's a very big economic question.
22 You know, enrichment is pretty much tied--
23 enrichment and power distribution assumed in design is done 24 both for economic and safety reasons. And the moment you' d 25 have to cut down in power distribution to put away stability r- Geritage Reporting Corporation
( )/ (202) 628-4888
185 f ,~- 1- issues is an enormous amount.
2 MR. LEE: I mn not saying that changing . enrichment
-t 3' is a simple issue whatsoever. -But by changing enrichment, 4 you can' change the magnitude of void correction to 5 reactivity which in my mind is one of the most sensitive 6 primaries effecting the stability.
7 MR. RAUSCH: Yes. I think you can have a bigger-8 change,-within the constraints of building a reload cycle 9 that: lasts fifteen months or whatever. You'can get the same 10 type of changes, as I said, the more advanced fuel designs.
'll MR. LEE: But at the present time no design 12 changes that will make the system more stable is seriously
-13 considered?
14 MR. RAUSCH: That's correct. And really the
-15 reason that was a quick conclusion to come to is that all 16 the. field vendors we know of had already been looking at 17 that over the years. They've been aware of at least of what 18 the direction of magnitude--relative magnitude and direction-19 of'all the various skill design primers are for-stability.
20 You know, there's all kinds of things. 9 by 9 is slightly 21 less stable because of quicker time constant, but that can )
22 be made up in one phase and two phase pressure drop 23 relationships. I 24 So these issues have been looked at by all the 25 field vendors we know of for a long time. So we already Heritage Reporting Corporation O. (202) 628-4888
.186 I L 1 know.about the relativa sensitivities to all these types of" 2 -things.
3 MR. LEE: But we also know that the possibility of.
y 4 this isolation existed several years ago, and I mean you are 5 discussing anticipated transient with our SCRAM too.
6 MR. RAUSCH: Yes.
7 MR. LEE: And someho? that issue was not pursued 8 by field designers and inventors. So I'm bringing up the 9 question again. Whether we should look at this mode of 10 making the system inherently more stable as one of a more
- 11. primary action item.
12 MR. RAUSCH: All I can do is give you my own 13 thoughts on it. And I think the risk and benefit is far out 14 of. balance. The cost benefit. I was a reload designer and
' O 15- there's enormous amounts of money associated with_even a .25 16 percent enrichment savings. And I'm sure we'd have to go
. 17 much further than that. The' average now going into the 18 reactors is in the low 3 percent range. 3.0 to 3.4, 19 .somewhere around there. I think you'd have to get a fair 20 amount of reduction. You know, each small reduction adds 21 eight fuel assemblies. Reload.
22 MR. WARD: Each what, did you say?
23 MR. RAUSCH: Each .1 or so enrichment reduction.
24 .2 adds maybe another eight bundles to the reload batch.
25 Perhaps more than that. Maybe sixteen. So it's extremely
~
. Heritage Reporting Corporation
- - (202) 628-4888 s
. . _ _ _ _ _ _ _ _ _ _ _ _ ._ _._______-_____.__________._._________________.____._____m______m,_____
( '.
} 187 1 expensive..
2 MR. LEE But if I may just pursue once more.. You-3 have to balance this against potential loss in your 4 operational flexibility.
l 5 MR. RAUSCH: Yes.
6 MR. LEE: So one additional' SCRAM could amount to 7 a good bit of money as well.
8 MR. RAUSCH: Well, believe it or not, one SCRAM is 9 on the order of one or two fuel bundles of cost. Fuel 10 bundles aren't cheap. So we feel pretty confident that both 11 this option and this one by plant time--particularly this 12 one--can have almost negligible plant trips associated with
-13 it. That. remains to be seen of course.
- 14 DR. KERR
- Well, that assumes then that you will 15 very seldom-ever encounter a situation in which that-SCRAM' 16 is dated.
17 MR. RAUSCH: Or it assumes that you'have a 18 stability monitor or you are smart enough to look--
19 DR. KERR: Well, in other words, you will avoid.
20 the instability.
21 MR. RAUSCH: That's right. That's correct. I l 22 think that-- l 23 DR. KERR: Which says in effect you make the 24 probability of instability very low. 3 25 MR. RAUSCH: Yes.
Heritage Reporting Corporation (202) 628-4888
188 fg 1 DR. KERR: In your operating situation.
N) 2 MR. RAUSCH: Right. Another way of looking at it 3 is the work we'll be doing will help define the areas enough 4 that people will naturally want to do what you say. They 5 will be smart enough to look at what could cause them to 6 trip--
7 DR. KERR: Yes.
I 8 MR. RAUSCH: And try to prevent it.
9 Another problem with the fuel enrichment changes, 10 for example, is again to get the significant--one reload 11 won't change it overnight. It's going to take several 12 years.
13 Thir is our simplified schedule for long-term 14 solutions. It's very similar to what we showed the NRC and G
l k-) 15 the staff on April 6th. That meeting is over with.
16 We've written bid specifications for the scoping 17 analyses we need to support those options I've shown you.
18 We gave a major presentation to what's called the Executive 19 Oversight Committee. The Owners Group.
20 We obtained some funding direction and some l
21 guidance on how we should proceed. It recommended that we 22 work primarily with General Electric in developing these 23 workscopes and we are essentially starting that right now.
24 Major funding comes tomorrow. We are keeping our 25 fingers crossed. We think we've done a pretty good sales
,f g Heritage Reporting Corporation
() (202) 628-4888 l
1
I '
189 1 job through the stability committee. Assuming the funding
.\
2 passes, we_ start work the next day and there will be 3 decision milestones along the way as to how far we proceed 4 with the various types of calculations.
5 And we are looking for the September time frame to 6 update the NRC and provide some guidance to the individual 7 . owners as to what type of solutions for their particular 8 plants they should proceed with. And we'll determine how to 9 fund it from there. And probably come down to whichever 10 ones look best. .There will probably be some differences of 11 opinions or differences by plant types.
12 Well, that concludes the portion, the discussion 13 of what the stability committee has been up to and where we 14 are going.
k- 15 I don't know if it's of any interest to you to get 16 into some details of NRC questions and answers. I think the 17 first couple I have here are operational details beyond the 18 level we've gotten into in discussion of the Bulletin. But 19 there are some questions related to ATWS that I think are 20 worth going into. And some NRC-suggested activities that I 21 can discuss just very briefly.
22 I don't have a lot of information on these but 23 I'll just give you our position.
24 There is an NRC-suggested activity in terms 25 related to understanding more about symmetric and asymmetric g- Heritage Reporting Corporation
( (202) 628-4888 i
190 h 1 oscillation modes.
2- I guess there's another issue. The Owners Group 3 plans to address this by really the second bullet. It's 4 looking at the experience available and making sure our 5 long-term solution includes the issue. In other words, we 6 cannot make a fix that works only for core-wide 7 oscillations. We_have to study--for example, if we are
,8 using a regional exclusion mode, we have to be confident 9 that it envelopes both core-wide and regional oscillations.
10 Now, that doesn't address the ATWS answer of that, 11 the research issue.
12 Similarly, the variation of oscillation amplitude 13 shape and period. Same thing. The long-term solution will'
') 14 address that. It's a thermal hydraulic oscillation. If.
(V 15 it's a suppressed-type device. We'll make sure it's moot as 16 to exactly the type of shape of the oscillations.
17 In terms of ATWS again, this doesn't address that 18 issue. We realize that.
19 We aren't trying to skirt the ATWS issues. We do 20 have to make sure of our limited resources and funding that l 21 we are proceeding with all due speed on the operating issue.
L 22 And we do have a somewhat formal clock. I'm not sure how l 23 I'd characterize it but the Bulletin does state that the l
h 24 Owners Group would be anticipated to have some type of 25 generic guidance available by about now. In June sometime.
O erie 9- verei e cernor tio-(202) 628-4888 l
__m_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _
)
191 i
)
('-) 1 And that's the purpose of telling the NRC in April'where we 2 were. And that by the end of the year plant-specific 3 solutions would available. I don't think it will be quite [
4 that soon although it's still possible. And that was the 5 purpose.of trying to demonstrate that we're going as fast as 6 we can and we are being prudent and that the interim 7 corrective actions are--
8 DR. KERR: Are you happy with the progress being 9 made?
10 MR. RAUSCH: Yes, I am. What's been a learning 11 experience for me is that this is twenty-three. utilities 12 involved and there's just as much effort going into keeping 13 everyone updated and convincing all the various owners that 14- this is an issue for them as well.as for the high power 15 density plants.
16 But, yes, we haven't really skipped a beat yet,
- 17. since we've gotten underway.
18 Well, if there's no further questions on my part 19 .right now, Hank Pfefferlen will go through a little bit of 20 rehashing and a little more information on ATWS.
21 MR. LEE: One question. -
22 MR. RAUSCH: Sure.
23 MR. LEE: And that is to do with the last copy of 24 the transparency which you did not use apparently.
25 MR. RAUSCH: Yes. Hank will be going into that.
O rit e- rerei-e cerror tie-(202) 628-4888
192
,s 1 About the code qualification?
'~
2 MR. LEE: One of the NRC-suggested activities?
3 MR. RAUSCH: Was code qualification? Is.that the 4 one?
5 MR. LEE: I guess.
6 MR. RAUSCH: Yes, Hank will address that also.
7 I'm sorry.
8 MR. LEE: Fine.
9 MR. RAUSCH: That's really a GE answer. He's 10 better qualified to answer that.
11 MR. WARD: Why don't we take a break until three 12 o' clock, if you don't mind.
13 MR. PFEFFERLEN: I don't have very much.
14 MR. WARD: Well, how much do you have?
q kJ 15 MR. PFEFFERLEN: I could probably do it in fifteen 16- minutes, ten or fifteen minutes.
17 MR. WARD: Oh, okay. Fine. Go ahead.
18 MR. PFEFFERLEN: My name is Hank Pfefferlen. I am 19 a Manager of BWR Licensing Programs for GE. And what I'd 20 like to do is to talk a little bit about the--I think much 21 of what I'm going to say about ATWS has really been 22 discussed, but let me put it up here and maybe we can get a 23 little further discussion to the degree that I'm able to 24 address it on the ATWS, the Phase IA, and then a little bit 25 about the qualification activities planned at GE for the 1
7- Heritage Reporting Corporation (3) (202) 628-4888 l
l 1
4
_ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ . - - - - _ _ _ _ __ J
193 1 ' codes that we've been using.
2 The question that has come up that really 3 stimulated the Phase'IA study was a question from the NRC
'4 -staff on what happens--actually it says, do the oscillations 5 cause a difference in the heat dumped into the suppression 6 pool? In other words, will the average. power increase 7 during an ATWS event because of the presence of the 8 oscillations and if they do, will that result in an 9 increased C load on a suppression pool which is already 10 pushing its limits.
11 And so we attempted to address that and I guess in 12 summary right now there's no additional work planned to go 13 further. We think that the work that has been done, both in 1 <4 the Phase IA study and work that has been done previously 4- x 15 provides an answer to this question, perhaps not at the 16 heart of what causes it or a lot of the details, but from an 17 operational and an ATWS point of view, provides sufficient 18 information.
-19 However, the Owners Group is sensitive to 20 questions that may come up in this regard and we're going to 21 work with the staff and try to respond to this issue because 22 clearly ATWS has to be solved before the stability issue 23 goes away.
24 DR. KERR: Okay. So the answer to the question, 25 if you have to give it right now, is that we don't know, is Heritage Reporting Corporation (202) 628-4888
194
,3 1 that correct?
! )
2 MR. PFEFFERLEN: No, we believe we know enough to 3 say--let me go through it. But I think that given the 4 system response, not the core and the stability 5 characteristics. But the total system response. We believe 6 that the ATWS analyses that have been done are adequate and 7 are still valid.
8 Now, this is uhere we are coming from. This is 9 what I'd like to address a little bit here.
10 DR. KERR: I don't know what's meant by "we 11 believed." Does that mean you have a lot of faith? Or that 12 you have some basis for the faith?
13 MR. PFEFFERLEN: We have operated BWRs. I have 14 watched BWRs operate for many years--
/^%
k-) 15 DR. KERR: The Russians operated the Chernobyl 16 type reactors for many years and they never got in trouble.
17 And that just doesn't leave me with a lot of--
18 MR. PFEFFERLEN: No, I understand. My point was 19 that the fact that the BWR, the system operation in a sense 20 of the feedback between power, water level, in-lead water 21 conditions, and the limits that it puts on what the core can ,
22 do, we believe--
23 DR. KERR: Is this engineering judgment or do you 24 have a model--
25 MR. PFEFFERLEN: No, it's based on analyses. On 1
.s
- Heritage Reporting Corporation
(_j (202) 628-4888 l
1
__._______________________a
l-I.-
l
[ 195 i) l' Lthe. analyses that--some of the analysis that was performed ~
2 for Phase IA. The analysis that was performed for the ATWS 3 study in NEDO 24222 back r.any years ago and an operating.
4 experience with BWRs and how the power is driven by the void 5 in the core flow.
6 So we take all of those and we put'those together 7 and I guess, you know, the bottom line may come out as an 8 engineering judgment because we haven't done the complete 9 analysis.
10 DR. KERR: And you are convinced there is a 11 limitation on the impetus of oscillation that you can 12 predict?
13 MR. PFEFFERLEN: Now, wait a minute. Let's make fS 14 very clear what we are talking about. The question here was
's) 15 an increase in the average power--
16 DR. KERR: That's right. But it's not a' symmetric 17 oscillation. And the bigger it gets the less symmetric it's 18 likely to be.
19 MR. PFEFFERLEN: I agree. But let me submit that 20 when we talked about increase in average power, what we are 21 doing is we are talking about the heat load in the 22 suppression pool.
23 DR. KERR: Exactly.
24 MR. PFEFFERLEN: And you talk about the maximum 25 amplitude of an oscillation, you are talking about what it
.O serie se negarti-e corear tie-(202) 628-4888
196 1- can do-to the core. And'I'see this as two separate things.
f-)
.g
'2 They are separable, I believe.
3 MR. CATTON: Not on the average.
4 MR. PFEFFERLEN: They both'have to be answered.
5 DR. KERR: ' Yeah, but if you have a non-symmetric
- 6. oscillation and you increase the amplitude, you are going to 7 increase the average power. I mean if it's non-symmetric in 8 the sense that the bottom is less than the top, and that's 9 what you are likely to get.
10 MR. PFEFFERLEN: Well, but what happens is that 11 th'e_ upside gets narrow and the boundside gets wider so that
'12 the average--
13 DR. KERR: Well, maybe.
14 MR. PFEFFERLEN: Well, that's right. And this g .
s/ 15 question was' aimed at answering the average power and not 16 the amplitude.
17- DR. KERR: But I have an idea that the amplitude 18 . will influence the average power. You think it won't?
19- MR. PFEFFERLEN: Let me tell you why I think it 20 won't. And I hear your question, but let me try our logic 21 here as to why that won't happen integrated over time and 22 then we can come back and discuss that.
I 23 MR. CATTON: Was that analysis the 1 1/4 runs with 24 TRAC?
25 MR. PFEFFERLEN: I'm sorry?
Heritage Reporting Corporation I
s,) ' (202) 628-4888 L - - - _ - - - - - - - - - - - - - - - - - - - - _ _ - - _ _ _ - - - - -
i 197 1 MR.'CATTON: The analysis that you referred to, is i 2 that the 1 1/4 runs that were made with TRAC?
3 MR. PFEFFERLEN: That's'one analysis, yes. But
- 4. .that looks at the response and there we did find the power 5 increased. And that's what I have here. That the Phase IA 6 studies did show this slight increase that Tom talked about, 7 about 7 percent.
8 But what we found--well, let me qualify that.
9 .That analysis--
10 MR. CATTON: That's a strong conclusion based on 1 11 1/4 calculations.
12 MR. PFEFFERLEN: No, I think it's a good--it's a 13 conclusion--the Phase I results showed that.
.,- . 14 Now, the question was asked does that mean that 7 15 percent is as-big as it ever gets? I would not draw'that 16- conclusion from that one analysis. I agree with you there.
17 I think I would feel that the increase would be 18 small, but you are right, I wouldn't limit the magnitude 19 increase based on that one calculation.
20 But that's not the reason I feel it's not a 21 problem. The analysis was done with the plant in level 22 control. Which meant that as the power wanted to go up, the 23 feedwater could accommodate the increased demand to maintain 24 a given level and therefore a given core flow.
25 That is not the situation in an ATWS--ATWS I'll l [~T Heritage Reporting Corporation b/ (202) 628-4888 1
198
(; 1 come to. But even with these boundary conditions of the 2 plant in level control, we found a duration was short, the 3 increase. It lasted about thirty seconds. And then it
^
4 leveled out. And it was self-terminating. There was 5 nothing done to turn it around.
6 Those are the results of the Phase IA study.
7 When I looked at ATWS, the assessment, the basis 8 for the ATWS rule contains the analysis done in NEDO 24222 9 was based on flow control, which means that he--
10 MR. CATTON: Could we hold just a second?
11 Increasing void decreases stability. That seems 12 to me if you start cutting back on the flow you are going to 13 increase the instability. That's counter to the argument
,A 14 you just gave.
'\-)
15 MR. PFEFFERLEN: If we cut back on the flow, there 16 is a tendency to increase--no. You increase the void, you 17 reduce the power. Now, the question is which is the 18 stronger effect? The increase in void which will decrease 19 the power, and we are talking about a power-to-flow ratio.
20 And as you get into a more adverse power-to-flow ratio, you 21 may become less stable.
22 A decreasing level decreases flow, increases 23 power, which increases void which decreases power. And the 24 question that is on the table then is how does that affect 25 the stability characteristics. And in some cases the i
l
()
Heritage Reporting Corporation (202) 628-4888
199
,- 1 decrease in power may increase the magnitude of instability.
\'~)
2 And others, the case that we ran- we found that reducing the 3 level actually stabilized the plant for awhile. And we 4 think we understand what goes into that.
5 But the power level comes down. The power level 6 comes down because you are increasing void. And here I am 7 talking just about the average power. So what happens is 8 that if the power--if the average power goes higher than was 9 analyzed in, say, the NEDO 24222 case, with the fixed inlet 10 flow, that means you've got more steam leaving the vessel 11 than you have water coming in. Which says that the level 12 has to go down. That's the only place the steam is coming 13 from. If the level goes down, that has to build in voids 14 and it has to reduce power.
4
-s
.. 15 Therefore, in the ATWS scenario, albeit maybe 16 there could be a transient increase in power that integrated 17 over time, the steam out has to become equal to the flow in, 18 and that is the basis for the ATWS assessment. And that's 19 why we say that as far as the average power and the heat 20 load on the containment, we don't believe that the 21 instability is going to affect the conclusions that were 22 drawn that the suppression pool is going to be able to 23 accommodate the results.
24 Now, as far as the magnitude, how big the 1 25 oscillation gets, that's a little different question. But l Heritage Reporting Corporation l f-)
s (202) 628-4888 l 1
l l
l l
____._..____.____________Q
200
,- ll as far as the average integrated power over time that is put k~
2 into the suppression pool, and that really is containment 3 now, we believe that the ATWS assessment for the reasons 4 that I've indicated here, provides an upper bound to those 5 results.
6 Now, Tom did indicate this. We did some 7 sensitivity studies--
8 Yes, sir.
9 MR. LEE: If this report on ATWS assessment is 10 several years old, perhaps that's the last time I had looked 11 at these analyses of ATWS for the oscillatory events, if I 12 remember correctly, the actual application of the computer 13 codes used was terminated during this oscillatory period of 14 events. And some analysis were performed based on assumed r^s
() 15 -- to come up with an estimate of temperature and other 16 things.
17 Am I off the mark completely?
18 MR. PFEFFERLEN: What the analysis--let me jump to 19 my next chart and see if I can--I'm not sure I fully follow 20 your question, but let me jump here and say that the basis ,
21 for the assessment--the analysis followed through these i 22 major steps. The recirculation pumps tripped to time zero.
23 The feedwater pumps were tripped at approximately 45--this 24 is the typical assessment I believe. The feedwater pumps 25 tripped at 45 seconds. That forced the level to start 7~s Heritage Reporting Corporation
() (202) 628-4888 l
-i
201 1 dropping. There's no water going into the core. The ECCS
(
s) 2 flow was initiated at approximately one minute. Now, these 3 are fixed flow systems. It's on or off. It doesn't try to 4 maintain level.
5 They are on at one minute. Boron begins to enter 6 the core at approximately three minutes. The water level 7 reaches a minimum and begins to come back up at about four 8 minutes, which indicates that the lowering of the level with 9 the attendant void increase and the boron contribution is 10 starting to catch hold of the power now and turn it around 11 and so now the level is beginning to come back up.
12 We reach hot shutdown at seventeen minutes.
13 That's way out in time, after all of this happens. The 14 containment temperature and pressure, when you integrate
(- 15 this whole effect as far as the steam that's going to the 16 containment, that peaks out at 28 minutea.
17 And what we are seeing here--our view is that if 18 you expect oscillations, it will occur in here.
19 MR. LEE: Right. That's exactly the period of 20 time which, as I remember it, in this NEDO document--
21 MR. PFEFFERLEN. That's right.
22 MR. LEE: Actually application of computer code !
23 was terminated--it was suspended. And some semi-argument 24 was provided.
25 MR. PFEFFERLEN: No. This analysis went all the c Heritage Reporting Corporation
(,) (202) 628-4888
202
'I way through. HIf you are talking about the stability
()~
j- 2 assessment itself, that may be the case. But this analysis 3, followed--we did run into some oscillations with our code at 4 that time it was not qualified for stability, but it went 5 through and they grew and then were terminated and we went 6 on.
7 But the real thrust here was not to look at the 8 stability characteristics, but rather to see what happens.
9 You know, with the plant response following a pump trip, 10 which says that the temperature really falls below the ill- equilibrium--I should say the power flow ratio falls below-12 the equilibrium point, it takes some time for the 13, oscillations to begin.
~
14 And I believe in the LaSalle case there was five
-O 15 or so minutes before the oscillations were observed. And by
.16 .that time we're putting boron into the core and so there is 17 a fairly small window where you would expect to see 18 oscillations before the boron would take effect and turn 19 .them around.
20 So purely from the point of view of the average 21 power increase, we have a fixed flow. We have boron coming 22 into the core, and the oscillations would be expected to 23 occur in this short period of time with the tail where again 24 you would be stable because you've shut the plant down--with 25 a tail putting more heat into the containment.
Heritage Reporting Corporation (202) 628-4888
l l
203
() 1 And when we follow this through, there may be a 2 transient response where the average power may increase, it !
3 has to decrease out on the tail such that the integrated 4 effect, the containment heat load, is going to be small.
5 That's average power.
6 Now, when you talk about magnitude, that's a 7 little broader. How big can they get? But here again--
8 well, two things. I believe that for very large 9 oscillations we have not seen that tendency to date for them 10 to get very large. Lowering the water level has a potential 11 for increasing the magnitude.
12 But what.we believe came out of our assessment was 13 that the plant doesn't move down an equilibrium line. What 14 happens is the power is affected more severely than the flow
(~)
v 15 and so you tend to move into a more stable region and then 16 relax back up to the more equilibrium point. And this is 17 what happens when you trip the pumps from rated you fall 18 below and then you relax back to the point and that's why it 19 takes time for the oscillations to begin.
20 And as you lower water level, in this process, the 21 same thing is happening. The plant tends to stabilize and 22 then turn around and come back up to that equilibrium point 23 which may be further off in the unstable region, but now we 24 are beginning to put boron in and believe that there is- you 25 know, the competing effect of shutting the plant down is l
() Heritage Reporting Corporation (202) 628-4888
204 1 taking effect very early here.
2 MR. LIPINSKI: What did the LaSalle oscillations 3 show in terms of an average change? You must have had 4 enough that you could run a line through them and determine 5 a new average and determine whether the average increased or 6 not. Did anybody analyze those oscillations in terms of an i
7 average?'
8 MR. RAUSCH: Oh, yes. In great detail. The.whole 9 part was dissected.
10 MR. LIPINSKI: What was the result?
i 11 MR. RAUSCH: The result is that the power level is 12 being driven by the major change in sub-cooling due to the 13 feedwater heater stripping. And the normal time it takes 14 for the sub-cooling to catch up for the power flow 15 conditions.
16 MR. LIPINSKI: Did the average- you didn't answer 17 the question. Did the average power increase over what you 18 obcerved.
19 MR. RAUSCH: Sure it did. But it would have 20 increased steady state without oscillations the same amount. l
'41 MR. PFEFFERLEN: It was the feedwater sub-cooling 22 and system effects that caused the power to respond the way 23 it did. Not the oscillations.
24 ( MR. LIPINSKI: Okay. I am looking at the shape of 25- the oscillations to find out whether there's more area under Heritage Reporting Corporation (202) 628-4888
[L 205 1 the positive peaks as there is on the valleys. I haven't 12 seen the trace so I don't know. That's why I am asking-you 3 the question.
4 MR. PFEFFERLEN: Again, let me say, that in the 5 LaSalle case they had feedwater.- Feedwater was driving was 6 flow, so they were analogous to the first case that I showed 7 here--whatever I did with it--where they were.in level 8 control. And, in fact, their feedwater was hunting and they 9 were adjusting the level and the power was responding with 10 that response. But that's all attributable to the system 11 effects, not the oscillations.
12 MR. LIPINSKI: Okay. There are two separate
-13 things. Let's talk about the first part and that is the 14 shape of an oscillation. Whether the area under the O' lL5 positive peak is greater than the. valley part when you come
-16 down to the bottom. And if the area under'the positive peak 17 is greater than it causes an average power increase.
18 You know, if I have a straight line and that's my 'l l
19 average power and suddenly I go into oscillations and I' 20 compute a new average power, it's a question is that new 21 ' average line above the original line that I started with.
22 And that answers the initial question that Dr. Kerr asked, 23- is there an increase in average power because of the shape
- 24. of the oscillation?
25 Now, in asking that question, on LaSalle I
' ~
Heritage Reporting Corporation (202) 628-4888
206
() ~1' understand that the average power went up for other reasons.
2 But just looking at the chape itself, that' tells me the 3 average power relative to neutron flux.
4 MR. RAUSCH: I' don't think you can do'what you are 5 postulating. I don't understand how you could simply look 6 at that and separate the two. You can try to--the power is 7 going to be driven--maybe have a 100 to 1 effect of the sub-8 cooling and the feedwater flow versus whatever this other
'9 effect might be. I don't see how you can separate it. The 10 average under the top and the bottom curves is really not 11 subtracting anything. The average of those two determines
- 12. your actual average power level. I don't see any simple way-13 of looking at that and seeing'what the oscillation did 14 -versus what all the driving forces were doing.
15 Analytically you can do_that.
16 MR. LEE: If I may interject. Somewhat in 17 response to your question, maybe it's difficult to separate 18 'those two events so clearly or cleanly, but in general the 19 area under the power pulse is a little bit larger than the 20 recovery phase, the area under the recovery phase, and that 21 indeed is what gives rise to the increase in the amplitude l 22 of the oscillation, until the limiting condition is reached.
23 Then the area under the power pulse becomes equal to the 24 area under the recovery phase.
25 Until that time, the average power keeps on O rie e- rerei 9 (202) 628-4888 cerrer eie-
f--
L:
207 e" --
1 increasing essentially.
2 ~ MR. PFEFFERLEN: And, again, see, now, you are 3 talking about the physics of the pulse itself, and I say 4 superimposed on that is this other effect that if the power 5 is increasing, the system response is going to lower the.
6 level and intend to have it find that limit cycles state 7 where the average power is consistent with the boundary 8 conditions or sub-cooling, pressure and what have you that 9 go into the core.
10 And in fact, that's what we found here. That when 11 we went through the cycle with the oscillations and they got 12 up to their new power level and leveled out, we went in and 13 we ran our 3-D core simulator for that case without 14 oscillations, and for the given boundary conditions we got p
15 precisely the same power.
16 So it went through that cycle of increasing where 17 the feedwater.could sustain it, It reached its limit' cycle- ,
18 and then it's back in.an equilibrium state with the boundary 19 conditions that it must reach.
20 And if it were in flow control, this would have to 21 be a very short transient because the level would come down
- 22 to compensate for that increase in power.
23 So, you know, we are stepping beyond the pure 24 oscillation of what can happen and looking at how does the 25 total system respond and that why I commented earlier that Heritage Reporting Corporation
(_ (202) 628-4888
L 1
" 208 l we are looking at more than just.the analysis. We are l 2 looking at years of experience with_the BWR and what does it 3 do in natural circulation conditions. You know, so it's 4 broader than just this one analysis.
5 But that's basically our view on the average power 6 integrated over time. And why the analysis covers it.
7 Now, with regards to the peak amplitude.of the 8 oscillation, I indicated there that we probably wouldn't 9 expect the oscillations to occur much before you start 10 putting boron in, of course.
So there is some mitigative j 11 effect. The calculations show that under the conditions i
12 that we analyze, that the oscillation magnitude went down at ;
13 least temporarily and in order to get very.large-14 oscillations I believe you would have to.have a plant that g-
\ 15 would be very unstable and I think there were some comments 16 about fuel vendors and fuel designs and I think clearly the 17 thrust is to not let the--we may not be able to cure it with 18 fuel design, but we certainly are not going to let it get i 19 any worse. !
20 So I think that the potential for large amplitude 21 oscillations at the normal corner of the natural-circulation 22 corner in the power flow map in not very high and lowering 23- the water level, we've seen evidence that that's not 24 necessarily going to increase them significantly.
j 25 And the time would be short when we go through i Heritage Reporting Corporation (A_/ (202) 628-4888 1
1
209-1- that period and I think we've got some boron in the core to O. 2 tend to mitigate it.
3 So it may not be a total answer on the' magnitude, .
'4 but I believe the likelihood of those oscillations becoming 5 unacceptably large and realize than in an ATWS we are not 6 talking about a safety limit. We are talking about big 4
7 oscillations before they really interfere with our ATWS 8 conclusions.
9 The ATWS study was done assuming that all of the 10 fuel-entered boiling transition was perforated. And there 11 were no consequences. And so we must maintain the 12 containment.
13 The next threshold for large oscillations would be 14 some kind of different failure mode where you are worrying 15 about coolable geometry and that's awful high. With the one 16 exception of how--and it was discussed at the aseting with 17 the NRC, what does it do to the operator? How does the 18 operator perceive what's going on? Does he get fooled? And 19 there's some work underway in-that that is being done at 20 EPRI and that we are watching very closely.
21 But for these reasons we feel that the ATWS 22 situation is not at an unacceptable state. And that we have 23 reflected that in the report that unfortunately you don't 24 have yet but that will be coming out in the next few weeks.
25 MR. LEE: I'd like to have just one quick comment.
f-Heritage Reporting Corporation (202) 628-4888
r 210 1
1- I hope you are not telling the operators that by lowering L-(f 2 the water level you can kill the oscillation or anything 3 like that?
4 MR. PFEFFERLEN: Oh , no, no, no. In fact, the 5 emergency procedure guidelines, those have words that say 6 "you may encounter oscillations, power-flow oscillations, 7 while you are going through this process. Go aheed and do 8 what you were told to do."
9 There's no attempt in the Rev 40 EPGs to let this
- 10 ~ interfere, so leng as he's got a signal. I believe that if 11 the power is oscillating, he probably has a stronger signal,
-12 you know, if he's down at a fairly low power anyway and he 13 sees it bouncing around, he will clearly understand that he-14_ is not shut down. And that may be good news.
15 What happens level-wise, we're looking at--EPRI--
16 industry is looking at this. Trying to understand is there 17 an impact on this level signal. We believe that, again, 18 because of the size of the vessel and the type of 19 oscillations that would occur that there will be very little 20 impact on his level signal. You know, a short duration, 21 such that he would not know what his level is doing, and I 22 think he would go through this region and follow the EPGs, 23 get the boron in the core, and shut it down.
24 And we are talking about a several-minute window, 25 as I see it. Where he may see these oscillations before the
() Heritage Reporting Corporation (202) 628-4888 I
I l
i i
211
!t . . . 1 boron takes effect and damps them out.
2 Okay. Those are our current views on the ATWS.
3 Let me just ju np, because I said I'd be finished fifteen l I
4 minutes ago and I apologize for that.-
5 Let me just very quickly address two questions 6 that came up from the NRC in that March 22nd letter dealing i
7 with the methods. And I think we've talked about a lot of 8 this already, but basically the questions came up about the 9 ability of the codes to predict the oscillatory behavior and 10 to predict boiling transition.
11 Let me just say that we have, when we did the
-12 22277 which was our stability report back in the 1979 time 13 frame, we did tests of the oscillation, either both flow and 14 power oscillations looking for critical power ratio, boiling 15 ' transition, and we determined that the GEXL correlation is 16 conservative for those--
17 DR. KERR: What does that have to do with 18 calculating oscillation, the fact there's conservative--
19 MR. PFEFFERLEN: What amplitude?
20 DR. KERR: Well, I mean you don't want a 21- conservative correlation. You want one that's correct, it 22 seems to me. If it's conservative, you may get results 23 about oscillatory behavior that are meaningless.
24 MR. PFEFFERLEN: Well, this is entering into the 25~ boiling transition and we found that the results were Heritage Reporting Corporation O (202) 628-4888
r----_-- - - - _
212 i
<^ 1 consistent with the correlations application of the kNl 2 transients and that'a kind of what I've said down here.
3 That really it's been tested against plant transient 4 response, LOCA, and ATWS conditions, and these are more 5 severe. The characteristic period of an oscillation is 6 like--
7 DR. KERR: If you want to avoid my question, okay.
8 I just don't see what conservatism has to do with the 9 current situation.
10 MR. PFEFFERLEN: The point--I guess the concern 11 was are we underpredicting the power or flow that it may 12 take to get a boiling transition and we feel that--
13 DR. KERR: If the correlation is incorrect, which 14 is what conservative means, it seems to me you don't know
'# 15 whether you are predicting it accurately or not.
16 MR. PFEFFERLEN: Oh, I think the co' 91ation is 17 conservative, albeit there's a known amount of conservatism 18 in the correlation, as it's applied to all safety 19 evaluation. The GEXL correlation. And I don't believe--
20 there's no significant--and now I'm not sure exactly how the 21 numbers stack up, but it's my view that there is not a 22 significant departure from the degree of conservatism that 23 is in the correlation as it's supplied to other situations 24 when you look at the stability characteristics. Because 25 they are not that different in some transients that the r-S Heritage Reporting Corporation
(_) (202) 628-4888 l
213 es 1 thing was set up to--
k' 2 DR. KERR: But see, conservative probably meant 3 conservative for calculating a LOCA. Now, what a 4 conservative for calculating a LOCA means the same thing as 5 conservative for calculating this is anybody's guess.
6 MR. PFEFFERLEN: No , the boiling transition 7 correlation was really developed for the plant transient and 8 protection of the same safety limit that we are talking 9 about here. Turbine trip, other events that cause a power 10 increase and that we do the analysis for to show that we do 11 not violate the safety limit and the stability response is 12 not that different. And this is what was demonstrated in 13 these tests back in 79.
14 Now, I agree with you. Conservatism isn't always l}
ks 15 good, but the conclusion is that--
16 DR. KERR: Conservatism in one situation is not 17 necessarily a conservatism in another. And it does not seem 18 to me you have demonstrated that it's conservative for this 19 calculation.
20 What you say up there is that you've found it to 21 be conservative for a number of other situations. Now, 22 maybe that isn't what you mean but that's what th'at says.
23 MR. CATTON: Have you done sensitivity studies and 24 pushed the GEXL correlation in the other direction to see 25 what it does to stability?
Heritage Reporting Corporation g-)s
(_
(202) 628-4888
l' ,
214 l( ) 'l MR. PFEFFERLEN: We've done that down here with 2 the plant transients.and--
3 MR. CATTON: That's a different--
4 MR. PFEFFERLEN: And the conclusion I was drawing 5 from this was that the transient response is consistent with 6 the stability response in that it's a power flow situation,
-4 7 and the two are not that different.
8 MR. RAUSCH: I think.there's a semantics- -
9 MR. PFEFFERLEN: But the direct answer is no, we 10 have not done a sensitivity.
11 MR. RAUSCH: I think we've got a semantics 12 problems here. The GEXL correlation is a dry-out 13 correlation. It's nothing more. It has no relationship to 14 stability other than its ability to. predict onset of
%)
r".3 15 transition boiling. That's all it is. It's a simply dry-16 out correlation.
17 MR. CATTON: --the void production. As if you go 18 through transition you reduce the heat transfer to the--
19 MR. PFEFFERLEN: Okay. If you are talking about 20 void feedback, no. This is.not--this has nothing to do-with 21 that.
22 MR. RAUSCH: No. GEXL correlation is just simple 23 parameters that impact dry out.
24 DR. KERR: So it doesn't have anything to do with 25 oscillation? -
O' aerit 9- rei - cerre (202) 628-4888 tie-
215-1 MR. RAUSCH: That's-correct.
2' DR. KERR: So it shouldn't be up'there?
3 MR. RAUSCH: Well, no, it was an asked question we 4' received from the NRC and that's our response. It's 5 important in the sense of how much real probt:,sility of u"
6 . damage is there if you have a dry-out condition.
7 MR. PFEFFERLEN: It's fuel performance relative to 8 boiling transition or dry out, not to void feedback.
9 Because that's this step down here.
10 'MR. CATTON: Do you deal with it outside of the 11 code? Do you use the--
12 MR. PFEFFERLEN: It's the correlation that when 13 you get a certain power flow condition--
14 MR. CATTON: Is it in the code or outside of the 15 code?
16 MR. RAUSCH: No. It's a loop of the code if you 17 want to. find your thermal margin. But if you are not 18 looking for thermal margins--
19 MR. PFEFFERLEN: You can turn it off and-still 20 . predict your stability characteristics. It doesn't feed 21 into the neutronic coupling. If that's your question, if I 22 understood your question right.
23 MR. CATTON: Does it feed into the heat transfer 24 package within the code? If it does, then it impacts the--
25 MR. PFEFFERLEN: No. It tells you when you pass Heritage Reporting Corporation (202) 628-4888 L
216 1 through a safety limit. If you turn it off, you'll do the
~) ]
%)
2 same calculation it's just that you don't know when you pass 3 through the safety limit. So it isn't to void feedback.
4 What I think you are asking is this. We are in 5 the process of qualifying TRAC and an improved frequency 6 domain code against the available data.
7 There was a lot of discussion about that earlier.
8 We agree we don't have all the answers. We drove the Phase-9 I results. We believe that TRAC is doing a good job in the 10 hydraulics. But you've got to make the full loop. You've 11 got to make the full cycle and that's what this effort is 12 intended to do. To take TRAC, to take our improved 13 frequency domain codes, and test it against the plant data 14 that's out there in the purest stability world and to end up D
k- 15 with some qualified tools. We are not tlm re yet.
16 We believe the code is doing a good job because of 17 the approach that was taken in developing the code, the 18 separate effects tests and the system test, the test loops 19 that we've had. But the final proof of the pudding will 20 obviously be when we test it against the plant data.
21 Now, the one touch point is the LaSalle case that j 22 we ran in the Phase IA. It was not a pure LaSalle but we i
23 set up TRAC with the boundary conditions and it predicted 24 the onset of oscillation and so we felt that, you know, the 1
25 TRAC is indeed doing a good job but by no means does that I
e- Heritage Reporting Corporation
()g (202) 628-4888
l 217 i
,- 1 represent a qualification. l V) 2 So that is yet to come. And our schedule for this 3 is in the October time frame we hope to have the data pretty l
4 well under our belt and some better answer to some of the
]
1 5 questions that you are asking about the code qualification. j 6 Yes, sir, i
7 MR. LEE: What does the document 22277 address? !
p 8 MR. PFEFFERLEN: That was the stability--we 9 submitted that to expand the response of stability to detect 10 and suppress. So it addressed the stability 11 characteristics, what was seen and what the instrumentation, 12 the neutron monitoring system would be expected to see.
13 And in that analysis we did some conservative 14 assessments of the heat flux variation that would get you to p.
\/ 15 boiling transition to see just where we stood. And these 16 tests were done in support of that assessment.
17 Now, what we've learned more recently is that 18 really the flow is more responsive than we had anticipated.
19 Originally we felt that the boiling transition was 20 approached only by the power response with the flow fairly 21 fixed.
22 What we found from the Phase I study is that 23 really the flow was much more responsive, especially in a 24 regional-type oscillation sense. Such that there isn't as 25 much margin to that safety limit as we may have at once fg Heritage Reporting Corporation
() (202) 628-4888
218 q 1- thought. Because of the flow response.
2 MR. LEE: So the first bullet that'you are 3 suggesting there perhaps is not that meaningful, because in' 4 a density of oscillation situation, the. flow would not be 5 uniform--
6 MR. PFEFFERLEN: Yes, but we oscillated flow in
'7 these. tests. We oscillated both flow and power to check on 8 our correlation and we were predicting--actually the way, as 9 I. recall now, .the way they predict a correlation over a 10 ' period of time, .they determined the number of times the 11 boiling transition versus-the number of times you predicted 12 you would enter boiling transition. So it's hard to 13 quantify the degree of conservatism.
14- But basically we looked at both flow and power
- h. 15 perturbations and concluded that the correlation is not
- 16. flawed in this regard. And is consistent with what we've 17 done before here,.you know. So now it's a matter of 18 understanding, if we know the power and we know the flow 19 situation, we believe that the correlation will tell us when 20 we reach boiling transition.
21 The question is what is the power flow? Are we 22 doing a good job of that? And that's what we have to wait 23 for this one to really tie the whole process together.
24 MR. LEE: I think you answered my question, but I 25 just want to make sure if I understood your answer h
v Heritage Reporting Corporation (202) 628-4888
219 1 correctly.
2 In this test that was reported in 22277. report, 3 you had test cases where the flow distribution along_the 4 length of the channel was not uniform.
5 MR. PFEFFERLEN: That's correct. That is correct.
6- HP. LEE: And oscillatory, essentially.
7 -MR. PFEFFERLEN: Yes, yes.
8 MR. LEE: Thank you.
9- MR. LIPINSKI: Now, in your last bullet you refer 10 to your frequency domain code. That's going to be a point 11 code as. opposed to a 3-D, correct?~
- 12 MR. PFEFFERLEN: That is not a 3+D. That may be a 13 1-D. I am not sure.
14 MR. LIPINSKI: If you are in the current domain, O
kl 15 you don't have much choice.
16' MR. PFEFFERLEN: Okay.
17 MR. LEE: You can go to a space dependent 18 ~ frequency domain analysis. It's possible, but more 19 complicated.
20 MR. PFEFFERLEN: The intent of the improved one is 21 to try and account for the regional situation where in the l 22 existing code that we have, we look at core-wide 23 oscillations where we include the recirc loop, the 24 stabilizing effect of recirc loop and we look at single 25 channels without neutronic feedback. What we are trying to-r Heritage Reporting Corporation
( (202) 628-4888 ,
i
l ' 220 H 1. do;here is to look'at single channels with neutronic LOS '
2 feedback to get a feel for the regional--without the
< 3' stabilizing effect of the loop, as well as the core wide r
4 with the loop.
5 And other things have been done to tune:this model:
6- up and we want to test it against the. data'in that.
7 And that concludes my. charts. I apologize.for 8 running past your time.
I' 9 MR. WARD: No problem. W were very interested,_
.e.
10 as you can'tell. That's fine.
11 Thank you very much.
- 12. Let's take a break now and come.back at.3:40.
13 (Whereupon,'there was a short recess.)
14- MR. WARD: The next speaker is'Mr. Phillips.
yO-15 MR.PHILLIPS: I'll follow along to address ~ pretty 16 much the agenda as it's been prepared.
17 Jose' March-Leuba is here from Oak' Ridge National 18- Laboratory and following my remarks, he will.give a brief 19 dissertation on what we have learned with the work that he 20 is doing for NRR.
21 Following the LaSalle event, of course we put out
~
22 a Bulletin 88-07 that said in substance'all licensees should 23 stop relying on decay ratio calculations alone and should 24 implement procedures to avoid regions of instability.
25 The Owners Group set off to do sone calculations
'O' Heritage Reporting Corport. en V (202) 628-4888 i
_ _ _ - _ _ -_-m. .m__________--um_-___ -
f 221 f'T V
1 and.they found that there were regions where contrary to 2 ' findings in their previous calculations, they did violate 3 safety limits and that was followed by the Bulletin 8807 4 Supplement 1 which in effect said modify your procedures to 5- make.them stronger, to include automatic--well, not 6 automatic, but to include more positive actions to avoid-7 possible violation of safety limits and more positive 8 actions included manual SCRAMS.
9- And BWROG set off to do some additional studies ;
10 seeking a more positive long-range solution with hardware 11 fixes as you've heard here in the previous presentation.
12 The Phase IA studies are the ones that you were-13 discussing earlier where a TRAC calculation that was 14 performed that would solve' conditions and the objective was 15 to determine if thermal power increased due to the l-16 oscillations.
17 The curve was presented to us last November and we 18 looked at-that curve and there was an integrator on the 19' curve and it seemed to us that eyeballing it we could see an 20 increase of about 4 percent in thermal power over a 50-21 second interval with the oscillations growing to peak power 22 level of 200 percent.
23 We weren't prepared and we still aren't to say
-24 whether this is significant or whether it's not significant. I 25 It depends partially on whether that rate increases.
, Heritage Reporting Corporation (202) 628-4888 c_ _ - _ - - _ - - _ - - - _ _ - - - - _ _
I 222 Jx 1 We had indicated to BWROG that we would like to
' _] '
'N 2 see some more calculations, perhaps extending the cales for 3 a longer period of time and also doing some at larger 4 oscillations to look at the effect on thermal power.
5 They indicated that the calculations are very 6 expensive, which they are. And that they had done some 7 heavy-duty analyses which indicates that the slope of the 8 curve had stopped increasing and was due to system effects.
9 We understand that will be in the report, which we 10 are expecting next month and we have also been doing some 11 studies of our own in this area and depending on the outcome 12 of our own studies in conjunction with what they have to say 13 in that report, we haven't taken a position on whether we
, 14 are going to need these additional calculations.
15 Additionally, they are doing some work at EPRI to 16 investigate the interaction of the operator response to ATWS 17 with power oscillations and we haven't had a lot of feedback 18 on that work yet.
19 We understand that is not in the report that we 20 will get next month. But we are very interested in it.
21 As I said, the report is overdue. We now expect 22 it next month. The recommendations for long-term solution, 23 as you saw, are expected in late 1989 with initial progress 24 and we hope some useful information in September.
25 And then following the BWROG recommendations,
' Heritage Reporting Corporation
(\ (202) 628-4888
l 223 )
1
-( g -
1 'there will have to be implementation of the long-range i AJ 2 solution on individual plants. And plant-specific schedules 3 will have to be set up for that by individual licensees and i'
4 it will be based on scheduled shutdowns and whatever 5 procurement, labor, and so forth, is involved.
6 As far as the response to Bulletin 88-07 7 Supplement 1, all they were required to do was to tell us 8 they have implemented the interim operating recommendations.
9 All of the licensees have responded. Some gave us more 10 information. We would consider that the responses met was 11 required of them.
12 There is an inspection program to look at the 13 plants to see how these procedures have been implemented 14 correctly, and we don't have results of that in yet.
()
15 As far as additional programs supported by NRR, we 16 have for over ten years supported work at Oak Ridge and to 17 assist us in licensing reviews with respect to stability and 18 this included development work and improvements on a simple 19 frequency domain code, Lapur, and Jose March-Leuba will 20 address some of the things that he has learned using that 21 code, and we have to say that he's done things with 't that 22 we didn't think were possible. And we've been very happy 23 with the support that we've received from Oak Ridge in this 24 area.
25 As far as future plan and schedules goes, we have
('} Heritage Reporting Corporation
() (202) 628-4888 i
L _ -- --- - - - - - -
224 1 to review the reports and are interested'in review them, b(~'
2- that we will receive next month. We will have to carry on a 3 continued interaction with our consultants e.t Oak Ridge.
4 Also at Brookhaven National Laboratory.
5 And we are participating very closely with RES who 6 is conducting research in this area and has formed a .
I 7 technical program group and we have meetings periodically, 8 fairly frequently, to discuss the calculations that need to 9 be done to investigate the unresolved areas.
10 Our interaction with BWROG on the ATWS power 11 increase possibility and its significance should be 12 hopefully concluded during 1989 also.
13 There are technical specifications involved which 14 govern the operating procedures to avoid operation in
(,)
15 regions which are susceptible to instability. And some 16 licensees have found some inconsistency in their tech specs 17 with the interim operating recommendations and have elected 18 to submit changes. Also in some few cases there have been 19 some changes to the interim recommendations involved because 20 of peculiar operating requirements at some specific plants.
21 So where that's the case we are involved in those reviews so 22 we have some of them that are ongoing now.
23 The long-term solution we hope will be proposed i
24 before the end of the year and that our review can be 25 complete by near the end of the year or early 1990.
1 A Heritage Reporting Corporation (202) 628-4888 ,
l 1
_.________._._.._______.______.___________J
225
_ (mj 1 We will, of course, again have to review the 2 individual licensees' implementation.
3 And we are now due to provide a final report to--
4 what we hope will be a final report--to the Commissioners at 5 the end of February 1990. And, of course, whether that will 6 be a final report or not will depend largely on whether the 7 long-term solutions have been resolved and the program is 8 well in place to our final solutions.
9 I might also say, just as an add on in connection 10 with the Oak Ridge work, that they have also developed a 11 noise analysis system for decay rate monitoring and I 12 believe that this is the first one that was developed and 13 that the European systems are basically feeding on that
/~N 14 technology.
\~)
15 And I'd like to introduco Jose March-Leuba from 16 Oak Ridge to discuss the work in more detail.
17 MR. MARCH-LEUBA: I would just to thank Larry for 18 giving me my full half an hour. I was kind of worried at 19 the beginning that we were not going to be able to get to 20 our presentation.
21 He knows me and he knows how I enjoy these 22 -presentations very much and I could talk for hours so he 23 especially told me to stay within your half an hour or they 24 won't allow for you.
25 So I have just presented some very little amount l
1 l
() Heritage Reporting Corporation (202) 628-4888
l 226 1 of work that we've done very recently. And judging from the d(T 2 discussions there was before, it seems to be fairly relevant 3 I'd say.
4 I wish I had known what questions you'd had before 5 and I would have addressed the problem differently, but 6 bas.4 , ally I think you'll be satisfied with some of the 7 answers. I hope you will. Okay.
8 This is an outline, more or less, of the two 9 things I'm going to cover. I'm going to first talk about 10 limit cycles. And I am going to kind of give you an idea, a 11 physical idea, of why the limit cycle occurs. I am not going 12 to get into why there are instabilities, okay. That would 13 be a completely different presentation and would take 14 another half an hour.
k ') 15 And so assuming that the reactor is going to be 16 unstable, why is oscillation not growing and why does it 17 limit, why doesn't it stay bounded.
18 And then we'll answer once we understand why that I
19 happens we will answer whether there is another power 20 increase due to oscillation.
21 And we'll go into the details later.
22 Then I have planned, if I have some time, to talk 1
23 something about bifurcations because it's very relevant to 24 this average power increase.
25 And I did my dissertation on this area so I kind l
l Heritage Reporting Corporation
(^)g
( (202) 628-4888 l \
227
'l of like$it. It's my son, or something. But-probably we L 12 won't have time to do.it. -We'll see.
- 3. . Then I'll tell you about some of the calculations 4 we have performed on what happens when an operator uses.the 5 water level at natural circulation.
6' Common wisdom says that if you reduce the flow, 7 you reduce the water level, it's going to be bad for your 8 stability. Well, common wisdom is usually wrong in this p
9 business. The only thing you can do is compute it and see 10 what happens.
11 'I will show that the reason the water level.can be 12 good or bad depending on what your regional conditions are.
13 So it is very hard. If you get something out of my-14 presentation, let it be this. It is extremely hard to do-L 15 generalities from anything related to stability. It is all li6 very extremely sensitive to initial conditions.
17 Given the questions that were before, a fourth or 18 a third term in there, which is a stability monitor, and the 19 thing that Larry referred to, we developed a noise' analysis 20 monitor and I heard some comments before that'it was kind of-21 magic'and all that. It really is not magic. There are very 22 important mathematics behind it. Very rigorous development 23 behind it. And it can be trusted when it's used properly.
24 A stability monitor is something--it's just black 25 box that you plug into an existing LPR detector. It tells Heritage Reporting Corporation (202) 628-4888
I
-i!
228 E 'l you how stable the reactor is.
- h. ~
2 -The reactor does not need to be unstable for it to-3 know. -It will tell you how far you are from stability. It 1
4 'will give you the decay ratio measurement. J 5 Whenever the decay ratio.is greater than'1, the 6 rer.ctor becomes unstable. If it is .5, you are very stable, 7 you are okay.- You can pull control rods. If you are at 8 .09, be careful. If you pull into a rod, you might get-9 instability.
10 So nothing-exists on commercially available 11 packages like that that can be sold.
12 And there was another comment that PCs are cheap 13 but once you put--but if you say that, that 's because' you -
14 haven't bought software recently. Software for PCs can cost 0- 15 you $50,000 easily. Whenever you have to put one or-two man 16 years in development or some business software required.100' 17 man years and you have to charge dearly considering there 18 are fifty plants. The real expense from the safety monitor 19 is not the hardware but the software and the developm nt 20 that goes with it.
21 So back to the problem at hand. And we'll start 22 with the limit cycles and why there is a limit cycle, okay.
23 And I'll have to do some kind of mathematics ahead
'24 of time to get to the results. And I want to show to you--
25 and I'm going to tell you first what I'm going to show and Heritage Reporting Corporation
^
(202) 628-4888
229
/m '
( ,) I then I'll go through the details--is that the time varying 2 portion of the feedback activity is essentially a sine wave. l
)
3 It does not have higher harmonic contamination. That is, 4 even though we don't know the limit cycle, the power trace 5 has very sharp peaks and shallow valleys, and it has lots of 6 harmonic contamination. It doesn't look at all like a sine 7 wave.
8 Once it goes through the hydraulic feedback and it 9 gets converted into reactivity, it is a sine wave. And this 10 might seem like a rather irrelevant finding, okay. But it 11 really simplifies dramatically--well, we can do that later 12 on. If you assume that your reactivity is a sine wave. It 13 allows you to do some calculations and to develop reactor 14 independent correlations that relate to the average
/")S
(
15 reactivity increase with the amplitude of oscillation.
- f. But first to the second point. Let's not get 17 ahead.
18 We will show that assuming that the reactivity is 19 a sine wave, that this reactivity oscillates as a sine wave i 20 and the power oscillations remain bounded. That is, when 21 you have a limit cycle, the thing is not diverging, then the 22 reactor must be critical. That is, there must be an average 23 value, an average or a DC component in the feedwater t
24 activity. And thus, this must be a power increase during 25 the limit cycle oscillations. I mean it is a must. If you O) i
'~'
Beritage Reporting Corporation (202) 628-4888 l -
1
230
([ } l- do not have a power increase, you will not have a limit 2 cycle. You will have a divergent exponential.
3 And I will show you more or less how this works 4 and you will see that based on these two points how we can S develop some correlations, much like the GEXL correlations 6 that we were talking about.
7 We can find some correlations that apply for all 8 reactors that will tell us given an amplitude of oscillation 9 how large is the average power increase.
10 And for.that we have to back up a little hit more 11 and give you some background. And in the view of the an 12 operator, a typical BWR is something that when he moves the 13 control rod he gets some power out of it. So it is nice to 14 put tha dynamics of a BWR in this form. Some people like to 15 think that the thermal hydraulics, which is the feedback in 16 this figure is the driving--that's the main box. And 17 deed 17 it probably is, okay. But basically it doesn't really 18 matter where you put your input and your output. This is a f 19 closed group system so everything is important.
20 And what I am presenting here is the transfer ---
21 That is, given a frequency of 1 hertz, you get 5 DBs out.
22 MR. LIPINSKI: Why is it shown that way as opposed 23 to being on the zero DB line? I assume this is your zero 24 power kinetics curve 25 MR. MARCP-LEUBA: Yes.
i (m) Heritage Reporting Corporation (202) 628-4888
231
~T 1 MR. LIPINSKI: Why is it biased upward?
(d 2 MR. MARCH-LEUBA: I don't know. I just wanted to 3 show you the shaps. I really don't--the amplitude--what I 4 gave is for the relative amplitude. Actually no answer.
5 These units are very weird. This was developed by a friend 6 of mine, Pedro Telwe, and he is Spanish, and as such, is a 7 little crazy. He's very peculiar. So he used percent of K 8 over K and power is normalized--all the units, you have to 9 really look in the manual and see where they are. These 10 have very good reasons behind it, okay.
11 MR. LIPINSKI: Okay.
12 MR. MARCH-LEUBA: But you cannot just find it.
13 The units are not what you think. That's the reason. It's
.- 14 probably a 6.5 or something like that. This is not in
15 dollars.
16 This is the important region. It's about half a 17 hertz. And a half a hertz shape's is broad and flat and 18 uninteresting. But the thermal hydraulics has a roll off of 19 about .1/.2 hertz and it breaks down very rapidly. It has a 20 tremendous filtering effect. You'll see that a half a hertz 21 right here, it's Ebout that point. We are 20 DBs below the 22 value at zero frequency. That is, we are down by a factor 23 of 10. At 1 hertz, that would respond to a second harmonic, 24 and you can see what I am going for with my sine wave, I was 25 telling you before.
(~% Heritage Reporting Corporation
(/ (202) 628-4888 l
l l
L _ _ ----- - ---
'232-1 Even if we have a limit cycle that'has tremendous
}
2- amount of harmonic contamination, this filter is going to 3 -kill it off. 'So when you are down at.1 hertz, you are down-4 almost 40 DBs plus 20 here. You are;60 DBs below. You have 5 a factor of 1000 down. So that harmonic,-you are not going R
Okay.
6 to see it.
7 And that's basically why we are going to have a 8 sine wave.
9 And I have another slide that proves it more or-ICL less. But when you multiply these two you get the upper 11 loop reactivity to power down four points. And this is a 12 typical that shows it.
13 The solid lines again and this is more line as a f- 14 the phase. First this is one of Dr. Lee's questions also.
'V 15 He asked many questions which were relevant before,_.is what' 16 can you do to stabilize a reactor. Why are we trying to 17' mitigate?- Why are we trying to avoid instability and.why 18 not avoid them for sure inherently?
19 If you remember any linear dynamics, 20 instabilities occur when your phase goes below 180 degrees.
21 That's right there. And your gain is written at 1. So that 22 of all the phase, all the frequencies you can have, only 23 because this little darn thing went down by 5 degrees below, 24 to 180 degrees, you are able to have instabilities. If this 25 shape was 10 degrees above this line, I mean it's nothing.
Heritage Reporting Corporation
]N- ' (202) 628-4888
233 1 =You would not be able-to have an instability. It's not that
/-i
%)
2 'you would be more unlikely. It's that it will be impossible 3 to have an instability.
4 Unfortunately, God was bad to us when he designed 5 BWRs-and.we copied them from him, and there's no way to 6 raise that thing without going into great expense as Tom 7 said before.
8 You could, if you had an active system, you have a 9 control rod, and you have an active controller, it would'be 10 very easy to raise it. But that wouldn't be.any easy thing.
11 You would have to.have a high -- control rod beam and a 12 frequency to raise the frequency.
13 Okay. Other things you can do--things-you can do.
14 easily is reduce the amplitude, reduce the gain of the --
'O 15 That you can fairly easily. But that is also expensive.
16~ You cannot change the phase. And it is very unfortunate 17 that just because of this damn 5 degrees we are--of course, 18 it has made my career. I'm living off of it. But this h
19 little thing in there.
20 So if somebody has some ideas of how to raise 10 21 degrees off a phase with a heating design, you will not win 22 the Nobel Prize but you can patent it and make some money on 23 it.
24 MR. LIPINSKI: How are you going to change the 25 gain in the physical system if you don't have a pot in L
"q- Beritage Reporting Corporation
() (202) 628-4888 L
E 234 2 there?
^
2 MR. MARCH-LEUBA: Well, you do change--when you 3 change the arrangement, for instance, you are. changing-the 4 work co-efficient, you are changing the game. There are 5 many' things you can do that change the game but very few 6 that change the face. Becauce that face is time lag. And 7 that time lag exists. I mean the voids have to move through 8 the core.
9 MR. LIPINSKI: The phase is a function of how 10 quick the gain curve is falling off. If I can change the 11 rate of change of the gain curve, I'd change the phase. So 12 if I'm going to change enrichment, I'll have some effect on 13 the rate of change of gain and that in turn will change 14 phase.
O- 15 MR. MARCH-LEUBA: I'm sorry to say no, but no.
16 MR. LIPINSKI: They go hand in hand. You cannot 17 have one independent of the other.
18 MR. MARCH-LEUBA: The phase always drop as one 19 decay by decay. Okay, that's your gain and your phase as 90 20 degrees. Okay. But you have to change--if you want to 21 change the shape of the curve--I was not saying change the 22 shape of the gain curve. I was saying keep the shape of the 23 curve and move it up a notch.
24 MR. LIPINSKI: And all I'm saying is if I don't 25 want the phase to be there, I can slow down the shape of the Heritage Reporting Corporation O (202) 628-4888 1
235-L(~SJ 1- gain curve and I'll effect a phase. The two go hand in
%)
2 hand. They are not.one independent of the other.
-3 MR. MARCH-LEUBA: Absolutely, absolutely. The 4 phase--once you draw me the behavior of the' gain,.any 5 freshman in electrical engineering will draw you the phase.
l 6 They'are perfectly related.
7 But the gain has an extra degree of freedom which 8 is the absolute amplitude. You cannot change that in a 9 phase.
10- MR. LIPINSKI: And that was my first question. If 11 you are simply going to make a gain change, how would you 12 physically accomplish a straight gain change? I think you 13 are going to effect-the shape by whatever you do.
14 MR. MARCH-LEUBA: I'm sorry. Larry is back there.
15 .You buy 9 by 9 fuel instead of 8 by 8. And then you 16 ' increase the heat transfer co-efficient. The heat transfer 17 co-efficient what it does is this -- which doesn't show 18 here very well, but there should be a point of 3 hertz 19 somewhere in here that breaks.
20 MR. LIPINSKI: You don't have to go into it now.
21 We are holding up everybody else in this meeting.
22 MR. MARCH-LEUBA: You can changs the break 23 frequency of the fuel time constant. You can change time 24 constant in the fuel. That increases the frequency toward 25 the right. You are erasing the gain a little bit, a half of O Heritage Reporting Corporation L /- (202) 628-4888
p' 236L 1:
g-) I hertz. But.the phase is still 90 degrees during the
%) ~
2 behavior.
'3 If your break frequency is two decays below where L 4 -you are interested, you can change the gain but the phase-is
.' 5 '90 degrees.
6 If you come up with a way of changing that, please-7 let me know.
8 MR. LIPINSKI: Well, you want metallic fuel rather 9 than oxide.
10 MR. MARCH-LEUBA: I thought I was going to make 11 .this presentation to a bunch of managers and they would be 12 more interested in results than how you get the results.
- 13. And that's why 1 put some figures and very few equations and 14 then I found out this was a really technical session and I 15 should have gone into more detail.
16 But basically to. prove the reactivity we must
~
'17 state the limit cycles have been observed. There is no 18 doubt that they exist. And we have observed also the limit 19 cycles are stationery singles. That means once you reach 20 the limit cycle they are periodic and they don't keep 21 growing and the amplitude is bound. And this is before the 22 bifurcation regiment and we won't go into detail until later 23 maybe.
24 Therefore as long as they are periodic, they can 25 be expanded and your limit cycling power can be expanded as f
Heritage Reporting Corporation (202) 628-4888
N F
237 L -
l7 f() ~1 a sum of sines times the basic frequency and'the harmonics
[ 2 of power.
3 MR. CATTON: Now, each cycle of bifurcation, d.o 4 you mean in time or in space or both?
5 MR. MARCH-LEUBA: If you hold on, I'll show you.
6 It is--okay. I'd better answer it.
7 What happens is the limit cycle itself can become j 8 unstable. Okay. The amplitude limit cycle would become 9 unstable and then you will have a limit cycle that has a 10 high peak and a low peak and then a high peak and a low 11 peak. Okay. The amplitude limit cycle becomes unstable 12 when it reaches a limit cycle of its own.
13 And if you put it in' terms of -- maps and one 14 dimensional map it becomes a bifurcation. And that 15 bifurcation is not that you have two possible solutions.
16 It's that your solution becomes unstable and now you have--
17 MR. LIPINSKI: I understand what a bifurcation is.
18 I was just--the one you described is time-wise. Do you get.
19 spatial bifurcations as well?
20 MR. MARCH-LEUBA: No, no, no, not spatial.
21 MR. LIPINSKI: Okay.
22 MR. MARCH-LEUBA: Okay. You lead me into it and 23 now you have to see it.
24 This is what happens to the limit cycle in a 25 simple model as you change that gain and you might not have O rit e- e rei 9 (202) 628-4888 cere r eie-
i i
i 238 )l e3 1 it there because I took it out because I wanted to stay in G
2 the half an hour. I have 300 figures in here. You can find 3 it in my dissertation or in some other papers I've written. i 4 What happens is, as you change one part, say, 5 enrichment, and you change the feedback gain, and here we 6 have a 1.2, 1.4 and 1.5 We're increasing it. We are making 7 the system more unstable.
8 First we have a limit cycle--as the discussion, it 9 grows and then it stabilizes. And if you look at the 10 amplitude, it looks smooth. There's a nice transition.
11 Here we make the limit cycle more unstable, you can clearly 12 see the oscillation starts to oscillate. The amplitude ;
13 starts to oscillate. And eventually converges and this 14 limit cycle just like that.
in k- 15 Now, in this case, what we have is a lot of 16 oscillation but it never converges. In a sense, the 17 amplitude of the limit cycle has become unstable. And has 18 reached the limit cycle of its own and now it is oscillating 19 among the two. And that's the type of bifurcation I am 20 talking about.
21 Now, these limit cycles, the one that is 22 superimposed on the original one, becomes unstable too. Or 23 has another bifurcation and another and another. And those 24 bifurcations occur closer and closer in time until you have 25 bifurcated an infinite number of times and your solution is j^
Heritage Reporting Corporation (m,) (202) 628-4888
I i
239 f . 1 aperiodic. It does not have a period. It's random.
( )'
2 Still, it is bounded within a stringent factor.
3 Okay.
4 We have this figure here. You can see that the 5 oscillations could be fairly large. Mostly what it's trying 6 to reach oscillation, you can have--you usually have a very 7 large peak, very early in the transient, that kind of puts 8 you in the -- state, and then it centers onto a limit 9 cycle.
10 So for safety reasons, we might be interested in 11 knowing what is the maximum peak we are poing to get during 12 the trans. tent before we reach the limit cycle.
13 Normally limit cycles are more like this one now 14 here. It kind of grows smoothly and nicely and settles
('s 15 down.
16 But if you run into a real bad one, and we are 17 talking real bad one here, 2,000 percent oscillation, it can 18 have a tremendously big overshoot at the beginning and then 19 settle down to a smaller one. And this is closed.
20 But you might even be prone critical at this 21 point. I mean your reactivity might get to be prone 22 critical and have a really large peak and then settles down 23 to e. small limit cycle.
24 j 25 g Heritage Reporting Corporation j i (202) 628-4888
240-93 1- MR. LEE: -I hope you are not implying that there's
~2 a good chance of observing bifurcation type of oscillations -
3 in operating boiling water reactors.
4 MR. MARCH-LEUBA: Yes, yes, I am saying that. If 5 you get a limit cycle that has more-than 500 percent of the 6 steady state operating condition of oscillation. Measure as 7 a serious state of a point. If he has more than 500 percent 8 of oscillation, it will bifurcate.
9 MR. LEE: I'm not saying that's it not impossible.
10 And I'm asking your judgment of how likely it is to get into 11 bifurcation type of oscillations in operating boiling water 12 reactors.
13 MR. MARCH-LEUBA: What I.am saying--okay. How 14 likely it-is'I cannot put likelihoods in this because I T -
15 don't know how crooked, how perverse, an actual power save 16 you can give me.
17 I know that I can imagine an actual power save, 18 and I have calculated that power save. And that actual 19 distribution so that 60 percent power and 40 percent flow, 20 which is within the operating range you will have 21 bifurcations. For sure. You are unstable like hell. Now, 22 that power saved is very unrealistic. So how likely is that f
23 power save to occur? I don't know.
24 But I can give you the paran.cters of a reactor 25 that will bifurcate. And that will be a commercial reactor.
7g Heritage Reporting Corporation s) (202) 628-4888 l
l
- l. .
241 j-( if Now, I don't know if you can go on and. generate
~
2 ' actual- power saved I am goir.;g to give you, it's looks awful.
3 It has a'3.2 factor on the operating mode. But if we are 4- ' awarded a.-- with a large amplitude of oscillation - , we 5' should understand the oscillation phenomena because if you.
Ei have a large oscillation, you will have bifurcation. Juud I
, 7 told you at more than 500 percent of the steady state,.you:
8 will start having them.
9 MR. LIFINSK1: In the early Borax experiments,
'10 reactivities were put in in large amounts in the non-linear 11 phenomena grew exponentially without ever bifurcating.
12 MR. MARCH-LEUBA: Yes. Let me show--
13 ~ MR. LIPINSKI: And that was metallic fuel though.
14 It was not an oxide fuel. So the time count is relatively 15 shortened in transfer.
-16 MR. MARCH-LEUBA: Something like this?
'17 MR. LIPINSKI: That's correct.
18 MR. MARCH-LEUBA: That correct? Okay.
19 Well, let me go first-to this one. So you believe 20 the reactivity is going to be a sine way. And you see that 21 it is. Here we have the power. You have a limit cycle of 22 500 percent. That's the largest limit cycle you can get 23 without bifurcations. And then you have a reactivity which 24 is on this scale down here that goes-from almost $4 peak to 25 peak. I mean that's a lot of reactivity. And still you can Heritage Reporting Corporation O- (202) 628-4888
____ _ m__.___________m_____m.__ - --_ ___.____.____
242 i
['j 1 see it's clearly a sine wave. It has some contamination I- :
s-2 believe, but on first examination we can imagine it has a 3 sine way.
4 Now, knowing thct in our reactor, the reactivity 5 is going to look like a sine wave, we can now study just the 6 effects of a sine wave on the dynamics of the neutrons. As 7 you said before, if we put a sine wave in a_particular 8 model, just simply point kinetics, you obtain an oscillating 9 behavior in the power.and another power that grows 10 exponentially also.
11 'And that is why the question of is there an 12 average power increase? If you see the figures, you say, 13 well, gee, they tell me they have 120 seconds and it has 14' stabilized. But is it really here and it's soon going to 15 take off or not? This is a very valid question.
16- The second argument that I have heard is that, 17 yes, this speaks of higher in the peak and lower in the 18 valley so they might not average out.
19 Indeed they tend to average out because when you 20 have a very large peak you tend to have a longer period for 21 the next peak. So even if you have a very sharp peak, it 22 will take longer for the next one to appear. So that the j i
23 average power, as you see, it increases. It has to 24 increase. But it's not tremendously large amount. Okay.
25 So if we now see, now try this model with a sine Heritage Reporting Corporation (202) 628-4888 i
I o____,__________ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ . _ _ _ _ _ _ . _ _ _ _ _ _ _ _ . _ _ _ _ _ _ __.__._________.____.________.____.___________________J
l l l
243 i 1
-( )- I wave that has an average, negative average, that's a l 2 subcritical reactor, -- a subcritical-reactor that has half 3' a'subcritical reactivity with a sine wave that is .7 or so.
- 4. And what we see is that the oscillations will die '
l 5 down. The reactor is subcritical and it will tend to dump l 6 out.
7 But if we happen to hit the right subcritical 8 value and that's this right here with just the right 9 suberitical value and just the right amplitude oscillation, 10 the oscillation-is maintained. And you have.your limit 11 cycle.
12 And that's a claim that if you do see a limit 13' cycle oscillation which is bounded, then I have proved that 14 the reactivity must look like a sine wave. And thus it must-15 have an average reactivity value for the oscillations to 16 stay bounded. Otherwise it will grow or go down, unless you 17 have the right amount of reactivity inserted. And this 18 amount of reactivity inserted is accomplished by increasing 19 the average power. And that is through your power co-20 efficient, power to reactivity co-efficient, you get this 21 amount of reactivity decrease.
22 For each amplitude of oscillation, there is only 23 one average reactivity for which you get a limit cycle 24 oscillation. Only one, because if you have more it grows.
25 If vou have less, it dumps out. And you can measure that m Heritage Reporting Corporation Ig -
(202) 628-4888
l
-244 L
1 and I have-run many times--this simply code and found the 2 correlation. I am sort of trying to show this by hand. I l-3' find that there's more correlation that tells me how-much-4 this value should be given the amplitude of oscillation.
5 And more than that, once I know this value of.
6 amplitude of oscillation, I can calculate what is the peak-7 to peak value of oscillation. So I can get the correlation 8 between the peak-to-peak value of oscillations and the 9 reactivity that has to go down.
10 And all this is general for any reactor. The only.
- 11 thing that might-change it.is changes in decay neutrons.
12 But for typical BWR plants, and we all know better 13 changes are in the cycle and all these things, we can get a
-14 correlation good enough for government work.
15 So what I did is I took a model that I had 16 developed some time ago that represents-fairly well the 17 dynamics of BWRs. This is just kinetics. I took another 18 slide.- I have now 60 groups if somebody is --
about it.
19 And then we have one node for the fuel and two nodes for'the 20 energy continuity questions. How good is this model? Well, 21 I have compared this model with the results of the Lapur 22 Code that I already ment'ioned. It's very detailed. We have 23 seven channels and 120 nodes per channel actually. It's 24 very detailed. And the solid line is this model.
25 By adjusting only these three parameters. A1, A2 Heritage Reporting Corporation O- (202) 628-4888
~1 l
245-(r
'/ -
1 and A3. So this model, within the linear domain, represents )
l 2 at least as well if I know the parameters that I choose as {
3 well the linear behavior of the BWRs as well as the 1000 4 node Lapur. Now, this codel that I'm going to start with. q i
5 That's going to give us the non-linear behavior. '
6 MR. CATTON: Linear stability gives you sure 7 instability.
8 MR. MARCH-LEUBA: Sure?
9 MR. CATTON: When you do a linear stability 10 analysis, you get a sure instability boundary. Anything 11 above it, it will go unstable.
12 MR. MARCH-LEUBA: That's correct.
13 MR. CATTON: If you don't go below it--
14 MR. MARCH-LEUBA: Below it you are stable.
()%)
15 MR. CATTON: Do you do finite amplitude 16 instability where you put a step in parameter and see i 17 whether it will go unstable or not? Not non-linear 18 stability.
19 MR. MARCH-LEUBA: No. The answer is no.
20 MR. CATTON: You just do linear stability and then 21 into finite amplitude behavior?
22 MR. MARCH-LEUBA: Yes.
23 MR. CATTON: So the boundaries ought to be real3y 24 grown over but bigger. Maybe a lot bigger.
25 MR. MARCH-LEUBA: That's a good question. Since Heritage Reporting Corporation
(~#)
(202) 628-4888
k 246 n/^j 1 you have lots of noise in the reactor. The APRM has 3 to 5 b
2 percent all-the time.- -- average power is 1500 megawatts, 3 and it's supposed to be plus or minus 3 percent of that 4- which is a significant amount.
5 So i:E you are trying .to tell me--you have noise, 6 you have unknowns, you have non-linear fix. Some kind of 7 nice words to talk about. But it just confuses the issu6.
8 Trying to determine this boundary with three 9 significant midgets--let me finish because it's the wrong 10 thing to do. I mean everybody knows that they can calculate 11 the results of a test with fairly good accuracy. But they 12 cann:t -- unstable for a cycle. Simply because they do not 13 know the conditions. They don't know that this particular 1
14 reactor will be playing with that sensor and.there was going 15 to be a -- for that particular condition and that 16 particular one.
17 So that given that the operating conditions are 18 not so well defined, trying to get within the 5 or 10 19 percent error, it's in my point of view, and this is my 20 personal point of view, is not relevant.
21 MR. CATTON: I was asking a question.
22 MR. MARCH-LEUBA: Yes, I know, I know. But there 23 is a tendency to--
24 MR. CATTON: It sounds to me like you don't have 25 to worry about non-linear--if that's the case, that's nice.
1 O Heritage Reporting Corporation (202) 628-4888 l
l
247 0
ps .1 It makes life much easier.
- %.))
2 MR. MARCH-LEUBA: I gave'you the wrong idea- If 3 you look at this, the Swedish have performed lots of tests 4 with flow because they have flow sensors. You have a 5 tremendous amount of nice academic research to be done in 6 there. A tremendous amount. I don't know how relevant it 7 is for safety, but before you get into-our clear cycle 8 definition, when you are in the low limit cycles, 5..to 10 9 percent amplitude, you have a tremendous amount of beating 10 effect. -You see the cycle goes like that and then goes 11 down and then goes like that and then goes down.
12 There are lots of parameters changing. Maybe it 13 is due to the fact that the flow in the channels is 14 competing and they have different frequencies. There is a 15 lot to be studied in that area.
16 What I am saying is we have limited resources.
17 Let's concentrate on what we think is relevant. And that's 18 why we have wise men like you to tell us'what you think is 19 relevant, among other things. That is why we have these 20 committees and these reviews.
21 MR. CATTON: I wasn't telling. I was asking.
22 MR. MARCH-LEUBA: I'm sorry. Okay.
23 MR. LIPINSKI: Where did you obtain that detailed 24 curve at the bottom? What produced that curve?
25 MR. MARCH-LEUBA: It's a frequency to main code.
.- Heritage Reporting Corporation (202) 628-4888
I l
l 248 )
'l 1 very similar to-- J
'n' xj.
2- MR. LIPINSKI: 'Okay. That's how you generated ,
)
3 that curve. Now, in your fuel equation, what are the 4- parameters K and the function FFT7
-5 MR. MARCH-LEUBA: Okay. This is an old figure and 6 I should not have used it. K is something that you change' 7 to change the dynamics of the system. You think of it as 8 being the -- co-efficient.
9 MR. LIPINSKI: But you are saying that's not one-10 of your parameters. Because you only said A1, A2 and A3 11 were your parameters.
12' MR. MARCH-LEUBA: Well, that was a mis--it is one 13 of the parameters.
14 MR. LIPINSKI: Okay.
15 MR. MARCH-LEUBA: This is one of the parameters to 16 get a good curve.
17 MR. LIPINSKI: And what is FFT?
18 MR. MARCH-LEUBA: Okay. FFT is a delta function 19 that we use to excite from the -- ' point.
20 MR. LIPINSKI: Okay.
21 MR. MARCH-LEUBA: It's a mathematical--
22 MR. LIPINSKI: It's not part of the closed loop?
23 MR. MARCH-LEUBA: No. Okay. Now, I didn't say K 24 was a parameter of the system because once I found the value 25 of K that fits this curve, I change K. So I do not use the Heritage Reporting Corporation O. (202) 628-4888
249
) 1 value of K that fitted that- curve. I'll go up. I need that 2 value to feed the curve. Then this is my degree of freedom.
3 I change K to change--
'4 Well, you see this model now. I can converse the 5 -limit cycles and calculate the average power. increase and 6 all that. So this is-the main result, which is a 7 correlation. An empirical correlation we must say, 8 empirical numerical correlation that we have found by 9 computing this limit cycle for many values of K.
10 And what we have is a value of the peak 11 oscillation power in percent of the steady state condition, 12 how much reactivity one needs to sterilize it in the cycle.
13 And you can see that originally assessment is 14 fairly linear and it's of the order of .3 cents for every--
l 15 well, whatever. There's a value. And then when your first 16 bifurcation starts, that reactivity changes and right here l 17 is where the periodic region begins. So that bifurcation --
18 has an important effect on this curve.
19 Now, this curve is for any BWR. For every reactor 20 that has a -- that I used. Okay. If you multiply that f
21 curve by the atypical power co-efficient, that's how much 22 reactivity you have by a 1 percent increase in power, you 23 can convert that curve to average power increase. So that 24 this curve is only valid for the particular reactor I used 25 the power co-efficient from.
! Heritage Reporting Corporation (202) 628-4888
250.
f-1 1. -And this curve tells you that for.'each peak of e 1 E 2 ' oscillation power, you must have a power increase of this 3 much.- And you see, it could be up to 20 percent,. for ;
'4 oscillation allows us 3,000 percent.
5 And this is where I say that-bifurcations and --
6 behavior is important because this curve really turns down..
7 I mean'it makes it a misinterpretation because it's not that 8 the curve goes down. . It's that the peak oscillation speeds 9 up. Whenever you get into the -- behavior, you have much 10 larger peaks with the same amount of increasing in power.
11 So that I believe this answers the question where 12 there is an average power increase or not. -- system 13 effects. And we all agree that system effects are something 14 to be taken into account. System effects are things 1that 15 affect the power which changes reactivity. Changes in water 16 level, fuel water temperature, and things of that sort. .The 17 operator will pull the control rod.
18 This is over and beyond the system effects.
19 And roughly it comes out to be in the linear part, 20 below 500 percent oscillation amplitude. It comes out to be 21 1 1/2 percent. The average power increase is 1 1/2 percent 22 the increase of the peak oscillation power.
23 So I have one or two minutes. I must say that I 24 have grown very fast through it but I think I have shown to 25 you that this average reactivity decrease that is observed
- - Heritage Reporting Corporation (202) 628-4888
___m. _ _ _ . . _ - _ _ _ _ _ _ - _ - _ _ _
251
" to the. power increaselis one of the stabilizers that are in
-l l'
~
2 -oscillation. That's-the one that bounds oscillations. And-
'3 this decrease is governed by. general correlations. The one.
4 I-just'gave you. And.it's approximately .3~ cents per:
5 percentage. That's what it comes out to.
6 The average power increase must exist if you want-J
- 7. oscillations to remain bounded. It depends on the particular 8 power feedback co-efficient.- Okay. So you just get the 9 average density -- feedback that'you need and you will --
10 power co-efficient to get your average power increase.
11 And typically it's on the order of 1 1/2. percent 12' of the oscillation amplitude.
'13 Now, very, rapidly--I'm already one minute over--l'
~
t 14 have here'some results of ATWS. And we were saying that
=W' 15 conventional wisdom says that using the water level should 30 be'destabilizing because you are using the flow and you'are 17- increasing the voids and everything.
18 But that's not the case, okay. I have more with
'19 Lapur. Our reduction in water level. And in Lapur you-20 don't have a= water level. You have a power and a flow and a
- 21 suppression loop constant'. But the way to haveLmore 22 reduction in water level is a reduction in power and a p
23 reduction in flow. The flow more or less occurs--just l
24 . keeping the power to flow very constant.
25- And what we obtained is a tremendous disparity of Heritage Reporting Corporation
_ (202) 628-4888
l n
/ ir 252 1 what happens to the ratio depending on what the actual power 2 is. For instance, if we have an extremely -- power shape, 3- like the one that was in LaSalle, reducing the water level 4 reduces the - ,'that is it stabilizes the reactor. And 5' that's what they generally observed when the reduced the 6 water level in the Phase IA calculations. So it is very 7 consistent.
B But they had had a sine power shape, the --
9 wouldn't have changed at all practically. If they had had a 10 uniform that's completely constant, the -- would have gone 11 down, but not as much. But if they had had what I call the 12 optimum which is a double humped power shape which is the 13 one that gives you the worst stability conditions that you 14 can imagine and it looks more or less like this.
15 I am introducing you right now to the worst 16 ' accident power shape you can have according to Lapur. I 17 have run trillions of analyses with Lapur and that's one of
- 18. the advantages of having a --- frequency to main core.
19 You car run it on the PC and get results like this. And if 20 you plug into any code that has a 2 peaking factor in the 21- first node,.it goes down to zero in the middle, and then a 22 3.2 in the 10th node, you will get -- that will astonish 23 you. I mean this thing will bifurcate--this thing will blow 24 up the reactor up to the earth.
25 So if we use this one or for that matter double Heritage Reporting Corporation E m) (202) 628-4888
t 253
.7 s, 1 humped power shape i
& )
2 MR. LEE: It shouldn't not have surprised you 3 whatsoever because you are indeed exciting harmonics, which 4 is indeed a dominant mode of stability associated with the 5 whole density of oscillation. So this is almost oovious.
6 MR. MARCH-LEUBA: Everything is obvious once you 7 ask. Yes. It was common thought that the more -- you have 8 a power shape. That's a measure of wisdom. The worst 9 stability. That's another case. I can show you another 10 figure and unfortunately we should have had more time for 11 this. I show that you can be too much -- and then the 12 stability reverses. I mean by going even lower, you have 13 . lower --
14 I just set up a loop in Lapur and I calculated--I 15 changed this 12 values in the input until I maximized the --
16 and that's what I did.
17 MR. LIPINSKI: Does Lapur solves linear equations 18 or is it capable of doing--
19 MR. MARCH-LEUBA: Linear.
20 MR. LIPINSKI: Then how do you account for 21 distribution in power shape in a linear system?
22 MR. MARCH-LEUBA: Because it's an input. It's not 23 a self-starting oscillation. I tell you what, the model 24 from the -- mode is a point kinetics mode.
25 MR. LIPINSKI: Yes. But I have to go from the y- Heritage Reporting Corporation (202) 628-4888
o.
, -I 254
/-s.
( ); 1 bottom of the core to the top of the core and I have to have E 2 a power distribution that's going to feed us a. function of 3 nodes into er ,oint
> along that pass.
4 (ARCH-LEUBA: Yes, that's' correct. And that's 5 how you- cap tra thermal hydraulics with a point 6- kinetics. Once you havc another power for the core,. you 7' multiply times the fundamental shape to feed the power to 8 each node in the thermal hydraulics. i 9 MR. LIPINSKI: But your linear system is connected 10 .by nodes.
11 MR. MARCH-LEUBA: That's correct.
12 MR. LIPINSKI: Okay.
13 MR. MARCH-LEUBA: The thermal hydraulics is a rs 14 nodal approach.
(>)
15 MR. LIPINSKI: Okay. It's not totally all points 16 coupled together. It's nodal distributed along the axis.
17 MR. MARCH-LEUBA: That's correct.
18 MR. LIPINSKI: Okay.
19 MR. MARCH-LEUBA: The neutronics is point kinetic.
20 The thermal hydraulics is nodal.
21 So if you want to get a scare, you just plug this 22 power shape and you see what happens. So if you plug this 23 power shape to the ATWS then what happens is that the water 24 level increases the -- very dramatically.
25 And there's a footnote here that to get this --
-( Beritage Reporting Corporation (202) 628-4688
255 1 with original values I had to divide by the reactor co-2 efficient by a factor of 8. Not by a factor of 1.2 By a l
3 factor of 8. It's because of the ratio of this condition 4 with this power shape is something like 3 or 4. It's not 5 something -- It's outrageous. Well, I guess that I've run 6 out of time. Is there any questions I could answer?
7 DR. KERR: Your conclusion then is that one can't 8 be sure whether you've got unbounded oscillations or not.
9 MR. MARCH-LEUBA: Okay. My conclusion is you will 10 have bounded oscillations always. Because, and maybe I 11 wasn't clear, what happens here--well, this was a point 12 kinetics model with--
13 DR. KERR: I shouldn't have said unbounded. But I 14 mean you can't predict what the bound is.
15 MR. MARCH-LEUBA: I cannot predict what the 16 maximum amplitude will be. Because it depends on how 17 perverse the reasonable conditiens are.
18 DR. KERR: Yeah. Okay.
19 MR. MARCH-LEUBA: I have seen 3,000 percent 20 oscillations in my little model with unreasonable 21 conditions. Very unreasonable but they are not un-physical.
22 MR. SCEROCK: Does what you've described relate to 23 the asymmetric oscillations?
24 MR. MARCH-LEUBA: Not at all because this would 25 apply completely to the asymmetric oscillations the same way
(" Heritage Reporting Corporation (202) 628-4888
-256 1 that it applies to the fundamental mode.
~
2 MR. SCHROCK: Do you have a concern about.the fact 3 that--
4 MR. MARCH-LEUBA: Unfortunately you have given me 5 eight hours for presenting the asymmetric oscillations.
6 Yes.- I have a tremendous concern with asymmetric 7 oscillations because as all calculations show they are the 8 ones that can cause a problem. And it is because the 9 asymmetric oscillation is not a neutronic-induced 10 oscillation.- It is a flowing-use oscillation which happens 11 to be feedback by a little bit of neutronics. And you have 12 violent flow oscillations between--what you have is flow of 13 distribution from the right channel to the left channel or 14 half--and if you'have any experience with power channels, O
A/ 15 you know that if you took two power channels, they are much 16 more stable than a single one.
17 What happens is that neutronic feedback is fed 18 back now through a -- neutronic mode which has less. gain 19 than the critical, obviously. There's a mode oscillation 20 that is the inface which is -- neutronic and a modal 21 oscillation that is out of phase which is moda11y flow and 22 they are both fighting. In LaSalle they are in Phase I.
23 So, yes, the core-wide is not only harder to p 24 detect because of the fact that it is a flow-induced 25 oscillation. You have, according to some calculations I've Heritage Reporting Corporation O (202) 628-4888
1 I
257 3 1 seen, a reverse flow at the end of the channel. You have a w/
2 point of stagnation somewhere in the channel. And you have 1 1
1 3 the-- much more amplitude oscillation. j 4 MR. CATTON: I still don't think I understood your 5 answer to my question. I asked whether or not you had 6 bifurcations to spatial, different spatial instability. Do 7 you ever go from symmetric to anti-symmetric or anti-8 symmetric to symmetric?
9 MR. MARCH-LEUBA: No.
10 MR. CATTON: No. Once the instability starts 11 that's it? That's the kind you are going to have. If it's f
j 12 symmetric it stays symmetric. Bifurcations are only in 13 time.
14 MR. MARCH-LEUBA: That's correct.
15 That is--
16 MR. CATTON: Why is that? Most physical problems 17 usually have bifurcations both ways. '
18 MR. MARCH-LEUBA: Yes. But you are thinking of a
- 19. real bifurcation. This is a mathematical bifurcation. You 20 are thinking of bifurcation in which you have two possible 21 --
points.
22 MR. CATTON: No. I am thinking of once I have 23 turbulent flow and I push it a little harder, it suddenly 24 develops a different length scale and time scale. I push it 25 a little harder, it does it again. Eventually it becomes t
Heritage Reporting Corporation O (202) 628-4888 U____-__________-_____
J 258 1 chaotic.
- 2. MR. MARCH-LEUBA: Yes.
3 MR. CATTON: But most instability mechanisms sort 4L of walk through these things. Some go two dimensional 5 steady. Three dimensional. steady. Periodic, aperiodic, 6 chaotic. You see this everywhere. Yet here I am'seeing a 7 fixed spatial and it goes from one sort of time: behavior to 8 another.
9 MR. MARCH-LEUBA: Uh-huh.
10 MR. CATTON: You see that also. But usually the 11 length scale or the cross sectional structure changes when l 12 these things happen.
13- Is this because of the way you look at the 14 problem?
15 MR. LIPINSKI: Yes. He's only doing a single 16 channel. He's only got one Z axis.
17 MR. CATTON: Well, okay.
1 18 MR. LIPINSKI: He-doesn't have coupled cores.
19 MR. CATTON: I understood that because he said--
20 MR. LIPINSKI: If he had coupled cores, then he'd 21 see another phenomena.
j 22 MR. CATTON: But if he looked at the asymmetric or 23 the symmetric with up on the center and down on the edges, 24 within his problem could bifurcate.
25 MR. LIPINSKI- That's right. But in his solution, Heritage Reporting Corporation O (202) 628-4888 L -_ _- -
259 l/ - 11 'he can't generate that because that information isn't 2 present.
3 MR. CATTON: Well, does anybody do relative 4 stability? To find out which one--
5 MR. MARCH-LEUBA: Well, it's not going to answer 6 ycur question directly but maybe it will clarify a'little 7 bit what happens.
8 By understanding how the outer phase mode behaves, 9 why we have an outer phase type of instability, I have 10 modified the Lapur codes that not only calculates the in 11 phase mode of oscillation, but also out of phase. <
12 MR. CATTON: When you say "out of phase," you mean 13 the neutronics are out of phase with the hydraulics.
14 MR. MARCH-LEUBA: No. Neutronics are always seen-15 -- with hydraulics. The right channel is out of phase with 16 in channel.
17 So now the Lapur mode code can, with some 18 approximation I must add, calculate a stability not only 19 from the -- mode but of the -- mode that is oscillation.
20 And I have calculated the boundary of a stability as a power 21 and flow. And this is the boundary of the core-wide which 22 is also called the in phase instability. And it is this-23 one. And I have calculated the boundary of the out of phase 24 instability which is this one. And don't take this--this j
25 was a partial type of approach. Everything constant. I Heritage Reporting Corporation O (202) 628-4888
260
() 1 only changed the power and flow. So that's a disclaimer.
2 MR. CATTON: So that's the type of bifurcation I 3 vould expect.
4 MR.-MARCH-LEUBA: What happens here is if you 5 increase power in this region, you will become unstable --
6 of the out of phase mode or you would be unstable in the i
7 core-wide mode.
8 MR. CATTON: Okay.
9 MR. MARCH-LEUBA: If you come this way, you will 10 be core-wide unstable but out of phase stable. If you are 11 in this region, you are both unstable. I mean you have a 12 mess. And if you are in this point, you have bifurcation of 13 the type you are talking about.
r 14 MR. CATTON: Okay, I understand.
k_'s 15 MR. MARCH-LEUBA: Now, don't take this out of 16 context. This again was a partial -- and we have now more 17 understanding of what -- was a core wide and all this kind 18 of things.
19 But basically maybe that gives you an idea.
20 There are two modes. And depending where you are, 21 one is stable and one isn't 22 MR. CATTON: Well, but now you have to look at 23 relative stability criterion and ask which one is going to 24 grow. That's a different kind of problem. If I pick one of 25 those as a parameter I'll track the core wide and then if
() Heritage Reporting Corporation (202) 628-4888
261 r~, 1 I'm increasing flow, I'll bifurcate to the other. And*
U. 2 that's what I would expect. And then go to the region eff 3 the left, you'll have to decide which one you are going to 4 see. That's the relative stability criterion. One may win.
5 MR. MARCH-LEUBA: I think you see both.
6 MR. CATION: Maybe not.
7 MR. MARCH-LEUBA: That is a very interesting area.
8 MR. CATTON: It's an area that in -- niechanics is 9 well studied.
10 MR. MARCH-LEUBA: Yes.
11 MR. CATTON: And they still don't know the answer.
12 It's a much more complicated problem so if you told me you 13 knew, I would be a little suspicious.
14 MR. MARCH-LEUBA: They knew it 300 years ago. I 15 know this six months ago. So, yes, we don't have an answer 16 for that.
17 MR. CATTON: It's a very interesting area of 18 study.
19 MR. MARCH-LEUBA: Absolutely. And it's relevant, 20 it's relevant. You can see that we need to solve this 21 problem 22 MR. CATTON: That makes it even nicer, doesn't it?
23 An interesting problem that's relevant.
24 MR. MARD: Okay. Any other questions?
25 Thank you very much.
Heritage Reporting Corporation O (202) 628-4888 i
262
.,s 1 MR. WARD: Next on the agenda, we have David b 2 Bessette.
3 MR. BESSETTE: I think it would be much 4 interesting to spend the rest of the time listening to Jose 5 than to some bureaucrat from Rockville, but I'll just 6 present this anyway.
7 This is the research--to give you an idea of the 8 research program on the BWR stability which Jose has a major 9 part of and have to give credit to NRR for discovering Jose 10 and utilizing him and, I think, you know, if you are to hear 11 him you'll have to come to our stability meetings and his 12 presentations are always interesting.
13 Our objectives are to determine--first of all 14 determine the effective oscillations of core outage power; k 15 determine the sensitivity oscillations to key parameters, 16 such as the power shape, inlet sub cooling, power to flow 17 ratio. Develop capability to predict the mode of 18 oscillation, you know, uniform versus non uniform and 19 basically to develop a understanding to assist review and 20 possible audit of industry submittals.
21 And the method is first of all to understand the 22 phenomena, to validate the NRC codes and to apply the NRC 23 codes to evaluate stability and particularly the ATWS 24 situation .
25 We formed a technical program group early this s Heritage Reporting Corporation (202) 628-4888
( 'l l-1, I
263
(~T 1 year. The membership is Jose March-Leuba from Oak Ridge,
\_/
2 Gary Wilson from EG&G, Wolfgang Woulff from Brookhaven and 3 the rest is NRC, Stan Papick from NRR and Larry Phillips and 4 Howard Rushings and so on. Novak Zuber is participating.
5 Harold Scott is the Program Manager.
6 Se far we have met twice and will probably meet 7 again the early part of July and, as was the case with the 8 CSAU work, we find this working arrangement really helpful.
9 MR. LIPINSKI: Your heading says, "To carry out 10 stability research," but what we heard this morning, I don't 11 recall that stability research was 'a part of your program, 12 was it?
13 MR. CATTON: It was buried in one of the items.
14 MR. LIPINSKI: Okay.
I
'u 15 MR. BESSETTE: Ne mentioned the stability research 16 this morning.
17 MR. LIPINSKI: But it wasn't as a line item on any 18 of your vu-graphs, was it?
19 MR. BESSETTE: It did show up--
20 MR. LIPINSKI: Okay.
21 MR. BESSETTE: I can't remember exactly what form, 22 but I certainly know it was there.
23 MR. LIPINSKI: Okay.
24 MR. WARD: Well he actually had it on the reactor 25 applications.
Heritage Reporting Corporation (202) ;28-4888
264
(( ) 1 MR. BESSETTE: You remember there was some 2 discussion of chifting money from some other thing into--
3 MR. CATTON: If you needed it.
4 MR. LIPINSKI: Okay.
5 MR. CATTON: It was the BWR stability reactivity 6 accident.
7 MR. WARD: Okay, thank you. Continue.
8 MR. BESSETTE: I shouldn't go into the Oak Ridge 9 core analysis since Jose covered that. This is--Now the 10 membership in this group, you don't have anyone from General 11 Electric or the Owners Group. Is that not feasible or --
12 MR. BESSETTE: We have considered this and also 13 EPRI--
14 MR. WARD: Yes.
(A_)
15 MR. BESSETTE: So far, we haven't had them to the 16 meetings. We may in the future, but--
17 MR. WARD: They might just know something about 18 it.
19 MR. BESSETTE: Yes. But on the hand, there's this 20 concern, well maybe we're better off keeping things separate 21 since this owners group analysis is under way, in response 22 to the bulletin and--
23 MR. CATTON: You have Gerry Lellouche from SLI and 24 that's Saul Levy's company and he is certainly close to CE, 25 indirectly, He has some influence.
l (') Heritage Reporting Corporation (202) 628-4888 l
i 265 (m) 1 MR. BESSETTE: I guess I forgot to mention Gerry.
2 He's almost like a one man committee, in himself. We may 3 have Di Fianos in the future also. We' re trying to get him 4 involved.
5 One thing Jose hasn't mentioned is that I believe 6 he is intending to calculate the Oskanshamn data.
7 MR. LEE: What is Oskanshamn data?
8 MR. BESSETTE: It's a Swedish plant. I think the 9 event was in January of-this year, is that right and--
10 MR. MARCH-LEUBA: It is a stability test from 11 Sweden that is very well recorded. The Swedish are going to 12 supply the date for our bench marker.
13 MR. CATTON: Was this a test or was it an event?
14 MR. MARCH-LEUBA: I think it was both.
15 MR. PHILLIPS: It was a test. The always test a i
16 reactor during start up, but it did occur in a normal 17 operating range. The got out of phase oscillation.
18 MR. CATTON: When you get out of phase 19 oscillations, how good is point kinetics?
20 MR. LEE: It is a point area for the second mode.
21 It is a mode of nutronic analysis. Point kinetics usually l
22 work for the fundamental critical mode. You can also 23 formulate equations for the second mode of nutronics, much 24 the way you do the analysis for action- you can formulate 25 equations on- you can use the same code, changing a very 1
()
Heritage Reporting Corporation (202) 628-4888 l
l
l 266
([ ) 1 very few--
2 MR. CATTON: So you do have the neutronics L 3 changing in different parts of the core?
4 MR. MARCH-LEUBA: Yes. Just because I know 5 neutronics, I have the amplitude of a' mole as my neutronics 6 and as long as I have the shape of the mold appropriate, 7 appropriately it changes, yes.
8 MR. CATTON: Okay.
9 MR. BESSETTE: Looking at Brookhaven, using their 10 BWR analyzer code HIPA. They have seen results as far as a 11 core power change, they have seen results similar to LAPUR 12 and I think when they separated out the system affects, they 13 find a similar power increase that is one and a half percent rs 14 per 100 percent oscillation range.
U 15 MR. LIPINSKI: This is the BWR analyzer?
16 MR. BESSETTE: Yes. And one of the things we're 17 planning is a LAPUR HIPA benchmark exercise.
18 MR. LIPINSKI: Is the LAPUR Code in the DOE Code 19 Center?
20 MR. MARCH-LEUBA: It will be soon.
21 MR. BESSETTE: Concerning the RAMUNA analysis, 22 there hasn't been anything done since the fall. We have been 23 working on code modifications, changing the drip flux in the 24 core and allowing for reverse flow and that work should be 25 complete very shortly and then we would validate the code
()~ Heritage Reporting Corporation (202) 628-4888
267 l against FRIGG data. That is FRIGG power to flow data. That
}
2 analysis is in progress right now with TRAC and should be 3 complete, I think, in June or July and then the last part of 4 the fiscal year we would calculate the LaSalle event with 5 TRAC.
6 MR. CATTON: Do you expect TRAC to do a good job 7 for you?
8 MR. BESSETTE: Well, we'll see.
9 MR. CATTON: The reason is is if you're doing a 10 linear stability, there's a problem where it's linear 11 stability. TRAC is so highly damped to keep it operational 12 hydrodynamically. Won't it wipe all these instabilities out 13 before they start?
7._
14 MR. MARCH-LEUBA: The answer to that is General J 15 Electric has already run TRAC and it didn't dump it out, so 16 we have hopes it will work. RAMONA has certainly worked.
17 We have never run TRAC yet, but that is why we are trying to 18 benchmark it against FRIGG with millions of points that it 19 can be benchmarked against.
20 And see that is all what GE got, we are hopeful it 21 will work, but again--
22 MR. CATTON: You can always go in and put some 23 sort of purtobation in it.
24 MR. MARCH-LEUBA: It will grow, no problem. If 25 you put the right purtobation, you will reach the limit
{}
~
Heritage Reporting Corporation (202) 628-4888
268
() 1 cycle faster, so it will cost you less money. In my simple 1
2 model,.I always put one cent purtobation and let it grow. !
j 3 It grows fast.
4 MR. BESSETTE: That's why we do FRIGG before we do 5 LaSalle and a because we want to see those results first.
6 MR. LEE: Ivan, are you talking about the fact 7 that TRAC would prefer, rather than place it, time wise 8 implicit treatment so that--
9 MR. CATTON: in order that TRAC run, the way the 10 put their numerical alterisms together, they're highly 11 damping, there's decay. Noise gets wiped out. So, if you 12 have a circumstance where you want to look at an 13 inst ability, I would always be concerned that the fs 14 instability would get wiped out and you would come to a R) 15 wrong solution.
16 MR. LEE: Do you have to use fine time stats?
17 MR. CATTON: Yes.
18 MR. LEE: Then the noise would not be wiped out?
19 MR. CATTON: Only if you changed the damping.
20 It's inherent in the code. If it's really a good 21 instability, it will grow in spite of it and that may well 22 be the case here. You may have very sharp boundaries and 23 very rapid growth, then you don't have a problem.
24 MR. LEE: Would you comment a little bit on the 25 preliminary RAMONA study of LaSalle event reported? I have
() Heritage Reporting Corporation (202) 628-4888 l
1 l
. _ --- .-= -
269 i
( ). 1 a' copy of the preliminary report that I receivsd'from Dr.
2 Shotkin? g 1
3 MR. BESSETTE: Comment on anything in particular 4 or--
5 MR. LEE: In particular, ifEI~ remember, correctly, 6- the report' indicates during the LaSalle event, perhaps it I l
7 could have undergone a state where minimum critical power 8 ratio was one or less, which means that the boiling 9 transition might have taken place.
10 MR.-SCOTT: I don't think it said that. It.said-11 if the oscillations got up to 200 percent of rated power,-
12 which they did not, they only went up to a 118 percent of 13 rated power.
14 MR. LEE: I-see. So for the actual LaSalle, what 15 was the predicted minimum value crticial power ratio? I 16 didn't read the report carefully.
17 MR. MARCH-LEUBA: 'Let me--from what I understand, 18 RAMONA has_not been completed for LaSalle. It was done for 19 Maine Yankee with LaSalle conditions, wasn't it?
20 MR. LEE: Whatever conclusions that were reached 21 were not--Whatever conclusions were made--is that correct?
22 MR. PHILLIPS: I don't recall that they did an 23 MCPR calculation, but we wouldn't have expected it to reach 24 boiling transition under those conditions.
)_ 25 MR. LEE: If I may read from Page 6 of the report, Heritage Reporting Corporation (202) 628-4888
L - ;. ._
270 1 :the last sentence of the third-paragraph says, "As.a result 1 ])
2 of the increasing local power and reviewed flow-during_these
' oscillations, MCPR may have approach one."
4- MR. PHILLIPS: Yes, that is for a case where you 5 -don't get a CRAM. If you didn't'get a reactor trip and_the 6 118 percent power-and oscillations continued to grow, that 7 was a case they ran.
~
8 Then you get into flow reversals situation and the 9 problem is that if he RAMONA breaks down, the code breaks
~
10 down at that point, the thermal hydraulic mode 1 and part of 11 the objective research program going on on a longer range is 12 to install a new thermal hydraulic model in RAMONA that will 13 allow an extension of those calculations.
14 MR. LEE: I should not trust the results of the
.O' 15 preliminary report--
lei MR. PHILLIPS: No, except -I think they' re in 17 general agreement with the TRAC results that you get.into l
18 conditions--you can't get in conditions where~you reach 19 boiling transition. That's the same conclusion that General 20 Electric reached in their Phase I study, a little different L
21 calculations, but it was with large oscillations.
22 MR. LEE: Thank you.
23 MR. BESSETTE: And lastly, we, we haven't yet used 24 NUFRIGG. We have had some review of a NUFRIGG MPW by Oak 25 Ridge and it was not--it didn't encourage us to go ahead and Heritage Reporting Corporation (202) 628-4888 L-__---_-______.-__ _ - _ - - _ _ _ _ _ - - - _ _ .- - - - _ - . _ - - - - - - _ - . - _ ___________________ _ _ _ _ _ _ _ _ _
i i
271: 1 L.
1- start.using NUFRIGG and we're having further review done by 2 . Sol Levy, Incorporated. That we will hear about.the'next 3' time this Technical Program group; meets.
4 MR. SCHROCK: Which NUFRIGG are you referring to?
6 MR. MARCH-LEUBA: NUFRIGG NPW'is a Westinghouse J
7 version of EMPRI and NUFRIGG NP which is now going to be ADB' l
8 A.TMS version of NUFRIGG. There might be a NUFRIGG MPW(a) or 9 something like that.
10 MR. WARD: Okay. Is that it?
11 MR. BESSETTE: Yes, that's it.
12 MR. WARD: Thank you. The agenda shows Mr. Scott 13 is going to say a few words.
14 MR.' SCOTT: I wanted to mention that one of the
.O. ~
15 aspects of this program is that we have really strong 16 interest from NRR and AUB compared to a lot of the other 17 research. programs we have had over the years. Some of them 18 we even had trouble getting the NRR people to come to our 19 Program Review Meetings, but they are very interested, we 20 have user needs from both AEOD and NRR and as David has 21 mentioned, we have these TPG, Tec'anical Progarm Group 22 Meetings, Program Review Group Meetings, in which our ,
23 purpose is to review the work going on, coordinate the work 24 between the various codes and laboratories and provide 25 technical assistance.
() Heritage Reporting Corporation (202) 628-4888
l 272 (p l' One of our problems in inviting people that are 2 not contractors and not ERC staff is they come under the 3 Federal Advisory Committee Act, which. requires us to 4 'announce the meetings, which we~can'do that as we need to,.
5 that was one of the reasons that we decided, at least at the
)
6 'beginning meetings to keep it amongst NRC and our own 7 contractors, that way they're operational meetings and they_
8 don't meet the FACA requirements.
9 As David said, we have had two meetings and he.
10 mentioned the people that have come. I just wanted to 11 quickly say that Oak Ridge and Brookhaven are working quite 12 well together. We think this is going to be a sample of a 13 research program where the various laboratories sort of 14 leave their pettiness aside and work with each other, and in
'15 particular the group membership helps to do this and as 16 March-Leuba has teld-you, he has sort of elucidated some of 17 these ideas about why oscillations affect core power.
18 As was mentioned here, RAMONA is being fixed to 19 change the full reversal at the core bottom. The code 20 assumes that water would always be going upward and the 21 bottom node of the core and that may not always be the case 22 so we need to make some changes in the code to do that and 23 we'll be implementing the drip flux correlation in RAMONA, 24 if it is the same that is in the plant analyzer.
25 It may also turn out that drip flux will do about O a rie 9- earti=9 coreor tie-(202) 628-4888
273 1 as well es two fluid formulations that are in TRAC, but we
("')
v 2 won't know until we have actually made some comparisons.
3 As David said, the TRAC calculations in La Salle 4 are coming up next.
5 At these meetings we develop various action items 6 for things to be worked on. I just wanted to throw up a 7 couple of these and mention that, in this case, like where 8 we are working with Brookhaven and Oak Ridge, here was a 9 case where the nodalization, this first one here is 10 significant. We have seen just today that actual power 11 profile is very important. We have also seen from some 12 Swedish calculations that the number of xy nodes in the 13 core, exactly how you space those out is important.
14 In other words, if you're going to do left -
15 right, so called nonuniform asymmetric, the question is, 16 would you--how fine do you need to notalize in order to get 17 good answers.
18 Another thing I think I wanted to mention that had 19 come up in your letter to us last year was, how are we going 20 to keep a cadre of people together and could we provide them 21 with problems that will provide interest and vitality.
22 This problem seems to be one that everybcdy 23 believes is really stretching our ability to understand and 24 to direct him to something that is not really mundane.
25 I also might put up here, just quickly, the same Heritage Reporting Corporation O- (202) 628-4888
274
{} 1- questions that came up, that we just talked about before, 2 only to mention that research and NRR work very carefully to 3 define these questions. We, in Research, said, you know, 4 these are the kinds of questions or words we said to NRR, 5 why don't you see if you can't get the owner's group to also 6 look at them, so we didn't have to do all the work 7 ourselves.
8 Number 5, of course, is an important one. We'll be 9 doing this as I mentioned with the out of piles single 10 bundle break test and by comparing with the LaSalle event.
11 This is probably something that NRC is not going 12 to work too much on. I think that is why we wanted the 13 owners group to be sure and tell us enough about it so the 14 NRR people would be happy with the answers. Thank you,
~
15 I think I will say too that we have found that 16 Jose March-Leuba at Oak Ridge has been able to, with simple 17 codes and using his head, explain things--there were a lot 18 of times we were so busy working with our great big heavy 19 duty codes that we don't even have time to hardly think 20 about.
21 MR. WARD: Any questions for Mr. Scott? Virgil.
22 MR. SCHROCK: In the discussion on RAMONA, you 23 mentioned RAMONA needed a fix to allow it to expel liquids--
24 is that something that you are far enough to know, is there 25 a concern here that when you go into parallel channel
^
k]
/ Heritage Reporting Corporation (202) 628-4888
1
-275 (J 1~ instabilities that you might get such a strong effect on one 2 bundle that you try to iry and never accomplished it?
3 MR. SCOTT: I' don't think that was very likely. I 4 don't think the power is going to be that peaked in one 5 channel. The code, of course,. can handle flow going up and-6 flow going down at any node in the middle of the core. The 7 question is, at this point here, if some time step, let me 8 push it up higher, you drop down to negative flow, at the 9 bottom of the core--what we're saying is that code always 10 assumed that the liquid was coming in. They proceeded to 11 change the logic in the code to, if the code calculates that 12 the fluid is going down at this node in the core, it allows 13 that to happen.
gm 14 MR. SCHROCK: The only way that can happen is if 15 the individual channel pressurizes and picks the lower 16 pressure.
17 MR. PHILLIPS: That's right.
18 MR. SCHROCK: Sort of in response to the local' 19 code situation. Once it boils liquid out both ends, getting 20 liquid to go back into--
21 MR. PHILLIPS: Cnce the flow rate is dropped, it 22 can figure the flow rate quite low, the voids will increase 23 the power dtop down in that channel and then liquid can go 24 back in.
25 '.fou say, what would cause the liquid not to go t
/~' Heritage Reporting Corporation (202) 628-4888 c
r 276-
[^ 1 back in?'
2'. MR. CATTON: You have to rewet it.
3- MR. MARCH-LEUBA: .Yes, but you don't have any 4' power as long as you have voids.
5 MR. CATTON: It's over heated--
6 MR. MARCH-LEUBA: That would--
7 MR. CATTON: If it doesn't have the reflux 8 capability and all that sort of nonsense--
9 MR.'SCHROCK: You've generated enough pressure to 10 completely turn the core around--there'sivery little liquid 11 -left in there, but enough left to continue to sustain high 12 pressure and prevent flow in.
13 MR. LEE: You're just postulating what would be 14 the failure mode of the fuel if the situation were large 15 enough. You' re right, that's the only way it's going to 16 fail.
17 MR. SCOTT: The code could calculate whether 18 boiling transition occurs, if we have the right kind of 19 correlation. What we said before was since the code doesn't 20 do very well when that occurs, when the flow rate goes 21 backwards at the inlet, the code sort of hangs up and we 22 haven't ever really been able to calculate what the minimum I'
23 critical power ratio is.
24 After we make this fix, then we can run some 25 calculations with large amplitudes. This only occurs with Heritage Reporting Corporation I
i (202) 628-4888 t _ _ _ _ _ _ _ _ . _ - _ _ _ _ _ _ _ _
277 1 like 4-or 500 percent--3 or 400 percent amplitude.
')
2 MR. SCHROCK: How many channels will this version 3 accommodate?
4 MR. SCOTT: You can have dozens of--
5 MR. SCHROCK: In your TRAC calculations, you used 6 what channels?
7 MR. SCOTT: GE said they.used 11 grades. I think 8 they can have one or more channels grade.
9 MR. SChROCK: Well, it doesn't occur in 10 symmetrical fashion. It occurs asymmetrical.
11 MR. SCOTT: RAMONA allows you to place hydraulic 12 channels at any place in the core, so we can have, you know 13 that--
14 MR. SCHROCK: What 1 am looking for, Mr. Scott is:
15 Is there a concern here about the real reactors ever getting 16 into this kind of a situation.
17 There is a lot of emphasis on the study of--it 18 seems implied there is a comparable amount that we haven't 19 heard on the other--
20 MR. SCOTT: Yes, there is a concern--
21 MR. SCHROCK: It seems to me there is the area in 22 which you ought to be most concerned about, is there a real 23 world problem out there for us. Are you ever going to melt 24 down a channel--
25 MR. SCOTT: That's what we believe we can
() Heritage Reporting Corporation (202) 628-4888
i 278
() 1 positively fix by predicting the regions where we might get q
)
2 into boiling transition and putting on SCRAMS in those 3 regions. The type of thing they're looking at are long term 4 solutions and interim operating procedures and in the 5 meantime to stay away from those regions.
l 6 But yes, GE too has calculated that high power, l
7 low flow conditions, you can reach boiling transition and--
8 RAMONA can handle this. RAMONA allows you to 9 place--we are definitely going to make those calculations, 10 but it won't be until fiscal year ' 90. There is a cash flow 11 problem that we have to deal with, we can do everything at 12 once.
13 MR. SCHROCK: Well it seems to me you ought to be 14 looking at what is first priority.
15 MR. SCOTT: We could, rather than going ahead with 16 these kind of applications, but we felt as soon as we do an 17 application, somebody will say to us, just like we're saying 18 to GE, where is your assessment, so we're sort of like, 19 which do we do first, the assessment or the application and 20 we chose to do the ussessment.
21 That's what I'm saying, the application that we're !
22 just talking about wouldn't be done until '90 because we are 23 goir.g to be doing the assessment in the remainder of ' 89 24 after the code is fixed.
25 I'm sure if I came and showed you the results
() Heritage Reporting Corporation (202) 628-4888
279 '
[)
v 1 you're asking for, then you might ask me, what assessment do 2 you have.
3 MR. WARD: OKay, thank you, Mr. Scott.
4 This portion now will be for the committee members 5 and consultants now.
6 The Commission has asked us to comment on the 7 program. You know, we know it's not a wholly developed 8 program, but they would like us to comment on things as they 9 stand now.
10 So right now we're scheduled to have an hour and a j 11 half to devoted to this at he June full committee meeting on 12 Thursday. I 13 But first, I would like, before talking about 14 that, I would like to ask the consultants and committee 15 members here a couple of questions. Give me your i
16 impressions of what you heard. 1 J
17 First of all, is this an important safety issue, 18 first question? Is the owner's group program, as they have 19 described it appropriate in direction and in intensity? Do 20 they seem to be putting the appropriate level of resources i 1
21 on it and there is both the set of immediate actions, of l 22 course, in response to what NRR has asked, but especially 23 the long term program and then, finally, is the NRC, the 24 Research Program that we have heard going to add anything?
25 Is that appropriate and is it necessary?
l
() Beritage Reporting Corporation (202) 628-4888 1
l
- - - _ - _ _ - _ _ _ _ - _ _ _ _ _ _ _ _ _ _ - _ _ _ \
280~
() 1 Let me tell you, I'll just summarize what' Walt 2- Lipinski said. .I may be at risk of doing damage to his 1
3' opinion.
1' 4 He thinks the most important part of a solution'is.
5 -really what I guess I would call the pragmatic solution,that l 6 'is to develop the regional trips and it's the direction that 7 the owner't group seems to be going.
i 8 Walt seemed to think all the analysis is 9- interesting and maybe necessary to develop a better 10 background of understanding, but that the real, you know, if 11 we perceive that there is a safety problem, the real wr.y to 12 deal with that is more pragmatically because of the l 13 complexity of it.
14 Lurking in the background is kind of a concern.
15 that, gee it turns out these boiling water reactors are
~
16 somewhat more complex than wee thought they were or hnve 17 been led to believe for the last generation and perhaps
- 18. that's a little troublesome. I don't -quite know what we 19 will do with that. I guess the last comment, I don't want 20 to blame entirely on Walt. That's perhaps more--I put a 21 little spin on what he said, i
22 So let me go around and ask, if you got something L
23 you would like to say at this time, we would like to hear it 24 because we have to decide, among other things, where we want 25 to go with the full committee. John, how about starting Beritage Reporting Corporation (202) 628-4888 L___-_:____-__________ - _ _ _ . _ - _ _ _ _ _ _ _ _ - _ _ _ _ . ___ ___ __ ___ _ _ __ - _____________ - _______
281
() 1 with you?
2 MR. LEE: I believe it a safety issue because of 3 all the discussions we had and what I have worked on myself.
4 It is very difficult to predict exactly the onset of 5 stability as well as the magnitude of limi cycle in 6 operating boiling water reactors for the various reasons we 7 talked about. Instability is very sensitive to all kinds of 8 operating processes.
9 In my opinion, I would put less emphasis pm 10 defining even the instability rather I would rather prepare 11 for handing the instabilities from an operating pragmatic 12 point of view in a more long term manner as well as short 13 term stability.
14 One of the solutions that hasn't been talked about 15 at all is from a pragmatic point of view, short of SCRAMING 16 the reactor, maybe you could make the system react just 17 slightly sub critical by killing the transient oscillation 18 and you can recall that at some time, perhaps.
19 And at the same time, as I pursued with, I guess, 20 BW Owners Group representative a little bit, over the long 21 term, I would like to somehow make efforts to make the 22 system inerrantly more stable and less susceptible to these 23 oscillations. Those are my comments.
24 MR. WARD: Okay. Thank you, John. Charlie, have 25 you got anything you want to add?
l
() Heritage Reporting Corporation (202) 628-4888 i
l l
282
) 1 MR. WYLIE: Well, yeah, I tend to agree with what
[/
w 2 has been said and think it 's a safety issue and I believe 3 the owner's group approach is headed in the right direction 4 and not sure about what the last question you asked about, 5 the NRC research adding to this, except insight, and to make 6 them take measures to protect against the problem.
7 MR. WARD: Thanks Charlie. Ivan.
8 MR. CATTON: I think I would agree with other that 9 it is a safety issue, particularly the high amplitude 10 asymmetric modes with the possibility of a reverse flow. It 11 sounde to me like dry out and a channel melting.
12 I don't know if it's possible, but certainly, I 13 was kind of led to believe there is a chance.
14 The planning and stability values, it seems to me 15 the appropriate second thing one ought to do after you put 16 in a stability monitor, but finding the stability 17 boundaries, I think you're going to have to go to analysis.
18 You certainly don't want to test the fact.
19 One of the graphs that we wero shown showed that 20 the--I think it was the third peak was 2,000 percent.
21 Granted this was a terrible powerful thing he put into it, 22 but still, trying to map out what is going to happen where, 23 is a difficult task. It means hundreds of calculations, 24 different kinds of distributions to try end formulate this 25 map and you're never sure you're going to get it all.
() Heritage Reporting Corporation (202) 628-4888
283 1 I was a little disappointed in what the owners
[w )\
2 group had to say. I think the one calculation, with a 3 practical certainty, doesn't tell me very much. I was very 4 impressed with the combination of research and NRR are 5 doing, particularly this process where they think a little 6 bit instead of just hammering with large code.
7 Some combination is where you need to be and it-8 seems to me that they're close.
9 MR. WARD: Thanks. Milt.
10 MR. PLESSET: I also think it's a safety issue of 11 some importance. I think you've got to understand very 12 well, because you can never tell what an operator will do.
13 We have seen the effects of that in other examples.
14 I agree with Ivan about the owners group needing a
(-)
(/
15 little more encouragement to be more intelligent, or 16 whatever. I don't know what adjective you would recommend, 17 Ivan.
18 MR. CATTON: I would recommend aggressive.
19 MR. PLESSET: Aggressive, all right, I'll buy 20 that. I think what NRR and Research are doing is very good.
21 I think it should be encouraged . We need more of it. I 22 don't think we should discourage it because that's the way 23 we're really going to understand some of this stuff. I 24 think we ought to be able to handle it effectively.
25 MR. WARD: Thanks Milt. Virgil.
() Heritage Reporting Corporation (202) 628-4888
284
() 1 MR. SCHROCK: Well Dave, I'm in conflict of 2 interest as far as GE is concerned. I wanted to say that 3 before I said anything.
4 MR. WARD: Okay. We should have--I think as long 5 as you have let us know you're in conflict, then we would 6 like to hear your vi'tws?
7 1R. SCHROCK: Well, I really don't have much to 8 add to what has been said. I pretty much agree with the 9 other comments that have been made.
10 I tend to have the impression from understanding 11 all of the mechanisms involved here is pretty much left to 12 the NRC function. It's not going to come out of the owner's 13 group activity and that may be as it should be.
14 MR. WARD: Okay. Thank you, Virgil.
15 We can go off the record now.
16 (Whereupon, at 5:46 p.m., the subcommittee 17 was adjourned.)
18 19 20 l
21 22 23 24 25
() Heritage Reporting Corporation (202) 628-4888
1 1
i
.n CERTIFICATE
/
U 1 l 2
3 This is to certify that the actached proceedings before the 4 United States Nuclear Regulatory Commission in the matter 5 of: A c & S - % b c ';$/t vi r/M Pben omena/ Care c 6 , Name:
7 8 Docket Number:
9 Place: h/ #5 '/ '
10 Date: f/y3/ff 11 were held as herein appears,. and that this is the original 12 transcript thereof for the file of the United States Nuclear 13 Regulatory Commission taken stenographically by me and, 14 thereafter reduced to typewriting by me or under the d(m 15 direction of the court reporting company, and that the 16 transcript is a true and accurate record of the foregoing 17 proceedings.
18 /s/ unA4 regan 6/ /
19 (Signature typed) : Irwin J. Coffenberry 20 official Reporter 21 Heritage Reporting Corporation 22 23 24 l 25 I
O Beritage Reporting Corporation (202) 628-4888 i
1 1
O THERMAL-HYDRAULIC RESEARCH PLANNING AND DIRECTION FOR
! 1989 AND BEYOND 1
PRESENTATION TO THE ACRS THERMAL-HYDRAULIC PHENOMENA /
CORE PERFORMANCE SUBCOMMITTEE MAY 23,1989 Brian W. Sheron, Director .
Division of Systems Research Office of Nuclear Regulatory .Research O
l O 2
- Brief History -
l o NRC has been developing T/H systems codes for 15
+ years.
o Code development and verification went from LBLOCA to SBLOCA to transients to front end of severe accidents.
O o Codes have now reached an acceptable level of accuracy and maturity for current generation LWR's.
further development not likely to produce major changes in our understanding of performance or consequences; CSAU has quantified uncertainty for LBLOCA.
SBLOCA uncertainty assessment planned.
O
O 3 ;
Planning and Direction Objectives Overall Objectives Bring major thermal-hydraulic code development program to successful conclusion.
Maintain capability for thermal-hydraulic analysis at minimum level.
Apply developed codes to reactor issues.
O ;
I O
-O 4 Planning and Direction Objectives Specific Objectives
- 1. Maintain Capability within Agency for Thermal-Hydraulic Analysis of LWRs !
a) Onerating Reactor Events (e.g., Davis-Besse, LaSalle)-
b) Licensing Issues (e.g., license amendments) c) Generic Research (e.g., front-end of S.A.
sequences)
O
- 2. Maintain cadre of experts (contractor and in-house) to achieve (1) above
- 3. Maintain code development /research at minimum level necessary to:
a) ensure codes are acceptable for advanced LWR '
analyses; 3,
':4 b) ensure new information does not invalidate current understanding of code accuracy; and c) achieve (2) above.
O .
O 5
- 4. Establish and maintain low-cost experimental capability at Universities through construction and operation of scaled loops representing maior U.S.
reactor tvoes.*
- 5. Retain involvement in international T/H activities provided resource commitment is minimized and there is substantial benefit to NRC.
- 6. Expand applications research using codes to systematically assess reactor behavior.
- a. operating LWR's O
- b. advanced LWR's
- Pending verification of usefulness of University of Maryland loop.
+
O
= !
O
\
I THERMAL-HYDRAULIC RESEARCH PLANS and FUNDING FY89-94 O louis u. SsOTxis, CsiEF REACTOR and PLANT SYSTEMS BRANCH DIVISION OF SYSTEMS RESEARCH OFFICE OF NUCLEAR REGULATORY RESEARCH PE OR NCE BA TTEE May 23,1989
THERMAL-HYDRAULIC PLANNING ASSUMPTIONS O
i 1
- 1. COMPLETE MOST MAJOR RESEARCH PROGRAMS BY FY92.
- 2. DEVELOP AND MAINTAIN EXPERTISE TO MEET FUTURE AGENCY' NEEDS:
- BASE-LINE ACTIVITIES PRIORITY ISSUES
- SMALL-SCALE INTEGRAL TEST LOOPS
- 3. T-H RELATED ISSUES WILL CONTINUE TO ARISE WITH REGULARITY.
O + CURRENT ISSUES:
- BWR STABILITY
- NRC SIMULATOR FIDELITY
- INTERFACING SYSTEMS LOCA
- OTSG BEHAVIOR UNDER SECONDARY DEPRESSURIZATION
- 4. NEW INITIATIVES WILL CONTINUE TO BE EXPLORED.
- CURRENT INITIATIVES:
- ACCIDENT MANAGEMENT: EFFICACY OF FRONT-END (PREVENTIVE) STRATEGIES
- ADVANCED LWR (600 MW) SAFETY ISSUES
- INTERNATIONAL CODE MAINTENANCE CONSORTIA O
I O NEW BUDGET STRUCTURE ELEMENT OLD NEW
- 1. PLANT PERFORMANCE B&W TESTING B&W TESTING PWR LARGE LOCA TESTING PWR SMALL LOCA TESTING EXPERIMENTS AND OTHER EXPERIMENTAL ANALYSIS PROGRAMS MODELING MODELING
- 2. REACTOR APPLICATIONS CONTAINMENT / BOP OPERATING REACTORS TECHNICAL SUPPORT CENTER LWR SYSTEMS STUDIES
^^ ' ^
NPA/ SIMULATOR EVENTS O
__ - - 1
c LOl NEW BUDGET STRUCTURE (cont'd)
PROGRAMS IN EACH ACTIVITY PLANT PERFORMANCE B&W TESTING: MIST OTSG SBLOCA CSAU EXP'TS AND ANALYSIS: 2D/3D ROSA-IV BETHSY SMALL LOOPS O
MODELING: TRAC RELAP ICAP ICAP' REACTOR APPLICATIONS OPERATING REACTORS: T-H TECH-SUPPORT CENTER SIMULATOR FIDELITY LWR SYSTEMS STUDIES: ADVANCED LWR'S ANAL. FOR OPERATING REA T R EVENTS: BWR STABILITY O REACTIVITY ACCIDENTS
9 0 0 0 5 5 5 5 9 0 5 5 7 7 7 2 1 3 1 1 3 -
1 1 3 0 0 0 0 0 0 0 9 0 0 0 0 5 5 5 5 9 0 5 5 7 7 7 2 1 3 1 1 3 1 1 2 0 0 0 0 0 0 0 0 S 9 9
0 0 9
0 5
0 7
5 7
5 7
5 7
5 2
E 1
1 4 1 1 3 1 1 I
T R 1 0 0 0 0 0 0 0 0 I A 9 0 5 5 0 0 0 0 0 E 9 5 2 3 0 7 6 V 9 1 Y 1 4 1 1 2 4 2 1 I
T L 0 0 0 0 0 0 0 0 0 C A C
S 9
9 0
2 5
4 0
1 5
6 0
3 0
5 0
5 0
3 A 2 5 2 1 I
1 5 1 1 1 F
O F 9
8 0
0 0
7 0
3 0
0 0
3 5
2 5
9 0
1 9 5 8 9 7 3 0 2 O 1 7 2 1 2 2 1 1 G
G S N
I T
N S S
R E
I A R
I N O D D E C
S I
S O I
T C
U T
E S P T N N Y T A S ONE A L A U M G A N
C I
E R
S M
R V O E F R O
N I
T A
L P G E T
F R F S & G P N S SO R E N A I T Y I
S T E T S I
L A S Y C R
P 'T E O R L A T W P D T E R A R E
N & X O C P W N A B E M A O L A O L P
E R
D T
I O L C I
B G N
A I T
L I
S I
X A E V E A .
S U
)
G W M
T N
N I
T N E 0 E M D 0 6 M E N ( E G A
U s G A N F R N A W A M eY B L A M T N
T R N .
E O D D F ED )P I
C E S I
CT C
C T E CS A C
I T A ,
R R
I E L I
RO O F C O F C F
F A L G F .
A T E E RM N
I T
S A S S
E T WP E M T F D A T H
A / OC C I
R T S S /
F I
C G W WS P E O E E A P N N L S ORP O N
O E N (R O
N
_ A N R OM G S .
S A I
G E R R S T P N
EG R I
N O O S T I E VO E R
EP R M B O D E L D O
_ D P E S N
E T
RE UL C D O NM TE E
R A TA F C E O ^UL UC O T D N
I T
A A F S TM P SE A E A D V E SL
.L
- S N AV C I L G N TA O L O E
NM I
RO P N I
T M G N E S S
R F O PT M D M M E M
I S O E EN V T I
E B E C E I
NE S O T E L RS D ED O C F R A GD E N MO P O
G S AE E N SC S
O O R N A P L N A EWP G O I
T T O R
A MG L
P S E S N N
I E
L P RN OT I T S
A E E:
E P M
FS E
A L T E OT B O C '
F O
T L P N D M S N E
O C
M A 9 8
2 9
0 9
2 9
9 8
1 9
R Y F
Y F
Y F
Y F
Y F
Y F
G O G N
R I T R R P S E
VY WW T I P BS - P D - S -
T G 3 AH C C A P S S / ST A AL A I T D OE R RE C M O 2 R B T TR I s_.
1 2
3 4
5 6
l .
- ,e a
REQUEST FOR ACRS ACTION COMMENTS REQUESTED FOR INCLUSION IN '
COMMISSION PAPER N
O
" STATUS AND PLANS FOR T/H RESEARCH {
I CONDUCTED BY NRC/RES" l
1 i
)
l 1
O l
b ___-- - - -
RESSE &
O FUTURE PROGRAM PLANS - BASE PROGRAM
- INTERNATIONAL COOPERATION PURPOSE: ACCESS TO FOREIGN FACILITIES AND EXPERTISE PLAN: MULTILATERAL PROGRAM BASED ON RELAP5/ MOD 3 AND TRAC-PF1/ MOD 2
- UNIVERSITY EXPERIMENTS
- REACTOR APPLICATIONS OPERATING REACTOR EVENTS ANALYSIS, e.g., LASALLE/BWR STABILITY ADVANCED LWR ANALYSIS REGULATORY ISSUE RESOLUTION SYNTHESIS OF RESEARCH RESULTS DEVELOPMENT OF PLANT MODELS
- NRC SIMULATOR SUPPORT O.
O PLANT PERFORMANCE PROGRAM ELEMENT THREE ACTIVITIES:
- 1. B&W TESTING
- 2. EXPERIMENTS AND ANALYSIS O
- 3. MODELING l
O I
O B&W TESTING l
I. Multi-Loop Integral System Test (MIST)
II. University of Maryland at College Park j (UMCP) 2x4 Loop III. Once Through Steam Generator (OTSG)
O Testins O 1 i
_ _ _ _ - - _ - _ - _ _ )
4 0 PLANT PERFORMANCE PROGRAM ELEMENT l l
THREE ACTIVITIES:
- 1. B&W TESTING
- 2. EXPERIMENTS AND ANALYSIS O
- 3. MODELING l
O
- _ _ - - _ - - - - i
O B&W TESTING 2 l
. presented to the j Advisory Committee on Reactor Safeguards i Combined T/H Phenomena and Core Performance Subcommittee Meeting May 23,1989 Bethesda, Maryland O
by D. Bessette Reactor & Plant Systems Branch Division of Systems Research Office of Nuclear Regulatory Research 1 2
revised 5/22/89 g 1
O l
B&W TESTING l
I. Multi-Loop Integral System Test (MIST)
II. University of Maryland at College Park (UMCP) 2x4 Loop III. Once Through Steam Generator (OTSG)
O Testing ,
I i
1
O I. MIST Status Objective: To obtain integral system- test- data- to study thermal hydraulic phenomena unique to the B&W NSSS
Participants:
NRC, B&W, B&WOG and EPRI o Phase-I: Design. Completed.
o Phase-II: Construction. Completed.
o Phase-III: -
Testing completed. Total of 50 tests conducted. O
- In addition, three tests were conducted for and funded by Toledo Edison.
A SBLOCA data base has been established to satisfy the TMI Action Plan Requirements (NUREG-0737) Item II K.3.30 that SBLOCA calculational models be compared to applicable data.
2 9
O ursr status (continuea) o Phase-IV: -Eight tests (mostly operational transients) performed. schedule
)
to conduct one additional test within the next two months.
o Contract termination in December,1989.
i O
O 3
1 1
MIST Status (continued) Gj i
Phase-III tests:
Mapping - 9 ~
Boundary system - 11 Leak-HPI configuration - 6 Feed and bleed - 4 Steam generator tube rupture - 6 Non-condensible and venting - 6 Reactor cooling pump operation - 6 Other - 1 1
Phase-IV tests:
SBLOCA w/o HPI - 2 O MIST scaling - 3 Station blackout - 1 Intermediate size break - 1 Steam generator steady state - 1 4 0
O Misr status (continued)
Reports:
All intermediate data reports for all the tests have been issued.
Data Analysis (Vol. 2 to Vol. 9 for Phase-III, Vol.11 for Phase-IV) reports are in preparation.
Code analysis (Vol.10 for Phase-III) report in preparation.
O Vl3t 9: Publication in April,1989.
Remaining schedule for publication in June,1989.
Post-test analyses:
RELAP5 analyses for 5 Phase-III completed. MIST report Vol.10. '
TRAC analyses for 4 Phase-III tests {
completed. Analysis of 2 Phase-IV tests '
scheduled for FY 1989.
O 5
i
)
II. UMCP 2x4 Loop g A review of UMCP analyses and analytical work concerning thermal-hydraulic scaling, l UMCP-MIST' counterpart tests, and-MIST' '
design atypicalities was conducted on !
May 9,1989. ;
participants were from the NRC, INEL,-
{
LANL, MIT and Purdue Univ.
UMCP presented the analytical scaling for single-phase subcooled break flow, and applied it to two counterpart natural circulation experiments at UMCP and MIST. The results showed good O agreement in the transitions between different modes (single-phase, two-phase, 1 boiler condensation) of thermal-hydraulic behavior for the two experiments.
l 6 9 ,
u , x i
L O UMCP 2x4 Loop (continued)
Evaluation of the influences that four MIST design atypicalities have had on the characteristics of SBLOCA transients.
Despite the distinct difference between the two facilities, similar thermal-hydraulic phenomena were observed.
UMCP concluded that these observations could be taken as a strong indication that the phenomena are indeed generic, and the UMCP small-scale, reduced-pressure facility is fully capable of simulating the energy transport capabilities of a B&W system over a O range of water inventories.
Reviewers concurred with UMCP conclusions.
A report has been prepared by the UMCP on the subject of data comparisons between MIST and UMCP experiments on SBLOCA behavior in B&W plants. The analysis concerns four of the identified MIST facility atypicalities. The contents of the report is shown below.
O 7
9
- 1. Introduction S 2 Comparisons of selected MIST and UMCP characteristics 2r RVVV~and downcomer characteristics 2.2 Pipe size 2.3 Metal heat capacity
- 3. Comparison of atypicality effects 3.1 Comparison criteria 3.2 Baseline concerns 3.3 Phenomenological comparisons 3.4 Effect of pipe size 3.5 Effect of metal heat capacity
- 4. Summary 0
- 5. References Post-test analysis:
l LANL completed a preliminary analysis of a UMCP test. This analysis is part of TRAC post-test analysis for MIST (draindown test 3004CC), UMCP and S'RI2. TRAC predicted the UMCP transient with reasonable agreement.
8 O
O,- UMCP 2x4 Loop (continued)
Table I: Test Performances in 1987-88 at UMCP Test ID Power Sec. Level Break Site (kwp 19r of fulH STR/AFWil25 155 55 18 STR/HPil020 160 55 l '8 STR/HPIl022 152 55 18 STR!HPille4 153.2 55 1/8 STR/HPill09 153 50 1/8 STR/HPilll8 153 50 1:8 STR/ITLO312 132 40 1.8 STR/ITLO318 132 40 !!8 STR/ITLD805 168 50 1/8 STR/ITL0807 168 50 !!8 STR/ITLil30 155 55 1/8 STR/STPillt 141 55 I/8 MISO 92188 75 60 1/16 MIS 100588 75 60 1/16 Test ID Power Sec Level Break Site (kwl (% of fulu STR/BCM0715 141 50 1/8 STR/BCM0731 142 50 !!8 STR/BCM0915 141 75 !!8 STR/BCM0917 141 50 1/8 STR/BCM0922 142 55 1/16 STR/BCM0924 141 55 1/16 STR/BCM1006 141 50 1/8 STR/BCM1008 142 75 I!8 STR/BCM1013 141 50 1:8 STR/BCM1015 143 75 If8 STR/BCM1219 140 75 1/8 AFW - Aux-Feed Water ITL - Integral Test 9 BCM - Boiling Condensing Mode STP - Step-Wise Test MIS - Counter-Part for MIST Test
O III. OTSG Testing o Air-water OTSG auxiliary feedwater (AFW) separate effects experiments at INEL o OTSG Technical Advisog Group (TAG)
O 10 0
O OTSG Testing (continued) ;
o Air-water OTSG AFW separate effects experiments at INEL i Objective: To obtain experimental data for improvement of code models during:
AFW injection in a B&W OTSG OTSG Phenomena studied:
- AFW penetration and distribution
- Tube wetting
- Flooding at tube support plate O - Liquid carryover
- Vapor generation
- Condensation
- Primary to secondary heat transfer Status: Testing completed. AFW model has been developed.
Implementation in RELAP5 and TRAC scheduled to be completed in FY 1989.
O 11
OTSG Testing (continued) $
OTSG TAG Objectives:
To investigate the thermal-hydraulic issues related to OTSG.
To identify whether additional experimental data was needed, and if necessary, define the required experimental design concept (s).
Membership:
O NRC, EPRI and B&WOG. Chaired by the B&WOG's representative.
l 12 O l
l
n
.~
O OTSG Testing (continued)
OTSG TAG (continued) o TAG Report published in March,1989.
o Major TAG
Conclusions:
3 In general, the current state of knowledge regarding most thermal '
hydraulic phenomena important to OTSGs is sufficient. However, certain OTSG geometry dependent thermal-hydraulic phenomena were f und t have data deficiencies.
O Detailed sensitivity studies using best-estimate codes can and have been used to understand the importance of certain phenomena in OTSG performance. Using this approach, the following phenomena were found to be unimportant even though geometry specific data was limited:
- 1) AFW flow distribution
- 2) AFW spray condensation O 13
OTSG Testing (continued) G OTSG TAG (continued) :
Seven phenomena (entrainment, deentrainment, liquid carryover, void distribution above mixture level, phase ;
separation, decontamination factor, and-flow-induced vibration) were found to have data deficiencies.
1 The majority of these phenomena are I important to the BE analysis of rapid OTSG depressurization and/or excessive steam flow. g BE transient predictions of SG depressurization, excessive steam flow, '
and SGTR (dose calculation) will require experimental data to develop and verify code models for the identified phenomena.
The verification of BE predictive models requires composite testing. Separate
! effects testing may be required to l generate data for model development.
- 14. O '
s O OrsG Testing (continued) 1 OTSG TAG (continued) i o NRC and the B&WOG met on March 17; 1989 to ' discuss the TAG findings.
The NRC and the B&WOG endorsed.
the TAG findings.
The B&WOG agreed to go forward in a joint effort with NRC to obtain the necessary experimental data.
g -
The B&WOG renewed their commitment to provide up to $2 million.
NRC is committing similar amount of $2 million.
Total cost of an appropriate OTSG experimental program is undetermined at this time.
O 13
OTSG Testing (continued) G-The MIST PMG was requested to examine facility designs and provide the-NRC and the B&WOG'with a recommendation on how to best proceed with an OTSG testing program. Effort is estimated to be completed- by mid--
, June,1989.
The PMG met on May 17,1989.
Reviewed proposed OTSG composite and separate effects facilities from B&W and INEL. INEL's proposed OTSG separate effects is satisfactory. MIST PMG requested additional information O on the proposed OTSG composite design from B&W.
o Future activities The NRC and the B&WOG will meet again to finalize the necessaiy agreements, and to initiate an OTSG experimental program.
16 0
i . !
t O MODELING I l
o RELAP5/ MOD 3 l
O O 17
i l
'l RELAP5/ MOD 3 Status 9 j o MOD 3 modifications scheduled for completion in June,1989.
Fix known MOD 2 deficiencies. '
Extend range of applicability of RELAP5 to large break LOCA.
Utilize code development expertise of ICAP members, o MOD 3 modifications Critical heat flux 0 Reflood heat transfer Critical flow Vertical stratification criterion ECCS jet condensation in large horizontal pipes Horizontal stratification criterion 18 O
O RELAP5/ MOD 3 Status (continued) q o MOD 3 modifications (continued)
Pipe offtake model Metal-water reaction Fuei model Radiation heat transfer model Non-condensible gas model O Juncti n based interfacial drag LBLOCA hydrodynamic phenomena Code portability O 19
RELAP5/ MOD 3 Status (continued) $
o Milestone release interim " frozen"~ code version )
with input description: June 1,1989 !
complete developmental assessment: )
and release of " frozen" version of MOD 3: September 30, 1989 Complete draft documentation:
January 1,1990 O
l 1
20 e
)
O EXPERIMENTS ~ and ANALYSIS-presented to the Advisory Committee on Reactor Safeguards Combined T/H Phenomena and Core Performance Subcommittee Meeting May 23,1989 Bethesda, Maryland O
by D. Bessette Reactor & Plant Systems Branch Division of Systems Research Office of Nuclear Regulatory Research O
- l
.i O
EXPERIMENTS ~ and ANALYSIS' I. 2D/3D II. BETHSY III. ROSA-IV O
1 0
O 20/30 Program status Facility Testing o CCTF Completed o SCTF Completed o UPTF 26 out of 30 tests completed. Remaining 4 B&W/ABB tests to be completed by 9/89 O
4 l
O 2
2D/3D Program Status (continued) e Test Data Analyses Status A. SCTF I test series Completed.
RIL issued.
B. Fluid Mixing in cold leg and Completed.
downcomer RIL issued.
C. Hot leg Completed.
countercurrent RIL issued. 1 flow for SBLOCA l O
D. SCTF II test series In progress.
RIL to be issued by 6/89.
E. CCTF II test series In progress.
l RIL to be i issued by 6/89.
F. Upper plenum In progress.
injection tests 3 0 ,
4 O Test Data Analyses Status G. SCTF III test In progress series H. Upper plenum deentrainment I. Downcomer ECC bypass J. Steam / water interaction in loop O K. ECC cold leg injection tests L. ECC downcomer injection / vent valve tests i
O 4
s 2D/3D Program Status (continued) $
Significant Results A.. A fluid mixing test series in the full scale UPTF showed that the cold ECC mixed well with the hot primary coolant so that the thermal shock of the vessel-wall due to ECC injection was not serious. This data helped close the PTS issue.
B. A hot leg CCFL test series in the full scale UPTF showed that under typical g PWR operating condition, steam / water countercurrent flows in the hot leg were very stable, far from the CCFL limit.
l Therefore, decay heat can be removed in a stable manner using a reflux condenser cooling mode during Small-break LOCAs.
i 5 O
O 2D/3D Program Status (continued)
C. Test series in the radially and axially full scale core test facility (SCTF) showed that there exist strong cross flows between the fuel bundles so that local overheating due to power density differences would not occur.
D. The tests data from the 2D/3D test facilities (UPTF, CCTF, and SCTF) provided a valuable basis for computer code uncertainty evaluations.
O O 6
2D/3D Program Status (continued) e Schedule for preparing final 2D/3D program summary report ITEM COMPLETION DATE
- 1. Agreement on report layout and October 1989 responsibilities
- 2. Availability of all December 1989 test data
- 3. Completion of all February 1990 g individual test reports
- 4. First draft of final summary report completion by primary May 1990 authors comments by July 1990 l l other parties ;
resolution of September I comments 1990 7 O
O: 2D/3D Program Status (continued) 4 ITEM Completion D' ate
- 5. Revised draft of final summary December 1990:
report i
- 6. Review of final draft summary January 1991 report O 7. Publication of ,
final summary February 1991 report O s l
f f ohh o l eavot/ 0 D s r eysli n o d o .ttt y p cbeet e2 2 n t m o l x hrcsn n aew ras di esT eioe e o d e oriynr pg b a a yl l
e sj eh h i
eif ec nbsmt t t l s eos st a
,s er d r . oot o ctbt a rrtd rsf f eoiu preeef e i u vosuuapr i
o "e s m o o o l nson ob scav eprneso y l s i r sl reop S oses h eoen pyscgrf s r N cicrt . nt rss r ,
rt no a a O ceps o b t o .
orf er m d nahee iff al ot sa" cpogg m I
S eo t oo l nc e. . t m u U l i yoxa t cs c ea ai ) rmd r s L itl ec e" eb roe ( s ounel psavl ell ral .
C aal y l t m ed r e i ab a . l m N t t ab ,d l auov v e grmsrm l a O enc me oel t o s s "s i r ma "s eusea ar C d eieat cmo venreut se t c eresvnevr rg eo sucb gl i caaoo mtl ra i
s"eer gaors pd s j e s git o eorr ee l og o OP vr h ors rs hhh anedi b hreoohhnr T psptTt aP T d puPi TtTl i ra" o -
d s s sn t t ea
, r n s ysu n d d .I y o e o n n e gP r p m l t s s i a s .U a e e u al s tk oI mI R l : r nuI l
a ao s iR ae gg t A s -
s ro t tE B( ,
mR y nn ss ey r aA t uE ntCRb o ul mJ ssr SA r ii ff sp g - o r tt o J a sz I sA 0 eos ik p s m i eRf s or rsp l i rns f cse f osoee ersm ar +T1 on l dP on niiR n f e pT R soeu a A S u/io ous i e yfiS em sff ni eooos T E t sti C Q yt I stRF g)l t bm N L S nat l ns hti T et ati e u i t ggsl i
E P aca yd a el crl" Pl rnrl s ds oosc M M sl l it t nnma roieU oAoie l i U A raPf n i aAuu pcn ut l l l un rn apcu t p on iaaco 1 C X ei i e l eeasnoeCeaa O E t ct cs i aCt a sRF soHRARF p wem ctt sC aaai L D t esee ctA sM e R a L epepr aaRn I T J s u FCCD LSTSP FDTI R " i c U M h o A A * * *
- I R * * *
- T d G * * * * * * * *
- O R
P 0
D 0 3
/ f S 0, D O T 0 2 N 1 R E 0
- 0 l UM l
i S 0 1 0 ~ 3 M 0 -
U 0 0 N 0 0, S
t s mr ao m mr rp a ae rp ge N r n 0 g ga o oR l
o oP i r
i r r/ Py yt s r P P Ps rat I t aur Da R Ds Dr 3 m 3 e 3 o ml o /m 0
/p map 0u 5 /l uve 0i De SER 2S 10 2f 2R E
T A t L
foI I D V N E
" I E t M 1 M RL l
I u 5t RL EL Vf L EE P PV i 01 tWV T V NE PE 0
1 I [ 0E UL
" 1 I I L
1 O
i II. BETHSY Status 1
BETHSY facility o Designed to simulate FRAMATOME PWR 3-loop system.
o Scaling factors:
- a. Power and volume: 1/100 O b. Elevation: 1/1 o SGs each have 34 U-tubes.
o Maximum core power: 10% nominal O 10
BETHSY Status (continued) $
Data from four experiments selected for study o Criteria used for selection by USNRC:
countercurrent data allowing scaling criteria study between facilities
" holes" in existing data base known code deficiencies O
BETHSY data schedule for Analysis o Two-phase natural circulation.
o Secondary behavior as function of changing secondary inventory.
o SBLOCA: 5% cold leg horizontally oriented.
o SBLOCA: 1-inch cold leg with no HPI.
11 0
O BETHSY Status (continued)
Program Tasks o Data from first experiment, i.e., two-phase natural circulation in-house.
o Data analysis and code asses.sment studies, o Study will focus on natural circulation scaling and quantifying ability of MOD 3 to match data.
O Completion scheduled for end of FY-89.
O 12
.\
BETHSY Status (continued) $
Future Tasks o R~emaining three experiments will~ be analyzed and used for code assessment j as received. Projected data arrival dates:
- 1. Secondary behavior as a. function of changing secondary inventory -
June 1989
- 2. SBLOCA: 5% cold leg - June 1990
- 3. SBLOCA: approximately 1-inch break with no HPI - December 1990 O o Overall task completion: mid-1990.
o Results will be factored into ICAP.
1 l
13 9
4 O
III. ROSA-IV Program Status O
O 14
.___-_______-__-_-_a
O
. E L E T T F S L N A E A A , R T C L ) . A A I S P % N P T L - 0 A O E A I 8 P . 1 N I S D C 4 O 1 E T A / O O MA A F F 1 L E T O L F S yU - O T A R A P E C L S U H U H R ) I E - O P ( P I F O T F C T B P
) N S D R F E I A N L T N
(
C A A I O T P J S Y . S O R A A L T S T L ,U Y T M ( - . E N B S T T B E 1 B E S T A I O O Y S : U M N N L W T U 1 T I E E I O T I O - R D I ,
C T L H G U E U S . A S I G N P L N G F D E C N I 1 X C A . E D A I L 4 E N R E T S U F T A 1 I T S U L S C 1 , E C TE S S 4 S L S T Y N SW A T A T I E N H D N N N WI TF O E E O E O L O G T MI M L I M I C I T I F C A E T S R L N A U R A R G A O V H D E R E E F O C I E C N P E P S R S T L A O X P X A S P A E E C E O E H T E L P C V G U oF I RM . . wF
- A I . .
T E A L S A B C D S
O R
O O l
C .
R F S N O O E S T I
- U N S T S I S E I R Y S U L E B Y E S I F L N S : C F D K G I R A O E C N O F D I I S F N E U L S S A E Q L E L U P N I R U O A .
D WD F E
I R
J S E D T N D A S A N N I E U V I E M T D M A A N G N S I X R A N E E R E T I E Y I E SC V W R T P E . NR O T A I X B S ON R E M L E E MS P B O M U
I C T N I A T EU D A S S A C C I H T E F U L E T A I D S I V I D D T N . E C AW U S O2 U A H M T E C9 S F E A S T 9 S E T . R O1 IV S A N G G V T I E R R O N I L " - N E E R I
- E I TA A P C P L A T T OS P O N A S I N HO A O O V C O S U " R J C C I S R -
A E S H . . . O .
T A B C R A O O O
E
_ D R E U T S E S L . E P N R M O P O I C T H A G
- L I U H M C A R D R I N G C A O
R L N P A O R I V U S I T S
- A E A N R S P O E E R S D .
A G Y H L N B P E I
- V R D O E A E W L E S T L S D C E D I R N U L D A Q A D I E A E L S L
S G E P C N R O I I O O P S C L O
T O O
H T T T E S O I L R TW T I U F T P N O S U E S E V N G A T Y E O S L W B .
I A L ,"
T M R S A A N W A O U I R T O O L R T H G A I L U F AT E D T F C N T A R " N D L E I N E . I E U S C O S T T C R I A N N C R E L T H E A I I V A A P M D C E R . L - I N E R U D U2 R I R L" T E C E . P A A T R G P A E R R T N C I N X N G E U I o E S
U D
N C I E E D E
R U F O L ME V T M O A R N E A O S D O N W 1 P H C R E O F E P E A E H N P S O
- S V N S P K - A T O T E E A C H W N R L H F Y A P D T E " P P O M R - E M T2 N D I D T N N O G N R E E O O T D G I A E W L I I N N S P O N T T Y A I E E X H I A A T R D L E S L V I S U G G U R L P D L N R AS C E I A E I u T R S C L% D S O A O I B A E0 O F D T C O F R5 M O O O 0
t i
,i i 9 W !
x Q&
w 9
u
_E<3 $
m w
y T- - ,._.
b
- v ~S 2a i,
" ( _
/c,
_4 v s - e w
f - -a
[ F- 8 w -
3 W
-d 8
~
8 (r d
-$m l Y
W N e a r
- h O
O I
dO od "
O
I U
D Q I
L A N S E
E A T S .
R ASO L E O N A T -
A T CI N P A ,T O E A ET S C D HN O S G I
O TE N P L SN I nL B O S A
T G NE I
SS E R L N N I
AO R A D N E R RD P E N T E M I
U D
T A E L A -
R E EC D C R P P,HO N E X U TLB O L OV I
E T F S L SO AO E A S E N G E A
V E I N HE R I S G E S PD S E L EE U R HIAN DI D D O P U R C
D P O L U ET E
K ND .
I VS R T AIORR N
U O E S A H S WE QL I
L A M H
T E S P C N
L DW I
I L I R
B UD L
O EE A O CC CL RS l l
R U : W D OI O E G FR M L T N E A F A
. I ,A E R R SJ R G K T A T T N A S E N Y S I E N L E B P R R O C M U E B M I D DM E L R E R I A D D
' A E T E S E E E P A P N T L E S X L O O S O .
M
~ E O R C O Y E P S P D C R H O E I S N B T C O S EA A U E S L E S TC S M H E AO T O G D T I LL N F E N .
N C UB E O G I A G N MS M R N E : I N N K O N I I L S R I O A N O Z C E O A I E O I Y I D O F R T R D S L F O T T . A B S A E M S N R L N E N D Y N E E UI O R A W T O D MR I .
P 2 O I I T N I E T E E F0 L L T E U SA C L D O0 F I I L Y J N B M B D P R R uA L E/ L A N O R O R aT E S5 A N O R A O O P V RP C I C D C P F 5 E E UA I P C L OL T A C CE I L A D R R . . E N I G C A B R U U N R Q I O
' I R J .
L U A .
D M 1 2 E
R O
C O
. I 1l l Illll1Il1
~
o N I N I
G W D N O I L E R F D A U E L L L A C C C N I I L .
T A E I G E T R N S A C I U R E P Q E B O E P O D U Y L A S L N . T D I TW N N C O E -
A L I L R E D F R N D E U O O C o I S
M R
P E L
N E R
S G O N E A T A A R R H P D D C 3 E N D E U L D S O A N L A F T A E H N V P 5 O ,
E R P Z 2 L E AI T L R ,
D N E O 1 I I R H U S .
Q M3 I ED L . . TO 3 4 IM E
R -
O C -
O o
1 l l l '
O T
T N ,S p E S E N D E C I I U N 0 C S E .
G C S I N A I C .
O I S T G Y F E N
- N T E V T E I E D I N M M N F T E S A R A E C .
M S R E S D E O S E G C O J I S S O N C C B R E S R O I O . A S A P C R D G N S E N X N E A L V S N A I I C A I D E R R S L T
- E G ,T A A N A E S A E E T E S N ,E M L R N M O N U C U E P R C O S T T M O R I S S L P P L Y N T I E A U O E B S A T E R L V U G Y S E E D I T M E V D E G T E A P B E S N I F R O U D 2 S I M A G O T D E T/ S O L 3 O R A T R T D M D L N DP M N O /
D E E E R E M 1 A R MTV E M / F E CI T U S P S Y G E - R P -
C D A PA G T A C I U N SS N S L A P T A UO O N E R O S M SR L I R T T
O O o O O ll
o T R
D G O O N . N G P T A N C I N P R R D I U S G E N U ES T N C S L B N I N U C L E R O N Y A M A C F I L C I E O . TI R L Y D ,NN E C T T E G E H P E R T N R C G X L F O C I R E N E A A P U R UT I E S P D A C R T S U N E S A S O L ,C A C E U P C C S I L D O S N E L C N O I I Y L S U .
O L L A Y A L C P T E E L R A MU S . T S A D E O R T E S A N Y o S P
TE A I
T E H
TK TE E
M I
T P
O E N E R U N M L A O R I O MI R J L O FW S R E M C T T E D H R D R D N TNT E E E D E N E A T T H E TI M G R S T T C H I N A E G E E A U T R O TD -
N U H I D I E L A N O Q WC NW P D U L E O O X R E S C T E : V D .
N S S SI E R C I A I E E U - I E RM R U E TAD T N A E E Q R AS U N S X H AE H TO T E U E T J R T SR SC O O O O O
A K
R O
E WO T
N I O R R
- E E D A H O N J C I A T
- R G E A S N L N E IF E S Z C Y I .
S L SD A E E N H .
R A TC U N T F Y Y P O S B .
U 9 R R .
Y E 8 E T L N -
E B R T O Y B MO N D F U U P E T N P R S F U R E O T D S U S N E C Y D E T N L N M C I Y A . E U U E NL R D T L AI Y T N N T ' B S O E O S N CM MU E E I I O C U I R . I U D R E E V D E P E O L A X .
R R I J E f P P R O O O 1
T P S O N E O A T L H
W N C O O O R U T L I E S F S V T 9 S E F N 8 E . E N O E 9 S A R M 1 A T P S R S H A E A O S ,
P D D H I E 1 - -
V S O S L5 A S E WT EP H A N T C VA E U E EL B L J G F : LE .
A N F G R D A T M I E N D E T N O S I I T A E R U E S U . C D M F T U Q A I O P O E E A D R E I T D L A E F
F L N O A D D R O E O . C P O E P L V D9 E C R G I O
E 8 S S O N Y D E9 S S C I L . B B1 E) S R L T 3 SF E A A U N E D L ,
ST S C E F E R O L0 AP S O L S M O M I3 T A L C S I C
/ W L( B E R S R L L S L C E F P K E I Y L A C P T A R B WT I % E U X P L OM I W 5 S S E T E WE I L R T R I L L P E C E L E A A N . .
e 9
8 .
9 S 1 T N
E 0 M 3 I T R N R E E E P M B X S M E S E E T N S P O S E I A S T A
L Y T A B : N .
T G E A N D N I T E E I R A M T S O D P E U O
L L
P E K
A F
F O E M D E O V O O R L E C C B I D O E S A B 2 B S C D E O E O L S L R M L S B O
/ I A S C 1 W L % F F
P K L 5 T
- R I . F C O W 0 T A W R L T L N L A . .
A L A a o o O
C Y R T NE E S F R U A U S .
T T H U R C K C F O I A D P R E
- P E . R E U N S B Z M S E E I A G U F M R O S O I G T ,S N O N I T I R D O C M N P E I Y E O I T T F O I V F A E F T T I I G F E A
- C I A N L O U o A S
O E T S P I S MD D
N A
I T
C R
R / E A I S T T I Z C Y T N C P I B S E E H R L E M P U A D U E S O S R E Q G U N S U S E A S E T .
S R N H . R A E A D T N P N R T M N I O E D N A WI D E D E T T S A MN ,
S A Y A I E S A T R H S R D E C N A P C E I U O E M -
_ I P C S L I I O P X C S B R R w O E A I S O P T T
- O O O O O
5 .
. T 0 N R A AI E L P V P GL E .
N N N D I E I E R V O D U O T L E D B E E A D V N N E E O L T L S I L A N T I L D O C T U A I I E S T E T P J S H C H N E O A I R P F 0 : U F R N D R S E O O E O S R T T H W T E U U A T O A R S H R I H L P S S E W S U E P M Y RI O S S U R PP A E C A . L F C I C M D Y O
- O L
R E
A I A R E R O A T N O
B S G P H M S - N I E I H I ,
F R S T N WA F P A A E U
_ E L
A A OB H H E V C FS S T D E S
I M
E . . .
S A B c o
U N N
,O O R O S B I O I .
T A T F T T N E A W C N E D L T O D A E I E B N L E ' I S R A E F E R S N R T I N O N A U P R L T A R C E O A E A R T C C C T R T O C K I A E S H A A T U P A G C G E I L O N U N S R R A H I S I E B C V F C D R V E O U N I G A IA T S I P N E GC S , E F H I L O E . K P G T C 3L W E R Y N G 0 C DB E D A T I N N I L OS N O M R I D A M C H D U Z H E E /% S C N D I T R S 55 S O N A R O I W
P P A
. P A0 S
. E T E B S Y A
M F O T : L - S S N N M S O O A S E FA N E O A U A L D E R O O D I S C B V I O T F O O D R I E N .
T C C I L L J E E T T O S I A I B T T C A I T D G ,
R S R A F E R T N D N R D O L A J T A E A I O E E O U B S L I W T R C P C S O N U S L O A I U L E O C N E L R U D D A T S M L A D L E Q O I C U I E A R O O P E R D N S D C T M F O R P
,I Y E .
M L D E A O .
0 N . .
C I 1 A A B C O 0 O
l
M O N A I O R T I G C T O N Ok R U T OP J P N H N V O C O" I C D I D -
T NA N E A ES I Z L O I U TR R M C N O I L O E I N A R T V I C F A A M " I H .
K R E S O R F P B I T A O O S M R " Y N H L P P L O C E P R A I mV A U N
- T L E B A A B L G "
Z N T I 3 E I R N R D C S O E U O N U T D A I
- S M E S L C
- / D S
- E S I I U C R P F S M A P A N Y U E L O L C E D E C A C R R N A E
- Y E A V R E S E E A C AH Z S
- M U EC .
Y I D RD A L H R O C T A T P R N F A N I P I O D AW O O
i O
1 MODELING2 !
i presented to the Advisory Committee on Reactor Safeguards Combined T/H Phenomena and Core Performance Subcommittee Meeting May 23,1989 Bethesda, Maryland O
by D. Bessette Reactor & Plant Systems Branch Division of Systems Research Office of Nuclear Regulatory Research l
2 revised 5/22/89
I
.\
l g
MODELING~ l l
I. TRAC-PF1/ MOD 2 II. RELAP5/ MOD 3 III. ICAP O
1 1 9
F l 0 TRAC-PF1/ MOD 2, RELAP5/ MOD 3 DEVELOPMENT o Development began in 1987.
o Consensus obtained among ICAP members on the code deficiencies that existed and the approach to resolution.-
o Contributions to development effort obtained from Japan, FR Germany, Sweden, United Kingdom.
o Cooperation between INEL and LANL.
O o Development complete June 1989.
o Meeting on June 7-8, 1989 to finalize developmental assessment plans.
O 2
TRAC-PF1/ MOD 2, RELAP5/ MOD 3 DEVELOPMENT g (continued) o Developmental assessment during June-October 1989. I o Code release in October 1989.
o Independent assessment by ICAP members. during 1990-91.
o No' further development planned (i.e. no TRAC-PF1/ MOD 3 or RELAP5/ MOD 4).
O I l
f 'I 3 e l
E_-_--_----------_--- - i
7
-O I. TRAC-PF1/ MOD 2 Status Cooperative development effort involving LANL, INEL, JAERI, CRAY Research, CEGB;-
and UKAEA.
1 Contributions: '
UKAEA and CEGB Post-CHF heat transfer and quenching.
Recommendation for interfacial drag i under wet-wall conditions.
Improved offtake model for O horizontal pipes.
JAERI Implicit axial heat conduction model.
Recommendations on core reflood l model.
Assist in code speed-up.
Assist in elimination of nonstandard FORTRAN. j 4
CRAY Research Assist in vectorization. {
Assist in parallelization. I i
i O 4
i TRAC-PF1/ MOD 2 Status (continued) $
Contributions:
INEL Provide to LANL documentation on code improvements to RELAP5/ MOD 3.
)
O 5 0
f.
J .
O TRAC-PF1/ MOD 2 Status (continued)
Schedule 1 Task,
)
Estimate finish datee
- 1. Provide the central 6/89 location for receiving updates, running final testing, performing and documenting quality i control reviews, and releasing of new versions to off-site developers.
- 2. Short-term development 2/89 (red star). Complete the implementation in MOD 2 c de.
O
- 3. Implementation of the 3/89 CCFL model into MOD 2.
- 4. Development of a general 6/89 model for interfacial heat transfer for all flow regimes, including the effects of noncondensibles.
O 6
o Task Estimate finish date g
- 5. Development of improved 2/89 reflood heat transfer by removal of Forslund-Rohsenow and the .
implementation of an appropriate film-boiling
. correlation.
- 6. Improvements in time-step 3/89 control.
- 7. Improved accumulator 2/89 l model. !
- 8. Break-flow time-step 5/89 sensitivity.
- 9. Flexible _ vessel noding. 6/89
- 10. Final Implementation and 6/89 9
testing of off-site developers models and updates.
- 11. Developmental assessment 10/89
& draft documentation.
I
~
l t
7 O
- . +
O TRAC-PF1/ MOD 2 Status (continued)
Model improvements in MOD 2 that are not.
available in MODI.
- 1. 3D-2 Step
- 2. Partial vectorization of vessel component
- 3. Inversion of the vessel data base
- 4. Generalized heat structures I l
O 5- Improved core void fraction for reflood
- 6. Conserving momentum flux solution developed for ID and 3D plenum i components
- 7. Consistent wall-shear between 1D and
! 3D 1
1 8. Improved valve model j l 9. Replace the Gauss-Seidel method with I
the Capacitance-Matrix method O 8
~
TRAC-PF1/ MOD 2 Stabus (continued)
Model improvements in MOD 2 that are. not; available in MOD 1; (continued)
- 10. Replace;the subcooled-boiling model
- 11. Capability to input magnitude and orientation of the vessel gravitational-acceleration vector
- 12. 60,120, and 180-degree rotational symmetry in cylindrical geometry
- 13. Improved offtake model for horizontal O pipes.
- 14. Implicit axial heat conduction model.
- 15. Irnplementation of ANS'79 decay heat standard.
9 O
O II. RELAP5/ MOD 3 Status o MOD 3 modifications scheduled for completion in June,1989.
.Fix known MOD 2 deficiencies.
Extend range of applicability of RELAP5 to large break LOCA.
Utilize code development expertise of ICAP members.
O M D3 modifications Critical heat flux Reflood heat transfer i
Critical flow Vertical stratification criterion ECCS jet condensation in large horizontal pipes Horizontal stratification criterion O 1o
O RELAP5/ MOD 3 Status (continued) 1 o MOD 3 modifications (continued) 1 Pipe offtake model.
Metal-water reaction Fuel model Radiation heat transfer model Non-condensible gas model Junction based interfacial drag O LBLOCA hydrodynamic phenomena Code portability 11 0
1 l
O RELAP5/ MOD 3 Status (continued) i o Milestone release interim " frozen" code version-with input description: June 1,1989 complete developmental assessment:
and release of " frozen" version of MOD 3: September 30, 1989 Complete draft documentation: '
January 1,1990 O
l O 12
III. ICAP Status 9 o 14 member countries are providing code assessment reports. The codes used are:
TR'AC-PF1/ MOD 1 and RELAP5/ MOD 2~
for PWR, and TRAC-BF1 for BWR.
o Reports are reviewed by INEL for RELAP and by LANL for TRAC.
o Code deficiencies are noted and provided to code developers for future improvements.
o Summary assessment reports will be published by the end of each fiscal year. O o An international code improvement consortium was formed within ICAP countries to pool resources together to reduce the NRC cost.
o Improved versions of codes (RELAP5/ MOD 3 & TRAC-PF1/ MOD 2) will be released in October 1989.
13 e l
E a r
S ,
B ER s RR t m9 N t ots9 e8 c y1 o YS S S ,
T E d t 3 I l o n2 L R r a y l
I B H aP a
& M A n T oi r
o St t c
a a n R e
R d i
Wr B o o
o C
n i
o
, r t S e l a
R w u C o m A p r t
af o
e l
. a e h i x
l x
u t a u s q a s l g
i f
o r f e e t
i 7
e b n r c 2 s h t
vt ei f l i
t - e f r r l i
Y e d 6 i v t a re o d w n 2 T e a t
a h t t
t c re
. w o he t I
M l
i t e e l n L , r p
n mf f f t e
I e e w w v B w t
a A s o ra e o wtn e e p oe A e n d r p
h s a p e l e
e i
Ti v d n h l
a e m l r a e s g l
a S e o r n tn t
a op l
e S
o e o a d r n a R c 3
- d i
t tr e e t
mu L R 2 u S . a o t a v siJ
, l a n o f m l u p Wa B r i wh c e r i cE nR l c
ai d
mi u c e r
o i
t o t n ad d i
o s
g v a c l c n e nl e s d y e c o o
i r M t n n r f at L e e i a a s n i
r d t t
P o d p N v i s o oe l
p l a
t l e
B y mn omuic l
s h i
i cRR d a l a n s i l
i s
t r N n s n o A e c a e g a a i
ai t N o e r c s n p ,
e a o i D L l i L l O cb C p t CO Nf o Ni c l
M A r e el l
- u e RE R r R s A h i
R
- P M NA Op O o R t wT
h R k
- S s r U s g a nl o
i G d P h t i l
N e f eA i o s i
I w d h t L w t u w T S e r my s s sL I
E e n d
e o r n l a nS E o o ui o r
o da e f h i M t n n d fr n t d ta d g o e e u a i vl n i a s p o l
u i u a nt t W b a i
a u b l c .
d c nl L E i r i z d aiy l
l a e ci l i
aN c
I c a l t
wt y of e B S V t
s u a i l f
E e do b a o l
p d i r
M R d n n L t N a v p b i t
e y r E
i nf uRt s i t r g c e pO s e de rat TM I i o o h n e i
n wl O A t at c t i e et .h e o wcl u R
N G t
n a f
o hl s p oc f a n Al a kn ar e p s t l e
O O I mml e t s LSi Pi e c N a mE x TR i
ui l t h c e y RL e e t ct r a i CP o h c a Of t at A d t t e t
o e t n h n u Y e gmi d e to i .
d e ooei T d ni vd n o o ws e
i n
cb ao I
L i
vd il oo r c di t i a
l l h or f nt v i nt I
B r ao p g p
R el a s s u n r et r
wo i A e n iUl l
c o o ci n r o l t
b ps l l i i T wP l l l
l i
at a i wst a LL A wcl u wdr Fa oo i i t
S l a r d R s l
u u N e LA c L g i R R ei mR h NP l
a N r eb e h l W N r s Ot BE c B r Dt O B * * * *
- s
. t u s o
e p y e t l
a nd r o h d e r fo n ie a t t o d e d w n a we
) y o s d i
1 u d
e l
t t n a i
F B g l n
u e
r n Y mos p u u e r n l
a b o i uo -
e b ah e d
Ce i
qiX l n n n m e t e o Ab r a c i
i s g l n xt t i n u n a a y i
n pt R) e a f i T n ol s c o c o d z o i
o el r e b i
c at Aht d C
(
i nl d
a h ei t a ).
t sb c et i o
r c u u e N e l
l o
t ma e h a oAn O )t e v wn i
t h r f sN f Muin o da t r .
S e t
i f
e n y eO o r A ph n ci o I
L r wfn e i
a e M r Ri at O M S c o i G c A e a E
L N
gG h )sed e t
h t Rb f i
h g et n
& d nI T BiRt eh e e m u e ub o e
I ,e r F t cf n r n n r d s L s e o( a n o i o i e mc l
N o p
d pt l ui s e mh r t ur o p N R Or o i
v o
p n u e a2 v( o n r
e8 h t e( wr e e
me st de/ h t t O ,p r s p s i
1 et d a ta r I
L d l t a o T Nt e n s l
i nl u a ru t
mh t e wemlc l l l
l c
C l
l i t B a i i i
s x wt o w o el fe
- w (a s e a l
A u af
-- a L p ap e e - r L mLk c N v c L p e L r E r E e r R o n N e n N ph N fo I
N d fo I O mo B r o B a s a O
t s e
a n c e i o
ar s h t c
k n e a r oGGOe r u t
c aMi h h or e s n R c t i r o cfePR o n n o A -
y( B DWoc i at a n r s d.A e e e nd t b p e g c c p s l u n s o e o u ni min o a i
t n o e r t a cd r e r o w r S o u e l sG s o l f
f Wt de l i
c a t at u
Tt s
a 's def o Ac e ocr l
i s h q
_ r d ? ot wen l
t s a o n s p p e e he o el gt
]
e p a e md O 9 wgt d r oi n 8 Og s g r n o 9
1 u s s s u
a n
o i nh t a f i
ms R e r f ci s u g u "e od n
i 2,Wr wn e r s d nd d s n e s i
2 B ud s a o p r r n n i oi s n n e s
h e o i p u o a [ i n
t ci r ch r t l
cd a u c t
c t e e e a oe n u l l l
i s oar g sd i v
o c e i t
Mt n w s h s nt e i t
n ws i t
c e
ot oiin n uf d )R At e a r i s
x o "o f t R a h e to t a e e?
i et d (N ). n s t wn o
i l
l i
h t n S gl e
r i a m r
e i
n ms a e pd c
s s oWT n a d a r
a r y e oi d a r s c
t o p A a e t t d
e a a a C a e D m i
l f I
s r O l h h n u .h a o 3 nh t i wb ATC a 1 d 2 f i i
O e t e e? g p s r n f a r e o o haot i i
l t h v i o
e s e st n a b d
t uibnl a
ooe mf b r h t
c w p n i
l n e i d o oo d d e o n ef l
mist r aaa e i t osiut a
p g r ( l n -
m a [r s l Ci t u v e t s e c si Aa mt i s Ri l
O i
ab l
xd o o T c l
xh i s e ct s -
a m wso e f c c od i
s o e
e nd p n e ore _
e y r d d h
t ai s d i
t i
p l l
n s sd viol e u?
i n o?
oi eb a
t wno s n
t i r
e ]e t a p )s o at a p m h a wit i ci e h t l oi sd i Wicd su l
W e du H n a n
. s n s .
h t .
r o 4 oa a 5 t s 6 t c O
1lll
T EM E F G'T E C OE S I AN N TDUIL O S MO C AS TAT N O AC N TNI F D I D/
S O L E T E A P RE I
GA L R I
OS S TM OS AL EU T T A A ATU S S L
C E :
ML I HF U T A I
E D UM A AC? T FR S GE C U E B N S DSD I WOEP L OE S DT S AIMRNO O E CX E EI T L
EL S STM A HU A AH H S UA S E NNE
.Q P MT T E FNR OET FOP I
E R I OG P A EO T EU M ,D O U C ,O R O8 N S NS O YT N R S I o R R8 F9 AT O WE I EC L
TI ND Ml ,ON P E. P F
O ,3BMITN A 1
F M2 - LAO L I S E T ED A T E E A TL T LME I N R T MNOE ENNICAM DE A MT I TE NV P S A UMR S R
T A
TY C EE P U
J AE O OL NS E T E N DR H XFE EA T
TE OS S U
I
,E N U NR OMD PWN FT FI E TO I
Q FA OEF H L
S EH PO T R O H A S E AD M TNO H I
T E H
F N T OT R T T SN ROE R E TG E E TC?
I R TFR E E R R K FT T .T SIN R A CFU R T AUIPA I TA R N HHLNTL OEUERU T LUY T EES S N A A SE A P SDE ESTA H EH H ENR H VF OV o TL N EOTE I WRT WRAP
1 l 1 i
{
04RR .
~
L 3WR S~A3_ Y 3ROGRAV l
l 3 R ES E\~A~ 0 s
~0 AC RS s
gy 3
_ARRY 3-
- S '0 \ R R1 ANJ tOSE VARC _E 3A
'OA<
R JGE N A~ 0s A _ _A30 RA~0 RY1 f
l V AY 23, ' E89 i' O
y l
i >
NRR COMMENTS ON BWR0G PROGRAM
- PHASE I STUDY IDENTIFIED THE NEED FOR REMEDIAL ACTION TO ASSURE COMPUANCE TO GDC 12 DURING REGIONAL IN BULLETIN 88-07 SUPPlIMENT 1 PROVIDES FOR INTERIM OPERATING RESTRICTIONS TO REDUCE PROBABluTY OF THERMAL UMITS V10LA110N l
ONG0ING BWROG PROGRAM IS SEEKING MORE POSITNE LONG RANGE SOLUTION O
l
- u . :.
- L- . _ . _ - - - . . .
J>
4 NRR COMMENTS ON BWROG PROGRAM (CONTINUED)
- PHASE lA (GE TRAC 3D) STUDY INDICATES APPROXIMATE 4 % INCRE IN AVERAGE THERMAL POWER DURING 50 SECOND INTERVAL WITH OSCluATIONS GROWING TO PEAK POWER EVEL OF 200 %
- EVALUATION OF THERMAL POWER RISE IMPUTATIONS ON ATWS EVENTS INCOMPETE
- STAFF REVIEW OF COMPLETED WORK AWAITING BWROG REPORT SUBMITTAL
- ONGOING BWROG/EPRI STUDY IS INVES11 GATING INTERACTION OF OPERATOR RESPONSE TO AlWS WITH POWER OSCIU.ATIONS 4>
t
. ..- . 4. ..::
1>
EXPECTED COMPLETl0N
- BWROG PHASE I/lA REPORTS OVERDUE (EXPECTED IN JUNE)
Ii
- BWROG RECOMMENDATIONS FOR LONG TERM SOLUTION EXPECTED IN LATE 1989
- IMPLEMENTATION OF LONG RANGE SOLUTION (PLANT SPECIFIC SCHEDULES BASED ON SCHEDULED SHUTDOWNS) l l
lI l
l.
't i 1 .
. )
i i
FUTURE PLANS AND SCHEDULE.
\
- REVIEW 0F BWROG REPORTS ON CALCULATIONS (BEGIN JUNE 1989)
- CONTINUED INTERACTION WITH CONSULTANTS,
~
CALCULATION RESULTS AND RES TPG (1989 - 1
a c
{
l i
FUTURE Pl.ANS AND SCHEDULE (CONTINUED)
- REVIEW UTIUTY TS CHANGES REl.ATED T0 INTERIM RECOMMENDATIONS (MID 1989)
- REVIEW BWROG PROPOSED LONG TERM SOLUTION (LTS)
(l. ATE 1989 - EARLY 1990)
- REVIEW UTIUTY LTS SELECTION, IMPLEMENTATION AND TS CHANGES (1990)
- REPORT TO COMMISSIONERS (FEBRUARY 1990) e
v' j\ F q OUTLINE Limit cycles Why is there a limit cycle ?
Is there an average power increase during large limit cycle oscillations ?
Bifurcations ???
O ATWS What happens when the operator reduces water level at natural circulation ?
' J
( ) M-L: ACRS 23-MAv.99 ]
1 l
4 O '
AVERAGE POWER INCREASE )
DURING BWR LIMIT CYCLES We will show that during limit cycle oscillations :
Time varying portion of feedback reactivity is essentially an undisturbed sine wave (i.e., little higher harmonic contamination)
O If the reactivity oscillates as a sine wave, and the power oscillations remain bounded by a limit cycle, then the reactor must be subcritical (i.e., there is an average or DC component in the feedback reactivity)
==>Thus, there is a power increase during limit cycle c. oscillations.
Reactor-independent correlations relate the average reactivity increase with the
)
amplitude of oscillation.
( J M-L: ACRS 23-MAY-89]
_ _ _ _ _ - -- - a
9 4
I TYPICAL BWR TRANSFER FUNCTIONS t
REACTIVITY " POWER s,
Lc ... u , s. 1.
O i
1
.L l
~,
% ... o i. i.
O '
( ) M-L: ACRS 23-f/.AY-89 )
2
_ _ _ _ _ - _ - _ - - - _ - . - - - - - l 4
9 O ' '
l l
BWR OPEN LOO?
l 1
TRANSFER FUNCTIONS i
l l
l 50 0 l
l 1
0- '
\- - -90 P G H
^ ^
O I - -
S N E D D B - E
-180 G
Ii
- l. ,,. .,, , , , , . , ,,,,, , .. "
l
-270 0.001 0.01 0.1 1 10 100 FREQUENCY (Hz)
O '
( J M-L: ACRS 23-MAY-89) 2 I,
(3 r ,
REACTIVITY FEEDBACK DURING LIMIT CYCLES IS SINUSOIDAL l Limit cycles have been observed (numerically and experimentally)
Observed limit cycles are stationary signals. Amplitude is bound and signal is periodic (before bifurcation regime)
O => Power can be expanded in Fourier series n(t) = E Au sin (kwt)
Thermohydraulics:are an excellent filter.
(1/10 at 0.5 Hz, 1/1000 at 1 Hz)
=> p(t) = I Hu Au sin (kwt)
.~ H o A o + Hi A i sin (wt) + e L "
( J M-L: ACRS 23-MAY-89)
_ _ _ _ _ - - - - _ . - l
O '
TYPICAL LIMIT CYCLE 1 600 6 i
k l l l
i 4007- -4 I
i R e
P 200 -
J2 C 0
i t O e i
'70 0
\ nnni 0 1
y l '
-200 - -
-2 1 ' ' ' '
l -400 -4 ,
l 0 2 4 6 8 10 l Time (s)
O ' 2
( J M-L: ACRS 23-MAY-89 }
l I;
F:.
O ' '
SINUSOIDAL REACTIVITY With zero average, Power increases 160 j i i Reactivity (S)
IO 4:
120 -
e . .
i 1 \ !
a 300 .
y i O
~
s0 .
e .
Time (s)
P 60 -
o W
e 40 -
gb dk 20 _ 1 0
M dh VUUV .
0 10 20 30 40 50 Time (s)
' 2
( ) M-L: ACRS 23-t 'AY-89 )
O - '
i SINUSOIDAL. REACTIVITY With too much average, j Power decays i 2
, Reactivity (S) j t
R e
.5 -
, n. g n g 1
j I a l n -..
i !
~'
l
~
v O i f t 1J P I hime(55 O s!
w fgIt e 0.5 ffff j UVVVdudVVVhg@9 e
)
0 10 20 30 40 50 Time (s) 1 4
' #l
( J M-L: ACRS 23-MAY-89) l
- _ _ - _ _ _ - _ _ - _ _ _ .0
V-O ' '
i l
1 l l SINUSOIDAL REACTIVITY With the right average, '
I Limit cycle is established j
)
l 5 Readwy (S)
- ~
R .
\
l e
1
. i a ,, .
t 3 -
i V Time (s)
,3, ,
z, <;,,, 1 ,
O W l r 1 -
0 O 10 20 30 40 50 Time (s) o o ,.._s 23_.ee >
I,
.N
+
v) <
- THE MODEL
- A P- @ n . x c. E dt A ^
POINT KINETICS di _ (h. ^
n )C de d--
d t.
K ( m f m) - as T FUEL 2
cl P
--.- 4 4 t dP . a, p T ENERGY AND CONTINUITY d0 dt MODEL PARAMETERS ARE OBTAINED BY FITTING THE RESULT
(] 0F A DETAILED NEUTRONIC-THERMAL-HYDRAULIC CALCULATION 101 E ' ' ' ""4 ""'4 ' ' ' ' " "I ' ' ' "" % ' ' ' " "4 ' ' ' " '! <
- : i 10 E E z : : l 2
e 2
-4 -; + -
I 10-1 E -
5 '
10 4 ' n'd i f ' id f ' ' ' " I "i ' ' ' " n'I '''
10-3 10-2 13-1 10 0 10 1 10 2 103 FREQUENCY (HZ) l
(~)
i ouucasu.v.v.ee l
i i
O <
l AVG. :1EACTIVITY DECR: EASE DUE TO OSC::LLATIONS 0
t
-0.5 r !
l R I e '
a -l[
c ,
O !
1
-1.5 - t Y l 1
1 y 2-6 S
-2.5 - !
l- a l
-3 O 500 1000 1500 2000 2500 3000 3500 Peak Oscillation (%)
( ) M-L: ACRS 23-MAY-89 )
I
r Gj ,
AVERAGE POWER INCREASE DUE TO OSCILLATIONS 20 i
I !
l l 15 ' !
i P i l
O i e 10 r l r !
5F
'l l
l 0
0 500 1000 1500 2000 2500 3000 3500 Peak Oscillation (%)
O <
g ,,. , ,s 23., _ . g i
i 1
y .
)
O ' '
l 1
POWER INCREASE
SUMMARY
1 I -
Average reactivity decrease stabilizes (i.e.,
bounds) limit cycle oscillations Governed by general correlations Approximately .3 cents /%_ steady _ state Average power increase must exist Depends on particular power feedback coefficient Typically, power increase is 1.5% of peak oscillation amplitude t "
( ) M-L: ACRS 23-MAY-89 )
_ _ _ _ _ _ - . _ _ - _ - - - _ .- -- - l
'L A
M N I T
3 6
3 9
7 9
3 ,
5 -
e s
O P -
- 0 0 0 1
- l u
I - a O - -
T -
v e
C -
l b
a U
- n o
D E M s
a e
E P R -
r o
R A H
O F 0 7 3 4 2
t o
I 7 4 1 -
i L S N 1 1 1 0
t a
r E L U y a
VY A I
c e
E LI T X
A -
d g
n L R =_
i b
r RI O E o
E B F 9 3 7 20 ._
. t TA O N . 9 I .
0 0 f 8
AT o I S _ 0 1 1 1 T r o
o WS N A
R f t
c a
S D a
_ W T
O Y A
C E
E K
A-b d
e y
c A D E u P- d e
r F M 2 6 7 t
n O O.5 T
3 1 8 3 0
ic e
1 1 0 i T T O
f f
e C B -
o c
E -
ty F -
i i
v F = t c
E W )h -
a e
Ob/ . 6 7 8
=
r Ll 2 9 ty FM( 3 1 is n
e D
R E) _ . .
WW 0 0 0 0 -
OM.0 P( . 2
. 0 0
5 1
0 0
1 0 ._
5 _
O " s
<EE Os y<> % <>
lc l
7.O BWROG/ACRS MEETING ON STABILITY MAY 23, 1989 O
O 051989.HCP
.s
.. O PURPOSE OF MEETING
-0 UPDATE-THE ACRS ON BWROG ACTIVITIES RELATED T0 RESOLUTION OF THE STABILITY ISSUE O PROVIDE STATUS REPORT ON THE DEVELOPMENT OF BWROG LONG-TERM CORRECTIVE ACTIONS 0 INFORMAL DISCUSSION OF NRC LETTER DATED MARCH 22, 1989
([) -
BULLETIN 88-07 SUPPLEMENT 1 GUIDANCE SUGGESTED BWROG ANALYTICAL ACTIVITIES NOTE: THIS MATERIAL REPRESENTS THE PRELIMINARY PERSPECTIVE OF THE BWROG STABILITY COMMITTEE.
IT DOES NOT REPRESENT A COMMITMENT ON THE PART OF THE BWROG OR ANY UTILITY TO A SPECIFIC COURSE OF ACTION.
O 051989.HCP
'o
%GENDA-1:15 INTRODUCTION AND OVERVIEW 1:30 BWROG RESULT TO DATE 2:15 BWROG LONG TERM CORRECTIVE ACTION STATUS 2:45 NRC LETTER TO BWROG 3:00 DISCUSSION O .
051989.HCP
.t
- D .B W R S T A B I L I T Y C O M I T T E E OVERVIEW PROGRAM OBJECTIVES:
PHASE I:
0 . INVESTIGATE PLANT RESPONSE TO REGIONAL INSTABILITIES 0 IDENTIFY APPROPRIATE REMEDIAL ACTIONS PHASE-IA:
O STUDY LARGE AMPLITUDE. OSCILLATIONS
.O O- DETERMINE AVERAGE POWER RESPONSE DURING ATWS POST PHASE I/IA ACTIVITIES:
0 REVIEW PHASE I/IA RESULTS l 0 PROVIDE GUIDANCE ON INTERIM ACTIONS 0 GATHER WORLDWIDE DATA 0 IDENTIFY VIABLE LONG TERM SOLUTIONS !
O -
051989.HCP
~O STABILITY COMMITTEE. CHRONOLOGY 11/09/88 BWROG/NRC MEETING IN WHITE FLINT, MARYLAND RESPOND TO NRC INFORMATION REQUEST !
PHASE.I/IA APPROACH /RESULTS OVERVIEW ;
INTERIM CORRECTIVE ACTIONS HAVE BEEN TAKEN BWROG STUDYING OPTIONS; WORKING TO RESOLUTION 11/18/88 CECO ENFORCEMENT CONFERENCE ON LSCS )
EVENT AND DECAY RATIO DESIGN CONTROL 11/22/88 BWROG/NRC MEETING RESOLVED 100% FCL RESTRICTION j CONCERN 1
() -
NRC TO REQUIRE SCRAM ON 2 PP TRIPS LONG TERM RESOLUTION SCHEDULE DISCUSSED 12/13-15/88 STABILITY COMMITTEE MEETING-IN ST. LOUIS FOUR SUBCOMMITTEES FORMED WITH CHAIRMEN ON A STEERING COMMITTEE:
0 PHASE I/IA CLOSURE 0 LONG TERM SOLUTION (DETERMINE ACCEPTABLE OPTIONS) 0 INTERIM CORRECTIVE ACTIONS REVIEW (OPERATING AND SAFETY) 0 STABILITY EXPERIENCE REVIEW (ACQUIRE AVAILABLE DATA) !
l O
051989.HCP l i
l
~
l O STABILITY C0letITTEE CHRONOLOGY (CONTINUED)~
u ESTABLISHED SOLUTION OBJECTIVES AND POTENTIAL SOLUTIONS 12/16/88 STABILITY COMMITTEE PRESENTATION TO FULL ACRS 4
- DISCUSSED PHASE I/IA SCOPE /RESULTS;-
FUTURE PLANS ACRS DOESN'T BELIEVE STABILITY TO BE SIGNIFICANT SAFETY ISSUE ACRS QUESTIONS ON ATWS/EPG IMPLICATIONS 1 i
12/30/88 BULLETIN 88-07 SUPPLEMENT 1 ISSUED SCRAM ON TWO PUMP TRIP FOR NON-FLOW 4 BIAS SCRAM PLANTS (S) -
INTERIM CORRECTIVE ACTIONS ATTACHED LONG TERM SOLUTION 6/12 MONTH SCHEDULE L/70-12/89 STABILITY COMMITTEE MEETING IN DALLAS '
UPDATE STATUS / PLANS OF SUBCOMMITTEES DEVELOPED DRAFT COMMENTS / CLARIFICATIONS MEM0 TO NRC 01/17-18/89 EOC AND FULL BWROG MEETING (NEW ORLEANS)
ENDORSED STABILITY COMMITTEE ORGANIZATION APPROVED BULLETIN SUPPL. 1 BWROG COMMENT MEMO AUTHORIZED FUNDS TO CONTINUE O. EFFORTS 051989.HCP
-() STABILITY COMMITTEE CHRONOLOGY (CONTINUED) 01/26/89 BWROG LETTER TO NRC COMMENTING ON BULLETIN SUPPLEMENT JAN-FEB SEVERAL NRC CONFERENCE CALLS BWROG COMMENTS EXPLAINED PHASE I/IA SCHEDULE DISCUSSED NRC QUESTIONS ON POWER LEVEL INCREASE DURING ATWS OSCILLATIONS EPRI/NSAC ATWS/EPG WORK DISCUSSED JAN-MARCH SEVERAL SUBCOMMITTEE MEETINGS LONG TERM SOLUTION AND PHASE I/IA COMPLETION SUBCOMMITTEES ESPECIALLY ACTIVE
() 02/27-3/2/89 ABB-ATOM EXPERIENCE REVIEW AND LONG TERM SOLUTION PLANNING / WORK MEETING 03/01/89 ABB-ATOM /NRC MEETING ON STABILITY EXPERIENCE 03/15-16/89 ANF/KWU EXPERIENCE REVIEW AND STABILITY COMMITTEE MEETING (ATLANTA)
ORNL (J. MARCH-LEUBA) STABILITY '
OVERVIEW POTENTIAL LONG TERM OPTIONS IDENTIFIED SCOPING AND FURTHER ANALYSES PLANNED O
051989.HCP
a l[)
STABILITY COMMITTEE CHRONOLOGY (CONTINUED) l 03/22/89 NRC LETTER TO BWROG RESPONDING TO 01/26/89 LETTER INTERIM CORRECTIVE ACTION COMMENTS BWROG ANALYTICAL PROGRAM !
SUGGESTIONS 03/28-29/89 FULL BWROG MEETING (ORLAND0)
DESCRIBE LONG TERM SOLUTION RESULTS AND PLANS PHASE I AND IA REPORT PLANS SUBCOMMITTEE ACTIVITIES REVIEWED 04/5-6/89 PRE-MEETING AND NRC MEETING
()' -
DESCRIBED LONG TERM SOLUTION PLANS AND STATUS UPDATED AND DISCUSSED PHASED IA ATWS RESULTS INFORMAL COMMENTS'ON 3/22/89 l NRC LETTER (SUGGESTED WORK SCOPES AND BWROG WRITTEN COMMENTS) 04/89 BID SPECIFICATION AND SCOPING STUDY PROPOSAL REQUEST DEVELOPED 04/26/89 BWROG EOC MEETING (CHICAGO)
APPROVED STABILITY COMMITTEE TO PROCEED WITH LONG TERM FIXES WITH GE AS CONTRACTOR
()> !
051989.HCP l
4
-~O - STABILITY COMMITTEE CHRONOLOGY (CONTINUED) 05/8-9/89 STABILITY COMITTEE MEETING (DALLAS)
DEVELOPED 4/6/89 NRC MEETING
SUMMARY
INCLUDING RESPONSE TO NRC 3/22/89 LETTER DEVELOPED METHOD / APPROACH OF WORKING WITH GE ON SCOPING STUDIES 05/24/89 BWROG FULL EXECUTIVE MEETING (CHICAGO)
O O .
051989.HCP
. - \
\
~O l
RESULTS TO DATE 1
O O -
051989.HCP
- I[)
PHASE I RESULTS 0 ENGINEERING SCOPING ANALYSIS PERFORMED TO EVALUATE THERMAL MARGIN WITH REGIONAL OSCILLATIONS 0- REGIONAL OSCILLATION MODEL BASED ON TEST DATA DEVELOPED TO DRIVE TRACG THERMAL HYDRAULICS 0 INDIVIDUAL CHANNEL AND NEUTRON MONITORING SYSTEM' RESPONSE STUDIED 0 CONCLUSIONS:
() -
POTENTIAL FOR SAFETY LIMIT VIOLATION PREDICATED UNDER SOME CONDITIONS SUBSTANTIAL LPRM AND APRM INDICATIONS AVAILABLE BEFORE SAFETY LIMIT IS REACHED RESULTS SUPPORT INTERIM CORRECTIVE ACTIONS l
l l
l C) 1 051989.HCP )
.o PHASE IA RESULTS 0 ENGINEERING SCOPING ANALYSIS PERFORMED TO EVALUATE IMPACT OF OSCILLATIONS ON AVERAGE i REACTOR POWER 0 CORE WIDE OSCILLATIONS STUDIED USING COUPLED 3D' FULL CORE NEUTRONIC TRACG MODEL 0 SMALL MAGNITUDE OSCILLATIONS PREDICTED AT STEADY STATE HATURAL CIRCULATION CONDITIONS 0 ALL CONTROL RODS WITHDRAWN TO INCREASE MAGNITUDE OF OSCILLATION 0
() CONCLUSIONS:
AVERAGE CORE POWER INCREASED SLIGHTLY
(< 7%)
INLET FLOW /SUBC00 LING INCREASE SUSTAINED POWER INCREASE (UNLIKE ATWS EVENT ASSUMPTIONS)
NEUTRON FLUX OSCILLATION MAGNITUDE REACHED APPR0XIMATELY 200% WITH ALL 4 CONTROL RODS WITHDRAWN REDUCING LEVEL DID NOT INCREASE OSCILLATION AMPLITUDE
()
051989.HCP
l'
~
O STABILITY EXPERIENCE REVIEW SUBCOMMITTEE
' PURPOSE:
L COMPILE OPERATING AND TEST DATA ON STABILITY OCCURRENCES WORLDWIDE TO ENHANCE OPERATOR KNOWLEDGE AND QUALIFY LONG TERM FIX.
L RESULTS TO DATE:
0 REQUESTS FOR EVENT DESCRIPTIONS DISTRIBUTED 0 REVIEWED EVENT DESCRIPTIONS (FEB. -
MARCH) 0 ABB-ATOM PRESENTATION TO STABILITY O COMMITTEE (2/24/89) 0 ANF/KWU EXPERIENCE PRESENTATION TO STABILITY COMMITTEE (3/15/89)
O CONTACTED STOLLER CORP. TO INVESTIGATE INFORMATION POSSIBILITIES FUTURE ACTIONS:
0 OBTAIN STOLLER REPORT l
O 051989.HCP
- _ _ - _ _ - _ _ - - _ _ _ - - _ - _ - - - - _ - - - - _ - - - - - -- -l
I[) ' CORRECTIVE ACTION REVIEW aan i EVALUATION SUBCOMMITTEE i
\
PURPOSE:
REVIEW INTERIM CORRECTIVE ACTIONS AND PROVIDE PROCEDURAL GUIDANCE FOR IMPLEMENTATION RESULTS TO DATE: !
o CLARIFICATION OF INTERIM ACTIONS SUMMARIZED AND TRANSMITTED TO NRC o BASES FOR INTERIM CORRECTIVE ACTIONS REVIEWED AND DOCUMENTED i
() o PROCEDURAL GUIDANCE DEVELOPED o INVESTIGATED POTENTIAL ADVERSE OPERATIONAL CONSEQUENCES OF INTERIM CORRECTIVE ACTION IMPLEMENTATION FUTURE ACTIONS:
o EVALUATE AND DOCUMENT REMAINING IDENTIFIED PROBLEM AREAS l
l o
r 051989.HCP.
I
.;o PHASE I/IA CLOSURE SUBCOMMITTEE PURPOSE:
VERIFY TECHNICAL ADEQUACY AND APPLICATION OF THE GE I/IA STABILITY ANALYSIS RESULTS AND REPORTS RESULTS TO DATE:
o COORDINATED EPRI PEER REVIEW o PERFORMED DETAILED REVIEW OF PHASE I/IA ANALYSIS o DETAILED REVIEW AND COMMENT ON REVIEWED PHASE I/IA DRAFT REPORTS o C0ORDINATED ASSESSMENT OF ANF FUEL USING GE PHASE I ANALYTICAL RESULTS FUTURE ACTIONS:
o FINALIZE REVIEW 0F PHASE I/IA REPORTS AND ISSUE IN MAY AND JUNE o FINALIZE ANF T/H ASSESSMENT (2) _
051989.HCP
.- 1 LONG TERM SOLUTIONS SUBCOMMITTEE IDENTIFY VIABLE PERMANENT SOLUTIONS TO THE STABILITY ISSUE WHICH APPLY TO ALL GE BWR PRODUCT LINES AND TO ALL FUEL TYPES RESULTS TO DATE:
o MET WITH NRC APRIL 6, 1989 o DETAILED CONSULTATIONS WITH MAJOR VENDORS AND CONSULTING FIRMS o A SPECTRUM OF POTENTIAL SOLUTIONS CONSISTENT WITH THE FULL STABILITY COMMITTEE GUIDANCE IDENTIFIED W RKSCOPE AND: ESTIMATED RESOURCES TO O ESTABLISH FEASIBILITY OF SOLUTION f OPTIONS DEVELOPED l o LONG TERM DEVELOPMENT AND HARDWARE COSTS ESTIMATED l
o PLAN FOR DEFINING FINAL SOLUTIONS BY PLANT TYPE ESTABLISHED FUTURE ACTIONS:
o IMPLEMENT CLOSURE PLAN l
O -
051989.HCP
- (]F CURRENT STABILITY STATUS 0 INTERIM CORRECTIVE ACTIONS IN PLACE AT ALL DOMESTIC BWRs MORE CONSERVATIVE THAN SIL-380 REY. 1 ADDITIONAL REQUIREMENTS TO SCRAM ON TWO PUMP TRIP ENHANCED TRAINING AND OPERATOR AWARENESS 1 0 INSTABILITY EVENT VERY UNLIKELY WITH RECOMMENDATIONS IN PLACE l 0 PROBABILITY OF THERMAL LIMITS VIOLATION EVEN :
LESS THAN EVENT OCCURRENCE 0 REGION OF EXCLUSION LARGER THAN REGION OF l (2) INSTABILITY OCCURRENCES INTERIM RECOMMENDATIONS HAVE GREATLY REDUCED PROBABILITY OF THERMAL LIMITS VIOLATION IN DOMESTIC BWRS, PROBABLY BELOW THE FREQUENCY OF ANALYZED OPERATIONAL TRANSIENTS.
O 051989.HCP
D LONG-TERM SOLUTION OBJECTIVES O CLOSE OUT STABILITY AS A PLANT OPERATIONAL j 1
ISSUE 0 MINIMIZE IMPACT ON PLANT OPERATION 0 RECOGNIZE PLANT TYPE DIFFERENCES AND PROVIDE SOLUTION FLEXIBILITY TO OPTIMIZE APPLICATION TO EACH PLANT l
O 051989.HCP
' ([) LONG TERM SOLUTION OPTIONS STATUS REPORT 0 PREVENTION BY REGIONAL EXCLUSION AUTOMATIC ACTION IF REGION ENTERED REGION DEFINED BY OPERATING EXPERIENCE AND ANALYSIS OPTIONAL USE OF STABILITY MONITOR TO REDUCE OR ELIMINATE EXCLUSION REGION OR INCREASE CONFIDENCE LEVEL l 0 PREVENTION BY REGIONAL EXCLUSION WITH LIMITED, CONTROLLED BYPASS ,
NORMAL SCRAM UPON ENTRY INTO EXCLUDED REGION CONTROLLED ENTRY PERMISSIVE WITH STABILITY MONITOR
() 0 MITIGATION BASED ON APRM RESPONSE AUTOMATIC ACTION FOR SPECIFIED APRM l SIGNAL TRIP POINT ESTABLISHED BY ANALYSIS OF APRM RESPONSE TO OSCILLATIONS OPTIONAL IMPROVEMENT OF APRM SENSITIVITY (I.E., NOTCH FILTER)
O MITIGATION BASED ON LPRM RESPONSE AUTOMATIC ACTION FOR SPECIFIED LPRM I SIGNAL UTILIZE APPROPRIATE NUMBER / DISTRIBUTION OF LPRMs MINIMAL ANALYSIS REQUIRED 0 INVESTIGATE CONSERVATISM OF PRESENT SAFETY L
O 051989.HCP L _ _ _ _ _ _ --_ _ __
l 1
LONG TERM SOLUTION NEAR. TERM SCHEDULE l
4/6 NRC UPDATE MEETING (COMPLETE) 4/10 WRITE BID SPECS FOR SCOPING ANALYSIS i (COMPLETE) 4/26 BWROG E0C PRESENTATION (COMPLETE) 5/18 STABILITY COMMITTEE PREPARED PLAN FOR GE WORKSCOPE DEVELOPMENT AND IMPLEMENTATION O 5/24 BWROG EXECUTIVE COMMITTEE UPDATE AND FUNDING 5/25 START WORK 1
7/1-8/21 DECISION MILESTONES (VARIOUS) 9/89 STABILITY COMMITTEE DETERMINE OPTION (S)
TO BE DEVELOPED FOR IMPLEMENTATION AND l l OBTAIN OG CONCURRENCE 9/89 NRC UPDATE MEETING O
051989.HCP L_____
4 4 .
INFORMAL ,
i COMMENTS ON NRC LETTER (3/22/89)
O l
l l
)
i l
l O
i 051989.HCP 1
D 0 BULLETIN 88-07 SUPPLEMENT 1 GUIDANCE:
ENTRY INTO REGIONS B & C (INTERIM CORRECTIVE ACTION #4)
EVIDENCE OF INSTABILITY (INTERIM CORRECTIVE ACTION #5)
O 1
I
)
I O -
1 051989.HCP
\ _ - - __--_ - _ -_ _
D INTENTIONAL ENTRY INTO STABILITY REGION O INTERIM CORRECTIVE ACTION #4 IMPLEMENTATION:
ISSUE:
UNDER WHAT CONDITIONS SHOULD ACTIONS WHICH MAY RESULT IN ENTRY INTO REGIONS OF POTENTIAL INSTABILITY BE RESTRICTED? !
RESPONSE
RESPONSE TO IMMINENT PLANT EQUIPMENT OR FUEL FAILURE SHOULD BE IMMEDIATE
() -
MOST CASES WOULD RESULT IN NO REQUIREMENT FOR SCRAM IF AS A RESULT OF AB0VE ACTION STABILITY REGION IS ENTERED, APPROPRIATE F0LLOW ON ACTION WILL BE TAKEN EXISTING EQUIPMENT / FUEL PROTECTION PROCEDURES AND TRAINING SHOULD NOT BE MODIFIED O
051989.HCP
o J
EVIDENCE OF INSTABILITY 0 INTERIM CORRECTIVE ACTION #5 IMPLEMENTATION ISSUE: )
CLARIFICATION OF WHAT CONSTITUTES " EVIDENCE
'0F INSTABILITY" NEEDED. ,
RESPONSE
LPRM ALARMS (UPSCALE AND DOWNSCALE) ARE INDICATIONS OF POSSIBLE INSTABILITY NEEDING EVALUATION EVALUATION OF APRM AND LPRM NOISE IS BASIS FOR OPERATOR ACTION TO SUPPRESS OSCILLATION l -
EVIDENCE OF INSTABILITY REQUIRES IMMEDIATE REACTOR SCRAM i
i l
i O
051989.HCP 1 L - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - _
}
' k) HRC SUGGESTED ACTIVITIES l ISSUE:
HEAT DUMPED TO SUPPRESSION POOL DURING ATWS l
RESPONSE
NO ADDITIONAL WORK IS PLANNED IN THIS AREA L 0 PHASE 1A RESULTS DID SHOW SLIGHT INCREASE IN POWER WITH PLANT IN LEVEL CONTROL (NOT ATWS ASSESSMENT BASIS)
SHORT DURATION OF INCREASE INCREASE WAS SELF TERMINATING 0 ATWS ASSESSMENTS (NED0-24222) BASED ON FLOW CONTROL MODE FLOW INTO VESSEL FIXED POWER (STEAM FLOW) INCREASES
(). ARE SELF LIMITING-(I.E., HIGH STEAM FLOW REDUCES LEVEL, CORE l FLOW & POWER)
O PHASE IA SENSITIVITY STUDIES F0R GIVEN BOUNDARY CONDITIONS POWER NOT EFFECT3E0 BY OSCILLATIONS REDUCING WATER LEVEL AT NATURAL CIRCULATION REDUCES OSCILLATION MAGNITUDE (FOR CASE STUDIED)
BOUNDING SUPPRESSION POOL HEAY LOAD IN EXISTING ATWS ASSESSMENTS NOT IMPAC1ED BY INSTABILITY O
051989.HCP
1
'O O
ATWS OVERVIEW 0 ATWS IS A BEYOND DESIGN BASIS EVENT 0 ATWS ASSESSMENTS HAVE BEEN PERFORMED 0 EMERGENCY PROCEDURES GUIDELINES DEVELOPED AND ARE BEING IMPLEMENTED O
1 l
l O
051989.HCP
.;D ATWS SCENARIO O ATWS ASSESSMENT BASIS IN NEDO 24222 (TYPICAL)
ATWS RECIRCULATION PUMP TRIP OCCURS (PLANT AT NATURAL CIRCULATION) 1 FEEDWATER PUMPS TRIPPED AT~45 SEC ECCS FLOW (FIXED FLOW SYSTEMS) INITIATED AT~1 MINUTE ;
BORON REGINS TO ENTER CORE AT~ 3 MINUTES WATER LEVEL REACHES MINIMUM AND BEGINS TO RISE AT~ 4 MINUTES HOT' SHUTDOWN AT~17 MINUTES CONTAINMENT TEMPERATURE / PRESSURE PEAK AT
~28 MINUTES
\
0 EMERGENCY PROCEDURES GUIDELINES REDUCE CONTAINMENT HEAT LOAD BELOW NEDO 24222 ASSESSMENT BY LOWERING WATER LEVEL SPECIFY ATWS ACTIONS BE TAKEN IRRESPECTIVE OF POWER / FLOW OSCILLATIONS l OS m e.NCP 1
,,--n, , - - - , - - - - - - - , - - - - - - - - - - - , - - - - , - - - - , -
O ATWS CONCLUSIONS 0 OSCILLATIONS WOULD BE SHORT LIVED AFTER BORON INJECTION BEGINS 0 AVERAGE POWER CONTROLLED BY INLET FLOW CONDITIONS O ,
POSTULATED AVERAGE POWER INCREASE SHOULD NOT IMPACT ATWS ASSESSMENT O
051989.HCP
b
! NRC SUGGESTED ACTIVITIES l
l ISSUE:
IMPACT OF INSTABILITY ON OPERATOR RESPONSE TO l ATWS
RESPONSE
! EPRI IS INVESTIGATING THIS AREA O
O 051989.HCP
40 NRC SUGGESTED ACTIVITIES ISSUE: f 1
UNDERSTANDING OF SYMMETRIC / ASYMMETRIC OSCILLATION MODES
RESPONSE
THE BWROG PLANS TO ADDRESS THIS ISSUE AS FOLLOWS:
O O WORLDWIDE OPERATING PLANT EXPERIENCE 1 WILL BE EVALUATED TO DETERMINE OSCILLATION CHARACTERISTICS 0 THE FEASIBILITY OF THE LONG TERM I SOLUTION OPTIONS WILL INCLUDE CONSIDERATION OF DIFFERENT OSCILLATORY MODES i
i l
O 051989.HCP
NRC SUGGESTED ACTIVITIES ISSUES:
MAXIMUM AMPLITUDE OF OSCILLATION AND POSSIBLE VARIATIONS IN SHAPE AND PERIOD
RESPONSE
THE BWROG PLANS TO ADDRESS THIS ISSUE AS FOLLOWS:
0 WORLDWIDE OPERATING PLANT EXPERIENCE-WILL BE EVALUATED TO DETERMINE OSCILLATION CHARACTERISTICS O
O THE FEASIBILITY OF THE LONG TERM SOLUTION OPTIONS WILL BE EVALUATED ON THE BASIS OF THEIR ABILITY TO PREVENT OR DETECT OSCILLATIONS OF DIFFERENT-SHAPE AND PERIOD AND TAKE ACTION TO LIMIT MAGNITUDE O
051989.HCP
Q p .
)
ry NRC SUGGESTED ACTIVITIES ISSUES:
0 CAPABILITY OF CODES TO PREDICT-OSCILLATION BEHAVIOR 0 CAPABILITY OF CODES T0 PREDICT BOILING TRANSITION UNDER OSCILLATING FLOW CONDITIONS iiESPONSES:
O O GE HAS PREVIOUSLY EVALUATED BOILING TRANSITION AS PART OF NEDE-22277-PA AND FOUND THE GEXL CORRELATION CONSERVATIVE O GE HAS EVALUATED THE CAPABILITY OF TRACG l
TO PREDICT BOILING TRANSITION UNDER MORE SEVERE CONDITIONS AND FINDS NO REASON TO QUESTION RESULTS L -
PLANT TRANSIENT RESPONSE LOCA ATWS 0 GE IS QUALIFYING TRACG AND THE IMPROVED FREQUENCY DOMAIN CODE AGAINST AVAILABLE OPERATING PLANT EXPERIENCE O,
0
- _ - - - - - - - _ _ - -