ML19126A321
| ML19126A321 | |
| Person / Time | |
|---|---|
| Issue date: | 04/16/2019 |
| From: | Kathy Weaver Advisory Committee on Reactor Safeguards |
| To: | |
| Weaver K | |
| References | |
| NRC-0296 | |
| Download: ML19126A321 (359) | |
Text
Official Transcript of Proceedings NUCLEAR REGULATORY COMMISSION
Title:
Advisory Committee of Reactor Safeguards Digital Instrumentation and Controls Docket Number: (n/a)
Location: Rockville, Maryland Date: Tuesday, April 16, 2019 Work Order No.: NRC-0296 Pages 1-359 NEAL R. GROSS AND CO., INC.
Court Reporters and Transcribers 1323 Rhode Island Avenue, N.W.
Washington, D.C. 20005 (202) 234-4433
1 1
2 3
4 DISCLAIMER 5
6 7 UNITED STATES NUCLEAR REGULATORY COMMISSIONS 8 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 9
10 11 The contents of this transcript of the 12 proceeding of the United States Nuclear Regulatory 13 Commission Advisory Committee on Reactor Safeguards, 14 as reported herein, is a record of the discussions 15 recorded at the meeting.
16 17 This transcript has not been reviewed, 18 corrected, and edited, and it may contain 19 inaccuracies.
20 21 22 23 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 www.nealrgross.com
1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 + + + + +
4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5 (ACRS) 6 + + + + +
7 DIGITAL INSTRUMENTATION & CONTROLS SUBCOMMITTEE 8 + + + + +
9 TUESDAY 10 APRIL 16, 2019 11 + + + + +
12 ROCKVILLE, MARYLAND 13 + + + + +
14 The Subcommittee met at the Nuclear 15 Regulatory Commission, Two White Flint North, Room T2-16 D10, 11555 Rockville Pike, at 8:30 a.m., Charles H.
17 Brown, Jr., Chairman, presiding.
18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
2 1 COMMITTEE MEMBERS:
2 CHARLES H. BROWN, JR., Chairman 3 PETER RICCARDELLA, ACRS Chairman 4 MATTHEW W. SUNSERI, ACRS Vice Chairman 5 JOY L. REMPE, ACRS Member-at-Large 6 RONALD G. BALLINGER, Member 7 DENNIS BLEY, Member 8 JOSE MARCH-LEUBA, Member 9 GORDON R. SKILLMAN, Member 10 11 DESIGNATED FEDERAL OFFICIAL:
12 KATHY WEAVER 13 14 15 16 17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
3 1 C-O-N-T-E-N-T-S 2 Opening Remarks . . . . . . . . . . . . . . . . . 4 3 NEI 96-07, Appendix D - Application of 10 CFR 50.59 4 to Digital Modifications and Draft RG 1.187, 5 Rev. 2 . . . . . . . . . . . . . . . . . . 7 6 Break 7 NEI 96-07, Appendix D - Application of 10 CFR 50.59 8 to Digital Modifications and Draft RG 1.187, 9 Rev. 2 (Continued) . . . . . . . . . . . . 78 10 Lunch 11 NEI 96-07, Appendix D - Application of 10 CFR 50.59 12 to Digital Modifications and Draft RG 1.187, 13 Rev. 2 (Continued) . . . . . . . . . . . 154 14 ACRS Members' Comments . . . . . . . . . . . . 288 15 Public Comments . . . . . . . . . . . . . . . 288 16 Closing Remarks . . . . . . . . . . . . . . . 300 17 Adjourn . . . . . . . . . . . . . . . . . . . . 303 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
4 1 P R O C E E D I N G S 2 (8:33 a.m.)
3 CHAIRMAN BROWN: The meeting will come to 4 order. The meeting will now come to order. This is 5 a meeting of the Digital Instrumentation and Control 6 Subcommittee. I'm Charles Brown, Chairman of the 7 committee. ASCR members in attendance are Jose 8 March-Leuba, Dennis Bley, Joy Rempe, Pete Riccardella, 9 Matt Sunseri, Dick Skillman and Ron Ballinger. Is 10 Vesna going to be on the phone or do we know?
11 MS. WEAVER: No, Vesna's not coming --
12 CHAIRMAN BROWN: Okay.
13 MEMBER RICCARDELLA: She's in New York.
14 CHAIRMAN BROWN: Oh, that's a good reason.
15 Anyhow, representatives who are attending today are 16 Stephen Geier, Steven Vaughn, let's see, Neil 17 Archambo. Did I get that right?
18 MR. ARCHAMBO: Yes.
19 CHAIRMAN BROWN: Thank you. And then, 20 Peter LeBlond. Right there. Okay. The designated 21 federal official for this meeting is Kathy Weaver.
22 The purpose of this meeting is for the 23 staff to brief us on their review of NEI 97-06, 24 Appendix D, supplemental guidance for application of 25 10 C.F.R. 50.59, digital modifications. And Draft NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
5 1 Regulatory Guide 1.187, Revision 2, which will 2 subsequently document the results of their review.
3 The ACRS was established by statute and is governed by 4 the Federal Advisory Committee Act.
5 That means that the committee can only 6 speak through its published letter reports. We hold 7 meetings to gather information to support our 8 deliberations. Interested parties who wish to provide 9 comments can contact our offices requesting time after 10 the meeting's Federal Register notice is published.
11 We also set aside 15 minutes for spur of the moment 12 comments from members of the public attending or 13 listening to our meetings. Written comments are also 14 welcome. The ACRS section of the NRC public website, 15 excuse me, provides our charter, bylaws, letter 16 reports and full transcripts of all full and 17 subcommittee meetings, including all slides presented 18 at the meetings.
19 Today we will hear presentations from the 20 NRC staff and representatives from NEI. The 21 subcommittee will gather information, analyze relevant 22 issues and facts. And formulate for post positions 23 and actions as appropriate for deliberation by the 24 full committee.
25 The rules for participation in today's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
6 1 meeting have been announced as part of the notice of 2 this meeting previously published in the Federal 3 Register. As an add-on to the two -- to the 4 particular previous two couple of sentences, all 5 public people that are in the audience as well as the 6 staff should be aware that comments by the 7 subcommittee members themselves, individually, are 8 their opinions, not those of the full committee. And 9 are provided for staff consideration and their further 10 work.
11 Currently, we have received no written 12 comments or request for time to make any oral comments 13 from members of the public regarding today's meeting.
14 As always, we have one bridge line established for 15 interested members of the public to listen in. Also 16 the bridge line will be open at the end of the meeting 17 to see if anyone listening would like to make any 18 comments.
19 A transcript of the meeting is being kept 20 and will be made available as stated in the Federal 21 Register notice. Therefore, we will request that 22 participants in this meeting use microphones 23 throughout the meeting room when addressing the 24 subcommittee. The participants should first identify 25 themselves and speak with sufficient clarity and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
7 1 volume so that they may be readily heard. Also, 2 please silence all cell phones, pagers, iPhones, 3 iPads, et cetera. Any other electronic miscellaneous 4 inconveniences that you may own or have in your 5 possession.
6 We will now proceed with the meeting and 7 I guess I will call upon Chris Miller for introductory 8 comments and opening remarks. Okay, Chris, fire on.
9 Push the little button underneath at the bottom, the 10 green light should come on.
11 MR. MILLER: Thank you, chairman, members.
12 It's good to be back down here briefing you and we 13 look forward to the discussion today on 10 C.F.R.
14 50.59 and NEI 96-07, Appendix D, as you mentioned in 15 your opening remarks.
16 Our presentation represent NRC and 17 industry progress over a two year period. To provide 18 clarity, as industry performs, 10 C.F.R. 50.59 19 screening and evaluations for potential digital I&C 20 plant modifications. This work supports actions 21 described in the integrated action plan to modernize 22 digital instrumentation and controls, regulatory 23 infrastructure.
24 From April 2016 through 2017, the staff 25 and industry participated in monthly public meetings NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
8 1 to resolve NRC comments on draft, NEI 96.07, Appendix 2 D. In December 2017, NEI and staff mutually agreed to 3 place the review on any NEI 96.07, Appendix D, on hold 4 to dedicate resources to the issuance of RIS 2002-22, 5 Supplement 1. That's a clarification on endorsement 6 of Nuclear Energy Institute guidance on designing 7 digital upgrades and instrumentation and controlled 8 system. RIS 2002-22, Supplement 1, was issued on May 9 31st of 2018.
10 Then in July of 2018, NEI provided an 11 update to any NEI 96-07, Appendix D. In August the 12 NRC provided a set of comprehensive comments to NEI 13 and began a discipline process for cataloging and 14 tracking the comment resolution.
15 Five public meetings were held with 16 industry to resolve these comments. Over 90 percent 17 of the comments were resolved using this process. NEI 18 submitted its final revision of NEI 96-07, Appendix D, 19 to the NRC on November 30th of last year and requested 20 endorsement on January 8th of this year.
21 So we have two divisions of NRR presenting 22 before you today. My division, I'm the director of 23 the Division of Inspection and Regional Support. And 24 we also have the Division of Engineering that will be 25 making comments.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
9 1 Staff members at the table here today, 2 Mike Waters, NRR, Division of Engineering, Phil 3 McKenna, NRR, DIRS, Division of Inspection and 4 Regional Support, Wendell Morton, NRR, Division of 5 Engineering. We also have in the audience a number of 6 staff members, Tara Inverso in DIRS, Dave Beaulieu, in 7 DIRS, Norbert Carte and -- Division of Engineering, 8 Dave Rahn, Division of Engineering, Erick Martinez, 9 Division of Engineering. I see Nancy Salgado, 10 Division of Engineering and Brian Smith, back -- in 11 the back there, so we look forward to addressing --
12 presenting and addressing your questions. So thank 13 you very much.
14 MEMBER BLEY: Chris. Can I take a minute 15 before you guys --
16 CHAIRMAN BROWN: Please.
17 MEMBER BLEY: -- get started because I've 18 gotten a little confused. And this is for all the 19 staff. Don't answer this now, but as the talks go on 20 today, if you can dig into this stuff it would help 21 me. I'm going to read off a few things that caught my 22 eye.
23 In the revised Reg. Guide, it says 24 Appendix D does not replace or supersede NEI 01-01, in 25 whole or in part. And licensees have the option to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
10 1 use either that or Appendix D. Of course, when I read 2 the description in NEI 96-07, Appendix D, they talk 3 about the fact that you had trouble with submittals 4 using NEI 01-01. So I'm a little confused about that.
5 The next thing is it says -- I lost it.
6 Appendix D is applicable to digital modifications only 7 and not generally applicable to 10 C.F.R. 50.59. Two 8 pages later, it says Appendix D generally is 9 acceptable as a means for complying with the 10 requirements of 10 C.F.R. 50.59. And then there's a 11 few more statements similar to that leaves me not 12 quite sure what you're saying.
13 And then you have a section adding 14 clarifications to Appendix D. This is in the Reg.
15 Guide. As I read those clarifications, at least for 16 a first time reader, they don't clarify anything.
17 They get muddy as can be. It seems, in fairness, that 18 all of these came about from reviews of examples or 19 submittals and you're trying to preclude things you 20 didn't like. I can't tell what those are when I read 21 that, so if you can emphasis that and tell us more 22 about the problems you ran into, it would be helpful 23 because at least for me, I don't know quite what you 24 want.
25 MR. MILLER: I think we'll be able to do NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
11 1 that. As we go through our presentation we'll address 2 those issues, so I appreciate the question --
3 MEMBER BLEY: Perfect.
4 MR. MILLER: -- and we will do that.
5 Thank you.
6 And so now I'll turn it over to Mike 7 Waters.
8 MR. WATERS: Good morning. This is the 9 agenda for today's meeting. We have a full packed 10 discussion for you, starting with the 50.59, NEI 96-07 11 as -- and it funnels down to a revision to Reg. Guide 12 1.187, Revision 2, for endorsement of Appendix D. I 13 won't focus on this too much so we can to the 14 presentation.
15 Next slide please. Just for background 16 for all the members. I know many of you have seen 17 this already. This is the overarching plan we're 18 working under to a modernized regulatory 19 infrastructure with digital I&C. We list here some of 20 the key directives from the commission. Addressing 21 50.59 guidance is one of the key activities which 22 we're talking about today.
23 I'd just like to note, highlight what 24 Chris said. You know, bullets two and three, engaging 25 stakeholders in identifying common priorities and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
12 1 solutions. And focus on a couple approaches to comply 2 with requirements. This has been a focal point of 3 this effort. We've had several meetings with industry 4 to address this guidance and explore it in great 5 depths different ways to address the DD-9 criteria for 6 specific digital technologies.
7 Next slide please.
8 CHAIRMAN BROWN: Can I ask a question 9 about that? On the technical, the technology neutral 10 focus. I guess I've -- when you talk about that 11 relative to the I&C and the regulatory guides and the, 12 at least the major rule, 603.1991, I believe, when I 13 read 19 -- 603, it's about as technology neutral as 14 you can get. It doesn't tell anybody what piece part 15 to use or whether it's got to be analog or digital or 16 what have you. It's just kind of here -- overall 17 basic concepts that you've got to comply with.
18 The Reg. Guides do provide some 19 specificity but they are not necessarily from reading 20 them, necessarily not technology neutral. So I --
21 everybody keeps, you know, talking about how we have 22 this disconnect and I, for the last ten years, I've 23 been trying to figure out where the disconnect is.
24 MR. WATERS: Yeah. So --
25 CHAIRMAN BROWN: So I haven't paid any NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
13 1 attention to that. I've just tried to use common 2 sense.
3 MR. WATERS: No, that's absolutely 4 correct. The IEEE standard has incorporated a role as 5 technology neutral. It applies to analog and digital.
6 The history of where of that came from is, as you 7 recollect, we had a rulemaking activity for a latter 8 version of IEEE, 603 and in that rulemaking proposal 9 to the commission, we had recommended additional 10 requirements conditions. Some focused on new reactor 11 technologies versus operating reactors technologies to 12 shorten it. And the commission did not agree with our 13 proposed rulemaking and out of that guidance came this 14 directive, make the rules technology neutral focused.
15 And the guidance can be tailored if necessary.
16 So that was the directive of the staff to 17 kind of keep the current framework on this, in my 18 words, of being, keeping Sector 3 technology neutral.
19 So, but this is a principle we try to apply to 20 everything we do, not trying to, you know, bifurcate 21 high level requirements specific for new reactors 22 versus operating reactors and so forth.
23 CHAIRMAN BROWN: Yeah, I remember the rice 24 bowl conflicts between the reactors and operating 25 reactors. Okay. Thank you.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
14 1 MR. WATERS: No problem. Finally, just to 2 help you navigate, this is -- we have the Integrated 3 Action Plan, Modernization Plans. I know there's a 4 lot of numbers here. Highlighted in blue is activity 5 A. We're talking about the data. We'll -- Phil will 6 talk in detail.
7 Just to explain to the members, we were at 8 last year and this year. Last year we did issue the 9 RIS supplement for 50 to 59, 50.59, that was MP 1A, 10 that was complete. In parallel, we also issued 11 ISG-06, Revision 2, for licensing which we did brief 12 the ACRS members on. This year we're doing 13 endorsement review of Appendix D to further extend 14 guidance for 50.59 and in parallel, as you know, we 15 are starting work on our branch technical position, 16 7-19, for licensing, which we will brief the ACRS 17 later this year, I believe, given current schedule.
18 So these are all the -- I can work on them.
19 CHAIRMAN BROWN: I think that's -- we're 20 going to work in the fall for that.
21 MR. WATERS: Yes.
22 CHAIRMAN BROWN: Okay. Sometime, whenever 23 it's the right time.
24 MR. WATERS: So unless there's any 25 questions on this, I'll turn over to Phil to begin NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
15 1 discussion of Appendix D.
2 MR. MCKENNA: Okay. Good morning. So I'm 3 going to pick up on this slide which talks about the 4 Modernization Plan, Number 2, which is assuring there 5 is adequate guidance through Appendix 96-07 for 10 6 C.F.R. 50.59 evaluations of digital I&C upgrades. And 7 the main goal, that was to reduce licensing 8 uncertainly and clarify the regulatory process for 9 digital I&C upgrades.
10 The following is accomplished so far which 11 we actually mentioned a few times already. We issued 12 the RIS Supplement 1 back in May. We had public 13 meetings to comment on Appendix D. We developed the 14 Reg. Guide, Revision 2, which you have. And we've 15 also conducted regional inspector training for Regions 16 1 and 4 in December. And we'll hit up Regions 2 and 17 3 inspectors in June 2019. And that training was 18 focused on the RIS Supplement 1.
19 So now we'll go into an overview of 10 20 C.F.R. 50.59 and I'll do this as quickly or as slowly 21 as you would like. But this takes us back from the 22 beginning.
23 So the rule was first promulgated in 1962 24 and modified in 1968. It allows licensees to make 25 changes to their facility without prior NRC approval.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
16 1 And the entering argument into that is they must 2 maintain acceptable levels of safety as documented in 3 the FSAR. The rule was revised in 1995 and issued in 4 1999, which increased the flexibility for licensees.
5 It allowed them to make changes that only have a 6 minimal increase in the probability of consequences of 7 accidents.
8 And back in November of 2000, we first 9 issued Reg. Guide 1.187 which is the current revision 10 right now. We are soon to issue Rev. 1 which will 11 updated guidance for SONGS, steam generator 12 replacement, lessons learned.
13 NEI 96-07, which is the document that the 14 industry had written, was originally endorsed by NSAC 15 125 -- or was originally issued as NSAC 125, but was 16 not endorsed by NRC. Industry came in with Revision 17 1 to 96-07 which basically is a fairly detailed 18 document that runs you through how to do a 50.59 19 process for a modification. The applicability stage, 20 the screening process and the evaluation process.
21 Again, that was endorsed by Reg. Guide 1.187. And in 22 our endorsement statement we said that it provides a 23 method acceptable to the NRC for complying with Rule 24 10 C.F.R. 50.59.
25 Again, I mentioned that Rev. 1 will be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
17 1 issued for that Reg. Guide based on the San Onofre 2 Generating Station, lessons learned. And it will 3 clarify 50.59 guidance on departures from a method of 4 evaluation and accidents of a different type.
5 So in the 10 C.F.R. 50.59, there is 6 relationships to other licensing processes. I list 7 those on the slide here. Amendments to the operating 8 license are covered under 10 C.F.R. 50.59. More 9 specific regulations apply over C.F.R. 50.59. For 10 example, quality assurance program, security, 11 emergency planning, program changes, all fall under 12 the conditions of licenses and 10 C.F.R. 50.54(f).
13 Exemptions to the licensing are processed 14 under 50.12. Maintenance rule, maintenance activities 15 are assessed and managed under 10 C.F.R. 50.65. And 16 if there's license conditions in a licensee's license, 17 they are controlled under the license condition and 18 not under 10 C.F.R. 50.59.
19 So just a little bit of an eye chart, but 20 this is taken out of NEI 96.07 which would basically 21 run somebody through the process of how to do a 50.59 22 from when you proposed the modification and through 23 the evaluation stage. And I'll quickly highlight some 24 things on here.
25 So first entering argument in this is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
18 1 licensee decides that the modification is safe to do.
2 And then if it involves a change to tech specs, 3 automatically kicks out 50.59, does not cover any 4 changes to technical specifications. And then it goes 5 down to the next section where I talked about the 6 other 10 C.F.R. processes that apply. That will kick 7 you out of it also.
8 And next you get into the screening 9 portion of it to where you screen a modification for 10 adverse or non-adverse. If you determine that the 11 modification is non-adverse you're then kicked out and 12 can proceed on with the modification. If you 13 determine that, or licensee determines that, the 14 modification is adverse then they go into the 15 evaluation phase of 10 C.F.R. 50.59. And at the end 16 of that evaluation phase, they either determine that 17 a license amendment is required or they can proceed on 18 with the modification. And I will go in further 19 detail in this process chart as we go on here.
20 Okay. So skipping ahead to the evaluation 21 criteria. So licensee has finished the screening.
22 They determined that the modification screened as 23 adverse, meaning they need to go onto the next step of 24 50.59 to evaluate it against each criteria. I 25 mentioned that if it's a change to technical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
19 1 specifications, you're no longer in 50.59. You have 2 to come in with a license amendment to the NRC. And 3 then there's eight screening questions or eight 4 criteria for evaluation criteria which you have to not 5 meet any of these criteria to go on and proceed 6 without a license amendment. And I will --
7 MEMBER BLEY: Are those in the rule or in 8 the guidance?
9 MR. MCKENNA: They are in the rule.
10 MEMBER BLEY: Okay.
11 MR. MCKENNA: So these are rule criteria 12 and I have the subparagraphs of the rule next to each 13 of the eight criteria in the side brief. And I'll 14 read each one.
15 So the first one is a result in more than 16 a minimum increase of frequency occurrence of an 17 accident previously evaluated in the FSAR. Criterion 18 2 is result in more than a minimum increase of the 19 likelihood of occurrence of a malfunction of an SSC, 20 important to safety previously evaluate an answer.
21 MEMBER SKILLMAN: Before you --
22 MR. MCKENNA: Yes.
23 MEMBER SKILLMAN: Phil, before you change 24 that. In the meetings that you've experienced with 25 industry and in the -- particularly the input you may NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
20 1 have received from your inspectors, what has been the 2 feedback from understanding of frequency of occurrence 3 versus the likelihood of occurrence?
4 I think back in all of the years I've been 5 involved in 50.59 and the understanding of what the 6 difference is between frequency and a likelihood can 7 be a conundrum for the evaluators for the owner. And 8 I'm just wondering, what feedback do you have from 9 this latest campaign where licensees are trying to 10 answer this question and your inspection team or your 11 residents are watching this and challenging this?
12 MR. MCKENNA: So further along in the 13 brief we will discuss problems that were in the 14 industry on executing a digital modification using 15 past guidance. I don't think they resolve so much 16 around those questions. They really resolve around 17 screening and proceeding on with the 50.59 section.
18 But maybe if anybody -- Dave is raising his hands.
19 I'll pass it over to Dave Beaulieu.
20 MR. BEAULIEU: Yeah. The two concepts are 21 related. Criterion 2 deals with the likelihood of a 22 malfunction which is a likelihood of failure. And 23 thats -- you'll find that qualitative assessment.
24 That's a common theme is that Criterion 1, 2, 5 and 6 25 all deal with the likelihood of failure. The criteria NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
21 1 use different wording but -- and have a different 2 focus, but they all stem from the likelihood of 3 failure. And so when it comes to likelihood of 4 failure, likelihood to Criterion 1, frequency of 5 occurrence of an accident, well, frequency is directly 6 related to the likelihood of failure.
7 Accident deals with the likelihood of 8 failure of a piece of equipment that can initiate an 9 accident, so it's narrow in scope. But Criterion 2 10 you'll see likelihood of a malfunction of a SSC so 11 that's a little, thats a little bit broader than the 12 accident, so.
13 MEMBER RICCARDELLA: Is the key difference 14 between those two not so much frequency versus 15 likelihood, it's the rest of the sentence, right?
16 It's whether it's --
17 MR. BEAULIEU: It's the --
18 MEMBER RICCARDELLA: -- an accident versus 19 failure of an SSC.
20 MR. BEAULIEU: Right.
21 MEMBER RICCARDELLA: Right. The feedback 22 that we --
23 MR. BEAULIEU: It has to do with a piece 24 of equipment that can initiate an accident. That's 25 what -- so if it doesn't impact equipment, piece of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
22 1 equipment, they can initiate an action or transient, 2 then you pass Criterion 1.
3 MEMBER RICCARDELLA: Dennis, help me. How 4 do you differentiate between frequency and likelihood?
5 MEMBER BLEY: I dont think you do up 6 here. Likelihood is a more general concept.
7 Frequency is a particular measure. Probability is 8 another measure. If you're doing a probability, it's 9 within some particular time period or some other 10 event. I think here the way they're using them 11 they're essentially equivalent.
12 CHAIRMAN BROWN: I'll just make one 13 observation on that. I understand your confusion 14 because I had it. But likelihood to me normally meant 15 there's not a calculated frequency for the failure or 16 that occurrence. An engineering judgment based on 17 largely qualitative --
18 MEMBER BLEY: As I said, likelihood's a 19 more general concept.
20 CHAIRMAN BROWN: Exactly. And --
21 MEMBER BLEY: Frequency is one measure of 22 likelihood. There are others.
23 CHAIRMAN BROWN: But yes, if you have a --
24 MEMBER BLEY: But it doesn't necessarily 25 mean you're just being qualitative.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
23 1 MEMBER RICCARDELLA: So --
2 MEMBER BLEY: And frequency is also a 3 measure of likelihood.
4 MEMBER RICCARDELLA: So frequency's a 5 measure of likelihood?
6 MEMBER BLEY: Yeah, can be.
7 MEMBER RICCARDELLA: Okay.
8 MEMBER BALLINGER: But the way I look at 9 this is Criterion 1 relates to an accident that could 10 occur and the steps to get to that accident are 11 multiple. Whereas, the likelihood of occurrence of a 12 malfunctioned Number 2, is a malfunction of a 13 component which may result in that accident occurring.
14 Is that not correct?
15 CHAIRMAN BROWN: Or, you --
16 MEMBER BLEY: But a system or a structure.
17 MEMBER BALLINGER: Yeah.
18 MEMBER BLEY: So a system has multiple 19 components.
20 MEMBER BALLINGER: Right, but --
21 MEMBER BLEY: And it's more complex 22 calculations.
23 MEMBER BALLINGER: -- they all relate to 24 the upper level accident.
25 MEMBER BLEY: That's what they say, yeah.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
24 1 MEMBER BALLINGER: Yeah.
2 MEMBER BLEY: I shouldn't be speaking on 3 their behalf because I don't know exactly how they 4 used they words. But the way I described it, I think, 5 is generally acceptable.
6 CHAIRMAN BROWN: I interrupted Wendell, so 7 he can --
8 MR. MORTON: That's okay. I just want 9 -- to directly answer your question, during the 10 process of reviewing Appendix D, and the comments we 11 got and feedback we received, we didn't receive 12 specific feedback or any confusions between the use of 13 the phrase increase in frequency versus increase in 14 likelihood.
15 MEMBER SKILLMAN: Thank you. You've 16 answered my question.
17 PARTICIPANT: That answered the question.
18 MEMBER SKILLMAN: And I appreciate your 19 saying, stand-by because we're going to get into some 20 of the examples later.
21 MR. MORTON: Yes.
22 MEMBER SKILLMAN: I got that.
23 MR. MORTON: Yes.
24 MEMBER SKILLMAN: But I will comment, 25 having spent a lot of time at sites and having been an NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
25 1 overview positions of multiple sites to determine 2 whether or not the 50.59s were, in fact, done 3 correctly, that is -- it's a screening for a license 4 amendment, not an mini-mod process. If you recall, 5 many sites were using 50.59 as kind of a mini-mod 6 process. The individuals that were doing the 7 screening would become confused here because of the 8 potential similarity of interpretation of frequency of 9 occurrence versus likelihood of occurrence. Just like 10 we had around the table here. So I appreciate you 11 saying that wasnt on the radar screen for the 12 comments but I will be curious how you speak of this 13 as you go ahead.
14 MR. MORTON: Right. That wasn't a point 15 of confusion or concern, at least that part of the 16 phrase -- of the rule of language. It's the latter 17 part of the sentence which was more the focus of 18 concerns and then the actual result with Appendix D 19 and R.
20 MEMBER SKILLMAN: Thank you. Again, 21 thanks.
22 MR. MCKENNA: I'm back on here. Okay. So 23 again, we're -- this is Criterion 3, results in more 24 than a minimal increase in the consequences of an 25 accident previously evaluated in the FSAR. Criterion NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
26 1 4, result in more than a minimal increase in the 2 consequences of a malfunction of an SSC, important to 3 safety. Strike accident there, it shouldn't be there.
4 To safety previously evaluated in the FSAR. It was a 5 typo. Create the possibility of an accident of a 6 different type then any previously evaluated in the 7 FSAR.
8 Create the possibility of a malfunction of 9 an SSC with a different result than any previously 10 evaluated in FSAR, that's Criterion 6. Criterion 7, 11 result in a design basis limit for efficient product 12 barrier as described in the FSAR being exceeded or 13 altered.
14 And finally, Criterion 8, result in a 15 departure from a method of evaluation described in the 16 FSAR used in evaluating the design basis for a safety 17 analysis.
18 MEMBER BLEY: Could you clarify a little 19 bit, Criterion 6? I know how I'm interpreting it, but 20 I don't know if that's how you intend it.
21 MR. MCKENNA: So interesting enough, 22 Criterion 6 is what we had the most issue with --
23 MEMBER BLEY: I can understand that.
24 MR. MCKENNA: -- in Appendix D. So I'll 25 pass it on to Wendell.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
27 1 MR. MORTON: Sure. Well might I ask, 2 what's your interpretation of Criterion 6?
3 MEMBER BLEY: Well, that makes it too easy 4 MR. MORTON: Before I -- because I can 5 --
6 MEMBER BLEY: -- for you to say oh, yeah.
7 Well, to me, what this means is if I'm substituting, 8 say component A for a new component B, if component B 9 has failure modes that create different effects then 10 the failure modes in component A, I would think that 11 would come under Criterion 6.
12 MR. MORTON: Yes, correct. Now --
13 CHAIRMAN BROWN: You want to say that 14 again?
15 MEMBER BLEY: Yeah. You got a widget and 16 then you got a new better widget.
17 CHAIRMAN BROWN: Yeah.
18 MEMBER BLEY: You think it's better, but 19 the original widget could fail to deliver a voltage 20 signal somewhere. The new widget could do that, but 21 it could also have some other failure mode --
22 CHAIRMAN BROWN: But don't you have --
23 MEMBER BLEY: -- that you haven't 24 considered.
25 CHAIRMAN BROWN: That's what I wanted to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
28 1 get at --
2 MEMBER BLEY: That the first one didnt 3 have.
4 CHAIRMAN BROWN: Okay. That's what I 5 thought you said.
6 MEMBER BLEY: So it might not be worse, it 7 might be better but it introduces something new and 8 that's saying if it does, then you can't live with 9 this anymore.
10 MR. MORTON: Well, no --
11 MEMBER BLEY: You have to justify it 12 through the --
13 MEMBER SKILLMAN: But the operative word 14 there is different, right?
15 MR. MORTON: Than previously analyzed.
16 MEMBER BLEY: Yeah.
17 MEMBER SKILLMAN: Then previously 18 analyzed.
19 MR. MORTON: Correct.
20 MEMBER SKILLMAN: So you need to know what 21 it was before so that you have a comparison basis.
22 MEMBER BLEY: Oh, absolutely.
23 MEMBER SKILLMAN: Bingo.
24 MR. MORTON: Correct.
25 MEMBER SKILLMAN: Thank you.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
29 1 MEMBER REMPE: So -- well, I have a 2 different question. Go ahead and finish this topic.
3 Are you done with it? Okay. When I look at the third 4 and fourth ones, if I think about a bunch of different 5 regional offices, it sure seems like that would 6 introduce a lot of regulatory uncertainty, especially 7 the probability of an accident of a different type 8 than any other previously evaluated in the FSAR. It 9 seems like people would be looking all over for new 10 and different things. And how does -- it seems like 11 there should be some sort of headquarters oversight of 12 something like that or something so you don't have 13 people going off the deep end looking for new things 14 that are different.
15 MR. MCKENNA: So in my previous job I was 16 an inspector in the field for the past eight years.
17 And this reported to headquarters back in October. So 18 if there's questions on the criteria that the licensee 19 is screening, those questions typically come in to our 20 branch to resolve. So they speak to our branch in 21 DIRS since we're the interface with the inspectors in 22 that region.
23 Now to answer your question about 24 Criterion 3 and 4, I would say they're probably no 25 different than the other criteria if you had a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
30 1 question in the field on how the licensee is 2 evaluating that modification in accordance with those 3 criteria.
4 MEMBER REMPE: So again, there's design 5 basis accidents that are evaluated in the FSAR and 6 suddenly you've opened it up to other events that 7 aren't considered in -- that may -- again, this is 8 with 10 C.F.R. 50.59, it's not specific to the topic 9 today, but when I looked at this I was just thinking 10 man does -- how do you rein in and keep people on the 11 same path of what can and can't be considered?
12 MR. MORTON: So one of the things to keep 13 in mind as part of this discussion is the RIS 14 Supplement 1 to 2002-22. One of the clarifications we 15 made is that when it comes to the qualitative 16 assessment, we actually put guidance within that RIS 17 Supplement that kind of points you in the directions 18 of what types of digital modifications could trigger 19 Criterion 5 in terms of creating an accident with a --
20 of a different type. In terms --
21 MEMBER REMPE: So it's not a wide open 22 field?
23 MR. MORTON: It's not a wide open field.
24 MEMBER REMPE: That's good to know. Thank 25 you.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
31 1 MR. MORTON: There is specificity. We're 2 telling the licensees how you should look at your 3 proposal mod to see if there are different types of 4 configuration. You have a digital introduction to 5 software, shared resources, because of the new 6 technology. How you could trigger Criterion 5 and 6, 7 for example. And we try to leverage that within 8 Appendix D through our clarifications in the Reg.
9 Guide. So we'll get more in details in that later.
10 MEMBER BLEY: Let me take you back to my 11 opening remarks.
12 MR. MORTON: Yes.
13 MEMBER BLEY: In the Reg. Guide you had a 14 discussion that, an example 4.19 of Appendix D, they 15 said the acceptability of new area radiation monitor 16 will be dictated by their reliability which is 17 assessed as part of Criterion 2, not Criterion 6. And 18 then you objected to that and said well, you need to 19 see if there are other failure modes applicable. I 20 would have thought they would have addressed that but 21 apparently they didn't in the example.
22 MR. MORTON: I guess we're going to jump 23 ahead a little bit in terms of that discussion. So 24 that's some of the places where staff and NEI part 25 ways in terms of the Criterion 6 guide. So that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
32 1 example is when the Criterion 6 guide inception for 2 Appendix D.
3 MEMBER BLEY: Okay.
4 MR. MORTON: With that particular example, 5 it was the staffs' position that it gives the 6 impression that potential software CCA is not --
7 MEMBER BLEY: In just talking about --
8 looking at this as an example.
9 MR. MORTON: Right.
10 MEMBER BLEY: Okay. I got it.
11 MR. MORTON: Okay.
12 MEMBER BLEY: Thank you.
13 MR. MORTON: Yeah.
14 MEMBER BLEY: And you don't -- you wanted 15 the example to be more general?
16 MR. MORTON: Well, the point exists --
17 MEMBER BLEY: Or at least to narrow its --
18 MR. MORTON: Not introducing guidance 19 within an example when it's not within the descriptive 20 material of Appendix D.
21 MEMBER BLEY: Okay.
22 MR. MORTON: Okay.
23 MR. MCKENNA: And also, when I was 24 answering your question, I was answering general 25 50.59. Wendell asked it if we're digital, so just to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
33 1 clear that. Although, this briefing is more digital.
2 MEMBER REMPE: And so is --
3 MR. MCKENNA: Yeah.
4 MEMBER REMPE: There is guidance somewhere 5 also for just generally what can and can't be included 6 or they have to interface with headquarters in your 7 branch here?
8 MR. MCKENNA: Well, the guidance for 50.59 9 is what I briefed first. So we have the rule 50.59.
10 There's NEI 96-07, Rev. 1, which is the industry 11 guidance which we endorsed through the Reg. Guide 12 1.187. So that's all the guidance. And then if 13 there's any specific questions on a 50.59 issue, then 14 they would come to headquarters to ask these questions 15 --
16 MEMBER REMPE: Yeah, I did look through 17 the --
18 MR. MCKENNA: Yeah.
19 MEMBER REMPE: -- 96-07 and I didn't see 20 anything that said you can't go off the deep end and 21 start making up new design basis events for evaluation 22 when you're trying to substitute or look at an 23 accident of a different type than previously evaluated 24 in the FSAR. That seems like it's very wide open.
25 MR. BEAULIEU: Yeah, the -- this is Dave NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
34 1 Beaulieu. In fact, we're issuing a -- it's about to 2 hit the street, Revision 1 to NEI -- or Reg. Guide 3 1.187 that clarifies that exact question. What is an 4 accident of a different type? And --
5 MEMBER REMPE: It limits it?
6 MR. BEAULIEU: Yes. And --
7 MEMBER REMPE: Okay.
8 MR. BEAULIEU: -- it really has to do with 9 it's actually not an easy question to answer, but it's 10 a question -- it's a -- the commission intended if it 11 was an accident, if it was -- the plant was being 12 designed today it would be an accident in the FSAR.
13 It's of a similar frequency and significance of those 14 in the FSAR, but they're -- but none of the methods of 15 that are evaluated the current methods are applied to 16 this new scenario, but it -- so we attempt to -- we 17 clarify that in Revision 1.
18 MEMBER REMPE: Okay. I'll look through 19 that again.
20 MR. BEAULIEU: That's a good question.
21 That's a -- the guidance isn't really particularly 22 clear regarding that.
23 MEMBER RICCARDELLA: Could -- just for my 24 clarification, just for my clarification, the six or 25 eight things that we're talking about here are general NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
35 1 50.59 stuff, not specifically related to this 2 MR. MCKENNA: That is correct. We have 3 not gotten to any digital briefing yet.
4 MEMBER RICCARDELLA: Right.
5 MR. MCKENNA: This is all general.
6 CHAIRMAN BROWN: Just for calibration, 7 when we get to Appendix D, Section 4.3 deals with one 8 through eight, 4.3.1, the way they structured it so 9 that they map up to Criterion 1 --
10 MEMBER RICCARDELLA: Yeah, I understand 11 that.
12 CHAIRMAN BROWN: -- 2, 3, 4, 5, 6, 7, 8.
13 That's the only point I was trying to make for 14 calibration part purposes. That made it easy, 15 somewhat easy.
16 MR. MCKENNA: So we're beginning to 17 approach the digital guidance. We're not there yet 18 though. So now I'm going to talk fairly quickly on 19 NEI 96-07 which, again, was the industry guidance that 20 we endorsed through the Reg. Guide. Okay. So NEI 21 96-07, Revision 1 provides the guidance for 22 implementing the 10 C.F.R. 50.59 rule and that is as 23 of 1999.
24 MEMBER BLEY: Can I ask a question? The 25 basic NEI 96-07, Rev. 1 is the guidance. The NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
36 1 Appendices, including Appendix D, we're talking about 2 just kind of refines how you meet what's in the main 3 document. Is that -- is that the correct 4 interpretation?
5 MR. MCKENNA: That's correct. So the 6 Appendices that are in 96-07, only supplement the 7 96-07 guidance.
8 MEMBER BLEY: Okay.
9 MR. MCKENNA: And the only -- I'll stop 10 there.
11 MR. WATERS: Yeah. Maybe I'll would 12 provide just a little bit of broader context. 50.59 13 applies to all modifications to plan, of course.
14 96-07 based document applies to all modifications, 15 including digital. The challenge that was faced is 16 when you put a digital mod in you may interconnect 17 systems, you may have software cross redundant 18 systems. You raise challenging regulatory and 19 technical questions. And the whole purpose of 20 Appendix D was to clarify those issues within a base 21 framework within the NEI 96-07. Is the challenge of 22 the digital technology when we're, you know, 23 integrating systems together, for example, we're 24 interconnecting them.
25 MR. MCKENNA: Okay. So NEI 96-07, Rev. 1 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
37 1 is set up, again, going through the applicability of 2 the modification, the screening, the evaluation 3 process, gives some guidance for applying 50.59 to 4 compensatory actions to address non-conforming or 5 degraded conditions. And it talks about how to retain 6 10 C.F.R. 50.59 evaluations in accordance with the 7 rule.
8 So now we're going to start going into the 9 digital part. So digital guidance was first addressed 10 through the industry in EPRI TR-102348 which was 11 issued in 1993 to establish guidelines for digital 12 upgrades. We endorsed that through a Generic Letter 13 95.02 and that was superseded when Rev. 1 was issued 14 to the EPRI guidance. And that was issued as NEI 15 01-01. So NEI 01-01 is still in effect today. We 16 endorsed that through RIS 2002-22 which is still in 17 effect today and you'll recognize that number. That's 18 what we issued the Supplement 1 to, to clarify 19 guidance for digital modifications.
20 So NEI 01-01 is what I'll focus on in the 21 next couple slides. Again, it was issued in March 22 2002 to help nuclear pen operators implement licensee 23 digital upgrades in a consistent, comprehensive and 24 predictable manner. It has -- it's a fairly extensive 25 guidance and it gives guidance on the technical side NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
38 1 of the modifications and guidance for how to perform 2 a 50.59 evaluation for a digital modification. And it 3 also gives guidance for a licensee amendment request 4 if you're in that -- if the licensee is in that step 5 of doing the modification.
6 So throughout the years we found that 7 industry was inconsistently applying the guidance in 8 NEI 01-01 in digital upgrades. Mainly because of what 9 I've listed here. The lack of industry guidance for 10 a technical evaluation of common cause failures. We 11 had a couple issues that were noticed in the 12 inspection program which I've highlighted on the 13 slide. The first one was the implementation of a 14 digital controls, a LaSalle Rod Control Management 15 System modification which we published in an info 16 note, 2010-10 to describe that issue. And also in 17 Harris 2013 violation in Region 2 where they did a 18 modification to their solid state protection system 19 and replaced some circuit boards that had complex 20 programmable logic devices in those circuit boards.
21 And didn't go through the evaluation phase of 10 22 C.F.R. 50.59.
23 MEMBER BLEY: This is kind of an overview 24 for us, I suppose. Were there many more violations 25 and other issues? Because something drove us to come NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
39 1 out with this Appendix D.
2 MR. MCKENNA: Right. So no other issues 3 than what I'm highlighting here, but I think what 4 happened back when we started to uncover the stuff is 5 -- and why we had to go into this process is licensees 6 reevaluated say, how are we going to do digital 7 modifications using the 50.59 process since we are, 8 potentially, not doing it correctly. So it slowed 9 down digital modifications.
10 MEMBER RICCARDELLA: So when you say there 11 was a violation, did the violation mean that they did 12 something wrong in the modification or they just did 13 --
14 MR. MCKENNA: The violation --
15 MEMBER RICCARDELLA: -- without -- did it 16 without coming for a license amendment?
17 MR. MCKENNA: The violation in this case 18 was a violation of 10 C.F.R. 50.59. So they did not 19 file -- they did not do the rule the correctly and 20 that's what the violation was implementing TLC. In 21 that case they should have either completed the 22 evaluation or should have come into the NRC with a 23 license amendment. In this case they didn't.
24 CHAIRMAN BROWN: So their assessment 25 should have said hey, we've got something different, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
40 1 we got the software, TLCs, and therefore, we need to 2 do an evaluation. And that would have kicked them 3 into Item 6, I guess, for what you're talking about in 4 this PLC issue. And that they should have said uh-oh, 5 we failed. In other words, we don't meet that, 6 therefore, we, I think this is a negative reversing.
7 So they should have come in with an LAR? Is -- that 8 was the conclusion you reached and that's why it was 9 a violation?
10 MR. MCKENNA: So I'll hand it over to --
11 PARTICIPANT: Norbert.
12 MR. MCKENNA: Norbert.
13 MR. CARTE: I'm Norbert Carte. So two 14 different types of violations. The first violation 15 was what we would call a D1 violation under Clause D1, 16 which was inadequate documentation. So they did not 17 provide sufficient documentation for the conclusion 18 that they reached.
19 The other violation was C26 violation, so 20 they had not eliminated from consideration common 21 cause failure. So again, based on the information at 22 hand, they inappropriately reached that conclusion.
23 The C26 violation says you should have 24 come in for a license amendment. The D1 violation 25 says basically, you should have done your paperwork NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
41 1 better. And to resolve the D1 you -- they did 2 additional analysis and demonstrated that there wasn't 3 a problem.
4 MEMBER SKILLMAN: So if I could ask, in 5 these two cases were in either case the licensee 6 gaming the system? In other words, were these errors 7 of commission or omission?
8 MR. CARTE: I think there was a confusion, 9 confusing technical and --
10 MEMBER SKILLMAN: Administrative?
11 MR. CARTE: -- licensing guidance. The 12 technical, what is a digital -- when do you need to 13 consider CCF and what is licensing guidance? I don't 14 think it was an intent to game the system.
15 MR. MORTON: Right. I believe in -- and 16 correct me if I'm wrong staff, but I believe in the 17 Harris example they actually screened that mod out.
18 They didn't actually get to the evaluation, correct?
19 MR. CARTE: Correct.
20 MR. MCKENNA: That's correct.
21 MR. MORTON: So there was -- I believe in 22 that case there was a failure to recognize how 23 software can effect this particular mod in the 50.59 24 criteria in itself, necessitating, maybe, some better 25 clarifying guidance because digital is complicated NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
42 1 when it comes to 50.59. The staff recognize it, 2 industry does as well, which proceeded to what we're 3 talking about today.
4 MEMBER SKILLMAN: Okay. Thank you.
5 MEMBER RICCARDELLA: So the resolution was 6 that then ultimately they did come in for license 7 amendment? The resolution of these?
8 MR. CARTE: So there were two different 9 resolutions. On the first one, the basic problem was 10 they installed a rod control system that had the 11 potential to withdraw multiple rods when the originals 12 did not. And that could have occurred due to a number 13 of errors, so -- but they never did an analysis to say 14 that the plant would be okay if that happened. So 15 they then subsequently did the analysis and said they 16 were bounded. And therefore, they would pass the 17 criteria. So that was a failure to do adequate 18 analysis and documentation. They believe that they 19 didn't need to do that analysis because the failure 20 was not incredible.
21 The second one was a little bit more 22 complicated. Multiple plans had installed these 23 cards. What we decided to do was issue an EGM, 24 Enforcement Guidance Memorandum, to withhold issuing 25 a bunch of violations until we resolved the issue. We NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
43 1 requested that the PWR Owners Group submit a topical 2 report and we wanted to address this generically. And 3 based on that topical report, we concluded that they 4 had done sufficient analysis of the testing to 5 demonstrate that common cause failure was not a 6 consideration under 50.59. And then the remaining 7 installations and modifications never came in for a 8 license amendment request.
9 MEMBER RICCARDELLA: Okay. So in both 10 cases, had they done the proper 50.59 evaluation it 11 would have come out that they didn't need a license 12 amendment, is that right?
13 MR. CARTE: Yes.
14 MEMBER RICCARDELLA: Thank you.
15 MR. MCKENNA: So again, I think why 16 Norbert hesitated on that is --
17 MEMBER RICCARDELLA: It's complicated.
18 MR. MCKENNA: -- they actually have to do 19 the evaluation before we can, you know, we can't 20 pre-judge the decision.
21 MEMBER BLEY: I'm jumping the gun a little 22 bit again. You're going to get to Appendix E in just 23 a minute. What is the logic and the need for 24 retaining the ability for a licensee to use NEI 01-01 25 given you have new guidance?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
44 1 MR. MORTON: Okay. To get to your point.
2 Earlier you had some questions, so we'll have to --
3 we're going to have to get into a little bit of a 4 history of what we're talking about. And I would 5 request that look at that last bullet on this slide 6 which is the staff issued a letter publicly into NEI 7 stating a number of different concerns we had with the 8 guidance because NEI 01-01 at this point, in 2013, was 9 fairly old at this point. There were a number of 10 technical concerns, but not with the -- with NEI 01-01 11 but not with the licensing piece. So if you -- right, 12 so if you have not read NEI 01-01, it has 50.59 13 licensing guidance and technical guidance interspersed 14 throughout the document. They are not separate 15 entities. It's all within one package.
16 The staff had a lot of concerns with a 17 01-01, the technical piece but not necessarily the 18 licensing piece which the 50.59 guidance. 01-01 had 19 screening guidance and has evaluation guidance. It 20 even has some guidance on addressing software common 21 cost failure as well.
22 Part of that concern is that, you know, if 23 licensees are having trouble. In the case of the SSPS 24 card issue not recognizing that maybe this should 25 screen in, either the base part of the process should NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
45 1 it screen in or not? If you're not seeing that that's 2 an issue with software then we've got larger issues 3 going on that maybe across the board. And that really 4 resulted in the letter signing off from staff saying 5 hey, there's some things we, they need to take a look 6 at.
7 So it's part of that and NEI can to speak 8 to that more when they actually speak. But as part of 9 the overall game plan they decided to come in with 10 sort of a two phased approach. One was splitting off 11 NEI 01-01 into two different documents. One was the 12 licensing document which is Appendix D. A second was 13 a technical document to replace the technical content 14 of 01-01. Together those documents would be 15 sufficient to supersede NEI 01-01 so that licensees 16 could put NEI 01-01 to bed and use these two adjoining 17 documents. That was the game plan presented to staff 18 back in 2016, I believe October.
19 So when we're reviewing this, we're 20 reviewing from the standpoint of hey, you've got a 21 technical document that's, it's backing up the 22 licensing consideration they're doing with Appendix D.
23 The technical document hasn't quite -- isn't quite 24 fully baked yet in terms of being provided to staff 25 for a full review endorsement as well. We don't NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
46 1 really have that in our hands --
2 MEMBER BLEY: You have had not had --
3 okay.
4 MR. MORTON: We've been given draft of a 5 document. NEI can speak to that more, but there were 6 concerns with that particular document and at this 7 point that review's been delayed in terms of that 8 document --
9 MEMBER BLEY: What kind of document would 10 this be? Would it be like the topically or something?
11 MR. MORTON: It could be a topic. It was 12 not submitted to us as a topical.
13 MEMBER BLEY: Okay. Hasn't been submitted 14 to you yet --
15 MR. MORTON: No, not --
16 MEMBER BLEY: -- in any form? Okay.
17 MR. MORTON: Just here, take a look at 18 this, see what you can review, see what you think 19 about it, to that part. So that --
20 MEMBER BLEY: So using Appendix D before 21 that technical report's in place sounds problematic.
22 I don't know if it is.
23 MR. MORTON: Ergo some of the 24 clarification exceptions we ended up taking because 25 without the technical document to sort of enforce some NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
47 1 of the things that's in Appendix D we're reviewing it 2 sort of in a vacuum.
3 So on the one hand that results in some of 4 the clarifications and exceptions you'll see. But on 5 the other hand, one of the particular one that you 6 cited was that we could not -- we couldn't find 7 Appendix D adequate to supersede NEI 01-01 by itself 8 without a technical adjoining document to go along 9 with it. That's why we'll still have 01-01 and the 10 RIS Supplement 1 on the streets with Appendix D 11 because we don't have a technical basis to administer, 12 to allow us to supersede that at this time.
13 And it's one of the reasons why the risk 14 supplement came into being is because we wanted to get 15 short term guidance for low safety significance 16 systems to get those mods modernized to plants. In 17 the meantime, we're still looking for the technical 18 base document that can -- when they help supersede 19 some of the older guidance.
20 MEMBER BLEY: Okay.
21 MR. MORTON: But until then, the game plan 22 is what you're seeing inside the Reg. Guide right now.
23 MEMBER BLEY: A little related issue for 24 me is given this separation into two documents, I'm 25 assuming there's also -- I haven't sat down and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
48 1 compared them. The hope is that the two new documents 2 will clarify some of the things that weren't clear to 3 licensees in the 01-01?
4 MR. MORTON: Yes.
5 MEMBER BLEY: Okay. But we aren't there 6 yet, so we're just kind of getting advance look. You 7 -- now you can use Appendix D, but you have to use the 8 guidance and the RIS as well.
9 MR. WATERS: I -- this is Mike. I think 10 we're part way there. I think the other part is we 11 issue the Supplement, which is two files in one, which 12 does provide the licensing interpretation with some 13 general guidance of how do you determine common cause 14 failure, sufficiently low. I mean, the think the 15 example Wendell's -- to me the primary thing was to 16 answer the criteria in 50.59 is you make a 17 determination, for example, a common cause failure, 18 sufficiently low, that makes sense.
19 You can define what that means from the 20 50.59 criteria standpoint. But then from a technical 21 standpoint, okay, what does that really mean to say 22 CCF is sufficiently low? What message do you look at?
23 What features you consider? How'd he get that 24 technical conclusion? And that's the struggle we had.
25 NEI 01-01 had a lot of technical guidance.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
49 1 Appendix D does not have a technical 2 guidance and that's how we got to this point in day 3 where he issued a RIS Supplement to provide the 4 licensee interpretation with some technical factors to 5 look at for the quality of assessment. And we're 6 trying to incorporate this thought processes into the 7 Appendix D endorsement review.
8 MR. MILLER: So if I could. This is Chris 9 Miller. The question was asked, so what drove us 10 there? And what exactly drove us there is because we 11 had, you know, licensees were saying well, wait a 12 minute, I'm not sure I'm a, you know, want to issue 13 any digital -- do any mods on the 50.59 because I'm 14 going to get challenged on it by NRC.
15 And so we had to bridge that gap. The 16 RIS, we think, bridged the gap and the guidance. And 17 so now you'll hear in this presentation later that 18 there's a number of mods being planned or in progress 19 that are using the guidance and the RIS. And so in 20 this -- so this effort is just to move on from that 21 guidance and the RIS and provide additional 22 clarification. But we do think that big picture wise 23 we solved a lot of the problems that we had with not 24 being able to issue digital mods under 50.59.
25 MEMBER BLEY: Two related things. Are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
50 1 licensees, any licensees, still making submissions 2 using 01-01? Do you have any of those coming in?
3 MR. CARTE: Norbert Carte. And so the 4 license gives them, under 50.59, the discretion to 5 make decision whether a change requires a license 6 amendment. So when they use 01-01 they -- you use it 7 essentially to pass 50.59 and not a license amendment.
8 You don't really reference 01-01 as part of a license 9 amendment.
10 MEMBER BLEY: Yeah, I get that.
11 MR. MORTON: I think, along with what 12 Norbert was saying, if the licensees are using RIS 13 2002-22, Supplement 1, which is a quality assessment 14 guidance, it's based upon the technical content of NEI 15 01-01. So if they're using the supplement they are, 16 in effect, still using 01-01. If that's -- I think 17 that's what you're asking.
18 MEMBER BLEY: Sort of, yeah. The related 19 question, and you may have sort of answered it 20 already, at some point -- well, at some point in the 21 future, maybe we'll only have one of these two things 22 available then. In the interim, when we have both 23 possibilities, you need -- and I assume the first 24 batch that you're going to be looking at are using 25 Appendix D and the RIS Supplement. At some point if NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
51 1 it stays this way, do you need some kind of extra 2 guidance for the staff on how to look at one that 3 comes in or gets inspected where they've used 01-01?
4 Are we losing something? From the inspector's point 5 of view, is it going to be obvious how to deal with 6 these, using the two different sets of guidance?
7 MR. MCKENNA: So I -- from our feedback, 8 to date, it looks like the licensees have shifted over 9 to use the RIS 2002-22, Supplement 1, and are not 10 using NEI 01-01 guides because --
11 MEMBER BLEY: That's kind of what I 12 expected, but then --
13 MR. MCKENNA: -- Supplement 1 --
14 MEMBER BLEY: -- two years from now if 15 we're still in the same boat, if somebody comes in the 16 other way do you need anything for the inspectors?
17 MR. MCKENNA: Well, there wouldn't be any 18 -- so the reason why we issued the RIS Supplement 19 again is the NEI 01-01 introduces the term qualitative 20 assessment.
21 MEMBER BLEY: Yeah.
22 MR. MCKENNA: Didnt really tell you how 23 to do -- places you've got to do it. The RIS --
24 MEMBER BLEY: Does.
25 MR. MCKENNA: -- gives you -- tells you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
52 1 how to do it.
2 MEMBER BLEY: Go ahead. Im sorry. I've 3 beaten this enough.
4 MR. MILLER: I'm going to beat it just a 5 little bit more just to say that you saw in Phil's 6 presentation earlier that we have provided training to 7 a couple of the regions. We're going to do some more 8 this spring and to the other two regions. And we've 9 also determined that when some of these mods are being 10 inspected under a 50.59 inspection, we in the program 11 office would write support to this initial ones that 12 are going out there. So that's going to help 13 calibrate us in did we get the guidance right or is 14 there something else that we need to change? So we 15 hope those will bear fruit in that area.
16 MR. MCKENNA: So I shifted to the slide as 17 we were talking because we were talking to NEI 01-01 18 and I think we hit a lot of the highlights of this 19 slide. I'll just leave it up there without going over 20 each individual bullet in case the members have any 21 questions. But these were all the concerns with NEI 22 01-01 and why we decided to issue the RIS and why 23 industry decided to write Appendix D.
24 Again, if you go down to like the seventh 25 bullet right there, it talks about NEI 01-01 contains NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
53 1 the two type of guidance embedded in the procedure for 2 the technical side and how to do a 50.59.
3 So the big point on what makes digital 4 modifications different from other modifications is 5 the common cause failure issue. You can have combined 6 function, shared communication, shared resources, 7 software or -- in the safety model of a nuclear power 8 plant when we had -- while we have hardware out there, 9 it's -- the likelihood of common cause failure is said 10 to be low because we maintain high standards. There's 11 physical separation between equipment. And the 12 degradation methods for a piece of hardware are slow 13 to develop and we test for it. We do surveillances.
14 In this case you can have a -- for 15 software you can introduce that error into the 16 software, have the software in both trains of the 17 safety system and, you know, not know about it until 18 it rears its head and it causes an issue. So --
19 CHAIRMAN BROWN: So the definition of a 20 common cause failure, based on what you just said, 21 would be you introduce an error into the software? In 22 other words, a design error. You don't consider 23 something -- I'm still trying to figure out what you 24 all -- how you all define a software common cause 25 failure.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
54 1 MR. MCKENNA: Well that's --
2 CHAIRMAN BROWN: Software's written, it's 3 programed. It's stored in memory. It doesn't just 4 disappear and the only -- in my -- this is my 5 particular -- this is -- now I'm not speaking for the 6 committee. This is just my own troglodyte, 7 Neanderthal thought process. Those, that stuff's 8 built in in the design. That's the way you program 9 it. And if you have a common cause failure it means 10 you have the same software process in each train.
11 And, therefore, it responds incorrectly in each train.
12 So that's a design error. It's not necessarily --
13 MEMBER BLEY: Or it could -- yeah, it is 14 a design error, but it could be that the plant has 15 brought itself into conditions that you hadn't quite 16 --
17 CHAIRMAN BROWN: That's a design error.
18 MEMBER BLEY: -- analyzed.
19 CHAIRMAN BROWN: You don't consider the --
20 MEMBER BLEY: And how it takes out more 21 than you expected.
22 CHAIRMAN BROWN: Well, if you don't 23 consider the plant conditions with the software.
24 MEMBER BLEY: Software doesn't wear out.
25 It doesn't have the characteristics of mechanical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
55 1 failure so it's got to be --
2 CHAIRMAN BROWN: Exactly. Well --
3 MEMBER BLEY: Because it's executed in a 4 different way then you expected.
5 CHAIRMAN BROWN: You can lose bytes and 6 bits. You can lose memory.
7 MEMBER BLEY: That's true.
8 CHAIRMAN BROWN: Okay. At cosmic rays 9 theory, right, you can -- well, actually, we know that 10 happens.
11 MEMBER BLEY: It can be, yeah.
12 CHAIRMAN BROWN: But it's not like, you 13 know, like, ten to the mat is 40 seconds or something 14 like that. If you're inside of a ship particularly 15 it's a little -- my point being is --
16 MEMBER BLEY: That's not quite applicable 17 here, Charlie.
18 CHAIRMAN BROWN: -- we're really talking 19 about you can have the same type of failure, common 20 cause failure, relative to analog designs. If you 21 design the analog system to not consider the 22 circumstances under which it's got to operate 23 correctly, then you have a problem. And that causes 24 the common cause from a train to train situation 25 there.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
56 1 So, I mean, the way I view you have 2 resolved this in the past is -- and others, at least 3 on the stuff we've looked at over the last few years, 4 is with the diversity issue, defense-in-depth 5 diversity issue. In other words, you have something 6 outside or off that has a different format then the 7 main pathways to deal with the common cause failure 8 issue.
9 Now you can debate whether you need to do 10 that or not. You can assess it whether you need to do 11 that or not and that, to me, is part of the assessment 12 and the evaluation you go through because the last 13 thing you -- in my own personal opinion, if you've got 14 -- like some examples you all gave, somebody replaces 15 a two train or two controller, redundant controller 16 type system with two PLCs. They have the same 17 software, they have the same inputs and outputs and 18 off they go. Well, they could both fail, they could 19 both do this.
20 But you don't necessarily want to have 21 them built and designed with different components 22 because that becomes an economic nightmare that won't 23 help. So there's a trade-off. You have to have some 24 assessment or evaluation that gets -- has a low 25 likelihood. Here's where we get into the stork dance NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
57 1 of likelihood. So I just -- this is a very difficult 2 process to go through.
3 MR. WATERS: I mean, from a highest level 4 we had these questions when we went down this path of 5 when we had Appendix D and then what we did through 6 RIS. And I think what you said about having diversity 7 and having that low assurance for protection systems, 8 absolutely correct. And I think we're maintaining 9 that philosophy and this will be a subject of -- BT 10 719 would be for later for, like, just amendment. For 11 the RIS Supplement we were focused on this over 12 auxiliary support systems. The chiller was a common 13 example. Whether you have, you know, just a 14 possibility of a multiple -- two chillers failing at 15 the same time from software, can you determine if CCF 16 is sufficiently low and address the client from that 17 standpoint. So this is really targeted at those type 18 of over stipulated systems and off systems outside the 19 core protection systems.
20 MEMBER RICCARDELLA: For non-electrical 21 engineer, non-I&C person like me, what's the essence 22 of the difference in common cause failure 23 considerations between an analog controls and digital 24 controls?
25 MR. MORTON: So one of the things that, I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
58 1 think, Appendix D does a good job of now is it 2 actually tells you --
3 MEMBER RICCARDELLA: Appendix D?
4 MR. MORTON: Appendix D.
5 MEMBER BLEY: The new appendix.
6 MR. MORTON: NEI 01-01 as well, that's the 7 previous document, is it why is digital different than 8 all these other disciplines? What makes it hard? So 9 the introduction of software. And as Charlie was 10 talking about, it's very difficult to demonstrate that 11 a complex software package is free of design errors or 12 bugs because we distribute it. And if it's 13 distributed within different redundant, different 14 redundant channels.
15 MEMBER RICCARDELLA: I'm familiar enough 16 with software to understand that.
17 MR. MORTON: That's just one aspect of it.
18 Another ability of digital versus the analog was 19 combining of different design functions on the same 20 process or chip. So as you had previous segregated 21 systems, even within the protection system or 22 non-protected systems like, say, safety chillers, for 23 example, in order to consolidate parts, reduce your 24 costs. You can consolidate functions, different 25 design functions, onto the same processor or series of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
59 1 processors. It gives you redundancy and reliability 2 and reduces costs.
3 MEMBER RICCARDELLA: Yeah.
4 MR. MORTON: But it also is doing 5 something that 50.59 will have a hard time with.
6 MEMBER RICCARDELLA: We're using the term 7 software but it's really firmware, right?
8 MR. MORTON: For the -- for the purposes 9 of this discussion we're just trying to keep it 10 succinct in terms of just software considerations.
11 But other half is you're introducing shared resources 12 in a way that you weren't previously doing which is 13 introduction to digital networking, introduction 14 interconnections between different systems that were 15 not previously interconnected.
16 These are all things that are common 17 within a distributed control system for any process 18 application within the nuclear context within 50.59 19 context. It makes it difficult to demonstrate that a 20 modern distributed control system does not create an 21 accident with a different effect or malfunction with 22 a different result. And it makes it difficult to say 23 that well, if I'm screening in that it's not adverse.
24 So there's a lot of consideration for 25 digital to make it more challenging than other NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
60 1 disciplines when it comes to the basic 50.59 process 2 as described in NEI 96-07 which was ultimately why 3 Appendix D was created to try and address why is 4 digital challenging and how to refine the guidance for 5 the base 50.59 guidance specifically for digital 6 considerations.
7 MEMBER RICCARDELLA: So I can I --
8 CHAIRMAN BROWN: Can you calibrate that 9 just a little bit? Take the -- there's two chillers 10 in the main -- for the main control room, I guess, the 11 examples that are used in Appendix D. Today they're 12 separate. You've got a box over here, a box over 13 here, they have their own controls. They operate 14 independently, totally. One of the thought processes 15 you can combine the controls for those two chillers, 16 the functional and mechanical parts, into one 17 processor control them both.
18 Now I've got shared resources, if you want 19 to call it that. I've combined everything. So now a 20 single failure can take out both of them and you may 21 or may not realize it at the time, you know, it can be 22 not obvious. And now you're in trouble.
23 MEMBER RICCARDELLA: So that would fail 24 the 50.59 --
25 CHAIRMAN BROWN: That would --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
61 1 MEMBER RICCARDELLA: -- test. And 2 therefore, would have to -- that change would have to 3 come in for a license amendment?
4 CHAIRMAN BROWN: Well, I --
5 PARTICIPANT: In theory.
6 CHAIRMAN BROWN: In theory. If that's a 7 safety related -- I think, is that a safety related 8 system with chillers in the main control room? I'm 9 trying to remember it right now. So that's a safety 10 related system because you got to keep relay rooms 11 cool and you've got to keep I&C stuff cool. I mean, 12 you -- that's in -- and people cool. So that's where 13 that one comes into play.
14 But you've got other non-safety related 15 systems where they have two sets of stuff. Is it okay 16 to combine those now because they're non-safety 17 related and if you -- now you look at other -- you can 18 look at other, what is it, qualitative assessments of 19 whether the -- what's the impact going to be? So 20 that's where this stork dance gets pretty convoluted.
21 You wanted to say something, Dennis?
22 MEMBER BLEY: For Pete, when you have the 23 two separate ones, the analog or whatever we're 24 talking about, the kind of things that introduce 25 common cause failures are common maintenance. The NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
62 1 same guy does maintenance on both of them screws it up 2 the same way. A lot of examples of that sort of 3 things, or wear out problems where they -- both age at 4 about the same rate.
5 CHAIRMAN BROWN: Or a particular component 6 may be a less reliable component and it's in both 7 things --
8 MEMBER BLEY: It's in both --
9 CHAIRMAN BROWN: -- and they both failed, 10 theoretically, at the same time. You've got to --
11 MEMBER BLEY: Within a narrow enough 12 window to take them both out.
13 CHAIRMAN BROWN: Narrow enough, exactly.
14 So you were -- somebody was going to say something 15 when we were --
16 MR. MCKENNA: I was just going to mention 17 we'll get on in -- later in the brief. But the 18 qualitative assessment is assessing whether -- what 19 the likelihood of failure of that component is. And 20 if it's a low likelihood of failure then you go ahead 21 and pass the evaluation course under the 50.59 without 22 a license.
23 MR. BEAULIEU: Can I shed some light on 24 this? Because what -- you guys ask great questions.
25 It boils down, what really makes digital different?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
63 1 And primarily, it's not the wires connecting the 2 systems because you could buy insulation devices.
3 It's software. So why is software different? Okay.
4 It's different because it's -- writing software is 5 very prone to human error. Like, you type in a Google 6 search. If you don't type it in exactly, it -- you're 7 not going to get the result. It's very prone to human 8 error. Okay. So you know the errors exist. And --
9 MEMBER BLEY: Sure.
10 MR. BEAULIEU: -- you can't test them. It 11 turns out you can't comprehend -- for, unless it's the 12 simplest of system, it turns out there are so many 13 different combinations, it would take years or 14 computer, you can't test every combination to identify 15 all the errors. So now you're end up -- you -- so now 16 you have, now you know you have errors. You don't 17 know exactly where they are.
18 And now you have to answer a 50.59 19 question. Say, is the likelihood of failure of that 20 component any worse than the analog system? And just 21 like well, how do you know? You don't know where it 22 -- really where the errors are or where they're going 23 to appear. You just know that if both redundant 24 trains receive a common, a similar input, they're 25 going to fail at exactly the same way, at exactly the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
64 1 same time because they don't wear out. They just --
2 they're going to fail to -- so that's what makes 3 digital different.
4 You have -- so it's not that software is 5 unreliable. It's the fact that you just can't write 6 that in your 50.59. Say, oh, software's pretty 7 reliable. You need to have a basis. You need to 8 describe why this software is not going to fail any 9 more likely then -- and that's where this qualitative 10 assessment comes in. There was no standard way of 11 documenting the logic of how you determine how the 12 likelihood is and to be able to answer the 50.59 13 questions.
14 MEMBER RICCARDELLA: But isn't the bigger 15 concern how you do the design, not the 50.59, but all 16 these things we're talking about really effect how you 17 design the digital equipment? And I'm assuming is 18 IEEE standards and other things that give guidance on 19 that. I mean, to me that's a much bigger concern then 20 whether you come in for a license amendment request or 21 not.
22 MR. WATERS: I think that's correct and 23 when Wendell gets his presentation, that was a big 24 focus of what are the areas you look for quality of 25 assessment. Design features is the most important NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
65 1 one, I would say, for us. What -- from really general 2 terms, right. What are key design features to take 3 credit for and consider for quality of assessment as 4 well as the quality processes which may be, you know, 5 rely on some filing codes and standards.
6 So you're absolutely right. That was part 7 of the conundrum here we had is it's easy to say it's 8 common cause failure, sufficiently low. But what does 9 that mean from a design and ethological approach and 10 make that determination? That's how we got here 11 today.
12 MR. MORTON: And we'll -- and we'll get to 13 that slide when we talk about the RIS Supplemental 14 where we actually speak specifically about the 15 engineering evaluation portion of it as it relates to 16 supporting the overall 50.59 evaluation.
17 MEMBER RICCARDELLA: Thank you.
18 MR. MCKENNA: So I moved ahead to the next 19 slide which just introduces the RIS 2002-22, 20 Supplement 1. We'll discuss it more towards the end 21 of the presentation, but needed to introduce it here.
22 So the Supplement 1 gives the clarifying 23 guidance for preparing and documenting a qualitative 24 assessment which is mentioned in NEI 01-01. It does 25 mention it right in the RIS supplement that it is not NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
66 1 for replacement of the reactor protection system, not 2 for replacement of the engineered safety features 3 actuation system and not for modifying the internal 4 logic of RPS or FSAR.
5 MEMBER BLEY: And that means if you're 6 going to do one of those three you have to come to the 7 staff?
8 MR. MCKENNA: It means that you can't use 9 a qualitative assessment to -- yes.
10 MEMBER BLEY: All right. Okay.
11 MR. MCKENNA: Yeah.
12 CHAIRMAN BROWN: Well --
13 MEMBER RICCARDELLA: It doesn't mean you 14 can't do digital?
15 MR. MCKENNA: Well, again, any licensee 16 can start the 50.59 process for any modification they 17 want to do. In the end, the end result would be if 18 you're going to modify your reactor protection system, 19 you're coming in for a license for that. Not to 20 prejudge it, but --
21 MR. MILLER: Or come up with some other 22 method -- this is Chris Miller. Or come up with some 23 other methodology that we would find acceptable.
24 MR. MCKENNA: Correct.
25 CHAIRMAN BROWN: The RIS didn't say it's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
67 1 not for replacement. That, those are kind of 2 interesting words. I'm just now reading it. It says 3 the RIS supplement is not directed toward reactor 4 protection systems and safety, safeguard systems or --
5 I've got to use the exact words.
6 MR. MCKENNA: I thought it used the exact 7 words, but you may be true, sir.
8 CHAIRMAN BROWN: I'm just pawing through 9 it. I just have this uncanny interest in --
10 MEMBER SKILLMAN: It's on Page 2 of 5 in 11 the second paragraph. I would like to ask a question 12 while you're rummaging through your paper. I thought 13 that this RIS Supplement is the -- is so well written 14 that it should be applicable beyond digital. And so 15 I'm wondering why it seems to be limited.
16 CHAIRMAN BROWN: What do you mean by 17 beyond digital?
18 MEMBER SKILLMAN: This RIS Supplement is 19 intended to address digital modifications, but may be 20 used for modifications for non-safety related systems 21 at the discretion of the licensee. It seems to me 22 that the tone and the thoroughness of this could be 23 used anywhere in the plant. So it just seems to me 24 that it's more, it could be more broadly applicable 25 then the way the RIS describes itself.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
68 1 MR. WATERS: It -- it had to do, there's 2 a history behind here. I mean, I don't think staff 3 would disagree. When we developed the RIS Supplement, 4 we like to get State Colder feedback and there was, 5 frankly, a concern that maybe there's -- some of the 6 guidance would not be appropriate for even lower tier, 7 non-safety systems where everything's well. So, you 8 know, or -- and we believe it was not an appropriate 9 for a reactor protection system during the safety 10 check system. But we did not want to be so absolute 11 on either side of the spectrum to prohibit it or allow 12 it.
13 So those are words that we chose and you 14 can use a RIS for a non-safety digital modification.
15 I think the concern expressed by State Colder is when 16 you go through it, you have to do everything in there 17 to address it for those type of systems. And this is 18 the language we fell upon to explain both the 19 non-safety and the other end of the spectrum for 20 reactor protective systems. Did I say it correctly, 21 Wendell?
22 MR. MORTON: That's correct.
23 CHAIRMAN BROWN: I will probably comment 24 later. One of the examples in Appendix D, since we're 25 talking about the RIS right now, and in the RIS -- I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
69 1 mean, excuse me, in Appendix D, there's a statement up 2 in the beginning of Appendix D that's more, far more 3 explicit, that says the qualitative assessments are 4 not to be used for reactor protection and safeguard 5 systems. But yet, the last example is a replacement 6 of a reactor protection system where they caught that 7 qualitative assessments reaching a not -- what's the 8 term for not adverse, it's the other words about it's 9 not --
10 PARTICIPANT: Sufficiently low.
11 CHAIRMAN BROWN: Not sufficiently low.
12 Thank you. And therefore, it defaults to the LAR 13 routine, I guess. So I was interested -- you don't 14 have to address it now. It just seemed to be that 15 NEI's own example after they said it's not to be used, 16 all of a sudden one of the examples uses it.
17 MR. MCKENNA: Yeah, I --
18 CHAIRMAN BROWN: So.
19 MR. MCKENNA: I think what they're trying 20 to demonstrate there is, again, you can use 50.59 for 21 any modification you're going to make in the plant and 22 the end result could be that you have to submit a 23 license amendment.
24 CHAIRMAN BROWN: Yeah. Well, it did come 25 to a conclusion, but it was based on a qualitative NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
70 1 assessment, not an -- it was part of the evaluation 2 but yet they used the words qualitative assessment as 3 arriving at the not -- what was the word again?
4 Sufficiently low. Thank you.
5 MEMBER BLEY: Even though at the top they 6 said you can't use it for the --
7 CHAIRMAN BROWN: Exactly. But way up in 8 the beginning of the Appendix. It's a little bit of 9 an inconsistency, so.
10 MR. RAHN: I could -- this is David Rahn, 11 out of the Office of NRR. I could shed a little bit 12 of light on this is that the original thought on the 13 RIS was it's to clarify a previous RIS which was 14 describing what was covered in NEI 01-01. NEI 01-01 15 does not outright prohibit using the guidance in NEI 16 01-01 for safety related modifications like the RPS 17 and SFAS.
18 But what it does say is that if you 19 attempted to do so you would likely run across 20 problems in responding to some of the 50.59 evaluation 21 questions. And the likelihood is that you wouldn't 22 actually make it all the way through the system. And 23 that's why the RIS used the words it's not directed 24 towards, still this time, some modifications mainly 25 because the likelihood of actually going all the way NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
71 1 though without problems is not high likelihood.
2 So we use the terms -- it's not directed 3 towards those and one could really attempt to use the 4 RIS and NEI 01-01 guidance for doing this type of 5 modification. But -- so anyway, I just wanted to make 6 sure that's the nuance on the word --
7 CHAIRMAN BROWN: That is -- it's an 8 interesting nuance you have there.
9 MR. RAHN: Right.
10 CHAIRMAN BROWN: It's -- because of the 11 protection system -- and I'll just say it right now.
12 You can't -- obviously, this is not a rule. But you 13 just say it's not directed at or you can't use the 14 qualitative assessment to assess or evaluate reactor 15 protection systems. It just seems like you ought to 16 make a flat out statement that you need an LAR to 17 replace reactor protection system. You don't have to 18 say you must -- you're required, but you can say you 19 should submit an LAR for protection systems and 20 safeguard systems.
21 I can't even conceive of having a licensee 22 replace an entire protection and SFAS system, the 23 electronics part of it, under 50.59. I just -- but 24 yet, it's not prohibited, if I'm not mistaken. Am I 25 correct? Did I say that right?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
72 1 MEMBER BLEY: Kind of saying you're 2 wasting your time is the way I took that.
3 MR. WATERS: Again, nothing is prohibited 4 from using 50.59. The end conclusion is going to be 5 you're going to have to submit a license amendment.
6 CHAIRMAN BROWN: Well, but you had the 7 circumstance where people evaluated it that it didn't 8 need to be submitted, and therefore, nobody found out 9 for a while. I can't imagine. That's pretty big 10 extensive replacement, so hard to be -- it's 11 invisible.
12 MR. MORTON: If you're referring to some 13 of the examples we had earlier in the presentation, 14 that's why the clarifications for the screening and 15 evaluation are being made to help licensees not make 16 those particular mistakes when comes to --
17 CHAIRMAN BROWN: I understand that point.
18 MEMBER BLEY: You know, that -- just on 19 the surface looking at the Criterion 6, if you're 20 doing that big a replacement --
21 CHAIRMAN BROWN: It seems obvious.
22 MEMBER BLEY: -- saying you're not going 23 to introduce anything new just, you'd never be able to 24 show that on here.
25 MR. MCKENNA: So member Bley --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
73 1 MEMBER SKILLMAN: Could I ask this --
2 MR. MCKENNA: Oh, sorry.
3 MEMBER SKILLMAN: So the RIS allows the 4 qualitative assessment. What attention is given to 5 the credentials, but more importantly, the 6 qualification of the individuals, the author, that 7 qualitative assessment? Because to me, that's where 8 the, that's where the quality is provided --
9 MR. MCKENNA: So a --
10 MEMBER SKILLMAN: -- in the argument.
11 MR. MCKENNA: So licensee would have their 12 own qualification process for people who are allowed 13 to do modifications and write 50.59s. And they'd also 14 have peer reviewed.
15 MEMBER SKILLMAN: Is that inspected? Do 16 the NRC residents or special teams approach those --
17 MR. MCKENNA: So that would certainly be 18 part that you --
19 MEMBER SKILLMAN: -- qualifications?
20 MR. MCKENNA: -- sample in doing an 21 inspection, right. The qualification of the person 22 who perform the 50.59. Yes, sir.
23 MEMBER RICCARDELLA: Is the 50.59 subject 24 to 10 C.F.R. 50, Appendix B, quality assurance 25 requirements?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
74 1 MR. BEAULIEU: The mod itself is.
2 MEMBER RICCARDELLA: The mod is.
3 MR. BEAULIEU: The date of assessment is 4 part of the mod and so the --
5 MEMBER RICCARDELLA: Okay.
6 MR. MCKENNA: So when we do a modification 7 inspection in the field, it covers the design of the 8 modification and the 50.59 process. It covers both of 9 those areas.
10 MEMBER RICCARDELLA: Thank you, again.
11 MR. CARTE: This is Norbert Carte. One 12 dimension -- so you could change a non-safety system.
13 So Appendix B would not apply.
14 MR. WATERS: This is Mike. I just want to 15 help answer the question from a different angle. I --
16 our focus was not on the credentials or expertise of 17 the licensees. They have the expertise. I think the 18 issue was even if two competent individuals you reach 19 the same conclusion for digital modification. So we 20 focus more on what things you look at. What type of 21 documentation do you need to have a -- lead competent 22 people to the same conclusion whether it passes or not 23 the threshold. So that was really our focus, was how 24 do you get a documented result and address the right 25 things to come to those kind of conclusions?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
75 1 MEMBER SKILLMAN: Yeah, fair enough. I 2 wasn't trying to be pejorative. It just seems to me 3 that when one talks about a qualitative assessment you 4 see this group of people at a licensees' headquarters 5 or at the site. Sometimes you wonder, do those 6 individuals really know what they're doing?
7 MR. MORTON: So as part of the 8 construction creation of the RIS, Supplement 1, we 9 tried to take into account let's not make this too 10 complex. Let's keep it pretty straightforward. Look 11 at the issue that we've heard from our licensees and 12 from our inspectors and from general staff. On the 13 areas we've had concerns of 50.59 screening and 14 evaluations when it comes to digital. So the 15 documents informed from that standpoint.
16 Because we try to take into account that 17 perspective of folks who may not necessarily be a 18 specific expert in either the technical side or the 19 licensing side. But to be very clear and concise on 20 the expectations we have with this RIS, Supplement 1.
21 So we try to take that respect into account as well.
22 MEMBER SKILLMAN: I think one of the 23 challenges is if a licensee is going to change any 24 kind of control system that was supplied by the NSSS 25 vendor. the thinking, the NSSS vendor, the metal NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
76 1 framework and the designer's mind's eye is not easily 2 duplicated at the site or at the applicant's 3 headquarters.
4 And so very often their -- the intentions 5 are noble, but the product might not connect with what 6 had been, if you will, the original functional 7 performance requirements of the protection system that 8 the NSSS vendor was trying to implement. A lot of 9 that's very subtle and unless the individuals are 10 doing the qualitative assessment have some at least 11 sensitivity to that, an important feature or important 12 features might be lost.
13 MR. MORTON: So one of the things we 14 clarified. I want to come -- and Norbert referred to 15 this in terms of our documentation aspect. We really 16 tried to nail home within the RIS, Supplement 1 is to 17 -- you need to understand what you're putting in your 18 plant.
19 You need to understand the specific design 20 features of the widget you're trying to install and 21 the particular failure modes of this particular device 22 you're trying to install. And how you're compensating 23 for those particular failure hazards within this 24 system with some sort of compensatory measure within 25 the system, self-testing features or any sort of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
77 1 independence or redundancy of which you're crediting 2 against that particular device's failure.
3 So as part of the failure analysis to 4 support the engineering evaluation that you're using 5 to perform the qualitative assessment, we tried to 6 cover all those different bases on where you're 7 looking in the actual engineering aspect for potential 8 hazards when you're installing digital networks, 9 potentially creating new types of accidents, for 10 example, especially if you're combining multiple 11 different system functions on the same platform. We 12 point those hazards out in very specific bright 13 detail.
14 And then that's where -- with the 15 intention of documenting what those hazards are as 16 part of your 50.59 overall evaluation during the 17 qualitative assessment. So we try to take that into 18 account to make sure hey, we covered a technical 19 aspect of it, potential hazards you may be 20 introducing. Alert you to what those potential 21 hazards could be and what you should be supporting in 22 an engineer technical evaluation. And documenting 23 them so that we're insuring you're meeting the D1 24 requirements for 50.59 by documenting what you should 25 be.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
78 1 MEMBER SKILLMAN: Okay. Thank you.
2 CHAIRMAN BROWN: Before we go on, did you 3 have something else to say on this?
4 MR. MCKENNA: I was just going to 5 recommend that this would be a good spot for break.
6 CHAIRMAN BROWN: I'm way ahead of you.
7 I'm probably older than you, so -- we will take a 15 8 minute break, come back at 10:20, so we're recessed 9 right now.
10 (Whereupon, the above-entitled matter went 11 off the record at 10:01 a.m. and resumed at 10:20 12 a.m.)
13 CHAIRMAN BROWN: The meeting will come 14 back to order, and you may proceed now.
15 MR. MCKENNA: Sure. So now, we'll get 16 into the part where we'll brief Appendix D. So that's 17 the introductory slide, we'll move on, backwards.
18 Okay.
19 So like we have discussed in previous 20 discussions, Appendix D, the RIS 2000-22 Supplement 1, 21 gives the technical aspect of digital INC 22 modifications, not the 50.59 process. So that is why 23 Appendix D was written, to give the 50.59 process for 24 digital INC modification.
25 CHAIRMAN BROWN: Excuse me, Appendix D NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
79 1 then is the process and the RIS is the technical part?
2 You talked about --
3 MR. MCKENNA: That is correct.
4 CHAIRMAN BROWN: -- technical advice 5 evaluation process.
6 MR. MCKENNA: Right. So Appendix D is 7 written in the same format as 96-07, so the paragraphs 8 actually align.
9 CHAIRMAN BROWN: Yes.
10 MR. MCKENNA: So it's supplemental 11 guidance for 96-07.
12 CHAIRMAN BROWN: Okay.
13 MR. MORTON: But I do need to make one 14 nuanced clarification to, in addition to what Phil 15 said, the RIS supplement does not provide screening 16 guidance. It's specifically for the evaluation 17 portion.
18 CHAIRMAN BROWN: Yes.
19 MR. MORTON: Just, just make it clear.
20 CHAIRMAN BROWN: Well but the RIS talks 21 about qualitative assessments --
22 MR. MCKENNA: So the qualitative 23 assessment portion is used --
24 CHAIRMAN BROWN: That, that --
25 MR. MCKENNA: -- so you can pass the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
80 1 evaluation questions.
2 CHAIRMAN BROWN: So that is -- but, but 3 yet, well how do you ever get to the -- I thought the 4 assessment, I thought you did. You had to do an 5 assessment before you get to evaluation.
6 MR. MCKENNA: You do screenings --
7 CHAIRMAN BROWN: Screenings also involve 8 qualitative assessments.
9 MR. MCKENNA: The qualitative 10 assessments --
11 CHAIRMAN BROWN: Qualitative assessments 12 also --
13 MR. MCKENNA: It's not used for the 14 screening portion. So you can --
15 CHAIRMAN BROWN: But -- well, well, how do 16 you do a screening then? Sit around and --
17 MR. MCKENNA: So this is what Appendix D 18 will cover.
19 CHAIRMAN BROWN: Yes.
20 MR. MCKENNA: Appendix D covers the 21 screening portion of 50.59.
22 CHAIRMAN BROWN: But I read the screening 23 portion and it certainly looked like you were making 24 qualified --
25 MEMBER BLEY: I think you were doing NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
81 1 qualified judgment, but he's going to tell us.
2 CHAIRMAN BROWN: Okay.
3 MEMBER BLEY: I think.
4 MR. MCKENNA: Remember for the screening 5 portion, you're only deciding if the modification is 6 adverse or not adverse. Right?
7 CHAIRMAN BROWN: But there was qualitative 8 assessment to do that. That's the way I read it.
9 MR. BEAULIEU: You're correct in -- it, 10 the initial part, it, it does deal with the same 11 aspects that are covered in the qualitative 12 assessment. However, it had some additional 13 considerations that give you an additional degree of 14 confidence that, that goes beyond just the qualitative 15 assessments.
16 It has to be like very simple design. It 17 has to be, I forget the additional considerations.
18 But you're right, the way it's written, it kind of 19 uses qualitative assessment as a starting point. But 20 you need greater assurance than that.
21 CHAIRMAN BROWN: Well isn't that section 22 of -- I'm going to skip a little bit here because 23 then, I, I really, I've really kind of get a feel for 24 how we get this screening done relative to that.
25 Because Appendix D leaps into and adds a new section NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
82 1 3.15, which is largely determining whether you're 2 sufficiently low.
3 It's a qualitative assessment, what the 4 title is. And then you get down and you determine 5 whether you're sufficiently low or not sufficiently 6 low, or what have you through the screening. But, and 7 then, the screening process follows in section 4.2.
8 MR. MCKENNA: So maybe this is a good 9 clarification for, in the rule, the only documentation 10 required is the --
11 CHAIRMAN BROWN: Is the 50.59.
12 MR. MCKENNA: 50.59 rule is the 13 evaluation.
14 CHAIRMAN BROWN: I got that part.
15 MR. MCKENNA: Okay. So if you do a 16 screening and you're using the qualitative assessment 17 portion of the screening, that documentation is not 18 required to be retained.
19 CHAIRMAN BROWN: Is not required to be 20 what?
21 MR. MCKENNA: Be retained. So it's part 22 of the rule. So you're screening as adverse or 23 non-adverse.
24 CHAIRMAN BROWN: Okay. But when, if --
25 MEMBER BLEY: So if, a just thought. If NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
83 1 you don't retain that, then you can't inspect, the NRC 2 can't inspect to see if they did a reasonable job of 3 screening. I don't think. Short cut -- or don't you 4 even inspection?
5 MR. MCKENNA: So I'll -- I don't want to 6 backtrack here, but I will backtrack a little bit. So 7 all, all licensees keep the screening portion. It's 8 in their procedure.
9 CHAIRMAN BROWN: That's what I would 10 think.
11 MR. MCKENNA: Yes, yes.
12 CHAIRMAN BROWN: But you're just saying, 13 the rule doesn't --
14 MR. MCKENNA: The rule doesn't required 15 it. Right.
16 MEMBER BLEY: The Appendix.
17 MR. MORTON: Let me turn to something that 18 Charlie was or Member Brown was asking. Don't, please 19 don't confused with the engineering 20 judgment/qualitative assessment you may be doing to 21 determine whether a modification is adverse or not 22 with the qualitative assessment as endorsed by the RIS 23 Supplement 1.
24 CHAIRMAN BROWN: Yes. The RIS Supplement 25 applies to the evaluation itself. Correct?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
84 1 MR. MORTON: Well --
2 CHAIRMAN BROWN: I got that --
3 MR. MORTON: -- you're making an 4 engineering judgment/qualitative assessment in the 5 screening section, but it's just to determine whether 6 the modification is adverse or not.
7 But they are different criteria in 8 screening that aren't necessarily covered by the RIS 9 Supplement 1, which is why we tried to be very clear 10 that it is not specifically for the screening portion 11 of 50.59. So I just wanted to clarify that point so 12 there wasn't confusion that it applies to the entirety 13 of the process.
14 MEMBER BLEY: So that's the C2 one to 15 eight issues?
16 MR. MORTON: Correct.
17 MEMBER BLEY: That's the evaluation.
18 MR. MORTON: That's the qualitative 19 assessment as per RIS Supplement 1, the 2002-22.
20 MEMBER BLEY: Okay. Evaluation?
21 MR. MORTON: Evaluation.
22 MEMBER BLEY: That's the evaluation 23 process --
24 CHAIRMAN BROWN: Don't you have to look at 25 those as well to see if you can screen it?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
85 1 MEMBER BLEY: You don't have to.
2 CHAIRMAN BROWN: How do you decide if it's 3 adverse?
4 MR. BEAULIEU: No because the weight, I 5 sorry, repeating here a -- the reason the screening 6 guidance is, just deals with the direction of, of, the 7 change in terms of, can it have an adverse impact on 8 the ability of the system to perform its design 9 function?
10 Some changes have a positive impact. So, 11 so you, it's that. So the screening deals with the 12 direction. Is, is it positive impact or negative 13 impact on the design function?
14 And if it's, if it doesn't have a negative 15 impact, then it screens out because there's no way 16 that, that any of the evaluation criteria could be 17 tripped.
18 MEMBER BLEY: I'm sorry but the, the 19 criteria that you use for evaluation, the eight 20 criteria from the rule, if you don't ask those kind of 21 questions when you're doing the screening, you might 22 not see that it was adverse. I mean those are kind of 23 general issues that you're trying to see, could this 24 move it into a bad direction?
25 MR. BEAULIEU: I suppose you could --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
86 1 CHAIRMAN BROWN: I tried to use, I tried 2 to use their examples to get a feel for this. They've 3 got several examples under the screening section. And 4 if you look at -- this is me talking now relative to 5 this and I've got it open.
6 They talk about the screening response, 7 digital transmitters use a relatively simple digital 8 architecture. Well that's, that's the same kind of 9 thought process --
10 MEMBER BLEY: Simple.
11 CHAIRMAN BROWN: -- I would use in a 12 evaluation, also. But I mean, it's, it is, it is not 13 a detail, it's just, I mean, you can have very simple 14 digital components, which with a limited programming 15 set that are fairly, that their architecture's pretty 16 straightforward.
17 And then, then they look at failures of 18 the device that were encompassed by the failures of 19 the existing analog device. In other words, it can't 20 file, fail differently than that one.
21 MEMBER BLEY: Right.
22 CHAIRMAN BROWN: But yet, if you look at 23 the rest of the paragraph. It certainly looks like 24 you're evaluating it from a -- there's qualitative 25 thought processes in there. And you know --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
87 1 MEMBER BLEY: Yes, yes, I keep going back 2 to the C2-6, which is introduce a new failure mode.
3 Very simple system could introduce a new failure mode.
4 CHAIRMAN BROWN: Absolutely. I guess --
5 MEMBER BLEY: Yes. So you must be using 6 that structure. We'll talk to the guys who represent 7 the people who do this.
8 CHAIRMAN BROWN: Any way, I just, I just 9 was trying to get a feel after reading any 96-07, the 10 Appendix D, the RIS, and there was oh, the Reg Guide.
11 I was trying to figure out then, and say, okay, how to 12 do parse screening from evaluation? And I --
13 Seemed to me screening was a qualitative 14 thing, but I want the assessment, as well. And the 15 3-15 was set of qualitative -- how do you do a 16 qualitative assessment.
17 Then you come along later in your Reg 18 Guide and say, throw 3-15 out. We don't recognize 19 that. The only qualitative assessment we recognize is 20 in RIS Supplement 1. Well does the 3-15, is that not 21 for the adverse? Or was that meant in the Appendix D 22 to apply to the evaluation part, 4.3?
23 MR. WATERS: So I mean, I think you're, 24 you're right. Technically speaking, the screening 25 portion is a qualitative judgment. Is it adverse or NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
88 1 not? From a process standpoint, under 50.59, you 2 know, a plant makes thousands of changes from small 3 piece parts to bigger ones.
4 And the question is, then you do a full 5 50.59 evaluation, it's only criteria for every single 6 change they make. So part of the process would be, 7 we've adopted under, by endorsement in 89607, before 8 digital was the ability to screen things out from that 9 full evaluation, the criteria being is it adverse or 10 not.
11 I think you are correct. It is a 12 qualitative judgment and in day. But you've picked up 13 on a struggle we had in this, this guidance here. Is, 14 what, what does it mean to be adverse for dozer 15 modification?
16 I think we started from a standpoint, from 17 a conservative standpoint where if it was digital, 18 usually screened in, you had to run the full 19 evaluation, even if it was small, small digital piece.
20 So part--
21 Part of what we did in working with our 22 endorsement review in Appendix D is, what digital 23 modification can screen out because we're not adverse?
24 And have you ran a full evaluation?
25 But you're right. We, we reserved the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
89 1 term qualitative assessment only for the evaluation 2 portion of the 50.59. Not for the screen, not for the 3 screening portion.
4 CHAIRMAN BROWN: Yes.
5 MR. MCKENNA: And, and one of the goal's 6 again for Appendix D was to remove the regulatory 7 uncertainty. And right now, there's no screening 8 guidance out there. So putting a screening guidance 9 into Appendix D removes some of the regulatory 10 uncertainty.
11 MR. MORTON: Right. And as Mike was 12 alluding to, one of the clarifications made, not 13 clarification in the Reg Guide, but what's actually 14 written I believe in that section that you're looking 15 at, is just because it introduced the software, the 16 modification doesn't necessarily mean it screens in, 17 doesn't necessarily mean it's adverse.
18 So some of the criteria you're looking at 19 in terms of a simple device, a small amount of 20 input/output, these are ways to add additional 21 criteria to determine well, okay, it may have 22 software, but it may be simple, highly testable, 23 something to that extent.
24 And it may be not necessarily be at 25 adverse. So that's what we're referring to. And NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
90 1 that's generally an engineering judgment of 2 qualitative assessment. So really we don't disagree 3 with that.
4 CHAIRMAN BROWN: Okay. Well 4.2 was 5 fairly, in the original 96-07, non-Appendix D, the 6 regular 96-07, very, very generalized. I mean, does 7 the activity decrease the reliability of an SSC design 8 feature? Well that's kind of a, that's kind of -- you 9 have to sit there and think about that.
10 MEMBER BLEY: Charlie, you've got somebody 11 wanting to talk. Excuse me.
12 MR. LEBOND: Hi, I'm Peter LeBlond and I'm 13 here primarily with Neil to talk about the 436 14 discussion. But perhaps, I can assist. I was one of 15 the authors of the original 1999 rule change.
16 First, sufficiently low, not sufficiently 17 low, it's really a surrogate for the language of 18 creating a possibility that's used in criterion five 19 and six.
20 MEMBER BLEY: Now is this screening or --
21 MR. LEBOND: No, this is the evaluation.
22 MEMBER BLEY: Okay.
23 MR. LEBOND: So, so when the folks say, 24 well not sufficiently low is primarily used in, in the 25 evaluation phase, it's really used as surrogate to, to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
91 1 answer the, the sub-element of criterion five and six, 2 create a possibility.
3 If it's sufficiently low, then from a 4 legal standpoint you can say it doesn't create a 5 possibility and the issue of what a different result 6 is doesn't come into play. So --
7 CHAIRMAN BROWN: And an LAR is not 8 required?
9 MR. LEBOND: Correct. Yes. In four --
10 CHAIRMAN BROWN: And I got that.
11 MR. LEBOND: -- five and six, it may be 12 required for one or the other criterion.
13 CHAIRMAN BROWN: But that's a different 14 circumstance.
15 MR. LEBOND: Different, different issue, 16 right. Right. Right, another rabbit hole. Okay. So 17 it's not used, that concept as Wendell and Phil were 18 saying. And they're right, it's not really used in 19 the screening process.
20 However, in 96-07, there is a statement 21 that was expanded upon in Appendix Delta. It says, 22 for those changes that introduce the possibility of a 23 new type of action or malfunction, they screen in.
24 And that is 421, if memory serves, and that's 96-07.
25 And on 4211, it repeats that logical NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
92 1 Appendix Delta that says, if you put software into 2 redundant systems then the assessment that you're 3 talking about is just a go, no go.
4 And the reason for that is because the 5 process doesn't handle when you say, adverse impact, 6 on a design function. Well if you introduce a new 7 possible hit on five and six, screening can't handle 8 it.
9 So you've got to disposition it under five 10 and six. And that's why the language was there in the 11 original 96-07. And that's why it's a go, no go in 12 Appendix Delta. Once you put in software, into a 13 redundant systems it says, --
14 CHAIRMAN BROWN: That was in 4.2.1, you 15 said?
16 MR. LEBOND: I'm looking at the second 17 paragraph under discussion 4211. On this basis, most 18 digital modifications, redundancy systems are adverse.
19 And that expands in that one sentence that's embed in 20 4.21 of 96-07 for the reasons we've been talking 21 about.
22 CHAIRMAN BROWN: What paragraph is that?
23 MR. LEBOND: Second paragraph. I'm on, 24 copy, I don't have my --
25 CHAIRMAN BROWN: I've got it open, so at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
93 1 4211. Right?
2 MR. MORTON: Well at open at --
3 MR. LEBOND: 4.211, discussion for 4 redundancy safety systems.
5 MR. MORTON: Charlie, I don't want to 6 interrupt, I just want to clarify for the record. Mr.
7 LeBlond is not a member of staff.
8 CHAIRMAN BROWN: I know.
9 MR. MORTON: Okay. I just wanted to make 10 sure.
11 CHAIRMAN BROWN: He's an industry --
12 MR. MORTON: Yes. Okay.
13 CHAIRMAN BROWN: -- industry person. I'm 14 not, I'm not in any --
15 MR. MORTON: That's, that's fine, sir.
16 CHAIRMAN BROWN: I understood that.
17 MR. LEBOND: Well I'm not taking issue 18 with anything anybody said, so --
19 CHAIRMAN BROWN: No, that's, that's fine.
20 You're allowed to talk.
21 MR. MORTON: But we're not either. I just 22 want to make sure that's clear.
23 CHAIRMAN BROWN: No. I, I get that.
24 I'm just trying to find what he said in 4211 of 25 Appendix, I'm looking at -- oh I'm sorry, I'm not NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
94 1 looking at Appendix D. I'm looking at the original 2 96-07, not the -- so you say that was amplified then 3 in Appendix D?
4 MR. LEBOND: Okay, the sentence that I 5 referred to in 96-07 is buried. The sentence -- I 6 don't have a copy of 96-07 with me.
7 CHAIRMAN BROWN: I --
8 MR. LEBOND: But embedded in a larger --
9 CHAIRMAN BROWN: Okay. Yes, I see. It's 10 the second paragraph under the use of software.
11 MR. LEBOND: -- paragraph there. It's one 12 or two sentences embedded in --
13 CHAIRMAN BROWN: Yes. I got it.
14 MR. LEBOND: Okay.
15 CHAIRMAN BROWN: I know I read through 16 that, but I guess I didn't highlight it. Okay. I 17 just -- obviously I'm sitting out here on the edge.
18 Glad I don't work in the plants. I'd fail open.
19 MR. MCKENNA: So the next slide is, is, 20 we've already hit a lot of these bullets. But this is 21 how we develop their, or discussed with NEI along the 22 road of submitting the document and holding the public 23 meetings, and how we got to this point right now.
24 Chris Miller covered this in his opening 25 set of remarks. So we received the Appendix D back on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
95 1 November 30th and request for endorsement on 2 January 8th, after our initial public meetings over 3 the past two years. Wrong way, yes --
4 Okay. So this is what Member Brown has 5 already started to discuss, the screening section of 6 Appendix D. Again, the scope of digital modifications 7 can be software related. They can be hardware 8 related. They can be Human System Interface related 9 activities. Again, the goal of screening in 50.59 is 10 to reach a conclusion of non-adverse or adverse.
11 And again, what Member Brown had mentioned 12 already was some of the things that are written into 13 Appendix D are the same criteria that we would use in 14 the formal qualitative assessment portion if we were 15 to get a screening of non-adverse and moving on 16 evaluation portion of 50.59.
17 CHAIRMAN BROWN: You said that you 18 screened as non-adverse?
19 MR. MCKENNA: If you --
20 CHAIRMAN BROWN: And move on to the 21 evaluation? You, you only do that if you're --
22 MR. MCKENNA: Sorry, I had I had it 23 backwards. Yes, if you're adverse.
24 CHAIRMAN BROWN: I thought I heard the 25 word.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
96 1 MR. MCKENNA: Yes, sir. Yes, we would not 2 throw it backwards. You heard me correctly. I said 3 it backwards. Again so in the screening section of 4 Appendix D. Some clarifications that the mere act of 5 the combining functions or components and systems does 6 not make the screen adverse. If it cause --
7 Obviously if it causes an adverse act on 8 the design function then it would be adverse. And 9 reductions in redundancy, diversity, or separation, or 10 independence of the, in the SR design function would 11 screen adverse.
12 The human factors section in Appendix D 13 was -- NEI worked closely with our human factors 14 personnel and NRC to develop that. And there's two 15 steps in that, identify the generic primary task and 16 mouth.
17 And then for all those primary tasks that 18 you identify, assess of that, if the mod negatively 19 impacts those primary tasks.
20 MEMBER BLEY: Let me take you back to the 21 Reg Guide and read a sentence there. And I want to 22 ask you a couple of questions about it. It says 23 regarding Appendix D, it's understood by the NRC staff 24 that screening human system interface changes is an 25 exception from the guidance contained in 96-07 Rev. 1, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
97 1 since HSI is not discussed in the document.
2 And then it goes on, digital is kind of 3 unique here. I, I get what it says here. But if I 4 have an analog system, the HSI is just as important as 5 digital. And if there's no guidance in the base 6 document it seems we're missing. We've got a gap 7 here, unless you don't figure we'll ever get anything 8 analog coming in again.
9 MR. CARTE: Norbert Carte. So 96-07 does 10 have a statement that says that basically if you 11 change the HSI it should come in, it should be 12 evaluated. So there is --
13 MEMBER BLEY: It just doesn't tell you how 14 to do it.
15 MR. CARTE: Right. It says, if you're 16 changing the HSI, do an evaluation. In other words, 17 it, it screens in and you do an evaluation.
18 MEMBER BLEY: Okay.
19 MR. CARTE: So what Appendix D does it 20 allows you to do an actual screening of the change and 21 make the determination. Rather than just saying, oh 22 we changed something in the HSI, go do a full 23 evaluation.
24 MEMBER BLEY: Right. Okay.
25 MR. MCKENNA: Now moving on to the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
98 1 evaluation section of Appendix D. The evaluation 2 section again, section 4.31 aligns with the main body 3 of 96-07. Again it's the supplement, the guidance in 4 the main body of 89, 96-07.
5 And for digital modification criteria one, 6 two, five, and six are the evaluation criteria that 7 apply for a digital modification. Criterion three and 8 criterion four, which are sections 4.3.3 and 4.3.4 9 state that they provide no new guidance for digital 10 modifications.
11 Section 4.3.6, does the activity create a 12 possibility of a malfunction of a necessity important 13 to safety for a different result? We will discuss 14 this more in detail in the next couple of slides.
15 Just highlighting some portion of Appendix D there.
16 There's a discussion on a design basis 17 functions and the connection between the design basis 18 function and the safety analysis result.
19 And the overall perspective out of section 20 4.3.6 is unless the equipment would fail in a way not 21 already evaluated in the safety analysis there can be 22 no malfunction of a necessity important to safety with 23 a different result. The current words in Appendix D.
24 So for Section 4.3.6 there's a six step 25 process. The process five and six which are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
99 1 highlighted in bold are the ones that we take 2 exception with in our endorsement in the Reg. Guide.
3 I'll let you read through the six steps.
4 And then, I'll go down to steps five and 5 six, which is, the way it's written in Appendix D is 6 identify all the safety analysis involved and then for 7 each safety analysis involved, compare the projected 8 possible results with the previous evaluator results.
9 So from step five in that six step process, 10 which I discussed, Appendix D states that if there's 11 no safety analysis involved, then there cannot be a 12 change in the result of the safety analysis.
13 Therefore, the proposed activity does not create a 14 possibility of a malfunction of an important safety of 15 a different result.
16 We disagree with this in 4.3.6 in that it 17 should determine the SSC malfunction instead of the 18 impact of the result on the safety analysis as a 19 facility on a whole.
20 MEMBER RICCARDELLA: That's a, that's a 21 subtlety. Could you help me with that, please?
22 MR. MORTON: Member Bley talked about the 23 example 4-19, about the radiation monitor example.
24 That's basically, that example is essentially 25 leveraging this piece of the guidance in Appendix D.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
100 1 So the way that you could interpret that and the way 2 they run through the six step process for that --
3 You have an analog system, software 4 comment cause failure would not necessarily be part of 5 its original licensing basis if it was an originally 6 an analog system, then it was not digital.
7 So therefore, a software comment cause of 8 failure would not have been part of the original 9 failure analysis, FMA, or whatever. Analysis would 10 have existed and wherever it would have existed in the 11 FSAR.
12 If you are upgrading that particular 13 system to a digital control system for this area 14 related issued monitors, you -- at a minimum you would 15 introduce one new particular failure by which is 16 something based upon software, your software comment 17 cause of failure.
18 The way you can interpret that particular 19 guidance is that if you don't have a software based 20 type of failure pre-existing within your licensing 21 basis to compare against the similar type of failure.
22 Then they're saying that there's not a pre-existing 23 analysis talking about that. Therefore, it can't 24 really be a different result.
25 And therefore, it's really a criterion two NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
101 1 issue, not a criterion six issue, which is part of the 2 reason we took exception to both the example and that 3 little note that's within the example because that's 4 a kind of two, two for one deal in terms of guidance.
5 And we disagree with and took an exception 6 to -- and then saying, that hey, this particular, in 7 this particular modification, criterion six would not 8 even apply or it would, it's really a reliability 9 criterion two issue.
10 Sort of, sort around the discussion, back 11 to what we talked about previously in terms of why we 12 got to point with that particular example.
13 MEMBER BLEY: I think I agree with you 14 folks.
15 MR. BEAULIEU: Could I elaborate on that?
16 To it's an important question, that you indicated it's 17 a subtlety. What's the difference between whether you 18 consider a safety analysis of each individual SSC, 19 which is the definition in which the regulations say?
20 Or do you say it's a safety evaluation means just the 21 plant as a whole?
22 That makes a huge difference because when 23 you say, if there's different result, you're comparing 24 the result. Are, are you looking just at the 25 accident, Chapter 15 accident analysis, from, to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
102 1 determine whether there's a different result?
2 Like, you know, do you compare everything 3 to a large break LOCA and then if you don't, if you're 4 within, if you're bounded a large break LOCA, there's 5 not a different result? Is that what the Commission 6 intended?
7 The answer's no because they could have 8 written it. They could have written the regulation 9 that way, but they didn't. They, they said, 10 malfunction of an SSC.
11 So the SSC previously evaluated and most 12 of the time those evaluations are, many times, they're 13 in a failure modes and effect analysis is a table. It 14 says, a pump, you have discharge valve. And if a --
15 Usually malfunctions are considered 16 single failures. So, so in the failure modes and 17 effect analysis, we'll say one train fails, and it'll 18 say the effect of that is we're good.
19 We have redundant trains, 100 capacity, 20 plant is safe. That's were that's a, that is where 21 that is previously evaluated is in the chapter for 22 that system. And often it's in the failure modes and 23 effects analysis for that system. So --
24 So therefore, if you have a common cause 25 failure, if you just assume the, there's a software NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
103 1 common cause failure, well that failure mode is --
2 that's a different result for every safety system.
3 The --
4 Virtually every system, safety system, the 5 system is credited in one way or another in the 6 accident analysis. The thing is it might not, every 7 system might not be explicitly described in the 8 accident analysis.
9 The accident analysis might not describe 10 control room chillers, for example. They might not 11 describe, you know. So, so the regulation says 12 anywhere previously evaluated in the FSAR, not the 13 accident analysis.
14 Anywhere in the FSAR, that's what they're 15 talking, that's, that's the really distinction between 16 the two, which, which, what do you compare as a 17 different result.
18 And it makes a huge, makes a huge 19 difference in terms of, like what I said, preparing 20 everything to large break LOCA. Is that what the 21 Commission intended? No. And it's not what the 22 regulation says.
23 CHAIRMAN BROWN: You guys went so fast 24 that I lost the bubble on something. If you could, if 25 not synopsis are not synopses. Can you go back Slide NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
104 1 26?
2 Here you say in 4.3 1, 2, 5, and 6 the 3 criterion and discuss the use of the qualitative 4 assessment outcome. So efficiently go -- now this is 5 the evaluation part now, this is not screening?
6 Correct?
7 MR. MCKENNA: That's correct.
8 CHAIRMAN BROWN: But yet in the document, 9 they've got a section in Appendix D called -- again, 10 back to 3.15, 315, rather 3.15. That was a 11 qualitative assessment.
12 And you all pushed that aside and said 13 only RIS 2002-22 was the only one we've evaluated in 14 terms of doing qualitative assessments. Now is 3.15 15 applicable to evaluations? Or is it not applicable to 16 evaluations?
17 MR. MCKENNA: So now I understand what 18 you're talking about? So that's where Appendix D 19 introduces the qualitative assessment. And does not 20 speak about it anymore. It's only in Section 3.1.5 is 21 where --
22 CHAIRMAN BROWN: Point .15 is that.
23 MR. MCKENNA: Right. Sorry. Yes.
24 That's, that's the only place that Appendix D talks 25 about it. But it's, like that qualitative assessment NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
105 1 is meant to be used in the evaluation process.
2 CHAIRMAN BROWN: And you all have said 3 the only one we've recognized is the RIS qualitative 4 assessments? So that effectively negates 3.15 from 5 the Appendix D from being utilized as part of an 6 Appendix D evaluation?
7 MR. MCKENNA: No. I believe Appendix D is 8 -- that's where they're talking about the qualitative 9 assessment process, in that section.
10 MEMBER BLEY: They say you have to do it.
11 They don't tell you how in there.
12 MR. MCKENNA: Right. That, that is not 13 part of the screening section. That is part before 14 the screening section. So they're introducing the 15 qualitative assessment portion in Appendix D there.
16 MR. MORTON: Right. So they introduce the 17 concept of the qualitative assessment. One of the 18 things you'll see with the Reg Guide is that 19 originally there was reference to a generic 20 qualitative assessment without putting context around 21 what does that mean.
22 So one of the clarifications made of Reg 23 Guide is to clarify RIS 2000-22, Supplement 1, is the 24 qualitative assessment method that we have 25 specifically endorsed --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
106 1 CHAIRMAN BROWN: Well --
2 MR. MORTON: -- as applied to it.
3 MR. MCKENNA: So Member Brown, that's, 4 that's in their definition portion.
5 CHAIRMAN BROWN: Yes. But the --
6 MR. MCKENNA: So they're defining --
7 CHAIRMAN BROWN: I'm, I'm reading your, 8 your item D, and 1.187 Rev. 2. It says, you all, you 9 provide additional guidance or a position, the names 10 have changed from, from older Reg Guides. And so you 11 quote a qualitative assessment.
12 This is a specific type of technical based 13 engineering evaluation useful to 10 CFR 50.59 14 evaluations. Okay? And applying to one, two 15 criteria, one, two, five, and six. Then you go on.
16 That's, that's the quote.
17 Then your statement, gone on and say. The 18 staff's position is that any NEI 0101 Section 5 is 19 clarified. Is the only guidance the NRC has reviewed 20 or endorsed. But when I read 3.15, it wasn't -- I 21 mean, it was a qualitative assessment. It talked 22 about doing a qualitative assessment. And it had, it 23 was not just blank. But I -- I'm trying to see how, 24 how the qualitative assessment in 3.15, if, it, has it 25 been screened out? Not screened out, has it been NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
107 1 overwritten by your position in the Reg Guide?
2 MR. MCKENNA: Again 3.15 is only the 3 definition section. It's not part of the process of 4 96-07.
5 CHAIRMAN BROWN: Let's go and look.
6 MR. MCKENNA: They're only defining the 7 qualitative assessment.
8 MEMBER BLEY: I don't think there's 9 anything in 3.15 that's --
10 MR. MCKENNA: It's not snap --
11 MEMBER BLEY: -- inconsistent with what's 12 in the RIS.
13 MR. MCKENNA: It's, it's not a step in the 14 Appendix D process. It's a definition in Appendix D.
15 MR. WATERS: All right. I, I think, well 16 maybe, I understand the question correctly. But I 17 think part of the history here is NEI, as authors of 18 this documents, hasn't endorsed. They want to leave 19 open the option that there may be than one way to do 20 a qualitative assessment.
21 And one day they may have detailed 22 guidance they may ask us to endorse. And they may 23 have something else. So generally, the quality of the 24 assessment is introduced in that manner. What we're 25 generally saying is, the only one when we recognize NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
108 1 right now is the --
2 CHAIRMAN BROWN: Okay. Hold on.
3 MR. WATERS: -- the one that we have that 4 is part of the RIS.
5 CHAIRMAN BROWN: Think back for a minute.
6 You say it just says, it's a definition. But in the 7 fourth, under the third paragraph.
8 It says, generally reassure, responsible 9 assurance of a low likelihood of failure is derived 10 from a qualitative assessment of factors involving the 11 design attributes of the modified SSC, the quality of 12 the design processes, and the operating experiences of 13 the software and hardware used, product maturity, in 14 service experience.
15 To me, that sounds like, here's some 16 attribute that you have to assess. And but yet, 17 you've said, no, RIS 2000-22 is a definition. I'm --
18 MEMBER BLEY: I think it's like we usually 19 see in their documents. This, this is kind of high 20 level. So if you do what's in the RIS, you will have 21 done these things. They're saying, right now, they 22 approve what's in the RIS.
23 If, if you want to submit something 24 different that meets this criteria and see if the 25 staff likes, you're free to do that. It's just a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
109 1 little riskier for you.
2 MR. MCKENNA: I just want to again clarify 3 that Section 3.15 is not a step in a Appendix D.
4 CHAIRMAN BROWN: I understand it's a 5 definition. Okay. Yes. I understand that and then 6 there's a discussion of it. All right. And then you 7 might want to answer one other one for me. No, that's 8 the wrong question. Where is it?
9 MEMBER SKILLMAN: May I ask a question?
10 CHAIRMAN BROWN: I'll, I'll wait until I 11 get to the Reg Guide, 1.17. It has to do with human 12 systems interface thing.
13 MEMBER SKILLMAN: May I ask, on Slide 29 14 you've identify the NRC's position that you're focused 15 on the SSC versus the overall system response. What 16 is the status of this disagree between the NRC and 17 NEI?
18 MR. MCKENNA: So we've discussed this 19 disagreement in every single public meeting we held in 20 the comment phase of the guidance. And we, there's 21 disagreement between the NRC and NEI. That's, that's 22 where it stands. And that's why we have why have an 23 exception in the Reg. Guide.
24 MEMBER SKILLMAN: So NEI will we hear from 25 you a little later, what you're going to do with this?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
110 1 MR. LEBOND: Yes.
2 MEMBER SKILLMAN: Thank you. Thank you.
3 MR. LEBOND: We're prepared to address 4 those issues.
5 MEMBER SKILLMAN: Yes.
6 MR. WATERS: And so process wise, the next 7 step is to issue the draft for public comment where 8 stakeholders have an opportunity to comment on our 9 endorsement review including this issue.
10 MEMBER SKILLMAN: Thank you, Michael.
11 MR. WATERS: Okay.
12 CHAIRMAN BROWN: So fundamentally, this 13 particular thing, just to make sure I understand.
14 This is just a, the disagreement is between whether 15 you focus on the safety analysis for determining, 16 determining whether something is sufficiently low, as 17 opposed to not? I mean --
18 MR. WATERS: So, I mean, there's many ways 19 to explain this. This is so, you got to realize, this 20 is a modification where the first thing you're able to 21 do is do a qualitative assessment to determine 22 software common cause failure is sufficiently low.
23 If you can answer that question as it 24 sufficiently low, you've addressed criterion six as an 25 unusual, new type of malfunction. This goes into a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
111 1 situation where you have a digital modification, I 2 guess, in there for some reason you cannot demonstrate 3 as a sufficiently low.
4 It may be a new type of malfunction. It 5 may not result in different results. So the question 6 is, what does is, what do we mean by different result.
7 What, what do you compare? I think that's what Dave 8 is trying to explain. In other words do you compare 9 that they're all plant response level or at a system 10 level?
11 Early on decisions, some interviews said, 12 hey the minute we assume multiple trains though, we 13 have a hard time answering this question as being a --
14 there's going to be different result. That's from the 15 feedback we got. And why we focused on the 16 qualitative assessment.
17 I believed NRC staff believes both based 18 on our engineering safety judgment and what the 19 commission intended, if you cannot demonstrate a 20 common cause for a sufficiently low for these systems, 21 and you may have a different result as far, at the 22 system level, you know.
23 Two trains do not work, for example. It 24 still may be safe. But that crosses a threshold where 25 that safety demonstration should come to NRC for come NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
112 1 a license approval. And he may well show that 2 multiple trains can fail and it's safe, but this is a 3 threshold test wherein they should come to NRC for 4 license, licensing approval.
5 CHAIRMAN BROWN: But that assumes that the 6 multiple trains failing wasn't as part of the initial 7 safety analysis, or FSAR?
8 MR. WATERS: Yes, I mean, I, I, and I 9 don't want to speak and please correct me that. You 10 know, we looked at a few FSARs and some say, for 11 example, a chiller fails everyone starts. That, that 12 could be the extent of the FSAR.
13 CHAIRMAN BROWN: Say that again. If a 14 chiller thing --
15 MR. WATERS: If one chiller one fails, a 16 second one will start and provide, provide the 17 cooling. That could be the extent of what the FSAR 18 says. It does, it -- and you know, that's, that's the 19 basis. Okay?
20 CHAIRMAN BROWN: And therefore, if now 21 both of them fail?
22 MR. WATERS: Well, that's --
23 CHAIRMAN BROWN: That's way outside --
24 MR. WATERS: -- that, that may be 25 different results providing that safe, that design NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
113 1 function.
2 CHAIRMAN BROWN: Right.
3 MR. WATERS: You can still make a safety 4 case that the show that is acceptable, but we believe 5 that cross is the threshold requiring a NRC review for 6 approval.
7 CHAIRMAN BROWN: Okay. So they were, they 8 were -- okay, I think I --
9 MEMBER BLEY: But that's, of course, two 10 different of kinds of safety analysis. In Chapter 15, 11 you have an event and you have under the worst single 12 failure.
13 CHAIRMAN BROWN: Yes.
14 MEMBER BLEY: If they did a PRA, you'd 15 look at those combinations, but that's not what 16 they're looking at. They're looking at the Chapter 15 17 kind of analysis.
18 MR. WATERS: Well in the --
19 MEMBER RICCARDELLA: So that's what I was 20 going to ask. On Slide 29, on the first bullet, where 21 you say safety analysis. Are you referring to the 22 whole FSAR or, or just some specific portion of it?
23 MR. MORTON: Within the context of 24 Appendix D.
25 MEMBER RICCARDELLA: This would NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
114 1 Appendix D?
2 MR. MORTON: Appendix D. When they're 3 referring to safety analysis, they're really referring 4 to your Chapter 15 or Chapter 6 asset analysis. Am I 5 correct?
6 MR. MCKENNA: Yes.
7 MR. MORTON: And our interpretation of 8 safety analysis, it's wherever the failure analysis 9 for that SSC resided. It could be in Chapter 15, it 10 could be Chapter 7, or Chapter 10 if you're --
11 MR. MCKENNA: Yes.
12 MR. MORTON: -- talking about temperature 13 of safety chillers or 20 percent redundant two-channel 14 safety chillers, for examples. That's the general 15 gist of the difference is where you're actually 16 analyzing for the different result to resolve 17 criterion six.
18 CHAIRMAN BROWN: So your point, just to 19 use your chiller example, if they combine the 20 functions and they would both fail at the same time, 21 they would, they would argue you can screen that out 22 based on -- not screen it out, wrong term.
23 They could evaluate that it's okay and 24 therefore, we don't have to talk to the NRC. You all 25 would say that no, because the component, the system NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
115 1 has now failed. That's not -- the total system has 2 failed, that's not considered in the accident or 3 safety analysis. Therefore, you should come to NRC to 4 get agreed with that.
5 MR. MORTON: Generally --
6 CHAIRMAN BROWN: Before you make the 7 change.
8 MR. MORTON: Generally speaking, yes. And 9 it's, in a pretty, strict concise way. If it was 10 analyzed only for a single failure, the safety 11 analysis Dave was referring for your two chamber end 12 safety chillers, main control safety chillers. The 13 safety analysis is crediting in an indirect way in 14 that you still have one train of this particular --
15 CHAIRMAN BROWN: I, I got that point.
16 MR. MORTON: If, if you didn't account for 17 both trains for being available in the original 18 analysis, wherever that existed, then that would be a 19 different result.
20 CHAIRMAN BROWN: And therefore, you should 21 see it. But they wouldn't -- their argument would be 22 that they can do the analysis and it shows that they 23 don't care whether the trains are there or not, and so 24 the six. For whatever reasons, it may be that there's 25 no effect for four hours and 59 minutes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
116 1 MR. WATERS: Correct.
2 CHAIRMAN BROWN: People can bring in fans, 3 they can open doors, they can do this, that, or the 4 other thing, and everything's okay. Therefore, we can 5 make the change without, without qualms and not bother 6 the NRC. And you all saying, no, it's a different, 7 within the -- actually within the --
8 MR. MORTON: -- within the act itself.
9 CHAIRMAN BROWN: The end result --
10 MR. MORTON: Right.
11 CHAIRMAN BROWN: -- is different. And 12 therefore, you should cede it. And that's where your 13 disagreement is.
14 MR. MORTON: Generally speaking, yes.
15 MR. WATERS: That's true and we --
16 CHAIRMAN BROWN: That sounds good?
17 MR. WATERS: That's true, we believe -- I 18 think one thing you hear every other --
19 CHAIRMAN BROWN: Like Dennis, I've had 20 hard time with this one. I was reading it --
21 MR. WATERS: I, I just want to reiterate 22 that the reason we do the RIS is you don't get to the 23 point in terms of common cause and sufficiently low by 24 looking at other design features. We don't 25 necessarily want to do a LAR review of chillers.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
117 1 I mean, we're not seeking that, but if for 2 some reason you can't determine it's not sufficiently 3 low and that's a threshold that crossed, come in for 4 an NRC review in theory. And hopefully relief should 5 be simple. I think you'd make a safety case for that.
6 MEMBER BLEY: This is really I think I'm 7 getting this, back to the eight criteria of 50.59 8 reading them essentially literally, which is where I 9 think the staff has in reading them somehow more 10 inclusively might be where we haven't heard from --
11 MR. MORTON: Right. So if --
12 MEMBER BLEY: -- folks yet.
13 MR. MORTON: -- yes, so if you look at the 14 criteria in six language, it's, as described in the 15 FSAR, is updated. It doesn't say a specific part of 16 the FSAR you're looking it. It says, in the FSAR as 17 updated. If you're taking a very little reading of 18 the rule and FSAR interpretation.
19 MEMBER RICCARDELLA: So if you replace 20 safety analysis with FSAR, you guys would be okay with 21 it. Right?
22 CHAIRMAN BROWN: I, I don't think so. I 23 think based on their other, I think based on the other 24 texts that's in there, FSAR's a combination of 25 nothing. It's just already covered. If they had NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
118 1 already covered the failure of both of them --
2 MEMBER RICCARDELLA: Yes.
3 CHAIRMAN BROWN: -- you know, in the 4 overall FSAR --
5 MEMBER RICCARDELLA: Yes.
6 CHAIRMAN BROWN: -- and it was benign, 7 then you would not care.
8 MEMBER RICCARDELLA: But it's the overall 9 FSAR, as opposed to just the Chapter 15 portion of the 10 FSAR. Right?
11 MR. MORTON: That would be correct. Now 12 using the phrase safety analysis isn't and of itself 13 a bad thing. If just when you get into the nuances of 14 50.59 world, what, what do actually mean by safety 15 analysis. And that's where we part ways a bit with 16 NEI on the point.
17 CHAIRMAN BROWN: Well the FSAR is what's 18 quoted in 50.59.
19 MR. MORTON: Right.
20 CHAIRMAN BROWN: Not the safety analysis.
21 MR. MORTON: Correct.
22 CHAIRMAN BROWN: So --
23 MR. CARTE: Norbert Carte, NRC. One 24 nuance -- so there are a couple of different things.
25 So this change or these words are supposed to cover NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
119 1 any type of change. So what we're discussing is say 2 two systems and what happens if you have CCF.
3 But there are other possibilities. And 4 part of the problem with this language is if you 5 create a condition that no one ever thought of there 6 will be nothing in the FSAR that describes it. So it 7 not just a question about if there's nothing --
8 The way this step five is written, if 9 there's nothing described in the FSAR, then it can't 10 be new, which is kind of crazy. So the whole point is 11 if it's new, it hasn't been considered, it hasn't 12 described in the FSAR.
13 So fundamentally, there's nothing in the 14 FSAR to describe it. And, and the reason why look at 15 it is in part to review that the methods they used to 16 address are adequate.
17 MEMBER BLEY: I, Charlie, point of order.
18 Is the staff asked or are intending to write a letter 19 on this at this time, or this information now for 20 some, something we'll dig into later.
21 CHAIRMAN BROWN: I, my general conclusion 22 right now is until get public comments there's a 23 number of issues that ought to be resolved before, 24 that we shouldn't --
25 MEMBER BLEY: I'm more comfortable with NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
120 1 that if that's okay.
2 CHAIRMAN BROWN: Yes. I, I don't think we 3 need to speak to this now. I just want to make sure 4 we haven't adequate description of what their 5 disagreements are so that when we see the resolution, 6 we can go back and look in the transcript and see what 7 our mindless comments were or something like that. To 8 see if we really understood what we were talking 9 about. So, no, but I think my -- after looking, going 10 through all the paperwork and everything, it seems 11 like there was enough here that we would have an 12 information brief for the full committee meeting 13 coming up in May. And then deal with this letter-wise 14 after they finish the public comment routine. I was 15 going to --
16 MEMBER BLEY: Thank you.
17 CHAIRMAN BROWN: -- discuss with everybody 18 here after we finish this. But that was, that's --
19 MEMBER BLEY: It's just starting to weigh 20 on me if we had to write something.
21 CHAIRMAN BROWN: Oh no, this would be very 22 difficult to try sort all the -- we shouldn't be in 23 the matter of sorting. We ought to let them sort 24 before we try to make any pronouncements. I think 25 they'll get enough of a sense of what were thinking NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
121 1 by what we're saying, whether it's intelligible or 2 not, is another question.
3 But they'll at least to get some point to 4 see if we're asking enough questions, anyway. All 5 right. I'm sorry to drag to this out, but I think 6 it's important for us to kind of understand these 7 nuances of what you think.
8 MEMBER SKILLMAN: I mean, let me ask this 9 question. The plants that remain, are aged and many 10 of the remaining plants are struggling to get parts.
11 And in many cases, they're cannibalizing spares. In 12 other cases, they're doing commercial rededications.
13 To what extent have you heard from those stakeholders 14 regarding this criterion six?
15 MR. WATERS: Well that's my colleague's 16 crypt of hoorah (ph.) I, it's very good question.
17 That's why we did the RIS supplement first to address 18 this real time need in 2018-2019 for ops list of 19 concerns for the vast majority of, you know, safety 20 supports systems modification systems.
21 And the focus of that was determine a 22 common cause that are sufficiently low and you can 23 adequately address your criteria. For its Appendix D 24 question and concerning the interpretation of a 25 different result, I, I do not know right now what type NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
122 1 of digital modification could not be made because of 2 that.
3 Or this interpretation is critical to 4 digital modification. I have not heard a specific 5 example. But if, if -- I haven't been to every 6 meeting, if Phil or Wendell heard a specific 7 modification, hey, we can't make this without this 8 interpretation. You asked that question, so that's --
9 MR. MCKENNA: I don't think we're aware of 10 any digital modification that would have gotten held 11 up by criterion six.
12 MEMBER SKILLMAN: Thank you.
13 CHAIRMAN BROWN: Maybe I have submit an 14 LAR for it.
15 MEMBER RICCARDELLA: Yes, but --
16 CHAIRMAN BROWN: It would not be --
17 MEMBER RICCARDELLA: -- I mean --
18 CHAIRMAN BROWN: -- it would just have to 19 go through the process.
20 MEMBER RICCARDELLA: -- yes, but --
21 CHAIRMAN BROWN: By definition they would 22 think that's being held up.
23 MEMBER RICCARDELLA: -- yes, I mean, that 24 makes a significant difference in the schedule to 25 complete a modification? Whether they have to do an NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
123 1 LAR or not?
2 MR. WATERS: Yes.
3 CHAIRMAN BROWN: Yes.
4 MEMBER RICCARDELLA: So it is eventually 5 held up.
6 MR. WATERS: Again though, this, were not 7 talking about protection systems per se. We're 8 talking about other systems where we believe there is 9 a pathway for the vast majority of the systems with 10 the purpose of consideration, which we'll, you know, 11 get to, to determine common cause for a sufficiently 12 low.
13 You addressed the first part of criterion 14 six and any type of malfunction. This, this is a 15 question of whether or not, for some reason, you 16 cannot for some reason answer that.
17 MEMBER RICCARDELLA: Okay. I see. Okay.
18 MR. MORTON: You can determine common 19 cause for sufficiently low, but there were reasons for 20 that. And now, I have to say the what if, what is the 21 different result.
22 MEMBER RICCARDELLA: Yes.
23 MR. MORTON: We're trying to say is we 24 don't, we haven't heard exact feedback of what type of 25 digital modification we cannot determine common cause NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
124 1 that are sufficiently low. And a different result 2 becomes that, that critical determination of whether 3 or not it has to become a law review. Not --
4 MEMBER RICCARDELLA: Yes.
5 MR. MORTON: -- there may be some out 6 there. We just haven't had that specific feedback.
7 MEMBER RICCARDELLA: You could fail --
8 CHAIRMAN BROWN: That's a much better --
9 MEMBER RICCARDELLA: -- and still this 10 criteria and still go ahead with the 50.59, if you do 11 that further evaluation.
12 MEMBER RICCARDELLA: Right. Some, you 13 know, I think someone said, a while that Appendix D 14 provides additional alternatives beyond CCF 15 sufficiently low. And this is where we have this 16 difference of interpretation on this criterion six, 17 different result.
18 CHAIRMAN BROWN: Wouldn't, isn't there a 19 way to categorize stuff? And this is, you know, 20 again, this just my brain moving around. Are 21 non-safety systems largely, couldn't they largely do 22 what they wanted with non-safety systems? Why would 23 you, why do you have to go through an evaluation to 24 deal with non-safety systems?
25 MR. MORTON: Well part of the RIS NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
125 1 Supplement is we cover, as an example of modifications 2 that should go pretty cleanly with the RIS without 3 being a LAR, would be non-safety related 4 modifications.
5 CHAIRMAN BROWN: Yes. I would think stuff 6 like TG voltage regulators, governor's, a controller 7 for miscellaneous pumps through the plant.
8 MR. MORTON: All those mods are things --
9 CHAIRMAN BROWN: Clean water control 10 system. I imagine a feed water control system. I'm 11 not, not, maybe not all the plants, but at least the 12 PWRs, I would think that. Unless you put all, if it's 13 a four loop plant, you put all of the controllers on 14 one chip.
15 MR. MORTON: All those have modification 16 card, generally determining --
17 CHAIRMAN BROWN: That's a total loss of 18 feed water so you probably cover in an SR, I would 19 think. Or some type of plant doing an accident 20 analysis. I, I just, it just seems to me that there 21 ought to be a way to categorize it. I'm, my thought 22 process is to make it as easy as for industry as 23 possible without compromising safety.
24 MR. BEAULIEU: That's, that's -- you're 25 absolutely correct on the feed water example. In NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
126 1 fact, 90, 50.59 does not distinguish between safety 2 related and non-safety related. It, because non-3 safety related systems can have a design function. So 4 it, that's what, that's what the key is for 50.59.
5 But you're absolutely right. Like for 6 feed water it's a non-safety related system, so 7 therefore, it's not credited in the accident analysis.
8 The system description assumes both trains fail. They 9 don't, it assumes a loss of feed.
10 So that's already addressed in 96-07. It 11 says that for a different result, feed water, you 12 might not have a different result because it's 13 already, the analysis of the plant already assumes a 14 loss of both trains, so -- so they won't trigger a 15 variance state.
16 CHAIRMAN BROWN: But, but you talked about 17 it. It's a design function. But when we talk about 18 a design function, is that a safety design function, 19 or is that a plant operation, business type operation 20 design function? I mean --
21 MR. BEAULIEU: Is the definition?
22 CHAIRMAN BROWN: Sounds -- design is a 23 kind of all-encompassing type thing that covers every 24 piece of equipment in the plant. So same if it's got 25 a design function, then you have to go through some NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
127 1 type of an assessment or evaluation. It's just, it 2 was a very general, so broad terms. I don't know, I'm 3 just thinking outside the box, right here, a little 4 bit. Maybe too outside the box.
5 MR. MORTON: There is a definition of 6 design function in NEI 96-07 and it's not specific to 7 a safety related device. It could be for non-8 stipulated systems, too?
9 MR. BEAULIEU: Correct.
10 MR. MORTON: And you're evaluated against 11 that in terms of the screening and the evaluation.
12 The RIS supplement, and you were talking about, 13 categorizing systems.
14 So we although we talked protection 15 systems protection systems not necessarily being the 16 focus of the RIS Supplement, it's intended for 17 everything beneath that, safety chillers, diesel 18 support systems, feed water control mods, integrated 19 non-safety related distributive control systems.
20 It's intended to cover all of those other 21 systems in terms of providing the technical rigor, 22 documentation rigor, and the qualitative assessment 23 itself.
24 Whether you're looking a design features, 25 operating history, and the quality design process to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
128 1 make an engineering judgment of the likelihood of 2 software common cause failure is sufficiently low that 3 I can answer criterion two, no. And I can answer 4 criterion six, no. But that's just for those 5 non-protection systems, generally speaking.
6 So it allows us to address common cause 7 software failure without having to go into BTP 7-19.
8 CHAIRMAN BROWN: Okay.
9 MR. MCKENNA: I'll read you the exact 10 words out of 96-07, maybe that ought to help. Design 11 functions are FSAR describe design basis functions and 12 other SSC function described in the FSAR that support 13 or impact design basis functions.
14 CHAIRMAN BROWN: Design basis functions.
15 Okay. Well that's, that's pretty specific. Okay.
16 That's, I take a design basis is fundamentally if the 17 life in basis function. If I'd, you can almost use 18 those interchangeably? A design basis function is 19 that within the licensing basis under that title also?
20 MR. MCKENNA: Yes.
21 MR. BEAULIEU: That's the definition for 22 design basis function too?
23 CHAIRMAN BROWN: Okay.
24 MR. MCKENNA: That's the argument --
25 CHAIRMAN BROWN: What?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
129 1 MR. MCKENNA: -- NEI's going to present 2 that if you -- several layers down under the 3 definition of design function, they narrow it down to 4 the, where it, you get this one, one leg of that, that 5 says accident analysis.
6 And they're going to say, well, that's, 7 that's, that's where it's accident, that's where we 8 limit it to accident analysis. And we just don't' 9 agree with that.
10 MR. MCKENNA: I'll just read the 11 definition. Design basis functions are functions 12 performed by SSCs that are required by or otherwise 13 necessary to comply with regulations, license 14 conditions, orders, or technical specifications, or 15 credit in the licensee safety analysis to meet NRC 16 requirements. So I think that answers that questions.
17 CHAIRMAN BROWN: That's in 96 -- that's in 18 50.59, that's in --
19 MR. MCKENNA: That's in 96-07 base, 50.59 20 guidance.
21 CHAIRMAN BROWN: Okay. All right. We'll 22 have to -- I'm sorry. Go ahead.
23 MR. MCKENNA: Okay. So we're going to 24 move on now to the development of the Reg Guide. And 25 this will talk about each individual clarification we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
130 1 made or exception. And I think we've had a lot of it 2 all ready.
3 So Revision 2 to the Reg Guide will endorse 4 Appendix D with exceptions. It will soon be issued 5 for public comment, most likely the end of April.
6 CHAIRMAN BROWN: For how long? 30 days?
7 MR. MCKENNA: 60 days.
8 CHAIRMAN BROWN: 60 days.
9 MR. MCKENNA: 60 days. And the first 10 exception that we'll discuss is Appendix D states that 11 the NRC, or sorry. The NRC staff considers Appendix D 12 to be applicable to only digital modifications, not 13 applicable to the whole 50.59 process. Just a 14 clarification. The second clarification --
15 MEMBER BLEY: I'm just curious about why 16 in the first section of Section C, you say, Appendix C 17 and Appendix D are generally acceptable as a means of 18 complying with the requirements of 50.59.
19 MR. MCKENNA: So you're looking at an 20 older version of the --
21 MEMBER BLEY: I am.
22 MR. MCKENNA: -- Reg Guide.
23 MEMBER BLEY: That's simple.
24 MR. MCKENNA: So Appendix C, Appendix C is 25 no longer discussed.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
131 1 CHAIRMAN BROWN: What was that? Say that 2 again.
3 MEMBER BLEY: We don't have the current 4 version.
5 MR. MCKENNA: Yes. So, so you required 6 that 30 days in advance. And we were still going 7 through the mods.
8 MEMBER BLEY: Well if, if it only speaks 9 to Appendix D, then why given what you have up here, 10 which is in the previous section that Appendix D is 11 applicable to digital modifications only, and not 12 generally applicable to 50.59. Over here it says, 13 acceptable as a means of a combined with requirements 14 with 50.59. Maybe you want to --
15 MR. MCKENNA: That, that wording has 16 changed substantially.
17 MR. MCKENNA: Okay. I got that --
18 CHAIRMAN BROWN: I totally lost the 19 bubble.
20 MR. WATERS: I think that sense is --
21 don't, please, I was just trying to talk, you. I 22 think that sense is, we didn't want to - this is 23 highly focused on digital technologies, how to address 24 it in 50.59. We didn't want to inadvertently set 25 precedence for some other type of modification outside NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
132 1 of that.
2 MEMBER BLEY: Well the first statement 3 makes that very clear.
4 MR. WATERS: Right.
5 MEMBER BLEY: The second one kind of 6 nullifies it.
7 MR. WATERS: It says --
8 MEMBER BLEY: If you changed the line, 9 maybe --
10 MR. MCKENNA: But that word, again, that 11 wording has changed substantially.
12 MEMBER BLEY: Okay.
13 CHAIRMAN BROWN: Where, where are you?
14 Where did you read that Dennis?
15 MEMBER BLEY: I've got a draft version.
16 CHAIRMAN BROWN: Are you talking at 1 --
17 MR. MCKENNA: C1.
18 CHAIRMAN BROWN: C1.
19 MR. MCKENNA: Heading, not A.
20 CHAIRMAN BROWN: C1. Blah, blah, blah.
21 Generally acceptable --
22 MR. MCKENNA: So --
23 CHAIRMAN BROWN: Which version are you 24 reading? The NRC staff considers the guidance --
25 MEMBER BLEY: The one you sent me, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
133 1 Charles.
2 CHAIRMAN BROWN: I can't read the header 3 on it. It got wiped out. Oh, April of 2019.
4 MR. MCKENNA: That header is always been 5 there in the development because that's when we've 6 thought that it would be published for public comment.
7 CHAIRMAN BROWN: And serves the guidance 8 of Rev. 1, Appendix C and D, generally acceptable. Is 9 that what you're talking about?
10 MEMBER BLEY: That's what I've been 11 talking about, but that's what he said has changed 12 now.
13 CHAIRMAN BROWN: What's changed?
14 MR. MORTON: The reference to -- oh, 15 sorry.
16 MR. MCKENNA: So that whole paragraph has 17 been --
18 CHAIRMAN BROWN: The lead in paragraph 19 before you hit A --.
20 MR. MCKENNA: That's correct.
21 CHAIRMAN BROWN: It's 1A, 1A is --
22 MR. MCKENNA: And it is currently under 23 revision in OGC to get better legal words. So I can't 24 even tell you --
25 CHAIRMAN BROWN: Okay.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
134 1 MR. MCKENNA: -- the exact wording right 2 now.
3 CHAIRMAN BROWN: Okay. That's, that's 4 fine. Okay. Thank you.
5 MR. MCKENNA: Yes.
6 CHAIRMAN BROWN: I just wanted to make 7 sure I understood what you were talking about.
8 MR. MCKENNA: The next clarification is on 9 Human System Interface, which we, we've already 10 discussed previously in that I think we discussed this 11 whole slide, actually. I'll just keep it up there in 12 case there's any more questions.
13 CHAIRMAN BROWN: You -- yes, I just had a 14 question on that. You, you're commenting that you 15 agree in doing this, but yet, when in Section 2C you 16 hedged it saying, it may or may not be appropriate to 17 be used.
18 MR. MCKENNA: Section 2C.
19 CHAIRMAN BROWN: Of the, of your comments.
20 MR. MCKENNA: C2?
21 CHAIRMAN BROWN: Yes. There was -- I'm 22 trying to find where the explicit --
23 MR. MCKENNA: This is the one I brought.
24 CHAIRMAN BROWN: The NRC, yes. The NRC 25 staff position, this note, there's a note. Oh, that's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
135 1 on the radiation letters.
2 MR. MCKENNA: Yes. We're, we're not 3 hedging this one at all. We're acknowledging is that 4 there's more guidance in screening for Human System 5 Interface in Appendix D than there is in the base 6 document of 96-07.
7 MEMBER BLEY: Norbert pointed, Norbert 8 pointed out earlier that that in the basis document, 9 it doesn't ignore it. It just mentions it and says 10 you need to do it.
11 MR. MCKENNA: There, there should be 12 hedging on it.
13 CHAIRMAN BROWN: I've got too many 14 documents open. If I go back and find that, I'll come 15 back and ask a question again.
16 MR. MCKENNA: Okay. Again, we're not 17 taking exception to it. We're just clarifying.
18 CHAIRMAN BROWN: It would, the way I read 19 this, it said that the Section 2C, C2B in the Reg.
20 Guide.
21 MEMBER BLEY: Do you want to read? I've 22 got it here, Charlie, if you want to read.
23 CHAIRMAN BROWN: The Human System 24 Interface et cetera, et cetera is -- in other words, 25 that digital interfaces are different, and may not be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
136 1 appropriate. That's, I took out of that section. It 2 said that, that -- are you all taking that out also?
3 Or are going to retain that information?
4 MR. CARTE: No. I think that that's a 5 slight Norbert Carte, NRC. I think the point is that 6 direct comparison between analog and digital may not 7 be appropriate on -- if you compare them on such 8 criteria as number of steps.
9 Because the steps you perform on a digital 10 display may be much easier and quicker to perform than 11 the steps you would perform using other forms of 12 controls. Go ahead.
13 You can't compare them directly on 14 abstract characteristics. And that's why we wanted a 15 human factors professionalist involved in the 16 assessment of whether it's adverse.
17 CHAIRMAN BROWN: Well in Appendix D there 18 was an example where it talked about using a touch, 19 touch screen for controlling an operation where you 20 then had to, it was like four steps.
21 Instead of turning a switch and the thing 22 started, you had to go through four steps. You had to 23 select, you had to select a screen. You had to then 24 find the, some, some other screen. Then you had to 25 find a third screen. Then you actuate it, which if NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
137 1 you've got a critical component, that's absolutely 2 stupid.
3 You ought to have a switch to turn it. If 4 you don't care how long it takes you to find the 5 information or to do it, then that's okay. So that 6 seemed to be a disconnect in my own edits. There's 7 an input --
8 To me, you've got to be very, very careful 9 with touch screens, fact is that, I just, in my 10 experience, from our plants that we first did this.
11 We only used it for recording and doing and checking 12 calibration data, or other type data in logs.
13 And any part that, any pump, any valve, 14 anything that had to be operated, you did it with a 15 switch just because we didn't want an operator taking 16 30 seconds to do something, or a minute as he was 17 scrolling through a menu, or scrolling through a list 18 of options.
19 So I, I take it you all are not 20 disagreeing necessarily with that. It's just that 21 you're saying that you have to have a separate HSI 22 evaluation of each and every circumstance? Is that 23 what I hear you saying?
24 MR. CARTE: We want to have an explicit 25 HSI professional involved in the basic determination NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
138 1 of whether it's adverse or not. Not necessarily that, 2 that the screener being an HSI, but there needs to be 3 an HSI evaluation that this is not adverse.
4 Because it, it's more complicated than 5 just the number of steps or things like that. So 6 it's, it's a, something we felt was, was not 7 appropriate for any 50.59 trained engineer to do. But 8 properly appropriate for an HFE professional to make.
9 CHAIRMAN BROWN: But does that drive you 10 into the LAR world? Or is that still within the 11 licensees ability to make that determination on his 12 own?
13 MR. CARTE: If they decide it's not 14 adverse, then evaluation and no LAR. If they decide 15 it is adverse, then they go into a full evaluation, 16 and possibly LAR.
17 CHAIRMAN BROWN: Yes. But you're going to 18 leave that at, you're going to leave that decision in 19 terms of the -- how, how components are actuated 20 explicitly, you'll leave that up to the licensee to 21 make that determination?
22 And whether it turns out you eventually 23 have to do something else as it comes up later?
24 You've made that decision in terms of your flexibility 25 assessments?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
139 1 MR. CARTE: Yes.
2 CHAIRMAN BROWN: Okay. I just wanted a 3 explicit --
4 MR. CARTE: It's in the screening section.
5 CHAIRMAN BROWN: Okay. Fine.
6 MEMBER SKILLMAN: Back on Slide 32, 7 please. Thank you. The Life Sentence, the staff 8 agrees that HSI maybe screened. Is the verb, 9 screened, in your definitions? Is there any ambiguity 10 about what that means?
11 MR. MCKENNA: So, so this slide is not the 12 language that is in the Reg. Guide.
13 MEMBER SKILLMAN: Right.
14 MR. BEAULIEU: No. The screen is, is the 15 term used in 96-07.
16 MEMBER SKILLMAN: It is?
17 MR. BEAULIEU: So there is no ambiguity 18 with respect to that term.
19 MEMBER SKILLMAN: Thank you, David.
20 Because --
21 MR. MCKENNA: I'm sorry I didn't answer 22 your question. I put your example.
23 MEMBER SKILLMAN: Thank you.
24 MEMBER BLEY: Appendix D has a whole 25 section on this.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
140 1 MR. MORTON: Yes.
2 MR. MCKENNA: Okay. To the next 3 clarification, this is discussed in Appendix D, but we 4 wanted to emphasis that the examples in Appendix D are 5 meant to illustrate guidance and not derive. You 6 can't derive guidance from the examples.
7 MEMBER BLEY: I'll tell you what, it makes 8 sense and after you explained with respect to one 9 example it makes sense. It's pretty cryptic to 10 somebody who wasn't involved in the history getting to 11 that point. You know, you might think of clarifying 12 the language a little, what you mean.
13 MR. MCKENNA: Yes, I -- those, these exact 14 words do not exist anymore, so. OGC has provided or 15 is providing some better language.
16 CHAIRMAN BROWN: I hope that helps. Is 17 that, in the Reg. Guide, is that C? C.G., C.G.?
18 MR. MCKENNA: Now --
19 CHAIRMAN BROWN: You talk about examples, 20 14 through 23.
21 MR. MCKENNA: Right. Yes.
22 CHAIRMAN BROWN: In the Reg. Guide.
23 MEMBER BLEY: I'm sorry. Oh, examples, 24 you got the numbers wrong.
25 CHAIRMAN BROWN: I went from item E, there NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
141 1 was no F, and there was a G. So I didn't know what 2 was in between. So I guess one of my questions there 3 was C deal with, or E dealt with overall perspective.
4 This is a paragraph in Appendix D.
5 And then at some point, you transition to 6 the determination on -- your position on determination 7 of safety analysis result impact. I thought that 8 might have been F, but there was no F by it. That's 9 on Page 10. Actually that's PDF Page 10. I think --
10 MEMBER BLEY: Just a short cut, I think 11 we'll be real interested in seeing how you've revised 12 this section.
13 MR. MORTON: Yes, because it's been 14 modified.
15 MEMBER BLEY: For me is was pretty cryptic, 16 those whole list of things under 2, C2.
17 MEMBER REMPE: Is there a date when you 18 expect the revised version from OCG, do you think?
19 I'm sorry --
20 MR. MCKENNA: Yes, so when we issue this 21 for public comment, at the end of April, we'll have 22 all the current OGC comments incorporated into the 23 Reg. Guide.
24 MEMBER REMPE: So typically, our meetings 25 always want 30 days advance notice, but you're going NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
142 1 to give us an information briefing in the 1st of May.
2 And so it might eliminate --
3 MR. MCKENNA: So as soon as --
4 MEMBER REMPE: -- some in the summary of 5 the discussion if we could see the revised version.
6 MR. MCKENNA: Right. Soon as we have the 7 revised version that will go out for public comment, 8 I will hand that ACRs.
9 CHAIRMAN BROWN: Yes, that's in our first 10 meeting in May?
11 MR. MCKENNA: May 2nd.
12 CHAIRMAN BROWN: For the full committee is 13 May 2nd? Yes. Okay. If you could --
14 MEMBER BLEY: That's only a couple of 15 weeks of away. We're not likely to see this before.
16 MEMBER REMPE: Well I think we should.
17 CHAIRMAN BROWN: If you've issued it, then 18 we should be able to see it.
19 MEMBER BLEY: Well, right, or they can 20 bring it with them.
21 CHAIRMAN BROWN: No. We're not going to 22 write a letter. But I mean, they, they could at least 23 use that in their briefing as opposed to this version.
24 MEMBER REMPE: And if we had time to read 25 it, it might eliminate a lot of questions if the text NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
143 1 is changed.
2 CHAIRMAN BROWN: But we'll have a sub --
3 after public comment, we'll have another meeting --
4 MEMBER REMPE: Right.
5 CHAIRMAN BROWN: -- to go through what all 6 the resolutions are. I presume that would nice. And 7 then we'll end writing a letter. Is that shaking your 8 head up and down? Or not shaking your head up and 9 down?
10 MR. MCKENNA: Keeping my head straight.
11 MEMBER BLEY: Maybe they'll go away.
12 CHAIRMAN BROWN: Oh, we won't go away.
13 MR. MCKENNA: Okay. The next section, so 14 we're just clarifying again that the RIS Supplement 1 15 is the technical basis for digital modifications, and 16 it's a clarification, not an exception. Okay.
17 Then we get into the part where we've been 18 discussing on Section 4.3.6. And again, we've 19 discussed this previously in that Appendix D is 20 written such that the determination of the impact is 21 done against the safety analysis.
22 Whereas our position is the result of any 23 malfunction previously evaluated in the SR must be 24 compared.
25 MEMBER RICCARDELLA: So the operative NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
144 1 words are what's in parenthesis there? I just want 2 to --
3 MR. MCKENNA: Yes.
4 CHAIRMAN BROWN: So that is yet to be 5 resolved due to your public comment, again?
6 MR. MCKENNA: Right, right. It has not 7 received public comments.
8 CHAIRMAN BROWN: Yes. Okay. So --
9 MR. MCKENNA: We've resolved it as far --
10 CHAIRMAN BROWN: You're concerned.
11 MR. MCKENNA: Yes. That's correct.
12 CHAIRMAN BROWN: So unless somebody really 13 comes up with something good, it's irrefutable? This 14 is your alls position, right now, relative to that 15 disagreement?
16 MR. MCKENNA: That's correct.
17 CHAIRMAN BROWN: With NEI? Okay.
18 MR. MCKENNA: And then the last, I believe 19 it's the last one. Yes. Examples -- the examples in 20 Section 4.3.6, so they all carry through the guidance 21 written in 4.3.6 in Appendix D. So all those examples 22 used that -- use that guidance.
23 CHAIRMAN BROWN: They used the RIS 24 guidance --
25 MR. MCKENNA: No. They used the Appendix NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
145 1 D guidance. So the this Appendix D where you're 2 evaluating against the safety analysis and not using 3 that malfunction of the SSC.
4 CHAIRMAN BROWN: Yes. So that would have 5 -- if your logic wins, or is retained, then those 6 would have to be rephrased, or reworked?
7 MR. MCKENNA: Well, Appendix C, is what it 8 is. We're going to issue the Reg. Guide and it would 9 say words to this effect, that you can't follow the 10 guidance --
11 CHAIRMAN BROWN: All right. So don't have 12 to revise it.
13 MR. MCKENNA: No.
14 CHAIRMAN BROWN: It would just your alls 15 -- so they would just have to -- whoever's doing the 16 analysis or the evaluations would have to do it with 17 respect to your position --
18 MR. MCKENNA: That's right.
19 CHAIRMAN BROWN: -- relative to what's in 20 Appendix C. Okay. I got that. Okay. Thank you.
21 MR. MCKENNA: Okay. So we've stopped 22 there on discussing the Reg. Guide and I'm going to 23 move in to discussing the RIS, which actually may go 24 fairly quickly.
25 MR. MORTON: Because we've covered a lot NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
146 1 of it.
2 MR. MCKENNA: Because we've covered a lot 3 of it already.
4 MR. MORTON: Yes.
5 MR. MCKENNA: So again, the qualitative 6 assessment of the RIS was originally discussed in 7 NEI 101-01. but there was limited discussion we had to 8 accomplish a qualitative assessment.
9 The RIS Supplement 1 is very good guidance 10 on his to do a qualitative assessment, which I'll get 11 into in the next slides. In order to support a 12 conclusion, that you have a low likelihood of failure 13 and you can now evaluate those sections, those 14 criterion of 50.59 in the evaluation section.
15 So the first input into the qualitative 16 assessment is to design attributes. What is built 17 into the modification, as far as, false detection, 18 failure management schemes, internal redundancy, 19 diagnostics. And you can have external items in the 20 modification, as far as mechanical stops or limiters.
21 MR. MORTON: Anybody want some?
22 MR. MCKENNA: Sure.
23 MR. MORTON: Actually, no.
24 MR. MORTON: So just as another nuance, 25 within the details of the qualitative assessment, when NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
147 1 we looked at the three qualitative factors, design 2 attributes, the quality of the design process, and the 3 operating history.
4 This particular slide is where we 5 leveraged the post credit for our expectations for 6 qualitative assessment, DI specific. Deterministic 7 features within the design, within the digitalized 8 architecture to address the specific hazard that you 9 have identified within the particular mod.
10 And we still began a few examples of some 11 testing features, internal redundancies, even 12 diversity if you want add that, within the channel 13 itself. What specific ways are you addressing direct 14 hazards with in the systems.
15 How are you addressing them? What factors 16 and to do that? So the documentation, as well. So --
17 we leveraged the most credit. Here, within the three.
18 So I just wanted to comment, make sure we're clear on 19 that particular piece.
20 MR. MCKENNA: This slide is just lists 21 some more design attributes that were not included in 22 the previous slide.
23 CHAIRMAN BROWN: Can you go backwards?
24 Typically, we obviously look at Watchdog timers and 25 unit directional, direct communications, relative to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
148 1 protection and safe guide systems. That's what we've 2 had most discussions on.
3 The universal, unidirectional 4 communications can apply from a control of access 5 standpoint because the network is used to consolidate 6 all information going from into a plant before it has 7 been sent from made successful to a main controlled 8 room, or other technical support facility, or 9 something like that.
10 So that becomes a critical point if you're 11 going to external, if you're going to access or 12 external to the overall plant, to the internet, in 13 other words.
14 I presume that's, so that's it's not just 15 applied to protection systems and safe guard systems.
16 That's a more universal -- Watch Dog Timers, are 17 fundamentally saying, hey did the processors stop or 18 not? And does it matter?
19 In safeguards, it does. In the protection 20 systems it does. But for, for other parts it may not.
21 Although it might be useful for other reasons because 22 if you just have it reset, you don't even have to shut 23 down, it resets them.
24 I don't know, the processor, the, the 25 platforms that they've been using, some of them take NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
149 1 five or six minutes to reboot, which really kind of 2 nasty.
3 MR. MORTON: Yes. This is not an all-4 inclusive list of futures --
5 CHAIRMAN BROWN: Yes. I understand, I 6 just want to --
7 MR. MORTON: And each one of these gives 8 you --
9 CHAIRMAN BROWN: -- it's just to --
10 MR. MORTON: -- a different benefit.
11 CHAIRMAN BROWN: -- it's quite a shopping 12 list is the point.
13 MR. MORTON: Yes. We've leverage -- this 14 list is based upon a lot of staff's own engineering 15 judgment and knowledge. And also, with the lifesaving 16 reviews you've done both in advanced reactors and 17 operating reactors, and the different design features 18 we've seen licensees and applicants apply to adjust to 19 some hazards.
20 One in particular, segmentation, which 21 keeps you a lot of bang for your buck in terms of 22 giving you a level of diversity in terms of signal and 23 process diversity, things of that nature. Mentioning 24 redundant, ring that rec, sir, connect with switches, 25 traffic control to try to prevent things such as NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
150 1 thunderstorms --
2 CHAIRMAN BROWN: I got it.
3 MR. MCKENNA: So we tried to give a 4 plethora of options so that people got the gist of it.
5 You don't have to just depend all the timers freezing.
6 CHAIRMAN BROWN: Okay. Thanks.
7 MR. MCKENNA: The next step in qualitative 8 assessment is the quality of the design process. How 9 the software was developed? The system designed? The 10 validation and the testing processes. And for safety 11 related modifications, the, the documentation is 12 available for referencing. And obviously for 13 commercial grade, the amputation may be not extensive 14 as a safety designed modification.
15 MEMBER SKILLMAN: Would a, would a 16 proposed modification to a nuclear safety grade system 17 necessarily come from a program that is under 18 regulated QA program, Appendix B to 10 CFAR 5, 50.
19 MR. MCKENNA: Well you answer that.
20 MR. MORTON: You're asking if, if you're 21 doing a modification with safety grade system, would 22 you be?
23 MEMBER SKILLMAN: Yes, yes.
24 MR. MORTON: It should have if it's 25 meetings it's requirements under Appendix B program.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
151 1 MEMBER SKILLMAN: Now if you --
2 MR. MORTON: Well if, if it's rate 3 degraded. That's a different aspect to it, as well.
4 But it really depends on the SSC you're modifying.
5 We put that last bullet in there, for 6 examples, because of a lot feedback that we received 7 from industry saying, hey, NRC, if you're doing 8 modifications to these non-safety related systems, 9 they're not Appendix B systems.
10 They don't necessarily have the quality 11 attached to them that safety related systems do. We 12 recognize that fact. So, as part They qualitative 13 assessment, we tried to bridge the gap between some 14 levels of demonstration of quality, building, 15 construction.
16 But necessarily having to be something 17 that a part of Appendix B, Reg. Guide endorsed 18 standards. So we said, you know, in the street 19 consensus standards, something to that extent for 20 non-safety related.
21 And for safety related, you are under 22 auspices of your Appendix B, quality assurance program 23 per the documentation, you have to meet, provider 24 standards that you have provide, as part of the 25 design.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
152 1 CHAIRMAN BROWN: Okay. Thank you. Go 2 ahead, I was just bemusing with Peter.
3 MR. MCKENNA: So that finishes the 4 discussion of the RIS. We're now going to go into 5 what started and planned modifications are started and 6 planned using the RIS in the industry. This is a list 7 provided NEI.
8 Just to give some examples of what is out 9 there. I think I have two slides on this. These are 10 the final two slides. So currently there is three 11 safety related digital mods, start in 2018. You can 12 read the list. These are generator controls, 13 breakers, circuit controls.
14 Same type of modifications. Plan to start 15 in 2019 with various completion times. I'll keep that 16 list up there without reading everything. And I can 17 go back and forth here. And then the last, you can 18 see, the types of modifications also.
19 CHAIRMAN BROWN: So are these planned 20 safety related mods? Are those LARs or are those ones 21 being made the 50.59 and don't require LARs?
22 MR. MCKENNA: So obviously, the industry 23 plans to use the 50.59 process on these, but we don't 24 know the outcome.
25 CHAIRMAN BROWN: Okay. So that's, that's, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
153 1 this is the plan in other words. Yes. Then we go --
2 so they will go through the process and you'll have a 3 determination made.
4 MR. MCKENNA: That is correct.
5 CHAIRMAN BROWN: But they've identified 6 what they're looking at?
7 MR. MCKENNA: Right. But there are most 8 likely ones that they're going to do a qualitative 9 assessment on. And that qualitative assessment could 10 see the final outcome of that would be low likelihood 11 of failure.
12 CHAIRMAN BROWN: On an assessment? On a 13 screening basis or --
14 MR. MCKENNA: Qualitative assessments done 15 on the evaluation.
16 CHAIRMAN BROWN: Evaluation. Okay. Okay.
17 Next slide? Backup slide.
18 MR. MCKENNA: So we're done. Our briefing 19 material.
20 CHAIRMAN BROWN: All right. We've got NEI 21 still to come. I don't think we will finish that this 22 in, there might be some discussion there. My 23 suggestion is that we break for lunch, and then have 24 NEI come in after lunch. And so we'll come back.
25 Well before we leave, before we recess, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
154 1 before we recess, are there any, anybody else have any 2 questions here around the table? I hear none. So we 3 will recess until 1:00 p.m. when we will restart. And 4 then NEI will come up to the table. Okay. Thank you.
5 (Whereupon, the above-entitled matter went 6 off the record at 11:44 a.m. and resumed at 1:02 p.m.)
7 CHAIRMAN BROWN: Okay. The meeting will 8 come back to order, and we'll proceed with NEI and 9 associates, LeBlond & Associates to give us our NEI 10 industry perspective on the Appendix D. So, who's 11 going to open?
12 MR. GEIER: Yes, I'll open. I'm Steve 13 Grier. I'm with NEI and my current position is Senior 14 Director of Engineering and Risk. And I'll give just 15 a real brief, a little bit of background on myself.
16 I've been with NEI about three and a half years.
17 Before that I worked 30 years at two different nuclear 18 stations, primarily in design engineering positions.
19 I did want to just provide just a few 20 opening remarks before I pass it off to my much more 21 qualified colleagues to talk on this issue. And 22 basically, I did want to say -- I do want to express 23 appreciation to ACRS to having industry come in and 24 talk and provide some of our prospectives.
25 You know, we are -- did want to say, you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
155 1 know, we're very much aligned on being able to move 2 forward with upgrades for digital. And primarily to 3 do some needed replacements of obsolete and some 4 challenges with reliability at our stations. We 5 strongly feel that digital upgrades can provide 6 improvements to plant safety and also, to the station 7 and system reliability.
8 And, you know, as was talked about during 9 the NRC portion there is an appetite for stations to 10 move forward and begin moving forward with their 11 digital upgrades. We got, we know there are several 12 fleets -- and Neil will talk a little bit about what 13 Duke Energy is doing -- and several other fleets that 14 are moving forward with smaller mods with but are also 15 looking for some of the major mods including SFAS and 16 RPS.
17 The good news is the RIS 2002-22, 18 Supplement 1 and the ISG 6 are really important 19 documents for the station. The risk is really 20 spurring some sub-lumen, some of the smaller less 21 safety significant mods. And then ISG 6, I think once 22 we get a little bit experience and people do some 23 planning we'll help some stations move forward with 24 some of the major again, our SFAS and RPS.
25 The Appendix D is really, we look at this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
156 1 as the third leg, regulatory leg, out of the -- that's 2 coming out the independent or integrated action plan.
3 It's really a critical piece to provide the guidance 4 to the industry to give them the confidence to move 5 forward with 50.59 for digital.
6 You know, the staff talked about is the 7 Appendix D has been more than three years in the 8 making. And we've had over 30 public -- around 30 9 public interactions to try to get a line around it.
10 I think the staff and the industry is largely aligned 11 on the guidance.
12 MEMBER BLEY: Steve, can I interrupt you?
13 MR. GEIER: Sure.
14 MEMBER BLEY: Because I didn't see it in 15 the rest of your slides. We heard this morning that 16 you did Appendix D but you had a parallel effort to 17 extract the technical details and you were going to 18 publish that later. Now are you guys reasonably 19 content with the RIS? Or are you planning to get that 20 other document out soon?
21 MR. GEIER: Right. So the -- back in the 22 2017 range, we were putting out a document that was 23 called NEI 16-16.
24 MEMBER BLEY: Okay.
25 MR. GEIER: And it was primarily to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
157 1 address the software, CCF.
2 MEMBER BLEY: Okay.
3 MR. GEIER: And so, it was a lot of time 4 and effort, a lot of public interaction on that. And 5 when we decided to move forward with the RIS, we put 6 that on hold. So now that the RIS is out and we've 7 got the ISG, we're reassessing what we need to do.
8 And there's some changes. We can talk a little bit 9 more specifics but, so there's changes in the guidance 10 stack that the industry is looking to use.
11 And primarily these are coming out of 12 EPRI. EPRI is developing several new products to help 13 with, you know, what we've talked about is a real 14 solid quality design process. So they have a design 15 engineering guide that's been issued last year.
16 They're also coming up with some new documents related 17 to addressing hazards associated with digital mods 18 that will include CCF but also will address other 19 potential hazards such as EMI, RFI, cybersecurity, 20 human factors. So that's kind of a new population of 21 information.
22 So we're going to reassess that NEI 16-16, 23 which we are, and we're going to develop a different 24 document that's going to leverage the EPRI products 25 that are coming out.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
158 1 MEMBER BLEY: Okay. So you'll make use of 2 those?
3 MR. GEIER: Yes. And we're looking at mid 4 to late summer to having the draft of that out.
5 MEMBER BLEY: Of this year?
6 MR. GEIER: Of this year.
7 MEMBER BLEY: Okay. EFPI has that second 8 document done?
9 MR. GEIER: Yes. So they've issued their 10 design engineering guide.
11 MEMBER BLEY: Right.
12 MR. GEIER: They've issued what they call 13 a HAZCAD document which is their Hazardous Analysis.
14 And their final document, that's going to come out, is 15 specifically to address CCF, it's called a DRAM.
16 MR. ARCHAMBO: And it's a --
17 MR. GEIER: Neil, can probably talk to the 18 actual acronyms of that.
19 MR. ARCHAMBO: Yes, it's Digital 20 Reliability Analysis Methodology, that's what DRAM 21 stands for --
22 MEMBER BLEY: Okay.
23 MR. ARCHAMBO: -- if you hear that term.
24 MEMBER BLEY: Okay.
25 MR. GEIER: Back in 2017, EPRI had NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
159 1 published a document that we called the CCF Guide. It 2 was --
3 MEMBER BLEY: I sort of --
4 MR. GEIER: -- Dependability and 5 Reliability of Digital Systems, a long name.
6 CHAIRMAN BROWN: I think we looked at that 7 anyway --
8 PARTICIPANT: Which one?
9 CHAIRMAN BROWN: The operating.
10 PARTICIPANT: I don't remember that.
11 MR. GEIER: And that actually formed the 12 basis for this NEI 16-16. We abstracted some of the 13 detail out of that. They're revising that with this 14 DRAM document. And so, we need to find the best way 15 to leverage that information.
16 MEMBER BLEY: Thank you.
17 MR. GEIER: Okay, next. So what I wanted 18 just talk briefly about Appendix D before I turned it 19 over to, Neil and Peter, is what we're looking for is 20 a clean endorsement. And quite honestly, anything 21 other than a clean endorsement of this would really 22 cause confusion and cause it to be problematic for the 23 industry.
24 MEMBER BLEY: I'm sorry to interrupt, but 25 endorsement of Appendix D -- are you saying without NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
160 1 the RIS or with the RIS?
2 MR. GEIER: Well, the RIS is already --
3 MEMBER BLEY: That makes it not quite 4 clean because it's adding the RIS to it.
5 MR. GEIER: So the Appendix D provides a 6 detailed guidance --
7 MEMBER BLEY: Right.
8 MR. GEIER: -- to licensee staff for 9 how to do 50.59s while taking -- while leveraging the 10 RIS and the assessment that's allowed by the RIS.
11 MEMBER BLEY: Okay.
12 MR. GEIER: So it's supplemental. It's --
13 they actually took these --
14 MEMBER BLEY: So you would see that as a 15 clean endorsement --
16 MR. GEIER: Yes, sir.
17 MEMBER BLEY: -- Appendix D with the RIS?
18 MR. GEIER: That's right.
19 MEMBER BLEY: That's what I was hoping, 20 you meant. Okay.
21 MR. GEIER: With the RIS, exactly. Yeah, 22 not standalone that's for sure.
23 CHAIRMAN BROWN: But without this other 24 controversy?
25 MR. GEIER: Without the other controversy.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
161 1 And that's -- and we put this together. That's really 2 what our slides are going to be about, is --
3 CHAIRMAN BROWN: To kind of gather --
4 MR. GEIER: -- kind of our perspective on 5 the controversy, where we feel we need to go. And so, 6 you know, I think -- the only thing I really wanted to 7 say before I moved on is that I think that if we don't 8 endorse it with the Section 4.3.6 the way it's written 9 really we'll comprise the benefit of the document.
10 And a lot of the feedback we're getting is 11 that a lot of station's then will not move forward 12 with their digital mods because it'll take a lot of 13 the benefit out of the 50.59 process. So I'm going to 14 turn it over next to Neil to talk about -- oh, go 15 ahead.
16 CHAIRMAN BROWN: You just -- I got a 17 little confused you said moving forward we'll take 18 something -- what out of the design -- if what, takes 19 something out of the design process?
20 MR. GEIER: If --
21 CHAIRMAN BROWN: It won't move forward?
22 MR. GEIER: Right. If we don't move 23 forward with the document, as written, meaning the 24 Section 4.3.6, the way it's written, and the exception 25 is taken, then that's going to take a lot of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
162 1 benefit. And a lot of the mods that could be done --
2 CHAIRMAN BROWN: On Appendix D?
3 MR. GEIER: -- under 50.59 will now not be 4 able to move forward that way. It'll likely require 5 a LAR. And the feedback we have, and I know from my 6 personal experience working at stations, is that if a 7 modification requires a LAR, that removes a lot of 8 that from approval from the station's approval 9 process.
10 CHAIRMAN BROWN: So you're saying if the 11 issue is not resolved with the words you want about 12 the safety analysis versus the --
13 MR. GEIER: Component.
14 CHAIRMAN BROWN: -- component thrust, it 15 will decimate the ability to make back fits?
16 MR. GEIER: That's correct.
17 CHAIRMAN BROWN: Using 50.59?
18 MEMBER SKILLMAN: I'd like to understand 19 that because I've been in the position that you have 20 been in for years, both as director of Design 21 Engineering and multiple tours of Planned Engineering.
22 And in the Design Engineering role, under 50.59, there 23 was no question that the assessment was at the 24 component level.
25 And when I think of the modifications for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
163 1 control systems, while it was complex, it was doable 2 to do the SSC evaluation and also the overall system 3 behavior evaluation. They're not mutually exclusive.
4 So I would like to understand the basis of what you're 5 saying.
6 MR. GEIER: And we're going to get to 7 that. And I'm --
8 MEMBER SKILLMAN: I hope so.
9 MR. GEIER: Yes.
10 MEMBER SKILLMAN: Because it kind of 11 sounds like --
12 MR. GEIER: That's exactly --
13 MEMBER SKILLMAN: -- blackmail --
14 MR. GEIER: That's really why we're here 15 just to talk about that.
16 MEMBER SKILLMAN: It sounds like 17 blackmail. I know it's not, but that's what it sounds 18 like. But I know, for a fact, that you can do both.
19 It takes more effort. But you end up with a more 20 thorough assessment of both, what the SSC behavior is 21 on what would be the outcome of the, if you will, of 22 the system behavior and the, if you will, the 23 applicability of the accredited devices for the 24 license.
25 MR. GEIER: Right.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
164 1 MR. ARCHAMBO: Yes. I think it'd become 2 clear as we go through this where the issues are.
3 MEMBER SKILLMAN: Please.
4 MR. ARCHAMBO: Okay.
5 MR. GEIER: Okay.
6 MR. ARCHAMBO: If you're ready -- okay.
7 MR. GEIER: So next I'll turn it over to, 8 Neil from Duke.
9 MR. ARCHAMBO: Neil Archambo with Duke 10 Energy. I appreciate the opportunity to be here.
11 I've been in the industry about 32 years, over 32 12 years, hard to believe -- hard to believe for me. I'm 13 a design engineer, spent most of my time as a design 14 engineer, and I write 50.59s. In fact, I just 15 finished writing one a few weeks ago. So this stuff 16 is near and dear to my heart.
17 I understand the issues and on top of 18 that, I get the opportunity to the review every single 19 50.59 screen and evaluation at Duke that's done on a 20 digital modification.
21 We have six plants, 11 units so, that's a 22 lot of readings. So, I get to see a lot of the things 23 that have caused us issues when trying to apply 50.59 24 and digital changes. And once in a while, I get one 25 from the industry to look at, so I have a pretty good NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
165 1 idea where the issues are at.
2 So I'm going to run through just a few of 3 them quickly and then we'll get on to the meat of this 4 discussion. And explain --
5 MR. GEIER: Can I go to my next slide 6 then?
7 MR. ARCHAMBO: and -- yes, just go on to 8 the next slide, please.
9 MR. ARCHAMBO: And explain what Appendix 10 D does for us. How it resolves that issue now. We've 11 had problems. We've had difficulties as licensees 12 identifying FSARs described as iron functions as they 13 apply to digital modifications. And, you know, why is 14 that? Why is that so much different? Well, they're 15 a lot more complicated.
16 We put in digital modifications now.
17 We're combining functions. We're maybe networking 18 things, putting them on a platform. It's not always 19 easy to understand exactly what you're affecting.
20 That's been an issue in the industry. And Appendix D 21 helps us walk us through that -- how to identify those 22 design functions that were otherwise -- little bit 23 difficult to understand.
24 It helps us to determine if a change is 25 adverse or not adverse in the screening process. Now NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
166 1 we have people. We have folks out there that are 2 improperly screening things out when they should 3 screen them in. And we have people screening things 4 in when they probably should've been screened out or 5 could have been screened out.
6 So there's licensees out there that just 7 thrown their hands in the air and say, I don't know.
8 I don't know if I should screen or I'm just screening 9 everything in. From the smallest change, I'll screen 10 it in, when clearly it could have been screened out, 11 they're just screening everything in.
12 So Appendix D helps us walk through that.
13 It gives us the guidance that we think we need to 14 determine, in a reliable fashion, whether something 15 screens in or screens out in a digital modification.
16 Next one is how do we address CCF in a 17 50.59 evaluation. How do we address it? You got 18 Questions 1, 2, 5 and 6 that we've talked about, that 19 really are the critical ones, the digital 20 modifications. Where do you address CCF? You know, 21 it's not necessarily in Criterion 1 or Criterion 2, 22 that's about reliability. It's more 5 and 6 when 23 we're talking about actions to different type, 24 malfunction with a different result. But folks are 25 confused I want to address Common Cause Failure.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
167 1 MR. RICCARDELLA: Excuse me. What's that 2 acronym mean? Is it ECF?
3 MR. ARCHAMBO: I'm sorry, CCF, Common 4 Cause Failure.
6 MR. ARCHAMBO: And predominantly software 7 Common Cause Failure, wherein the process do we 8 identify and address software Common Cause Failure or, 9 any digital Common Cause Failure. And we believe 10 Appendix D addresses that issue. Gives us an idea 11 where we're supposed to address those within the 50.59 12 process. Next slide, please.
13 One thing I want to, maybe before we go to 14 the next slide, is it's hopefully, abundantly clear in 15 Appendix D. There's a couple boxes with caution 16 statements that says, this is supplemental guidance.
17 You've got to use NEI 96-07 Revision 1 in conjunction 18 with Appendix D. We don't want licensees stripping 19 out just Appendix D and walking away with it when they 20 do their digital modifications.
21 So we try to make that abundantly clear 22 that you still have to use the main body, the main 23 guidance in order to successfully go navigate through 24 the process.
25 MEMBER BLEY: And that was real clear. In NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
168 1 fact, sometimes, I'm -- why is he saying that, that 2 seems obvious.
3 MR. ARCHAMBO: That's why. There's a lot 4 of information in the mother document that we don't 5 want to lose. You know, we just didn't want to 6 re-write everything out of the main document make it 7 an unmanageable appendix. So next slide, please.
8 So I just talked about addressing Criteria 9 1, 2, 4 and 6. Now 1 -- Question 1, we're talking 10 about accidents, right, accident frequency. In 11 Question 2, it's likelihood of malfunctions. We 12 talked about frequency and likelihood a little bit.
13 That's a reliability issue. Questions 5 and 6, 14 accidents of different type, malfunction with a 15 different result. People have difficulty in digital 16 modifications addressing those particular questions, 17 those criteria.
18 So we believe Appendix D provides that 19 guidance that they need for digital-based activities 20 with examples to how you apply the guidance.
21 MEMBER SKILLMAN: Neil is the difficulty 22 they're not understanding how the plant behaves based 23 on the body of documentation that they have?
24 MR. ARCHAMBO: Some of it is, yes. And 25 we'll talk about that in just a few seconds. What I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
169 1 see, when I review a lot of these documents, as people 2 be addressing Common Cause Failure in Criterion 1.
3 Criterion 2, well, that's really not the place to 4 address it. That's where you address the reliability.
5 Have I increased the likelihood of a 6 malfunction, well, that's a reliability issue. It's 7 nothing to do with Common Cause Failure, until you get 8 to Criterion 5 and 6. That's where Common Cause 9 Failure comes in. So we see that over and over that 10 people are addressing the wrong things in the wrong 11 places.
12 MEMBER SKILLMAN: Well, let me ask my 13 question a different way. Is this because the station 14 does not have an accurate design basis document? Or 15 is it because the station doesn't have an accurate 16 final safety evaluation report?
17 MR. ARCHAMBO: Well, I wouldn't say 18 they're not accurate. I mean they've all, of course, 19 been licensed to those particular documents.
20 MEMBER SKILLMAN: Are they not thorough?
21 MR. ARCHAMBO: That comes to that 22 Criterion 6 and the question is -- or the comment is 23 we have varying degrees of detail in FSARs. You know 24 we have six plants, 11 units within our fleet. We 25 have some older ones that have very little detail, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
170 1 very little descriptive material and our newer plants 2 have all kinds of detail and descriptive material.
3 MEMBER SKILLMAN: It could be twice as 4 much?
5 MR. ARCHAMBO: Yeah, it could be twice as 6 much. It could be three times as much. So what we 7 run into is applying -- this is where Criterion 6 8 really comes in. This is the meat of Criterion 6, is 9 you're going to have uneven application of 50.59 on 10 plants that have older FSARs versus plants that have 11 newer FSARs because there's going to be a lot more 12 descriptive materials in the newer FSARs.
13 So under the guidance, the way it's 14 interpreted by the staff, I could probably do a 15 modification under 50.59 in an older plant but I might 16 have to get a license amendment request for that exact 17 same modification in one of my newer plants simply 18 because it has more descriptive material.
19 And that problem was actually solved about 20 20 years ago and, Pete's going to talk about that. So 21 we're kind of drudging up a problem that's already 22 been solved. And that's what we hope to walk you 23 through as we go through this particular section of 24 the presentation. Does that address your question 25 adequately?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
171 1 MEMBER SKILLMAN: I'm thinking now you've 2 got McGuire and Catawba on one end and you've got 3 Oconee on the other.
4 MR. ARCHAMBO: Robinson --
5 MEMBER SKILLMAN: So you have some moldy, 6 moldy, oldies there.
7 MR. ARCHAMBO: Sure.
8 MEMBER SKILLMAN: Where the information's 9 very sparse.
10 MR. ARCHAMBO: That's right.
11 MEMBER SKILLMAN: So I certainly 12 appreciate that.
13 MR. ARCHAMBO: Yes. That causes issues.
14 Certainly causes issues. The next item down there is 15 recognizing, you know, the impact on our plant license 16 basis when we combine functions. You know, a lot of 17 our plants were built and large stuff is separate.
18 Maybe the only reason it was separate was because we 19 didn't have the technology to put them together. Now 20 we can. And in some cases, we are. We're putting in 21 distributors control systems platforms. And we're 22 migrating systems onto those platforms where we're 23 combining functions.
24 CHAIRMAN BROWN: What kind platform?
25 MR. ARCHAMBO: Distributors control NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
172 1 systems platforms.
2 CHAIRMAN BROWN: You say you're migrating 3 from -- I missed that sentence, sir.
4 MR. ARCHAMBO: Okay.
5 CHAIRMAN BROWN: You had a run-on a set of 6 words there for me.
7 MR. ARCHAMBO: In a number of plants, and 8 we're included in that, we're putting in platforms, 9 distributed control system platforms and we're 10 migrating to balance the plant systems on to those 11 platforms. Now when you do that, you know, you're --
12 CHAIRMAN BROWN: There's a lot of common 13 software in that?
14 MR. ARCHAMBO: Yes. You're combining 15 functions and it's -- so recognizing your impact on 16 the licensing basis, when you do, that's key. That's 17 the key issue. 96-07, the base document, doesn't 18 necessarily address that. So we're trying to bring 19 that out in Appendix D. Here's what you have to look 20 for when you're combining functions in such a manner.
21 And we hope we did that with -- through the guidance 22 and examples.
23 And the last one is the main one, and we 24 really want to spend the balance of our time talking 25 about. That's Criterion 6, which you've heard a lot NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
173 1 about today. Where do we address malfunction results?
2 Is it at the system level? Or is it at the plant 3 level?
4 Now we talked briefly about the FSARs.
5 That's one of the key components. A lot of the older 6 plants have very little descriptive material, very 7 little descriptive material. So applying 50.59 to 8 some of those plants is going to give you a different 9 result than if you apply it to one of the older 10 plants.
11 So for digital activity, I just want to 12 make it clear, that's the critical -- that has been 13 the most difficult criterion to address, is Question 14 6 because that's where Common Cause Failure, software 15 Common Cause Failure comes into play.
16 MEMBER BLEY: Can I pin you down a little 17 because I think I'm understanding, but I want to make 18 sure I'm right.
19 MR. ARCHAMBO: Okay.
20 MEMBER BLEY: And this is the difference 21 between applying a defined process and I'm kind of 22 thinking more generally about what one would you like.
23 From the more general side, I would think if we have 24 less detail, the older FSARs, then there ought to be 25 more question of does this new thing do something NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
174 1 different.
2 But since it's written in terms of does 3 this change something that was in the FSAR, if the 4 FSAR didn't even talk about it, then it doesn't change 5 it. But if it's something that wasn't considered then 6 how could it be important? It seems like you oughtn't 7 get out of it so easily.
8 On the other hand, if you've got a lot of 9 detail in the FSAR and this changes something that's 10 not too significant, then you're paying a price for 11 having more information there. And I think that's 12 what you're saying.
13 MR. ARCHAMBO: Right, yes.
14 MR. LEBLOND: Well, you can see its a 15 problem, right? You can see it's an issue.
16 MEMBER BLEY: I can smell it's a problem.
17 I haven't actually tried to do this. But I can see 18 it's, you know, I've done lots of other kind of 19 analyses on old plants, new plants. And, yes, most of 20 the time we'd have more trouble with the old plant 21 because we had to dredge up the information that 22 wasn't easy to find, sometimes hand over hand looking 23 for it.
24 But this is a different problem. This is 25 -- I'm working from this document and now do I have a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
175 1 change or no change. And if there was nothing there 2 -- the way it's being interpreted, I think, on both 3 sides is you don't need to dig on it. If there is 4 information there then the question is how much of a 5 change is really significant enough to push this 6 outside of the 50.59 process. Am I saying that close 7 to right?
8 MR. ARCHAMBO: Yes. There is a method and 9 again, this was actually addressed 20 years ago. Pete 10 was in the midst of it. They thought about this.
11 They considered this, the uneven application of 50.59.
12 Because of different details in your FSARs, you know, 13 it was resolved. And so there is a method to take 14 care of that to level the playing field, right.
15 MR. LEBLOND: Right, and see --
16 MR. ARCHAMBO: So everybody has to address 17 the same things no matter how old your FSARs is, no 18 matter how detailed it is or, how not detail it is.
19 What they solved 20 years ago, we're going to talk 20 about and explain -- Pete will -- is how we level the 21 playing field on that.
22 MEMBER BLEY: So the claim is --
23 MR. ARCHAMBO: And what we don't want --
24 MEMBER BLEY: -- it back then is 25 incorporated in Appendix D?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
176 1 MR. ARCHAMBO: Yes. And what we don't 2 want to do is undo things that were done 20 years ago 3 because that's the way it's being taught today.
4 MR. LEBLOND: Right.
5 MR. ARCHAMBO: And that's the way people 6 have been doing 50.59 so we got to consider that, you 7 know. So you know, stay tuned. Pete will certainly 8 educate you on that.
9 MEMBER BLEY: I'll be quiet for that.
10 MR. ARCHAMBO: No, it's quite all right.
11 MR. LEBLOND: Well with that I'm just 12 going to leave. I'm done. I just figure I got nothing 13 to talk about.
14 MR. ARCHAMBO: Well, so, having said that 15 I'm going to turn it over to Pete. I'm done unless 16 you have any questions specifically for me.
17 MR. LEBLOND: Next slide, please, thank 18 you. Hi, I'm Pete LeBlond. And I'm a member of the 19 team. And before I go any further it would be a sin 20 if we didn't mention Kati Austgen. I'm a very poor 21 fill in for Kati Austgen. She's really our leader.
22 And if she wasn't on vacation with her kids in spring 23 break -- who'd believe that -- she'd be here. So 24 she's been our leader and she's really the person that 25 would be doing this most likely I suspect.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
177 1 So, having said that, my name is, Pete 2 LeBlond. I've been in the business about now 42, 43 3 years that sounds like an awful long time. I spent 22 4 years with Exelon. I was at Zion for many years. And 5 then I quit 20 years ago. I've been on my own and 6 doing this operability 50.59 design basis control 7 stuff.
8 And I can't believe I still have to bring it up 9 now, but I was one of a group of five or six people 10 that helped craft 96-07. And I was involved with the 11 negotiations that we're talking about and the kind of 12 issues that you bring up. And we're talking about 13 leveling the playing field. I'm going to address that 14 right now. I hope, I hope, I hope.
15 So we have four points that I want to make 16 and I'm going to have a slide for each one to expand 17 on each one. But before I do start marching my way 18 through these four points I want to highlight kind of 19 a common theme. You heard Steve ask for a clean 20 endorsement. And I'm going to say that absent a clean 21 endorsement you run the risk of undoing the fixes to 22 the problems that were 20 years old.
23 We've touched on some of them right here 24 and I'm hopefully, going to give you some more detail.
25 So what we've done in Appendix Delta is entirely NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
178 1 consistent with the rule as it was crafted 20 years 2 ago and has been implemented since then, as I've 3 taught at Fermi last week -- literally taught at Fermi 4 last week.
5 So with that, what are the four points 6 that I'm going to highlight? Number 1, we talked all 7 morning using the word malfunction as if it's in 8 Webster's. It's actually a malfunction of an SSC 9 important to safety. One of the -- we have 24 10 objectives 20 years ago. And one of them was there is 11 no definition for a malfunction of an SSC important to 12 safety.
13 So 96-07 created it. And it's definition 14 3.9 which says it's a failure to perform a design 15 function. That will then start us on a march, which 16 I will cover in the next -- in the slide that comes 17 after this.
18 But Point Number 1, there's definitions, 19 approved regulatory definitions, in the Reg Guide --
20 two Reg Guides, 1.186 for 50.59 and 1.187, or the 21 other way around, for design basis. We're following 22 them. And I hope to demonstrate that we're following 23 them in a very clear and unambiguous way.
24 Secondly, the rule making is clear. I 25 heard a request for irrefutable evidence. When we go NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
179 1 back and look at the evidence from 1999 for the 2 rulemaking record for the rule we have today, it's 3 absolutely clear. And to be even more direct, there's 4 no semblance of an opening for an alternative view.
5 It focuses on the safety analysis which is taken by 6 your comments distinguishing between the safety 7 analysis report and the safety analysis. The record 8 is clear, its the safety analysis.
9 Thirdly, and we just touched on it here 10 what, Neil described, Robinson, and I think your words 11 were, sparse. Was that not? Well, some of the sites 12 are sparse. Some sites have a factor of two or three.
13 You compare Robinson to --
14 MR. ARCHAMBO: McGuire.
15 MR. LEBLOND: -- McGuire. Maybe two, 16 three times level of detail. That's not a new 17 problem. That's a new manifestation of a problem that 18 was fixed 20 years ago. It was recognized 20 years 19 ago, and we're running the risk of undoing that fix 20 right here. And finally --
21 MEMBER SKILLMAN: Well, describe that fix, 22 please.
23 MR. LEBLOND: Well, I'm -- I'll do it now 24 if want but there's a slide coming.
25 MEMBER SKILLMAN: Okay, I'll wait.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
180 1 MR. LEBLOND: That talks about --
2 MEMBER SKILLMAN: Thank you.
3 MR. LEBLOND: -- yes. And just so like 4 what I'm telling what I'm going to tell you then I'm 5 going to tell you in the -- a next slide for each one 6 of these. And finally, you'll see that Appendix Delta 7 relies on a technique where you look at the levels of 8 change and then you cascade upwards to a higher 9 functional level. That was taken with your comments 10 earlier today about, oh, you have a new result. Next 11 question is, where is that result.
12 And that gamut has been used successfully 13 for 20 years within the other criteria -- Criterion 3, 14 and 4 for malfunctions, Criterion 7 for design and 15 basis and, in some cases, Criterion 2 for malfunction 16 of SSC. We're simply using it again under 4.3.6.
17 So with that, let's start with the 18 expansion of Point Number 1 of having -- next slide.
19 First, a malfunction is a failure to perform a design 20 function. That's a defining term. You might go to 21 Webster's and see, oh, that's the failure of a 22 component to work. That wasn't adequate.
23 There was 24 issues that we were given 20 24 years ago that say fix these problems. And one of the 25 problems was we made a definition of a malfunction of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
181 1 an SSC important to safety. So in order to understand 2 what a malfunction is, you need to know what a design 3 function is.
4 So what's a design function? Well, what's 5 a malfunction? A failure to perform a design 6 function. A design function has four constituent 7 parts. It either performs a design basis function, 8 which I'll demonstrate as a very high level function.
9 It either supports, a design basis function, impacts 10 a design basis function or is a transient action 11 initiator.
12 I was taken with some of your comments 13 about well, how do you partition these things? That's 14 a term that's used. It's a defined term. It has four 15 constituent parts. These were negotiated between the 16 team and the NRC during this time period.
17 MEMBER BLEY: My sense is there's no 18 disagreement between the staff and you, as 19 representatives of the industry, on this definition of 20 a malfunction. Is that correct?
21 MR. LEBLOND: Well, they didn't mention 22 it. They didn't malfunction of an SSC. They never 23 mentioned definition 3.9, the entire morning. So I 24 don't know I can't speak for --
25 MEMBER BLEY: They haven't objected to it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
182 1 in anything that I've read.
2 MR. LEBLOND: I haven't -- I can't speak 3 to what they think. I just know what we're doing. A 4 design basis function, it was properly quoted this 5 morning, required by regulations, licensee condition 6 or orders or tech specs or accredited safety analysis.
7 However, when you read that definition 8 there's a footnote, which was not mentioned, and the 9 footnote says this definition comes from Regulatory 10 Guide 1.186. So you can't understand a design basis 11 function unless you go read the Reg Guide from which 12 it generates it.
13 If I go to the Reg Guide, Appendix Bravo 14 to NEI 97-04, it says, what's a design and basis 15 function? To understand a design and basic function 16 we need to understand the underlying general design 17 criteria which are very high level functions. There's 18 further direct guidance that says no individual 19 component performs a design basis function. They're 20 functionally far above any individual component.
21 The converse to that is no individual 22 component, by itself, performs a design function --
23 part of a collective whole. Doesn't mean that a mal 24 -- and when something breaks, it can't propagate to a 25 malfunction. It means that no individual component by NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
183 1 itself performs a design function.
2 And on Page Bravo 5 of that definition, of 3 that document, it says, how do you figure out whether 4 or not a component has a design basis function. It 5 says the safety analysis, not the safety analysis 6 report. The safety analysis provides the context.
7 I was at Fermi all week last week and I 8 did a two-day initial class on screening. People have 9 been in the industry two, three years and now they got 10 to learn how to screen. Can you understand the 11 definition of design function without understanding 12 the Reg Guide? No.
13 So when they get taught this, they're 14 given a copy of these pages that say, here, this is a 15 design basis function. Here's how you orient 16 yourself, design basis function on top, underneath, 17 support, then impact and then action initiators off to 18 the side. It's the very basics of 50.59.
19 It does not mean, in any sense, would you 20 ever ignore a component contribution. The gamut, the 21 technique that's used has been used for 20 years and 22 we're using it here, is that you look at the 23 contribution of a component at a lower functional 24 level. And then you look at the impact of that 25 component as you go up the functional ladder.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
184 1 So the example on the slide is if I'm 2 altering a level indicator on a surge tank on a diesel 3 generator or jacket water, the question is how is 4 jacket water cooling impacted? And as those jacket 5 water cooling goes to diesel. And as goes to diesel, 6 goes to safety analysis assumptions. So in no case, 7 in no case for 20 years, if it's done properly, do we 8 ever ignore the contribution of a component at the 9 level of a change -- ever.
10 And that is interspersed in multiple 11 locations throughout 96-07. And it's specifically 12 identified in Steps 1 and 2 of Section 4.3.6. If I --
13 isn't that right, I think?
14 MR. ARCHAMBO: Yes.
15 MR. LEBLOND: I believe that's right, yes.
16 So Point Number 1, we're just following our notes, 17 lower left hand side. There's approved regulatory 18 guidance 1.186, 1.187. There's definitions there.
19 We're going from Definition 39, what's a malfunction 20 -- failure to perform a design function -- to 21 Definition 33. That's a design function.
22 And there's a footnote on that definition.
23 To understand design function you go to Reg Guide 24 1.186. We're just simply following our way through 25 approved regulatory guidance. I'm done with the --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
185 1 next slide if you're ready.
2 MEMBER SKILLMAN: I -- well, let's stop.
3 MR. LEBLOND: Okay.
4 MEMBER SKILLMAN: This is dandy. Talk to 5 me about a reactor coolant pump.
6 MR. LEBLOND: What would you like to know?
7 MEMBER SKILLMAN: Is it a -- does failure 8 to function, failure to pump --
9 MR. LEBLOND: Good.
10 MEMBER SKILLMAN: -- constitute loss of a 11 design function or loss of a safety function?
12 MR. LEBLOND: Well, safety function is not 13 defined in the regulation. So I don't know how to 14 answer the second part. Reactor coolant pumps perform 15 design functions. Why? Because if they fail to 16 function they produce an accident or transient 17 initiator.
18 Further, there's a whole plethora of basic 19 high level design basis functions that are involved 20 with cooling the core, maintaining pressurizer spray 21 if it's that loop, that would support those design 22 basis functions.
23 So right off the top of my head, what do 24 RCPs do? There's two hits on a design function.
25 First accident initiator, secondly supporting a wide NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
186 1 range of design and basis functions. So what I'd like 2 to come back to, and you gave me the opportunity, is 3 since a malfunction is a failure to perform a design 4 function, to be able to make this partition, you need 5 to make it expertly.
6 Now we can now ask ourselves what are the 7 malfunctions I have to think about, because in the 8 case of a RCP, there's now two malfunctions I have to 9 think about. Accident initiators, if the pump trips 10 then we'll have a loss of flow accident. That's an 11 initiator. Or supporting the wide range of design 12 basis functions -- core cooling, you know, cladding 13 and cooling, pressurizer spray control that's another 14 GDC.
15 So each of those, since a malfunction is 16 defined, then as a failure to perform a design 17 function, then we look at the converse -- it's not 18 performed -- as I march through the Definition 39.
19 Steps 1 through 6 of 4.3.6 accomplish that function.
20 That partitioning and then I am parsing out the pieces 21 and then saying well, now let's think about it, if 22 those functions are preserved or not. Am I answering 23 your question?
24 MEMBER SKILLMAN: Somewhat. Let's change 25 channels.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
187 1 MR. LEBLOND: Okay.
2 MEMBER SKILLMAN: PORV, PORV.
3 MR. LEBLOND: PORV? What would you like 4 to know? Which kind of PORVs? I mean air operated or 5 safeties?
6 MEMBER SKILLMAN: Let's take a Target 7 Rock.
8 MR. LEBLOND: Say what?
9 MEMBER SKILLMAN: Target Rock.
10 MR. LEBLOND: Well, that's a trade name.
11 Do you mean --
12 MEMBER SKILLMAN: It's a --
13 MR. LEBLOND: You mean like an air 14 operated one?
15 MEMBER SKILLMAN: Let's take one that's 16 pilot operated, spring, and pilot valve.
17 MR. LEBLOND: Okay. So it's not the 18 safety? It's the atmospheric?
19 MEMBER SKILLMAN: Neither.
20 MR. LEBLOND: It would go directly to the 21 --
22 MEMBER SKILLMAN: It's a pilot-operated 23 valve like Oconee I, II, and III.
24 MR. LEBLOND: Okay. Well, it would be 25 another class if it were probably --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
188 1 MEMBER SKILLMAN: Your theorem is no 2 single component is that function. In that case, in 3 a TMI, it's a single device failed and loss to 4 function.
5 MR. LEBLOND: Well, what I said exactly 6 was no single component performs a design basis 7 functions. I did say --
8 MEMBER SKILLMAN: But this one does and so 9 do the codes.
10 MR. LEBLOND: Well, let me just see.
11 However, the failure of a component of that component 12 may propagate to a malfunction. So in the case of the 13 PORV the failure itself would propagate to a 14 malfunction. So you're actually sort of raising the 15 issue -- sometimes the failure of individual component 16 will propagate up to a malfunction. Sometimes it 17 won't. Sometimes other pieces of the system will 18 accommodate it.
19 So it's hard to say one size fits all. We 20 need to design a process that can accommodate all 21 these variations. So when I say no individual 22 component can perform a design function, I mean at the 23 PORV is part of a larger system that goes directly to 24 coolant drain tank. It comes off the top of the 25 pressurizer.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
189 1 I didn't say if the PORV fails I don't get 2 a malfunction. I didn't say that. I said that the 3 design function itself is at a higher functional level 4 than the function of the valve, which is to open.
5 And so this idea of what functional level 6 do I want to have this, the -- what we're approaching 7 here is what functional level should the rule talk 8 about. What functional level do I want to make the 9 rule control? And what I'm about to talk about is 10 that that's a 20-year-old issue. I helped solve that 11 problem 20 years ago and here we are.
12 Now in the case of this instance, if the 13 PORV that you're asking about opens up, that will then 14 propagate to a malfunction. That will propagate to a 15 malfunction. That's right.
16 MEMBER SKILLMAN: I would certainly agree 17 with that.
18 MR. LEBLOND: That's right.
19 MEMBER SKILLMAN: Let's go on that.
20 MR. LEBLOND: That's right. So there's no 21 attempt to try to minimize that. And in the case of 22 --
23 MEMBER SKILLMAN: Well, what I'm 24 challenging, Peter, is this. Your theorem that the 25 current wording in 6, SSC is really pointing to a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
190 1 larger group of functional performance requirements 2 that need to be considered to define the safety 3 analysis. That's the theorem.
4 MR. LEBLOND: No, that's not the theorem.
5 The theorem is that you look at the impact of every 6 component, no matter how functionally low. And then 7 you make the decision what's a different result. You 8 find out if the results at the safety analysis level 9 are preserved. That's the question.
10 So it's not that you don't -- if there's 11 something wrong with the PORV, then that's going to 12 propagate up to a malfunction at different -- well, I 13 shouldn't be bit. The question here is at what 14 functionally level do I make the decision. Do I make 15 the decision at the component level or at the safety 16 analysis level? Because at the safety analysis you 17 take the impact of the component the smaller level and 18 drive it up to the safety analysis to ensure that the 19 safety analysis remains valid. They're not --
20 MEMBER SKILLMAN: Okay, now isn't this the 21 crux between your interpretation and the staff's 22 interpretations?
23 MR. LEBLOND: Yes. Yes, yes. They're 24 saying -- well, I don't want to say what they're 25 saying. I mean --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
191 1 MEMBER SKILLMAN: I think they know -- I 2 think --
3 MR. LEBLOND: Yes.
4 MEMBER SKILLMAN: -- we've touched on it.
5 MR. LEBLOND: Yes. They're saying you 6 make the evaluation at the component level. And you 7 look at the words, and yes, sir. And I'm about to 8 show you that the rulemaking directly says that you 9 don't do that and this is an old solution that was 10 solved 20 years ago.
11 It's got nothing to do with digital. It's 12 the same problem we had 20 years ago, and I'm about to 13 show you where it was assigned to us 22 years ago to 14 fix -- and we did. And now we're reopening it. We're 15 reopening that -- reestablishing an error in the 16 regulation that was fixed 20 years ago. And that 17 seems untenable. And I'll let Steve talk about it.
18 MEMBER BLEY: If you would. The fact that 19 it was 20 years ago, I mean, you can leave that 20 behind. The real question is what matters here.
21 MR. LEBLOND: Yes.
22 MEMBER BLEY: And if you focus on the 23 engineering, I'd appreciate it.
24 MR. LEBLOND: Yes. Well, what I'd like to 25 --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
192 1 MEMBER BLEY: A little too preachy.
2 MR. LEBLOND: -- highlight is that the 3 implications here aren't digital. We're here for 4 Appendix Delta. The implications are, here, is what 5 if I want to add a new mechanical interlock to a 6 diesel generator air receiver? So the problems were 7 initially challenged for, well, what if I have now a 8 description of a mechanical component? How do I 9 process any change? So the implications are much 10 bigger than simply digital.
11 So that's why I'm going back to -- the 20-12 year discussion is to say that it's not just digital, 13 it's everything. It's everything. Next slide is 14 coming.
15 MR. GEIER: Yes, I'd like to suggest, 16 Pete, at this point --
17 MR. LEBLOND: To move ahead.
18 MR. GEIER: -- completely move ahead to 19 the next slide, I think so. That would be great.
20 Thanks.
21 MR. LEBLOND: The old regulation use to 22 use the word type.
23 MEMBER BLEY: I'm a little lost, which old 24 regulation?
25 MR. LEBLOND: From pre-1999.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
193 1 MEMBER BLEY: Okay.
2 MR. LEBLOND: So that meant if you had a 3 new failure mod that, required prior Commissioner 4 approval. So the proposal was to change the 5 regulation from type to result, which it did. The 6 words you have here are from the Notice of Proposed 7 Rulemaking written in 1999.
8 Now we've been chastised for saying we 9 don't care about the proposed rule. We want the final 10 rule. This is the final rule. When the regulation 11 was proposed in 1999, there were minor changes from 12 the 1999 version to the statements of consideration 13 but not on Criterion 6.
14 So the words that are here are the intent 15 of the regulation when it was written in 1999. And 16 there it says, unless the equipment would fail in a 17 way not already evaluated in safety analysis -- not 18 the safety analysis report -- there is no need for the 19 NRC to review of the change that led to the new type 20 of malfunction. You wanted -- what was your wording, 21 inconvertible or -- irrefutable. It'll never get 22 clearer than this.
23 It's talking about what's the criteria, 24 where is the level of the result that I think about 25 for when I trip the trigger for Criterion 6. And it NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
194 1 says in a way not already evaluated in a safety 2 analysis. So where do you make the decision? At the 3 safety analysis level. Doesn't mean that you don't 4 ignore what happened to the PRV but you drive the PRV 5 up to the higher functional level and say, does this 6 now produce a result that's not already evaluated in 7 a safety analysis.
8 I mean, I don't know how it can get 9 clearer than this. So since 1999, it used to say 10 something different. It used to say type. We're about 11 to talk about that in the next slide. I don't see any 12 other way to read these words. I don't see any other 13 way to read these words. Once again --
14 MR. GEIER: Next slide or?
15 MR. LEBLOND: Next slide I think, yes, 16 unless there's a question. Following on the heels of 17 the paragraph I just gave you, remember this also 18 comes from the Notice of Proposed Rulemaking these 19 words are written literally came to the next 20 paragraph. You know, we can only put so much on one 21 slide.
22 Four years earlier, the NRC staff had 23 given guidance inside Generic Letters 95-02 they said 24 here is how we want you to access failures of 25 components. And they said, taken by your words, we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
195 1 need to access at the component, at the level of the 2 equipment being replaced.
3 This is the old way of doing it, if you 4 will. Now the regulation is changing from type to 5 result. So the only legal issue in question here is, 6 what result are we talking about? What's the 7 functional level? Exactly what you're suggesting, 8 where in this perspective in this hierarchy, do I talk 9 about result.
10 And it goes on to say this is how we used 11 to say it. Here's how we did it in 95-02, and then 12 the words that start with "unless" were added as part 13 of the Notice of Proposed Rulemaking.
14 So what's the only functional level 15 discussed? The SAR analysis, not the safety analysis 16 report -- analysis -- was truly binding and 17 applicable. So once again, in the course of two 18 paragraphs, two direct citations to the SAR analysis.
19 Next slide, I think unless there's questions?
20 MEMBER BLEY: Well, a couple of things I 21 don't mean to pick. SAR is Safety Analysis Report?
22 MR. LEBLOND: Analysis.
23 MEMBER BLEY: Analysis?
24 MR. LEBLOND: Right.
25 MEMBER BLEY: It is Safety Analysis Report NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
196 1 and you're saying it's the back up to it. When you 2 say this is taken from the Notice of Proposed 3 Rulemaking, I assume these were parts of the 4 statements of consideration? Is that correct?
5 MR. LEBLOND: Well, the -- right, it's 6 actually the Notice of Proposed Rulemaking then starts 7 the public notice period. Then it goes to statement 8 of consideration for the final rule. So this is part 9 of the Notice of Proposed Rulemaking. And then the 10 statement of consideration says if they make any 11 changes from the Notice of Proposed Rulemaking. And 12 so for Criterion 6, there were changes.
13 MEMBER BLEY: Okay.
14 MR. LEBLOND: So this is, I mean, you 15 know, I don't want to get too legal here, but this the 16 intent --
17 MEMBER BLEY: Pretty legal.
18 MR. LEBLOND: -- of our -- these two 19 slides, when you use the word result, that's what was 20 being adjudicated -- type to result. SAR analysis, on 21 the previous page, says safety analysis. I don't know 22 how it can get clearer than that. So I would say 23 that's irrefutable, in my view.
24 The other point that I struggle with, 25 candidly, is I've tried to look at the alternative, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
197 1 I've tried to say well, okay, where have I made a 2 mistake where have I missed something? There's no 3 evidence for using the entire descriptive material.
4 There's no evidence in this to say that result mean 5 anything other than what we just talked about. If you 6 can -- can you back up?
7 MEMBER BLEY: I was hoping when you said 8 you had thought about it you had thought about could 9 we have -- not with respect to the literal language 10 that's here -- could we have missed something that 11 could be important to safety by coming to this level 12 rather than the lower level?
13 MR. LEBLOND: Well --
14 MEMBER BLEY: I was hoping that's what you 15 would focus on.
16 MR. LEBLOND: This is one of eight 17 criteria. Each of the criteria has a role and this 18 criteria has two parts to it. Greater possibility, 19 there's an element of plausibility, and then that's 20 the sufficiently low we've been talking about.
21 And then, if it's plausible, then can it 22 produce a non-bounded result. So within the narrow 23 range that Criterion 6 applies, it does do its job.
24 It does do its job. Now but it doesn't do its job 25 without the help of the other seven criteria. Each of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
198 1 the seven criteria have got a role, and Criterions 6 2 does play its role.
3 MEMBER BLEY: I'm just hanging a little 4 bit. These words seem straightforward.
5 MR. LEBLOND: They do don't they.
6 MEMBER BLEY: But there is some assurance 7 that the mode of failure can be detected and that 8 there are no consequential effects. Those have to be 9 consequential effects that would affect the safety 10 analysis.
11 MR. LEBLOND: Absolutely.
12 MEMBER BLEY: Yes, I would agree with 13 that.
14 MR. LEBLOND: And we agreed that -- that 15 is --
16 MEMBER BLEY: But the real key you have to 17 look for consequential effects. They don't --
18 MR. LEBLOND: Absolutely.
19 MEMBER BLEY: -- have to be in the 20 original analysis.
21 MR. LEBLOND: Well, you have to -- that's 22 the interface between the engineering work.
23 MEMBER BLEY: This is what would take you 24 beyond the original safety analysis.
25 MR. LEBLOND: Well, this would take you in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
199 1 the engineering work to say well, you know, do I 2 really understand this mod. So when I make a change, 3 the PORV isn't a good one because as soon as it fails, 4 then we have an issue, yes. But --
5 MEMBER SKILLMAN: That's why it's a good 6 example.
7 MR. LEBLOND: Yes, right. But I mean in 8 terms of propagating upward. But taking that, do you 9 understand what -- you know, you're making some 10 changes to the PORV. Do you understand how that can 11 be manifested -- this is an engineering word -- before 12 you come to this effort. So, you know, that's really 13 where Neil would really expand to say, well, you know, 14 have you considered these elements? Is this a part of 15 the engineering work? Have you considered more than 16 that? So you know how to answer this question.
17 Once you get to this point those technical 18 questions, and I'm trying to give Neil a second to 19 think, these questions should be answered. And, Neil, 20 I don't know do you agree?
21 MR. ARCHAMBO: Yes, fully.
22 MR. LEBLOND: All right. So again, it's 23 not -- I've said it so many times but it probably 24 bears repeating. It's not that you overlook these 25 effects. It's that you make the final decision when NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
200 1 it says different result. You make the decision at 2 the SAR analysis level. That's the only issue in play 3 here.
4 MR. ARCHAMBO: And if I'm not mistaken, 5 Pete, you know, 20, 22 years ago, when they were 6 talking about types, types of failures, you know, if 7 I had a valve in my FSARs that failed open and now I 8 put in a valve that fails closed, by the staff's 9 definition that would be a malfunction with a 10 different result.
11 MR. LEBLOND: Different type.
12 MR. ARCHAMBO: Different type, but by 13 their definition laid out today that would come in for 14 NRC approval. Even if it had no effect whatsoever on 15 the safety analysis, it failed at the system level in 16 a different way. So that was -- in their definition, 17 I would call that a malfunction with a different 18 result. What we're saying is that's not where you 19 look at it. And 22 years ago that's where it was 20 looked at and decided that's not the proper place to 21 look at it.
22 MR. LEBLOND: Right. 20 years ago, 23 there's an example that said if I go from, if I 24 recollect, it's an oil-filled switch to a diaphragm 25 switch or from a switch --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
201 1 MR. ARCHAMBO: Right.
2 MR. LEBLOND: -- to -- I don't remember the 3 example.
4 MR. ARCHAMBO: Transmitter.
5 MR. LEBLOND: Transmitter. And they said, 6 well, that's a malfunction of a different type and 7 that requires licensing. So that was the issue at the 8 time. Any new failure node at the component level 9 requires license amendment. And so the problem to be 10 fixed was type to result. And, candidly, people 11 declared success.
12 Nobody thought, you know, these words, you 13 know, my next tag line of my little talk here is if 14 you combine Point Number 1 and Point Number 2, well, 15 this issue of what's a different result has been 16 hiding in plain sight all these years. We just never 17 went and looked at it because, until you get to a 18 digital modification, these issues don't become 19 important.
20 So going from Definition 39 to 33, follow 21 the footnote. Go to Reg Guide 1.186, look at the 22 Notice of Proposed Rulemaking. It's been there all 23 these years. But until we come to Appendix Delta, we 24 will never had to say well, what functional level does 25 result really mean.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
202 1 MEMBER BLEY: So if you started where the 2 staff seems to be and said, gee, I put in a valve that 3 fails in a different -- fails open instead of close --
4 MR. LEBLOND: Okay.
5 MEMBER BLEY: -- that's certainly 6 something different. But then you're arguing -- you 7 acknowledge that's something different and you follow 8 the safety analysis to see if it --
9 MR. LEBLOND: Correct.
10 MEMBER BLEY: -- puts you in --
11 MR. LEBLOND: Right. A change. And --
12 MEMBER BLEY: Now, see, 20 years ago we 13 all had PRAs of all the plants and you could do that 14 real, I will say, precisely. If we reflect to the 15 safety analysis in the FSAR, mostly Chapter 15, 16 sometimes it was a different chapter, some of these 17 things weren't even considered in the model. So 18 that's where I'm kind of hanging. You'd had to look 19 and say, if this new valve that fails in a different 20 way carry that up and say --
21 MR. LEBLOND: Right.
22 MEMBER BLEY: -- does that situation --
23 should that affect my safety analysis.
24 MR. LEBLOND: Right. And to be fair, you 25 know, the three people that really generated what's in NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
203 1 the 4.3.6 is myself, Neil, and Kati. And when we 2 first came up with 4.3.6 we had a table of all the 3 permutations. Because there's a lot of different 4 possibilities about, you know, what kind of function 5 is it; is it quickly manifested.
6 Does it proceed slowly, you know is there 7 single failure or is it redundant -- go through all 8 the permutations. And I think we came up with eight 9 or nine categories, something like that. Well, we 10 ended up having to say, well, we can't put that many 11 examples in 4.3.6, so some of them fell by the wayside 12 some of them got combined.
13 But this whole idea of starting with Step 14 Number 1, what are you doing. Step Number 2, go up 15 the functional ladder find one of those four classes 16 of a design function -- design basis function, 17 support, transient initiator.
18 Which one am I involved with? And then 19 from that point the next four steps go different ways 20 based upon what kind of function you've identified.
21 And, you know, without -- between, Neil and I, we'll 22 be happy to work through any permutation.
23 But, we certainly can't say it's one size 24 fits all because, you know, we had different, we had 25 eight different -- if memory serves, eight different NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
204 1 combinations of oh, it can go this way, it can go that 2 way. And that's why there's so many examples.
3 Next slide? If they're ready, I'm ready.
4 This is the point we've been talking about from -- in 5 1997, and I know you're going to -- this is old school 6 here. One of the problem was, if you tie the scope of 7 59 to the SAR, not the SAR analysis, because people 8 have differing levels of descriptive materials you get 9 different answers.
10 Well, you say well, some of them are 11 sparse. For Neil, say well, compare Robinson to 12 McGuire, a factor of two, factor of three. So now, 13 let's say McGuire has a description of some widget --
14 a compressor or a transformer -- and it fails. And 15 Robinson doesn't. Maybe it's digital-related maybe 16 it's not.
17 Well, you want to level the playing field.
18 Shouldn't we have the same change treated the same at 19 two different sites? And that's what that means, it 20 says look by focusing on the words, we focus on the 21 words we get uneven application. And Dresden can do 22 something that Comanche Peak can't. Well that's not 23 right. That simply isn't right. How does that make 24 sense for regulations? So we were asked to fix that.
25 And how do we fix it. We focused on NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
205 1 verbs, functions. What kind of functions, high-level 2 functions, design functions. We're back to the first 3 slide. So now, maybe not everybody has a description 4 of a failed component. Maybe Robinson doesn't and 5 McGuire does. But everybody's got these high-level 6 design functions credited in their safety analysis.
7 Everybody has those functions. So that is 8 a way to level the playing field, so we have the same 9 change, treated the same as Dresden, as Comanche Peak 10 at Robinson versus McGuire. And notice this 11 discussion is not necessarily digital-specific. So 12 that's the solution. The solution is to focus on 13 verbs, high-level verbs.
14 And the gamut that I'm about to get to, 15 which was executed in all the other criteria, is to 16 say, if you're going to make the decision at the 17 higher functional level then you got to be sure that 18 when you make a mod way down here, you drive it 19 properly up to the higher functional level so the 20 decision can be made properly.
21 And that is written in multiple places in 22 96-07 and it's expressly written in Steps 1 and 2 of 23 4.3.6. Because if you don't do that it all falls 24 apart. It all falls apart.
25 MEMBER SKILLMAN: I'm going to repeat that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
206 1 back. So it's clear --
2 MR. LEBLOND: Yes, sir.
3 MEMBER SKILLMAN: -- in my mind. What you 4 just said is even though you're going to drive up to 5 the verb, the design function, you're going to begin 6 at the SSC level --
7 MR. LEBLOND: Level of change.
8 MEMBER SKILLMAN: -- and examine that 9 through the entire --
10 MR. LEBLOND: Process.
11 MEMBER SKILLMAN: -- upward tier to that 12 function?
13 MR. LEBLOND: Exactly, precisely correct.
14 MEMBER SKILLMAN: 10/4.
15 MR. LEBLOND: Precisely correct.
16 MEMBER SKILLMAN: Okay.
17 MR. LEBLOND: And what I want to highlight 18 is that gamut, that technique, that approach was 19 developed 20 years ago, when the next slide is going 20 to say we use it on Criterion 3, Criterion 4, 21 Criterion 7 today with the same rule language as we 22 see in 6.
23 So we're just simply executing the same 24 approach in Criterion 6 that we used in 7, 3 and 4 25 because 3, 5, 7 and sometimes 2 make the decision at NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
207 1 the safety analysis level.
2 For example, Criterion 7 says, is the 3 design basis limit which is at the safety analysis 4 level, is exceeded or altered. There's a paragraph in 5 4.3.7 that says, be careful you got to cascade upward, 6 and we couldn't agree more.
7 Finally, on this bullet --
8 MR. GEIER: If I could just --
9 MR. LEBLOND: Yes, sir, I'm sorry.
10 MR. GEIER: If could just add where you're 11 getting at is the whole point is you're not ignoring 12 what happens at the SSC level. It's still being 13 evaluated. It's still being identified. What has 14 changed? Because this is all applicable to mods, 15 changes in the plan.
16 And so, you still identify it. It's just 17 that you're taking that, you know, what has changed 18 and you're applying it up higher.
19 MEMBER SKILLMAN: Well, I --
20 MR. GEIER: And in your yes/no, you know, 21 on the question, that's being applied at the high 22 level just not at that first tier.
23 MEMBER SKILLMAN: Well, it's clear to me, 24 in order to pull this off, the evaluator has to have 25 a thorough knowledge of the SSC in question. And if NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
208 1 the people are good at their art, they have a 2 documentation trail that's bulletproof.
3 MR. LEBLOND: I agree.
4 MR. GEIER: That is what the design 5 process requires in a 50.59 process.
6 MR. ARCHAMBO: And that's why we set up 7 the six-step process, to walk people through that, 8 because that's absent in the other guidance.
9 MR. LEBLOND: And lastly --
10 MEMBER BLEY: I get all of that but you --
11 I just went back and looked up the rule. And the 12 material you gathered from the proposed rule change --
13 MR. LEBLOND: Yes, sir.
14 MEMBER BLEY: -- has -- is a very nice 15 basis.
16 MR. LEBLOND: Thank you.
17 MEMBER BLEY: When you read the actual 18 rule, they don't say final safety analysis or safety 19 analysis or safety analysis. They say final safety 20 analysis report on every one of the six, eight 21 criteria.
22 MR. LEBLOND: That's going to be my next 23 point. That's Point Number 4.
24 MEMBER BLEY: Okay.
25 MR. LEBLOND: Yes. I agree, I agree.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
209 1 MEMBER BLEY: So even though they wanted 2 it to be the other way, they wrote words that allow 3 some confusion.
4 MR. LEBLOND: Well I'm going to take that 5 point on directly in the next -- as soon as I'm done 6 with my last bullet here, then that's Point Number 4.
7 MR. GEIER: Point Number 4?
8 MR. LEBLOND: Well, Point Number 3 or 9 whatever.
10 MEMBER MARCH-LEUBA: I wanted to ask a 11 question. Back to this question, the analyst has to 12 have the general knowledge of the SSC, but they also 13 need to have a detailed knowledge of the SAR analysis 14 to propagate it.
15 MR. LEBLOND: Well, they have to 16 understand the -- they have to identify the high-level 17 verb. That's what they have to do. So, for example, 18 in a class I just held last week for initial people, 19 the high-level design basis function is to ensure 20 monitored releases.
21 What's the change? This is initial class.
22 I'm changing the control mechanism on a balance and 23 control damper. So, the class is expected to go from 24 a balance and control damper control, to balance flow 25 to flow from lesser to greater, to no unmonitored NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
210 1 release, to achievement of the design basis function 2 up to ensure all the releases are monitored.
3 So, that's for a class that, four hours 4 ago, they couldn't say 5059. So, that's how they're 5 taught. They're taught to do this cascading up. And, 6 if they don't, it all falls apart.
7 MEMBER MARCH-LEUBA: But my point is, it's 8 one thing if you tell -- you have to 16 hours1.851852e-4 days <br />0.00444 hours <br />2.645503e-5 weeks <br />6.088e-6 months <br /> of this 9 analysis.
10 MR. LEBLOND: Yeah.
11 MEMBER MARCH-LEUBA: Another thing is 12 being able to do it.
13 MR. LEBLOND: Yeah.
14 MEMBER MARCH-LEUBA: I'm being cognizant 15 of how a failure of a -- And, there is always 16 unintended consequences that you don't prepare 17 properly.
18 MR. LEBLOND: I agree. I agree. And, 19 I'll tell you, with a class of 29, a fraction of them 20 can't do it. Some of them can't make the connection 21 between some small component and a higher function in 22 a definition of design function. Some -- you know, 23 roughly, 20 percent, say I don't get it. I don't get 24 it.
25 MR. GEIER: But if I can just interject?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
211 1 So, the question is, okay, how does a person, you 2 know, that's new, how would they do that? And, the 3 thing is, they go find the right people that can 4 answer those questions.
5 Because most stations with fleets have 6 safety analysis individuals that that's what they do, 7 is they maintain the safety analysis. And, the design 8 process points them to go there and consult with those 9 people if they can't answer those questions 10 themselves.
11 MEMBER MARCH-LEUBA: But my concern is, I 12 mean --
13 MR. GEIER: But it's not done in a vacuum.
14 MEMBER MARCH-LEUBA: We're supposed to 15 think of safety at a higher level. And, my concern in 16 this process will be that it is very clear that this 17 damper goes into the selection towards that function.
18 And, you forget there is a little branch 19 that, because the temperature around here or the --
20 something else happened and you never analyzed it.
21 That a failure causes an unintended consequence that 22 you never think of.
23 So, unless the process is thorough, I'm 24 telling you there's some good engineers with a lot of 25 experience, there is that risk.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
212 1 MR. GEIER: That's why it's really 2 important that design, particularly for designs with 3 potential safety significance, it's a team effort.
4 And you need to draw on the expertise of the 5 organization.
6 MR. ARCHAMBO: Yeah. And one comment to 7 make. When an individual writes a 5090 evaluation at 8 a site, it's reviewed. It's approved. And also --
9 and I can't speak for every site. I'll speak for 10 Duke.
11 Every single 5059 evaluation goes through 12 a challenge board. Every single one, whether it's 13 digital or non-digital, it goes through a challenge 14 board of subject matter experts to review it.
15 And so, you never have a situation where 16 one individual writes a 5069 and files it away and 17 nobody ever looks at it. So, there's checks and 18 balances, as there is to other parts of our design 19 process.
20 MEMBER MARCH-LEUBA: And, that's good to 21 know. But for consideration that there has to be an 22 emphasis on unintended consequences.
23 MEMBER RICCARDELLA: Are they done under 24 10 C.F.R. 50, Appendix B?
25 MR. ARCHAMBO: That's correct. Yes. We NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
213 1 do have our -- our design process is. And, they have 2 to have qualifications, you know? That's why Pete 3 goes around and teaches. And, I've taken Pete's 4 course many years ago.
5 MR. GEIER: And, you have to take tests.
6 MR. ARCHAMBO: And then we'd have, you 7 know, every year we have refresher training on 5059.
8 Because, quite frankly, it's nothing a lot of people 9 do all the time.
10 That's why we have challenge boards.
11 That's why we have subject matter experts review that 12 work. Because, you know, it might be six months, a 13 year, between the time you did your last 5059 14 evaluation.
15 MEMBER BLEY: I heard of something from --
16 can I mention it? I won't say where it was. But Ron 17 participated on a panel which -- challenge board would 18 be the right name for this.
19 But they assigned two groups, one to --
20 for why this was good and one to show why it was 21 different and argue it out. Is that a common kind of 22 practice or this unusual? I've never seen anybody do 23 that.
24 MR. ARCHAMBO: Well, yeah. I was -- yeah.
25 MR. GEIER: I can hit this. One of my --
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
214 1 I participated in the development of, you know, an 2 engineering error-type process in back in the early 3 2000s. It was with INPO.
4 And, one of the different tools, depending 5 on, you know, if you're a knowledge worker and, you 6 know, what the potential error-likely situation is.
7 But the use of a devil's advocate is what is actually 8 a defined error prevention tool.
9 It's in there. And, I think these 10 challenge boards have really, you know, kind of taken 11 that on. And, we've done that. I've done that on 12 root causes, where you have a root cause and you take 13 somebody off and have them off to the side.
14 Your job is to prove why our conclusions 15 are wrong. And, it's a pretty common tool, not just 16 for 5059, but also for design, for design boards, as 17 well as root causes, corrective action within the 18 station. It's become a pretty common tool.
19 MR. LEBLOND: I would say, if the 20 underlying technical work is accurate, you know, what 21 I would preach -- what we were trying to do is that 22 you can -- different groups should come to the right 23 answer, the same answer, in the same way. That's what 24 we hoped to get to. However, if a mistake is made at 25 the engineering level, if you fail to understand those NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
215 1 effects we were talking about, there's nothing in this 2 process that catches it.
3 I mean, and I use the example of San 4 Onofre, as the extreme. Those 5059s looked -- they 5 certainly didn't catch the fact that they'd have fluid 6 elastic instability. If they did, they would've had 7 a different result. But they didn't know, and so here 8 we are.
9 MEMBER BALLINGER: I would contend that 10 that analysis was flawed and that a proper, murder 11 board-type analysis would've picked that up because --
12 MR. LEBLOND: Me too.
13 MEMBER RICCARDELLA: A proper what type of 14 analysis?
15 MEMBER BALLINGER: Well, we were talking 16 about it today. I call it a murder board. We're 17 trying to put a -- wanting to put a detector on the 18 pressurizer surge line, very unusual, to detect 19 cracks. And so, it was non-intrusive.
20 It was not going to be used for any kind 21 of action that was going to take place. It was just 22 a monitoring thing. And so, we had two groups that 23 were -- one was assigned to say you can't do this.
24 And then, the other one was assigned to 25 say you can do this and we fought it out. We, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
216 1 meaning, I was a bystander. I was the person that 2 wanted to do it.
3 The utility was the one that did the 4 analysis. And that's how that worked. And, if that 5 kind of approach had been taken at San Onofre, they 6 would've picked that up.
7 MR. LEBLOND: I agree with that.
8 MEMBER BALLINGER: There's no doubt that 9 they would've picked that up, I think.
10 MR. LEBLOND: I agree.
11 MEMBER RICCARDELLA: But again, we're 12 talking 5059. And, it's not clear to me that even if 13 the San Onofre hadn't been done under a -- had been 14 done under a licensed member request, the same thing 15 might've happened. There was a chance the same thing 16 might've happened.
17 MR. LEBLOND: The fundamental flaw is the 18 technical issue and it passes it along.
19 MEMBER RICCARDELLA: I mean, we've had 20 other big issues like that, the Mark I -- the BWR Mark 21 I containment program comes to mind. I men, there 22 were errors made in that analysis. And that had 23 nothing to do with 5059, it was just bad engineering.
24 MR. LEBLOND: Exactly. So, I mean, that 25 gets back to your observations, which I think are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
217 1 well-taken. You need to know those affects so you 2 know what to cascade upward.
3 If you don't know those affects, I can't 4 help you. I just can't help you and it's not going to 5 work. I will just tell you right now, it's not going 6 to work.
7 MEMBER MARCH-LEUBA: Let me put something 8 else on the table that I'm thinking of and it's really 9 different. And, it's what I call time for recovery.
10 It's something that's specific for --
11 We're all used to working with computers 12 and the computer starts working unusually slow, and 13 you do control, alt, delete, reboot, and it starts 14 working. Take for example, a system data conditioning 15 in the control room.
16 MR. LEBLOND: Yeah.
17 MEMBER MARCH-LEUBA: It's a really bad 18 system and you need it for habitability. But if this 19 system were to freeze, the operators would notice 20 immediately. They would get on the horn and they 21 would reset it, and it would start working. So, with 22 this, I'll give you the credit for time for recovery.
23 MR. LEBLOND: Yes. Interesting. Go 24 ahead. Well, do want -- well --
25 MEMBER MARCH-LEUBA: That's it. Go ahead.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
218 1 MR. LEBLOND: One of our examples is the 2 -- two, safety-related, air compressors -- is that the 3 right term?
4 MR. GEIER: Chillers.
5 MR. LEBLOND: Chillers that provide 6 cooling to the control room. Okay.
7 MR. GEIER: Control mainframe.
8 MR. LEBLOND: Pardon me?
9 MR. GEIER: Control mainframe -- right.
10 So, we've had a lot of conversation with this. You 11 know, what if you have a software -- now, we talked 12 about common platforms. And, during those 13 conversations, Neil nudged me and said, well, we could 14 make a common platform. We could put the same 15 software and two components. That's true, right?
16 MR. ARCHAMBO: Yeah. Let me just clarify 17 that.
18 MR. LEBLOND: So, clarify that.
19 MR. ARCHAMBO: You have two trains of 20 chillers, two safety-related trains. There's -- you 21 cannot legally take one controller to control both of 22 those trains.
23 I want to make that clear -- without 24 getting a license and memory request. Because it's 25 not nothing to do with digital. I could do that with NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
219 1 an analog controller. But once I do that, I've 2 reduced -- or independence.
3 I've reduced independence. And, 9607 will 4 tell you, a single reduction in independence goes for 5 a license and memory request. So, what we're talking 6 about is two analog chillers, separate, completely 7 separate. We put digital controls on each one of 8 those, not the same digital controls.
9 MEMBER MARCH-LEUBA: But they're exactly 10 digital controls?
11 MR. ARCHAMBO: The exact same.
12 MEMBER MARCH-LEUBA: Completely random, 13 but not diverse.
14 MR. ARCHAMBO: And just to set this 15 example up before Pete goes, is we -- it's a real-life 16 example. You see, we have an -- at one of our plants 17 that talks -- that has enough on the A table that we 18 say is descriptive material.
19 But it has enough on the A table that 20 says, if one train fails, the other one starts. So 21 now, if I put in both digital controls, and I can't 22 say I'm not susceptible to CCF, where both trains 23 could fail, the staff would say that's a different 24 result.
25 And, we'd say, not right away it isn't.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
220 1 We've got to see. We've to see. We've got to 2 propagate that up to the higher functional level. And 3 I'll let Pete finish that, but that sets up the 4 example.
5 MEMBER MARCH-LEUBA: My claim is that you 6 break the receipt and advance into two classes, the 7 ones that will -- the software failure will be so bad, 8 that it will prevent the second one from starting, and 9 the ones that -- blue screen up there that we said we 10 would fix. And you still have the common cause 11 failure, but it has a much lower priority, if you 12 allow for time to go and push the reset button.
13 MR. LEBLOND: And, that's what we shared 14 when we went through this is, we actually identified 15 eight, separate classes and permutations of things to 16 come. The solution to that one is, well now, as I 17 climb the ladder, what design basis function do I get 18 to?
19 And, the example says, you get to the 20 cooling of the reactor protection racks. How long do 21 you have from the time that those two coolers fail to 22 start, to the time where those reactor protection 23 racks exceed their designery.
24 And, I think in the example we said 20 25 minutes, didn't we? Anyhow, so the system has to be NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
221 1 designed, as you directly pointed out, with some way 2 to say, hey, how come these things didn't start? I 3 don't know. Mash the bypass button.
4 Get one of them running. And, we have 5 another and off you go. So, that then says, you have 6 a soft. So, now in the language of 5059, do you have 7 a malfunction? Yes, both fail. But does it propagate 8 to a different result at the safety analysis level?
9 No.
10 MR. ARCHAMBO: And, one thing I'd like to 11 point out about that is that's -- you know, that was 12 giving an example of things that -- equipment that 13 don't show up in a safety analysis. But as Pete just 14 pointed to you, that doesn't mean we don't consider 15 it.
16 MR. LEBLOND: Exactly.
17 MR. ARCHAMBO: It failed. We have to see 18 what affect that has on equipment that is in the 19 safety analysis. So, it doesn't get ignored, just 20 because it doesn't show up in a safety analysis.
21 MR. LEBLOND: And, let's say, somebody 22 proposes a mod to say, I don't know. I'm not going to 23 put that bypass in. I'm just going to put in my two 24 little software panels.
25 And, I'm going to have a little screen NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
222 1 here, and a screen here. Well, it becomes dumb-mod, 2 just like any other dumb mod. And now, they've 3 designed a mod that they can't implement. So, another 4 common problem is, people designing a modification 5 that they can't implement.
6 And so, there would just be another 7 example of saying, well, you've got to come up with 8 one way. You know, at lunch, Steve said, well, you 9 could just always go mash them, MCC, if it comes to 10 it. But if you design it so you can't do that, then 11 you've designed a mod you can't manipulate on your own 12 and, just like any other mod, it's a dumb mod.
13 MEMBER SUNSERI: But in those examples 14 that you're citing that -- where now you're 15 introducing a human element, where before there wasn't 16 one, right? So, I mean, you're going to propagate now 17 too, right?
18 MR. GEIER: So, part of the process -- and 19 this is actually before you even get into the 5059.
20 If you do the technical evaluation of this particular 21 mod and you look at that and you say, okay, can you --
22 in that type of scenario, can you rely on manual 23 action, or operator action to go do that within that 24 time period?
25 And, you have to do a human reliability NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
223 1 evaluation to determine of that's the case. But you 2 also look at other types of things, whether it's 3 defensive measures or coping. And, quite honestly, 4 you know, we talked -- when I was talking, we talked 5 about this technical document that we need to replace 6 NEI 16-16.
7 That's the hole that that's going to fill 8 is what type of design attributes, in addition to the 9 ones that are already listed, that can be utilized to 10 able to result and to say that, even if I have common 11 cause failure, I can cope with it or I can take an 12 action so that it will still perform its function.
13 MR. LEBLOND: And, in the point of the six 14 steps where we -- remember, I told you, after you get 15 to step two, they go different ways. And, if you 16 follow the path, it goes through this.
17 There's a statement that says that any 18 actions have to be what's called interdependent with 19 the mod. So, if you create a mod that's got a start 20 bypass, you've got to have a procedure for start 21 bypasses, for starting it, for operating it.
22 And, you have to train the people on 23 operating it. So, those are requirements of the 24 underlying mod. Those are all interdependent. And 25 so, therefore, you can -- you know, that that passes, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
224 1 whatever you elect to use has to be interdependent.
2 Now, let's say, oh, I want to open that 3 door and put a fan in it? That's not interdependent.
4 That's a comp measure. That's not allowed. So, that 5 distinction is made in the six steps.
6 MEMBER SUNSERI: Right. But -- and I'm 7 not a human factors expert. But that -- the first 8 part that you just described, just seems to me as 9 though it's independent of what else might be going on 10 as a result of that trip.
11 So, the operator might be distracted. The 12 workload may have exceeded -- this new element may 13 have exceeded their capacity to act, right?
14 MR. LEBLOND: Yeah. That's the HRA 15 assessment that you'd end up doing, by looking at all 16 of those.
17 MEMBER BLEY: And, I see you are an 18 expert. You've got it all.
19 MR. ARCHAMBO: Well, the truth is, there 20 are procedures for that. Even with our analog 21 chillers, we have a procedure in case both of them 22 fail, that tells them exactly what to go on.
23 And, the first step they do is go out and 24 try to restart it. That's the first step. I mean, 25 those exist. That's already taken into consideration.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
225 1
2 MR. GEIER: And then, quite honestly, 3 your, if you lose both, you know, you lose both your 4 303 and shut down anyway. So --
5 MR. LEBLOND: In 96-07, there's guidance 6 as to how extreme and how, I think, about expended 7 reliance and manual action. It's under criterion two 8 and the guidance in the six-step process points back 9 to that to say, make sure that whatever manual actions 10 you ascribe, they're reasonable.
11 So, when I roll play act, what would say 12 that this manual action has to occur within two 13 minutes. It would fail that test. Oh, I've got 20 14 minutes to make this test. Well, it would pass that 15 test.
16 So, that decision point is written in the 17 process. Step four, if I remember correctly. The 18 third bullet on this slide, Reg Guide 1.186. Its 19 purpose is to distinguish between descriptions that 20 are part of the safety analysis and descriptions that 21 are part of the descriptive material.
22 And, it says that the plant's response to 23 an individual comp own's failure is part of the 24 descriptive material, not part of the safety analysis.
25 So, in the past, we've heard -- and I think it's Mr.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
226 1 Riccardella had made that distinction between safety 2 analysis and the --
3 In earlier meetings we had heard, well, 4 that descriptive material is part of the safety 5 analysis. I didn't hear that today. That Reg Guide 6 provides a direct example, on point, that says, if you 7 have a component that fails, and here's what the plant 8 does, that's described as descriptive material.
9 MEMBER BLEY: All very interesting. You 10 promised me when you got to this slide though, it 11 would explain why, despite all the language in the --
12 MR. LEBLOND: It's the next slide. We've 13 been on this slide a long time.
14 MR. ARCHAMBO: Your time has come. Your 15 ship has come in.
16 MR. LEBLOND: But you are correct. I did 17 promise that. You are exactly correct. I've 18 described an approach where we start with a low-level 19 component and cascade upward. And, we heard this 20 morning well, look at the language of the rule.
21 It doesn't fit. You know, it says, as 22 evaluated by the UFSAR. Criterion 3 and 4 focus 23 solely on those analysis. Seven focuses solely on RCS 24 barrier accent analysis, but yet, look at the 25 language, the language, as described in UFSAR.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
227 1 Malfunction previously evaluated in the, 2 previously evaluated. So, those three criteria, and 3 sometimes also, Criterion 2, depending upon the type 4 of change, they cascade up to the safety analysis.
5 But yet, the rule language has this global wording.
6 So, the question now is, how do I read that? Do I say 7 --
8 CHAIRMAN BROWN: Hold it. You say the 9 rule language. Are you talking about --
10 MR. LEBLOND: This is the rule language, 11 the 59-rule language.
12 CHAIRMAN BROWN: So, that's the one you're 13 referring to?
14 MR. LEBLOND: Right. Right.
15 CHAIRMAN BROWN: In other words, Criterion 16 6 you're talking about? I see the first three. But, 17 I mean, when you say, you're trying to relate it also 18 -- this same language to Criterion 6 as well?
19 MR. LEBLOND: Criterion 6 has the same 20 kind of language. It says, than any previously 21 evaluated. That's what it says.
22 MEMBER BLEY: In the FSAR.
23 MR. LEBLOND: In the FSAR. And you'll see 24 the same language here, previously evaluated in the 25 FSAR -- previously evaluated in the FSAR -- as NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
228 1 described in the final signal analysis report.
2 MEMBER RICCARDELLA: But what about 5, 3 which is of a different type than any previously 4 evaluated in the FSAR.
5 MR. LEBLOND: Well, 5 -- an action of a 6 different type would be an action for which you don't 7 have a safety analysis. So, it doesn't really make 8 our observation here. But what I'm trying to say is 9 that this approach of starting at a lower functional 10 level and going to a higher functional level has been 11 around for 20 years.
12 But Criterion 5 would be, I don't have a 13 safety analysis to compare to so, therefore, it's an 14 accident of a different type, so it doesn't really fit 15 this slide.
16 MEMBER RICCARDELLA: Well, but it -- I 17 understand, but I mean, that seems like it opens the 18 door to having to consider accidents that aren't 19 considered in the FSAR.
20 MR. LEBLOND: Well, it does. It certainly 21 does.
22 MEMBER RICCARDELLA: Okay.
23 MR. LEBLOND: That's the point of 5 is 24 that, let's say that you've never brought a gas line, 25 a natural gas line, on site before and you want to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
229 1 bring in a 10-inch line. There's a Reg Guide that 2 lets you bring small lines.
3 Well now, have you now created the 4 possibility of a natural gas explosion on site? Yes.
5 Well, what do I bound that with? What safety analysis 6 do I look at? Well, there is none. So, that becomes 7 an accident of a different type. That's why, in 3, 4, 8 and 6 -- just look at 7.
9 It says, a design basis' limited efficient 10 product barrier is described in the FSAR. If you go 11 to the guidance for that it says, go find the safety 12 analysis that demonstrates the integrity of that 13 barrier. So, is that described in the final safety 14 analysis report? Yes.
15 But it's not the description. So, it's 16 the same technique. It's the same approach, where you 17 start with a component at the lower functional level 18 and go all the way to a higher functional level, to 19 the safety analysis level.
20 So, there's nothing new. There's nothing 21 in the language of Criterion 6 that impedes us from 22 doing that. If you want to say Criterion 6 impedes 23 us, well then, 3, 4, and 7 are also in trouble too, 24 because they use the same global language, inside the 25 rule language itself.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
230 1 So, a regulatory guide is an approved way 2 to meet a regulation. We're just following the 3 definitions. We're following our notes. And now, in 4 hindsight, this issue of where are the results for 5 Criterion 6? It was hiding in plain sight all of 6 these years.
7 We just never went and looked for it. I'm 8 ready for the summary if you are. So, the big 9 finishes. We're using previously approved 10 definitions. And they were created 20 years ago, and 11 they admit the rule.
12 The rulemaking record is clear. There is 13 no evidence for a contrary view. It says, the safety 14 analysis is the level at which you make the decision, 15 unless the equipment would fail in a way not already 16 evaluated in a safety analysis, there's no need for 17 on-site review.
18 I don't know how it gets clearer than 19 that. Reliance on safety analysis results levels the 20 playing field, as Neil said. Do some sites have 21 sparse description, other sites not? How do we make 22 sure we have consistent treatment?
23 That solution -- and that's not a new 24 problem. It's an old problem that was fixed 20 years 25 ago. And finally, finally, finally, this approach of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
231 1 starting at a lower component level, making sure you 2 have all of the technical information, driving to the 3 safety analysis level and say, is this analysis still 4 valid?
5 It's omnipresent in 9607 today. So, we're 6 just using the same approach in Criterion 6, that we 7 used in 2, 3, 4, and 7. Not every change is at 2.
8 So, with that, Mr. Geier, I'm done.
9 MEMBER BLEY: While I don't disagree with 10 your line of reasoning throughout, I see how others 11 would say, despite all of these words from the old, 12 throwaway discussion that led to the change in the 13 rule that used safety analysis or the SAR analysis, 14 when they actually wrote the rule, they didn't write 15 it that way.
16 They wrote it FSAR and left it. So, those 17 arguments, as good as they sound, don't seem to have 18 been carried through when they wrote the rule. And I 19 don't know why that happened and probably none of us 20 do.
21 MR. LEBLOND: I do.
22 MEMBER BLEY: Okay.
23 MR. LEBLOND: I do.
24 MEMBER BLEY: Then tell us.
25 MR. LEBLOND: Then, go back to Peter's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
232 1 slide, to the -- can you back up?
2 MEMBER BLEY: Was there a substantive 3 reason or was there --
4 MR. LEBLOND: Well, it's because this is 5 the language that was similar to the old language, to 6 the old rule. And then, I mean, just take Criterion 7 3, it talks about consequences, okay?
8 9607 said, the same issue we had, had to 9 resolve. At what level are you going to make the 10 answer? At what level are you going to make the 11 decision? And the answer was, in every case, you 12 drive the answer up the safety analysis level.
13 So, Criterion 2, c(2)iii and iv, they 14 focus on the dose analysis described in -- So, where 15 do I find the consequences of an actually previously 16 evaluated? I find them -- 9607 is an approved way to 17 meet a regulation.
18 It says it on the bottom of every Reg 19 Guide. It's an approved way to implement those words.
20 You go to the safety analysis, which is where you find 21 the consequences of an accident previously evaluated.
22 But what if I find some words? And 9607 says, they 23 may be in the FSAR, but it's not the consequences of 24 an accident previously evaluated. The Reg Guide 25 defines that to be the safety analysis.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
233 1 And that approach was used for these 2 criteria and we're just using it again for Criterion 3 6. It dovetails with this idea of the design function 4 has got to be a higher functional level.
5 MEMBER BLEY: I think your technical 6 argument's pretty good. I don't know if you're a 7 lawyer or not, but I'm not sure if the legal arguments 8 are as good.
9 MR. LEBLOND: I'm not a lawyer, but I 10 believe the arguments are --
11 MR. GEIER: Actually, we didn't bring our 12 lawyer with us. Actually, the NEI, Legal, and OGC 13 have been talking about this and I think they're 14 reaching alignment as well.
15 MR. LEBLOND: I'm not a lawyer but --
16 MEMBER BLEY: Charlie, I don't want to 17 jump the gun, but I'm going to, a little.
18 CHAIRMAN BROWN: Go ahead.
19 MEMBER BLEY: We agreed. Well, you and I 20 agreed, sitting at the table earlier today, that 21 probably this isn't the time for a letter. If this 22 disagreement is pushed through and comes out one way 23 or the other, it's fairly significant.
24 And, I'm not sure that I agree with myself 25 that we shouldn't write a letter addressing that one NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
234 1 issue. I have no idea where we'd fall out as a group.
2 But we could have some impact in that area if we wrote 3 something now. We wouldn't later.
4 CHAIRMAN BROWN: Well, if you expect me to 5 write that in two weeks, you've lost the bubble, 6 because I'm lost again.
7 MEMBER BLEY: We can write it later than 8 two weeks. We don't have to write it at this minute.
9 CHAIRMAN BROWN: Right now, I have totally 10 lost the bubble on the interactions of these.
11 MEMBER BLEY: Maybe we ought to have a 12 session at the full committee meeting ourselves to 13 talk about this and see if we have a consensus that 14 would drive us to want to weigh in on this issue.
15 Because I think, assuming we can weigh in 16 later, might be wrong. We might have missed the boat.
17 And let's talk about it some more later. I'm sorry 18 for the diversion, folks.
19 MEMBER REMPE: Could you clarify your last 20 comment about your lawyers and OGC lawyers are getting 21 closer to alignment?
22 MR. GEIER: Right. So, because we know 23 that a lot of this is very legal language, we have --
24 our legal staff, we've briefed them, and they've 25 contacted their counterparts at OGC to talk through NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
235 1 this particular issue to see if they can reach 2 alignment. And my understanding is they're getting to 3 that point.
4 MEMBER REMPE: So, the decision may 5 dissipate in the next couple of weeks?
6 MR. GEIER: I don't know. I just know 7 from what our NEI legal has told me that -- but I 8 don't know what that means in NRC process, as far as 9 if they reach agreement or alignment.
10 MEMBER MARCH-LEUBA: Shouldn't you 11 convince the technical stuff and then talk to the 12 lawyers?
13 MR. GEIER: Well, we've tried. We 14 wouldn't be here if were successful in that.
15 Yeah. This has all been multiple -- and 16 we understand that OGC is a step in the process to a 17 review of the Reg Guide.
18 MR. LEBLOND: This is all old news. This 19 is not a new -- I mean, those folks back there could 20 probably give this story as well as I could.
21 MEMBER BALLINGER: Let me ask a 22 metallurgical question. Is there any practical 23 difference between the two, with respect to safety?
24 In other words, if you take industry's proposed 25 approach versus the staff's opinion, is there any real NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
236 1 difference between the two, with respect to what could 2 happen?
3 Could we miss something by taking one or 4 the other of approaches that would impact safety? I 5 mean, I know -- are you in raging agreement, but 6 disagreement?
7 MR. ARCHAMBO: No. Nothing could impact 8 safety, but the implication, if we went the way that 9 the staff is proposing, it may open up modifications 10 that would have to come in for license amendment 11 requests that would really push the --
12 MEMBER BALLINGER: Okay. If there's no 13 impact on safety, then the choice is pretty easy. But 14 my question is, is one or the other approach -- is 15 there a possibility that one or the other approach 16 would result in something significantly impacting 17 safety down the road?
18 MR. ARCHAMBO: Yeah. I think what --
19 maybe what you're asking is, say, using our approach, 20 could we possibly miss something that the other 21 approach wouldn't miss?
22 MEMBER BALLINGER: Or the other way 23 around.
24 MR. ARCHAMBO: Or vice versa -- and I 25 think the answer is no. I think the issue that we NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
237 1 have is two things. One is, the way Pete laid it out 2 is what was decided 20-something years ago, and it's 3 the way it's been being taught and used for the last 4 20 years.
5 The proposal that the staff put on the 6 table would likely put a number of modifications that 7 we might want to do, come up with a malfunction with 8 a different result, at the system level, not at the 9 safety analysis, but at the system.
10 MEMBER BALLINGER: That's -- again, that's 11 the -- to me, that's the nuts and bolts level. Now, 12 it may be very expensive in one case.
13 MR. ARCHAMBO: That's what's critical to 14 the industry. That's what's detrimental to us though, 15 as an industry.
16 MEMBER BALLINGER: Yeah.
17 MR. ARCHAMBO: That's why when we say --
18 if we get this document endorsed, with the exception 19 of Section 436, it really diminishes the usefulness of 20 this document.
21 MEMBER BALLINGER: Maybe I'm just not 22 being clear. Somebody else has got to say this.
23 MEMBER REMPE: So, let me try and ask your 24 question. When you discussed this with the staff, 25 could they cite one example saying you -- if you did NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
238 1 give an example, you'd miss it with your approach.
2 Our approach won't miss it and it impacts safety. Did 3 they ever give you a concrete example?
4 MR. ARCHAMBO: No. And -- no, they have 5 not. And, they're sitting behind you. They might 6 want to chime in. But to the best of my knowledge, 7 no.
8 MR. LEBLOND: Let me try to make their 9 argument for them.
10 MEMBER BLEY: To me, they're here. We 11 could hear it from them.
12 MR. GEIER: The term miss is -- you know, 13 remember, I think in your term, it misses going in for 14 an LAR, versus being evaluated.
15 MEMBER REMPE: No. I'm talking about 16 safety. They talked about a couple of prior examples 17 where they said, you missed the need to document it.
18 They documented it, as Pete asked. It wasn't 19 determined to be a safety issue. I'm asking, did it 20 impact safety?
21 MR. LEBLOND: Well, I think I can make 22 their argument for them.
23 MR. CARTE: There's a couple of premises 24 wrong with that question. We actually haven't put 25 forward a position. So, what we're discussing is NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
239 1 NEI's position and our objections to it.
2 And they're speculating our position, 3 based on our objection. So, it's a little funny like 4 that. But no. We have not given that. But part of 5 the issue is, what's coming up is, if you create a new 6 situation that's adverse, we review, in general, as 7 our role, the measures to mitigate that situation.
8 So, if you create a situation and you say, 9 yeah, but it's not really that bad because we do this, 10 this, and this to mitigate it, then we're really not 11 evaluating or -- what they do in response to adverse, 12 new adverse condition.
13 So, it's not that -- I disagree with a lot 14 of their characterizations of our positions. And I 15 disagree with their use of quotes, because I think 16 they took them out of quote context.
17 But one of the problems in this discussion 18 is we've always partitioned this thing in terms of two 19 types of discussions, licensing, which is a bunch of 20 logistic-type discussions and technical. And we've 21 never really gotten down into technical.
22 And if you get down into the weeds of 23 technical, then you can start parsing these 24 malfunctions in terms of safety system malfunctions, 25 non-safety system malfunctions. But we never go to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
240 1 those discussions. We're just always arguing at this, 2 one syllogism applies to all cases, when you have very 3 different, specific cases in the plant.
4 So, the short answer is no, we have not 5 pointed that out. But we've never really engaged in 6 detailed technical discussions, because we've never 7 gone down that path.
8 MR. MORTON: I want to emphasize something 9 that Norbert has said in terms of, one of the 10 challenges we had when reviewing Appendix D, is that 11 oftentimes, Appendix D is taking things out of the 12 full context of partial quotations.
13 So, to take a look at the entire 14 definition of safety analysis in 9607 and the entire 15 definition of design function. Because the 16 presentation kind of leads you down one specific 17 pathway, with a consideration of design functions 18 being the sole subset focus of design functions.
19 That's where the staff part -- because 20 there's an entire definition of design function and 21 it's not just about design pieces function, similar to 22 the definition of safety analysis.
23 There's other aspects to safety analysis, 24 not just some of the descriptive material and/or 25 accident analysis. And, taken within that context, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
241 1 leads us down to some of the reasons why the staff 2 disagrees with the totality of the approach in that, 3 with this process. So, I just wanted to clarify that.
4 MEMBER RICCARDELLA: Excuse me. While 5 you're up, could I -- why has this issue on 3.6 come 6 up solely in the aspect of digital INC and Appendix D?
7 I mean, it seems to me this question of 8 whether a component gets considered at the component 9 level versus at the systems safety analysis level, is 10 broader than -- so, has this come up in considerations 11 of the regulation in general?
12 MR. MORTON: You nailed it. And, that's 13 one of the concerns the staff had is that this is the 14 Appendix, specific to digital INC. And addressed the 15 challenge of digital INC as with regard to 16 implementing digital INC models under 5059.
17 Much of what's in the guidance and 18 criterion for the Criterion 6, under 436, is not 19 specific to digital INC. They are generic 20 interpretations of the basic rule language. So, that 21 it's beyond the scope of digital we're talking now.
22 This entire meeting is really about interpreting 5059, 23 not digital INC in terms of the exceptions we took.
24 Take generic interpretations within an 25 appendix of the larger body of 5059 guidance. So, you NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
242 1 kind of hit the nail on the head, in terms of where.
2 MEMBER BLEY: It seems to me that you 3 folks would serve yourselves well if you wrote some 4 type of white paper, really taking this apart and 5 looking at it. And, kind of the same -- I mean, you 6 presented slides here, but I don't know if you have a 7 white paper on this from NEI as well.
8 And, they pointed out that the two Reg 9 Guides, 186 and 187, 1186 and 1187 -- I don't recall 10 seeing those called out in your document, but I could 11 be wrong about that.
12 But I think that if either of you wanted 13 to prevail on this for whatever reasons you want to 14 prevail, getting some clarity on what's the same or 15 what's different across this discussion would be very 16 helpful, to us and to the commission one day. Yeah.
17 You've got to do your name.
18 MR. BEAULIEU: Yeah. I'm Dave Beaulieu.
19 I'm the Agency's 5069 person.
20 MEMBER BLEY: We need it for the 21 transcript. That's all.
22 MR. BEAULIEU: There's a couple of things 23 that they leave out. You raised one point about that 24 the 5059 rule says what it says. It says, any 25 previously evaluated in the FSAR.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
243 1 It doesn't say in the act of an analysis.
2 It doesn't say -- it doesn't limit it. It just says 3 in the FSAR, okay? That's for one. So, if they want 4 to narrow it, that would require a rule change, for 5 one.
6 And, not only in 5059, but 10 C.F.R. 5034 7 is for the regulation for the FSAR. And that -- the 8 regulation for the FSAR gives a definition here of 9 final safety analysis report. And, it says that each 10 applicant for an operating license shall include a 11 final safety analysis report.
12 It says, the final safety analysis report 13 shall include information that describes the facility, 14 presents the design basis and limits on its operation, 15 and prevent -- and presents the safety analysis of 16 structure systems and components and of the facility 17 as a whole.
18 They say -- it's two pieces, which is 19 consistent with any NEI 9607, the definition 312. It 20 says, also within the meaning of this definition, for 21 the purposes of 5059, is supporting FSAR analysis that 22 demonstrate SSE design functions will be accomplished 23 as credited in the accident analysis.
24 So, they make a distinction between the 25 accident analysis and the safety analysis are the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
244 1 supporting FSAR analysis. That's exactly what the 2 5034, C.F.R. 5034 would say. And so, they would need 3 -- to take their approach, they would need role-making 4 because what they're saying is contrary to that as 5 well.
6 MEMBER BLEY: I'd come back to, there's a 7 lot of disagreement about what words in the rule mean 8 and how to interpret them. It seems the two sides 9 here could come together on what's important to safety 10 and then figure out what needs to be done with respect 11 to the rule. Otherwise, somebody's going to be really 12 mad when this is over.
13 MR. BEAULIEU: You're correct. So, we're 14 saying -- our approach was saying, if you have a -- if 15 the chapter says, malfunctions are evaluated, 16 generally evaluated, as potential single failures.
17 That's true.
18 So, you go to the system. You go to the 19 system description and it says single failure. If you 20 have a common cause failure of software, that's a loss 21 of the entire safety system. That's a non-functional 22 safety system.
23 And that requires -- that's a different 24 result. That's a safety issue. That requires prior 25 NRC approval. That's a common cause failure of a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
245 1 safety system. It's not functional.
2 MEMBER BLEY: But the arguments we've been 3 hearing don't seem to deal so much with common cause 4 failure as with how you parse out the results on a 5 particular SSC.
6 MR. BEAULIEU: Right. Because the result 7 -- Criterion 6 is really not -- why is this a digital 8 issue? It's because of software common cause failure.
9 And other mechanical systems are not faced with common 10 cause failure as directly or as commonly.
11 MEMBER BLEY: But we have a record of them 12 on common cause failure. We have data. We understand 13 those common cause failures better than we do the 14 digital ones.
15 MEMBER BLEY: Yeah. Right.
16 MEMBER RICCARDELLA: Excuse me. Wouldn't 17 the concern that you just talked about come out of 18 this cascading up concept that NEI talked about? I 19 mean, if what you said happened, and you have common 20 cause failure that affects multiple systems, I mean, 21 I would think that if their evaluation right, as they 22 cascade up, they'd say, yeah. And, it would affect 23 the safety evaluation.
24 MR. BEAULIEU: Well, a malfunction is a 25 failure to perform design function, previously NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
246 1 evaluated. The guide says, malfunctions are 2 previously evaluated, generally single failure.
3 You go to the chapter and it says, oh, a 4 single failure? It has the failure modes. It 5 typically has failure modes and affects analysis. It 6 says, hey, single failure? What if you lose one 7 train? The results of that are -- it says, you can 8 withstand that because you have a redundant train at 9 the 100 percent capacity.
10 So, that's what's described in the NSAR.
11 And now, you have a different result, a loss of a 12 safety system. And, it is printed in the accident 13 analysis. And, it might not be described in the 14 safety analysis.
15 NEI 9607 doesn't -- also mentions that.
16 It might not be specifically described in the safety 17 analysis, but it is discussed as like, single failure 18 of a system, as a system description.
19 And, that's the safety issue is a loss of 20 an entire safety system without NRC approval. It's 21 the same as -- that's not -- a loss of an entire 22 safety system is not a different result as it just --
23 it doesn't pass the straight base test.
24 MEMBER RICCARDELLA: But that's what I 25 assumed they meant when they said they were going to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
247 1 consider that failure as it cascades up and has an 2 effect on the -- that's what I thought.
3 MEMBER BALLINGER: Another metallurgical 4 comment. Are we getting into a situation where it's 5 my PhD against your PhD? I mean, should you guys all 6 go to Starbucks and argue this out?
7 I mean, I just don't think that there's 8 much of a difference other than this is the way it is, 9 and we think it should be different. The end result 10 is exactly the same. And so, it seems that there 11 should be a way to --
12 CHAIRMAN BROWN: No. It's not, exactly.
13 If you listen to -- you all guys are so far ahead of 14 me that I couldn't write a letter on this if I had to, 15 if I can't get it down to what the hell the darn 16 problem is to start out with, okay?
17 It's all results from evaluating common 18 cause failures as applied to digital instrumentation 19 that's installed in various systems in the plant, 20 regardless of the characteristic, whether it's a 21 protection system, which we've kind of shoved off to 22 the side, or whether it's a control system, a chiller 23 system, radiation detectors, individual little 24 pressure sensors and whatever else.
25 That's what we've gotten down to. The NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
248 1 5059 reads very clearly. It talks about malfunctions 2 of an SSC, with a different result, whatever that 3 means. But my meaning is, I have SSCs that operate a 4 certain way.
5 I put in digital stuff, it doesn't operate 6 that way anymore. It fails in a manner that we have 7 not considered. Is that -- that's kind of where --
8 did I get that right? That's where --
9 MR. ARCHAMBO: And, that's part of the 10 problem.
11 CHAIRMAN BROWN: Oh, don't give me part of 12 it. Bob, forget process. I don't care about process 13 right now.
14 MR. LEBLOND Well, you have it wrong.
15 CHAIRMAN BROWN: Well, I don't have it 16 wrong. You fail something. It fails and with a 17 different result. It doesn't say who's result. It 18 doesn't say whether it's a result on the FSAR.
19 It doesn't say whether it's a result on 20 the SSC. It doesn't say. It just says, fails with a 21 different result. Me, in my simple, I would think the 22 SSC has failed with a different result, in terms of 23 its performance.
24 MR. ARCHAMBO: And, that's where we were 25 22 years ago.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
249 1 MR. LEBLOND: That's where you were 22 2 years ago.
3 CHAIRMAN BROWN: Okay. Now, the conflict 4 here comes about in that, if you get something that's 5 got a couple of trains and you have a common cause 6 failure, and it now has -- it's failed in a different 7 manner than what's been considered throughout the 8 entire set of whatever FSAR we have, they're looking 9 at this as, okay, hold it.
10 That's a different result. Therefore, 11 bang, you fail. You've got to come in. NRC ought to 12 look at it. You're saying, King's X, that's not --
13 we're going to elevate to the next level up, look at 14 the FSAR or the safety or whatever analysis -- I get 15 confused between safety analysis and FSARs.
16 They are different things. Chapter 15 17 vice the overall FSAR. You all are elevating that in 18 my mind to the next level up and say, NRC, you don't 19 need to talk to us or we don't need to talk to you, as 20 long as the next level up analysis, whatever it is, 21 safety, FSAR, says, it doesn't matter. It's okay.
22 We're bounded by something, but we've completed our 23 analysis. That's my concept of what's going on here.
24 MR. GEIER: You still take the failure.
25 CHAIRMAN BROWN: Huh?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
250 1 MR. GEIER: You still evaluate the 2 failure. You've elevated up at this decision.
3 CHAIRMAN BROWN: No. That part of your 4 evaluation is you recognize that as a failure, but you 5 finish your evaluation and say, hold it, I move up 6 one. And you escalate everything, everything, to the 7 FSAR level. And, if that's okay, then fine.
8 We go ahead and do our job and NRC is back 9 over here doing whatever they normally do. And, 10 they'll get -- they'll find out you've changed the 11 system, whatever it is. Did I actually put this in a 12 framework that's --
13 MR. LEBLOND: And they would say, make the 14 --
15 CHAIRMAN BROWN: Hopefully, I will get a 16 transcript that I can remember this.
17 MR. LEBLOND: They would say, make the 18 decision at the component level, at the descriptive 19 words.
20 CHAIRMAN BROWN: Yeah. They want to say, 21 because you've now failed at a different mode, now 22 forget the FSAR. You have to come in and tell us what 23 you're doing, and we'll get you -- that means an LAR.
24 MR. LEBLOND: And, what we tried to say 25 is, that issue propagates everything. It propagates NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
251 1 to everything, not just digital stuff. It propagates 2 to everything.
3 CHAIRMAN BROWN: I'm not so sure. No.
4 The natures of common cause failure is, I mean --
5 MR. ARCHAMBO: No. If I have a valve 6 today, mechanical, nothing digital about it and it 7 fails open and I want to replace that with a new 8 valve, a mechanical valve, nothing digital, and that 9 new valve fails closed. It might not have any impact 10 on the plant, nothing.
11 CHAIRMAN BROWN: That's a different --
12 MR. ARCHAMBO: They would say, that's a 13 malfunction with a different result. We're saying 14 it's not. You look at -- it might be, but you've got 15 to propagate it up. Now, it may be in a non-safety 16 system that I don't care how it fails. I don't care.
17 But that would have to come in for NRC approval.
18 That's what was hashed out 22 years ago 19 because that's a different type, a different type of 20 malfunction. That was a different type. That's what 21 we're trying -- again, 22 years ago, there was a group 22 of people that hashed all this out.
23 Pete was in there. And, we're going back 24 to 22 years and we're throwing everything out the 25 window that was discussed 22 years ago and resolved.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
252 1 CHAIRMAN BROWN: You're going back to the 2 slide where it talked about the words in 5059 were 3 changed for the type?
4 MR. ARCHAMBO: Yeah. Well, the actual --
5 yes.
6 CHAIRMAN BROWN: Well, there's still a 7 type in there on one of these.
8 MR. ARCHAMBO: Yeah. That's an accident 9 of a different type. That's a different --
10 CHAIRMAN BROWN: It's an accident of a 11 different type. Yeah. Hold it. I'm just trying to --
12 this conversation has been so far, you know, that we 13 went through. I just had to get it down to the 14 simplistic, what are the pieces we're talking about?
15 What's the mode of failure, you know, or 16 the different result that we're talking about? And 17 now, you're considering it, elevating it. They're 18 considering stop right here. Do not pass go. Don't 19 collect $200. Send us a letter.
20 MEMBER BALLINGER: Let me pull the string 21 a little bit further then. Let's say you get a 22 different result. You have to come in for a license 23 amendment.
24 CHAIRMAN BROWN: Yeah.
25 MEMBER BALLINGER: As part of the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
253 1 evaluation for the license amendment, is that more 2 like to end up the next level up, satisfying a 3 criterion, just like they're saying?
4 CHAIRMAN BROWN: They could come in with 5 a license amendment and NRC says fine, go ahead and do 6 it. That's exactly what the process would be. Or the 7 NRC could come back and say, King's X.
8 Naughty, naughty, mustn't do. That's 9 another foul on you and you can't do it. I'm sorry.
10 I had to have a little humor in here. I was just --
11 my brain's frying right now.
12 MEMBER BLEY: You missed some of it.
13 CHAIRMAN BROWN: Well, I wasn't going to 14 do that to you because, you know, you've got to help 15 me. I just wanted to make sure I had it characterized 16 right because I got lost in all of the elevated 17 discourse with all of these other rules. I had to 18 take it down to some simple, very simple-minded 19 concept, that an old guy like me can comprehend.
20 MR. LEBLOND: You could summarize it by 21 saying, if I changed the words of the FSAR, then I 22 have to ask for amendment.
23 CHAIRMAN BROWN: You've lost me on that.
24 What do you mean if I --
25 MR. LEBLOND: If I change the description NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
254 1 of a failed component in the FSAR, then I need an 2 amendment.
3 CHAIRMAN BROWN: Unless the description is 4 not in the FSAR.
5 MR. LEBLOND: Unless -- well, that's --
6 you don't need an application then. But now you can 7 see, with that uncertainty, what you --
8 CHAIRMAN BROWN: I understand your 9 conundrum.
10 MEMBER BLEY: Can I ask you folks 11 something completely different?
12 CHAIRMAN BROWN: Hold it. Before you ask 13 that, are we stopped on this subject?
14 MEMBER BLEY: I think they're done with 15 it. I don't know if we are.
16 MEMBER REMPE: I think there -- I get the 17 point that there's a financial concern, is why you're 18 up here talking about this. But do you -- I think I'm 19 hearing from the staff, and it's just a legal thing, 20 that if what you're proposing is inconsistent with the 21 existing words in the regulation and you understand 22 their conundrum that it's hard to regulate in 23 something that's not quite consistent with the 24 regulation.
25 Is that also part of this puzzle that NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
255 1 we're gather today? I mean, they cited 10 C.F.R. 5034 2 and other places, not just 5069. And, it sounds like 3 that they're struggling with how we'd implement this 4 if we go with what NEI says.
5 MR. LEBLOND: A regulatory guide is an 6 approved way to meet a regulation. So, if we throw 7 out previous Reg Guides, well then, it's anarchy.
8 Then, you're back to reading the rule regulation. And 9 the rule regulations, in general -- that's why you 10 have guidance.
11 That's why you've got consensus standards.
12 And so, that's -- so on the bottom of every regulatory 13 guide, it says, this is an approved way to meet a 14 regulation. And we're following that. We've taken 15 that seriously.
16 MEMBER REMPE: But you don't think they 17 have a valid point at all?
18 MR. LEBLOND: What?
19 MEMBER REMPE: I think you're saying they 20 don't have a valid point at all.
21 MR. LEBLOND: No. They don't. I don't 22 think so at all and I think that we are unwinding the 23 clock.
24 MEMBER BLEY: And they've anchored back to 25 Reg Guides 1187 and 1186.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
256 1 MEMBER REMPE: And 1186 just basically 2 endorses this NEI document.
3 MEMBER BLEY: And staff is referencing 4 other things so they're kind of at loggerheads, at 5 this point.
6 MEMBER REMPE: They definitely are at 7 loggerheads, but I just -- I'm trying to understand if 8 there's any way for compromise.
9 MEMBER SKILLMAN: Let me make a comment, 10 Charlie. Hey, Charlie, let me make a comment. When 11 you say they're winding back the clock, where were you 12 22 years ago.
13 MEMBER BLEY: I was working on this.
14 MEMBER SKILLMAN: Twenty-two years ago, I 15 was Director at TMI. And, my colleagues and I were 16 required, under oath and affidavit to sign a 5054F 17 letter to confirm that TMI 1 was in conformance with 18 its FSAR.
19 That meant, under 50.9, go to jail if 20 you're lying. That's what that meant. And the reason 21 the NRC required that is because, prior to that time, 22 well-meaning people, like us, were doing 5059s and, 23 many times, doing them incorrectly.
24 And like, 1,000 paper cuts can kill you, 25 little by little, the design and licensing basis of NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
257 1 the plants, all of the plants, were departing from one 2 another. And the NRC said, time out. If you don't 3 want to go to jail, conform your FSAR.
4 You've come the 22 years ago so many 5 times, the one stunning event that was in that time 6 period was the requirement for injury to sign the 5054 7 F letters, for a good reason.
8 When you say, they're winding back the 9 clock, I would say, they're saying, keep your 10 licensing and design bases certified. And, when we 11 say -- and I'm speaking for the staff, but I'm 12 speaking for Dick Skillman over 50-something years.
13 When they're saying SSCs, they're saying, 14 be certain that your knowledge of the detail is 15 sufficient so that when you do make that change, that 16 change is consistent with the overall safety envelope 17 of the plant. I think that that's all that they're 18 saying.
19 MR. LEBLOND: We agree with that. We 20 don't disagree with that.
21 MR. ARCHAMBO: No disagreement.
22 MEMBER SKILLMAN: So, I'm kind of with 23 Ron. What's the fuss? Why can't this be aligned in 24 a way that you are successful, and the staff is 25 successful? Because you want the same thing. And NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
258 1 when you say, if there's a pushback on the wording, 2 industry's going to back away, I would say, I don't 3 think so because there are smart people in industry 4 who will say, as long as we understand, they are 5 requiring that the configuration management program 6 clearly identifies the change.
7 Then, I would think the NRC staff would be 8 satisfied. So, it seems to me that the two teams are 9 so close together, that there should not be a whole 10 lot more effort to heal this. But as you speak, it 11 sounds like it's a colossal wall.
12 MR. LEBLOND: Well, I think it is.
13 MEMBER SKILLMAN: It doesn't need to be.
14 CHAIRMAN BROWN: Quite frankly -- and this 15 is my opinion now, Dennis, based on -- I really hate, 16 at this stage, to sit here and try to put -- to have 17 us frame this discussion down one side and the other.
18 And then, say here -- put them together 19 and say, whatever. I would rather have NRC and 20 industry put together the one, two, three, and why 21 they can't come together. In other words, what is the 22 logjam? What is the thing that we lose? And so, they 23 both -- and so, we understand that.
24 MEMBER BLEY: I would too, unless there's 25 a forcing function here that's going to toss it up for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
259 1 a commission decision soon that we could have some 2 impact on. And I don't know what the schedule is for 3 this going forward.
4 CHAIRMAN BROWN: Well, my point is -- my 5 reason is to try to -- I mean, literally, the one hour 6 between everybody else circling around here, I lost 7 the bubble, literally. I have lost what I have --
8 MEMBER BLEY: You would have lost it if 9 was two hours. So --
10 CHAIRMAN BROWN: You're right. I'm just 11 -- my point being is, unless we can frame it in a very 12 simple, straightforward way -- if we have to branch 13 here and then, branch there, and then branch here and 14 refer to 22-years ago.
15 And, you know, it's going to get lost.
16 The point that should be made will get lost, if you 17 can't have a nice, clear transition from the specific 18 problem. They want to stop at the different failure 19 mode, end-result failure mode of the SSC. And that 20 requires an LAR. Industry wants to stop at the safety 21 analysis if there's no impact in the plant.
22 What is the compromise? It's either you 23 agree with one or the other. The commission would 24 have to do that. But why isn't the compromise, in my 25 own mind, ensuring that the details of that are NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
260 1 embodied, somehow, in the licensing basis, the new 2 licensing basis, so it's very clear and it doesn't get 3 lost.
4 So, it's part of -- even though they don't 5 submit an LAR, there ought to be a formal attachment 6 or something that goes into the licensing basis that 7 says, this was this. It is now this. We've evaluated 8 it. It's okay, based on these analyses. That, to me 9 --
10 MR. LEBLOND: Well, the issues going to be 11 updated. That does -- I mean, all of these needs to 12 go back.
13 CHAIRMAN BROWN: But if you back years ago 14 -- like you say, if you rewind for -- is it 22 years?
15 That was luck. People had to go back and find that 16 and dig it out then because it was not consolidated.
17 MEMBER BLEY: Because 40 years ago, nobody 18 kept these up to date.
19 CHAIRMAN BROWN: Oh, absolutely.
20 MEMBER REMPE: Charlie, there's an 21 individual from the staff who's been standing back 22 there.
23 CHAIRMAN BROWN: I know. I just --
24 MEMBER REMPE: Okay.
25 CHAIRMAN BROWN: I've got eyes in the back NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
261 1 of my head. You've already told me twice. I'm not 2 going to lose Wendell. He'll be okay. I get -- all 3 I'm trying to do is to get the point across that there 4 can't be five pages-worth of convoluted discussion, 5 trying to get to the end-point game.
6 MEMBER BLEY: You're trying to write the 7 letter before we even talk about it.
8 CHAIRMAN BROWN: If I've got to write the 9 letter, it better be easy for me to write, because 10 otherwise, it's not going to get written. I couldn't 11 go through all of that dog and pony show that you guys 12 do.
13 MEMBER SUNSERI: Can I make a point?
14 CHAIRMAN BROWN: No. Let Wendell --
15 MEMBER SUNSERI: This will be about 16 whether we have a letter or not.
17 CHAIRMAN BROWN: Oh, okay. I'm sorry, 18 Wendell. You'll get a shot here.
19 MEMBER SUNSERI: Essentially, what I hear 20 the discussion is about is setting the threshold for 21 when a change goes to NRC for notification and 22 proving, before it gets implemented, right?
23 CHAIRMAN BROWN: LAR -- yeah.
24 MEMBER SUNSERI: That's what we're talking 25 about. I think I've heard it pretty clearly expressed NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
262 1 in here, that ultimately, safety is going to be 2 preserved because either the licensees will have a 3 process that is acceptable, or it will come over to 4 the NRC to determine whether it's acceptable. And, in 5 no case, will an unacceptable change be made.
6 CHAIRMAN BROWN: That's the theory.
7 MEMBER SUNSERI: So, my question then is, 8 from a nuclear safety perspective, is this an issue 9 that the ACRS should be weighing in on, because all 10 we're talking about is the threshold on when it comes 11 over to the regulator?
12 CHAIRMAN BROWN: We've got to talk about 13 that ourselves.
14 MEMBER SUNSERI: Okay.
15 CHAIRMAN BROWN: Okay, Wendell, sorry.
16 MR. MORTON: Not a problem. Wendell 17 Morton, NRC. I just wanted to get some additional 18 context because we're getting very deep into the weeds 19 here, but we need to really remember what was Appendix 20 D submitted to us for, and it was to address 21 challenges in 5059 for both screening and the 22 evaluations.
23 CHAIRMAN BROWN: For?
24 MR. MORTON: For digital modifications to 25 plants. And the screening section of Appendix D, it's NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
263 1 actually very specific on various aspects of digital 2 and how they affect the screening. It talks about 3 combining of design functions and how you do screen 4 criterion for that.
5 It talks about introduction of software 6 and saying, "Well, just because you introduce software 7 doesn't mean it's adverse, but here are some 8 additional considerations." It talks about a number 9 of other things, HSI, for example, so it's actually 10 very specific to things that are challenging about 11 digital and design and how you address them for 5059.
12 When you get to the evaluation -- and it 13 does a good job of doing that, and there's exceptions 14 or clarifications there in that section of Appendix D.
15 It's when you get to the evaluation criteria that 16 we're kind of getting beyond what is this doing here 17 for digital I&C? What specific challenges with 18 digital I&C are you addressing and how are you 19 connecting them?
20 And that's when we get into the 21 conversations about what Pete was talking about, 22 safety analysis, design basis, function. These are 23 not things specific to digital I&C, but they're being 24 discussed within a digital I&C appendix.
25 Really that's more appropriate for NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
264 1 discerning whether you should be addressing NEI 96-07 2 itself if you think these interpretations need 3 clarifications there, not in an appendix to the Reg 4 Guide on that particular --
5 MEMBER BLEY: I don't think you guys said 6 that anywhere that we've had the opportunity to --
7 (Simultaneous speaking.)
8 CHAIRMAN BROWN: In the NEI --
9 MR. MORTON: Correct.
10 CHAIRMAN BROWN: -- document, not the Reg 11 Guide.
12 MR. MORTON: Correct, because mainly these 13 are just generic 50.59 issues. These are not specific 14 to some challenge in digital I&C. So when you're 15 talking about the definition of safety analysis, well, 16 that's not necessarily digital. That's a general 17 issue no matter what design discipline you're talking 18 about.
19 That's better handled within 96-07, not in 20 an appendix to the Reg Guide for it, and that's where 21 we come to part ways with NEI on a number of these 22 different topics, and that's the beginning of where we 23 start when you're getting into debating whether you're 24 including design basis function of a subset of design 25 functions or if you're including all design functions.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
265 1 That's not a digital I&C specific 2 discussion, yet it's in Appendix D now, and we've been 3 challenged to deal with it in a generic context in 4 something that was specifically targeted toward 5 digital I&C mods.
6 I just want to make sure the ACRS has that 7 context for Appendix D that in one regard, the 8 screening section does a very good job of connecting 9 specific digital I&C challenges and how you address 10 50.59. The evaluation goes into a much broader 11 context beyond just digital. I just wanted to 12 clarify.
13 CHAIRMAN BROWN: While you're up there --
14 I don't know if you're the right one to be up there.
15 I think I heard Stephen say that they expect 16-16 to 16 have a draft out this summer?
17 MR. GEIER: Right, the replacement for 16-18 16 later this summer.
19 CHAIRMAN BROWN: And that would be a final 20 report?
21 MR. GEIER: And that would be the 22 technical information primarily on how to address CCF 23 using additional design attributes.
24 MEMBER BLEY: What's the schedule going 25 forward for this process? Does it go to the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
266 1 Commission? Where does it go and when does it go?
2 MR. MORTON: You mean with regards to the 3 technical document they're referring to?
4 MEMBER BLEY: The whole thing.
5 CHAIRMAN BROWN: You mean Appendix D?
6 MEMBER REMPE: Appendix D.
7 MEMBER BLEY: Appendix D. Start with 8 Appendix D.
9 MR. MORTON: Well, that, right, we don't 10 have a technical basis document right now for us to 11 really include in the schedule, so the review --
12 (Simultaneous speaking.)
13 MEMBER BLEY: Well, you have your own RIS 14 that you're counting on, but you have a draft Reg 15 Guide that says, "We endorse this, but we're worried 16 about the following 10 things," or whatever it is in 17 there, and one of those 10 is the one we were just 18 talking about for the last two hours. Is that about 19 ready to -- does it go to the Commission when you're 20 done with that Reg Guide?
21 MR. CARTE: So the current schedule is for 22 it to go out for public comment, a draft for public 23 comment at the end of April.
24 MEMBER BLEY: And that's the Reg Guide?
25 MR. CARTE: And that's the Reg Guide, and NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
267 1 so then there's 60 days for a comment period, and then 2 I don't know after that.
3 CHAIRMAN BROWN: Is the Reg Guide in 4 Appendix D?
5 MR. CARTE: The Reg Guide qualification in 6 Appendix D, the qualifying endorsement of Appendix D, 7 and then it has 60 days for a public comment period, 8 and then it's a little bit up in the air in terms of 9 depending on what the comments are, but if there are 10 --
11 (Simultaneous speaking.)
12 MEMBER REMPE: Well, what's going on 13 between the lawyers? We heard from them that the 14 lawyers and OGC lawyers were interacting. Is that 15 going to affect the release?
16 MR. WATERS: So this is Mike Waters.
17 First, Appendix D is not going to the Commission for 18 any type of approval. We're following our normal 19 regulatory guide update process.
20 CHAIRMAN BROWN: But the Reg Guide --
21 MR. WATERS: We're endorsing a proposed 22 guidance and taking exception. We're following that 23 process right now, so there's no plans to go to the 24 Commission on that.
25 We have noted in the past that we're NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
268 1 bordering on policy issues, much broader policy issues 2 than the ones we're trying to get at. This is a 3 broader policy type of interpretation, so it's not a 4 program for just digital I&C. It's something bigger.
5 We want to avoid that to facilitate the near-term 6 digital modifications that we need.
7 With respect to legal counsel, I'm not 8 aware of any discussions among our legal counsel and 9 their legal counsel.
10 MEMBER REMPE: Coming to any resolution, 11 so that's a surprise to you.
12 MR. WATERS: Yes.
13 MEMBER BLEY: But if you sent the Reg 14 Guide to the Commission without a SECY that laid out 15 these broader issues that they might weigh on first --
16 CHAIRMAN BROWN: No, they weren't going to 17 the Commission with the Reg Guide.
18 MEMBER BLEY: The Reg Guide doesn't get 19 approved by the Commission?
20 CHAIRMAN BROWN: No, they were just going 21 to issue it after they --
22 MEMBER BLEY: That's what I was asking.
23 CHAIRMAN BROWN: Yeah, the public comment 24 --
25 MEMBER BLEY: They just issue it?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
269 1 CHAIRMAN BROWN: Yeah, they would --
2 MEMBER BLEY: Okay.
3 CHAIRMAN BROWN: That's their normal 4 process for a Reg Guide. They're endorsing it as to 5 how to take action and that's it. It was not going to 6 the Commission.
7 MEMBER BLEY: Okay.
8 CHAIRMAN BROWN: So, but, I mean --
9 MEMBER BLEY: That's what I was trying to 10 get at.
11 CHAIRMAN BROWN: The broader issue, the 12 policy issue that you're talking about, the example 13 was he's got to put in a valve that's normally open 14 and then you take it out. You put in one that feels 15 normally closed. That's the broader policy issue in 16 terms of it got a different result. Now that's into 17 the mechanical world as well as -- it's not just CCF 18 is the point.
19 MEMBER BLEY: Where I'm hanging up, 20 Charlie, and why I suggested we might want to write 21 something is right now, this Reg Guide has the 22 exception we've been talking about for two hours.
23 CHAIRMAN BROWN: Yeah.
24 MEMBER BLEY: And at the same time it goes 25 out for public comment, there's going to be pressure NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
270 1 on multiple sides coming on the staff and the 2 Commission, I assume, because we've heard NEI's side 3 of this being that you're not going to get any 4 submittals for -- nobody's going to go in for mods, 5 digital I&C mods with it written this way. That's 6 probably an overstatement, but for the trickier ones, 7 that might be true. I'm just --
8 CHAIRMAN BROWN: Well, it's all wrapped up 9 in the integrated actions, the action plan that's all 10 about this --
11 MEMBER BLEY: Yeah, but that's --
12 CHAIRMAN BROWN: -- modernization and --
13 MEMBER BLEY: That's going slower than 14 this is going.
15 CHAIRMAN BROWN: I know. I'm well aware 16 of that. If you want, we can talk -- do you want to 17 talk about -- do you want to decide in full Committee 18 whether we're going to write a letter or do you want 19 to try to do it now?
20 MEMBER BLEY: I want to talk about it 21 during full Committee week sometime.
22 MEMBER RICCARDELLA: We can talk about it, 23 but, I mean, the subcommittee has to make a 24 recommendation.
25 CHAIRMAN BROWN: I'm just saying the NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
271 1 letter is not going to be prepared for May 1.
2 MEMBER RICCARDELLA: No, no.
3 MEMBER BLEY: There are things I don't 4 want to talk about here that I would talk about there.
5 CHAIRMAN BROWN: That's fine. I have no 6 problem with that.
7 MEMBER RICCARDELLA: I guess I have a 8 question. Under the current 50.59, the way, and 9 forget about digital I&C, and the associated NEI 96-10 07, would the example that was cited about a valve 11 that fails open versus fails closed, would that 12 require a license amendment request?
13 CHAIRMAN BROWN: This is I&C. This is not 14 common cause failure. This is a different result.
15 MEMBER RICCARDELLA: Yeah, would that 16 automatically require a license --
17 (Simultaneous speaking.)
18 MR. BEAULIEU: The guidance is very clear.
19 A new failure mode is not a different result. It's 20 explicit.
21 MEMBER RICCARDELLA: Okay.
22 CHAIRMAN BROWN: So that was a red herring 23 that you just threw out there?
24 MR. BEAULIEU: Yes, that's right.
25 CHAIRMAN BROWN: Okay, I just wanted to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
272 1 make sure.
2 PARTICIPANT: Well, no, no, no.
3 CHAIRMAN BROWN: Well, that's what he's 4 saying.
5 MR. BEAULIEU: Just so it's clear, it's a 6 different failure mode. The valve fails open and now 7 it can fail open or closed. That's a different 8 failure mode. Under the old 50.59 prior to 1999, that 9 used to be a license amendment, but now no longer. It 10 only matters whether the result of that new failure 11 mode is bounded --
12 CHAIRMAN BROWN: Failure mode.
13 MR. BEAULIEU: -- is bounded, is a 14 different result.
15 MEMBER RICCARDELLA: Bounded by what?
16 MR. BEAULIEU: By any previously evaluated 17 in the FSAR.
18 MEMBER RICCARDELLA: Okay.
19 MR. BEAULIEU: Anywhere in the FSAR.
20 MEMBER RICCARDELLA: So then why is that 21 any different than what they're proposing?
22 PARTICIPANT: It's not any different.
23 PARTICIPANT: It's not.
24 MEMBER RICCARDELLA: Why is that different 25 than what they're proposing?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
273 1 MR. BEAULIEU: Well, they want to say, 2 "Oh, it's only accident analysis evaluation."
3 CHAIRMAN BROWN: No, no, no, no, step 4 back.
5 MR. BEAULIEU: Okay.
6 CHAIRMAN BROWN: Why is a CCF that has a 7 different result, which is a different mode of 8 failure, if all of the trains fail as opposed to just 9 one train, different than having a valve fail closed 10 as opposed to open?
11 MR. BEAULIEU: Oh, because a different 12 result. Before, it's a failure of one system, and now 13 -- and that's what's analyzed, the failure of one 14 system. Now it's a failure of the entire safety 15 system. That function and that safety system is gone.
16 That is a -- that has not been evaluated --
17 (Simultaneous speaking.)
18 CHAIRMAN BROWN: Well, no, I'll take issue 19 with that, okay, because you could have a four train 20 system.
21 MR. BEAULIEU: Right.
22 CHAIRMAN BROWN: It might not be a 23 protection system. It could be some other system 24 where you really wanted it, and because of the common 25 cause failure, you could fail two trains as opposed to NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
274 1 four.
2 MR. BEAULIEU: Yeah --
3 (Simultaneous speaking.)
4 CHAIRMAN BROWN: Because the -- well, no, 5 no, they operate asynchronously. Is the data all the 6 same? The specific things that cause -- if you're 7 talking about a computer getting confused, they do get 8 confused, okay, particularly the platforms can 9 depending on the nature and how they're programmed, 10 but the idea that all four of those are going to get 11 confused at the same time, I don't know. If that's 12 the likelihood --
13 (Simultaneous speaking.)
14 MR. BEAULIEU: The technical argument.
15 Yeah, that's the --
16 CHAIRMAN BROWN: It's a likelihood or 17 possibility.
18 MR. BEAULIEU: Bingo, it's a qualitative 19 assessment.
20 CHAIRMAN BROWN: It's a qualitative 21 assessment, therefore --
22 (Simultaneous speaking.)
23 CHAIRMAN BROWN: My point being is that 24 common cause failure sounds like a similar 25 circumstance to a mode of failure, not necessarily a NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
275 1 fatal failure, but similar to the mode that a valve 2 fails. If you're trying to make the mechanical versus 3 a digital electronics type of comparison, then that's 4 the way I would broach that argument.
5 I mean, you could make the argument that 6 there's a lot of different ways --
7 MR. BEAULIEU: Yeah.
8 CHAIRMAN BROWN: -- to assume whether you 9 should have different software and different piece 10 parts within even a reactor protection system.
11 MR. BEAULIEU: Yeah, the only mechanical 12 type equivalent would be a cross connect between two 13 mechanical systems where a single failure could 14 disable both trains, so that's this kind of thing.
15 You're right. A different failure mode --
16 CHAIRMAN BROWN: Well, I don't know. If 17 you had a bunch of relief valves, and you had replaced 18 them all, and now the new failure, there was a common 19 cause failure mode for those, and why couldn't they 20 all fail at the same time if it's another design that 21 hadn't been used before?
22 MEMBER BLEY: Or if you go back 50 years, 23 scram relays.
24 CHAIRMAN BROWN: Circuit breakers, scram 25 breakers.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
276 1 (Simultaneous speaking.)
2 MEMBER BLEY: -- had a whole set of them 3 at a whole plant in German all melted.
4 CHAIRMAN BROWN: Oh, is that right?
5 MEMBER BLEY: Yeah.
6 MR. LEBLOND: Just the facts that Neil 7 gave started with the premise that says, "I had a 8 valve that used to fail open. Now it fails shut."
9 The words changed. It's a different result.
10 So the question now begs do you make the 11 decision based upon that description or some other 12 functional level? That's the issue. So you want a 13 simple one sentence statement, that's it.
14 So if the result of a description of a 15 component failure changes, does that mean a different 16 result or do you drive that impact up to some higher 17 level like you do the other criteria?
18 MR. ARCHAMBO: That's correct. That's it.
19 No red herring.
20 CHAIRMAN BROWN: No, I understand that, 21 but I was just trying to compare the valve 22 circumstance relative to what we perceive a CCF is 23 because all you hear about is common cause failure of 24 software.
25 MR. ARCHAMBO: Right, and let's say this NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
277 1 --
2 CHAIRMAN BROWN: I'm talking about in the 3 digital common cause failure world. I mean, that's 4 the software. That's the one we're working on right 5 now. Of course there's common cause failures in other 6 worlds.
7 MR. LEBLOND: Now you can see why we were 8 so worried because that logic applies to everything, 9 not just common cause.
10 MR. ARCHAMBO: Yeah, but say that valve 11 has a design function that's spelled out in the FSAR.
12 It has a design function. It doesn't have to be 13 safety related. It doesn't have to be train related, 14 non-safety, it has a design function, and the 15 description in the FSAR said this valve fails open.
16 Just because I put a description in there, 17 that's just the way it failed. Somebody decided to 18 write that description in there and now I want to put 19 one in because that one --
20 CHAIRMAN BROWN: We're trying to help you, 21 by the way.
22 MR. ARCHAMBO: I understand. I'm just 23 trying to drive this home.
24 CHAIRMAN BROWN: We're trying to come up 25 with the right result and trying to help in at least NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
278 1 making a decision. That's all.
2 MR. ARCHAMBO: But if I put one in there 3 that fails closed, that's a malfunction, failure to 4 perform a design function perhaps, could be, with a 5 different result.
6 MEMBER BLEY: You know, I think this issue 7 of level playing field, although that's not the way I 8 would have put it, that a plant that goes to the 9 effort to give more description has to do more when it 10 comes to making a change is --
11 CHAIRMAN BROWN: Gets penalized.
12 MEMBER BLEY: -- is counter-safety. It's 13 not helpful. Somehow dealing with that is important.
14 CHAIRMAN BROWN: It's not counter-safety.
15 It's just harder to make changes.
16 MEMBER BLEY: Well, it is for the plant 17 that doesn't have all of that detail, you know, they 18 don't even have to consider whether --
19 CHAIRMAN BROWN: From that standpoint, 20 it's a counter-safety.
21 MEMBER BLEY: Yeah.
22 CHAIRMAN BROWN: Potentially a counter-23 safety.
24 MEMBER BLEY: Potentially.
25 CHAIRMAN BROWN: I believe this dead horse NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
279 1 doesn't have any legs, skin, or --
2 (Simultaneous speaking.)
3 PARTICIPANT: It's still alive. It's 4 still alive.
5 MEMBER BLEY: I did want to ask these 6 gentlemen --
7 CHAIRMAN BROWN: Are you going to shift 8 subjects? Are we done with this now for right now and 9 we'll discuss this at the open meeting?
10 MEMBER BLEY: That's what I would 11 recommend.
12 CHAIRMAN BROWN: I'm just glad I've got a 13 transcript because I know have at least a half an hour 14 bubble when I may even understand what I said.
15 MEMBER BLEY: Given the extreme discussion 16 for the last couple of hours, clearly you've read the 17 Reg Guide, the draft.
18 PARTICIPANT: No, we haven't.
19 MEMBER BLEY: Oh, you haven't seen it?
20 It's not out?
21 CHAIRMAN BROWN: No, it's not up here.
22 MEMBER BLEY: We're the only ones who have 23 seen it? Well, now it's public, right? No, it's 24 still not public?
25 MR. LEBLOND: No, we haven't seen it.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
280 1 CHAIRMAN BROWN: It's with OGC, right? Is 2 it with OGC?
3 MEMBER BLEY: Well, then the question I 4 was going to ask you is irrelevant, so.
5 MR. LEBLOND: We can give an irrelevant 6 answer.
7 CHAIRMAN BROWN: I have not asked any 8 questions on one point. I had a couple of questions 9 myself on one point, but it's really -- this has been 10 sucked up in this black hole.
11 MEMBER RICCARDELLA: Will it be public 12 soon, by the time of the full meeting?
13 CHAIRMAN BROWN: Yeah, they're going out 14 for public comments at the end of April, and our full 15 committee meeting is the first week, so we'll have --
16 just let us know if it doesn't go out so we don't --
17 but you're going to be at the meeting. You're going 18 to give us --
19 MEMBER RICCARDELLA: Well, we have it.
20 The question is will they have it?
21 MEMBER REMPE: Well, what we have though 22 is going to be different than what it currently exists 23 at.
24 CHAIRMAN BROWN: Information meeting still 25 for the full committee meeting?
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
281 1 MEMBER BLEY: Everybody who cares is 2 probably here, well, not quite.
3 CHAIRMAN BROWN: Yeah, but we still have 4 other people that have to vote on something by the 5 time we --
6 MEMBER BLEY: I would think the only 7 reason we would have the full committee brief is if 8 there's any possibility we would want to write 9 something.
10 MS. WEAVER: This FRN has already gone 11 out.
12 MEMBER RICCARDELLA: I'm sorry?
13 CHAIRMAN BROWN: The FRN has already gone 14 out she says.
15 MEMBER BLEY: That's the reason why we'd 16 do it.
17 MS. WEAVER: You could cancel.
18 MEMBER RICCARDELLA: No, I think we 19 should. I'd like to hear it. I'd like to get --
20 CHAIRMAN BROWN: Well, we've got, what --
21 MEMBER REMPE: But again, they're going to 22 be revising what we've seen for the Reg Guide, so 23 there will be a new one come out the week before, and 24 so it would behoove us all to take a look at that 25 document so some of our questions can be focused.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
282 1 MEMBER BLEY: Yeah, that would be nice if 2 we got it a week ahead. Even if we don't though, it's 3 a small enough document. If the staff could walk us 4 through the changes, that would be helpful.
5 MR. WATERS: We'll be happy to give you 6 the document as soon as it's publicly released, and to 7 the extent practical, we'll step you through any 8 substantive changes between that version and the 9 version you have right now.
10 CHAIRMAN BROWN: Okay, how do we get NEI's 11 position in the full committee? Do we have them --
12 MEMBER BLEY: They could come and make 13 comments after they hear it for the first time, right?
14 CHAIRMAN BROWN: You mean -- well, they 15 have already heard it for the first time with us.
16 They've heard it once.
17 MEMBER BLEY: Well, they haven't read the 18 whole Reg Guide.
19 CHAIRMAN BROWN: Oh, oh, from a Reg Guide, 20 but, I mean, the issue is not -- you all haven't 21 actually seen the words in the Reg Guide yet.
22 MR. GEIER: It's not a new issue. This 23 has been discussed, you know, for several months.
24 CHAIRMAN BROWN: Well, the issue is in a 25 separate letter of comments.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
283 1 MR. GEIER: But we haven't seen the actual 2 write up then, yes.
3 CHAIRMAN BROWN: Was that letter with the 4 comments, was that public?
5 MS. WEAVER: Yes, it was, and I sent it to 6 everyone.
7 CHAIRMAN BROWN: But not the Reg Guide?
8 I know, I've got it.
9 MS. WEAVER: Yes.
10 CHAIRMAN BROWN: It's in my little brain 11 thing here.
12 MS. WEAVER: Yes.
13 MR. GEIER: Yes, there's a letter that was 14 written in December, mid-December that we have where 15 they described their position.
16 CHAIRMAN BROWN: Mine doesn't have a date 17 on it, so I have no idea when it was written.
18 MR. GEIER: It was December 15, 14, 15, 19 something like that.
20 CHAIRMAN BROWN: Okay, where you all 21 discussed this different --
22 PARTICIPANT: Summarized.
23 CHAIRMAN BROWN: Summarized comments and 24 the major disagreement, right, that was public?
25 PARTICIPANT: That's correct, yes.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
284 1 CHAIRMAN BROWN: So they have that?
2 PARTICIPANT: Yes.
3 MR. GEIER: Yes, we have that.
4 MEMBER RICCARDELLA: This isn't going to 5 the Commission, so if we were to write a letter, it 6 wouldn't be to the Chairman. It would be to the EDO, 7 right?
8 MEMBER BLEY: Let's talk about that.
9 CHAIRMAN BROWN: The full committee, 10 right? We'll make a decision on that later. You 11 don't have to decide that now.
12 MEMBER RICCARDELLA: No, I just --
13 CHAIRMAN BROWN: It sounds like --
14 MEMBER RICCARDELLA: For my edification --
15 (Simultaneous speaking.)
16 CHAIRMAN BROWN: -- for a letter. It's 17 just a matter of when we --
18 MEMBER BLEY: I think it would -- most 19 likely, we would write a letter after public comments 20 if we wrote one at all.
21 MEMBER RICCARDELLA: Well, earlier you 22 said --
23 (Simultaneous speaking.)
24 PARTICIPANT: It would really behoove you 25 guys to get together and talk about it.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
285 1 MEMBER RICCARDELLA: -- you said you 2 wanted us to chime in before though.
3 CHAIRMAN BROWN: After our brief with the 4 full committee --
5 MEMBER BLEY: I said there are reasons why 6 we might want to.
7 (Simultaneous speaking.)
8 CHAIRMAN BROWN: We don't have to fight 9 about it.
10 MEMBER BLEY: And I think we have to talk 11 about them in closed session.
12 CHAIRMAN BROWN: No, but they haven't gone 13 after this particular --
14 MEMBER BLEY: That's just me.
15 CHAIRMAN BROWN: -- tooth and nail, I 16 mean, you know, a couple of bottles of wine or 17 whatever it is.
18 MR. GEIER: We'd be happy to get together 19 and keep discussing this, and hopefully reach a 20 resolution.
21 MR. WATERS: So part -- well, I don't want 22 to get into the meeting to be honest, but part of the 23 challenge is we have to talk in generalities. We talk 24 about valve changes. We talk about diesel generator 25 water levels. These are electronic engineers. We're NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
286 1 not mechanical engineers.
2 We have asked to have specific examples 3 where it's not feasible to demonstrate common cause 4 failures sufficiently low. That's the first way you 5 demonstrate it. What are the systems talking about?
6 What does the FSAR say? And walk us through why and 7 how you address this with a different result.
8 We have not had that conversation to my 9 knowledge, so part of our challenge is this talking in 10 generalities and making broad decisions that may have 11 different precedent setting for all systems, talking 12 about specific systems. So we're on schedule to 13 issue, kind of doing facility implementation, but it's 14 challenging.
15 And this is Mike Waters' personal view, 16 not of the NRC, that it's hard to talk in generalities 17 without talking about specific upgrades and needs or 18 criteria that becomes important in the decision.
19 Again, we focused the last year and a half 20 on the pathway to demonstrate the common cause failure 21 is sufficiently low for these digital systems and you 22 don't get to this criterion six conundrum. That's 23 what we had focused on.
24 MR. ARCHAMBO: Yeah, one thing that would 25 be helpful too is if the staff could come up with an NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
287 1 example where, using our six-step method, something 2 would fall through the cracks. That would be very 3 useful because we don't see it.
4 In fact, we think that six step method is 5 very robust and that things won't fall through the 6 cracks, so we would be interested in that feedback as 7 well.
8 We do provide some examples in Appendix D 9 that show things that go through that six-step process 10 that screen in and some that, or, I mean, need a 11 license amendment request and some that don't, but it 12 would be very helpful to see something of the 13 converse.
14 MEMBER RICCARDELLA: You're using the word 15 screen, but --
16 MR. ARCHAMBO: I'm sorry, evaluate.
17 MEMBER RICCARDELLA: Evaluate.
18 MR. ARCHAMBO: It's been a long day.
19 CHAIRMAN BROWN: Still got fuzzy logic 20 from that standpoint.
21 MEMBER REMPE: We still have people on the 22 line.
23 CHAIRMAN BROWN: Yeah, I know. We're not 24 finished yet.
25 MEMBER REMPE: Okay.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
288 1 CHAIRMAN BROWN: Okay, any other 2 diversions at this particular point? Have we finished 3 the process we've gone through? If so, the -- where 4 is Kathy? Where did she go?
5 PARTICIPANT: Are you ready to open the 6 line?
7 CHAIRMAN BROWN: Yeah, open the line and 8 we'll go to the phone first here. We have smoke 9 signals. In the age of technology, we have smoke 10 signals and Morse code on the glass window.
11 PARTICIPANT: While you're doing that, you 12 could ask for comments from the public in the room.
13 CHAIRMAN BROWN: Well, I guess I could.
14 Is there anybody in the room that would like to make 15 a comment, please? I think the answer to that is no, 16 correct? Okay, now we will wait for the phone. Is 17 there anyone on the phone line? Could you just say 18 something so we know the phone line is actually open?
19 PARTICIPANT: Yes, the phone line is open.
20 CHAIRMAN BROWN: All right, would anybody 21 that's on the line like to make a comment? Okay, 22 hearing none, Kathy, go close the phone line. One 23 more -- what have I missed? I don't think I've missed 24 anything. We'll just go around one more time, Ron, 25 we'll start with you.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
289 1 MEMBER BALLINGER: Yeah, I think there are 2 some larger issues that we need to discuss. I think 3 I'm probably reading Dennis's mind a little bit. So, 4 but I thought the discussion was very enlightening and 5 really solidified the main issues, which I think was 6 a good thing. I think that we -- I think there's 7 resolution.
8 CHAIRMAN BROWN: Okay, Dick?
9 MEMBER SKILLMAN: Thank you, Charlie. I 10 want to commend the NRC staff, NEI, and Pete there for 11 your work. Thank you very much. As contested as this 12 might seem, this is how the ACRS completes its work, 13 so thank you.
14 CHAIRMAN BROWN: Matt?
15 MEMBER SUNSERI: I'd like to also extend 16 appreciation for both the staff and the industry for 17 the unvarnished discussion that we had today. Thank 18 you.
19 CHAIRMAN BROWN: Pete?
20 MEMBER RICCARDELLA: You know, I guess I'm 21 trying to see both sides of the picture, you know, and 22 when you say well, we're going to restrict the 23 evaluation on item six, question six, to things that 24 are considered in the safety evaluation, which I 25 interpret as Chapter 15, that, to me, is a very NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
290 1 prescriptive set of events and accidents that may or 2 may not be relevant.
3 But I wonder why question five doesn't 4 then come into play where it says, "Create the 5 possibility of an accident of a different type than 6 previously evaluated," and so it seems to me you've 7 got both bases covered if you consider the question 8 five along with question six.
9 CHAIRMAN BROWN: Well, it's FSAR. Is 10 there a difference between the FSAR and a safety 11 analysis?
12 (Simultaneous speaking.)
13 MEMBER BLEY: FSAR is final safety 14 analysis report.
15 CHAIRMAN BROWN: And that's safety 16 analysis Chapter 15, right?
17 MEMBER BLEY: Well, that's part of the 18 FSAR, but --
19 CHAIRMAN BROWN: I know it's part of the 20 FSAR, but it's a part within the FSAR that's the 21 Chapter 15 accident analysis, the design basis 22 analysis.
23 MEMBER BLEY: Yes.
24 CHAIRMAN BROWN: Okay. Okay, so I'm not 25 sure what you meant by restricted. I thought it was NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
291 1 within the FSAR based on the way the words read in 2 item six, in all of the items.
3 MEMBER RICCARDELLA: Yeah, but industry is 4 telling us, well, no, they want to interpret that as 5 the safety evaluation within the FSAR and not the 6 entire FSAR.
7 CHAIRMAN BROWN: So I missed the point.
8 MEMBER RICCARDELLA: Oh, yeah.
9 CHAIRMAN BROWN: That's another fine point 10 that I missed.
11 MEMBER RICCARDELLA: It's not very fine.
12 CHAIRMAN BROWN: You mean just on the 13 safety analysis, the Chapter 15 analysis?
14 MEMBER RICCARDELLA: That's the big issue, 15 yeah, but I guess I'm less concerned about that when 16 I consider questions five and six together.
17 MEMBER MARCH-LEUBA: Can we follow this 18 discussion because I think you want to expand it? I 19 mean, what you're saying is their language says if 20 it's in the FSAR, if it is written down in the FSAR, 21 you should consider.
22 What you're saying is if it was considered 23 during the analysis to generate the FSAR, then we'll 24 consider that analysis, which is always larger than 25 what was documented in the FSAR. That's where we go NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
292 1 into more detail or less detail.
2 MR. LEBLOND: Well, our position is that 3 the decision is made that the safety analysis was 4 Chapters 15, 6, and sometimes 3.
5 MEMBER MARCH-LEUBA: But I thought you 6 said --
7 MR. LEBLOND: So that's our position is in 8 the safety analysis contained in the FSAR.
9 MEMBER MARCH-LEUBA: But I thought you 10 said that some plants don't document all of their 11 safety analysis and that some have more detail than 12 others, and therefore by saying the safety analysis, 13 you are doing more than was documented.
14 MR. LEBLOND: Some plants have much more 15 detail beyond that. Everybody is pretty consistent on 16 the safety analysis in 6, 15, and 3. Everybody is 17 pretty consistent with that. That's why you level the 18 playing field by using that subset.
19 Where the big variation comes in is this 20 plant will have a description of maybe a failed 21 component. This one has five tables of it.
22 MEMBER MARCH-LEUBA: Right, but you want 23 to restrict it into just Chapter 15?
24 MR. LEBLOND: And 6 and 3, right.
25 MEMBER MARCH-LEUBA: Then I'm with Pete.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
293 1 I don't like it.
2 MEMBER RICCARDELLA: But again, what I 3 said was consider that in conjunction with item five, 4 question five.
5 MR. ARCHAMBO: Now, just to be clear, 6 we're not saying that we want to restrict it to this 7 point. We want to follow the rule, the guidance as it 8 was applied 22 years ago. We're not changing that, so 9 this is nothing new. You know, we're not trying to 10 restrict it. This is nothing new for us in Appendix 11 D.
12 CHAIRMAN BROWN: Well, but the rule says 13 FSAR. It doesn't say safety analysis. The safety 14 analysis is part of the FSAR. So I don't know what 15 was 22 years ago, but it was the FSAR then. All you 16 did was change different type to different result.
17 MR. ARCHAMBO: That's why this argument 18 was --
19 CHAIRMAN BROWN: You keep saying 22 years 20 ago, but it says FSAR, not, and based on the 21 discussion, safety analysis. It was a subset.
22 MEMBER BLEY: Charlie, the first stuff 23 they did was laid out, the reasoning language that 24 went with the rule change, at least the part of it 25 they wanted us to see, that kind of defined what those NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
294 1 words meant, and those definitions were analysis.
2 MR. LEBLOND: And we used a Reg Guide 3 which implements the rule, so that's what a Reg Guide 4 is for is to implement a rule.
5 CHAIRMAN BROWN: No, I know sort of what 6 we do with Reg Guides. I just don't have the 40 years 7 of background in the commercial world. Joy?
8 MEMBER REMPE: So I want to also express 9 my appreciation to the staff and NEI for the 10 presentations and explanations today. I'm glad I was 11 here for the entire meeting, but as I think about 12 what's going to happen at the full committee meeting, 13 I'm guessing you'll have a couple of hours at most for 14 this discussion and it would be good for us to think 15 about how to best focus that presentation.
16 I'm guessing since the staff is going to 17 be having a new Reg Guide, but they said earlier they 18 may focus on the remaining exemptions or exceptions to 19 the guidance in your document, but I think you might 20 want to ask members, especially I think Dennis may 21 have some suggestions on what should be focused on in 22 those couple of hours.
23 CHAIRMAN BROWN: Yeah, obviously we can't 24 do the whole 46 pages of their presentation. It needs 25 to be reduced in terms of what gets presented. I NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
295 1 really want them to focus on this particular issue in 2 one way and frame both sides of the argument, not just 3 their own.
4 MEMBER BALLINGER: Might we consider 5 delaying the full committee meeting by a month or so 6 because the Reg Guide is going to be coming out almost 7 coincidentally with the full committee meeting?
8 MEMBER REMPE: I think that would be good.
9 MEMBER BALLINGER: And somebody -- you 10 know, maybe there's a little bit of time to read 11 things and --
12 MEMBER RICCARDELLA: That's a good 13 suggestion, Ron. So you're suggesting we postpone it 14 until June then?
15 MEMBER BALLINGER: Yeah, I mean, I don't 16 know what the time is that we have allocated during 17 the June full committee meeting.
18 CHAIRMAN BROWN: But to me, we could do 19 that. I mean, this is a Reg Guide that wasn't going 20 to the Commission. They're going forward. They're 21 going to go get public comments. It's going to be a 22 60-day public comment, so that's May through June.
23 So we would at least have the documents 24 and give them time to tailor your presentation, and 25 you're not dependent upon the Commission to validate NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
296 1 this unless we write a letter which says something to 2 the contrary.
3 MEMBER BALLINGER: But it also gives us a 4 chance to have our discussion, separate discussion.
5 MEMBER BLEY: I hope we have a discussion 6 the first week of May.
7 MEMBER BALLINGER: But it could be a 8 separate discussion, not a presentation.
9 CHAIRMAN BROWN: You mean the staff 10 discussion as well? We ought to have an internal 11 discussion. No, no, I think we can do that. It's a 12 matter of -- we've got time to do it. There's only 13 one letter right now. We'll put the letter off for 14 this. It would be after the June meeting if we were 15 going to do anything.
16 MEMBER RICCARDELLA: For those of you in 17 the ACRS in the room, that transformation topic has 18 been postponed until October, so we won't be spending 19 a lot of time on that.
20 MEMBER BLEY: Then I guess what I did with 21 my airplane will work just fine.
22 MEMBER BALLINGER: October of which year?
23 CHAIRMAN BROWN: Okay, let's finish.
24 Dennis, do you have anything to add at the end here?
25 MEMBER BLEY: Nothing more to add. Thanks NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
297 1 for the whole day. I think it was very useful.
2 CHAIRMAN BROWN: Jose?
3 MEMBER MARCH-LEUBA: Yeah, I had not been 4 able to prepare for this meeting the way I would like 5 to, but I have a couple of concerns. Number one is 6 what Peter was talking about. If a failure of the new 7 control system produces a different type of transient, 8 it should really trigger without having to elevate it 9 to analysis.
10 And the second concern I have is that 11 maybe the criteria, the screening criteria, we're 12 setting it so high that we would only ask for LARs for 13 things that we know we're not going to get because 14 they do affect the safety of the reactor. So if it 15 triggers the criteria, you're not even bothered to do 16 it because you're not going to get it, so we are not 17 going to get anything.
18 Maybe we're setting the criteria -- that's 19 what I'm thinking right here. Maybe we're setting the 20 criteria so high that it will only trigger when you 21 shouldn't be doing it anyway, and think about it.
22 CHAIRMAN BROWN: Well, a new protection 23 system or safeguard system to go all digital would 24 obviously be an LAR and it would come in for full 25 review, so.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
298 1 MEMBER MARCH-LEUBA: And it would probably 2 not be approved.
3 CHAIRMAN BROWN: I wouldn't say -- I 4 disagree with that. We've approved -- they've 5 approved several in the new plants, so there's no 6 reason why it couldn't be, and Diablo Canyon was 7 approved.
8 MEMBER MARCH-LEUBA: Yeah, but it would 9 have to have --
10 (Simultaneous speaking.)
11 CHAIRMAN BROWN: Oconee has a new system, 12 so.
13 MEMBER RICCARDELLA: It's just that it 14 couldn't be done on the 50.59.
15 CHAIRMAN BROWN: It won't be a 50.59. It 16 will be done as a --
17 MEMBER MARCH-LEUBA: I'm not sure that if 18 you have a completely new digital protection system 19 that is properly diverse, redundant, and with all of 20 the bells and whistles that it would trigger the 21 criteria.
22 MEMBER RICCARDELLA: There's a --
23 CHAIRMAN BROWN: An assessment.
24 MEMBER RICCARDELLA: Something that's done 25 very early in the game that if it's a plant protection NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
299 1 system or any one of two or three categories --
2 CHAIRMAN BROWN: It's got to come in.
3 That's already --
4 MEMBER RICCARDELLA: -- that automatically 5 doesn't get a 50.59.
6 CHAIRMAN BROWN: They don't even have to 7 go through the historic bands. They might try, but --
8 MR. LEBLOND: It would be L06 criteria.
9 MR. ARCHAMBO: Yeah, what happens with a 10 protection system, for a digital protection system is 11 the new ones have cross channel communications, and 12 once you do cross channel communications, you've 13 reduced independence, and once you've reduced 14 independence --
15 (Simultaneous speaking.)
16 CHAIRMAN BROWN: Not supposed to be 17 anywhere but in the voting units, so.
18 MR. GEIER: And it's acknowledged that 19 it's going to require an LAR, and that's why ISG 6 was 20 written and approved so that it streamlined the LAR 21 process.
22 It actually helps the utilities have more 23 certainty and predictability before they make the 24 major investment and purchase the system. So that's, 25 quite honestly, for fleets that want to go that way, NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
300 1 stations that want to go that way, they've got a path.
2 CHAIRMAN BROWN: We started that nine 3 years ago.
4 MR. GEIER: And we know two to three 5 fleets right now are considering initiating projects 6 for that.
7 CHAIRMAN BROWN: Good, that's a good plan 8 if it was through everybody else. Yeah, I just want 9 to echo this was a very robust and animated 10 discussion. I think it brought out a lot of good 11 points. I may even have a halfway decent 12 understanding if people expect me to write a letter on 13 this at some point as long as the transcript is clear.
14 I would still encourage, once this thing 15 is issued at you're out for public comment, you'd 16 really save yourself a world of hurt if you could 17 somehow get NRC on your side via some, you know, just 18 absolutely irrefutable, as you would phrase it, 19 thought processes in terms of how this should get 20 applied.
21 I tried a little bit of thought process of 22 why is CCF in this particular circumstance? Forget 23 the bigger policy, but is there a bigger policy issue, 24 and if there's examples where you can show where the 25 bigger policy issue would be impacted by this decision NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
301 1 to not accept your words because they may have a more 2 generic application.
3 Because that's the point is they have been 4 dealing, the staff has been. They're focused on the 5 digital I&C and not necessarily the whole 96.07 6 itself, and if this would impact that, then some 7 examples of where that would happen based on this 8 decision for Appendix D certainly ought to be brought 9 out. You could help your case if you had some solid 10 examples.
11 So, other than that, do you all have any 12 other comments at this thing? I think I've done that 13 before.
14 MS. WEAVER: I need to ask you.
15 CHAIRMAN BROWN: Yes?
16 MS. WEAVER: Just so I understand what the 17 subcommittee wants to do, you want to defer the May 18 meeting and have me set it up in June. Is that 19 correct?
20 MEMBER RICCARDELLA: Yes.
21 CHAIRMAN BROWN: I'll look at the chairman 22 here.
23 MEMBER RICCARDELLA: Yes, but then we do 24 want to have some discussion among ourselves probably 25 in closed session maybe.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
302 1 MS. WEAVER: We can add it to the PMP if 2 you would like.
3 MEMBER RICCARDELLA: PMP, yeah, add it to 4 PMP.
5 CHAIRMAN BROWN: Or even -- yeah, it could 6 be any time.
7 MS. WEAVER: Okay, and we have time to 8 repost the May agenda. That's why I'm asking right 9 now.
10 CHAIRMAN BROWN: Okay, yes, so we ought to 11 delete that.
12 MEMBER BLEY: Kathy, that ought to be 13 closed. It shouldn't be part of the public PMP.
14 MS. WEAVER: Okay.
15 MEMBER RICCARDELLA: And we can maybe 16 close the PMP.
17 MEMBER MARCH-LEUBA: Also the agendas from 18 June and July are getting very full with NuScale. You 19 checked it?
20 MEMBER REMPE: But there is one letter in 21 June. We checked.
22 MEMBER MARCH-LEUBA: Okay.
23 MEMBER REMPE: The rest of it's review 24 info.
25 MEMBER MARCH-LEUBA: July is the one.
NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
303 1 CHAIRMAN BROWN: Okay, and there --
2 MS. WEAVER: Okay, I'll put it in June and 3 out of May.
4 CHAIRMAN BROWN: And a potential letter in 5 June depending on how we decide. At least put it on 6 the list so we can plan on it, and then we can make a 7 decision.
8 MS. WEAVER: Okay, I will do that.
9 CHAIRMAN BROWN: Is that agreeable to the 10 committee, subcommittee? Okay, with that in mind, has 11 that answered your questions, Kathy?
12 MS. WEAVER: Yes, the staff knows.
13 CHAIRMAN BROWN: Okay, with that in mind, 14 the meeting is adjourned. Thank you very much.
15 (Whereupon, the above-entitled matter went 16 off the record at 3:43 p.m.)
17 18 19 20 21 22 23 24 25 NEAL R. GROSS COURT REPORTERS AND TRANSCRIBERS 1323 RHODE ISLAND AVE., N.W.
(202) 234-4433 WASHINGTON, D.C. 20005-3701 (202) 234-4433
NEI 96-07 Appendix D and RG 1.187 Revision 2 Chris Miller, NRR/DIRS Director Michael Waters, NRR/DE/EICB Tara Inverso, NRR/DIRS/IRGB Philip McKenna, NRR/DIRS/IRGB Wendell Morton, NRR/NRR/EICB Norbert Carte, NRR/DE/EICB David Rahn, NRR/DE/EICB David Beaulieu, NRR/DIRS/IRGB Eric Martinez, RES/DE/ICEEB Advisory Committee on Reactor Safeguards DI&C Subcommittee Briefing April 16, 2019
Purpose
- Brief ACRS on the process for evaluating and documenting digital instrumentation and controls (I&C) modifications using the 10 CFR 50.59 rule
- Progress on Digital I&C Integrated Action Plan
- Overview of 10 CFR 50.59
- NEI 96-07, Guidelines for 10 CFR 50.59 Implementation: What it is and what it does
- NEI 96-07 Appendix D, Supplemental Guidance for Application of 10 CFR 50.59 to Digital Modifications: Digital and Most Important Criteria
- Draft RG 1.187 Revision 2, Guidance for Implementation of 10 CFR 50.59, Changes, Tests and Experiments: Content and Exceptions
- Started and planned modifications by industry using RIS 2002-22, Supplement One guidance 2
Commission Direction on Digital I&C (SRM-SECY-15-0106 & SRM-SECY-16-0070)
- Develop an integrated strategy to modernize the DI&C regulatory infrastructure
- Engage stakeholders to identify common priorities, problems, and potential solutions to address them
- Focus on acceptable approaches to comply with requirements
- Technology neutral focus; guidance can be tailored if necessary
- Evaluate potential policy issues 3
IAP - Modernization Plans
- Modernization Plan (MP) #1 - Protection against Common Cause Failure
- MP #1A - Regulatory Issue Summary (RIS) 2002-22, Supplement 1, Clarification on Endorsement of NEI Guidance in Designing Digital Upgrades in Instrumentation and Control Systems
- MP #1B - Review of NEI 16-16 Guidance for Addressing Digital Common Cause Failure
- MP #1C - Implementing Commission Policy on Protection against CCF in DI&C Systems
- MP#1D - Update to BTP 7-19 for Diversity and Defense in Depth Against CCF
- MP #2 - Considering Digital Instrumentation & Controls in Accordance with 10 CFR 50.59
- MP #3 - Acceptance of Digital Equipment (Commercial Grade Dedication)
- MP #4 - Assessment for Modernization of the Instrumentation & Controls Regulatory Infrastructure
- MP #4A - ISG-06 Revision
- MP #4B - Broader Modernization Activities 4
IAP Modernization Plan #2
- The Integrated Action Plan (IAP) established the following objectives:
o Ensure there is adequate guidance within NEI 96-07 for 10 CFR 50.59 evaluations of digital I&C upgrades to:
o Reduce licensee uncertainty o Clarify the regulatory process o Ensure common understanding for the use, interpretation, and application of guidance.
o The following has been accomplished so far:
o RIS 2002-22, Supplement 1 issued on 5/31/18 o Public meetings to comment on NEI 96-07, Appendix D o Development of RG 1.187, Revision 2 o Regional inspector training for Regions 1 and 4 in December 2018.
Regions 2 and 3 will have training in June 2019 5
NEI 96-07 Appendix D and RG 1.187 Revision 2 Overview of 10 CFR 50.59 6
50.59 Rule History
- First promulgated in 1962 and modified in 1968.
- Allows licenses to make changes to the facility without prior NRC staff approval
- Must maintain acceptable levels of safety as documented in the FSAR
- Rule was reviewed for revision in 1995; issued in 1999 which increased flexibility for licensees:
- Now allows changes that only minimally increase the probability or consequences of accidents
- Nov 2000: NRC issues RG 1.187
- Endorses NEI 96-07, Rev.1, Guidelines for 10 CFR 50.59 Implementation 7
- NEI 96-07 was originally as NSAC-125, but not endorsed by NRC
- NEI 96-07, Revision 1
- Applicability
- Screening
- Evaluation Process
- Endorses NEI 96-07, Revision 1 - Provides methods that are acceptable to the NRC staff for complying with the provisions of 10 CFR 50.59
- Revision 1 to RG 1.187 will be issued based on San Onofre Generating Station (SONGS) lessons-learned. Clarifies 50.59 guidance on:
- Departures from a method of evaluation
- Accidents of a different type 8
10 CFR 50.59 Relationship to Other Licensing Processes
- Amendments to the operating license (including technical specifications) are obtained under 10 CFR 50.90
- More specific regulations apply over 10 CFR 50.59 (e.g.
quality assurance, security, emergency planning (EP) program changes under 10 CFR 50.54(a))
- Exemptions are processed IAW 10 CFR 50.12
- Maintenance activities are assessed and managed under 10 CFR 50.65
- License conditions (e.g. fire protection under GL 86-10) are controlled under the license condition and not 10 CFR 50.59 9
50.59 Process Chart 10
50.59 Evaluation Criteria
- A license may make changes in the facility as described in the FSAR (as updated), make changes in the procedures as described in the FSAR (as updated), and conduct tests and experiments in the FSAR (as updated) w/o obtaining a license amendment only if:
- A change to Technical Specifications is not required
- The change, test, or experiment does not meet any of the following criteria:
- Result in more than a minimal increase in the frequency of occurrence of an accident previously evaluated in the FSAR (50.59(c)(2)(i))
- Result in more than a minimal increase the likelihood of occurrence of malfunction of a structure, system, and component (SSC) important to safety previously evaluated in the FSAR (50.59(c)(2)(ii))
11
50.59 Evaluation Criteria
- Result in more than a minimal increase in the consequences of an accident previously evaluated in the FSAR (50.59(c)(2)(iii))
- Result in more than a minimal increase in the consequences of a malfunction of an SSC important to safety accident previously evaluated in the FSAR (50.59(c)(2)(iv))
- Create the possibility of an accident of a different type than any previously evaluated in the FSAR (50.59(c)(2)(v))
- Create the possibility for a malfunction of an SSC with a different result than any previously evaluated in the FSAR (50.59(c)(2)(vi))
- Result in a design basis limit for a fission product barrier as described in the FSAR being exceeded or altered (50.59(c)(2)(vii))
- Result in a departure from a method of evaluation described in the FSAR used in evaluating the design basis or in the safety analysis (50.59(c)(2)(viii))
12
NEI 96-07 Appendix D and RG 1.187 Revision 2 NEI 96-07, Guidelines for 10 CFR 50.59 Implementation 13
NEI 96-07, Rev. 1
- Provides guidance for implementing the revised (as of 1999) 10 CFR 50.59 rule.
- Definitions and applicability of terms for 10 CFR 50.59
- Implementation guidance
- Applicability
- Screening (Adverse or Non-Adverse)
- Evaluation Process
- Applying 10 CFR 50.59 to Compensatory Actions to Address Nonconforming or Degraded Conditions
- Disposition of 10 CFR 50.59 Evaluations
- Documentation and Reporting 14
Digital I&C 10 CFR 50.59 Guidance
- EPRI TR-102348
- Issued in 1993 to establish guidelines for digital upgrades in the context of 10 CFR 50.59
- Endorsed by NRC GL 95-02
- Use of NUMARC/EPRI Report TR-102348, Guideline on Licensing Digital Upgrades, in Determining the Acceptability of Performing Analog-to-Digital Replacements under 10 CFR 50.59
- EPRI TR-102348, Rev. 1 issued to address revised 10 CFR 50.59 rule in 1999
- Issued as NEI 01-01
- Endorsed by NRC RIS 2002-22 15
- Issued in March 2002 to help nuclear plant operators implement and license digital upgrades in a consistent, comprehensive, and predictable manner
- Guidance for important steps in the design and implementation process to ensure digital upgrade issues are adequately addressed.
- Guidance for performing the 10 CFR 50.59 evaluation for a digital modification
- Guidance for a license amendment request 16
- Industry inconsistently applying guidance in NEI 01-01 in digital upgrades
- Lack of industry guidance on the technical evaluation of common cause failures
- NRC Info Note 2010-10: Implementation of a Digital Control System Under 10 CFR 50.59 (La Salle Rod Control Management System (RCMS) Modification)
- Harris 2013 violation: SSPS control circuit boards replaced with digital complex programmable logic device (CPLD)-based boards
- NRC Letter to NEI: Summary of Concerns with NEI 01-01 dated 11/05/13 (ML13298A787) 17
Concerns with NEI 01-01
- Definitions have changed
- Guidance documents have changed (e.g., ISGs )
- Operating experience (e.g., LaSalle, Harris)
- Description of diversity and CCF
- Need to address Embedded Device RIS in examples
- NEI 01-01 contains two types of guidance
- 1. Guidance for digital modifications
- 2. Guidance for implementing 50.59
- SECY 93-087 states that CCF should always be considered while NEI 01-01 does not
- Digital modifications to non-safety systems
- Characterization unanticipated behaviors of digital systems 18
Digital I&C Mods
- What makes these different?
- Common Cause Failure (CCF) (Due to combined functions, shared communications, shared resources, and software error in redundant channels)
- Safety model of nuclear plant
- Defense in depth and redundant equipment
- Hardware: Likelihood of CCF acceptably low
- High quality standards in development and manufacture
- Physical separation of redundant equipment
- Degradation methods slow to develop (i.e. corrosion)
- Software: Special cause of single failure vulnerability
- Software resides in redundant channels of the system
RIS 2002-22, Supplement 1
- NRC issues RIS 2002-22, Supplement 1 in May 2018 to clarify RIS 2002-22
- NRC continues to endorse NEI 01-01
- RIS 2002-22, Supplement 1, clarifies guidance for preparing and documenting Qualitative Assessments
- Not for Replacement of:
- Reactor Protection System (wholesale)
- Engineered Safety Features Actuation System (wholesale)
- Modification/Replacement of the Internal Logic Portions of These Systems 20
NEI 96-07 Appendix D and RG 1.187 Revision 2 NEI 96-07 Appendix D, Supplemental Guidance for Application of 10 CFR 50.59 to Digital Modifications 21
NEI 96-07, Appendix D
- RIS 2002-22, Supplement One gives guidance on the technical aspect of digital I&C modifications, not the 50.59 process
- Appendix D gives digital I&C modification screening and evaluation guidance
- The format of Appendix D is aligned with NEI 96-07, Rev. 1 text for ease of use
- NEI 96-07, Appendix D does incorporate some RIS 2002-22, Supplement One guidance on qualitative assessments 22
NEI 96-07, Appendix D
- From April 2016 through 2017, the NRC staff and industry participated in monthly public meetings to resolve NRC comments on draft NEI 96-07, Appendix D
- In December 2017, NEI and the NRC staff mutually agreed to place the review of NEI 96-07, Appendix D on hold to dedicate resources to the issuance of RIS 2002-22, Supplement 1
- RIS 2002-22 Supplement 1 was issued on 05/31/18
- In July 2018, NEI provided an update to NEI 96-07, Appendix D
- In August 2018, the NRC staff provided a set of comprehensive comments (85 total) to NEI, and began a disciplined process for cataloging and tracking comments for resolution
- Public meetings were held with industry on 8/30/18, 9/11/18, 10/11/18, and 11/14/18 to resolve these comments. Over 90% of the comments were resolved using this process
- NEI submitted its final revision of NEI 96-07, Appendix D to the NRC on 11/30/18 and the letter requesting endorsement on 1/08/19 23
NEI 96-07, Appendix D Screening Section
- Scope of digital modifications:
- Software-related activities
- Hardware-related activities
- Human-System Interface-related activities
- To reach screen conclusion of non-adverse:
- Physical characteristics of the digital modification
- Change has limited scope
- Relatively simple digital architecture
- Limited functionality
- Can be comprehensively tested
- Engineering Evaluation Assessments
- Quality of the design process
- Single failures encompassed by existing failures of the analog device
- Has extensive operating history 24
NEI 96-07, Appendix D Screening Section
- Combination of Components/Systems and/or Functions
- Mere act of combining does not make the screen adverse
- If it causes an adverse act on the design function, then adverse
- Reductions in the redundancy, diversity, separation, or independence of a UFSAR design function screen adverse
- Human Factors Engineering Evaluation
- NEI worked closely with NRC human factors personnel on this section
- Two steps:
- Identify generic primary tasks involved
- For all primary tasks, assess if the mod negatively impacts the primary task 25
NEI 96-07, Appendix D Evaluation Section
- Guidance in sections 4.3 aligns with main body of NEI 96-07 and there is a caution that Appendix D is intended to supplement guidance in main body of NEI 96-07
- Sections 4.3.1, 4.3.2, 4.3.5, and 4.3.6 (which align with the Criterion in the evaluation paragraph of 10 CFR 50.59)
(50.59(c)(2)) discuss the use of the qualitative assessment outcome (sufficiently low or not sufficiently low) to answer the evaluation questions
- Sections 4.3.3 and 4.3.4 state that they provide no new guidance for digital modifications
- More than a minimal increase in the consequences of an accident
- More than a minimal increase in the consequences of a malfunction 26
NEI 96-07, Appendix D Evaluation Section
- Guidance in section 4.3.6 (Does the Activity Create a Possibility for a Malfunction of an SSC Important to Safety with a Different Result):
- Discussion on design basis functions
- Connection between design basis functions and safety analysis result
- Overall perspective in section 4.3.6:
- Unless the equipment would fail in a way not already evaluated in the safety analysis, there can be no malfunction of an SSC important to safety with a different results (emphasis added) 27
NEI 96-07, Appendix D Evaluation Section
- Six Step Process in Section 4.3.6
- 1. Identify the functions directly or indirectly related to the proposed modification
- 2. Identify which of the functions from Step 1 are Design Functions and/or Design Basis Functions
- 3. Determine if a new Failure Modes and Analysis (FMEA) needs to be generated
- 4. Determine if each design bases function continues to be performed/satisfied
- 5. Identify all safety analyses involved
- 6. For each safety analyses involved, compare the projected/postulated results with the previously evaluated results 28
NEI 96-07, Appendix D Evaluation Section From step 5:
- If there are no safety analyses involved, then there cannot be a change in the result of a safety analysis. Therefore, in this case, the proposed activity does NOT create the possibility for a malfunction of an SSC important to safety with a different result
- NRC Staff disagrees in that 4.3.6 should determine the impact of the SSC malfunction instead of the impact on the results of the safety analysis on the facility as a whole
- Section 4.3.6 is inconsistent with NEI 96-07, Section 4.3.2 which states: The safety analysis assumes certain design functions of SSCs in demonstrating the adequacy of design. Thus, certain design functions, while not specifically identified in the safety analysis, are credited in an indirect sense 29
NEI 96-07 Appendix D and RG 1.187 Revision 2 Draft RG 1.187 Revision 2, Guidance for Implementation of 10 CFR 50.59, changes, Test and Experiments 30
RG 1.187 Rev. 2
- Endorses NEI 96-07 Appendix D with exceptions
- Will be issued for public comment in same FRN that issues RG 1.187 Rev. 1 (Clarifications based on SONGS Lesson Learned)
- Exceptions:
- States that NRC staff considers NEI 96-07, Appendix D, to be applicable to digital modifications only and not generically applicable to the 10 CFR 50.59 process
- Basis: NEI 96-07, Rev. 1 base document provides guidance to the 10 CFR 50.59 process. This is a clarification 31
RG 1.187 Rev. 2
- Exceptions (Cont):
- Acknowledgement that Human System Interface (HSI) is now discussed in Appendix D whereas NEI 96-07 base document does not discuss HSI. The staff agrees that HSI may be screened
- This is a clarification 32
RG 1.187 Rev. 2
- Exceptions (Cont):
- That examples in Appendix D are meant to illustrate guidance provided and should not be used to derive guidance
- This is a clarification 33
RG 1.187 Rev. 2
- Exceptions (Cont):
- That NEI 01-01, Section 5, as clarified by RIS 2002-22 Supplement 1, is the only guidance the NRC has reviewed or endorsed as providing an acceptable technical basis to determine that the likelihood of software CCF is sufficiently low for the purpose of 10 CFR 50.59 evaluations
- This is a clarification 34
RG 1.187 Rev. 2
- Exceptions (Cont):
- NEI has written Appendix D such that the determination of the impact is done against the safety analysis (which they attempt to redefine as only the accident analysis), whereas, the staffs position is that the results of any malfunction previously evaluated in the UFSAR must be compared. By following the guidance as written, the staff believes that licensees can arrive at different conclusions in evaluations against 10 CFR 50.59(c)(2)(vi).
- NRC interpretation of the rule 35
RG 1.187 Rev. 2
- Exceptions (Cont):
- Examples 4-17 through 4-23 of NEI 96-07, Appendix D, use the evaluation criteria in section 4.3.6 of NEI 96-07, Appendix D, and it is possible to obtain a different result using this criteria
- From Example 4-19 of NEI 96-07, Appendix D:
- Although the software CCF likelihood was determined to be not sufficiently low, there are no safety analyses that directly or indirectly credited the design basis function or contain expected responses of the radiation monitors. Thus there cannot be a different result when comparing to a pre-existing safety analysis since none exists 36
RIS 2002-22 Supplement One The following discussion will involve qualitative assessment information from RIS 2002-22 Supplement One 37
Qualitative Assessment
- Originally discussed in NEI 01-01 (Section 4&5 and Appendices A&B), but limited guidance on how to accomplish
- RIS 2002-22, Supplement 1
- Evaluate the likelihood of failure of a proposed digital mod to accomplish designated safety function
- Evaluate the likelihood of common cause failure
- Used to support a conclusion that a proposed digital I&C modification will not result in more than a minimal increase in:
- The frequency of occurrence of accidents (50.59(c)(2)(i))
- The likelihood of occurrence of malfunctions (50.59(c)(2)(ii))
- Create the possibility of an accident of a different type (50.59(c)(2)(v))
- Create the possibility for a malfunction of an SSC with a different result (50.59(c)(2)(vi))
38
Qualitative Assessment Factors
- Design Attributes
- Can prevent or limit failures from occurring
- Focus primarily on built-in features:
- Fault detection
- Failure management schemes
- Internal redundancy
- Diagnostics within the integrated software and hardware architecture
- Can be external:
- For example: Mechanical stops or speed limiters 39
Qualitative Assessment Factors
- Typical Design Attributes
- Watchdog timers that function independent of software
- Self-testing and diagnostics capabilities
- Use of highly testable devices (i.e. breakers, relays)
- Elimination of concurrent triggers
- Segmentation
- Redundant networks
- Unidirectional communications
- Network switches with traffic control
- Use of redundant controllers, I/O, power sources, etc.
- Internal or external diversity
- Use of isolation devices
- Extensive testing 40
Qualitative Assessment Factors
- Quality of the Design Process
- Software development
- Hardware and software integration processes
- System design
- Validation and testing processes
- For Safety Related:
- Development process is documented and available for referencing in the qualitative assessment
- Commercial grade:
- Documentation may not be extensive
- Qualitative assessment may place greater emphasis on design attributes and OE 41
NEI 96-07 Appendix D and RG 1.187 Revision 2 Started and Planned Modifications by Industry using RIS 2002-22, Supplement One 42
Digital I&C Mods
- The following Digital I&C Mods are either started or planned based because of RIS 2002-22, Supplement One Issuance:
- 3 safety-related digital mods started in 2018 and planned to be complete in 2019:
- Diesel Generator Controls
- Digital Breakers
- Chiller Controls
- 8 safety-related mods planned to start in 2019 and completion in 2020, 2021, and 2022
- RWCU Instrumentation
- Chiller Controls
- EDG Sequencer
- Digital Inverters
- Control Room HVAC Controls
- Low Voltage MCC Breakers
- Radiation Monitoring System (2 mods) 43
Digital I&C Mods
- Planned Digital I&C Modifications (Contd)
- 3 safety-related digital mods with a start date TBD
- HPCI/RCIC Speed Control
- Single Loop Controllers (AFW, HPCI, RCIC)
- Incore TS and RVLIS Upgrade
- 6 non-safety related mods started in 2018 and 2019
- Turbine Controls
- Plant Computer System
- Feedwater Control
- Fuel Handling
- Rod Control (2 mods) 44
NEI 96-07 Appendix D April 16, 2019
©2019 Nuclear Energy Institute
NEI 96-07 Appendix D - Benefit to Industry With current guidance, licensees have Appendix D difficulty Identifying pertinent UFSAR-described Guides the 50.59 practitioner on design functions and descriptions affected identification of UFSAR-described design by digital activities functions relative to digital activities Determining if a digital change is adverse to Provides specific screening guidance for a design function or method of performing digital activities, including comprehensive or controlling a design function Human-System Interface guidance Deciding how and where to address CCF, Guides the 50.59 practitioner on how and including software CCF, in the 50.59 review where to address CCF in the 50.59 review process process and how to apply CCF assessments to justify conclusions Appendix D is supplemental guidance to be used with NEI 96-07, Rev. 1
©2019 Nuclear Energy Institute 2
NEI 96-07 Appendix D - Benefit to Industry With existing guidance, licensees have Appendix D difficulty Addressing Evaluation Criteria 1, 2, 5, and 6 Provides specific digital activity-based as they apply to digital activities guidance and examples on application of 50.59 Criteria 1, 2, 5, and 6 Recognizing the plant licensing bases Provides detailed guidance and examples impact when combining previously separate on the combination of design functions design functions Identifying malfunction results when Provides detailed guidance on how to addressing Evaluation Criterion 6 identify malfunction results even when challenged with UFSARs of varying levels of detail and descriptive material For digital activities, Criterion 6 has been the most difficult to address
©2019 Nuclear Energy Institute 3
Criterion 6 - Four Major Points
- 1. NEI 96-07, Definition 3.9, malfunction of an SSC important to safety is used within Section 4.3.6 of Appendix D consistently
- 2. The rulemaking record is clear - the rules intent when looking for different result is the safety analyses, not the descriptive material
- 3. Consistent with NEI 96-07, Rev. 1, Section 4.3.6 of Appendix D avoids uneven application of 10 CFR 50.59
- 4. Section 4.3.6 of Appendix D is consistent with the other 10 CFR 50.59 Evaluation criteria
©2019 Nuclear Energy Institute 4
Point 1 - A Malfunction is Defined A Design Function is either:
A Design Basis Function is either:
A malfunction is a A Design Basis Function Required by regulations, license failure to perform a Supports or impacts a conditions, orders, or TS Design Function Design Basis Function Credited in the safety analysis Accident/transient initiator App B to NEI 97-04 (endorsed by RG 1.186) states that Design Basis Functions are:
Derived primarily from the GDCs Functionally far above individual SSC functions Safety Analyses provide context In every instance, the Evaluation All of the information on this slide is begins at the lower SSC level and found in approved regulatory guidance or assesses the impact at the safety the regulation itself. analysis level.
(e.g., D/G jacket water level D/G)
©2019 Nuclear Energy Institute 5
Point 2 - Rulemaking Record Refers to Safety Analysis Level for Different Result From the Notice of Proposed Rulemaking for the current regulation:
The final change is being proposed in response to the comments on the staff proposed guidance (NUREG-1606) on the interpretation of malfunction (of equipment important to safety) of a different type However, the Commission recognizes that in its reviews, equipment malfunctions are generally postulated as potential single failures to evaluate plant performance; thus, the focus of the NRC review was on the result, rather than the cause/type of malfunction. Unless the equipment would fail in a way not already evaluated in the safety analysis, there is no need for NRC review of the change that led to the new type of malfunction. Therefore, as the third change in § 50.59(a)(2)(ii), the Commission is proposing to change the phrase of a different type to with a different result.
different result with respect to safety analyses - the focus since 1999
©2019 Nuclear Energy Institute 6
Point 2 - Rulemaking Record Refers to Safety Analysis Level for Different Result Proposed rule discusses earlier GL 95-02 guidance generated for applying the pre-1999 rule language of type The staff has provided guidance on this issue in Generic Letter (GL) 95-02, concerning replacement of analog systems with digital instrumentation.
The GL states that in considering whether new types of failures are created, this must be done at the level of equipment being replacednot at the overall system level. Further, it is not sufficient for a licensee to state that since failure of a system or train was postulated in the SAR, any other equipment failure is bounded by this assumption, unless there is some assurance that the mode of failure can be detected and that there are no consequential effects (electrical interference, materials interactions, etc.), such that it can be reasonably concluded that the SAR analysis was truly bounding and applicable.
Guidance generated for where to apply result in the revised rule
©2019 Nuclear Energy Institute 7
Point 3 - Avoid Uneven Application of 10 CFR 50.59 From SECY 97-035:
Plant SARs vary in depth and completeness. In general, the level of detail of information contained in an SAR for later facility applications was much greater than that for the earlier licensed plants. Thus, tying the scope of 10 CFR 50.59 to the SAR results in uneven application of 10 CFR 50.59.
- The solution in the current rule was to focus on Design Functions and not the descriptive material contained in the UFSAR
- Since individual sites have varying degrees of UFSAR descriptive material, this is necessary to avoid having the same change treated differently
- App B to NEI 97-04 (endorsed by RG 1.186) provides guidance that the response to an individual SSCs failure is part of the descriptive material and not part of the safety analysis
©2019 Nuclear Energy Institute 8
Point 4 - Section 4.3.6 Consistent With Other Criteria
- 10 CFR 50.59 c(2) iii states:
accident previously evaluated in the final safety analysis report (as updated)
- 10 CFR 50.59 c(2) iv states:
malfunction of an SSC important to safety previously evaluated in the final safety analysis report (as updated)
- 10 CFR 50.59 c(2) vii states:
as described in the FSAR (as updated) being exceeded or altered
- Criteria 3, 4, and 7 all rely solely on the results of safety analyses
- The guidance contained in NEI 96-07 is endorsed in Regulatory Guide 1.187 and is an approved way to meet the 10 CFR 50.59 rule
©2019 Nuclear Energy Institute 9
Summary
- Section 4.3.6 of NEI 96-07, Appendix D, utilizes previously approved definitions from NEI 96-07, Revision 1
- Section 4.3.6 of NEI 96-07, Appendix D relies on the 1999 rulemaking record to understand different result
- The rulemaking record establishes that [u]nless the equipment would fail in a way not already evaluated in the safety analysis, there is no need for NRC review of the change that led to the new type of malfunction.
- Reliance on safety analysis results versus descriptive material avoids repeating the problem of uneven application
- The logic and treatment of Section 4.3.6 of NEI 96-07, Appendix D, is consistent with the application of other 10 CFR 50.59 Evaluation criteria.
©2019 Nuclear Energy Institute 10