ML18186A571
| ML18186A571 | |
| Person / Time | |
|---|---|
| Issue date: | 05/15/2018 |
| From: | Zena Abdullahi Advisory Committee on Reactor Safeguards |
| To: | |
| Abdullahi Z | |
| References | |
| NRC-3731 | |
| Download: ML18186A571 (100) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION OPEN TRANSCRIPT & SLIDES Title: NuScale Subcommittee Open Session Docket Number: N/A Location: Rockville, Maryland Date: 05-15-18 Work Order No.: NRC-3731 Pages 1-100 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005 (202) 234-4433
1 1
2 3
4 DISCLAIMER 5
6 7 UNITED STATES NUCLEAR REGULATORY COMMISSIONS 8 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 9
10 11 The contents of this transcript of the 12 proceeding of the United States Nuclear Regulatory 13 Commission Advisory Committee on Reactor Safeguards, 14 as reported herein, is a record of the discussions 15 recorded at the meeting.
16 17 This transcript has not been reviewed, 18 corrected, and edited, and it may contain 19 inaccuracies.
20 21 22 23 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 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 + + + + +
4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5 (ACRS) 6 + + + + +
7 NUSCALE SUBCOMMITTEE 8 + + + + +
9 OPEN SESSION 10 + + + + +
11 TUESDAY 12 MAY 15, 2018 13 + + + + +
14 ROCKVILLE, MARYLAND 15 + + + + +
16 The Subcommittee met at the Nuclear 17 Regulatory Commission, Two White Flint North, Room 18 T2B1, 11545 Rockville Pike, at 1:00 p.m., Walter L.
19 Kirchner, Chairman, presiding.
20 COMMITTEE MEMBERS:
21 WALTER L. KIRCHNER, Chairman 22 RONALD G. BALLINGER, Member 23 MICHAEL L. CORRADINI, Member 24 VESNA B. DIMITRIJEVIC, Member 25 JOSE MARCH-LEUBA, Member 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 1 JOY L. REMPE, Member 2 PETER C. RICCARDELLA, Member 3 GORDON R. SKILLMAN, Member 4
5 ACRS CONSULTANT:
6 STEPHEN P. SCHULTZ 7 DESIGNATED FEDERAL OFFICIALS:
8 ZENA ABDULLAHI 9 MIKE SNODDERLY 10 ALSO PRESENT:
11 BRUCE BAVOL, NRO 12 MICHAEL BRADBURY, NuScale 13 ALLYSON CALLAWAY, NuScale 14 TIMOTHY DRZEWIECKI, NRO 15 CHRIS KIRBY, NuScale 16 VICK NAZARETH, NuScale 17 JENNIE WIKE, NuScale 18 *Present via telephone 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 1 C-O-N-T-E-N-T-S 2
3 Chairman Opening Remarks. . . . . . . . . . . . 4 4 Overview. . . . . . . . . . . . . . . . . . . 7 5 Power Nuclear Analysis Codes. . . . . . . . . . 9 6 Code Description. . . . . . . . . . . . . . . . 13 7 Core and Reflector Description. . . . . . . . . 16 8 Code Validation and Benchmarking. . . . . . . . 19 9 Nuclear Reliability Factor Generation. . . . . . 30 10 Code Application. . . . . . . . . . . . . . . . 34 11 Critical Heat Flux Correlation. . . . . . . . . 38 12 Introduction and Background.. . . . . . . . . . 40 13 Correlation Development. . . . . . . . . . . . 46 14 VIPRE-01 Code. . . . . . . . . . . . . . . . . 57 15 Overview: Power Nuclear Analysis Codes 16 and Critical Heat Flux Correlation. . . . . . 59 17 Public Comment.. . . . . . . . . . . . . . . . 72 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 1 P R O C E E D I N G S 2 12:59 p.m.
3 CHAIRMAN KIRCHNER: The meeting will now 4 come to order. This is a meeting of the NuScale 5 Subcommittee of the Advisory Committee on Reactor 6 Safeguards. I am Walt Kirchner, Subcommittee Chair 7 today for the review of the two NuScale topical 8 reports, one TR-0616-48793, Nuclear Analysis Codes and 9 Methods Qualification, and two, TR-0116-21012, 10 Revision One, NuScale Power Critical Heat Flux 11 Correlations.
12 ACRS members in attendance today are Ron 13 Ballinger, Gordon Skillman, Michael Corradini, Pete 14 Riccardella, Jose March-Leuba and Joy Rempe. We also 15 have a former member and consultant, Stephen Schultz, 16 with us today.
17 Dr. Rempe is conflicted on matters related 18 to Studsvik scanning power that's mainly related to 19 the neutronic methods and therefore will recuse 20 herself from review of the neutronic methods topical 21 report and any associated letter writing.
22 Zena Abdullahi and Mike Snodderly are the 23 designated federal officials for this meeting.
24 Today the staff, the NuScale applicant and 25 their consultants will brief us on the content of two NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
5 1 topical reports and their applicability to the NuScale 2 design and operating conditions. The staff will 3 establish the basis for their findings on these two 4 reports.
5 We have one bridge line arranged for 6 interested members of the public to listen in during 7 the open portion of the meeting. In order to minimize 8 noise, this line will kept in mute. At the end of the 9 open portion of the meeting, we will request if anyone 10 listening would like to make further comments.
11 We have not received written comments or 12 requests for time to make oral statements from members 13 of the public regarding today's meeting.
14 A separate closed bridge line is available 15 for NRC and applicants and their consultants. And 16 please place your phones on mute to minimize 17 interference. Also state your name, the organization 18 you are representing before we commence the closed 19 portion of the meeting. That's I think, going to be 20 in reference to those who are coming in on the bridge 21 line for the contractor and the staff.
22 I request all parties to acknowledge and 23 confirm that the participants in the phone conference 24 are members of staff and/or the NuScale applicant and 25 their consultants. And that's obviously relevant to 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 1 the closed portion of the meeting.
2 As the meeting is being transcribed, I 3 request that participants use the microphones located 4 throughout this room when addressing the subcommittee.
5 Participants should first identify themselves and 6 speak with sufficient clarity and volume so that they 7 can be readily heard.
8 Let me remind you to please ensure that 9 all electronic devices have been placed in silent mode 10 to minimize disturbances during the meeting.
11 We will now proceed with the meeting and 12 I call -- oh, pardon me, Vesna Dimitrijevic has also 13 joined us for the record.
14 MEMBER RICCARDELLA: Walt?
15 CHAIRMAN KIRCHNER: Yes, sir.
16 MEMBER RICCARDELLA: I am also conflicted 17 on this topic so I will not participate in the 18 deliberations.
19 CHAIRMAN KIRCHNER: On both neutronics and 20 CHF? Both topics?
21 MEMBER RICCARDELLA: I believe so.
22 CHAIRMAN KIRCHNER: Okay. Okay. Thank 23 you. And now we will proceed. Who is going to be up 24 first? Allyson, are you starting? Or Jennie? Okay.
25 Please.
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 1 MEMBER CORRADINI: You need the green 2 light to shine.
3 CHAIRMAN KIRCHNER: Yes. And pull it 4 closer to you if you can. That will help.
5 MS. WIKE: I apologize for the brief noise 6 as I'm moving. So my name is Jennie Wike. I am the 7 licensing manager for NuScale's DCA review. So I'll 8 kick off our first presentation.
9 You know, as Chairman Kirchner mentioned, 10 NuScale is here today to present on two topics. We're 11 going to start out with the topical report on nuclear 12 analysis codes and methodology.
13 First, we're going to do some 14 introductions of who's presenting on this topic. So, 15 you know, I already introduced myself.
16 I'm the NuScale licensing manager. My 17 background, I've been with NuScale for roughly four 18 years. Previous to this, I worked at Xcel Energy at 19 the Monticello and Prairie Island Nuclear Generating 20 Plants. And prior to that, I did some R&D work 21 because I graduated from the University of Wisconsin 22 with a materials science degree and did some R&D work 23 before moving into nuclear power.
24 CHAIRMAN KIRCHNER: Does that conflict our 25 chairman?
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 1 MEMBER CORRADINI: No.
2 CHAIRMAN KIRCHNER: No. Thank you. I just 3 wanted to understand.
4 MEMBER CORRADINI: She's in the wrong 5 department.
6 MEMBER REMPE: You'll hear him say it's 7 okay since she didn't go to that department that he's 8 in.
9 MS. WIKE: But we were right next door, 10 connected by the skyway. All right. Allyson.
11 MS. CALLAWAY: So my name is Allyson 12 Callaway. I'm the supervisor of the nuclear analysis 13 group and the interim supervisor of the core form 14 hydraulics group.
15 I've been with NuScale for eight years.
16 Before that, I came from Oregon State University from 17 their nuclear sharing department. Go ahead.
18 MR. KIRBY: My name is Chris Kirby. I'm 19 a consultant with Structural Integrity Associates.
20 I've been doing work for NuScale for about five years 21 in the nuclear analysis and core design group.
22 Before that, my experience is with San 23 Onofre Generating Station for about six years during 24 core design and nuclear analysis in a nuclear fuel 25 management group. And before that I worked at Bettis 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 1 Atomic Power Laboratory for a year working on the core 2 designs.
3 And my educational background is a degree 4 in physics and a degree in nuclear engineering from 5 the University of Michigan.
6 MR. NAZARETH: And I'm Vick Nazareth. I'm 7 the director of nuclear fuel at Structural Integrity 8 Associates. My background is I worked at Combustion 9 Engineering for 10 years. And then I worked at San 10 Onofre for 25 years. I was the last director of 11 nuclear fuel at San Onofre. And unfortunately Harold 12 isn't here or he'd remember that.
13 So I've worked in the industry for 35 14 years, all in nuclear fuel, different aspects of 15 nuclear fuel.
16 MS. CALLAWAY: Okay. So with that we'll 17 get started with the open session for the nuclear 18 analysis codes and methods topical report.
19 A quick agenda. This is the agenda for 20 the open session. Some of these topics we'll cover in 21 more detail in the closed session.
22 So we'll go over an overview of the 23 purpose of the topical report and then a high level 24 brief overview of the code system that we're 25 qualifying.
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 1 We'll give a review of some of the 2 relevant aspects of the NuScale reactor design. And 3 we'll talk about the overall validation approach and 4 the benchmarking we've done as far as the topical 5 report, generation of our nuclear reliability factors 6 and some of the applications that NuScale is going to 7 use the code for.
8 So this is just a quick overview of what's 9 in the topical report. So the topical report is 10 seeking approval to use CMS5 code suite from Studsvik 11 for physics design and analysis for the NuScale 12 reactor.
13 So within the topical report, we've 14 benchmarked this code for NuScale applications and for 15 the NuScale design. With the benchmarking we've 16 developed biases and bias uncertainties and developed 17 nuclear reliability factors out of that which will 18 mostly be referred to as NRFs from here on out by us, 19 I think.
20 And then the culmination of the topical 21 reports is the validation of the code for physics 22 applications for NuScale.
23 MEMBER MARCH-LEUBA: Let me ask a 24 question, a different question. I'm asking about 10 25 CFR Appendix B, particularly when the typical paradigm 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 1 is for vendors to have their own codes, which they 2 have licensed and approved and they maintain. It's 3 not unusual for Studsvik for be part of that.
4 But here, who is going to be the holder of 5 the license? Who's in charge? Who is responsible to 6 do Appendix B on those codes?
7 MS. WIKE: Sure. So NuScale has an 8 approved QAPD topical report. And so we do have a 9 license. Though I say license, just utilizing your 10 words, we just have an approved topical report to 11 implement Appendix B at NuScale for our applications 12 across --
13 MEMBER MARCH-LEUBA: I'm more interested 14 on the codes themselves. When you're not the final 15 repository for the code, like I know, this student in 16 Nigeria finds a problem with CASMO5 and reports it to 17 Studsvik, how does that change translate into the 18 planned operation?
19 MS. WIKE: Right. So we would be notified 20 through a Part 21 notification because we would be 21 using that code.
22 MEMBER MARCH-LEUBA: So you're going to 23 get periodic updates from Studsvik? You'll always be 24 using the latest version or the version that you froze 25 in 2018?
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 1 MS. CALLAWAY: We'll be using the 2 dedicated version, the version that we've qualified.
3 And then we would requalify any new version that we'd 4 like to upgrade to.
5 MEMBER MARCH-LEUBA: So once you qualify 6 the code, it becomes your version of the code even 7 though you didn't change any code. And that's the one 8 that is approved, right?
9 And then that code derived from a 10 repository somewhere in Studsvik, that somebody found 11 a mistake on that one. And that needs to eventually 12 migrate into your version.
13 MS. WIKE: Mm-hmm. So what we're seeking 14 approval of in our topical report, and let me know if 15 this better answers your question, we're approving the 16 application of the code to the NuScale analysis.
17 MEMBER MARCH-LEUBA: But not the code.
18 MS. WIKE: But not the code itself.
19 MEMBER MARCH-LEUBA: I'm worried a little 20 bit it's a different problem, but the application of 21 Appendix B to codes that don't belong to me.
22 MS. CALLAWAY: So in general, if an error 23 is identified, we would either reconcile the error or 24 update to a newer code version where that error has 25 been resolved and then revalidate through our 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 1 dedication process.
2 MEMBER MARCH-LEUBA: I want you to say 3 that you will keep the codes up to date not -- or at 4 least you tried to for your mistakes. You have to 5 evaluate it if the change is significant. And if you 6 change it, at which point do you reevaluate?
7 MS. WIKE: Yes. Our Appendix Bravo 8 Program would require us to evaluate any errors. So 9 if Studsvik found an error in the code and they 10 communicated to that us, usually through a Part 21 11 notification, then we would evaluate it under Appendix 12 Bravo. And at that point we'd have to, as Allyson 13 mentioned, either resolve it because maybe the error 14 doesn't impact how we're applying the code. But if it 15 does impact what we're doing, certainly we would have 16 to address that.
17 MEMBER MARCH-LEUBA: Yes. I have one or 18 two reservations that when there's multiple ownership 19 that those things become difficult.
20 MS. CALLAWAY: Okay. So we'll cover those 21 codes a little bit more in the closed section. This 22 is the initial high level overview of the codes so 23 starting CASMO5, that's a lattice physics code. It's 24 using ENDF/B-VII.1. And it's the 2D transport solver 25 that develops the nuclear data in cross-sections.
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 1 The CMSLINK compiles all the cross-2 sections into a library and that library is the 3 primary input into SIMULATE5. SIMULATE5 uses the 4 diffusion equation in a 3D model basically built up of 5 the cross-sections from the library using 3D core 6 simulation.
7 So that's the general code flow. CASMO5 8 and SIMULATE5 are the latest versions of the codes.
9 CASMO4 and SIMULATE3 are fairly prominent in the 10 industry. And so these are the latest versions.
11 Studsvik has just also submitted a generic topical 12 report for us.
13 MEMBER MARCH-LEUBA: Are you already 14 making commercial use of CASMO5 approved to date in 15 the U.S.?
16 MS. CALLAWAY: Duke uses CASMO5 currently.
17 I believe they've licensed that themselves. And I 18 think that Dominion is or will be going through an 19 8311 with Studsvik's general code.
20 MEMBER MARCH-LEUBA: The way I worded the 21 question is CASMO5 on paper is infinitely superior to 22 CASMO2 for sure, right? The question is we've been 23 tweaking those old codes forever and not in just 24 special factors here and there to make them work.
25 And now when you work, like, on 5 maybe 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 1 those three factors don't apply. So I'm interested to 2 follow-up on benchmarking against reoperating 3 conditions in real plants.
4 How do they -- better code follows the old 5 plant. And obviously you're not going to have a 6 NuScale core to do that, but you should be plugged in 7 on experiences from Duke or whoever has it or European 8 experience to make sure that you don't make the same 9 mistakes.
10 DR. DRZEWIECKI: I'm sorry. This is Tim 11 Drzewiecki. I mean, that's something that will be 12 covered in our presentation. There is a generic CMS 13 fact topical report. And there is plenty of operating 14 plant benchmarking that's reflected in that topical 15 report, over 63 plant cycles, I believe.
16 MEMBER MARCH-LEUBA: Mostly foreign 17 plants?
18 DR. DRZEWIECKI: No. It's U.S. plants.
19 MEMBER MARCH-LEUBA: A U.S. plant, yes.
20 Thanks.
21 MEMBER SKILLMAN: Does Studsvik maintain 22 a user's group for their codes? And I presume if they 23 do, NuScale is a member?
24 MS. CALLAWAY: Yes.
25 MEMBER SKILLMAN: For both.
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 1 MS. CALLAWAY: Yes to both.
2 MEMBER SKILLMAN: Both of their codes.
3 Thank you.
4 MS. CALLAWAY: Okay. With that, I will 5 turn it over to Chris, and he's going to cover some of 6 the more relevant aspects of the NuScale design.
7 MR. KIRBY: So this slide just presents 8 some general parameters of the NuScale core and fuel 9 design. And as it relates to the codes, modeling of 10 the NuScale core is not much different than modeling 11 typical PWR except for the size and the power level.
12 As you can see, the core is composed of 37 13 assemblies with 16 CRAs. The fuel assemblies are 14 standard 17 x 17 design, 24 guide tubes and one 15 instrument tube.
16 The fuel has a maximum UO2 enrichment of 17 95 percent, UO2 and its encapsulated M5 cladding. An 18 active fuel length is two meters.
19 The control routes have a split 20 composition of 37 centimeters of silver indium cadmium 21 at the bottom and that's followed by B4C at the top 22 for the remainder of the rod. And then gadolinia is 23 used as a printable absorber, which is combined 24 homogeneously with the UO2 and fuel pellet.
25 And one of the features of the NuScale 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 1 design, as it relates to safety, is its resiliency to, 2 or stability with respect to xenon oscillations.
3 So analysis that we've done with SIMULATE 4 has shown when we perturb the core to cause a xenon 5 transient that the xenon oscillations are quickly 6 damped out and the core returns to equilibrium, 7 usually within one oscillatory cycle. So it's 8 different than a big PWR. It's very stable.
9 MS. CALLAWAY: If I could interrupt 10 quickly, I think you said it was 95 percent max 11 enrichment. You meant --
12 MR. KIRBY: 4.95 percent, yes. Thank you 13 for that.
14 MEMBER REMPE: If I could interrupt for a 15 second, just for clarification. When I was looking at 16 this report versus the CHF report, which clearly said 17 a particular fuel type, I didn't see a specific fuel 18 type identified. And somewhere I thought in some of 19 the material we were given there was a time where 20 NuScale didn't have a fuel or they switched from one 21 fuel to another.
22 And I was just curious, is this topical 23 for the particular fuel type that you are now going 24 with? Or what's the story on the fuel? Is this 25 topical for it?
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 1 MS. CALLAWAY: So the generic parameters 2 of the fuel type that apply, whether we are using a 3 prior fuel design or this fuel design, is the general 4 applicability of this topical report. So the 17 x 17.
5 MEMBER REMPE: So when we write, or when 6 the Committee writes a letter on this, they should not 7 reference the specific fuel type on this topical 8 report. You want it to be more general for more than 9 the AREVA fuel type that you're currently going with.
10 Correct?
11 MS. CALLAWAY: Yes.
12 MEMBER REMPE: Okay.
13 DR. DRZEWIECKI: This is Tim again. This 14 is something which is addressed in the staff's safety 15 evaluation that there is some flexibility in terms of 16 your fuel type, but there's also a limitation provided 17 in the SE in terms of there's only certain materials 18 that are inside of the CMS5 code suite. And so they 19 are limited, obviously, to those.
20 MEMBER REMPE: Thank you.
21 MR. KIRBY: So as you can see in this 22 slide, it's a depiction of the heavy reflector around 23 the core. That reflector contains cooling channels to 24 remove the heat generated by the gamma radiation. And 25 the heavy reflector for NuScale was designed to 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 1 control leakage, improve fuel utilization and evenly 2 distribute power throughout the core.
3 As can be seen in the figure, the 4 reflector thickness varies around the core and ranges 5 from about 10 centimeters to about 30 centimeter.
6 Standard reflector modeling in CASMO5 and 7 SIMULATE5 is used to model the reflector, but the 8 number of reflector sections model depending on the 9 reflector geometry.
10 Modeling multiple reflector segments based 11 upon position around the core and material composition 12 of the various segments was done to maintain the 13 fidelity of the model, keeping the distinct regions of 14 the reflector separate.
15 The ability of SIMUALTE5 to accurate model 16 the SIMULATE reflector has been demonstrated by our 17 core benchmarking, which we'll go into a little bit 18 later and specifically comparisons to MCNP.
19 And now we'll go back to talk a little 20 about the validation of the code with Allyson.
21 MS. CALLAWAY: So we used three different, 22 I guess, primary categories for benchmarking. The 23 first one was code-to-code benchmarking with MCNP, 24 empirical benchmarks of critical configurations and 25 experimental reactors and commercial reactor benchmark 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 1 to TMI1.
2 MEMBER MARCH-LEUBA: So you say you're 3 doing a full core MCNP?
4 MS. CALLAWAY: Yes.
5 MEMBER MARCH-LEUBA: Full three?
6 MS. CALLAWAY: Mm-hmm.
7 MEMBER MARCH-LEUBA: You have plenty of 8 servers to run it on?
9 MS. CALLAWAY: It takes awhile to run.
10 MEMBER MARCH-LEUBA: It takes awhile, 11 right?
12 MS. CALLAWAY: Mm-hmm. So those 13 benchmarks comprise our code validation. And the 14 results of the code validation are used to develop 15 biases and bias uncertainties.
16 MEMBER MARCH-LEUBA: Just curious, on MCNP 17 model, to what level do you homogenize? You don't 18 model every pin do you?
19 MS. CALLAWAY: Yes, we do.
20 MEMBER MARCH-LEUBA: You model every pin 21 on every fuel assembly?
22 MS. CALLAWAY: Yes.
23 MEMBER MARCH-LEUBA: I'd love to see that 24 model.
25 MR. KIRBY: Remember, it's only 37 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 1 assemblies.
2 MEMBER MARCH-LEUBA: Yes. But there is 3 the 17 x 17 pins. I know MCNP allows you to repeat 4 the limits, but, geez, you're talking dissertation for 5 if you bring it out, you get your Ph.D. just out of 6 it.
7 MS. CALLAWAY: Okay. So the results of 8 the benchmarking we used to develop biases and bias 9 uncertainties are tolerance limits as its given here.
10 And those values we develop nuclear reliability 11 factors to encompass them and the nuclear reliability 12 factors and the validated code are ready for 13 downstream application.
14 So that's the general process that we 15 followed for validation and the basic flow of the top 16 core part that we're talking about.
17 Okay. So a little bit more on code to 18 code --
19 MEMBER REMPE: I have another question on 20 that slide, too. And this is more because, again, of 21 the interface with the thermal hydraulics. NuScale is 22 going to operate at lower pressures and lower flow 23 conditions.
24 And, again, I'm not an expert what all 25 Studsvik did to validate their codes, but in all of 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 1 these benchmarks, did they consider a case that would 2 give you appropriate benchmarks that would consider 3 void coefficients and moderate greater coefficients 4 because of the different conditions that you're 5 operating at?
6 I mean, I know the benchmark of BWR 7 conditions as well as PWR conditions. But did you 8 look and say, oh, yes, this one really will give us 9 the right void coefficient and moderator coefficient?
10 MS. CALLAWAY: So the closest benchmark 11 that we have for that is the commercial benchmark.
12 MEMBER REMPE: PMI1?
13 MS. CALLAWAY: And it's 400 psi, I think, 14 lower pressure, but generally, fairly similar 15 operating conditions. Also, with the code-to-code 16 benchmark with MCNP, you know, pressure doesn't 17 necessarily draft input. But we looked at a number of 18 different densities of the coolant as part of that 19 benchmark. So we get those implicitly in a couple of 20 different ways.
21 MEMBER REMPE: Because again, when I think 22 about things we've heard so far, ATWS concerns and 23 things like that are more important. And so that's 24 why I'm real curious on that.
25 MS. CALLAWAY: And so moderator 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 1 coefficients would be similar with the benchmarking we 2 looked at.
3 MR. NAZARETH: And so let me add a little 4 bit more to that. Typically when you do a code like 5 this, am I on or off? Sorry.
6 Typically when you do a code like this, 7 you benchmark to operating data. That's the key 8 aspect of what it is, but NuScale has no operating 9 data. So the benchmarking done here that's described 10 over here is very extensive to make up for whatever it 11 is you're going to have.
12 So the MCNP, you have to go into that 13 level of detail because you have to have two codes 14 that match up perfectly to say that the benchmarking 15 was okay. Okay?
16 And then same thing with the empirical 17 benchmarks with the criticality and so on. Now TMI 18 data is the closest data we could find that will give 19 us the kinds of things we wanted to do for which 20 you're talking about. And so we used that.
21 And then ultimately we set up, we compared 22 against the industry standards just to see that we're 23 not off, that we're consistent with what the industry 24 standard was for those kinds of item.
25 And so, these in the end, we added 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 1 conservatisms to account for things we didn't know.
2 Okay? We expect to verify that during startup testing 3 in this middle half of this table over here.
4 When it says update methods, we actually 5 put the methods in and how we validate it after we get 6 actual startup data and make sure it's still 7 conservative. Okay? But we had to do much more 8 extensive benchmarking because of the fact that the 9 NuScale conditions are unique.
10 MEMBER REMPE: Thank you.
11 MEMBER SKILLMAN: Let me expand on Dr.
12 Rempe's question, please, Vick.
13 MR. NAZARETH: Sure.
14 MEMBER SKILLMAN: The first couple cycles 15 of TMI1 were at 2413, not 2568, four pumps running.
16 Very high Reynolds numbers.
17 MR. NAZARETH: Yes.
18 MEMBER SKILLMAN: And an enrichment that's 19 very different than yours.
20 MR. NAZARETH: That's correct.
21 MEMBER SKILLMAN: And early, early TMI 22 enrichments were 3.5, 4.1 max. And you're 4.95 with 23 a 24 month fuel cycle. Why is the TMI data 24 applicable?
25 MR. NAZARETH: So the TMI -- okay. So the 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 1 TMI data is applicable to the point of simulating 2 actual operating data. Okay? It could have taken 3 many different things like that. The specific things 4 with enrichment and all of that, we did the code-to-5 code benchmarking to match that up really well. Okay?
6 And the TMI data is like the last step to 7 close the ranks to make sure we're actually using an 8 operating data from a plant to match up as well as we 9 can. And then we added some more conservatism to the 10 top of that. Okay?
11 So we brought in -- there is many -- in 12 the Studsvik generic topical report, they brought up 13 data from many different plants, most of them Dominion 14 plants here today. Okay. So they had a broader based 15 look at all the plants. Okay?
16 So what we did in the end is we came 17 together with that and said, well, for whatever we 18 don't have -- because you can run infinity with the 19 amount of data that you wanted to. Whatever data we 20 don't have, we'll make certain conservative 21 adjustments to make sure we're not -- we're within the 22 realm of what we expect to be for this particular run.
23 So it's the best we got, really, to match up with as 24 many things as we could.
25 MEMBER SKILLMAN: Thank you.
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 1 MR. KIRBY: Can I clarify something? The 2 two year cycle length doesn't have a blanket 4.95 3 percent enrichment. That's just the maximum 4 enrichment from the scale. For the two year cycle, 5 the average enrichment is 3-1/2, 4 percent, something 6 right around those numbers. I just wanted to clarify 7 that.
8 MEMBER MARCH-LEUBA: But do you have 9 gadolinia?
10 MR. NAZARETH: Yes.
11 MEMBER MARCH-LEUBA: That's something that 12 broadly the very early TMI code didn't have.
13 MR. NAZARETH: That's correct.
14 MEMBER MARCH-LEUBA: That would be the 15 biggest difference.
16 MR. NAZARETH: And this is why did the 17 code to kill kind of thing, right? Because we needed 18 to be able to duplicate that we could model those 19 kinds of things. This is why we also compared it to 20 the industry data to make sure they were not stepping 21 out of things that are different. But gadolinia is 22 fairly well known now. Framatome uses it all the time 23 so it's not that unusual.
24 MEMBER MARCH-LEUBA: Going back to my 25 original comment which you probably forgot already is 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 1 this are you relying on the Studsvik valuation also?
2 MR. NAZARETH: Okay. So let me add 3 something more to that. Okay? So Studsvik is an 4 independent resupplier. So if there are errors in 5 their code, they're required to tell NuScale about the 6 errors. And then NuScale evaluates that to make sure 7 it complies with NuScale's Appendix B program. Okay?
8 So for things that are nice to have, 9 NuScale doesn't have to take them in or not. With 10 error reporting, it has to be done and correctly.
11 That is a requirement for an Appendix B supplier. And 12 there is what's called a NUPIC audit that goes back to 13 suppliers of codes and validates them.
14 MEMBER MARCH-LEUBA: I'm not picking on 15 you.
16 MR. NAZARETH: No, no, no.
17 MEMBER MARCH-LEUBA: I'm trying to 18 understand --
19 MR. NAZARETH: I'm just explaining the 20 system of our -- because many utilities now are doing 21 independent analyses, right? APS is doing their.
22 Duke is doing it. Dominion is doing it. We did it.
23 It's ours.
24 So the same kind of thing. We take over 25 whatever the vendor codes are. It doesn't have to be 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 1 a field vendor anymore. Take over the vendor codes.
2 An Appendix B supplier, they're required to provide 3 you with error reporting.
4 MEMBER MARCH-LEUBA: And then are you 5 required to validate your code independently from the 6 vendor?
7 MR. NAZARETH: You assess it and you 8 evaluate it independently, correct. You are the 9 licensee. So this gives the responsibility to 10 validate.
11 MEMBER MARCH-LEUBA: You cannot rely on 12 the validation that the vendor, Studsvik, probably 13 knows more about this than you do.
14 MR. NAZARETH: No. You do rely on it 15 because they give you all the information relative to 16 that. Right? The impacts and all of that, they do 17 the impact assessment and all of that. Okay?
18 So you apply it to yourself what is the 19 homogeneric application?
20 MEMBER MARCH-LEUBA: Well, what gives me 21 a very nice warm feeling is well, we just held there 22 are 65 cores that we have validated, that's validated 23 on their own. You've done one.
24 MR. NAZARETH: Yes.
25 MEMBER MARCH-LEUBA: So I don't have 65.
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 1 I have one. Let's degrade for it.
2 MR. NAZARETH: That's why we got all this 3 benchmarking, right? Tons of it because we try to 4 fill in as many gaps as we can.
5 CHAIRMAN KIRCHNER: Just out of curiosity, 6 has Studsvik used this code suite on boilers, on BWRs?
7 MR. NAZARETH: They have. The answer to 8 that is yes.
9 CHAIRMAN KIRCHNER: Okay.
10 MR. NAZARETH: Which ones, I'm not quite 11 sure. But the answer to that is yes. It's 12 generically --
13 CHAIRMAN KIRCHNER: So that I would guess 14 they would have benchmarked. And therefore, they 15 would have data at 1000 psi and different flow 16 conditions than typical of the PWR. So it may provide 17 some confidence that the code scale through a range of 18 pressures and flows and associated neutronic 19 parameters that are typical.
20 It's almost as if, without making any 21 judgment, it's like the NuScale design is somewhere 22 between a BWR and a PWR with your natural circulation 23 and your lower pressure. And I guess we'll hear more 24 about that when we get to critical heat flux.
25 But if Studsvik has benchmarked against 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 1 boilers, that's enhanced competence that the code 2 scales through substantially different thermodynamic 3 and neutronic conditions.
4 MR. NAZARETH: Yes. That's interesting.
5 CHAIRMAN KIRCHNER: Okay.
6 MS. CALLAWAY: Okay. So just a bit more 7 detail on the code-to-code benchmarking to start. As 8 we have said, the code-to-code benchmarking is CMS5 9 compared to MCNP6. And the benefit here is that we're 10 comparing two different methods of deterministic 11 versus a stochastic method of very detailed MCNP6 12 model.
13 Because it's a code-to-code benchmark, 14 we're not relying on a fixed number of data points.
15 We have flexibility in the number of data points that 16 are available to us. So most of the physics 17 parameters that are important to us for the validation 18 we have the ability to benchmark through this method.
19 We can move on to the next slide unless 20 anybody has any questions on that one. So the 21 empirical benchmarking, the commercial reactor is 22 included in this but we did do a large amount of 23 critical experiments and experimental reactor 24 benchmarking as part of our empirical benchmarking as 25 well.
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 1 And the empirical benchmarking and the 2 commercial reactor benchmarking gives us real data to 3 compare against a limited number of data points and 4 still a substantial amount, we have actual measured 5 data that we start to get to compare to.
6 MEMBER MARCH-LEUBA: What I don't see on 7 that slide is core depletion. I mean, it feels like 8 you consider it more to a particular space, the 9 operating condition, and calculated it to power 10 distribution. Did you do the core depletion to match 11 the K effect as a function of time?
12 MS. CALLAWAY: Yes, yes. We get that out 13 of the TMI benchmark.
14 MEMBER MARCH-LEUBA: That's the TMI?
15 MS. CALLAWAY: Mm-hmm. Yes. So there's 16 two cycles of core deletion.
17 MEMBER MARCH-LEUBA: That's much harder to 18 do than a single point.
19 MS. CALLAWAY: Yes. And so commercial 20 data benchmarking, we've covered this quite a bit 21 already. So TMI, you know, on cycles 1 and 2 are 22 modeled. We get all of our key physics parameters 23 benchmarked through this commercial reactor 24 comparison.
25 So the commercial benchmarking, the other 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 1 empirical benchmarking, the code-to-code benchmarking, 2 as a recap, we use the results of those in combination 3 with developed biases and bias uncertainties and use 4 those to develop NRFs for NuScale's application of the 5 code.
6 And then the final recap of the process.
7 I guess, I mostly just said all of this. So once the 8 NRFs are developed, we use those for the initial 9 NuScale design. And as soon as we're starting up, we 10 have the ability to validate those NRFs and then start 11 collecting additional data to either confirm or update 12 the NRFs once we have NuScale's specific data 13 available to us.
14 MEMBER MARCH-LEUBA: And these are 15 penalties when they are effective? Give me, like, an 16 example of an NRF.
17 MR. NAZARETH: Can we talk about that more 18 in the closed session? Because we're actually 19 presenting some of those things there.
20 MS. CALLAWAY: I can give you an example 21 for now would be a penalty to use your words on 22 control out growth.
23 CHAIRMAN KIRCHNER: And so since we're in 24 open session, I think it's useful for the record and 25 the public to just compare your NRFs that you 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 1 generated with utility. Could you highlight that?
2 Are they bounded by typical values used by utilities?
3 You have a table on Page 78 of your report, industry 4 standard values.
5 MS. CALLAWAY: Yes. So that's --
6 MR. NAZARETH: Yes, so.
7 CHAIRMAN KIRCHNER: That's public 8 information. Can you just go qualitatively, just say 9 how your NRFs compare with industry benchmarks or 10 standards?
11 MR. NAZARETH: Okay. Our NRFs are more 12 conservative than the industry benchmarks then. Okay?
13 So the specific details we will provide in the closed 14 session.
15 CHAIRMAN KIRCHNER: Sure.
16 MR. NAZARETH: But in general, that middle 17 bullet that says as conservative factors, they're set 18 up to be conservative relative to all the types of 19 benchmarking that we did, okay, at a 95/95 probability 20 confidence level.
21 And we'll talk about -- we have some of 22 the statistics in the topical report for anything else 23 you want to see. But they are more conservative than 24 in the industry.
25 CHAIRMAN KIRCHNER: And so going into 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 1 startup testing, you're using more conservative 2 nuclear reliability factors than are typical in the 3 industry?
4 MR. NAZARETH: Yes.
5 CHAIRMAN KIRCHNER: That's what I wanted 6 you to say.
7 MR. NAZARETH: Yes. There is one 8 parameter that is more conservative, and I'll have to 9 explain it.
10 CHAIRMAN KIRCHNER: Okay.
11 MR. NAZARETH: In the closed session.
12 Okay?
13 CHAIRMAN KIRCHNER: All right. Thank you.
14 MS. CALLAWAY: Okay. I'll let Chris take 15 over the last couple of slides for this session on the 16 code application.
17 MR. KIRBY: Okay. So NuScale was planning 18 to use the CMS5 to perform core designs and to provide 19 input to safety analysis. And in doing so when they 20 performed these calculations, they will be applying 21 NRFs to the CMS5 results in order to ensure a 22 conservative result on the analysis.
23 They also would like to use CMS5 to 24 perform startup physics testing and also do core 25 follow-up predictions. And in using the code in this 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 1 manner, that would also help to validate the NRFs that 2 have already been established and provide data to 3 update their NRFs when an enough data has been 4 presented.
5 So in summary, NuScale has performed 6 extensive benchmarking into the CMS5's code suite for 7 application to the NuScale design. The benchmarking 8 process has resulted in the development of nuclear 9 reliability factors for use in the application of the 10 code system to perform nuclear analysis. And these 11 NRFs will be validated and updated when operating data 12 from NuScale modules becomes available.
13 MS. CALLAWAY: That is the end of our open 14 session of material for nuclear analysis, codes and 15 methods topical report.
16 CONSULTANT SCHULTZ: Allyson, I have a 17 general question. And it has to do with NuScale 18 support of the overall application and basically who 19 did what?
20 MEMBER MARCH-LEUBA: Steve, can you put 21 the green light? Oh, you have? Sorry.
22 CONSULTANT SCHULTZ: Oh, it's on. And the 23 question is, have you some consultants in your -- as 24 I understand it in your application, you relied upon 25 Studsvik methodology. I don't know how much Studsvik 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 1 participated in this particular application. What I'm 2 most interested in is going forward.
3 NuScale is going to have a staff that's 4 going to move this in to the next phase of the 5 licensing process. And I wanted to get an 6 appreciation for what NuScale is doing to make sure 7 that the staffing that they have is well-experienced 8 in doing that application.
9 It has certainly benefitted from 10 consultants in developing the licensing application.
11 But going forward, you might leave those consultants 12 behind and have a group that's going to make this 13 satisfactorily move forward. And where are you now 14 and where are you going?
15 MS. CALLAWAY: Okay. So Studsvik didn't 16 participate at all in the development of this work 17 short of obviously supplying the software. But the 18 rest of the validation work and the contents of the 19 topical report and the rest of the work that supports 20 what's in the topical report.
21 CONSULTANT SCHULTZ: But you must have 22 consulted with them to at least develop an 23 understanding that what you were doing was an 24 appropriate application of the code?
25 MS. CALLAWAY: Yes, yes. So they do 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 1 provide training on the code, yes.
2 CONSULTANT SCHULTZ: Good.
3 MS. CALLAWAY: But otherwise the use and 4 application of it is then independently owned as a 5 function by NuScale. Does that fully answer your 6 question?
7 CONSULTANT SCHULTZ: And that included the 8 development and submittal of the application and the 9 RAI responses and so forth?
10 MS. CALLAWAY: Yes.
11 CONSULTANT SCHULTZ: Okay.
12 MS. CALLAWAY: Yes. So it will be 13 NuScale's staff or augmented staff. But the work and 14 the process and the technology and ownership of the 15 actual property is also --
16 CONSULTANT SCHULTZ: Augmented staff, and 17 is that augmentation going to continue or is it 18 envisioned to be more of an in-house activity?
19 MS. CALLAWAY: The scope of the topical 20 report was for approval for NuScale Power to apply 21 these methods. And certainly we do intend to be a 22 design center for our future plants. And dependent on 23 how many plants we have --
24 CONSULTANT SCHULTZ: Understood.
25 MS. CALLAWAY: -- that will determine our 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 1 staffing. But certainly we would hope to have 2 hundreds. If there's any potential customers --
3 CONSULTANT SCHULTZ: Thank you. That 4 clarifies my concern and your consideration. Thank 5 you.
6 CHAIRMAN KIRCHNER: Okay. Why don't we 7 continue with the CHF correlation presentation.
8 MS. WIKE: All right. So we're moving on 9 to the next NuScale open presentation on the critical 10 heat flux correlation topical report. So we do have 11 a new presenter joining us at the table. So I'll have 12 Mike introduce himself, give a little bit of a short 13 bio before we roll straight into the PowerPoint 14 slides.
15 MR. BRADBURY: Okay. Good afternoon. My 16 name is Mike Bradbury. I'm staff aug to NuScale from 17 ISL. And I was the lead of the CHF correlation 18 development.
19 With some of my prior experience, I worked 20 on mPower mostly thermal hydraulics stuff. I worked 21 at AREVA for six or seven years with PPR and such 22 things. And then --
23 CHAIRMAN KIRCHNER: Michael? Can I 24 interrupt and ask you just pull your microphone a 25 little closer and speak out a little more?
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 1 MR. BRADBURY: Okay.
2 CHAIRMAN KIRCHNER: Please.
3 MR. BRADBURY: And then prior to AREVA, I 4 worked at Cappell for six or seven years as well. So 5 I think that's me.
6 All right. So for the agenda, I'll just 7 do a quick overview of what the CHF topical is, some 8 introduction background information, the overview of 9 the development of the NSP1, 2 and 4 correlations and 10 then a little bit of a discussion about VIPRE and our 11 use of it.
12 Okay. So the overview, essentially 13 NuScale was seeking NRC approval to use both, or 14 either, the NSP2 or NSP4 CHF correlations in safety 15 analysis. The CHF testing that we have was performed 16 -- we had actually two kind of rounds of testing.
17 The first was done at Stern Labs and then 18 we have additional from the Framatome's KATHY facility 19 in Germany. And VIPRE was used in the correlation 20 development and is also used as the analysis tool for 21 safety analysis.
22 So this slide, we'll just kind of get into 23 a little bit what makes, you know, NuScale special.
24 The NuScale SMR operated under some unique conditions, 25 most notably the low flow from natural circulation 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 1 since we don't have forced flow pumps.
2 MEMBER CORRADINI: So maybe this is in 3 closed session. So you tell me to stop. Where do you 4 measure the flow in the design? We can wait.
5 MR. BRADBURY: In the design --
6 MEMBER CORRADINI: In other words in 7 actual operation.
8 MR. BRADBURY: In actual operation, I 9 would have to get back to you. I'm not sure exactly.
10 MS. WIKE: Yes. Let us get back to you in 11 the closed session.
12 MR. BRADBURY: So, you know, we needed a 13 NuScale specific CHF correlation basically because our 14 flow rates are very low. You know, the PWR, the 15 typical test in flow rates go down to about 1/2 16 million pounds per hour foot squared. And we need 17 data below that.
18 So that's, you know, we went off and 19 tested our -- at those conditions, which isn't 20 necessarily easy. CHF testing at low flows can be 21 kind of tricky.
22 We also have a shorter heated length, 23 which does play a little into the testing. You know, 24 ours is about a 6-1/2 feet length bundle versus 12 or 25 14 for conventional PWR.
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 1 So, like I said, we ran two different 2 campaigns. The first was at Stern and that was back 3 in 2012, 2013. And then we did a prototypical NuFuel-4 HTP2 bundled design test at KATHY in 2016.
5 MEMBER REMPE: So, I know Stern and KATHY 6 have now been used quite a bit in various 7 applications. But I still am interested in how you 8 have confidence in the calibration of the sensors.
9 And I know the report said you had diverse means.
10 And I guess I'm curious if you know of any 11 cases where they found that a sensor went bad and they 12 replaced it. Because, you know, if you have two 13 sensors and one is reading different from the other, 14 you don't know which one is right. And so I'm curious 15 about that.
16 Is there a potential that they both could 17 have drifted? And I know there's something in the 18 report that even said, well, all we care about is the 19 difference in temperature. And, again, if things are 20 drifting, I don't have a lot of confidence in the 21 difference in temperature or flow or whatever the 22 parameters you're measuring.
23 And so can you give us any of your 24 insights about why you are comfortable and confident 25 of the calibration of the sensors?
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 1 MR. BRADBURY: I'm not aware that any 2 sensors failed during our test campaign. I don't --
3 MEMBER REMPE: In the past, when you 4 talked to them, before you gave them money to do the 5 test, did you ask them about those kind of questions?
6 MR. BRADBURY: I believe, yes. And we 7 came away with confidence that they would be able to 8 adequately test. I can't speak too much to the Stern 9 data in that regard because that was before my time.
10 But at least with KATHY testing we, you know, 11 discussed those things and --
12 MEMBER REMPE: And what gave you 13 confidence that they were taking care of it?
14 MS. WIKE: Do you mind if I jump in?
15 MR. BRADBURY: Oh, absolutely.
16 MS. WIKE: So we do have, like I 17 mentioned, an approved Appendix B Program. And we did 18 do a QA audit of the testing facilities to ensure that 19 their controls and how they were running the 20 facilities met our Appendix B expectations for 21 quality.
22 MEMBER REMPE: But I just am curious in 23 just knowledge of what they did -- and I don't know 24 what Appendix B folks who were auditing would do to 25 give them that confidence. But what do you do? I 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 1 mean, I ran tests at a laboratory. And I know some of 2 the things we did and I just am curious in what they 3 did and why you felt comfortable.
4 MR. BRADBURY: Yes. I'm not sure that I'm 5 the right person to answer that.
6 MEMBER REMPE: If you have time, maybe 7 there's time to get back to us. I'm just curious if 8 you could write a couple paragraphs and let us know.
9 MR. BRADBURY: Absolutely.
10 MEMBER REMPE: Thank you.
11 MS. WIKE: Okay. Next slide.
12 MR. BRADBURY: Sure. All right. So a 13 little background on the NuFuel-HTP2 design itself.
14 We covered this a little bit in the nuclear analysis 15 code of methods. You know, it's a conventional 17 x 16 17 arrayed fuel bundle. Like I said earlier, it's 17 shorter than normal for typical PWRs.
18 We are using Framatome's HTP and HMP 19 spacer grids. And those have been used extensively 20 throughout the industry, both in B&W-type plants and 21 Westinghouse type plants. So there's quite a history 22 with those.
23 MEMBER SKILLMAN: Could you please go 24 back? Were the tests that were conducted tests that 25 were conducted at the Stern and KATHY tests that used 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 1 an assembly that looked like this?
2 MR. BRADBURY: The one at Stern --
3 MEMBER SKILLMAN: Let me say it different.
4 MR. BRADBURY: Sure.
5 MEMBER SKILLMAN: An assembly that was 6 assembled like this.
7 MR. BRADBURY: The one at Stern was not 8 because that was prior to us having Framatome as a 9 fuel vendor. So the one at Stern, you know, while the 10 lattice was the same, you know, the pitch and the 11 diameters and broads and stuff that are the same, the 12 grids were just simple non-mixing space requirements.
13 So they would not represent this product.
14 Now what was tested at the KATHY facility 15 was this actual product with the correct spans and, 16 you know, like, for instance, the bottom, you know, 17 grid before even the heated length, HNP, and then 18 there's HTPs and there's an HMP at the top. I mean, 19 down to the that, you know, they put all the right 20 spacer grids in all the right places for the KATHY 21 test.
22 MEMBER SKILLMAN: So I hear you say for 23 the KATHY test it is virtually an identical --
24 MR. BRADBURY: Yes.
25 MEMBER SKILLMAN: -- device?
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 1 MR. BRADBURY: Yes.
2 MEMBER SKILLMAN: Thank you.
3 MR. BRADBURY: Okay.
4 CHAIRMAN KIRCHNER: I think we're -- I'm 5 just scanning your slides, I don't think you are 6 showing us in your slide deck that's for the open part 7 the actual test array. It that 5 x 5 or?
8 MR. BRADBURY: Yes. It is a 5 x 5.
9 CHAIRMAN KIRCHNER: All right.
10 MR. BRADBURY: Which is pretty standard 11 for --
12 CHAIRMAN KIRCHNER: Okay.
13 MR. BRADBURY: All right. So a little 14 background on the operating conditions. When we 15 compared it to traditional PWRs, NuScale relies on a 16 much lower flow, obviously due to the lack of poor 17 circulation. And commensurate with that we also have 18 a much lower power density for linear heat generation 19 rate, whichever way you'd like to look at that.
20 So, you know, in this plot you can sort of 21 see NuScale as an outlier away from the other PWR 22 plants.
23 Now one of the things we look at is Power-24 to-Flow just to see, you know, it kind of gives you an 25 indicator on how aggressively you're going to be 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 1 pushing the thermo-hydraulics.
2 And NuScale's is a little higher than, you 3 know, other designs like EPR, AP1000. But when you 4 compare it to the typical Westinghouse Plant for the 5 AP1000, it's not drastically higher. So, you know, 6 that gives us confidence that we're not pushing it so 7 hard that we're going to, you know, be pushing 8 ourselves into a constant CHF problem.
9 CHAIRMAN KIRCHNER: In addition to linear 10 heat rate, what about typical heat fluxes?
11 MR. BRADBURY: The average heat flux is --
12 I want to say it's very low. I'm not sure of an exact 13 number.
14 MEMBER CORRADINI: We can get to it in the 15 closed session.
16 MR. BRADBURY: Yes, that's fine. I mean, 17 we can talk about it there.
18 MS. WIKE: Okay.
19 MR. BRADBURY: Sure. All right. Okay.
20 So it's a little convoluted how we got to our CHF 21 correlations due to what we were talking about earlier 22 with having -- you know, the Stern test doesn't model 23 exactly, you know, what we're planning on putting in, 24 et cetera.
25 So our initial correlation was this NSP1 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 1 correlation. And that was built off of the Stern data 2 that we had gotten back in 2013.
3 Like I said it had a generic non-mixing 4 spacer grid, but it was -- the testing was very 5 expansive in terms of conditions, like very low 6 relative to where the NuScale reactor is fairly high, 7 you know, mass fluxes. Pressure range was very large, 8 covering almost, you know, way below where a BWR would 9 go to until, you know, up to where a PWR would 10 normally be tested, things like that.
11 So it was a very thorough database, but, 12 you know, it just wasn't the final design. So we used 13 that to develop this NSP1 correlation. And then in 14 early 2015, we changed fuel vendors to Framatome. And 15 we were going with this NuFuel-HTP2 design, which 16 then had the HTP and HMP grids and that.
17 And so what we did was to make sure that, 18 you know, the correlation we would have would work 19 with that. We were able to obtain a limited amount of 20 data from AREVA that was relevant.
21 And we could look at that with this NSP1 22 correlation and we did. And it wasn't -- the 23 correlation didn't do a great job of predicting it 24 partly because it wasn't tuned to predict that.
25 So we developed this NSPX factor which it 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 1 was intentionally designed to have conservatism. It 2 addressed the trend difference that we saw with that 3 small set of data from Framatome. And it also added 4 an intentional, you know, like 5 percent extra 5 conservatism. Because the way we were working at the 6 time, we were going to have to be using the 7 correlation while we were in parallel getting the data 8 for the actual bundled design.
9 So we built in a little bit of risk 10 mitigation by adding conservatism. And then once we 11 actually received the data for the NuFuel-HTP2 design 12 in 2016, we used that set of data to validate the 13 combination of NSP1 and NSPX, which is what we refer 14 to as NSP2.
15 MEMBER CORRADINI: So what is NSP4?
16 MR. BRADBURY: I will get to that, I 17 believe, on the next slide.
18 MEMBER CORRADINI: Oh, sorry.
19 MR. BRADBURY: Oh, no, that's fine.
20 MEMBER SKILLMAN: Okay. Just before we 21 change slides, NuFuel-HTP2 trademark.
22 MR. BRADBURY: Yes.
23 MEMBER SKILLMAN: If you will codify it 24 such that it is married to the NuScale design forever 25 and ever?
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 1 MR. BRADBURY: Okay. So when you say 2 that, do you mean, would Framatome use it another 3 plant or do you mean it (simultaneous speaking)?
4 MEMBER SKILLMAN: For any NuScale plant 5 must the new NuScale plant use only NuFuel-HTP2 TM?
6 MR. BRADBURY: If they are -- if their 7 safety analysis is performed with either the NSP2 or 8 the NSP4 correlation, than yes. Now, if we switched 9 fuel vendors 20 years from now --
10 MEMBER SKILLMAN: A whole new package.
11 MR. BRADBURY: Right. We'd have to --
12 MEMBER SKILLMAN: Re-analysis.
13 MR. BRADBURY: Right. We'd have to get a 14 whole new correlation and such, yes.
15 MEMBER SKILLMAN: Okay. So what we're 16 really talking about here is this 37 fuel assembly 17 plant with 16 control rods up to 4.95 percent 18 enrichment using NuFuel-HTP2 TM.
19 MR. BRADBURY: For the CHF correlation, 20 yes. As far as -- I'll let you jump in on the other.
21 MEMBER SKILLMAN: The analogy in my mind 22 is when you buy a vehicle from a vehicle manufacturer, 23 you get an engine, probably designed and built by that 24 manufacturer for that vehicle. You don't get a 25 different engine.
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 1 And so what I'm coupling in my mind is, 2 okay, I got it. I understand what they're doing. But 3 for the suite of analyses and for the effort that is 4 going into this, the buyer of the first NuScale design 5 uses this fuel and only this fuel unless there's a re-6 analysis for fuel?
7 MS. WIKE: Correct.
8 MR. BRADBURY: Correct.
9 MEMBER SKILLMAN: It seems crazy, but I'm 10 just trying to get it set in my mind. Thank you.
11 CHAIRMAN KIRCHNER: Let me ask a different 12 question. NSPX is that a constant or a variable 13 correction?
14 MR. BRADBURY: It's a variable.
15 CHAIRMAN KIRCHNER: Like a Tong factor, 16 then?
17 MR. BRADBURY: Kind of.
18 CHAIRMAN KIRCHNER: Has it any physical 19 bases?
20 MR. BRADBURY: Yes. In developing it, I 21 think we'll get a little more into this in the close.
22 But in developing we, you know, looked at 23 sensitivities to various -- you know, I mean, in the 24 main frame there's a pressure, flow, et cetera, et 25 cetera. So, you know, we analyzed those things to 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 1 see, okay, you know, what seems to have the most trend 2 here? And then we picked the one that it seemed to 3 have the most trend with and used that.
4 And we designed it so it's conservative.
5 Like, we added the conservatism in. We designed it so 6 that you'll never get a benefit from it. So it's only 7 ever the worst -- or at best it's a penalty. You 8 won't ever gain from it, I guess.
9 CHAIRMAN KIRCHNER: Now stop me if I get 10 into something closed, but normally, in the industry, 11 the history is the spacer grid designs have been used 12 to enhance heat transfer.
13 And typically that gives you the ability 14 to run the fuel at a higher heat flux and still have 15 a significant margin in your margin to either, in your 16 case critical heat flux or DNB in the larger PWRs. So 17 I'm trying to just understand your language here.
18 You were saying repeatedly that the NSPX 19 factor conservatively extends the NSP1 correlation.
20 MR. BRADBURY: Right. Okay. So to make 21 it a little clearer.
22 CHAIRMAN KIRCHNER: Because the reason I'm 23 struggling with the verbiage here is that NSP1 was 24 developed with, you know, the plain vanilla non-mixing 25 grid. Once you go to your actual fuel with its nice 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 1 mixing veins and all of that in the grids, then one 2 expects better performance thermo-hydraulically 3 speaking.
4 MR. BRADBURY: I can discuss that. I 5 think that probably needs to be in the closed session.
6 CHAIRMAN KIRCHNER: Okay. I'll wait.
7 MR. BRADBURY: I'm ready to discuss it but 8 -- okay. I'll take a note and I'll bring it up at --
9 MS. WIKE: Okay. NSP4.
10 MR. BRADBURY: All right. The NSP4 11 correlation. So NSP2, if you look at it, we developed 12 it based on the Stern data really and a small subset 13 of data from Framatome. And then we validated it with 14 the NuFuel-HTP2 data. Okay. That's a non-traditional 15 of doing it. Normally, you would --
16 MEMBER CORRADINI: But when you say 17 validate, based on how you describe NSPX multiplying 18 NSP1 getting NSP2.
19 MR. BRADBURY: Mm-hmm.
20 MEMBER CORRADINI: It should, if I had a 21 45 degree line, always underpredict or predict CHF 22 before you actually saw it experimentally.
23 MR. BRADBURY: Correct.
24 MEMBER CORRADINI: Okay.
25 MR. BRADBURY: Yes. So the NSP2 was more 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 1 of a convoluted way of getting there. And that was 2 just due to scheduling essentially. NSP4 was a clean 3 slate. You know, we developed it off of the NuFuel-4 HTP2 specific CHF data.
5 We validated it with the same data. It's 6 more the normal process you would expect to go through 7 for developing a CHF correlation. So that's what the 8 NSP4 correlation is.
9 And what that gets us is it turned out the 10 NSP2 was very conservative. So NSP4 helps regain some 11 of those margins for other things, you know, whether 12 it is to do something that, you know, where you have 13 to accommodate a higher heat flux or, you know, 14 whatever could come up.
15 MEMBER CORRADINI: But let me ask a 16 different question. And, again, I'm not sure what you 17 showed on a slide somewhere, some box somewhere, about 18 flow versus power.
19 MR. BRADBURY: Oh, yes.
20 MEMBER CORRADINI: The design point didn't 21 change or did it?
22 MR. BRADBURY: No. The design point does 23 not.
24 MEMBER CORRADINI: So essentially, you've 25 regained margin to be used or to be known, but not to 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 1 be changing the design that you --
2 MR. BRADBURY: Correct.
3 MEMBER CORRADINI: -- now submitted.
4 MR. BRADBURY: Correct. Right. We would 5 not change anything with the DCI or anything like 6 that.
7 MEMBER CORRADINI: Okay. Fine. Thank 8 you.
9 MR. BRADBURY: So basically, you know, the 10 process for NSP2 was kind of convoluted. You started 11 with one set of data. You validated with a different 12 set of data. And then the process for NSP4 was more 13 traditional where, you know, you created the 14 correlation to the actual data you care about and 15 validated it as such. Okay.
16 CHAIRMAN KIRCHNER: But, again, it might 17 explain, I think, gets to what I was thinking that 18 your terminology -- you have a static design. But 19 your finding improved with your -- not improved data.
20 With more data that's more prototypical of your 21 design, your margins are greater with the static 22 design.
23 MR. BRADBURY: Right. Given the 24 particular design, we'll call it the NSP2 correlation 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 1 CHAIRMAN KIRCHNER: Yes.
2 MR. BRADBURY: -- because that's what I've 3 been calling it. That one predicts very 4 conservatively. Like, you know, if you pop up the CHF 5 data and you've got that 45 degree line, like, all of 6 it will be below it.
7 CHAIRMAN KIRCHNER: Yes.
8 MR. BRADBURY: And it's just very, very 9 conservative. And that's fine. I mean, you know, in 10 the Rev 0 or whatever of the DCA, that's what we used 11 and, you know, we were able to survive, you know, 12 transient scenarios. You know, but the desire was 13 there to, you know, recover margin in case things come 14 up in the future or, you know.
15 CHAIRMAN KIRCHNER: In case you have spare 16 change in your pocket and you want to come back for an 17 upgrade.
18 MR. BRADBURY: Maybe. Something. There's 19 lots of options where certainly margin never hurts you 20 so that's --
21 MEMBER CORRADINI: But I think just so 22 we're clear, you quantified the margin. You didn't 23 change the design.
24 MR. BRADBURY: We did not change anything 25 in the design.
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 1 MS. WIKE: That's correct.
2 MEMBER MARCH-LEUBA: The thing I don't 3 really understand on this little bit higher level is 4 why are you asking approval for NSP2 if NSP4 works 5 just as well and is much better than NSP2? Why are 6 you wasting our time?
7 MR. BRADBURY: We'll let Jennie.
8 MS. WIKE: Yes. Well, the answer is we 9 had originally submitted the topical report with NSP2.
10 MEMBER MARCH-LEUBA: So you worked very 11 hard on it so you're making us work hard on it?
12 MS. WIKE: We didn't want to restart the 13 review. There was a lot of overlap in our correlation 14 development between NSP2 and NSP4. And we wanted to 15 use that and so we added NSP4 to this existing 16 topical. And that's why we kept NSP2.
17 MEMBER MARCH-LEUBA: It was supplemental 18 rather than (simultaneous speaking).
19 MS. WIKE: Correct. When you see us again 20 for the applicable chapters in the DCA, you'll see 21 that we use, like for example, in Chapter 14, NSP4.
22 So while this topical covers both NSP2 and NSP4, in 23 our DCA we've adopted NSP4. Next slide.
24 MR. BRADBURY: All right. So both of 25 these correlations, NSP2 and NSP4, are implemented in 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 1 VIPRE. You know, VIPRE is a pretty ubiquitous 2 Subchannel T/H code. It's used by numerous vendors 3 and utilities. It has the generic SER.
4 And we did, you know, do all the use case 5 stuff. I'm not going to talk about it here because 6 that was in a separate topical report. That was in 7 Subchannel Analysis Methodology Topical Report.
8 But the point is that, you know, we took 9 the generic SER. We do show that it is applicable for 10 NuScale. And then, you know, one of the key things 11 here is, you know, all of the constituent of models 12 and such that you would pick for using an analysis are 13 identical to what was used in the CHF development.
14 So the CHF correlation itself is acting as 15 a closure model, kind of tying the application back to 16 testing.
17 MEMBER MARCH-LEUBA: So, yes, honestly, 18 I'm not picking on you. I'm just trying to learn. Is 19 EPRI an Appendix B supplier for you?
20 MR. BRADBURY: EPRI is not the -- the code 21 is owned at this point by Zachary.
22 MEMBER MARCH-LEUBA: And they are an 23 Appendix B supplier?
24 MR. BRADBURY: Yes.
25 MEMBER MARCH-LEUBA: Okay.
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 1 MEMBER REMPE: Well, if they're going to 2 be overlapped, we're going to do a separate review on 3 using VIPRE for NuScale and, I assume, each of the 4 documents references certain of each other's document.
5 MR. BRADBURY: Pretty much.
6 MEMBER REMPE: And if there's something 7 you6 find wrong in here, you'll have to update the 8 VIPRE document and vice versa. And that's the plan.
9 You'll just do a rev.
10 MR. BRADBURY: Yes. They are pretty 11 inexplicably tied.
12 MEMBER REMPE: Yes.
13 MEMBER CORRADINI: So just for my memory, 14 did Zachary used to be Nuclear Numerical Applications?
15 MR. BRADBURY: Well, yes, that was --
16 MEMBER CORRADINI: They're renamed.
17 MR. BRADBURY: That wasn't the -- who was 18 it? CSI, that's right.
19 MEMBER CORRADINI: Okay, fine.
20 MR. BRADBURY: Yes. Computer Simulation 21 something, I think.
22 MEMBER CORRADINI: Thank you.
23 MR. BRADBURY: Okay. So effectively we 24 have two CHF correlations with the NSP2 and the NSP4 25 that we developed. We feel they're both applicable to 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 1 NuScale. The NSP2 is conservative, which was good in 2 some ways. And the NSP4, you know, we can regain 3 margins.
4 You know, the topical report provides the 5 test data, provides the details of the correlation 6 development, provides the limits for the correlations, 7 which tie back to GDC10. And it also gives the 8 applicability range and criteria. And that's what 9 we're seeking approval of.
10 CHAIRMAN KIRCHNER: Okay. Any questions 11 at this point? Okay. Let's take a brief pause here 12 in situ and have the staff come up and present their 13 evaluation in open session. Go ahead, please.
14 MR. BAVOL: Okay. Thank you and good 15 afternoon. My name is Bruce Bavol. I'm a project 16 manager for the NuScale project. These are two of my 17 topical reports that we're reviewing. Currently, we 18 completed our review on.
19 And to my right is Tim Drzewiecki. And I 20 just wanted to let you know that per our agenda that 21 in a closed session, we are going to cover each topic 22 separately to add continuity to the closed material.
23 The other contributors, Becky Karas is 24 here. She's a branch chief of the Reactor Systems.
25 Jeff Schmidt, who is not. And then we had some 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 1 additional support for both the nuclear analysis and 2 CHF by Boyce Travis, Josh Kaizer and David Heeszel.
3 The timeline for the codes and the methods 4 qualification, this particular topical report, as was 5 discussed, came in on August 30, 2016. We had an RAI-6 8807 came in on May 9, 2017. We got a response back 7 from NuScale. That response was reviewed and closed 8 resolved.
9 Our future plans are to have an ACRS full 10 committee scheduled currently at June 7. It's a 11 Thursday morning. And we plan to issue a -- our plan 12 is to issue a final SER late 2018. And then publish 13 our approved version early October.
14 With that, I'm going to turn it over to 15 Tim.
16 DR. DRZEWIECKI: Thanks, Bruce. So in 17 terms of the findings made for the nuclear codes and 18 methods topical report we're not making findings, of 19 course, on any of the GDCs at this point. Those are 20 made during a DCA review.
21 But where this methodology is used once 22 approved, it will form a partial basis for those 23 findings.
24 What was asked for is staff to first 25 approve the application of the CMS5 code suite for the 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 1 new scale core design as well as the base NRFs which 2 are the numbers that are used to account for the 3 uncertainty in the prediction of those parameters as 4 well as an NRF update methodology.
5 Key areas of the staff review, and this is 6 kind of how the SE is laid out. One is basically 7 modeling capabilities of the CMS5 code suite, in 8 particular can you mode the appropriate geometry, the 9 material? And do you have the appropriate physics 10 modeling capabilities in this code to do the job?
11 The assessment of the CMS5 code suite, and 12 that includes the code-to-code comparisons as well as 13 comparisons to any kind of benchmarks and operating 14 data, the NRF update methodology and development, 15 that's mostly a math problem.
16 And then the last one was the uncertainty 17 of management during operation. And this is even if 18 you have confidence in your NRFs and your update 19 methodology, what if you're still wrong whether you do 20 that? So that was something that we had fleshed out 21 as part of the SE.
22 Okay. So the areas in which, you know, 23 staff had spent some more time and which you would not 24 do in a traditional review, one as far as the geometry 25 of this plant the fact that you have several modules 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 1 that are close together, we wanted to ensure that 2 there was no kind of feedback from module to module, 3 especially given the location of the ex-core 4 detectors.
5 The physics modeling, we wanted to make 6 sure thermo-hydraulic modeling capabilities inside of 7 SIMULATE5 were appropriate. And this goes into 8 things, you know, like your void feedback. Make sure 9 that you can actually capture, you know, your void 10 fraction inside of SIMULATE5.
11 A validation of analysis, we made sure 12 that, you know, some of the key components here that 13 were different from a plant or the fact that it's a 14 smaller core, higher leakage and it's got a heavy 15 reflector. So we made sure that they had validation 16 data that can address those aspects of this design.
17 Things staff had considered as part of 18 this review was not only the information provided in 19 this topical and obtained in the REI process as well 20 as the audit of the supporting calculations, but we 21 also looked at the review of the generic CMS5 report 22 as well as the manuals and information that was 23 supporting that review.
24 Also, we had looked at similar types of 25 reviews that were done in this area, like Duke Power 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 1 design methodology on any APS reactor physics 2 methodology.
3 So as far as staff conclusions, on this 4 SER we did find the CMS5 code suite can model the 5 NuScale core design subject to one limitation. And 6 that was that there was a list of materials used in 7 the core design that were fleshed out in the CMS5 8 generic review. And they are going to be limited to 9 those same limitations here obviously.
10 But other than that, they are not 11 particularly wedded to a design that would have, like, 12 an M5 cladding. If that were to change in the future, 13 they would not be bound by this topical to use M5.
14 The base NRFs --
15 CHAIRMAN KIRCHNER: May I interrupt you, 16 Tim, just on the public record, limitation to 17 materials. Now, if I recall, CMS5 is using ENDF/VII, 18 whatever mod.
19 DR. DRZEWIECKI: That's right, yes.
20 CHAIRMAN KIRCHNER: So I would assume 21 that's all encompassing of the materials that are 22 being modeled or are there other materials in this 23 design that that cross-section database doesn't cover?
24 DR. DRZEWIECKI: So it's not limited by 25 the cross-section so much as the closure of equations 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 1 that are being used to model some of the materials 2 inside of SIMULATE5. Because SIMULATE5 does account 3 for things like fuel burn-up and change in the thermal 4 conductivity. It accounts for things like changes in 5 geometry from swelling. And those have an impact on 6 reactivity. And because of those equations, is why --
7 CHAIRMAN KIRCHNER: So it's more in 8 reference to the SIMULATE5 --
9 DR. DRZEWIECKI: Yes.
10 CHAIRMAN KIRCHNER: -- not the basic 11 neutronics. Thank you.
12 DR. DRZEWIECKI: Sure. Okay. The second, 13 which is the base of NRF's proposed are acceptable.
14 And going back to one question that was stated before, 15 the base NRFs all account for more uncertainty than 16 the generic CMS5 NRFs.
17 And last is that the NRF update 18 methodology that was proposed in its topical report 19 are acceptable subject to one limitation and the 20 condition.
21 And one limitation was that any kind of 22 updates to of a delayed neutronic parameter cannot 23 reduce its magnitude at below 5 percent. And that's 24 based on the data that we had seen. It was kind of 25 limited to that. So we couldn't see -- I mean we 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 1 could get it lower than that.
2 And the one condition is that any updates 3 to the peaking factors, they must account for the pin-4 to-box bias and fixed in-core detector bias in 5 accordance with the generic CMS5 topical report.
6 CONSULTANT SCHULTZ: Tim, that condition 7 is there because more information needs to come in 8 during operation in order for the overall evaluation 9 to be complete in these areas?
10 DR. DRZEWIECKI: You mean as far as this 11 condition for the peaking factors?
12 CONSULTANT SCHULTZ: Yes. This is 13 provided because there wasn't sufficient information 14 based upon lack of operation of the NuScale 4.
15 DR. DRZEWIECKI: No. It has to do with 16 how these factors were generated in the generic 17 methodology in the sense that when you come up with a 18 pin-to-box bias, that has two factors in there.
19 One is a comparison of CASMO5 to MCNP6 --
20 I'm sorry, actually, no. One of the benchmarking of 21 the CASMO5 code to data to basically see how that code 22 actually compares against data and the other was a 23 SIMULATE5 to CASMO5 to look at the uncertainty that 24 you impose for your pin power reconstruction 25 algorithm. Those are fixed values that you really 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 1 can't reduce by getting more data. And those were set 2 by the code vendor. And so we are requiring that 3 those be maintained.
4 MEMBER CORRADINI: Oh, I got you. It's 5 the bias that you're maintaining.
6 DR. DRZEWIECKI: That's right.
7 MEMBER CORRADINI: As whatever new data 8 will move the performance, you want to maintain the 9 bias, if I understood the condition.
10 DR. DRZEWIECKI: That's right. That's 11 right. Yes. And there's also a bias associated with 12 having a fixed core instrumentation system, which 13 affects your 3D power peaking factor.
14 CONSULTANT SCHULTZ: All right. So in 15 spite of what might be developed in the future by 16 NuScale, the obligation is to meet condition one?
17 DR. DRZEWIECKI: That's right.
18 CONSULTANT SCHULTZ: Thank you.
19 MR. BAVOL: Okay. We're going move on to 20 the critical heat flux, the staff reviewed time line.
21 This particular topical report Rev 0 was updated to 22 Revision 1. The revised submittal implements the 23 additional NSP4 critical heat flux correlation and 24 incorporates changes associated with the RAIs that 25 this staff had submitted to point.
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 1 There were two key RAIs. RAI-8795 and 2 RAI-8931. ML numbers are provided. Those two are as 3 responded to, completed, resolved and satisfactorily.
4 The staff plans to also have an advisory committee 5 represent the safeguards at full committee June 7, 6 same time. We'd like to have an issue of a final SER 7 late July 2018 and publish the approved version of the 8 topical report also in October 2018.
9 DR. DRZEWIECKI: So similar to the last 10 topical report, we are not making findings against 11 specific GDCs here and approve the CHF topical report 12 can be used to establish a partial basis, which is 13 used in review of the DCA to make findings against 14 things like GDC10, 12 as well as any kind of 15 omnivaluation of accidents looking at those 16 consequence of analyses.
17 NuScale had requested this review go 18 through and approve the use of NSP2 and NSP4, the CHF 19 correlations for the safety analysis of NuScale power 20 module with NuFuel-HTP2 fuel over a specified range of 21 applications and with specific CHFR limits.
22 Key areas of staff review, of course, was 23 data collection, CHF model generation and a large 24 focus of this review was on quantification of the CHF 25 model error, which is how you come up with your CHFR 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 1 limits.
2 Areas in which the staff spent more time 3 was on the development of a 95/95 CHFR limits as well 4 as we did look closer at the correlation of behavior 5 because its trends was somewhat different from what we 6 had seen in the traditional PWR. And so we wanted to 7 make sure that we understood why we were seeing that, 8 or at least, you know, know that there was enough data 9 to back up those trends.
10 I think the staff had considered obviously 11 the information in this topical as well as the RAI 12 responses in the audit of the supporting calculations.
13 There was an inspection done at Stern 14 Laboratories for 2013 when we were collecting data.
15 There was also a program, a report, that was submitted 16 to the staff in 2014, preapplication space, that 17 pretty much had laid out what their program was going 18 to look like, where they expect in terms of what kind 19 of flow rates and pressures they expect to see in 20 their transient and accident analyses as well as there 21 was an audit done of a KATHY facility in 2016. And 22 then, of course, the information that was obtained 23 from previously reviewed CHF correlations.
24 And as part of this SE, there was an 25 appendix that kind of lays out the review framework 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 1 that was informed by several of the past CHF reviews.
2 MEMBER REMPE: So I'll ask you the same 3 question I asked the applicant about the data and when 4 instrumentation goes bad and what you've seen in the 5 past with your reviews of Stern data and KATHY data.
6 Again, you know, have they had some errors where they 7 detected it? And how was that detected and how was it 8 replaced, the censors replaced and how were they 9 qualified?
10 DR. DRZEWIECKI: So I have not seen issues 11 with them seeing failures in their, you know, in their 12 calibrated instrumentation.
13 Something that does give us confidence is 14 both at Stern and at KATHY, they do repeat test points 15 to ensure that they can get, you know, the same points 16 over and over again. Now the way it's done at Stern 17 and KATHY are different. And that's fleshed out some 18 in the closed session. I can explain how that's done 19 at both facilities.
20 MEMBER REMPE: Okay.
21 DR. DRZEWIECKI: And so if they fell out 22 of calibration, there are certain reference points 23 where they should be able to go back and check and see 24 that they're still getting the results.
25 MEMBER REMPE: I'd be interested. Thank 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 1 you.
2 DR. DRZEWIECKI: Okay. So the staff of 3 SER conclusions is that the NSP2-CHF correlation is 4 acceptable for use in performing the safety analysis 5 of NuScale power module with NuFuel-HTP2 fuel with a 6 limit of 1.17 over a range of applicability provided 7 in the table, in that topical report, subject to two 8 limitations.
9 Similarly, the NSP4 correlation is also 10 acceptable for use with a limit of 1.12 over a range 11 of applicability, which are provided in the topical as 12 well.
13 Those two limitations are that the non-14 uniform flux factors, or your Tong factors, have to 15 always be greater than or equal to one. Basically, it 16 has to always be -- it can't be a credit to CHF 17 margin. It's always applied as a penalty.
18 And that the use of these correlations 19 have to be used in accordance with the Subchannel of 20 Analysis Methodology to be used in accordance with the 21 validation of these CHF correlations.
22 MEMBER CORRADINI: Which is a separate 23 topical we'll see.
24 DR. DRZEWIECKI: Yes. That's right.
25 Acceptable topical we'll see.
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 1 MEMBER CORRADINI: So, I mean, that makes 2 sense. I'm just trying to find out how this is 3 linked. So how can one agree to use the CHF 4 correlation until we see the subchannel analysis.
5 They seem inextricably linked.
6 DR. DRZEWIECKI: Well, they are linked.
7 The important thing is that the models used for all of 8 the data reduction calculations are frozen. That's 9 the important thing.
10 That's what we want to do is to ensure 11 that those models and those code options are frozen 12 which is, you know, that's what's laid out in that 13 subchannel analysis methodology. And so, basically, 14 when you do any kind of subchannel calculation, you 15 know, you have confidence that you're going to get the 16 same kind of -- yes.
17 MEMBER CORRADINI: And then the next 18 question is so, as I understand the limitation, if 19 used with an approved subchannel analysis methodology, 20 NuScale can apply either to or for?
21 DR. DRZEWIECKI: That's right.
22 MEMBER CORRADINI: Okay.
23 MR. BAZOL: And that concludes our open 24 part of the presentation.
25 CHAIRMAN KIRCHNER: Okay. Any questions 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 1 from the members at this point? Let us then turn to 2 any comments from the public and see if there are any 3 members of the public that wish to make a comment?
4 I'm not seeing any present.
5 Can we -- yes, I know. We'll wait for the 6 bridge line to be opened and see if any members of the 7 public are out there. If there are any members of the 8 public listening in, and if you wish to make a 9 comment, please identify yourself and make that 10 comment. Hearing none, we can close the bridge line, 11 please.
12 And with that, I propose that we take a 13 break at this point. I assume that we'll have 14 comments from the committee and our consultant after 15 the closed session. So with that we are recessed.
16 Let's come back at quarter of on that clock.
17 (Whereupon, the open session of the matter 18 went off the record at 2:31 p.m.)
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
NuScale Nonproprietary Nuclear Analysis Codes and Methods Qualification TR-0616-48793 ACRS Presentation OPEN SLIDES Allyson Callaway Chris Kirby Vick Nazareth ACRS Open Session - May 15, 2018 1
PM-0518-59766-P Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Acknowledgement & Disclaimer This material is based upon work supported by the Department of Energy under Award Number DE-NE0000633.
This report was prepared as an account of work sponsored by an agency of the United States (U.S.) Government. Neither the U.S.
Government nor any agency thereof, nor any of their employees, makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the U.S.
Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the U.S.
Government or any agency thereof.
2 PM-0518-59766-P Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Agenda
- Overview
- Code Description
- Core and Reflector Description
- Code Validation
- Nuclear Reliability Factor Generation
- Applications
- Summary 3
PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Overview
- NuScale seeks NRC approval to use Studsvik Scandpowers Core Management Software (CMS5) code suite for the nuclear physics design and analysis of the NuScale reactor
- Code suite has been benchmarked for NuScale application
- Nuclear reliability factors (NRFs) have been generated to ensure conservative application for downstream calculations
- Codes have been validated to generate nuclear physics parameters for use in core design, safety analysis, startup testing, and operations support 4
PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Code Description
- CMS5 code suite includes:
CASMO5
- Lattice physics code (ENDF/B-VII.1)
- Uses multi-group 2D/3D neutron transport SIMULATE5
- Core simulator
- Uses 3D steady-state, multi-group, nodal diffusion theory CASMO5/SIMULATE5 is the latest Studsvik neutronics code suite version
- Earlier and latest versions of this code suite are widely used in the nuclear industry for core physics calculations 5
PM-0518-59766-P Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Core, Fuel, & CRA Description Core Description Power Level, MWt 160 Number of Fuel Assemblies 37 Number of Control Rod Assemblies 16 Fuel Assembly Description Fuel Rod Array 17x17 Number of Guide Tubes 24 Number of Instrument Tubes 1 Fuel Rod Description Material UO2 Maximum Enrichment, wt% 4.95 Clad Material M5 Active Fuel Length, in. (cm) 78.74 (200)
Control Rod Assembly Description Clad Material Stainless Steel Poison Material AIC, B4C Integral Burnable Poison Description Absorber Material Gd2O3 6
PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Core Radial Reflector Radial Reflector Cooling Channels Core Barrel 7
PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
CMS5 Validation Flow Chart 8
PM-0518-59766-P Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Code-to-Code Benchmarking
- CMS5 code-to-code comparisons are performed with MCNP6
- MCNP is considered a higher-fidelity code
- MCNP6 model is both detailed and specific to NuScale design
- Parameters benchmarked with this process include keff, peaking factors, power distribution, critical boron concentration, reactivity coefficients, and control rod worth
- Data from code-to-code comparisons consists of tens to hundreds of data points for each parameter being examined
- Data points represent either an absolute or relative difference between CMS and MCNP 9
PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Empirical Benchmarking
- Empirical benchmarking by code simulation of criticality data and comparing it to data results
- Types of Data
- Critical Experiment Data
- Experimental Reactor Data
- Commercial Reactor Critical Data
- Parameters benchmarked with this process include keff, power distribution, critical boron concentration, reactivity coefficients, and control rod worth 10 PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Commercial Data Benchmarking
- Commercial Data Benchmarking
- Use available and relevant operating data to demonstrate ability to accurately predict neutronics parameters with CMS5, as no NuScale-specific operating data exists
- TMI-1 Cycles 1 & 2 were modeled with CMS5 suite
- Comparisons were made between actual operating data and CMS5 predictions
- Important neutronics parameters were compared to derive biases and uncertainties
- Results from commercial data benchmarks, empirical benchmarks, and higher-fidelity code comparisons are used to determine bias and bias uncertainties, and in turn develop NRFs for NuScale applications of CMS5 11 PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Reliability Factor Generation
- The benchmarking data discussed previously was used to determine bias and bias uncertainties
- Develop a set of conservative factors called Nuclear Reliability Factors (NRFs) for initial use in the NuScale design
- NRFs will be validated and updated during the startup testing process and module operation 12 PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
CMS5 Application
- CMS5 is used to perform core physics calculations for multiple applications in the NuScale modules:
- Designing and optimizing the nuclear core for initial core and reloads
- Providing input parameters for design basis safety analysis
- Generation of startup physics testing predictions
- Generation of core follow predictions
- Developing plant physics data books 13 PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Summary
- NuScale has demonstrated acceptability of Studsviks CMS5 code suite for the nuclear design and analysis of the NuScale design.
- Code qualification is supported by benchmarking against critical experiments, experimental reactors, commercial reactor data, and a higher-fidelity code.
- Qualification has resulted in the determination of conservative NRFs for use in the NuScale analyses.
- Physics parameters generated using this code suite will be used in core design, safety analysis, startup testing, core follow and operations.
14 PM-0518-59766-P T Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Portland Office Richland Office 6650 SW Redwood Lane, 1933 Jadwin Ave., Suite 130 Suite 210 Richland, WA 99354 Portland, OR 97224 541.360.0500 971.371.1592 Arlington Office Corvallis Office 2300 Clarendon Blvd., Suite 1110 1100 NE Circle Blvd., Suite 200 Arlington, VA 22201 Corvallis, OR 97330 541.360.0500 London Office 1st Floor Portland House Rockville Office Bressenden Place 11333 Woodglen Ave., Suite 205 London SW1E 5BH Rockville, MD 20852 United Kingdom 301.770.0472 +44 (0) 2079 321700 Charlotte Office 2815 Coliseum Centre Drive, Suite 230 Charlotte, NC 28217 980.349.4804 http://www.nuscalepower.com Twitter: @NuScale_Power 15 PM-0518-59766-P Revision: 0 Copyright 2018 by NuScale Power, LLC.
Template #: 0000-21727-F01 R3
Presentation to the ACRS Subcommittee Staff Review of NuScale Topical Reports TR-0616-48793, REV 0, NUCLEAR ANALYSIS CODES AND METHODS QUALIFICATION
&
TR-0116-21012, REV 1 NUSCALE POWER CRITICAL HEAT FLUX CORRELATIONS OPEN SLIDES Presenters:
Bruce Bavol - Project Manager, Office of New Reactors Timothy Drzewiecki, Ph.D.- Reactor Systems Engineer, Office of New Reactors May 15, 2018 (Open Session)
Non-Proprietary 1
NRC Technical Review Areas/Contributors Reactor Systems NRO/DSRA/SRSB:
Jeffrey Schmidt Timothy Drzewiecki Additional Support on Nuclear Analysis Methodology Boyce Travis, NRO/DSRA/SCVB Additional Support on CHF:
Joshua Kaizer, NRR/DSS/SNPB David Heeszel, NRO/DSEA/RGS 2 Non-Proprietary
Staff Review Timeline TR-0616-48793, Nuclear Analysis Codes and Methods Qualification NuScale submitted its Topical Report (TR) TR-0616-48793, Nuclear Analysis Codes and Methods Qualification, on August 30, 2016 (Agencywide Documents Access and Management System (ADAMS) Accession No. ML16243A517), as supplemented by letter dated July 6, 2017 (ADAMS Accession No. ML17187B240).
Staff issued request for additional information (RAI 8807) on May 9, 2017 NuScale responded to RAI 8807 on July 6, 2017 (ADAMS Accession No. ML17187B239)
Staff plans to brief advisory committee on reactor safeguards (ACRS) full committee on June 7, 2018 Staff plans to issue its final SER in late July 2018 Staff plans to publish the -A (approved) version of the TR in early October 2018 3 Non-Proprietary
Scope of the Staff Review
- Regulatory Basis
- An approved nuclear analysis methodology is used to establish a partial basis for compliance with several general design criteria (GDC) of 10 CFR 50, Appendix A
- Applicant requested approval of:
- 1. Applicability of Studsvik Scandpower, Inc. (Studsvik), Core Management Software, Version 5 suite (CMS5) to the NuScale reactor core design (RXC)
- 2. Base nuclear reliability factors (NRFs)
- 3. NRF update methodology
- Key areas of NRC staff review:
- Geometric, Material, and Physics modeling capabilities of CMS5 (applicability to NuScale RXC)
- Assessment of CMS5 suite (code-to-code and empirical data)
- NRF development and update methodology
- Uncertainty management during operation 4 Non-Proprietary
Scope of the Staff Review (cont)
- Areas requiring additional review (due to unique design features)
- Geometry Modeling: Multi-module effects and modeling of heavy reflector
- Physics Modeling: Thermal-hydraulic modeling capabilities in SIMULATE5
- Validation analyses to address high leakage cores and heavy reflectors
- Information considered by NRC staff:
- TR-0616-48793, Nuclear Analysis Codes and Methods Qualification, and supporting information obtained via RAIs and audit of supporting calculations
- Documentation provided by Studsvik in support of generic CMS5 topical report:
- SSP-14/P01-028-TR-NP, Generic Application of Studsvik Scandpower Core Management System to Pressurized Water Reactors (ML15355A285)
- CASMO5, SIMULATE5, CMSLINK5 manuals
- Similar topical reports/safety evaluations (Duke Power Company Nuclear Design Methodology, Arizona Public Service Company PWR Reactor Physics Methodology) 5 Non-Proprietary
Staff SER Conclusions
- 1. CMS5 code suite is applicable to the NuScale RXC design, subject to Limitation 1:
- Limitation 1: Application of TR-0616-48793 is limited to the materials identified in the safety evaluation for the generic CMS5 methodology
- 2. The base NRFs proposed in TR-0616-48793 are acceptable
- 3. The NRF update methodology proposed in TR-0616-48793 is acceptable, subject to Limitation 2 and Condition 1
- Limitation 2: Updates to any delayed neutron parameter NRF cannot reduce the magnitude of the NRF below 5 percent
- Condition 1: Updates to the pin peaking NRFs (FH and FQ) must include the pin-to-box bias and fixed in-core detector bias in accordance with the generic CMS5 topical report.
6 Non-Proprietary
Staff Review Timeline TR-0116-21012, Revision 1, Critical Heat Flux Correlations NuScale submitted its Topical Report (TR) TR-0116-21012, Revision 1, Critical Heat Flux Correlations, on November 30, 2017 (Agencywide Documents Access and Management System (ADAMS) Accession No. ML17335A089), this revision replaced the submittal dated October 5, 2016 (ADAMS Accession No.
ML16279A363), NuScale Power Critical Heat Flux Correlation NSP2.
The revised submittal implements an additional NSP4 critical heat flux (CHF) correlation and incorporates changes associated with NRC requests for additional information (RAIs).
Staff issued two requests for additional information (RAI 8795 on May 8, 2017 -
ADAMS Accession No. ML17128A468) and (RAI 8931 on August 21, 2017 -
ADAMS Accession No. ML17233A127)
NuScale responded to RAI 8795 on (July 7, 2017 - ADAMS Accession No.
ML17188A461) and RAI 8931 on (September 25, 2017 - ADAMS Accession No.
Staff plans to brief advisory committee on reactor safeguards (ACRS) full committee on June 7, 2018 Staff plans to issue its final SER in late July 2018 Staff plans to publish the -A (approved) version of the TR in early October 2018 7 Non-Proprietary
Scope of the Staff Review
- Regulatory Basis
- An approved critical heat flux (CHF) correlation is used to establish a partial basis for compliance with several regulations including the general design criteria (GDC) of 10 CFR 50, Appendix A:
- GDC 10 and GDC 12 (margin to fuel design limits), 10 CFR 52.47(a)(2)(iv) and GDC 19 (radiological consequences)
- Applicant requested approval of:
- Use of NSP2 and NPS4 CHF correlations for the safety analysis of the NuScale Power Module
- NuFuel-HTP2TM fuel
- Over a specified range of applicability
- With specified CHF ratio (CHFR) limits
- Key areas of NRC staff review:
- Collection of experimental data
- CHF model generation
- Quantification of CHF model error (development of CHFR limits) 8 Non-Proprietary
Scope of the Staff Review (cont)
- Areas requiring additional review
- Statistical Analysis - Development of the 95/95 CHFR limits
- Correlation behavior differs from trends observed in previously reviewed CHF correlations for pressurized water reactors
- Information considered by NRC staff:
- TR-0116-21012, Revision 1, NuScale Power Critical Heat Flux Correlations, and supporting information obtained via RAIs and audit of supporting calculations
- Inspection report for CHF testing at Stern Laboratories, April 18, 2013
- NuScale CHF test program technical report, January 24, 2014
- Audit report for CHF testing at KATHY, August 12, 2016
- Information obtained from previously reviewed CHF correlations as documented in Appendix A of the staffs SE 9 Non-Proprietary
Staff SER Conclusions
- NSP2 CHF correlation is acceptable for use in performing safety analyses of the NPM with NuFuel-HTPTM fuel
- With a CHFR limit of 1.17
- Over the range of applicability provided in Table 8-2 of TR-0116-21012, Rev. 1
- Subject to Limitation 1 and Limitation 2
- NSP4 CHF correlation is acceptable for use in performing safety analyses of the NPM with NuFuel-HTPTM fuel
- With a CHFR limit of 1.21
- Over the range of applicability provided in Table 8-4 of TR-0116-21012, Rev. 1
- Subject to Limitation 1 and Limitation 2
- Limitations
- Limitation 1: Nonuniform flux factors must always be greater than or equal to 1
- Limitation 2: CHF analyses using the NSP2 and NSP4 correlations must be performed in accordance with TR-0915-17564, Subchannel Analysis Methodology 10 Non-Proprietary
Questions/comments from members of the public before the closed session starts?
11 Non-Proprietary