ML19046A392

01/23/2019 Meeting Transcript

ML19046A392
Person / Time
Issue date:	01/23/2019
From:	Office of Nuclear Regulatory Research
To:
Kenneth Hamburger 415-2022
Shared Package
ML19046A388	List: ML19046A389 ML19046A390 ML19046A391 ML19046A392 ML19046A393
References
20181317, NRC-0048
Download: ML19046A392 (170)

Text

Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSIONTitle:Informational Update on High Energy Arc Faults (HEAF) Involving AluminumDocket Number:(n/a)

Location:Rockville, Maryland Date:Wednesday, January 23, 2019Work Order No.:NRC-0048Pages 1-169 NEAL R. GROSS AND CO., INC.

Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W.

Washington, D.C. 20005 (202) 234-4433 1 UNITED STATES OF AMERICA 1 NUCLEAR REGULATORY COMMISSION 2+ + + + +3 INFORMATIONAL UPDATE ON HIGH ENERGY ARC FAULTS 4 (HEAF) INVOLVING ALUMINUM 5+ + + + +6 WEDNESDAY, 7 JANUARY 23, 2019 8+ + + + +9 ROCKVILLE, MARYLAND 10 The public meeting convened in the 11 Commissioners' Hearing Room at the Nuclear Regulatory 12 Commission, One White Flint North, 11555 Rockville 13 Pike, at 9:00 a.m., Ken Hamburger, Facilitator, 14 presiding.

15 PRESENT: 16 KEN HAMBURGER, RES/DRA/FXHAB, Facilitator 17 MICHAEL CHEOK, RES/DRA 18 TOM BOYCE, RES/DE/RGGIB 19 MICHAEL FRANOVICH, NRR/DRA 20 STANLEY GARDOCKI, RES/DE/RGGIB 21 NICK MELLY, RES/DRA/FXHAB 22 KENN MILLER, RES/DE/ICEEB/EET 23 MARK HENRY SALLEY, RES/DRA/FXHAB 24 GABE TAYLOR, RES/DRA/FXHAB 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 2 C O N T E N T S 1 Page 2 Opening Remarks.................3 3 Introduction & Meeting Objectives........6 4 Generic Issue Program Status for PRE-GI-018...6 5 Preliminary Observations of Sep 2018 Full-Scale 6 Confirmatory Tests on Aluminum HEAFs...44 7Future Test Planning and Equipment Selection..87 8 NRC/EPRI Working Group Charter.........99 9 Public Question and Comment..........116 10 Concluding Remarks..............169 11 Adjourn....................169 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) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 3 P R O C E E D I N G S 1 9:01 a.m.2 MR. HAMBURGER: Good morning, everybody.

3 It's nine o'clock, so we're going to get started.

4 My name is Ken Hamburger. I work in the 5Office of Nuclear Regulatory Research. I'm an NRC 6 Facilitator, and I'm here today in that capacity.

7 Mostly just to keep us on time and on 8schedule. And make sure everybody has a chance to say 9 what they came here to say.

10A couple of logistical notes. For 11 visitors, while you're here, please keep your visitor 12badges visible. Keep them on you. You shouldn't need 13 an escort wherever you are today.

14 This room is fine. The hallway out here 15is fine. The cafeterias are fine. You don't need an 16 escort for any of that.

17 If you need to access any of the other 18 floors, please find an NRC staff member to escort you.

19 The bathrooms are through this door, make a left.

20Women's bathrooms are on the left. Men's bathrooms 21 are on the right. Again, don't need an escort.

22 The cafeteria is also open to you if you'd 23like to grab coffee or a snack. It's out that door to 24the left. Follow the hallway down, and the cafeteria 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 4 is on the left.

1 If there is any sort of fire or emergency, 2 you have emergency exists in the rear of the room.

3 And we have the two exits in the front of the room.

4We are transcribing today's meetings. We 5have our Court Reporter in the back there. So, a 6 couple of things for those in the room.

7 Please identify yourself when speaking.

8 Use your name so he can capture that in the 9 transcript.

10Please speak into the microphones. He has 11 an audio feed, so he can only hear what's said over 12 the microphones.

13And we have webinar participants. So they 14can only hear what's said into the microphones. So 15 please don't forget.

16I'll try to remind you. Even if you're 17 right up here in the front and you're speaking to 18 somebody over here, the webinar hears silence if you 19 don't have one of the mics.

20 So, we have an aisle mic over here and 21over here. I have two wireless mics. If you have any 22 trouble getting to the aisles, you can just raise your 23 hand, I'll bring over a wireless mic to you.

24 And for those on the webinar, best way to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 5 ask a question or make a comment is to use the raise 1hands' feature on the webinar. Or use the chat 2feature and you can ask Tom. Tom is manning our 3 webinar.4 And he will either relay the question to 5 us. Or open up the phone lines. We do have a phone 6 line in the room so you can ask the question yourself.

7 And we have you muted at the moment, but 8 periodically we'll open up the phone lines. Anybody 9 has a comment or question.

10 So I suggest that you keep yourself muted 11 until you're ready to speak in case we open up the 12 phone lines.

13 We have a public comment period scheduled 14for the last two hours of today. That's open for 15 anybody to present or say anything they would like to 16 say.17 I do think we're pretty lightly loaded in 18terms of the agenda. I expect to have two hours plus 19 for that public comment period.

20 But just so that we can get through all 21 the items on the agenda and make sure that everyone 22 has a chance to speak, I may ask you to hold certain 23 comments or questions until that public comment 24 period.25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 6 Just to make sure we get through 1everything we have to get through. Again, I expect 2 more than enough time for that public comment period.

3 If anybody has any questions or issues 4 with today's meeting or with the webinar, you can 5 reach out to me, Ken Hamburger. I'll put my contact 6 information up on the board during the breaks.

7 And with that, I would like to introduce 8 our Division Director, Mr. Michael Cheok.

9MR. CHEOK: So thank you Ken. And good 10morning. Welcome. And thank you for participating in 11 this public meeting on the Pre-Generic Issue on High 12 Energy Arching Faults involving aluminum.

13Again, my name is Mike Cheok. I am the 14 Director for the Division of Risk Assessment in the 15 Office of Nuclear Regulatory Research.

16 Today we will provide a brief status of 17where we are in the generic issue. We will also talk 18 about our proposed next steps.

19 But most importantly, we would like to 20 hear your comments, questions, or feedback on the 21 process.22 As Ken said, I would like to welcome those 23on the Webinar. And I would like to say that we would 24also love to hear from you. For those of -- for those 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 7 people who are not available today, this meeting is 1 being transcribed.

2 So, today you will hear two main teams 3throughout the meeting. First, you know, we will 4follow the GI Process, the Generic Issue Process, 5 which has several well-defined steps.

6 The first step that you'll hear about is 7the screening process. And you will hear about the 8 results of that screening process.

9 We are now in the assessment phase of the 10process. And you know you will again hear about what 11 we have done so far and where we plan to go.

12We will have a methodical process. We 13 will be methodical in getting to the most realistic 14 and the most applicable offering experienced data and 15 information.

16 We will also have to be methodical in 17terms of defining our tests going forward. We will 18 have to have realistic and applicable tests to get the 19 best information on how we can characterize the HEAF 20 phenomenon.

21 We will -- we will have to do this so 22 that, you know, when they come up with our 23recommendation, the recommendation will be robust. It 24 will be supportable. And it will be defendable.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 8 Also, you know, I would like to say that 1 not all HEAFs are the same. We have to characterize 2 the different classifications or categories of HEAFs 3 correctly.

4 We have to determine the corresponding 5 frequencies of these classes of HEAFs correctly. We 6 have to model the effects of these HEAFs on plants 7 correctly.

8Also, not all plants are the same. Not 9all plants have aluminum. Plant configurations and 10 plant equipment are different.

11 Plant procedures and plant practices are 12different. So, you know, we have to get this right in 13 terms of the generic issue.

14 A single solution for all plants may not 15be applicable. We may have to have solutions that 16 maybe different for different plants.

17 So the second thing you will hear today is 18 that we will need to continue to improve on our 19communications. We will need to involve all the 20 public stakeholders throughout the process, early and 21 often.22 We plan to hold as many public meetings as 23needed. And we hope that you will show up and provide 24 us with the feedback.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 9 So with that, I would like to go ahead and 1 turn the meeting over to Mark Salley for his --

2 bringing us through the Agenda.

3MR. SALLEY: Thanks Mike. To pick up 4 where Mike left off there, this is about 5 communications today.

6 And as a part of the Generic Issue, the 7 Generic Safety Issue process, there is a requirement 8 that they have periodic updates.

9 And this fulfills the periodic update.

10But more importantly, when we talk to Tom Boyce and 11 Stan, there was additional information that we wanted 12to share today. So, we wanted to share as much 13 information as possible.

14 You know, also with communications, good 15 communications are very important. We've done a lot 16of internal briefings inside the NRC through the 17 Commissioner assistants, up through the DEDOs.

18 And it's surprising some of the questions 19 and information we hear that's floating around in the 20 industry.21 Without getting into that, we'd like to 22 take the high road here and to answer any questions 23you have. And to communicate anything we can with 24 you.25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 10 Also, with good communications is a two-1way street. We need dialog. So, at the request of 2 the stakeholders and industry, we've allowed two hours 3 this afternoon as Kenny said, for any open discussion.

4So again, we want to hear from you. So 5 today is about communication.

6 I want to give you a quick overview of 7what we're going to try to accomplish today. And then 8 we'll get right into it.

9 Again, this is part of the Generic Issue 10Program. We're doing this update. We've also 11 performed some testing in September.

12 And we'll refer to this as confirmatory 13 testing. If you remember the first phase of testing 14we did with the Internationals, the OECD NEA Program, 15 we had two tests, Test 23 and Test 26 that kind of 16 stuck out a little bit.

17 And we had some ideas and some thoughts 18that aluminum could be the issue. But we wanted to be 19-- we don't want to cycle industry.

20We want to be sure. We want to be 21 thoughtful in how we move forward. And we wanted to 22 do some confirmatory testing.

23We did some in September. We did the 24 medium voltage. It's basically broken up into three 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 11 pieces, medium voltage, low voltage, and the bus 1 ducts.2 We did four tests with the medium voltage.

3Frank Cielo is here from KEMA. If anyone has any 4 questions about the testing lab, you can grab Frank at 5 a break and talk with him.

6 But we did the confirmatory testing, and 7Nick's going to give a presentation. And we're going 8 to share some of the insights.

9There's a lot of data. And it's going to 10 take time for us to analyze the data and go through 11 it.12 But I think at this point we can share 13 some observations and let you know what we see.

14Again, it's about giving you the information and 15 sharing what we have.

16 We'll also be moving into the next series 17of tests. We hope to run the low voltage and the bus 18 ducts this summer.

19 So again, we'll talk about that. If you 20 remember the meeting back in April, we had two and a 21half days, very good discussion with everyone. We're 22 moving along with that matrix and some of the testing 23 there.24 And finally, the last thing that we'll 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 12 talk about after lunch is, NRC has an MOU with EPRI to 1 work on technical matters in a number of areas. One 2 of them is fire risk.

3 Under the fire risk we've put together a 4 working group and a charter with EPRI that we'll look 5 at the data and do some of the technical analysis.

6 We'll discuss that this afternoon.

7 So that's the basic overview of what we're 8going to have today. Again, this afternoon I want to 9 emphasize that we've got a two-hour slot for any of 10 your questions and open discussion.

11 I see that Victoria, you've got a 12presentation from NEI, which we'll put up first. And 13 Kelly, I believe you have one too that we'll have.

14 So, with that, I'll turn it back to you 15 Kenny and get started.

16MR. GARDOCKI: Well good morning. My name 17is Stan Gardocki. I work with the Generic Issues 18 Program in the Department of Research here at the NRC.

19 If you have any questions dealing with the 20 generic issue process, you can always give me a call 21 anytime. Or send me an email.

22 I want to first step back a little bit for 23 those people that weren't here at the last public 24 meeting, and introduce the Generic Issues Program and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 13 how it -- the generic issue processes through the 1 program.2 It basically has three stages, as you can 3see on the slide there. Now the first time we see the 4 GI, it comes into what they call the screening stage.

5 And we take a first cut at it to see if it 6meets the seven criteria. If it does not meet all 7 seven criteria, it doesn't proceed forward.

8 We spend as much time on this screening 9stage as necessary to make a good evaluation. Because 10 we don't want to waste, you know, resources, time and 11 personnel doing something that's not necessary.

12 So, we've revamped the program several 13years ago and came up with these seven criteria. And 14 we think it's a pretty good screen to go to the next 15 stage.16 After it goes through the screening stage, 17we go to what they call the assessment stage. And we 18 spend about one to two, maybe possibly, you know, two 19years on the issue to determine the risk and the 20 safety significance of the issue.

21 This is important for us here to determine 22 whether it goes to the next stage, call the Regulatory 23Office of Mutation Stage. Which, you know, we have 24 the applicable NRC office that's working on the issue.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 14 This particular issue would go like to 1NRR. So we don't want to waste NRR's time. So, we do 2 what they call an assessment.

3 We do the testing up front. We do the 4modeling up front. And we do what they call an 5 assessment report at the very final part of it.

6And we put it out to the public. We have 7 meetings with ACRS and we discuss what they call the 8regulatory analysis. We'll get into that a little bit 9 later on.10 And then if it passes the screen -- the 11assessment stage, it goes into a regulatory office 12 stage. And this is where the regulatory office here 13 would be like NRR.

14It develops generic communications. They 15might send out generic letters, bulletins, INs. They 16meet with the industry. Develop an action plan if 17 there's any necessary actions that industry has to 18 take.19 That process could take several years.

20 Once you get into a regulatory office, depending on 21 how quickly the industry reacts.

22 And we stay in the regulatory office stage 23 until all actions have been implemented for all plants 24affected. And that the NRC has verified that all 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 15 actions have been taken.

1 The next slide here is a generic process 2overview. It's a very encompassing slide. But it 3 captures everything from the screening to the 4 regulatory office.

5And it shows who's responsible. Who's 6working on it. And at the bottom you can see all the 7 notices and the public involvement and ACRS 8 involvement.

9So it's a pretty good slide. It gives you 10 an overview of the entire process.

11 Where we are now, is we went through the 12screening process. And we're in the assessment stage.

13 You can see there the drop downs that we put out the 14 necessary reports.

15 And during the assessment stage, you can 16see the research still has the issue. So we're still 17 in charge of the research office.

18 And we have what they call an assessment 19team working on it. These personnel are just, you 20 know, working on it.

21 And they go back and report to the generic 22issues, you know, group. And put out an assessment 23 report at the end. Okay. Next side.

24 Specifically where we are on this generic 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 16 issue is, when we first get the generic issue into the 1 NRC, we screen it out to see if it's an immediate 2safety concern. We ask the regulatory office to 3 document that.

4 And we don't want to procrastinate through 5 a process if there's an immediate safety concern in 6 the plants.

7NRR has the LIC-504. And I don't know if 8anyone's been introduced to that process. But it's 9 pretty quick.

10 You know, they can say, you now, we have 11 a problem with a valve, or a valve stem or whatever it 12 is. And the plants need to go out there and replace 13 immediately.

14 But, if the NRR comes back with immediate 15 safety concerns and says it's not an immediate safety 16 concern, we don't know how much of a safety concern it 17 is, so put it in the generic issues process.

18Do the entire process. The in depth 19development and the assessments. And see how much 20 risk and safety importance it has.

21So they've done that. You can see that 22back in March 2016. In May 2016 we put it in the 23 program.24 You can see the mark -- the ML numbers at 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 17the end there. If you ever want to go into the ADAMS, 1 and you can see all these documents.

2 We also have a generic issues' dashboard.

3It's on the public website. You can follow the issue.

4 It can show you where we are.

5 All these information documents are tabbed 6at the bottom there under documents. And then once 7 you get into regulatory office space, you can see 8 which plants are affected.

9 You know, how they're progressing. What 10actions they need to take. What steps need to be 11 taken to take the generic issue to completion. It's 12 a pretty good tool.

13All right. Once we get through this 14 screening process, you can see May 2016 we did a 15generic issues program staff screening. Which is just 16 myself and my d epartment and my supervisor and 17 division director signing off to see if there's any 18specific screening criteria not being met. I can 19 screen it out.

20 So if this only applies to one plant in 21 the United States, it doesn't meet the criteria of two 22 or more plants. It's not a generic issue.

23 So, I or my supervisor at that point 24cannot waste putting together a committee. Review 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 18 panel, they call it G-I-R-P, GIRP to waste time and 1 effort.2So it's very economical this program. We 3 take various steps to say push it forward or step 4 back, depending on the criteria.

5So, we took a look at it. And we said, we 6think it meets all seven criteria. We formed a review 7 panel.8 We did form a review panel. And you can 9 see the screen report produced by the GIRP was on July 10 2017.11 So you can see we've gone through about a 12year to get the screening report done. And now we're 13 progressing into the assessment stage.

14We issued an assessment plan. And you can 15see up there. The -- if you want to take a look at 16 that.17 It pretty much highlights what the 18 activities are going to be done during the assessment 19stage. And it gives it kind of a time line so we can 20 kind of hold people, you know, responsible to get 21 something done at a certain time.

22 If we think it's going to be protracted 23 and go beyond two, three, four or five years, then we 24might pull back and say we might take it out of the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 19 program. Put it in a study. And then come back in.

1 So that's what the assessment stage does.

2We try to put a time. And we put milestones. And we 3 try to meet those and keep people to those milestones.

4 Okay, next.

5 What you can look forward with this 6generic issue coming up. We're going to put out -- or 7 hope to put out an assessment report near the end of 8 the year.9 We should get all the tasks and milestones 10done. And if it meets the risk and safety criteria to 11 go on into the regulatory office, we think it's a 12 safety or risk important item, then we formally call 13 a transition team.

14 We don't want to just hand it over to 15another department without a smooth transition. So we 16 do a transition team.

17 We put members on the team that are on the 18GIRP review panel. And then we put people from the 19 office it's going to on the team.

20And we hand over all the information. And 21 the Office Director will send a letter to the other 22Office Directors saying, you have it. And you have 23 sufficient resources, you know, knowledge about the 24 issue.25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 20 So we just don't want to hand it over to 1-- across the fence and say it's now -- it's yours.

2 So, it's a very smooth transition.

3 Once it's transitioned to the regulatory 4 office, the regulatory offices, like I said before, 5 will determine what regulatory actions need to be 6 taken.7 It could be, like I said, INs, GLs, 8generic letters. It could be even orders issued by 9 the Commission.

10 They're looking forward to see if there's 11 anything that the industry needs to do. To look and 12see if there's any, not loopholes, but missing 13 regulatory items and regulations that need to be 14 adjusted.15 And once they've determined that, they put 16out that information to the industry. They could put 17 it out in a 54F letters to say industry, here it is.

18 And you know, respond back as appropriate.

19 And one industry has taken all the 20actions, we review them. We verify them. And then we 21 close out the generic issue.

22I think that's the last slide, right? Oh, 23 one more. Okay.

24All right. The assessment plan. This was 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 21 a -- you can find this inside the assessment plan.

1And this is a very good overview of what milestones 2 and actions are being taken during the assessment.

3 You can see which parts have actually been 4completed. And you can see what actions we plan on 5 going forward.

6 I will ask Nick to chime in a little bit 7 on what activities he foresees and adjustments to the 8 schedule.9MR. MELLY: This is Nick Melly in the 10 Office of Research. I just wanted to kind of key in 11 on some of the important things that you're seeing up 12 here in terms of the time line.

13 And in terms of some of the things that 14Stan was talking about in terms of the risk numbers 15that are used in the assessment plan. And what causes 16things to move forward. And what causes things to 17 fall out of the generic issue plan.

18 So, currently right now, we're looking up 19 there, and we're right around the first of the year 20 we're looking at the Phase Two, large scale test plan.

21 Right now we've finished four tests since September on 22 the medium voltage.

23 And we're looking to also get low voltage 24 testing done and bus duct testing done so that we can 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 22 have the information we need in order to create a zone 1 of influence model for aluminum moving forward for 2 these risk studies.

3 That's very important that we have all the 4data we need. So that we're not cycling the industry.

5 Or we're not creating overly conservative models for 6 the damage states.

7 To accomplish that, the NRC is working 8with EPRI to create the zone of influence. And we're 9 going to have discussions here about a charter that's 10in the process so that the NRC can look at all the 11 data with EPRI, and create a zone of influence that's 12 appropriate for aluminum taking into account many 13 different factors.

14 We want to look at the frequency of the 15event occurrence. Couple that with plant 16 configurations such as a unit connected design and 17 things like that, in order to create a good picture of 18 what the zone of influence is.

19 We do not want to move forward in the 20 generic issue with a conservative model giving us 21 conservative risk estimates, because it's just not 22 realistic.

23 That's where the pilot plant selection, 24 the pilot plant studies come into play. We really 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 23 need help from industry because if we use our in-house 1 modeling techniques such as the SPAR to move things 2 forward as we have done for quick assessments, it may 3 be conservative because there are a lot of large 4 assumptions that have to be made.

5 If we can leverage the plant design, the 6 plant PRA models when we move forward using pilot 7 plants, we can get a better picture of what the risk 8 is at the plants.

9 And we want to link that with moving 10 forward through the generic issue, so that we can 11 adequately assess the risk at plants.

12 It's an aggressive schedule right now.

13And we believe that we can accomplish these. We have 14 a lot of things moving in parallel in terms of the 15 testing, the data analysis with the working group.

16 We believe we are close to being able to 17 select pilots, or hope to get pilot plants in this 18 process, so we can start figuring out what needs to be 19done in order to adequately assess the risk. Which is 20 necessary for the assessment plan right now.

21MS. ANDERSON: Are we -- are we taking 22 questions?

23 MR. GARDOCKI: One last comment. Like I 24 said, the dashboard has all the information as -- in 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 24 regards to this generic issue.

1 You can always refer to that on the NRC 2public website. And in the future, we try to hear 3 periodic meetings, you can probably expect another 4 meeting right before we start the assessment report.

5 And we're going to have a public meeting 6 with ACRS if they want to have a meeting to introduce 7this. And they kind of get involved with the 8 regulatory analysis part of the assessment report.

9 So he can expect those as future 10 activities as far as generic issues program public 11 involvement. That concludes my presentation.

12 MR. HAMBURGER: Yes. So we will welcome 13 questions from those in the room first.

14MS. ANDERSON: Victoria Anderson from NEI.

15 Nick, you mentioned that one of the things you'll be 16 doing is working on making sure you have all the data 17 you need to develop an aluminum zone of influence.

18 Are you looking at developing a single 19zone of influence? Or a spectrum of ZOIs based on 20 different physical conditions?

21 MR. MELLY: A spectrum.

22 MS. ANDERSON: A spectrum?

23MR. MELLY: We're not trying to just do a 24 one worst case zone of influence approach for 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 25aluminum. We're trying to link it in more a 1 probabilistic manner so that we can look at the 2 potential at the plant in order to get a more dynamic 3 zone of influence based on important parameters such 4 as the current, the voltage, plant design.

5 We don't want one conservative zone of 6influence that would be used for the study. We're 7 really trying to get a -- to move the method forward.

8 MS. ANDERSON: Okay.

9MR. TAYLOR: And if I could -- this is 10Gabe Taylor from the NRC. If I could just add to 11 that.12 At the workshop last April we talked about 13different approaches that we could do. And I think 14 the idealized approach would be the dynamic ZOI that 15 Nick alluded to.

16It's similar to what the IEEE does for 17their arc flash calculations. So if we could get 18 there, that would be great.

19 If we can't, there's other fall back 20 options where we basically bin different categories 21 based on the hazard and the frequencies. And have a 22 ZOI for each individual bin.

23 And then, you know, the lease detail would 24be something like we have today. Which is, you know, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 26 a one size fits all, which is kind of not where we 1 want to go.

2 But what path we go hasn't been determined 3 yet.4MR. MELLY: And it's actually very 5 important to understand that when even looking at the 6 test program that we have developed. Because we are 7 testing at two seconds and four seconds.

8 And for the low voltage cases, we're going 9to be doing minimal tests at eight seconds. But we're 10 not doing -- we're doing that for various reasons just 11 so we can look at the spectrum of possible fall cases 12 in the plant.

13 We're not trying to say, the four second 14test had this damage state. So now we're going to 15 link all zones of influence to that four second test, 16 which was the worst case.

17 That is not what we're trying to do with 18 the test program.

19 MS. ANDERSON: Okay. Are we going to be 20 talking more later about the considerations that go --

21 that will go into that development of the ZOI 22 spectrum?23MR. MELLY: Right now I believe we're 24 trying to work that through the EPRI working group --

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 27 MS. ANDERSON: Okay.

1MR. MELLY: Charter. In that that is 2 essentially what the whole conversation of the charter 3 is going to be.

4 MS. ANDERSON: Right.

5 MR. MELLY: Is that -- we're going to be 6 looking at the data that we have in-house, and coming 7 up with a method to develop that.

8MS. ANDERSON: All right. If we don't get 9 into it in detail today, I think it may be useful to 10 have another public interaction on that topic. Even 11if it's outside of the generic issue process. I think 12 there may be some substantial technical input that 13 stakeholders may have.

14MR. MELLY: I agree. And we've had 15 initial discussions already that we're -- we're 16looking at the data now. And we've just developed 17 this charter.

18 We do plan on having several more 19 interactions with the public when we move a little 20 more forward in the process of developing an initial 21 method. Looking at the data.

22 We plan on sharing the working group 23 conclusions or the working group assessment with the 24 public before any finalized method is developed.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 28 MS. ANDERSON: Okay. Thank you.

1 MR. HAMBURGER: Thanks Victoria.

2MS. UHLE: Hi. This is Jennifer Uhle from 3NEI. So, my question here is talking about this 4 approach, which is great to hear.

5 However, the amount of test data you would 6 need to really develop that doesn't seem to jive well 7 with the, you know, having the, you know, four tests.

8 So, -- or then I would ask in reverse, if 9 you plan on developing these zones of influence, then 10 shouldn't that inform your method in how you develop 11 them?12 Shouldn't that inform the tests that you 13do run? So it -- you know, where are we in likelihood 14 of getting to this state?

15MR. MELLY: So to respond to that, we did 16 hold a PIRT panel to discuss the test results of the 17 Phase One testing.

18 And that was really to inform the Phase 19 Two of testing of what we would need to do in order to 20 create this spectrum of zone of influence, and what 21 data we would need.

22 That's really what went into the 23 development of the test matrix that is in the current 24 test plan.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 29 And while there have only been four tests 1 performed right now on the medium voltage at two and 2 four seconds and various currents, we're really trying 3 to isolate parameters in this test series so that we 4 can look parameter to parameter differences as we move 5 forward.6 Also, we are looking at the next series of 7 testing this summer to have four additional tests at 8low voltage, various parameters. As well as four 9 tests looking at the bus ducts.

10 Additionally, we're trying to look at when 11 we can perform decrement curve testing, which I'll get 12 into a little bit later, in order that we can have 13 enough information to supplement the operating 14 experience data that we do currently have in-house, as 15 well as a wealth of information from previous testing 16 at shorter durations using the IEEE 1584 data and 17 things like that.

18So we're trying to bring all of that 19 together to create this. But as you say, it is a 20 limited number of tests, but it is a resource issue.

21 Testing is fairly expensive, fairly slow, 22 and there's a lot that goes into the testing itself.

23 MR. SALLEY: Yeah, Jennifer, and just to 24 put the final piece on that, as a researcher you know 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 30 you can never have too much data. The more data the 1 better.2 I think we've -- from what we've worked in 3 the April workshop, we have a pretty good matrix.

4 Which we'll talk about in a little bit, as to how we 5 parse this out.

6 And at the end of this afternoon's session 7 on the working group, you'll see how that it fits into 8 the NRC's scheme with the OECD NEA and how we bring 9 that to bear too, with it.

10So, we have parts long presentation. And 11 I think it will provide you more information.

12 MR. MELLY: Yeah. And this was also one 13 of the main issues that was brought up by the GIRP 14 panel for the generic issue.

15 Is that we didn't want to come out with a 16 conservative zone of influence for aluminum before we 17did this assessment plan. So, the real discussion by 18 the members was, what test do we need to accomplish in 19 order to create a realistic zone of influence?

20 And it was decided that we needed medium 21 voltage tests, both low voltage and bus duct testing 22 if we wanted a whole picture of what was going to be 23 necessary for the working group to establish 24 something.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 31MS. UHLE: Okay. And then you're, I would 1 expect working with EPRI to plan these tests and 2 ensure that they are prototypic in their set up?

3 MR. SALLEY: Yes. Like I said, we -- in 4April we've done that. And as a matter of fact for 5 the September tests, we had members from EPRI, witness 6 the testing with us.

7 And that and we plan to do the same in the 8spring and fall. That we have the working group that 9 will be there to see the tests and get the information 10 first hand.

11MS. UHLE: Yeah.

It's not so much the 12 witnessing the test as is, you know, taking the 13 comments that are proposed and factoring them in then, 14 rather than dispositioning.

15 You know, going forward I think that would 16 have to be done to really get to the point that we 17 would, I think, all like, with the multiple zones of 18 influence.

19MR. SALLEY: Yeah. And I agree. And it's 20 one of the reasons we made all the public comments and 21 how we changed the test plan as a public document.

22 So you can -- the ML number is out there.

23You can see the effects it had from the meeting. And 24 the changes it made.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 32 You know also, one of the things is, we 1never really were looking at the decrement curve. And 2 that was something that Ken brought up in the -- in 3 the April meeting.

4 And we've expanded our test program to 5 include that for the value we hope to bring to the 6 testing. So we are listening.

7 MS. UHLE: Thanks.

8 MR. HAMBURGER: Thanks Jennifer. I know 9we have a question on the webinar. Do we have any 10 more questions from those in the room before we jump 11 over to the webinar?

12 (No response.)

13MR. HAMBURGER: Okay. And just before we 14 go to the webinar, both Nick and Stan will be here all 15 day.16 So if any questions occur to you as we 17 move forward, please you can jump in and ask them or 18 find them at one of the breaks. Yes. You may.

19 MR. AIRD: We have one question from the 20webinar from Mohammad Mustafa. He asks, did the GIRP 21 review include existing codes and standards?

22 For example, National Electric Code, 23IEEE's et cetera, related to aluminum? Are they 24 inadequate?

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 33MR. SALLEY: I'll start the answer and 1I'll let me guys finish. You know, one of the things 2 that we did when we were doing internal briefings, 3 when Fred Brown was the DEDO and we first started 4 going through the internal briefings with this, you 5 know, Fred made the comment to us that the high energy 6 arc faults are not unique to nuclear power plants.

7 You know, this is an electrical 8phenomenon, you know, it's not unique. Now granted in 9 a nuclear power plant, you're all familiar we do move 10a lot of electricity around. We've got a lot of 11 things happening.

12 So, it may amplify a little bit, or we've 13picked up on it. But, one of the things again with 14 Fred, was to reach out and to see that very question.

15 If you remember the April workshop, we had 16 the chief electrical engineer from the NFPA come down.

17 And of course in the common vernacular in the industry 18 is arc flash.

19 You know, you'll hear this referred to a 20 lot as arc flash and arc flash events. So, we are 21 working with the NFPA.

22 And again, I think too, with the test lab, 23 the different scenes that we're taking it a little 24 further then the arc flash events that commonly occur.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 34 And you see them here.

1We see them in the Metro in D.C. We've 2lost it. The Atlanta airport has had the one a couple 3 of years ago.

4Around Christmas you just saw it in New 5 York City with the blue in the sky again. So, these 6 events are happening out there.

7 We're going to talk a little bit about the 8 differences in how we bin them in '16 with that. So 9 that will be part of the discussion with looking 10 between an arc flash, an arc blast, and a HEAF.

11 And again, working with the NFPA, one of 12 the things Ken Miller and us have worked on, is 13definitions. To clearly define which of the different 14 phenomenon.

15 So, that's kind of where we're at.

16 Anything you guys want to add?

17 MR. TAYLOR: This is Gabe Taylor, Office 18 of Research. So, I was on the GIRP panel.

19 And the information that we looked at was 20 the operating experience that we communicated in the 21Information Notice 2017-04. As well as other 22 international operating experience that was documented 23 in the OECD reports.

24 In addition to that, Nick mentioned, we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 35 looked at the SPAR models, SPAR all hazards model to 1 see what the potential risk increase could be.

2 Understanding that they are somewhat course, and not 3 as detailed as the plant's PRA models.

4 We also looked into the NFP 805 transition 5 plants, into their SEs and information available there 6 to see what the potential HEAF contributions were.

7 And that's not -- that's using the 8 existing models, so it wouldn't include the 9 increasing, if there is an increase for the aluminum 10 type HEAF.

11 So that was kind of the core information 12that we looked at to evaluate the HEAF and the 13 screening, the aluminum HEAF and the screening phase.

14 As far as the NEC, that's more of a 15 utility specific, whether they follow that code or 16 not. And I really can't make any comment to whether 17 it's adequate.

18MR. MELLY: This is Nick Melly in the 19Office of Research. Yeah, as far as the NEC codes and 20 the IEEE, they did a lot of the wealth of their data 21 collection on copper.

22 They haven't specifically looked at 23 aluminum. They do have some theoretical comparisons 24 between copper and aluminum.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 36 But their main focus is on personnel 1protection. Whereas we are looking a little beyond 2 that for extended duration faults.

3 They are following closely what we're 4doing in terms of definition space. And they're --

5 we're going to be looking at their data in a little 6 bit more detail as we move forward.

7 But in terms of saying whether their 8 regulations are inadequate based on what we've seen, 9 we can't really make that determination.

10MR. SALLEY: And I guess a final thing on 11the aluminum, Frank Cielo is here. Frank, if I could 12 ask on you, you do testing every day.

13 Is there any insights or anything you can 14 provide to that?

15MR. CIELO: Yeah, this is Frank Cielo from 16KEMA labs. I was just talking to my colleague Alex 17 Feldman.18 We do arc flash testing on a regular basis 19 for lots of manufacturers. And it is absolutely for 20 personnel protection.

21 And we've done your testing of course.

22 And we've done high energy arch fault testing for some 23 other customers.

24 It's just completely -- the consequences 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 37 are just -- they're completely different. In orders 1 of magnitude different.

2 So, and one of the things you may want to 3 think about is showing somebody a video of a regular 4 arc fault test for, you know, to comply with the IEEE 5standard versus the HEAF. It's just like, pfft. I 6mean, it's so obvious, the amount of destruction. The 7 amount -- the impact on our laboratory, the physical 8 impact on our laboratory and what actually happens, 9 it's just -- it's just absolutely incredible.

10 And you know, we test for every -- we're 11an independent test lab, all right. We're ISO/IEC 12 17025 certified.

13 We have no affiliation with any 14manufacturer, anybody. And we test for every 15 manufacturer basically on the planet between our three 16 laboratories.

17 And most of the arc fault testing, we do 18in the U.S. against the IEEE standard. And you really 19 have to see it to kind of appreciate what this is.

20 Because the first time we did this for the 21 NRC, we weren't sure we were going to continue, 22 because of just the amount of destruction that we 23experienced at our lab. And the damage that we 24 experienced at our lab.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 38But it's important testing. And it needs 1to be done. And we, you know, we just continue to 2 work through it. Hope that helps.

3MR. GARDOCKI: And this is Stan Gardocki.

4 Back to your original question about the generic 5 issues program involvement with industry standards.

6 We do look at the NRC endorses the 7industry standards to the regulations. So we'll find 8 out if there's any adjustments that need to be made in 9 the industry standard.

10 And then they have to go back and re-11 endorse it for the regulations.

12 MR. HAMBURGER: Before we get to you, we 13 have a follow up question from the webinar.

14MR. AIRD: Yeah. We have one follow up 15question. It goes, is this an electrical circuit 16 protection issue? Or an aluminum conductor issue?

17 There are fast acting devices to preclude 18 arc flash and protection schemes.

19MR. TAYLOR: So at this stage we're trying 20to understand the hazard. We are aware that there are 21 some protection schemes out there that are available 22commercially that could be use to fastly open up a 23 protective device and clear the fault and the cycle, 24 a few cycles.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 39So, we're aware of that. And you know, 1 once we understand what the hazard is, and the 2 frequence is, and understand what -- where the risk 3 falls out in the delta risk, then we'll be able to 4 have a clearer picture of where it goes moving 5 forward.6 And if it does make it into the regulatory 7office limitation stage. At that point there would 8likely be a lot of involvement between the NRC and 9 industry in understanding the ways to protect the 10 plant from the type of hazard, whether that be 11 prevention, protection, or mitigation features.

12 So, we are aware of those technologies.

13 But as far as this program, we're not trying to 14 evaluate the adequacy of those feature devices.

15 And then regarding the comment on 16 aluminum, you know, we did see the two tests in the 17 first phase that showed some much larger damage areas.

18 There were a few aluminum tests that didn't create 19 any, you know, additional hazard, then what we're 20 seeing in the copper.

21 They actually didn't even hold the fault 22in. So, not all aluminum type events may lead to some 23 type of larger damage footprint then what we're seeing 24 form the copper.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 40 But from the September tests, when we ran 1 those, the four tests that we ran, and Nick will talk 2 about those in the next presentation, all four -- they 3were medium voltage. But all four did last for the 4 entire duration.

5 And we did see, you know, a much larger 6damage footprint. And more so the energy release, 7 because the energy is related to the duration of the 8 arc.9And when you're running them for two to 10 four seconds, there is a difference between that.

11MR. MILLER: This is Ken Miller, 12Electrical Research. As Gabe said, we've about 13 recognized that protection schemes in fact are 14 supposed to clear these kinds of faults within very 15 short periods of time.

16We understand that. And but of course 17 operating experiences has shown that sometimes that 18doesn't occur. There's failures that occur. And 19 these kinds of HEAFs, you know, occur as a result of 20 that.21 And again, the important point here is to 22 study and understand how destructive those can be as 23 far as effecting the zone of influence.

24 So, again, we understand that protection 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 41normally would take case of something like this. But, 1 we're seeing cases where it doesn't.

2MR. MELLY: Yeah. And to the second part 3 of your question as well, this is Nick Melly from the 4 Office of Research.

5 In terms of aluminum, we're also not 6 saying at this moment in time that the -- if you have 7 aluminum as part of your conductive pathway that it's 8 more prevalent to have a higher G arching fault.

9 In terms of frequency, we're not seeing 10that from the data yet.

It is an issue that we're 11 going to be looking at when we take a deeper dive into 12 the data.13 But the frequency of occurrence from 14 aluminum versus copper, right now we're not seeing as 15 a change.16MR. HAMBURGER: Okay. Let's take the next 17 question from the room, please.

18MR. FLEISCHER: This is Ken Fleischer from 19EPRI. Not so much a question, but I did want to share 20 some OE of one major U.S. manufacturer of switch gear 21 in regards to type testing of copper bus bar switch 22 gear versus aluminum bus bar switch gear back in the 23'70s.24 Now again, to reemphasize that these were 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 42 two second tests, because the IEEE standard requires 1switch gear to be rated to two seconds. They noticed 2 no discernable difference between the two tests.

3 The only thing that the test did to 4 demonstrate was that the aluminum bus bar would have 5 to be a little larger than the copper bus bar with a 6little bit of different spacing. But they did not see 7 any of the extra aluminum byproduct from those two 8 second tests.

9 We are also working to contact a few more 10additional manufacturers. But we did get one that 11 actually told us they did the type tests with 12 aluminum.13 MR. HAMBURGER: Thank you Ken.

14MS. VOELSING: Hi, this is Kelli Voelsing, 15also from EPRI. So, I think we need to clarify so 16that we're not mixing an apples and oranges 17 discussion.

18 The reason that Ken shared that 19 information that we received from a manufacturer 20 regarding the difference between aluminum and copper 21 type testing, is because from the comments from the 22 gentleman from KEMA, we assumed when he said the 23 testing is completely different, that he was referring 24 to testing between copper and aluminum.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 43 In offline discussions, he clarified that 1 he was talking about the difference between flash 2 testing and arc fault testing. So the duration of 3 testing.4 So I did want to clarify that apparently 5 he was intending to communicate a difference in the 6duration of tests. Not a difference between copper 7 and aluminum, which is the generic issue that we're 8 supposed to be talking about here today.

9 And I'd also like to clarify that I think 10 we have to be careful about referring to the damage 11 footprint. Because we haven't necessarily validated 12 any damage.

13 What we've validated is, you know, 14 byproducts being deposited at a certain radius. Not 15 damage.16MR. TAYLOR: Thank you Kelli. And that's 17a -- that's a good point. I was searching for the 18 right words.

19 And it's really the observations from what 20 we saw between the copper and the aluminum, where at 21 the two aluminum tests, well, six aluminum tests now 22 showed the larger area.

23MR. CIELO: Yeah. This is Frank Cielo.

24Thank you for clarifying that.

I didn't intend to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 44 cause any confusion there.

1 It was the difference between the regular 2 arc fault testing and this testing. Thank you for 3 clarifying that.

4 MR. HAMBURGER: Okay. Any more comments 5 from anyone in the room?

6 (No response.)

7 MR. HAMBURGER: Tom, do we have anything 8 else on the webinar?

9 (No response.)

10MR. HAMBURGER: Okay. We are ahead of 11schedule. So Nick, if you're ready to go, we'll talk 12 about those preliminary observations.

13MR. MELLY: All right, next slide. Again, 14 this is Nick Melly from the Office of Research. I'm 15 going to be discussing the testing that was performed 16 in September.

17 Kind of a big picture look at some of the 18 observations, some of the preliminary insights and the 19 project goals.

20 I'm also going to be talking about some of 21 the challenges, the measurement devices that we're 22 using, and the general observations.

23 I also want to discuss the future testing 24 parameters and schedules and some important milestones 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 45 as we move forward. Next slide.

1 So to start this all off, we did share the 2 draft test plan to our OECD NEA members, as this 3 initially started as an international project on June 4 30 of 2017.

5 We also put this out for public comment in 6 the Federal Register in August for a month long public 7comment period. That closed on September 1, 2017.

8 Next slide.

9 As part of that initial test plan, we did 10receive five comments from NEI. And we had 32 11 comments from our OECD members.

12 Since then we have received numerous 13 comments both from NEI, EPRI, as well as some of the 14industry itself. Which we have discussed at the 15 public workshop that we held in April 2018.

16 And we've addressed all of these comments 17and dispositioned them. Formally put that into our 18 public document system, ADAMS, essentially addressing 19 all comments and how the NRC planned to address them.

20 Whether they were going to be put into and 21incorporated into the test plan. And if any physical 22 changes were made to the test plan, both parameter and 23testing method. We have the ML link that is up there.

24 So right here on the screen you're seeing 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 46 the Phase Two HEAF test program. This is the matrix 1 that we discussed earlier.

2 This came about after the PIRT panel 3 phenomenon ranking table -- parameter identification 4 and ranking table exercise that we held in January 5 2017.6 This was our best guess of, these are the 7 amounts of tests that need to be conducted in order to 8 answer the question and get the data that we need.

9 What you're seeing on the screen here is 10four box tests. Test 2-19, 2-21, 2-22, and 2-24.

11 These are the tests that were performed in September 12 that I'm going to be discussing today.

13 These were all medium voltage, 6.9 kV 14 tests at 25 kA and 35 kA for two seconds and four 15 seconds respectively.

16 Another aspect that you're going to see on 17 the slide here, you see the two tests that are on the 18far right-hand side with Xs. These were identified 19 that we may need to do additional testing.

20 And we wanted to be budget ourselves to be 21 able to perform tests as things came up.

22 MR. TAYLOR: Nick, just to clarify. This 23is the -- showing 35 kA for the 6.9 kV test. We 24 actually tested at 32 kA.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 47 So this was from the original test plan.

1 MR. MELLY: And that is because the test 2 facility itself, there are some limitations on the 3capabilities of the generator. So we actually had to 4 reduce the current in order to achieve the duration of 5 testing that we wanted to achieve.

6 The two tests that I was discussing 7 earlier, the SPAR tests on the right-hand side, we 8 have allocated those for the decrement curve testing 9 that was a direct result of the comments from EPRI.

10 These are going to be tests where we're 11 actually going to be simulating actual operating 12 experience event at a unit connect design plant where 13 the generator essentially fed the fault.

14 This is not a trivial exercise. We have 15 a separate contracting action with the laboratory in 16 order to develop the capabilities needed to do this 17 test.18KEMA has never done a test like this 19before. So, getting their generators ready to 20simulate this, is a separate effort. Which is why we 21 did not do them in September.

22 And we're currently pushing that forward 23 right now, working with EPRI to establish what needs 24 to happen. And the KEMA test engineers.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 48 Another thing that you'll notice on this 1 test matrix is that there are some tests in blue and 2some tests in orange. We're doing this as part of an 3 international program.

4 The tests in blue represent tests that are 5 going to be funded through the OECD NEA test program.

6 And largely these are copper tests versus the ones in 7 orange, which are the aluminum tests.

8 We're performing the orange tests first, 9 because they're associ ated with the need of the 10generic issue program itself. So we are prioritizing 11 the generic issue and performing the test that you see 12 in orange before we do any OECD NEA tests.

13Next. I'll take questions during the 14 presentation. So, sure.

15MR. HAMBURGER: Sure. This is a long 16 presentation. So if you --

17 MS. UHLE: Okay. Thanks. Jennifer Uhle 18from NEI. So, I want to go back to a question that I 19 previously asked.

20 And I think in your discussions about this 21 test -- particular test series -- excuse me, I have a 22 cold. You indicated that the industry comments, the 23 EPRI comments, you dispositioned.

24 And this is the point about that we really 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 49 want to make or get across so that it's clear to 1 everybody. The industry comments that were given to 2 NRC, they were, I agree, dispositioned.

3 But that does not mean addressed in a 4substantial manner. There were numerous comments that 5 the industry felt essentially affected the proto-6 typicality of the tests that were

-- we got the 7 answer, we'll deal with this in a next test series.

8 And so, we've been saying that industry 9has -- comments have been addressed. But, that's not 10 our opinion with regard to the test series.

11 And that's why when we want -- we want to 12 highlight the importance of truly listening to the 13comments. And not dispositioning, meaning, I heard 14you. I don't agree with you. And I'm not changing 15 the test set up.

16 MR. MELLY: Do you have any specifics in 17 terms of which comments were not addressed?

18 MS. UHLE: Yeah. And I'm going to --

19MR. HAMBURGER: Well, let's -- if we're 20 going to discuss those, I suggest we wait for the 21public comment period. Because that might be a 22 detailed discussion.

23 And so it's --

24MS. UHLE: Well, I just -- if you don't 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 50 mind, this can be very quick if we just -- and then 1 maybe we discuss it later.

2 MR. HAMBURGER: Okay. Sure.

3 MR. FLEISCHER: Ken Fleischer from EPRI.

4 One particular comment that actually we also 5verbalized during the April workshop, is we still 6 didn't really see the benefit or -- really the benefit 7 for the low voltage eight second test.

8 Unless that test could be justified by 9some means. And I know that it was originally 10 mentioned at the April workshop that it was probably 11going to be discontinued and go back to the four 12 seconds.13 But we see that it's still considered as 14 eight seconds. I don't know if that ever really got 15 finally dispositioned as one example.

16MR. MELLY: So this has been 17 dispositioned. And the final test plan has not been 18 released, because it is part of the OECD agreement.

19 However, you still see the eight second 20test as part of the low voltage. We did reduce the 21 number of eight second tests.

22 However, we've added to the test plan that 23 the eight second tests are not going to be used to 24 inform the eight second or the low voltage zone of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 51 influence spectrum.

1 However, we are trying to still do an 2 eight second test only in term -- so we can draw 3 comparisons between the low voltage and medium voltage 4tests where we might see a medium voltage case hold 5 for longer than four seconds.

6 So the eight voltage second test is kind 7 of just a place holder, because we are at the limits 8of the capabilities of the KEMA facility. So we 9 cannot do an eight second test at KEMA for a medium 10 voltage case.

11 MR. FLEISCHER: That's interesting.

12MR. MELLY: However, we're trying to draw 13 some comparison of what we see from the low voltage to 14 what we can maybe expect to see in a medium voltage 15 case where it does last for potentially longer then 16 four seconds.

17 So, we have made that distinction in the 18 test plan which we plan on releasing very shortly.

19 That is the only reason that you see that eight 20 seconds on the low voltage case currently.

21 MS. VOELSING: Kelli Voelsing from EPRI.

22 And Ken can probably speak to this a lot more 23 eloquently.

24 But, I'm not -- we see in the OE very 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 52 different situations that lead to potentially high 1 energy arc faults in low voltage versus medium voltage 2 equipment. So very different set of conditions that 3 lead to those.

4 And so I'm not sure that there's a solid 5 technical basis for extrapolating the results of an 6 eight second low voltage test and saying that that's 7 somehow applicable or representative to what you might 8 see for medium voltage if the facility had the 9 capability to do that.

10 And I would say that the generator voltage 11to k curve is another example of a comment. And I 12 realize, you know, you're planning to address that in 13 the future.

14 But, it's not going to be particularly 15 relevant to the low voltage testing that's up coming.

16 And so, is there a time line when those two reserve 17 testing places are planned?

18 Because it would seem that the results of 19 those tests might be critical to the resolution of 20this issue. And I'm wondering how those are going to 21 happen before the planned time line for the first 22 presentation of resolving that by December.

23MR. TAYLOR: Okay, Gabe Taylor from 24Research. So, I'll touch on the last part of your 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 53 comment first.

1 The two orange boxes on the far right, the 2 optional tests, as Nick mentioned, we're going to save 3 those for the decrement curve evaluation. And later 4 in the presentation we're going to talk about the 5 decrement curve.

6 One thing that -- that's the type of 7 experiment that hasn't been done at KEMA, trying to 8 match that type of decay response from the generators.

9 Typically they use super excitation to maintain a 10 constant power source.

11 So because of that, there's a lot involved 12 on their end analytically. And also validating that 13 they'll be able to perform that type of testing.

14 So because it's not in our current 15 contract, and there's the work involved that it 16 requires us to go into our contracting process to make 17changes that are new. A new contract to get the work 18 done.19 So, through that process, it's not a --

20 it's going to take some time to get that in place.

21 MS. VOELSING: I think the complexity is 22 understood, I'm just asking about the time line, with 23 respect to the resolution of the GI that you propose 24 by the end of December.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 54 MR. TAYLOR: Yes, I apologize for the long-1 winded explanation, I was trying to provide as much 2detail as possible. So, when we get that process 3 completed, we'll be able to perform the testing.

4 So, because where we're at in the 5 contracting process phase, I don't have an exact time 6frame for you. But if I were to guess, it would be 7 sometime later this year.

8 MS. VOELSING: Do we want to talk about the 9 extrapolation of eight-second?

10 MR. TAYLOR: Yes, that was the next place 11that I was going to pick up. So, you remember from 12 our April workshop, we talked about some small-scale 13 testing and modeling. And over time, that has -- we 14 have changed directions and the process that we're 15 going through, because of some subject matter expert 16 changes.17So, currently, the plan is to work with 18 Sandia, using their area model, and basically, be able 19to develop that to run simulations to estimate 20 different physical parameters, such as heat fluxes, 21 pressures, particle ejecta, that sort of thing.

22 And by being able to perform these small-23 scale, as well as medium-scale, tests, we could have 24 models and simulations that could -- basically, the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 55 data would validate the models and then, you could use 1 the simulations to develop ZOIs or that sort of type 2 of information that would be useful in informing the 3eventual model that's used in the PRA risk 4 assessments.

5 MS. VOELSING: So, obviously, modeling and 6 simulation is always a valuable part of the process, 7 but developing models and validating them based on an 8 extrapolation, I'm not sure that that provides solid 9 technical basis for the models.

10 MR. TAYLOR: They actually already have the 11 model in use out there, and I believe EPRI is actually 12 using it for some of their DC work.

13 So, what we really need to do is provide 14 some changes to the sub-model, to accurately represent 15what we're trying to characterize. So, it's a multi-16 pronged approach, to try to understand this hazard.

17 We see some value in performing the small-18scale work, but also, regardless of that pronged 19 approach, there's also some value in having more than 20 two points and you're trying to extrapolate for low 21 and medium voltage.

22 So, while we did reduce the number of 23 eight-second tests that were originally planned in the 24 small-scale test, or in the full-scale test plan, we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 56 still see value in performing that type of duration.

1 And in addition, we're seeing some 2 operating experience that shows longer duration low 3voltage OpE. However, I think we are in agreement 4 that the frequency of low voltage versus medium 5 voltage and the hazard that it contributes are 6 different between the two.

7 MR. HAMBURGER: If we have more questions 8 about the small-scale testing or about the resolution 9 of specific technical comments, let's hold those and 10 let Nick --

11 MR. MELLY: I'd like to answer one thing.

12 The extrapolation is important, we think, because we 13just simply don't have the ability to do an eight-14 second test at the KEMA facility.

15 We are trying to match, we would like a 16 data point, so that we can develop this spectrum type 17 approach to the zone of influence models.

18We did see some OpE, such as Robinson, 19 where it was between an eight-second to 11-second arc, 20 in medium voltage. We do not have the capability at 21 the KEMA facility to replicate this event.

22 So, we are looking at how we can make 23 comparisons between the low voltage energy release 24 versus the medium voltage energy release, in terms of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 57 duration. We do see that as a key important factor.

1 The PIRT documented duration as one of the 2 most important factors for the high energy arcing 3fault and we're, again, seeing that from testing. So, 4 we do see limited number of low voltage tests as an 5 extrapolation point, an important test parameter.

6 Okay, moving forward.

7 I want to discuss some of the challenges 8 that we saw with the Phase 1 testing, and some the 9 measurement limitations that we're running into. We 10 had been testing with plate thermocouples, essentially 11using an Inconel plate to measure the heat flux and 12 the temperature.

13 And you can see the picture on the left, 14 the before image, versus the picture on the right, 15 after it was subjected to Test 23, where we completely 16 melted the Inconel plate and were not able to record 17 any data at the three-foot location.

18 This was a limitation that we had not 19 anticipated and we were not able to get any data from 20 that test, because we damaged our instrumentation.

21 This was a main issue that we were trying to deal with 22as we went towards Phase 2, is how are we going to 23 collect the data that we need to inform the model?

24 Next slide.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 58 So, what we came up with with NIST was a 1 tungsten slug calorimeter, essentially, a more robust 2 measurement device using one-inch diameter tungsten 3 slug.4 It was very durable and the hope was that 5 it would survive a direct impact from the high energy 6 arcing fault at the three-foot location and beyond, or 7 wherever we put these devices.

8 We were using a calcium silicate 9 insulating board, a Marinite type board, around the 10 tungsten slug, to essentially protect the device and 11 get the data that we needed.

12 You see a picture on the left there, that 13 is a picture of it under the heat release rate hood at 14 NIST itself, cone calorimetry, testing the principles 15 and developing that tool, so that we can use it in 16 Phase 2 testing. Next slide.

17 Another challenge that we faced with the 18 Phase 1 testing was, how to get the pressure results 19 from the events themselves.

20 We had a lot of interference with 21 electromagnetic interference for our pressure devices 22 that we were using in the Phase 1 testing, so we 23 didn't fully believe the results that we were getting, 24 because of this large interference.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 59 Pressure is a very important aspect of 1 these high energy arcing faults. As we said, we can 2 have blast effects with these, where we can create 3 pressure waves within a compartments and potentially 4 damage boundaries between fire barriers, or between 5 fire areas, or have pressure effects on other 6 equipment.

7 A good example of this is the event that 8 occurred at Turkey Point in 2017, where we did breach 9 a fire barrier and damage a fire door between two 10 adjacent switch gear rooms.

11 This is one area that we're looking at and 12 we really needed to dial in on how we were performing 13 these pressure measurements, so that we can get a 14 better picture of what is going on. Next slide.

15 So, as we move forward to Phase 2, we 16 decided to move to a fiber optic measurement device.

17 KEMA has upgraded, essentially, their facility and 18 we're getting much cleaner results.

19 All the pressure measurements that we have 20 received in the Phase 2, the first four confirmatory 21 tests, do appear to have worked adequately and just as 22 we expected and things are looking good in that area.

23 Next slide.

24 We are also looking at video as data, as 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 60part of these test series. It's a very fast event.

1 We're using a lot of different videography tools, such 2 as our infrared cameras, essentially to look at the 3 temperature of the event as a whole.

4 We are -- there are some challenges with 5 this, because it is such a quick event, and we're 6 seeing temperature ranges going from ambient to 35,000 7 degrees at the initiation of the arc itself.

8 So, there are some compromises that we 9 needed to look at, in terms of our infrared cameras, 10 such as the field of view, the speed, and the 11 temperature ranges that we can achieve. Next slide.

12 We have improved the infrared camera 13 technologies that we're using, both through NIST and 14 Sandia, trying to look at a dynamic picture of what is 15 going on during the test itself.

16 For our Phase 2 camera capabilities, we 17 have a greater temperature range and higher 18 resolution.

19 And all of these infrared techniques that 20 we're using will be shared, potentially, in a separate 21 NUREG report, that our colleague Dave Stroup is 22 working on currently, and we're going to be pushing 23that forward to our Phase 2 test program. Next slide.

24 Another important tool that we're using 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 61 and that we have recently developed for this test 1 program is our data acquisition system.

2 Right now, we're using an isolated data 3 acquisition system with an independent power supply.

4 We've seen in other test programs that the arc itself 5 can short out the data acquisition.

6 And like we said, these are fairly slow 7 tests, we can only perform a few a week, and if we 8 lose a data acquisition system and all the data, it's 9 a complete waste of a test.

10 So, we have spent a lot of effort in how 11 we can create redundant systems in order to collect 12all the data that we need. Our current system has 72 13 channels, we're only using about half of those 14 channels currently.

15 And I'll discuss different ways that we're 16 adding capability to this data acquisition system to 17 potentially collect more information, as we move 18 forward. Next slide.

19 So, what you're currently seeing on the 20 screen is our test setup for the Phase 2. We have 21 different instrumentation racks, located at three-foot 22 on all sides of the cabinet and above.

23 And additionally, we have a six-foot rack 24 located in the direction where we believe the arc is 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 62 going to travel. We've had fair success in the past 1 with predicting the arc path, based on the electrical 2 setup of the cabinet itself and where we're initiating 3 the arc.4 However, it's not an exact approximation 5for every test case. Sometimes, the arc will tend to 6 migrate, as we've seen done in previous tests, so we 7 really need to have a way to collect information on 8 all sides of the cabinet itself, so that we do not 9 miss the arc ejecta and all the energy release, to 10 have a useful test in all cases.

11 Some of the challenges with this is that 12 it's a lot of instrumentation, it's a lot of data, and 13 it's a lot of time to set up the tests. Next slide.

14 What you're looking at on the screen here 15 is what one of those typical racks looks like and the 16types of instrumentation that is on each rack. We 17 have ASTM E thermocouples that we're using, that link 18 to the IEEE standards.

19We have plate thermocouples. We have the 20tungsten slugs that we discussed. We also have a lot 21 of passive instrumentation on these test racks, such 22as cable coupons, aerogels, and carbon tape, that 23 we're using to measure the ejecta from the event 24 itself, looking at the byproducts that are created at 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 63 different distances away from the arc.

1 We're hoping that all of this will feed 2 into the model that we select, as we move forward with 3 the working group.

4 You can see here, this is a lot of data.

5 So, we're trying to spend our resources wisely, we 6 want to work with our modeler and figure out what data 7 exactly they need before we go and post-process all 8 data and all collection points that we took during the 9 test. Next slide.

10 Additionally, we're working with Sandia, 11 trying to leverage a lot of their DOE advancements and 12 their camera capabilities, high speed capabilities, 13 leveraging that, so that we can use it for this test 14program. We have various cameras located away from in 15 front of the test cell itself.

16 And additionally, we've developed a system 17to put a camera inside the test cell, using mirrors 18 and, essentially, to look at the arc location without 19 damaging their expensive cameras.

20 Worked fairly well in the first series of 21 tests and we hope to continue that as we move forward, 22 because, again, the video data is very important as we 23 try and assess what happened during the event, both 24 the infrared, as well as the high speed. We're able 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 64 to understand things now that we were not able to see 1 in the first series of testing. Next slide.

2 Another thing that we're doing in this 3 test series is, we're doing the full weights and 4 measurement of all of the test equipment before we 5 test and after we test.

6 We want to get a picture of how much 7 material was vaporized, both in the conductive 8 pathways, the aluminum in the different phases of 9 power, as well as how much was vaporized of the 10 cabinet itself, if we had cabinet breach.

11 This is something that was not done in the 12 first series of testing and does take considerable 13 resources post-test arrangement. Next slide.

14 This is the piece of equipment that we 15 used in the first four tests that were conducted in 16September. It was a single compartment GE-Magne-Blast 17 Metal-Clad Switchgear, where aluminum busbar was 18 located in the rear of the compartment that you're 19 seeing there.

20 This was the equipment that was discussed 21 in the April workshop and was selected with some 22interaction with the industry. It was deemed to be a 23 typical cabinet that we would see in the U.S.

24 operating plants. Next slide.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 65 This is the actual in-KEMA design setup 1before one of the tests. You're looking at the 2cabinet. The front door of the cabinet is located on 3your left-hand side. You see pressure transducers 4located on the sides of the cabinet, under that 219 5 symbol.6 As well as, you're seeing our protected 7 instrumentation racks located, again, at the three-8 foot and six-foot locations, in order to collect the 9 information that we needed from the ejected arc.

10 This was a -- for each one of these test 11 series, you see the racks there, themselves, we could 12 not reuse a single instrumentation rack, because of 13 the damage from the event itself.

14 There is a considerable amount, about a 15 week of prep time that went into building these 16 instrumentation devices at the KEMA facility before 17 they are used.

18 So, it is a considerable amount of work 19 that goes into a single test and they're not reusable, 20 so that kind of explains why these test take as long 21 as they do to prepare for, as well as to run a single 22 test. Next slide.

23 What you're seeing here, and I'm not sure 24 if you can make it out clearly, is the arcing wire 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 66 that we're using to initiate the arc itself across 1three phases of power. It's a thin -- Ken, can you go 2 over that with the mouse by any chance?

3 MR. TAYLOR: Yes, it's a 24-gauge single-4 strand tin-copper wire, per the C37.27 standard, IEEE. 5 MR. MELLY: And this was a change that we 6 made based on some of the comments we received during 7the public workshop. We initially had been initiating 8these arcs using the low voltage standard, because 9 they had a rationale for the ionization of the wire 10 that was created itself.

11 However, from one of the comments, we 12 decided to change that to the medium voltage case that 13was used, and we went with a thinner gauge wire. That 14 wire was able to maintain and hold the arc in all 15 cases during the test week. Next slide.

16 So, here are the four tests that we did 17perform. These are pictures of the rear of the 18cabinet itself, post-test. The first upper left-hand 19picture you see there is our Test 219. This was a 20 two-second arc at our 25kA.

21 And moving across, you have our two-second 22 arc at 35kA. On the bottom left, you have our four-23 second arc at the 25kA. And the bottom right is our 24 four-second arc at the 35kA, 32kA, I'm sorry.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 67 You can see the progression in damage 1 states, however, we did breach the cabinet in all 2cases. We do see this linking up with operating 3 experience, where an arc occurs in the rear of the 4 cabinet itself, it does create a cabinet breach.

5 From the video itself, we can actually 6 witness the metal in the back of the cabinet 7 essentially vaporizing and ejecting the energy towards 8 our instrumentation devices.

9 One thing to note here is that we do see 10fairly repeatable results. That was a challenge when 11 we had equipment donated from a ton of other 12 countries, where we essentially didn't have the same 13 piece of equipment.

14Which is why, in the second phase of 15 testing, we're trying to use the same make, model, and 16 initiate the arc in a similar location, so that we can 17 have comparable data as we move forward.

18 We were successful in locating our 19 instrumentation and we're using the arc initiation 20 point as our main parameter to decide where we 21 locating things.

22 In all four tests, we did see the door in 23 the front of the compartment open, due to pressure 24effects. It was a single-latching door. We were able 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 68 to re-close the cabinets after the experiment itself.

1 We saw minimal bending, but we didn't see many thermal 2 effects at the front facing of the cabinet itself.

3 Next slide.

4 Another aspect of this I want to discuss 5 is some of the observations we're seeing from our 6 measurement devices themselves. We were again using 7 that calcium silicate Marinite board for our ASTM E 8slugs, as well as our tungsten slugs. And we saw some 9unusual effects occurred during the test series. Next 10 slide.11 What you're seeing there is a before and 12 after picture of one of our tungsten slugs that was 13located on the three-foot rack. On the right-hand 14 side, you see this green silicate glass-like material, 15 that has bubbled up from the Marinite itself.

16 And we're looking into trying to match 17 material property characteristics for the Marinite to 18 the temperatures of the exposed surface, with the data 19 that we've collected from that tungsten slug.

20 We have -- those devices did survive and 21 we were able to actually collect valuable temperature, 22 as well as heat flux information from the majority of 23 our tungsten slug devices.

24 So, we do see that as a success moving 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 69 forward in the ability to capture these high energy 1cases at a close proximity to the cabinet. Next 2 slide.3 Again, our plate thermocouples did not 4 survive the highest heat fluxes that we saw there, 5 which did expose us to a new problem that we had not 6 anticipated, which was once the plate thermocouple 7 vaporizes, the wires and the wire pathway that is 8 behind that device becomes exposed, which damages all 9 the other wires behind it.

10 So, there were some cases where we still 11 lost data because of the high energy that's released 12 from the cabinet, and we're currently working at ways 13 to enhance the protections, so that we do not lose any 14 of the valuable data. Next slide.

15 Again, another thing that we're looking at 16 to improve our cost-efficiency of the test program is, 17 we have been using the ASTM E slugs.

18 That is largely so that we can have a 19direct comparison between testing that was done as 20part of the IEEE work, as well as ASTM work. However, 21 we may see this as a redundant testing device, which 22 we do not need to use moving forward.

23 It's one area that we hope to bring up 24 with the working group, after we can look at all of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 70 the data for these devices, to see if we even need to 1 use these moving forward and maybe make each test a 2little bit more cost-effective. They're fairly 3 robust, however, they did not survive the high 4 intensity arc itself. Next slide.

5 Another big ticket item that we're looking 6 at for this test program is the deposition of ejected 7 byproducts, as we move away from the cabinet itself 8 for these aluminum tests.

9 What you're seeing here in these two 10 pictures is some of that ejected material, which is 11 the white powder coating that is on the wall of the 12 test cell itself.

13 We did take black carbon tape samples of 14 this material and we're going to be performing post-15 test analysis at Sandia to look at exactly what that 16 byproduct was, both at the wall, on the three-foot 17 test stand, and the six-foot test stand.

18 One important note is that we didn't 19 really see this type of ejecta during the two-second 20 arcs that we had in this test program. The majority 21 or the bulk of the material was during the four-second 22 test.23 So, that's another issue that we're 24 looking at, that these aluminum tests might not have 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 71 the amount of byproduct material ejected, depending on 1 the duration of the arc itself.

2 So, this was a question that came up just 3 recently in this series of testing that we're going to 4 be evaluating. We have a question?

5 MS. VOELSING: Yes, sorry, this is Kelli 6Voelsing from EPRI, again. I'm sorry, I may have been 7 confused, but I thought the latest information I 8 received was that, although you took samples with the 9 carbon tape, you did not plan to have Sandia analyze 10 them, because of resources, and you were counting on 11 the small-scale test results to do the material 12characterization. Did I misunderstand what I was 13 previously told?

14 MR. TAYLOR: Yes, there must have been some 15 misunderstanding. So, we put samples in a number of 16locations. On every rack, we had a number of 17 different types of samples, aerogel, black carbon 18 tape, PVC samples, and we have hundreds of samples to 19 analyze.20 There's probably five or six different 21ways that can analyze each sample. And really, 22 putting the samples in the testing is not the 23 expensive part, the expensive part is the post-test 24 analysis.25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 72 So, what we're trying to do right now is 1 understand exactly what the model needs and what 2 samples need to be analyzed and how they need to be 3 analyzed, before we move forward and expend the 4 resources to analyze the samples.

5 So, they've done some preliminary, picking 6 off one on each rack and analyzing it for different 7 parameters.

8 But as far as analyzing the samples that 9 are needed to inform the working group's work or the 10 modeling work, we haven't initi ated that type of 11resource expenditure. So, we're trying to be as 12 efficient and as effective as we can to get the 13 information that's needed.

14 MS. VOELSING: So, to date, either from the 15 small-scale or from this testing, there's no material 16 characterization results available?

17 MR. TAYLOR: Not to the public, no.

18 MS. VOELSING: Or to the working group?

19 MR. TAYLOR: Or to the working group, yes.

20 They have done some preliminary analysis, but even on 21 the small-scale, the small-scale is really focusing on 22 the modeling.

23 And until they have -- because the changes 24 in direction on the modeling and the approaches, until 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 73 we have the modeling approach nailed down, we didn't 1 want to analyze the samples yet.

2 So, all the samples are still out at 3 Sandia, there's some preliminary information that was 4 provided to the NRC, as just a quick look, but as far 5 as a comprehensive analysis of the samples, we don't 6 even have that yet.

7 MS. VOELSING: It might be helpful to the 8-- even to have the quick look, to know what the 9material was characterized at. And will there be any 10 potential for stakeholder engagement on these models 11that you're working on with Sandia? Because that 12 doesn't seem to be a part of any of the discussions.

13 MR. TAYLOR: I'll let -- do you want to 14answer that or -- okay. So, yes, when we get far 15 enough down the line, I think that's, as Victoria made 16 mention, it would be very valuable to have feedback on 17 the modeling and what we're getting out of that.

18 MS. VOELSING: It might be helpful to have 19 feedback on the front-end, to make sure that what 20 comes out the back-end is appropriate.

21 MR. TAYLOR: Agreed.

22 MR. MELLY: And we do see this as a very 23important piece, as we move forward. Again, the 24 material composition, as well as the conductivity that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 74 we're talking about affecting equipment is a big part 1 of the aluminum issue.

2 And we recognize that and we have been 3 working towards ways of improving the data that we've 4collected, so that we can draw this conclusion. We do 5 see this as an important aspect. Next slide.

6 And again, this is a picture of some of 7 the metal ejecta that we did take samples of at 8 various distances from the cabinet itself, that's what 9 you're seeing on the ground here. Next slide.

10 And to even push that forward, we have 11 looked into new ways to actually address, actively, 12 conductivity measurements within the room itself, 13 during the test that we do conduct.

14 We're going to be using a developed tool, 15 instrument, in order to look at the conductivity 16 during a test cell, as we do these four low voltage 17 and four medium or bus duct tests in the summer.

18 These are essentially resistant devices, 19 where their initial intent was to measure the 20corrosion effects in smoke environments, to look at 21 how they affect conductivity changes in real-time.

22 And we're hoping that, if we do use these 23 as we move forward, as we coat these in byproducts, we 24 can get a real-time estimate of conductivity changes 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 75 within the test cell itself at various distances.

1 This is a new device that we've not tested 2 before, we're currently working with NIST, not right 3 now, because they're furloughed, but we're working 4 with NIST to add these into the test program itself 5 and put these in the future tests that we're running.

6 We're doing some shake-out tests right now 7to see if these will be useful and the type of data 8 that we can collect using these.

9 As I said earlier, the next four tests 10 that we have planned, or the next eight tests that we 11 have planned, include four low voltage tests, at 15kA 12and 25kA, the tests that are identified here. A two-13 second test and eight-second test is slated for the 14 next series of testing. Next slide.

15 And additionally, we have our bus duct 16tests. We're going to be doing a copper bus duct with 17 an aluminum enclosure, which is typical that we see 18 and that we saw from a survey of how much aluminum is 19 out there, that was performed by NEI.

20 As well as, we're going to be looking at 21aluminum bus in an aluminum enclosure. So, these are 22 the tests that we have slated for the spring/summer.

23 Next slide.

24 Some of the important OECD, the next phase 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 76 actions and some of the important milestones that we 1have upcoming. We've completed most to this date. We 2 also are going to be getting some international 3equipment for the OECD NEA tests. We have just 4 received that in-house.

5 And the next big milestone is the second 6 series of tests that we're going to be performing in 7the spring, May, summer time frame. We have not 8 established a firm date yet, we're going to be working 9 with KEMA, as well as NIST, when they're off furlough, 10 to establish the dates for the upcoming test series.

11 Any questions on anything that I've went 12 over in the last portion of this presentation?

13 MR. HAMBURGER: We'll take any questions 14 from the room first.

15 MR. SCHAIRER: Yes, this is Mark Schairer 16with EPM. Can you elaborate a little bit more on how 17 the data you're collecting will support the zone of 18 influence spectrum?

19 In particular, are you going to 20 differentiate between ignition and damage of cables 21 and different cable types, like thermoplastic, 22thermoset, Kerite? So, is the data you're collecting, 23 will that be sufficient to support that spectrum?

24 MR. TAYLOR: So, we're -- Gabe Taylor, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 77 Office of Research. So, that's one thing that we've 1 been trying to get a grasp around is, we're doing a 2 lot of work to understand the hazard, but on the back-3end, the fragility that you're speaking to, does 4 equipment or cables become damaged or do they ignite?

5 That's an area where -- the applicability 6 of the current data, that's more thermal fires and 7 damage via these high energy short duration exposures.

8 We're trying to get a grasp of how we would go about 9 doing that.

10 So, again, we're looking at possible heat 11 conduction type modeling, similar to the THIEF model 12 that we have in FDTs, to try to characterize what the 13 thermal fragilities, from a function standpoint and 14 from an ignition standpoint, are.

15 But if there's -- it's an area where we're 16 definitely seeking feedback from anybody that has 17 expertise in this area, to help inform the process 18 going forward, because that's a very important concept 19 of determining what the ZOI is, is what those 20 thresholds or fragilities are.

21 MR. SHUDAK: This is Tom Shudak with 22 Nebraska Public Power. As far as your other testing 23 that you're going to do on copper buses, were you 24 anticipating doing those before you released your 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 78 report or are you going to do some comparison between 1 the copper and aluminum from just a zone of influence 2 from these control tests?

3 MR. TAYLOR: So, we have the original 4 testing program for the Phase 1 OECD test. So, they 5potentially provide us, for a comparison. But as far 6 the GI Program, that test data from Phase 1 and 7 anything else that we can come up with from either 8 NFPA or IEEE to compare it from, we will.

9 As far as the other tests in this matrix, 10 we don't envision those testing being performed before 11the assessment phase would need to be completed. So, 12 the comparison of those wouldn't be available for the 13 GI Program.

14 However, later on, when we complete all 15 the testing and we work with EPRI through the working 16 group and come up with kind of the updated guidance 17 for 6850 Appendix M replacement, that obviously would 18 be a very valuable comparison.

19 MR. PRAGMAN: Chris Pragman, Exelon 20Nuclear. I just want to follow up on Tom's question, 21 because my understanding of the GI process is that the 22 decision is based on a change in CDF, or delta CDF.

23 So, if you don't have the base model involving copper, 24 how do you quantify the delta CDF for aluminum?

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 79 MR. MELLY: So, the CDF or the delta that 1 we're going to be using right now is compared to the 2current 6850 Appendix M zone of influence model. So, 3 we're not changing that.

4 However, if we need to look at the zone of 5 influence as potentially greater, through the working 6group or through the charter, that is how we would 7 establish a delta.

8 From the current modeling techniques that 9 are used, that three-foot, five-foot, with the 10 caveats, versus what we establish for the aluminum, 11 that would be the current delta that we're talking 12 about for the Generic Issue assessment phase.

13 MR. PRAGMAN: But it seems like to make an 14 apples-to-apples comparison, you'd have to compare 15four-seconds copper to four-seconds aluminum. 6850 16 isn't four-seconds anything, it's more a empirical 17 number based on past experience.

18 MR. MELLY: And that's why we're not trying 19 to simply say, it's going to be a four-second model.

20 We're going to be developing a model specifically for 21 aluminum, using this data, and then, applying that to 22 the plant.

23 However, the current model that is in 24 regulatory use is that 6850 model, so that would be 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 80 the baseline risk, what plants have currently modeled 1 to, versus what we come up with as a working group for 2 the generic assessment phase.

3 The eventual models, as we move forward 4 and complete the entire test program, will be looking 5 to change both the 6850 and the aluminum, so we may be 6 altering the copper zones of influence as we move 7 forward.8 However, for the Generic Issue assessment 9 phase, the baseline is the current modeling techniques 10 that are part of the regulatory basis versus what the 11 working group comes up with for the zone of influence 12 of aluminum.

13 Again, not a single four-second 14 conservative model, but a spectrum that uses the 15 frequency of occurrence, plant design, as well as all 16 the data available, both the spectrum of two-second, 17 four-second tests.

18 MR. PRAGMAN: So, it sounds like we're 19 saying we're going to -- the baseline is 6850, which 20 is based on experience. The GI process, we'll use a 21 model -- I mean, when we're using models as licensees, 22 we're expected to benchmark them somehow.

23 So, I don't understand how you would 24benchmark your model versus experience. I don't feel 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 81 like you've explained that well enough yet.

1 MR. MELLY: And hopefully, that will be the 2 next meeting that we have, where the working group 3 establishes the basis for this new model that we've 4 created, using both data, operating experience from 5 the plants and the frequency to explain what we will 6 be using moving forward as a zone of influence for 7 these aluminum cases. Again, we're not changing the 8 baseline, which is the Appendix M of 6850, though.

9 MR. PRAGMAN: Thanks.

10 MR. TAYLOR: I just had one thing to add to 11the previous comment from Mark. One thing I forgot to 12mention, in our -- and Nick probably brought it up.

13 But on our instrument stands, we also place cable 14 samples, so we can look at those afterwards and see 15 the effects from the event or the test.

16 MR. MELLY: Yes.

17 MR. TAYLOR: Cable coupons, yes.

18 MR. MELLY: We use various materials of 19cables as well on those. Tom, do we have questions on 20 the webinar?

21MR. AIRD: We have one question on the 22 webinar, from Mark Hewlett, APS. He asks, are the 23 connectivity sensors mounted in a manner 24 representative of location inside a nearby cabinet or 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 82 direct exposure?

1 MR. MELLY: Currently, the thought is that 2 we're going to put them direct exposure at various 3 distances away from the cabinet, such as on our 4 instrument racks at the three-foot, six-foot, and 5 maybe beyond, where we anticipate deposition of 6 materials.

7 However, the current test plan does have 8adjacent cabinets as a potential. And if those tests 9 are performed, then we can put the conductivity 10 measurement inside of that cabinet, to see if we're 11 getting any influence inside the cabinet itself from 12 deposition of materials.

13 MR. TAYLOR: And from those sensors being 14 used in the smoke type testing, orientation plays an 15 impact on the measurement.

16 So, if there's feedback on -- we have a 17 solid state protection cabinet that's got circuit 18 cards in this orientation or we have this other 19 cabinet that has circuit cards in this orientation, 20 that would be useful information for us to incorporate 21into the testing, to ensure that we're being 22 representative of the configurations.

23 MR. HAMBURGER: Any other questions from 24those in the room? Anything on the webinar. Okay.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 83 And again, I think all of the staff here is going to 1 be here for the rest of the day, so if any questions 2occur to you, don't hesitate to ask. And I think Mike 3 Cheok has a --

4 MR. CHEOK: So, Ken, before we go to the 5 next presentation, I would just want to revisit what 6 I said at the beginning and then, we can go from 7 there.8 So, I think what I said at the beginning 9 was we are going to have to be very methodical in 10doing our tests and evaluating this issue. And the 11 second thing was, improving our communications. So, 12 I will, I guess, say it again.

13So, in being methodical, we have to do 14things correctly. We have to do tests that are 15 realistic, that are in conformance to some standards.

16 We have to do tests that are practical.

17 So, what we heard this morning was some 18 discussion of the tests, but on the -- and the second 19 part of the Generic Issue is not just the testing, but 20 to look at operating experience. And we haven't got 21to that part of the test yet. And that becomes a very 22 important part of resolving this Generic Issue.

23 So, we also need to be very realistic, in 24 terms of looking at OpE, operating experience, so that 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 84 we can be realistic and so that we can be using 1 parameters that are plant-specific, that we see in our 2 plants.3 So, what's going to happen -- what we 4 continue to have to do is to marry the results from 5 the testing with a frequency and a look at what we 6 have out in the plants.

7 So, what we are hearing this morning is 8only a smaller part of the whole Generic Issue. We 9 will continue to have to involve all our stakeholders, 10 in terms of trying to be looking at operating 11 experience.

12 We have to be looking at plant-specific 13information. Like I said this morning, we have to 14 look at the design, plant-specific designs, 15configuration-specific designs. We have to look at 16 plant practices.

17 And so, on the other hand, on testing, we 18 also have to understand what the zone of influence is.

19 And we will continue to work with our testing partners 20 and with EPRI on, once we get the tests back, we will 21 have to look at what the results show us, how we look 22 at the raw data, and how we can interpret the raw data 23 into something that could be useful.

24 And so, yes, Sandia modeling on the data 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 85 will be available to us, hopefully soon, and when we 1 get it, we'll be glad to share it with EPRI and the 2 rest of the industry: this is what the data is showing 3 us, this is what we will -- this is how we will use 4 it, in terms of modeling.

5 And at that point, we -- it's going to be 6 a public meeting, everyone's going to be asked for 7 feedback on how they think the data and the modeling 8 is going to match.

9 Again, so, we are a little bit ahead of 10the schedule. I mean, off-the-cuff right now, we are 11hoping to get some data soon. NIST being not in a 12 work status now kind of delayed some of our data 13 collection and analysis.

14 But once we get that also, obviously, we 15 will walk that through the work group, which you will 16 hear this afternoon, and then, with the rest of the 17 public in a forum where we will present the data for 18 everyone to analyze.

19 So, the second part of what I said this 20morning was communications. And I would like to have 21 everyone's input, and I'm not sure I got that feeling, 22 and I'm not sure I want to be talking past some of the 23 questions and comments this morning.

24 So, I think what I heard was, we said we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 86 dispositioned a lot of the industry comments, but that 1 in dispositioning the comments, we may not have come 2 to a -- we did not address them, for lack of a better 3 term.4 So, I think what we need to do, maybe this 5 afternoon or maybe after this meeting, we will sit 6 down with our staff and we'll -- if you all will 7 provide us with very, more specific comments on, this 8 is what we thought we gave you, we did not see the 9 resolutions to this, we'd be glad to discuss this 10 again in the next public meeting.

11 We'll be very specific as to, this is your 12 comments, this is how we will address this in the next 13 phase of testing or in the next task, in terms of 14 frequencies or in the characterization of the zone of 15 influence.

16 So, I think we can be more specific, in 17 terms of addressing comments in a public forum, so 18 that we understand your comments better and you can 19 understand our responses better.

20 So, let's see what else I've got in my 21notes from the last two hours. I think I got it all.

22I mean, so, we will talk about the decrement curve.

23 We will talk about using the need for an eight-second 24 duration test.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 87 I understand that there's some 1sensitivities to test extrapolation. So, things like 2 that, I think we can address and we will address it 3 specifically in the public forum.

4MR. HAMBURGER: Thank you, Mike. If we 5 don't have any more questions from the room, I think 6we're overdue for a break. So, let's come back at 7 11:00. Come back at 11:00.

8 Again, bathrooms are out the door on this 9side of the room, to the left. Cafeteria is open, 10 again, that's to the left, down the hallway, and that 11will be on your left as well. My contact information 12is up on the screen. If you have any issues, give me 13 a call, it's probably the best way to get a hold of me 14 at the moment.

15 (Whereupon, the above-ent itled matter went 16 off the record at 10:42 a.m. and resumed at 11:07 17 a.m.)18MR. HAMBURGER: All right. So, we're going 19 to resume with Gabe Taylor.

20MR. TAYLOR: Okay. Welcome back. So, 21 myself and Kenn Miller are going to go through future 22 test planning for the aluminum HEAF testing and the 23 equipment selection. Next slide.

24 So, during this presentation, we're really 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 88trying to communicate what we're planning on doing, 1the upcoming tests, and also, receive feedback on 2 anything that we can do to improve how we go about 3 testing the equipment as we move forward.

4 Just as a refresh, I know Nick brought 5 this up in this past presentation, but if we look at 6the overall test program, including the OECD test 7 report, you can see that we've completed the four 8 medium voltage 6.9kV tests, in the fall, September.

9 One clarification is that it was 32kA, not 35, as the 10 slide shows.

11 So, for the aluminum HEAF, we're really, 12 on this enclosure testing matrix, we're really looking 13 at the right portion of the matrix, and the individual 14 tests that are highlighted in the peach or orange 15 color are the ones that are NRC-sponsored.

16 So, that leaves us with four low voltage 17 tests, at 480 volts, two current levels, 15kA and 1825kA, and durations of two and eight seconds. Note 19 that we also will be planning on running four-second 20 tests as part of the OECD program.

21 On the far right, you can also see two 22 tests that aren't connected to any part of the test 23 tree matrix.

24 And as we communicated previously, we plan 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 89 on using those two spare or optional tests to evaluate 1the decrement curve. The decrement curve will be 2 evaluated at the 6.9kV, and it's yet to be determined 3 what current level we'll be testing that at.

4 But those are information that we're 5 trying to work with EPRI and understand the best way 6 to represent realistic plant decrement type 7 conditions. So, just -- and I have a later slide on 8 the decrement curve. So, that's low voltage.

9 If we look at the equipment that we're --

10we haven't procured any equipment yet. We've put out 11 bids.12 And right now, what we're look at for the 13 aluminum bus low voltage enclosures, either a 14 Westinghouse DS type switchgear or a General Electric 15 AK series switchgear. So, these are the low voltage 16 switchgear load centers that will be just downstream 17 of your step-down transformer.

18 As far as frame size, we're looking at 191600 amp or 200 amp continuous frame size rating. And 20 then, also, for interrupting capacity, we're looking 21 at anywhere between 42kA and 65kA.

22 And those numbers really came out of the 23 evaluation we did to support the workshop last April, 24looking at what was common for plants out there. And 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 90 although our sample size wasn't the entire fleet, it 1 was about a quarter of the fleet that we sampled from 2 the information that we had.

3 And as far as -- these were the common 4ranges of equipment that was out there. So, if you 5 looked at an average, interrupting capacity was right 6 around 50kA for what we looked at. Next slide.

7 Okay, moving on to the medium voltage bus 8 ducts. For this type of testing, we're looking at a 9 combination of enclosure and busbar material types.

10 And also, there's a combination of NRC solely-11sponsored tests and OECD-sponsored tests. The voltage 12 level we chose for this was 4160 volts and a single 13 current at 25kA.

14 The combinations, as you can see, are 15 copper bus/steel enclosure, which is OECD only.

16 Copper bus/aluminum enclosure, which is similar to the 17 test we ran in Phase 1, the Zion bus duct experiment, 18 so we have two tests there at two and four seconds.

19 Then, we have an aluminum bus/steel 20enclosure. In that case, we'll run one of the two-21 second tests, there's an OECD test for four seconds.

22 And then, aluminum bus/aluminum enclosure, we have two 23 tests there for two and four seconds.

24 We do have two spares test, and at this 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 91time, we haven't determined a need for those yet. But 1 there could be a potential there to evaluate decrement 2 curve or some other type of configuration, as we see 3 fit.4 So, if there's feedback to support those 5 optional tests, we'd be interested in learning that.

6 So, and one thing, just to be clear, these aren't the 7 iso phase bus ducts, these are typically non-8 segregated phase bus ducts. Next slide.

9 As far as the general characteristics of 10 the medium voltage bus ducts, again, non-segregated 11 phase bus ducts. I gave you the voltage already.

12 As far as the rating, we're looking at 13 1200-2000 amps rating on the busbars and then, also 14 from a withstand rating anywhere from 31.5 to 63kA, 15 which that's typical out of the standard that's 16 identified there for medium voltage bus ducts.

17 One area that's still uncertain at this 18 time is, for our measurements, we need to stabilize 19 the arc in one location.

20 And if you're familiar, from the testing 21that we did in the first phase, the OECD program 22 testing, because of electromagnetic effects, it 23 basically propelled the arc out the end of the bus 24 duct configuration.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 92And if you look at the OpE, there's 1 typically some feature that stabilizes the arc, 2 whether it's a bolted connection or a grounding bar or 3 some other cause.

4 Recent OpE, there was an event that 5 happened exterior to the plant, they had an exterior 6 bus duct, and because of the coatings that they had on 7 the busbar, that acted to stabilize the arc in one 8 location.9 So, right now, what our thought process is 10 is what's shown in the graphic there, is that if you 11 see the power supply from the laboratory, coming down 12 the left side of the image, the copper colored 13 conductors, it then runs horizontally into the bus 14 duct.15 And in the bus duct, we make a physical 16 separation of the bus duct and, on the left side, we 17 place a shoring wire at that location to generate the 18arc there. The idea there is that would stabilize it 19 in that general location, where we'd then have the 20 instruments located above and below.

21 Another option could be to use the epoxy 22 insulation that was used in the industry testing.

23 What we found in some communications with external 24 parties is that that type of insulation isn't common 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 93inside the plant. So, that's one non-conformity that 1 we would potentially introduce into the testing, if we 2 performed that.

3 Another option that we also had some 4 feedback from the Japanese experiments is that they 5 put a grounding bus perpendicular to the orientation 6 of the phases, the bus phases, and that helped to 7stabilize the arc as well. But then again, that's 8 something that's not common to the types of 9 configurations that are in the plant.

10 So, just to kind of summarize, right now, 11 we're looking to put a gap in the bus ducts to 12 stabilize the arc, but if there's any feedback or 13 logic to support some other configuration that would 14 provide a more reliable means to maintain the arc in 15 one location, so that our instruments could capture 16 reasonable and usable data, we'd be very interested in 17 that type of feedback.

18 And then, from informal surveys of some of 19 the equipment vendors, common enclosure housing 20thickness, looking at 11-gauge aluminum housing. The 21 conductor for aluminum, for the aluminum experiments, 22 to follow the ASTM standard that's specified there.

23 So, if there's any other types of design 24 considerations that are out there we should be 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 94 considering, we're very interested in learning those.

1 If not, we'll work with our electrical contractor to 2 try to find and procure bus ducts that are 3representative of the types of installations in the 4 field.5 Okay, now it's moving on to the decrement 6curve. And we talked about this a little bit earlier.

7Last January of 2018, EPRI published some white 8 papers.9 In the one white paper, they went through 10a number of different scenarios. And one of the 11 scenarios they to be most common to exhibit a long-12 duration arc was cases where it's a generator-fed 13 fault.14 So, basically, you'd have a fault between 15 a generator and a protected device, so you wouldn't 16have a main generator breaker to clear the fault. And 17then, the generator would basically feed the fault 18 until the exciter voltage dropped out and stopped 19 providing energy to the fault.

20 So, we found that information very useful 21 and, as I mentioned earlier, we're trying to work with 22 our testing facility to be able to replicate some of 23 the field experience that was provided to us via EPRI.

24 It does require new contracting, which takes time.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 95 And also, it requires some analysis and 1 verification on the laboratory's part, to make sure 2 that they can run the experiment that meets our needs, 3 to replicate that type of exposure.

4 So, it's not a rudimentary process of 5 going out and doing this, this isn't very common or 6 we're not aware that it's been done in the past, but 7 we do have confidence that it can be done moving 8 forward.9 And that's feedback we received from EPRI 10 and during the workshop and during the public 11 comments, and we're definitely taking that into 12 consideration and making changes to improve our test 13 program.14 So, that's one example where we made 15 changes, but we're also looking for any other feedback 16 to improve our testing, because in the end, we have to 17 have something that's usable to make realistic methods 18 and models to inform our risk analysis.

19 So, just to summarize, our initial focus 20 here is to support the Generic Issue Program. We're 21 focusing on the NRC-sponsored aluminum type tests, to 22 support the GI Program in its assessment phase, to 23 provide useful information to determine what the ZOI 24 or the ZOI model is, to support the risk assessment.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 96 And although the decrement curve, the 1 process for getting all that is slow, we're trying to 2 expedite it as much as possible.

3 There could be possibilities, if they're 4 interested, for industry to per form that type of 5testing. We could support it on the measurement side 6 as well, if that is a faster means of getting it done.

7I think that's all I had. So, I really 8 appreciate any feedback on our planned approach this 9 coming year.

10 MR. HAMBURGER: Gabe, if people have 11 feedback that occurs to them after this meeting, can 12 they get in touch with you by email?

13MR. TAYLOR: Right. You can email 14gabriel.taylor@nrc.gov. My information's on the 15meeting notice. As well as Nick Melly, 16 nicholas.melly@nrc.gov.

17 MR. HAMBURGER: Okay. Any questions from 18those in the room? About any of the equipment Gabe's 19 presented?

20 MR. TAYLOR: I'll just make one additional 21thing, while we're waiting. We'll run it no 22 differently than we've done previously.

23 In the first phase, we identified what 24 equipment was available and then, we worked with EPRI 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 97 to make sure that was consistent with the type of 1 equipment that's found in the field.

2 So, we do have some initial leads on 3 equipment last week, but there's a few bids still out 4there to see. And once we get that information, we'll 5 work with the working group to understand which 6 equipment is most applicable for the series of 7 aluminum tests.

8MR. HAMBURGER: Okay. Any questions or 9 comments? Tom, anything on the webinar? Okay.

10 Well, I think the next presentation is 11 slated for an hour and I think there's going to be 12 some comments and questions on that, so I don't want 13 to get into that before lunch, because I have a 14 feeling that may delay lunch.

15 So, unless anyone objects strenuously, I 16 think we can go to lunch now and take a little 17extended lunch. Come back at 12:45? 1:00? You okay 18with 1:00? Mark, unless you think you can get it done 19 in -- okay, we'll come back at 1:00.

20 (Whereupon, the above-entitled matter went 21 off the record at 11:21 a.m. and resumed at 1:00 p.m.)

22MR. HAMBURGER: All right. Thank you all 23 for returning promptly.

24 Just a reminder, for questions and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 98 comments, please state your name before your comments, 1 so the court reporter can transcribe it properly, and 2 try to use the microphones for any comments or 3 questions, so other webinar participants can hear it.

4 Mark?5MR. SALLEY: All right. I take it 6everybody had a good lunch. We've got a lot to cover 7 this afternoon. So, we'll pick it up.

8 Back on May 11th, 2017, in this room, 9 there was a briefing for the Commission. And one of 10 the industry execs was at the table and briefly 11 described the HEAF and the work that was going on.

12 And he made sure to make the point that the industry 13really wants, speaking for the industry world, the 14 industry really wants to work with NRC research on 15resolving this. The vehicle that we typically use in 16 research is the MOUs I described earlier with EPRI.

17 And we try to get the scientific arms together to do 18 that work.

19For those of you that have been around 20 fire protection and have been here a while, if you 21 think back 20 years ago or so, again, in this room we 22 had a whole different topic in fire protection that we 23 were working, and that was multiple spurious --

24 everybody remember that? -- with the hot shorting of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 99circuits and all that. It was a big issue in the late 1 nineties and at the turn of the millennium.

2Today we don't talk too much about it. It 3seems to be solved. Everybody's comfortable doing it.

4The utilities know what they need to do. The 5 inspectors know how to inspect it. And we have done 6 a lot of work.

7 And as a part of the success of that, it 8was that MOU between research and EPRI. If you 9remember, we did a whole series of tests. The 10 CAROLFIRE program was the first one, and we worked 11 through it, again working together to get a good 12 scientific solution.

13 So, we've seen success with this before, 14and it's worked quite well. I think we really 15advanced the science, and it's good science. So, 16 again, with that in mind, we want to reach out and 17 work with EPRI on the technical solution to the HEAF.

18 Next slide.

19 So, what we did in the past few weeks, 20 working with Kelli and her folks and mine on this 21 side, is we had a number of calls and discussions, and 22 we thought we'd put together a Charter for what this 23 Working Group is going to attempt to do. We held it 24 out as a draft until today, again, for any feedback 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 100 from stakeholders or industry on where we're looking 1 to move forward.

2 Again, the basic mission statement here, 3I think it's quite obvious: to improve the 4 understanding of risk from electrical arcing fault 5hazards in nuclear power plants. And you'll notice we 6 chose the words "arcing fault hazards" and not 7directly to HEAFs. I think that's part of it and the 8 fact that we want to get the correct binning and get 9 the correct selection of the events. So, that's the 10 basic Charter for the group.

11 Next slide.

12 So, the goals of this Working Group:

13 Obviously, a better understanding of the 14 key factors contributing to the occurrence and 15 severity.16 You can read them here.

17 Advancing the HEAF fire PRA modeling.

18 Again, we have the single model out of 6850 for the 19zone of influence in Appendix M. We also have the one 20FAQ for the bus ducts. We think we can do a lot 21 better with that.

22 And then, analyzing the plant impacts and 23the risk implications. If you look at that, that 24should look very familiar to everyone here. That's 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 101basically HEAF in the form of a risk triplet. So, 1 again, following the sound principles is how we want 2 this group to move forward.

3 Next slide.

4 So, this is where it kind of gets a little 5interesting. There's a lot of famous quotes and a lot 6 of people have said things about, you know, getting 7the right group of people together. I think one pops 8into my mind, Lee Iacocca.

If you remember Lee 9 Iacocca from the automotive world, this is the man, 10 basically, that came up with the Mustang in 1964, and 11then, he saved Chrysler later. So, he was a pretty 12 heavy hitter in there.

13 He had a saying that kind of went 14 something like this: I hire people smarter than me, 15and then, I get out of their way. So, that's kind of 16the philosophy that I want to look at with this 17 Working Group, and this is the Working Group that 18 we've assembled between Kelli and myself working for 19 some of our best experts out there. You may of the 20 people. Some or a lot of them are in this room here 21 today. And again, they're very good people.

22 Also what's here, we put two levels of 23management. The Project Managers, again, Kelli and 24myself. If you have questions or want to talk to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 102someone, we're there. And then, our sponsors that are 1 above it for the whole program, which is Mike Cheok, 2 my supervisor, and then, Tina Taylor.

3 A lot being said with putting the right 4 people together and getting out of their way. We've 5 talked a lot about zones of influence and things of 6that nature, and how we want to do this. And I really 7 think that, as we're putting this group together --

8and they're just starting out, by the way. So, we 9want data, we want this, we want a lot of things. But 10this group is just getting together. Like I said, the 11 Charter is a draft. Any input from anyone today was 12 welcome, and then, we'll get started with this.

13 But looking at HEAF -- and I'll set the 14 bar real high for this group -- don't just lock me 15into a ZOI or something like that. I mean, the 16 sheet's clear. We're starting out.

17 If we look at HEAF and we talk about the 18 tests and why we're doing some of the tests and 19 different things, we basically have one equation with 20six unknowns. Okay? And if you think back to college 21and your early days in engineering, most of us went 22 through you could do one equation, two unknowns, and 23 work it through and substitute, and bring it back, and 24 all that.25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 103But this is a complex phenomena. I mean, 1 what effect does the voltage play? What effect does 2the current play? And what effect does duration play?

3What effect does the material play? What effect does 4the enclosure do? We know there's a lot of variables 5 here.6 And I think, as a part of this Working 7 Group, or what I challenge them to do is tell me this.

8Does voltage matter? Does current matter? We've 9fought a lot about duration. We're seeing the 10 material properties seem to matter between copper and 11 aluminum.12 So, this group has a lot of work ahead of 13 them to do it. And I don't want to tie their hands, 14 but I want to let them do their thing.

15In a perfect world, if this group was, 16 I'll say it as a manager, very successful, we would 17 have an equation where you could plug in the voltage, 18 you could plug in the current, the enclosure, and come 19 up with the zone of influence.

20 And maybe it's not a box like we've been 21 doing 3 foot x 5 foot. If you look at some of the 22 data we've gotten from video, it's more of a sphere.

23 Is that the correct thing?

24 Or the FAQ group, when they worked the bus 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 104 duct out, I think, Industry, what do you call that, 1"the cone of death," or something, for the cone that 2comes down from the bus duct? Again, I don't want to 3limit the group. I want to let the physics and the 4 science do what it needs to do.

5 So, these are the people that we've 6 selected. I think we've got a good group here, very 7 capable. And we want to let them get on with it.

8 Let me see what I've got. Next slide.

9Again, deliverables. To improve the risk 10 models. The PRA stuff is driving it. That's what's 11got us here. We want to look at the frequency and the 12binning and the zone of influence. And again, I'll 13zone of influence loosely. I think the group can come 14up with what they need to for that. It may be a 15 sphere. It may be an equation. It may be different 16 things they can come up with.

17 Again, our goal was to look at the risk to 18the plants. We talked a little bit about pilots, and 19 that's something, again, where we'll need some help 20from industry. If you go back and look at 6850 and 21the history of that, again, for the folks that have 22 been around since the 2000 timeframe -- we published 23 that in 2005 -- one thing that we constantly at the 24 end of AFP-805 we're hit with is, you know, we never 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 105really piloted it. We piloted it in pieces. We never 1did a full pilot. If hindsight is 20/20, everyone 2 agrees back in 2000 we would have piloted 6850, and it 3 would have been a better tool for all to use.

4 So, here's an area, too, where we're 5looking at pilots. Again, we're still in the 6 discussion phase as a part of the Working Group.

7Remember, this is just starting up. But that the 8 Working Group would develop the models and the tools 9and the data and how to do that, and then, let the 10 licensees or the consultants implement it in the plant 11to do the pilot. So, that's kind of where we're at 12 with that right now.

13 Again, we really hope to advance the 14 science, to reduce uncertainty, to improve realism, 15all the things we strive for, and to update the 16current guidance. It is funny, when we started out in 17 the testing way back with the internationals, our goal 18 was to go and to provide the confirmatory test that 19 Appendix M, 6850, was a good model.

20 And for the most part, we were doing 21 really great and we were thinking, hey, this is a good 22-- man, if you had to pick a one-size-fits-all HEAF 23 model, up until about that 22nd test, we're saying 24this was right on. And we were, you know, Nick and I, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 106 patting each other on the back that this is really 1 going to be successful until those later tests came 2 and showed us that there's more to it than that.

3 So, again, we want to get that accuracy 4 and we want to get that fidelity in it; and again, 5 understand the variables and the parts they play, 6 which, again, will be the update to the guidance for 7 6850, Appendix M, and the supplemental, the FAQ -- I 8 think it's 37 or 47 -- for the bus duct.

9 Communications, we talked this morning is 10important. We will strive to have open communications 11 with you. And I guess, as the Working Group, like I 12 said, the key is we've got to let them do a little 13work. It's just getting together. And I think as 14 they get to key points in that, we can agree upon 15 things that we want to share with the larger group as 16 to how we're moving forward.

17 Next slide. So, I've got one more slide 18 here.19Priorities. And these are the NRC's 20 priorities as to how this Working Group moves forward.

21I kind of want to give you the big picture of this.

22 This is not a short-term thing that we do in six 23 months or a year, but it's a little bigger.

24 Obviously, the first thing we want to look 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 107at is the aluminum HEAF. That's the driver for us in 1 the GSI program, and we want to solve that portion 2 first.3 You saw the matrix that Gabe and Nick used 4 earlier this morning, how we laid all the testing out.

5 And we want to pick those aluminum pieces up first and 6 get this Working Group going.

7 Again, the decision from the GIRP was that 8we want to go out with the most accurate model. We 9 don't want to go out with something that's very 10 conservative, do some more research, have it combine, 11 and have people do the work two and three times, but 12 to come out with the most accurate, realistic model as 13 possible. That's the goal that we're striving for.

14It's a timely goal. If we get our testing 15done, we get this group going, we believe we can 16achieve it, and we have the schedule. So, that's our 17 first priority.

18 You also know that the way we originally 19 started this program was with an international 20program. HEAF is not unique to the United States 21nuclear plants. It's worldwide. We see these, 22 approximately 10 percent of the most significant 23 events, we see with our international partners are 24HEAF. There's a lot to learn, and everybody pretty 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 108 much understands that.

1 So, the OECD/NEA group, we're in the 2 process of signing out the agreement now, where we 3 want to go forward with the second form of testing.

4 And this will do a lot more of the copper and it will 5 be a lot more confirmatory, and again, bring a lot 6 more of the realism into the total understanding of 7 HEAF.8 As a matter of fact, the German equipment 9 has already been shipped, and we have already received 10equipment from Germany. So, there's a lot of interest 11 from our international partners to work on this.

12 Our goal with our international partners 13 was to keep this Working Group going because they 14 would have momentum, they would have understanding, 15 they would have experience, and to look at refining 16 the other models, again, the Appendix M, for the 17 copper and the different conditions.

18 We could, then, take that and put it out 19 for public comment and get our OECD partners, our 20 experts in the foreign countries, and let them do a 21 peer review of it, so that we could have a good, high-22 quality product.

23 So, that's kind of the big picture looking 24 forward, that we go for the aluminum piece first, and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 109then, we work into the second phase of testing with 1 the OECD and the NEA.

2 Again, the whole goal at the end of the 3 day is to improve the realism and fidelity of our 4 models.5Kelli worked very closely with us. Kelli, 6that's kind of the end of my presentation. If you 7 have anything you'd like to say on the side of EPRI or 8 anything, this would be a good time.

9MS. VOELSING: Are you saying you want me 10 to give my presentation now?

11MR. SALLEY: If you're ready. Is it on 12 the Working Group?

13 MR. HAMBURGER: Sure. That's fine.

14 Before you give your presentation, does 15 anybody have any questions for Mark?

16 MS. ANDERSON: I have a question.

17 MR. HAMBURGER: Oh, okay.

18 MR. SALLEY: Okay.

19MR. HAMBURGER: And, Mark, would you mind 20 just telling people where they can find the current 21 version of the Charter?

22MR. SALLEY: Yes. If you look at the 23 meeting notice today, the one -- we updated it last 24week. It's a two-page Charter. It's a link at the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 110very bottom. You can just load it. It's an EPRI 1 Dropbox, is it?

2 MR. HAMBURGER: I believe it's a link to 3 the EPRI box.

4MR. SALLEY: Box, and it's in there, a 5copy. Again, it's a draft. So, we'll take any 6 comments/suggestions.

7 MR. HAMBURGER: Victoria?

8MS. ANDERSON: So, this is Victoria 9 Anderson with NEI.

10That was actually my first question. You 11 said that stakeholder input on the Charter was 12welcome. And I checked out the meeting notice this 13 morning, and I don't think I noticed that link. So, 14 even if it is available, maybe there's a way we can 15better publicize it. Maybe make it available in 16 ADAMS, if we really want to get as much public 17feedback as possible. I mean, I was hunting for it 18 and I couldn't find it.

19 MR. HAMBURGER: I believe it's a link at 20 the bottom of the meeting description.

21MS. ANDERSON: Right. But, I mean, again, 22 I was motivated --

23MR. HAMBURGER: I'll put that down as --'

24 MS. ANDERSON: -- to try to find it, and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 111 I still struggled.

1 MR. HAMBURGER: Okay. So, I'll put that 2 down as an action item for NRC.

3MS. ANDERSON: If we're really looking for 4 broad stakeholder input, that's not really going to 5 help it along.

6MR. SALLEY: The millennials got me on 7 that one, Victoria.

8 MS. ANDERSON: What's got you?

9MR. SALLEY: The millennials got me on 10 that.11MS. ANDERSON: I'm not a millennial. So, 12 I can't help you there.

13 MR. SALLEY: My millennials.

14 MS. ANDERSON: Okay. They'll help you.

15MR. HAMBURGER: And of the documents, 16 including the Charter and the presentations from today 17 will be packaged up and placed into ADAMS and made 18 publicly available.

19MS. ANDERSON: Okay. So, I think the 20 sooner that can be done with the Charter, the better 21 to help with soliciting feedback.

22MR. GARDOCKI: And we'll try to put stuff 23 like that on the GI Dashboard.

24MS. ANDERSON: Excellent. That's an 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 112 outstanding idea.

1MR. GARDOCKI: So, that's a great place 2 for all this information. Okay?

3MS. ANDERSON: Outstanding idea.

4 Definitely fully support that.

5 How will other stakeholders and the public 6be able to interact with this Working Group? Like 7 will information come out of this group? Will there 8be draft reports put out? Will there be public 9 meetings on where they're going?

10MR. SALLEY: Yes, we can decide that 11however it's best for everyone. And again, I'll work 12 with EPRI and Kelli. I think that once we maybe get 13 a product that we think is far enough along would be 14 a good time to put a draft report out, give folks 30 15days to comment on it, whatever. If we see 16 significant comments, we could hold a public meeting 17 and discuss it.

18 Or, again, how we talked with the pilots, 19 and I think that the Working Group itself could do 20 some presentations as to what their logic was, how 21they came up with it, and what the intent was. So, we 22 can do that in a public forum, if you like.

23MS. ANDERSON: Great. You mentioned 24 Kelli, and she showed up behind me.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 113 (Laughter.)

1MS. VOELSING: Yes, I was going to say, 2 the names that are on the list for the Working Group 3predate me or my involvement. And to your point, 4 Mark, if there's other industry stakeholders and 5 experts that have technical expertise to contribute to 6 the Work Group, I would think we would want to 7consider their participation. So, there should be 8 room for that.

9 MS. ANDERSON: Okay. And just one other 10question or I suppose a question/comment hybrid is:

11 you underscored the need to have the pilots to support 12 this Working Group and the pre-GI resolution. Right 13 now, I think plants would be hesitant to volunteer for 14 that because we heard this morning is we're not 100 15 percent sure what we're going to be evaluating or what 16 we're looking at, or how to deal with the fragilities, 17 or how to deal with this or how to deal with that.

18 The licensees in the audience can correct 19 me if I'm wrong, but, until some of that is better 20 defined, I don't think you're really going to get a 21volunteer. So, I think the sooner we can better 22define some of that, and we can get some realistic 23 inputs, the sooner we'll be able to get some pilots to 24 volunteer.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 114MR. MELLY: Yes, we absolutely understand 1that. And the Working Group is currently also working 2 towards a pilot plant reso urce need or resource 3 allocation, what would be expected of the pilot 4plants. We drafted that late at the end of last year, 5 and we're working that through the Working Group to 6 say, would this be the type of resources that any 7 pilot plant would need to allocate to doing this?

8 What type of information are we looking for? How it 9would move through the process. So, we're still 10 fairly early in that step, and that draft is being 11 shared within this Working Group to see, is this the 12 type of information we need to collect and how would 13 it move forward?

14 MS. ANDERSON: So, I think you correctly 15 identified the resources as one specific concern the 16 pilot plants would have. I think another concern we 17 need to pay attention to is whether or not they're 18 going to be asked to consider information that they 19 believe, based on OE, to be unduly conservative.

20 I don't think a plant is going to 21 volunteer to put their information out there to be 22 evaluated with information or analysis based on 23 extremely conservative testing or results, to possibly 24come up with erroneous conclusions. I just can't see 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 115 a plant being willing to do that.

1MR. TAYLOR: Yes, I agree. I think one 2 thing from other projects we worked on, such as the 3 spurious operation one, where they basically use data 4 from all the tests, we use operating configurations, 5 and then, the experts' judgment to modify or adjust 6the data. So, I agree, we want to get a realistic 7 model to characterize the risk. And I think letting 8 the Working Group work through that process should be 9 able to get us there. But we're too early to --

10MS. ANDERSON: And I think that 11 underscores the need to have continuing public 12 interactions, so that potential pilots have more 13 confidence in that. And then, there's potential for 14 them to volunteer.

15MR. HAMBURGER: Thank you. Any more 16 questions for NRC staff about the NRC/EPRI MOU or the 17 Working Group?

18 (No response.)

19MR. HAMBURGER: Any questions on the 20 webinar?21 (No response.)

22MR. HAMBURGER: Okay. And just a note for 23 those on the webinar. Before the end of the day, we 24 are going to open up the phone lines to make sure 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 116 nobody has any questions that they haven't put down in 1 the chat.2 All right. We are now officially in the 3public question and comment session. We have two 4 prepared presentations.

5 Kelli, would you like to go first?

6Okay. The first presentation was provided 7 to me by Kelli from EPRI.

8 And you can use the podium, and I can step 9 through. Or, if you want to take the buttons?

10MS. VOELSING: First of all, thanks for 11the opportunity to comment. My name is Kelli 12 Voelsing. I'm the RSM Program Manager at EPRI.

13 So, we have been participating and 14 following the development of what's going on with high 15 energy arcing faults and in preliminary discussions of 16 the Working Group Charter with Mark and his team, and 17 with Mike Cheok and Mark Thaggard as well.

18 I will say that at a high level I think we 19 are aligned on the tasks in the Working Group Charter, 20 the issues that need to be worked through in order to 21 have a realistic understanding of the risk and the 22 potential contribution to risk from aluminum. But I 23 think there's significant work and technical 24 challenges that we're going to have to work through in 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 117 order to get to that answer.

1 So, maybe perhaps for those of you that 2 don't know, my technical lead on fire PRA is out right 3 now on a medical leave, having just had a baby. And 4 so, I have been given the opportunity to learn a whole 5 lot about high energy arcing fault in a very short 6period of time. And the analogy that keeps coming in 7 my head, as I think about these, I thought it might 8 resonate with some of you all, which is, when we see 9the videos, when we see the pictures, when we talk 10 about these enormous volted three-phase long-duration 11 faults, I absolutely understand how the first reaction 12to seeing those is obviously one of concern. That 13 could be a significant issue with significant damage 14 at a nuclear plant.

15 But I think back to loved ones -- and 16 hopefully, none of you -- but if you've been given a 17 diagnosis and you hear the words "You have cancer" 18 from a doctor, the first thing that's going to go 19 through your head, oh, you know, invasive surgery.

20 We're going to cut out large portions of tissue and 21 poisonous chemotherapy and radiation, and obviously 22 the worst-case scenario. It's very easy to envision 23 that.24 But I think very quickly you get into the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 118 details, and you start hearing from the doctor about 1 locations and size and phases and genetic markers of 2 the tumor, and all of the discriminating factors that 3go into the appropriate treatment of a cancer. And I 4 think what I've learned about high energy arc fault is 5 that there's a lot of those discriminating factors in 6 not only what we see in the testing, but what we see 7 in the OE and in the experience of our plants as well.

8 And just like proceeding with a worst-case 9 cancer treatment, sometimes proceeding with a worst-10 case treatment or a one-size-fits-all model or 11 approach to understanding HEAF would ultimately do a 12 disservice to the patient; and that for the best 13 outcomes, we have to understand those factors and we 14 have to take them into account, as we understand both 15 the risk and the mitigation plans for how we deal with 16 those.17 So, I guess I was just saying, it keeps 18 coming into my head that I understand why the concern.

19 And obviously, these worst-case HEAFs would be a 20 significant or could be a significant safety impact, 21 depending on the configuration, the location of the 22 plant, the equipment that could in a zone of 23influence. But there's a lot of other factors that we 24 have to consider before we get to that conclusion.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 119 So, if we could go to the next slide?

1 I'll start here, and I think we've already 2heard this today. But I just wanted to put in context 3 that, currently, the Bin 16 treatment in NUREG-6840 or 4 EPRI 1011989, the Fire PRA Methodology in Appendix M, 5 that we've been referring to is a one-size-fits-all 6for everything in Bin 16. There's one zone of 7influence. There's one initiating event frequency.

8 And it's certainly easy to imagine that, if you double 9 or triple or quadruple the zone of influence, you're 10 talking about taking out entire rooms or large zones 11of potentially impacted equipment. So, we have to 12 consider what all goes into Bin 16 and is it really a 13 some-size-fits-all approach that we should be 14 pursuing.15 So, if we'll go to the next slide?

16 Some differentiating factors that may come 17into play here. We do see distinct differences in the 18 types of events, the damage profiles of events between 19 medium voltage and low voltage equipment. I'll talk 20 some more about that.

21The location. It's probably true that 22 there is aluminum out there in bus ducts, whether non-23segregated or iso phase bus ducts. But that equipment 24 is outside or in the turbine deck, and doesn't impact 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 120 safety-significant equipment, it may not still be a 1 significant contributor to risk.

2 We see differences within Bin 16 between 3 switchgear and motor control centers, bus ducts, that 4 all of those have different probabilities in terms of 5 is this likely to be a long-duration or short-duration 6 event; is it likely to be a flash or a blast versus a 7HEAF? And I think we can start to draw some insights 8 there.9 Source of the fault may be another 10 important consideration that we haven't really talked 11about yet. We saw a picture of the wire that was 12 being used to generate the fault on the aluminum 13 busbars, which would be, obviously, on the switchgear 14on the load side of the breaker. Whereas, a good 15 portion, more than half of the industry experience is 16 that the fault is generally generated on either the 17 breaker side or at the stabs, which is generally not 18where the aluminum is present. So, that may be an 19 important consideration in terms of testing, in terms 20 of where do we put the fault.

21Electric configuration. At the time of 22 the white papers last year, we had reviewed a good 23portion of the industry. We've now reviewed all 103-24 104 units to determine which units have safety-related 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 121buses or buses that are subject to unit-connected 1 generator design without a generator breaker, such 2 that you would have a long-duration fault that has to 3spin down. And that applies to some plants. It also 4does not apply to a lot of plants. And so, I think we 5 need to consider that as well.

6Material. Obviously, the reason we're 7 here today is the presence and location of aluminum.

8 And then, again, operational issues. We 9 would be remiss to not acknowledge that the best way 10 to prevent the risk of a HEAF is to not have a HEAF.

11 And although there are not one smoking gun, we do see 12 in the OE that a large majority of HEAFs have, 13 although not the same preventable cause, they do have 14 preventable causes that relate to cleanliness, 15 maintenance practices, human performance.

16 So, next slide.

17Again, we are in alignment on the major 18tasks of the Working Group Charter. We think those 19 are the right tasks that need to be worked through and 20appreciate the commitment from Mike Cheok and Mark 21 Salley and their teams to work through that in a 22 methodical fashion that considers the spectrum of risk 23 and the probabilistic aspects of that spectrum of 24 events.25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 122 So, the first task is to improve the risk 1 models, supported by all available information, be 2 that testing, data, whether type testing, 3 manufacturing data or manufacturers' data, or the 4 current test programs, or other testing.

5 OE from the industry as well as expert 6 engineering judgment because the bottom line is I 7 don't think we're ever going to have enough resources 8 to test all the configurations, all the possible 9 configurations that we might like to test.

10 And then, there are basically two tasks 11that we've aligned on that support that. The first is 12 a realistic representation of the frequency of events 13 represented in Bin 16 and an appropriate modeling of 14 the various zones of influence for the events included 15 in Bin 16.

16 And then, following that, once we have 17 developed those methods and agreed and aligned on 18 those approaches, and tested them, and make sure that 19 we feel like they're practical and implementable, 20 then, obviously, that translates into updated 21guidance. So, I think, in general, we are aligned on 22 the major tasks of the Working Group. Perhaps the 23 timeline and the ability to get through those tasks I 24 think will be an evolving issue, as we really start to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 123 dig into the technical details.

1 So, I have a slide with a lot of words for 2 kind of what -- and again, I don't have answers for 3 all of these or necessarily any of these, and 4 certainly some technical experts in the crowd who can 5 support, as necessary, but we wanted to identify some 6 of what we think, in working through these tasks in 7 the Working Group, what are some of the key insights 8 that we think we need to take into account as the 9 Working Group, as we work through these?

10 First, with respect to the frequency work, 11 you've heard about binning or sub-binning or this idea 12 that Bin 16 includes multiple different kinds of 13events represented. And so, I just wanted to give you 14 a flavor of what we're talking about there.

15 So, right now, the total population in Bin 16 16 which is the frequency bin and the fire events 17database that supports the arcing fault inputs into 18the fire PRA. So, from 1964 to 2017, the total 19 population of that bin is 28 events across the U.S.

20 nuclear industry.

21 It does include a wide variety of 22 equipment. There's not a lot of events, but there's 23a lot of potential sources that come into that bin.

24So, it includes medium voltage, low voltage. It 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 124includes switchgear. It includes motor control 1centers. It includes bus ducts, and it includes --

2 basically, in the Charter we refer to it as the 3 arching fault hazard because it includes things that 4would be characterized as flashes or blasts that 5 really didn't do damage outside the component and did 6 not, essentially, have a zone of influence or create 7 a post-event fire, as well as events that are 8 definitely high energy arcing faults and do have that 9 larger zone of influence and immediate damage or fire 10 or immediate damage to components.

11 The data includes events with and without 12 post-event fires and with and without post-event 13damage. And perhaps it's obvious to say, but with 14 only 28 events, it may be that we have to use some 15 engineering judgment and for different types of 16 components or different aspects, what we actually know 17 about the probability of having a flash event or a 18blast event versus a HEAF event. Because with 28 19 events, you can only break that out down into so many 20 sub-bins. And you've got 30 bins and three-quarters 21 of them have one event or zero events in it, I'm not 22 really sure how much that tells you from a statistical 23perspective. So, we're going to have to be willing to 24 bring in other expertise and data and judgment into 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 125 that consideration.

1 But I do think, nevertheless, the OE in 2 the ISAs database and the OE in Bin 16 does give us 3 some pointers about things that are critically 4 important. So, I'll highlight a few of those here.

5 First of all, medium voltage equipment, 84 6percent of the events in Bin 16 that were damaging 7 occurred in medium voltage equipment. So, we're not 8 seeing large numbers of these events in low voltage or 9 high voltage equipment.

10 I'll also add that, for low voltage 11 equipment, there were two damaging events in low 12voltage equipment in this population of 28 events. In 13 both of these cases, they were allowed to exist for a 14 prolonged period of time because there was not enough 15 energy to trip the protection features in the design.

16 So, they simply didn't have enough current to cause a 17 trip, and that's why they persisted for a long period 18 of time.19 And so, although they're in the bin and 20 they do indicate post-event damage in some cases, we 21 need to do some more digging into what that damage 22actually was. And it appears that the damage was due 23 to a post-event fire and not related to the HEAF 24 itself.25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 126 And then, of the events with external 1 damage, the graph at the lower right of the slide 2 shows a breakdown of these events. In Bin 16, there 3were 11 of them that caused external damage. Seven of 4 11 of those were in bus ducts, either iso phase or 5 non-segmented bus ducts.

6 And so, I would point out that our 7 experience in fire PRAs tells us that, generally, due 8 to the location of the bus ducts -- you know, they 9 tend to be outside or in the turbine building, and not 10 impacting safety-significant equipment -- so, they 11tend to be lower contributors in the fire PRA. So, 12 more than half of those causing external damage would 13 have been in that type of application.

14 So, if I can go to the next slide?

15 Similarly, some early thoughts or insights 16 related to the zone of influence work that we've been 17referencing. And I think we've pointed to this as 18 well. Currently, the zone of influence that is used 19 is based on one event that at the time it was created 20was kind of the worst-case known event. It was a 21 SONGS event, and it's the basis for the 3x3x5 zone of 22 influence that's currently used.

23 So, just consideration of the test 24 protocol that we're involved in now and trying to come 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 127 up with a zone of influence versus what has been done, 1 it's important to recognize the differences in those 2 approaches if you're trying to come up with a 3 reasonable delta CDF.

4 And then, secondly, to date, small-scale, 5 no large-scale, test results on HEAF or the byproduct 6 characterization have been available for analysis.

7 And so, to fully have a robust set of comments about 8 future test plans and the data that needs to be 9 collected to make sure that we can make good 10 decisions, we don't really feel like we can do that 11 until we've been able to analyze the existing data.

12 So, useful insights which might play into 13 that feedback on future testing related to zone of 14 influence work is, first of all, aluminum oxide is an 15 excellent insulator material with very low 16conductivity. And aluminum in most situations will 17 oxidize almost immediately to aluminum oxide.

18 And so, one of the reasons we're 19 particularly interested in the small-scale or material 20 characterization properties is because, if the white 21 dust or very fine particles that we're observing at 22 distant locations are primarily aluminum oxide, and 23 they're not conductive, it may be that they don't 24 really -- you know, they may be very fine and spread 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 128 easily like volcanic dust, but if they don't 1 contribute to damage, then that may not constitute a 2 zone of influence necessarily in a PRA.

3 Finally, we know from testing and OE that 4the arc will travel away from the source. And so, in 5 the testing that's been done, as we said, we induced 6 the arc at the switchgear, at the busbars, and it's 7 going to immediately blow out the back of the cabinet.

8 And if you put cable coupons directly in front of 9 that, is that -- we've talked about the location of 10 the arc and the potential impact that that could have 11on the testing. But also, when you're trying to 12 understand the fragility of components, it's important 13 to represent cables and cabinets and other SSEs in a 14 manner that's consistent with how they might be 15 configured in the plant.

16 Again, just the fourth bullet there is 17kind of a similar concept.

I know some of the OECD 18 testing on bus ducts, they, I believe, had the bus 19 duct on the ground. And so, if you're talking about 20 a conical zone of influence, which is what we 21 currently use in the fire PRAs for bus ducts, well, if 22 you did the testing on the ground, then you would, 23 obviously, not be able to get that downward zone of 24 influence that you would expect to see in the plant.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 129 And that wasn't part of the NRC's current 1 test plan, but I just point that out as indicative of 2 the kind of things we need to make sure that, when 3 we're configuring this equipment and collecting test 4 data, that the data we're collecting is representative 5 of the way the equipment would be installed and 6 operated in the field.

7 And then, we've alluded to this as well.

8 Bin 16 does include significant frequencies that 9include arc blasts and arc flash events. Of those 28 10 events, we have looked at them, and at least 10 of 11 them -- there's a few that we want to understand more 12 about -- but at least 10 of them would be classified 13 as flashes or blasts, which, per the definitions 14 proposed by the NRC, would cause no damage external to 15 the component, and therefore, essentially, a 16 negligible zone of influence versus the one-size-fits-17 all zone of influence.

18Early insights on plant risk. I think 19 this is going to be a challenge for the Working Group.

20 Although EPRI is not involved in the generic issue 21 process, it's certainly something we're going to have 22 to be able to address as part of the generic issue.

23 This idea of an average plant risk is an 24 elusive concept, especially in this situation, because 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 130 there are some drivers, some designs, some risk 1 factors that would certainly make you more susceptible 2 to this type of risk, the most obvious being, do you 3 have aluminum installed in these components or not?

4 A good fraction of the industry may not have aluminum 5 or may have aluminum only in bus ducts where it's not 6 particularly significant to risk.

7 A review of all of the U.S. units shows 8 that 70 percent of plants are not susceptible to a 9 long-duration generator-fed fault on a safety-related 10bus. So, we've identified this as one of the primary 11drivers. You know, where can you get a long-duration 12fault? When there's only one breaker and that's where 13 the fault occurs, then you have to wait for the 14 generator to spin down while you're still feeding that 15fault. But 70 percent of the U.S. industry would not 16 have a safety-related bus that is even susceptible to 17 that kind of fault.

18 Even when the long-duration generator-fed 19 fault can occur, the following factors must be 20considered. So, we've talked about the voltage decay 21as the generator spins down. And I'm glad to hear 22 that we're looking at how do we make that in future 23 tests reasonably represented.

24 The presence or absence of aluminum and 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 131 the proximity of aluminum to the fault location, and 1 then, other equipment in the zone of influence, and 2 the actual fragility of that equipment.

3 With respect to the bus ducts, I will say 4 our not 100-percent survey of the U.S. industry, but 5 our reasonable survey of the industry suggests that 6it's not rare to have aluminum in bus ducts, mostly 7due to weight and cost. So, we probably are going to 8 find aluminum there, but the risk is dependent upon 9the SSEs within the zone of influence. And aside from 10 some spatial considerations at very specific plants, 11 it's generally a low contributor to risk in most fire 12PRAs. So, having a larger zone of influence in those 13situations is probably not going to make much of a 14 difference.

15Testing considerations. We talked at the 16 beginning about differentiating factors, the breaker-17initiated versus switchgear-initiated fault. And I 18 have a side views of a breaker, and in the NRC's 19 presentation they showed the shortening wire across 20 the switchgear busbars that they used to initiate the 21 fault.22 I will say, we found that more than 50 23 percent of the experience in the industry is that it 24 is the moving parts where you rack breakers in and out 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 132 where the fault is likely to initiate, either due to 1 maintenance or human performance, or other causes.

2 But it generally will tend to be on the breaker side 3or at the breaker stabs or where those moving parts 4 come together.

5And those parts of the breaker are made 6 out of copper, and they travel through some porcelain 7 bottles and through a torturous path, through some 8 more porcelain components before they move into the 9 segment of the cabinet where the switchgear busbars 10 would be.11 And so, we don't know, but it could be 12 important to consider that because, you know, if the 13 fault initiates there, it has to sustain itself and 14 move far enough along, and continue to sustain energy 15 in order to make it to the aluminum busbars, and then, 16 still have enough energy to cause this larger-than-17expected oxidation. That may be important to 18 consider.19 I think we've touched several times on the 20data for evaluating damage. Just the mere observance 21 of damage to instruments at 3 feet or presence of 22 substance at a distance from the test does not, in and 23of itself, dictate a zone of influence. To agree with 24 the NRC, that's not an easy question to answer. You 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 133 know, how are we going to take these -- you know, what 1instrumentation, what equipment? How are we going to 2take measurements? How are we going to know what part 3 of that is due to aluminum versus everything else?

4That's part of the switchgear. How are we going to 5 have a baseline to compare to in order to come up with 6a zone of influence? I think we have to consider that 7 as part of the design of the future test program.

8 And I am personally not familiar with the 9 Sandia model. So, I don't have any comments on that 10 model or the fidelity of the model, or the ability of 11 the model to predict damage, and therefore, a 12 representative zone of influence.

13 So, target selection is critical to assure 14 the results are representative of plant equipment.

15 Usually, in a switchgear room you're going to have 16multiple cabinets. And therefore, does the byproduct 17 penetrate through the cabinet? Does it cause damage 18inside another cabinet? Does it move through the 19 louvers? That's a little different than just having 20 a rack directly in front of the blast.

21 Actual location and data collection to 22assess impact on operation. To date, the OE does not 23 suggest large-scale deposition of byproducts or damage 24outside the cubicle. So, when there has been 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 134 byproducts, you know, in general, we have been able to 1 wipe them off and close the cabinet and everything 2still works. So, we need to make sure that we're 3 using representative equipment to assess fragilities 4 and zones of influence.

5 And then, clarity on what conductivity and 6 what material is being measured will be a challenge 7 for the test program. And along with representative 8 conditions that lead to the longer-duration faults, 9 this is why I asked earlier about the extrapolation of 10 an 8-second fault in low voltage equipment to be 11representative of high voltage equipment. And so, 12 this really gets to that point.

13 In medium voltage equipment, we see very 14 clearly in the OE that the dominant contributor there 15 is this unit-connected design where the generator has 16 to wind down as the voltage collapses and continues to 17 supply energy into the fault for a long-duration 18 period of time.

19 For low voltage equipment, what we see as 20 the contributor most frequently there is it's a long-21 duration event simply because there's not enough 22energy to trip the protection scheme. And therefore, 23 the current is just below what you need to cause a 24trip in the circuit. And so, it just continues to 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 135propagate. So, there may be long-duration arcing 1 fault events, but they may not have the high energy, 2 because, by definition, if they did, the equipment or 3 the protection system would be expected to trip.

4 And so, I think you can see how what leads 5 to a long-duration fault in low voltage is a different 6 condition than leads to a long-duration fault in 7 medium voltage. And so, the extrapolation of one to 8 the other I think would have to be reviewed by the 9 Working Group to make sure that we're comparing apples 10 to apples in our extrapolation.

11 So, to kind of circle back to where we 12 started, at least our hope, like what the NRC said, is 13 that we come out of this with a refined Bin 16 14 treatment and supplemental guidance to update 15 NUREG-6850, and an acknowledgment that all arcing 16 fault events are not the same, and that risk should 17consider the probability of various types of events 18and associated zones of influence, neither of which 19 are probably a one-size-fits-all.

20 In the diagram there, maybe the cabinet in 21 the middle is a low voltage cabinet and probably is 22 most susceptible to high energy -- or not high energy 23-- but an arcing fault with a relatively or non-24existent zone of influence. And maybe there's some 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 136probability of a larger zone of influence. And then, 1 maybe it's the bus duct with aluminum that has a 2 larger zone of influence because it does have a low 3minimum. And we do see in the OE damaging types of 4 HEAFs occurring in that type of equipment.

5 So, this is by no means intended to 6illustrate the answer. Just that we think that 7 there's a nuance in the answer, and that we hope what 8we end up with at the end of this is a reasonable 9 representation based on all the input and the expert 10 judgment of the probability of the various types of 11events represented in Bin 16. That's kind of 12 component- or application-specific. And then, also, 13 a representation that zones of influence for each of 14 those probabilities is also probably variable.

15 So, with that being said, we look forward 16to continuing to work with the Working Group. As I 17 said, it's been a good collaboration on the Charter, 18 making sure that we had the right tasks identified in 19 order to get to a technically-solid basis for 20 improving our state of knowledge and our modeling of 21 the risk of high energy events or arcing fault events.

22 MR. HAMBURGER: Thank you, Kelli. Would 23 you be willing to answer any questions from the 24 audience, if they have any?

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 137 MS. VOELSING: I'll try or --

1 MR. HAMBURGER: Okay.

2MS. VOELSING: -- I'll defer to the 3 technical experts. But sure.

4 MR. HAMBURGER: Okay. Any questions for 5 Kelly from the room?

6 (No response.)

7 MR. HAMBURGER: From the webinar?

8 (No response.)

9 MR. HAMBURGER: Okay. Thank you.

10 And I have one more presentation that was 11 provided to me in advance by Victoria Anderson from 12 NEI.13 And after this presentation, the floor is 14 open for any questions that anybody might like to ask.

15 MS. ANDERSON: All right. So, we wanted 16to talk a little bit about the context of the 17 technical information that this effort is gathering 18 and the operating experience and the technical 19 information we have so far, a nd put that in the 20 context of the risk evaluations that are done as part 21 of the pre-GI process.

22 Because any decisions associated with the 23 generic issues process are designed to be risk-24 informed, and really the specific thing we're looking 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 138 at here, as I'll get onto in the next slide, is the 1 change in core damage frequency due to potential 2 aluminum HEAFs needs to be treated as the core damage 3 frequency due to aluminum HEAFs minus the core damage 4 frequency baseline. And I think one thing we talked 5 about a little bit this morning was that we aren't 6 really sure yet how we're defining that baseline, and 7 it's very important that we define it appropriately.

8 Next slide, please.

9 So, most past generic issues have been 10 dispositioned according to delta CDF instead of CDFs.

11 If you look at this chart here that talks about which 12 issues you exclude from further consideration and 13 which ones you continue, the X-axis is core damage 14 frequency and the Y-axis is change in CDF.

15Based on our knowledge of the fire PRA 16 CDFs out there in the industry, which would really 17 tell us about our HEAF baseline, we believe the 18 results for this pre-GI evaluation would be in this 19 green region. We don't think anything would be over 20 1E minus 4 on core damage frequency. So, therefore, 21 the determination is going to be based on delta CDF, 22 which really underscores the importance of properly 23 defining that CDF baseline.

24 In either factor in that determination of 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 139 the delta CDF, any conservative assumptions could 1result in an inappropriate decision. It could wind up 2putting you in the incorrect region. And if we try to 3 make a decision based on a single assessment of the 4 baseline risk or we just use one specific event, 5 initiating event, frequency, or consequences, we'll 6 wind up with an inaccurate and unrepresentative 7 answer, which is not what you want in a generic issue 8 program.9 Next slide, please.

10 So, this chart gives sort of a pictorial 11 of where we think we might be and what kind of 12 decisions we might be making. So, I'll just give an 13 explanation of what the different dots are for those 14 in the room.

15 The orange dots are perhaps guesses of 16 where we think we might be, based on some of the 17 initial bounding testing that's been done. The blue 18 dots are where we think perhaps realistically some of 19 the plants may lie in that delta CDF versus CDF region 20 if we applied some of the insights from the tests done 21to date. And then, that green dot is perhaps what we 22 would decide our assessment would be, based on just 23 those orange dots, if we aren't really adequately 24 considering all the input from various plants.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 140 So, there's really two things to consider 1from this chart. The first thing is, should we be 2 making determinations based on just a couple of small 3 samples interpreting data from bounding results so 4 far, when that may actually not be representative of 5what we see in the industry? And the other 6 consideration is, are those orange dots themselves 7based on conservative assumptions? And perhaps we 8 don't have anything outside of the excluded region.

9 I'm not saying that these are the answers, but this is 10 a stylized representation of where we may be finding 11 ourselves and underscores what it's very important to 12 make sure that we are accurately assessing both the 13 baseline risk and the potential delta risk from 14 including aluminum.

15 So, moving on to the next slide, we have 16 been talking about this in terms of, well, what have 17we done with past GIs? It wound up being a matter of 18 delta CDF.

19 So, if you look at what we did with the 20 seismic GI, when that came out, we didn't even have 21 the amount of information relative to seismic PRAs 22that we have relative to fire PRAs right now. But, 23 even with that limited risk-informed approach, there 24 was a pretty high variability in the decisionmaking 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 141 that we came up with.

1 So, again, if you go back to that previous 2 chart, even without detailed information, we've found 3a lot of variability. So, we anticipate that you 4 would find, with additional plant detail, you would 5 continue to find more variability.

6 So, we believe that it's extremely 7 important that we include realism in the assessment 8and in resolution of the pre-GI. It's important to 9 make sure that we're really isolating the specific 10 impacts associated with the inclusion of aluminum 11 because that's a scope of this GI, and appropriately 12 shore up what that CDF for that baseline is, so that 13 we can accurately calculate a delta CDF and accurately 14 assess this issue.

15 We believe that a lot of plants may not 16require action. We think a lot of plants will fall 17 into that region to exclude from further 18consideration. If we have a properly-conducted pilot 19with proper insights and proper consideration of OE 20 and properly-constructed tests, I don't think that 21 we'll find ourselves in that delta CDF region that has 22 us moving forward.

23 So, if we move on to our last slide, I 24 think what our priorities for the evaluations are 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 142-- and these are not just priorities in terms of what 1 we think is most important, although they are that, 2 too -- these steps really need to be taken 3sequentially for us to get to a proper answer. So, we 4 need to make sure that we fully understand and 5 evaluate data from previous tests. I think we heard 6 a few times today that some of the previous test 7 results weren't fully understood; we didn't have full 8material characterizations complete yet. And we think 9 that that's critical to get that done, so we can get 10 a better picture of what phenomena we're looking at.

11 We need to achieve consensus on the 12 technical approach for frequency and zone of influence 13work. It sounds like the Working Group will be taking 14 that on. So, we fully support continuing with that.

15 Once that's complete, I think that would 16 be the appropriate time to move on to any additional 17 testing to resolve any identified gaps, so that we 18 make sure that our tests are carefully constructed to 19 close those gaps.

20 And finally, complete a realistic 21 evaluation of the plants using all that test data in 22 OE.23 So, I think just one more thing I want to 24 underscore in my presentation is that it was mentioned 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 143 earlier that the industry comments were dispositioned.

1 We turned in comments following the workshop last 2April. I know that there still have been some changes 3to the test plan made. But, based on the comment 4 resolution we saw, it doesn't look like the test plan 5 would really support a fully realistic evaluation of 6this pre-GI. And we do think it's important to 7 perhaps look at those comments again before we proceed 8with further testing. Because, right now, we don't 9 think that the testing plan fully aligns with what is 10 needed for a realistic evaluation.

11MR. HAMBURGER: Would you be willing to 12 answer any questions?

13MS. ANDERSON: I'll answer any questions.

14MR. HAMBURGER: Okay. Anybody in the room 15 who has questions or comments for Victoria?

16 MR. HYSLOP: Hi, Victoria. J.S. Hyslop, 17 NRR.18 You talked about the baseline and not 19really having a baseline. I know we had some comments 20earlier about 2 seconds versus a different model. Is 21 that what you're getting at when you say that?

22 Because there is a baseline, at least one that's been 23 used in regulatory analysis, and the NRC has accepted 24it, and licensees have acted on it. So, can you 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 144 clarify that for me?

1MS. ANDERSON: What I mean is, if we're 2 trying to determine what the delta CDF is for the 3 specific impact of aluminum involvement in a high 4 energy arching fault event, our baseline needs to be 5 isolating out -- needs to be defined such that we're 6 isolating out the unique impacts of having aluminum 7 involved. And right now, I don't think that we have 8 the detailed information on the phenomena associated 9 with the testing results we have. That's not really 10reflected in the baseline. If I understand, right now 11 the baseline is simply what we get from 6850 right 12now, which is based on one event. So, we wouldn't 13 really be comparing an appropriate baseline to an 14 appropriate CDF from the aluminum impacts.

15 MR. HYSLOP: So, does that mean that, if 16 you feel that the testing and the analysis truly 17 represented aluminum, that you couldn't use the 6850 18 baseline? Is that what you're saying?

19MS. ANDERSON: I don't think that you 20would be able to just take the 6850 baseline by 21itself. I think you need to do a more nuanced 22 evaluation once you have a better understanding of the 23phenomena that we're observing in those HEAF tests 24 that involve aluminum.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 145MR. HYSLOP: So, you're recommending, 1 after we understand the issue, just looking back at 2 6850 and verifying that's a suitable baseline or 3 making a decision --

4 MS. ANDERSON: Or adjusting it. I don't 5think it's going to be a matter of verifying. I think 6 it's going to be a matter of adjusting it in some way 7 to make sure that you're uniquely reflecting just the 8 impacts of aluminum.

9MR. HYSLOP: I guess that might be a 10 plant-specific analysis as well.

11MS. ANDERSON: I think you could do it on 12 a generic basis or using representative plants.

13MR. HYSLOP: Yes, that sort of gets to the 14 pilots --15 MS. ANDERSON: Yes.

16 MR. HYSLOP: -- because, earlier in your 17 slide, you talked about having not just one PRA.

18 MS. ANDERSON: Correct.

19MR. HYSLOP: And certainly we need several 20 pilots, so that we can look at this and not just have 21 to rely on one answer.

22 MS. ANDERSON: Yes, absolutely. But, as 23 I noted, I think we need to have a better idea of what 24 we're going to be asking the pilots to do before we 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 146 get any volunteers.

1MR. MELLY: Hi, Victoria. This is Nick 2 Melly in the Office of Research.

3 Can we go back to the slide with the dots, 4 the third slide?

5MS. ANDERSON: Yes. Which, by the way, 6are not representative of anything. This is a 7 pictorial -- yes, it's a cartoon.

8MR. MELLY: All right. I was kind of 9 hoping that there were numbers behind them right now, 10 but --11 MS. ANDERSON: No, no. It's a cartoon.

12 (Laughter.)

13 MR. MELLY: Okay.

14MS. ANDERSON: I mean, aside from the fact 15 that we know that almost all the fire PRAs sit in that 16 CDF region.

17 MR. MELLY: Right. I wanted to speak to 18 this, just to reiterate the fact that the pilots are 19really necessary as part of this process. The generic 20issue assessment stage needs a risk evaluation. If we 21 do not have the pilots to do that, it falls back on 22 the NRC staff in order to make approximations, to make 23 assumptions on what that risk would be at various 24 plants.25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 147 Some ways that we've done this in the past 1 are the SPAR model, and there are conservatisms within 2 and there's some limitations on the NRC staff at what 3resources we have available to make our assessment.

4 For instance, spatially in the room it's very 5 difficult to tell what other CCDP targets would be 6 involved in a SPAR model versus the wealth of 7 information that the plant PRA models actually have.

8 So, that's just I wanted to stress that the pilots are 9 very important if we're looking to get a realistic 10 representation of the increase or the delta risk.

11MS. ANDERSON: And I'll say it again, that 12 we're not really going to be able to in good faith 13 encourage a licensee to volunteer as a pilot until we 14 have confidence that the evaluation that's being done 15 is based on realistic evaluation of the operating 16 experience and realistic testing.

17MR. MELLY: And I hope that, with the 18 products that the EPRI Charter and the NRC Working 19 Group come up with in terms of frequencies, zone of 20 influence, operational experience to support that, 21 will bolster some confidence in that approach.

22 MR. GARDOCKI: Hey, Ashley, this is Stan 23 Gardocki. Oh, this Victoria. I was thinking that's 24 Ashley. Sorry.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 148 (Laughter.)

1 MS. ANDERSON: That's Kelli.

2MS. ANDERSON: Ashley is at home with her 3 baby. She's beat all of us today.

4MR. GARDOCKI: Okay. Just one comment on 5this graph. We've used it in the Generic Issues 6 Program, and people understand that it's not a go/no-7 go definitive line there.

8 MS. ANDERSON: Yes.

9 MR. GARDOCKI: It's a gradient --

10 MS. ANDERSON: It's soft line, yes.

11MR. GARDOCKI: It's a gradient line. And 12 if you go back to the PRA Reg Guide 1.197, you'll see 13 that notation in there that says you take a gradient 14 approach when you get near that line. So, it gets a 15 little bit, you know --

16 MS. ANDERSON: Yes.

17MR. GARDOCKI: It gives that go/no-go 18 approach. Okay?

19MS. ANDERSON: Yes. I should take a 20 blurring tool to that and sort of make it nice and 21fuzzy. If you're right on the line, you still have 22 further evaluation to do probably.

23MR. GARDOCKI: You can try. I'm doing 24 that in updated documents that we have --

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 149 MS. ANDERSON: Yes.

1 MR. GARDOCKI: -- to try to show that.

2MS. ANDERSON: Actually, I believe NEI put 3 in a comment to the effect that that should be done --

4 MR. GARDOCKI: Yes.

5 MS. ANDERSON: -- in NRC documents.

6MR. GARDOCKI: One question I do have is, 7we put an IN out about a year ago. And based on 8 operating in a plant, you have to take an IN that the 9 NRC puts out and evaluate for your plant to see if it 10has impact. But the NRC doesn't have resources to go 11 out and see what reaction the plants had to the IN 12that we put out. I don't know if your agency can 13 somehow get that information to see how the plants 14reacted to that IN. If they put it in a corrective 15 action base, and they say, well, does it apply to our 16 plant, does it not apply to our plant, and what action 17 plan they have and correspond to that IN, that would 18 be a resource available maybe for your reach-out to 19the plants that already out there. They don't have to 20do anything. They already have done that, or 21 supposedly have done that.

22MS. ANDERSON: Yes. From what I 23 understand, I believe most plants looked at that 24 testing and they believed that it was not 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 150 representative of their plant configurations and I 1 think determined that there was no action needed on 2 their part. So, are there any licensees with action 3 to the contrary? There are a couple of licensees in 4 the room, but I don't see any of them getting up to 5correct me. So, I do not believe any plants made any 6 physical modifications as a result of that Information 7 Notice.8PARTICIPANT: I would say, because of the 9 prototypicality of the --

10MS. ANDERSON: Yes, yes. I think that was 11the determination, was that the tests that the 12 Information Notice was based on, most plants believed 13that the data didn't have applicability to their 14plant. So, they did not take any action to make 15physical modifications. But, again, there are many 16operating plants throughout the fleet, and I don't 17 have full knowledge of what all of them have done, and 18 no licensees have come to correct me yet.

19MR. BOYCE: Hi. Tom Boyce. I'm the 20 Branch Chief for Regulatory Guidance and Generic 21 Issues.22 These might be just feel-good comments, 23 but I'm observing this and I just wanted to share 24 that, from the program perspective, having the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 151participation today is really very positive. The 1worst scenario for the GI program is that we don't 2 even have public meetings and we don't get this input, 3 and we do operate with one-size-fits-all and perhaps 4overly conservative type of outcomes. So, I think 5 this is very positive.

6 And the best thing for the program and for 7 safety, honestly, is if all those green dots and those 8orange dots became blue dots. So, we're not in it 9just to do something. We're in it to make sure safety 10 is accomplished.

11 MS. ANDERSON: Yes, we concur with that.

12 MR. HAMBURGER: Okay. Anything else for 13 Victoria?14 (No response.)

15 MR. HAMBURGER: Okay. Thank you.

16 MS. ANDERSON: Thank you.

17MR. HAMBURGER: Anything from the webinar?

18 (No response.)

19MR. HAMBURGER: Okay. So, the floor is 20 now open for anyone who would like to make any 21comments about any of the presentations we've given 22 today or anything tangentially related, HEAF-related.

23MR. BOYCE: We just got a question from 24Paul Gunther on the GI process. So, his question is, 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 152"Will the GI process look at any age-related 1 degradation mechanisms that might contribute to the 2 risk frequency for HEAF events?" 3 MR. GARDOCKI: This is Stan Gardocki.

4 As part of the Generic Issues Program 5 itself, we would look at the initiators of the event 6 and the contributors to the probability of the event 7in the risk analysis. As far as aging management 8 specifically, the NRC has its own aging management 9 program for life extension, AMR process, and they have 10 specific engineers that work on cabling, submerged 11 cabling, and they put out the report on that as 12 influence to aging management. I don't know if that 13 addresses it correctly.

14 Do you want to add some more?

15MR. MELLY: This is Nick Melly from the 16 Office of Research.

17In addition to when we do the frequency 18 work, we also trend the frequencies of fire events 19 within power plants and the most significant events.

20 So, we're trying to capture any aging-related issues 21when we're looking at the root cause of the events 22 themselves, as well as the frequency of occurrence in 23 the plants.

24 Currently, we have seen no increase in 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 153 frequency of high energy arcing fault events that are 1solely attributed to aging concerns. When we look at 2 the root cause, it's kind of all over the board of 3 different factors, such as human-related failures, 4design specifications. Aging is also in there as well 5as foreign material. It is an aspect that we're aware 6 of, and we look for it when we evaluate the frequency 7 of these events.

8MR. TAYLOR: Yes, this is Gabe Taylor, 9Office of Research. I'll just add one additional 10 piece of information that kind of goes on what Nick 11 said.12 Back when we issued the Information Notice 13last year, although we didn't do any type of formal 14 trending analysis, I did do an informal kind of back-15of-the-envelope-type evaluation. And from the data 16 that we had, the operating experience that we had, we 17 didn't see any increase, any decreasing trend versus 18 time. So, it looks like it's somewhat of a constant 19 hazard or a constant failure rate type of event.

20 MR. HAMBURGER: Any other questions from 21 anybody in the room?

22 (No response.)

23 MR. HAMBURGER: You have one more?

24MR. BOYCE: Yes. We have one more 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 154question from Rob Cavado. His comment or question is, 1"I want to reinforce the comment earlier that we need 2 a public meeting prior to finalizing the actual PRA 3 modeling to ensure the updated HEAF modeling is usable 4 and realistic." So, more of a comment.

5MR. TAYLOR: Yes, I don't think we've 6changed from this morning. I think we agree, agree 7with that. And when we get to that stage, we'll 8 probably engage the stakeholders.

9 MR. HAMBURGER: Okay. And just for --

10MR. CHEOK: Actually, I also have one 11 question. I said this morning that we need to get a 12 clarification of industry and other stakeholder 13 comments. I know that we have received an abundance 14 of comments, and I think the staff feels like we have 15dispositioned them. I get the sense that a lot of the 16 disposition is not satisfying. Put it this way.

17 So, I'm going to offer again -- I heard it 18 from Victoria again, which is what reminded me of 19this. That one of the steps that was proposed by NEI 20 was that we resolve all the comments before we start 21on the next set of testing. So, I don't want to do 22 the next set of testing and hear that we did not 23 resolve all the comments. I think that would not be 24 fair to the staff or to all of our stakeholders.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 155 So, again, I think I want to work with the 1 staff and work with NEI and EPRI on maybe a date as to 2 what we can do in terms of resolving all the remaining 3 comments, all the comments that need to be resolved 4 before we start our next set of testing.

5 MR. HAMBURGER: Go ahead.

6MS. UHLE: Okay. Jennifer Uhle from NEI.

7So, thanks, Mike. We are very interested 8in working together collegially here to address the 9issue. And we recognize the importance of a pilot in 10 determining the delta CDF. And so, yes, we hear you 11 and appreciate that.

12 But there's also the element of the 13 modeling that has to be, I think, discussed because I 14think at least what I've heard is that we'll be given 15the results of the modeling. And that, to us, is not 16 pleasing, considering the modeling -- having a 17 modeling background myself -- can go in a very 18conservative direction. It can go in a very non-19 conservative direction.

20And so, if we are going to be as an 21 industry -- you know, the generic issue disposition is 22going to be impacted by the modeling. We would like, 23 the industry would like to play a role in at least 24 commenting on the modeling approach, rather than just 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 156 the data or the results that get generated through the 1 use of the model.

2 So, the more cooperation, and cooperation 3-- obviously, NRC still maintains its independence --

4 cooperation is what I mean when I say "communication" 5and justifying a variety of assumptions. The more 6 that occurs, I think the more confidence that people 7 will have in participating in a pilot, which is going 8 to be a critical part of this.

9 But, if we are just given a model and it's 10 used, and we haven't participated or aware of how it 11 was developed, that's going to be problematic from our 12 side of the issue.

13MR. TAYLOR: Yes, I'll go ahead and 14respond to that. I think I agree with what you said.

15We need to get involvement early on. Probably I 16 misspoke earlier about not getting involvement early 17 on.18 MS. UHLE: Okay.

19MR. TAYLOR: As I said earlier, we're just 20changing approaches. Just recently, that happened.

21 So, I think once we get moving, we'll obviously be 22 working with EPRI through that format. And then, as 23 we move along, we'll identify points and try to bring 24 in other stakeholders via public meetings or other 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 157 forums.1MS. UHLE: Okay. Great. I appreciate 2 that, Gabe.

3 And so, if I can just continue and say 4 that, getting industry information to give to the NRC 5is obviously work on our part as well as your 6licensees. And so, having assurance that this 7 information is going to be used in a realistic manner 8 would also help in encouraging the industry to be more 9 willing to participate in the process.

10 Thanks.11MS. VOELSING: This is Kelli Voelsing from 12 EPRI again.

13 Mike, I really appreciate your commitment 14 to go back and make sure that we addressed and 15resolved all comments. And at least speaking for 16 EPRI, we'll make sure that we get you any that we 17 think are still outstanding in a concise format.

18 I would, however, say that I think there's 19a strong probability that there will be additional 20 comments that we want to take into account in planning 21 future test phases based on the results of previous 22 tests. And so, until we have those test results and 23 the ability to spend the time and look at them, we 24 probably don't have a complete set of comments. But 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 158 we'll certainly get you the previous ones in a concise 1 and consolidated fashion.

2MR. CHEOK: So, at lunch today, Stan 3 Gardocki came over to my office, and coming over to 4 this meeting, Tom Boyce was also talking to me. And 5they looked at the generic issue timeline, right? And 6we saw that this morning. And rightly so, the generic 7 issues staff would like to see things move on, so they 8do not drag on forever. And if it's something that's 9 a safety case, we should not take forever to resolve 10 something that potentially could be safety.

11 But, on the other hand, we also have got 12 to get it right, right? So, I think when we talk to 13 Stan and Tom, we understand that this process maybe 14 may take a little longer because we want to get it 15 right, but we still cannot push it out too far. But 16 we understand the sequence of events that you laid 17 out, and we'll talk, talk over this with the staff on 18 the effect of the test schedules, the availability of 19 the lab, and things like that.

20 But those are comments that are good 21 comments. We will take that into account.

22MR. GARDOCKI: To point out -- this is 23 Stan Gardocki -- point out exactly what this 24conversation was, we had the pilot plants. You could 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 159see on the screen the assessment plant timeline. And 1 the people on the phone, if they want to go to that in 2 my presentation?

3 We plan on completing the assessment by 4 the end of 2019, and then, rolling it all up and have 5 the assessment report done in the first quarter or two 6of 2020. It's a very aggressive schedule based on 7 getting all this testing complete.

8 Based on the comments we are having today, 9 we see the pilot plant study is being done right after 10June, July, and completed by August. And that's in 11 the summer of this year is when we're going to do that 12 additional testing with the bus ducts and the low 13 voltage. To complete the pilot plant study prior to 14 that testing is going to be a challenge.

15MR. CHEOK: Again, we will work this out.

16I think it's more important to get this right. I 17think everyone wants to get this correct. So, we will 18 take everything, all the comments we heard today, into 19 consideration, and we will work out a schedule that is 20not going to drag this thing out too much. But we 21 will work to do the right, to get the right 22 information, so that we can resolve the issue that has 23 a recommendation that could be defendable and that is 24 robust.25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 160 MR. HAMBURGER: I think we have one more 1 question on the webinar?

2MR. BOYCE: Yes, we have a brief question 3 from Mohammad Mustafa, Entergy.

4 His question is, "Do existing Appendix R 5 and NFPA 805 separation criteria remain adequate?" 6MR. HYSLOP: I don't know. We'll have to 7see what the testing does. Certainly, this generic 8 issue is looking at a PRA and looking at risk results, 9but I don't think it's limited to that. I think we 10 talked about Appendix R, the impact on that as well.

11 So, we just have to see, and I can't speculate any 12 further than that.

13MR. HAMBURGER: I just want to do one 14final sweep of the webinar for anyone who may have 15called-in only. Can we open up all the phone lines on 16 the webinar to see if anybody has any questions they 17need to ask? And if you don't have a question, now is 18 a great time to put yourself on mute.

19 Just give us a second to open up all the 20 lines.21 Anybody on the webinar with a question?

22 (No response.)

23MR. HAMBURGER: Okay. And again, if 24 anything occurs to you later, you can get in touch 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 161with anybody on the meeting notice. We will sure the 1 comment or question gets to the right place, and we'll 2 get back to you.

3 Any last questions from anyone in the 4 room?5 (No response.)

6MR. HAMBURGER: Okay. And just a note.

7 I have two actions here for me. One is to circulate 8 the Draft Charter.

9MR. BOYCE: We have, I think, one question 10from Rodney Pletz. If you have your question, you can 11 ask it now. I unmuted you. Go ahead, Rodney.

12 (No response.)

13MR. HAMBURGER: We have to open up the 14 phone lines again in the control booth. No, he says 15 we're good.

16 Rodney, do you have a question for us?

17 (No response.)

18MR. HAMBURGER: Okay. Well, if you can 19 type it, we'll answer it.

20 So, I'm going to make sure that the Draft 21 Charter gets circulated and placed in public ADAMS for 22 industry comment before that becomes finalized.

23 And I will also publish today's meeting 24 package that includes all the presentations, the 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 162 transcripts, and everything that was presented today.

1 So, if nobody has anything else, I want to 2turn it over to Mr. Michael Franovich. He has some 3 closing remarks.

4MR. FRANOVICH: Good afternoon. I'm Mike 5Franovich. I'm the Director of Risk Assessment in 6 NRR.7I am Co-Chair of the GIRP. Mike and I co-8chair the GIRP. There's a reason we've structured it 9 that way. We want to make sure that when or if this 10 issue migrates to NRR in terms of a transition, that 11 the expectations are really met.

12 Stan this morning described in detail the 13 process for the GI program. An issue migrating over 14 to NRR in the Program Office, there are a series of 15 processes that we have in the Program Office.

16 Certainly, if there's a viewpoint that there's an 17 imposition of a new requirement or a new position, we 18will go through a very structured process. Even if 19 it's a generic communication such as a Regulatory 20 Information Summary, we do have processes to ensure 21 that we're not imposing undue burdens.

22 We have revamped those processes, 23 actually, as many of us have gone through, or, 24 actually, all of us have gone through backfit reset 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 163training. And I want to emphasize that it is not 1 success for us to have incomplete information moving 2 in the regulatory program space and trying to move 3 forward with some types of expectations for licensees 4 to do more work.

5 Certainly, some of the issues discussed 6 today about risk modeling, fire modeling, the testing 7 and how representative it is, those are the types of 8 issues that would come up in our internal review in 9NRR. Because when we look at it, we actually bring 10 those types of issues to our leadership team. The 11 leadership team is looking at it from a resource 12 impact, not just a safety perspective, but a resource 13 impact, as we have to approve resources to be able to 14 move forward. And that looks at both how long those 15 processes can involve and what will be the ultimate 16 end state.

17 We've talked about a lot of testing today 18and the need for more communication. And I won't 19reiterate that. There are some very good 20 characterizations of that.

21 I think it was recognized early on in the 22 GIRP when we did the As sessment Plan Report in the 23 summer timeframe that this is living plan; there would 24 be give-and-take back and forth.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 164 If you notice in that plan, there are a 1 number of highlighted engagement points, in 2particular, with ACRS. ACRS is a vital stakeholder 3for us and they will scrutinize the pedigree of the 4 work and the fidelity, any kind of outcomes that come 5from the research activity. And also, they want to 6know which direction we'll be moving in. Those 7 processes engaging with the ACRS are very public, and 8 information is released in advance of the meetings at 9ACRS. So, there are other engagement points where 10 information will be made available to other 11 stakeholders, not just the ACRS.

12 But point well taken about more 13interaction. I think the interest we've had both here 14 in the room and on the webinar illustrates the need 15 for more dialog.

16 The other thing I want to really point to 17 is, ultimately, this comes down to a risk management 18equation. And in risk management equation space 19-- set aside the regulatory piece -- there's both 20prevention and mitigation. And I think we need to not 21 lose perspective, that if more can be done with 22 prevention in terms of looking at predictive 23 maintenance and preventative maintenance practices, at 24 least that can help shape the risk management approach 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 165 to the issue and resolution.

1We had some dialog. Kelli shared some 2 information about parsing more of the data and looking 3more closely at the OpE. But we may not be able to do 4 statistical changes based on some qualitative 5 analysis, but if there are better practices that are 6 going to be promoted, we can take those matters into 7 consideration.

8 Victoria had some graphs up with the lines 9that aren't really fuzzy, at least drawn on yours. We 10 do have some documents that do have some fuzzy lines, 11by the way. NUREG-1855 on the treatment uncertainties 12 actually some illustrative points there, how to weigh 13 those in decisionmakings when you're not clearly on 14 one side or one regime versus another regime in terms 15 of thresholds.

16 But, to that point, we would consider 17qualitative type of information there to say which 18 direction are we taking. The other thing we need to 19 keep in mind is, if there are other improvements in 20 the fleet, and we talk about FLEX, although FLEX was 21 not designed specifically for this type of hazard, 22 there is an optimization process that the industry has 23 promoted that we're well aware of and we're engaged.

24 And that's in NEI 1606 and 1608.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 166 Some plants have gone forward and actually 1 done some modifications for other hazards to mitigate 2those risks by modifying FLEX. With very subtle mods, 3 they've been showing -- granted, it's plant-specific 4 type of risk -- that there can be significant 5 reductions in risk.

6 So, irrespective of the regulatory piece 7for a moment, just keep those in mind. If you're out 8 looking at doing modifications of the plants and you 9 see somewhere where there's an opportunity here to do 10 that, it may be a great opportunity in the span of 11 doing other work at the plant.

12 But I do want to note again, when we get 13 to the transition point, there will be open questions 14 that we need to make sure we have a smooth transition 15 of the issue to the Program Office.

16 There will be a risk-informed resolution 17 of the issue one way or another. That's the climate 18 we're working in. It's a smart climate.

19 And with that -- we are well ahead of 20 schedule, actually -- I will open it up for any last 21 questions here in the room and on the line.

22MR. HAMBURGER: All right. We have one 23 question in the room, and then, we'll get to the last 24 question on the line.

25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 167MS. VOELSING: Again, sorry not a 1 question. Kelli Voelsing with EPRI.

2 Mike, I did want to give you confidence 3 regarding the holistic treatment, that at least from 4 an EPRI perspective, we've gone back and looked at is 5 there more guidance needed around these maintenance 6and operational best practices. We found that there's 7 plenty of historical documents on that research, and 8we've engaged with our advisors in both the 9 maintenance organization and the engineering 10 organizations, discussed it with executives, and we 11 highlight those in our pointers to those best 12practices. So, that is definitely work that is out 13 there in the industry and ongoing. And the holistic 14 treatment at least is not lost on EPRI, and I think 15 the industry is well aware of that through our 16 advisory structure and their interactions with NEI as 17 well.18 MR. FRANOVICH: Okay. Appreciate that.

19MR. BOYCE: We have one question from the 20webinar. Rodney managed to get his question in. So, 21here's his question. He says, "This question is more 22for participants from the industry. As we work to 23 refine and define the impact of HEAF, are any sites 24 taking action with their operators and fire responders 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 168 to train them in the potential for increased ZOI from 1these events? In other words, the survey that we did 2over a year ago gave at least preliminary 3 identification of where HEAFs with aluminum are 4located and could happen. Are any sites improving 5 their fire response training or fire preplan 6 information to account for these events? Do we need 7 any or want any industry alignment on this during the 8 interim period while the GI process continues?" 9MR. HAMBURGER: I can't make anybody field 10 that question, but if anybody would like to?

11MS. UHLE: This is Jennifer Uhle from NEI.

12 I don't have an answer to that question.

13 As with any issue, there's plant interdependency -- or 14 excuse me -- plant-specific dependencies of the impact 15of the current knowledge. But, going forward, we 16 certainly, as an industry, are getting our views 17together and ensuring that there is a common 18 understanding at both the staff and management level.

19 And we can certainly bring that up for consideration 20 on the industry side.

21MR. BOYCE: We have a comment from Rob.

22 He says that, "We already have procedures in place 23that address full compartment losses. This bounds the 24 issue." 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.(202) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433 169MR. HAMBURGER: Okay. Any last questions 1 or comments before we close the meeting?

2All right. Thank you all for your 3 participation.

4 (Whereupon, the above-entitled matter went 5 off the record at 2:31 p.m.)

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) 234-4433WASHINGTON, D.C. 20005-3701(202) 234-4433