ML19333B830
| ML19333B830 | |
| Person / Time | |
|---|---|
| Issue date: | 10/03/2019 |
| From: | Derek Widmayer Advisory Committee on Reactor Safeguards |
| To: | |
| Widmayer, D, ACRS | |
| References | |
| NRC-0619 | |
| Download: ML19333B830 (103) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Docket Number:
(n/a)
Location:
Rockville, Maryland Date:
Thursday, October 3, 2019 Work Order No.:
NRC-0619 Pages 1-77 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005 (202) 234-4433
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com 1
1 2
3 DISCLAIMER 4
5 6
UNITED STATES NUCLEAR REGULATORY COMMISSIONS 7
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 8
9 10 The contents of this transcript of the 11 proceeding of the United States Nuclear Regulatory 12 Commission Advisory Committee on Reactor Safeguards, 13 as reported herein, is a record of the discussions 14 recorded at the meeting.
15 16 This transcript has not been reviewed, 17 corrected, and edited, and it may contain 18 inaccuracies.
19 20 21 22 23
1 UNITED STATES OF AMERICA 1
NUCLEAR REGULATORY COMMISSION 2
+ + + + +
3 667TH MEETING 4
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5
(ACRS) 6
+ + + + +
7 THURSDAY 8
OCTOBER 3, 2019 9
+ + + + +
10 ROCKVILLE, MARYLAND 11
+ + + + +
12 The Advisory Committee met at the Nuclear 13 Regulatory Commission, Two White Flint North, Room 14 T2D10, 11545 Rockville Pike, at 8:30 a.m., Peter 15 Riccardella, Chairman, presiding.
16 COMMITTEE MEMBERS:
17 PETER RICCARDELLA, Chairman 18 DENNIS BLEY, Member 19 CHARLES H. BROWN, JR. Member 20 WALTER L. KIRCHNER, Member 21 JOSE MARCH-LEUBA, Member 22 DAVID PETTI, Member 23 JOY L. REMPE, Member 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
2 DESIGNATED FEDERAL OFFICIAL:
1 DEREK WIDMAYER 2
3 ALSO PRESENT:
4 STEPHEN M. BAJOREK, RES 5
JAMES CORSON, RES 6
AMY CUBBAGE, NRO 7
RICHARD LEE, RES 9
BOYCE TRAVIS, NRO 10 KIM WEBBER, RES 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
3 C-O-N-T-E-N-T-S 1
Opening Remarks 2
by Peter Riccardella, Chair........
4 3
Advanced Reactor Computer Codes 4
Remarks by Mr. Dennis Bley, member 6
5 Briefings and Discussion 6
by Mr. Steve Bajorek, RES.........
6 7
Fuel Performance 8
by Mr. James Corson, RES
......... 46 9
Source Terms 10 by Mr. Hossein Esmaili, RES........ 57 11 Conclusion and Summary 12 by Mr. Steve Bajorek, RES......... 71 13 Adjourn..................... 77 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
4 P R O C E E D I N G S 1
(8:30 a.m.)
2 CHAIRMAN RICCARDELLA: The meeting will 3
come to order. This is the second day of the 667th 4
meeting of the Advisory Committee on reactive 5
safeguards. I am Pete Riccardella, Chair of the ACRS.
6 The ACRS was established by the Atomic 7
Energy Act and is governed by the Federal Advisory 8
Committee Act or FACA. The ACRS section of the USNRC 9
public website provides information about the history 10 of the ACRS and provides factor-related documents such 11 as our charter, bylaws, Federal Register Notices for 12 meetings, letter reports, and transcripts of all full 13 and subcommittee meetings, including all slides 14 presented at the meetings.
15 The committee provides its advice on 16 safety matters to the Commission through its 17 publically available letter reports. The Federal 18 Register Notice announcing this meeting was published 19 on September 18th. And provided an agenda and 20 instructions and for interested parties to provide 21 written documents or request opportunities to address 22 the committee as required by FACA. In accordance with 23 FACA, there is a designated federal official for 24 today's meeting. The DFO for this meeting is Mr.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
1 During today's meeting, the committee will 2
consider the following: Advanced Reactor Computer 3
Codes and NuScale certification, application, safety 4
evaluation, and also we have ACRS retreat topics 5
scheduled for today as well. And finally preparation 6
of ACRS reports. As reflected in the agenda, portions 7
of the sessions of both of these topics may be closed 8
in order to discuss and protect information designated 9
as sensitive or proprietary. There's a phone bridge 10 line. To preclude interruption of the meeting, the 11 phone will be placed in a listen only mode during 12 presentations and committee discussions.
13 We have received no written comments or 14 requests to make oral statement from members of the 15 public regarding today's session. There will be an 16 opportunity for public comment as we've set aside ten 17 minutes in the agenda for comments from members of the 18 public attending or listening in on our meeting.
19 Written comments may be forwarded to Mr. Derek 20 Widmayer, the designated federal official. A 21 transcript of the open portions of the meeting is 22 being kept. And it's requested that the speakers use 23 one of the microphones, identify themselves, and speak 24 with sufficient clarity. I would also ask everyone to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
6 silence their cell phones so we don't have 1
interruption of the meeting.
2 And the first topic is Advanced Reactor 3
Computer Codes. And I will turn the meeting over to 4
our Subcommittee Chairman, Dennis Bley.
5 MEMBER BLEY: Thank you, Mr. Chairman.
6 This is a culmination of a series of meetings we've 7
had. A while back, we had two meetings -- with --
8 subcommittee meeting where we learned about the codes 9
that they're developing. And we've had two meetings 10 with the staff on Strategy 2, which is deciding what 11 computer codes they're going to use and how they're 12 going to use them. They've given us Volumes 1, 2, and 13 3 as an introduction. And there's two more volumes 14 we're going to see at some point later in time. But 15 today we're going to hear about the first three 16 volumes. And then we're going to write a letter on 17 this one. I'm going to turn it over now to Steve 18 Bajorek.
19 MR. BAJOREK: Thank you very much and good 20 morning, everyone. My name is Steve Bajorek and I'm 21 joined here this morning with James Corson and Hossein 22 Esmaili. And our goal today is to brief the committee 23 on the work we've been doing for Advanced Reactor 24 Computer Codes.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
7 As Dennis pointed out, our topic is 1
referred to often as Strategy 2 of the Implementation 2
Action Plan. The IAP as we call it was started 3
roughly two years ago by the NRC staff in order to 4
prepare ourselves for non-light water reactors. We 5
realized at the time that we were largely a water 6
centric organization. And we had to do a number of 7
different things in order to prepare ourselves for the 8
reviews, which are imminent on us.
9 Strategy 2
involves development, 10 identification of computer codes, and the tools that 11 we're going to use if needed to do an independent 12 evaluation and analysis of the advanced non-light 13 water reactors. They key word for what we're doing 14 right now and the main objective is that of readiness 15 and preparation. We have been looking at all of the 16 different designs. And our goal is to make sure that 17 the staff has a set of tools that we'll be able to use 18 during the review of any of these types of designs.
19 Right now, our mission is to try to develop these 20 tools to accommodate all of the various plant designs.
21 And I'll show you a figure on how this has been 22 changing.
23 We consider all of these as equally likely 24 to come in for a design certification. Although we do 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
8 see a couple of frontrunners right now that we do 1
think will come in very early. This is still about 2
mile three of the marathon. And others may eventually 3
catch up and pass those up.
4 The licensing approach may or may not use 5
the Licensing Modernization Plan. So when we're 6
developing these codes, we're trying to keep a broad 7
view in that the licensing strategy that these 8
applicants may pursue may use LMP or they may come in 9
with a more deterministic type of approach.
10 Schedule is very important to us. We have 11 been informed that there are some applicants that may 12 come in for design certification or some type of 13 review as early as this December. And others may be 14 following in --
15 MEMBER MARCH-LEUBA: Sorry, 2019?
16 MR. BAJOREK: 2019, yes.
17 MEMBER MARCH-LEUBA: In two months?
18 MR. BAJOREK: Yes, two months. That 19 remains to be seen whether they stay to that schedule.
20 But every indication that it will be -- it will be 21 fairly near term. And there's others that are looking 22 at 2020 and 2021, which is still fairly short.
23 The other aspect in terms of preparedness 24 is that we expect an expedited review schedule. We 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
9 hope that these plants have large margins of safety.
1 We've seen from previous reviews of advanced light 2
water reactors that there's a large demand from both 3
the applicants and the commission that, that review 4
schedule be much, much shorter than it has been in the 5
past. That means for us when we're developing these 6
tools that sometimes take several years to develop and 7
become acquainted with, get used to, we have to begin 8
a lot of that work well in advance of the actual 9
application.
10 A lot of the designs are currently under 11 development and what we refer to as the landscape 12 continues to evolve. Some of the designs --
13 MEMBER REMPE: Steve, before you leave 14 that slide --
15 MR. BAJOREK: Sure.
16 MEMBER REMPE: I didn't want to interrupt 17 you, but if you'll go back. Earlier a couple of weeks 18 ago, we had a discussion about activities going on in 19 the Agency with improving Part 50 and 52. And during 20 that discussion, Member Ray brought up, he said do you 21 guys ever try really to encourage some of these folks 22 that Part 50 is really a better way to go when you 23 don't know what you're doing with the design? And not 24 to be rude to these people, but the first time, there 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
10 are going to be changes. And I know if you don't have 1
insight, it's hard to do Part 50.
2 (Off mic comment.)
3 MEMBER REMPE: But it really is -- I mean 4
even with some of our certified design, when changes 5
are made, it's expensive.
6 MS. CUBBAGE: So this is Amy Cubbage, NRO 7
staff. Way outside the scope of this meeting. That 8
said, in all of our interactions with developers, we 9
inform them on all the available options. And we do 10 advise on pros and cons. Ultimately it's their 11 decision of which path they want to pursue. But I 12 agree with you, there's certain attractiveness to Part 13 50 for new technologies. Anyway, sorry.
14 MR. BAJOREK: Okay. No, it's a good 15 point. And as we look at the designs, some are 16 relatively mature. We have a pretty good idea of what 17 the design is going to be, it's power level. Some of 18 the designs are not far removed from cocktail napkins.
19 They're still working on that. And we expect those to 20 change and mature in the years coming.
21 This shows what we call the advanced 22 reactor landscape. If you haven't seen this for 23 several months, you'll notice it's changing. We're 24 seeing more emphasis on what we're terming 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
11 "microreactors." The ones that have talked to us so 1
far are cooled by heat pipes. There are others out 2
there that may use some type of a gas as a coolant.
3 So there's diversity in designs in there.
4 The point to be made on this, we have a 5
variety of coolants. We have neutron spectrums that 6
go from thermal, epithermal, to FAST. Fuels can be 7
TRISO. It might even be an oxide type of fuel that's 8
still under consideration of a couple. Metallic or, 9
you know, perhaps the most intimidating, the liquid 10 salt fuel types of reactors. Nitrite fuels are out 11 there. So there's a diversity in the fuels.
12 Power levels go from as low as about one 13 megawatt thermal to well over 1000 megawatts. So you 14 can see that, you know, the potentiality of a review 15 may take us in the area of looking at something that's 16 on the order of a research and test reactor. And it 17 might be able to be approached on that type of a basis 18 to something that's very large -- has a very large 19 burn-up, fission product inventory that will have to 20 be treated maybe more in a conventional sense as a 21 large power reactor.
22 Submitted right now to the subcommittee 23 and available for public viewing is our four --
24 (Sound system failure.)
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
12 MR. BAJOREK: Okay, are we good to go?
1 Thank you. Okay. Strategy 2 right now has four 2
documents that have been available for review. The 3
first one, the introduction, kind of outlines the use 4
of the codes. I'll talk more about that later on in 5
the next steps. We feel that one does need some 6
significant revision. Volumes 1, 2, and 3 lay out the 7
codes and approach that we are using for plant system 8
analysis, Volume 1. Fuel performance in Volume 2.
9 Source term and consequence analysis in Volume 3.
10 We expect to follow these up with two additional 11 volumes. One, to look at licensing, site dose 12 assessment in Volume 4. Volume 5, fuel cycle-related 13 topics, those are under development. I don't have a 14 date for you of when I think those are going to be 15 complete.
16 We focused on the topics in Volumes 1, 2, 17 and 3 because we saw these as potentially needing the 18 longest lead time in order to get us ready for review.
19 The idea of each of these volumes is one, to point out 20 the specific tools that we're developing right now.
21 And that we intend to use for non-LWRs. Point out 22 technical gaps. And those can be either in how you 23 use these codes to model a certain type of accident 24 scenario. Features they need to have. Gaps that may 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
13 be there in the assessment. V&V type related 1
activities. And to the extent possible at this point, 2
gaps in the experimental data.
3 And those gaps can kind of come in three 4
different areas. There's data that may exist. It may 5
not be in a format that we can readily use right now 6
because, at least the U.S. data is quite old. There's 7
data that may have very high uncertainty and may need 8
to be improved upon. And there are some phenomenon 9
where the gaps and experimental data say you don't 10 have much, if anything, to go on. And we're working 11 with Department of Energy now to try to categorize 12 these and point out where we think some of those major 13 needs --
14 MEMBER REMPE: Steve, first of all, you 15 said something that's not quite true. I'm glad you 16 changed the title of Volume 1. I want to say that 17 first. But apparently, it was reissued on October 1st 18 and you changed the title. So you don't say it's 19 design basis analysis anymore, which I'm glad. We've 20 harped about that for a while. But what other changes 21 did you make? This just got issued a couple of days 22 ago. Right?
23 MR. BAJOREK: Oh, that's basically it.
24 And next steps, I'm going to talk about what we're 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
14 going to do to these volumes to come up with a new 1
revision to each one of these.
2 MS. WEBBER: But Joy, it wasn't reissued.
3 He just changed the title slide to indicate that we're 4
moving in that direction.
5 MEMBER REMPE: Well it says Rev 29, 6
October 1st, 2019.
7 MS. WEBBER: That's just the title page.
8 MR. BAJOREK: That's just a very small --
9 that's just a very small to distinguish --
10 MEMBER REMPE: So really the volumes that 11 we think we have are what we have --
12 MS. WEBBER: Correct.
13 MR. BAJOREK: Right.
14 MEMBER REMPE: -- and we should be 15 reviewing it.
16 MS. WEBBER: Correct.
17 MR. BAJOREK: So the references are going 18 to be kind of weird in our letter. Right?
19 MEMBER BLEY: Joy's vision is much better 20 than mine. I can't read the slide --
21 MEMBER REMPE: I can't read it either.
22 (Simultaneous speaking) 23 MS. WEBBER: It's a vision of the future.
24 MR. BAJOREK: Yeah. We intend to change 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
15 the title of that one. And refer to it more as plant 1
systems analysis tools as opposed to design basis.
2 Okay? We will be revising that and the other volumes 3
-- different comments that we've received from you and 4
the other stakeholders and come up with a revision.
5 But right now, what you see in those ADAMS documents, 6
that's what's available.
7 MS. CUBBAGE: Yeah and Steve if I may, I'm 8
assuming that when you changed the title to make your 9
graphic your slide, it put an auto date in there.
10 MR. BAJOREK: Yeah.
11 MS. CUBBAGE: There was no October 1 --
12 MR. BAJOREK: No, no, no.
13 MEMBER REMPE: That's makes me feel 14 better. Thank you.
15 MS. CUBBAGE: I'm sorry. But I was going 16 to interrupt you there, but you were on a role and I 17 didn't --
18 MR. BAJOREK: I'm sorry. That was --
19 MEMBER BLEY: As far as our use, we'll use 20 21 MEMBER REMPE: That's what I would think.
22 I mean 30 days is the requirement. Go ahead.
23 MEMBER MARCH-LEUBA: On the slide you've 24 got, you talk about independent analysis at the bottom 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
16 of the slide -- to which level do you consider being 1
independent from the applicant, a requirement? And 2
let me bias the answer for you. In 1960, we were 3
designing reactors within the slide rule. And 4
necessarily you do have to do very simplified 5
assumptions. And that almost required you to do a 6
separate analysis that was completely independent so 7
you could verify the assumptions. Now we're going to 8
the Monte Carlo analysis. We won't have any 9
assumptions. And if you can agree that the code is 10 good enough, what you need to review in the 11 confirmatory is the input deck and the way they use 12 it, not the code itself. So did you give any thought 13 to that?
14 MR. BAJOREK: Yeah. That's an important 15 topic because I think we'll see more of this. We have 16 talked to Technology Working Group and several of the 17 potential applicants. In some cases, they're using 18 their own tools -- legacy tools that's easier for them 19 to understand license and protect. Okay? They want 20 to keep these proprietary as well. Then when we use 21 the tools that we're using, okay, we've got that 22 independence. We're independent at least on the codes 23 that we're using. They both may be assessed against 24 the same set of data and that could be limited. And 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
17 that's also something that we're going to have to 1
consider. Have we tuned them both to the same type of 2
information? So that will be concerning.
3 The other aspect with these tools is that 4
they have become in general, much more flexible and 5
complex than what they were back when -- I actually 6
ran WFLASH at one point. Okay? It was hardwired.
7 There was only one way to do it. The only thing you 8
could change were the inputs. Now there are some many 9
switches and options and nodalization and correlations 10 that you use. It's still very easy to be independent.
11 And even when people have been using the same tool, 12 studies have shown that this group of experts in the 13 tool and this group of experts will take the same 14 information and they will get a different answer.
15 Our view right now is that we're okay with 16 that. Okay? We'll be able to be independent by us 17 taking information and doing our evaluation. Letting 18 the applicant do theirs and their way and then 19 comparing things. Because at the end of the day, 20 these tools are used, one to inform the staff, educate 21 the staff on how the machine works or should work in 22 the case of the scenario. Ask intelligent questions 23 of the applicant because it's their tool after 24 assessment and review, which becomes the analysis of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
18 record. So that's the key.
1 Right now we have not seen anyone who's 2
going to say we're going to use identically the same 3
thing that the staff is proposing. As we start to go 4
towards severe accidents, everyone will probably be 5
using MELCOR. Okay? So that concern is valid.
6 MEMBER MARCH-LEUBA: But what I'm saying 7
is do you see independence as an absolute requirement?
8 Something nice to have.
9 MR. BAJOREK: No. It's a nice thing to 10 have, but I think there's enough flexibility in what 11 we're doing in order to maintain.
12 MEMBER MARCH-LEUBA: And I support that.
13 I'm talking about biasing you in the answer.
14 Independence is not an absolute requirement.
15 MR. BAJOREK: Okay. The range of 16 capabilities that are outlined in Volumes 1 through 3 17 and the preparedness here, I like to characterize as 18 looking at the problem from two different angles.
19 One, the Volume 1 codes are really looking at adequacy 20 of the safety systems. Are we operating the machine 21 within safe limits? I used the word "safety 22 functions" as opposed to "safety systems" or an ECCS 23 system as we're used to light water reactor. Because 24 a lot of these plants are going to be utilizing 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
19 passive safety features, natural forces in order to 1
remove the energy and to shut down the reactor.
2 Regardless, the staff is going to be 3
interested to make sure that those safety functions 4
perform adequately, are the limits at which they 5
operate the reactor in terms of burn or power 6
distribution, power shifts that may occur. Are those 7
acceptable? Are the Advanced Reactor Design Criteria 8
satisfied? And in doing so, does the staff understand 9
how this can -- this machine is going to work during 10 a hypothetical accident scenario?
11 On the other end, okay if you do have a 12 severe accident, the important things that need to 13 come out of the Volume 3 methods are first of all, 14 when is that fission product inventory? Some of these 15 plants are going to have very high powers. They may 16 utilize a breed-and-burn type of operation where the 17 fission product inventory can be large. There may be 18 a large source term. Looking at what the source term 19 is. Where the fission products go. And the dose is 20 looking at that from the opposite end.
21 Volume 2 is sort of in the middle. And 22 that's kind of good because if we have problems, we 23 can always point to Dr. Corson over here and blame 24 things on him. Because we're both going to be looking 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
20 at Volume 2 to help us establish the initial 1
conditions. What is the initial fuel temperatures?
2 What are the initial average fuel temperatures -- peak 3
fuel temperatures, power distributions, oxide 4
distribution should that occur in one of these types 5
of systems.
6 We want to make sure we're consistent with 7
material properties so that we're using the same type 8
of thermal conductivity degradation in both of these 9
tools. Because we'll frequently have our own fuel 10 elements, either in MELCOR or the Volume 1 tools.
11 Those should be very closely represented by a tool 12 that can accurately show what those are.
13 And
- finally, what are the failure 14 mechanisms? What are we looking for when we do the 15 Volume 1 type analysis? Are we interested at 16 temperatures at which the TRISO particles may start to 17 come apart or other failure mechanisms that start to 18 release fission products and go to the fission product 19 inventory.
20 The role of the NRC codes with non-LWRs as 21 part of the review, this remains to be defined. Okay?
22 We're trying to be very general at this point. Be as 23 generic as possible not only with the application of 24 these codes, but also to the potential reviews that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
21 will occur. And what these codes will be needed for.
1 We look at those as being based on the applicant's 2
submittals. What type of a licensing strategy that 3
they want to pursue. The perceived safety margin. We 4
think it's going to be large. The applicants say it's 5
large. We hope that's what that is. But that's going 6
to be up to us to help verify.
7 MEMBER BLEY: Excuse me. Somebody on the 8
phone needs to mute their phone. We're getting noise 9
from you.
10 MR. BAJOREK: Okay. And finally, what are 11 the needs of the User Offices? We may have different 12 requests from NRR, NRO, NMSS as we start to look at 13 fuel cycle. We are starting to see some indication of 14 what might occur for those which are using a Licensing 15 Modernization Project, going along that plan. And you 16 know, we say some of them will probably go along this 17 path. And what's kind of interesting is the points 18 that they're reporting in these pilot studies kind of 19 come in two different categories. We're seeing a 20 large number of them concentrate on what I'll call the 21 Y-axis. Okay, where they're saying that there's no 22 dose. There's no source term. There's lots of safety 23 margin. You don't have any kind of release.
24 And what we see coming out of this are 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
22 questions that involve the safety systems, those 1
systems which may or may not be safety significant as 2
credited to the analysis. And verification of that 3
margin in order to ensure that, that point is really 4
on the Y-axis. And that you don't fail an additional 5
barrier. And that point move over to the right close 6
to this not an acceptance curve, but the frequency 7
consequence target. There will be other cases out 8
there --
9 MEMBER PETTI: Steve, can I just ask a 10 question? I understand your thinking on this. So 11 here's a case, you know, there's no release. Okay?
12 Let's say the fuel does release something, but the 13 system doesn't, you know, open up and so a valve 14 doesn't open. It stays below a set point. In your 15 thinking of the confirmatory, would you just then look 16 at that and say well all I have to do is the thermal-17 hydraulics as the first cut. And if I can show the 18 pressure doesn't leave the relief valve, then I'm good 19 enough? Or will you guys go and also try to do the 20 fission product release as well? Because you know, 21 they're acceptable, wherein light water reactor, they 22 may not be. So have you thought about how you're 23 going to really slice that up?
24 MR. BAJOREK: We think that using the LMP, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
23 you do enter a potentially grey area where there is 1
potentiality that the systems analysis codes, they may 2
say sorry, we have large uncertainty in the 3
performance of the safety system. It's large enough 4
that we feel that looking at the source term and 5
tracking the release is very much warranted. And then 6
of course, we would do that to full source term type 7
of calculation using MELCOR. I think the real answer 8
is going to come in how much of that margin is really 9
there, okay? And --
10 MEMBER MARCH-LEUBA: By margin, you mean 11 Y-axis or X-axis? Frequency or consequence and don't 12 tell me both.
13 MR. BAJOREK: Possibly both. Possibly 14 both. For those on the Y-axis, it's going to be more 15
-- it will be margin to whatever that next barrier or 16 barriers happens to be. We would anticipate the 17 questions.
18 MS. CUBBAGE: I wanted to add something 19 here. Are you speaking more from the perspective of 20 what the staff will do or the applicants or both?
21 MEMBER PETTI: The staff --
22 MS. CUBBAGE: What we will do. So I think 23 given that the staff tends to explore sensitivities 24 and explore margins and maybe go a little beyond, we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
24 may want to look at additional studies, even if it's 1
shown that the system is intact. But I mean I would 2
expect that applicants are going to truncate and say, 3
okay we haven't breached any barriers. You're not 4
going to feel a full blown -- you could stop.
5 MEMBER PETTI: Right, but this is a 6
multidimensional problem. Right? I mean you could 7
decide just to exercise the system code --
8 MS. CUBBAGE: You could.
9 MEMBER PETTI: -- convince yourself you 10 understand the --
11 MS. CUBBAGE: You could.
12 MEMBER PETTI: -- reactor.
13 MS. CUBBAGE: In a perfect world, that's 14 the end.
15 MEMBER PETTI: Right, that could be the 16 end. As opposed to then going in this other dimension 17 and parameterizing all the fuel behavior, the fission 18 product release.
19 MS. CUBBAGE: Right.
20 MEMBER PETTI: I mean, you know, this is 21 all in the context of the quicker licensing. You 22 know, you've got to think about -- think about things 23 in a different way in what you're going to do and what 24 you're not going to do.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
25 MS. CUBBAGE: I think typically we'd be in 1
where your thinking is. But then if we are seeing 2
that there's not a lot of margin, then we might start 3
to explore.
4 MR. BAJOREK: Yeah, I think we're going to 5
-- it's going to depend a lot on the submittal and 6
where we see that fitting in. And as I said, there 7
may well be that grey area where we're going to have 8
to do both in order to ensure that there is the public 9
safety. And that we're confident that the system is 10 working the way it is.
11 MEMBER MARCH-LEUBA: Even for advanced --
12 MEMBER PETTI: I just think, you know, 13 you're going to have to really recalibrate your 14 thinking in how your approach a problem. Right?
15 Because it's not like a light water reactor at all, 16 any of these.
17 MR. BAJOREK: And they're each going to be 18 different in and of themselves.
19 MEMBER PETTI: Right, yeah.
20 MR. BAJOREK: And so it's almost, you've 21 got to, you know, handcraft the approach to the 22 technology.
23 MEMBER PETTI: And that's what I was sort 24 of getting at in the earlier slide. That you may have 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
26 some that look much like a research and test reactor.
1 Okay, there's very, very little burn-up. There's huge 2
margins. You may be able to limit the review in 3
looking at a certain class of accidents. There's 4
going to be others, maybe out here by the case here 5
that I label as a non-zero dose where you know, say 6
example in a gas-cooled reactor, you're going to have 7
some circulating activity. You may have graphite 8
dust. You've got a high initial system pressure.
9 In some of those scenarios, yes, you are 10 going to have a
significant consequence and 11 potentially a dose. In that case, the question's 12 going to be what are the uncertainties one, in 13 evaluating with that source term, okay, and its 14 consequences. That sort of uncertainty on the Y-axis.
15 And what is the uncertainty in my PRA, which is on the 16 Y-axis? Because the staff is going to be concerned 17 that in that little square box that's drawn there, is 18 that star closer to the lower left-hand corner or 19 potentially over in the upper right-hand corner?
20 Because it's more frequent than we anticipated.
21 The consequences are worse such that it 22 now becomes a risk significant type of scenario. And 23 that's probably one where if we see things like that, 24 I would anticipate that the focus of the activities 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
27 then goes on source term consequence and dose. We may 1
not be as concerned on the systems analysis at that 2
point except for other types of scenarios.
3 MEMBER MARCH-LEUBA: But even for advanced 4
LWRs that we're seeing right now, all of this is 5
coming out not only on the Y-axis, they're coming on 6
the white area.
7 MR. BAJOREK: Right.
8 MEMBER MARCH-LEUBA: Basically if you 9
believe that, then you don't need to do any 10 confirmatory calculation.
Because there's no 11 probability or frequency, and no consequence. How 12 much confidence do you have that the analysis that the 13 applicant performed is not really a box like the one 14 you're showing on the document.
15 MR. BAJOREK: That's where the challenge 16 is going to come in the review.
17 MEMBER MARCH-LEUBA: Is there a way to do 18 a simplified cheap calculation, bounding -- more like 19 in Appendix K, there's some best testament without 20 having going to into a Level 3 BRA with 150,000 21 milligrams?
22 MR. BAJOREK: If you hold that thought, 23 I'm going to be talking -- What we're going to do is 24 I'm going to talk a little bit about Volume 1, then 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
28 we'll do Volume 2 and Volume 3. I'm going to show an 1
example in how we may be able to step into that type 2
of a --
3 MEMBER MARCH-LEUBA: I'm not against 4
cheating. I mean if you can do a simple calculation 5
of this, which says for sure, it's not graded on this.
6 I don't know how low it goes, but it's not graded on 7
this.
8 MR. BAJOREK: We are not opposed to doing 9
a hand calculation if that's sufficient. In fact, I 10 think it was a couple of weeks ago, we gave one of the 11 subcommittees a report on GSI-191. And one of the 12 sections in the back, how much entrainment do you need 13 to show that you aren't going to precipitate? That's 14 a hand calculation. Okay and if we can do that, we'll 15 certainly go ahead and do it.
16 Let's go on then unless there are -- any 17 more questions on this? If that's the case, what 18 we'll do is we'll start --
19 MEMBER KIRCHNER: I mean just an 20 observation, not a question, Steve. If you look at 10 21 CFR 52, which a number of people are considering as 22 the regulatory basis for applying, no matter the 23 option you pick, basically you make the assumption of 24 a major hypothetical accident. And then you do a dose 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
29 assumption subsequent to release. So I think that's 1
the back of the envelope simplified assumption to 2
bound a problem when there's large uncertainty or the 3
PRA isn't mature enough. Yesterday we had a full 4
presentation on PRAs for these applications.
5 MS. CUBBAGE: I'd like to address --
6 MEMBER KIRCHNER: A long list of all the 7
issues in terms of uncertainly and how mature the PRA 8
is as you enter into this kind of approach if you're 9
using the LIM option.
10 MS. CUBBAGE: I'd just like to address 11 that for a moment. The regulation that you're 12 speaking to, while it doesn't explicitly say it's 13 applicable only to LWRs, we're exploring whether 14 advanced reactor applicants would need exemptions to 15 that.
16 MEMBER KIRCHNER: I would guess they 17 would.
18 MS. CUBBAGE: Yes. Okay, so I just wanted 19 to make sure it was clear that it's not 100 percent 20 decided that, that regulation and the way it's worded 21 would apply. Because while it doesn't say LWR only, 22 the language is very LWR-centric talking about 23 demonstrable leakage into a containment, et cetera.
24 So it's not directly applicable.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
30 MR. BAJOREK: Okay, what I'm going to do 1
next is talk very briefly about the plant system 2
analysis codes that we have discussed in Volume 1.
3 We've researched a very large number of codes; USNRC, 4
some international codes. And tried to select ones --
5 a limited set that will allow us the flexibility to do 6
a very broad number of designs. And we've decided 7
that essentially a combination of some of the NRC 8
tools and tools primarily by NEAMS because they focus 9
more on Gen IV systems in their applications. This 10 one appears to be one that would give us the 11 capability for these broad number of designs and meet 12 the schedule that we anticipate.
13 Volume 1 documents what we can BlueCRAB 14 suite of tools. That's what we're calling it. And 15 that's just a pallet that shows you everything that 16 potentiality would be used. We wouldn't use all of 17 those for a particular design or evaluation model.
18 It's a limited subset. And I'm going to step through 19 an example on that.
20 The main purpose of Volume 1, as well as 21 the other volumes are to try to identify what are the 22 technical gaps? What do we need to resolve here over 23 the next year or two so that as these applications 24 come in and we get in to the review, that we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
31 understand what needs to be resolved. We've based 1
this on review of available PIRTs where experts have 2
gone and looked at each one of these systems for 3
plausible scenarios -- accident scenarios. What are 4
the physical phenomenon and what are the features that 5
you need to be able to analyze to in order to get a 6
reasonable answer? So we've used those as the basis 7
to identify these gaps in terms of the modeling, 8
experimental data, and V&V.
9 We feel that the codes, which are outlined 10 in BlueCRAB are applicable to all the expected 11 designs. And we outline ten unique design types that 12 are potentialities at this point. I'm going to step 13 through one we're working on right now. And as of 14 yesterday, we are running in-house on a microreactor.
15 The topic has come up on a few times on complexity.
16 And we understand, you know, the concerns on here.
17 Because I think in the developer's for some of these 18 tools like to show how detailed they can be. It's a 19 feat of software engineering. But it's not something 20 that we have to use as part of the review. So when we 21 develop these models, we do not expect to develop high 22 resolution models where you're modeling every pellet 23 as you've some of the CASTLE tools being used for. We 24 would expect to do things where we only put the detail 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
32 where that is needed. That might be the hot spot, 1
either in the vessel or the fuel, depending on what 2
that technical or regulatory issue is. So the guiding 3
principle is that of KISS, Keep it simple, Steve. And 4
then add the detail as -- I have to be careful when in 5
the record here -- But add the detail only as the 6
regulatory issue or the technical concern demands it.
7 If there is something that gives you lots 8
and lots of margin, you don't need that geometric 9
detail. If there's lots and lots of margin, it also 10 makes life simpler because there's some physical 11 phenomenon that might be very difficult for us to 12 analyze that we may be able to be bound or use a 13 simplistic model. So that's our going in approach for 14 applying these types of tools.
15 This figure shows the tools which we are 16 working on and developing. It addresses the features.
17 It uses the MOOSE framework as a way of coupling 18 things together for those applications where tight 19 coupling is needed. The applicants in our review of 20 the PIRTs saying there are some cases of molten fuel 21 salt reactor for example, where having this type of 22 coupling is very important.
Looking at a
23 microreactor, also has its needs for a coupling. But 24 I'll show you how we can simplify this. So we would 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
33 start with this and then pare this down for the given 1
application to come up with an evaluation model.
2 MEMBER MARCH-LEUBA: While we have this 3
slide, let me go back to a question in this area. The 4
white codes or the DOE codes --
5 MR. BAJOREK: Right.
6 MEMBER MARCH-LEUBA: -- and the reactors 7
are somehow supported by DOE and national labs. And 8
they're likely to use the same code. At which point 9
do you see the fact that they're using the same MOOSE 10 framework that you are for confirmation as a problem.
11 MR. BAJOREK: You see it in some cases, 12 but I think of the applicants we've talked to, most of 13 them are taking their own approach. Okay? There are 14 a couple which are using something very similar to 15 this. But there again --
16 MEMBER MARCH-LEUBA: Do you see a problem 17 if an applicant -- we used to call Jo's reactor 18 decided to use BISON and we use BISON too. Would that 19 be okay?
20 MR. BAJOREK: If we need to go into that 21 detail. One of the -- One of the avenues we have --
22 there's two fuel performance codes on there; BISON and 23 FAST. And we're still working to get FAST coupled in 24 with the MOOSE framework. If someone wants to use 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
34 BISON, we can go and use FAST. If someone uses --
1 MEMBER MARCH-LEUBA: Let me just give away 2
my final -- I don't think there's a problem with using 3
the same code as long as the code is validated and 4
hopefully is. Ninety percent of the sources of error 5
is not the codes, it's the user.
6 MR. BAJOREK: It's the user.
7 MEMBER MARCH-LEUBA: So just having a 8
different user. You've seen in the past, applications 9
where we use two different codes. One at NRC and one 10
-- And we take the same today and convert it.
11 (Simultaneous speaking) 12 MS. CUBBAGE: I'd like to chime in on a 13 licensing perspective. We also use applicant's code 14 and applicant's input data and we exercise them and 15 make sure we understand how they work and we do 16 sensitivities. So that is definitely a viable option 17 for the licensing office.
18 MEMBER MARCH-LEUBA: We don't see a 19 problem with doing that.
20 MEMBER REMPE: I agree with you. And I've 21 heard Amy say in the past, NRC doesn't have to have a 22 validated code. But if an applicant comes in and --
23 MS. CUBBAGE: I didn't say we didn't have 24 to have a validated code. I said we may or may not do 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
35 calculations. And we are not subject to Appendix B, 1
et cetera, et cetera. But if we're going to use a 2
code, we need to have comfort that it's been 3
validated.
4 MEMBER REMPE: You have to have comfort 5
that's it's been validated. And that's I think a very 6
important point because if an applicant with Jo's 7
Reactor comes in and uses the BISON code, which has 8
not been validated for the conditions of interest and 9
NRC can't use FAST because it's not validated for a 10 molten salt something or other --
11 MEMBER MARCH-LEUBA: And the applicant 12 cannot use it here.
13 MEMBER REMPE: Yeah. And so that really 14 needs to be emphasized if we go with this approach, 15 that applicants better be careful that the code is 16 validated because this will catch us. And that 17 warning just needs to be emphasized throughout this 18 discussion.
19 MEMBER MARCH-LEUBA: And the point I'm 20 trying to make is there are two issues. There's 21 validation. There is independence.
22 MEMBER REMPE: I agree with you.
23 MEMBER MARCH-LEUBA: And if you have to 24 weigh them, you'd have 90 percent validation, 10 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
36 percent independence or 95/5.
1 MR. BAJOREK: Validation is vital. I mean 2
we're working on that. We'll be doing validation to 3
our satisfaction that we need to have our codes 4
validated to. Applicants are likewise going to have 5
to validate their tools. If they validate some subset 6
of this, the burden's on them to make sure it's 7
eventually satisfactory to the staff.
8 MEMBER REMPE: And I know you're doing it 9
from the reports. But when you're trying to consider 10 all the field of reactor types and things, it gets to 11 be very long list of things. And boy, it would be 12 nice to have --
13 MR. BAJOREK: Yeah, it's difficult for us 14 15 MEMBER REMPE: -- prioritization.
16 MR. BAJOREK: -- to try to swallow the 17 whole hog at once.
18 MEMBER MARCH-LEUBA: But that's where the 19 20 (Simultaneous speaking) 21 MR. BAJOREK: Right. And we're working 22 with Department of Energy. And we're pointing out 23 where the validation -- the assessment needs to be.
24 And for the designs that look like they're coming up, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
37 they're working on those. And we're working with 1
them. They're doing much of the assessment. We're 2
doing some in-house because that's how we educate 3
ourselves by setting up some of our own models.
4 MS. WEBBER: But Joy, to your comment.
5 This is Kim Webber. To your comment about 6
prioritization, so we do have a prioritization. We 7
have limited resources. And so by the very nature of 8
limited resources, we have to prioritize. And so I 9
think the introduction does characterize our general 10 approach of prioritizing those resources. And so your 11 point's well taken that we just don't -- we can't work 12 on everything. But I think what Steve's laying out 13 here is a framework that we will work on over a period 14 of time as resources become available. And as 15 applicant's identify themselves as serious.
16 MEMBER REMPE: And we've discussed this 17 before. I'm just reacting to the bullet -- or sub-18 bullet that said all plant designs considered equally 19 likely for DC. And so just kind of be careful on how 20 your characterize it is where I'm coming back with it.
21 Then also, sometimes I hear well okay, so 22 the codes haven't been validated yet but we get 23 wonderful insights when we use these multi scale codes 24 on what experiments are needed while relying on 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
38 invalidated codes for those insights could be a trap.
1 And that's a point that I think ought to be brought up 2
at the full committee meeting, which I know we've 3
mentioned it before.
4 MR. BAJOREK: That's why when we start to 5
look at assessment, we go to the PIRTs first. And if 6
the PIRT team has identified certain phenomenon or 7
processes that you need to validate against, that's 8
what we're expecting to come out of the code 9
assessment, at least first format. It's not really 10 using the tool -- the code to do a prediction. And 11 then say that was important because the code said so.
12 No, it really starts from the PIRT and hopefully the 13 code is giving you the same thing. Okay --
14 MEMBER MARCH-LEUBA: I apologize. I was 15 not in the subcommittee meeting, so I'm wasting some 16 of your time now. I'll go real fast. So I wanted to 17 bring a different subject. You're the code guy and 18 I'm a code guy. So we concentrate on fidelity or 19 demonstrate conservatism. As long as that happens, 20 you're happy. But if you talk to the Chapter 15 21 review guys, they've spent enough time, mad, arguing 22 about the failing calculation, but arguing about that 23 sequence of events. Did they assume all the failures 24 that were supposed to assume? Did the boron dilutes 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
39 in this corner and it doesn't go. And that's 1
something that we should be able to do. And that's 2
something that we end up spending more time than 3
running the calculations.
4 MR. BAJOREK: Yeah.
5 MEMBER MARCH-LEUBA: You should be ready 6
for it.
7 MR. BAJOREK: Well I guess sort of -- and 8
this is a little bit of my philosophy is that when you 9
do an analysis, you want to start off with sensitivity 10 studies and certainty calculations if you can do that.
11 And hopefully from those, they start to point to 12 certain processes and phenomena that tended to 13 dominate the calculation. We may not necessarily 14 believe the code in all cases, but that can help guide 15 the review to look at certain things. So as we start 16 to use these sensitivities and uncertainties, are 17 certainly going to be one of the things.
18 MEMBER PETTI: So Steve, have you given 19 any thought to really doing a pilot on this to run 20 front to back on a couple of concepts?
21 MR. BAJOREK: Yeah, we --
22 MEMBER PETTI: -- because a lot of this is 23
-- it's sort of at 100,000 feet. And you know where 24 the problems are where the rubber meets the road.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
40 MR. BAJOREK: We made a suggestion a few 1
weeks ago after looking at some of the LMPs that 2
because this is such a different departure from the 3
way the staff has done things in the past, the staff 4
should do a pilot study. Step through a Level 3PRA, 5
apply these tools to some design that's out there that 6
we gain information through. And step through the 7
process.
8 When the staff did the 1988 Appendix K 9
rule change to go from Appendix K to best estimate, 10 they used the CSA use study as a way of providing 11 guidance, not only for what industry was doing, but 12 also kind of to help when we came in with the 13 application later on. The reviewers always were going 14 back to that initial pilot study as a way of how they 15 should even conduct the review. So I agree. I think 16 something like that should be going on.
17 MEMBER PETTI: The only other thing I'd 18 say -- You know, you mentioned earlier about using the 19 PIRTs -- I call them the headlights to figure out 20 where you're doing.
21 MR. BAJOREK: That's a good point.
22 MEMBER PETTI: I worry that the one that 23 I was involved with on TRISO is now approaching 20 24 years old. And frankly, much of what we thought back 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
41 then, we've proven to not be correct. So somehow we 1
have to have a feedback loop. You know, because 2
you're going to look at them and I'm going to go well 3
no, well that's not right. Thinking has changed today 4
to tell us something different. So there's another 5
inter loop around here that adds complexity.
6 MR. BAJOREK: Well we very frequently --
7 We're using these PIRTs because that's what out there.
8 In a couple cases, we've done some studies on our own 9
with molten salts to get us up to speed. But we would 10 anticipate as the design comes in, we may do another 11 independent PIRT. We'll look at what the applicant 12 has to say because they're at least using the most 13 recent design information. We may do our own internal 14 one. So those PIRTs will be revisited as part of our 15 exercise.
16 MEMBER PETTI: I mean it's the one area 17 where there's actual R&D going on. And so it is now 18 in the field of reducing uncertainties.
19 MS. CUBBAGE: One thing I wanted to 20 mention is that Hossein can provide additional detail, 21 but we are -- the Office of New Reactors is 22 contracting in consultation with Research to do some 23 sample calculations using MELCOR for the three 24 technologies to run it all the way through. So we're 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
42 planning to do that in FY20.
1 MR. BAJOREK: Okay, let me run through --
2 MEMBER KIRCHNER: Can I just ask a 3
question?
4 MR. BAJOREK: Oh, sure.
5 MEMBER KIRCHNER: I should have pointed 6
this out at the subcommittee meeting. I think you 7
have this. Do you have a finite element code in-house 8
or commercial product that you have access to? Do you 9
have a license?
10 MR. BAJOREK: We use COMSOL as one. I 11 think --
12 MEMBER KIRCHNER: I was going somewhere 13 with this. And that is with the microreactors, what 14 I found, we used ABAQUS on --
15 (Simultaneous speaking) 16 MR. BAJOREK: Yeah, yeah, yeah.
17 MEMBER KIRCHNER: -- to do the detailed 18 analysis. So when you have a quasi solid state 19 reactor, then it may prove that a finite element tool 20 is the most useful way to do your bounding 21 confirmatory analysis.
22 MR. BAJOREK: MOOSE is a finite element 23 tool.
24 MEMBER KIRCHNER: Forget Moose. There are 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
43 good commercial products that are simpler to use.
1 MR. BAJOREK: You know, that gives us the 2
capability. Now I think that if I were off doing my 3
own microreactor, COMSOL would probably be the way to 4
go. Okay? Because remember, we're using MOOSE not 5
only for the finite element capability and the tensor 6
mechanics. But also to handle all the data transfers 7
between these other tools. And it's already --
8 they're available to us.
9 All right, well -- Okay --
10 MEMBER KIRCHNER: There are commercially 11 available tools that are simple to use that may be 12 just what you need to do your confirmatory analysis 13 without getting involved in MOOSE.
14 MR. BAJOREK: There are other ways to do 15 these, yes.
16 MEMBER KIRCHNER: Okay.
17 MR. BAJOREK: Now just as a quick example 18 on how we would simplify this, we are looking at a 19 microreactor. We've developed a model, worked with 20 ARGON. We're running this model now. It does not use 21 all of the tools that are available in BlueCRAB.
22 Okay? We use SERPENT to develop the cross sections.
23 That feeds the MAMMOTH code for doing the neutronics.
24 Now a microreactor is interesting because what really 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
44 shuts down is the radial and the axial expansion of 1
the core as it heats up. That gives you the negative 2
reactivity that shuts the system back down. So you 3
need to have this active coupling between a kinetics 4
tool and your tensor mechanics.
5 We put SAM on here. It's not really doing 6
a system or a hydraulic. But what we're using that 7
for is a very simple model of the heat pipe. Okay?
8 Rather than trying to model a heat pipe with a SOCKEYE 9
or another sophisticated code, we said hey, it's just 10 a superconductor. So we take a heat structure in SAM, 11 we put in a very high thermal conductivity to give us 12 the effect because one of the accident scenarios, we 13 want to look at anyway is going to be loss of a 14 shutdown -- or loss of your shutdown heat removal.
15 The heat pipes don't do anything except act as a 16 source of stored energy in that. So you know, we 17 simplify what we do in SAM to get the heat removal.
18 We understand the tensor mechanics and the neutronics.
19 I wouldn't say it was trivial to set up, but we 20 already have the model. We're up and running. We're 21 doing that right now.
22 Now do we need to use FAST or BISON on 23 this? Well at this point, we're saying no. We can 24 handle the material properties through our common heat 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
45 structures that are available in the other tool. If 1
we get to the point where we have to have the details 2
from a FAST or a BISON, we could turn on that link.
3 But in the interest of simplification and keeping it 4
simple at this point, we're not building that 5
complication yet. We'll wait for that.
6 We would expect however that some systems 7
-- and boy, the title's really got clobbered on this 8
-- they might come in with a reactor cavity coolant 9
system. We've already completed the link to MOOSE 10 with the TRACE code. So if we're looking at an RCCS, 11 typically water cooled, we can handle that with TRACE.
12 We know how to model those types of things. And we're 13 confident that TRACE has the basic assessment to 14 handle water boiling in a tube. If you have a gas 15 cooled reactor, you might use some different tools and 16 we have a Rankine Cycle. And that's where we bring in 17 TRACE.
18 So in our evaluation, we have all the 19 features that we need to deal with this broad range of 20 plant designs, which are coming in. Our approach is 21 to keep it simple. Build in the complexity as we need 22 to as we go along. And be ready for when any of these 23 review eventualities.
24 With that, unless there are any more 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
46 questions on Volume 1, I'm going to turn it over to 1
James Corson. And he is going to bring everyone up to 2
speed on fuel performance. Thank you.
3 DR. CORSON: Okay. Yeah, I just have a 4
few slides on fuel performance. I feel like I was 5
just here recently with the subcommittee. Some of you 6
weren't here, but -- some I'm going to go through kind 7
of a high level overview of what's in Volume 2. And 8
if you have any questions, of course, feel free to 9
stop me.
10 So first to start, I just wanted to go 11 over on this slide why we're doing this or generally 12 how we do a fuel review. So for advanced reactors, 13 we've put out Reg Guide 1.232 and that proposes some 14 advanced reactor design criteria for non-LWRs. And in 15 a lot of ways, they just mirror the general design 16 criteria. So of course, you know, vendors can use 17 these ARDC, or they can develop their own plant-18 specific ones. But generally what we would do when 19 we're reviewing an application is we verify 20 compliance. And for the fuels area, the main criteria 21 are no fuel failure during normal operations. And you 22 always have to be able to shut down and cool the 23 reactor. So that's what we're really verifying when 24 we look at the fuel review.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
47 And typically we use fuel performance 1
codes to help with these reviews. We don't 2
necessarily need to do this. As Steve says, sometimes 3
we can just do hand calculations or simple analysis.
4 But oftentimes it helps to have a fuel performance 5
code to look at some sensitivities or uncertainties.
6 So how do we typically support our user 7
offices, NRR and NRO? So the way we plan on using 8
FAST for non-LWRs is going to be very similar to how 9
we use it for LWR analysis. So we have two main 10 approaches that we take. The first approach is the 11 single element just looking at one fuel rod or one 12 fuel element, whatever it is for a non-LWR. And 13 making sure that the safety limits are met during 14 normal operations or an AOO or a mild transient. So 15 we don't see that changing for most of these non-LWRs.
16 We still anticipate having to do this sort of check 17 that their fuel doesn't fail during normal operations.
18 The other way we use it is in conjunction 19 with Volume 1 codes or with TRACE in the case of LWRs 20 where we're providing initial conditions for an 21 accident full-core type analysis. So again, we don't 22 see that really changing as Steve really went over in 23 his presentation. We still expect the fuel 24 performance code to support the accident analysis.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
48 Now the way that FAST interfaces with 1
other codes, typically for LWR is we just manually 2
transfer information. So we run, you know, get the 3
initial conditions and manually feed those into TRACE.
4 That may be sufficient for some of these non-LWRs. We 5
may be able to take this very simplified approach.
6 But we're also working on directly coupling the codes 7
as Steve has mentioned, in case we need a more 8
detailed analysis. Right now we don't know exactly 9
what the applicants are going to come in with, so it's 10 beneficial for us to be ready to be able to do this.
11 So right now in the fuels area, our main 12 focus is on metallic and TRISO fuels. So molten salt 13
-- I said this at the subcommittee meeting, but for 14 people who weren't there -- for the molten salt fueled 15 reactors, we're not using a fuel performance code like 16 FAST or BISON. Those are thermal mechanical codes 17 where you have some solid rod, whatever. So we're not 18 looking at that with FAST. Of course, you know, Steve 19 and Hossein are talking about what we'll do for those 20 types of fuels.
21 MEMBER MARCH-LEUBA: So how does FAST 22 calculate on this side of a fuel failure? And let me 23 bias you on this. Let's consider the reactivity event 24 section with the famous 280 calories per gram or 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
49 whatever number is today. How would FAST calculate 1
that? It relies on experiments.
2 DR. CORSON: Yeah, so I mean there's 3
different like failure criteria for these scenarios.
4 There might be like a maximum strain. So you 5
calculate a certain amount of strain in your rod. And 6
if it exceeds that, you say it failed. And that's 7
based on -- yeah, like the clad for an LWR. So again, 8
that's based on a lot of experiments. I mean you 9
could have something simpler like does your fuel melt?
10 I mean that's a little more straight forward.
11 MEMBER MARCH-LEUBA: You don't need FAST 12 for that one.
13 DR. CORSON: Yeah, you can just do a 14 simple how much enthalpy did you add?
15 MEMBER MARCH-LEUBA: That's --
16 DR. CORSON: Yeah.
17 MEMBER MARCH-LEUBA: So the point is, what 18 Dr. Rempe says all the time. You do need experiment.
19 You can have experiment with a code. And typically 20 once you have sufficient experiment on it, you end up 21 with rules of thumb like 280 calories per gram.
22 DR. CORSON: Yes, exactly. Exactly like 23 what you said. And you know for like TRISO fuel, you 24 would use something like the Weibull statistical 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
50 parameter. And again the parameters in that 1
distribution are from experiments that we find and 2
narrow down and then apply to our code.
3 So the last slide I have is just talking 4
about where we're currently at with FAST. So it's 5
underacted development. Metallic fuels, we're in very 6
good shape. We've already done some preliminary 7
assessments. There are some models that we can 8
improve. But for the most part, we're in really good 9
shape for metallic fuels and don't have too much more 10 to do.
11 So for TRISO fuels --
12 MEMBER PETTI: So can I just -- Are you 13 able to calculate the swelling and interconnection and 14 fission gas release?
15 DR. CORSON: So we have the very simple 16 model.
17 MEMBER PETTI: -- the big safety, but 18 you've at least got something in there that --
19 DR. CORSON: Yeah. So for fission gas 20 release, we just say it releases X amount. So it's 21 like 70 percent. And generally that is actually 22 pretty good.
23 MEMBER PETTI: -- when you get to a 24 certain swell.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
51 DR. CORSON: Yeah. So we do have 1
simplified models for that. That I think is the main 2
area for improvement -- to improve those fuel swelling 3
and fission gas release models. But we do have 4
simplified models in there right now.
5 So TRISO fuels we're further behind. But 6
we have a plan for how to deal with that. And 7
fortunately we have the PARFUME code to rely on that 8
has a lot of the necessary material properties, the 9
necessary models that we can use and apply to FAST.
10 We think we're in very good shape as far 11 as experimental data out there for validation. Of 12 course, EBR2 ran a large number of metallic fuel, 13 U10s, or cases, pins. So A&L has this experimental 14 database. Right now, we're looking to get access to 15 that. We think that will probably be sufficient. And 16 then we can perform additional sensitivities and 17 uncertainties to explore, you know how manufacturing 18 might affect things or so on.
19 MEMBER PETTI: I mean, you know that those 20
-- the radiations of the thermocouples.
21 DR. CORSON: Yeah. So that is something 22 I thought I'd add.
23 MEMBER PETTI: I saw that in the write-up.
24 I mean, how are you going to deal with that?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
52 DR. CORSON: It's difficult. I mean, you 1
know, so many of these things are temperature 2
dependent. So if you don't know exactly what the 3
temperature is, it places a large uncertainty on your 4
results. And I think that's how we're going to have 5
to handle it is with some uncertainty. Yeah, I didn't 6
point that out there, but yeah, I mentioned that in 7
the report.
8 And then yeah, for TRISO fuel, we have the 9
ongoing AGR fuel qualification program. So we're 10 going to rely on that quite a bit. And some of those 11 are radiations and PIE have already been completed for 12 that. So we think we're pretty good.
13 And then the last thing here, of course we 14 are still working with the BISON team. The one area 15 where we really think we can collaborate a lot is on 16 validation. So there's still -- DOE is still doing 17 their own development. But we really want to work 18 together with validation to make sure that we're 19 confident in both of these codes. Because in some 20 cases, we might want a more detailed code like BISON 21 as Steve had in his CRAB diagram.
22 So that's all I had on fuel performance.
23 If you have any questions, I'd be happy to take them.
24 If not --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
53 MEMBER PETTI: So let me ask you a 1
question.
2 DR. CORSON: Okay.
3 MEMBER PETTI: I'm been involved in the 4
AGR program. There's a chance that they'll come in 5
purely with the data and develop what I'd call 6
empirical models. And just argue that the data show 7
is the failure rate. You know, I have a model that 8
does all the thermal mechanics, but I'm not going to 9
go there because it's difficult to validate, you know, 10 these little coding layers. And they're not really 11 adding to our understanding, we've done all these 12 radiations. So you guys ought to have an option to be 13 able to go straight to the source term. Because if 14 you assume the failure rate on input, you bypass the 15 thermal mechanics. Just assume it fails at that rate 16 and go from there.
17 DR. CORSON: Yeah, and I think we do have 18 that approach.
19 MEMBER PETTI: I mean how did the simpler 20 approach, whether -- you're not going -- Because you 21 know, there's not an ARDC about fuel on HTGRs. It's 22 all about the circulating -- the activity. Right? So 23 it's all about the source term. So there's a less of 24 a reliance in that community. The whole thing is to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
54 show the fuel doesn't fail in the envelope. And 1
you've done that. And then you go from there.
2 MR. BAJOREK: Would you expect to fail 3
during operation or do you think most of those 4
failures are going to come from manufacturing defects?
5 So you may know that right off, but --
6 DR. CORSON: You know that right off. You 7
test as best you can and that statistically limits 8
you. You see zero out of how many? There's your 9
initial condition. And actually in accidents, they 10 don't think they fail anything either. You know, so 11 it's just a different mindset in terms of how you have 12 to think about it to make sure the codes have switches 13 so that you can do these simpler things --
14 MEMBER KIRCHNER: I think Dave, that's 15 good advice because in the final analysis, it's all 16 going to be the manufacturing as you suggest and the 17 quality.
18 MEMBER PETTI: Now the one thing --
19 MEMBER KIRCHNER: And what you start with 20 is going to then create your circulating inventory 21 laid out, your others -- But if you've bound the 22 problem based on experimental data, you probably don't 23 need a detailed model of the TRISO particle.
24 MEMBER PETTI: Right. The other --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
55 MEMBER KIRCHNER: And you do macro 1
calculations thermally to show that you're within your 2
database.
3 MEMBER PETTI: What about molten salt 4
cooled TRISO designs? Have you guys given any thought 5
to some of the differences of modeling that? Because 6
you said you got TRISO fuel models. But I would 7
anticipate that you might see the molten salt cooled 8
option coming in earlier than others.
9 DR. CORSON: So I'd say as far as fuel 10 performance, it doesn't change a whole lot. So you 11 have a different thing at the boundary. You have a 12 different coolant. And so you do have to consider 13 whether you get penetration of the salt into the fuel 14 that would affect things.
15 MEMBER PETTI: That's the thing.
16 DR. CORSON: And now, from what I've seen 17 from MSRE, they didn't see a whole lot of penetration 18 of salt into the graphite. But of course --
19 (Simultaneous speaking) 20 MEMBER PETTI: -- and the Chinese -- you 21 should check the literature. The Chinese have done a 22 tremendous amount on this infiltration of the salt 23 into the pebble -- into this porous graphite. I think 24 it's a problem that the designer has to -- the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
56 applicant's going to have to deal with, whether it be 1
another coating on their pebble or something. But 2
it's something you guys ought to be aware of.
3 DR. CORSON: Yeah.
4 MEMBER PETTI: Because that would be a 5
really difficult model.
6 MEMBER KIRCHNER: And is that database 7
available?
8 MEMBER PETTI: I've seen it -- I've seen 9
the stuff published.
10 MS. WEBBER: Dave, it would be helpful if 11 you could just point us in the right direction.
12 MEMBER PETTI: I just in general looked at 13 materials. There's a number of those.
14 DR. CORSON: Yeah, we're generally aware 15 of the issue. I haven't seen that particular --
16 MEMBER PETTI: There's a couple papers out 17 on that.
18 MEMBER MARCH-LEUBA: We have a limit of 19 the staff that wants to contribute?
20 MR. TRAVIS: Yes, so this is Boyce Travis, 21 NRO. I want to correct something that was said 22 earlier with regards to ARDC on fuel for MHTGR. I 23 mean there is ARDCs related to fuel for MHTGR, 24 specifically talking about
- SARDLs, specified 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
57 acceptable radionuclide design limits. And that could 1
include the reactor coolant boundary. It could 2
include just the fuel boundary. It would depend on 3
the approach the designer takes.
4 Separately from that, we have a topical 5
report in-house from EPRI for qualifying TRISO fuel.
6 And I think it sort of lays the foundation for the 7
approach that Member Petti was talking about with 8
regards to using the data set on a more empirical 9
approach. And that's something we're fully prepared 10 to pursue, depending on the outcome of that report.
11 MR. ESMAILI: Okay, good morning. So we 12 talked a lot of about source terms. And I'm just 13 going to actually talk about source terms here.
14 This is just a very summary of what we 15 talked about on May 1st. And I only have three 16 slides. So I'm keeping it simple, Hossein. So when 17 we was talking about the source terms, we are actually 18 considering radioactive release of materials from an 19 overheated fuel, either into the containment or the 20 environment. And so what you're concerned about is 21 what is the timing -- the duration of this release?
22 What is the chemical form of the radionuclides? And 23 what is the magnitude of this source term? Because 24 that has quite a bearing on, you know, what the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
58 consequences are.
1 The importance of this source term is well 2
known. And we have talked about this over the past 3
year. And we use it for safety and environmental 4
reviews. For LWRs, we have a Reg Guide 1.183. It 5
provides a good summary of, you know, what the 6
regulations are and how to do different things in 7
terms of controlling equipment qualifications, et 8
cetera. So when we do source them, how we evaluate 9
the source term has evolved over the years. Now we 10 actually have best estimate computer codes that can 11 mechanistically model the release and track the 12 movements of this radionuclides that goes from the 13 primary system into the containment and finally into 14 the environment.
15 Of course we rely on experiments because 16 we have to validate that model. We have to have some 17 confidence that the code is doing what we expect. And 18 the risk-informed approach requires selection of 19 accident scenarios. We kind of talked about this on 20 the EPRI diagram. So that relies on the PRA. And in 21 my opinion, what LMP does is just brings the PRA up 22 front. But in terms of what you need to do in terms 23 of quantification of source term, et cetera, we are 24 just doing what we are doing for the LWRs.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
59 So in terms of characterizing the source 1
term, they are phenomenal and actually there are basic 2
modeling frameworks that are common to both LWRs and 3
non-LWRs. I talked about that on May 1st. And it is 4
here that we can leverage, you know, many decades of 5
model development, research, validation, et cetera and 6
extend it to non-LWR applications. So that's why we 7
think -- So when we approached this a few years ago, 8
we think that for source term and consequence 9
evaluation, the best course of the action is to use 10 our existing tools. And these are SCALE, MELCOR, and 11 MACCS.
12 For some technologies, HTGR, the models 13 are ready to be tested. We spent quite some time, you 14 know, ten years ago, developing these models. And 15 right now, we are ready to test these models. Of 16 course, we need data. But we just want to see how 17 these -- how this machine works.
18 So our approach -- our technical approach 19 facing framework is similar to LWRs. This is what 20 James was talking about. We do need to worry a little 21 bit about adjustments for different technologies. For 22 example when it comes to HTGRs, I have a more 23 elaborate initialization of the problem because I have 24 to do some processing of the fuel and some processing 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
60 of, you know, what is the gaps in the primary system, 1
et cetera. But the basic idea is the same.
2 SCALE, it will provide us the necessary 3
input. This is the fission product inventory, the 4
decay heat. This is going to go both into MELCOR and 5
MACCS. For MELCOR, we need you know, inventories in 6
terms of actual masses and decay heat. For MACCS, we 7
need inventories in terms of activity, et cetera.
8 MELCOR will then predict the fission product release 9
if there is one from the fuel. And the release to the 10 environment, that becomes a source term input to 11 MACCS. And then MACCS does what it does best, 12 predicts the consequences of an accident and gives us 13 health and economic consequences.
14 In Volume 3 of the report, and we've gone 15 into some details during the May 1st meeting, it 16 provides a detailed development plan. As I said, what 17 is our selection criteria? You know, we have staff 18 familiarity, domestic international use. We have a 19 large number of code users, worldwide that are 20 currently using the code. Life cycle development and 21 maintenance once we can actually use the code for both 22 LWRs and non-LWRs. So whenever we do changes, it's 23 going to benefit both technologies.
24 We talked about existing capabilities, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
61 modeling gaps. This is basically using the PIRT. But 1
you also have expert judgements. You know, we don't 2
go over everything that's in the PIRT. We just pick 3
and choose what is important. Data needs modeling 4
parameters and validation basis, these are all 5
discussed in Volume 3 of the report.
6 So what are the code attributes? I guess 7
I refer to this. So when it comes to this code suite 8
that we are planning to do, it's technology-inclusive.
9 That means I can apply it to LWRs and all the other 10 technologies. It's the same code. If I want to go 11 from an LWR to an HTGR in terms of even the input 12 deck, there are minimal input deck changes. Because 13 the code knows that either you're talking about the 14 LWR versus HTGR or SFR, et cetera.
15 It's a best estimate state of art. You 16 know, we are keeping it simple again. This reflects 17 the current scientific knowledge. This is supported 18 by experimental observation. The code is integral and 19 you have integral code because there are feedback 20 among immunological models. The current focus, and 21 this is -- I think Amy talked about this. So right 22 now, as we are -- So in the Volume 3, we discuss, you 23 know, what are the models that will be required to put 24 in. So we broke it up into FY18, 19, 20. There are 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
62 some modeling work that's still going on. But as we 1
are putting some of these models, we are also planning 2
to do entire full plant demonstration calculation.
3 This is going to help us in terms of you know, testing 4
the models, et cetera. And it involves development of 5
the input
- models, running simulations, doing 6
sensitivity calculations. And I think this is very, 7
very important. This provides, under role of this 8
modeling parameters and system interactions, I want to 9
see, you know, are there like clippage effects? You 10 know, if I choose this parameter if I don't have the 11 data in terms of effective temperature for example, if 12 I change it, am I expecting something different? You 13 know, how does that --
14 So this gives us a lot of information.
15 And the other activity we are involved in is data 16 transfer interface between SCALE and MELCOR. This is 17 data that's coming from MELCOR is going to be written 18 in a form that MELCOR can understand. So we are also 19 maintaining what we are actually doing right now in 20 terms of --
21 And so MELCOR, MACCS interface is already 22 an existing capability. So we don't have to worry 23 about that. So the source term that's coming out of 24 MELCOR goes directly into MELCOR and to MACCS.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
63 And for some of this, I can jump on. At 1
May 1st, I talked in detail about this. And this is 2
something that we are currently working on, near-field 3
atmospheric transport is exploring, you know, options, 4
et cetera. And what is the best way of treating this 5
within the MACCS scope.
6 MEMBER REMPE: I don't know if this is a 7
good time to interrupt you or --
8 MR. ESMAILI: I'm done, so you're not 9
interrupting me.
10 MEMBER REMPE: Okay. Now I wanted to go 11 back to your current focus about the full plant 12 demonstration calculations. Can you elaborate, are 13 you doing a microreactor that's a molten salt one?
14 And can those calculations help you gain insights, as 15 well as the other ones if you go through for each 16 technology what the demonstration calculation is 17 evaluating in an overall global sense. You think it 18 will give us some ideas of when you might need more 19 detailed modeling. And when you think, this is going 20 to be good enough unless --
21 MR. ESMAILI: That's the whole idea.
22 That's what I was saying about the sensitivity 23 calculations will tell you how this system works. And 24 do we need data? Do we need additional validations, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
64 et cetera?
1 So right now as a first cut, the 2
technologies that we are looking at is the 3
microreactor. We don't have -- the extent that we can 4
get design data. You know, we had a public meeting at 5
the NEI headquarters last month that we discussed if 6
the vendor's coming to get some design information, we 7
take that design information. If not, we use public 8
information. For example, megapower, the UC Berkeley, 9
you know, for the FHR. And we already have a detailed 10 model for HTGR. So we're going to build on that. So 11 I think what I --
12 MEMBER REMPE: Go slow now. You've got a 13 megapower microreactor. Is that what you said?
14 MR. ESMAILI: Yes.
15 MEMBER REMPE: And you've got a molten 16 salt, the FHR, and some sort of NGNP --
17 MR. ESMAILI: PBR-1400.
18 MEMBER REMPE: PBR-1400, okay. And then 19 I did hear you say the importance of getting data.
20 What I didn't hear you say is it could tell you 21 whether you need a finer tuned model or not. But you 22 think it will give you some insights on that.
23 MR. ESMAILI: This is part of the 24 exercise. I mean I heard all the comments. I think 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
65 right now, we are putting a lot of effort in terms of 1
making our codes ready. And as a part of this 2
process, we are learning what is important, what is 3
not important. You know, do we need experimental 4
data? What type of experimental data is important?
5 Can I do away with experimental data as you said that 6
I'm going to rely on sensitivity analysis. Is it good 7
enough?
8 So we are learning a lot of stuff. But 9
right now, we have to put everything together. And 10 this is what we are planning to do in the next year or 11 so. And see how this system works. What is the 12 source term? You know, what is going to -- you know, 13 what are the clippage effects as I mentioned before?
14 MS. CUBBAGE: I wanted to just add one 15 point. We've heard loud and clear from the committee 16 in previous meetings on the LMP citing the PRA 17 discussions yesterday that there's concern about do 18 applicants do know how to do source term? Does there 19 need to be source term guidance? So one of the duel-20 purposes of this exercise and why NRO is sponsoring it 21 is you know, we are using tax payer dollars for this.
22 We're going to use publically available information on 23 the designs, rather than a proprietary design. And 24 we're going to make information publically available.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
66 Have workshops so that we can explain to developers 1
how we've exercised MELCOR because we do understand 2
that many developers do plan to use MELCOR or similar 3
tools.
4 MEMBER REMPE: So that was the -- I'm glad 5
you said you were going to have workshops because I 6
have a lot of interactions with them as you go along.
7 And so if they bring up a valid point. Oh, we're not 8
considering this. This is a good way to influence 9
this. But I think this is a good approach and I'm 10 glad to hear about it.
11 MR. ESMAILI: So I didn't want to go into 12 a lot of details. As I said, yes, we're going to do 13 this. We're going to have workshops for NRC staff.
14 And then we're going to go public to say this how we 15 are doing things. This is what we think is happening.
16 And this is how probably -- how we are going to review 17 whatever is coming.
18 MEMBER MARCH-LEUBA: How confident are you 19 of MELCOR probabilities for molten salt? I'm thinking 20 fission product retention. Do we even have 21 experimental data for that?
22 MR. ESMAILI: When you start talking about 23 molten salt, you're talking about like FHR type of --
24 MEMBER MARCH-LEUBA: When you have molten 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
67 salt, if you release it from TRISO or from inside 1
itself, you are going to have retention of most of the 2
fission products.
3 MR. ESMAILI: So if it comes out of the --
4 If it's a pebble bed, you know, just like a pebble bed 5
and they have TRISO, we are exercising the same model.
6 (Simultaneous speaking) 7 MEMBER MARCH-LEUBA: But do you have the 8
fundamental data to know that silver doesn't get 9
retained, but iodine does?
10 MR. ESMAILI: This is something that we 11 are working on. This is something that we have 12 stressed -- you know, we are working with DOE to have 13 the information and you know, the specificity, et 14 cetera.
15 MEMBER MARCH-LEUBA: To you, you are much 16 more simplified example. Can we get a beautiful 3D 17 neutronics model versus point kinetic. But if I don't 18 know what better effective is, I cannot do the 19 calculation. Same thing, you can have a beautiful 20 MELCOR model, but if you don't know how much iodine 21 the molten salt
- retains, you cannot do the 22 calculation.
23 MR. ESMAILI: Yes. And those are some of 24 the unknowns. We know the basic framework of the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
68 model. What we don't have is the data of what's put 1
into the model. For example, when HTGRs set up, I 2
have this diffusion equation, but I don't have the 3
statistical information that goes in and says how 4
things are diffused. But these are some of the things 5
that we can give -- So we are doing all of these 6
things in peril. We are developing the models. We're 7
putting into the code. We're doing full front 8
calculation. And at the same time, do require data.
9 I think this is the same data that I think that the 10 vendors or whoever is going to -- they would need too.
11 So in LWRs, they're sharing the same data. We are not 12 going with one independent, you know, experimental 13 assessment.
14 MEMBER PETTI: Are you aware there's some 15 new data coming out on salt -- fission product release 16 from salts by the Canadians?
17 MR. ESMAILI: Yes. And I mean I don't 18 know if they published it yet.
19 MEMBER PETTI: I don't think they have.
20 MR. ESMAILI: Are you guys all linked in 21 with the GAIN workshops or the technologies? That's 22 probably how you can keep your finger on the policy.
23 MR. BAJOREK: Yeah, I think there's -- I 24 think there's a couple of NEUP studies going on right 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
69 now that are also sort of addressing that. But that's 1
a phenomena that we don't have a lot of data for yet.
2 But we know everything's working.
3 MEMBER PETTI: Well what I hear is that 4
the results do not -- are not showing ideal behavior.
5 MR. BAJOREK: Yes.
6 MEMBER PETTI: Which means greater release 7
than thermodynamic --
8 (Simultaneous speaking) 9 MR. ESMAILI: That is why we need to have 10 these codes ready. Because I was actually at CNL, you 11 know, last year and Mike Salay -- he's a colleague of 12 mine. He's involved in the PIRT. So we are aware of 13 those experiments that show potential releases. And 14 they are interested in the same thing. So we are also 15 interested. But I think what I'm saying is that we 16 are working on the basic models. Because we don't 17 think the basic modeling is going to change. It's 18 just that what are the clippage? What is the 19 temperature at which you expect this enhanced release, 20 et cetera?
21 PARTICIPANT: We're trapping.
22 MEMBER PETTI: Yeah. You need the 23 constitutive --
24 (Simultaneous speaking) 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
70 MR. ESMALI: That's right and that's why 1
we need the data, correct. So that's why I said that, 2
you know, it goes three years. You know, we're still 3
working on this. But this is providing some 4
additional information.
5 MR. LEE: This is Richard Lee from 6
Research. Dave, we have access to the CNL information 7
through the CSAC program. But I think the temperature 8
they run is very high. I told them that they need to 9
conduct additional tests to come down in temperature 10 for the releases.
11 MR. ESMAILI: Yeah, this was not a very 12 comprehensive experiment. It was very, very small 13 scale. They had these little bowls, et cetera that 14 they mixed the fuel. So we have to understand. And 15 right now, I don't want to make it, you know, one way 16 or another.
17 MR. LEE: But it's the first data that 18 I've seen on this stuff.
19 MR. ESMAILI: The first data, yes. Yes.
20 MR. LEE: So at least they're thinking in 21 the right direction.
22 MR. ESMAILI: Yeah.
23 MS. WEBBER: I think Steve has two more 24 slides.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
71 MR. ESMAILI: I'm done.
1 MR. BAJOREK: Okay, what I'd like to do 2
now is -- This is Steve Bajorek again. What I'd like 3
to do now is just sort of wrap up and talk about just 4
some overall summary and conclusions as we see it.
5 And talk about some of the next steps that we're going 6
to be doing.
7 Right now, we see Volumes 1 through 3 of 8
the Strategy 2. They document the codes, the outline 9
the analytical approach that we would use for 10 independent NRC analysis. We think it covers all of 11 the different design technologies that we're going to 12 have to face. There are uncertainties in you know, 13 constituent of models. And we know that assessment 14 needs to be done. Using the codes and the models 15 right now, at least starts to help to develop the 16 staff expertise and understand how these new designs 17 should really operate.
18 We've spent a lot of our time in 19 developing Volumes 1 through 3 and trying to identify 20 the technical gaps in terms of
- modeling, 21 experimentation, assessment. We're working right now 22 with Department of Energy to identify some of those 23 experimental data needs which are missing and need to 24 be refined.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
72 We do reserve the right to become smarter 1
as times goes on. As the designs mature and they come 2
in with new information, we may identify some 3
additional gaps that need to be resolved. So this is 4
based on what the applicants have told us right now.
5 A lot of the public information that's out there for 6
these various types of designs.
7 The scope and the depth of the analysis 8
that we would do is going to depend largely on what 9
the application is, the review strategy that the 10 applicant wants to pursue, ideas on the safety margin, 11 power
- level, burn-up, initial fission product 12 inventory, and the needs of the User Offices. So 13 those --
14 MEMBER BLEY: I'm not sure what you mean 15 by the review strategy the applicant chooses to 16 pursue.
17 MR. BAJOREK: Whether it is more --
18 (Simultaneous speaking) 19 MR. BAJOREK: Well they can come into our 20 Part 50, Part 52.
21 MEMBER BLEY: Oh, okay. If that's what 22 you mean --
23 MR. BAJOREK: I believe they can use the 24 LMP --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
73 MEMBER BLEY: -- I understand.
1 MR.
BAJOREK:
can be more 2
deterministic. We can't say we're marching down one 3
path. We need to hear from them.
4 Next steps, we're going to revise each of 5
the documents based on what we've heard from the 6
subcommittee, the full committee, your letter, and 7
from other stakeholders who are involved with -- we've 8
met with. We've met with some of the technology 9
working groups. We've met with stakeholders on a 10 periodic basis. And we have comments from them.
11 We would hope for each of these volumes to 12 come out with a Rev 1, maybe the end of the year.
13 That's going to depend on resource and personnel 14 availability. The one that needs the most work is the 15 introduction. This was put together rapidly towards 16 the end of getting Volumes 1 through 3 before the May 17 1st meeting. And it needs to better articulate what's 18 our anticipated use of the analysis codes in the 19 review and with the LMP? How do we sort of see these 20 fitting? We want to emphasize that our approach is to 21 one, right now remain flexible. Simplify our review 22 to the extent possible and right-size it to the 23 analytical -- right-size that analytical approach to 24 the needs of what that application really entails.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
74 Finally, we need to point out that these 1
codes, the analysis that we do, they educate us. They 2
help us inform ourselves on how the machine should 3
work. What we should focus on in the review. Ask 4
intelligent questions of the applicant when they come 5
in. But they're not going to be the analysis of 6
record. Now that doesn't mean that hey, we cannot 7
assess them and we just take whatever results. Yes, 8
we realize that we have to do an amount of assessment 9
-- amount of V&V that puts us at least on the level 10 playing field as the applicant. So we intend to do 11 that. But at the end of the day, it's the applicant's 12 analysis that will become the analysis of record.
13 We'll be working on Chapters 4 and 5 on 14 dose assessment and fuel cycle related topics. I 15 don't have a date for you when I think those are going 16 to be complete. In the meantime, we're going to 17 continue to work on these full plant models and 18 exercise what I call the reference plans. We'll put 19 things together that we think look and behave much 20 like the applications plan is going to be. But we'll 21 base it on publically available information.
22 The megapower for the microreactor that 23 we're running right now, it looks very much like one 24 of the applications. We have another microreactor 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
75 model that we anticipate being able to start using in 1
December. It looks like another one of the 2
applications. And this is what we'll use to try to 3
find out what are -- first of all, we try to run it 4
and from a code developer's point of view, I want 5
people to take it and break it. Okay? I want to know 6
what are the weak parts of this model, so we can 7
improve that now before the application in. But start 8
doing sensitivities to see what really has a big 9
impact on the results? So we either can refine those 10 models, refine our methodology, or improve the 11 constituent of models in order to do a better job on 12 the calculation.
13 We're both working on these Volume 1, 3, 14 and I think 2 with the stand alone models for the fuel 15 elements, which I hope to see in the near term. So we 16 can one, investigate how the codes run. Make sure 17 they're running on our computational platforms. We're 18 doing that right now with the microreactor to see if 19 there's any glitches. And be prepared so that when 20 the application comes in, we're able to take those 21 reference models, complete the assessment, and then 22 modify them to look like the applicants design so we 23 can go ahead and accommodate the User Offices as part 24 of the review.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
76 But first I'd like to thank everyone's 1
questions and comments on this. It's been brought up, 2
some of this is confusing. It probably should be 3
because these designs are new to, I think, everyone.
4 And anything you could point out in the direction of 5
new data and where we should go with these is very 6
desirable. Thank you very much.
7 MS. WEBBER: And this is Kim Webber. I 8
also want to echo the thanks. Because I think even 9
since we had the meetings with DOE, our thought 10 processes have evolved with our interactions with you.
11 And so I really appreciate the interactions, the 12 comments, the letter. And I think at the end of this, 13 you know, we're going to take the information that we 14 have from your letter and include that when we revise 15 the introduction in all parts quite frankly. So thank 16 you. I appreciate your time and dedication to doing 17 this for us.
18 MEMBER BLEY: Okay. Well thank you, the 19 staff for really good discussions here and during the 20 subcommittee meeting. Before I turn it back to the 21 chairman, I should ask for public comments? Can we 22 get the phone line open? Is there anyone in the room 23 who would like to make a comment? If so, please step 24 up to the microphone and identify yourself. Do we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
77 know if the phone line's open? Could somebody go over 1
and check and make sure they have it open?
2 Is there anybody on the phone line?
3 Please speak up and let us --
4 PARTICIPANT: It's open. This is Matt.
5 The line's open.
6 MEMBER BLEY: Thank you. Hi, Matt. Okay, 7
is there anyone on the phone line who would like to 8
make a comment? Again if so, please state your name 9
and give us your comment. Okay. With that, Mr.
10 Chairman, back to you, a full three minutes late and 11 my apologies.
12 CHAIRMAN RICCARDELLA: It's good because 13 we have another topic starting at 10:15. And --
14 MEMBER REMPE: Is the NuScale session 15 going to be transcribed? It is going to be 16 transcribed.
17 CHAIRMAN RICCARDELLA: And is it open?
18 MEMBER REMPE: It's open.
19 CHAIRMAN RICCARDELLA: Okay. Okay, so 20 with that, we will recess until 10:15.
21 (Whereupon, the above-entitled matter went 22 off the record at 10:03 a.m.)
23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
Advanced Reactor Computer Codes Stephen M. Bajorek, James Corson, Hossein Esmaili Office of Nuclear Regulatory Research United States Nuclear Regulatory Commission Ph.: (301) 415-2345 / Stephen.Bajorek@nrc.gov Advisory Committee on Reactor Safeguards Meeting October 3, 2019 RES Implementation Action Plan for Advanced Non-LWR ; Codes and Tools
Slide 2 Introduction 2
- Strategy 2 of the Implementation Action Plan (IAP) is directed at identification & development of computer codes and tools to prepare the staff for evaluation of advanced non-LWRs.
- Objective of the IAP Strategy 2 is readiness.
- All plant designs considered equally likely for DC.
- Licensing approach may, or may not, use LMP.
- Near-term submittals & shorter review schedules expected.
- Many advanced designs are under development and the landscape continues to evolve.
High-Temperature Gas-Cooled Reactors (HTGR)
Liquid Metal Cooled Fast Reactors (LMFR)
Molten Salt Reactors (MSR)
GE-H (VTR)
ARC TerraPower Westinghouse Columbia Basin Hydromine Framatome X-energy StarCore General Atomics Kairos Terrestrial Thorcon Flibe TerraPower Elysium Liquid Salt Fueled TRISO Fuel Sodium-Cooled Lead-Cooled Alpha Tech Muons Micro Reactors Oklo Advanced Reactor Landscape Stationary Transportable Others Others Westinghouse
Strategy 2 - Computer Code Reports Introduction Volume 1 Volume 3 Volume 4 = Licensing and Siting Dose Assessment Volume 5 = Fuel Cycle Topics Under Development ML19093B266 ML19093B322 ML19093B404 4
ML19246C319 Volume 2 These Volumes outline the specific analytical tools to enable independent analysis of non-LWRs, technical gaps in capabilities, V&V needs. Gaps in experimental data is currently being identified.
Slide 5 Range of Capabilities in Vol. 1-3 5
Are safety functions/systems adequate?
Are the operating limits acceptable?
Are the ARDC satisfied?
How does the machine work?
What is the fission product inventory?
What is the Source Term?
Where can the fission products go?
Volume 1 Volume 3 Volume 2 Initial Conditions Material Properties Failure Mechanisms Initial Conditions Material Properties Failure Mechanisms
Slide 6 Role of NRC Codes with non-LWRs...
6
- Remains to be defined and will be based on applicant submittals, perceived safety margin and User Office needs.
- Recent LMP pilot studies provide insight on potential review questions.
Y-axis case: No FP release, questions to involve adequacy of safety systems, verification of safety margin claims.
Non-zero dose: Source term, with questions to involve FP transport, event frequency &
their uncertainties.
Independent analysis with NRC codes develops staff understanding & expertise.
Slide 7 Volume 1 Computer Code Suite for Non-LWR Plant Systems Analysis
Slide 8 Volume 1 8
BlueCRAB represents an integrated code suite for nuclear plant systems analysis. It is a combination of NRC codes and limited number of NEAMS based codes.
Technical gaps are outlined and are based on review of available PIRTs. V&V requirements are also specified.
Flexibility: BlueCRAB is applicable to all of the expected applications - however not all codes are used for a particular design type. (See microreactor example on following slides.)
Complexity: We do not expect to develop high resolution models, or utilize detail where not needed. KISS - and then add detail as the regulatory issue or technical concern demands.
Slide 9 TRACE System and Core T/H MOOSE Tensor Mechanics, Data Transfers PARCS Neutronics SCALE Cross-sections FAST Fuel Performance BISON Fuel Performance PRONGHORN Core T/H SAM System and Core T/H Nek5000 CFD DOE Code NRC Code MAMMOTH Neutronics Comprehensive Reactor Analysis Bundle SERPENT Cross-sections Intl Code FLUENT CFD Commercial Planned Coupling Completed Coupling Input/BC Data Current View; Sept. 2019 BlueCRAB
Comprehensive Reactor Analysis Bundle BlueCRAB DOE Code NRC Code Intl Code Commercial Planned Coupling Completed Coupling Input/BC Data Comprehensive Reactor Analysis Bundle BlueCRAB - MicroReactor MOOSE Tensor Mechanics, Data Transfers SAM System and Core T/H MAMMOTH Neutronics SERPENT Cross-sections
Comprehensive Reactor Analysis Bundle BlueCRAB DOE Code NRC Code Intl Code Commercial Planned Coupling Completed Coupling Input/BC Data Comprehensive Reactor Analysis Bundle BlueCRAB - MicroReactor w/RCCS TRACE System T/H MOOSE Tensor Mechanics, Data Transfers SAM System and Core T/H MAMMOTH Neutronics SERPENT Cross-sections
Slide 12 Volume 2 Fuel Performance Analysis for Non-LWRs
NRC non-LWR Computer Code Development Plans for Fuel Performance Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission October 3, 2019
Fuel Review Methodology 14
- Reg. Guide 1.232 proposes Advanced Reactor Design Criteria for non-LWRs
- Vendors may adopt ARDC or develop their own plant-specific design criteria
- NRC fuel reviews verify compliance with fuel-related design criteria
- E.g., no fuel failure during normal operations and AOOs, ability to shut down reactor always maintained
- Fuel performance codes often used to aid fuel design reviews
Supporting Fuel Design Reviews 15
- Planned approach for using FAST will be similar to approach for LWRs
- Perform single-fuel-element standalone calculations to verify fuel safety limits are met
- Perform full-core analysis in conjunction with neutronic and thermal hydraulic tools
- FAST interfaces with other codes may be manual (current approach for LWRs) or direct code-to-code coupling
- Current focus is on metallic and TRISO fuels
- We will re-prioritize as new information becomes available
FAST Status 16
- FAST is currently under active development for non-LWR fuels
- Significant progress made on metallic fuels
- TRISO fuel models less developed, but will be ready to support licensing activities
- Most data needs expected to be met by existing database (metallic fuel) and by planned DOE programs (AGR)
- NRC staff is also maintaining awareness of BISON and its capabilities
Slide 17 Volume 3 Plans for Severe Accident Progression, Source Term, and Consequence Analysis
NRC non-LWR COMPUTER CODE DEVELOPMENT PLANS FOR SEVERE ACCIDENT PROGRESSION, SOURCE TERM, AND CONSEQUENCE ANALYSIS Office of Nuclear Regulatory Research U.S. Nuclear Regulatory Commission October 3, 2019
Regulatory Source Term
- Accident source term
- Release of radioactive materials into the containment or the environment
- Used for siting, EPZ sizing, control room habitability, and equipment qualification
- Estimate risk-informed source term
- Use best estimate computer codes
- Phenomenology (e.g., radionuclide release from fuel, aerosol dynamics)
- Experimental assessment
- Selection of accident scenarios 19
Evaluation Model Leveraging decades of physics model development and validation that can be extended to non-LWRs
- Most efficient approach to support confirmatory analysis using NRC developed computer codes SCALE/MELCOR/MACCS
- For some technologies (e.g., HTGR), the models are ready to be tested Technical approach and basic framework
- Similar to LWRs
- SCALE reactor physics analysis provides necessary input (e.g., fission product inventory and decay heat) to MELCOR and MACCS
- MELCOR accident progression analysis predicts fission product release from the fuel and release to the environment (source term input to MACCS)
- MACCS predicts dose, health effects, and economic/societal consequences
- Vol. 3 report (ML9093B404) provides a detailed development plan
- Selection criteria (e.g., staff familiarity, domestic and international use, and life cycle development and maintenance costs)
- Existing capabilities and modeling gaps (using PIRTs)
- Data needs (modeling parameters and validation basis) 20
21 Code Development
- Code Attributes
- Technology-inclusive (LWR, HTGR, SFR, MSR, FHR, HPR)
Best-estimate & State of the Art - Reflects current scientific knowledge supported by experimental observations Integral - provides feedback among phenomenological models
- Current Focus
- Full plant demonstration calculations - involves development of input models and running simulations.
Sensitivity calculations can provide insights on the role of modeling parameters and systems interactions.
- Data transfer and interface between SCALE &
MELCOR
- MELCOR/MACCS interface is an existing capability
- Near-field atmospheric transport
Slide 22 Summary & Conclusions
Slide 23 Summary & Conclusions 23
- Volumes 1 - 3 of the IAP Strategy 2 document the codes intended for independent NRC analysis of non-LWRs in support of licensing reviews. Use of these codes also develop staff expertise.
- Technical gaps; modeling needs and capabilities, necessary V&V, experimental data are outlined in the reports. Additional gaps may be identified as new design information becomes available.
- Scope and depth of the independent analysis using these tools is TBD, and will depend on the application, perceived safety margin, and needs of the User Office(s).
Slide 24 Next Steps...
24
- Strategy 2 documents (Introduction, Volumes 1-3) to be revised based on ACRS and Stakeholder comments.
Rev. 1 of each Volume expected by end of year.
- Introduction revision to better articulate...
- Anticipated use of analysis codes with LMP
- Flexibility & simplicity to right-size the analytical approach
- NRC codes and analysis are to educate staff and inform the review, and are not the analysis of record.
- Complete Chapters 4 (Licensing and Siting Dose Assessment) and 5 (Fuel Cycle Related Topics).
- Develop & exercise reference plants for non-LWRs for early insights.