ML20127B956

Transcript of 930106 ACRS Subcommittee on Advanced Reactor Designs in Bethesda,Md.Pp 1-148

ML20127B956
Person / Time
Issue date:	01/06/1993
From:	Advisory Committee on Reactor Safeguards
To:
References
ACRS-T-1945, NUDOCS 9301130229
Download: ML20127B956 (183)

Text

{{#Wiki_filter:il gw y f g: w ' OFFICIALTRANSCRIFT OF PROCEEDINGS i p a ,i E LK-q; g e~' $g/ o4 away TR (ACRS) p sydf;.. RETURN ORIGINAL TO Q 1',. gb G.J.WH7TE, ACPS-P-315 [~ THANKS! Bart:ard Jo $f(!' @j ~ E

1. 27288 fgp ikg f;p,'.

Agency: u.s. wuclear aegulatory commission i gj*g 6 Advisory comittee on Reactor safeguards y JRL% "' " "^" I w = p Docket No. Mf EE l-gy = 'N95fMM Nse ! 6 ' iis e ? 2 -

Raggg g gi IDCATION Bethesda, Maryland y.

DATE Weddesday, January 6, 1993 PAcn 1 - 148 lgg i

dV e

f 8 -! a. Ap40l vimC u l t' /" * (' !T , m-. - $[F 10H:he Life of R3 Coirtha c. n1 gg b ? *f{ ' p%; < 9 k-ANN RILEY& ASSOCIATES, LTD. 1612 KSt. N.W, Suite 300 ' ?gj; 13 g g Mshington, D.C. 20006 ISDy! 9301130229 930106 (202) 29F3950 W PDR ACRS 44 T-1945 PDR gy .y ' ft % ,.9f I - *i-f; i

RCRST l945 ~ J 1 G \\'A ~ OFFICIAL TRANSCRIPT OF PROCEEDINGS g I TR04 (ACRS) RETURN ORIGINAL 10 s 8.J. WHITE, ACP9-r-315 THANKS! Bar! ara 10

1. 27288 Agency:

u.s. nuclear negulatory commission Advisory Committee on Reactor Safeguards N0* Subcommittee on Advanced Reactor Designs Docket No. LOCAT10th Bethesda, Maryland DATE: Weddesday, January 6, 1993 PAGES;I - 148 I l NDQ,OJJ'n.s O m.1, C,'. L, 1 o k uiL.- 50'theC,idv. ..I ANN RILEY & ASSOCIATES, LTD. 1612 K St. N.W Suite 300

130ggg, mshingtcn.o.c 2 coo 6

?gg11pgg99a0106 (202) 293-3950 T-1945 PDR

PUBLIC NOTICE BY THE UNITED STATES NUCLEAR REGULATORY COMMISSION ADVISORY COMMITTEE ON REACTOR SAFEGUARDS DATE: Wednesday, January 6, 1993 The contents of this transcript of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor Safeguards, (date) Wednesday, Jantiary 6, 1993 , as Reported herein, are a record of the discussions recorded at the meeting held on the above date. This transcript has not been reviewed, corrected or edited, and it may contain inaccuracies. ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 l Washington, D. C. 20006 j (202) 293-3950

1 1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3 4 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 5 SUBCOMMITTEE ON ADVANCED REACTOR DESIGNS 6 Nuclear Regulatory Commission 7 Room P-110 8 7920 Norfolk Avenue 9 Bethesda, Maryland 10 11 Wednesday, January 6, 1993 12 13 The above-entitled proceedings commenced at 8:30 14 o' clock a.m., pursuant to notice, Ernest J. Wilkins, Jr., 15 Subcommittee Chairman, presiding. 16 PRESENT FOR THE ACRS SUBCOMMITTEE: 17 Paul G. Shewmon, Vice Chairman 18 James C. Carroll, Member 19 Thomas S. Kress, Member 20 Peter R. Davis, Member 21-Carlyle Michelson, Member 22 David A. Ward, Consultant 23 Medhat El-Zeftiu/, Designated Federal Official 24 25 O ANN RILEY & ASSOCIATES, Ltd. Coun Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 e1-

2 1 PARTICIPANTS: O 2 3 M. Slosson E. Throm 4 J. Donoghue J. Donohew 5 T. Cox B. Wetzel 6 S. Scaletti B. Pierson 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 Q ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

3 1 PROCEEDINGS 2 (8:30 a.m.] 3 MR. WILKINS: Good morning. The meeting will-now 4 come to order. This is a meeting of the ACRS Subcommittee 5 on Advanced Reactor Designs. I am J. Ernest Wilkins, 6 Chairman of the Subcommittee. ACRS members in attendance 7 are Carlyle Michelson, Pete Davis, Tom Kress, Paul Shewmon 8 and J. Carroll. Ivan Catton is supposed to be here, and may 9 show up a little bit later. ACRS Consultant in attendance 10 is Dave Ward, who has a long and illustrious membership on 11 this Committee. We are glad to see him back in a different 12 capacity. 13 The purpose of this meeting is to discuss the key l 14 policy issues that the NRC staff has identified for the l 15 MHTGR, PIUS, PRISM and CANDU 3 advanced nuclear power plant 16 designs. Dr. Medhat El-Zeftawy is the cognizant ACRS staff 17 member for this meeting. 18 The rules for participation in today's meeting 19 have been announced as part of the notice of this meeting, 20 previously published in the Federal Register on December 23, 21 1992. A transcript of-the meeting is being kept and will be 22 made available as stated in the Federal Register Notice. It l 23 is requested that each speaker first identify herself or 24 himself, and speak with sufficient clarity and volume so 25 that he or she can-be readily heard. We have received no Q ANN RILEY- & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

4 1 written comments or requests for time to make oral -O 2 statements from_ members of the public. 3 The Subcommittee will review ten specific issues 4 that have been identified by the staff in a draft SECY, 5 copies of which members of the Subcommittee have received. 6 Each of those ten issues may or may not be relevant to each 7 of four separate reactor designs. So, thess's a potential 8 for 40 separate discussions, and we have to do all of this 9 today before approximately 2:15 or 2:30. 10 It is therefore apparent, that we will not be able 11 to get into a great deal of depth on any of them. I have 12 spoken to Ms. Slosson who will lead the discussion for the 13 NRC, and she and I have agreed that we will try to keep this ( 14 meeting on schedule even if that means curtailing the well 15 known tendencies of members of the ACRS to wax eloquent and 16-to pursue issues into considerable depth. 17 I believe it will certainly be necessary for this 18 Subcommittee to have other meetings as new material becomes 19 available to the staff. It will become apparent to us, I -20 think, that many of the positions the staff will take are 21 based on the information they have today, and that there's a 22 lot of other information that they are. attempting to get. 23 I think that's all I really_ wanted to say. Do any 24 of the members of the Subcommittee have any comments they 25 would like to-make? ~O ^"" ai'ev a Associates. 'ta. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

5 1 MR. CARROLL: Good luck. 0 2 [ Laughter.] 3 MR. WILKINS: We will proceed with the meeting. I 4 will call upon Ms. Slosson of NRR tt, begin. q 5 MS. SLOSSON: Good morning. 'fy name is Marylee 6 Slosson, and I am the Acting Project firecte. for the 7 Advanced Reactor Project Director at NRR. My branch is a responsible for conducting the pre-application reviews for 9 the FIUS, the MHTGR, the PRISM and CANDU 3 designs for NRC. 10 It is also the branch that prepared or was the primary 11 author of the draft paper that you have before you. 12 The paper is in draft form. We have, released the 13 paper for public comment and have sent it to the pre-14 applicants for their comments. We have requested their 15 comments by January 25th, and are requesting ACRS review and 16 comments on the issues in the paper. once we receive the 17 comments back from the pre-applicants or anyone else who 18 should comment and then once we received ACRS comments, we 19 will finalize the paper and go forward to the Commission 20 looking for their approval of our recommendations or other 21 guidance. 22 What I would like to do this morning is just go 23 through some background of why we are conducting these 24 reviews, the pre-application reviews, how we went through 25 the process to develop these policy issues. Then I would Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 - Washington, D. C.' 20006 (202) 293-3950

6 1 like to turn it over to some of my staff, who will go O 2 through the individual issues, as you can see from the 3 agenda. We didn't plan to go through any detailed 4 discussions of the plant designs today, but the project 5 managers are here. If any questions come up they should be l 6 able to either answer the question or should be able to get ) 7 the information for you. 8 [ Slides.] 9 MS. S LOSSON: As I discussed previously, we are 10 conducting the pre-application reviews of the four reactor 11 designs. We are doing these in accordance with the advanced 12 reactor policy statement. The objectives, you can see on 13 the viewgraph here, primarily are to get early staff 14 interaction with the industry on these -- early staff /NRC 15 interaction -- on these designs. It's to perform a basic 16 licenseability type review. The applicants come in and give 17 us their positions on NRC criteria and give us information 18 in a preliminary safety information document on the plant 19 designs themselves. 20 We are in various stages of review of the four 21 designs but we are furthest along on the PRISM design. 22 For your information these are the schedules. The 23 schedules you see there on the left are our previous 24 schedules. We just issued within the last month or so, a 25 new paper with new schedules for the completion of the pre-Q ANN RILEY & ASSOCIATES, Ltd. Coud Reporters 1812 K. Street, N.W., Suite 300 Washington, D. C. 20W (202) 293-3950

m-i 7 1 application reviews. The dates you see there are to be the O 2 final issuance date. We expect about six months before that l 3 we will go to the Commission with a draft and hope to go out 4 with public comment also. 5 MR. MICHELSON: Excuse me. How do these dates 6 relate to what you call the application date. As far as 7 regulations there's an application date. How do these 8 relate to that? 9 MS. SLOSSON: These are all prior to that. There 10 is no -- 11 MR. MICHELSON: There is no application date in 12 mind yet, c 13 MS. SLOSSON: Not yet. We expect that after the I 14 pre-applicants see the positions, then they will make a 15 determination whether they are going to come in with a PDA. 16 MR. MICHELSON: Thank you. 17 MR. WARD: The priority there is kind of puzzling 18 to me. I guess you have a certain input from DOE. I don't 19 know-who is responsible for setting priorities and this sort 20 of thing. I am sure the NRC will say it isn't for the 21 overall. program. You get the priorities you said are going 22 to influence things. 23 MS. SLOSSON: Yes. We expect that'we will be-24 getting in the near term, some information from DOE 25 discussing how some milestones for how they are going to Q ANN RILEY & - ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

8 1 meet the Energy Policy Act which requires in 1996 for them 2 to come in with PDA: application -- correct me if I am wrong 3 -- for the PRISM and MHTGR. 4 The dates you see here may need to be re-5 evaluated, depending on what DOE tells us with respect to 6 the MHTGR in particular. I think the PRISM dates are in 7 line with what they are doing. 8 The way we came up with the dates though was based 9 on inputs from DOE and the pre-applicants, primarily with 10 the level of information that they have available today and 11 information on whether they believe they are going to come 12 in with an application, eventually an application for a 13 - design certification. We didn't have information yet as to 14 whether DOE -- they were not going to make a decision I 15 guess, until the August of 1993 timeframe when we were 16 preparing this paper -- as to when or if they would come in 17 with a design certification application for the MHTGR. 18 MR. WARD: I think these sort of reflect-19 recommendations made by a National Academy Committee which I 20 am not sure are the world's best recommendations, in my own 21 opinion. I-wonder if you have been influenced by that or 22 just by -- is this more driven by what you expect to get .23 - from the applicants? 24 MS. SLOSSON: I-believe-it is what we expect to 25 get from the applicants. I have not been doing this very Q ANN - R! LEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington,- D. C. 20006 . (202) 293-3950 --y-. e.- g wwlr

9 1 long. Perhaps Tom Cox who is the Section Chief for the 2 Projects Section, may be able to address that. 3 MR. COX: I think you are probably referring to 4 PRISM being first. 5 MR. WARD: Yes. 6 MR. COX: The simple fact is we had all of the 7 information we needed to complete the pre-application review 8 of that project. It was well along at the time that we were 9 starting the serious consideration of shuffling schedules 10 and figuring out what to do. That naturally came along-11 first. 12 We also had meetings with each of the pre-13 applicant's organizations around March to June of 1992 and 14 considered what they had to say and what their plans were, 15 as Marylee has said, and put this all together on that 16 basis. Still, PRISM remaining first, simply because at that 17 point we were well through it and all the information is 18 there. 19 One other note. You will notice that even-the 20 latest one there, MHTGR at.12-95, is still at least a year 21 before they or the DOE would.be required under the Energy 22 Policy Act to submit a PDA application. 23 MR. DAVIS: It's not quite a year. I believe it's-24 9-30-96. 25 MR. COX: Right. That's right. O ^"" ai'ev 5 ^ssoci^Tes' 'id-Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 t r- + ew -ry.w m-e

--r--p cr---

f y-=v= 7vr'-*t' P- '4

10-1 MR. DAVIS: It's nine months. 2 MS.. SLOSSON: We did indicate in the paper that we 3 may need to re-examine this in light of the Energy Policy 4 Act. 5 MR. WILKINS: It's still' fair to say that this 6 listing does not reflect an opinion of the NRC as to the 7 merits or demerits of any of these reactor concepts. 8 MS. SLOSSON: Yes, that's correct. 9 MR. WILKINS: The DOE decision on MHTGR may be 10 driven more by defense considerations than by commercial 11 reactor considerations because they have to make some 12 decisions on production reactors. The technology for the 13 MHTGR is perhaps being supported by that program. 14 So that, they will be giving themselves ample time 15 to make that kind of decision. Until they have made that 16 decision this last date of December 1995 is sort of kind of 17 flaky. 18 MR. WARD: We might have to wait a long, long time 19 for that decision. 20 MR. WILKINS: We might, indeed. If we do, then 21 -the 12-95 will slip and DOE may find itself in difficulty 22 with Congress on the Energy Policy Act. That's their-23 problem, it's--not NRC's problem and-certainly-not ours. 24 Please continue, Ms. Slosson. 25 (Slides.] O ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington,- D. C. 20006 (202) 293-3950 rm.m,-w,--rev er .e*<mn-i.e<1 e==w er- -e--*we-*'"*e* +

~ -. 11 1 MS. SLOSSON: In keeping with the reviews that we 2 have done so far and with Commission guidance to the staff 3 to identify issues where pre-applicants indicate that they 4 may want to deviate from NRC regulations, requirements, 5 criteria, we have developed the key policy and technical 6 issues in this paper. The way we did this was to identify 7 the issues within my branch and to sit down with NRR 8 management, the division directors and above in the March 9 timeframe, to come up with some positions of which way we 10 would like to recommend the Agency go on these issues. 11 Then, we put together and draft and went through 12 some more reviews. Then, we sat down in November again with 13 Dr. Murley, Frank Moralia and Bill Russell and the division 14 directors again, and hammered out these issues to make sure 15 that we were speaking as the one NRR when we put this paper 16 out in a draft. 17 We identified issues that -- two categories of 10 issues. Issues that the staff agrees that NRC should 19 consider departura from the current requirements, and those 20 issues that right now we don't believe that the NRC should 21 consider -- at least at this pre-application stage -- should 22 consider a departure. We will go through our explanation 23 for that and those issues as we go through each of the 24 individual issues. 25. MR. MICHELSON: In the process of arriving at this Q ANN RILEY & ASSOCIATES, Ltd. Coud Repoders l 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 - -~ --~.v r .e w,

12 1 list of course, I guess you looked predominantly at what was d 2 being done on the present applications. But did you look to 3 see how the USI's -- the resolution of USI's and GI's might 4 apply differently to this group of reactors versus the light 5 water reactors? 6 MS. SLOSSON: I don't believe we did. 7 MR. MICHELSON: Don't you really have'to do that, 8 to see if you have new issues or not? You have to look at 9 your USI's and GI's. They have to be resolved at every 10 plant. They may bring up things on these plants that 11 weren't brought up on the light water reactors. I don't 12 know. If you haven't done it, you can't say. 13 MS. SLOSSON: I don't know that we have gone into 14 that detail yet. I am saying that we may in the future need 15 to do something like that. As of now, we have not done 16 that. 17 MR. MICHELSON: One of the issues of particular 18 interest which the Committee is going to discuss in February i -19 is the resolution of A-17 which was really never resolved. 20 It's the systems interaction.- It was resolved by agreeing 21 that one piece of it we thought we agreed on. The rest of 22-it is into the multiple system response program. 23 The results of that program which we will 24 hopefully hear about in February, will determine 25 considerably what other issues may come up on these. That's ] ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202)_293-3950

13 1. why I wondered if you started thinking about it. 2 MS. S LOSSON: We haven't done that yet. 3 MR. MICHELSON: Okay. 4 MR. DONOHEW: The material we have is very 5 preliminary., The systems in some cases haven't been 6 completely designed. I don't think it's a fact that the 7 USI's are going to be disregarded. 8 What we have if we picked items which are very 9 clear to be issues which will make a difference between 10 licenseability on the designs or not and going through 11 those, I think what we have done is picked the ten most 12 important. When we get into the preliminary review and when 13 we get design information, I think then we can do that type 14 of review. l 15 MR. WILKINS: Carl, I can remember hearing you say 16 some things in another context which apply even more 17 forcefully here. You don't have information, and you had 18 even less information. l l 19 MR. MICHELSON: I was mostly wondering whether 20 they tried to look at the resolution of the old -- 21 MR. WILKINS: I think you raised a valid point -- 22 MR. MICHELSON: -- and so forth, to see if they_ 23 applied differently-to these reactors'than they.did -- 24 MR. WILKINS: That's a perfectly fair question. I 25 think for the time being, we have to' accept their answer l l Q ANN RILEY & ASSOCIATES, Ltd. COud Repoders 1612 K. Street, N.tN., Suite 300 Washington, D. C. 20006 (202) 293-3950

14 e 1 that: one, they haven't; two, they will. 2 MR. CARROLL: They say they may. 3 MR. WILKINS: They may. 4 MR. MICHELSON: 1 think there is no -- 5 MR. CARROLL: I think they should say they will. 6 MR. MICHELSON : They have no choice, if it's under 7 Part 52. It's required by regulation, 8 MR. WILKINS: They will, if there is an 9 application. 10 MR. MICHELSON: If there is an application. 11 MR. WILKINS: At the present stage there isn't an 12 application, and what they are doing is that they have 13 identified these issues as worthy of attention..By so () 14 identifying them, they do not say there are not other issues 15 that may show up later on. 16 MS. SLOSSON: Right. There are additional issues 17 that would come up as part of or have already come up as 18 part of the evolutionary and passive reviews, and we tried 19 not to repeat those issues-as part of this review. 20 MR. CARROLL: A few minutes ago you were telling

21 us about how this all took place within NRR.

22 MS. SLOSSON: Right. I 23 MR. CARROLL: I guess I have the impression that 24 Research had a lot of involvement in these advanced concepts 25 also. ( ANN RlLEY & ASSOCIATES, Ltd. COud RepOders - 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

15 1 MS. SLOSSON: Certainly. O-2 MR. CARROLL: How are their views being factored 3 into this? 4 MS. SIASSON: They are and have been, and continue 5 to be-involved in the meetings that we have. They are 6 conducting research programs with this. They prepared the 7 draft of the pre-application safety evaluation reports that-8 were done for both the PRISM and the MHTGR in the 1980's. 9 We do continus to work with thsm and not igno'J9 10 them, and I misspoke if I said it was only NRR. 11 MR. MICHELSON: One of your earlier statements 12 left me a little confused. If an_ issue is already looked at 13 on LWR's, do you not then look at it-on these as well? 14 MS. SLOSSON: No, it will be looked at. It is 15 just not being identified as a key policy issue as part of 16 this paper. 17-MR. MICHELSON: The suggestion on control rocm 18 design, is there some kind of a different issue than was 19 raised on LWR's; is-that how I should--interpret this? 20 MS. SLOSSON: Not necessarily. In seme of the 21 cases, I think, if you look through'this paper -- this paper 22 has had-some iterations. If you look through, _ we wc;,, ;ack 23 and said under the recommendation that we would continue 24 with the LWR. There are differences in the designs that we 25 pointed out. -] ANN FilLEY & ASSOCIATES, Ltd. Court Reporters 1612 K Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 e. .,,.e n

~. 16 1 MR. WILKINS: -In some cases, Carl, the 2 recommendation will be, we will conform our policy for these 3 new reactors to the one that is not yet been finalized for 4 the evolutionary and passive plants. 5 MS. S LOSSON: There is still some. overlap, but we 6 try not to repeat everything. 7 MR. WARD: I have one question. You're about 8 finished, aren't you? 9 MS. SLOSSON: Yes. 10 MR. WARD: You have separated out CANDU from the 11 other three in the nature of the review. CANDU review is 12 going to be similar to the evolutionary designs. I guess it 13 makes a certain amount of sense. Apparently,-one difference 14 is that it probably won't be -- a prototype w7n't be 15 required for CANDU. What other differences are there? 16 MS. S LOSSON: Ed, can you address that? 17 MR. THROM: What you will see as we go through the 18 day, especially if you look at the first four categories up 19 there, a lot of that came out of the submittals by DOE on 20-PRISM, MHTGR and earlier SAFR, and PIUS. Under the Advanced 21 Reactor Policy Statement they want to take credit for-more 22 passive systems, reduced operator actions and-they would-23 like to demonstrate enhanced safety which, under the policy l l_ 24 paper, the Commission had encouraged. 25 The Canadian design is not as p.asive in its l l Q ANN RILEY & ASSOCIATES, Ltd. Coun Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 =

~ 17 1 characteristics as the other three. It is based on the O 2 heavy water technology that the Canadians have-had ongoing 3 for quite a number of years now. In their eyes they 4 consider it an evolutionary type design with respect to the 5 CANDU 6. 6 The problem you have here at the NRC is the l 7 knowledge base on the heavy water reactor design, the 8 particular characteristics of the pressure tubes, the 9 horizontal flow problems and some of the regulatory 10 philosophies that the Canadians would apply to licensing 11 their plant. 12 It's Kind of in a gray area in the NRC's 13 perspective, as we can't really say it's fully evolutionary. O" 15 We could with our knowledge base of light water reactors, 16 there should be some credit for what has happened in the 17 operating history with the heavy water reactors that the-18 Canadians have. 19 MR. WARD: I think I understand the issue. I am 20 trying to understand -- the sentence in the draft paper 21 introduction on page~four says the staff proposes to apply 22 pre-application review criteria to the CANDU 3 reactor that 23 are consistent with evolutionary LWR review requirements. 24 When I read the paper the only difference I can really find 25 is that it is on prototyping. Q ANN RlLEY & ASSOCIATES, Ltd. Gourt Reporters 1612 K. Street, N.W., Suite 300 Wasnington, D. C. 20006 (202) 293-3950

- _ - ~.. -. 18 1 I was wondering what other differences is the O 2 staff proposing in the review requirements, pre-Opplication 3 review criteria. 4 MR. COX: I don't think we are proposing any 5 specific differences other than what we see in.this paper 6 right now. We haven't really gone into the detailed 7 technical review of this CANDU 3 design. 8 MR. WARD: I guess what you are really saying is 9 that it's going to be easier to review, but that's not what 10 this sentence says. t 11 MR. COX: We hope to apply criteria that are more 12 like those of evolutionary requirements than something you 13 would apply to PRISM or MHTGR because of the vast 14 differences. PRISM and MHTGR are much farther out in 15 technology than is the CANDU 3 design. 16 MR. THROM: Maybe as we go through the 17 presentations today, I think you will see that you can see 18 from the list that some of the reactors get pulled out of 19 issues, and maybe we can make some of that a little clearer 20 by looking at what the proposals were from the DOE and PIUS 21 reactor versus what the proposal from the CANDU 3 reactor 22 is. 23 I think a lot of it is going to tie into our 24 interpretation of the advanced reactor policy statement and 25 the methodology for the review that was laid out in NUREG Q ANN RlLEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

19 1 1226 by research, as to how to go about this type of' review. 2 3 As I said, it's a problem that we have right now in that 4 it's a gray area. I think otte of the biggest problems and 5 why it's in this group right ncw is the fact that it is 6 different from what the NRC's knowledge base on light water 7 reactors is. 8 It will be a fairly mixed review, in terms of - 9 evolutionary requirements versus really advanced 10 requirements. 11 MR. WILKINS: I think you have addressed Mr. 12 Ward's question adequately. We are running behind already. l 13 MS. SLOSSON: That's it. 14 MR. CARROLL: I did have one other question. 15 MR. WILKINS: Go ahead. 16 MR. CARROLL: How does the fee issue impact on 17 these rcviews. Who is paying for this? 18 MS.-SIDSSON: For the PIUS and the CANDU reviews 19 the pre-applicants are paying. For the DOE reviews it goes 20 into the -- correct me if I am wrong, Tom -- it goes into 21 the overall NRC. 22 MR. WILKINS: You don't bill DOE direct. 23 MS. SLOSSON: We don't bill DOE-directly.- 24 MR. CARROLL: Does that mean that the utility rate l l 25 - payers are paying for it? l l-O ^"" ai'ev & Associ^Tes' 'id-Court Reporters _1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950- ,-n e-N. N-n m .m y

20 1 MS. S LOSSON: I believe so. Am I correct? 2 MR. COX: I am not really qualified to say how the 3 money is coming into the NRC or general fund. I would have 4 to get the controller people. 5 MR. WILKINS: I think the answer to your question 6 is yes. 7 MR. CARROLL: I wonder if they know this. 8 MR. WILKINS: If it's not the utility rate payers 9 it will be the U.S.A. taxpayers, which is almcst the same 10 group of people. 11 MR. DAVIS: There's a rather significant piece of 12 research that is going on in support of the review of these 13 applications. That's fairly expensive, I would guess. ~ 14 MR. WILKINS: We are not going to settle this one. 15 16 Ms. Slosson, are you finished? 17 MS. SLOSSON: I am finished. I am ready to turn 18 it over to Ed Throm, who will go through the first issue. 19 MR. CARROLL: I would ultimately like to hear more 20 about that questica. 21 MS. S LOSSON: All right. 22 [ Slides.) 23 MR. THROM: I would like to open by saying that 24 the first four topics are very interrelated. The accident 25 selection criteria would determine how you would look at Q ANN RILEY & ASSOCIATES, Ltd. l COud RepOders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

21 1 source term. Source term and containment performance would 2 be how you would get to evaluate emergency planning. It 3 would be perhaps a necessity to put off to the next topic, 4 some of the discussion points that may come up. It is a 5 very interrelated process. 6 In a way of background, the advanced reactor 7 policy statement was put forth by the Commission as Marylee 8 said, to encourage early interaction with licensees, 9 government and the NRC on these advanced reactor designs. 10 As I just mentioned, NUREG 1226 was put together in the 11 office of research where this work had stated back in 1986 12 and 1987, as the general guidelines upon which the NRC would 13 try to do its review. 14 The intent is to build up on the light water 15 reactor experience. The current regulations will determine 16 where those regulations would apply to an advanced design 17 where perhaps new regulations would have to be developed or 18 where a particular pre-applicant could come in and try to 19 make an argument for an equivalent level of safety. In 20 other words he had another method or technique for assuring 21 the same level of safety as you would find in a current 22 light water reactor. 23 Also, encouraged by the policy statement was to 24 encourage the designers to go to passive designs, remove the 25 operator, some of the man-machine interface problems that we Q-ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

22 1 had, simplify the designs and make the plants easier to 2 maintain and operate. For the most part the pre-applicants 3 are trying to come in with propoeals on those lines. 4 The work that we are doing is basically taking the 5 preliminary safety information documents that are submitted 6 by the pre-applicants and reviewing what they are trying to 7 do, and trying to see how it fits into the current 8 regulations as we have them now where they deviate from 9 regulations, what kind of information would you need to 10 accept that. Probably more important in the long rua is 11 where there are technical areas that need resolution, to 12 understand and evaluate research and development programs 13 that the pre-applicants have underway to assure that when 14 they come in for design certification they know the type of 15 information the staff is looking for. We would have a 16 feeling that the type of information that we think is 17 necessary to support some of their contentions on their 18 design would be available to NRR-when the design 19 certification came in. 20 The MHTGR, PRISM, PIUS pre-applicants have 21 submitted evaluation schemes wnich you would consider to be 22 quite more conservative than we have under current light 23 water reactors. In looking at the safety goal policy 24 statement what they were proposing initially is to have a 25 design such that all events that they would put into design Q ANN RILEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

23- _1 basis accident space down to about ten to the minus six in 2 probability of occurrence would lead to offsite consequences 3 substantially lower than Part 100. 3 4 That was the justification that they would try to 5 apply to having reductions in energency planning. 6 MR. DAVIS: Excuse me. I could find no reference 7 in your draft policy statement to the safety goals. Instead 8 what I did find was a statement -- it appears on page 10 -- 9 that these reactors need to be at least as safe as the 10 current generation of reactors. 11 If one looks at the Pw4 results for the current 12 generation of reactors you see a rather wide range of risk 13 results, all of which are below the safety goals. When you 14 say you want something at least as safe as the current 15 generation, I am not sure which member of the current 16 generation you are talking about. Also, I am wondering why. 17 wouldn't it be acceptable just to say that these plants must 18 meet the safety goals. 19 Do they not apply to these plants, or what is the 20 relationship here? 21 MR. THROM: I think the safety goal is one of the 22 important Commission guidance that we think is applicable-to 23 any design. There is an item in the list that talks about1-24 - I will get to it'in a_ couple of minutes -- the guidance 25 that the Commission has that we need to use to set Q ANN RILEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950

24 l 1 acceptable fuel, core damage, offsite consequences. I think ' -O 2 that is tied into the safety goal and the severe accident 3 policy papers. 4 What I was trying to say here is that at least the 5 three designs, PRISM, MHTGR and PIUS, when they came in 6 their interpretation was that if they could -- down to a ten 7 to a minus sixth type of range and possibly a little lower 8 to make sure there weren't any outliers in there -- 9 demonstrate not only that they could meet the safety goal 10 but look at offsite doses that were substantially less than 11 what one might consider as the safety goal which is not 12 finalized yet, that would be a justification for the 13 reduction in offsite planning. ( } 14 We don't plan to say that the safety goal is not 15 applicable. It is definitely applicable. 16 MR. DAVIS: What if you meet the safety goal but 17 you are not as safe as some current reactors. 18 MR. THROM: I guess that's a difficult thing. I 19 think what the Commission was attempting to say at that 20 point is that the staff would not accept anything that they-21 thought was not as safe as at least what in the policy paper 22 which is called the current generation of reactors which is 23 inferred to as the current evolutionary and passive designs 24 that are under review, not necessarily the ones that are out 25 there operating right now. O ^"" ai'av a Associates' 'id. Coud Rapoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

25 1 MR. KRESS. If you indeed met the PAG guidelines 'l 2 for accidents at ten to the minus six probability or less, 3 then there's no way you would not be better than the 4 existing. 5 MR. DAVIS: Which existing ones? I mean, some 6 existing plants have extremely low risks, partly because of 7 favorable sites and they are well below the safety goals. I 8 am not sure -- 9 MR. KRESS: You are right on that. 10 MR. DAVIS: You really need to be -- 11 MR. KRESS: For the majority of them you would be 12 well below that. 13 MR. SHEWMON: It sounds like a ratchet by your i 14 explicit policy unless you say something about the safety 15 goals. That's just one man's opinion. Two, I am told. 1C MR. WILKINS: The only reason that I don't add my 17 name to that list is because I don't think they have really 18 committed themselves to anything yet. 19 MR. ShEWMON: There is a draft. 20 MR. THROM: Again, there is also the pre-21 application review. 22 MR. WILKINS: Of course,.this illustrates a trap 23 into which I think this Committee has pointed out to-the f 24 staff many times. Just because the applicant or a pre-25 applicant proposes to do something that is better than they Q ANN RlLEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

26 1 have done in the past does not mean that the NRC should 8 7 require them to do it. 3 MR. THROM: No. 4 MR. WILKINS: That's the concern here. 5 MR. TilROM: That is part of the concern that 6 perhaps puts the CANDU 3 design into a little bit more of a 7 gray area, in that the other three reactors have como in 8 basically and have said in response to the advanced reactor 9 goal, we do believe that we do have enhanced margins and 10 enhanced safety in these designs. This is what we would 11 propose to show you as to how we would be safer than the 12 current generation of reactors, where the CANDU design being 13 basically an evolutionary design they have not come in and 14 said that they want to be safer than. 15 They believe that they are within what the context 16 of what the safe as is going to be defined in terms of NRC. 17 It's a very nebulous thing to characterize. 10 HR. SilEWMON: That. s what the safety goal is for, D is to sort of define how safe is safe enough and stop thie 20 perpetual ratcheting that you seem to be urging. 21 MR. DAVIS: Exactly. 22 MR. COX: In fairness to Ed here, I don't think 23 that we are urging that all plants be orders of magnitude -- 24 all future plants -- be orders of magnitude safer than the 25 safety goal. What we are engaged in here is a pre-Q ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950 ~

27 1 application review, where we examine plant design to see 2 what they have proposed or what they claim is -- 3 MR. SHEWMON: CANDU is in a gray area because it 4 didn't promise orders of magnitude more like everybody else 5 did. 6 MR. COX: Maybe you want to address what you meant 7 by gray area. I don't think you meant that we are going to 8 ratchet them up to a higher icvel of safety necessarily. 9 MR. SHEWMON: It sounded like it. It's wasn't 10 doing anything until they came back with more promisos, is 11 what I heard. Onward. 12 MR. WILKINS: Why don't you continue, Mr. Throm. 13 (Slides.) 14 MR. THROM: The issue as is defined is, how you go 15 about selecting accidents and acceptance criteria for these 16 advanced reactors and the CANDU 3 for this pre-application 17 review. The recommendations are a little better worded in 18 the paper than put up here on the viewgraphs. 19 What we are recommending is that the events be 20 selected deterministically and supported or supplemented by 21 PRA insights as best as you can get them at this stage of a 22 design. 23 MR. KRESS: Just exactly does it mean, to select 24 events deterministically? 25 MR. THROM: That's to look at the design, look at Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Stieet, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

1 28 1 the systems that the designor is saying are important to 2 mitigating initiating events, and analyzing those events 3 without saying that they in a ten to the minus sixth range 4 an6 not wanting to look at them based on purely looking at a 5 probability of occurrenco. 6 In other words, we don't want to just rely on 7 trying to do a full PRA of the initiating events and the 8 systems that come into play to mitigate them and only 9 analyze based on a fixed perspective of thn likelihood of 10 that event, the consequences. We think it's important to 11 look at the types of events that seem to be important, to 12 look at the systems that are going to be relied on in those 13 plants, to mitigate initiating events, and look at those } without getting into the game of trying to say that's a ten 14 15 to the minus six ovent and wo don't have to look at it or I 16 think it's ten to the minus seven, or I don't think you havo 17 enough information on reliability for now systems that don't 18 have any operational experience to say that you can exclude 19 certain events from consideration purely on someone's 20 estimate of what they think the likellhood of the event is. 21 MR. KRESS: I neo. Someone came up with an event 22 that clearly was ten to the minus tenth, you wouldn't throw 23 it out just because it was ten to the minus tenth? 24 MR. Tl! ROM: No. You have to look at it and see 25 where that ten to the minus tenth came from. If it was O ANN FilLEY & ASSOCIATES, Ltd. Court Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 ,---n ,.rne- ,..-g.

29 1 relying on say multiple failure probabilities of 2 undocumented equipment performance or undocumented system 3 performance. 4 MR. KRESS: You would throw it out because it's 5 probably not a likely event. I am trying to decide what 6 deterministically really means. It means you probably chose 7 realistic events that have some probability of occurring. 8 MR. THROM: Yes. You might, at this particular N 9 stage of a review, want to in the case of a designer saying 10 he had enhanced margins of safety ask him or look at events 11 that were again maybe one more failure past where he had 12 gone, to see what type of timeframe is available. 13 If he's got under his design basis accident he has 14 400 degrees fahrenheit for example margin to a fuel failure 15 criteria, he might want to look at an additional failure on 16 that just to see whether an additional failure would really 17 make it a lot worse or how long it would take to get worse. 13 What we need to do is assess licenseability. 19 You can't go in on a full PRA perspective and say 20 we know everything well enough that there's nothing that is 21 going to catch us in design certification that we didn't 22 look at. 23 MR. KRESS: I think I understand what you mean 24 now. 25 MR. MICHELSON: In the light water reactor case Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

h 30 1 you have a standard review plan which identifies a number of 2 things you have to look at including certain types of 3 events. Are you going to use that set as a starting point 4 for evaluating these plants, or are you going to dream up a 5 new set? 6 MR. TilROM: To the extent that that set is 7 applicable to the designs, they would be used. Also, there 8 has to be new onen for the gas technology, the liquid metal 9 technology. 10 MR. MICllELSON: For instance, and example is 11 oscillation of pipe breaks. That's something you do 12 deterministically in present day and not on a probabilistic 13 basis. In the future or plants you don't have much 14 experience with certain of these types of flows and so forth 15 so it's hard to say whether you have corrosion problems or 16 whatever. 17 That would be a case, I think, of doing it 18 deterministically. It is just simply arbitrarily 19 pcstulating pipe breaks at certain locations for the purpose 20 of protection of the plant. 21 MR. IIEATil: Yes. That's done. The design basis l 22 accidents -- l 23 MR. MICHELSON: I wondered if you were goinc, to 24 use those same rules for these plants or whether you are 25 coming up with a new set of rules. O ANN RlLEY & ASSOCIATES, Ltd. Court Reporters - 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950 t ---

~ _ = _ i 31 1 MR. THROM: It's hopefully the same rules. I 2 think that's what we are trying to do as best we can, apply 3 the current rules to see how well they fit the designs and 4 the concepts and then see where things would need to be 5 changed. 6 MR. WILKINS: Mr. Throm, why don't you pick up the 7 pace a little bit, please. 8 MR. WARD: I hate to slow things down -- 9 MR. WILKINS: We have a choice, colleagues. We 10 can spend an hour on this one and then no time at all on 11 some of the later ones. That's the inexorable nature of the 12 arithmetic. I won't say I'm indifferent, because I do care 13 what we do. Really, the choice is yours. 14 MR. MICHELSON: What's sacred about 2:00 o' clock? 15 MR. WILKINS: Because at least three members of 16 this subcommittee including its Chairman, have a meeting at 17 3:00 o' clock. We have already used up -- 18 MR. MICHELSON: I was just pointing out -- 19 MR. WILKINS: -- 25 minutes of the 45 on the first 20 issue. 21 MR. THROM: We fully expected on these first four, '22 that -- 23 MR. WILKINS: Let me make those speeches. You go l 24 ahead and talk about -- 25 MR. THROM: Let me try to wrap up. I think what i O ANN RILEY & ASSOCIATES, Ltd. Court Reporters l 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

1 32 1 we are trying to do here in evaluating the pre-application O 2 designs is apply the current light water reactor regulatory 3 space if I can use that expression, to make the first pass 4 at how these designers are muoting those requirements, the 5 GDC's. Where they have alternatives to meet the GDC's, we 6 vant to look at those. 7 We want to base the methodology pretty much on 8 what one would consider standard light water reactor 9 practices in that, we would look ALO's, we would look at 10 DBA's. We would probably look at DBA's to lower frequencies 11 that you might see in the traditional light water reactor. 12 Again, that's because we want to assure that we are not 13 missing anything by trying to cut off something in a ( ) probability space. 14 15 We also think it's important to look at other 16 types of events that would significantly challenge the 17 designs, to come up with a source term to look at 18 containment performance for example. This is something that 19 we would not consider in probability space but we would look 20 for some type of scenario that would lead to core damage or 21 could lead to core damage, and then look at that type of an 22 action and to evaluate containment performance. 23 A lot of the information is preliminary right now. 24 25 There are a lot of systems, the metal fuel program that O ANN RlLEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 {- 4 e., ,,e-

33 1 PRISM is relying on. These things are still under O 2 development. At this stage what we need to do is put a fair 3 amount of conservatism into the evaluation that we do to 4 assure that there is nothing that would come up in an R&D 5 program such that a significant oversight would make a 6 design fall into a fairly hard licenseability frame. 7 We want to be able to come to a preliminary safety 8 evaluation report that really puts a good perspective on 9 what we think the licenseability of the design is. 10 MR. KRESS: one more question. I am quite 11 intrigued by your last sub-bullet. I got the impression 12 from reading this information that you had in mind 13 developing a set of accident sequences that would be 14 evaluated, you would eventually pin some sort of probability 15 of occurrence to those and bin them by probability of 16 occurrence. 17 And then, use some sort of a consequence analysis 18 for each of those that would have acceptance criteria that 19 would be different in terms of consequence for each of those 20 probability levels that would be keyed to something like an 21 acceptability which may be a safety goal or may be something 22 else. Is that the correct interpretation of this? 23 MR. THROM: Yes. 24 MR. KRESS: Isn't that quite a radical departure 25 from what you have been doing, or am I wrong in that? Q ANN RlLEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

34 1 That's sort of a mini-PRA that uses acceptance criteria of 2 safety goals or something. 3 MR. THROM: The problem that you have today is, 4 you don't have a complete design and a good knowledge base 5 to base that solely on a risk assessment. What we are 6 trying really here to do is look at the new technologies, 7 look at the design features, and identify areas where these 8 designs need more work in R&D. 9 one of the problems you have is, how do you come 10 to a point where you can write back to a pre-applicant a 11 safety evaluation report. You need to use something as a 12 gage. The gage that we would come up with, be it fuel 13 damage limits or offsite consequences or levels of risk, ( ) 14 what's a DBA, what's a severe accident, what would you do l 15 with a severe accident as you really thought it was a low 16 probability event. 17 That's the things that I think we will be coming 18 back to not only the ACRS but the Commission with when the l 19 pre-application SER's are put forth. That's where on a l 20 design specific level we will have to have something in a 21 frequency range, something as to what we think are 22 applicable guidance for acceptable fuel damage, what types 23 of offsite consequences do we think are reasonable, what are t 24 the offsite consequences that the applicant is proposing to 25 be marked against. O ^"" ni'ev a AssocieTes' 'td. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293-3950

35 1 Also, at the same time, to keep abreast of what's O 2 going on with the evolutionary and passive designs because a 3 lot of the things that these people are wanting to take 4 credit for have similar things like, how do you evaluate a 5 passive safety system that has no active components. 6 What do you do about the auxillary systems that 7 are not safety grade but supported. We will be addressing 8 some of that later this afternoon. When I talk about 9 licenseability, what we are trying to do is look at the 10 design, look at the technology, identify those areas where 11 we think an applicant at that time would have to come in 12 with data R&D and some kind of experimental test program and 13 data, to support the contentions that he's trying to make. 14 We need to understand that when he comes in with an 15 assumption for example on a source term, that there is 16 something to it; that it's not just pulled out as someone's 17 guess. 18 It looks reasonable, you can get some people from 19 the national lab or contract to look at it, look at the 20 international database and say yes, it looks like it's in 21 that ballpark. Understand that he has an R&D program to 22 obtain the data that is necessary. You don't want to fix 23 yourself to the point where, if he's off by ten percent in 24 his source term, he's all of a sudden on the edge of a drop 25 off where he would have to do some radical design changes or Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

36 1 one would possibly question licenseability against any kind 2 of guideline that you might want to apply to that 3 determination. 4 MR. KRESS: Do you have any notion at the moment 5 on what basis these consequence acceptance limits will be 6 placed? Will those be keyed to the safety goals, for 7 example? 8 MR. THROM: Yes. I think that for the design base 9 accidents we are right now looking at the Part 100, Part 50 10 rule changes for the interpretation of what acceptable 11 offsite dose limits are. In terms of the safety goal, wo 12 think that they would be looked at against the safety goal. 13 Whether you could do that today not fully 14 understanding the full risk type numbers of the plant, 1 15 don't think you can do that today or feel comfortable with 16 someone coming in and saying look at all of these passive 17 systems that are ten to the minus sixth failures. We don't 18 have right now, the information to say that's absolutely 19 right. 20 Design certification will be a very, very detailed 21 review. I think that's the place where we would start 22 picking up not only the current regulatory space but USI's 23 as Dr. Michelson has alluded to, generic issues. As a 24 matter of fact on the SAFR submittal that DOE had in a 25 number of years ago there were sections in there that went (] ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 Y

37 1 over all of the USI's and the TMI action items to look at 2 that. 3 That needs to be looked at in design 4 certification. I don't think a lot of detail of that has to 5 be looked at today. Surely, that knowledge base would be 6 applied to the design certification of any of these plants 7 as they apply. 8 MR. WILKINS: llow much longer do you have? You 9 have at least one more slide, or you don't have another 10 slide? 11 MR. THRoM I prefer to skip it. I think the 12 words in the paper are a little bit better. 13 MR. WILKINS: In fact, I think the words in the 14 paper are better than what the words on the slide are. I 15 would like to remind the Subcommittee that what we are 16 expected to do is to comment to the Commission if we wish to 17 do so, on the staff recommendations. The staff 18 recommendations are in the paper. 19 In this particular issue what they have proposed 20 to do is develop an approach which as certain 21 characteristics, and these are thi characteristics that were 22 on that slide and the one he didn't show. We must not 23 forget that we are not reviewing a completed design and 24 neither is the staff reviewing a completed design. In all 25 the discussions that the committee has had with respect to O ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950 ~

38 1 DAC's and ITAAC's and all that kind of business will come O 2 later, at a much later stage in the discussions on these 3 designs if they come in at all. 4 Mr. Throm, are you prepared now to proceed to the 5 source term? 6 MR. THROM Yes. 7 (Slides.) 8 MR. THROM: Under source term, we believe that all 9 of the pre-applicants -- that's all four of the designs 10 that we are looking at -- are proposing a mechanistic source 11 term, in that for a particular scenario accident sequence. 12 They would evaluate what the source term for that accident 13 was in terms of fuel failure, core damage, how much gets } into the coolant and how much gets into containment, and how 14 15 much is released from containment. 16 We are aware of an ongoing activity in the staff 17 in NUREG 1465, where they are re-evaluating accident source 18 terms. We are keeping abreast of what's going on under that 19 issue. That kind of provides some guidance as how you would 20 do a mechanistic source term in deference to the old TID 21 methodologies that would be applied to siting for example. 22 We also know that there are revisions underway to 23 separate some of the dose requirements from Part 100 into 24 Part 50, and some of those administrative changes so to 25 speak where the regulations require -- will be changed. The Q ANN RlLEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 l Washington, D. C. 20006 (202) 293 3950 1

r 39 1 guidance that we would pose is to be consistent with that 2 approach. 3 MR. DAVIS: Excuse me. One of the recommendations 4 associated with that issue -- and 1 am talking about item 5 three on page eight -- says that events considered in the 6 analysis to develop the set of source terms for each design 7 are selected to bound credible severe accidents. 8 Do you have a definition of credible in this 9 context? Is it a probability number or is it a judgment 10 call, or what is it that you had in mind? 11 MR. THRoM: At this point it's more a judgment 12 call. 13 MR. DAVIS: Is it clear that the applicant knows ( 14 what your judgment is on where the cutoff is on credible and 15 incredible, or is that something you work on as you go 16 along? 17 MR. THRoM: That, we will work on as we go along. 10 We think it's important to understand through our 19 interpretation of thG severe accident policy statement at 20 least what the consequences of a significant core damage 21 accident would be. 22 Once you gain that understanding you then would 23 have to consider what you really thought the probability of 24 the event was, what the consequences really look like in 25 terms of offsite consequences, and any kind of dose to the O ^"" ni'ev a Associates' 'td. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

1 40 1 public. Once you have that picture, determine what you O 2 wanted to do about severe accidents in terms of whether you 3 would. want to go in and require any design changes. 4 It's not our intention to come up with something 5 that would require a design change purely for the reasons of b wanting to do that. But severe accident policy statement is i 7 interpreted to mean that you should understand how a core 8 uamage event would occur and what the consequences of it 9 are. We are not going to weight that on a full probability 10 assessment. 11 We do need something to look at containment 12 performance with, and if a designer comes up and tries to 13 say there's just no way that down to ten to the minus sixth 14 that I can have a core damage therefore this containment 15 design that I am proposing is acceptable, I don't think we 16 are ready to fully accept that without understanding what 17 the possible consequences would be. 18 MR. DAVIS: Thank you. 19 MR. WARD: I think that number three needs to be 20 described differently in some way. It's not a very helpful 21 statement or recommendation. I think the use of the word 22 " credible" as an adjective in a sense has sort of lost 23 credibility. 24 MR. WILKINS: I don't really quite know how you 25 are going to bound design dependent uncertainties either. Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

l 1 41 1 MR. THRoM: No one said this was an easy job. For 2 an example, if a design is relying on a single safety grade 3 passive cooling system -- of course, to use PRISM's reactor 4 sir cooling system as an example -- that's the last thing S that you have in the design for unprotected transients to 6 remove decay heat. 7 If you are willing to give them 100 percent credit 8 for it, then for a lot of very low likelihood events you t 9 would say that that plant had very good performance 10 characteristics. You also have in that design -- which we 11 will discuss a little later -- a potentially unfavorable 12 attribute. That's the positive void reactivity coefficient. 13 14 If that system does not work you can get yourself into a 15 core disruptive accident. 16 The designer has looked at that. He has not said 17 what the credibility of it is or how he would get there. He 18 did take the initiative and did look at a core disruptive 19 accident to look at the containment performance. It wound 20 up being okay. If it did not wind up being okay what would 21 we have done, I am not sure. I think we need to have the 22 full picture and understand what's going on and understand 23 the consequences before we can make any determination on 24 features of the design. 25 We don't want to just bypass it at this particular Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D, C. 20006 (202) 293-3950

i i 42 1 stage as being fully incredible. 2 MR. WILKINS: Your recommendation is that the i 3 source term be developed on a mechanistic analysis provided 4 that, and then you list three items. 5 MR. THROM: Yes. 6 MR. WILKINS: But you don't say what happens if 7 those provisos are not, either all of them or one of them is 8 not met. I would have understood this better if you would 9 have recommended that the pre-applicants do the work 10 required to meet the proviscs. It's not cicar to me that 11 your recommendation doesn't amount to that anyway. 12 MR. THROM: It does. 13 MR. WILKINS : You might want to consider rewording 14 it then, so that it's much more explicit in that regard. 15 There is some performance data on fuel of the MHTGR fuel, 16 from Fort St. Vrain for example, and other such places. But 17 I would understand it if you said that was not enough. 18 But then ultimately, you need to say how much more 19 is enough. 20 MR. THROM: Yes. I think that is one of the 21 objectives of the preliminary safety evaluation report. For 22 example -- I am sorry, but PRISM is the one we have been -23 working with -- is the metal fuel and sodium. There is some 24 information on it. Unfortunately, source terms in most 25' metal sodium reactors are based on oxide fuel. Q ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

43 1 PRISM is using plutonium, uranium, zirconium metal O 2 fuel. It has very different characteristics. It has 3 different molting temperatures. It has a different makeup, 4 different fission products to worry about. They have come 5 up with an understanding of what they think the source term 6 is. 7 We have told them that we have looked at your R&D 8 program where they will be doing follow on testing of 9 actually running some fuel damage transients in their test 10 facility to support that data. That's what we mean here, 11 and you very well have to know what that is. 12 At this stage we want to understand what they are 13 trying to tell us is the source term and feel relatively 14 comfortable with it, and make sure that they know when they 15 come in with design certification what type of information 16 they need to put forth to the staff to get fully accepted. 17 MR. WILKINS: I notice that PIUS was not listed as 18 one of the reactor concepts of which this was an issue; why? 19 MR. THROM: It's very typical light water reactor 20 fuel. We think that the ongoing efforts under 1465 might be 21 something that the applicant would apply, in which case 22 there is nothing in there that would be significantly 23 different from what the staff would be familiar with. That 24 particular design uses very typical light water reactor fuel 25 in terms of its composition. It's in a water environment. O ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

44 1 The perspective of the fission products that you 2 have to deal with, I 3 MR. WILKINS: You may want to consider putting 1 4 those two sentences or three sentences in this document. In 5 fact as a general remark, Ms. Slooson, you might wish to 6 explain all of the gaps in the table to make sure that they 7 are not overlooked. 8 MS. SLOSSON: All right. 9 MR. THROM: We have gotten comments back, and wo 10 expect comments back from some of the licensee pre-11 applicants. I think these things are coming up as important 12 probably to put into the paper where things are different 13 and why they don't fall into the box. ( } 14 We are trying to keep it at a very policy level. 15 Most of the issues are policy-related. 16 MR. WILKINS: What is the status of the revisions 17 to 10 CFR Part 50 and 100 that deal with separating siting 18 from source term? We have had some presentations to the l 19 Committee on that subject, but I don't know off hand. I 20 MR. THROM: I know that it's still being worked 21 on. I don't think it has been finalized yet. I don't think 22 the specific details, I could not bring up to you right now. l 23 MS. SLOSSON: We can try to get that information 24 for you tomorrow. 25 MR. w1LKINS: Someone should have a general l O ^"" si'ev a Associates' 'id. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950

45 1 statement to make tomorrow on that. 2 MR. KRESS: When you speak of source terms, you 3 are referring to what escapes the containment and not what 4 goes into contninment in this context. 5 MR. TIIROM: I think when we talk about the 6 app 4M2h cf a e chanistic source term, of course, what is 7 important tc, evaluating the goal of the nealth and safety of 8 the public it's what gets out of containment. 9 Built into that is also what is the fuel damage 10

e. mount, timing, what gets out of the fuel and gets into the 11 coolant, what gets out of the primary boundary, what gets 12 into containment, what featutes are in containment that you 13 can credit filters, sprays or whatever.

14 That's what we talk about in a mechanistic source 15 term. What the applicants are talking about is, you don't 16 develop a GO percent core damage source term and apply it to 17 all events. You look at opecific events or categories of 18 events and look at the timing, the amount, what barriers are 19 available to mitigate the_ transport into and out of 20 containment to develop -- for the lack of a better word -- a 21 realistic assessment of the offsite consequences. 22 The source term -- there's also the severe 23 accident source term, one that if it didn't come out of the 24 design base accident evaluation would come out of this 25 credible event that we would try to look at to assess the-Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950

46 1 consequences of core damage if at this point the pre-O 2 applicant thought that that was a very highly unlikely 3 ovent. We think it needs to be looked at. 4 Again, what we would do with it from a perspective 5 of regulatory design change, you would have to know what was 6 really going to happen before you could make any 7 determination on that. I don't think it's something that 8 should be left out of the pre-application review. It helps 9 significantly to understand the overall plant performance. 10 MR. CARROLL: We used a couple of words here -- 11 you have -- credible and highly unlikely. What do those 12 mean? 1 13 MR. WILKINS: He already answered the credible 14 question, didn't he, when Pete asked it. 15 MR. CARROLL: Did you? I am sorry, I left the 16 room for a minute. 17 MR. DAVIS: He talked about it. 18 MR. WILKINS: That would be better than saying he 19 answered it. He talked about it. What was the other one? 20 MR. CARROLL: Highly unlikely. 21 MR. WILKINS: Highly unlikely. 22 MR. CARROLL: Same question. 23 MR. THROM: Right now we are not trying to focus 24 on what a specific numerical value would be. If I thought 25 ten to the minus sixth was highly unlikely you might think Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950

47 1 it is ten to the minus seven. You might think that ten to 2 the minus three, ten to the minus four, is highly unlikely. 3 We are not in a position to try to quantify things 4 at that level of detail at this particular time. Maybe it 5 affords us a little leeway to go a little bit beyond where 6 someone else might want to go in looking at things. 7 MR. CARROLL: Going back in history -- 8 MR. WARD: That's a little bit of a cop out 9 though. 10 MR. CARROLL: Going back in history we started in 11 this business talking about maximum credible accidents, and 12 found that that was a bad word, credible, because to one 13 person something is maybe very credible and to another it's 14 something else. We abandoned that and stopped talking about 15 maximum credible accidents. 16 Now, we are back using that bad terminology. I i 17 guess I have a problem with that. I think we have replaced 18 it ultimately with the safety goal policy, that's what 19 credible is. I guess my recommendation would be that we 20 expunge the word credible from our vocabularies and talk 21 about consistent with the safety goal policy or something 22 like that. Credible caused a lot of litigation. 23 MR. WARD: I think Ed, the problem with saying you 24 don't know enough about these designs -- the decision on 25 some of these policies about what level of probability you (~] ANN RlLEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

I 48 1 are going to exclude or something, can be viewed as O 2 independent of what the design is. If you somehow can 3 derive something from the safety goal policy you come up 4 with a number which is some sort of a guidance. 5 It's difficult, I agree. I don't think it's going 6 to be any easier when you get more detailed design. We 7 haven't gotten to this, but the same policy general issue is 8 going to be facing you. The fact that you have some design 9 details to wallow in maybe is going to obscure it, but it 10 isn't going to solve the problem that you have. 11 MR. THROM: I understand that. What we will be 12 bringing out at the staff level in the SER's for these 13 designs is some of that type of guidance. Right now at NRR (} apparently is not wanting to go with that specific guidance 14 15 at this time. I think in the paper it talks about the 16 commission guidance in establishing what the acceptance 17 criteria is. It doesn't allude to the fact that we are 18 talking about the safety goal or Part 100 or something else. 19 That is what will come out in the SER's on these 20 specific plants. It's built into the process. 21 MR. KRESS: That's what I was getting at when I 22 was asking about those consequence limits. 23 MR. WARD: They are really talking about a risk 24 level -- 25 MR. KRESS: Risk level, risk hase -- Q ANN RILEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

49 1 MR. WARD: Which makes sense. But if you do that, O 2 you have to bite the bullet and say we are going to talk 3 about numbers. 4 MR. KRESS: You can establish those criteria 5 without knowing a thing about the reactor. 6 HR. WARD: That's right. 7 MR. KRESS: They ought to do that. I agree with 8 you. 9 MS. S LOSSON: I think at this stage, we were just 10 trying to do the higher level to get initial guidance, to go 11 through and say this is our approach. Later on as we get 12 further along in the review, go to the more detailed -- 13 MR. WILKINS: As a matter of fact, the paper poses 14 the issue slightly differently than the slide. The paper 15 says should mechanistic source terms be developed in order 16 to evaluate these things. That's a question that has an 17 answer, yes or no and then give some background discussion 18 and so on. i 19 I asked earlier whether that really meant that the 20 pre-applicant had to do the technical work required to 21 understand these three items. I guess they don't have l 22 bullets on them, those bottom three items. Mr. Throm said 23 in effect, yes. I don't think it's fruitful for us to beat 24 them over the head with these numbers yet. 25 They are prudent individuals and they will O ^"" ai'ev * ^ssoc'ATes' 'id-Court Reporters 1612 K. Street, N,W., Suite 300 Washington, D. C. 20006 (202) 293 3950

50 1 recognize that before this clears they will get beat over 2 the head by this Committee with respect to those numbers, so 3 they will be prepared to fend off the well meaning and well 4 reasoned attacks from the Committee. 5 MR. THROM: I would also like to hope that this 6 guidance, when it does get final, is good information for 7 the number five pre-applicant. 8 A lot of it is based on our dealings with these 9 four pre-applicants right now, and trying to apply the 10 policy issues that the Commission has and looking at 1226 11 for guidance in the review. What we are trying to do is, up 12 at a policy level type of venue, identify to potential pre-13 applicants the types of information we would need. 14 The fifth person coming in may not have even 15 thought about we really have to have data to support our 16 contention. We are trying also to get that message across 17 in terms of I think we preface it with the goal here, of 18 trying to establish the policy criteria for the review of 19 the next generation of plants. 20 MR. WILKINS: Mr. Throm, are you about ready to 21 proceed to the next one? 22 MR. WARD: On the release of transport research 23 that is necessary here at Oak Ridge there is some work for 24 the gas reactor. I think you said at Idaho there's work for 25 the metal fuel now. O ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 m.. .n.i. m. r mu u siid

51-1 MR. THROM: It's at Argonne. There's a lot of 2 work on metal fuel. 3 MR. WARD: Specifically addressing this? 4 MR. THROM: Yes. 5 MR. WARD: The release of transport. 6 MR. THROM: Yes. 7 MR. WARD: PIUS is covered by LWR. What about-8 CANDU, have they done this sort of work or do they plan to 9 do this? 10 MR. THROM: What we know right now'is that in the 11 information that we have -- I have to defer to the project 12 manager -- is that what they do for sets of accidents is 13 look at deterministic mechanistic source terms that are 14 predicated on what the course of the event is. 15 We need to look at with them, since they are kind-16 of the last design that came in is, can they_do this? Do 17 they have data-to support -- 16 MR. WARD: You don't-know whether they do or not? 19 MR.-THROM: 'No. 20 MR. WILKINS: .The paper says that the_ staff has 21 not at this time,_ evaluated the CANDU 3-codes and methods. l 22 MR. THROM: Right. 23 MR. WARD: I am wondering about -- the 24 - experimental data _is hard to come by sometimes. 25 MR. THROM:.It's very difficult to generate a -Q ANN RlLEY'& ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 1 Washington, D. C. 20006 (202) 293 3950

_. _ = _ _ - _ _ _ _ _... _ _ _ _ _. 52 1 computer code without comething to verify it against. 2 MR. KRESS: They may have a lot of data on fission 3 product release from their fuel, and it's the same as the 4 CANDU 6 fuel. They have a lot of data on that. 5 MR. TIIROM: I guess what's interesting to us is 6 the design of release paths. The timing of releases is 7 somewhat difforent I think than what we would be accustomed 8 to in our light water reactors. It needs a review. We need 9 to understand what they are doing and how they are doing it, 10 and how they support their contentions. 11 MR. WILKINS: Why don't you proceed. 12 MR. TilROM: Joe Donoghue will be talking about 13 containment for a few minutes, and then I will be back up 14 here to discuss the emergency planning. 15 MR. WILKINS: I think the committee might wish to 16 note that there are two different J. Donoghue's. Their 17 names are spelled differently. As I understand it, they are 18 pronounced essentially the same. Do the J's stand for the 19 same thing? 20 MS. S LOSSON: No. Joe and Jack. 21 (S1idos.] 22 MR. DONOGHUE: Good morning. I am Joe Donoghue, 23 from the Technical Review Section of Ms. Slosson's branch. 24 I am here to talk about containment performance for the l 25 advanced reactors and CANDU 3. i Q ANN RlLEY & ASSOCIATES, Ltd. l Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950

~ _ _. _ _ _ _ _. _ _. _ _ _. _ _ _ _ _ _ _ _ _ _ _ _ _....... 53 1 A little bit of background here. The advanced O 2 reactors, MitTGR and the CANDU 3 design, propose containment 3 structures with designs that are a little bit -- in CANDU 4 3's case and much different in the MHTGR case -- than the 5 essential leak tight structures that we are used to se01..;. 6 The other two designs remain essentially leak tight. 7 Ilowever, they include features or their approaches in some 8 way much different than what we are used to seeing for 9 conventional LWR containments. 10 The changes that we are seeing are a result of -- 11 what we think are a result of what the pro-applicants 12 propose, are a result of the approach taken by especially 13 the advanced reactors as opposed to the CANDU 3; that, 14 accident prevention is more important than accident 15 mitigation. They are arguing more in the line that the 16 advanced features of these designs make strict containment 17 requirements less important than their designs. 18 The advanced reactor policy statement which Ed 19 talked about at some length in his first presentation was 20 supportive of such simplifications and changes of the 21 designs to make them to more simplify the designs. The 22 policy statement at the same time though, also recommended 23 that there be multiple fission product barriers to a 24 containment in its design. 25 (Slide.] Q ANN RILEY & ASSOCIATES, Ltd. L Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

54 1 MR. DONOGHUE: This frames the issue and we are -O 2 trying to focus on the essential leak tight aspect of it 3 which is required by GDC 16, that questior. when posed has 4 some lead ons to other issues such as what auxillary systems 5 are required to support the containment and how the 6 containment is to be tested. Those are strict requirements 7 that we impose now. When the initial issue is posed, those 8 other ancillary things come up. 9 MR. CARROLL: Am I correct in saying that the 10 GDC's are for light water reactors? 11 MR. DONOGHUE: The GDC states in the beginning of 12 Appendix A, that there are proposed for light water 13 reactors. I think in the same part of the Appendix A, it 14 goes on to talk about they could possibly be applied to 15 advanced reactors as well. 16 MR. CARROLL: It's not just fair to take each GDC 17 and say these advanced reactors which are not light water 18 reactors have to meet each and every one of these to the 19 letter. 20 MR. DONOGHUE: Exactly. That's the whole idea of 21 this meeting, I suppose. 22 MR. CARROLL: All right. 23 MR. DONOGHUE: Based on that, the recommendation 24 that we propose is that because the advanced designs 25 consider such different approaches in some cases to Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

55 1 containment function and design and the fact that we don't O 2 really know all of the contributors to the defense-in-depth 3 for the designs and how they are going to perform in 4 containment of fission products, we propose that we move 5 away from prescriptive containment requirements; that it's 6 possible to have a design that could have a containment such 7 that it's different or maybe radically different from 8 current requirements but still do the job with the other 9 features in mind. 10 When we do this we realize that the containment 11 must meet some basic requirements as far as release limits 12 go, it has to do that job for design basis accidents. But 13 for advanced reactors we go on to propose that for a -- this 14 goes back to what Ed was talking about earlier in his 15 presentations -- we propose to select a severe accident 16 event which is going to challenge the containment. 17 Some of the designs have proposed the possibility 18 -- you were talking about credibility but I don't want to 19 use that word -- the likelihood of an accident is so low 20 that they are not going to worry about it as much as they 21 did in the past possibly, what the consequences would be of 22 a release of the containment. But at the same time they 23 have a containment design. 24 What we think we have to do is challenge that 25 containment design. That's why we propose to choose such an Q ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

56 1 event. Once we propose that event the requirements as we 2 depict them here, are that the design be cble to perform for 3 an initial period of time. It has to be able to do what it 4 was designed to do with whatever value the. design proposes. 5 After that, the containment must be able to at least prevent 6 uncontrolled releases of reactivity, to at least maintain 7 some level of safety. 8 MR. CARROLL: The distinction you are making here 9 between the first 24 hours and what the design has to do 10 after that, you don't know at the moment what uncontrolled 11 leakage is. 12 MR. DONOGHUE: Not to any specific extent, no. 13 The first bullet here was taken primarily from words used in 14 90-016 and some follow on. The rationale there is talked 15 about in the severe accident policy which was just put out 16 in the Federal Register. They have to be able to maintain 1/ the integrity long enough to allow the fission products to 18 decay so that if they are released somehow, that it won't 19 cause immediate health effects. 20 MR. CARROLL: I understand the 24 hours. I don't 21 know -- following this period the design must prevent 22 uncontrolled releases. 23 MR. DAVIS: It could be large, as long as the 24 controlled. 25 MR. DONOGHUE: We haven't placed a limit on that O ANN RlLEY &. ASSOCIATES, Ltd. Court Reporters 1612 K. Sweet, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 L

57 1 yet. 2 MR. KRESS: I'm struggling a little with this 3 also. It's perfectly reasonable to think that the MHTGR 4 could make a safety case that it meets the safety goals and 5 well beyond without having any containment at all. Perhaps 6 even PRISM and PIUS almost could do that. 7 How are you going to deal with that concept, that 8 the don't really need a containment to meet the safety goals 9 based on some realistic PRA type analysis. 10 MR. CARROLL: They may need one to be consistent 11 with defense-in-depth. 12 MR. DONOGHUE: Exactly. We are proposing that 13 they have to have some design there. 14 MR. SHEWMON: In both cases though, the main 15 pressure containing boundary is your pressure vessel. I 16 wanted to get off on the bugaboo of before you got into this 17 business was what you could get out of an accident in a fast 18 reactor that would blow things apart. I have the impression 19 that in your design and analysis we aren't talking about how 20 many mega jewels of energy we have to cope with in the 21 pressure vessel itself -- whatever you call that vessel 22 around pot type reactor. 23 I don't know that I want to get into that today. 24 Basically, how are you dividing the containment function 25 between the pressure vessel -- what I will call the pressure (] ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293-3950

58 1 vessel -- this pot in a PRISM reactor and whatever comes 2 outside of that. 3 MR. DONOGHUE: You are using the PRISM example, 4 and they have changed their design since they initially came 5 to the NRC, and included containment boundary. They 6 specifically name it as such and treat it as a containment 7 boundary. 8 MR. WARD: That containment boundary is just the 9 original overflow tank and then they put a sealed cover on 10 it so it has some pressure capacity. But the volume is very 11 small, compared to what you think of it is. What asn't 12 been justified is what -- 13 MR. SHEWMON: The vapor pressure is exceedingly 14 low, too. So, the main pressure you have to cope with has 15 to do with what sort of nuclear event you talk about and not 16 the vapor pressure from over heating the coolant. 17 MR. WARD: That's right. I don't think we have 18 been told what an analysis shows. 19 MR. SHEWMON: My question to him though was, to 20 what extent can we-talk about the containment being a 21 stainless steel boundary around the sodium, to what extent 22 whatever comes outside of that. 23 MR. DONOGHUE: The designer proposes a containment 24 design. What we are trying to accommodate is a difference 25 as major as the one that PRISM has proposed. Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

59 1 MR. SHEWMON: But they are calling it containment, 2 is what comes outside the stainless steel pot; is that what 3 you are saying? 4 MR. DONOGHUE: Outside of the reactor vessel and 5 the primary coolant system, yes. There's another stainless 6 steel pot that we are talking about is the containment 7 vessel; containment dome, and there are some systems that 8 are there that support the containment. That is a 9 containment design. 10 What we are trying to do is accommodate such a 11 different design by,-instead of having strict requirements 12 as we have had for light water reactors where it ends up 13 having a large reinforced concrete structure, instead of (} 14 telling them they have to have that no matter what they 15 design we are proposing that they have -- 16 MR. DAVIS: To me, the critical issue here is what 17 is a deterministically selected severe accident event, and 18 how does one go about selecting that. In reading the policy 19 statement it looks like you need to select something that is 20 even less likely than a design basis accident for current 21 reactor designs. That may put you down into very low 22 probabilities space. 23 I am wondering if there is really adequate 24 information to decide on what that accident really needs to 25 be. I guess you are going to work this out as you go along, O ^"" nev & Associates. 'id-Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

_ _. _ = _.. 60 1 presumably. O 2 MR. DONOGHUE: Exactly. The idea is, yes, to go 3 beyond design basis. We are trying to assess the margins of 4 the designs and ensure that the containment is going to 5 work. If they later on in the certification review the 6 determination is made that yes, indeed, the design is such 7 that there's going to be a very low likelihood of a 8 challenge of release of specific amount of fuel to challenge 9 the containment we still want to be able to assess the 10 ability of the containment to function as it is designed. 11 MR. WILKINS: Of course, this issue is directly 12 related to Issue A that Mr. Throm talked about earlier this 13 morning. I don't know what uncontrolled release is, either. 15 You may want to expand on that when you re-write this. 16 MR. WARD: Pete, the point that you raised, I 17 don't think they have come to grips with it here. Of 18 course, 90-016 didn't come to grips with it either. We 19 commented on that; that, 90-016 presented alternatives. You 20 could either have a CCFP of.1 or this thing up here. 21 I think we pointed out that those weren't 22 necessarily -- it depends on what you mean by CCFP. They i 23 aren't necessarily sensible alternatives. In fact, they 24 might just be appropriate as supplements to each other. If, 25 for example, you look at the 90 percent or one over CCFP, I l l O ANN RILEY & ASSOCIATES, Ltd. O Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

61 1 think in some letters were have tried to make the case and 2 there's been some discussion that what that means is that 90 3 percent of the possible containment challenges will be 4 accommodated by the containment system, 5 That's one way of getting at the se?.ected severe 6 accident event. If you look at that, then the other thing 7 they got up there would just be something at another level 8 below that. If you look at the 90 percent as somewhere from 9 God or something you get deterministically selected severe 10 accident event, then you could say the containment has to be 11 able to survive that 90 percent of the time or this could be 12 a kind of an alternative for that. 13 It depends on what you means by the CCFP. I don't 14 think -- just to continue with this muddle, I don't think-is 15 very helpful. 16 MR. WILKINS: Dave, the letter does have some 17 language addressing the reasons why they did not wish to use 18 the CCFP in this context. I don't know whether you find 19 their language persuasive or convincing -- 20 MR WARD: No, I don't. 21 MR. KRESS: Dave, I will tell you-what bothers me 22 about this. You take four different reactor concepts and 23 you are going to.say each one of them has to have a 24 containment on it, and it has to be well enough designed to 25 withstand some beyond design basis accident. The design ]- ANN RILEY & ASSOCIATES, Ltd. \\ Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3850

62 1 basis accidents for each reactor is sort of arbitrarily 2 chosen, looking at the design and what could fail and what 3 couldn't. 4 Then, you are going to try to say now we are going 5 to make sure it's sufficiently challenged so we are going to 6 choose another arbitrary accident well beyond that design 7 basis and come up with pressure loads and fission products. 8 It's inconsistent. It has no underlying principles that 9 guide you on it. It's arbitrary and ad hoc for each 10 reactor. 11 It always bothers me when I see that kind of lack 12 of principles guiding what you are doing. 13 MR. WARD: I don't think you ought to have this 14 two stage design, design basis and beyond design basis. 15 That doesn't make sense. 16 MR. KRESS: That's what I am saying, I don't think 17 so either. 18 MR. WARD: What you want is a containment that 19 will mitigate most things that can happen but not everything 20 anyone could think of. We have tried to define that in some 21 places by this 90 percent CCFP or 10 percent CCFP. It's 22 very difficult to do when you get down to the details. 23 Still, conceptually, that's the sort of thing you are trying 24 to do. 25 If the requirements get too far away from a Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

63 1 sensible conception you are going to continue to be adrift 2 on this, I think. I don't think this is very close. 3 MR. CARROLL: I am amused by the statement in the 4 draft that says further, the CCFP is grounded in a firm 5 understanding of light water reactor safety systems and 6 accident progression. 7 MR. WILKINS: That may be a stronger assessment of 8 the NRC's understanding than what we have heard before. 9 MR. DONOGHUE: It's at least firmer. 10 [ Laughter.] 11 MR. WILKINS: Am I correct in assuming that this 12 is an area where the staff recommendation may be orthogonal 13 to the pre-applicant's desires? O 14 MR. DONOGHUE: Not at this stage, no. They have LJ 15 all proposed some design features that are going to contain 16 fission products and deal with the fission product release. 17 MR. WARD: They love this, Ernest. If you let 18 them select-the severe accident ovent they can do the second 19 part casily. 20 MR. WILKINS: The MHTGR, they may want to argue 21 that their triso particles are se good that there is no way 22 that any damage can come to the public health and e.afety. 23 Anything you want us to put up as a containment, se will put 24 up a few two by four and call that a containment. 25 MR. WARD: They admit there is some quality Q ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293-3950

64 1 control problem on the particles, and they are allowing just 2 for that. I am having a little trouble with the/c. 3 MR. DONOGHUE: let me interject. The proposed the 4 fuels are going to be very high quality. They do have a way 5 to deal with fission products. They have a volume which 6 does release to the atmosphere; however, they think the 7 source term is so low that there won't be adverse 8 consequences. That's a little different than saying that 9 there is a lack of a containment function. It is definitely 10 much different than what we used to know. 11 MR. CARROLL: That all assumes that the plate 12 doesn't fail and you end up with a uncoolable geometry. 13 MR. WILKINS: I've heard that MHTGR has so much 14 heat capacity that you can go home and come back the next 15 day -- 16 MR. WARD: If it stays -- 17 MR. CARROLL: If air gets in there, it's just -- 18 MR. WILKINS: Start a fire -- 19 MR. SHEWMON: The air comes in and burns on the 20 way in. You have to talk about how the air gets down there. 21 22 There's a lot of graphite that it has to get to before it 23 gets to the bottom of the reactor. 24 MR. WILKINS: We are not going to design the MHTGR 25 for them. ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 i

65 1 MR. DONOGHUE: That's all I have for containment. O 2 MR. WILKINS: Are there any further questions on 3 this one? I would remind the speaker and Ms. Slosson, that 4 Mr. Davis asked earlier about the item on page 10 in the 5 third paragraph, and whether there was an implicit 6 comparison with the safety goals. I don't remember what 7 answer you gave to that, incidentally, when he asked it in 8 the first place. That was some minutes ago. 9 Are there any further questions from the members? 10 [No response.] 11 MR. WILKINS: We are only about 15 minutes late, 12 14 minutes late by my watch. Let's take a 10 minute break, 13 and return at 10:15. That will put us 15 minutes late. ( 14 [Brief recess.] 15 MR. WILKINS: Let us continue. Mr. Throm, you are 16 ready for the fourth item? 17 MR. THROM: Yes, I am. i 18 [ Slides.] 19 MR. THROM: The fourth item which is on the agenda 20 today is emergency planning. The emergency planning really 21 goes back to the original proposals by the DOE and IS 22 designers. With their conceptual design, their accident 23 evaluation, source term containment proposals that they had, 24 they were proposing reductions in emergency planning. 25 What their objective was, was to demonstrate that Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

66 -l 1 the offsite releases were within the lower level guidelines O 2 that the EPA has for protective actions at the exclusionary 3 boundary. Of course, this is for design base accident 4 space. The question is, what is the level of emergency 5 planning that is required for advanced reactor designs. 6 (Slides.) 7 MR. THROM: At this point, the staff is 8 recommending that the licensees develop emergency offsite 9 plans, provide provisions for testing these, consistent with 10 what we do right now. But we are willing to consider 11 relaxation at some later date in the future when we can 12 understand what the accidents are, what the source term is, 13 what containment performance is, to the point where you 14 would feel comfortable with an evaluation that would 15 demonstrate offsite consequences were extremely low. 16 Their proposal is the EPA guidelines which would 17 be one rem whole body and five rem to the thyroid as 18 compared to Part 100-right now, which-would be 25 rem whole 19 body and 300 rem to the thyroid. You can see that there is l 20 quite a bit of reduction in what they are proposing for as i 21 the ofisite dose consequences. As a result of that, they 22 would like some type of relaxation in emergency planning. 23 We think that in the future there may be some i l 24 cause to do that and look at it more closely. The things 25 that we would be considering would be relaxing the need for O ANN RILEY & ASSOCIATES, Ltd. Coud Reponers l 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

67 1 prompt notification, possibly the frequency of drills, and O 2 what the size of the emergency planning zone might be. The 3 approach that we will be taking will be consistent with 4 that, that is being taken right now on the passive light 5 water reactors. 6 We are just not quite there yet at this pre-7 application stage, feeling that we could provide them 8 guidance or feedback that says the Commission is ready to 9 accept that departure. But, we are willing to listen to it 10 at design certification if the accident evaluation or the 11 detailed design, we understand what the consequences and 12 what the source terms for these designs would be, what the 13 containment performance is and make an envelope is and 14 demonstrate to us that at least with these three designs -- 15 the PRISM, MHTGR and PIUS designs -- indeed, you feel that 16 the offsite consequences are significantly reduced. 17 Also, taking into account things of the long-18 delays in releases from these plants. PIUS, for example, 19 has a pool around the reactor that would take seven days to 20 basically uncover the core if you had a loss of heat sink. 21 The PRISM. reactor, you could operate under a severely 22 degraded mode for quite some time before you would get core 23 damage. It's built into the timing of the release, what the 24 release is. 23 We haven't really shut the door on some of the ] ANN RlLEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

68' H 1 proposals that they are asking for. I don't think anyone is (2) 2 truly asking for no planning. I think they would know what 3 had to be done. Local officials would be involved with it, 4 but they don't necessarily want to have the detailed test of 5 actual evacuation of this site. I think that's one of the J 6 key things that they would be looking for in this 7 relaxation. 8 It's an issue that will be ongoing. It's just 9 that we are not here today based on our understanding of the 10 designs, to say that we are willing to basically give them 11 relaxation at this point. 12 MR. KRESS: If you were going to have a criteria 13 that says you either do or do not have to have an emergency } 14 plan, that criteria would have to have some probability base 15 and it would be a risk criteria. Does such a criteria 16 exist? 17 MR. THROM: Today? 18 MR. KRESS: Yes. 19 MR. THROM: I'm not sure. I don't believe so. I 20 think it's within whatever winds up being defined as the 21 design base accident space. 22 MR. KRESS: Generally for LWR's that has some -- 23 you can pin a probability to that. Then, it has these 24-arbitrary source terms associated with this, so it's not 25 all together kosher to tie them together. Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C.' 20006 (202) 293-3950

l l 69 q I have problems knowing how you are going to do l 2 this relaxation without having some criteria to guide you. 3 That's my whole problem. 4 MR. WARD: Yes, or without a program to develop 5 those instead of just waiting for the specifics. What 6 information is available about LWR's in PRA space on the 7 contribution of emergency planning the risk reduction? Do 8 you know anything about that? 9 MR. TH' ROM: No, I don't. 10 MR. WARD: Does anyone know about that? 11 MS. SLOSSON: No, we don't, here. I think at this 12 stage, we are just trying to say that we are not ready yet 13 to make any relaxations. We are not closing the door yet, 14 to say that we will never consider them. 15 MR. WARD: I don't see how you can even think 16 about this without considering what you know now about the 17 risk importance of emergency planning for existing reactors. 18 19 It seems to me that background of information is essential. 20 This is 1993. 21 MR. CARROLL: Some risk studies were done post-22 TMI, to develop whatever the NUREG number is. I don't think 23 they were very rigorous or very good, but at least an 24 attempt was made to give some credibility to the Ten Mile 25 EPZ and all that stuff. O ^"" ni'ev & Associates. 'id-Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

70 1 MR. DAVIS: Yes, there has been quite a bit done. C) ( 2 There have been PRA's that have looked at no evacuation 3 versus optimum evacuation. There is information out there. 4 MR. KRESS: It is really risk significant, if you 5 look at it from that standpoint. 6 MR. CARROLL: Absolutely. 7 MR. WARD: Somebody made the statement earlier, 8 that the PRA's we have indicace that most present plants 9 would meet this safety goal. But, would most present plants 10 meet it without evacuation or with a different sort of -- 11 emergency planning and evacuation aren't quite the same 12 thing, of course. That's what these applicants are saying, 13 you can have evacuation without emergency planning. 14 MR. DAVIS: Yes, I think most of them would meet 15 it without evacuation. 16 MR. WARD: It's something that I think should be 17 discussed and analyzed are argued over in the course of 18 developing these advanced reactor recertification reviews. 19 MR. WILKINS: It's not clear that this branch of 20 NRR is the branch that would do that, though. 21 MR. THROM: No, we are not. 22 MR. WILKINS: You are saying it is clear that you 23 wouldn't. 24 MR. Tl! ROM: Organizationally, we are tasked with 25 trying to look at the technologies and understanding what ]' ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293-3950

71 1 they are, and working at a semi-informal level with staff 2 but at a very formal level with the other NRR managers under 3 Dr. Murley who actually would do the design certification 4 review, to assure that what we are doing is consistent with 5 what they are doing and that we would be in agreement with 6 our positions in not only the policy paper but in the 7 preliminary SER's that go out to the applicant in terms of 8 telling them what our best thoughts are, what we think they 9 would need to make a point or justify in this case for an lo emergency planning, 11 MR. SHEWMON: Do you think you can do that without 12 having specified a set of severe accidents that you are 13 going to examine to and they have to design to? 14 MR. THROM: No. 15 MR. SHEWMON: You will come up with a set of 16 severe accidents before you get done with this? 17 MR. THROM: Yes. Whether or not the pre-18 applicants would be able to evaluate them at this stage to 19 the satisfaction of the staff, it's not clear. I think for 20 example on the Canadian design, we will be requesting some 21 severe accident work from them. Whether they would be able 22 to complete or do a real good evaluation on it in the 23 timeframe we had put up to get a pre-application SER out is 24 questionable. 25 Clearly, the pre-application SER would indicate Q-ANN RILEY & ASSOCIATES, Ltd. k Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

. ~ - 72 I 1 that the staff is looking for this kind of information in 2 order to help them make that determination. 3 MR. SHEWMON: Since there will be other kinds of 4 severe accidents for the other types of reactors, you also 5 specify or reveal some severe accident sequences there too? 6 MR. THROM: I am not sure that I understand the 7 question. 8 MR. SHEWMON: You said you would do this for 9 CANDU. We both agreed that -- 10 MR. THROM: All ot them. I am just relating a 11 present experience right now. 12 MR. SHEWMON: Fine, okay. 13 MR. THRGis' In dialogue with the -- 14 MR. SHEWMON: At some page here will I find the 15 list of these severe accidents, or does that come later? 16 MS. SLOSSON: It comes later. 17 MR. WILKINS: The promise that this will be done 18 is part of Item A. 19 MR. THROM: Yes. 20 MR. WILKINS: If I remember correctly. There are 21 some characteristics of these accidents and so on that are 22 listed as the recommendations. 23 MR. PIERSON: This is Bob Pierson, from the 24 Standardization. We are working on the advanced light water-25 reactors. Let me. clarify that a little bit. Q ANN RlLEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950

73 1 We are working on a couple of papers that we call 2 90-016 and the subsequent follow on papers, which we were 3 trying to give to the Commission to give them some input on 4 the policy issues associated with the advanced light water 5 reactors. This issue for emergency planning is one of the 6 central issues for the passive light water reactors. It's 7 one of the issues that we hope to follow up with a 8 Commission paper sometime in the future, tying that together 9 with the source term. 10 At that time our emergency preparedness people and 11 our people who are doing radiation evaluation source term 12 evaluation will have some sense of what values they want to 13 use to bound this. I think that it's probably simplistic to 14 think that we are going to have a series of numbers and say 15 if you got this it's too miles and if you get that it's one 16 mile. I am maybe incorrect about that. 17 I think what it will probably come down to is that 18 there will be some goals, some guidelines that we have 19 established and ultimately come down to just a fundamental 20 policy issue that the Commission is. asked to make a 21 judgment, do you want to do this or do you not want to do 22 this. 23 MR. WARD: You sort of denigrate the idea of 24 specific quantities being used, that it's simplistic. But l l 25 somehow the Commissioners have to make up their mind based i ' Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293-3950 l I

74 1 on some understanding of things just very much like those 2 numbers. 3 MR. PIERSON: The thing that I think is important 4 -- and I am just very peripherally involved with this 5 because we are just in the beginning development of this -- 6 it's going beyond the technical issue. It's a political 7 technical issue. I think it becomes more involved than just 8 simply extrapolating numbers and saying you have to abide by 9 a certain criteria for certain numbers. 10 MR. WILKINS: I sort of hoped that Charlie was 11 going to be here today, because he's concerned with some of 12 these passive LWR policy issues. This is one of those. 13 MR. PIERSON: Unfortunately, this one is not going ( 14 to be included in the summary paper that we are preparing 15 right now. This is going to be in -- 16 MR. WILKINS: This is being held for still later. 17 MR. PIERSON: That's right. These people are 18 probably the wrong people to pick on. You should probably 19 wait until he comes down. 20 MR WARD: Except that, they are talking about the 21 need or issue of emergency planning. 22 MR. WILKINS: Their proposal is for right now, 23 that they will require emergency planning. 24 MS. SLOSSON: That's correct. 25 MR. PIERSON: Their proposal is consistent with (O ANN RILEY & ASSOCIATES, Ltd. COud RepOners 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

- _ = 75 1 what we are doing on light water reactors. All I am saying -O 2 is that there is a provision and we are working on a paper 3 where we are examining these issues right now. 4 MR. WILKINS: They leave the door open in case 5 future developments may make it possible, future information 6 and future developments may make it possible to relax the 7 requirements. 8 If I were a pre-applicant, I am not sure I would 9 be completely happy with this recommendation. I would 10 believe that I had given you enough information to make up 11 your mind. 12 MR. WARD: That's it, what future information. 13 MR. WILKINS: What more do you want, yes. That 14 gets to the question of what are your criteria. 15 MR. THROM: We are beating the horse here again. 16 They have proposed a set of accidents that they have 17 evaluated, a source term that they think is reasonable and 18 realistic. They have a concept for a containment, and they 19 have elected to use the lower level EPA guidelines as a 20 marker. 21 With what they have submitted they demonstrate 22 that. The question is, have they missed any accidents in 23 design basis space. Have they got the correct source term. 24 Is their containment performance what they really expect. 25 Not only is the lower level EPA guidelines the right marker Q ANN. RILEY & ASSOCIATES, Ltd. Coud Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

76 1 to use but what would happen if, after they finished their O 2 R&D program and they found out the source term was about a 3 factor of ten higher than they thought -- a lot of them are 4 using very low source terms. 5 Now, they are back up into perhaps Part 100 space. 6 7 Maybe at that position they might say we really can't 8 demonstrate that the types of reductions in emergency 9 planning would be justified. 10 I think we need to be able to put the entire 11 picture into perspective to understand that within the 12 design space on design review, that they can meet those 13 criteria and evaluate those at that time. We don't close 14 the door on them. But if they were asking for a position 15 today, we would have to say we are not comfortable with 16 going out right now and saying we think you can do it or 17 going to the Commission and asking for that type of a 18 change. l 19 MR. KRESS: The protective action guidelines that 20 are on the books now are strictly for present day LWR l 21 reactors. If one were to go to a PRA for the various 22 reactors and tie to the risk level at which they meet those l l protective action guidelines, one.could end up with a risk 23 24 level below which that guideline represents for the present 25 day reactors. {' ') ANN RILEY & ASSOCIATES, Ltd.. Coud Reponers l 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

77 1 That would be a criteria then, where if those O 2 present day reactors met that risk level they would be 3 excluded from emergency planning. You have built in your 4 system already a criteria that says you would, however, not 5 have an emergency plan. Why not extend that and be 6 consistent, atid say the same thing for the advanced 7 reactors. Let them tie it to a risk level that would be the 8 same risk level. 9 It's built into the system already, and why not 10 just go on and extend it that way? 11 MR. CARROLL: I think you have to separate this 12 thing into the technical risk regime and the political 13 regime. I am afraid that these reactors are ultimately 14 going to end up if they are ever built, with the same sort 15 of emergency planning even though it doesn't make sense from 16 a risk point of view, that we have on light water reactors. 17 I don't think the public is going to accept the 18 idea that if they can do it for these kind of reactors, why 19 can't they do it for the reactor they are building in my 20 back yard. 21 MR. KRESS: Why do we have PAG's then? 22 MR. WARD: That may be true, but it seems to me 23 it's the responsibility of the technical community, 24 especially the NRC and ACRS, to present the best technical 25 case that can be made. If a different political judgment O ANN RILEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street,' N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

78 1 has to be made at some point, so be it. \\# 2 MR. CARROLL: I agree. 3 MR. WARD: We shouldn't cop out on doing our job 4 because we say hey, there's no point in it because of the 5 politics. Do our job the best that we can, and hope that 6 will make the politics more sensible. 7 MR. CARROLL: In that area I am encouraged. I 8 read the prospectus of the National Academy study that was 9 floating around in the last month or so which is going to 10 try to bring together people to look at the question of how 11 do you get the public to accept technically based risk 12 arguments better. 13 I certainly think that's a major issue. It's not 14 just dealing with nuclear power. It's technically based 15 risk arguments, in all aspects of our lives. 16 MR. WILKINS: Like second-hand smoke, for example. 17 18 Let's continue. 19 MR. THROM: That completes this particular 20 discussion. I believe Dino Scaletti is next, to talk about 21 reactivity control. 22 MR. WILKINS: Before you walk away, this is an 23 area that did not apply to CANDU. The reason for that is 24 because CANDU is so similar to the -- 25 MR. THROM: PIUS. Q' ANN RlLEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293-3950

79 1 MR. WILKINS: Wait a minute. I am right, CANDU. 2 MR. THROM: CANDU has not come in at least to the 3 best of my knowledge, indicating that they were looking for l 4 any reduction in emergency planning. I think they plan to 5 be consistent with the current practice. 6 (Slides.) 7 MR. SCALETTI: My name is Dino Scaletti. I am the 8 NRR's Project Manager for the PIUS reactor design which is 9 in for a pre-application review. The issue before us now is 10 reactivity control. As you see here by this little policy 11 issue applicability chart, PIUS is the only one that is in 12 that category. 13 What I am trying to do is just briefly go over the ( } 14 background, the issue, and then go over quickly some of the 15 details of the PIUS design in the reactivity control systems 16 just very grossly. There has been some changes in PIUS 17 since the paper was written, and I will highlight the 18 changes in the next slide that I have coming up after these. 19 The issue is rather straightforward. The issue 20 is, should the NRC accept and review a design that does not 21 have control rods. GDC 26 requires that design shall have 22 two reactivity control systems, one of them shall be control 23 rods. The other one shall be capable of planned reactivity 24 changes and also not protect the fuel and don't exceed the 25 fuel limits. O ^"" ni'ev a Associates. 'id. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

80 ,f s The staff have looked at the PIUS design, and our 1 2 recommendation is that yes, we believe that a design without 3 control rods may be acceptable if the applicant demonstrate 4 the equivalent level of safety given in these existing 5 categories here. That part of the review has not been 6 undertaken yet. We plan on doing that when we have all the 7 information before us, which is scheduled to review is 8 officially scheduled to start sometime in the near future. 9 As I told you we have some changes that are coming 10 to us from ABB Combustion Engineering relative to the scram 11 systems. The PIUS design as it previously existed, utilized 12 -- they didn't have the scram valve. They utilized the main 13 coolant pump trip to disrupt the thermal density locks here } 14 and got a natural circulation of highly borated water 15 circulating through the core and shut the core down to 16 control the reactivity. 17 They also had the operational control was you 18 could inject boron into the primary loop and could have a 19 control shutdown that way. 20 They have added to the design, scram valve. This 21 is one of the loops. There are four loops. There are, in 22 essence, two scram valves in each loop, single failure 23 proof. There are eight scram valves in all. These valves 24 are automatically actuated as well as can be manually 25 actuated, is our understanding of it. These. valves would be Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

.. _. _. - -. = -. 81 1 activated -- if you wanted to scram the reactor they would 2 be activated for a three minute period, in which time 3-reactivity would be shut down in the core, thermal 4 transients would be reduced as opposed to using this 5 mechanism here and the reactor would be shut down. 6 MR. SHEWHON: Would you take me by the hand 7 through that? It looks like the scram valves only bring 8 water into what is your normal roturn from the steam 9 generator anyway. How does that scram? 10 MR. SCALETTI: This is a highly borated pool. 11 MR. SHEWMON: I understand that. 12 MR. SCALETTI: In essence, about 2,880 gallons of 13 borated -- 14 MR. SHEWMON: The feed to the scram valve is 15 borated water, fine. 16 MR. WILKINS: He says you can stop. 17 MR. SCALETTI: Okay. 18 MR. DAVIS: What was the inspiration for naking 19 this design change. Was it something that the staff 20 questioned, or was it a totally voluntary change, or was 21 there some -- 22 MR. SCALETTI: It was a totally voluntary change. 23 Other than asking questions we are doing some -- some 24 evaluations are being done at some of the national labs. 25. The PSID for PIUS shows that on scram there are some ] ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

82 1 oscillations using the pool itself. This way you smooth off 2 the oscillations and you reduce the thermal transients in 3 this part of the reactor. 4 It's quite a temperature difference between the 5 pool and the -- 6 MR. DAVIS: Whose calculations showed the 7 oscillationsi 8 MR. SCALETTIi Both the calculations done by ABB 9 and the calculations done by Brookhaven. 10 MR. DAVIS: Thank you. 11 MR. WILKINS: lias there been any experimental 12 testing. It would have to be nuclear testing, but any 13 experimental testing of these density locks? 14 MR. SCALETTI: Yes. We were just recently in 15 Sweden in November. We had an opportunity to observe the 16 magna loop which is a proof of principle loop that ABB has 17 set up over there tc demonstrate that the thermal density 18 locks do work. This is done at elevated temperatures. I 19 think atmospheric pressure was slightly above - 'ertain 20 pressure. 21 They do elso have a test loop which is I believe 22 approximately - 8.t is not a full scale to the PIUS design, 23 maybe 100 feet in length -- which they have done their 24 testing of their validation of their code against the PIUS l l 25 design. There has been work done and has been other work (' ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 i \\

83 1 done. O 2 MR. WARD: Is the speed of the scram -- could you 3 give un an idea of the scram rate or the reactivity 4 reduction rate for the old systems with the new systems. 5 MR. SCALETTI: I don't think there -- my 6 understanding is that with the scram valves, even though 7 there is apparently a delay time in getting the boron to the 8 core from the pool, it overall -- my understanding is that 9 it is quicker than using the pool. 10 Exact times, I don't know. We don't have that 11 detail. 12 MR. WARD: Do you have any rough idea? Are wo 13 talking about minutes or seconds? 14 MR. SCALETTI: I can't answer that question. 15 MR. THROM: I am trying to remember some old 16 calculations that we looked at. I think with the density 17 locks you were getting the scram on the order of ten to 20 18 seconds. I think with the scram valves here, in the few 19 second range. It's like an order of magnitude. I think it 20 was in the ten to 20 second range when using the density 21 locks to get the scram functioning. Of course, it's only 22 the loop transport time now with the full pump -- with the 23 valves, a couple of seconds I believe. 24 MR. WILKINS: How does that compare with typical 25 LWR scram times usi.ng control rods? Q ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 - Washington, - D. C. 20006 (202) 293-3950

84 1 MR. SCALETTI: Much sirwer. 2 MR. WILKINS: Is it fast enough? 3 MR. SCALETTI: That's one of the things we are 4 trying to hopefully determine through the course of the 5 review. There is being work done. We have thermal 6 hydraulics modeling being done at Los Alamos. Brookhaven is 7 doing some work for us. We will make a decision on whether 8 it's fast enough. 9 Right now, as I said before, the scram valves are 10 on for three minutes. At that time they have reached the 11 reactor scram. It is a longer period of time than control 12 rods. The exact amount of time, I don't know yet. 13 MR. DAVIS: Does this system depend on the 14 continued operation of the mbin coolant pump? 15 MR. SCALETTI: The spare valves would, yes. 16 MR. DAVIS: It's not a coast down thing. 17 MR. SCALETTI: I don't have those details yet. 18 But it would -- 19 MR. DAVIS: For most IRR transients the main 20 coolant pump will trip, but I presume this is entirely 21 different. Is it a safety grade typt system that has an T 22 emergency power to the pump? 23 MR. SCALETTI: They are trying to find that 24 question out now. I don't have that. Given that they have 25 designated -- and the information that I have -- they have l O ^"" ai'ev a Associates' 'ta-Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D.- C. 20006 (202) 293-3950 l

85 1 designated these as safety grade single failure proof would O 2 almost imply that the reactor coolant pumps were safety 3 grade. For sure, that hasn't been identified yet. In the 4 old design they were not. 5 MR. DAVIS: I knew that. That's why I was asking 6 the question. 7 MR. WILKINS: The present draft of the paper says 8 that a failure of the density locks would not only prevent a 9 scram but would interrupt the only safety grade core cooling 10 mechanism. Is it a reasonable inference that the presence 11 of the scram valves is intended to ameliorate that problem? 12 MR. SCALETTI: That's correct. It's one of the 13 stated reasons, that it would reduce the ncrd for the scram 14 valve. They could be totally blocked and still scram the 15 reactor. 16 MR. CARROLL: Yes, but you wouldn't get cooling. 17 MR. SCALETTI: Yes, you wouldn't get cooling. 18 MR. CARROLL: If they were totally blocked. 19 MR. SCALETTI: Yes, you are right. You wouldn't 20 have the -- 21 MR. WILKINS: There's no flow someplace. 22 MR. SCALETTI: You wouldn't have this recirculated 23 principle here, the ultimate safety supposedly of the PIUS 24 design if you blocked the pool, that's correct. 25 MR. WILKINS: What you propose to do is to say Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 l Washington, D. C. 20006 (202) 293-3950

86 i show me, basically. O 2 MR. SCALETTI: Presently, yes. 3 MR. WILKINS: It's some legal questions but I 4 don't want to get into those. I didn't know what efficacy 5 meant when you said reliability and efficacy of the scram 6 function. 7 MR. SCALETTI Those are someone else's words. 8 Efficiency of scram, I guess. The way it works. 9 MR. WILKINS: That will deal with such matters as 10 scram rates or reactivity reduction rates and things of that 11 sort. 12 MR. SCALETTI: Yes. 13 MR. WARD: Do you have any idea of whether -- 14 there are eight valves I suppose, and seven of them will do i 15 the job. Will one of them do the job? 16 MR. SCALETTI: You mean, one valve and one loop. 17 one valve in each loop is supposedly will do the job, I 18 understand. 19 MR. WILKINS: One valve will -- I see what you are 20 saying. It's not obvious that you need all four loops. 21 MR. SCALETTI: Those details, we don't have yet. 22 We are expecting a submittal which would depict these 23 details sometime in the very near future. 24 MR. THROM: Maybe I can help Dr. Wilkins keep us 25 on track. The question is, should we be looking at a design ]\\ ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

87 1 without control rods. The things that have to be considered O 2 in that, that's the fundamental question that we want to get 3 answered by the Commission is, should we look at design 4 without control rods. 5 We need to have that allowance, we think, at a 6 policy level. Then, we will go on with the detailed review 7 of whether the pumps need to be safety grado, how many of 8 the loops, how many of the valves, whether it's physically 9 feasible system, what the failure of the density locks would 10 be. Maybe that's the severe accident space for this plant. 11 The quantion is, should we be looking at a design 12 that doesn't have one of the attributes that is called out 13 in the GDC's, namely control rods. 14 MR. CARROLL: In the light water reactor GDC's. 15 MR. THROM: Light water reactor GDC's. 16 MR. CARROLL: Actually, it's the light t'ater 17 reactors of the type that are presently being built or 18 something like that. I notice on your matrix that you have 19 an "x" for reactivity control. You have no "x" for positive 20 leak coefficient. 21 How do you rule that out in a PIUS? 22 Mh. SCALETTI: We are told that it does not have. 23 The PSID clearly states that it does not have a positive -- 24 MR. WILKINS: There will be a presentation by 25 someone else. Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

88 1 MR. SCALETTI : Right. /- 2 MR. WILKINS: On positive void reactivity. 3 MR. CARROLL: Yes, but it's not in that 4 discussion. 5 MS. S LOSSON: It's not part of this. 6 MR. CARROLL: Simplistically, boron is a poison. 7 If you displace it with a steam void it seems to me you are 8 adding positive reactivity. 9 MR. WILKINS: The boron isn't there normally. 10 MR. CARROLL: Yes, it is. 11 MR. KRESS: Around the outside. 12 MR. WILKINS : It's on the outside. 13 HR. SCALETTI: There is a certain amount of boron. 15 I think in the beginning of the fuel cycle it's like 300 16 plus parts per million concentration in the primary loop. 17 MR. KRESS: You are right, it's a balance between 18 the absorption characteristics of the coolant as opposed to 19 its moderating characteristics. It's-a balance between 20 those, whether it's positive or not. 21 MR. WARD: One time early in the cycle it was a 22 potential for -- 23 MR. KRESS: Yes. 24 MR. CARROLL: I would like to hear an answer to 25 that question eventually. O ANN RlLEY & ASSOCIATES, Ltd. Coud RepOders - 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

89 1 MR. SCALETTI: Okay. .O 2 MR. WILKINS : Are there any other comments on this 3 issue? 4 (No response.] 5 MR. WILKINS: Let's move on to the next one. 6 MS. S LOSSON: Ed Throm, even though your agenda 7 says Beth Wetzel, I believe will make this presentation, 8 (Slidos.) 9 MR. THROM: The next issue deals with operator 10 staffing and function. It's going to take me a little while 11 longer to get through this one, because I'm not as familiar 12 with the details as Both would have been. 13 Basically, within our current requirements in () 14 50.54, there are requirements for the number of operators 15 you need at a reactor to safely operate the plant. The 16 reactors that we are looking at are proposing slightly 17 different number of operators per module, power block, 18 whatever you want to couch the design as. The modularity of 19 the designs would tend to perhaps reduce the number of 20 operators in terms of what you would find in 50.54 in terms 21 of the number of senior reactor operators that you would 22 need at the station. 23 MR. CARROLL: Does 50.54 have a light water 24 reactor of the present design caveat -- 25 MR. THROM: Absolutely. I am not sure. I guess O ^"" ai'ev a Associ^Tes' 'id-Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 w m y

90 1 what it basically deals with is the type of reactors we are O 2 used to, largo, multi -- 3 MR. CARROLL: I understand, but does it have the 4 caveat expressed in words. 5 MS. SIMSSON: I don't think we know the answer to 6 that. 7 MR. CARROLL: You may need to change the rule if 8 you are going to consider this. 9 MR. THROM: That's part of the issue. Now, what 10 the applicants are basically proposing is to look at what 11 the operator needs to do best based on a task function 12 analysis, look at the design, look at the things that could 13 go wrong in a unit, how many things could go wrong in a 14 certain period of time, start evaluating the human machino 15 interface from that perspective and studying what level of 16 operator actions would be needed for certain scenarios. 17 Then, to look at the way they are proposing to 18 control them from the control room, be it for example PRISM. 19 20 Right now, we would say they would have one reactor operator 21 monitoring the performance of three nuclear reactors that i 22 made up a power block which supplied a single turbine. 23 The question there is, in a faulted unit who is 24 responsible for monitoring the faulted unit while taking 25 care of the-other two units in that power block that might LQ ANN RlLEY & ASSOCIATES, Ltd. I Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

91 1 be working, getting them to cold shutdown or whatever is 2 deemed necessary to be done on those. When you get into 3 PRISM if you went to a full module plant you have three of 4 these control units, how many people do you need to 5 potentially monitor all of the permutations that could go on 6 based on what the operating specs for the power plant would 7 be. 8 We need to be able to offectively respond to worst 9 caso power measurements, refueling maintenance activities, 10 accident conditions. 11 MR. DAVIS: I have a problem with that statement. 12 What do you mean by worst case accident conditions? Are 13 those worse than the -- 14 MR. THROM: Worst case power maneuvers. 15 MR. DAVIS: Right. 16 MR. CARROLL: Worst case modifies maneuvers only? 17 MR. THROM: No. I believe it's only supposed to 18 modify the first. 19 MR. DAVIS: In what accident conditions does that 20 refer to design basis accidents? 21 MR. THROM: Yes. DBA's -- suppose you had in a 22 unit you have three reactors. Suppose one of the units 23 develops a fault. A pump trips, there is an inadvertent 24 scram, there's a break. If you have one person there who is i 25 looking at these three units and one of the units goes to -- Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293 3950 l {~.

_... _ ~ _ _... - - i 92 1 I will call it a faulted conditions -- who is responsible 2 for that unit, maintaining it in a safe condition and 3 monitoring its perfarmance. I 4 For instance, who is now looking at the other two 5 that would be possibly associate 1 with the same control 6 panel. 7 MR. DAVIS: Just a comment. I have a problem with 8 using worst case, because that means different things to 9 different people. In the extreme it could be the worst 10 possible thing you could think of in terms of power 11 maneuvers, independent of the likelihood. 12 That may be a more severo case than you want them 13 to consider. 14 MR. THROM: I am not sure that's what is in here. t 15 That is under operating a plant. A load rejection -- 16 MR. LAVIS: Also, in the recommendation itself, 17 the wording is that the applicant must demonstrate and 18 confirm through test and evaluation these following 19 conditions. How do they do it by test? Is this something 20 you expect them to do on a simulator? 21 MR. THROM: Yes. That's the bottom line. At the 22 initial conceptual review you would say okay, if a person 23 was -- for example the DOE and Oak Ridge have right now a 24 computer base simulation which uses a 19 inch monitor. It 25 puts out caricatures. What they can do on that is O ANN RILEY & ASSOCIATES, Ltd. Court Reporters i 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950 .-g+-m-n< r-- - - --- y yypm-,y,--,-y ~,- y 9 ,,,y y ,,,,n,-w,-pneg.- 4 y-,w- , age.emvm

93 1 demonstrate the functionality of the plant control system O 2 and to some respect the operator interface. 3 He is interfacing with the computer terminal so to 4 speak at that time. We think it's necessary that once you 5 have done that, you have identified the task function 6 workload that the operator needs to handle, then you need to 7 go to a prototype simulator. 8 MR. DAVIS: That includes all modules, 9 simultaneously? 10 EMR. Ti! ROM: We would have -- if necessary, we have 11 to look at that when it comes in and we get a clearer 12 picture of what they are trying to do. If there is multi-13 module interactions they would need to be able to do the 14 multi-modulo demonstration on a simulator or prototype. 15 If there were multi-unit interactions and they 16 were part of the expectof metrology of handling operators, 17 they would have to add into that the additional control of 18 the other modules to make sure that the control room 19 operations were understood and supported by their design 20 contentions, 21 MR. DAVIS: Thank you. 22 (Slides.) 23 MR. TilROM: On the second slide there is a couple 24 of items there that are a little bit more specific to at l 25 least some of the things that we know right now need to bc l ! Q ANN RlLEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 l

94 1 included in these task and function analyses. That's the O 2 ability to take all of the site reactors down to safe 3 shutdown during complete loss of computer control 4 capability. 5 Most of these designers are going to be using a 6 computer controlled system for their plant safety and some 7 reactor protection. Complete station blackout and design 8 basis seismic events. The bottom line here is -- 9 MR. CARROLL: You have left out something very 10 important. The size of operating crews in most of our 11 existing plants is dictated by fire brigade requirements. 12 You should have that on your list. 13 Assume a fire brigade is out or assume you have to 14 constitute the fire brigade to deal with a fire. Are there 15 enough people left to safely shut down the plant under those 16 conditions. 17 MR. THROM: I think we would be conscious. I 18 think what we are looking at right now here is, who is 19 necessary in the control room. If part of it is that you 20 are expecting some of these operators to do that, then we 21 need to look at and try to identify what the minimum 22 staffing in the control room has to be. 23 MR. MICHELSON: I think the point was that fire 24 has to be added as one of your events, during which you must 25 have adequate control room -- O ^"" ai'ev a Associ^Tes. 'id-Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

1 95 1 MR. TilROM: Okay. 2 MR. MICllELSON : And then, they will figure out how 3 to do it. 4 MR. TilROM ' Right. 5 MR. CARROLL: Typically, you do take people away 6 from the control room to establish the fire brigade in 7 present day plants, i 8 MR. THROM: I didn't mean this list to be all-9 inclusive. 10 MR. WILKINS: Your point is that after you have 11 done that, you still have to be able to operate -- 12 MR. DAVIS: Simultaneous occurrence of the fire 13 plus one of these events, is that -- 14 MR. CARROLL: No. As a result of the fire, you 15 may want to be able to shut down one or more of the modules. 16 MR. SiiEWMON: The one operator can't go and fight 17 a fire. 18 MR. CARROLL: That's a good point. That's the way 19 to say it. 20 MR. WILKINS: Particularly if he's the shift 21 supervisor. 22 MR. MICl!ELSON: They provide a separate man fire 23 support. 24 MR. CARROLL: Some present day utilities are 25 moving in that direction. Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

96 1 MR. MICHELSON: These skinny designs for manning, 2 it seems unlikely they will have a fire brigado around full l 3 time. 1 4 MR. CARROLL: That's correct. 5 MR. DAVIS: Item two on page 18 does in fact 6 include taking to a cold shutdown condition from a variety 7 of potential operating conditions including a fire in one 8 unit. It seems to me like it's covered there. 9 MR. CARROLL: It wasn't on this list. 10 MR. DAVIS: No. 11 MR. WILKINS: It's not on the slide but it's on 12 page 18. 13 MR. MICHELSON : Does it require multiple units be 14 taken down for a fire in one unit. 15 MR. WILKINS: What is the significance of the less 16 one in that same sentence? An accident can be mitigated 17 with the proposed number of licensed operators less one. It 18 wasn't clear to me what happened to that one. 19 MR. MICHELSON: Heart attack. 20 MR. WILKINS: If that's what they meant and that 21 is a possibility -- 22 MR. CARROLL: I don't think that is covered,-Pete, 23 because it doesn't recognize the fact that you are going to 24 constitute your fire brigade from your operating crew. 25 MR. DAVIS: You may not have to. Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950 l

1 97 1 MR. CARROLL: Why? 2 MR. DAVIS: You may have a separate fire brigade. 3 MR. CARROLL: I doubt it. I mean, these guys are 4 trying to cut down on costs by reducing the number of people 5 on shift. 6 MR. DAVIS: To me, it's covered. 7 MR. CARROLL: I think they have a -- 8 MR. WILKINS: Particularly when they are sitting 9 around just waiting for a fire to happen. 10 MR. CARROLL: I think they have a real conflict 11 with the fire brigade problem. 12 MR. MICHELSON : How do they constitute their fire 13 watches and whatever, if they have difficulty. 14 MR. CARROLL: Those typically are operators. 15 MR. MICHELSON: No, but they have to have enough 16 people and warm bedies around to do it. 17 MR. CARROLL: They put in ten years from now, 18 equivalent of thermal lag insulation. 19 MR. THROM: There will be a whole new set of 20 lessons learned, yes. Again, the question here is as we 21 see it, we think that the advanced reactor designers are not 22 quite consistent with what we think is the requirements for 23 staffing. Based on the design characteristics of 'he plant, 24 the modularity, the multi-module units, we are basically 25 asking the Commission if it's okay to let the people put O ANN RlLEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

98 1 together a tasking function analysis to demonstrate what the O 2 correct level of staffing needs to be. 3 Then, to assure that they do have plans to test 4 this not only in a controlled environment perhaps computer 5 simulation but to commit to some type of prototype testing 6 on basically a hands on type simulator. 7 The number of units that need to be handled will 8 be design dependent and based on what we think is the types 9 of situations that a power station could get into, be it 10 multiple power blocks or only within modules in a power 11 block. 12 MR. WILKINS: I note that 50.54 now requires that 13 a senior operator must be present in the control room at all 14 times. As you have described the position of the pre-15 applicants, that possibly could not happen. I don't say it 16 couldn't possibly happen but it possibly couldn't happen. 17 If there was only one senior licensed operator as there is 18 for the MHTGR and he had to leave the room for -- then, he 19 couldn't leave the room for any purpose. 20 There is a policy issue there. 21 MR. THRoM: Yes. I think the continuation of that 22 statement is where I might see more at variance, where if 23 you have three power producing modules -- for example in 24 PRISM -- only one individual at the controls, one senior 25 reactor operator per three fueled nuclear power units. ' Q ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

99 1 That's kind of where we also get into -- O 2 MR. WILKINSt llow does 50.54 address that issue? 3 MR. WILKINS They will probably never have to 4 deal with it. 5 MR. WILKINSt Control rooms only control one unit. 6 MR. TilRoMt Right. 7 MR. WILKINS: At present. 8 MR. THROM Right. 9 MR. CARhoLL: No, that's not true. 10 MR. WILKINS I didn't think it was true. 11 MR. THROM I guess there are some reactors out 12 there that have the mirrored control rooms for dual units. 13 They have their independent staffing, and they should be 14 able to work either unit. I think the question here is a 15 little bit different than that in terms of interpretation. 16 Again, what we are trying to do is go into the 17 current regulations and look for those areas where we think 18 that there's a potential problem that needs to be raised. 19 This is one, where it would appear that the staffing that 20 they are proposing is not quite consistent with 50.54. We 21 think that they should be allowed to continue their R&D 22 programs to look at what's going to be required of the 23 operators and what is the proper staffing, and to be able to 24 demonstrate it on an actual prototype type of control room. 25 MR. WILKINS: You would answer the question raised O ^"" ai'Ev 5 Associates' 'td-Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C, 20006 (202) 293-3950

100 1 in the draft paper, should the designs be allowed to operate 2 with the staffing component less than that currently 3 required, as possibly. 4 MR. TilROM: Yes, if they can manage it. 5 MR. WILKINS: It will be up to the applicant at 6 some point, to demonstrate that this is reasonable and does 7 not produce a decrease in public health and safety somehow. 8 MR. THROM: Right. 9 MR. WILKINS: Are there any further quections on 10 this item? 11 [No response.) 12 MR. TIIROM: We have Joe Donoghue again, on 13 residual heat removal. 14 (Slides.) 15 MR. DoNOGHUE: The next item here is residual heat 16 removal systems for the advanced reactors. Several of the 17 3dvanced reactor designs propose completely passive heat 18 removal systems, completely passive meaning that they 19 require no operator action and there are no components that 20 need to be activated to have the systems operate. 21 Essentially, always operating when the plant is on line and 22 always removing some amount of boat from the reactor. 23 Because of that, the designers are proposing that 24 they are able to continuously monitor their performance of 25 these systems, ensuring that some amount of heat is being O ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1812 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293 3950 l

101 1 removed so that the system can operate when a demand to 2 remove decay heat is going to be required. Their design 3 proposals include the argument that this will help allay 4 reliability concerns in heat removal requirements. 5 There are some related issues for passive LWR's. 6 I listed a couple of those here. It's not an all-inclusive 7 list at all. Those lusues may or may not, depending on the 8 design that is being considered, may or may not be 9 applicable when the designs come under certification review. 10 11 The concerns are centered around diversity, testability of 12 the system, reliability. These are things as I said, the 13 designs have tried to address. 14 [S1idos.) 15 MR. DONOGHUE: This poses the issue that we are 16 trying to specify or concentrate on, the RilR systems -- not 17 so much the ancillary issues that are being considered in 18 passive LWR's -- whether or not a completely pacsive system 19 in a single system, a safety grade system, can be relied on 20 for residual heat removal. 21 There are arguments raised about some of the 22 designs, whether or not they are diverse enough and how they 23 are designed. We are not going to answer that question 24 here, today, but it is a question that has been raised; 25 whether or not for example a design with four heat removal O ANN RILEY & ASSOCIATES, Ltd. i Coud Repoders l 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 i (202) 293-3950 i

102 1 panels that use a natural convection of air is a diverse 2 system. 3 MR. MICHELSON: Don't you have to decide what kind 4 of external events are credible before you decide whether a 5 single train is adequate? What are your thoughts about 6 external events? 7 MR. DONOGHUE: That's definitely a concern, and 8 that's one of the issues that was left off the list. As I 9 said, the list wasn't all-inclusive. That's definitely an 10 issue that is still being considered for the passive LWR's. 21 MR. MICllELSON: That's where you are going to get 12 into the pipe breaks, the fires, the floods, the whole bit. 13 If you aren't going to have any of those things, then maybe l l 14 the argument will hold. If you can design so that none of 15 these happen, then it's about time we apply those design 16 principles to present day plants and get rid of all of this 17 redundancy if it's poccible to do. 18 MR. DONOGHUE: It's possible for a design to 19 propose such a thing. I don't think they have yet. We are 20 still wrestling with the question of how to deal with 21 external events and effects on these passive systems. Wo 22 just don't know exactly how -- 23 MR. MICHELSON: Some of these passive plants have 24 unique materials that might be a little bit of a problem if i 25 they get away too, such as sodium. You have to decide if a O ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (200) 293-3950 he um r we w =r wp -, 9

103 1 sodium leak is an external event or not, and whether the O 2 consequential fire is an external event and so forth. 3 That looks like that's the fundamental thing you 4 pursue is, what kind of external events can you have.

Then, 5

you decido how much redundancy you need to take care of it. 6 This looks like it's going the other way around. 7 MR. DONOGHUE: We are taking the approach of 8 before we try to figure out what is the effect on the 9 passive system, we want to at least see if it's acceptable 10 to consider a design that rolles on a single passive system. 11 12 I think that's even more basic. We are just trying to see 13 if it's even worth considering instead of going through the 14 trouble of trying to determine what external cvents should 15 be considered. 16 MR. WI LKINS : As I understand your recommendation, 17 all you are really saying is we pass, and we will wait to 18 see what they do on the passive light water. 19 MR. DONOGHUE: We are saying that we think this 20 may be an acceptable approach on the designer part. We are 2.' not taking a position there. Yes, we do defer to work that 22 is being done a little ahead of us because we should be 23 consistent with that, because a lot of the significant 24 issues such as external events are being considered almost 25 as we speak, on passive systems and non-safety interactions Q ANN RILEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

104 1 with them. 2 All the designs that use passive means for such 3 things as decay heat removal which we are considering here, 4 do have some kind of active system that performs the same 5 function but they are not considered to be safety grade. 6 They are not relying on it in their analysis all the time. 7 All those questions are being considered for 8 passive plants, and we should be consistent with them. 9 That's what we intended to say here. 10 MR. DAVIS: When do you expect that decision to be 11 resolved? 12 MR. DONOG110E: I really couldn't tell you. I know 13 that tracking the meetings that I have been sitting through, 14 1 would venture to think, a while. They are having meetings 15 this month on this issue. 16 MR. WILKINS: You might ask Charlie that tomorrow. 17 la Am I not right, that's charlie's Subcommittee. 19 MS. SLOSSON: Bob Pierson is here, also. 20 MR. DAVIS: This may affect your schedule. 21 MR. DONOGilUE: That's true. 22 MR. PIERSON: We are working on this for the 23 advanced light water reactors, and we are continuing to 24 pursue it. The treatment of highly reliable non-safety 25 related backup, that is a central issue that are talking O ^"" ai'ev a Associates' 'id-l Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 l

105 1 about. In fact, as he said as we speak now, we have an EPRI O 2 meeting that we are going to later this month. We are going 3 to sit down and give them the final decision based on what 4 we have reviewed and give them a chance to address that 5 decision, and then pursue it at that point. 6 If you are looking for a timeframe, March or 7 April. 8 MR. MICHELSON: What your views in that case, on 9 external events, so you can decide whether non-safety backup 10 is acceptabic. You have to know what kind of events that 11 non-safety -- 12 MR. PIERSON: You are absolutely correct. The 13 external event dominates it. The one thing that I would 14 like to point out that's differing between these reactors ) 15 and the passive light water reactors is that in these 16 reactors the system is working essentially all the time. 17 You are losing that as basically a heat loss. On the PRISM 18 reactor you are losing your five percent that would be 19 resulting from your decay heat and a continuing process. 20 on the light water reactors the systems are 21 slightly different, in that when the reactor goes into an l 22 accident transient that's the first time that the passive 23 system is being called upon to operate, j 24 In effect, one could make the argument that ( 25 perhaps these are maybe more reliable in the light water O ANN RlLEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 - Washington, D. C. 20006 (202) 293-3950

106 1 reactors because you can measure the heat loss that is 2 coming through the system. That's another thing that is 3 going to have to be factored into this equation. 4 MR. WILKINS: I would state that somewhat 5 differently. It's not that they are more reliable. It's 6 just that you know that they have failed. 7 MR. PIERSON: We know that there is heat 8 transferred through that system. 9 MR. WILKINS: Or, you know that there isn't. 10 MR. PIERSON: Or, you know that there isn't. 11 MR. WILKINS: You know that almost immediately 12 after it stops. UJ MR. PIERSON: Whereas, in the case of the light 14 water reactors we are really tre.veling into an unknown area. 15 16 We can do the testing and verify it on models, but until we 17 have some sort of transient that we don't want to have wo 18 will never know whether the system actually works or not. 19 MR. KRESS: Your present position on the non-20 safety related backups for the passive LWR's is that you 21 would possibly tie to the probability -- 22 MR. PIERSON: I really hesitate to get into that 23 realm. There are a number of riaestions about what the 24 interaction between active syatem and the passive system and 25 what implication that has and what we can draw from that. I (} ANN RILEY & ASSOCIATES, Ltd. b Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 v ,,-.3 w c.,--.- 7 y y y,--

107 1 guess what I would try to say is that we are looking for a O 2 highly reliable safety passive system, but we don't think 3 that we can decouple the active systems from that. 4 How we provide that coupling or what we look at in 5 terms of what sort of oversight or what sort of operating 6 constraints, we are still trying to decide that. 7 MR. KRESS: Well, what I struggling with is that I 8 see a rule possibility that your decision on the passive 9 backups is that you will have some requirements that they be 10 safety grade. This bullet says that the treatment will be 11 consistent in this case. 12 They are different. I am worrying about that 13 bottom bullet. 14 MR. PIERSoN: I am concerned about that, too. If 15 it turns out that way, then we will have to do something 16 about that situation. 17 MR. WILKINS: I find, to my complete astonishment, 18 that I didn't need to be quite as ruthless between 8:45 and 19 9:45 as I was, because now at 11:15 we have finished the 20 morning's agenda anyway. 21 MR. WARD: Could I ask a question about this? 22 MR. WILKINS: Yes. You are not finished with 23 this? 24 MR. WARD: No. 25 MR. WILKINS : By all means, Dave. O ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950

108 1 MR. WARD: I think it's a real good feature that O 2 the emergency heat removal systems are on line and you can 3 measure the performance or get some measured indication of 4 performance. But for the PRISM and I think for the MHTGR, 5 the ultimate heat removal system really isn't in the same 6 form during normal operation as it would be in an emergency, 7 is it? 8 Isn't there something in the PRISM? Doesn't the 9 sodium level have to swell up and spill over and go down 10 that annulus or something? 11 MR. THRoM: That's the design for the sodium to 12 spill over and enhance the heat transfer under loss of 13 circulation, the natural circulation you get. Analysis will {} show that if you have a low sodium level in the reactor and 14 15 you do not get that over swell with the RVAC's working based 16 on its expected performance, you still meet what doe or GE - 17 - the pre-applicant is looking at is acceptable temperatures 18 for as far as ASME. It's not absolutely necessary. That's 19 part of the way it would be expected to work. 20 The only thing that has to happen is, the core 21 temperature or the inner temperature of the vessel has to 22 increase for it to start working and always in a natural 23 circulation mode. It's an air cooling system. 24 MR. WILKINS: Again, CANDU is not -- this is not a 25 CANDU issue because CANDU is very much like the existing . Q ANN RlLEY & ASSOCIATES, Ltd. l Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293-3950

- _ _ = _ _ 109 1 light water reactor. O 2 MR. DONOGHUE: Did not propose passive residual 3 heat removal. 4 MR. WARD: CANDU has the same achilles heel that 5 modern reactors have. 6 HR. WILKINS: Yes. 7 MR. WARD: These don't have them, and that's why a we are going so slow in getting them licensed. 9 MR. WILKINS: That's not correct. There has been 10 no application. We have an option. It is 11:20, and the 11 option is to proceed with the first item on this afternoon's 12 agenda. Are you prepared to support that, Ms. Slosson? 13 MS. SLOSSON: Yes. 14 MR. WILKINS: Then, I suggest we do that. 15 (Slidos.] 16 MR. THROM: I am now on my second issue. I may 17 not be as adequately prepared as Beth would have been. This 18 is one that was brought up a couple minutes earlier, about 19 whether or not PIUS has a positive coefficient to be 20 concerned with. 21 This issue has been identified right now on two 22 reactors, the PRISM and the CANDU 3 design. Basically, 23 where we identify this issue it comes out of the GDC 11 24 requirements that are under normal operating conditions. 25 The inherent reactivity feedbacks in the core should O ^"" ai'ev a Associates' 'id. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20000 (202) 293-3950

110 1 compensate for reactivity insertion. 9% 2 Under those conditions I think these two plants 3 would meet that. The problem is, for some very specific 4 events wherein the CANDU 3 design, a loss of coolant 5 accident with a failure or delayed scram would lead to a 6 large positive reactivity insertion. In the liquid metal 7 reactor if you Icse sodium, again, you could have a large 8 positive reactivity insertion. 9 In PRISM, we are talking about some un-scrammed 10 event with total loss of heat sink of probably -- again, I 11 know you don't like the words -- I guess we talk low 12 probability of very unlikely event. Nevertheless, it is 13 something that is of concern to the staff. 14 The question is, given the interpretation of GDC 15 11 and ' ' tat the possible consequences of one of these prompt l 16 reactivicy excursions could be as a result of the inherent 17 physics cr 7t'ristics of these cores, should we continue 18 to review or _.nould we be reviewing these designs that have 19 that attribute. 20 MR. KRESS: I thought the PRISM design which 21 included their gas expansion modules basically precluded the 22 positive void coefficient. Are there events you can 23 postulate still, that will give you the positive void 24 coefficient? 25 MR. THROM: Yes. If I have a basically -- if I Q ANN RlLEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950

~ -. l 111-1 can get to the stage where I can boil off all the sodium and O \\/ 2 void the core. The gas expansion modules help with the loss I l 3 of a reactor coolant pump. PRISM uses electromagnetic ] i 4 pumps. They don't have an inherent coast down capability. 5 They have a machine that backs it up to supply a program 6 coast down. 7 They put the gems in 'casically for an unprotected 8 loss of flow transient. Without a scram loss of flow the 9 gems gi'/e them a lot. They put them in basically for that 10

reason, 11 MR. KRESS:

How do reach boiling of the whole core 12 without losing flow? 13 MR. THRoM: You have to have loss of flow and loss I 7f residual heat removal. It's a very incredible, unlikely, 15 low probability event. The consequences could be severe. i 16 The PRISM group, the ALMR group, they have been responsive 17 to the concern. They have done a hypothetical core 18 disruptive accident. I think Dr. Shewmon was alluding to it l 19

earlier, l

20 They looked at a 500 mega jewel energy deposition l 21 as a disruption of the core. They look at the primary 22 reactor coolant boundary and said it could handle that load. I l 23 24 They went in further to assume a failure in the upper head j 25 region to get a sodium fire to consume the oxygen in the O ANN RILEY & ASSOCIATES, Ltd. Coud Reponers-l l 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

112 1 containment, to look at the containment performance for the 2 low leakage containment dome they put on the system. 3 So, they have done very much what's up here in a 4 policy issue. What we are basictlly probably doing here is 5 identifying that on the Canadian design or any other design 6 that might come in the future that it might have this type 7 of characteristic, there is a need to understand it. 8 There's a need to understand what the consequences are. 9 There is a need to understand what kind of events could lead 10 to that situation, to assure that you haven't missed 11 anything in the design that could be put into the design to 12 help mitigate it or reduce its probability even further if 13 it can be done. 14 In the CANDU design, not only is it necessary to 15 have the scram, it's necessary to have it fairly prompt 16 scram for a large LoCA. We think we should be looking at 17 delayed scrams just to make sure we know how likely or 18 unlikely hopefully this event may be. It may be the severe 19 accident that would be useful to evaluate containment 20 per formance. 21 The question we are putting forth to the 22 Commission is not the nuclear physics of this situation. We 23 want to know basically, is it acceptable to continue 24 reviewing these designs to root out all of the potential 25 problems, to assure that we understand what the consequences ]' ANN RILEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

113 i 1 of this characteristic are. Again, it goes back into the O 2 containment performance. 3 If this becomes the event that challenges 4 containment do we necessarily have to change the design. 5 Maybe not. Once we understand what the likelihood of 6 getting to this situation is and what the real consequences 7 are, then we need to step back and make a decision on the 8 specifics of the design. 9 PRISM has gone a little bit further in actually 10 trying to look at it. They haven't said what it is that 11 leads to the hypothetical core disruptive accident, they 12 have just said we will go out and analyze it. They analyze 13 the structures and then they analyzed the containment design 14 against even a further failure over what they would expect 15 .given the hypothetical core disruptive accident. 16 They are kind of doing exactly what we would 17 recommend. Again, this is to recommend for future designs 18 as well as what we are looking at. 19 MR. SilEWMON : PRISM, I notice they still have what 20 I think they used to call a heterogeneous core, one that 21 mixes blanket as well as fuel in the center of the core. The 22 purpose of that was to reduce the sodium void coefficient. 23 Is that one of the purposes, wasn't it? 24 MR. TIIROM: What I kraw is, if you lose all of the 25 fuel -- if you melt the fuel for example, there is a Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

114 1 contribution to it. If you lose the blanket material O 2 there's a contribution. Whether or not the particular 3 design -- I can't answer your question. 4 MR. SHEWMON: What they were talking about was the 5 prompt ejection of sodium, I think, is what got them from 6 the homogeneous to the heterogeneous. What you are now 7 postulating is a slow boil off and redistribution of fuel in 8 core. 9 MR. THROM: No. You asked what was the situation 10 that would get you to that. 11 MR. SHEWMON: I asked about the purpose of the 12 heterogeneous core which I thought was to reduce sodium void 13 coefficient, and why that hadn't been mentioned or what kind ( 14 of an accident you were postulating which didn't benefit 15 from this. 16 MR. THROM: I can't answer that, the physics of 17 that question. What I can-remember from the numbers are if 18 the core voids including the blanket material --'that's 19 worth about five dollars worth of reactivity. 20 MR. SHEWMON: Voids. This is over days it can be, 21 or over milliseconds. 22 MR. THROM: Milliseconds. It is very, very fast. 23 MR. SHEWMON: We are still talking about the same 24 rate for void. How you get rid of that solid material in 25 milliseconds is a more difficult -- O ^"" ai'av a Associ^Tes' 'id-Coud Repoders l 1612 K. Street, ' N.W., Suite 300 l Washington, D. C. 20006 (202) 293-3950 l

115 1 MR. THRoM: That's right. What the calculations 2 that the designer has done and some independent work that we 3 had done by Brookhaven using what they call the Beta Tape 4 analysis -- this is getting very technical -- those types of 5 assumptions were used. I guess the reactivity insertion 6 rates come out to like two hundred dollars a second or 7 something like that for what I believe people think are the 8 dynamics of the situation. 9 The reality of the situation may be that in the 10 real world the net prompt reactivity might be less than 11 what's being assumed at these somewhat simplified 12 calculations. As a matter of fact, if you boiled off slowly 13 for the first part of that transient you would expect to see 14 increased leakage out of the core before you got the sodium 15 level down far enough where you would start getting a void 16 coefficient. 17 Right now, their estimate of a net reactivity 18 insertion due to core voiding, if you look at competing 19 effects on the temperature rise, axial expansion of the 20 core, radial expansion of the core, some of the other 21 physics characteristics would come in that the prompt 22 reactivity insertion might only be like a dollar forty-23 three. It's still a core disassembly concern. 24 MR. WILKINS: That dollar forty, that's twice as 25 much as the gems would give. O ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

= 116 i MR. THROM: Right. Again, the gems were basically O12 put in the for the unprotected loss of flow. 3 MR. SHEWMON: I havon't heard of Beta tape in a 4 long time. 5 MR. THROM: One of the questions we had when they 6 came in with the analysis that posed 500 megawatts per jewel 7 as the mega jewels of energy deposition into the thing, we 8 weren't sure exactly how that number came out as an FFTF 9 metal oxide data. We wanted to see how good reasonable 10 looked. 11 We went back and did the quickest type of 12 calculation which would be Beta tape at this particular 13 time, to try to get a handle on whether or not that order of 14 magnitude was about right. It looks to be reasonable. It 15 looks to still be conservative. 16 Again, I guess the physics of these systems is a 17 wonderful thing to discuss in detail. We are not prepared 18 to get into it today. 19 MR. SHEWMON: Let's not talk about it then. 20 MR. WARD: Could I talk about-it a little more? 21 MR. WILKINS: Sure. 22 MR. WARD: There is an interesting paragraph on 23 page 23 in the paper. It says attempts to modify the 24 designs to reduce the effects of these positive coefficients 25 may result in other consequences, potentially serious. Q' ANN RILEY & ASSOCIATES, Ltd. - Coud RepOders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

117. 1 Then, you go on to say in the PRISM design, if you make~I 2 guess bigger rods to offset the effect of the void you could 3 end up with more severe rod withdrawal accidents. 4 There is a possible trade off. 5 MR. THROM: That's right. 6 MR. WARD: In the CANDU, as I understand it, 7 there is nothing inherent in heavy water reactor that means 8 you have to have a positive void coefficient; that CANDU 9 reactors are -- I guess physicists don't like that term -- 10 over moderated. -l 11 MR. THROM: That's right. 12 MR. WARD: There really probably wouldn't have 13 been a negative safety tradeoff. It may be an awful 14 economic tradeoff, 15 MR. THROM: It possibly could. I guess the things 16 that could be considered are enriching the fuel -- that 17 would be an additional cost in operating the plant. We are 18 not as far along on there, whether exploring alternatives 19 that might reduce that problem in that design. 20 MR. WARD: That's one of the things that has been 21 done, hasn't it, with the RBMK's, to get around some of the -22 positive void coefficient there -- 23 MR. THROM: That's conceivable. I don't know much 24 about those. 25 MR. KRESS: That's one of their fixes. That's the l} ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

4 118 1 one that they had the most effect, as a matter of fact. O 2 MR. THROM: We asked PRISM what they could do to 3 reduce it, and they have done that study. 4 MR. WARD: Have you asked CANDU. 5 MR. THROM: No, not yet. I think it's been 6 explored at the level we have been going on and may be 7 something we want to look at. I think there are some papers 8 in the international community that they looked at some of 9 these alternatives. They haven't come in suggesting that 10 they are going to change the design. 11 We also have to consider for this reactor as we 12 understand it right now, this is a large LOCA with a loss of 13 two completely redundant independent shutdown systems. They 14 have a rodded system and a liquid poison system that are 15 both safety grade. They are diverse and redundant in 16 shutdown capability. What's the credibility -- excuse the 17 word -- of getting into that situation. That has to get 10 factored in somewhere along the line. 19 No, we have not gone into that level of review 20 with them and explore? those types of alternatives, at.least 21 to the extent that-we have with the liquid metal design. 22 MR. WILKINS: Your recommendation is fundamentally 23 the top bullet, the positive void coefficient should not l 24 necessarily disqualify a reactor. 25 MR. THROM: Right. IlQ ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

119 ,-s 1 MR. WILKINS: You would have to analyze these f ) \\# 2 things, and whatever else you think might be appropriate. 3 MR. THROM: Right. 4 MR. WILKINS: To analyze. 5 MR. THROM: Right. 6 MR. WILKINS: Are there additional questions or 7 comments from the committee? 8 (No response.] 9 MR. THROM: Joe is back on now, to discuss control 10 room design. 11 MR. WILKINS: We have 25 minutes. At the rate we 12 are going, I think we could finish control room design in 13 that time. Are you willing to proceed Ms. Slosson? 14 MS. SLOSSON: Yes. ( 15 MR. WILKINS: Why don't we do so. 16 MS. SLOSSON: I guess it is to be noted too, we 17 have entered the realm of category two issues as opposed to 18 category one issues for this paper. 19 MR. WILKINS: Thank you. 20 MR. WILKINS: We will recall,-category two are the 21 issues where the staff position is that they should not 22 Jevi ate from existing policy or procedures. 23 MS. SLOSSON: Right, that we are not willing to 24 consider that at this time. l 25 MR. WILKINS: Yes. Rather, your recommendation to 1] ANN RILEY & ASSOCIATES, Ltd. Coud RepOders 1612 K. Street, N.W., Suite 300 l Washington, D. C. 20006 (202) 293-3950

120 1 the commission is that they-instruct you not to consider. O 2 MS. SLOSSON: Exactly. 3 [ Slides.) 4 MR. DONOGHUE: This is part of the second-category 5 of issues. I won't belabor that. The issue here about 6 control room and remote shutdown area design, the title is 7 worded a little bit differently than in the table to fit it [ 8 into that table. It's not just control room design. It's 9 considering the relationship between the control room and 10 remote shutdown requirements. 11 Again, consistent with the advanced reactor policy 12 statement that we keep bringing up today, the designers are 13 proposing changes for what we are used to in regulation, to 14 simply the design and make it -- to try to meet some of the 15 intent of the advanced reactor policy statement. 16 One of the things we are trying to do here is to 17 reduce the requirements on the control room design. One of 18 the things they are trying to do to allow that is to provide 19 a remote shutdown facility or remote shutdown area that is 20 safety grade. 21 The two specific designs that we are considering 22 here are mostly centered on this issue are PRISM and MHTGR 23 which have indeed proposed non-safety grade control rooms 24 and a safety grade remote shutdown area. CANDU has -- we 25 have some questions _about how they define safety grade in Q ANN RILEY & ASSOCIATES, Ltd. COud RepOders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

121 1 general. We will get to that later. They don't propose all 2 the same requirements for their control room. PIUS is the 3 closest to what we consider to be at current requirements in 4 that they have a safety grade control room and remote 5 shutdown area. It is not so much of a problem with their 6 design approach as with the others. 7 As I said, those control room requirements aren't 8 considered as necessary as they may have been in the past by 9 the designers. This approach is similar to some issues that 10 again have been raised with the passive LWR's that concern 11 the control room. They are not centered on remote shutdown 12 ability but are central to the question of the safety 13 related or safety requirements of the control room. 14 [ Slides.] 15 MR. DONOGHUE: As I said, we are considering the 16 specific issue of whether the remote shutdown requirements 17 by their approach is not the other ancillary questions 18 although they are connected. The staff -- we already-stated 19 this is part of our second category of issues. 20 We don't think it is at this time warranted that 21 the designs be permitted to deviate extensively from what we i 22 consider conventional requirements for control rooms. 23 MR. WARD: Why not? 24 MR. DONOGHUE: Up to now, the staff has taken a l l 25 very strong line that the operators are central to safety of O ANN RILEY & ASSOCIATES, Ltd. l Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 l (202) 293-3950 i i

122 a plant. Right now, we are not convinced -- and we are not 1 2 sure that depending on the design review having changed. 3 Right now, we are not sure that we could accept the argument 4 that the operator is not a central component to plant 5 safety. 6 MR. WARD: I don't think the applicants are 7 proposing that, are they? They are proposing the operator 8 might have to go to a different place to do his job. 9 MR. DONOGHUE: Right. The first part of my 10 statement was that the operator should be required to stay 11 in the area he is used to that has the central controls for 12 the plant. We shouldn't plan on an event requiring the 13 operators to go a remote location. We don't think that 14 would serve safety intent. 15 MR. KRESS: If the safety case for the reactor 16 design doesn't require an operator action, then why should 17 NRC require it? 18 MR. DONOGHUE: We are not proposing it as a 19 requirement. We are saying that at this stage when we are 20 looking at the preliminary designs, that we don't think that 21 right now we could go past what we currently require.

True, 22 they could make the case and could successfully convince the 23 NRC staff that we could change the requirements.

24 Right now, we don't think that it is warranted at 25 this time. O ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

-.= l 123 1 MR. KRESS: Except, your recommendation once again _O 2 is tied to what the outcome of the passive LWR policy is. I 3 think these are different than the LWR in this particular 4 case also. 5 MR. DONOGHUE: In making our recommendation we 6 realize as I said on the first slide, that there are 7 connected issues with control room safety requirements and 8 remote shutdown requirements, diversity of the systems, 9 habitability and so forth. 10 We wan' to keep an eye on that because whatever 11 determination we are going to make on this, we have to be 12 consistent with that. That has to be central to our 13 recommendation. l 14 At the same time, we realize that the operator -- 15 we believe that the operator is an essential component to l 16 plant safety right now. We are not sure that an argument-17 can be made to obviate that need in our eyes yet. 18 MR. KRESS: I thought that was one of the real 19 strong justifications to go to a passive LWR, so you could 20 have a foolproof design that didn't require innovative 21 operator actions to really get it through the safety case. 22 Wasn't that the reason to go that way? 23 Now we are saying okay, but we are not going to l 24 believe it. l 25 MR. DONOGHUE: Back again to the advanced reactor l Q-ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 .,o ,,o-,m,,y. ,m,.,,_- -,---w.,-y- ..-,,.y

i 124 1 policy statement, we are trying to satisfy the Commission's 2 encouragement that designers propose simpler designs. You 3 have to meet safety requirements. We think that the 4 operator and where the operator is going to do his job or 5 her job is central to assuring plant safety. 6 We are not sure now in the pre-application stage 7 as we are with other issues where we can see there's a 8 possibility to leave the door open on some of these things, 9 that -- it's a can never say purely technical but it's more 10 tentatively based argument where they can change their 11 design, we might be willing to change our requirements. 12 On this issue we don't think that we will be 13 changing our requirements, at least not until we are more ( 14 strongly convinced. 15 MR. MICHELSON : Isn't the key, again, what 16 external events this plant will have to face as to what 17 these operator stations might be? 18 MR. DONOGHUE: Absolutely. 19 MR. MICHELSON: Until you start getting serious 20 about your external events you don't know whether you need 21 one or two locations to operate the plant, because the 22 external event is going to happen in one of them. 23 MR. DONOGHUE: That's true. Again, we are just 24 considering on the face of the designs without any detailed 25 reviews being done of them all, our position right now is O Ann ni'ev a Associateste. Coud RepOders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 l

125 1 that the current requirements should be applied with the r. 2 caveat under our recommendation here, that we are going to 3 he making sure that with the approach we take when we get to a the certification stage that it's going to be consistent S with what has been done before us. 6 MR. WILKINS: This is a policy which you will 7 apply to these pre-application applications, so to speak. 8 MR. DONOGHUE: Right. 9 MR. WILKINS: It may well be that by 1996 when you 10 get a real application -- if you get a real application in 11 1996 -- you may want to reconsider this recommendation. 12 MR. DONOGHUE: Absolutely. 13 MR. WILKINS: You may decide that the passive LWR 14 policy doesn't apply either because there are enough 15 differences. 16 MR. WARD: Meanwhile, the design applicants are 17 throwing up their hands and saying what's the use in coming 18 up with all this wonderful system innovation because NRC 19 isn't going to give us -- 20 MR. WILKINS: Give us any credit for it. 21 MR. WARD; Yes. 22 MR. MICHELSON: Are you going to rule out fire, 23 for instance, which is one of the main reasons for this in 24 the control room. No more fires in the control room, that 25 helps. That's the issue you have to face. Q-ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

126 1 MR. WARD: I think what they are proposing is to 2 take advantage of the remote shutdown areas. 3 MR. WILKINS: The applicants are. 4 MR. WARD: The applicants are. 5 MR. WILKINS: Yes, they are. 6 MR. MICHELSON: But apparently then, a safety 7 grade control room or what? 8 MR. WARD: Right, that's what they are suggesting. 9 MR. DAVIS: The control room is proposed to be 10 non-seismic. The concern that I have is, if that's the case 11 it's possible that a seismic event would prevent the 12 operators from leaving the control room. 13 MR. MICHELSON: Right. The ceiling falls in. 14 MR. DAVIS: Either that, or blocking the doors or 15 jamming the doors. 16 MR. WARD: We will make the doors seismic, if 17 that's the issue. I mean, don't make everything -- 18 displays and everything else seismic if all you are worried 19 about is the door opening. 20 MR. DAVIS: I am worried about getting the l 21 operators out of there to the remote areas, whatever that j 22 takes. 23 MR. WARD: If that's the issue -- 24 MR. DAVIS: It is an issue. 25 MR. CARROLL: There's another issue. I don't know l 1 l . O ^"" ""' ' ^SS '^'S S ' d-Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

127 1 if it's changed since the last time we were briefed on the [V 2 gas cooled reactor. At that time they were arguing that the 3 control room was not a vital area from a security point of 4 view. 5 The operators are prevented from leaving it by the 6 bad guys. You don't need a seismic event, you just need 7 some bad guys to take over the control room. 8 MR. WARD: I think that was the PRISM, wasn't it? 9 They changed it. 10 MR. CARROLL: Did they change it? 11 MR. WARD: That's what I recall. 12 MR. CARROLL: That's something that your write up 13 doesn't reflect, is the issue of what kind of a security 14 area this is or what kind of security -- 15 MR. DAVIS: Being non-seismic you also have the 16 possibility of injury to the operators from falling light 17 panels, for example. That has been found to be the case at 18 some plants even. To me, there's more to it than just what 19 we have talked about. 20 You have to make sure the operator is still 21 capable and that he can get to the remote area if you are 22 going to use that as your main backup during a seismic 23 event. 24 MR. WARD: It may be a reasonable argument, but 25 that doesn't mean that all kinds of instrumentation and Q' ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

128 1 displays and everything in the control have to be seismic, O 2 for example. A lot of them are anyway. i 3 MR. MICHELSON: You have to address the system 4 interaction question then, as to whether or not the seismic 5 event and the falling panels or whatever ja actuating 6 switches that may hinder the ability to operate from the 7 other center. You can't avoid the fact that if you are 8 designing for seismic you have to address those issues, 9 whether it's so-called safety or non-safety areas. 10 MR. WILKINS : In contrast to some of our other 11 issues, it seems to me that what the Committee is saying is 12 that they agree with the staff. 13 MR. KRESS: Some of the Committee. 14 MR. WILKINS: Some of the Subcommittee. 15 MR. WARD: I don't think so. 16 MR. KRESS: I don't think I do. 17 MR. DAVIS: I think it's about a three to two 18 split right now. 19 MR. WILKINS: Three to two out of seven, that's 20 pretty good. 21 MR. DAVIS: There are a couple of uncommitted. 22 MR. MICHELSON: Do you want to take a vote on it? 23 MR. WILKINS: Maybe that will be at the end of our 24 agenda. 25 MR. DONOGHUE: Those are all the remarks I have Q ANN RILEY & ASSOCIATE" Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

129 1 about this issue. Are there any more questions? 2 MR. WILKINS: You didn't on that slido at least, 3 talk about the electrical requirements, this so-called class 4 lE. Would you mind saying a few words about that and your 5 position in that area? 6 MR. DONOGliUE: It's not a -- 7 MR. WILKINS: That's a GDC 17, apparently. 8 MR. DONOGlIUE: Yes. The requirement as it stands, 9 we think serves safety functions that we think are 10 important. Right now, although we haven't done the reviews 11 of these areas yet, we think that we would be applying those 12 requirements to ensure that the systems function when they 13 are required to function. 14 That's why the basis of our recommendations that 15 right now we don't think a change is warranted, because 16 those requirements h3ve served us well to now. We see no 17 reason to change them based on what we have seen so far in 18 the pro-applications. 19 MR. CARROLL: Common modo failures of digital 20 systems are concerned you are sort of following the lead of 21 what's going on with the light water reactors? 22 MS. SIDSSON: Yes. 23 MR. MICllELSON: Except that the light water 24 reactors are putting all that equipment in seismically 25 qualified areas. We start talking a non-seismic control O ^"" ai'ev a Associ^Tes' 'td. Court Reporters 1612 K. Street, N.W., Suite. 300 Washington, D. C. 20006 (202) 293-3950 r c- ..-m-.

130 1 room, you have to follow some other guidance. A 2 MS. SLOSSON: I think what the staff is saying 3 here is, we want them to maintain current requirements with 4 respect to control room. 5 MR. SHEWMON: Seismically qualifying a PC oug'tt to 6 be a lot simpler than a relay, isn't it? 7 MR. MICHELSON: No, I think it's a more comp lex 8 issue than just whether the equipment is seismically 9 qualified. It's all the system interaction, the falling 10 debris on the equipment, the fire that is generated by 11 seismic -- 12 MR. CARROLL: My question really had to do with 13 the issue that is still floating around on analog backup to 14 the digital system because of concerns of common mode 15 failure. 16 MR. WILKINS: That issue just hasn't been resolved 17 yet, to the best of my knowledge. 18 MR. MICHELSON: -The problem with making -- 19 MR. CARROLL: The staff has taken a position, I 20 think, to the effect that they will consider simple digital 21 systems -- 22 MR. DAVIS: Redundant. 23 MR. CARROLL: -- as a backup to the nain control 24 room protection system. 25 MR. WILKINS: I don't want to start the last one. (]' ANN RILEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

131 1 I am going to save something for after lunch. We will give \\-- 2 ourselves an extra seven minutes, 3 MR. CARROLL: Are you going to allow us to -- 4 since we did so well this morning -- go back and revisit 5 some of the earlier issues? 6 MR. WILKINS: What I would propose this afternoon 7 is, we have the staff presentation on this last issue. 8 Then, we need to as a Subcommittee, to discuss the nature of 9 the recommendations we will make to the full Committee 10 tomorrow and what sort of presentation. We have an hour 11 tomorrow for that. 12 MR. CARROLL: The reason I raised the question is, 13 maybe some of these people can go back to work if we are not (~h 14 going to revisit some issues. V 15 HR. WILKINS: I would hate to think you were 16 implying that they weren't in fact working this morning. 17 [ Laughter.] 18 MR. WILKINS: I suspect we ought to at least keep 19 them around for the very early part of the first part of the 20 afternoon. I know how much their chomping at the bit to get 21 back to their desks. Whether they work when they get back 22 there is another matter, and not our concern. 23 Let's recess until 1:00 o' clock. 24 [Whereupon, at 11:52 a.m., the Subcommittee 25 recessed, to reconvene at 1:00-p.m., this same day.) (D ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N W., Suite 300 Washington, D. C, 20006 (202) 293-3950

132 1 AFTERNOON SESSION 2 (1:00 p.m.] 3 MR. WILKINS: We will reconvene the meeting of 4 this Subcommittee. I believe we are now prepared to hear 5 from Mr. Jack Donohew this time. 6 [ Slides.] 7 MR. DCNOHEW: I am the project manager in NRR for 8 the MHTGR reactor design. What I am here to discuss is l 9 safety classification, which is the methodology by which the 10 designer would decide which system structures and components 11 would be safety related. This is a category two item in 12 which we have decided we have reached a point in our review 13 of the pre-applicants -- I guess I should point out that 14 there's only one pre-applicant that proposed a difference 15 from what would be current light water reactor practices. 16 This has been looked at, and we have reached a 17 point where we decided that we would propose that the 18 Commission would continue with what our current light water 19 reactor practices as opposed to changing to what was 20 proposed by the pre-applicant. 21 What I have here as background is the current 22 light water reactor requirements, of which I think the point 23 to be made is that here in Part 50 and Part 100 there are 24 what you could say are a definition for what is a safety 25 related systems structure and component. O ANN RILEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

133 1 The three items are those which are needed to 2 maintain the reactor coolant pressure boundary, needed to 3 shutdown the reactor and maintain it in a safe condition as 4 to pose safe conditions, and needed to prevent or mitigate 5 accidents that could result in dose consequences. It says 6 here comparable to Part 100 guidelines, because that's what 7 is stated in Part 50. Part 50 and Part 100 as they exist 8 now, there will be changes to both Part 50 and Part 100 9 where these words are and it would change. 10 MR. WILKINS: Excuse me. My recollection is that 11 those changes deal only with where things are located within 12 the rules, and do not represent a substantive change in the 13 rules. 14 MR. DONOHEW: That is correct. That is there is 15 not any point being made about that, other than to say where 16 they are located particularly with Appendix A of Part 100 it 17 will be somewhere else that will be in Part 50. 18 [ Slides.] 19 MR. DONOHEW: What this is, is defense-in-depth. 20 It's maintaining three barriers to the release of activity 21 from within the fuel to the public. You have the reactor 22 coolant boundary, you have the protection of the fuel, and I 23 will say this would include the fact of having a containment 24 to protect the releasea from beyond those two barriers to 25 the environment. Q ANN RlLEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

134 1 MR. DAVIS: What do the words, " comparable to" 2 mean? 3 MR. DONOHEW: What they have been used in the fact 4 of current light water reactors, if you needed a system, 5 structure, component taking credit for that structure, 6 system or component, to mitigate your doses or prevent the 7 doses from being above Part 100. Or, comparable in the fact 8 that in some cases accidents -- the staff has applied like 9 ten percent or 25 percent of Part 100 depending on 10 probability of the design basis event. 11 If they were needed to get down within below that 12 value then they had to be classified as safety related. 13 MR. DAVIS: That's what I am concerned about. 14 What is the value when you say comparable to? 15 MR. DONOHEW: At this point -- this is light water 16 reactor practices. What they did is -- for example on a 17 refueling accident where the probability was considered 18 higher than for a loss of coolant accident. The numbers 19 like ten percent of Part 100 were used in reviewing the 20 plant in terms of granting a license. 21 That ten percent of Part 100 would have been used 22 in looking at that accident to decide what were the 23 structures, systems and components that needed to be safety 24 related. What we are proposing to the Commission is that we 25 do not change the criteria that the pre-applicant should be Q ANN RILEY & ASSOCIATES, Ltd. Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

132 1 AFTERNOON SESSION 7 2 (1:00 p.m.] 3 MR. WILKINS: We will reconvene the meeting of 4 this subcommittee. I believe we are now prepared to hear 5 from Mr. Jack Donohew this time. 6 (Slides.] 7 MR. DONOHEW: I am the project manager _a NRR for 8 the MHTGR reactor design. What I am here to discuss is 9 safety classification, which is the methodology by which the 10 designer would decide which system structures and components 11 would be safety related. This is a category two item in 12 which we have decided we have reached a point in our review 13 of the pre-applicants -- I guess I should point out that 14 there's only one pre-applicant that proposed a difference 15 from what would be current 1.ght water reactor practices. 16 This has been looked at, and we have reached a 17 point where we decided that we would propose that the 18 Commission would continue with what our current light water 19 reactor practices as opposed to changing to what was 20 proposed by the pre-applicant. 21 What I have here as background is the current 22 light water reactor requirements, of which I think the point 23 to be made is that here in Part 50 and Part 100 there are 24 what you could say are a definition for what is a safety 25 related systems structure and component. (' ANN RILEY & ASSOCIATES, Ltd. \\~ Coud Repoders 1812 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

~.- 133 1 The three items are those which are needed to O 2 maintain the reactor coolant pressure boundary, needed to 3 shutdown the reactor and maintain it in a safe condition as 4 to pose safe conditions, and needed to prevent or mitigate 5 accidents that could result in dose consequences. It says 6 here comparable to Part 100 guidelines, because that's what 7 is stated in Part 50. Part 50 and Part 100 as they exist 8 now, there will be changes to both Part 50 and Part 200 9 where these words are and it would change. 10 MR. WILKINS: Excuse me. My recollection is that 11 those changes deal only with where things are located within 12 the rules, and do not represent a substantive change in the i 13 rules. 14 MR. DONOHEW: That is correct. That is there is 15 not any point being made about that, other than to say where 16 they are located particularly with Appendix A of Part 100 it 17 will be somewhere else that will be in Part 50. 18 [ Slides.) 19 MR. DONOHEW: What this is, is defense-in-depth. 20 It's maintaining three barriers to the release of activity 21 from within the fuel to the public. You have the reactor 22 coolant boundary, you have the protection of the fuel,.and I 23 will say this would include the fact of having a containment 24 to protect the releases from beyond those two barriers to 25 the environment. 'Q ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

134 1 MR. DAVIS: What do the words, " comparable to" 2 mean? 3 MR. DONOHEW: What they have been used in the fact 4 of current light water reactors, if you needed a system, 5 structure, component taking credit for that structure, 6 system or component, to mitigate your doses or prevent the 1 7 doses from being above Part 100. Or, comparable in the fact 8 that in some cases accidents -- the staff has applied like 9 ten percent or 25 percent of Part 100 depending on 10 probability of the design basis event. 11 If they were needed to get down within below that 12 value then they had to be classified as safety related. 13 MR. DAVIS: That's what I am concerned about. 14 What is the value when you say comparable to? 15 MR. DONOHEW: At this point -- this is light water 16 reactor practices. What they did is -- for example on a 17 refueling accident where the probability was considered 18 higher than for a loss of coolant accident. The numbers 19 like ten percent of Part 100 were used in reviewing the 20 plant in terms of granting a license. 21 That ten percent of Part 100 would have been used l 22 in looking at that accident to decide what were the 23 structures, systems and components that needed to be safety 24 related. What we are proposing to the Commission is that we l 25 do not change the criteria that the pre-applicant should be 1 Q ANN RILEY & ASSOCIATES, Ltd. Coud - RepOders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 1 l l

~. 1 135 1 using to decide what are the safety related structures, t 2 systems, components. 3 The issue is, what criteria should be applied by 4 the advanced reactor designs. The recommendation is to 5 continue the light water reactor practices. I want to make 6 one point. These two are exactly the same as written 7 before. This one is written exactly as before but it-8 shouldn't have. The thing being is, we have not decided 9 what would be acceptable dose consequences. 10 Really, what this last one should say is needed to 11 pr9 ent accidents that could result in dose consequences 12 witnin acceptable limits. We are in fact not going to be 13 talking about comparable Part 100. We are going to talk 14 about what would be considered acceptable dose consequences, 15 which gets back to the very first issue that we discussed. 16 Granted, there was not any specific dose limits put in 17 there. 18 The work in that area on the individual designs, 19 those dose limits that would be considered as the values 20 that would be considered as necessary for being acceptabb 21 then would used here as the third criteria. 22 MR. KRESS: Those limits would be related to the 23 probability of the accident. You have different limits for 24 different probabilities. 25 MR. DONOHEW: I think so. That's what was done Q ANN RlLEY & ASSOCIATES, Ltd. i l Court Reporters i 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

i 136 1 for light water reactors. I really don't see that change. 2 I think one point I haven't told you what was actually 3 proposed to us. I think what the MHTGR program proposed was 4 just to have this third criteria as being the only criteria. 5 MR. MICHELSON: Aren't you trying though, to 6 reduce your emergency planning levels for these plants? 7 MR. DONOHEW: Okay, but now we get back to -- 8 MR. MICHELSON: That relates to Part 100 and 9 probably does require emergency planning. 10 MR. DONOHEW: Now we get to a point of let's say 11 what would be acceptable for emergency planning 12 considerations, and what would be acceptable for licensing. 13 I think the thing is that yes, I think that one possibility 14 is the fact that if you are -- if the staff would accept 13 reduced requirements on emergency planning, then when wo 16 come to here we might say okay, maybe this should be like 17 the PAG guidelines. 18 MR. MICHELSON: In other words, they would be 19 consistent with the emergency planning provisions and so 20 forth. They would have to be. ?1 MR. DONOHEW: I guess there are different ways of 22 saying that things can be consistent. I think that yes, it 23 would take that into account. I think you may be saying if 24 we did accept that, then obviously we would just on the 25 third criterion lower that to the PAG guidelines. I don't Q ANN RlLEY & ASSOCIATES, Ltd. l Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 I (202) 293-3950

137 1 think that may be true because you also have probabilities 2 in there. 3 Consistent, in the sense that it has to take that 4 into account. The fact that a plant with reduced 5 requirements on emergency planning, things would be 6 different than a plant that did not have these requirements. 7 MR. KRESS: I think he's saying that if a system 8 is made to get below PAG's then it ought to be part of your 9 safety classification. 10 MR. DoNoHEW: I think at first blush I think 11 that's true. I could envision situations where you look at 12 a very low probability event. For an example with the MilTGR 13 program they submitted design basis events, they had severe 14 accidents for a PRA look. They also had what they call 15 emergency planning accidents. They had separate accidents 16 if the probabilities on the emergency planning ones were 17 very low probability. 18 What we could end up as we do on severe accidents 19 about what equipment do we allow them to take credit for, 20 it's a possibility we might -- to give credit for a very 21 high reliable non-safety related system that would 22 compensate for the fact of being safety related. I don't 23 think we are at all ready to say anything other than to say 24 yes, it would be consistent in the sense that we would have 25 to take that into account. We would have to look at -Q ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

138 1 overything and not -- this isn't like a separate issue 2 totally onto itself. 3 It's involved with accident selection in emergency 4 planning and so forth. 5 MR. WARD: Would there be provision in this for 6 sort of mixed requirements or partial -- instead of a piece 7 of equipment having to meet the full range of traditional 8 safety grade requirements it may be that a picco of 9 equipment in a given plant has a safety function only for 10 certain kinds of accidents. Perhaps an accident that 11 doesn't have anything to do with an carthquake. 12 Instead of just saying this piece of equipment has 13 to be safety grade in the traditional sense of EQ and QA and 14 seismic resistant, maybe it just needs to be EQ and QA. Do 15 you see what I mean? 16 MR. DONOllEW: I see what you mean. I am not sure 17 what distinctions were done with light water reactors for '8 exactly that situation. 19 MR. WARD: I am not sure there were many. 20 MR. DONOHEW: In my mind there was, I think, 21 particularly like EQ was, for specific accidents you were 22 dealing with the worst accident that could occur in that 23 cubicle. 24 I really think we -- 25 MR. CARROLL: It may have been a mistake, by the Q ANN RILEY & ASSOCIATES, Ltd. Court Repo;ters 1612 K, Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

139 I way. 2 MR. DoHoHEW: Let's say to the best of our 3 ability, I think we would deal with what accident the 4 equipment was there for. I guess I am not sure we in fact 5 would do things really differently than how we have 6 developed them for light water reactors. I think this is an 7 area for which I don't think just because you have a sodium 8 cooled reactor or a gas cooled reactor in terms of safety 9 classification, I really don't think you are going into new 10 areas. 11 I think what you got is -- what we want to start 12 with is defense-in-depth, protecting three barriers, and I 13 think applying it in the way we have in the past. ( } 14 MR. CARROLL Let me give you an example of why I 15 said what I did. I am aware of some present day plants that 16 have qualified and have done the EQ on containment 17 electrical penetrations. The way that they pass EQ is to 18 pressurize the penetration with nitrogen to keep moisture 19 out. 20 Guess what? If the containment goes above design 21 pressure that whole system falls apart. They can go above 22 design pressure under severe accident conditions. There is 23 an example of where we talk design basis and over here is 24 something called severe accidents and never the twain shall 25 nie et. O ^"" ni'ev a Associ^Tes' 'ta-Coud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293 3950 m 4 m- -m

140 1 MR. DONOllEW That may be misapplication of what 2 we are -- 3 MR. CARROLL: That's the kind of trap you fall 4 into when you start drawing these severe accident -- design 5 basis accident boundaries and acknowledge that there is 6 something worse than this. 7 MR. DONOllEW' The thing being, this critoria is 8 being directed towart; #cign basis events and not severe 9 accidents. You have to look at what equipment is going to 10 be there when you have the severe accident. I guess I am 11 not sure how to really answer your question because I'm not 12 sure there's an answer, other than the fact that as we look 13 in those areas you have to be very careful about what you 14 do. 15 MR. CARROLL: One thing I do not find in your 16 viewgraphs or in your write up is any consideration of 17 shutdown and low power operation risk. 18 MR. DONOllEW: I guess in -- 19 MR. CARROLL: That has become a real issue. 20 MR. DONOllEW: I think that's because -- 21 MR. CARROLL: There are those that say it may be 22 the dominant risk. 23 MR. DONOliEW: I think the thing is that -- I guess 24 maybe I can't really answer that. 25 MR. WILKINS: Let's let Ms. Slosson answer it. Q-ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 ,-,----,r---., .,,,.g,,, gy4 y ,m; _,-,,p.p ,_.~.7q.___ .m-- .y y,--, r.--. _.v

I 141 1 She is trying to. 2 MS. SLOSSON: That's one of the issues that we are 3 looking at with respect to the passive and the evolutionary 4 reactors. 5 MR. WILKINSt Yes, that's what I thought. 6 MS. S LOSSON: I guess it is something that we 7 would follow along with them, and not point out separately 8 for these reactors yet. 9 MR. DONOHEW: I think also -- looking at USI's and 10 no forth, I think the fact where the light water reactors 11 were designed to this and as you are pointing out these 12 things have been brought up and have been addressed. I 13 think that issue may be addressed separately to safety 14 classification. I don't think the issue is going to be 15 overlooked. 16 I just don't think what it will be is addressed in-17 the fact as coming up with the first -- they use this to 18 como up with their list of safety equipment. Then, the 19 individual plant responses to individual accidents is looked 20 at. I think that area you are talking about will be looked 21 at separately. 22 MR. CARROLLt But should it? Your third bullet 23 talks about accidents, but I guess your pre-applicants are 24 probably reading that to mean traditional kinds of accidents 25 which haven't included low power shutdown risk. O ANN RILEY & ASSOCIATES, Ltd. Coud Reponers 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 o

1 MR. WILKINSt If those accidents could result in 2 doses that are comparable to Part 100 guidelines, they 3 should be looking at them. 4 MR. DoNollEW: Also, don't you have the problem on 5 the second one, too, to keep it in a aafe condition? I 6 mean, if you do that then you don't have any fuel damage. 7 You are keeping the fuel within its design limits. You are 8 not releasing activity above what would be normal coolant 9 activity. 10 MR. WARD: I think that Jay's point is that 11 following the old rules and practicca has given us a 12 generation of reactors which there may be -- there are 13 indications there is a disproportionate risk during 14 chutdown, a disproportionately high risk during shutdown. 15 Now, if you are coming up with a new series of 16 reactor designs the question is, should your interest as 17 regulators be in getting the designers through regulations 18 or requirements or something, to explicitly consider 19 shutdown risks in their design. It doesn't seem like you 20 have done that. 21 MR. DONOllEW t I guess I am not sure if the fact 22 proposed in these criteria, that we are avoiding the fact of 23 any particular accident. How you apply it and how you look 24 at the accidents, I think is separate. I don't think we are 25 talking about throwing away what has been learned on light ] ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

143 1 water reactors. O 2 I still say things like USI's when that is done as 3 pointed out as Part 52, when that is done I think things 4 like that will como out then. 5 MR. KRESS: The way you would handle that if you 6 knew about it, was make the shutdown accidents one or more 7 of the design basis accidents. Nothing here precludes them 8 from doing that. 9 MR. CARROLL: It doesn't preclude them. It 10 doesn't really hit them over the head and say -- 11 MR. KRESS: It doesn't say -- 12 MR. CARROLL: You should really be looking at that 13 question. 14 MS. SLOSSON: It certainly is something that wo } 15 will need to look at. I don't think we meant to exclude it 16 purposely. That's a good point, and we should consider it. 17 MR. DONOHEW: Is this not a question let's say, 18 maybe asking Ed Throm again on accident values. 19 MR. CARROLL: Sure, it belongs there too. Every 20 time you talk about preventing or mitigating accidents it 21 seems to me you ought to say including accidents that can 22 occur during shutdown or under low power operating 23 conditions. 24 MR. WILKINS: Of course, this issue is one of how l 25 you identify the structure, systems and components. i i Q ANN RlLEY & ASSOCIATES, Ltd. COud Repoders 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950 l

144 1 HR. CARROLL: It's just that I didn't think of it O 2 when we were talking about the -- 3 MR. WILKINS: The accident evaluation. 4 MR. CARROLL: The accident evaluation issue. 5 MR. WILKINS: Let me see if I can ask a different i 6 kind of question. i 7 MR. CARROLL: Before you do that, one of the 8 things that I worry about are uomo of these plants at least 9 to the extent that we have been briefed on them, have sort 10 of strange refueling schemes and that sort of thing. 11 Certainly, that's a shutdown risk that has to be considered. 12 MR. WILK1dS: CANDU has a different kind of 13 refueling scheme which is not a shutdown risk but is a very 14 real one just the same. 15 MS. SLOSSON: Right, it is. 16 MR. PIERSON: We are going to be doing that, of 17 course. As in the case with the passive reactors, in some 18 cases you find that as you are doing these refueling modes 19 you are not only -- as in the case of some light water 20 reactors -- you are violating the containment. You are 21 violating the pressure vessel, you are violating-the 22 integrity. So, some of the accident consequences-could in 23 fact be very serious. 24 I think what they are trying to do here is what is 25 different, noticeably different from the light water O ^"" ai'ev a ^ssoci^Tas. 'ta-Court Repo-ters 1612 K. Street, N.W., Suite 300 l Washington, D. C. 20006 (202) 293-3950 l \\

145 1 reactors. I think the case for the shutdown power in low 2 power operations is at this time, we would expect that we 3 would pursue it along the same lineu that we are going for 4 the passive light water reactors, looking at the 5 containment, looking at the refueling and looking at this 6 process. 7 I don't think anybody is implying that we want to 8 avoid or not include that. 9 MR. CARROLL: My only point is, why don't you say 10 it. 11 MR. WILKINS: The question I was going to ask 12 deals with the fact that in your writo up you refer only to 13 the MHTGR and you say they are the only one who are not 14 using the current LWR criteria both for safety 15 classification. 16 Is it not a fact that the others will label some 17 SSCs as non-safety related on the grounds that they don't 18 believe that they are needed to maintain RCP integrity for 19 example, whereas the identical component or the component 20 doing the identical function at least would so be labeled in 21 an LWR. 22 MR. DoNOHEW: I think the basic answer to your 23 question is yes. The thing is what we are trying to convey I.- 24 here is the fact that the other pre-applicants have proposed l 25 criteria which basically either is in fact the restatement-(] ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

146 1 of this or is consistent with this. Now, you have to go O2 through and do an evaluation of the specific proposal. 3 What we did find out is that there was one pre-4 applicant which was just so significantly different that we 5 could deal with this issue on the iact that they were not 6 being consistent with the LWR practices. After going 7 through and looking at their design and having discussions 8 with the pre-applicant, we felt we had reached a point whero 9 we could propose to the Commission that we should continue 10 with the current practices. 11 I guess it's the defense-in-depth, that we did not 12 see a justifiable reason to go away from current practices. 13 MR. WILKINS: Although, you hold out the 14 possibility that you will be consistent with whatever 15 position is later adopted for passive LWR's, which may 16 differ from current practices. 17 MS. SLOSSON: Yes. 18 MR. CARROLL: On page 27 on the first paragraph 19 under pre-applicant position the second sentence says somo 20 of those systems are entirely passive. Passive is a term 21 that has caused a lot of confusion in the past. You say 22 entirely passive, with no moving components or operator i 23 action. 24 MR. DONOHEW: Correct. 25 MR. CARROLL: Is a DC powered valve -- not a valve O ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

147 1 2 MR. DOHOllEW: Would you like me to explain which 3 system I was thinking about when we said it? 4 MR. CARROLL: Sure. 5 MR. DONOllEW: I would say that's for PRISM it's 6 RVAC's, reactor vessel air cooling s/ stem. For MHTGR it's 7 RCCS, reactor cavity cooling system. You have an enclosed 8 area, topologically separate from what would be called 9 containment, separate from containment. It's not inside 10 containment. You have -- it operates continuously. 11 It's not like you can start it. 12 MR. CARROLL: I know what it is. 13 MR. DONOHEW: There are no Valves that move, air 14 falls down when it's -- cool air falls down and hot air 15 rises. 16 MR. WARD: What about the PRISM or the PIUS ram 17 valves? 18 MR. DONOHEW: I guess I have to ask Dino to answer 19 that. I am not familiar enough with that. 20 MR. WILKINS: I wouldn't call it entirely passive. 21 MR. CARROLL: No. Nor, are they the subject of 22 this particular section. This only deals with the gas 23 cooling -- 24 MR. DONOHEW: Correct. That's correct. 25 MR. WILKINS: Are there further questions on this ]\\ ANN RlLEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

148 1 topic? O 2 (No response.) 3 MR. WIIZINS: What I would like to do next in for 4 the Subcommittee to have a brief internal discussion about 5 the substance of its recommendations to the full Committee 6 and perhaps suggestions to the drafter of whatever letter 7 the full Committee might write. That in I. We can go off 8 the record now. 9 [Whereupon, at 1:25 p.m., the tranneribed portion 10 of the meeting concluded.] 11 12 13 O" 15 16 17 18 19 20 21 22 23 24 25 Q ANN RILEY & ASSOCIATES, Ltd. Court Reporters 1612 K. Street, N.W., Suite 300 Washington, D. C. 20006 (202) 293-3950

REPORTER'S CERTIFICATE This is to certify that the attached proceedings before the United States Nuclear Regulatory Commission In the Matter of: liAME OF PROCEEDIIICI Subcommittee on Advanced Reactor I)esigna DOCKET NUMBER: PLACE OF PROCEEDI!1G: Bethesda, Maryland were held as herein appears, and that this is the original transcript thereof for the file of the United States liuclear Regulatory Commission taken by me and thereafter reduced to typewriting by me or under the direction of the court reporting company, and that the transcript is a true and accurate record of the v foregoing proceedings. . i l <.'r uJ ) { l : <> ( Official Reporter Ann Riley & Associates, Ltd. O v ________________________-._-______________-_----------------_--------_._______________________________.-----__________a

INTRODUCTORY STATEMENT BY THE CMAIRMAN OF THE ADVANCED REACTOR DESIGNS SUBCOMMITTEE 7920 NORFOLK AVENUE, ROOM P-110 BETHESDA, MARYLAND JANUARY 6, 1993 The meeting will now come to order..This is a meeting of the ACRS Subcommittee on Advanced Reactor Designs. I am Ernest Wilkins, Chairman of the Subcommittee. The ACRS Members in attendance are James Carroll, Ivan Catton, Pete Davis, Thomas Kress, Carlyle= Michelson and Paul Shevmon. ACRS Consultant in attendance is David Ward. The purpose of this meeting is to discuss the key policy issues that the NRC staff has identified for the MHTGR, PIUS, PRISM, and CANDU 3 advanced nuclear power plant designs. Dr. Medhat El-Zeftawy is the Cognizant ACRS Staff Member for this meeting. O The rules for participation in today's meeting have been announced as part of the notice of this meeting previously published in the Federal Register on December 23, 1992. A transcript of the meeting is being kept and will be made available as stated in the Federal Register Notice. It is-requested that each speaker first identify himself or herself and speak with sufficient clarity and volume so that he or she can be readily heard. We have received no written comments or. requests for time to make oral statements from members of the public. (Chairman's Comments-if any) ' We will proceedr with the meeting and I call upon Mrs. M. Slosson of NRR to begin. l. I O 1 W,d.g.-we--gr e,m .%..-y--- e-y.e.- 9.,. p p.w.w=ryr _fgmm--g-r-t ei*y ---en+e y 9-=w_--wr*- w --r---r.m-rwe- -w-+m'ymr-w+MT- -c'----+c*m-y. yer e-** wu D --us W1y q tr $g--F-r-g

h O O. O,O ~ NRR STAFF PRESENTATION TO THE ACRS [ ADVANCED REACTOR DESIGNS SUBCOMMITTEE j t i ADVANCED REACTOR AND CANDU 3 POLICY ISSUES l January 6,1993 1 i Marylee Slosson Beth Wetzel j i. Edward Throm Dino Scaletti l Joseph Donoghue Jack Donohew i 2 l t

Contact:

Marylee Slosson 1 Phone: 504-1111 l 1 j 4

i-7 i ADVANCED REACTOR POLICY STATEMENT OBJECTIVES Encourage earliest possible interactions of applicant, vendor, and e government agencies with the NRC; 1 e Provide all interested parties, including the public, with the Commission's views concerning the desired characteristics of advanced reactor designs; and 1 Express the Commission's. intent to issue timely comment on the e implications of such designs for safety and the regulatory l process. 2 l

O O O PREAPPLICATION SAFETY EVALUATION REPORT (PSER) SCHEDULES SECY-91-161 SECY-92-393 PRISM 11/92 12/93 CANDU 3 06/93 12/94 PIUS 07/93 04/95 MHTGR 12/92 12/95 3

o o o i i POLICY.lSSUE APPLICABILITY u Category issue PRISM MHTGR CANDU 3 PIUS A. Accident Evaluation X X X X B. Source Term X X 'X CL Containment Performance X X X X D. Emergency Planning X X X. f Category 1 E. Reactivity Control X t F. Operator Staffing X X X X f [ G. Residual Heat Removal X X X H. Positive Void Reactivity X X I Category

1. Control Room Design -

X X X X { 2 J. Safety. Classification X L 9 4

O O O ACCIDENT EVALUATION

Background

Advanced Reactor Policy Statement,1986. NUREG-1226, " Development and Utilization of the NRC Policy Statement on the Regulation of Advanced Nuclear Power Plants," 1988. Safety Goal Policy and Severe Accident Policy. PRISM, MHTGR, and PIUS preapplicants submitted accident evaluation schemes more conservative than LWR practices. Goal: to support reduced emergency planning requirements. t 5 i I

O O O ACCIDENT EVALUATION (continued) e issue Selection of accidents and acceptance criteria for the advanced reactors and CANDU 3. Recommendations Select events deterministically; supplement with PRA insights. Establish event categories consistent with LWR practices; encompass less likely event sequences. l Establish consequence acceptance limits consistent with i Commission guidance with appropriate conservatisms to account for uncertainties. i 6 1--ep-. + - - y + w-g + = 3 ~ m ,.,wg+e .,-w.

O O O ACCIDENT EVALUATION (continued) Develop methodologies / assumptions consistent with LWR practices. Determine source term in accordance with proposed SECY paper Section B guidance. Select set of events deterministically to assess safety margins, i determine source term, and identify containment challenge. Select external events deterministically, consistent with LWR practices. 4 Evaluate multi-module reactor designs considering scenarios allowed by proposed operating practices. 7 1

O O O SOURCE TERM DETERMINATION

Background

Preapplicants proposed mechanistic source terms. l i NUREG-1465, " Accident Source Terms for Light-Water Nuclear Power Plants," Draft,1992. Recommendations will be consistent with the revisions to 10 CFR Part 50 and Part 100. 8 i

o o o j SOURCE TERM DETERMINATION (continued) t e Issue I Determination of source term. i i s' Recommendations j Develop source term based on mechanistic analysis provided l that: i i Performance data on fuel during normal and off-normal j l4 conditions is well understood. l t Fission product transport can be modeled adequately. Events are selected to bound credible severe accidents and design dependent uncertainties. ( i 9 l t l'

O O O L CONTAINMENT PERFORMANCE .*. Background l MHTGR and CANDU 3 designers propose containments different from traditional essentially leak-tight structures. i PRISM and PIUS maintain leak-tightness, but include non-traditional approaches to containment system design. Changes are result of designer emphasis on accident prevention versus accident mitigation along with incorporation [ of passive, decay heat removal methods. A'dvanced Reactor Policy Statement encourages enhanced safety. margins suggesting that designs with simplified or passive systems should be considered. The Policy also recommends a multiple barrier approach to radiation release. 1 5 10

O O O CONTAINMENT PERFORMANCE j (continued) Issue 4 Should advanced reactor designs be allowed to employ l alternatives to " essentially leak-tight" containment structures? e Recommendation Move away.from prescriptive containment design criteria and utilize a performance standard based on accident evaluation cdteda. l' Containment must be adequate to meet onsite/offsite radionuclide release limits for design basis accidents. t i 11 i i l

LO O O CONTAINMENT PERFORMANCE (continued) Containment will be evaluated for a deterministically selected severe accident event: For approximately 24 hours following the onset of core damage; leakage no greater than the limiting leak rate used in, accident' evaluation, and structural. stresses within acceptable limits (ASME Level C). Following this period, the design must prevent uncontrolled releases of radioactivity. 12 - _ _ _ = _ -.

O O O EMERGENCY PLANN!NG l t

Background

[ Based on conceptual design coupled with accident evaluation, l source term, and-containment, the preapplicants (PRISM, MHTGR, and PlUS); proposed reduced levels of emergency planning. l Designer's objective is to show offsite releases are within the lower-level EPA protective action guidelines (1 rem whole-l- body, 5 rem thyroid) at the exclusion area boundary. i issue l h Level of emergency planning required for advanced reactors. 13

L O O O-t i EMERGENCY PLANNING (continued) Recommendations Advanced reactor licensees should develop offsite emergency plans.: Provisions.for periadic emergency exercises should be l L developed. i Relaxation from existing LWR requirements may be considered l based on accident evaluation review. Relaxations may include prompt notification requirements, EPZ l size and frequency of exercises. Approach will.be consistent with passive LWR decisions f 14

o-o o } REACTIVITY CONTROL SYSTEM i o

Background

GDC 26 requires two independent' reactivity control systems. One system shall use control rods. l 4 1 The other system shall be capable of controlling planned reactivity changes to assure fuel limits are not exceeded. j f Issue Is a reactivity. control system without control rods acceptable. Recommendations A reactivity control system without. control rods should not-1 necessarily; disqualify a reactor design. 15. i i 1 1

O O O l REACTIVITY CONTROL SYSTEM (continued) A design without control rods may be acceptable if the applicant can demonstrate an equivalent level of safety for l reactor control and protection: Reliability and efficacy of scram function. Suppression of osciliations. Control of power distribution. 1 Shutdown margin. I Operational control. 16 c

I O O O REACTIVITY CONTROL SYSTEM (continued) 880 m 2200 ppm Boron M gmy R CT GENERATOR g di = SCRAM VALVES h .. 2-iib! "E"euue Coo CONCRETE PRESSURE VESSEL PlUS SCRAM SYSTEM i 17 q 1

O. 'O O s OPERATOR STAFFING AND FUNCTION !ssue i -e 1 Acceptability of reactor staffing requirements less than specified-for LWRs. Recommendations jj Operator staffing may be design-dependent based on function and task analyses considering the following:- F Effective response to worst case power maneuvers, refueling and maintenance activities, and accident conditions. Ability to handle accident on one unit with proposed number i of operators, less one, while other units are taken to cold j shutdown. i 4 18-

O O O OPERATOR STAFFING AND FUNCTION (continued) Function and task analyses testing needed to assure that All site reactors can be shut down safely during; l Complete loss of computer control capability. Complete station blackout. Design basis seismic event. Adequacy of analyses shall be tested on a control room prototype. f 19 i

.) O O O l RESIDUAL HEAT REMOVAL

Background

Advanced reactor designs use completely passive, safety-grade RHR designs. l Designers contend that the capability to continuously monitor system performance helps allay redundancy and diversity concerns. I Related issues for passive LWRs: Ability of passive systems to achieve safe shutdown. a. b. Definition of passive failure. Treatment of non-safety systems in their relationship with c. passive systems challenges. 20 l

O O O' j RESIDUAL HEAT' REMOVAL (continued) 1 1 Issue { Are' single, completely passive, safety-related: residual heat i removal systems acceptable for advanced reactor designs? 3 Recommendations 1 j a 1 Reliance on a. single, completely passive, safety-related RHR-j system;may be' acceptable due to unique advanced reactor design features. I Treatment of highly reliable, non-safety related: backup j systems will be consistent with decisions on passive LWR design requirements. 21

O O OJ POSITIVE VOID REACTIVITY COEFFICIENT l

Background

i GDC 11, Reactor inherent Protection. 1 Two reactors (PRISM and CANDU-3) have positive void 1 coefficients.. Issue Acceptability.of design in which the overall inherent reactivity. f tends to increase under specific conditions or accidents. l i 22

O O 0 1 f POSITIVE VOID REACTIVITY COEFFICIENT (continued) Recommendations A positive void coefficient should not necessarily disqualify a reactor design. PRISM and CANDU 3 preapplicants should analyze the consequences of events such as ATWS, unscrammed LOCAs, delayed scrams, and transients that could lead to core damage as a result of positive void coefficients. Consideration of changes in designs to mitigate consequences j of these accidents should depend on estimated probability of i accidents and severity of consequences. 1 ? 23 l t ~, -...

1 O o O i CONTROL ROOM AND REMOTE SHUTDOWN AREA-

Background

Some advanced reactor designs have proposed to shift safety emphasis from the control room to a remote shutdown area. Applicants suggest that stringent control room requirements l i are obviated by safety grade remote shutdown arer, increased automation of plant. controls and passive safety systems. Related issues-raised for passive LWRs include: i

a. Control room habitability.

i l

b. Duration of control room operability.

i

c. Digital systems common mode failure.

24 ~' ~

o o 6 CONTROL ROOM AND REMOTE SHUTDOWN AREA l (continued) l l l 8 lssue Are current control room and remote shutdown requirements i fulfilled by a safety grade remote shutdown area and a non-safety grade control room? l e Recommendation 3 Apply current LWR requirements / guidance until passive LWR-policy.for design requirements of control rooms and remote shutdown facilities is finalized. L L i 1 25 -{ l t 1 ? i 4 . 1 1,

ELMf]LM O O O SAFETY CLASSIFICATION 1 l e

Background

i Current LWR Requirements: GDC 1, Quality Standards and Records. 10 CFR 50.55a - Identifies those ASME Section 111, Code Class 1 components of LWRs important to safety. 50.49(b)(1) and Appendix A.Vi(a)(1) / Part 100 List the following criteria to define safety-related SSCs: Needed to maintain RCPB integrity. Needed to shutdown the reactor and maintain it in a safe conditions. Needed to prevent / mitigate accidents that could result in dose consequences comparable to Part 100 guidelines. 26

O O O SAFETY CLASSIFICATION (continued) l Issue What criteria should be applied to advanced reactor designs to determine safety-related structures, systems, and components. Recommendations Apply the following criteria for determining safety-related l structures, systems, and components; Needed to maintain RCPB integrity. Needed to shutdown the reactor and maintain it in a safe conditions. Needed to prevent / mitigate accidents that could result in dose consequences comparable to Part 100 guidelines. 27 i T~}}