ML18275A083
| ML18275A083 | |
| Person / Time | |
|---|---|
| Issue date: | 08/24/2018 |
| From: | Kent Howard Advisory Committee on Reactor Safeguards |
| To: | |
| Howard K | |
| References | |
| NRC-3863 | |
| Download: ML18275A083 (290) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee on Reactor Safeguards Plant Operations and Fire Protection Docket Number: (n/a)
Location: Rockville, Maryland Date: Friday, August 24, 2018 Work Order No.: NRC-3863 Pages 1-290 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005 (202) 234-4433
1 1
2 3
4 DISCLAIMER 5
6 7 UNITED STATES NUCLEAR REGULATORY COMMISSIONS 8 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 9
10 11 The contents of this transcript of the 12 proceeding of the United States Nuclear Regulatory 13 Commission Advisory Committee on Reactor Safeguards, 14 as reported herein, is a record of the discussions 15 recorded at the meeting.
16 17 This transcript has not been reviewed, 18 corrected, and edited, and it may contain 19 inaccuracies.
20 21 22 23 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 + + + + +
4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5 (ACRS) 6 + + + + +
7 PLANT OPERATIONS AND FIRE PROTECTION SUBCOMMITTEE 8 + + + + +
9 FRIDAY, AUGUST 24, 2018 10 + + + + +
11 ROCKVILLE, MARYLAND 12 + + + + +
13 The Subcommittee met at the Nuclear 14 Regulatory Commission, Two White Flint North, Room 15 T2B1, 11545 Rockville Pike, at 1:00 p.m., Gordon 16 Skillman, Chairman, presiding.
17 COMMITTEE MEMBERS:
18 GORDON R. SKILLMAN, Chairman 19 RONALD G. BALLINGER, Member 20 DENNIS C. BLEY, Member 21 CHARLES H. BROWN, JR. Member 22 WALTER L. KIRCHNER, Member 23 JOSE MARCH-LEUBA, Member 24 JOY L. REMPE, Member 25 MATTHEW SUNSERI, Member NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
2 1 DESIGNATED FEDERAL OFFICIAL:
2 KENT HOWARD 3
4 5
6 7
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
3 1 CONTENTS 2 Opening Remarks and Objectives . . . . . . . . . 4 3 RES/DRA HEAF Presentation . . . . . . . . . . . . .
4 Break 5 RES/DRA HEAF Presentation (cont.) . . . . . . . 112 6 Public Comments . . . . . . . . . . . . . . . . 191 7 Adjourn . . . . . . . . . . . . . . . . . . . . 194 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
4 1 P R O C E E D I N G S 2 12:58 p.m.
3 MEMBER SKILLMAN: Ladies and gentlemen, 4 this meeting will come to order. This is a meeting of 5 the Plant Operations and Fire Protection Subcommittee.
6 I'm Gordon Skillman, I'm Chairman of the 7 Subcommittee. ACRS members in attendance are Ron 8 Ballinger, Dennis Bley, Matt Sunseri, Joy Rempe, Jose 9 March-Leuba and our Designated Federal Official is 10 Kent Howard. And we've been joined by Walt Kirchner.
11 The ACRS reviews and advises the 12 Commission with regard to the licensing and operation 13 and production of utilization facilities and safety 14 related issues.
15 The adequacy of proposed reactor safety 16 standards, technical and policy issues related to the 17 licensing of evolutionary and passive plant designs 18 and other matters referred to it by the Commission.
19 The purpose of this information briefing 20 is for the staff to brief the Subcommittee on the 21 Phase II High Energy Arc Fault Test Plan.
22 The Subcommittee will gather information 23 analyze relevant issues and facts and formulate a 24 proposed position and action as appropriate for 25 deliberation by the full Committee, if needed.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
5 1 The meeting will be open to public 2 attendance and we have received no written comments or 3 requests for time to make oral statements from members 4 of the public regarding today's meeting.
5 The transcript of the meeting is being 6 kept and will be made available as stated in the 7 Federal Register Notice, therefore, we request that 8 participants in the meeting use the microphones 9 located throughout the meeting room when addressing 10 the Subcommittee.
11 Participants should first identify 12 themselves and speak with sufficient volume and 13 clarity so that they can be readily heard.
14 A telephone bridge line has been 15 established for this meeting. To preclude 16 interruption of this meeting, I ask those on the phone 17 to please mute your individual lines during the 18 presentations and the Committee discussion.
19 I also ask that all participants please 20 silence all electronic devices.
21 I would like to make an addition to the 22 participants, Dr. Pete Riccardella is joining us by 23 phone. Pete, you out there?
24 (No response.)
25 MEMBER SKILLMAN: He may be on mute.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
6 1 We will now proceed with the meeting and 2 I call upon Mr. Mark Thaggard to please make 3 introductory remarks.
4 MR. THAGGARD: Okay, thank you.
5 Good afternoon. My name is Mark Thaggard.
6 I'm the Deputy Director, Division of Risk Analysis in 7 the Office of Research.
8 I'm going to be brief because we have a 9 fairly lengthy briefing this afternoon.
10 I just want to say, first of all, that I 11 appreciate the Committee giving us an opportunity to 12 talk to you this afternoon about our High Energy Arc 13 Fault project.
14 You may recall back in your February 2018 15 memo to the Commission that was on the biannual review 16 and evaluation of the research program, you agree with 17 our decision to better understand the results from the 18 phase one testing.
19 But, recommended that additional 20 assessments should be undertaken before we start the 21 next phase of testing.
22 Today's briefing is intended to provide 23 you some insights into what staff has done leading up 24 to the next phase of testing that we're planning to 25 do.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
7 1 We're currently working with the NEA to 2 get a formal agreement in place for the phase of the 3 project.
4 In addition to the international program, 5 we also have a smaller NRC set of activities that 6 we're doing specifically to deal with aluminum.
7 We're currently scheduled right now to 8 begin the Phase II testing the week of September the 9 10th and looking at four different -- four aluminum 10 bus -- switch guard -- switch gear pieces of 11 equipment.
12 So, with that, I will turn the 13 presentation over to Mr. Salley and his staff.
14 MR. SALLEY: Thank you, Mark.
15 And you can see the purpose of the meeting 16 that Mark covered there, the high level objectives, 17 what we're going to do today.
18 Next slide, Dick?
19 You can see the agenda, we have a lot to 20 cover, a lot of material to cover here, so it'll be 21 pretty fast paced.
22 What I'd like to do is just take a second 23 and introduce the people who will be presenting to you 24 and it'll be going back and forth between them.
25 We have Kent Hamburger, Fire Protection NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
8 1 Engineer in my branch.
2 Nick Melly, Nick is our principle on this 3 project. He's the person we sent to the OECD/NEA 4 meetings. So, Nick is pretty much our technical lead 5 on this project.
6 Next to him is Kenn Miller. It's 7 interesting that we're also breeching or branching out 8 inside research across different divisions. Kenn 9 comes to us from the electrical engineer branch, so we 10 want to give that full picture from the electrical 11 engineering branch, hence, the team lead over there.
12 Also, for completeness, we have Dave 13 Stroup. He's a Senior Fire Protection Engineer in my 14 branch. And we -- you'll hear a lot about the thermal 15 fires that were the big risk driver in the 805s. We 16 put a small piece in here just, again, for 17 completeness. Dave had done a lot of the work on 18 there and he can give you the insights on that.
19 We also have one member on the phone, Gabe 20 Taylor is another Senior Fire Protection Engineer.
21 Gabe is vacationing in lovely Pennsylvania up at Cook 22 Forest, but I did talk him into doing part of the 23 presentation. And we have Sandia on the phone should 24 we need them for some of the small-scale testing.
25 So, you'll be hearing from these NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
9 1 individuals and that's the roles they play in this 2 project.
3 We also have one more, Stan Gardocki.
4 Stan is the Project Manager for the Generic Issue 5 Program. And, again, for completeness, Dick, we'll 6 have Stan come in and talk about the Generic Issue 7 Program.
8 So, that's kind of the broad overview of 9 who you'll be hearing from today.
10 And, with that, next slide there?
11 MEMBER SKILLMAN: Thanks, Mark. Let's go.
12 MR. MELLY: Let me take this moment to 13 check if we have our phone participants able to speak.
14 Gabe, if you're on the line and can speak? Time in?
15 MEMBER SKILLMAN: Let's stand by just for 16 a second to let Kent make sure the line is open. This 17 is an open meeting, so we should be able to have our 18 participants.
19 MR. MELLY: So, Gabe, just testing.
20 MEMBER SKILLMAN: Hooked in and on mute.
21 Let's just wait for a second, please.
22 MR. MELLY: Okay.
23 Gabe said the line went dead when the 24 meeting was opened. Gabe won't speak for a little bit 25 through the presentation anyway and I know we're short NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
10 1 on time, so we can begin now.
2 So, we'll just start with some of the 3 background of the current HEAF treatment in current 4 PRAs. And, to do that, we're going to walk back to 5 NUREG-6850. Basically, this is the guidance document 6 for PRAs in relation to fire.
7 6850 forms the basis for the nuclear power 8 plant fire PRAs. And, essentially, we're going to be 9 discussing these in terms of bins.
10 Fire initiators are lumped into their 11 initiator bins that's associated with a frequency.
12 We're going to be primarily talking about Bin 15 and 13 Bin 16.
14 Bin 15 is the electrical enclosure fires.
15 Bin 16 is specifically for HEAF, High Energy Arcing 16 Faults.
17 We'll also discuss a little bit about the 18 lessons learned in NFPA 805 such as Bin 15 covers all 19 electrical enclosures whether it be a small fire 20 protection panel on the wall to a larger switch gear 21 unit in a 4160 room.
22 We've been working to enhance that 23 guidance and Dave will talk a little bit about that as 24 we move forward.
25 We see the same sort of thing when we look NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
11 1 at Bin 16 for HEAFs. The methodology in 6850 2 essentially says, it's captured in Appendix M of 6850.
3 And it says that all HEAFs are treated equally, 4 whether it be from 440 volts to 4160 or above.
5 They're all lumped into one category in terms of 6 damage states.
7 We've seen that we need more flexibility 8 and a better understanding so we can separate the 9 voltage levels as appropriate for the type of damage 10 they can be.
11 (Telephone interruption.)
12 MR. MELLY: I'll give this a second.
13 (Telephone interruption.)
14 MR. MELLY: Gabe, do we have you on the 15 line?
16 MEMBER SKILLMAN: Hello? Can we ask who's 17 joined, please?
18 Pete, are you with us? If someone is on 19 the line, will you just please say hello?
20 MR. MELLY: Did someone just join the 21 line?
22 MR. TAYLOR: This is Gabe.
23 MR. MELLY: Okay.
24 MEMBER SKILLMAN: Thank you.
25 MR. MELLY: The line's working and we do NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
12 1 have Gabe. All right, thanks for the text, Gabe, we 2 appreciate it.
3 All right, moving forward, this is kind of 4 what I discussed earlier.
5 You can see on the left here, thermal 6 fires, that is Bin 15. And those are treated in the 7 methodology like a classical fire where you see that 8 you can have different zones of influence from fires, 9 classical fires based on the heat release rate 10 associated with that fire, what's in the cabinet 11 itself and how many targets can be damaged above that 12 electrical enclosure.
13 That's one methodology and it's treated in 14 6850 in Appendix G.
15 And, on the right, you see we have our 16 High Energy Arcing Faults Program which we're going to 17 be discussing here today. It's treated in the 18 deterministic methodology using a zone of influence.
19 In general terms, it's three-foot 20 horizontally from the side of the cabinet from the 21 side of the panel and five-feet vertically is the zone 22 of influence of initial damage state when one of these 23 events occurs. There are some --
24 MEMBER MARCH-LEUBA: And that's 25 independent of the voltage?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
13 1 MR. MELLY: That is independent of the 2 voltage whether it be 440 or 4160, 6.9 and above.
3 That's one of the areas that we wanted to look at and 4 we'll go into some of the things we've learned from 5 that as we move forward.
6 I'll also, in a later slide, we have 7 separated the frequency of occurrence from low voltage 8 to medium voltage because we did see that medium 9 voltage tends to be more prevalent.
10 There's also some stipulations as to how 11 to do the damage outside of that three-foot, five-12 foot. For instance, if cable trays or protected or 13 not and how the fire will spread.
14 On the left hand side for the thermal 15 fires, you have a growth profile associated with the 16 first itself. So you have a T-squared ramp 17 essentially, zero to 12 minutes to your peak.
18 However, on the right hand side for the 19 high energy arcing faults, you reach your peak at time 20 T equal zero from that methodology.
21 So, that's kind of how they're split right 22 now, two different treatments for the two different 23 portions of electrical cabinets.
24 MEMBER BLEY: We've had some arc faults 25 that don't originate inside a cabinet, too.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
14 1 MR. MELLY: Yes.
2 MEMBER BLEY: Do you treat those 3 differently?
4 MR. MELLY: No, we treat them the same in 5 the methodology. For instance, the Robinson event 6 occurred at the --
7 MEMBER BLEY: Yes, for instance.
8 MR. MELLY: For instance. It occurred at 9 the cable that was entering into the cabinet.
10 However, Appendix M does discuss that it can occur 11 within the cabinet itself or usually at the 12 termination point.
13 So, that would be included in the current 14 methodology in that bin because it was on the cables 15 entering into the cabinet. So, we do treat that 16 similarly.
17 However, Appendix M only gives one zone of 18 influence for a high energy arcing fault. And what we 19 saw at Robinson was that it was a little bit more 20 severe. We damaged targets outside of that zone of 21 influence. I don't know for sure, but probably 22 because it did not have the enclosure to hold in some 23 of that energy and was able to --
24 MEMBER BLEY: It lasted --
25 MR. MELLY: Yes, it did last for 8 to 11 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
15 1 seconds.
2 In addition, not depicted on the slides, 3 we also do cover bus ducts in the high energy arcing 4 fault bin. There's four categories and bus ducts are 5 included. And, typically, those are not handled in an 6 Appendix R analysis as initiators of fire sources.
7 And we'll go into a little bit of that as we move 8 forward.
9 MR. STROUP: As Nick's -- I have a lot of 10 things, I just can't read apparently.
11 As Nick said, when 6850 was being 12 developed in the early 2000s, as has happened several 13 times in the past with entities that want to go to a 14 risk-based, performance-based kind of situation, 15 especially involving things like fire modeling, they 16 found there was a limited amount of available data.
17 And the data that was available was 18 typically skewed towards worst case, trying to answer 19 the question of what's the worst case that can happen 20 if a -- if you get a failed electrical cabinet or 21 something within the nuclear power plant.
22 So, that ultimately gave rise to a lot of 23 the concerns that we see here recently about fire PRA 24 being overly conservative.
25 The -- that there's a peak heat release NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
16 1 rate that's recommended for looking at electrical 2 cabinets from a thermal fire standpoint in 6850 that's 3 702 kilowatts.
4 That's a rather severe fire threat for 5 pretty much anything in the room. The old Appendix R 6 days would have you basically fair the entire room.
7 With a 702 kilowatt fire, you're probably, 8 if you try and calculate that scenario, you're 9 probably going to come up with the same answer.
10 Also, the other recommended peak heat 11 release rates in 6850, for examples, for pumps and 12 motors, there was no data to support the numbers that 13 are in the existing version of 6850. They were 14 derived by the expert panel looking at maybe this pump 15 that's sort of burned like an electrical cabinet.
16 So, there was a reason why the fire PRAs 17 appeared to be overly conservative, but those numbers 18 for the heat release rates were driven that way.
19 So, in order to try and address that 20 problem, we conducted a series of fire tests in around 21 2016 aimed at trying to supplement the available data 22 and look at more realistic ignition sources.
23 In the past, typical ignition sources for 24 electrical cabinet fire tests were things like 25 flammable liquids or large gas burners and the like.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
17 1 Because, you didn't want to invest a lot of money in 2 conducting a test and then have it not firm. So, you 3 wanted to make sure that your ignition was a 4 significant threat.
5 But, that led to concerns that you were, 6 again, skewing the results in a worst case and not 7 really looking at what might be considered typical.
8 So, we got some electrical cabinets out of 9 Bellefonte. We conducted a 112 tests measuring the 10 heat release rate of non-energized cabinets. And that 11 resulted in the HELEN-FIRE Report.
12 MEMBER SKILLMAN: David, what was the 13 source of the ignition? What did you burn in order to 14 get a heat rate?
15 MR. STROUP: For the HELEN-FIRE test?
16 MEMBER SKILLMAN: Yes.
17 MR. STROUP: The HELEN-FIRE test, we tried 18 an assortment of low energy ignition sources ranging 19 from a 300 watt electrical heat up to a 5 kilowatt gas 20 burner, something that would give you, say, like a two 21 or three-inch flame. Something more typical of a 22 failed switch.
23 MEMBER SKILLMAN: In the enclosure?
24 MR. STROUP: Yes.
25 MEMBER SKILLMAN: In the enclosure?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
18 1 MR. STROUP: Yes.
2 MEMBER SKILLMAN: Okay.
3 MEMBER BLEY: On thing you didn't mention, 4 6850 doesn't require you use just the peak heat rate.
5 People do that for convenience. You can use the 6 distribution treated probabilistically. And my 7 understanding is, very few people have taken advantage 8 of that.
9 MR. STROUP: Right, that's also another 10 consideration. And so, working under the Memorandum 11 of Understanding with EPRI, we gathered together a 12 working group composed on NRC, some consultants to NRC 13 and industry representatives.
14 We looked at the HELEN-FIRE data as well 15 as the older data and, out of that developed new 16 classifications for the electrical cabinets. The 17 original classifications, the electrical bins Nick was 18 talking a couple minutes ago about the bins being too 19 coarse.
20 Well, in the Bin 15 for electrical 21 cabinets, the way the heat release rate information 22 was originally characterized was based on whether the 23 cabinets had qualified or unqualified cable and single 24 or multiple bundles.
25 Well, kind of difficult to figure out how NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
19 1 to classify -- how to identify it by looking at the 2 exterior of the cabinet whether it's got single or 3 multiple bundles.
4 So, what we did with the RACHELLE-FIRE 5 exercise is, we reclassified the cabinets and went 6 from the original five bins to 15 bins. And, those 7 bins are based on the functions of the cabinets and --
8 the functions of the cabinets, the -- also the sizes 9 and, if you've got potentially risk significant 10 cabinets, we also came up with a methodology where you 11 could open those cabinets and evaluate the fuel load 12 within those cabinets and potentially reduce the peak 13 heat release rate if it was justified.
14 So, as part of the RACHELLE-FIRE group 15 took advantage or attempted to take advantage of the 16 fact that you had a fire in a cabinet. So, the fact 17 that the fire is burning inside an enclosure itself 18 should be -- should help improve the realism in your 19 scenario if you account for that fact.
20 So, we came up with like a 40 percent 21 reduction in the potential temperatures above a 22 cabinet based on analysis that we conducted as part of 23 the RACHELLE-FIRE, one activity.
24 We're currently working on a second 25 RACHELLE-FIRE report which will further expand the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
20 1 accounting for the cabinet as well as do a better job 2 of providing peak heat release rates for pumps and 3 motors.
4 And also we're looking at better ways to 5 model the fire in a -- in the main control board as 6 well as improved transient heat release rates.
7 And, with that, I'll turn it back to Dick.
8 MEMBER SKILLMAN: Isn't there a lot --
9 maybe that's the wrong way to ask the question. Do 10 the standards -- the fire protection standards provide 11 a lot of information that is valuable in informing 12 your documentation?
13 MR. STROUP: You mean in terms of the heat 14 release rates or --
15 MEMBER SKILLMAN: Yes. It would just seem 16 to me that the fire protection codes, all of the work 17 that's been done over the last 50 years to understand 18 fire and heat and heat release rates, it's probably 19 codified in tables somewhere. And, I'm just wondering 20 if, for better or for worse, you're duplicating 21 information that's readily available.
22 MR. STROUP: It's not so much -- I mean, 23 I've been doing this for close to 40 years now. And 24 I originally started working for the National 25 Institute of Standards and Technology back in 1980.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
21 1 And, at that time, the big focus was on 2 the residential fire problem. So, you saw -- that was 3 about the beginnings of readily available fire models.
4 And, to support those fire models, they 5 needed the heat release rate information as input 6 data. So, you saw a lot of beds and couches and 7 furnishings typical of a home being burned.
8 In the last 19 -- or in the early 1990x, 9 I moved to the General Services Administration where 10 we were trying to do a performance-based approach to 11 the safety and government buildings. The idea being 12 that we were responsible for a lot of safety and 13 environmental management issues, if you will, and we 14 wanted to look at an integrated approach to safety in 15 the building, not do one modification one day and then 16 have to come back and do another modification to 17 change the modification because it needs to address 18 something else.
19 So, in that case, we started looking at 20 the fire threat from office-type workstations, 21 computers and things like that.
22 Now, I've come to GSA and I find myself or 23 I've come to NRC and found myself in the same 24 position. We had a lot of worse case data for 25 cabinets and to some extent cables and stuff, but NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
22 1 we're trying to go back and refine that now.
2 So, we tend to be really on the cutting 3 edge of trying to develop the heat release rates.
4 MEMBER REMPE: So, on the other side, what 5 about applicability of this work to other industrial 6 facilities? Other types of power plants or whatever?
7 Does this work get used by other non-nuclear 8 organizations?
9 MR. STROUP: Yes, I mean, beyond the heat 10 release rates, one of the most significant 11 contributions that NRC has made to the general fire 12 protection community is the verification and 13 validation of the fire models.
14 In the past, fire models that were 15 developed by different people, the extent to which 16 they were verified and validated, you know, compared 17 to experimental data was typically the author or 18 developer of the program would compare against some 19 set of tests and say, the agreement was good or within 20 acceptable levels of uncertainty or some relatively 21 arbitrary comparisons like that.
22 But with the advent of the requirement in 23 NFP 805 for fire models to be verified and validated, 24 we needed to find an actual methodology that would 25 work and quantify that.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
23 1 And, you've seen that in the NUREG-1824 2 reports and the supplements how we've developed the 3 actual applications of the fire models and the 4 supporting validation of them. And, that's being used 5 throughout the industry now.
6 MR. MELLY: And, I do know from 7 interaction with other consultants that a lot of the 8 heat release rates that have been developed for 9 electrical cabinets through our work have been applied 10 to other industries, telecommunications, other power 11 plants, coal fired and things like that.
12 So, it is being broadly applied.
13 MR. STROUP: And the other thing -- the 14 other part of that, too, is the 6850, for all its 15 issues, provides a good methodology for how you do an 16 analysis of a particular fire problem. That overall 17 analysis framework is missing in the rest of the fire 18 protection community.
19 Typically, it's just a let's look at this 20 thing, try and make a decision as to what kind of fire 21 might occur and that there's no real framework as to 22 how you do that.
23 I actually serve on the Society of Fire 24 Protection Engineers. One of their task groups 25 looking at trying to develop what they call design NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
24 1 fires and taking some of NRC's experiences and saying, 2 can we come up with a set of design fires that you 3 would use to model fire safety performance of a 4 particular building.
5 MEMBER SKILLMAN: So, this would be a 6 categorization of, if you will, design basis fires for 7 certain types of structures or buildings or that type 8 of thing?
9 MR. STROUP: Yes, like, for example, the 10 National Fire Protection Association produces NFPA-101 11 which is the life safety code. And, you would have a 12 set of different types of recommended fire scenarios 13 that you would have to, though modeling or other 14 means, show that you're protection systems could 15 handle that particular fire.
16 For example, if you had a fire in an 17 atrium, would your smoke control system keep the 18 smo9ke layer above the level of egress that people 19 needed in order to escape? Things like that.
20 MEMBER SKILLMAN: Well, thank you. Let's 21 keep on going, thanks.
22 MR. MELLY: All right, moving forward, 23 kind of building off of what Dave just spoke about, on 24 the screen here, you see actually a slide that was 25 presented by EPRI and prepared by EPRI which looks at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
25 1 27 different plants and does a categorization of their 2 contribution to total risk fire CDF.
3 MEMBER SKILLMAN: Nick, please speak into 4 your microphone.
5 MR. MELLY: Oh, yes, I'm sorry.
6 MEMBER REMPE: There's a mouse.
7 MEMBER SKILLMAN: There's a mouse 8 somewhere that you can -- there you go.
9 MR. MELLY: Okay. So, it's broken down in 10 terms of the bins that we discussed earlier. You can 11 see the top contributor to this fire risk overall is 12 the electrical cabinets. Then it goes transients and 13 HEAF is the third highest risk contributor.
14 So, one of the major programs that we did 15 was based on electrical enclosures to focus in on the 16 highest risk driver of the plants. Where can we make 17 the biggest risk improvements to realism? And, we're 18 trying to get the most bang for our buck here.
19 High energy arcing faults are a subset of 20 that group. Typically, they damage the same targets.
21 However, the high energy arcing faults are more severe 22 and they're going to damage more targets at once.
23 So, that's why we are seeing it show up 24 high on the list.
25 MEMBER SKILLMAN: When you say the biggest NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
26 1 bang for the buck, whose buck?
2 MR. MELLY: In terms of I wouldn't say in 3 monetary value that we're discussing here but we want 4 to make the biggest leap forward in the terms of 5 methods, tools and data that we can.
6 MEMBER SKILLMAN: So, this is not a 7 financial issue?
8 MR. MELLY: No.
9 MEMBER SKILLMAN: This is getting the 10 maximum benefit possible --
11 MR. MELLY: Yes.
12 MEMBER SKILLMAN: -- from this research?
13 MR. MELLY: Yes. And the modeling. We're 14 trying to --
15 MEMBER SKILLMAN: And modeling, okay.
16 MR. MELLY: -- gain insights from the PRAs 17 that have been done as to where are these risk 18 drivers? And, we hear a lot of the realism and 19 uncertainties with the fire PRAs, and we were trying 20 to dive in further as to, do we believe this fire 21 risk? Can we do more research to increase the level 22 of realism and trust in the fire PRAs?
23 MEMBER KIRCHNER: Can I -- may I ask 24 something? The way you bin things has -- causes me a 25 little bit of a problem in confusing cause and effect.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
27 1 MR. MELLY: Okay.
2 MEMBER KIRCHNER: So, a transient, to me, 3 would be high amperage and a cable that would --
4 MR. MELLY: So, in terms --
5 MEMBER KIRCHNER: -- destroy the cable and 6 create a fire, whatever. HEAF is another source.
7 But, the electrical cabinets is the, I think in other 8 reports you've done, is the target.
9 So, how are you binning things here?
10 MR. MELLY: This current graph shows it as 11 fire initiators. So at -- if the electrical cabinet 12 is the fire initiator that would be included in this 13 bin.
14 However, in transients in this case, we 15 mean transient components throughout the plant. Such 16 as, if you have a -- accounting for a trash can fire 17 or a movable piece of equipment, those are transient 18 fires in the term of this chart.
19 MEMBER KIRCHNER: Okay. As opposed to 20 what?
21 MR. MELLY: As opposed to electrical 22 transients. This would be treated in 6850 as a moving 23 fire with a 317 peak heat release rate.
24 We do have an additional project looking 25 at those currently with EPRI. We're actually doing NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
28 1 fire testing this week -- this month looking at 2 typical fuel packages that you would see in an 3 electrical -- or in a nuclear power plant such as PPE 4 clothing, trash can fires, vacuum clear fires and 5 things like that.
6 MEMBER KIRCHNER: Yes, that I understand, 7 but now, just zeroing in on HEAF, which is where 8 you're going --
9 MR. MELLY: Yes.
10 MEMBER KIRCHNER: Again, to me, HEAF is 11 like a source term, a cabinet is where it takes place 12 or for a motor or transformer, those are pieces of 13 equipment.
14 HEAF is an energetic source.
15 MR. MELLY: This is broken down in terms 16 of frequency. So, each one of these bins, when 17 looking at the operating experience, is compiled to 18 create a frequency for each one of these bins.
19 So, how frequent are electrical cabinet 20 fires from the recent data?
21 NUREG-2169 complies all fires that 22 occurred in electrical cabinets or all high energy 23 arcing faults all transient fires and assigns them a 24 unique frequency to apply to your scenarios in your 25 plant. And, that is what this graph is showing, the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
29 1 frequency of occurrence against the CCDP of the 2 scenarios at your plant.
3 MR. HAMBURGER: In this case, the 4 difference between electrical cabinets in HEAF is the 5 difference between Bin 15 and Bin 16.
6 MR. MELLY: Yes.
7 MR. HAMBURGER: And, the 15 are those 8 traditional thermal fires in the cabinets. Whereas, 9 the HEAF is that energetic explosive event.
10 So, perhaps it's slightly misnamed on this 11 graph, it doesn't -- it's not electrical cabinets, 12 it's that traditional thermal fire we see in the 13 electrical cabinets.
14 MR. MELLY: So, each one of these bins on 15 the right is associated with a frequency of 16 occurrence.
17 MEMBER KIRCHNER: Thank you.
18 MEMBER SKILLMAN: Is the frequency of 19 occurrence, excuse me, is the event the information 20 that is brought from the root cause of the corrective 21 action activities at the site? Is that -- it almost 22 appears that that is what this is.
23 MR. MELLY: In part, it is. It's a part 24 from the root case as well as the damage state that 25 has been seen from the event.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
30 1 MEMBER SKILLMAN: Okay.
2 MR. MELLY: And then, that's the 3 evaluation done by both the NRC and EPRI to -- in 4 order to bin it into these bins.
5 MEMBER SKILLMAN: Okay.
6 MR. MELLY: Which is the appropriateness 7 as well as whether it's a challenging fire or non-8 challenging in terms of PRA.
9 MEMBER SKILLMAN: The --
10 MR. MELLY: And that is in NUREG-2169.
11 MEMBER SKILLMAN: The 4 events out of 27 12 of the diesel generators kind of catches my attention.
13 Can you speak to those? Is this an exhaust fire? Is 14 this the alternator?
15 MR. MELLY: Or these --
16 MEMBER SKILLMAN: No lubricating oil set 17 the whole thing on fire? Is it lubricating oil and 18 the engine lit itself off? What is that?
19 MR. MELLY: It is based on operating 20 experience, just how many diesel fires we have --
21 diesel generator fires we have seen. It could be --
22 what we see a lot is the lagging catch fire when 23 you're doing the test runs.
24 MEMBER SKILLMAN: So, it's lagging on the 25 exhaust system?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
31 1 MR. MELLY: It could be that or it could 2 be a fire in your diesel generator.
3 It's essentially when your diesel is 4 operating, has it caught fire? And, we have seen 5 several cases where that has occurred. It could be 6 the lagging, it's any component associated with that 7 diesel generator.
8 MEMBER SKILLMAN: Fair enough. Okay, 9 thank you.
10 MR. MELLY: Any other questions on this 11 slide?
12 Okay, moving forward. And, we can see 13 that it is a -- the high energy arcing faults do show 14 up as a fairly high risk contributor to the plant's 15 CCEP and CDF.
16 And this goes into a little bit of that 17 frequency that we've discussed. You can see the Bin 18 15 frequency from NUREG-2169. The mean is 3.0 times 19 10 to the minus two.
20 And, our high energy arcing faults, you 21 see that at least in order of magnitude, lower than 22 that if not two. And, that's because these high 23 energy arcing faults are relatively rare events.
24 However, they can cause very complicated 25 shutdowns. You can have human error failures during NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
32 1 these events. And, while they are rare, the CCDP 2 contributions associated with them are larger than 3 that we see in Bin 15.
4 MEMBER SKILLMAN: So, is the frequency on 5 a per year basis or what?
6 MR. MELLY: It's on a per year basis.
7 MEMBER SKILLMAN: Per 7,000 hours0 days <br />0 hours <br />0 weeks <br />0 months <br /> or for 8 365 days?
9 MR. MELLY: It is on a 365 days per plant.
10 MEMBER SKILLMAN: Thank you. Okay.
11 MR. MELLY: And, this would be a generic 12 frequency at the plant which then gets divided amongst 13 their electrical cabinets as done in every other bin 14 for 6850.
15 And, what you're seeing on the left hand 16 side is the U.S. operating experience and how those 17 events are binned, broken by low voltage events, 18 medium low voltage events. And, it's not up to date 19 currently, but this is what's included in NUREG-2169.
20 It goes from -- essentially up to 2010.
21 We are currently working on updating that 22 frequency to bring in the most recent years' data.
23 On the right hand side, you're going to 24 see what kind of the impetus for some of this project 25 is we see these events occurring internationally quite NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
33 1 frequently. And, we are -- we've been collecting the 2 data from all those events through a topical report 3 and interaction with the NEA OECP.
4 What you see on the right hand side is 5 just a small portion of the events that we've 6 collected internationally and we're trying to 7 understand those events in terms of duration, 8 severity, damage targets and things like that which 9 will bring us into essentially the international 10 program that we have developed. This is an issue 11 brought for --
12 MEMBER BLEY: There's nothing related to 13 nuclear plants about this?
14 MR. MELLY: No.
15 MEMBER BLEY: Have you -- I mean, you've 16 got nuclear data from the U.S., nuclear data from 17 elsewhere in the world. But, there is an awful lot 18 more electrical stuff that's not nuclear. Have you 19 tried to get access to that?
20 MR. MELLY: We have. Not only through 21 informally, but we've received many videos, pictures 22 and things like that when these events occur 23 internationally from some of the work that we've been 24 doing.
25 We'll get emails from Egypt, for instance, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
34 1 that occurred at one of their oil plants and they sent 2 us a smattering of information.
3 We looked to deal with the IEEE which 4 we'll discuss a little bit and NFP as to how often do 5 these events occur in the U.S. at other power plants?
6 MEMBER BLEY: Okay.
7 MR. MELLY: We've also seen this --
8 MEMBER BLEY: Do they have anything like 9 a consistent database on that?
10 MR. MELLY: So, they don't and that's 11 actually one of the issues that Ken's going to 12 discuss, it's the definition of these events. They 13 all -- they classify all these as just arc fault 14 events or arc flashes.
15 So, they don't distinguish between the 16 ones that are very severe versus the ones that are 17 not very severe. And, they're more thinking personnel 18 protection. So, we are working with Ken as well as 19 IEEE on that.
20 I don't want to steal any of his thunder, 21 but we're working towards defining these potentially 22 in other standards to make it uniform across industry 23 so we can get a better feel for when these occur, a 24 database for them and investigations base.
25 MEMBER SKILLMAN: Might you have a sense NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
35 1 that when the data is brought up to date that there 2 might be a change in the frequency?
3 MR. MELLY: Yes, I believe that there will 4 be and we've been handling that typically through the 5 PRA maintenance process of them upgrading their 6 frequencies when they do their periodic maintenance.
7 So, it's just a moving scale.
8 When we get more data -- we're on a 9 currently five year cycle to review the data and then 10 implement it into new frequencies.
11 MEMBER SKILLMAN: Will the frequency 12 increase a lot? A little? Go down a little? Go down 13 a lot?
14 MR. MELLY: I don't know. I do know of a 15 few events that have occurred recently. It's just how 16 are we going to classify those and count those.
17 One that comes to mind is the Turkey Point 18 event that we'll discuss a little bit later. The 19 question becomes is that a high energy arcing fault in 20 the way that we're modeling it because it did not 21 create the zone of influence of damage associated with 22 three-foot, five-foot. However, it did have other 23 repercussions.
24 So, that's another issue that we're trying 25 to deal with in definition space.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
36 1 We have an ongoing project or we're going 2 to begin an ongoing project under the EPRI MOU to take 3 a deep look at the way we're classifying these events, 4 the frequency of these events, the most recent data 5 and we're going to be looking at things like that.
6 MEMBER SKILLMAN: Thank you. All right.
7 MR. SALLEY: So just to take a second here 8 and the new term HEAF that really started appearing in 9 6850 around the 200 time frame, as we look at the 10 exiting regulation, was this covered? Did we look at 11 this? And, we took a second here and did that.
12 If you go back to the general design 13 criteria, Appendix A and 10 CFR 50, GDC-3, of course, 14 is fire protection. And, in the general fire 15 protection you talk fire and explosions.
16 The typical fire protect engineer, though, 17 he would look at the fires, the combustibles. If you 18 remember we looked at a lot of combustible loading, 19 where the combustibles were in the plant.
20 We looked at the cables. When we look at 21 electrical cabinets, we looked inside the cabinet, how 22 much combustible material is in here to have a classic 23 fire like Dave Stroup discussed.
24 So, we really didn't hit that explosion 25 piece of the energetic fault of the electricity. So, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
37 1 the fire protection community really never looked at 2 that.
3 Likewise, the electrical engineers at GDA-4 17 which is their single failure criteria. So, the 5 regulations speak to it, but the HEAF events kind of 6 fell through the cracks a little bit.
7 MEMBER BLEY: And most of the explosions 8 I remember in the older database weren't high energy 9 arc faults, they were faults inside canned equipment 10 that had oil in them that blew up. So, it was an 11 explosion nonetheless.
12 MR. SALLEY: Sure, hydrogen is the other 13 classic one --
14 MEMBER BLEY: Yes.
15 MR. SALLEY: -- that the fire protection 16 engineers looked at.
17 So, the reason I bring this up as we move 18 forward in that is toward the generic issues that 19 there are things in the regulation that kind of, sort 20 of spoke to it but it didn't call it out explicitly.
21 And, that gets to your question about the 22 energetic fault that we never looked at the energy, I 23 mean, we looked it from a fire protection standpoint.
24 The electricity formed the ignition source that looked 25 at combustibles on fire, put the fire in motion and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
38 1 that's how we would have calculated it in our model.
2 So, the HEAF brings out electrical energy 3 piece into it and that's what makes this somewhat 4 even.
5 MEMBER SKILLMAN: Proceed, let's go.
6 MR. HAMBURGER: So, an existing follow up 7 to some discussion of the single failure criteria in 8 GDA-17. What confidence that we have figures --
9 MEMBER BROWN: Can I interrupt? Can I 10 interrupt, Dick?
11 MEMBER SKILLMAN: Absolutely, sure.
12 MEMBER BROWN: I just pawed through all of 13 the view graphs just for fun. It's not the testing, 14 categorizing, different types and everything else, the 15 reason I ask if, have there been any efforts of how to 16 prevent these from happening at all?
17 Now, let me categorize that, because in --
18 I came from the naval nuclear program. And I had two 19 circumstances were a four-star admiral is ripping my 20 toenails off because, in one of the plants an arc 21 fault, we never even knew it was going on. It started 22 in one cabinet, ate its way across through the wall, 23 through another wall and almost came into the 24 operating space on a very critical piece of ship.
25 And that was kind of the start of a review NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
39 1 I was prodded along with the rest of the NAVSEA, 2 NAVSEA 05 to go generate an arc fault detection system 3 which we designed and they've installed, or at least 4 at the time I retired 17 years ago was in the process 5 of installing that worked very well.
6 And, we were even able to then develop 7 some stuff to work on the aircraft carriers as well, 8 even though they're more vented than the submarine 9 cabinets are.
10 And so, I was kind of taken aback by 11 looking that there's actually we're testing the heck 12 out of everything, but it doesn't look like there's 13 been an effort to try to identify how do we put in 14 system -- because these are nasty. We also killed 15 people in both circumstance.
16 One occurred on the carrier. Is so 17 happened the chief petty officer bent over to tie his 18 shoe at the time of 4160 volt arc fault occurred in an 19 exciter. Blew right over his head, singed his hair.
20 He didn't grow it back to contrary to my --
21 MR. MELLY: Yes.
22 MEMBER BLEY: And it precipitated a great 23 deal of -- I mean, we did a lot of testing with photo 24 sensors, pressure sensors and which you can do one 25 type but you can't do another, so we had to adapt NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
40 1 them.
2 So, there's a lot of -- there's some 3 background for how do you take care of that. And, it 4 would seem to be that some emphasis as opposed to 5 testing and identifying all these, it would be a good 6 idea -- because they're nasty when they occur. You 7 know, they cause a lot of damage to try to prevent 8 them.
9 MR. MELLY: I think Mark wanted to take 10 over.
11 MR. SALLEY: Yes, you know, as we look at 12 the final resolution for the HEAF work, obviously, 13 we're going to look at in defense in depth.
14 And the first step is going to be 15 prevention and I think that's something that Ken 16 Miller can probably speak to what the electrical 17 engineers have done. I know a lot from my TVA years 18 what they would do as far as their maintenance.
19 MEMBER BROWN: Yes, loose connections are 20 a real problem.
21 MR. MELLY: Yes.
22 MEMBER BROWN: That's how they generally 23 get initiated.
24 MR. SALLEY: Right. So, you know, we 25 understand that would be the first step in this. I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
41 1 guess this program is trying to understand the latter 2 half for the potential for damage.
3 And, again, this is almost a newer topic 4 for us that we really hadn't seen in the classical 5 work we've done before. And, I guess, Ken, you can 6 speak to what the electrical engineers to as far as 7 the preventive maintenance.
8 MEMBER SKILLMAN: Don't you what the 9 potential type of damage it is? I mean, these are 10 really damaging. I mean, if the extent is very, very 11 extensive.
12 MR. SALLEY: Yes, and we're going to get 13 in some of the photos and some of the other things.
14 MEMBER SKILLMAN: All right, I just --
15 it's here.
16 MEMBER BROWN: I actually think --
17 MEMBER BLEY: Did you share a lot of 18 information with the rest of world? My understanding 19 when I left that information was not classified. I 20 mean, I t was not it was shared.
21 MEMBER BROWN: It was classified but I 22 don't think it was shared very well.
23 MEMBER BLEY: No, we -- but the people we 24 talked to, I think it was. Okay? I retired 17 --
25 like I said, almost 18 years ago. So and we were in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
42 1 the process of installing it in submarines as well as 2 finishing up the stuff for the aircraft carriers.
3 We had to adapt the process for the 4 aircraft carriers.
5 MR. MELLY: And we do know that this --
6 these types of systems such as photo operated 7 detection very quick and very quick time frames is 8 used in the Navy in almost every breaker unit.
9 MEMBER BLEY: That's right.
10 MR. MELLY: In the U.S. Nuclear, that 11 would be considered backfitting every single 4160 12 breaker unit.
13 In addition to that --
14 MEMBER BLEY: Well, there's a different --
15 I mean, we're ships at sea. There's a different 16 metric, you don't want to stall airplanes in the air.
17 MR. MELLY: Right.
18 MEMBER BLEY: You don't was submarines to 19 go there. They'd have dire consequences.
20 Here, it's a fire, you put it out and then 21 you recover. But the plant can be impacted pretty 22 heavily.
23 MR. MELLY: Exactly.
24 MEMBER BLEY: So, and the number you're 25 had or that you document is not insubstantial over the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
43 1 last few decades.
2 MR. MELLY: Right. And other countries 3 are addressing this in potentially different ways as 4 well. Japan, for instance is starting regulation 5 where they may be requiring their power plants to 6 switch to all digital circuit protection schemes.
7 And, they are doing a lot of work on this 8 high energy arcing fault as well. And, we've been 9 working with them that Ken will speak to a little bit 10 because of the high energy arcing fault that happened 11 at Onagawa. They have an enhanced interest as well.
12 There are ways to take care of this 13 through physical design changes. The primary goal 14 with this testing, though, is, however, if we do see 15 these failure in the current technology that's out in 16 the fleet, or if that design were to fail, how do we 17 classify the risk in terms of a fire PRA?
18 MEMBER BLEY: I just -- just off the top 19 of my head, I hadn't heard about that with Onagawa 20 before. But, I would be careful on digital because I 21 don't know what the impact of -- there's some pretty 22 unique EM fields come out of these things. I don't 23 know how they cook through that.
24 MR. MELLY: Mr. Brown I think --
25 MEMBER BROWN: The electric -- yes, I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
44 1 wanted to emphasize Dennis's point. I mean, the 2 electric field inside, and if you look at the new 3 carrier, which is an even higher -- the medium voltage 4 --
5 MR. MELLY: Yes.
6 MEMBER BROWN: -- 13 point AKB electric 7 plant which I was fortunate enough to initiate in 8 1996. So, we recognized that issue at the time.
9 I don't know whether we've done anything 10 for those switchboards or not. Because I retired four 11 years later. But the fields are huge in their -- and 12 they -- with the it was just the stress that you have 13 to deal with. I mean, we had to go with the ozone 14 protection and everything else to get the fields down.
15 So, anyway, I just wanted to make sure 16 that -- I was hoping somebody up there had some 17 connection with the Navy experience. Because I was 18 pretty sure it's not hidden under a rock.
19 It's not like we're going to leap out and 20 tell people about it, but it certainly hasn't been 21 hidden.
22 MR. SALLEY: Yes. And it's not. I 23 remember the first time we briefed Commissioner 24 Ostendorff on this, you know, that was his exact 25 comment that you made was --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
45 1 MEMBER BROWN: He knew about it.
2 MR. SALLEY: Yes, hey, why don't you just 3 change all these breakers out and put these design 4 with the arc shoots and all these other things.
5 MEMBER BROWN: Well, it's not the breaks, 6 it's a separate little box.
7 MR. MELLY: Yes.
8 MEMBER BROWN: You stick it in there with 9 the sensors and stuff. It's not a -- you don't have 10 to change the breakers.
11 MR. SALLEY: And, of course, in our 12 environment, this would be a backfit. So, again, 13 we're working through the generic issue program and 14 where ever the generic issue program takes, it will 15 stay in process.
16 So, we recognize what you're saying. And, 17 like I said, we've heard that from Commissioner 18 Ostendorff, but we're trying to stay through our 19 process as we go through this.
20 MEMBER BALLINGER: And, I might add that 21 the current 04 and 05 admirals are nuclear engineers.
22 MEMBER BROWN: How does this put in the 23 framework I just finished a discussion, the safety 24 posture? I mean, it's -- is this a safety threat to 25 the plant or it's just, what do you call it, the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
46 1 balance of plant type --
2 MEMBER BLEY: It can hit anywhere, right?
3 MEMBER SKILLMAN: Both, it's clearly both.
4 MEMBER BLEY: So, they've got to 5 understand the frequency of it, the range of impacts 6 they've seen. It's not just high energy arc faults, 7 but heavy faults have not only caused a fire at the 8 immediate location, but they -- it causes remote fires 9 before they get tripped out and sometimes two or three 10 fires around the plant.
11 So, there's a pop hazard from it as well 12 as the explosion is.
13 MEMBER SKILLMAN: Charlie, the way I see 14 what the research is doing here is, if you recall 15 about a year ago, AIM or whatever the money reduction 16 project was was to get rid of this.
17 When we dug in our heels and said, no, 18 keep it.
19 MEMBER BLEY: To get rid -- oh, the 20 program? Oh okay.
21 MEMBER SKILLMAN: And, it is -- we're 22 hanging on to this HEAF research, but the real goal is 23 to get quantification for risk so we can look at 24 nuclear, not nuclear, both sides of the plant, make an 25 informed decision of what further activities we need NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
47 1 to do.
2 And so, these gentlemen are here to really 3 paint the picture in terms of how they've come to 4 where we are, what information they have and where 5 they want to go in the future.
6 MEMBER BLEY: But, sort of a point to 7 where you were talking --
8 MEMBER SKILLMAN: Yes, it's right where 9 you want to go.
10 MEMBER BLEY: Like, even though we see a 11 pretty good of them. We say, well, that happens 12 fairly often, if you live in one plant, you probably 13 never seen one of these.
14 MEMBER SKILLMAN: That's right.
15 MEMBER BLEY: And, you don't even think 16 about it unless you're sharing information and really 17 --
18 MEMBER SKILLMAN: But, we have not had a 19 lot of those in the Naval Ships. There were very few.
20 But, the potential --
21 MEMBER BLEY: Had some very interesting --
22 MEMBER SKILLMAN: But, the potential for 23 the end result, and we ended up -- it's not -- was not 24 expensive. I mean, we're talking a few hundred 25 thousand dollars, you know, for each ship to go put NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
48 1 the sensors and stuff in.
2 MEMBER BROWN: For the whole ship?
3 MEMBER BLEY: Yes, you want me to --
4 MEMBER BROWN: The accounting might be 5 different then.
6 MEMBER BLEY: No, I -- the base plants are 7 different. There is more switch gear there to higher 8 levels and we had a limited number of switchboards 9 that we had to go deal with. So, I mean, it's a 10 smaller thing.
11 But, I mean it was a matter of I'm more of 12 a solution -- once I know it's a problem, I don't need 13 more research, I just go try to fix it.
14 MEMBER BROWN: But the other side is --
15 MEMBER BLEY: If you find you need to fix 16 it is the point.
17 MEMBER BROWN: And, you said this, a power 18 plant loses a little bit of the plant, hopefully not 19 people, but they could 20 MEMBER BLEY: Yes.
21 MEMBER BROWN: The ship, you lose a whole 22 ship if you don't have -- if you go dead in the water.
23 You know, eventually you're --
24 MEMBER BLEY: A submarine is a submarine.
25 (Simultaneous speaking.)
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
49 1 MEMBER BROWN: Surface ship's not too darn 2 good if it goes dead in the water.
3 MEMBER BLEY: That's right.
4 MEMBER BROWN: But, a carrier is not going 5 to --
6 MEMBER BLEY: But there were 50 airplanes 7 at $50 million apiece in the air.
8 MEMBER SKILLMAN: All right, let's let the 9 guys go. Let's go on, team, let's move on.
10 MR. HAMBURGER: All right, so on the one 11 of the comps that we've received is that in our 12 investigation and analysis of these HEAF events we're 13 actually assuming two failure.
14 We're assuming that the arc initiates and 15 we're also assuming a failure of a protective breaker.
16 And, as the thinking goes, if the breaker 17 protection scheme is functioning as designed, it will 18 either eliminate or it will limit the duration of 19 these events.
20 Unfortunately, that's not borne out by 21 what we're seeing in operating experience. These 22 things are happening. They're happening fairly 23 frequently, as Nick's slide showed.
24 And, in 2008, the Division of Fuel 25 Facility Inspection did a roundup of root cause NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
50 1 failures for these breaker mechanisms.
2 So, they're listed up here on the slide.
3 There's a variety of causes and I think that's 4 important. There's no single reason that these 5 breakers are failing. There's no easy fix.
6 MEMBER BLEY: We know, though, we know 7 Robinson lasted a long time because the breaker didn't 8 trip.
9 MR. HAMBURGER: Yes.
10 MEMBER BLEY: Do you have a good handle on 11 how many of the other arc faults you've look at? You 12 also had breaker trip problems and they'd lasted 13 longer than one would have hoped from selective 14 tripping.
15 MR. HAMBURGER: So, in almost all of the 16 events that we've counted in our Bin 16, our heat 17 frequency, there has been some associated breaker 18 failure.
19 MEMBER BLEY: I'm really curious now. Has 20 it been associated with currents of the magnitude 21 people weren't thinking when they looked at the trip 22 mechanisms in the breaker or was it a pre-existing 23 problem in the breaker?
24 MR. HAMBURGER: We've seen both. So, 25 we've seen breakers that have failed to trip.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
51 1 MEMBER BLEY: Yes.
2 MR. HAMBURGER: We've also seen breakers 3 that tripped as they were intended, but the set points 4 were not sufficient to prevent the damage that 5 occurred as a result of the HEAF.
6 MEMBER BLEY: And the set points were 7 where you wanted them to be?
8 MR. HAMBURGER: Yes.
9 MEMBER BLEY: Where they were supposed to 10 be?
11 MR. HAMBURGER: Yes, and there have also 12 been design changes that resulted in set points that 13 were no longer appropriate for the function of that 14 breaks.
15 MR. MELLY: We've also seen faults in 16 locations where the circuit protection worked as 17 designed. However, the design has led to a generator 18 set fault even after the spin down.
19 So, the breaker clears, however, the fault 20 is still being fed by spin down.
21 MEMBER SKILLMAN; I would add one very 22 important feature in the U.S. nuclear industry is the 23 unending pressure on the sites to use the corrective 24 action program. And, when they identify an issue, 25 they are obligated to report in operating experience NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
52 1 space.
2 And, for better or for worse, that feeds 3 gobs of information into Atlanta that can be mined by 4 the industry.
5 And so, as difficult and as time consuming 6 as that is for the station staff, it enables you to 7 get real information, virtually within hours of an 8 event that could be quite serious.
9 So, there is every reason to believe that 10 your data will continue to get more current and more 11 accurate. Because the tools that are being used are 12 beginning to -- are more sophisticated than they were 13 a long time ago. That's a very good thing.
14 MR. MELLY: Yes, and that's exactly what 15 we're seeing with the new frequencies. We're actually 16 pulling directly from condition reports from the plant 17 to get this operating experience has a whole lot of 18 information that we're looking for such as the 19 duration of the event, the power levels and we're 20 requesting even more associated with this that we're 21 looking for.
22 So, it has come leaps and bounds forward.
23 Another thing that leads into this, we 24 took a look at the events that constitute the Bin 16 25 fires as to that root cause information that we were NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
53 1 just discussing, what has been provided from the 2 plants looking for is there a single failure cause for 3 these high energy arcing faults and what we're seeing 4 is that it's all for the board.
5 Some of them are unknown root cause, 6 because, as you'll see from some of the damage states, 7 the whole cabinet is destroyed. It's very difficult 8 to do a root cause analysis when you're starting from 9 a pile of scrap.
10 We have human error, such as the Robinson 11 event, breaker failures, poor electrical connections, 12 over current conditions, aging and degradation moving 13 forward, loose or foreign materials such as the Turkey 14 Point event or design deficiencies.
15 And some of them were a combination of 16 both. So, it's not a quick fix. It's not one thing 17 that we can change or focus in on to try and alleviate 18 these from occurring.
19 We're going to briefly now discuss the 20 definition of what the HEAF hazard is. I'll try and 21 go through this one relatively quickly.
22 MR. HAMBURGER: So, the HEAF is a fairly 23 complex phenomenon and it doesn't have a single mode 24 of damage. It has many modes of damage. And, often 25 what we've seen is that the damage state is a result NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
54 1 of a confluence of these factors.
2 So, the arc strikes, it produces 3 tremendous temperatures that really we don't see too 4 many places on earth, 3,500 degrees Fahrenheit is --
5 it can't be overstated how hot that is.
6 And, as a result, you have resistive 7 heating of the air between conductors. You have a 8 rapid expansion of air. It erodes the conductors 9 which causes vaporization of the metal conductors.
10 We have a factoid here that when copper 11 expands from solid to vapor --
12 MEMBER BLEY: I've never seen that one 13 before.
14 MR. HAMBURGER: -- phase, it expands by a 15 factor of 67,000 times. This is a very rapidly 16 expanding ball of molten and vaporized metal.
17 Some of this is ejected from the cabinet.
18 We have plasma coming out of the cabinets, ionized 19 gases and, in some case, we've even seen detonation 20 like effects.
21 So, we've seen shockwaves, we've seen 22 overpressures that are impacting equipment that's 23 fairly distant from the cabinet of origin.
24 And, we also have missiles, shrapnel, 25 things being thrown from the cabinet such as doors.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
55 1 So, there's a lot of potential damage modes for our 2 targets.
3 MR. SALLEY: As we heard from Dr. Bley 4 that, you know, does this occur outside of nuclear.
5 And, the answer to that is yes. And, we've been 6 reaching out to that.
7 As a matter of fact, as we were briefing 8 the EDO and especially Fred Brown in particular of 9 DEO, you know, he looked at it and it almost, you 10 know, why do have to find this in nuclear? I mean, 11 this is something that would be happening elsewhere.
12 And, you're absolutely correct. We've 13 reached out to folks like the NFPA who has the 14 National Electric Code and the IEEE groups who have 15 the standards and the committees for this.
16 We've included them with some our public 17 workshops.
18 One of the things we've notices, and 19 again, we've learned from what we've done in 805 and 20 6850 is that Bin 15 was way too course. You saw that 21 as the big driver for the fires and skyscraper chart.
22 But, the piece you don't see is that when 23 they did the cabinet counts that a small cabinet that 24 had communication -- low voltage communication versus 25 a 6.9 KB breaker, each counted as one and they split NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
56 1 the same frequency.
2 And, we all know that it just doesn't add 3 up. So, as we're smarter, we're moving forward with 4 Bin 16, we want to be more precise. We want to be 5 more accurate. We want to get more, the buzzword from 6 industry, realistic figures.
7 And, with that, one of the things we 8 needed to do was to get good definitions. Now, that 9 sounded trivial and easy to start up, but as Stan 10 Miller's going to tell you in a minute, that's 11 something that we're still working on because the 12 electrical community and industry, the commercial 13 industry, you'll hear arc flash.
14 And, you can go on YouTube and you can see 15 a bunch of different arc flashes and they're big 16 things. Nick said earlier, it's personal protective.
17 And, they tend to limit things at two 18 seconds. That tends to be what their standards use.
19 And, for the personal protection, the two second is 20 with a gear wear.
21 What makes the HEAF a little bit more 22 unique and it does occur outside. As a matter of fact 23 it happens here on the Metro is when the events to 24 last longer than that two seconds. And we've seen 25 them out to eight and 11 seconds.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
57 1 And, that duration tends to be the thing 2 that drives the HEAF.
3 So, with that background, I'd like to turn 4 it over to Ken. And some of the work we're doing with 5 NFPA and the IEEE groups for the definitions.
6 Ken?
7 MEMBER BLEY: Before he goes ahead --
8 MR. MELLY: Yes?
9 MEMBER BLEY: -- because you hit on Bin 15 10 which is the enclosure fires, I remember four or five 11 years ago, maybe not quite that long, we had a big 12 meeting in here and EPRI was going to put some of 13 their old hands that did fire risk work together to go 14 clarify the frequency of -- and divide it up out of 15 all those bin fires to what's important and what's not 16 and how much.
17 They never got there. They never finished 18 that work. They never came back. And so, it's not a 19 trivial task, I take it to work this out.
20 MR. SALLEY: No, and it goes back to the 21 whole re-baselining and some of the premises that it 22 was originally started out with.
23 And, you hear the argument the number of 24 cabinets. We have the frequency and you divide by 25 more cabinets, you get the lower numbers and some of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
58 1 the challenges to the PRA. Nick, you can speak to 2 that, we've talked a couple, three times with EPRI to 3 restart that efforts and --
4 MR. MELLY: Yes, it's been the current 5 discussion to move away from -- to move towards 6 component based frequencies. That's the fusion 7 argument that will be in 40 years. They've just never 8 gotten around to it and they don't have a -- it's not 9 one of their big ticket items to do right now.
10 MR. MILLER: So, again, referring to the 11 slide here, we're talking about definitions and you 12 see the terms, arc flash, arc blast and HEAF. And, 13 again, flash being the term that we've heard quite a 14 bit. A lot of that associated with personnel 15 protection and they -- there's been a lot of focus on 16 arc flash and there's an IEEE standard on calculating 17 it for determining protective device clothing for 18 people that are working it.
19 The arc blast gets into more of the high 20 energy kind of event where you could have more than 21 just the regular energy, but the chemical energy that 22 comes out of an event like this.
23 And, then, of course, the HEAF where 24 typically it's dealing with a much longer duration.
25 And, typically, you know, failure or protection that's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
59 1 allowing the event to run so long.
2 So, the next three slides basically, the 3 approach with the definition and maybe binning the 4 HEAF event is putting it in three categories based on 5 the outcome of the event.
6 Class 1 being the least severe event, 7 typically defined -- typically the damage is within 8 the component of origin.
9 Yes, as it says here, associated with 10 minor damage and mineral bus bar degradation from 11 melting and vaporization..
12 NFPA, electric code, you know, there's a 13 lot of discussion about this level of event. And, if 14 you go on to the second level, here in the Class 2, 15 not only is the damage with the component of origin, 16 but because of the blast, there are -- there's also 17 damage occurring to components in the area. And, 18 again, a much more severe event.
19 Typically, at this level, you're starting 20 to get into cases where, you know, there's some issue 21 with the protection misbehaving and not eliminate or 22 opening the fault soon enough.
23 Of course, the pressure effects that have 24 occurred within the area are dependent on the, you 25 know, the room configuration and the characteristics NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
60 1 of the equipment.
2 And, then finally, in the Class 3, this is 3 where, again, you have all the effects, the component 4 of origin is severally damaged. There's surrounding 5 components that are damaged and you can also have 6 effects that affect fire zone barriers.
7 And, there have been events like that, 8 too, where doors separating two fire zones have been 9 pushed open because of the shockwave and, again, 10 getting to the mechanical design of the room and its 11 ability to relieve pressure and can contribute to, you 12 know, the event becoming more -- affecting more than 13 a larger area.
14 MEMBER BLEY: I'm just --
15 MR. MILLER: Yes?
16 MEMBER BLEY: I'm mixing two things in my 17 head, especially from transformer fires, big ones, and 18 exposed and associated with them the fault, I know 19 have led to two or three fires around the plant 20 sometimes which is difficult to deal with if you have 21 more than one.
22 Has that happened with the arc faults as 23 well? Multiple fires coming or has it all mostly 24 local damage?
25 MR. MELLY: I can take this one. We have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
61 1 seen several events were there have been multiple 2 fires. The Robinson event is a good example of that 3 where we actually had two separate high energy arcing 4 faults that are connected in their root cause.
5 However, typically, we won't see these 6 creating fires in separate rooms right initially. Not 7 like the main generator fires and the generator or the 8 transformer fires are separately from the main 9 transformer. They're not included in the frequency of 10 the high energy arcing fault, they're in a separate 11 bin.
12 MEMBER BLEY: Understand.
13 MR. MELLY: But, typically, we do not see 14 multiple fires --
15 MEMBER BLEY: I mean, there's a place I 16 remember seeing multiple fires --
17 MR. MELLY: Yes.
18 MEMBER BLEY: -- because the currents are 19 pretty big in lots of places.
20 MR. MELLY: Yes. And, that has occurred, 21 but for the high energy arcing faults and the 22 methodology, what we've seen from operating 23 experience, there may be multiple fires within a 24 single room.
25 MEMBER BLEY: Okay.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
62 1 MR. MELLY: However, it's typically not 2 end in multiple rooms.
3 MEMBER BLEY: Okay. And the energy 4 release and the currents elsewhere.
5 MR. MELLY: Yes.
6 MR. MILLER: So, this slide is just, 7 again, trying to show you the three classes, the Class 8 1 and 2 on the top are there again associated with 9 normal -- more proper electrical protection operation.
10 And then, the third class being the lower 11 bar and showing the, you know, extensive damage with 12 a Class 3 type HEAF event.
13 So, again, the point of the three 14 classifications is to bin the HEAF events based on the 15 severity and how much the surrounding equipment is 16 affected and/or the Class 3 level potentially 17 affecting other fire zones which makes electrical 18 engineers nervous also because of the potential to 19 affect different trains of equipments like another 20 safety train being affected.
21 MEMBER SKILLMAN: Can the image on the 22 lower right hand is a fire door between compartments?
23 That is a large double door. And, that door has been 24 deformed.
25 MR. MILLER: Yes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
63 1 MEMBER SKILLMAN: How far away from the 2 HEAF is that door?
3 MR. MILLER: I don't know the specifics.
4 MR. MELLY: This is actually an event that 5 occurred in Germany. I don't have the exact distance 6 away for this fire door, the double doors. However, 7 the event above it in the arc blast category, this 8 door depicted here with the deformed latch, that was 9 actually from Turkey Point. And that event -- or that 10 was 14 and a half feet away from the arc initiation 11 point.
12 MEMBER BLEY: Is that from the pressures 13 or did that come from flying debris?
14 MR. MELLY: The pressure.
15 MEMBER BLEY: The pressure?
16 MR. MILLER: More like a shockwave, yes.
17 MR. MELLY: Yes.
18 MR. MILLER: And, again, getting into the 19 design of the ventilation of the room, that can affect 20 the --
21 MEMBER BLEY: Oh yes.
22 MR. MILLER: -- severity of this as well.
23 MEMBER BLEY: Well, I'm assuming that 24 there was a faction was or wave to push this door 25 open. And if probably, it would probably --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
64 1 MR. MELLY: Yes.
2 MEMBER BLEY: Probably yank it off i8ts 3 hinges or deform the latching system.
4 MR. MELLY: And, there actually is 5 information that we're going to discuss in a bit about 6 that Turkey Point event citing that this pressure 7 damage to that door occurred at 14 and a half feet 8 away from the arc.
9 Our current guidance in Appendix M only 10 would assume damage to that door if it was in that 11 three-foot zone of influence inside the cabinet. That 12 is one area that we're taking a look at ion the Phase 13 II of testing, looking at the pressure created by one 14 of these arcing events within the cabinet.
15 And I'll show you some of the difficulties 16 with that pressure measurement that we saw from Phase 17 I partially due to some EMI interference of the 18 instrumentation that we're trying to move forward with 19 the test laboratory.
20 MEMBER BLEY: Now that you've partitioned 21 these into the classes based on damage, did you have 22 enough information to go back through the data you had 23 on all the arc faults you had and partition them as 24 well? What fraction are in each of those three 25 categories?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
65 1 MR. MELLY We have not done that exercise 2 yet --
3 MEMBER BLEY: Do you have enough 4 information to do it?
5 MR. MELLY: I believe that we will have 6 enough info9rmation either anecdotally from the 7 condition report to the LARs or potentially other 8 special investigations.
9 That is an exercise that we're going to be 10 doing with EPRI under the MOU to create these 11 frequencies.
12 MEMBER BLEY: Okay, good.
13 MR. HAMBURGER: It's also work noting that 14 not all of these are currently classified as Bin 16 15 events. Some of the smaller events are classified 16 either under Bin 15 as being thermal fires or they're 17 considered non-challenging and they're not counted at 18 all. So, this would really be an exhaustive review of 19 all those events to try and re-bin them.
20 MEMBER BLEY: Okay. And all of -- even 21 those small ones are in the data you showed us 22 earlier?
23 MR. HAMBURGER: Yes.
24 MEMBER BLEY: Okay.
25 MR. MILLER: And, again, you know, this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
66 1 definition activity has been something we've been 2 working very methodically with and trying to harmonize 3 with NFPA and IEEE to make sure that we get mutual 4 agreement on defining these things so that then it 5 would make sense to use them once we're all agreed.
6 MEMBER BLEY: I apologize for grinning, 7 but I spent more than ten years on an IEEE committee 8 trying to define terms in --
9 MR. MILLER: Yes.
10 MEMBER BLEY: -- an availability related 11 program. And, it finally just shutdown, they gave up.
12 MR. MELLY: Well, we have spoken with 13 them, they are very interested in this and they asked 14 us what our criteria were.
15 MEMBER BLEY: We're very interested in the 16 other --
17 MR. MELLY: Yes, and they were saying, 18 well, do you have criteria because we don't. And so, 19 now, we're collaborating working with them to develop 20 it.
21 MR. MILLER: Yes, when Mark asked me to 22 come in and help him with definitions, oh that can't 23 be too bad. But it's been a lot more than I thought.
24 MEMBER SUNSERI: I'm not trying to be too 25 picky here, but I've read the information notice on it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
67 1 and it clearly identifies that as a high energy arc 2 fault.
3 You're calling it an arc blast up here, so 4 I know you're going to talk about this a little bit.
5 So, my question is going to be, as we go through 6 characterizing these things, are we cutting this thing 7 too finely and missing some big point here? Because 8 this Turkey Point thing seems like a big deal to me.
9 So, just keep that in mind when you get to 10 the Turkey Point discussion.
11 MR. MELLY: Yes, I'm not sure if we have 12 a slide dedicated to Turkey Point later on, so I may 13 as well discuss it now.
14 MR. MILLER: My own proponent as Turkey 15 Point would probably be more like a Class 3 just 16 because, you know, this door that was wedged is 17 between fire zones.
18 And, again, as soon as you get out of a 19 fire zone, in my mind, that makes me worried.
20 Because, again, I'm thinking in terms of electrical 21 distribution. If you're affecting multiple trains of 22 equipment then you're getting into trouble with the 23 general design criteria.
24 MEMBER SUNSERI: Right. I mean, it goes 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 1 MR. MILLER: Electrically.
2 MEMBER SUNSERI: The information notice 3 says there was smoke intrusion into the other rooms.
4 MR. MILLER: Which can -- and then, even 5 the smoke can cause issues also with electrical 6 equipment.
7 MR. MELLY: Yes, that's a definite 8 discussion that will be taken up as to how do we 9 classify this one?
10 Some of the challenges associated with 11 that is, if this does go to full classification Class 12 3 counted in that frequency, the way that it's 13 currently modeled would assume that this did have an 14 ensuing fire immediately off the bat and a three-foot, 15 five-foot damage zone which we didn't see during this 16 event.
17 So, it's going to have to link up with the 18 say that we model it. There's a challenge we're going 19 to be taking up with EPRI.
20 MEMBER SUNSERI: Okay, thanks.
21 MEMBER SKILLMAN: Okay, let's go.
22 MR. MILLER: So we --
23 MR. MELLY: I --
24 MR. MILLER: Go ahead, next slide.
25 MR. MELLY: We're going to be discussing NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
69 1 the background of the program now, what led us to our 2 work.
3 MR. SALLEY: Yes, so quickly, by way of 4 background, one of the programs we belong to with the 5 OECD NEA is the fire events database.
6 This is a nice little program that a lot 7 of the countries who don't have a lot of operating 8 experience in frequencies are trying to get a database 9 to use for their fire PRAs.
10 We contribute to it and it's generally the 11 more significant fires. For example, the threshold 12 for the United States that we report the things that 13 are LER and above. So, it's typically, 5, 7, 9 events 14 a year.
15 In discussions --
16 MEMBER BLEY: Can I interrupt you on that 17 one?
18 MR. SALLEY: Sure.
19 MEMBER BLEY: And, I haven't looked at the 20 rules on LARs lately, but there was a point in time 21 when they shifted from kind of anything that caused a 22 reactor trip to anything that affected more than two 23 trains of a single system.
24 So, if you have one of these that only 25 affects one train, it wouldn't necessarily be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
70 1 reported. Is that still true or is --
2 MR. SALLEY: Nick can take that.
3 MR. MELLY: In addition to that criteria, 4 there's also a timing associated with events. So, 5 there's a threshold time to 15 minutes that if they --
6 fire lasts longer than 15 minutes it does get reported 7 as an LER.
8 And, typically, these high energy arcing 9 faults, the fire spreading efforts all last longer 10 than 15 minutes.
11 MEMBER BLEY: So, you're pretty confident 12 --
13 MR. MELLY: Pretty confident that all --
14 MEMBER BLEY: -- that the big ones --
15 MR. MELLY: -- reported.
16 MEMBER BLEY: -- are picked up in there?
17 Okay.
18 Sorry for the interruption.
19 MR. SALLEY: Not a problem.
20 MEMBER SUNSERI: You could probably get 21 into that in another way, though. I mean, it's going 22 to probably cause the equipment to be inoperable which 23 would be reportable, too, right?
24 MR. MELLY: Yes.
25 MR. SALLEY: An interesting point about NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
71 1 this program, again, like I said, we have the EPRI 2 database that we use so we're pretty good in our 3 numbers.
4 What this program allows us to do is, it 5 allows us to look worldwide at the other countries and 6 say, okay, hey, we're having fires with this equipment 7 or this problem, are you having fires similar to that 8 in France, in Germany, in the UK?
9 And, it kind of gives us a check, you 10 know, internationally to what we're doing. So, it 11 serves a good purpose in that way.
12 And, one of the things that was 13 interesting is when we brought up the issue of high 14 energy arc faults, every country, all of a sudden 15 everybody's head kind of popped of and, oh yes, we've 16 had that and we've seen this.
17 When we started to look at the numbers, 18 and this data was taken probably early 2010 is about 19 when this collected and 2012, was 48 out of 415 events 20 were high energy arc faults internationally.
21 So, that kind of lit the lightbulb above 22 everybody and said, hey, this is significant. Okay, 23 so the next question was, the next slide, is, what do 24 we know about these events and how do we model them in 25 the PRAs and for risk applications?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
72 1 And, a group went off and looked at it and 2 basically they came back and parroted what we had in 3 6850 Appendix M, which we said, hey, you know, it's 4 based on one well documented even that we just 5 happened to find when we wrote it.
6 So, we clearly saw the need here to have 7 a better understanding of the phenomena and come up 8 with more accurate risk pools.
9 And, I'll just put one anecdote here 10 before I turn it back to Nick is that when we 11 originally started out with the testing, we were going 12 in a confirmation role that we were thinking that we 13 were going to say, okay, San Onofre was a pretty good 14 HEAF.
15 And, we're going to go and do some testing 16 and we are going to show that our Appendix M method 17 that everybody's using in the world is a relatively 18 good, conservative method and we feel comfortable with 19 that.
20 So, that was where we, you know, kind of 21 like Columbus where we set sail, you know, not knowing 22 that we were going to make it around or go off the 23 edge.
24 So, with that, Nick picks the next one up.
25 MR. MELLY: Yes, that's a good segue.
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 1 We have been looking at these 2 international events. And, some of the more 3 interesting ones that have occurred was the Onagawa 4 event in 2011.
5 The shaking -- Onagawa's actually the 6 closest nuclear power plant in Japan to the epicenter 7 of the earthquake and they had magna glass breakers, 8 or is it magna?
9 MR. SALLEY: Magna.
10 MR. MELLY: Magna glass breakers which are 11 the vertical racking breakers. They actually saw the 12 shaking misalign some of the stabs and caused a high 13 energy arcing fault in their 6.9 KB room.
14 They believe that spread and they actually 15 had two arcing faults within that room and the fire 16 lasted for seven hours.
17 The fire was so severe, and you'll see 18 some from the videos later on, you create so much 19 smoke, the onsite fire brigade could not fight the 20 fire or even find the fire.
21 And, the offsite fire brigade couldn't get 22 to the site. So, the fire lasted for seven hours.
23 I believe Mike Webber said that this was 24 actually the first fire that was reported to the NRC 25 when we were doing information sharing. So, he did 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 1 have experience with this event in real time.
2 MEMBER BLEY: Is there a good report on 3 Onagawa fire?
4 MR. MELLY: We do have a good report on 5 this fire. We also have a few presentations that have 6 been given by Japan on the fire.
7 And, additionally --
8 MEMBER BLEY: Some of us would love to see 9 those if you can make them available --
10 MR. MELLY: Yes, we can make them 11 available.
12 MEMBER BLEY: -- to us.
13 MR. MELLY: Additionally, Ken will speak 14 to later, we have a NUREG-IA with Japan based on some 15 of their testing that has a lot of details of the 16 Onagawa fire as well.
17 On the right hand side, you see a picture 18 of the San Onofre fire which occurred in 2001 which 19 was coincidentally when the NUREG-6850 writing team 20 was creating the methodology for 6850.
21 So, this was used as the template that is 22 fed into Appendix M, the three-foot, five-foot zone of 23 influence because they did have a large amount of 24 information from this event.
25 We were able to speak to real time plant 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 1 personnel and they developed a model based on this 2 event.
3 MEMBER SKILLMAN: I'm really impressed at 4 this Onagawa experience. Because one would think with 5 that oscillating current load and the I square R 6 losses that are accompanying that, that those breakers 7 would have stripped themselves.
8 You would have thought their own 9 protective devices would have stripped them.
10 MR. MELLY: Probably there are some --
11 MEMBER SKILLMAN: And, what you're saying 12 is that they did not and the current pulsations 13 propagated it.
14 MR. MELLY: They did. They believe they 15 had two arcing faults.
16 MEMBER SKILLMAN: Thank you.
17 MR. MELLY: And, as we discussed earlier, 18 it's not just the electrical enclosures that are 19 causing this problem, our testing is also looking at 20 bus ducts.
21 You see two examples from operating 22 experience on the screen. On the left, the Diablo 23 Canyon event and the Columbia event of the bus duct on 24 the right in 2009.
25 What you see in this picture on the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
76 1 Columbia event as well is a white -- essentially a 2 white room and cables above it.
3 And, during the discussion of that event, 4 there were eight feet of bus duct material vaporized 5 during that and four feet of the conductive material 6 with inside the enclosure vaporized.
7 And, if you actually go back and read the 8 LER and the root cause, there is a discussion on they 9 were actually looking for where these four feet of 10 material went and what was this white powder? And, it 11 was all vaporized.
12 So, it's an interesting read. And, until 13 we wrote the information notice and had the testing on 14 aluminum, we didn't put it all together because these 15 events are so few and far between.
16 MEMBER SKILLMAN: What is the compound?
17 MR. MELLY: Well, right now, we're looking 18 into that. The enclosure itself is made of aluminum 19 and the conductors in the Columbia event were aluminum 20 as well.
21 So, we've looking into whether it's pure 22 aluminum oxide or whether it's some in between and out 23 of the aluminum.
24 And -- or Gabe will discuss that in a 25 small scale test breaker.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
77 1 MEMBER SKILLMAN: Okay, thanks.
2 MR. SALLEY: You know, as a side point, 3 thinking back to the GDC-3 that we talked about for 4 fire protection, when a fire protection engineer would 5 walk by these bus ducts, they'd say, okay, there's a 6 copper bus or an aluminum bus, both of them 7 noncombustible, check, keep moving.
8 And, he wouldn't even have evaluated any 9 more for that explosion. But, again, that's what I 10 was trying to say earlier about, yes, it's in the 11 regulations, but no one really thought about it when 12 you're walking the plant down.
13 MEMBER SKILLMAN: So, he walks by and he 14 says, that's a potential thermite reaction, that's a 15 potential thermite reaction, that's a potential 16 thermite reaction.
17 MR. SALLEY: Well, it gets more complex 18 because how the NRC defined noncombustible, any metals 19 would turn out to be noncombustible so you would check 20 the noncombustible block and keep moving.
21 MR. MELLY: Yes. And, the center picture, 22 I'll just talk a little bit more about that when we 23 get to the video. But, that is the center picture of 24 the bus duct that we pulled out of Zion, the conductor 25 material within the enclosure itself was copper, the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
78 1 square peg conductors and the enclosure was made of 2 aluminum.
3 In that test, and you'll see the video, we 4 vaporized one and a half inches of the copper 5 conductor and seven inches of the enclosure material 6 around the copper conductor.
7 If Gabe Taylor's on the line, I believe --
8 MR. SALLEY: Yes, Gabe can take this.
9 MR. MELLY: -- this is his first slide.
10 MR. TAYLOR: Yes, this is Gabe Taylor.
11 I'm a Senior Fire Protection Engineer in the Office of 12 Research. Can everybody hear me all right?
13 MEMBER SKILLMAN: We can hear you well, 14 Gabe. Please continue.
15 MR. TAYLOR: Okay, all right, thank you.
16 Okay, so, leading into the testing program 17 that we did Phase I internationally, we knew that we 18 were going to pursue that testing.
19 We had the backing of our national 20 partners to do that, but we had asked ourselves, how 21 are we going to make the thermal and pressure 22 measurements to get data that's useable for us to 23 assess the hazards?
24 What we did is that we contracted with 25 Sandia National Laboratories to explore a number of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
79 1 different devices that could measure both the pressure 2 and the thermal type of exposure that we're getting 3 from this event.
4 They looked into passive devices such as 5 the bikini pressure gauges which is a piece of pipe 6 metal that sandwiches a piece of aluminum with a 7 number of different sized holes and, you know, an 8 aluminum break, it signifies a different pressure 9 threshold that's been reached.
10 We also looked at the thermal lacquers, 11 the cable ones and we also looked at different types 12 of active measurements, measurements that we could 13 connect to a data acquisition system and get an active 14 response, different types of thermal couples, space 15 thermometers, garden gauges and infrared temperature 16 sensors.
17 All that information is documented in a 18 Sandia report.
19 And, what we really found from that 20 endeavor was that there are a few devices that really 21 work. The plate thermometers worked well for the 22 active response measurements.
23 And, although the bikini pressure gauges 24 looked like they worked well, when we actually went up 25 and used them during some of the initial Japanese NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
80 1 tests, what we found is that a lot of the debris form 2 these events would break the foil off before the 3 actual pressure could affect the foil.
4 So, they didn't work as well in actual 5 testing.
6 They also looked at a pressure -- active 7 pressure measurements using a pencil probe. And, 8 again, while it worked well in the preliminary 9 endeavor, when we went and used them in full scale we 10 were getting a lot of electromagnetic interference on 11 the signal and it ultimately didn't work that well.
12 And, Nick will get in a little later on 13 how we tried to improve that and some of the things 14 that Japanese have done to get cleaner pressure 15 measurements.
16 But, from this type of testing we 17 basically found that, using flood parameters and 18 pressure -- or excuse me -- flood parameters and plate 19 thermometers were the best approach to look at the 20 thermal response.
21 MEMBER BLEY: Is this Sandia report in 22 ADAMS?
23 MR. MELLY: Yes, it is.
24 MR. TAYLOR: It is available from Sandia, 25 the report, and the link's provided there. I'm not NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
81 1 sure if it's in ADAMS. I believe it is.
2 MR. MELLY: Yes, it is.
3 MEMBER BLEY: Thank you.
4 MR. HAMBURGER: It's also on the --
5 MEMBER MARCH-LEUBA: There's an email 6 number.
7 MR. HAMBURGER: -- that we've give.
8 MEMBER BLEY: Oh, it is? So, we already 9 have it.
10 MR. MELLY: You should have it on the DVD 11 as well.
12 MEMBER BLEY: Are you going to talk about 13 the left side of this slide or shall I just ask you 14 about the exit solar tower?
15 MR. TAYLOR: So, the solar tower was what 16 we used to provide the thermal flux. It's an array of 17 mirrors that they have out at Sandia National 18 Laboratory. They consolidate all the rays on the 19 mirrors to a tower and they can basically open a 20 shutter and behind that shutter would be where are 21 gauges would be so we can control the duration of the 22 exposure --
23 MEMBER BLEY: Oh, that's pretty neat.
24 MR. TAYLOR: -- and they can calibrate the 25 mirrors to get -- we tested .5 megawatts per meter NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
82 1 squared up to 2 megawatts per meter squared.
2 MEMBER BLEY: Wow. So this stuff must 3 have come out of their solar energy program?
4 MR. MELLY: Yes.
5 MEMBER BLEY: Over the years. Okay.
6 MR. MELLY: Okay, moving forward into the 7 Phase I testing now that we've had a little 8 background. And, we'll cover some of that information 9 or instrumentation as we move forward into Phase II 10 and some of the improvements.
11 MR. SALLEY: Yes, and we'll look at the 12 testing, Dennis, to answer your question, this package 13 of information we use, we share this with a lot of 14 people. And, we try to put the references, you'll 15 notice the links are there to download it.
16 MEMBER BLEY: Oh great.
17 MR. SALLEY: So, this whole package has 18 got a lot of the links in there that are embedded to 19 download.
20 You'll also notice that --
21 MEMBER BLEY: Did we -- we got this 22 electronically I hope?
23 MEMBER MARCH-LEUBA: Yes, on Monday.
24 MEMBER BLEY: Oh, okay, great. Thank you.
25 Oh, it's on -- got you, got it. Go ahead.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
83 1 MR. SALLEY: And, this is the final report 2 that we put together. And, again, it's not a NUREG.
3 You'll notice that we're doing quite a bit here with 4 the OECD and the NEA. So, these are being published 5 internationally.
6 One of the things that we've talked about 7 with them is when we do this, you know, it's a lot of 8 times that those programs, they'll hold that data for 9 five to seven years for the countries that pay for it.
10 With this program and the safety 11 significance that we think that goes with it as well 12 as the open philosophy of the NRC being transparent, 13 we've been publishing things as soon as we've been 14 getting them done and we've been putting them out in 15 the open literature.
16 So, again, this is what the final report 17 looks like and it has our testing. So, I'm kind of 18 putting that out in front of you and then we'll work 19 back as to what's inside the report.
20 MR. MELLY: So, this is a parameter 21 o9verview of what we looked at in the Phase 1 testing.
22 We looked at both low voltage and medium 23 voltage equipment.
24 We also looked at a range of currents for 25 those equipments associated with our -- the operating NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
84 1 experience that we looked at.
2 And, we looked from anywhere from 4160 to 3 6.9 KB.
4 And, you can see from this chart here that 5 we were getting equipment through our Phase 1 program 6 almost from the international community through 7 donation. And, that was an effort to keep the cost 8 down and lower for this because, after you test one 9 piece of equipment, you cannot test it again based on 10 the damage.
11 So, we were requesting equipment and the 12 contribution around the country was just through 13 donation.
14 And, you can see that there were a lot of 15 different parameters were varied in this testing which 16 made it difficult to do a one to one comparison.
17 So, as we move forward, you're going to 18 see some of the changes that we've done in the efforts 19 to address that.
20 Additionally, we looked at one of the most 21 important factors that we have coming out of this 22 testing is the duration of the event itself. So, that 23 was important.
24 And, we tested anywhere from our goal of 25 two seconds to eight seconds.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
85 1 You do see several tests, five, six and 2 seven that did not meet the two second threshold and 3 that was because the arc extinguished prematurely.
4 That was a configuration issue and these were actually 5 some of the first tests that we ran. So, it may have 6 been an experience issue with the testing as well.
7 These tests were when we had visitors and 8 that could have been it as well.
9 MEMBER BLEY: On those -- oh no, five, six 10 and seven? Yes, five, six and seven, you had 11 substantial current. It's -- what snuffed them, do 12 you know?
13 MR. MELLY: Yes, they were low voltage 14 cabinets --
15 MEMBER BLEY: Okay.
16 MR. MELLY: -- and it's more difficult to 17 maintain an arc in a low voltage cabinet.
18 MR. HAMBURGER: I think those are target 19 currents as well, not measured currents.
20 MR. MELLY: They were the stipulated 21 currents that we gave to KEMA to initiate the arc.
22 MEMBER BLEY: Oh okay.
23 MR. HAMBURGER: So, we asked them to 24 provide a current, a voltage and a duration, yes, and 25 those tests didn't hold in.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
86 1 MEMBER BLEY: Okay, thanks.
2 MR. SALLEY: So, now, the challenge 3 becomes how do you do the testing? You know, this 4 isn't something classical fire protection where I want 5 to do an assembly and I get ASTME 119 and it's very 6 structured. It tells me what I have to do, how high 7 I have to measurement, where I put thermal couples, et 8 cetera.
9 We're kind of going off very exploratory 10 here as to how do we do this. Okay?
11 So, we're going to hook a piece of 12 equipment up and we're going to put a bolted fault to 13 it. All right, that much we know.
14 We use a piece of wire to do that to short 15 three phases, but how do we measure it and what do we 16 do with the measuring?
17 So, we looked here at a lot of different 18 partners and we primarily locked in with NIST -- NIST 19 and Sandia to help us with this measurement science 20 and to bring it together.
21 Before I turn Nick loose explaining all 22 the testing, I'd like to just give you an overview of 23 what the test assembly looked like.
24 Now, it evolved as we got smarter --
25 MEMBER BLEY: Can I ask you about where NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
87 1 you just were? I know you used the KEMA facility.
2 Are the KEMA folks over in Holland involved? Because 3 they've got some people who do this kind of stuff.
4 MR. SALLEY: That's kind of the mother 5 ship, the DNGBL I believe, if I got the acronym right, 6 is over there and --
7 MEMBER BLEY: Yes.
8 MR. SALLEY: -- KEMA is one of their 9 subsets over here. So, this is a subset of that, a 10 smaller laboratory. But, it's very convenient to us 11 up here in Pennsylvania.
12 And, what's also interesting --
13 MEMBER BLEY: No, what I was trying to ask 14 was the KEMA folks in Holland, you know, this is right 15 down their alley. So, they've probably got experience 16 in trying to set up oddball experiments because they 17 do them for people everywhere.
18 I'm wondering if they're part of your 19 group of international participants?
20 MR. SALLEY: And, the answer to that is, 21 yes. And, KEMA has communicated a lot with them. As 22 a matter of fact, when we had the public workshop over 23 here --
24 MEMBER BLEY: Yes?
25 MR. SALLEY: -- in April, their lead NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
88 1 expert came over and spoke with us.
2 I guess the thing that makes it unique 3 from what we're doing with KEMA, from what they are 4 seeing every day, is they are there trying to rate 5 equipment and qualify equipment.
6 MEMBER BLEY: That's correct.
7 MR. SALLEY: We're taking things to a 8 different level of failure and actually, when we did 9 this, it opened their eyes up a lot. And, when the 10 folks in KEMA Pennsylvania corresponded back to the 11 home office, there was a lot of interest over in the 12 UK and over in The Netherlands as to what we're doing.
13 And, they started following our project.
14 Because, they see some benefit in what 15 we're doing also.
16 And, the key gets really down to that 17 duration. I mean, most of their world is less than 18 two seconds. We're taking things, we're stressing 19 them well beyond that. And, that's not something 20 that's been done a lot out there.
21 MEMBER BLEY: Well, I interrupted you as 22 you were telling us how you're going to try to plan 23 these experiments. So --
24 MR. SALLEY: Okay.
25 MEMBER BLEY: -- try to catch where you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
89 1 were and go ahead.
2 MR. SALLEY: I'm just trying to give you 3 a complete answer, Dennis.
4 MEMBER BLEY: I appreciate it.
5 MR. SALLEY: So, we go with what we know 6 on this and we set a piece of equipment up, we hook 7 the electrics to it so that the power supply is there 8 for when they run the generator up and they drop it 9 in.
10 What do we measure? How do we measure it?
11 Well, as you heard Dave speak earlier that, you know, 12 then in the world of fire protection engineering, you 13 want to know the size of the fire, the power of the 14 fire.
15 And, we speak of that in terms of heat 16 release rate. And, the way we do heat release rate is 17 with a hood and we do oxygen count, telemetry, 18 consumption, we come up with the heat release rate and 19 we can get a nice profile of the fire.
20 So, what we had was NIST took a smaller 21 hood from one of their furniture calorimeters and we 22 set it up there and we were going to try to give it a 23 go to see if we could measure this in classical terms 24 of fire protection. So, that was one of the first 25 things we did.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
90 1 You heard Gabe speak to work we were doing 2 at Sandia trying to understand how plate thermometers 3 and things of that nature work.
4 So, what we did there was we put side 5 racks. And, interestingly enough, if you remember 6 Appendix M, it said that three feet away was the end 7 of the zone of influence, right? So, where do we set 8 those racks? Three feet.
9 So, we're trying to, again, qualify our 10 model in Appendix M. And, we hung our instruments on 11 there.
12 Also, if you go back in time, a couple 13 hundred years and look at fire test energy, you had to 14 do a lot of things empirically that, you know, what 15 did you see?
16 For example, what U119 talks about 17 firewalls. And, early on, they would take cotton 18 waste and when they would do a test, they'd put it on 19 the backside of the cotton waste to see that, hey, did 20 it ignite or did it not? Because the warehouse fires 21 and configurations and numerical was happening at the 22 time.
23 Similar to that, we put a cable tray in 24 that five-foot zone above it. And, the question was, 25 would we ignite the cables?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
91 1 So, again, it's a representative of what 2 we're trying to do.
3 One of the most important things we did 4 was the cameras. And, the camera, just to look at it 5 and to see it was going to be a very big 6 understanding.
7 We also have another technology that 8 Dave's going to speak to, and that was trying to use 9 IR to get an idea on those numbers. You heard Ken, he 10 -- every time I say that 35,000 degrees Fahrenheit, 11 it's just so off scale from what we're used to working 12 with in fire protection engineering.
13 So, that's the base layout and when you 14 see the tests as Nick starts walking through them, 15 keep in mind that that's our layout we're looking at.
16 And, if no questions, I'll turn it over to 17 Nick and he can go through some of the tests.
18 MEMBER SKILLMAN: Quick question.
19 MR. SALLEY: Yes?
20 MEMBER SKILLMAN: Mark, how does the test 21 leader know that the voltage and current, and hence, 22 the energy which is the time, are accurate? How do 23 you know you've got the right voltage? How do you 24 know you've monitored the right current for the right 25 duration so that you really get a full understanding NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
92 1 of the energy that you've imparted to the fault?
2 MR. MELLY: So, all of the electrical 3 energy, results and analysis are provided to us by the 4 KEMA Laboratory post-test. We have the full wave 5 forms of the duration during the event, the arc 6 voltage that they saw during the event itself as well 7 as what was supplied to the -- through the cabinet.
8 All of that information is part of KEMA's 9 normal, everyday deliverable for the rating testing 10 that they do. And, they provide very detailed 11 information. And, it is all included in our final 12 report.
13 MR. TAYLOR: If I could just add, this is 14 Gabe, they also do a number of calibration tests of 15 their equipment to make sure that they're hitting the 16 parameters that we're asking them for before we 17 actually even do a test.
18 MEMBER SKILLMAN: Yes, you know, I was 19 curious if they say they've got 16 point or 6 point 20 for a KV, you know it really is 6,400 volts and they 21 say they have 8,5000 amps, you know it really is 8,500 22 and not 6,500 that's been chinned up to 8,500. So, 23 you're really getting a profile of the true imparted 24 energy for the duration of the fault?
25 You're saying it's calibrated. They've NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
93 1 got certifications for the calibrations. You're 2 comfortable with those certifications. And, for all 3 intents and purposes, you know it to be accurate?
4 MR. MELLY: Yes.
5 MEMBER SKILLMAN: All right. Okay, 6 thanks.
7 MR. MELLY: Moving forward, this is some 8 of the Phase I testing. This test is test number 9 three. This was a cabinet that was donated to us by 10 Korea.
11 It was a low voltage cabinet. We tested 12 at 35 KA and this one was set in for 8 seconds. The 13 interior -- the conductive material was copper itself.
14 And, one important thing to note is when you see the 15 color of the smoke that is generated from this copper 16 fault.
17 Another one anecdote about this cabinet, 18 this is a very old vintage cabinet. It was built like 19 a battleship, very thick steel walls. And the 20 insulation inside the cabinet was actually mahogany 21 wood.
22 (Laughter.)
23 MR. MELLY: So, it gives you an idea of 24 sort of the vintage of the cabinet. And again, we got 25 all these donated to us.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
94 1 You see in this picture right here, you 2 see the heat release rate hood that Mark was 3 discussing. And the incoming low voltage power supply 4 was coming off the right hand wall.
5 We initiated the arc in the center cubicle 6 or the center right cubicle right here. And I will 7 play the test.
8 (Video played.)
9 MR. MELLY: So, as you can see, there are 10 a lot of flames generated and it looks like a very 11 exciting picture. However, this did not breach the 12 cabinet wall itself. We did not damage any equipment 13 beyond the zone of influence. This was a low voltage 14 test.
15 It was seen as a success and the arc did 16 hold in for the full eight seconds that we were 17 looking. So, some of the first testing, we were 18 seeing that we had low energy events for the low 19 voltage cabinets, things were looking good.
20 MR. HAMBURGER: While we're on this slide 21 and we have this up, I just want to note that that 22 beige skirt we have around the hood, that is just to 23 try to capture as much of the smoke and affluent as 24 possible.
25 And, as you saw, that initial arc is --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
95 1 there's enough pressure that it really doesn't capture 2 all the initial smoke that comes off of that event.
3 MR. MELLY: And you can also imagine that, 4 if this smoke was created in a small switch gear room, 5 you're filling the entire switch gear room immediately 6 with full smoke, which is why firefighting in these 7 events is so difficult.
8 MR. SALLEY: Notice the color of the 9 smoke, too.
10 MR. MELLY: Yes, it's a dark black.
11 MR. SALLEY: It was a dark brown.
12 MEMBER SKILLMAN: Is the vent duct off the 13 top of the canopy mechanically ventilated?
14 MR. MELLY: It was. We did have a fan on 15 that had been ducted and we were taking oxygen 16 consumption color and tree measurements throughout the 17 test.
18 MEMBER SKILLMAN: Okay, thank you.
19 MR. MELLY: So, it was we did get pretty 20 good results when we had the enduring fire. So, we 21 had very good heat release rate data for the enduring 22 fire.
23 However, the initial blast was off scale 24 and we didn't collect all the smoke. So, that initial 25 peak was not recorded.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
96 1 Now we move forward to a medium voltage 2 cabinet. This was used for breakers donated from 3 Germany. We tested at 10 KV for 15 KA for three 4 seconds.
5 These were oil filled breakers when they 6 were donated. However, we had to remove the oil for 7 concerns of explosions and a fireball and things like 8 that.
9 KEMA has had that happen at their 10 laboratory before. And, from their experience, they 11 said the fireball could be twice the size of their 12 facility or 80 feet. And there's a Pennsylvania SEPTA 13 line running right behind their facility.
14 We had enough fire department calls based 15 on our testing, so we decided to remove the oil and 16 alleviate that concern.
17 MEMBER BLEY: Nothing else in there, just 18 air?
19 MR. MELLY: Just air.
20 MEMBER SKILLMAN: But, if this had been an 21 arc fault on an oil filled breaker, of course burning 22 down in some remote cubicle in some nuclear power 23 plant, then what we are seeing here is just a minor --
24 MR. MELLY: Yes.
25 MEMBER SKILLMAN: -- reflection of what NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
97 1 really would have occurred?
2 MR. MELLY: Yes.
3 MEMBER SKILLMAN: Okay.
4 (Video played.)
5 MR. MELLY: I'm going to pause it right 6 here. Again, we had copper there and you can see that 7 the event did hold in for the time allotted. And we 8 had immediate ignition of all the cabinet internals 9 which were cables and into an air drop configuration 10 into the cabinet.
11 Additionally, we had full ignition of the 12 cable tray above the cabinet within 30 seconds of the 13 event.
14 And, again, this is medium voltage. We 15 are still seeing that it -- we're fairly bounded by 16 the 6850 methodology. We ignited, we did breach 17 portions of this cabinet.
18 Again, we're feeling fairly confident.
19 Again, the dark colored smoke.
20 Now, we're moving forward to one of the 21 later tests. This was a cabinet donated from Finland.
22 It was 480 volts. We tested at 40 K for seven 23 seconds.
24 One of the main differences of this 25 cabinet was that it had aluminum bus bars. And we'll NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
98 1 show a picture of that as we move forward.
2 That was not a parameter we initially set 3 out to investigate. It was give us what you have and 4 we will do the testing.
5 We initiated this arc in the front of the 6 cabinet itself, the lower front portion. And we will 7 discuss a little bit about the results of this when we 8 play the video.
9 (Video played.)
10 MR. MELLY: That was a shock to us in the 11 control room. We were taking bets on whether we would 12 even be able to maintain the arc because it was a low 13 voltage cabinet and we had difficulty in the past. We 14 did not expect this.
15 You will also not the color of the smoke.
16 It's a white plume that came out and it coated the 17 entire facility, the left hand side which is the 18 medium voltage incoming power supply which we'll show 19 a picture of as we move forward.
20 But, they -- we were shorting out the 21 equipment and the incoming power supply so we couldn't 22 even do any more testing after this test.
23 Additionally, we melted the 24 instrumentation test stand at three feet away from the 25 cabinet which had our in canal plate thermometers on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
99 1 them. We completely vaporized and destroyed our 2 instrumentation.
3 So, we weren't able to collect any data 4 from this test and it was a complete outlier from 5 everything that we'd been seeing in the previous 6 testing. And it was a low voltage cabinet, we didn't 7 expect this type of damage and we were still 8 scratching our heads as what just happened?
9 Additionally, the arc migrated from the 10 initiation point to a more substantial portion of the 11 cabinet following the electrical pathway. It migrated 12 to the upper portion of the cabinet which we'll show 13 in later pictures.
14 This was in October and it began what 15 happened during this test.
16 We then moved on to test the only bus duct 17 that was in this program. This was the duct donated 18 by Zion.
19 And, as you can see, we do not have the 20 heat release rate hood in this test because we had to 21 come up later, a few months later, we came up with the 22 Japanese and squeezed it into their test program 23 because we couldn't test anymore after that last test 24 that you just witnessed.
25 MEMBER BLEY: Exactly what kind of set up NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
100 1 did you have to initiate the arc?
2 MR. MELLY: To the --
3 MEMBER BLEY: I assume it's the same in 4 all of these?
5 MR. MELLY: Yes, it's the same for all 6 arcs that we initiated. We're following in the IEEE 7 standard, 20.37. I think I'm missing some numbers in 8 there. It's documented.
9 And is a -- for medium voltage moving 10 forward, we're going to be using at 24 gauge wire, 11 copper wire, copper tin wire. And, for the low 12 voltage we are going to be using a tenth gauge class 13 K stranded wire.
14 MEMBER BLEY: And you just set it up at a 15 distance and then hit it with the current or does it 16 actually start together and open up?
17 MR. MELLY: It starts there. So, we 18 actually go in and connect all three phases with this 19 wire, with this strand of wire. We'll drill holes 20 into the conductors themselves and wrap the cable 21 around.
22 MEMBER BLEY: Oh, through it? Okay?
23 MR. MELLY: Yes. So, the initial 24 vaporization of this wire is what creates the pathway 25 to initiate the arc.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
101 1 And this test, we initiated that in this 2 lower T portion of the connection here. And one 3 interesting note was, this was copper conductor. We 4 did not think that this was going to be an energetic 5 event at the time that we tested.
6 We hadn't even looked into the material 7 that the cabinet or that the enclosure was made out of 8 because we didn't think it was going to be important 9 factor to the damage of the result.
10 MEMBER BLEY: That last one you showed us 11 with the aluminum bus bars, you knew ahead of time 12 those were aluminum bus bars?
13 MR. MELLY: Yes, we knew ahead of time 14 that those were aluminum bus bars.
15 We lost sound on this one.
16 (Video played.)
17 MR. MELLY: I always have one video that 18 doesn't work.
19 So, that was a three second arc and the 20 video of it almost doesn't even do it justice for how 21 explosive and how energetic it was from even the 22 control room.
23 We had people ducking away from the 24 control room screen.
25 MEMBER BLEY: And you get paid for having NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
102 1 this kind of fun.
2 MR. MELLY: Yes.
3 (Laughter.)
4 MR. MELLY: So, we again saw the white 5 colored smoke during this event which left us 6 scratching our heads and then investigating that the 7 material was made of aluminum.
8 MR. HAMBURGER: So, now we have two points 9 to draw our line. We had the initial Finland cabinet 10 test and we had this bus duct test. And when we 11 realized that the housing was aluminum, the aluminum 12 was the standout result from this test series. It 13 clearly has an impact on the incident energy that 14 we're observing.
15 So, in terms of what the implications are 16 for our PRA and for our future testing, we're looking 17 at a potentially larger zone of influence from 18 cabinets and components containing aluminum than we 19 had postulated in 6850 Appendix M.
20 Given the temperatures and how far these 21 events were throwing molten metal and ionized gas.
22 We're also anticipating a higher risk of fire 23 propagation from these events containing aluminum.
24 And just what you're looking at here, on 25 the left side, you have the undamaged equipment, on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
103 1 the right side is the post test. You can see that we 2 basically vaporized the entire upper quadrant of this 3 cabinet.
4 So, it's gone. It's on the walls of the 5 cell that shorted out their incoming electricity 6 supply. And, as Nick mentioned earlier on the bus 7 duct, we eroded four inches of copper, the copper 8 conductors and the seven inches on the half of the 9 copper, seven inches of --
10 MR. MELLY: One and a half, seven inches 11 of the aluminum duct enclosure.
12 Next?
13 MEMBER SKILLMAN: Wait, before you change 14 that image, on the lower presentation, that appears to 15 be a duct and you've got treated wood as what? As 16 just cribbage or that's bracing or what is the wood 17 doing there?
18 MR. HAMBURGER: It's braking. It's 19 essentially because of the orientation that we had the 20 duct located in. We didn't want the arc to act like 21 a rocket engine and essentially damage or push the 22 whole cabinet backwards to damage the incoming power 23 supply feeds that we had coming in.
24 So, that was in an effort -- that was 25 actually KEMA that restrained this piece of equipment.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
104 1 MEMBER SKILLMAN: That was a low buck 2 restraint for that?
3 MR. HAMBURGER: Yes.
4 MEMBER SKILLMAN: Okay, that's fair 5 enough.
6 MR. HAMBURGER: And it's connected to 7 their floor bracing system that you see there.
8 So, one of the other standout results from 9 the test series was the mystery white material that we 10 had painted the walls of the cell with that ultimately 11 ended our October test series when we shorted out the 12 incoming electricity supply.
13 You can see it's quite stark here. On the 14 right, for Test 26, that brand new blue air filter 15 that they had installed just before we started 16 testing, it's not brand new anymore. It's covered in 17 some sort of byproduct and it's --
18 We are not a 100 percent sure what it is.
19 We have our suspicions. This is going to be one of 20 the major focuses of our Phase II testing. And, 21 unfortunately, it didn't occur to us at the time to 22 sample that material.
23 This type of failure mechanism, this 24 shorting out of whatever this byproduct is has also 25 been observed in operating experience.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
105 1 In the Fort Calhoun HEAF event in 2011, 2 they note that the -- there are adjacent cabinets that 3 were affected. They attribute the impact to this 4 conductive, combustive byproduct. So, this has also 5 been observed in operating experience as well.
6 MEMBER BLEY: Now, in the cases where they 7 used aluminum wire and houses, at least for a while in 8 California, I was actually bought one of those by 9 accident.
10 The aluminum oxide that occurs after 11 relaxation of the aluminum is not -- at least 12 partially nonconductive and it's very high resistance.
13 Is this -- well, you don't know, you 14 didn't sample this stuff. If this is a different form 15 of oxide, is it conducting at all? You don't know any 16 of that yet, that's what you're going to look for?
17 MR. HAMBURGER: Yes.
18 MEMBER BLEY: Well, you know, whatever it 19 is, it's conducting because it --
20 MR. MELLY: It's at least semiconductive.
21 MR. HAMBURGER: Yes.
22 MEMBER BLEY: Aluminum oxide is up your 23 power source, okay.
24 MR. HAMBURGER: If you had asked me before 25 I knew it had shorted out the equipment, I would have NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
106 1 said, it's probably all aluminum oxide. But, given 2 that it did short out the equipment, we think there 3 might be some oxidized aluminum in there as well.
4 MEMBER BLEY: Okay. So, it could just be 5 metallic powder.
6 MR. MELLY: And it's one of the main 7 focuses that we'll discuss in the small scale testing 8 program as well as we've added additional 9 instrumentation for the Phase II to specifically look 10 at what this material is as well as the conductivity 11 of it.
12 MEMBER BLEY: Did you know about other 13 aluminum fires before you got into this stuff? I 14 mean, there were things with deck plates and some 15 ships, that's why they're all steel now because it 16 makes a really nasty fire if you can get it going.
17 MEMBER BALLINGER: There was an infamous 18 aluminum fire in Connecticut underneath what used to 19 be called the Charter Oak Bridge in Hartford.
20 There was a scrap metal place that used to 21 get stuff from Pratt & Whitney who was nearby. And 22 they noticed -- they put the different piles of 23 material in different types and different piles.
24 And the aluminum pile started steaming one 25 day and then it caught fire and the fire melted NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
107 1 trailer truck terminals -- trailer trucks over a 100 2 feet away. And they had to re-survey the Charter Oak 3 Bridge because the fire went up under the girders on 4 the bridge.
5 MR. MILLER: How did it start?
6 MEMBER BALLINGER: Huh?
7 MR. MILLER: How did it start?
8 MEMBER BALLINGER: By itself. It self-9 initiated. The --
10 MR. MILLER: Wow.
11 MEMBER BALLINGER: -- they -- it was a hot 12 day, it rained and the aluminum was oxidizing and as 13 you get a bigger and bigger pile, you reduce the --
14 you increase the thermal resistance of the center.
15 And so, it ignited and turned into the -- turned into 16 a several hundred ton road flare.
17 MR. MELLY: Yes, we've seen several 18 international incidents. France has one associated 19 with aluminum in a reprocessing facility I believe.
20 We were aware of the risks with aluminum, however, we 21 did not foresee it to be such a large contributor.
22 MR. HAMBURGER: And this is the top of the 23 next section after these results, we went back and 24 looked at some of these HEAF event reports to see if 25 the aluminum was mentioned there. And we did find NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
108 1 several which ultimately led to the issuance of the 2 information notice regarding aluminum involvement in 3 HEAFs.
4 So, we did go back and look for this.
5 MEMBER SKILLMAN: Please proceed.
6 MR. HAMBURGER: This slide is really just 7 sort of a perspective on just how much of an outlier 8 those aluminum tests were.
9 On the left side, we have what we would 10 anticipate the incident energy to be at three feet 11 using the IEEE calculation for incident energy in arc 12 flash testing.
13 On the right, we have what was measured 14 with our slug calorimeters and you'll notice the 15 scales are different because for orientation effects.
16 But, if you look at Test 23, for example, 17 we were expecting that to be on the order of some of 18 our other copper tests. And if you look at the actual 19 observed incident energy, it exceeds the copper 20 testing by almost an order of magnitude.
21 So, this is a significant outlier in terms 22 of anticipated and observed incident energy at three 23 feet.
24 MEMBER BLEY: Is there some meaning to 25 your color coding?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
109 1 MR. HAMBURGER: Orange is aluminum.
2 MEMBER BALLINGER: Orange is -- so, five 3 and six, even though they're --
4 MR. HAMBURGER: Those are the arcs that 5 didn't hold in, but yes.
6 MEMBER BALLINGER: Oh, that's right, they 7 went out on you, yes.
8 MR. HAMBURGER: And on the lower two 9 graphs for our medium voltage tests, the Test 26 which 10 was our Zion bus duct, it's hard to say just how much 11 of an outlier that was because we destroyed our 12 instrumentation. So, we're assuming it's quite an 13 outlier in terms of incident energy based on video 14 footage.
15 MEMBER BALLINGER: It's really not an 16 outlier. All you need to do is ask yourself what the 17 free energy of formation of aluminum oxide is with the 18 copper oxide.
19 MR. HAMBURGER: Yes.
20 MR. SALLEY: So, Mr. Brown spoke earlier 21 about, okay, you've got a problem, solve it. And I 22 guess that's what engineers do, you see a problem and 23 you solve it.
24 With this issue, if you read some of the 25 correspondence with industry, we're still in the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
110 1 recognition identification that you do have a problem 2 phase, if you will.
3 But, folks are trying to solve things.
4 And one of the things we hear about is HEAF shields.
5 Hey, you know, if I'm going to have this HEAF and it's 6 going to take this cable tray out above and that's 7 going to give me some response to the plant that I 8 don't want, I'll just install a HEAF shield there.
9 And, in theory, it sounds pretty good.
10 Would it be workable? Sure, but, again, you know, all 11 the questions that come with the design, I mean, you 12 know, do I have to say thermal lag?
13 Because this is exactly how we went down 14 that thermal lag road. I mean, what's the -- what are 15 you -- what's going to be a qualifying test? You 16 know, how do you do that?
17 Where are, you know, are all the 18 parameters to standardize it to make sure that HEAF 19 shield does what it's supposed to do?
20 You know, these questions are unknowns.
21 So, again, is there a possible solution here? Sure, 22 there is, but there's a lot of work that needs to be 23 done to bring this to bear.
24 Likewise, again, engineers looking at 25 things want to do intuitive and 8610 evaluations and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
111 1 solve problems. And, they'll say, hey, okay, I got to 2 solid top to the cabinet. So, the HEAF's not going to 3 come out the top, it's solid, it's not open.
4 But, when we do the test, we see that a 5 lot of times the HEAF will burn through the top of the 6 cabinet. I mean, this is why we do the tests.
7 Likewise, we've heard arguments with 8 louvers. Hey, look at the cabinets, look how the 9 louvers always point to the floor. If I have a HEAF 10 inside, let's think it's a laminar flow and the 11 combustible material is going to come up and it's 12 going to hit the louver, it's going to direct it down 13 to the floor away from my cable trays and it's all 14 going to be wonderful.
15 And you could write a calculation up and 16 make it look very good with pictures and colors and 17 graphs. But, when we run the experiment, you can see 18 that the louvers don't survive. So, the whole 19 argument goes out the window.
20 So, again, we put these two slides in just 21 as a cautionary that would seem to want to make sense 22 or that you could do a good engineering evaluation.
23 When we actually do the testing, we see different 24 results.
25 So, that's kind of where we're at with NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
112 1 that.
2 MEMBER SKILLMAN: Okay, let's do this.
3 Thank you for moving along before Dr. Rempe left, she 4 asked me to please say thank you very much for your 5 work and for the presentation. I probably speak for 6 Walt, too.
7 With that, we are going to take a recess 8 for 16 minutes on that clock. Please come back at 9 1513, 3:15.
10 (Whereupon, the above-entitled matter went 11 off the record at 2:52 p.m.)
12 MEMBER SKILLMAN: Ladies and gentlemen, 13 let's bring the meeting back to order. Let's proceed.
14 MR. SALLEY: Okay. And the next piece 15 we're looking at here is the generic issue. Clearly, 16 we think we've got a generic issue. We're working 17 through the program.
18 And Tom Boyce and Stan Gardocki are both 19 here. So, the next couple of slides, it's best to 20 hear from them. So, Stan, are you going to take it?
21 MR. GARDOCKI: Right. I'm Stan Gardocki.
22 I work for Research.
23 The first slide here shows the process 24 overview of the Generic Issues Program. It's a pretty 25 busy slide, but I'll basically show you where we've NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
113 1 gone and where we are, and where we're going with this 2 generic issue.
3 But we've been working on this issue for 4 about a year now. We have gone through the screening 5 phase, and we issued a report last August on whether 6 it met the seven screening criteria to continue into 7 assessment stage. There's three stages: the 8 screening, assessment, and ROI, Regulatory Office 9 Implementation stage. So, right now, we've progressed 10 from the screening stage into the assessment stage, 11 and we produced an assessment plan we'll get to in the 12 next slides.
13 But, right now, I want to just point out 14 at the very bottom of this slide, in the middle, you 15 see ACRS. I'd just point out to your attention you 16 are ACRS and this is kind of where you're going to get 17 involved a little bit more. We send you the cc of all 18 our reports, so that you can see that we've got into 19 the assessment stage, and it's going to be reports 20 sent to you to say, do you want to get a little bit 21 more involved in briefings like you're having today?
22 You can request some briefings from Mark Banks to me, 23 and we can come down to discuss if you need to. We 24 look forward to putting out an assessment report in 25 about late 2019/early 2020. Okay.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
114 1 MEMBER BLEY: We appreciate that, and the 2 first showing up of ACRS is kind of new on your chart.
3 We have asked for that. So, we appreciate that you're 4 going to keep us hooked in. We are interested in 5 this.
6 MR. GARDOCKI: Okay. This slide we put in 7 here just to give you some idea where the ML numbers 8 of the reports. The screening report, August 21st was 9 published.
10 Okay. Next slide.
11 This is a little bit busy slide, but this 12 one is where we are right now. This is called the 13 assessment stage. And per MD 6.4, we put out a plan, 14 how we're going to finish the assessment in 15 approximately a two-year span.
16 And it's a pretty busy slide. On top you 17 can see important milestones about when we initiated, 18 and when the assessment report, we plan on getting it 19 issued. And down on the bottom part of it we put 20 milestones of tests and studies and pilot plant 21 studies. So, it gives you a pictorial of what we need 22 to do to get this assessment done. It's a lot of 23 work, and like I said, our guidance tries to get it 24 done in about a two-year span.
25 MEMBER BLEY: If I could throw something NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
115 1 in here? As I recall, there's a risk piece of that 2 assessment. Before that's all complete, we would 3 probably like to take a look at that and maybe offer 4 you some suggestions.
5 MR. GARDOCKI: Okay. The assessment 6 report has a couple of different parts to it. One is 7 risk assessment --
8 MEMBER BLEY: Right.
9 MR. GARDOCKI: -- safety assessment, and 10 a regulatory analysis.
11 MEMBER BLEY: And some of it has been kind 12 of dismissed on risk arguments that we thought were a 13 little too -- I won't put a word on it. We would like 14 a look before it gets closed, or whatever happens at 15 that point.
16 MR. GARDOCKI: Okay.
17 MEMBER SKILLMAN: Would you just talk us 18 through the next couple of events that seem to be on 19 this chart? One is the phase II large test, the 20 University of Maryland study, what appears to be the 21 scale for the aluminum HEAF, and the ZOI Working 22 Group. Can you walk us through that, tell us what 23 you're really up to?
24 MR. MELLY: Yes, I can walk you through 25 some of this. I'm going to start with the public NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
116 1 workshop that we held in April -- we'll cover that in 2 a little bit more detail -- when we kind of pinged the 3 industry on parameters and things like that to ensure 4 testing was going to be realistic.
5 We also have just recently completed our 6 small-scale testing that Kenny is going to discuss or 7 Gabe will discuss in a little bit. So, that's going 8 to elaborate to the small-scale testing that we've 9 done at Sandia.
10 We're, then, moving to our phase II 11 large-scale test program, which is focused on the 12 aluminum, to answer questions for the GIRP panel 13 members. That's going to be conducted on September 14 10th, the week of, to specifically look at four 15 cabinets: the medium-voltage, aluminum. We're also 16 going to discuss that in a little bit.
17 You can see that this milestone here, 18 though, goes out through the spring, in that we also 19 are going to be doing testing in the spring to try to 20 get a full picture, maybe look at low-voltage bus 21 stocks to get a larger picture of the zone of 22 influence and some of the insights from additional 23 aluminum results.
24 The University of Maryland study that you 25 see on there is we are in the process of contracting NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
117 1 with Jose Torero, who may or may not be on the line 2 right now, to take a look at the aluminum influence 3 and try and do some modeling work, applying the 4 results of the small-scale test program to models that 5 are currently available or the creation of new models.
6 So, we're looking into that.
7 In addition to that, you see the NRC/EPRI 8 Zone of Influence Working Group. The intention behind 9 that is to take the insights from the large-scale test 10 program and develop -- I'm going to use this word 11 loosely -- an interim zone of influence or a zone of 12 influence that we can, then, apply to the pilot plants 13 that we have discussed to get a better understanding 14 of the risk at those plants, using a zone of influence 15 that's more associated with the insights that we see 16 from aluminum.
17 We want to do that with industry, so we 18 don't use conservative assumptions in our modeling, 19 and we don't want to take full room damage in all 20 cases, and things like that. So, we want to work with 21 EPRI to do that and leverage some of the current 22 plants' PRAs to more adequately understand the risk 23 associated with potential increase on events from 24 aluminum.
25 MEMBER SKILLMAN: Nick, what's the current NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
118 1 thinking in terms of what those dimensions will be?
2 MR. MELLY: We've done some work in the 3 area looking into theoretical, such as what's a 4 theoretical energy increase from aluminum versus 5 copper. And if we were going to go very coarse, we 6 could say that the energy release could be anywhere 7 four times the zone of influence, so a factor of four 8 larger.
9 However, we think that may be too 10 conservative, which is why we're holding off on the 11 creation of that until we actually get more test data.
12 We've actually invited EPRI to witness the testing, 13 and potentially members who will be on that Working 14 Group, to witness testing and share the data, so that 15 we can get a better picture of what the actual zone of 16 influence will be, increase will be, taking into 17 account not only the increased damage potential 18 initially, but also the conductive byproduct. We want 19 to tackle that as well through this Working Group.
20 So, I don't want to speculate on what it 21 could be, but I would say it could be within the 22 bounds of a factor of four on the larger scale 23 potentially, and the conductive byproduct is still 24 such an unknown that we need more testing to make an 25 assessment of what that could do.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
119 1 MEMBER BLEY: That test you showed us on 2 the Zion duct --
3 MR. MELLY: Yes.
4 MEMBER BLEY: -- that filled the whole 5 breadth of that cubicle.
6 MR. MELLY: Yes, it did. It damaged and 7 singed cables that were 25 feet away on the adjacent 8 wall. We also melted our cabinet -- or a camera that 9 was 20 feet away.
10 MEMBER BLEY: Yes.
11 MR. MELLY: However, the zone of --
12 MEMBER BLEY: That's kind of pushing your 13 factor of four.
14 MR. MELLY: Yes. Yes. Well, yes, I am, 15 but that was a bus duct test. There is separate 16 guidance for bus ducts that's in NUREG-6850, 17 Supplement 1. It is FAQ, I believe 35. It covers the 18 zone of influence for bus ducts. And it is more of a 19 conical zone of influence in a downward direction that 20 does have an increased length of damage beyond the 3 21 foot/5 foot.
22 MEMBER BLEY: I guess we were a little 23 tricked by the duct being a few inches off the floor 24 --
25 MR. MELLY: Yes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
120 1 MEMBER BLEY: -- because that spread it 2 out very --
3 MR. MELLY: Exactly. And we are going to 4 be addressing that in the next phase of testing.
5 MEMBER BLEY: Okay.
6 MR. MELLY: And that kind of brings us to 7 the pilot plants that we're going to be trying to work 8 with industry, leverage their PRAs, working with EPRI 9 to get a better understanding of what the risk 10 profiles will be from an altered zone of influence for 11 aluminum.
12 MEMBER SKILLMAN: So, let's pull the 13 thread a little bit further. So, it was 3 feet, 3 14 feet, 5 feet --
15 MR. MELLY: Uh-hum.
16 MEMBER SKILLMAN: -- at the very top. Now 17 it could be 12 feet, 12 feet, 20 feet. Who's got that 18 much real estate --
19 MR. MELLY: Not many plants.
20 MEMBER SKILLMAN: -- in an old plant?
21 MEMBER BLEY: But you have to look at what 22 happens if all that stuff gets wiped out. You don't 23 have to protect it all, I assume.
24 MR. MELLY: Right.
25 MEMBER BLEY: Unless there's a need to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
121 1 protect it all.
2 MR. MILLER: Make sure our fire barriers 3 aren't destroyed, in fact, and other trains.
4 MR. MELLY: Additionally, We have to take 5 into consideration that, even for plants that 6 potentially have tight configurations -- so, if they 7 have cable transit a foot off the top of the cabinet, 8 and they have a stack of three of them, if their 9 current model, the 3 foot/5 foot, already damages 10 those cables, and their model assumes damage of those, 11 the delta risk is going to be zero because they have 12 already damaged those cables.
13 So, there's a lot of things to take into 14 consideration when we have the pilot plant selection 15 as well as how we're going to address any potential 16 increase in risk. It's going to be a challenging 17 project.
18 MEMBER SKILLMAN: So, one specific 19 strategy might be the definition of the divisions and 20 the separations of the divisions. So, one could 21 conceivably say, even if the zone of influence is 22 increased by a major factor, we have already accounted 23 for a loss of the entire division. That's why we 24 have, at least in the new plants, four divisions or 25 two half, a pair, two pair of half divisions. So, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
122 1 it's already accounted for, even though it's kind of 2 an awkward way to do it.
3 MR. MELLY: Yes, and that's what I'm 4 hoping will come out of the benefit of leveraging our 5 current PRAs, is that their CCDPs, based on the 6 targets that they're damaging, will reflect just that.
7 MEMBER SKILLMAN: Okay. Thank you. Okay.
8 MR. GARDOCKI: If we can go back to the 9 slides, I'll just finish up.
10 So, right now, once we finish our 11 assessment plan that you just saw, we'll be at that 12 bar going from assessment into Regulatory Office 13 Implementation. And at that point, it transitions 14 basically from the Research Office into the 15 Implementation Office. It will be like NRR. And what 16 we do is we form a transition team to smoothly 17 transition the issue all the way over into NRR, make 18 sure nothing is dropped. And we will be issuing a 19 report like for ACRS there at the end when the 20 transition occurs.
21 So, this is not, the assessment is not the 22 end of the generic issue. It's transitioned to the 23 next phase, where NRR has a ball, like we have 24 currently GI-191 and the other ones from seismic and 25 flooding PRAs going on. So, we track it in a Generic NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
123 1 Issue Program until it's actually finished and closed 2 out, and all the plants have taken the actions.
3 Any other questions on the Generic Issue 4 Program?
5 MEMBER BLEY: I just have one. I have 6 one. That second occurrence of the ACRS on your chart 7 --
8 MR. GARDOCKI: Okay.
9 MEMBER BLEY: -- right as you finish the 10 assessment, we get a look whether it's approved to go 11 forward or the decision is leaning towards stopping?
12 MR. GARDOCKI: Correct.
13 MEMBER BLEY: That's all I wanted to hear.
14 Thank you.
15 MEMBER SKILLMAN: In your quiet time, what 16 is your thought about how long this might take to 17 conclude an appropriate zone of influence and, then, 18 have industry implement? It would seem to me we're 19 talking years. This is not an activity that is going 20 to be resolved with any speed.
21 MR. MELLY: In my quiet time -- (laughter) 22 -- I would agree. However, I think that the way the 23 Generic Issues Program is set up, I believe that there 24 are avenues to speed things up potentially. If, 25 through the pilot studies, we do see that there is a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
124 1 need and a very large risk increase, or potential 2 safety issue that comes out of this pilot program 3 study, there are avenues to have generic 4 communications with the public as part of this Generic 5 Issues Program, which could go anywhere from 6 information notice to orders. So, there are avenues 7 to do it, but, right now, yes, the research and the 8 testing is fairly slow because these are complicated 9 tests.
10 MEMBER BALLINGER: From the standpoint of 11 just energy density for the same amount of fuel, this 12 zone of influence should be the square root of 10 13 larger for aluminum versus copper, because the free 14 energy of formation of aluminum oxide versus copper 15 oxide, it's a little over a factor of 10. And so, 16 it's just an R-squared thing, if you want to do that.
17 The zone of influence is the first order of the square 18 root of 10.
19 MR. MELLY: Yes, there is that issue to 20 contend with, and it's something we're definitely 21 looking into. Additionally, it's very much an 22 orientation question.
23 MEMBER BALLINGER: Yes, I mean, that's why 24 I said, for a fixed amount of fuel --
25 MR. MELLY: Right.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
125 1 MEMBER BALLINGER: -- the same amount of 2 fuel in a sphere, the square root of 10.
3 MR. MELLY: Yes, we're really playing the 4 line right now of making sure that we have a basis for 5 the assumptions that we are making going into creating 6 guidance for the increased zone of influence rather 7 than being completely reactionary and going the other 8 way with an overly-conservative result. We've had 9 many meetings with this with NRR through the 10 assessment process, trying to establish at what level 11 of information are we comfortable making our 12 assessment and making these recommendations.
13 There's also been several steps and stages 14 that we've gone through, such as the NRR immediate 15 safety concern evaluation. That was the entrance into 16 the screening program, which tried to take an initial 17 look at risk to make sure we're taking appropriate 18 actions we need to in a timely manner.
19 MEMBER SKILLMAN: So, the questionnaire 20 begins with, do you have aluminum, yes or no?
21 MR. MELLY: Right.
22 MEMBER SKILLMAN: And the second question 23 is, if yes, where? And then --
24 MR. MILLER: And how much?
25 MEMBER SKILLMAN: And how much? And that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
126 1 starts the engine. Is that pretty much the direction?
2 MR. MELLY: Exactly. And we have taken 3 steps informally right now to answer that question for 4 the screening process. One of the entrances into the 5 screening process was, do multiple plants have this 6 issue? NEI issued an informal survey to the plants to 7 answer just that question.
8 We drafted up a questionnaire for the 9 plants asking where they had aluminum, and it was a 10 subset, many questions to the plants. And we got a 11 response from NEI, which is included and referenced in 12 the screening report, that there is a large amount of 13 aluminum out there in the U.S. fleet. A lot of it is 14 in bus duct housing due to weight and concerns like 15 that. So, we do see that there is aluminum out there, 16 and the U.S. may a country that's used the most of it, 17 as we see from our international OECD/NEA 18 collaboration.
19 MEMBER SKILLMAN: But wouldn't that also 20 raise the question, if there is so much, why haven't 21 we seen major incidences that this testing points to?
22 MR. MELLY: And that is a segue that we'll 23 get to in, I believe it's actually the next slide.
24 MEMBER SKILLMAN: Okay. Okay, good.
25 Let's keep going. All right.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
127 1 MR. SALLEY: But, before you jump into 2 that, though, Nick, one other thing to answer for 3 clarity, you've got multiple programs going here, 4 remember. You know, we've got the International HEAF 5 Program that we're working and, also, the aluminum 6 piece. And as you said, the aluminum clearly showed 7 the United States. So, a lot of the countries really 8 aren't participating with that.
9 We're trying to move that forward and get 10 the aluminum piece done first. So, we're giving that 11 priority, again, to work it through the Generic Issues 12 Program.
13 There will be follow-on because everybody 14 wants a better model to do HEAF, even if you were 15 doing copper. So, again, that will be follow-on work 16 that will be with the International. So, we're trying 17 to give priority to the aluminum. We're trying to 18 move things forward.
19 MR. MELLY: And to the question that you 20 just asked, why haven't we seen this, it's that 21 question that these events are relatively rare, and 22 they don't occur at the same plant. And if you don't 23 really connect the dots, you don't see the influence.
24 So, as one of the exercises after the test 25 program where we did see the aluminum have a direct NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
128 1 influence, one of our activities was to go back and 2 look at these high energy arcing faults as well as 3 look at the operating experience database from the 4 frequency, for the creation of frequency in the 2169, 5 and we did word searches such as "aluminum high energy 6 arc fault," things like that.
7 And as some of the events that we have 8 seen, I found six events that did have aluminum as a 9 part of the arcing event itself. And in those six 10 events, there was always a predominantly larger zone 11 of influence attributed in the event itself, as well 12 as some potential consequences of this conductive 13 material that we saw at Fort Calhoun. It is a limited 14 subset of events due to some events not having enough 15 details from past history to figure out if aluminum 16 was actually present or not.
17 That's typically not something that would 18 be recorded in an LER. But, from the events that we 19 could look at and dig into, we did find six events 20 that did show this trend towards a higher energy 21 situation, which is why we issued the Information 22 Notice and have moved it forward into the Generic 23 Issue Program.
24 MR. HAMBURGER: In February of 2017, we 25 held an International PIRT with all of the members who NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
129 1 had participated in the phase I testing. And this 2 report I think answers the question that we had at the 3 time, which was, okay, now what? We have data from 4 phase I. We had the videos. Informally, we had drawn 5 our conclusions that aluminum plays a large role in 6 this. We needed to formalize those conclusions, and 7 we wanted the input of the participants in phase I to 8 record those conclusions and build a roadmap for where 9 we were headed next.
10 So, we held this Phenomena Identification 11 and Ranking Table Exercise. We presented several 12 scenarios and asked the participants to rank various 13 phenomena by their impact on the damage dates. This 14 document was selected for ACRS quality review. So, I 15 think you guys are probably working your way through 16 this as we speak.
17 MEMBER BLEY: Yes, three of us are.
18 MR. HAMBURGER: Yes.
19 MEMBER BLEY: The whole Committee will see 20 it in October, yes.
21 MR. HAMBURGER: I don't want to ruin any 22 surprises for you.
23 (Laughter.)
24 MEMBER BLEY: So, we won't talk about that 25 here.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
130 1 MR. HAMBURGER: But some of the things 2 that rose to the top of the list were the aluminum 3 oxidation and byproducts. This is not surprising.
4 After we had connected those dots from tests 23 and 5 26, this was the question on everybody's mind.
6 The international members particularly 7 were interested in pressure effects from these high 8 energy arcing faults. One of the interesting topics 9 that rose to the top was the characterization of 10 targets. In our current PRA treatment of these 11 events, all targets are the same. We have a zone of 12 influence. It doesn't matter what kind of target you 13 are, if you're in that zone of influence, you're 14 damaged; if you're outside the zone of influence, 15 you're safe.
16 One of the things that came out of this 17 PIRT was that's not necessarily true; it depends on 18 target fragility. An unprotected cable tray may 19 respond very differently than a metal-enclosed 20 switchgear. So, even if you could characterize that 21 arc with 100-percent accuracy, characterizing those 22 targets might also be important in predicting damage 23 dates.
24 Another thing that came up was those 25 mitigating factors that we talked about like the HEAF NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
131 1 shields. They have been implemented in some plants 2 already, but we don't know what they are, whether 3 they're effective, if there's any qualifications for 4 those shields. So, investigating the efficacy of 5 those HEAF shields also came to the top of our list of 6 potential things to investigate.
7 Next slide.
8 This is a NUREG-IA that we have published 9 jointly with JNRA as they worked their end of the HEAF 10 program. The driving force behind Volume 1, which has 11 been published, is the Onagawa HEAF event. So, their 12 goal was to understand that event, to clarify what 13 exactly happened in that event. And they used a bank 14 of cabinets and ignited some HEAFs in those cabinets.
15 This is also the origin for the soundbite 16 that you'll hear about the rocket fuel. They used 17 rocket fuel to simulate the energy release of an arc.
18 And they did that because they needed a larger 19 calorimetry hood that KEMA could not provide, and they 20 didn't have the energy source where that hood was 21 located. It was purely a logistical PIRT. So, they 22 used a surrogate energy source, which is that rocket 23 fuel.
24 The conclusion is that it is possible to 25 simulate the arc with the rocket fuel. They didn't NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
132 1 see the same damage states that they saw when they 2 used the electricity source at KEMA Laboratories. But 3 we've heard some soundbites from industry at 4 Commission meetings that we don't have rocket fuel in 5 our cabinets. This was a Japanese test program. It's 6 not something that we're adopting.
7 But we have had a lot of benefit from the 8 hindsight and experience that we've gained from 9 working with the Japanese, as you'll see when we talk 10 about our instrumentation and measurement science.
11 So, working with them has been a big benefit to us.
12 MR. MELLY: Additionally, we will provide 13 you with the information from that Onagawa fire.
14 There's a large event report that we can send to you 15 as well.
16 I also forgot to mention one of the 17 events; Nathan reminded me. There is a very 18 interesting event from Maanshan which was a high 19 energy arcing fault associated with a station 20 blackout. They are not part of the OECD fire group, 21 which is why it's not included on that previous list, 22 but it is a very informative and interesting read as 23 to how a high energy arcing fault can lead to a 24 station blackout and the consequences associated with 25 that. So, we will also send you that information when NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
133 1 we send you the Onagawa fire event.
2 MR. HAMBURGER: So, in addition to working 3 with the Japanese, we also went back and looked at our 4 test program, looked at what data we collected, what 5 data was collected successfully, and moving forward, 6 what we would need to do to collect the data we need 7 to answer the questions that the GI poses and to 8 better characterize these events. So, I'm going to 9 touch on various types of measurements, the 10 temperature heat flux incident energy, what we're 11 doing with the oxygen consumption calorimetry hood, 12 pressure video, and our data acquisition system.
13 So, this is one of our more important 14 measurements. This is really what's going to inform 15 how we decide what is damaged at a distance from the 16 arc.
17 In phase I testing, we used two types of 18 temperature and heat flux sensors. We used the ASTM 19 F1959 Slug Calorimeter. This is the slug calorimeter 20 that is prescribed in measuring arc flashes, again, 21 mostly for personnel safety purposes.
22 KEMA Laboratories provides these slug 23 calorimeters for us, and they take care of the data 24 acquisition. So, we don't have to do that ourselves.
25 They provide that as part of the test report NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
134 1 deliverables.
2 We looked into using water-cooled 3 Schmidt-Boelter gauges for heat flux. This was 4 difficult logistically because they require running 5 water. And there was a couple of reasons we didn't 6 want to do that. One is we need to provide running 7 water, but the other is it may interfere with other 8 measurements if that equipment is damaged during 9 testing. We didn't want water running in the cell.
10 So, for logical reasons, we decided not to use the 11 water-cooled heat flux gauges.
12 Then, we have our plate thermometers.
13 These are very low uncertainty measurement devices.
14 They're thin. They have a known surface area and 15 emissivity. And so, from the temperature that's read 16 on the back-end of the plate, we can determine what 17 the heat flux is at the face of the thermocouple, at 18 the face of the plate.
19 Unfortunately, these are limited by range.
20 As you saw in phase I, we melted our plate thermometer 21 at 3 feet for test 23. You can see the beads of 22 Inconel on what used to be the backing. So, Inconel 23 melts at 2400 degrees Fahrenheit, which is a 24 significant amount of heat. But we lost our 25 measurements, and we were unable to calculate heat NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
135 1 flux for two of our tests.
2 So, moving forward, we have actually 3 worked with NIST to develop a new method of measuring 4 heat flux at these high ranges. And this is another 5 slug calorimeter. The slug is tungsten, and we are 6 expecting that this will endure and survive some of 7 the higher heat fluxes that we're anticipating seeing 8 when we do these upcoming tests.
9 The operating theory here is, basically, 10 we have a tungsten slug of known mass properties, 11 emissivity, and we're measuring the temperature at the 12 back-end of the thermocouple. It's slower to respond 13 than the plate thermocouples because of its thermal 14 inertia. It's much thicker; it's much heavier. But 15 what we're trying to get out of this is an average 16 heat flux over the course of the arc. So, it doesn't 17 have such fine temporal resolution, but it will 18 withstand the higher heat flux levels, and we'll get 19 an average heat flux over the course of the arc. What 20 we're really hoping for -- and you'll see this 21 reflected in the design of our instrumentation racks 22 -- is a direct hit on one of these tungsten slugs.
23 And that will give us the best indication of what our 24 maximum credible heat flux is.
25 In phase I, we used that oxygen NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
136 1 consumption calorimetry hood. We have decided to do 2 away with that for phase II, for a variety of reasons.
3 One, as we talked about earlier, the initial event, 4 which is really what we're trying to characterize, 5 overwhelms that hood. It's not big enough, and the 6 KEMA facility can't support a larger hood. We had 7 trouble getting this one in position.
8 It's extremely costly in terms of setup, 9 time to set it up, effort to set it up. It took us 10 two full days to construct that scaffold, to put all 11 that stuff in the air. It takes a lot of time to do 12 the calibration burns, to transport up the bottles of 13 compressed gases that we need to do the analysis.
14 It's a very time- and labor-intensive measurement.
15 And we didn't feel that we were getting 16 out of it what we needed. The HEAF was overwhelming 17 the hood, and we're not as concerned with the ensuing 18 fire. We really want to characterize that initial 19 arcing fault. So, we've decided not to use the oxygen 20 consumption calorimetry hood moving forward.
21 As Gabe mentioned earlier, during the 22 Sandia scoping tests, we have looked at a number of 23 ways to measure the pressure from this event. The 24 bikini gauges were pretty much shredded by the molten 25 ejecta. The pencil probes proved unreliable.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
137 1 What we did in phase I was we created a 2 hole in the cabinet. We bolted a strain gauge-type 3 pressure transducer to the cabinet itself. So, what 4 we're measuring is that initial overpressure inside 5 the cabinet.
6 We didn't get great data readings from 7 this, and we're attributing that largely to 8 electromagnetic interference. And the reason we think 9 that is, if you look at our data, we have these large 10 positive and negative pressure peaks at just the time 11 that you would expect the EMI to be the most severe, 12 when they close the breakers and when the HEAF event 13 ends. So, we're getting those, what we assume are 14 erroneous peaks. We don't expect pressure to be 15 negative. And they're occurring right when we would 16 expect the EMI to be the highest. The data is very 17 noisy. It looks unreliable.
18 So, we have been working on new techniques 19 to measure, moving forward into phase II. And this is 20 one of those benefits from working with Japan.
21 They've had a lot of success moving towards a 22 piezoelectric pressure transducer rather than the 23 strain gauge type. It's much less sensitive to EMI.
24 It doesn't use a whetstone bridge.
25 So, there, you can see we have some NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
138 1 preliminary data from their unpublished reports.
2 We're getting much smoother, much more 3 reliable-looking pressure measurements from the 4 quartz-type piezoelectric gauge.
5 We've also replaced some of the twisted 6 pair wiring with fiber optic cable. So, we're 7 expecting that to be somewhat more impervious to the 8 EMI that we had experienced in phase I.
9 Dave?
10 MR. STROUP: In addition to regular video 11 and high-speed video, we also looked into the use of 12 infrared technology as a way to do non-intrusive 13 temperature measurements and to give us an opportunity 14 to look through the smoke and perhaps follow the 15 ensuing fires that occurred after the HEAF.
16 I've been involved in infrared camera 17 technologies since the days when they had to be filled 18 with liquid nitrogen. We purchased a 19 commercially-available camera and tried to balance the 20 requirements of you've got a high-speed, 21 rapidly-occurring event like a HEAF, so you need the 22 camera to respond very quickly and to be able to 23 record that data quickly. You've got a large 24 temperature range, in excess of 2,000, 30,000 degrees.
25 And we also needed a large field of view.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
139 1 Sometimes you'll see infrared cameras used as 2 diagnostic tools where they'll be looking at things 3 like an overheated relay or an overheated switch. And 4 it is a very small field of view. Other cameras that 5 are not as quantitative may look at a larger field of 6 view, like you may have seen TV videos where the 7 police are looking for a suspect, or something, and 8 they're using an infrared camera from a helicopter 9 looking at a large area. So, we were looking to 10 balance these competing needs.
11 Next slide.
12 So, our initial camera that we used for 13 phase I had the high-speed recording capabilities, it 14 had high resolution, but in the end it had a limited 15 temperature range. While the camera was capable of 16 recording temperatures from 20 degrees C up to in 17 excess of 2,000 degrees C, the way it did that was it 18 had a filter wheel. It had a filter that it placed in 19 front of the camera lens, and that filter only moved 20 once. So, you had to decide, am I going to focus on 21 the higher temperature range or the lower temperature 22 range, at the beginning of the test, and you were sort 23 of stuck with that throughout the test.
24 The improved camera that we're looking at 25 getting now has a little slower speed of recording.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
140 1 It still maintains a high resolution, but gives us a 2 greater temperature range, in that the filter wheel is 3 continuously revolving and it's tied to the camera's 4 internal circuitry, so that it knows what filter wheel 5 is in place and can correlate that with the 6 temperature range that it's looking at at the time.
7 Here's a comparison video from one of our 8 first phase tests. The right side image is the 9 regular video compared to the infrared image there.
10 And you can see, even with a maximum range of 2,000 11 degrees, we're still oversaturating the camera during 12 part of the HEAF.
13 Now one of the things I'm currently 14 involved in, working with the National Institute of 15 Standards and Technology, is we've done some non-HEAF 16 cabinet tests and looked at the temperatures that were 17 developed during fires in those cabinets and recorded 18 infrared data. And we know what the temperatures 19 were. So, I'm trying to correlate the temperatures 20 with the camera's indicating versus the data that 21 we've recorded, to get a better measure of the 22 accuracy of these cameras.
23 MEMBER MARCH-LEUBA: Yes, but when you get 24 to those temperatures, you should go to this roll, 25 into the one on the right. When you're above 2,000 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
141 1 degrees C, you're not at infrared anymore.
2 MR. STROUP: Right.
3 MEMBER MARCH-LEUBA: That's what the 4 surface of the sun is. I mean, the one on the right 5 tells you this information.
6 MR. STROUP: Right. Yes. Yes, in some 7 cases you'll miss, like the regular video will miss a 8 lot of the projectiles that are being thrown off.
9 Whereas, the infrared camera will see a lot more of 10 that. So, it's really not an "either/or". There's a 11 lot of information that you can gain from all of 12 those. And both the regular and the high-speed 13 regular video also take it all together and try and 14 analyze it from that perspective.
15 MEMBER MARCH-LEUBA: I take it back. The 16 surface of the sun is 6,000.
17 MR. STROUP: Okay.
18 MEMBER MARCH-LEUBA: 57K.
19 MR. HAMBURGER: We've also made some 20 upgrades to our data acquisition system. We've 21 increased the number of channels. We intend to 22 capture more data the next go-round. We've also 23 isolated the deck and we've made some redundant 24 systems available.
25 One of the things we observed in one of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
142 1 the Japanese tests was they shorted to some of their 2 instrumentation, and they actually lost all the data 3 from that test because it shorted out their data 4 acquisition system. So, we've been working on making 5 sure that this maintains its availability throughout 6 our tests.
7 MR. MELLY: Now, if Gabe is still on the 8 line and able to speak, we're going to go over the 9 Small-Scale Research Program.
10 Gabe Taylor, do we have you still?
11 MR. TAYLOR: Yes, I'm still online. Can 12 you hear me?
13 MR. MELLY: Yes. Yes, we can.
14 MR. TAYLOR: Okay, great.
15 So, I'm going to talk a little bit about 16 the small-scale research that we've been doing; 17 actually, just completed last month out at Sandia 18 National Laboratories.
19 And from the first phase of testing, one 20 of the major outcomes from that was aluminum 21 contributing to the additional energy that we're 22 seeing from these events. And the question that comes 23 to mind, okay, why is that? Can we model it? What 24 existing literature or methods out there are available 25 to try to characterize this?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
143 1 We started asking people who are very 2 knowledgeable in the combustion science area, one of 3 those being Jose Torero over at the University of 4 Maryland. And we provided him a number of information 5 and videos from the first phase. When we met with 6 him, the thing that came to his mind is that a lot of 7 people have been saying it's the oxidation reaction 8 that is causing the additional energy. And what he 9 said was somewhat, well, quite a bit different.
10 He was looking at that atmospheric oxygen 11 isn't very efficient at oxidizing aluminum to create 12 that amount of energy in that short of time. So, the 13 time scale was just too short for that amount of 14 energy being completely from the oxidation induction 15 reaction.
16 So, what he looked into a little bit was 17 the electrical nature of the event and that 18 contributing to the additional energy, as well as the 19 aluminum material causing the additional energy. And 20 he found a number of literature sources and models 21 that may be able to be tweaked to help us estimate the 22 amount of additional energy that's coming from this 23 aluminum.
24 But, to help develop his model or a 25 modified model, there was a number of information NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
144 1 about the particle sizes and characteristics that he 2 needed to support that development. So, we looked at 3 possibly doing it at Maryland, but we came across 4 Sandia having this lightning simulator at their 5 facilities that would provide the needed information, 6 as well as a lot of very expensive instruments that 7 they could use to characterize the type of information 8 that Jose would need.
9 So, we ended up going to Sandia and doing 10 a focused-scope small-scale testing program to try to 11 develop that information. So, if we're still on slide 12 60, trying to characterize the particles, how fast 13 they are produced for the production rate, size 14 distribution, density, trajectory, and whatnot, as 15 well as the different types of emissions from varying 16 the electrical input. So, voltage levels, current 17 levels, durations, how those parameters, electrical 18 parameters, affect the particle characteristics.
19 So, if we move to the next slide, 61, on 20 small-scale approach, we used the Sandia lightning 21 simulator. When you go to any type of testing, you're 22 going to have some limitations, pros and cons to doing 23 that. One of the things that we had to do in limiting 24 our approach was, instead of testing all three phases, 25 we basically just did a single phase of two poles of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
145 1 the system. So, that is one limitation and difference 2 from what you would likely be seeing in real-world 3 applications.
4 And then, we did have variables such as 5 voltage. We tested at low voltage and a medium 6 voltage. We varied current, durations. And although 7 our focus was on aluminum, we included a number of 8 copper tests as well. So, we had copper buses in some 9 tests and aluminum in other tests.
10 And then, we used a couple of different 11 techniques to collect the particles. We used 12 aerogels, which are a high-purity silicon material 13 that can collect the materials, and then, after the 14 test, we can analyze them. We also used black carbon 15 tape, which basically looked like duct tape, but a 16 high-purity carbon associated with that tape, to 17 collect the particles. And then, also, we used 18 high-speed photography with different types of 19 neutral-density filters on the photography, or 20 videography, to be able to see the particles and the 21 trajectories, and estimate the speed. And we'll have 22 a video of it, hopefully that plays later on, with 23 that.
24 So, if we move to slide 62, this is just 25 providing a general view of the experimental setup.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
146 1 So, in the photo on the left, we had two vertical bus 2 bars separated by approximately a centimeter between 3 the faces of the bus bars. Near the top of the bus 4 bar, they drill a very small hole on both buses and 5 put a small wire, almost like a filament, between the 6 two buses. So that, when they energize the buses, 7 that's where the arc initiates. And it's a little bit 8 hard to see in that photo, but, behind there, there is 9 a aerogel, so a small silicon coupon, and a black 10 carbon tape in that.
11 And the illustration on the left is trying 12 to show the same concept, but that was an earlier 13 design, and it changed somewhat differently. But one 14 thing that that does show is the video alignment. So, 15 we had three different axises orthogonal to each other 16 to capture particle movement at a high speed of 17 recording. Anywhere from 100,000 frames per second up 18 to 1 million frames per second was used in this 19 program.
20 So, we need slide 63. This is just giving 21 you a broad overview of the summary of this; 22 basically, what I talked about already. Videography, 23 high speed, to get the trajectory. Particle 24 collection using the two types of methods that I 25 discussed previously. And then, the particle NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
147 1 analysis, this is getting into the different types of 2 technologies that they have available to analyze the 3 particles that are collected post-test.
4 Some things that are of interest in this 5 analysis is that they will be able to differentiate 6 the amount of oxidation that actually does occur 7 between the particles. They'll also be able to --
8 things that we need -- identify the dimensions, the 9 morphology. Whether it's vaporized or melted 10 materials, that's collected, and that sort of 11 information, to support Jose as he refines and 12 develops the model that he's working on.
13 So, if there's no questions on that, we 14 can move to slide 64, which is a video.
15 MR. MELLY: It is working.
16 MR. TAYLOR: Okay. So, I don't have it in 17 front of me, but what this video is trying to show you 18 is some of the tests, well, all the tests they did, 19 but this is just an example. They took the data that 20 they collected to their videos. They ran into an 21 algorithm in one of their computers, high-performance 22 computers, and they are able to track the different 23 particles as they move away from the arc initiation 24 location. And they're basically able to measure, 25 through that analysis, the peak speed of, the maximum NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
148 1 speed of the particles as well as the average speed of 2 the particles.
3 MEMBER BLEY: What size particles are we 4 looking at?
5 MR. TAYLOR: I actually don't have that 6 information in front of me. I could get it for you 7 or, if Chris LaFleur from Sandia is on the line, she 8 possibly could provide that.
9 MS. LaFLEUR: I am on the line.
10 It's micrometers. It's small.
11 MEMBER BLEY: Thank you.
12 MS. LaFLEUR: Yes.
13 MR. MELLY: And what you're seeing here on 14 the video is that actual trajectory and the analysis.
15 MR. TAYLOR: Yes. So, it's one piece of 16 information that's going to help Jose. The energy is 17 being released close to the arc. We don't see energy, 18 at least from the videos, being released. As you get 19 farther away, it should dissipate. So, the trajectory 20 was one important parameter that he was interested in.
21 And then, if we can move to slide 65?
22 Just for everybody's awareness, we did put a draft 23 test plan out for public comment in The Federal 24 Register notice. The information is provided there.
25 And we did resolve all comments, and we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
149 1 actually did make changes to the test plan as we moved 2 forward. The April workshop helped us as well because 3 we had a lot of discussion there to make changes 4 before we went ahead and did the program. So, all 5 that information on the comment resolutions is also 6 publicly available.
7 MR. MELLY: Any questions on the 8 small-scale test program?
9 (No response.)
10 Okay. Moving forward.
11 MR. HAMBURGER: In April of this year, we 12 held a public workshop for anyone and everyone who was 13 interested in forming the small- and large-scale test 14 plans that we were planning.
15 We had received a number of comments and 16 criticisms from industry that the tests we had done in 17 phase I were not representative of their 18 configurations in operating facilities. Another 19 subset of those comments reflected a clear 20 misunderstanding of what we were trying to accomplish 21 through our testing. So, we thought that the best way 22 to resolve this, and to go over the comments we were 23 receiving on the draft test plans, was to have an 24 in-person workshop and resolve the comments from the 25 draft test plans, which we had published via The NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
150 1 Federal Register; get their input on what types of 2 equipment and how they are configured in the actual 3 operating facilities, so that our test plan is as 4 realistic as possible, and just update them on where 5 we were in the GI process and the research program.
6 MR. TAYLOR: Okay. So, I think this is 7 mine, "Needs and Objectives".
8 MR. MELLY: Yes.
9 MR. TAYLOR: Okay. One of the things 10 before we went into the workshop, we didn't want to 11 just have everybody show up, and then, just talk about 12 things. We wanted to try to get people prepared and 13 understand what we were interested in. And to do 14 that, we actually put out a document ahead of time, 15 ahead of the workshop, that kind of gave it a very 16 high generic-level what the needs of the program were, 17 what the objectives and goals of the second phase of 18 the HEAF testing was all about.
19 And the second part of that document was 20 getting into different types of electrical 21 configurations inside the plant and trying to identify 22 those configurations and basically request, without 23 putting out a survey or a formal request, that people 24 attending this meeting could bring information on what 25 fault currents are on these buses, the types of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
151 1 clearing times and durations that they would see if 2 they did experience some type of fault in that 3 location.
4 So, from what I saw in the workshop, I 5 think not everybody looked at it, but a number of 6 people did look at it, and it definitely helped 7 support the discussions that were held during that 8 April workshop.
9 MR. HAMBURGER: So, ultimately, the 10 workshop was a success. I think we achieved some 11 level of consensus with EPRI and NEI about what it is 12 we're trying to do and how we're approaching this 13 problem. We also got some good feedback on what types 14 of equipment is typical in an operating facility, and 15 we've incorporated that into our equipment selection 16 for phase II.
17 And all of the presentations and 18 discussions that were held, including the transcripts, 19 at this workshop are being published in a NUREG CP.
20 This is currently in publication. So, it will be out 21 any year now.
22 MR. MELLY: Okay. So now, we're going to 23 go into the phase II parameters and testing, things 24 that we've seen that are going to be important that 25 we're going to be focused on for the phase II.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
152 1 One of the important parameters, one of 2 the most important is the actual arcing time and 3 duration. And it's one of the criticisms that we've 4 gotten from industry that some of our arc times are 5 too long or their breakers will clear prior to the 6 length of duration that we're testing at KEMA.
7 And our response to this is we're trying 8 to back up the duration that we're going to be testing 9 based on operating experience. We've gone back and 10 looked at these events and how long they've actually 11 lasted in real plant applications to try to influence 12 our test program.
13 So, for this upcoming test series, we're 14 going to be testing at 2 to 4 seconds for 15 medium-voltage cases, and we're going to be testing at 16 2 seconds, 4 seconds, and 8 seconds for low-voltage 17 cases.
18 Out of discussion at the workshop, it came 19 to our attention from operating experience as well as 20 the way that the breaker coordination/circuit 21 protection is set up that, for low-voltage cases, 22 typically, we don't see these long-duration, 8-second 23 arcs. However, at the KEMA facility, we cannot 24 produce an 8-second arc in a medium-voltage power 25 supply because of the size of their generator and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
153 1 limitations of their facility itself.
2 So, the intention for us to do an 8-second 3 arc for a low-voltage piece of equipment is so that we 4 can, hopefully, make some sort of comparison between 5 the low-voltage results at extended duration and more 6 medium-voltage data.
7 MEMBER BLEY: Can I interrupt you a little 8 there?
9 MR. MELLY: Yes.
10 MEMBER BLEY: I'm glad you're doing the 11 longer ones. But I come in, and I tell you my breaker 12 coordination/selective tripping is going to keep me 13 out of this. But, if you're seeing a substantial 14 fraction of the real ones out there having breaker 15 problems that don't open as expected, that seems to 16 mean you need to look longer, because it's happening, 17 no matter what people say they've got for breaker 18 coordination.
19 MR. MELLY: Yes, that is something we're 20 trying to take into account when we're going to look 21 at the overall methodology. Because if breaker 22 coordination works as designs, and these events are 23 very short, we're not going to see high energy arcing 24 faults at all create damaging events. So, it's 25 something that we want to investigate, which is why NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
154 1 we're getting the range of data from 2 to 8 seconds.
2 And we think that is important to understand these 3 extended-duration events.
4 MR. HAMBURGER: Successful breaker 5 operation is accounted for in the frequencies that 6 we've assigned to bin 16. Because bin 16 are 7 typically those events where breaker protection 8 doesn't work and you do have an extended duration.
9 MEMBER BLEY: I was worrying that we were 10 letting claims of perfect breaker coordination keep us 11 from getting the test results we ought to be getting.
12 MR. MELLY: No, we're trying to avoid that 13 from occurring.
14 MEMBER MARCH-LEUBA: In real life what's 15 the percentage of the energy release that is 16 electrical? What I'm thinking is I don't care how 17 long the arc is. Once you ignite the aluminum, is 18 that 90 percent of the energy? Or is the energy 19 coming now from the electrical arc?
20 MR. MELLY: That's something that we're 21 going to be investigating, right, currently,.
22 MEMBER MARCH-LEUBA: Because, clearly, 23 from the test 26 or 23, the electrical energy wasn't 24 that much, and the blowup was --
25 MR. MELLY: And one of the actual insights NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
155 1 that we gained towards that effect was, when the 2 Japanese and JNRA actually did their testing using 3 rocket fuel as the source of energy, when they used 4 the rocket fuel that was strictly associated with the 5 energy release from KEMA, reported from KEMA's 6 electrical energy supply, it was nowhere near enough 7 to see the damage that they saw at KEMA.
8 They actually had to increase the amount 9 of energy for the aluminum rocket fuel --
10 MEMBER MARCH-LEUBA: The high arc is 11 igniting more stuff.
12 MR. MELLY: Yes. So, that is something 13 we're taking a look at moving forward.
14 MEMBER MARCH-LEUBA: Yes, that does make 15 sense.
16 MR. MELLY: So, that is the duration 17 question, and that's one of the biggest hitters, the 18 most important factors that you'll see from the PIRT 19 as well as a consideration for our test program moving 20 forward.
21 We've covered a lot of the parameters that 22 we are going to be doing, 2 to 4 and 8, 2 to 4. And 23 for the bus ducts, we can go 1 second, 3 seconds, and 24 5 seconds maybe, depending on KEMA's capabilities.
25 Again, we're basing that on operating NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
156 1 experience, as this is a subset of some of the events 2 that we've seen: the Robinson event, 8 to 10 seconds; 3 Diablo Canyon event that lasted for 11 seconds; 4 Prairie Island, from the LER, we only know it lasted 5 -- that it was in excess of 2 seconds. Again, with 6 San Onofre, in excess of 2 seconds, and the Fort 7 Calhoun was a 42-second arc in low voltage, and it had 8 to be terminated by a manual operation. So, this was 9 an interesting event, low-voltage. Whether it held in 10 for 42 seconds as a direct fault or for the sputtering 11 fault is still unknown right now.
12 Again, we've discussed this. The intent 13 of the long durations is just to account for these 14 operating experience events. We're going to get more 15 information. We want to know what is an arc flash 16 versus the high energy arcing fault. It's really what 17 sets our test program apart from what's been done by 18 NFPA. They are typically locked in at 2 seconds, 19 which is a human response time estimation. We're 20 looking at a little bit longer durations.
21 Again, we've had discussions to look at 22 the voltage differences. We're going to be looking at 23 low voltage, 480 volt, and medium voltage, 6.9 kV.
24 This is in an effort to not only get a large range of 25 data, but also it satisfies a lot of the international NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
157 1 concerns. They run internationally at 6.9 rather than 2 4160, more than that. So, we're playing a balancing 3 game right now with making sure that we're covering 4 the needs of our international partners while still 5 understanding the risk.
6 The current, on this slide, it shows at 7 low voltage we're going to be operating at 15kA and 8 25kA, and medium voltage at 25kA and 35kA. The next 9 few slides, I will discuss where those values came 10 from, partially from the "Needs and Objectives" 11 documents as well as our discussion with the industry.
12 Now there is one -- you see the star 13 there, depending on the power delivery system 14 capabilities. This is based on KEMA's generator size.
15 So, since we have created this document for you guys, 16 we've had discussions with KEMA. We cannot achieve 17 35kA at their facility for 4 seconds, based on they 18 don't want to damage their facility. So, we are only 19 going to be testing at 32kA to maintain that 4-second 20 duration. We had the option of either going 35kA for 21 3.2 seconds or 32kA for 4 seconds, and we chose the 22 duration parameter as the most important. So, we 23 selected and altered our current to accommodate it.
24 MEMBER MARCH-LEUBA: So, what's the 25 generator, a battery or --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
158 1 MR. MELLY: It's a large -- essentially, 2 they spin up a generator and they have inertia running 3 it. And then, they disconnect from the grid itself to 4 power our arc, yes.
5 MR. HAMBURGER: It's a 2200-MVA generator.
6 And they use a process of super-excitation to maintain 7 steady voltage and current levels throughout the 8 energy delivery.
9 MR. MELLY: Yes, it's very fancy.
10 MR. HAMBURGER: Yes.
11 MEMBER MARCH-LEUBA: Just an inertia 12 wheel?
13 MR. MELLY: Yes.
14 MR. HAMBURGER: Right.
15 MR. MELLY: And that is one good segue.
16 I hadn't discussed it previously, but that 17 super-excitation will maintain the current throughout 18 our event. One of the comments we received from the 19 public workshop was that in a real-life situation, if 20 it's a generator-fed fault, that the generator has 21 tripped, they are not going to see this steady 22 current. They're going to have the generator decay 23 curve coming in. And that is something we are working 24 currently with EPRI to get real generator data from an 25 actual event in order to create a decrement curve that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
159 1 we can, then, use at KEMA to simulate these --
2 MEMBER BLEY: KEMA can do that?
3 MR. MELLY: We are in current discussions 4 with them as to what the capabilities are to try and 5 mimic that decay curve.
6 MR. MILLER: Yes, their setup is a lot 7 more than just a generator. They've got some real 8 fine controls and the ability to do some very fancy 9 things.
10 MR. MELLY: Yes. So, that's another area 11 that we're working with EPRI collaboratively to ensure 12 that our test program is realistic and can answer 13 questions that they have.
14 Again, on the test parameters, one of the 15 important ones is current, and one of the soundbites 16 or criticisms that we got is that not all arc faults 17 are three-phase faults, as we're initiating in this 18 test program.
19 However, as we see in operating 20 experience, with these long-duration events, because 21 of the amount of ionized gas you're creating in the 22 cabinet, even if it's initiated as a phase-to-ground 23 or phase II phase arc, it will quickly progress to a 24 three-phase arc for these high energy arcing faults.
25 At trip point, it did initiate as a phase II ground NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
160 1 fault, and it went to a three-phase fault during the 2 course of the --
3 MR. HAMBURGER: In nearly every LER that 4 I've reviewed, something that starts as a 5 phase-to-phase or phase-to-ground progresses to a 6 three-phase fault.
7 MR. MELLY: Yes, for these HEAF events.
8 So here, we're looking at a viewgraph that 9 shows a sample from the U.S. plants. We're trying to 10 understand what fault currents we can anticipate based 11 on their system design.
12 Gabe actually put this together as part of 13 the "Needs and Objectives" when we're trying to figure 14 out what currents would be representative of plants 15 out there. And we put together a sample from the 16 plants if we are unable to understand that. We, then, 17 had discussions with the industry counterparts at this 18 meeting, and that's what led us to our test-level 19 parameters at 15kA and 25kA to cover this range.
20 MEMBER SKILLMAN: Are these currents from 21 ETAP or some type of a program similar to ETAP?
22 MR. MELLY: Gabe, do you want to take that 23 one?
24 I believe that --
25 MEMBER SKILLMAN: These are not measured?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
161 1 This is estimated, I presume?
2 MR. MELLY: Yes.
3 MEMBER SKILLMAN: What is the basis of the 4 estimating? I'm thinking it's something like --
5 MR. MILLER: Probably from fault studies, 6 yes, which would be done with something like ETAP.
7 MR. MELLY: Yes, that's typically what I 8 would assume that this was created. I'm not sure if 9 we still have Gabe. But I believe that it was based 10 on ETAP.
11 MR. TAYLOR: I'm online. Are you asking 12 about the estimate that I put in the --
13 MR. MELLY: Yes, what's the values?
14 MEMBER SKILLMAN: Yes, where did your 15 estimates come from, please?
16 MR. TAYLOR: Sure. So, I looked at all 17 the plant data that we had access to in our internal 18 system, and I basically did an infinite Voss-type 19 calculation, not taking into account the energy from 20 running motors back feeding into the fault. So, I ran 21 that type of calculation to come up with that 22 estimate.
23 MEMBER SKILLMAN: Thank you. Okay.
24 MR. MELLY: And we did that again for the 25 medium-voltage cases at 4160, and I believe the next NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
162 1 slide shows 6.9 kV.
2 So, this effort to collect data from the 3 sample plants as well as getting information from 4 those members who did attend the public workshop is 5 what led to our test levels of 25kA and 35kA. We 6 wanted to make sure that this was realistic of actual 7 fault currents that could be seen in operating 8 experience.
9 MR. HAMBURGER: So, throughout this 10 process, I think we've tried to be as transparent as 11 possible and include as many people as possible. For 12 both the small-scale and the large-scale test plans, 13 we published Federal Register notices; we had 30-day 14 comment periods, and we have responded and 15 dispositioned all of the comments that we have 16 received.
17 So, on this large-scale phase II draft 18 test plan, we did receive 64 comments in total. And 19 after the deadline had closed, we received EPRI 20 comments as well that we have addressed.
21 The majority of these comments were those 22 that I had mentioned earlier, that industry 23 representatives felt as if our testing was not 24 representative of configurations at the plant. And 25 that's one of the things that led to that public NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
163 1 workshop that we held. So, we've addressed the 2 comment we have received. We've incorporated some 3 suggestions, including the decrement curve that we're 4 going to try to implement. This is what led us to put 5 an actual breaker inside the cabinets that we're going 6 to be testing. So, we've tried to be as responsive as 7 possible to concerns.
8 Next slide.
9 So, moving forward into our phase II 10 testing, the way it's organized is we have the 11 small-scale testing being conducted at Sandia, and we 12 have the large-scale testing that we're going to be 13 conducting, starting in two weeks, at KEMA's 14 laboratory.
15 Because the aluminum is primarily a U.S.
16 concern -- and I say that because in some countries 17 they don't have any aluminum at all; in some 18 countries, like Japan, they are handling this from a 19 regulatory perspective by replacing their breakers 20 with digital systems, and they use steel often for 21 seismic concerns rather than aluminum.
22 The OECD countries were not all interested 23 in paying for our focus on aluminum. So, we split the 24 large-scale testing into two parts. One is the joint 25 NRC/OECD/NEA tests, and those include 16 enclosures NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
164 1 and three bus ducts. And then, we have the ones that 2 we are shouldering the burden on alone with the focus 3 on aluminum, and that consists of eight electrical 4 enclosures and five bus ducts. And both of these test 5 programs have been designed so that we have points of 6 comparison between them. So, we're leveraging the 7 international cooperation to get the copper bus work 8 done, and we're going to do some supplementary 9 aluminum work to try to resolve this GI issue of the 10 aluminum contribution.
11 Next slide.
12 MR. MELLY: So, this is kind of a visual 13 depiction of what that means in terms of a test 14 program. You can see that any of the tests are 15 indicated in blue. That is the OECD test program.
16 And those in the orange are the NRC-only Generic Issue 17 Test Program.
18 As you can see, the way that we split it 19 up is we've attempted to have one-to-one comparisons 20 between each test, focusing on single-parameter 21 changes between the copper and aluminum as these 22 various power levels, durations, and the currents.
23 Additionally, the tests that Kenny 24 discussed that are going to be conducted next week are 25 actually indicated here. Test 219, 6.9 kV, 25kA, for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
165 1 2 seconds; test 221, 6.9 kV, 25kA, for 4 seconds, and 2 their parallels at the 35kA amperage. So, those are 3 the first four tests that we're going to run.
4 That is largely due in part that they're 5 driven by the Generic Issue Program, as well as 6 coordinating with all the member countries is a little 7 bit slower in terms of getting an agreement in place.
8 So, we can't test until we have firm signatures from 9 a lot of people, and we will get into that we have a 10 little bit of an increased interest in this phase of 11 testing. So, we're going to be adding member 12 countries.
13 Again, Kenny spoke that we are splitting 14 into the bus duct and the enclosures. We are going to 15 be taking more of the burden for the aluminum 16 enclosures that we see on bus ducts because it is seen 17 currently as a primarily U.S.-centric issue. We used 18 a lot of aluminum for housing material of these bus 19 ducts, and these tests are currently not slated at a 20 firm date, but they will be done in subsequent test 21 series.
22 MEMBER BLEY: They're not a firm date, but 23 what kind of time scale are you looking at?
24 MR. MELLY: So, the current time scale 25 that we're looking at is we have a test this fall for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
166 1 the four tests that we're running, and we plan on 2 doing a test this spring as well. However, we're 3 going to be expanding the number of tests that's done.
4 So, this test series, we're doing four. It's one week 5 of testing. Next test series and the subsequent 6 tests, we're planning on doing two weeks of testing 7 back to back, so that we can get it done in a quicker 8 timeframe.
9 This first test series in September, we 10 wanted to limit it to one test week to ensure that we 11 have our instrumentation at the correct location, that 12 we're doing everything to collect the data, that 13 everything is working properly, so we can move forward 14 at a --
15 MEMBER BLEY: Will all the aluminum tests 16 be done before the Generic Issue Paper is finished?
17 MR. MELLY: Not all of them.
18 MEMBER BLEY: Not all of them?
19 MR. MELLY: But we hope to have --
20 MEMBER BLEY: Enough to help that along?
21 MR. MELLY: We hope to have enough variety 22 of tests from low voltage, medium voltage, and bus 23 ducts to inform that working panel to set the 24 assessment on the right path.
25 MR. HAMBURGER: And we're selecting the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
167 1 order of our tests with that in mind.
2 MEMBER BLEY: With that in mind?
3 MR. MELLY: Yes.
4 MEMBER BLEY: Okay.
5 MR. HAMBURGER: In terms of our 6 experimental approach, this is significantly different 7 than what we went into with phase I. Phase I was 8 confirmatory. It was a potluck. We had the countries 9 donate whatever they could. If there was a mod going 10 on in a plant, they would give us their cabinet. And 11 it was sort of a smattering of -- I don't want to call 12 it random, but we were testing without tight 13 experimental controls, just to see if these cabinets 14 would respond as we had hoped they would.
15 Phase II is going to be much more focused 16 in terms of experimental controls. We're looking to 17 vary single parameter, and to that end, we have 18 secured duplicate cabinets that we're going to test.
19 So, currently, we have sitting in a warehouse four 20 identical cabinets for our next week of testing, so 21 that we can try to isolate what factors have the 22 largest impact on incident energy.
23 MEMBER BLEY: Were you able to get these 24 donated?
25 MR. HAMBURGER: No.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
168 1 MEMBER BLEY: No? I didn't think so.
2 MR. HAMBURGER: No, and that's one of the 3 difficulties we had with phase II. All the 4 international participants want to donate cabinets in 5 lieu of monetary contributions. But, in order to get 6 these good experimental controls, we really need 7 duplicate pieces of equipment. So, we're working 8 through that.
9 MEMBER SKILLMAN: What are the primary 10 characteristics of these four cabinets?
11 MR. HAMBURGER: These are metal-enclosed 12 switchgear. They are Magne-Blast breakers, and they 13 have aluminum bus ducts inside.
14 MEMBER SKILLMAN: Thank you. Okay.
15 MR. MILLER: Very similar to ducts used at 16 current U.S. plants minus the aluminum bus work.
17 MR. HAMBURGER: So, in terms of layout, 18 it's going to be similar to what we did in phase I 19 with a couple of updates. We are still looking at 20 putting our instrumentation racks at 3 and 6 feet.
21 We've made some pretty significant modifications to 22 the types of instruments and the instrumentation racks 23 themselves.
24 When we get to the bus duct testing, as 25 Mr. Bley noted, we had that bus duct on the ground, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
169 1 which I think maybe altered what we saw in terms of 2 direction and magnitude of the heat. For our phase II 3 testing, we're looking at putting that bus duct up in 4 the air with a vertical stacking of racks to try to 5 measure, more in line with how the PRA model 6 anticipates it will be positioned in a plant.
7 We talked about most of these measurements 8 when we discussed lessons learned from phase I. We're 9 still measuring temperature and heat flux. We're 10 adding in that tungsten slug for some of the 11 high-level fluxes. We're using the piezoelectric 12 pressure transducer.
13 Some of the things that we're going to be 14 recording is qualitative data in terms of what was 15 damaged and how far away was it damaged. So, we're 16 looking at things that are thrown from the cabinet, 17 how far they're thrown, where they land, what their 18 state is, and other observations like smoke damage, 19 ensuing fires.
20 We spent a week up at the warehouse where 21 these cabinets are being housed, taking them apart and 22 weighing each component of them. So, we've removed 23 the bus bars from the cabinets and we've weighed them, 24 to, hopefully, post-test, look at how much mass was 25 lost during these tests. So, we took the whole NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
170 1 cabinet apart. We weighed the panels. We weighed the 2 bus bars. We even weighed the bolts and nuts. So, 3 we're hoping to get a very good handle on just how 4 much material is vaporized or oxidized during our 5 future testing.
6 Another qualitative data point, we're 7 placing cable coupons on these instrumentation racks.
8 This is just to observe what happens to these cable 9 samples as they're hit with the plasma and the ionized 10 gas.
11 And Sandia will be outfitting our 12 instrumentation racks with the aerogels and the carbon 13 tape that they're using in the small-scale testing.
14 They're going to bring them to our large-scale testing 15 to try to capture some of the particulate and ejecta 16 that's deposited during the large-scale testing. And 17 they'll be doing some post-test forensic analysis on 18 that byproduct and ejecta.
19 MEMBER SKILLMAN: Will the fixation of 20 those coupons adulterate your results?
21 MR. HAMBURGER: That was a topic of 22 discussion. We've placed them on the instrumentation 23 rack, so that if they do catch fire, they will have 24 not a significant impact on the data we're collecting.
25 And I'll get to that when we --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
171 1 MR. MELLY: It's on the next slide.
2 MR. HAMBURGER: Is it the next slide?
3 MR. MELLY: It's the next slide, yes.
4 MR. HAMBURGER: So, that was one of the 5 considerations when we put together this 6 instrumentation layout. And if you look at the very 7 top corners, you will see those cable coupons. We put 8 them as far out as possible, so that if they do catch 9 fire, they have as little impact as possible on the 10 rest of the instrumentation.
11 The two racks that we see here, one of 12 them is placed at 3 feet from the cabinet; one of them 13 is placed at 6 feet from the cabinet. And we've 14 designed them such that the front rack does not shield 15 the rear rack from the effects of the HEAF. So, we've 16 offset these horizontal pieces, so that we get the 17 full blast at 6 feet, where possible.
18 The tungsten slugs, the ones that are 19 measuring those high-level heat fluxes, we've placed 20 in sort of a circle in the center of the rack. The 21 idea is that we're going to place that center bar 22 where we anticipate the arc to initiate, and we're 23 trying to get as much of a direct hit as possible on 24 that center bar. So, we can adjust those horizontal 25 pieces up and down as necessary to center them on the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
172 1 point of arc initiation.
2 Just to go over everything that's on this 3 rack, we have the ASTM Slug Calorimeters that KEMA is 4 providing. We have the cable coupons, carbon tape.
5 We have aerogel. We have plate thermocouples, and we 6 have our tungsten slugs.
7 MEMBER SKILLMAN: Question: did you learn 8 anything on phase I small-scale tests or phase II 9 small-scale testing that's going to adjust this, the 10 testing we just finished?
11 MR. MELLY: I was going to try to let Gabe 12 take that. But we're still evaluating the small-scale 13 test program. We know that we did collect a 14 significant amount of information from the aerogels, 15 and we're going to be bring that forward to this test 16 program. We're also evaluating putting the aerogels 17 or carbon tape within the cabinet itself, which we 18 can't do with our electric instrumentation due to risk 19 of shorting. However, these passive techniques, such 20 as the aerogel and carbon, we can put into the 21 cabinet, in trying to understand distance-related 22 effluent, if there's any differences in the 23 concentration.
24 Additionally, we're going to be bringing 25 the high-speed video from the small-scale to the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
173 1 large-scale test program. So, we're hoping that the 2 cabinet survives the testing. But we do think that 3 we'll be able to use a lot of the information from the 4 small-scale towards the large-scale results.
5 MR. HAMBURGER: So, that high-speed video 6 that we saw and the small-scale sample video, they're 7 going to be bringing that camera and placing it in the 8 cell. And we're working on an array of mirrors, so 9 that we can protect the camera from the blast and 10 still get a good visual in high speed.
11 MR. MELLY: Yes. One of the main 12 questions, though, between small-scale and large-scale 13 is, what is the survivability of these carbon tapes 14 and aerogels? Because during the small-scale testing, 15 we're on the order of microseconds -- or milliseconds, 16 I believe, that we're testing, due to the power 17 capability of Sandia's lightning generator.
18 Chris LaFleur, what was the exact 19 timeframes that we went for the small-scale?
20 MS. LaFLEUR: This is Chris.
21 We were looking at up to 8 microseconds.
22 We got larger than that. I think we got to 32 23 microseconds.
24 MR. MELLY: So, that's the main difference 25 of are these going to survive in a large-scale NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
174 1 environment where we're testing on the order of 2 seconds. So, those are some of the questions we'll be 3 answering in two weeks.
4 MEMBER SKILLMAN: I see that you have an 5 inventory of 500 total instruments and you have these 6 identified for certain locations for the racks.
7 MR. MELLY: Uh-hum.
8 MEMBER SKILLMAN: Do you have extras just 9 in case the day of the test someone shows up and says, 10 "You know what? Two of those are really not the 11 correct ones" or "The one I was supposed to put in 12 this afternoon I left at home in Cincinnati."?
13 MR. MELLY: For some of the 14 instrumentation, we do have extras. For others, we do 15 not. And we will, hopefully, be able to repurpose 16 some of the instrumentation if it is not damaged. And 17 what I mean by that, if we look back at the picture on 18 slide 86, which is Gabe's slide, 85, a few of the 19 racks will not be in the direct direction of the arc 20 pathway, hopefully or potentially. If those do not 21 see a direct hit and they are not damaged by a single 22 test, we may be able to repurpose them for later tests 23 by either painting the surface, so that we have the 24 same emissivity as in the first test, and we're going 25 to, hopefully, be addressing those things as we move NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
175 1 forward.
2 One of the constraints of this test 3 program was that we did need to test this September to 4 meet up with KEMA's schedule, as well as potentially 5 avoiding any testing in October, because we have had 6 experience with running into government shutdown 7 issues with test programs, which is not ideal in any 8 sense of the word. So, we have been stretching the 9 capabilities of the creation of these devices both at 10 NIST and KEMA.
11 MR. HAMBURGER: But we're going in with 12 the assumption that we have enough equipment, new 13 sensors for each and every test. We're going to take 14 the old rack out and we're going to put the new rack 15 in. We haven't accounted for the ability to reuse any 16 of the equipment. So, even where there are not spares 17 -- and for most of the instruments there are -- we're 18 hoping that enough will survive that, if we had to 19 fabricate one or two new racks, we could do that 20 without a problem.
21 MEMBER SKILLMAN: Yes, I was just thinking 22 it would be very unfortunate to have gone through all 23 of this effort, and at the last minute find that one 24 or two instruments that you really need are not 25 available. The kind of things that happen in real NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
176 1 life and you say, "Golly, how did I forget that?"
2 MR. MELLY: One other thing that is not 3 depicted on the test setup here is that we are now 4 installing a rack positioned horizontally above the 5 electrical enclosure. So, in addition to the vertical 6 racks on the side, we have one rack mounted to the top 7 of the cabinet at a 3-foot distance to try to capture 8 anything that comes up.
9 Okay. For the phase II testing, I had 10 mentioned previously that we do have somewhat of an 11 enhanced interest from the different member countries.
12 We are looking at having Belgium, Canada, the Czech 13 Republic, France, Germany, Korea, Japan, the 14 Netherlands, Spain, and the USA participate in this 15 phase. Belgium will be a new participant as well as 16 the Czech Republic and the Netherlands. The only 17 country that we are missing from the previous phase I 18 test program is Finland, and that was due to a 19 personnel resource issue, that they just did not have 20 the resources to devote someone to this project.
21 So, we are looking at getting the 22 international agreements signed, and it's hopefully 23 moving forward and will be signed this fall.
24 We also, in addition to the international 25 partners, we're looking at having U.S. interagency NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
177 1 participation. So, we've been reaching out through 2 the Federal Fire Working Group meetings that are held 3 twice a year to discuss this high energy arcing fault 4 program with our federal partners, because, like we 5 said earlier, this is not just a nuclear problem or a 6 nuclear issue; it's broad.
7 So, we're looking to share our 8 understanding of this occurrence, and we've reached 9 out specifically to the ATF. They have some portable 10 Gardon gauges that might not have the logistical 11 issues of running water that we might be able to use 12 for future testing, as well as NASA is very interested 13 in this test program because they have had several 14 high energy arcing faults at their facilities. And 15 they have some interesting camera capabilities and 16 technology that we might be able to apply to looking 17 at the event itself and using some filtering 18 algorithms to actually look through the arc and look 19 at what's happening to the cabinet itself. So, that's 20 some of the interagency working that we are going to 21 be doing.
22 MR. HAMBURGER: NASA also has a lot of 23 experience with aluminum combustion and agglomeration.
24 They use it in their rocket fuel. So, on the 25 launchpad they've done studies of the aluminum that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
178 1 precipitates out of the rocket engine and how it 2 agglomerates, how it oxidizes on the launchpad. So, 3 they have a lot of experience with aluminum 4 combustion.
5 MR. MELLY: Yes. And as we previously 6 described, we are working with NFPA and IEEE on the 7 definition front, and they are fairly interested in 8 this project as well. And they may come witness some 9 of the testing.
10 So, one of our main challenges with the 11 international agreement, and the reason it's probably 12 not signed right now, is that we have a lot of lawyers 13 involved with the different member countries. And the 14 OECD has moved forward to a template that they want 15 to, then, use for all their international agreements.
16 And in that template there is a clause that 17 essentially indemnifies the owner of the project to 18 all liabilities. We cannot sign that, per the OGC, as 19 the NRC, per the Antideficiency Act.
20 So, we're currently in discussions with 21 the member countries who want to participate, the OECD 22 and NEA, as to how we can move forward without having 23 that clause in there. I believe that we're coming to 24 an agreement on that currently. We actually had a 25 meeting an hour before this meeting to, hopefully, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
179 1 finalize where that is going. And we will be pushing 2 forward.
3 MEMBER SKILLMAN: So, that will be some 4 form, you hope, of an indemnification waiver or 5 everybody's an even partner, and you come with what 6 you've got and you take what you get, and if anything 7 happens, it's not our fault, that type of thing?
8 MR. MELLY: I would hate to speculate on 9 what the lawyers will come up with on the record.
10 (Laughter.)
11 But we do have the OGC working on a final 12 resolution, and it is something to that effect --
13 MEMBER SKILLMAN: Okay.
14 MR. MELLY: -- trying to stipulate where 15 the liability does belong.
16 MEMBER SKILLMAN: Fascinating. Who would 17 have thunk it?
18 (Laughter.)
19 MR. MELLY: Timeline of actions. We have 20 completed many things. We've had the official OECD 21 comment period. We have had our HEAF Working Group 22 meetings. We held our HEAF workshop in April. We had 23 an additional OECD/NEA HEAF meeting April 23rd, the 24 subsequent week after our U.S. workshop, where we 25 shared insights to those member countries who could NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
180 1 not participate. We have completed our comment 2 resolution, and we are making that publicly available, 3 the full comments that we received on the test plan.
4 We hope to have our signed international agreement 5 this fall. We have finalized the test plan and shared 6 that with the international community as well as 7 publicly.
8 And we have our first test series that we 9 have discussed September 10th through the 14th, where 10 we will be testing four pieces of equipment. We're 11 looking to do the second series in the spring and the 12 remaining tests throughout 2019-2020.
13 MEMBER SKILLMAN: I would ask, if you are 14 able, to send to Kent Howard the details of those 15 tests up at KEMA. I don't know that members will be 16 able to attend. We have restrictions on our travel 17 and our time. But some members might find a path 18 forward to come and witness one or several. So, if we 19 just know about it, we can at least give it 20 consideration. I think the members might appreciate 21 that.
22 MR. MELLY: Okay.
23 MR. SALLEY: Yes, and we've had a lot of 24 requests like that from different groups. And the 25 12th and the 13th would look to be the best dates, if NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
181 1 you would like to potentially pencil that in on your 2 calendar.
3 You can do it as a day trip from here, if 4 you're down here for anything. The KEMA facility is 5 about 15-20 minutes away from Region I offices, just 6 to give you a perspective. And we can work with the 7 tests. We're trying to do one a day. So, they will 8 be in the morning doing a lot of setup, getting that 9 ready. Obviously, the tests are a couple of seconds, 10 but we'll try to run that in the afternoon. So, it 11 allows you to leave here from Headquarters, drive up 12 in a van, see the test, get back in and make it down 13 here in a day. So, it is very doable. Like I said, 14 there will be different groups from NRC heading up, 15 too.
16 MR. MELLY: Yes, and I can provide the 17 details to Kent, as in what you would need to witness 18 the testing, where it is, and details like that.
19 We do also plan to have the OECD 20 international meeting for all the member countries.
21 They are going to be coming in the spring of this year 22 -- or the spring of next year, to actually witness 23 these tests. We're going to hold our biennial meeting 24 either at the Region I offices or at the KEMA 25 facility, so that we can witness tests.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
182 1 MR. SALLEY: And again, the first two 2 days, the 10th and 11th, we want to make sure the 3 tests go off well before you come up and see it.
4 We've learned from experience. So, the latter two 5 days would be --
6 MR. MILLER: That's a Wednesday and 7 Thursday.
8 MEMBER SKILLMAN: We will be in session 9 the several days after Labor Day for our September 10 full Committee meeting. The following week is the 11 week that you are speaking about. And the week that 12 follows that, we've got some heavy-duty lifting in 13 this room.
14 So, I don't want to be suggesting the 15 members can come. We have members from the West 16 Coast, from the deep Southwest. For members to come, 17 it's a major issue. But just that they know when 18 these tests are, one or some might find it an 19 opportunity. Thank you.
20 MR. MELLY: Okay. Yes, we can provide 21 that here and moving forward as well.
22 So, I think this may be the last slide 23 before closing with questions and comments. These are 24 some of the planned publications that we have coming 25 out of this work.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
183 1 We're looking at getting the NUREG CP 2 published on the workshop that was held in April.
3 That will include the workshop presentations, 4 feedback. We're also looking at publishing a NUREG 5 Sandia report on the small-scale testing and an 6 updated, more expansive literature review.
7 We're looking at publishing a NUREG with 8 EPRI which will cover the definitions that we 9 described earlier and the frequencies. That's going 10 to be part of the Memorandum of Understanding, the 11 MOU.
12 We're looking at publishing two additional 13 NUREG IAs which cover subsequent Japanese testing that 14 we were able to witness and be a part of. We're also 15 looking at publishing the second phase of HEAF testing 16 in the same manner that we published the first, as a 17 CSNI report. We are going to be publishing the 18 aluminum-only tests as part of the generic issue, as 19 a NRC NUREG, not as a CSNI report.
20 We're, then, going to also be publishing 21 a NUREG on the data analysis of the full series, both 22 OECD and the aluminum tests. And finally, we're going 23 to be publishing a NUREG/EPRI document which is a 24 joint NUREG to cover the methodology changes on 25 changes to zone of influence. And anything that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
184 1 touches 6850 has typically been done in a joint NUREG 2 fashion, and we're looking to continue that with EPRI.
3 We have proposed that this joint NUREG will be 4 reviewed by the full OECD group because they do bring 5 a lot of experience to the table as being members of 6 the program.
7 And with that, I think that we are done, 8 pending any further comments or questions.
9 MEMBER SKILLMAN: Well, let's just hold 10 for a second here. Let me ask my colleagues around 11 the table here if they have any further questions.
12 Yes, do you have any questions? Dennis? Matt?
13 Charlie? Any questions for the guys in the front?
14 MEMBER BROWN: Well, I was just on the 15 internet a little bit. There is an IEEE document that 16 was issued in 2015. And I found the history on the 17 Navy one where they talk about three different 18 sensors, the photosensors, pressure sensors, as well 19 as thermal ionization detectors, where they credited 20 on Navy ships 11 circumstances in that particular Navy 21 writeup.
22 MEMBER BLEY: Interesting. My apologies 23 for my earlier comment.
24 MEMBER SKILLMAN: It's widespread.
25 MR. MELLY: Yes, and in addition to that, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
185 1 you'll also see that some of the recent things that 2 have been done is arc-resistant cabinets. Actually, 3 the physical design of the cabinet is so that, if you 4 do have an arc within the cabinet, it follows arc 5 shoots and goes in a specific location, where it's not 6 going to come out of the front of the cabinet itself 7 and damage workers.
8 However, we are dealing with cabinets that 9 are '70s vintage in the plant. So, I would love to 10 say, rip out all your old cabinets and put in these 11 new ones with the Navy technology. However, I think 12 that's out of my purview.
13 MEMBER BROWN: Well, and the other 14 argument, when we look at your slides, where you've 15 talked about like 48 out of 415, or some number like 16 that --
17 MR. MELLY: Yes.
18 MEMBER BROWN: -- that were credited 19 towards a HEAF circumstance from '87 to recent, 20 whatever that is, which works out to be like 1.7 per 21 year, or something like that, on average. I don't 22 know what the damage estimate was, but it looks like, 23 on a risk basis, it's not huge, when you look at the 24 total number of plants.
25 And are you all just looking at nuclear NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
186 1 plants for your data or are you looking at -- I mean, 2 there's only 20 percent of the electricity that is 3 generated by nuclear plants. There's another 80 4 percent, which means there's a lot more plants out 5 there, like 400 or 500 of them, that are coal, oil, or 6 gas. They've got all the same high-power switchgear 7 and everything else in them. Are they off putting in 8 arc fault stuff? Or they just live with the results?
9 Why would nuclear be different than them?
10 MR. MELLY: For the frequency values that 11 we are using for the evaluation, we are only looking 12 at nuclear. And that is partially because there are 13 arguments that the maintenance may be different 14 between a nuclear power plant and other types of 15 electrical generation. Whether that argument holds 16 water or not is a larger discussion.
17 MEMBER BROWN: I can pour water on that 18 one.
19 MR. MELLY: Yes.
20 MEMBER BROWN: I would personally.
21 (Laughter.)
22 MR. MELLY: But for the frequency, yes, 23 it's only nuclear. Also, for regulation, these are, 24 again, '70s, '60s vintage plants or pieces of 25 equipment; whereas, other newer power plants will be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
187 1 using newer equipment, possibly to different 2 standards, using this newer technology.
3 MEMBER BROWN: Well, I can guarantee the 4 Navy stuff does not look like the greatest and latest 5 new tech. Switchboards, they're steel and copper bus 6 bars in most circumstances. There is aluminum some 7 places. We tried to stay away from that, at least in 8 the submarines. I don't know whether that's changed 9 or not since I retired.
10 MR. SALLEY: With the data from other 11 plants, I can speak to my time at TVA. When you have 12 something like this happen in a nuclear plant, 13 obviously, there's a lot of people looking at it, and 14 we do root-cause evaluations and we tend to be very 15 thorough. And that's what you see here.
16 When we would see this in a fossil plant, 17 basically, it was how quickly can we get this plant 18 back up.
19 MEMBER BROWN: Well, that sounds like a 20 good plan.
21 (Laughter.)
22 MR. SALLEY: Well, that was the plan. So, 23 you went to the extent of damage. You cut it out.
24 You pulled the original drawings. You pulled the 25 building material. How fast can we get this stuff in?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
188 1 How fast can we hook it up, and how quick can we get 2 going again?
3 MEMBER BROWN: Yes. Yes. I guess I was 4 just raising the issue of why -- I mean, you look at 5 where all the electric plant equipment is. Basically, 6 it's separated from the boilers and all the other 7 fossil fuel stuff. Switchgear on these, every time 8 I've looked at a diagram, it's pretty far away 9 relative to the nuclear plant, even those that I've 10 walked through.
11 So, I'm just trying to look at 12 similarities. There's a lot of hurdles to justify.
13 And I'm not trying to pour wet water -- wet water?
14 How about that? You can tell I'm an electrical guy.
15 (Laughter.)
16 Pour water on it.
17 MR. SALLEY: It's firefighting foam, but 18 go ahead.
19 MEMBER BROWN: Exactly, whatever.
20 MEMBER SKILLMAN: Well, there is a huge 21 difference. The equipment that we're concerned about 22 may be credited, yes, for exit analysis. And so, 23 let's say you're chief engineer at a nuke and you've 24 had an arc fault. You may have taken out two strings 25 of 4160 that is essential for ECCS. And so, while you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
189 1 might be able to overcome that with an exigent tech 2 spec request, you are now in a degraded condition that 3 will probably force you to shut the plant down. And 4 depending on the extent of damage, it could be down 5 for a week, two weeks, a month, two months. And with 6 the current environment, that is a kiss of death 7 financially.
8 So, there are some real practical 9 differences between what you might do with a breaker 10 in, say, your cabinet strength in a nuke versus the 11 identical equipment that might be in a gas turbine 12 plant or a fossil plant.
13 MEMBER BROWN: Well, the Navy doesn't do 14 this with breakers. These are just installed down in 15 the bottom of the cabinets, and then, the photosensors 16 are located, as well as the TIDs, to detect the 17 thermal ionization, and what have you, and to plot 18 these suckers. I mean, you don't replace any of the 19 hardware. The sensors go inside the boxes or outside.
20 MR. MILLER: And then, they provide 21 redundant trip functions to the breakers?
22 MEMBER BROWN: Yes.
23 MR. MILLER: Yes.
24 MEMBER BROWN: And they trip downstream --
25 upstream breakers.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
190 1 MR. MILLER: Right. Just trip everything 2 around it.
3 MEMBER BROWN: You can do --
4 MR. MILLER: Yes.
5 MEMBER BROWN: I mean, there's a bunch of 6 different -- it depends on the arrangement and what 7 you're powering, and everything else.
8 MR. MILLER: Right.
9 MEMBER BROWN: But, I mean, there's no --
10 the internal equipment does not change. It's a matter 11 of going into the switchboards and putting the stuff 12 in. And the Navy switchboards, let me tell you, are 13 very difficult because they're very compact. We pack 14 that stuff in like gangbusters because you just don't 15 have enough -- you don't have a lot of room.
16 MEMBER SKILLMAN: Yes. Let's go around.
17 MEMBER BROWN: So, anyway, I just wanted 18 to toss some other information out; that's all.
19 MEMBER SKILLMAN: Good, Charlie.
20 And Jose, any questions for the team here?
21 MEMBER MARCH-LEUBA: Well, no. No 22 questions.
23 MEMBER SKILLMAN: I'm making sure that the 24 members that have persevered to be here have a chance 25 to ask questions.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
191 1 (Laughter.)
2 Now, at this point, I would like to ask if 3 there is anybody on the phone line that would like to 4 let us know you're there. Please say "hello," and if 5 you have a comment, please make your comment.
6 (No response.)
7 Hearing none, we can close the phone line.
8 Are there any individuals in the audience 9 that might like to make a comment, please?
10 (No response.)
11 Hearing none, let me make several closing 12 comments.
13 I want to thank my teammates for staying 14 on what is really our last full working day this 15 summer. I mean, it's rare for us to be here on a 16 Friday like this. Gentlemen, thank you.
17 Thank you to RES, to Mark, your team, to 18 Mark Thaggard. Thank you, sir, for this every 19 extraordinary presentation. This has been, I think, 20 a wakeup for all of us.
21 I would like to say, when we recognized 22 that AIM was going to pull this, almost all of us 23 said, "Time out. That's not a good idea." And to see 24 that this is moving forward lets me think maybe the 25 ACRS has a little bit of a flag that someone might NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
192 1 recognize here. So, I thank the Commissioners and 2 others for not shutting this down.
3 And I wish you well on your endeavors as 4 you go ahead.
5 Final comments, anybody?
6 MR. SALLEY: I have a few comments, if 7 that's okay.
8 MEMBER SKILLMAN: Shoot, Mark. Go ahead.
9 MR. SALLEY: Yes, thank you.
10 Again, August is the time most people take 11 off, and I really appreciate you making the time for 12 us and hearing us before we go to the next round of 13 testing.
14 I think you're okay with what you see and 15 where we're going --
16 MEMBER SKILLMAN: Yes.
17 MR. SALLEY: -- with the next round of 18 testing.
19 And again, thank you for your support when 20 the AIM 2020 was going and we were in danger of losing 21 this project. Your support helped us greatly in 22 there. So, again, thank you for that.
23 Just a couple of comments that I would 24 like to make in closing on our side, a couple of notes 25 I jotted down.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
193 1 Overall, we really want to try to stay in 2 process, and that's why we had Tom Boyce stand here 3 and work this through the Generic Issue Program. I 4 believe this is a generic issue. We just need to stay 5 in process and work it through.
6 We want to continue to work with our 7 international partners. They came to the dance. You 8 know, we brought them to the dance. They've worked 9 with us. So, even though we've got the aluminum HEAF, 10 we still want to work with them to understand the 11 greater risk with it. So, we plan to do that.
12 As you heard in the presentations, we've 13 tried to be as transparent and as open and public as 14 we possibly can with the FRNs, the workshops. And we 15 really wanted to be collaborative with industry 16 anywhere along the road.
17 For the last five-six years, going to the 18 fire forums, I've been encouraging them, "Please 19 contribute. Please attend this." And we've really 20 been trying our best to do that.
21 Communications, very important. You're 22 starting to see some Information Notices coming out, 23 the things that we're putting out, the test reports 24 and such. Again, it's transparent. We're putting it 25 out there, and that's how we want to go forward with NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
194 1 this.
2 The RIC, we've had two RIC sessions where 3 we've put this. So, again, we're being very 4 transparent.
5 With that, thank you again for taking the 6 time and doing this with us. And we will move forward 7 as we learn more in the next series of tests, and we 8 can come back and brief you again and let you know 9 where we're going.
10 MEMBER SKILLMAN: You're welcome.
11 Thank you.
12 And I want to acknowledge Kent and your 13 effort to help pull all this together. Thank you, 14 Kent.
15 With that, ladies and gentlemen, we are 16 adjourned.
17 (Whereupon, at 4:51 p.m., the meeting was 18 adjourned.)
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
High Energy Arcing Fault (HEAF) Research Update for ACRS Nicholas Melly Gabriel Taylor P.E.
Kenneth Hamburger David Stroup P.E.
Mark Henry Salley P.E.
Kenn Miller Office of Nuclear Regulatory Research Division of Risk Analysis August 24, 2018 1
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Purpose
- Provide information to the ACRS PRA Subcommittee on the NRC Fire Research Activity for the High Energy Arcing Fault (HEAF) Program
- RES commitment to brief ACRS prior to next phase of testing (ML18057B139)
- Provide status of aluminum HEAF Generic Issue (Pre-GI-018)
- Provide status of OECD/NEA international HEAF research program 2
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Agenda
- Definition of HEAF hazard
- Background of HEAF research program
- Phase I testing
- Generic Issue, IN, PIRT
- Phase II measurement improvements
- Small-scale research program
- Public workshop
- Phase II parameters and testing 3
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Background of Current HEAF Treatment in PRA 4
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
NUREG/CR-6850 EPRI 1011989
- NUREG/CR 6850 forms the basis for nuclear power plant (NPP) Fire PRAs
- Fire initiators are broken down by bins
- Bin 15 Electrical Enclosure Fires
- Bin 16 HEAF
- NFPA 805 Lessons Learned
- Bin 15 is too broad
- Low voltage controls modeled the same as medium voltage switchgear https://www.nrc.gov/reading-rm/doc-
- Clear distinction between arc collections/nuregs/contract/cr6850/
flashes and HEAFs 5
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
NUREG/CR-6850 Electrical Enclosure Failure Modes 6
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Electrical Enclosure Fire Experiments (Bin 15)
- Heat Release Rates of Electrical Enclosure Fires (HELEN-FIRE)
- 112 full-scale electrical enclosure fires
- Developed a series of heat release rate (HRR) profiles
- Non-energized https://www.nrc.gov/docs/ML161 1/ML16110A037.pdf 7
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Electrical Enclosure Fire Methodology
- Refining And Characterizing Heat Release Rates From Electrical Enclosures During Fire (RACHELLE-FIRE) Volume 1: Peak Heat Release Rates and Effect of Obstructed Plume, Final Report (NUREG-2178, Volume 1, EPRI 3002005578)
- NRC/EPRI working group
- Classification of electrical enclosures (function, size, content, ventilation)
- Determined HRR probability distributions for corresponding categories https://www.nrc.gov/reading-rm/doc-
- Characterization of obstructed fire collections/nuregs/staff/sr2178/
plumes
- NIST Fire Dynamics Simulator (FDS) 8 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
PRA Risk-Significant Contribution
- Presentation by EPRI during the 2018 Regulatory Information Conference (RIC) Session TH30 - Improving Realism in Fire PRA 3rd highest contributor 9
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
International & Domestic HEAF OpE Sample International Operating Expierience NUREG - Fire Country Date Type Canada 04/15/2005 High Voltage HEAF 2169 Bin Ignition Source Frequency Canada 10/15/2005 Low Voltage HEAF 15 Electrical cabinets 3.00E-02 Czech Republic 02/03/1994 Medium Voltage HEAF 16.a (non-HEAF) 1.52E-04 Czech Republic 02/17/2010 Low Voltage HEAF 16.b HEAF for medium-voltage electrical 2.13E-03 Finland 04/12/1991 Medium Voltage HEAF cabinet (>1000 V) Finland 09/27/2006 Medium Voltage HEAF 16.1 HEAF for segmented bus duct 1.10E-03 France 01/19/2001 Medium Voltage HEAF 16.2 HEAF for iso-phase bus duct 5.91E-04 France 07/25/2010 High Voltage HEAF France 09/18/2001 Medium Voltage HEAF France 08/30/2003 Medium Voltage HEAF U.S. Operating Expierience France 09/29/2004 High Voltage HEAF Plant Name Date Type Bin France 10/30/1990 Medium Voltage HEAF Yankee Rowe 02/08/1984 Low Voltage HEAF 16.a Germany 09/08/1989 Medium Voltage HEAF Germany 03/14/2008 Low Voltage HEAF Oconee 01/03/1989 Medium Voltage HEAF 16.b Germany 10/30/2002 Low Voltage HEAF Robinson 07/13/1990 Medium Voltage HEAF 16.b Germany 08/23/2004 Medium Voltage HEAF Davis Besse 10/14/1991 Medium Voltage HEAF 16.b Germany 05/30/1986 Segmented Bus Duct Waterford 06/10/1995 Medium Voltage HEAF 16.b Germany 06/28/2007 High Voltage HEAF Prarie Island 08/03/2001 Medium Voltage HEAF 16.b Germany 02/08/1996 Low Voltage HEAF Robinson 03/27/2010 Medium Voltage HEAF 16.b Germany 08/11/2002 Low Voltage HEAF Robinson 03/27/2010 Medium Voltage HEAF 16.b Germany 09/09/1987 Medium Voltage HEAF Germany 04/19/1988 High Voltage HEAF Limerick 04/15/1980 Segmented Bus Duct 16.1 Germany 05/17/1989 Low Voltage HEAF Unknown 03/19/1987 Segmented Bus Duct 16.1 Germany 08/11/1979 Medium Voltage HEAF Kewaunee 07/10/1987 Segmented Bus Duct 16.1 Japan 03/11/2011 Medium Voltage HEAF Kewaunee 03/02/1988 Segmented Bus Duct 16.1 Japan 09/07/1996 Low Voltage HEAF Japan 08/31/1985 Medium Voltage HEAF TBD 07/06/1988 Segmented Bus Duct 16.1 Japan 07/16/2007 Medium Voltage HEAF Diablo Canyon 05/15/2000 Segmented Bus Duct 16.1 Korea 04/22/2002 High Voltage HEAF Browns Ferry 07/27/2008 Segmented Bus Duct 16.1 Korea 01/30/2001 High Voltage HEAF Spain 06/22/1988 High Voltage HEAF TBD 07/15/1988 Iso-Phase Bus Duct 16.2 Spain 08/20/1988 High Voltage HEAF Sherron Harris 10/09/1989 Iso-Phase Bus Duct 16.2 Spain 12/02/1988 High Voltage HEAF Vermont Yankee 06/18/2004 Iso-Phase Bus Duct 16.2 Sweeden 10/30/2002 Low Voltage HEAF 10 Sweeden 09/15/2006 Medium Voltage HEAF Sweeden 05/10/2011 Low Voltage HEAF Sweeden 11/14/2006 Medium Voltage HEAF Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
NRC Regulations
- 10CFR 50 Appendix A General Design Criteria (GDC)
- GDC 3 Structures, systems, and components important to safety shall be designed and located to minimize, consistent with other safety requirements, the probability and effect of fires and explosions.
- GDC 17 The onsite electric power supplies, including the batteries, and the onsite electric distribution system, shall have sufficient independence, redundancy, and testability to perform their safety functions assuming a single failure.
11 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
NRC Information Notice 2008-21
- Typical Breaker Failure Mechanisms
- Deficient fit-up with cubicles
- Worn or misadjusted linkages
- Inadequate or inappropriate maintenance practices
- Configuration control errors
- Deficiencies from original design and refurbishment
- Design changes
- Foreign material entry Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018 12
Potential HEAF Root Cause Unknown 10% 15%
Human Error 10%
Breaker Failure 10%
Poor electrical connection Over current 20%
15% Aging/Degredation Loose Material 5% Design Deficiency 15%
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018 13
Definition of HEAF Hazard 14 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
HEAF Hazard Ref. UAW Electrical Safety in the Workplace 15 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
HEAF Definition (Bin 16)
- Event frequency classification
- Need for clear definitions & subdivisions
- Arc Blast
- HEAF
- NRC currently working with NFPA/IEEE
- Continued discussions to finalize definitions for arc fault events 16 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Arc Flash-Working Definition
- Arc Flash - Damage is contained within the general confines of the component of origin
- These events are associated with minor damage and minimal bus bar degradation from melting/vaporization
- NFPA 70 National Electrical Code (NEC) terminology
- Research directed toward personnel safety
- Active NFPA/industry research topic 17 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Arc Blast- Working Definition
- Arc Blast- Damage is contained within the general confines of the component of origin; however, arc blasts have the potential to damage surrounding equipment through pressure rise effects (i.e. severe equipment deformation, thrown doors, degraded fire barriers)
- Typically do not create ensuing fires
- Typically associated with electrical coordination and breaker performance as designed
- Pressure effects are highly dependent on room configuration and electrical characteristics of the event 18 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
HEAF- Working Definition
- HEAF - Damage includes the component of origin as well as spread to surrounding equipment within the fire zone. This damage includes pressure rise effects (i.e. severe equipment deformation, thrown doors, degraded fire barriers) which can potentially effect equipment in other fire zones
- These events are typically accompanied by ensuing fire conditions
- Typically indicative of a level of circuit protection failure and/or design flaw allowing for extended duration arc events
- Pressure effects are highly dependent on room configuration and electrical characteristics of the event 19 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Arc Fault Classifications 20 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Background of HEAF Research Program 21 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Background of the HEAF Program
- OECD Fire Incident Records Exchange Project (FIRE)
- Analysis of High Energy Arcing Fault (HEAF) Fire Events, NEA/CSNI/R(2013)6
- 48 of 415 fire events collected represent HEAF-induced fire events (over 10%)
- International Partners https://www.oecd-
- Canada, Finland, France nea.org/nsd/docs/2013/csni-r2013-6.pdf Germany, Japan, Korea, Spain, U.S.
22 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Background of the HEAF Program CSNI WGIAGE Task on High Energy Arcing Faults (2009 - 2013)
- Task Report A Review of Current Calculation Methods Used to Predict Damage from High Energy Arcing Fault (HEAF) Events, NEA/CSNI/R(2015)10
- Insights from operating experience with HEAF events http://www.oecd-
- Literature study on methods for nea.org/nsd/docs/2013/csni-r2015-10.pdf predicting HEAF consequences 23 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Example of Recent Electrical Enclosure HEAF Experience SONGS, 2001 San Onofre; 2001 Onagawa; 2011 24 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Recent Bus Duct HEAF Experience & Testing TESTING Diablo Canyon Bus Duct (OpE) Columbia Bus Duct (OpE) 2000 2009 Zion Bus Duct (testing) 2016 25 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Test Instrumentation Preliminary Evaluation
- Preliminary evaluation to understand the most practical data collection methods
- Solar tower for quick timeframe/high energy exposures
- Blast testing for pressure https://www.nrc.gov/docs/ML1803/ML18036A448.pdf 26 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I Testing 27 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I HEAF Testing Report
- Report on the Testing Phase (2014-2016) of the High Energy Arcing Fault Events (HEAF)
Project: Experimental Results from the International Energy Arcing Fault Research Program, https://www.oecd-NEA/CSNI/R(2017)7 nea.org/nsd/docs/2017/csni-r2017-7.pdf 28 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I Parameter Overview 29 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I HEAF Testing
- 26 full-scale experiments carried out at KEMA high energy test facility between 2014-2016.
30 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I HEAF Testing Test #3: 480 V, 35 kA, 8 seconds Copper Bus Bars 31 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I HEAF Testing Test #15: 10 kV, 15 kA, 3 seconds Oil-filled breaker (oil removed), copper bus bars 32 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I HEAF Testing Test #23: 480 V, 40 kA, 7 seconds Aluminum bus bars 33 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I HEAF Testing Test #26: 4.16 kV, 26 kA, 3.5 seconds Bus Duct, copper bus bars, aluminum housing 34 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I HEAF Testing Results
- Material impact of aluminum
- Potentially larger ZOI
- Potentially greater likelihood of maintaining an arc at low voltages
- Higher risk of fire propagation 35 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I HEAF Testing Results Test 23 Test 26
- New Failure Mode:
Conductive Products of Combustion
- Conductive AL byproducts coated facility
- Shorted out equipment and damaged electrical circuits
- Fort Calhoun HEAF event-June 7, 2011
- Adjacent cabinets affected by HEAF byproducts 36 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I Incident Energy Summary Low Voltage Incident Energy @ 3ft Low Voltage Incident Energy @ 3ft Calculated (IEEE) Measured via Slug Cal 3.500 1800 3.000 1600 1400 2.500 1200 2.000 kJ/cm2 1000 kJ/m2 800 1.500 600 1.000 400 0.500 200 0.000 0 5 6 7 25 1 2 23 3 5 6 7 25 1 2 23 3 Test # Test #
Medium Voltage Incident Energy @ 3 ft Medium Voltage Incident Energy @ 3 ft Calculated (IEEE) Measured via Slug Cal 0.400 3000 0.350 Instrument Damaged 2500 0.300 2000 0.250 kJ/cm2 0.200 kJ/m2 1500 0.150 1000 0.100 500 37 0.050 0.000 0 26 17 18 11 10 8 12 20 19 9 13 14 22 21 16 15 26 17 18 11 10 8 12 20 19 9 13 14 22 21 16 15 Test #
Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018 Test #
HEAF Shields
- Use of shielding to prevent damage to targets from a HEAF event
- objective is to prevent damage to risk-significant targets beyond the faulted component and/or damage and ignition in overhead cable trays
- Acceptability of HEAF shields should consider:
- What is the design basis?
- What is the acceptance/rating/qualification test method?
- How does the installed HEAF shield match what was tested?
- Why should this engineered feature be treated any different than fire barriers (walls/floors), fire doors/dampers, electrical raceway fire barrier systems, penetration seals, etc?
38 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Solid Tops / Louvers Misconceptions:
- Solid tops on switchgears always contain the HEAF and prevent damage to targets above
- Energy will only travel in direction of the vents and will prevent significant energy/mechanical damage targets located above or away from the vent path 39 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Generic Issue, IN, PIRT, & NUREG-IA 40 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Pre-GI-018 Process Overview 41 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Aluminum HEAF Generic Issue
- Generic Issues Program Pre-GI-018
- The NRC has performed a screening review as part of the GI process related to HEAF events involving aluminum components
- The generic issue review panel (GIRP) determined that the seven screening criteria were met in accordance with management directive 6.4 (ML14245A048) and is in the process of finalization and release of the screening phase document
- The staff has recommended a two phase approach to address the generic issue and identified both short term and long term actions
- GIRP memo issued (ML16349A027)
- Moving into next phase of Generic Issue Program 42 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Information Notice (IN) 2017-04
- High Energy Arc Faults in Electrical Equipment Containing Aluminum Components
- OECD/NEA international test program insights
- 6 U.S. operating experience events involving aluminum components Plant Date Fort Calhoun June 7, 2011 Columbia August 5, 2009 Diablo Canyon May 15, 2000 Zion April 3, 1994 Shearon Harris October 9, 1989 Kewaunee July 10, 1987
- Issued August 21, 2017 43 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
HEAF PIRT
- International Phenomena Identification and Ranking Table (PIRT) exercise held in February 2017
- Insights:
- Aluminum oxidation and byproducts
- Pressure effects
- Target characterization and sensitivity
- Mitigating factors (HEAF shields) https://www.nrc.gov/reading-rm/doc-
- Selected for ACRS quality collections/nuregs/staff/sr2218/
review 44 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
International Agreement Report
- NUREG/IA-0470 Volume 1 Nuclear Regulatory Authority Experimental Program to Characterize and Understand High Energy Arcing Fault (HEAF) Phenomena
- International Partnership with Japanese Regulator
- Secretariat of Nuclear Regulation Authority S/NRA/R https://www.nrc.gov/reading-rm/doc-collections/nuregs/agreement/ia0470/
- Volume 2 and 3 are currently in draft 45 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase II Measurement Improvements 46 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Measurement Techniques Lessons Learned Phase I to II
- Temperature, Heat Flux, Incident Energy
- Heat Release Rate (HRR)
- Pressure
- Video (visible spectrum and infrared)
Temperature & Heat Flux Phase 1 Testing
- ASTM F1959 Slug Calorimeter
- Range 84-25,120 kW/m2
- Primarily used to determining the Arc Rating of Materials for Clothing
- Water-Cooled Schmidt-Boelter / Gardon Gages
- Not used in phase I due to limited dynamic range, logistics, and safety 10 kW/m2 to 200 kW/m2 uncertainty of 6 kW/m2 (coverage factor of 2, 95
% confidence interval)
- Possible collaboration with ATF for Phase II
- Plate Thermometer
- Quick Response vs. Robustness
- 63 % of the incident heat flux in approximately 0.7 s. 48 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Measurement Limitations
- Melting point of Inconel approximately 2,400°F (1,325°C)
- No measurement locations beyond 3 ft.
49 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase 2 Improvements Tungsten Slug Calorimeter Structural
- 1 in. diameter tungsten slug Insulation Insulation Board
- Durable for high incident heat fluxes in direct arc plume Thermocouple environment q total 25 mm T (oC)
- Calcium silicate insulating board Metal Cylinder 50 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Heat Release Rate
- Phase 1
- Useful for enduring fire
- Not practical for initial energy release
- Timely setup and removal
- Alternative approaches investigated by JNRA (i.e. rocket fuel testing)
HRR_O2 (kW) 900 800 700 600 HRR (kW) 500 400 300 200 100 0
0 500 1000 1500 2000
-100 Time (s) 51 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Pressure: Active/Passive
- Bikini gauges
- Molten ejecta pierced the sheets
- Pencil probes
- Unreliable data
- In-cabinet measurement
- Strain gauge-type pressure transducers
- Heavily compromised by EMI 52 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase I Pressure EMI
- EMI tends to be most severe during large changes in current, voltage, and arc activity, and these are the same periods where large changes are expected in enclosure pressure
- Positive and negative pressure peaks occur at the onset of the arc, and are of similar magnitude
- New techniques have been developed for Phase II of testing 53 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Pressure Phase II
- Strain-gauge type sensor used in
- Quartz type gauge
- Test Cell 7
- Test Cell 9-Fiber Optic Cabling
- Dynisco Pressure Transducer PT150-50
IR Camera Capabilities
- Non-intrusive temperature measurement
- Visualization through smoke
- Speed - rapid event
- Dynamic temperature range
- Ambient to >2000 oC
- Compromises
- Resolution - field of view
- Speed
- Temperature Range (Dynamic Range) 55 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
IR Camera Improvements
- Phase I Camera Capabilities
- High Speed Recording
- Limited Temperature Range
- High Resolution
- Phase II Camera Capabilities
- Lower Speed Recording
- Greater Temperature Range
- High Resolution 56 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Infrared Videography
- Test #21
- 7.2 kV
- 29 kA
- 2.5 seconds
- Copper bus bars
- Vertical arc orientation 57 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Data Acquisition
- Phase II of testing will use an isolated data acquisition system with an independent power supply
- Redundant systems will be available for possible arc shorting consequences
Small-Scale Research Program 59 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Small-Scale Testing
- Purpose
- Characterize HEAF event particle characteristics near arc
- Objectives
- Investigate particle parameters
- Density, conductivity, size distribution, rates of production, composition, trajectory, morphology
- Evaluate physical parameters
- Temperature, heat flux, HRR, mass loss
- Measure particle emission characteristics and electrical data 60 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Small-Scale Approach
- Sandia National Laboratories lightning simulator
- Single phase arcing between two vertical bus bars
- Variables
- Voltage, current, duration, material
- Particle collection and post-test analysis
Small-Scale Apparatus 62 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Small-Scale Measurements
- Videography
- High-speed infrared (IR) imaging
- Trajectory
- Particle collection
- Aerogel plates (99.999% SiO2)
- Carbon tape
- Particle Analysis
- Energy dispersive x-ray analysis (EDXA)
- Electron energy loss spectroscopy (EELS)
- Scanning electron microscopy (SEM)
- Raman spectroscopy
- X-ray photoelectron spectroscopy 63 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Small-Scale High-Speed Video 64 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Small-Scale Test Plan Public Comment
- Federal Register Notice (FRN)
- www.regulations.gov
- Docket ID: NRC-2018-0040
- Aluminum High Energy Arc Fault (HEAF) Particle Size Characterization Test Plan - DRAFT
- ADAMS ML18163A423 https://www.nrc.gov/docs/ML1803/ML18036A448.pdf 65 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Public Workshop 66 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
April 2018 HEAF Public Workshop
- Held April 18th and 19th at NRC headquarters
- Objectives
- Update stakeholders on the status of the research program and GI process
- Discuss and resolve comments received on the draft test plan
- Solicit input on equipment and test parameters 67 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Needs and Objectives
- Document to support April Workshop
- General discussion of hazard and identification of information needed to inform testing to be https://adamswebsearch2.nrc.gov/w realistic ebSearch2/view?AccessionNumber
=ML18081B300 68 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Public Workshop Outcome
- Outcomes
- Further revisions of the draft test plan to reflect realistic configurations
- Better alignment with stakeholders on the objectives and methods of the research program
- Results, transcripts, and presentations to be preserved in a NUREG/CP 69 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase II Parameters and Testing 70 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Test Parameters
- Arcing Time (Duration)
- Electrical protection clearing times for primary and secondary protection
- Worst case bolted fault conditions may not produce bounding incident energy
- Should also evaluate clearing times for arc conditions with limiting source
- With and without considering failure of 1st upstream circuit protection 71 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Proposed Arc Durations
- Experimental parameters
- Electrical Enclosures
- Low Voltage
- 2 and 4 seconds (select tests to 8 seconds)
- Medium Voltage
- 2 and 4* seconds
- Bus Bar Duct
- Medium Voltage
- 1, 3, 5 seconds
- Depending on power delivery system capabilities 72 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Durations Based on OpE Plant Name Date Arc Duration (seconds)
Robinson 03/2010 8 - 10 Diablo Canyon 05/2000 11 Prairie Island 08/2001 >2 San Onofre 02/2001 >2 Fort Calhoun 06/2011 42 73 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Intent of Long Durations
- Short arc flashes lack sufficient energy to cause thermal damage to other equipment
- Total energy (thermal source term) dependent on duration
- Long durations and their damage footprint are showing up in operating experience
- Arc flash vs HEAF 74 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Proposed Test Parameters
- Voltage
- Low voltage: 480Vac
- Medium voltage: 6.9kVac
- Current
- Low voltage
- 15kA, 25kA
- Medium voltage
- 25kA, 35kA*
- Depending on power delivery system capabilities 75 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Proposed Test Parameters
- Current
- Bolted fault current
- A short circuit or electrical contact between two conductors at different potentials in which the impedance or resistance between the conductors is essentially zero
- Arcing fault current
- A fault current flowing through an electrical arc plasma Ref. IEEE 1584 76 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Fault Mean Median Low voltage Current Bolted (kA) 27.3 (kA) 24.6 460-480 Vac Arcing 14.4 13.3 Sample from US plants Fault Current 480V Sample (n=18) 14 12 10 8
Count 6
4 2
0 0-10 10-20 20-30 30-40 40-50 50-60 Current (kA)
Bolted Fault Arcing Fault Test Levels: 15kA and 25kA 77 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Fault Mean Median Medium Voltage Current Bolted (kA) 31.0 (kA) 30.8 4.16kVac Arcing 29.5 29.3 Fault Current 4.16kV Sample (n=23) Sample from US plants 10 9
8 7
6 Count 5 4
3 2
1 0
0-10 10-20 20-30 30-40 40-50 50-60 60-70 Current (kA)
Bolted Fault Arcing Fault Test Levels: 25kA and 35kA 78 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Fault Mean Median Medium Voltage Current Bolted (kA) 32.8 (kA) 33.6 6.9kVac Arcing 31.2 31.9 Fault Current 6.9kV Sample (n=9) Sample from US plants 6
5 4
Count 3 2
1 0
10-20 20-30 30-40 40-50 50-60 Current (kA)
Bolted Fault Arcing Fault Test Levels: 25kA and 35kA 79 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase II Draft Test Plan
- Public Comment Period
- OECD/NEA Phase I members for comment on June 30, 2017
- Federal Register notice (82 FR 36006) published on August 2, 2017
- Public comment period closed September 1, 2017
- 64 comments received in total + 27 EPRI comments 80 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase II Testing
- Large-scale testing
- Joint NRC-OECD/NEA tests - 16 electrical enclosures & 3 bus ducts
- NRC-only tests with a focus on aluminum - 8 electrical enclosures & 5 bus ducts
- First set of tests to take place the week of September 10, 2018 at KEMA Laboratories in Chalfont, PA
- Small-scale testing
- Sandia National Laboratories is conducting small-scale tests that look at aluminum particle size, morphology, speed, and trajectories
- Small-scale data will be used to develop model to estimate the energy release from aluminum interaction during a HEAF 81 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
HEAF Phase II Test Matrix-Enclosures Enclosure Testing Copper Bus Bars Aluminum Bus Bars 480 Volt 6900 Volt 480 Volt 6900 Volt 15kA 25kA 25kA 35kA 15kA 25kA 25 kA 35 kA 2s 4s 8s 2s 4s 8s 2s 4s 4s 2s 4s 4s 2s 4s 8s 2s 4s 8s 2s 4s 4s 2s 4s 4s X X 2-1 2-2 2-3 2-4 2-5 2-6 2-7 2-8 2-9 2-10 2-11 2-12 2-13 2-14 2-15 2-16 2-17 2-18 2-19 2-20 2-21 2-22 2-23 2-24 *
- Legend OECD/NEA HEAF Phase 2 Tests US NRC Specific Spplemental Testing driven by Generic Issue Aluminum HEAF Program
- Uncommitted tests to explore unanticipated results/enhance repetition if necessary 82 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
HEAF Phase II Test Matrix- Bus Ducts Bus Duct Testing 4160 Volt /
25 kA Copper Bus Copper Bus Aluminum Bus Aluminum Bus Steel Enclosure Aluminum Enclosure Steel Enclosure Aluminum Enclosure 1s 3s 1s 3s 1s 3s 1s 3s 5s 5s 2-25 2-26 2-27 2-28 2-29 2-30 2-31 2-32 *
- Legend OECD/NEA HEAF Phase 2 Tests US NRC Specific Spplemental Testing driven by Generic Issue Aluminum HEAF Program
- Uncommitted tests to explore unanticipated results/enhance repetition if necessary 83 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
HEAF Phase II Experimental Approach
- Key to experimental approach is to quantitatively characterize the phenomenon produced by the HEAF
- Limit test variables to understand the importance of specific variables on the severity of the HEAFs
- create a dynamic model based on scenario specific factors
- Repeatable arc location and plasma ejection direction
- repeatable tests using the same enclosure configurations
- Instrumentation will be the primary means of data collection at multiple distances from the HEAF origin
- No cable trays or external combustibles will be used
- Exception- 4 in. x 4 in. Cable Coupons on instrument rack
- No testing to be performed will subject any equipment to conditions that exceed equipment ratings Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
HEAF Phase II Experimental Approach Enclosures Bus Ducts 85 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018 13
HEAF Phase II Measurement
- Temperature and Heat Flux
- Quantities will be measured at multiple distances away from the arc point
- Will aid in a dynamic ZOI creation
- Pressure (improved measurement techniques developed)
- Potential to measure impact on room pressure currently being explored
- Damage Zone
- Furthest extent of damage
- Thermal (i.e. ensuing fire damage / smoke damage)
- Physical ( i.e. thrown cabinet door, shrapnel) 86 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
HEAF Phase II Measurement
- Mass of Material Vaporized
- Measurements pre- and post-testing to validate computer models and theory equations of vaporized material
- Potential to develop approximate energy release models from classical energy conversion models
- Cable Sample Material
- Cable samples placed at varying distances away from enclosure (to be tested for damage and electrical continuity)
- Byproduct Testing
- Conductivity measurements for aluminum deposited on surfaces
- Spectroscopy
- Heat Release Rate (HRR) will not be measured during experiments based on lessons learned in phase I testing 87 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Phase II Instrumentation Thermal Capacitance Slug (Tungsten) 88 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
OECD -Phase II HEAF Expected International Members
- Belgium-
- Japan
- The Federal Agency for
- Central Research Institute of Nuclear Control (FANC)
Electric Power (CRIEPI)
- Canada-
- Japan Nuclear Regulatory Authority
- Canadian Nuclear Safety (NRA)
Commission (CNSC)
- Netherlands
- Czech Republic
- The Authority for Nuclear Safety
- State Office for Nuclear Safety and Radiation Protection (ANVS)
(SÚJB)
- Spain
- France
- Consejo de Seguridad Nuclear
- The Institut de Radioprotection (CSN) et de Sûreté Nucléaire (IRSN)
- Germany
- United States Nuclear Regulatory
- Gesellschaft für Anlagen- und Commission (USNRC)
Reaktorsicherheit (GRS) mbH
- Korea (Republic of)
- Institute of Nuclear Safety (KINS) 89 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
OECD -Phase II HEAF Possible U.S. Interagency Participation
- NRC provides presentations to the Federal Fire Working Group biannually
- ATF & NASA
- NFPA & IEEE - Small scale testing and literature search survey and HEAF definitions and classification 90 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
OECD/NEA Challenges
- International Agreement NRC Office of General Council (OGC)Review
- Article 9 Clause (b)
- shall indemnify the other Parties for any such actions, claims, costs and expenses that may involve the other Parties.
- NRC cannot indemnify other parties against unspecified liability or unspecified sums of money.
- Agreement with clause removal is currently being reviewed by potential member countries 91 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Timeline of NRC/OECD Phase II Actions
- Public Comment Period Closes................................................ September 2, 2017 (Completed)
- OECD Comment Period...................................... August 31 / September 15, 2017 (Completed)
- OECD HEAF Meeting................................................................. October 12, 2017 (Completed)
- HEAF Workshop ......................................................................... April 18-19, 2018 (Completed)
- OECD HEAF Meeting....................................................................... April 23, 2018 (Completed)
- Comment Resolution ....................................................................... May 11, 2018 (Completed)
- Final Test Plan.................................................................................. June 13, 2018
- Signed International Agreement ...............................Fall 2018 (Target)
- Equipment Delivery...................................................................................Fall 2018
- Initial Test Series............................................................... September 10-14, 2018
- Second Series of Tests (To correspond w/ International OECD Meeting)................................. Spring 2019
- Remaining Tests................................................................................... 2019/ 2020 92 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Planed HEAF II Publications
- NUREG/CP
- Workshop presentations and feedback
- NUREG/SANDIA Report
- Small-scale testing
- Update and expand literature review
- NUREG/EPRI
- Definitions/frequency refinements
- NUREG/IA (2 additional)
- International reports containing JNRA information
- Same format as HEAF Phase I
- NUREG- NRC Aluminum Series Tests
- Supplements the CSNI report
- NUREG- Data Analysis Report
- NUREG/EPRI Document
- New Methodology for HEAF and HEAF aluminum phenomena
- Dynamic ZOI (preferred)
- Reviewed by OECD/NEA group 93 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018
Questions ?
94 Office of Research Office ACRS PRA Subcommittee Briefing August 24, 2018