ML19108A426
| ML19108A426 | |
| Person / Time | |
|---|---|
| Issue date: | 03/22/2019 |
| From: | Office of Nuclear Regulatory Research |
| To: | |
| K. Hamburger 415-2022 | |
| Shared Package | |
| ML19108A420 | List: |
| References | |
| NRC-0205 | |
| Download: ML19108A426 (76) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
High Energy Arcing Fault Large-Scale Test Plan Comment Resolution Docket Number:
(n/a)
Location:
Rockville, Maryland Date:
Wednesday, March 20, 2019 Work Order No.:
NRC-0205 Pages 1-75 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005 (202) 234-4433
1 UNITED STATES OF AMERICA 1
NUCLEAR REGULATORY COMMISSION 2
+ + + + +
3 HIGH ENERGY ARCING FAULT LARGE-SCALE TEST PLAN 4
COMMENT RESOLUTION 5
+ + + + +
6 WEDNESDAY, 7
MARCH 20, 2019 8
+ + + + +
9 ROCKVILLE, MARYLAND 10
+ + + + +
11 The public meeting was held in Room O-12 11B04, Nuclear Regulatory Commission, One White Flint 13 North, 11555 Rockville Pike, at 1:00 p.m., Mike Cheok, 14 Deputy Director, Division of Inspection and Regional 15 Support, presiding.
16 17 NRC STAFF PRESENT:
18 MIKE CHEOK, Office of Research 19 TOM BOYCE, Generic Issues Program Manager, Office of 20 Research 21 STAN GARDOCKI, Generic Issue Program, Office of 22 Research 23 KENNETH HAMBURGER, Office of Research 24 NICK MELLY, Office of Research 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
2 BRIAN METZGER, NRR Fire Protection 1
KENN MILLER, Office of Research 2
MARK SALLEY, Office of Research 3
GABE TAYLOR, Office of Research 4
MARK THAGGARD, Office of Research 5
6 ALSO PRESENT:
7 VICTORIA ANDERSON, NEI 8
JANA BERGMAN 9
PAUL GUNTER, Beyond Nuclear 10 CHRIS LAFLEUR, Sandia National Laboratories*
11 CASEY LEJA 12 CHICO PELIZZARI*
13 TONY PURTORTI, NIST*
14 MARKO RANDELOVIC, EPRI 15 RONNIE REYNOLDS, Exelon 16 MIKE SHAIRER*
17 OZZIE VIDAL*
18 KELLI VOELSING, EPRI 19
- present by telephone 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
3 C-O-N-T-E-N-T-S 1
Welcome and Opening Remarks 2
Mike Cheok...................
4 3
Introduction 4
Mark Salley 5
5 Marko Randelovic................ 20 6
Nicholas Melly................. 25 7
Discussion and Comments............. 65 8
Adjournment................... 75 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
4 P-R-O-C-E-E-D-I-N-G-S 1
12:56 p.m.
2 MR. CHEOK: Thank you for coming today.
3 As you know this is a series of -- one in a series of 4
public meetings. We will -- we had and will continue 5
to have -- talk about the aluminum HEAF generic issue.
6 Again, these meetings that we have periodically -- in 7
between meetings we will continue to work in the work 8
group with EPRI on things like the testing procedures, 9
comments on doing risk analysis, on doing frequency 10 determinations, and we'll continue to have these 11 public meetings to basically inform everybody else 12 what we do on -- on what we're doing and to get 13 comments in a public forum.
14 So today's meeting is mostly to hear any 15 additional public comments we have. We intend to be 16 mostly in a listening mode because I don't think -- we 17 got some of the comments beforehand, but at this point 18 I think we want to listen to what you all have. And 19 we'll offer a second meeting to be able to come back 20 and respond to any comments you may have.
21 And the one thing I want to say is we have 22 to keep in mind that this generic issue process is 23 about, right? So we would like to have a schedule or 24 type of -- we would like to finish things in a certain 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
5 schedule. And also as part of the generic issue 1
process it's not -- we don't intent for a lot of 2
things to be research projects as such. And so we 3
will listen to all the comments on doing testing. We 4
will try to incorporate them without having to impact 5
on our schedules by too much. Plus we don't want this 6
generic issue to continue for a long time. We would 7
like to get it finished in time that we will -- it's 8
supposed to get finished by.
9 And so we will try to incorporate a lot of 10 the comments as much as we can and we'll keep 11 everybody involved. And we will keep you updated as 12 to what our responses would be and we will get your 13 feedback into our responses. So it will be a 14 continuing feedback process.
15 So with that I'll just let --
16 MR. SALLEY: Okay. So I guess -- good, 17 Mike. Thank you. I want to give -- just give you a 18 quick overview of this, and the meeting is for you 19 all, so we'll turn it over to you. We'll just do a 20 little quick intro, or a quick overview of the 21 project.
22 As we've been going through and doing the 23 testing we've tried to be as transparent as possible 24 in both the large and the small-scale test plans.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
6 We've put those out in the Federal Register notice and 1
we've asked for public comments. And you can just see 2
just a sample there of some of the comments we have 3
received. You'll notice the OECD ones there that we 4
also have a follow-on project that we're doing 5
internationally with 10 countries to further look at 6
HEAFs in total.
7 Again, a lot of the comments we received 8
a lot of interest, and it's a little bit of a hard 9
problem because we don't have an easy test standard.
10 We just can't pull a test standard off the shelf and 11 here is the IEEE or NFPA how to test a HEAF and get 12 all the information you need, much like a fire barrier 13 or something simple, penetration seal. So we're 14 having to put the test plan together and do it as we 15 move along.
16 There's an IEEE standard on arc faults.
17 I believe it's C37.20.7, and that's a good starting 18 point and it helped us out with some of the things 19 like the shorting wire, where to locate it, the size, 20 the ASTM slug for the calorimetry, but we're also 21 relying heavily on the National Institute of Standards 22 and Technology, NIST. Their thing is measurement.
23 They've got a whole fire measurement laboratory to 24 work with us.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
7 We're also reaching to Sandia to get some 1
of the experience from Sandia to help us put this 2
together. And most recently we've been working very 3
closely with EPRI under the MOU to again bring 4
additional talent to this to bring this together.
5 So that's a little bit by way of 6
background and our transparency. And another step to 7
the transparency is the comments that we've gone 8
through. We actually made the dispositions in that 9
and publicly released it. You can see the ML number 10 there for that.
11 January we held a public meeting. Kenny 12 did a very nice job of putting the package together.
13 Again, the ML number is there. It's in the 14 presentation and you can download everything from the 15 public meeting. And one of the things we walked away 16 from with the public meeting was that -- Victoria, you 17 asked that we'd have more interaction, which is the 18 purpose for today's meeting, to have additional 19 interaction to discuss the testing.
20 Kelly?
21 MS. VOELSING: Mark, I was wondering if I 22 could ask on the -- I realize we're talking about 23 comments on testing on a test plan that already 24 occurred and Mike mentioned -- you know, you mentioned 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
8 timeliness and transparency, but as far as I know we 1
still haven't seen any data or results from a previous 2
testing. So where are we on that?
3 MR. MELLY: We have put -- at least as far 4
as the working group is concerned, we put a lot of the 5
raw data associated with the temperature probes, the 6
-- from NIST onto the Box sharing site that we're 7
using collaboratively. There are certain aspects of 8
the testing that occurred in September where -- that 9
we have not fully processed all the data yet to put 10 onto that Box site, primarily in terms of the 11 conductivity sampling as to the carbon tape and the 12 Aerogel probes.
13 A lot of that hasn't been done yet because 14 we wanted to identify how it was going to be used in 15 the modeling space before we do a lot of the resource-16 intensive legwork of actually processing all that 17 data. We wanted to -- it to inform our modeling and 18 before we figured how we're going to incorporate it in 19 the modeling. We didn't want to just go batch and 20 sample 200 or so sampling points because we used a lot 21 of different locations to collect data. We want to 22 make sure that we're using the smartest technique to 23 process that data.
24 MS. VOELSING: But some of it's pretty 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
9 fundamental and easy material characterization testing 1
that hasn't been done.
2 MR. MELLY: Some of it is.
3 MS. VOELSING: And it seems pretty 4
important to informing future comments on future 5
testing.
6 MR. MELLY: Yes, and we've done some 7
preliminary work to putting it out, but it hasn't been 8
processed in totality yet. So we haven't put it on 9
the Box.
10 MS. VOELSING: Is there a timeline for 11 that?
12 MR. MELLY: Gabe, is there a timeline?
13 MR. TAYLOR: No, like Nick said, the 14 modeling is driving the analysis, so the preliminary 15 stuff is on the Box, everything that we have.
16 MS. VOELSING: And what about the data 17 that was lost or missing?
18 MR. TAYLOR: So there was data lost during 19 one of the experiments due to human error, so we won't 20 capture that data. There was an issue with the data 21 acquisition.
22 MR. MELLY: And we have addressed that 23 with the test laboratory as to how to prevent that 24 moving forward. It happened in one out of the three 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
10 of -- or one out of four of the days. We --
1 essentially human error. Someone pressed the button 2
twice to record the data and we did not record the 3
data for the AASTM probes for that test.
4 We did have redundant data from the NIST 5
thermocouples as well as tungsten slugs, however, that 6
was on one of the test days with the largest energy 7
release, so some of their data was lost due to 8
damaging the instrumentation wire and things like 9
that. So it was unfortunate that the NIST -- or that 10 the KEMA data was also lost on that day. But we're 11 dealing with all of that as it comes in and I believe 12 the raw data from KEMA has been shared on the Box site 13 as well.
14 MS. VOELSING: I think it would be very 15 helpful if we could list all the types of data that 16 you're expecting and whether or not you currently have 17 plans to analyze it and when we expect results to be 18 available. And then we could comment and say, well, 19 we understand that modeling is driving it, but we 20 think this one is really important to our future 21 comments if -- and understand when there's a plan to 22 have that information available.
23 MR. MELLY: Okay.
24 MR. TAYLOR: Kelli, I just got feedback 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
11 from people on the phone. They're having trouble 1
hearing us --
2 MS. VOELSING: Oh, sorry.
3 MR. TAYLOR: -- so if we can move up 4
closer.
5 MS. VOELSING: Might have trouble hearing 6
me anyway.
7 MR. MELLY: Moving forward if we had 8
comments from outside the room, I'll try and repeat 9
them for the folks on the phone.
10 MR. SALLEY: Yes, and Kenny's going to 11 keep a running tab here of what we get at the meeting 12 as far as comments and such, so you'll see him going 13 back and forth to screen.
14 And, Kelli, that's a valid point. And 15 again, we had human error with some of the operators 16 that came on one of the biggest tests, which is -- it 17 hurts us a bit as well as we lost I believe some of 18 the NIST data, too, with -- we actually burned through 19 our leads.
20 MR. MELLY: Yes.
21 MR. SALLEY: -- which anybody who made the 22 RIC, if you saw the presentations from Sandia that 23 Anthony made on using the video as data, that will be 24 a third source of information. Hopefully we'll be 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
12 able to recover some of this from Sandia's video. So 1
again, we're looking at it from multiple points.
2 MR. MELLY: But as part of the working 3
group with EPRI we did identify a lot of action items 4
that came out of our in-person meeting. That wasn't 5
one of the primary ones, but maybe it's one that we 6
need to add as to establishing a firm timeline for 7
measurements taken and when data will be available on 8
the Box web site.
9 MS. VOELSING: Yes, Marko and I were just 10 talking about it this morning. And I understand this 11 meeting we're really talking about comments that have 12 already been provided and how those comments are 13 dispositioned and are we all in alignment, but we were 14 just kind of discussing -- we kind of have to reserve 15 the right to say -- our comments are out there until 16 we seen some of the test results that we haven't seen 17 at all yet. So that was the reason for asking the 18 question.
19 MR. MELLY: Right.
20 MR. CHEOK: Fair enough.
21 MR. TAYLOR: Can I ask a question?
22 MR. SALLEY: Sure, Gabe.
23 MR. TAYLOR: So is there any test data in 24 particular that you're interested in?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
13 MS. VOELSING: Material characterization.
1 MR. TAYLOR: Right. So what type of 2
characterization are you looking for because if you 3
look at the small-scale tests, I mean, there is 4
probably six or seven different analysis techniques 5
that they can do it and characterize various aspects 6
of the particles. So maybe if we'd get some feedback 7
on exactly what are you looking for and where, is it 8
near-field, far-field, that might help us better 9
prioritize getting certain data sets together. So if 10 we could work on that.
11 MS. ANDERSON: I mean, I think chemical 12 composition was the big one, right? Like the chemical 13 composition we'd want both near-field and far-field.
14 I mean --
15 MR. TAYLOR: Right, and there's different 16 ways to analyze that. So if you said you wanted to 17 use a scanning electron microscope or whatever the 18 other approaches were, that would help us better 19 prioritize. But I don't need an answer right now.
20 MS. ANDERSON: I don't know if we've 21 gotten from -- I don't know if that's been presented 22 to -- like I don't -- I mean, maybe that's presented 23 in the working group, like here --
24 MR. MELLY: It was in the small scale.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
14 Small scale tested it.
1 MS. ANDERSON: -- are the analysis 2
techniques available.
3 MR. TAYLOR: It was in the small-scale 4
test plan, but the methods have been available. So if 5
you could get back to us on exactly what you're 6
looking for: size, speed, species, whatever it is, we 7
can kind of see how we'd move forward.
8 MR. SALLEY: So picking it back up, as we 9
look at the operating experience on the HEAFs, again 10 the root cause and what we've seen looking at 11 different events, there's a number of different 12 failures that can influence the test and how it is 13 performed. Again, with our testing is we're trying to 14 generate data to better understand the physics and the 15 phenomena of the high energy arc fault which we're 16 working with Sandia to hopefully have a model that we 17 can dial in.
18 Right now the way we're doing it with 19 6850, Appendix M, it's the 2001 SONGS events, so it's 20 a one-size-fits-all. We believe we can do a much 21 better resolution with this if we get a good model and 22 we inform it.
23 Again, with just these -- some of the 24 failures; and you've seen these if you've looked at 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
15 the OpE, there's also a variation in electrical 1
equipment, and that's something that we need to think 2
about.
3 Next slide. Previously we talked a lot 4
about duration, so I really don't want to dwell on 5
that, but we did again look at the OpE and for the low 6
and medium-voltage we can see that the duration is 7
more than milliseconds in cycles, that the HEAF is 8
unique because we are looking at some form of failure 9
where it stays locked in.
10 Next slide. A question or a comment we 11 had from some of the industry execs was to get the 12 risk piece up front. And of course that starts 13 looking at things like the frequency of the event.
14 Again, working with EPRI; Kelli, you probably know 15 more about this than I do, is getting this together 16 with the frequency of it.
17 One way that we're trying to work a lot 18 smarter than we did in 6850 was with Bin 15 was so 19 inclusive of electrical enclosures that the HEAFs were 20 trying to subdivide different things with that so that 21 we can get a more defined resolution of the HEAF 22 events.
23 Kelli?
24 MS. VOELSING: I was going to say that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
16 would -- I think the working group has made good 1
progress on this one. I've been hearing good 2
discussions going on and good alignment. And we 3
definitely think the frequency is a piece of it --
4 MR. SALLEY: Definitely.
5 MS. VOELSING: -- but the frequency is not 6
going to allow us to evaluate risk without an 7
appropriate zone of influence. So it's only one piece 8
of it and we still have to have the zone of influence 9
in order to evaluate the risk.
10 MR. SALLEY: We agree.
11 MR. CHEOK: So I think I agree with you 12 all that the frequency is really important and the 13 zone of influence is just as important. And what I 14 said earlier was I think when we comment on the test 15 schemes and things like that, I think we might want to 16 focus on comments on the testing that would jive with 17 the frequencies of the HEAFs.
18 So in other words, I know when I talk to 19 Nick or Gabe and we're talking about some test 20 parameters and I -- my question to them always has 21 been, okay, so in the end -- or to guide us we still 22 will need to get the risk out of this. So to get a 23 risk out of this we need to have some frequency 24 definitions to be consistent with the parameters of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
17 the test.
1 So in other words, so let's do tests that 2
are now consistent with how we are going to be 3
defining the frequencies. I mean, that's kind of what 4
I meant by let's focus our comments on getting the 5
test parameters and the test characteristics so that 6
we can match the frequency.
7 MS.
VOELSING:
The intention was 8
considering -- start with the end in mind --
9 MR. CHEOK: Yes.
10 MS. VOELSING: -- and considering that up 11 front. Absolutely I'm in agreement with that. I just 12 was concerned when you said we got a -- an the 13 executive for saying we need to calculate the risk up 14 front. I was pointing out that we're still going to 15 need the zone of influence.
16 MR. CHEOK: Yes.
17 MR. SALLEY: Next slide. And moving 18 along, again with the testing, too, and we want to do 19 the testing in the most realistic manner as possible.
20 A couple things here that -- with our test program 21 we're not trying to recreate any specific event.
22 We're not trying to say, okay, SONGS 2001, let's 23 recreate this failure and see if we can do that.
24 Now that's a little bit different than 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
18 what our colleagues in Japan are doing. We're working 1
very close with our colleagues in Japan. And their 2
thing of course is the Onagawa event. And they really 3
want to understand that and they tend to really want 4
to focus in on that one event.
5 Ours is a little more macroscopic rather 6
than microscope, looking at a specific event, to look 7
at a larger phenomenon of HEAF, but still we want to 8
stay grounded that when we do our testing there's a 9
form of realism to it, that it's a realistic test.
10 And looking at some of the post-test photos we can see 11 that we think we're achieving that.
12 Next slide, Kenny.
13 MS. ANDERSON: I mean, don't think you can 14 necessarily just look at photos and say that the 15 testing -- that that proves that the testing was 16 realistic just because the visual end-state might kind 17 of be similar. Maybe your eyeballs are better than 18 mine, but I don't think you can discern that the 19 inputs and the conditions were necessarily reflective 20 of realism just because it looks the same.
21 MR. SALLEY: You're absolutely right, 22 miss, and you can take a look at this next test for 23 example when we get to bus ducts. There's no way the 24 KEMA facility can generate the power that a nuclear 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
19 power plant can, so we can never achieve those types 1
of events.
2 An important thing with the -- you can see 3
the damage in the duct. And important thing too and 4
what we learned with the Zion bus is we had copper 5
buses and we thought, okay, we're good with the 6
copper, but then it migrated to the aluminum. So we 7
saw that phenomena there that brought that to the 8
forefront.
9 Next slide. Again, the latest thing is; 10 and we talked about this in January, was that we were 11 teaming up under the MOU with EPRI to bring all the 12 best people together to work on this. We have the 13 group in place. In February they spent a week 14 together and had a full week-long meeting. I think a 15 lot of good stuff got accomplished; a lot of your 16 frequency stuff, Kelli, as well as starting to look at 17 the physical models of this. So that's working well.
18 And I want to close up just on a final 19 note here and I guess compliment EPRI on a document 20 that was released last week. It's EPRI Technical 21 Update 3002015459. It's publicly available. It's 22 titled The Critical Maintenance Insights on Preventing 23 High Energy Arc Faults dated March 15th, 2019. And I 24 see that as a very, very positive step moving forward.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
20 And again, we've heard from a lot of the 1
executives at different RIS meetings that prevention 2
is very key. Really want to -- the real success of 3
HEAF is you could possibly do a very good job 4
preventing them. I believe this is a step in the 5
right direction. And if you think back, when we had 6
the last Fire Forum in Charlotte I believe in 2017, we 7
talked about -- one of the things was a long-term 8
solution to HEAF. And again, and the way we do 9
business it's a defense-in-depth and I believe this 10 comes into the first layer of prevention where we talk 11 about things like safe work practices, maintenance and 12 arc-proof cabinets, if needed, but that prevention 13 element.
14 So again, Kelli, I think your folks did a really 15 good job getting that out and it's very timely.
16 So that's kind of the 10,000-foot quick 17 overview. And again, this is your meeting, so with 18 that I will turn it -- Marko, I believe you have some 19 stuff on the working group?
20 MR. RANDELOVIC: Yes. So my name is Marko 21 Randelovic. I am senior technical leader for EPRI.
22 MR. CHEOK: Mr. Marko, would you please go 23 on -- can we go around the room and just identify 24 everybody for -- since we didn't do that? We did 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
21 it --
1 MR. HAMBURGER: Sure.
2 MR. CHEOK: -- for the phone people, but 3
not the room people.
4 MR. HAMBURGER: Sure. We'll go around the 5
room and everybody will please state their name and 6
what organization who you're here representing. Start 7
with the outside.
8 MS. VOELSING: Kelli Voelsing with EPRI.
9 MS. BERGMAN: Jana Bergman.
10 MR. LEJA: Casey Leja, electrical for 11 Exelon.
12 MR. THAGGARD: Mark Thaggard, NRC, Office 13 of Research.
14 MR. BOYCE: Tom Boyce, NRC, Office of 15 Research, Generic Issues Program Manager.
16 MR. GARDOCKI: Stan Gardocki, Generic 17 Issues Program.
18 MR. METZGER: Brian Metzger, NRR Fire 19 Protection.
20 MR. GUNTER: Paul Gunter, Beyond Nuclear.
21 MR. MILLER: Kenn Miller, NRC, Office of 22 Research, Electrical Engineering.
23 MR. TAYLOR: Gabe Taylor, Office of 24 Research.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
22 MR. CHEOK: Mike Cheok, Office of 1
Research.
2 MR. MELLY: Nick Melly, Office of 3
Research.
4 MR. SALLEY: Mark Henry Salley, Office of 5
Research.
6 MR. HAMBURGER: Kenny Hamburger, Office of 7
Research.
8 MR. REYNOLDS: Ron Reynolds, Exelon.
9 MR. RANDELOVIC: Marko Randelovic, EPRI, 10 Risk and Safety Management.
11 MS. ANDERSON: Victoria Anderson, NEI.
12 MR. HAMBURGER: Okay. So, Marko, back to 13 you.
14 MR. RANDELOVIC: Okay. So I prepared 15 couple of slides to go through the status of the 16 resolution of EPRI's comments that we have been 17 working on with the NRC Research for the last couple 18 of months. And since the effort kind of required 19 significant collaboration between EPRI and NRC, I 20 welcome Nick jumping in the presentation and provide 21 some additional details as we are discussing different 22 items.
23 So we initially had 29 comments and 25 out 24 of 29 comments were addressed and resolved. There 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
23 were four comments that were left and needed 1
additional discussions between us. Since the initial 2
review EPRI has done additional research on the HEAF 3
events in the U.S. and with that growing state of 4
knowledge we had some additional recommendations and 5
suggestions for the next round of testing.
6 So with the old comments and the new 7
suggestions we decided to schedule a workshop in 8
February. We had colleagues from the NRC Research, 9
EPRI, industry experts, electrical, fire PRAs. We had 10 Sandia. And we worked for four days on different 11 topics including heat frequency and the testing. And 12 overall the meeting was very productive. I was -- I 13 personally liked the dynamics between members and we 14 ended up following the meeting assigning the action 15 items and we are making good progress regarding the 16 testing setup and the next testing.
17 MR. MELLY: Yes, and these -- the working 18 group meeting, the in-person meeting did go very well 19 and there were a lot of issues that needed to be 20 discussed kind of on a technical level in a lot more 21 detail than can be done in a simple -- here's our 22 comment to the test plan, here's a response to the 23 test plan. So we went over each of these four 24 comments that were previously on that comment 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
24 resolution sheet in a lot of detail. We discussed 1
working group consensus moving forward and we 2
discussed kind of how we're addressing those comments 3
either explaining what NRC Research is doing, how it 4
relates to the test plan or ways that we can answer 5
those questions.
6 So we will be making that discussion and 7
the elaborate -- or elaboration of those issues 8
publicly available post-this type of meeting once we 9
have NEI's issues also more well-understood.
10 So we can move on.
11 MR. RADELOVIC: To go to the next slide, 12 yes.
13 So here are the four comments that 14 required additional discussions where we actually 15 brought more technical research and findings. So the 16 first one was EPRI's position -- initially EPRI's 17 position was that -- measuring the end-state of the 18 combustion cloud may not be sufficient to really 19 characterize the electrical properties of the fine 20 particles as they travel. And we suggested that we 21 perform the dynamic conductivity measurement.
22 So in the same time while we were looking 23 into this the NRC Research, the Testing Team has 24 already been looking into that. So when we met at the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
25 workshop the Testing Team had already had a design in 1
mind and we discussed the performance of the 2
conductivity probes, the issues that we might be 3
facing and how to interpret the results.
4 So, Nick, if you want to provide some 5
updates and --
6 MR. MELLY: Yes, this gets back to an 7
issue we were discussing earlier as to what are the 8
material properties of the ejecta that we're seeing on 9
these carbon tape samples or the Aerogel samples, or 10 is there a better way to measure the conductivity of 11 the cloud, of the ejected material from one of these 12 events and to figure out if it's going to have an 13 impact on other electrical systems within the room, 14 either energized or relay cards or things like that.
15 So we were trying to enhance the way that 16 we're doing our measurement techniques in terms of 17 gaining conductivity results either in a delta format 18 or static or dynamic, or what are better ways that we 19 can assess the condition if we had a conductive 20 environment.
21 So on the screen here we say that we do 22 have a need for dynamic conductivity measurements.
23 We've been talking with NIST and other -- and Sandia 24 and we're trying to work out the kinks of how we can 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
26 do that. We have presented some ways that we can move 1
the conductivity measurement forward. And EPRI also 2
volunteered a conductivity type of approach for an 3
alternate power supply that they'll be discussing 4
later in the presentation which we think could be 5
valuable to getting more information as to the 6
environment in a HEAF event.
7 So we'll dive into that a little bit 8
later, but it's something we are actively moving 9
forward and we're enhancing the test plan to focus on 10 this area.
11 MR. SALLEY: You know, one thing, if I 12 could add to it, the dynamic measurement is going to 13 be very difficult. The after-test measure is a lot 14 more doable, and again it's a lot more cost-effective.
15 We've reached out in the past week or so to our 16 electrical colleagues. Kenn Miller is here and Tom 17 Koshy who have a lot of experience with this and they 18 have some ideas and some suggestions on how we could 19 analyze this for conductivity. We're trying to get 20 that together with Tom; he's in an IEEE meeting in 21 Germany, when he gets back. And we're going to work 22 with Sandia to just try some proof of concept, if you 23 will.
24 Again, this is an area where not a lot has 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
27 been done. So we're trying that and when we find that 1
out, we'll be sharing it with the working group.
2 MR. RANDELOVIC: So, yes, because of that 3
lack of experience regarding the conductivity probes, 4
how they perform, how to read the results, how to make 5
sense of the results, we kind of took an option to 6
design a mock switchgear. I actually have a slide on 7
this, to be more representative of what we are seeing 8
in the plant. And we -- I will explain a little bit 9
later.
10 The second comment was regarding the 11 methodologies for the evaluation of the zone of 12 influence and the fragility assessment for cable 13 trays, cables and electrical components. We -- the 14 working group has not seen at that point the 15 methodology, how we are going to use the test data to 16 establish this. So we wanted to have a little bit 17 more input from the NRC Research and Sandia on how the 18 data will be used. And so, Sandia had a presentation 19 and NRC Research presented the overall methodology, 20 which sounded reasonable.
21 Nick, if you want to --
22 MR. MELLY: Yes, this gets down to linking 23 the data that we received from the modeling in terms 24 of our heat flux that we're receiving at our 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
28 measurement devices to a fragility or a target 1
fragility, which was ranked high in the PIRT 2
assessment as to I have a heat flux from my measured 3
test. How do I link that with am I going to damage a 4
cable?
5 We know there are thermal response times 6
to damaging cable in and of itself, and we're talking 7
about a quick temperature rise, a quick influx of 8
heat. So we're going to be working with Sandia to 9
link the data that we're receiving from testing to 10 fragility assessment stages either using their solar 11 tower facility or other techniques that Sandia has 12 available to them to do more of a small-scale testing 13 approach to link the data that we receive from actual 14 testing to fragility values.
15 We're currently in the contractual phase, 16 right, Gabe --
17 MR. TAYLOR: Yes.
18 MR. MELLY: -- to get this out the door 19 and we're going to be sharing all that information 20 with EPRI and the working group as we move it forward.
21 MS. VOELSING: Will that information on 22 the front end include a V&V plan for how this model is 23 going to be validated?
24 MR. MELLY: Do we have a V&V plan right 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
29 now?
1 MR. TAYLOR: We don't even have their 2
proposal yet on the models, how they're going to be 3
modified. I mean, the models that they currently have 4
do give V&V, so until we have their proposal we can't 5
really answer your question.
6 MS. VOELSING: Okay. But I would think as 7
part of planning this that there would need to be a 8
plan for how V&V those models for the specific 9
application that you're planning to use it for.
10 MR. TAYLOR: Right.
11 MS. VOELSING: And you need to know what 12 that looks like so that you collect the right data to 13 support that. So I'm just saying V&V has got to be 14 considered on the front end.
15 MR. MELLY: Yes, absolutely. And this is 16 actually an item that came up during the working 17 group. We wanted to make sure that we did have a V&V 18 approach in there.
19 And, Chris, I don't know if you're -- you 20 want to elaborate a little bit on this as to the V&V 21 that is publicly available on the models that we were 22 discussing using.
23 MS. LAFLEUR: Yes, this is Chris. Yes, 24 our Aria programmer has told me that he'll be sending 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
30 me publicly available V&V documentation. And I'll 1
forward that to the working group and put it on that 2
Box so that it's available to everyone.
3 MR. SALLEY: There's also another data 4
point, too, as an alternative to this. We've talked 5
with NIST. If you remember Kevin McGrattan and the 6
Carroll Fire Program created the THIEF model that we 7
use for cable fragility for the standard fire 8
dynamics. We've met with Kevin and we've asked him to 9
look at the process that he developed THIEF with, if 10 there's something he could do similar to HEAF -- for 11 HEAF. So again, NIST as an alternative is also 12 looking at this.
13 MR. RANDELOVIC: So the overall outline on 14 how we are using -- going to use the data to what the 15 codes should do and how they would perform as I said 16 is reasonable. Now the work has to be done and we 17 have to see how these codes are working, how they are 18 benchmarked, and that is an important step and that 19 would be under the purview of the working group who 20 would be following each step as the process goes.
21 The next comment was regarding the 22 extrapolation of the low-voltage test results to 23 medium-voltage switchgear testing. Our comment was 24 that because the energy levels are different between 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
31 medium-voltage and high-voltage. The geometry. The 1
load center is in the switchgears are not exactly the 2
same and we were just -- we want more discussion and 3
details on how -- what this approach looks like and so 4
that we can judge if it's feasible or not.
5 Nick and the NRC provided a high-level approach.
6 Nick, if you want to explain a little bit?
7 MR. MELLY: Yes, we just kind of just at 8
the working group meeting discussed how we were 9
planning on using the eight-second low-voltage test to 10
-- as a data point to inform the medium-voltage 11 longer duration test because KEMA doesn't have the 12 ability to test at longer duration for medium-13 voltage. It's not something that we can do as a one-14 to-one evaluation because there are a lot of things to 15 take into consideration: the differences in power 16 levels, the differences in the arc voltage itself.
17 So this isn't a simple approach and it's 18 something that we're going to be relying on the 19 working group to evaluate the data to see if this is 20 even feasible or possible once we have the data.
21 So we do understand the hesitation with 22 saying that this longer-duration, low-voltage test ]KC 23 be directly comparable to a medium-voltage test and 24 we're not recommending that that's how we look at 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
32 things. We're going to be getting the data from this 1
test and making the decision can this be used at all 2
in an extrapolation method?
3 Some of the other interesting things that 4
came out of the working group was that we had a lot of 5
discussion that these low-duration -- or I'm sorry, 6
these low-voltage, longer-duration events are 7
relatively rare or not feasible, but from looking at 8
the operating experience we actually did find several 9
events. The Fort Calhoun event lasted 42 seconds.
10 And there was also a -- I believe it was 11
-- was it Robinson? No, it wasn't Robinson. The 12 River Bend event we identified a low-voltage event 13 that had been previously missed, classified as a 14 medium-voltage event that actually did last for 12 15 seconds.
16 So we are seeing that this eight seconds 17 for low-voltage may be appropriate. And one of the 18 other action items we were taking into account is 19 let's look at the realistic current value that we can 20 expect if we have a stuck-in, low-voltage event. So 21 we're trying to take that into account when performing 22 the testing. It may be at a lower current and it 23 could be based on the set points of the breaker 24 themselves. So that actually gets to the next point.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
33 MR. RANDELOVIC: That does get to the next 1
point.
2 MR. MELLY: Yes.
3 MR. RANDELOVIC: So that we are investing 4
currently those events and if the current is low, we 5
agree that the next round of tests we'll basically 6
replicate what we are seeing in the OE. So it should 7
be fairly prototypical.
8 MR. MELLY: Yes.
9 MR. RANDELOVIC: Next slide.
10 MR. MELLY: And so those are the easy 11 comments. Those the ones that were previously --
12 (Simultaneous speaking.)
13 MR. RANDELOVIC: Those are the previous 14 ones.
15 MR. MELLY: Now there's a whole new set of 16 new ones that we'll address. This was a very 17 productive meeting. It was four full days of 18 discussion.
19 MR. RANDELOVIC: As I said, as we -- EPRI 20 was investigating and digging deeper into the 21 operating experience and the HEAF events in the U.S.,
22 we actually learned a lot and formed the technical 23 basis for some recommendations regarding the test 24 changes and regarding the use of the additional 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
34 instrumentation. So we -- during the workshop we 1
brought the technical basis with us and we discussed 2
with the team the proposed changes for the medium-3 voltage test configuration and fault location; and I 4
have a slide on this, and the low-voltage load center 5
test design.
6 And as I said, the conductivity probes, we 7
have low state of knowledge on how they perform and 8
what kind of data we can gather from them. So EPRI 9
took a lead on designing the representative energized 10 mock switchgear that would be used for the measurement 11 of the conductivity in the combustion cloud. So we 12 are providing the support for design and for the 13 technical specification. And as I said, the working 14 group acknowledge the need and the value for such a 15 device.
16 Next slide, please. So research on the 17 medium-voltage switchgear suggests that the majority 18 of the medium-voltage switchgear events occurred in 19 the supply switchgear configuration and the majority 20 of the faults occurred at the supply breaker stabs and 21 the main bus bars. So you can see on the figure --
22 I'm sorry for those on the phone. So we have the 23 arrows where it shows that the -- where the majority 24 of the faults occurred.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
35 So our recommendations to the Testing Team 1
is to perform at least one test with the arc wire on 2
the main bus bars in order to replicate the majority 3
of those events and to be able to understand the 4
difference in geometry in the arc itself and the 5
importance of the cubicle where the arc is located 6
because currently we have the tests with the arc wire 7
and the main bus bars on the -- the bus bars over 8
here. What we're suggesting is to raise the arc wire 9
over here to perform a test and to see the 10 differences. Is there any difference in the zone of 11 influence or the arc behaviors, arc dynamics?
12 MR. MELLY: Yes, so this one is kind of on 13 the cutting room floor. We identified this at this 14 workshop and we put some more work into looking into 15 this. And the NRC is looking right now at the 16 feasibility of doing a confirmatory test to see is 17 this a primary impact driver for what the zone of 18 influence or the resultant energy release could be?
19 So we are looking into how we can 20 reconfigure one of the cabinets that we tested in 21 September. In a future test in a confirmatory manner 22 see is this really going to drive a difference in the 23 energy release? It's kind of flipping where we put 24 the power from our test facility into the cabinet 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
36 itself and then switching the arc location. So we are 1
going to look into is this possible in terms of how 2
much money we have to do testing, timeline of testing, 3
availability, knowing that there's a limited number of 4
tests we can do. There's limited budget. And what 5
are our primary interests in testings?
6 One thing we're not talking about here is 7
we also have already said that our spare tests and the 8
test matrix were going to be used to evaluate the 9
decrement curve approach. So right now we're looking 10 into that decrement curve and the additional tests 11 that we've already budgeted for is do we need four 12 decrement curve tests? Can we do two decrement curve 13 tests and maybe do two confirmatory tests? Or what 14 other tests in our current budget can be used in a 15 confirmatory nature when issues like this come up?
16 MR. RANDELOVIC: At the end we will have 17 to do the tests that satisfy and then remove any 18 uncertainties in how the data could be used in the 19 actual replication.
20 MR. MELLY: Yes.
21 MR. RANDELOVIC: If we can justify 22 technically that not doing a test -- not to do the 23 test because we have a strong technical justification, 24 then the working group will decide and we can proceed.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
37 But if we feel that we don't have enough knowledge to 1
explain the differences, then again we will have to 2
evaluate at the working group level and how to 3
proceed.
4 MR. CHEOK: So, Marko, thank you for that.
5 I think when you heard Nick say confirmatory test, 6
that's kind of what I was alluding to. I don't think 7
we can afford money-wise or time-wise to replicate all 8
our previous tests moving the location of the arc. I 9
think we -- let's do one confirmatory test, too --
10 MR. MELLY: Yes.
11 MR. CHEOK: -- to see if this is going to 12 have an effect before we move forward. For now the 13 working group suggests only one confirmatory test on 14 the main bus bar location. And then I think from 15 there we can look at the differences between these two 16 locations and we can determine if there is a big 17 difference. If there is no big difference, then it's 18
-- could be okay.
19 MR. TAYLOR: Do we know how common it is 20 for those main buses to be aluminum versus copper?
21 Because that might also influence whether we test with 22 the OECD Program or the ones we're sponsoring.
23 MS. ANDERSON: I thought it was pretty 24 uncommon.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
38 MR. RANDELOVIC: So you have two. So you 1
have the aluminum and you have copper. I just can't 2
tell you how --
3 MR. TAYLOR: That's -- I just --
4 MR. RANDELOVIC: -- how many plants have 5
the bus bars, aluminum or copper. I don't have that 6
information yet.
7 MR. TAYLOR: I mean, if we could get that 8
information, I think --
9 PARTICIPANT: We didn't do a study?
10 MS. ANDERSON: Yes, I thought it wasn't 11 that many.
12 PARTICIPANT: It wasn't that many.
13 MR. REYNOLDS: No, there should be no 14 copper on a main bus bar.
15 MS. VOELSING: Yes, on the supply side it 16 was mostly all copper and some on the right side had 17 aluminum bus bars. There are a few but copper is 18 still predominant. But that doesn't mean --
19 MR. TAYLOR: I mean, it would help us just 20 to determine what tests we should do that, one with 21 the copper and one versus -- sorry, one with the 22 aluminum, because when we were looking at procuring 23 the low-voltage stuff, we had a piece of gear that --
24 it was a split. It was half copper and half aluminum.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
39 And from the vendor they said that's a pretty common 1
design. Now this was low-volt. It was not medium-2 volt. So I mean, it would help us out if we could --
3 MR. RANDELOVIC: We can bring this at the 4
working group level and we would investigate.
5 MR. MELLY: Yes, it could also help our 6
understanding of is this an issue that we want to 7
handle in testing configuration space or is this 8
something that we think we can handle in the modeling 9
space? So can we account for the fact that will I 10 have copper in -- on the supply side or look at a 11 cabinet configuration-wise and I will only have copper 12 on my breaker? So maybe the aluminum --
13 MS. VOELSING: But aren't you tuning the 14 model for accurate testing?
15 MR. MELLY: No, because if we tune the 16 model to say that I have only copper in the breaker 17 stabs -- so the data that's -- I'm gathering from 18 aluminum, if I look at my cabinet in a different 19 aspect, I don't want to apply aluminum-type zone of 20 influence to my breaker cubicle itself if I know I 21 don't have aluminum there.
22 MS. VOELSING: You're going to need a 23 copper zone of influence, too.
24 MR. MELLY: And we think that we have data 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
40 from the Test Series 1 to inform the model for copper-1 initiated arcs.
2 MS. VOELSING: Then why haven't we already 3
validated the model and verified that this all work 4
before we move forward with testing?
5 MR. MELLY: It's being done as we speak.
6 MS. VOELSING: But you don't have a V&V 7
test plan.
8 MR. MELLY: All of it is -- this is ally 9
running in parallel right now.
10 MR. SALLEY: I'll caution you, too, to 11 mixing the different materials up and go back and look 12 at Test 26 from the first phase, the Zion bus duct, 13 because that was one that caught us looking where we 14 had copper conductors, if we had the aluminum 15 enclosure, and we saw the interaction between the 16 copper and the aluminum. So the answer may not be 17 obvious.
18 MR. REYNOLDS: What's the weight of actual 19 location of the arc as opposed -- like you're showing?
20 What weight is put on that in respect to testing?
21 MR. MELLY: As a parameter of interest 22 when we look at this the aspect of the PIRT it wasn't 23 a very important factor. The breakers locked in in 24 testing, so whether we come in from the rear or the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
41 front of a cabinet, in my opinion I don't think it 1
will have a very large impact. It will affect the 2
directionality of the arc itself, however, until it's 3
tested there's no way to say it's going to be a one to 4
one ratio versus a -- or if it has a bigger impact 5
that I'm not anticipating.
6 MR. REYNOLDS: Is that the intent of the 7
testing for that is --
8 MR. MELLY: Yes.
9 MR. REYNOLDS: -- also to determine where 10 that --
11 MR. MELLY: Right.
12 MR. REYNOLDS: -- impact of the location 13 is?
14 MR. MELLY: Yes.
15 MR. RANDELOVIC: Right, so if you look at 16 the bus bars here, the arc -- these are the horizontal 17 bus bars, so the arc is horizontal. When you look at 18 this one, so those are the three bus bars, because you 19 see the arc. There is a difference in geometry in the 20 arc. Cubicle itself is not the same. Space, the 21 location of the arc. The other side, you have the 22 differences in geometry in the arc itself and the 23 cubicles where the arc is. So how -- what is the 24 difference between those two, I don't know and I don't 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
42 think that anyone can say with 95 percent confidence 1
level they are the same.
2 MR. REYNOLDS: Without testing.
3 MR. RANDELOVIC: Without testing. So that 4
is why we were suggesting if we performed this one 5
test with the arc location there and the previous OECD 6
tests I think you guys had placed the arc wire, the 7
breaker stabs. And we are looking to -- gathering the 8
data from the first round of tests, if feasible.
9 MR. TAYLOR; So this is just me personally 10 speaking. I think there's a valid and reasonable 11 comment in the feedback on the material types that 12 would probably help us better inform what we want to 13 do moving forward. So I don't think we can say, yes, 14 we're going to do it, but we're definitely considering 15 how we can adjust the test plan to move forward.
16 The other thing I just want to point out 17 on this is that those main bus configuration is a 18 little different from like other vendors, most of the 19 other vendors for medium-voltage. Now these are out 20 in the field, but a lot of the other vendors have 21 three in the same plane, three phases in the same 22 plane.
23 MR. RANDELOVIC: But I mean, so you still 24 have --
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
43 MR. TAYLOR: Well, these are kind of --
1 you got two in one plane and then --
2 MR. RANDELOVIC: Right, right.
3 MR. TAYLOR: -- two in the other plane.
4 So it's a
triangle.
So there's just some 5
configurations, I think, and in the end it does 6
probably impact on the severity of the hazard. Yes, 7
we'll look into that.
8 MR. RANDELOVIC: Next slide. For the low-9 voltage events our research shows that all low-voltage 10 events occurred in the load center supply cubicle and 11 the all faults were initiated at supply breaker stabs.
12 So our recommendation for the Testing Team is to 13 supply -- to procure actually the load center supply 14 cubicles to maintain the prototypical geometry and to 15 place the arc at the same location. That's what we 16 are seeing in the operating experience. And I think 17 NRC Research is working on the procurement activities 18 right now and is considering EPRI's comment.
19 MR. MELLY: Yes, we're currently working 20 through procurement for -- of the equipment and we're 21 working closely to make sure that we'll have a 22 realistic set of equipment to test.
23 MR. SALLEY: If industry has anything 24 they'd to donate as far as equipment, we'll always 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
44 welcome that.
1 MR. RANDELOVIC: And the last suggestion 2
was the use of the mock switchgear test unit because 3
as we already discussed the conductivity probes, the 4
issues with performance and the test data. So we have 5
designed a mock switchgear unit and the purpose is to 6
see if there is a certain amount of particle 7
-- fine particles that could actually create a 8
secondary fault in the medium-voltage switchgear.
9 So this is still in preliminary design of 10 the device. It's a small box: 36 by 30 by 10. You 11 have five different ventilation panels. This is the 12 bus bar, 8 kilovolts, that will be maintained with two 13 insulators. And so as the dust cloud moves in through 14 the ventilation panels, the idea is to see if the 15 cloud itself could actually provoke a Phase II ground 16 fault. Or you have -- you collected sufficient 17 material on this -- on the insulator to create a low-18 resistance path for the Phase II ground fault.
19 MR. SALLEY: Will that be natural 20 ventilation or forced?
21 MR. RANDELOVIC: Natural. We will just 22 place -- the placement is going to be a point of 23 discussion. We may need to look at some videos from 24 the previous round of tests and located the device 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
45 somewhere where it's far from the molten ejecta 1
because we don't -- that's not the purpose of the 2
test. We were just looking at what's in the cloud 3
itself. But we are going to have the screens here 4
anyway on these ventilation panels to prevent any 5
entrance of the molten ejecta. And so this -- the bus 6
bar is going to be located a little bit further up, so 7
you actually get a cloud going in. And then we will 8
see if the cloud is conducting or you have the 9
sufficient material collecting on the internal lean 10 surface.
11 MR. MELLY: Yes, so the initial discussion 12 that we had at the working group -- this -- place this 13 at the -- behind the six-foot instrumentation rack in 14 the path of the ejecta.
15 MR. SALLEY: Okay. But did you have 16 screens or louvers up? Would it be more common or 17 more representative to have louvers?
18 MR. MELLY: It would be. And like we said 19 this is a -- this came -- yesterday this came 20 together.
21 MR. RANDELOVIC: It's a big room, so 22 having small louvers you may actually not be able to 23 get any cloud inside. So we are trying to get the 24 cloud inside the box. So we think we can actually 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
46 measure it. If we have the small louvers, then -- in 1
a sort of big space where you actually -- with the 2
device, you may not collect --
3 MR. MILLER: Not going to simulate the 4
actual equipment as opposed to actually exposing the 5
cloud to the voltage to see if that produces a fault?
6 MR. RANDELOVIC: Yes.
7 MR. MILLER: Okay. So said that's going 8
to be eight kV?
9 MR. RANDELOVIC: Eight, yes. Which was 10 calculated because in the plans the most typical 11 faults are phase to phase and with 13.8 kV, if you 12 divide by the square root of three --
13 MR. MILLER: Right.
14 MR. RANDELOVIC: -- you actually get 8,000 15 kilovolt for Phase II ground fault. So that's --
16 8,000 is equivalent to 13.8 in the plan. The currents 17 that we have is -- there are about 30 milliamps, so 18 very small. If you have a fault -- we have a -- I 19 think the spec requires like 5,000 kilohertz 20 frequency, so you're going to be able to get probably 21 5,000 data points in a second. And if you have a 22 flashover, then we would be able to see the current 23 going from 0 to 30 milliamps, which is the maximum 24 current for this transfer.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
47 MR. MILLER: So you're just doing like an 1
accelerograph measurement of leakage current?
2 MR. RANDELOVIC: Right. Yes.
3 MR. MELLY: Yes, and this is -- we will 4
have discussions on this at the working group as to 5
criteria, how we're going to measure, whether we want 6
it fused or things like that. So it's still on the 7
table right now, but this is the idea, to get another 8
data point as to the potential impact of the cloud 9
itself.
10 MR. MILLER: In parallel with the dynamic 11 conductivity measurements, of that's possible?
12 MR. MELLY: Yes, we also plan on putting 13 black carbon tape or Aerogel inside of this 14 measurement device, non -- hopefully not affecting 15 anything that we're of primary interest to, any of the 16 spacing or distance to the ground to see if we can get 17 a comparison data point, if we do see faulting or if 18 we don't see faulting.
19 MR. RANDELOVIC: That is basically all I 20 have for now.
21 MR. HAMBURGER: Okay. Any questions for 22 Marko or Nick on some of these items that were 23 discussed in the current status of 150?
24 (No audible response.)
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
48 MR. HAMBURGER: So we are now going to open 1
up the meeting. The rest of the afternoon is reserved 2
to solicit input and feedback from those in 3
attendance. And I think NEI sent us a list of flagged 4
comments that they wanted to discuss. Would you like 5
me to pull up the flagged comments and --
6 MS. ANDERSON: I mean, I think they 7
essentially boil down to two specific concerns, if you 8
look at them. I mean, so the first concern is that 9
the testing is realistic in the sense that all the 10 inputs match up with each other with respect to the 11 OE. So we're not taking the worst case duration that 12 we've seen in OE and the worst case voltages we've 13 seen in OE and the worst case material that we've seen 14 in OE, because they don't all necessarily go together.
15 So, and it sounds like the working group has worked on 16 looking at the OE more carefully and matching it up a 17 little bit better.
18 So like one of the comments we flagged was 19 with respect to the Robinson event. I think that was 20 our comment No. 20. And so it sounded like there was 21 some overstatement of the zone of influence that was 22 attributed there and that event is being -- was being 23 used to justify some of the testing parameters, but it 24 seems like there's some insight that maybe should be 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
49 considered with that specific event that would lead 1
you to not really support those test parameters. So 2
I think that's -- some of the work that the working 3
group has done really alleviates a lot of our 4
comments.
5 The other aspect, the other sort of flavor 6
of our comments is this testing plan is supposed to be 7
supporting the pre-generic issue on the impact of 8
aluminum, and sometimes it's not clear how we're 9
specifically teasing out that impact. And we talked 10 earlier about needing to do material characterization, 11 but I think it seems like maybe we can work on getting 12 some more input to you via the working group on 13 specifically where we see aluminum and what kind of 14 configurations we see it in.
15 And maybe the tests can be better 16 configured to help tease out those effects versus just 17 doing HEAF experiments and make sure that if we do see 18 any impact on the zone of influence, that it's 19 uniquely attributed to aluminum if that's what the 20 true purpose of the pre-generic issue is. We 21 definitely want to see realism in our PRAs and fully 22 account for all the potential hazards, but that being 23 said, we need to make sure that if we're saying that 24 there's an -- if we do decide there's a potential 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
50 increase in zone of influence due to aluminum, that 1
that's based on sound testing that teases out the 2
individual effects of aluminum. It is not just due to 3
testing factors. We need to make sure that we're 4
really carefully looking at what the unique of 5
aluminum is.
6 So it sounds like both of those are going 7
to be -- have been or are going to be addressed in 8
future tests via the working group.
9 MR. MELLY: Hopefully, yes. On the 10 Robinson event which parameters specifically were we 11 overestimating --
12 (Simultaneous speaking.)
13 MS. ANDERSON: So if you look at that 14 comment in the resolution, I think there was a comment 15 that the damage and the sort of the zone of influence 16 that was being assumed there were not actually 17 reflective of the OE. It sounded like perhaps the 18 vertical distance between the cabinet assumed in 6850 19 doesn't really correspond with the actual OE. So I 20 think they're saying that they noted that some of the 21 cable just three inches from the cabinet were 22 undamaged, yet we're relying on data that -- or we're 23 relying on a model from 6850 that assumes a larger 24 zone of influence.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
51 MR. MELLY: Yes. And I think the comment 1
resolution that I provided there was more so to say 2
that we did do a comparison -- we did get cables from 3
Robinson and were able to look at those for damage 4
states, however, the cables that we received, there 5
was no way to actually map those cables to the 6
location in the plant where the cables were received.
7 So it made it a little bit murky in saying that --
8 there was nothing we could do to make a determination 9
that, yes, the 6850 model was correct or, no, the 6850 10 model overestimated the risk without having that 11 mapping ability.
12 So and I think that was in the report that 13 we issued on the Robinson event.
14 PARTICIPANT: I think -- I mean, we can 15 get that information.
16 MS. ANDERSON: Yes, so I mean, I think 17
-- I understand why you did what you did because you 18 didn't have specific information, but we can work with 19 the licensee to get that information and improve the 20 realism there. We're happy to -- through the working 21 group or if you need to go through NEI, we can do 22 that. But --
23 MR. MELLY: Well, we also wanted to make 24 sure that we're not -- we're trying to be better than 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
52 the SONGS event in saying that all HEAFs are this 1
model.
2 MS. ANDERSON: Right.
3 MR. MELLY: So there is some benefit of 4
looking at the operating experience, however, what 5
we're seeing from the high energy arcing fault event 6
testing is if I direct my energy, I know that I'm 7
going to direct it all horizontally away from the 8
cabinet based on the configuration of where the arc is 9
initiated and based on where the energy is going. I 10 can run 15 tests in that orientation and I'll never 11 damage potentially cables in a vertical orientation, 12 however, if I just move the arc initiation point or 13 move the configuration of the cabinet, I will.
14 So we can gain insights from OpE, but the 15 testing is trying to get a generic approach for the 16 potential of an arc occurring in all locations within 17 the cabinet so we can try and take that into account.
18 So while we do have some anecdotal 19 information from the OpE from the Robinson event, 20 there's very little we can do to say this happened at 21 the Robinson event, therefore all potential HEAFs are 22 going to be this.
23 MS. ANDERSON: Right. And that's what we 24 just cautioned against before.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
53 MR. MELLY: Yes.
1 MS. VOELSING: But I guess could you 2
elaborate on how -- I understand you want testing to 3
represent all possible --
4 MR. MELLY: Yes.
5 MS. VOELSING: -- configurations. How 6
would you propose putting that distribution into the 7
PRA, because assuming the worst is not the right 8
answer either.
9 MR. MELLY: So we don't plan on assuming 10 the worst. That's definitely not what the working 11 group is trying to establish when we're creating this 12 model. We have a skeleton outline that's on our 13 working group agenda to describe how we plan on 14 bringing the OpE data that we have, bringing the test 15 data together in a format that can facilitate a more 16 dynamic zone of influence. So we can take into 17 account things like your plant design, your circuit 18 protection design, where your arc would be occurring 19 in the plant to try and inform potentially how many 20 layers of circuit protection I have, how long the arc 21 duration will last, and then how we link that with an 22 energy release that can tie to a zone of influence.
23 That's the current thought.
24 We're also going to try and take things 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
54 into account like where the arc is initiated in your 1
cabinet, the configuration of your cabinet. It's 2
lofty goals, but that's what the working group is 3
striving towards accomplishing.
4 MR. RANDELOVIC: So if the plant doesn't 5
have the aluminum, then you are going to have a zone 6
of influence without the aluminum.
7 MS. VOELSING: Right, but I guess there's 8
still multiple places that an arc could be initiated 9
10 MR. MELLY: Right.
11 MS. VOELSING: -- in a configuration.
12 MR. MELLY: Yes.
13 MS. VOELSING: So how are you going to 14 represent that?
15 MR. MELLY: Through the evaluation of how 16 much energy is potentially released timed back to 17 duration hopefully. And it may not be a three-foot, 18 five-foot zone of influence. It's going to try and 19 look probabilistically of what the energy release is 20 in your cabinet to more of a -- I don't want to assume 21 anything, but a spherical zone of influence based on 22 energy release, and then trying to use engineering 23 judgment to take into account things like where in the 24 cubicle -- or where in the cabinet itself the arc is 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 being initiated to potentially putting factors on 1
that.
2 MR. RANDELOVIC: I think this is where the 3
confirmatory test is crucial because now you're going 4
to relocate the wire --
5 MR. MELLY: Right.
6 MR. RANDELOVIC: -- on the main bus bars 7
and then you can compare the energy levels and 8
behaviors and zone of influence and other variables.
9 MR. CHEOK: So let me kind of restate what 10 you just said, I think, so that I understand it 11 better, right?
12 We will have enough tests hopefully to 13 replicate different cabinet and plant configurations.
14 We will not take the worst case and apply it to 15 everybody. We will -- we understand that HEAFs and 16 the risks from HEAFs is really plant and 17 configuration-specific. We will -- given the 18 different configurations that we are testing under we 19 will only apply the test for a certain configuration 20 to the plant if that configuration is applicable to 21 that plant.
22 MR. MELLY: Yes, and for instance, that's 23 the reason we're trying to isolate single parameters.
24 So that's the reason why I'm testing at two seconds 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
56 and four seconds. I'm not going to -- the goal is not 1
to link every single plant's zone of influence to the 2
most bounding test case of this is what I saw from my 3
four-second duration testing, so now this is the zone 4
of influence. The plan is to say this is the data 5
that I've obtained at two seconds, four seconds and 6
here's what it potentially could look like at eight 7
seconds, and then trying to link that with in my plant 8
configuration I will most likely have a 3.5, 4-second, 9
5-second arc. And then I can tie that energy release 10 to a more realistic energy output from the event 11 itself.
12 MR. SALLEY: You know one of the things I 13 guess is direction that I gave to our side on the 14 working group was that 6850, Appendix M, that 15 information is almost 20 years old now and that three-16 foot, five-foot box was a nice convenient screening 17 tool for PRA that they came up with back in the 2000 18 time frame to not limit themselves to a square three-19 foot, five-foot box. But I know Gabe has done some 20 looking at some IEEE standards and what they're 21 showing is more of a sphere or a different geometric 22 configuration that more accurately realistically 23 represents the HEAF. So again, that's some of the 24 stuff that I've asked them to look at.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
57 Gabe, what was the standard you were 1
looking at?
2 MR. TAYLOR: So it's our calculation it's 3
1584. But I think the point is whether it's a sphere 4
or some other shape. It's geometry-dependent. It's 5
cabinet configuration-dependent. So I think what we 6
saw from the fall test as well as Phase I of the OECD 7
Program that they're getting these directional energy 8
output and they do breach the enclosure. And I think 9
the model that we develop moving forward needs to 10 capture that directionality in some form to 11 realistically represent the energy release.
12 MR. MELLY: Yes, and one of the things 13 that we have identified from looking at the OpE is we 14 see a lot of these longer-duration high energy arcing 15 fault events that make us the frequency either being 16 generated or fed where they have no certain protection 17 to limit the fault energy or places where you have 18 circuit protection failures.
19 So our goal is to try and take into plant 20 configuration where these faults can happen in your 21 lineup at your specific plant into account very 22 heavily to try and link that with what is the 23 probability of having an extended-duration event 24 versus having a very quick fault that will only last 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
58 potentially on the milliseconds or half-a-second-type 1
approach.
2 So we're going to try and rely more 3
heavily on plant configuration in terms of 4
establishing the zone of influence as we move forward.
5 MR. CHEOK: And you're going to marry that 6
with the potential frequency of those HEAFs on an OpE 7
8 MR. MELLY: That's the plan.
9 MR. CHEOK: -- splitting fractions.
10 MR. MELLY: We'll see what we can do.
11 MS. ANDERSON: And the future test plans 12 are still going to be used for public comment, 13 correct?
14 MR. MELLY: Yes. The hesitation with --
15 so we've -- a lot of these have been incorporated into 16 the test plan as it lives right now. The shift there 17 is that the test plan in the format that it was in a 18 few months ago was linked to the OECD Agreement, so we 19 didn't want to have multiple versions of a test plan 20 out there; one on agreement; one on non-agreement 21 while we were in the signature process with our 22 international members.
Once that program is 23 officially kicked off, we will again release the 24 comments that we discussed with EPRI through the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
59 working group as well as any insights from this 1
meeting and an officially updated test plan as we move 2
forward.
3 MR. HAMBURGER: Victoria, did you want to 4
go through any of the comments since --
5 MS. ANDERSON: No.
6 MR. HAMBURGER: No? Okay. Can you --
7 would you mind just -- I'm trying to succinctly 8
summarize everyone's concerns so that we have it on 9
record and NRC can ask questions if they have any.
10 Did I accurately capture your two comments in No. 3 11 and 4?
12 MS. ANDERSON: Well, it's the pre-GI.
13 Yes, I think No. 3 is essentially more or less what 14 I'm saying.
15 MS. VOELSING: Maybe If I could --
16 MR. HAMBURGER: Certainly.
17 MS. VOELSING: -- add one more based on 18 what I'm hearing. I understand the intent and the 19 marrying of right frequencies within appropriate zone 20 of influence, but I also am hearing about this -- the 21 Sandia model being utilized to develop those zones of 22 influence. And so in order to accurately comment on 23 the test plan I think it's really important for our 24 stakeholders to be able to understand what feeds the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
60 Sandia model, how is that Sandia model actually 1
validated for this specific application and what 2
outputs are expected from the Sandia model, because 3
those things drive what we -- what data we need to get 4
out of the testing. And so to move forward with 5
testing without understanding that limits our ability 6
to comment on that.
7 And I'm sorry for the voice.
8 MR. HAMBURGER: You want to get Chris to 9
give an overview?
10 MR. TAYLOR: I can give high-level and 11 then Chris jump in, if she wants to.
12 MS. VOELSING: And I'm just saying I don't 13 know that this meeting is the right format, but --
14 MR. MELLY: No, I think --
15 MS. VOELSING: -- it needs to be published 16 and it needs to be put out for public comment.
17 MR. MELLY: I think we anticipated doing 18 all that and then holding an additional public meeting 19 that was requested at the --
20 MR. TAYLOR: So from the meeting we had in 21 January we committed; and I think it was Jennifer's 22 comment, that we need more engagement with the 23 stakeholders on what we're doing. So once we get 24 everything ironed out in our proposal with the work 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
61 with Sandia, we're going to have that interface and 1
interaction. And the things that you bring up we 2
hopefully will be able to communicate at that time.
3 MS. VOELSING: But I was just saying that 4
what we learn from that may influence --
5 MR. TAYLOR: Right.
6 MS. VOELSING: -- more comments that we 7
need to provide on the test plan.
8 MR. TAYLOR: Understood.
9 MS. VOELSING: We kind of got to have that 10 discussion first.
11 MR. TAYLOR: I agree.
12 MS. VOELSING: Okay.
13 MR. TAYLOR: And as I think talking to 14 Chris a lot of the stuff that they need they're doing 15 on a small scale basis. They're running a few 16 preliminary small scale tests to look at how Aria is 17 going to model the arc and a few other pieces. So 18 they should have that --
19 MS. VOELSING: And I know you said they're 20 already using it to model copper. It would be really 21 helpful to lay out how they're doing that, how it's 22 validated so that we understand the basis for the 23 model.
24 MR. TAYLOR: Chris, did you get all of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
62 that?
1 MS. LAFLEUR: Yes, absolutely.
2 MR. TAYLOR: Okay.
3 MS. LAFLEUR: I did send some documents 4
that I was able to find that were -- one around the 5
verification of the Aria model and it's like basic 6
equations and the user manual that goes through in 7
detail like how it calculates everything.
8 So I definitely hear we're definitely 9
going to look at how it compares to this specific 10 application, so that will be part of the whole model 11 plan.
12 MR. TAYLOR: Okay. Sounds like a good 13 starting point and we'll work towards what's being 14 asked for. Thanks, Chris.
15 MS. LAFLEUR: Thank you.
16 MR. HAMBURGER: Kelli, would you mind just 17 making sure I've captured the comment here in No. 5?
18 MS. VOELSING: I could be wrong, but I 19 think it -- I thought it was more than just the Aria 20 model. It was putting a couple of pieces together.
21 MR. TAYLOR: Just put modeling because 22 there's a number of models.
23 MR. MELLY: Okay. Yes, our model will see 24 a different model.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
63 MS. VOELSING: And that's an appropriately 1
validated tool.
2 MS. LAFLEUR: Yes, so the Aria model has 3
confirmation of mass, species, momentum, energy. It 4
also calculates current and voltage. And that will be 5
used to model the arc itself, but then what we see 6
happening is the arc then feeds this more sooty fire 7
plume. And so that is more appropriately modeled with 8
our Fuego model which isn't a mass and heat transport 9
model. And so those will -- coupled together where 10 the model of the arc will feed the development of this 11
- plume, sooty plume which radiates the energy 12 completely differently than the arc, which is just --
13 it's just a different energy release.
14 MR. SALLEY: And, Chris, for those base 15 models Sandia has done a lot of V&V work on those.
16 It's correct, isn't it?
17 MS. LAFLEUR: Absolutely, yes. They are 18 used and they were developed for modeling harsh 19 environments that nuclear weapons are exposed to, the 20 fires and lightning strikes and all types of things.
21 And so I mean, our thermal test facility here was 22 built specifically to validate all those models. And 23 so, there's a lot of documentation I just have to get 24 for the Fuego and I have to find the unlimited release 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
64 document, but that -- it's tested in minutiae detail 1
for the NMSA.
2 MR. SALLEY: That would be great, Chris.
3 If you could kind of put a literature thing together 4
on what's publicly available for that to support the 5
V&V --
6 MS. LAFLEUR: Yes.
7 MR. SALLEY: -- that would be good. Thank 8
you.
9 MR. HAMBURGER: Do we have any other 10 comments on any of the previously dispositioned test 11 plan comments?
12 (No audible response.)
13 MR. CHEOK: Maybe go around the room for 14 comments.
15 MR. HAMBURGER: Sure. If anybody in the 16 room has comments on any of the previously 17 dispositioned comments that were made available 18 through the public meeting notice and that have been 19 publicly available for some time now or if anyone on 20 the phone has any comments on those, we'd be happy to 21 take that input. And if anyone has any new comments?
22 (No audible response.)
23 MR. HAMBURGER: Okay. Well, I'll give 24 everyone a minute to think about it. And just to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
65 summarize what we have here for our action items and 1
comments, the first was a request from EPRI that we 2
list all the types of data collected from our fall 3
series of testing and come up with a timeline for when 4
it will be made available either in its preliminary or 5
final forms. And there was a specific interest in the 6
material characterization of that ejecta that we 7
collected.
8 MR. MELLY: We gave an action item for 9
them to potentially look at the Small Scale Test 10 Program and determine which specific type of analysis 11 12 MR. TAYLOR: Yes, we can probably work 13 that through to the working group.
14 MR. MELLY: We could do that.
15 MS. ANDERSON: You want to do that, 16 because I wasn't sure if you wanted to do that through 17 the working group. I was going to ask if you wanted 18 to do it through the working group or --
19 MR. MELLY: Let's put it on there.
20 MR. SALLEY: Well, we got a working group.
21 Let's make it work.
22 MR. MELLY: And they can add it to the 23 working group agenda.
24 We have an action item, and this hasn't 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
66 been assigned to anyone in particular, but to 1
determine the prevalence or even the existence of the 2
aluminum main bus bar where we're proposing the 3
confirmatory test be parked. So --
4 MR. TAYLOR: You want to make it medium-5 voltage. I think we know those combinations for low-6 voltage.
7 MR. HAMBURGER: Okay. So determine the 8
prevalence of main bus bar aluminum versus copper 9
material in medium-voltage enclosures for our 10 confirmatory test configuration.
11 And then we have NEI's two overarching 12 comments.
13 MR. SALLEY: So No. 2; backing up, who has 14 the action on that one, Kenny?
15 MR. HAMBURGER: We can -- want to work 16 that through the working group as well?
17 MR. SALLEY: Or is that something you 18 could do a survey, Victoria, and help us on like 19 you've done in the past before when we were looking 20 for aluminum?
21 MS. ANDERSON: It's not as easy as it 22 sounds, but I think we've -- I think we have some of 23 that data sitting around. So we -- yes.
24 MR. RANDELOVIC: Yes, 50 plants and I 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
67 don't know the level of details for the main bus bars.
1 I'd have to look.
2 MS. ANDERSON: Yes. I'm not sure how --
3 MR. RANDELOVIC: We can just --
4 MR. MELLY: I know from the surveys that 5
we at least saw I wouldn't be able to pull that type 6
of information out of them, but you may have more that 7
I don't know about.
8 MS. ANDERSON: Yes. Well, we know which 9
-- who -- which each of the plants are, so we can 10 query them --
11 MR. MELLY: Right.
12 MS. ANDERSON: -- more deeply.
13 MR. MELLY: So think this would be an --
14 I don't know if it's really appropriate for the 15 working group because we wouldn't -- the NRC side 16 wouldn't be able to help with this action item.
17 MR. SALLEY: No, this I think has to come 18 on your side.
19 MR. HAMBURGER: I can -- how about we put 20 this down as something that NRC will work with NEI to 21 see if we can query the fleet for.
22 MR. MELLY: I think it would just be an 23 NEI action item.
24 MR. SALLEY: You could, Victoria, because 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
68 I mean, we want to get into 5054(f) letters and that 1
kind of thing.
2 MS. ANDERSON: Yes.
3 MR. SALLEY: Yes. No.
4 MS. ANDERSON: NEI won't be able to get 5
you the full fleet.
6 MR. SALLEY: Could you get something 7
representative like you did last time?
8 MS. ANDERSON: I can do best we can, 9
but --
10 MR. SALLEY: Okay. Yes, I think that's 11 kind of what we're -- Gabe, that's what we're looking 12 for as far as details?
13 MR. TAYLOR: Yes, I mean, it's -- we don't 14 have to know the exact number, but if there's aluminum 15 out there, then it would probably make sense to test 16 aluminum. If there's only copper, then it makes sense 17 only to test copper. That's kind of what I need to --
18 I don't need to know 48 percent as this versus the 19 other half.
20 MR. SALLEY: Nick, is it worth reaching 21 out to our international folks to see if we can get 22 data from them on that?
23 MR. CHEOK: Again, how important is this 24 information, because if you were going to be doing it 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
69 as -- going to be --it only applies to a certain 1
configuration.
If you have data for both 2
configurations, you're only apply the pertinent data 3
to the correct configuration.
4 MS. VOELSING: But it also needs to be 5
-- what we're trying to understand, does it matter?
6 Right? And so as long as you do apples to apples, it 7
doesn't really matter what you do. If you're 8
comparing -- if you test aluminum, compare it to the 9
aluminum. And the other like just --
10 MR. TAYLOR: Right. I think we need to 11 understand whether there are even any little amount 12 there on the main buses. If there are, then it makes 13 sense to do it like you said and do the comparison, 14 but if there aren't, then we're testing something 15 that's not out there. So that's kind of what -- at 16 least I'm --
17 MR. HAMBURGER: I think what Gabe is 18 saying is that the apples to apples would best be done 19 with copper if the aluminum configuration doesn't even 20 exist.
21 MR. LEJA: So copper on the main bus bars 22 and the aluminum on that? Okay.
23 MR. TAYLOR: Or the aluminum on the main 24 and aluminum on the secondary.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
70 MR. LEJA: Yes, I don't think you want to 1
separate much on medium-voltage. Aluminum wears a lot 2
faster than copper. Just to -- putting the breaker in 3
and out, that would wear it out. And also the --
4 MS. VOELSING: I thought it was pretty 5
clear that the looping side was mostly all copper.
6 MR. LEJA: It is mostly all copper.
7 MR. RANDELOVIC: Yes, but I also talked to 8
Kenn and he was saying that he knows the existence of 9
the main bus bars are aluminum. So what I am saying, 10 that you just need to --
11 (Simultaneous speaking.)
12 MS. VOELSING: I think it gets impossible, 13 but predominantly it's copper.
14 (Simultaneous speaking.)
15 MR. LEJA: The other thing is aluminum 16 contracts and expands, so that it's also not a good 17 application for it. In fact, you wear it out even 18 faster. So you guys can verify. I'm pretty sure 19 you're not going to find very many applications.
20 MR. HAMBURGER: Okay. We have NEI's two 21 more general comments, and I think the first one we've 22
-- the NRC has committed not to doing this and the 23 working group is proceeding to everyone's 24 satisfaction.
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
71 No. 2 is the -- I guess just the idea that 1
these tests need to be designed to isolate the impact 2
of aluminum if that is the goal of the pre-GI. Do you 3
have anything to add or change?
4 (No audible response.)
5 MR. HAMBURGER: And then the last comment 6
was regarding the modeling effort that it's important 7
that we understand and that we make sure that the 8
working group has access to the modeling approach and 9
the fragility determination for targets. Is that what 10 you meant by fragility determination? Targets?
11 Potential targets?
12 MS. VOELSING: The zone of influence is 13 where the fragility is going to be affected, too, 14 so --
15 MR. HAMBURGER: Okay. I want to make sure 16 I'm understanding what you mean there. And the 17 model's validation inputs and outputs to make sure 18 that they are appropriate tools and that they're 19 properly validated for what we intend to apply it to 20 in terms of zone of influence calculations. And EPRI 21 has reserved the right to submit further comments 22 after looking at the details of the modeling approach.
23 Any other comments from anyone in the room 24 or on the phone?
25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
72 MS. VOELSING: I guess I'll just say 1
briefly obviously I've learned way more on this than 2
I wanted to and it -- I give credit to the team and 3
the working group. This is really -- this is not 4
simple and it's really new science. It's not stuff 5
that's out there. So I think they've done an 6
admirable job and identified a lot of issues that need 7
to be investigated, I think doing a great job working 8
together and working through the issues. I just want 9
to make sure that we don't get ahead of answering some 10 questions we need to answer in the interest of moving 11 forward on testing if we're not ready for it.
12 MR. HAMBURGER: Okay. Hand it back over 13 to our division director. Do you have any --
14 MR. CHEOK: Actually just thank you for 15 coming. Thank you for your interest. And I think 16 Mark wants to say something.
17 MR. THAGGARD: Well, I just wanted to say 18 we don't want to try to set a time frame for when we 19 want to have the next meeting. Are we going to just 20 figure out what works best?
21 MR. MELLY: So that's something I can 22 touch on at least in terms of the working group at our 23 level. Like we said, we discussed here a lot of the 24 discussions that we held at the last in-person 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
73 meeting. We did discuss and think that it would be 1
very valuable to hold another possibly shorter in-2 person meeting prior to the low-voltage and bus stop 3
testing that will be done in the late summer months.
4 So we do plan on meeting again once the 5
equipment for low-voltage has been procured, once the 6
instrumentation has been developed, once these devices 7
have been built and just so we can have everyone's 8
alignment on what we plan on doing, what we plan on 9
testing, what we plan on measuring, the new devices 10 that we've built and used. So we will be as a working 11 group meeting prior to testing. And there has even 12 been discussions of having it be in Pennsylvania to 13 actually see the equipment that we'll be tested.
14 The working group members will also be 15 invited to the testing itself. It's not easy testing; 16 there's a lot of setup that goes into it. So we do 17 foresee the NRC NIST Sandia Working Team potentially 18 spending a month at KEMA, so there will be plenty of 19 opportunities for interaction and visits.
20 MS. VOELSING: So do you think we can have 21 a public meeting on that No. 5 here sooner than later?
22 MR. TAYLOR: So that's what I was going to 23 say is I think once we get prepared for No. 5, we can 24 probably just have one public meeting that covers the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
74 modeling and the fragility and these other items as 1
well. That seems reasonable. I don't have a date yet 2
for when that will happen though.
3 MR. SALLEY: And to understand the timing 4
and how we're trying to schedule and bring this 5
together, remember we have timeliness for the pre-GI 6
to support that program. And we all agreed that the 7
more data, the more information we have, the more 8
realistic we can do the analysis for the pre-GI.
9 Scheduling that with the test lab, we're 10 trying to look at the what, August, September time 11 frame as when we want to be in there. And we kind of 12 got that blocked with KEMA. Again, they've got a lot 13 of clients. We need to kind of get in there and say 14 this is our time frame and we'll be ready. So that's 15 the target we're working to. There is a little time, 16 but there's not years. It's months. Okay? So that's 17 kind of what we're up against. And again, supporting 18 the Generic Issue Program.
19 MR. CHEOK: So we'll meet again. We will 20 discuss the follow-up discussion on several of these 21 items, notably No. 5. And I thank you for coming out 22 and thank you for your interest.
23 MR. SALLEY: Mike, it would probably be 24 good if had the model and any final comment resolution 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
75 at the same meeting. We could do that, put it 1
together like that.
2 MS. VOELSING: I don't think we can do 3
that until we know about the model. I can then 4
provide some comments.
5 MR. SALLEY: We can get the model stuff.
6 (Whereupon, the above-entitled matter went 7
off the record at 2:22 p.m.)
8 9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433