ML20137U235
| ML20137U235 | |
| Person / Time | |
|---|---|
| Issue date: | 12/03/1985 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1472, NUDOCS 8512090237 | |
| Download: ML20137U235 (317) | |
Text
{{#Wiki_filter:, c 06tSt' M 7c2 l 1 UN11ED STATES C- NUCLEAR REGULATORY COMMISSION , l l IN THE MATTER OF: DOCKET NO: ADVISORY COMMITTEE ON REACTOR SAFEGUARDS . SUBCOMMITTEE ON DECAY HEAT REMOVAL SYSTEMS - OR G N AL I s
~ *We -
LOCATION: WASHINGTON, D. C. PAGES: 209 - 446 DATE. TUESDAY, DECEMBER 3, 1985 ACRSomCECOPY i a m st m ess m e-ACE-FEDERAL REPORTERS, INC. OfficialReporters 444 North CapitolStreet t' Washington, D.C. 20001 es12090237 eb1203 n ' PLg7gCRS PDR , (202) 347-3700 NATIONWIDE COVERACE
._ - . . . _ , , _ _ . . _ _ . . _ _ . _ . ~ . . . . ,
c-PUBLIC NOTICE BY THE UNITED STATES NUCLEAR' REGULATORY COMMISSIONERS' ADVISORY COMMITTEE ON REACTOR SAFEGUARDS TUESDAY, DECEMBER 3, 1985 The contents of t_his stenographic transcript.of the proceedings of the United States Nuclear Regulatory Commission's Advisory Committee on Reactor. Safeguards (ACRS), as reported herein, is an uncorrected record of the discussions recorded at the meeting held on the above date. No member of the ACRS Staff and no participant at h this meeting accepts any responsibility for errors or inaccuracies of statement or data contained in this transcript. 9 9 9 b
. _ . _ . . _ . .._. . _ . ~ - . . _ _ . _ - . . . _ _ . _ . _ _ . _ , . . . _ . _ . . - . . , . . , . . _ . . _ . . . . . . _ , , . . . . - - _ , _ , ,
/
l 1 CR25279.0 209 l CPT/dnw I UNITED STATES OF AMERICA s 2 NUCLEAR REGULATORY COMMISSION 3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4 SUBCOMMITTEE ON DECAY HEAT REMOVAL SYSTEMS S Nuclear Regulatory Commission 6 Room 1046 1717 H Street, N.W. 7 Washington, D. C. 8 Tuesday, December 3, 1985 9 The subcommittee meeting convened at 8:30 a.m., 10 Mr. David A. Ward presiding. II ACRS MEMBERS PRESENT: 12 MR. DAVID A. WARD x.) 13 MR. JESSE EBERSCLE 14 MR. HAROLD ETHERINGTON 15 MR. CARLYLE MICHELSON 16 MR. GLEEN A. REED j 17 MR. CALTON, Consultant MR. DAVIS, Consultant 18 MR. MARCHESE, Consultant MR. ERICSON, Consultant 19 20 21 22 23 24 e9 Reponm, Inc. 25
25279.0' 210 o ( V[RT-1 P_ R_ O_ C E E D_ I N_ G S, 2 MR. WARD: The meeting will now come to order. 3 This is the second day meeting of the Advisory Committee 4 Reactor Safeguard Subcommittee on decay heat removal, decay
- 5' heat removal systems. I'm David Ward, the Subcommittee 6 Chairman. We'll continue the meeting with the agenda item 7 on status of USI A-45 resolution.
8 I would like to remind the committee that this 9 is not our final meeting on the topic. The Staff has not 10 yet finished their work, but we've asked them to come in 11 and give us a status report so that hopefully we can more
.-r'sb 12 quickly react or get on board with the final report coming
~13 up in a few months, we hope.
14 I would like to ask Mr. Marchese of the Staff to 15 make some remarks. 16 MR. REED: Phr. Chairman, unfortunately I didn' t 17 make it yesterday. I'm very sorry. I wanted to be here 18 for the auxiliary feedwater part of the meeting. 19 I do want to make a point that, from the 29 operator's point of view -- and I'll try to represent 21 operators -- auxiliary feedwater systems, that technique is 22 the technique for removal of decay heat, is a bit complex. 23 It involves four systems functioning: service water, 24 auxiliary feed, the circulation primary system and various (). 25 electric systems. And it bothers me that we have had so ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
25279.0 211 RT 1 many near n.isses in the last few years. 2 I think you get this because of the complexity, 3 trying to move steam and water in common pipes and the 4 water hammer events and all the other events and the 5 leakback events and the valve problems that go along with 6 it. 7 I guess from yesterday's meeting you didn't find 8 out too much about the reliability, but I like to think 9 that auxiliary feedwater is a key system to decay heat 10 removal. I'm not so sure that its reliability is good 11 enough to take care of this very serious core heat need. 12 MR. WARD: Thank you. . 13 MR. MARCHESE: Good morning, everyone. e My name 14 is Andy Marchese, I'm the task manager for the USI A-45 15 resolution and chief on shutdown heat removal requirements. 16 As Mr. Ward mentioned, this is a status briefing on where 17 we stand in the program at this particular point in time. 18 As *you know, we have had numerous meetings with 19 the Subcommittee -- I think a total of about 16 over the 20 past several years -- and I envision having quite a few 21 more in the months ahead. 22 (Slide.) 23 This is a list of the items that we plan to 24 cover today. We are going to start out with a discussion n () 25 of the plant analysis revision activities that have taken ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800-336-6646
25279.0 212 RT 1 place over the last several months; specifically, since we 2 met last May. And then we are going to focus in on the 3 results of the revised Point 3each and Quad Cities plant 4 analyses. 5 After that, we'll review the results that have -- 6 or the analysis results that have been generated on two 7 specific systems that I think you've asked for. One is the 8 dedicated primary blowdown system that Mr. Reed has
.9 proposed, and coupled with that, we'll also discuss the 10 preliminary analysis results that have been generated on a 11 high pressure RHR system. We'll discuss those together
('S;. 12 later on. 13 The next item, you have also asked how input 14 from A-45 will be factored into reaching a decision on the 15 need for PORVs on those cluster engineering plants that 16 don't have them. I'm going to try and explain how the 17 results of this program will be factored into that decision 18 process. And after that we are going to discuss the 19 sabotage analyses that have been conducted under A-45. 20 By the way, this part of the meeting will have 21 to be closed since there is safeguarded information that 22 will be presented. 23 And, finally, we'll give you the latest e, stimate 24 of the major schedular milestones on the program. () 25 (Slide.) ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336 4646
F-25279.0' 213 T 1 Okay. By way of introduction, as you'know, 2 ' rough draft reports for Point Beach and Quad Cities were 3' issued in May. We had a meeting with the Subcommittee on 4 May 30th. During the month of June, there were detailed 5 comments on those two reports that were issued by the Staff; 6 specifically, the reliability and risk assessment. Staff 7 members of NRR did a detailed review of those reports and 8 they had quite a few detailed comments. 9 There was a lot of Staff discussion of these 10 comments with Sandia, meetings, numerous meetings were held
- 11. in terms of trying to reach agreement on how best to handle
(~Nj 12 these comments in terms of the analyses that were being 13 done on Point Beach and Quad Cities. 14 Finally, agreement was roached on how to handle 15 these items, and it turns out that, as Dr. Ericson will 16 explain a little later, to factor these comments into the 17 analyses took a lot of additional work, which is one of the 18 reasons why we are back here again, discussing these 19 reports, because the results have changed significantly 20' because of the comments that were factored into the 21 analyses. You'll be hearing more about this later on in 22 terms of how the results have changed. 23 (Slide.) 24 This is a list of the items that the Staff took ( [ (,) - 25- exception to in the review of the original plant analyses ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 6646
25279.0 214 RT 1 'for Point Beach and Quad Cities. As I mentioned, a 2 considerable amount of time was spent-on reaching agreement
.3 : on-how best to handle-these areas.
l4 .The most important areas that required 5 significant additional analyses, let me just point a couple 6 of these out:- The treatment of common mode failures took a 7: . lot of addicional work. I think the Subcommittee at the 8 May-30th meeting was also concerned about this. 9- Support system initiators was another important 10 'one; for example, loss of DC bus as an initiator -- that 11 . required a lot of additional work. Ef -v 12. , Treatment of feed and bleed -- and this relates
- ' (_)
13 ~to how we handle the auxiliary feedwater system in terms of 14 common modes -- also took a lot of additional analyses and 15 a lot of -- had an important effect on the results. 16 The next-one, insensitivity of person-REM to 17 source term used, this has to do with the way the CRAC0 18 handles calculation of off-site releases, and we'll be 19 getting into that a little later on. 20 And.then finally, the other important one was 21 there was some inconsistent treatment in the way station 22 blackout was handled in Point Beach and Quad Cities; and 23 now we feel both of the reports are handling station 24 blackout sequences in a consistent manner. We'll go () 25 through some of these as we go through the plant analyses. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coserage 800 336 6646
;25279.0' 215 f RT
_ () 1 I'm going to come back later on and talk about 2 the CE/PORV issue, schedule and so forth at the end of the 3 day. But at this point, I'd like to turn the meeting over 4 to.Dr. Dave Ericson, from Sandia, who is a program manager 5 at Sandia Labs, and Dave is going to discuss the plant
~
-6 analysis revision activities and plant analysis status at 7 this point in time.
8 MR. WARD: Andy, first a question. Is someone 9 going to talk about the coolant pump seal LOCA, further? 10 MR. MARCHESE: In terms of how we are handling 11 it in the program?
~q 12 MR. WARD: Yes.
D' 13 MR. MARCHESE: Yes. I think we can get into 14 that a little later on. 15 MR, WARD: Okay. 16 MR. DAVIS: Excuse me, Andy, I also had a quick 17 question. Did the utilities have an opportunity to review 18 the draft? 19 MR. MARCHESE: Yes. 20 MR. DAVIS: Did you receive any comments from 21 them? 22 MR. MARCHESE: No. We submitted the revised 23 reports to them and asked for their review, I think by 24 mid-December. (). 25- MR. DAVIS: Thank you. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 80G 336-6646
25279.0 216 RT 1 MR. WARD: Any other questions for Mr. Marchese? 2 Thank you. 3 MR. ERICSON: I will attempt, in these few 4 introductory remarks, to be even briefer than Andy and 5 maybe we can get a leg up on our schedule today. 6 (Slide.) 7 As Andy has indicated, as a result of the review 8 by the Staff and then the subsequent dialogue with us, we 9 have reworked the internal analysis on Quad Cities and 10 Point Beach, and at the same time adjusted and made some 11 adjustments into the analyses we are doing on the other cm 12 plants. 13 Basically, if you go back through that list of 14 questions, what it has amounted to is that we have gone 15 back with a broader treatment of common mode, some 16 additional treatment of the way -- or some additional 17 inclusion of actuation in the modeling; specifically, 18 looking at long-term blackout and, where appropriate, we 19 havo gone back and done a modest reexamination of human 20 factors and recovery issues. Most of that in the basic 21 risk assessment mode. 22 The special emergency people -- that would 23 include the seismic, the fire, spray, flood -- have indeed 24 gone back as a result of some of the comments. We have n 25 done some review and have modified a few things. In ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
25279.0 217 RT 1 addition, the seismic area, modeling was coming along, and 2 we've changed a few things to better reflect a way to do 3 that modeling. 4 In the sabotage area, what we discussed in an 5 earlier meeting was a very qualitative approach. We have 6 since gone back and done what I choose to call a modest 7 conditional quantification. We'll get into more of that, 8 obviously, later this afternoon. You are aware, of course, 9 that there are a number of efforts going on, attempting to 10 quantify or put -- perhaps put sabotage into more of a risk 11 perspective. We have done some things in that regard on a
~
12 conditional basis, and we'll get into those details later. ( 13 (Slide.) 14 As to where we are, then, in the entire scheme 15 of prerating the specific analyses for submission to the 16 Staff for their use in preparing the subsequent 17 documentation, the resolution package, Quad Cities has been 18 submitted. The revision is in to them. They have had it 19 for a while now. It has also progressed through our own 20 internal management channels; copies are at Commonwealth 21 Edison for comment. We requested their comments by about 22 the middle of this month. 23 After receiving those, the document will be 24 submitted to the Staff for publication. It is, so far as () 25 we are concerned, as a study it is ready to go. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80 4 336-6646
25279.0 218 RT 1 Point Beach is in a somewhat similar situation. 2 Our draft is in. It is now in our line management approval 3 process at Sandia. A copy has gone to Wisconsin Electric 4 for comment, and because we have just a lot of cleanup 5 things, in terms of artwork in order -- for publication, 6 the actual printed version in terms of the NUREG CR 7 probably won't be able to go until probably the February 8 time frame. 9 It indicated here that the Cooper draft would be 10 submitted. I had planned to bring it along today, however, 11 we still have a few tables in the typewriter when Steve and 12 I left yesterday, so it will be coming in to the Staff next s 13 week, and also entering the internal Sandia review process 14 at the same time. This will be the first time that the 15 Staff has seen the Cooper report. That will be a new 16 document to the Staff. 17 Turkey Point is the -- the draft of that is 18 imminent. Gary Sanders is responsible for that analysis 19 and he's in the process of wrapping up the final sections 20 of the value impact now. In fact, I think most of it is at 21 home, waiting to be typed. When we get back we'll go 22 through the final reviews ourselves and double-check 23 numbers and this sort of thing. So we expect to have that 24 back in here before the holidays, certainly. r') 25 Saint Lucie is under way. In fact, the internal ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage M33M666
25279.0 219 RTt 1 PRA-type things are in the first stages of quantification. 2 The internal emergency people are off and running on their 3- things. .And our target date for submitting a draft is' 4 February; part of-the reason for that being -- is that in-5- order to do the job right we make a second plant visit, you 6' will recall, after we propose the modifications. And.due 7, to plant schedules, holidays and that sort of thing, we 8 won't be able to do that, until January. We are shooting 9 for early January to do that and so the report, then, 10 subsequently will have to come in February. 11 MR. WARD: That's a more pleasant time to visit 12
. Saint Lucie.
13: MR. EBERSOLE: I'd much rather go to Saint Lucie 14 in January than Point Beach. I think Chuck is here this 15_ morning; it's kind of grim up there these days. In fact, t
-16 Mr. Reed could probably tell us a lot of stories of what 17 it's like on the shores of Lake Michigan in January.
18 ANO .1 is on way; we have the f ault trees 19 available for some time initially and entered into computer. 20 Wally has been going back through those, double-checking 21 them, making sure they are consistent with what we've added 22 since we went back through the Point Beach analysis, and he 23 will be beginning to crank out numbers from the internal 24 part of that analysis when we go back next week. The 25 special emergency work is in progress, and, again, we have ( ]) ACE-FEDERAL REPORTERS, INC.
. 202 347-3700 Nationwide Coverage 800 33H446 l
25279.0 220 RT 1 a target date of February because we won't make a plant 2 visit until after the first of the year, probably. 3 Trojan is temporarily on hold at the moment. 4 That's primarily an internal problem for me. Some of my 5 younger staff members have been lured by the siren song of 6 other opportunities and so we are down temporarily on 7 staffing, but we expect to have that one back up and 8 running very directly. We put it on hold because it is, of 9 course, another Westinghouse plant, so we felt we should 10 get the others done first and then come back to that. 11 MR. WARD: What was unique about Trojan? 3 12 MR. ERICSON: Trojan just completes the family. 13 Then we go 2, 3, and 4 loopers. 14 MR. EBERSOLE: Mr. Ericson, could you comment 15 about how you can draw a line down between the total 16 efforts between the boilers and the PWRs and how the work 17 is going and what's the extent of it in front of us and 18 behind us? 19 MR. ERICSON: As far as specific plant analysis 20 we are done with boilers. We have had two and they are. 21 done. At this time. 22 MR. EBERSOLE: Okay. 23 MR. ERICSON: And -- well, the next viewgraph -- 24 let me comment on where we are in terms of pulling things 25 together. ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-336 M86
25279.0 221 RT 1 MR. WARD: Let me ask you just to clarify -- 2 maybe it's our record. At one time I had the impression 3 that the combustion engineering plant you were going to do 4 was SONGS. In fact, we quote you in a set of our minutes 5 as saying it was SONGS. That may be an error, that's why I 6 want to clarify. 7 MR. ERICSON: At one time I recall we discussed 8 rather extensively doing "a newer design," and we were 9 asked to look at the possibility of doing several new ones 10 and we suggented, I think, if we did SONGS and perhaps 11 Grand Gulf o'r something like that might be the ones to do.
-s 12 But we have not specifically put those on the agenda at the 13 present time.
14 Our model -- we've always had Saint Lucie as the 15 one for the detailed analysis at this point in time. 16 MR. WARD: Okay. Insofar as studying the impact 17 of PORV or no PORV, you can do that with any plant, I guess, 18 is that your argument? 19 MR. ERICSON: Yes. In fact, I think Andy may 20 have some comments, and we will, too. In fact, Wally will 21 show you a sensitivity to feed and bleed on Point Beach. 22 We are doing the Saint Lucie sensitivity on all the 23 pressurized systems. 24 MR. MARCHESE: Let me just comment, we I ', 25 originally had nine plants in the program. The last two ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336-6646
25279.0 222
,_BRT
'( )
v 1 were more recent vintage. We were considering, as Dave 2_ mentioned, doing San Onofre or Palo Verde, which are - more 3 -recent Combustion Engineering plants. We also had a more 4
- recent vintage BWR, such as Grand Gulf. However, about six 5 months ago we decided to drop those two plants because of 6 budget and schedule problems that the program is having.
7 So right now they are not in the program. 8 MR. EBERSOLE: May I ask this question? In the 9 case of the two boilers you looked at, General Electric has 10 finally, after some 15-odd years, agreed to an ultra-simple 11 open cycle boiling process they refer to as UPPS, which
,~3 12 appears to have some of the merits of an extremely simple i
\" 13 boiling process just to keep the core covered and do 14 preliminary cooling through the special cooling and 15 subsequent discharge to the atmosphere, which implies the 16 only active cooling you have to have to continue the 17 cooling indefinitely is a low pressure pump and source of 18 water. Was that incorporated into your studies?
19 MR. ERICSON: We did not attempt to do a cost 20 estimate of backfitting. It was commented on and looked at 21 in both plants but we did not go into a detailed 22 examination of backfitting. 23 MR. EBERSOLE: The degree this would be able to 24 cope with most emergencies has not been defined. (~) t_/ 25 MR. ERICSON: In addition to these plant studies, ACE. FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336 6646
25279.0 223 BRT O 1 of course, there are a number of what I have chosen to call 2 "special topics." 3 (Slide.) 4 First of all, let me say that as far as a 5 summary report, pulling these seven plant studies, which 6 all make rather handy doorstoppers when they are laid down 7 on the floor, we will, of course, eventually pull the 8 summary report together. 9 As a first step of that, we are beginning to put 10 together, now, insofar as we can basad on the few plants, 11 some generic conclusions, observations, call them what you
- 12 will. And we will be providing that to the Staff in 13 January, not so much as a report per se but as input to 14 some of the work going on. It will be more in letter 15 report format. But to begin to help Andy and his 16 colleagues put some things together.
17 We have also looked at some special issues. For 18 example, what happens, given that some people have argued, 19 if there should be another major incident and somebody says, 20 well, let's close everything up? What are the costs, the 21 values associated with that? We have done some looking at 22 that with our consultants at UCLA. We have had a rough cut 23 at that, or first cut. I shouldn't say " rough"; a first 24 cut. That is being revised at the present time and we hope ( '; 25 to have that available to the Staff sometime after the ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-336 M16 l
25279.0 224 RT 1 first of the year. 2 We are also assisting as we can -- the 3 regulatory analysis is a Staff function, but we are 4 providing all the assistance we can in terms of getting 5 numbers in the right format for them and attempting to pull 6 together insights that we see in the data or in the 7 information we generated. 8 We also have a summary report on value measures 9 and the value impact structure, which is in progress. UCLA 10 is preparing this for us. 11 The value impact approach we have taken, of r 12 course, is that -- is based upon the , B&L handbook, internal 13 NRR guidance that they have been using since late 1984. We 14 have kind of pulled a lot of that together. If you will, 15 in some instances, a cookbook format for our uses. It has 16 been documented, so the approach will be available; and if 17 someone really wants to dig into how we are doing it, it's 18 there. 19 Of course, in each of the plant reports there is 20 an appendix that describes these things, but we will be 21 pulling a lot of this stuff together in one place. 22 MR. WARD: Dave, is this -- I mean the B&L 23 handbook -- does that include just on-site costs? Or 24 off-site costs? Does it include on-site costs, loss of f 25 replacement power -- ACE FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 4 645
25279.0 225 1 MR. ERICSON:' I think it does, but certainly not 2 to the extent that we have done it and the options 'we have 3 looked at. We have added a lot of the on-site stuff as we 4 have been seeing things, of course, in the course of the 5 . program. UCLA has assisted us in pulling those things 6 together in terms of a consistent format. 7 MR. WARD: But in your treatment you are 8 displaying both -- in both ways? 9 MR. ERICSON: Value impact based on off-site 10 dose only and display value impact based on off-site and 11 on-site together. 12 MR. WARD: Okay. Thank you. 13 MR. ERICSON: We display both. And now we'll be 14 displaying several alternatives. 15 MR. ETHERINGTON: Do you and your contractors 16 have several groups working in parallel on different 17 projects? 18 MR. ERICSON: Yes. UCLA is working -- at the 19 present time we have, essentially, about four contract 20 groups. We have outside assistants in the special 21 emergency areas with seismic analysis -- 22 MR. ETHERINGTON: That wasn't quite the question. 23 Do you have groups working in parallel on different 24 projects or do you clean up one project first? () 25 MR. BRICSON: Do you mean plant studies? ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33666M
25279.0 226 RT 1 MR. ETHERINGTON: Yes. 2 MR. ERICSON: Yes. Two in parallel, and right 3 now, three in parallel. 4 MR. ETHERINGTON: And the same with the 5 contractors -- 6 MR. ERICSON: It's an integrated team. 7 Contractors and Staff are working together on a plan. Some 8 parts of the analysis, special purchases in particular, we 9 have contract support. The internal PRA work we are doing 10 in-house at Sandia, using our internal resources. 11 MR. ETHERINGTON: I see. m 12 (S1ide.) 13 MR. ERICSON: At this point, then, I'm going to 14 ask -- unless there are other questions, I'm going to ask 15 Mr. Wally Crammond from our organization, from our Staff, 16 to come up and begin a look at the Point Beach work. 17 I might point out in the handout we'll switch 18 the order in a couple of places, but Wally will point that 19 out as we go. We have decided to combine things slightly 20 differently. 21 (Slide.) 22 MR. CRAMMOND: My name is Wally Crammond and I 23 have been responsible for the Point Beach analysis and in 24 tha beginning, some of the methodology of how we would go (i 25 about doing the analyses. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 8003364 46
l l25279.0 227 fBRT
- (J -
1 (Slide.) 2 As Dave pointed out, we have submitted this a 3 second time. I'll put this viewgraph up just briefly to 4 .show the connections between the transient event trees, and
.5 the LOCA event tree; and the branches that we do 6 sensitivity analysis on, the secondary blowdown branch and 7 the feed and bleed branch.
8 In the second go-around of the analysis, we have 9 included loss of a DC bus and loss of an AC bus. 10 (Slide.) 111 That's essentially the same as the other 12 transient event trees.
- (:) 13 This is just showing the containment event tree, 14 and one thing that's a little different about our analysis 15 than some analyses is the fact that we separated the 16 containment events from the core melt events. So the first 17 thing we do is come up with a core melt frequency and then 18 we associate with each sequence that's significant, all the 19 possible containment failure modes. And then from there we 20 obtain, or we determine which sequences fall in which 21 release categories and extrapolate into the risk space.
22 (Slide.) 23 MR. MICHELSON: Your containment failure modes 24- are as a consequence of core melt? Is that what you are () 25 saying?
. ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide coverage 800 336 6646
25279.0 228 RT 1 MR. CRAMMOND: Yes. We only look, in fact, for 2 the significant ones to cut down on computer time. 3 MR. MICHELSON: Did you also look at a 4 possibility of having a core melt in conjunction with a 5 failure to isolate the containment to begin with? In other 6 words, the purge valves would not close -- 7 MR. CRAMMOND: No. One thing we did do, we 8 assumed that failure of the containment would not in itself 9 lead to core melt, as has been assumed in the past. 10 MR. MICHELSON: No. I think that's probably not 11 too bad an assumption. My postulation is that you have a 12 core melt scenario developing, and during the process you -) 13 were unable to isolate the containment which was, perhaps, 14 normally opened. 15 MR. CRAMMOND: We didn't consider that. 16 MR. MICHELSON: Is there some reason why that 17 wouldn't be a fairly high probability scenario? On what 18 basis did you discount it? 19 MR. MARCHESE: Mr. Michelson, this program, I 20 think we mentioned this at some other meeting, we are 21 concentrating on small break LOCAs and transients. We are 22 examining the systems that respond to small breaks and 23 transients. We specifically excluded at TWS, large LOCAs, 24 event --
' 25 MR. MICHELSON: The containments have to bottle ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 8043364M6
25279.01 229 BRT'
.m U, -
1- up for small LOCAs, too, very definitely. 2- MR. MARCHESE: 'I think you are talking about RHR. 3: MR. MICHELSON: I'm talking about the purge 4 valves,' containment ventilation. If they don't close, it's 5 a loss of containment.- 6 HMR. MARCHESE: We did not look at that. 7 MR. DAVIS: I might mention,-that has been
.8 looked at, Carl,.in some of the recent PRAs.- No one has 9 found it .to be a significant risk contributor yet.
., 10 Generally.the probability of isolation failure is 10 to the
- 11 minus 3.
gs 12' MR. MICHELSON: The problem is, I think, you
# 13 don't-have any reliability data whatsoever on the ability 14' of these valves to close under the dynamic conditions of an 15 accident.
16 MR. DAVIS: For the accidents they are looking 17 at there won't be much in the way of dynamic conditions. 18 .These are small breaks. 19 MR. MICHELSON: If they are small breaks they 20 should not be much worse than a normal test, that's right. 21 I assume you use normal surveillance data for your 22 reliability numbers. 23 MR. REED: I could be in conflict of interest on 12 4 Point Beach, but I think I can give a factual statement. ((- 25 The purge valves at Point Beach are run in closed condition ACE. FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 33Ml646
25279.0 230 BRT O 1 and are never opened above 200 degrees Farenheit primary 2 coolant. 3 MR. EBERSOLE: I would like to have you discuss 4 just for a minute why you believe a containment failure 5 could not initiate a core melt. In view of the fact you 6 have abnormal environmental conditions in the containment, 7 you might have destructive rupture of the containment which 8 would destroy piping and components and equipment. I don't 9 see how you can eliminate containment failure unless you do 10 a very selective identification of how it fails as an 11 initiator of core melt. As a matter of fact, in the boiler -s 12 it's the royal problem -- because of their self-contained 13 heat sink and loss of AC power -- 14 MR. CRAMMOND: The basis we used for not 15 considering it was the traditional idea that if the 16 containment f ailed the sump would flash and you would lose 17 the water for recirculation and fail the low pressure 18 recirculation pumps. That we concluded, after talking with 19 the plant and other people, that that really isn't the way 20 it goes, in fact. 21 MR. EBERSOLE: That's just a point failure of a 22 singular kind. There are many other ways of failing those 23 systems besides that. 24 MR. CRAMMOND: In terms of the conditions in the (~) 25 containment we didn't addres,s that. ACE-FEDERAL REPORTERS, INC. i 202 347 3700 Nationwide Coverage 80433 M 646
.25279.0 231 RT r
1 MR. EBERSOLE: I don't see that your analysis is 2 complete by any means. 3 MR. CRAMMOND: There are a lot of things that 4 are -- we didn't do for which you can claim we are not 5 complete, ATWS and interfacing ATWS -- 6 MR. EBERSOLE: Or any containment failure 7 initiated meltdowns. 8 MR. CRAMMOND: Yes. That has to be true. 9 MR. MICHELSON: Just for clarification now, what 10 you-are going to present to us applies to Point Beach 11 'specifically, but are these techniques the same ones 12 applied across the board or just to apply to Point Beach? 13 MR. CRAMMOND: Essentially the same techniques 14 that we use for all our analyses; in particular, the PWRs. 15 The BWRs would be slightly dif ferent, but -- 16 MR. MICHELSON: Is it safe to assume that as you 17 go from one plant to another that you adjust these 18 techniques for the plant-specific conditions? 19' MR. CRAMMOND: If there are plant-specific 20 conditions. 21 MR. MICHELSON: For instance, if purge valves 22- are normally open, you would adjust your reliability study 23 accordingly. If they are normally closed, you would leave 24 it out, obviously. But these are general techniques, I r 25 ascume? ACE. FEDERAL REPORTERS, INC. 2 347 3700 Nationwide Coverase 800 336 6646
p \-
- 25279.0 232 l RT I' 1 MR. CRAMMOND
- Yes.
2 MR. MICHELSON: Okay. So in general you did not 3 consider the purge valve question in your studies; is that , ! 4 correct? 5 MR. CRAMMOND: That's true. 6- MR. MICHELSON: Thank you. I 7 MR. MARCHESE: Let me comment, I think I 8 containment-initiated core melts are more likely for the 9 BWRs. Later on when we talk about the Quad Cities results 10 we'll have more to say on that. l 11 MR. EBERSOLE: Okay. 12 MR. CRAMMOND: The two things are containment
-) 13 failure due to overpressure and post-accident radioactivity 14- removal. Those two functions.
15 (Slide.) ! 16 This shows the results, the dominant accident 17 sequences and the probabilities before and after recovery 18 is applied; and essentially this is the number that we are 19 looking at: 1.4 times 10 to the minus 4 for the internal 20 events alone. 21 MR. DAVIS: Excuse me, Wally, I had a question. 22 On the top sequence I think is a failure to recirculate, 23 and you don't have any adjustment for the recovery part of 24 that even though there's two to four hours, I believe, for () 25 the operator to do something. Could you explain what you I
- ACE FEDERAL REPORTERS, INC.
, i 202-347 3700 Nationwide Covernoe 80M36 6646 [ __ __- -
l 25279.0 233 RT 1 used as a recovery criteria and why that particular 2 sequence doesn't change? 3 MR. CRAMMOND: I probably couldn't remember the 4 details of that. But, on the recovery we applied a number 5 of recovery factors based on -- we looked at each of the 6 cut sets in each of the sequences and we looked at the 7 elements in the cut set, and we had recovery factors for 8 diesel generators, for batteries, for loss of off-site 9 power, for LOCA faults where it could be remedied from the 10 control room, or where you'd have to go down to the place 11 and change a valve, for example. And each of those, we had 12 four or five different times: 30 minutes, an hour, two 13 hours, four hours, eight hours; and we had a matrix of 14 numbers that you can find in the report that we used. 15 In this particular instance, I don't recall why. 16 I believe that was probably the f ailure to change over to 17 recirculation at the proper time and the probability was 18 all wrapped up in the one number. Instead of applying 19 another human factor number, would he -- if he failed, he 20 failed. And that was in the original number. We didn't 21 give him another number, another chance to fail, because it 22 was already in the initial number. Some of them did change 23 and some of them didn't. 24 MR. WARD: Does that make sense to you, Pete? 25 MR. DAVIS: Yes. ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coserage 804336r446
25279.0 234 BRT O 1 MR. WARD: Okay. 2 (Slide.) 3 MR. CRAMMOND: Now, we obtained a number of 4 excellent comments from the NRC as a result of their review 5 of the first analysis. These are not necessarily changes, 6 but in some cases, changes -- in many cases, additions to 7 the analysis to account for their comments, including 8 long-term station blackout, control circuit failures to 9 pumps and valves, and test and maintenance down time, which 10 we hadn't included, and some past PRAs hadn't included 11 either when we reviewed that. And so on. I won't read all 12 these. But all of these changes and additions, I believe, 13 improve the analysis immensely but also took a lot of time. 14 (Slide.) 15 MR. DAVIS: Excuse me again, Wally, I want to go 16 back to Dave's question on the pump seal LOCA. I don't see 17 that on this list. 18 MR. CRAMMOND: That was one of the comments that 19 we addressed and decided that we would not change anything . 20 to reflect that. 21 MR. MARCHESE: Well, wait a minute. We assumed 22 that the other generic issue on sealed LOCAs would fix that 23 problem. 24 MR. CRAMMOND: That's right. So we did not need (~ 25 to address it. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6 4 6
25279.0 235 RT 1 MR. DAVIS: You assumed there would not be a 2 pump seal LOCA associated with loss of AC power? 3 MR. CRAMMOND: It was such an important issue 4 that it would be solved by the other generic ir, sue, so we 5 assumed it would be solved if it were a prchlem, for our 6 analysis. 7 MR. DAVIS: To my knowledge it hasn't been 8 solved, and I'm not sure what the disposition is. 9 MR. WARD: I think the latest indications are 10 that it may not be'a problem. But I'm not sure the Staff 11 has accepted that yet. 12 I'm not sure that's true. Does anybody have a 13 position on that? 14 MR. MARCHESE: I think they are still 15 negotiating with Westinghouse on the sort of' testing to be 16 performed. I don't think is finally resolved. 17 MR. WARD: I thought there were some tests which, 18 foreign tests which showed -- a year or two ago there was 19 some evidence that the leakage might be as high as 250 or 20 300, but there were some French tests, analyses, which 21 showed it much lower than that. But I'm not sure that 22 applies to all the plants. I guess the Staff hasn't yet 23 accepted that? 24 MR. MARCHESE: Right. I might point out that 25 (]) one of the dedicated system options that we are'considering E i ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwde Coverage 800 336-6646
25279.0 236 RT 1 has the capability to cope with small break LOCAs, so if 2 theleaki$githinacertainrangethatsystemcancope 3 with a seal LOCA. 4 MR. WARD: In any rate, Pete, it seems to me 5 this is a reasonable approach; although it hasn't been 6 resolved it looks like they are moving in that direction -- 7 MR. EBERSOLE: When you look at the PRV 8 performance or, for that matter, the performance of any 9 electric at apparatus which is in the nonqualified 10 configuration, in the open standard inductrial 11 configuration, I noticed that NRC has adopted the somewhat 7- 12 permissive philosophy that you can immerse those devices in
~
13 steam and suffer the consequences of surface condensation ,, 14 on terminal blocks and equipment and go through the rather 15 speculative process of examining whether the leakage 16 currents can lead to short circuits and failures of the 17 electrical process in question. Which view did you taka 18 here? 19 MR. CRAMMOND: We didn't address the 20 environmental qualification of equipment in the containment 21 par se. 22 MR. EBERSOLE: How can you rationalize you can 23 even prolong PORV functions to keep them open? 24 MR. CRAMMOND: We didn't take that as an issue. (} 25 Andy, do you have any comments on that? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
25279.0 237 RT l~ MR. ERICSON: Mr. Ebersole, we.have addressed 2 this, as we have stated from the very beginning, using 3 generic data. We have not gone into plant-specific: Where 4 is the PORV, where might it get this, where might it get 5 that? 6 MR. EBERSOLE:. Do you consider that open 7 standard exposed apparatus will work in a condensing 8 environment? 9 MR. ERICSON: We have not addressed that 10 question, what condenses on it or what doesn't. 11 MR. EBERSOLE: Okay. That's a no -- f- 12 MR. C,RAMMOND: We have to see what works. (_)/ 13 MR. EBERSOLE: Sure. 14 (Slide.) 15 I will show this very briefly. This is a 16 comparison between the initial analysis, or the original 17 analysis, and the current analysis. The dominant sequences 18 change slightly and the core melt frequency increases 19 somewhat. 20 (Slide.) 21 Again, very briefly, the next two viewgraphs 22 show the dominant action sequences grouped. .a 23 (Slide.) 4 24 And, with the failure probability for each group, () 25 its contribution. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80433Mi646
25279.0 238 RT 1 .(Slide.) 2 But more important than that is the 3 vulnerabilities. And the following viewgraphs show the 4 vulnerabilities. I was the failure to switch over from 5 emergency core injection to recirculation. That shows up 6 in three accident sequences for a core melt probability of 17 3 times 10 to the minu- 5. One of the more important 8 vulnerabilities. 9 The vulnerabilities aren't numbered by 1 10 importance.- They simply got numbered. So the next one is 11 slightly lower, the station blackout due to battery failure, e : 12 and station blackout due to diesel generator failures, is
'~'
i 13 _down to the 10 to the minus 7 range. 14 MR. MICHELSON: Excuse me, before you leave that 15 slide, your circulation, for instance, where you have to 16 make a switchover, involves the operation of a number of 17 probably-motor-operated valves; in some cases they may be'
- 18. air-operated.
19 What do you use to measure the reliability of 20 .these various valves? What's your source data bank for 21 knowing how good the valves really are under the conditions 22 that-they will be experiencing? 23 MR. CRAMMOND: The data we used was data from IREP 24 and ASAP; and as far as LOCA faults were concerned, and we () 25 used the human reliability number _from some of Swain's work. I'
~
ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-3364646
- _ = - -- _ _..
,m - , . _ , . . . , , , , . , . , _ . . . , , , , , . . , _ , , _ , . _ , ... . m,,,,,_
25279.0 239 RT 1 MR. MICHELSON: The IREP data, was it available 2 for Point Beach? Was that one of the plants analyzed? 3 MR. CRAMMOND: Yes -- oh, no, no, not for Point 4 Beach specifically. But generic data. 5 MR. MICHELSON: You are using generic data from 6 some'other plants, then, for each of these? 7 MR. CRAMMOND: For the most part we used generic 8 data. In cases where generic data doesn't exist and we 9 have information from the specific plant, we use that. 10 MR. MICHELSON: When using generic data from 11 other plants, I assume you were careful to see if these r-' 5 12' were comparable valves and so forth? 13 MR. CRAMMOND: No. .We didn't go in that detail. 14' MR. MICHELSON: You mean you just take the 15 numbers? 16 MR. CRAMMOND: We took the numbers from the -- 17 MR. MICHELSON: If they were air-operated you 18 certainly didn't use them for motor-operated, did you? If 19 ~ there were air-operated valves on a particular plant you 20 didn't use motor-operated valve reliabilities did you? 21 MR. CRAMMOND: No. We did different numbers for 22 air-operated valves, motor-operated manual valves, check i 23 valves -- 24 MR. MICHELSON: But only those developed from () 25 the IREP data where it was available? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 33(Hi646 p_
25279.0 240 1 MR. CRAMMOND: Really it's ASAP data and that 2 depends on IREP data. They are all folded together. So we 3 looked at several data sources and tried to come up with 4 the best, most representative number that we could come up 5 with. 6 MR. MICHELSON: Okay. Thank you.
;7 MR. MARCHESE: I would like to point out, 8 Mr. Michelson, that in a generic program like this we feel 9 it's appropriate to use a generic data base.
10 MR. MICHELSON: I have no quarrel with it. I 11 was just trying to determine whether you had to fall back 7y 12 to even older, less useful numbers or whether these are ( '/ 13 state of the art numbers. I gather-it's state of the art 14 numbers. 15 MR. MARCHESE: I gather you'll find in some of 16 the more recent PRAs coming out, where they are doing a 17 level 3, detailed ones, they are trying to get into l L 18 plant-specific data if it exists. 19 MR. CRAMMOND: We also talked to GPU people on L 20 their numbers, realizing they are doing a BWR; and for a PWR f H21 there are some similarities, like valves, for example. We l 22 tried to use the latest data we could. l 23 MR. DAVIS: On that question, Wally, the number l 24 1 sequence is the small break LOCA in terms of core melt (J~l. 25 probability. And you used a probability of .02 per reactor
- ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800-336 4646
25279.0 241 RT l 1 year, which seems to be quite high. It's quite a bit 2 higher than what was used in the IREP studies the NRC used 3 previously. Do you know what haa happened to cause that 4 number to go up? It's also more than a factor of 10 higher 5 than WASH-1400. 6 MR. CRAMMOND: That's the same IREP number. 7 MR. DAVIS: Do you know what's that based on? 8 That would mean for 100 reactors you'd have a couple of 9 events per year. It doesn't seem to me like we are having 10 that many. 11 MR. CRAMMOND: Okay. In the IREP data that ,r~ 12 comes more from ANO, the IREP procedures guide, the .02 is 13 the total of two small LOCAs: one that goes up to 14 something less than an inch, and then one that goes from 15 that number to 0.166 inches. So we said that our S-2 LOCA 16 was going to be the sum of both of those. So really it's 17 the lower, the smal.1. , small LOCA, the very smallest one, 18 less than an inch, that contributes the most to the .02 19 number. Actually, the slightly larger LOCA, which is still 20 a small LOCA, is probably in the order of magnitude -- an 21 order of magnitude lower than that. That's probably due to 22 PORVs, the majority of that .02. But that' effectively, 23 right out of the IREP. - 24 MR. DAVIS: Thank you, t () 25 (Slide.) ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Owerage 800-336-6646
25279.0 242 4" 1 MR. CRAMMOND: Number 4 is failure of the ECC I 2- recirculation due to RHR pump cooling failure caused by a 3- ' valve in the service water cooling system. 4 Vr.lnerability 5 -- that's a fairly big one, 1 5 times 10 to the minus 5. 6 Failure of the emergency core cooling injection
.7 due to the component cooling water failure, that's a 1 E to 8 the minus 6, not a particularly large one.
9 I might point out on all these vulnerabilities, 10 I'll point out later, we don't address all of them as far 11 as modifications because some of them we don't know 12 anything to do about them. Anything we thought of didn't
)
13 do anything to prove them. 14 (Slide.) 15 Here the major one, common mode, that affected 16 'every. major sequence, 2.4 times 1 to the minus 5, and it's l'7 one we had no modifications that we could place in the 18 alternatives.- But it's nevertheless a vulnerability. g 19 (Slide.)- l-l 20 Number 7, common mode failures of safety system 21 valves.- That one you might be able to do something about. 22 The failure of the low pressure injection system in the 23 recirculation mode, and then the failure of the aux l 24 feedwater turbine-driven pump. 25 I) (Slide.) ! ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800-33M646 L
i i l 25279.0 243 GT '
~V l
1 What we found in the second time around, that we 2 had more vulnerabilities and the probability, core melt 3 probability was distributed in more vulnerabilities than it 4 was before, and the probability also went into more cut 5 sets a little deeper than it had before. So when you 6 disperse them it makes it more dif ficult to come up with 7 discrete modifications to address those vulnerabilities. 8 Number 10, failure of the component cooling 9 water pumps. 10 And'then last but not least, the long-term 11 station blackout caused by depletion of the batteries or
~T 12 .the condensate storage tank, for almost 4 times 10 to the (G
13 minus 5. That was one of the changes that we had made from 14 the original analysis, or additions. 15 All'this, so far, has been the internal analysis. 16 The internal analysis models have been used by the external 17 analysis people and it changed to fit their particular 18 analysis. In particular, the seismic people used it 19 . extensively. This is the seismic analysis. This is simply - 20 showing the frequency of the earthquakes, various levels, 21 and the- three kinds of seismic initiating events that were 22- considered, small LOCA and two kinds of transients. 23 (Slide.) 24 MR. WARD: The earthquake frequencies, what are
) 25 the numbers there? I to 2 SSE? What does that mean?
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80 4 336-6646
25279.0' 244
;- RT ~\)
l' MR. CRAMMOND: Somewhere between 1 and 2, say 2 shutdown earthquake level, 2 to 3, 3 to 4, 4 to 5. 3 Why he didn't just say 1, 2, 3, 4? I don't know. 14 It's something between a 1 and a 2. Anything below a 1 we 5 are not considering as significant, obviously. 6 MR. WARD: But if you had a curve there'd be a 7 precise number on a curve. 1.5 times SSEC? 8 MR. CRAMMOND: That would be the interval 9 between a 1 and a 2. 10 MR. WARD: All right.
~11 (Slide.)
12 MR. CRAMMOND: The seismic core melt freq'uencies 13 for those 4 levels, and the 3 transient initiates are shown 14 .here, and the sum being 6 times 10 to the minus 5. The 15 vulnerabilities that contribute to this are the refueling 16 water storage tank, and then there's three vulnerabilities 17 included here on the second line: the anchorage of 18 electrical buses, transformers, inverters, and then 19 batteries, and then the battery racks themselves and the 20 instrument air system for the PORVs. 21 MR. MICHELSON: Excuse me. As you go to the
-22 much larger level earthquakes, how do you know the response 23 of the. relays and instruments to possibility of a relay 24 chatter at these higher levels -- these are well beyond, O
1 ,) 25 perhaps, the data you have available. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33Md46
I 25279.0 245 ,c" y;RT . 1 MR. CRAMMOND: Go ahead, Dave. 2 MR. ERICSON: The question of relay chatter has 3 come up several times, Mr. Michelson. We have not looked 4 at it here. It is being examined by -- under a separate
- 5. effort I'm aware of now. In fact, the people looking at 6 that have been looking at what we were doing under the 7 south ERA -- we have not looked at relay chatter here.
8 That can be an issue. 9 MR. MICHELSON: In terms of mountings within 10 racks, how do you even know the equipment stays in place, 11 because you haven't necessarily shaken the equipment at 12- these kinds of levels. (~}
%j -
13 MR. ERICSON: That's right. However, it's our 14 understanding through our seismic people who have been 15 involved with the broader seismic issues, A-46 and related,
.16 that what we see tends to be whole cabinets moving rather 17 than components jumping out of cabinets and this sort of 18 thing.
19 MR. MICHELSON: That's to some extent probably 20 . correct, but there have been a number of exceptions on 21 occasion where we find that little holding devices don't 22 hold the relays'in the cabinets at higher levels -- 23 MR. WARD: Wally -- Dave -- on the reactor 24- coolant plant or seal LOCA, you deferred to a generic issue ( 25 to solve that. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 336-6646
25279.0 246 1 For the seismic vulnerabilities, such as, well, 2 we are talking about anchorage of cabinets and so forth, 3 there is a generie issue, I guess, a USI A-46 on that. How 4 come you haven't treated that the same way? You haven't 5 credited resolution of A-46, where you apparently have 6 credited resolution of the generic issue on the seal LOCA. 7 MR. CRAMMOND: I think because it's not clear 8 the seal LOCA would change our analysis and it wouldn't be 9 included in the LOCA that we have, whereas in the seismic 10 analysis, I believe our seismic person, Mike Bond, felt he 11' -could address seismic in a very satisfactory way. I don't em '12 know the details of his analysis, but I think.he was pretty.
.O 13 confident-that he had covered things adequately.
14 MR. ERICSON: Wha?. you are seeking, Mr. Ward, is 15 just that: that the kinds of things we have seen are being 16 identified elsewhere, you know, tied down in cabinets, 17 additional anchorage. That seems to be what's coming out, 18 as I understand, most of the seismic work is anchorage 19 problem.- So it's straightforward enough that we have 20 included it. 21 MR. WARD: You have included it as being 22 anchored and as being a problem? 23 MR. ERICSON: Well, if we felt they were based 24 on our analysis we felt it -- we sought additional tie down. () 25 MR. WARD: But under A-46 there may be a fix to ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 4 646
25279.0 247
?~;.T
- N_.]
1 all plants to eliminate problems of relay cabinets tipping 2 over. So you have wouldn't -- if you credit that, then the 3- risk you've shown here of 6 times 10 to the minus 5 may be ! '4 quite smaller. J' 5 MR. ERICSON: In one of our modifications we do 6' go in and tie thocr :%aa and show you specifically the 7 change of risk. 8 MR. MARCHESE: Modification is consistent with 9 what A-46 has recommended. 10 MR. CRAMMOND: Here it is. 11 (Slide.) f'Y J-12 This is seismic also. And it shows ths core 13 melt probabilities, given that you apply the modifications, 14 and it's down an order of magnitude. And the modifications 15 are essentially replacing the battery racks and safety 16 class nitrogen bottle and tie down the buses and inverters, 17 and alternate source coming f rom the spent fuel rather than 18 the refueling water storage tank and the appropriate timing 19 to do that. 20 So, it can be solved. And I guess you are right, 121 there is a slight difference in our approach there. This 22 one we could address if we knew what to do and we did it. 23 In the case of the pump seals, we weren't even sure whether 24 that was going to b5 an issue or not. We decided if it
) 25 were to ima important it would get solved. If it weren't ACE-FEDERAL REPORTERS, INC.
202 347-3700 . Nationwide Coverage 800-336-6646
n 25279.0 , 248 RT 1 important than we didn't need to address it anyhow. 2 MR. MICHELSON: How did you deal with the 3 nonseismically qualified equipment, wh]se failure, 4 particularly in multiplicity, could lead to further
.5 ' difficulties with shutdown during these very large 6 earthquakes that you are predicting -- trying to calculate 7 here?. How did you deal with failure modes and effects of 8 nonqualified equipment?
9 MR. ERICSON: The seismic analysis has done a 10 fairly extensive look at individual component fragilities 11 where those show up at the cut sets. _ 12- MR. MICHELSON: Well, did the interactive r I 11 3 effects of those components even enter into your fault 14 trees? A water tank on the roof will fall through the roof 15 and so forth -- you have considered all the effects of the 16 nonqualified -- 17 MR. ERICSON: I'm not going to be so bold as to 18 say "all." I know better than that. Yes. Commonly 19 located failures in a room and that sort of thing. 20 MR. MICHELSON: Okay. Now, in dealing with that 21 question, then, did you consider the possibility that more 22 _than one thing will happen at a time during the earthquake 23 to nonqualified equipment? In other words, not just one 24 tank will rupture, but perhaps the tank will rupture and 25 several other pipes will' break and so forth in the process, {} ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6 & 6
25279.0 249 1 or did you deal only with one at a time? 2 MR. ERICSON: I don't know the detailed answer . +
'3 to that.
4 MR. MICHELSON: That's generally where you get. 5 into the difficulty. Dealing with one at a time is usually 6 not too bad to handle, but dealing with several of these at 7 once, it can be difficult. 8 MR. ERICSON: This is Steve Hatch from our Staff. 9 MR. HATCH: Components were linked, fragilities 10 of components were linked in the seismic analysis. If two 11 components would see the same acceleration and had the same r3 12 fragility, their the failures would be coupled in the V > 13 analysis. 14 MR. MICHELSON: Perhaps you answered the 15 question. Maybe I didn't understand it. In essence you 16 are saying-that you looked at the possibility of failurea 17 in various parts of the building occurring concurrently? 18 MR. HATCH: Yes. 19 MR. MICHELSON: Okay. The other question you. 20 get into is some of these failures might involve equipment, 21 oil lines and so forth, which then lead to fires as a 22 result of these failures of nonqualified equipment. Have
.23 you dealt with the fire consequences in this analysis?
24 MR. HATCH: I'd say in general we did not look () 25 at seismically induced fires or seismically induced floods ACE FEDERAL REPORTERS, INC. ' 202-347-3700 Nationwide Coverage 800 33(H5646
/25279'0
. . 250 l 1 for the most' part.
l 2- MR. MICHELSON: -Wait a minute, I thought you
- 3. just told me you dealt with tanks ~ falling and so forth. I 4 -hope.you dealt with the contents of the' tanks as well?- You 5 ;know the: flooding consequences of these failures. is perhaps 6 the more troublesome one in many cases, particularly when 7 water'gets around into electrical equipment that you are
'8 ~ having to depend upon. I assum'e that you chased the water
.9 and not just chased the pipe? If you haven't, you haven't 10~ really dealt.with'the issue at all, I don't believe. If
.11 you'are just considering it a missile only.
12 MR. HATCH: The only case,where I know that came
-13 up was the Cooper analysis; we did have a damn upstream of-t i
14 the plant failing due to a seismic event, but I think that
-- 15 ' was the only case where we looked at --
~16 MR. MICHELSON: Within the plant? If, for some '
17 strange reason h'ad you a nonqualified water pipe running 18 through the control-room and ruptured and dumped the water
'19 onto the bench board, you certainly would want to deal:with j 20- the consequence.
L 21' MR. CRAMMOND: We looked at that rupture in the
- 22 flood whether caused by seismic or not.
-23 MR. MICHELSON: The problem is in the past we
} 24 have dealt with these questions one at a time. A long time 25 ago it was asked to be dealt with but only one event at a ACE-FED ERAL REPORTERS, INC. 202-347-3700 Natioawide Coverage 800 336 6646
- A u ,, -, _ ,_ .. _ _ . _ _ _ . _ . .- _ - _ _ . _ - _ _ _ _ _ _ _ ,_
25279.0 251 ( lt
.AJ l time, and now in the seismic case, several of these 2 nonqualified pieces of equipment are going to be challenged 3 concurrently. I was kind of wondering what kind of 4 assumptions you make concerning the numbers of such 5 failures you have to deal with and how you handle this in 6 your analysis?
7 MR. CRAMMOND: Let me ask Gary Sanders, who is 4 8 also working on this project doing the Turkey Point 9 analysis, if he can add anything to that. 10 MR. SANDERS: Gary Sanders. Internal analysis 11- goes through -- internal flood analysis goes through and 12 looks at each specific water tank, where it is, where the ('v) 13 tanks run,-what components are vulnerable, what cabinets 14 are vulnerable and identifies vulnerabilities throughout 15 the plant. Some points, liL' Turkey Point, everything sits 16 outside. If you have a seismic-induced flood, it doesn't 17 go anywhere except back into the Atlantic Ocean, so they 18 can couple them any way they want and not get -- 19 MR. MICHELSON: Okay. This is the standard 20 approach to dealing with pipe breakouts the containment. 21 You looked at these individually. My question wasn't that, 22 but rather: In the case of a seismic event, now, how many 23 of these breaks do you assume are going on concurrently, 24 and that you do the analysis of the consequence on? When (O
./ 25 several of these breaks happen at once instead of just one.
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
I l
.25279.0 252 BRT (Il .
1 MR. SANDERS: From the internal flood analysis 2 they know which pipes are going to affect which cabinets, 3 components. In the seismic analysis, as Steve Hatch 4 indicated, they have fragilities on several pipes and they 5 looked at common mode failure of several pipes at one time. 6 I don't know what the number of pipes are concurrently, but 7 I know they do look at concurrent failures of pipes. 8 MR. MICHELSON: I guess I can find that 9 somewhere in the description. I wasn't able to find it, 10 but I assume you can find me some material to read on how 11 you deal with the assumptions concerning the number of 12 concurrent failures during an earthquake. If you just tell 7-)s 13 me where to read it, I'll be in good shape. 14 MR. MARCHESE: There is an unresolved safety 15 . issue, A-17, systems interaction, that we have been 16 struggling for a couple of years with this very question 17 you are asking. I think the conclusion they are coming out 18 with is it's pretty difficult, based on the analytical 19 tools that are available, to handle these kind of things. 20 MR. MICHELSON: I agree with you completely, but s 21 I thought I was getting answers like you were handling -- 22 I'd like to read about it. And if you aren't handling it, 23 ~just tell me so we know what the assumptions are in the 24 study. 25 MR. MARCHESE: We are approximately handling it, (O"l' ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6686
s-
~25279.0 253 f
1 but I would point out that one of the options that we're 2 handling that. would handle these kinds of things is the 3 dedicated system that's complete self -- 4 MR. MICHELSON: I'm just trying to determine 5' what the scope of your study is and how you handle this 6 particular issue, that's all. 7 MR. CRAMMOND: I think the answer to this is 8 'that I would like to think that we have but I'm not sure. 9 I don't want to say that we have because I'm not.sure. But 10 we had a rather long seismic analysis appendix, and I just 11 don't remember. 12- Gary? Can you add to that? 13 MR. SANDERS: On the internal flood analysis the 14 plants we looked at so-far, we haven't found more than two 15 or three places in a plant where there is large enough 16 water sourcos to damage the equipment. It usually comes 17 :down to two or three specific places. So in the seismic 18 . analysis I know they are looking at two and three pipe 19 breaks simultaneously. Beyond that, if they have 4100 pipe 20 breaks, the other 97 don't seem to be affecting safety 21 grade equipment. 22 MR. MICHELSON: I would strongly urge you read 23 your LERs more currently. There's numerous LERs of late 24 where they are broken, just leaks, and they get into the ( )- 25 electrical equipment and strange things happen, relief ACE-FEDERAL REPORTERS, INC. 202-347-3700 . Nationwide Coverage 2 336 6646
25279.0 254 RT 1 valves open, a number of things. 2 On a probability basis this ought not to be 3 happening very often if we know that most all breaks will 4 not cause of any of these consequences. So I'm really 5 wondering how good we understand the effects of water 6 getting into our systems. 7 MR. WARD: Carl, I think at the onset of this 8 program in A-45 I think there was agreement that the 9 analysts were going to use the existing technology for ERA 10 and FMMEA, and accident analysis, source terms, but they 11 weren't going to be able to or be responsible for advancing 12 the state of the art in each of those areas. That's just 13 too much. We'd never get on with the issue of A-45. 14 MR. MICHELSON: I don't disagree with that 15 provided they make it clear that this is what they are 16 doing. When you come to your bottom line you should always 17 indicate what some of the things are you have left out. 18 You can't quantify them, obviously, because you don't know 19 how to yet, but it's very important to indicate what your 20 assumptions are on earthquake effects when you start 21 talking about 4 or 5 SSE. 22 It's crazy to talk about such big earthquakes 23 without recognizing that it's only a very limited study of 24 those effects they have done here; mainly it's kind of a -m
) 25 structural study, apparently.
ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
25279.0 255 1 MR. MARCHESE: I think that's a good point. 2 We'll try to do a better job on qualifying what the 3 analysis includes and what it doesn't and specifically 4 giving -- what we are giving in the nature of assumptions. 5 MR. CRAMMOND: I'd be surprised if it's not 6 discussed in there, but I don't remember. 7 (Slide.) 8 In the fire analysis there were two rooms that 9 were vulnerable: one the aux pump feed room and the other 10 the 415 switch room. Those reduce the probability from the 11 10 to the minus 5 range to the 10 to the minus 7 range, 12 those modifications. 13 MR. MICHELSON: In looking at vulnerability to 14 fire, did you look at vulnerability to inadvertent 15 actuation of the fire protection features that might cause 16 a loss of the functionality? 17 MR. CRAMMOND: I don't remember that. 18 MR. MICHELSON: Because, again we have seen some 19 inadvertent actuations in the last year or so which lead to 20 equipment not functioning the way you might have thought. 21 MR. CRAMMOND: I believe we did consider that in 22 the internal flooding; in particular, in the pump house. 23 There's a fire main that runs across the top of the pump 24 house and a rupture in that line spraying on the service ( )) 25 water pumps could fail them, and that's the next ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
'25279.0- 256 RT.
1 ' vulnerability. 2 (Slide.) 3' MR. MICNELSON: Fire protection itself in-
?
4~ inadvertently actuated in any area of the plant where 5 deluge systems are used, which I find are a little more
- 6. prevalent than I thought: When a deluge system is' set off
-7 equipment becomes involved. The real danger is'during an 8 earthquake or extensive fire, deluge systems will be set 9 off in an area where there are no fires. If.this were to 10$ happen, and it's not qualified not to happen, you've got to 11 ' deal with a far more extensive loss of equipment than you- >l .
12 -might have first thought, and therefore these probabilities 13- of core melts and so forth, I would think, might change. 14 MR. CRAMMOND: I. believe that was covered in the
- 15 internal flood analysis.
16' MR.-MICHELSON: I would:like to read about it 17- because it's anJimportant subject. If you just, again,.get 18 me the reference where you cover the inadvertent actuation 19 of fire protection and how you deal with that on a 20 probabilistic' basis, that would be very helpful. 7
- 21 MR. MARCHESE: In terms of a quantitative-22' treatment, if it's in the data base we have attempted to 23- treat-it quantitatively. However, I think you are bringing 24' up a number of events that either the data doesn't exist or
;[) 25 the analytical tools are not available. In this case, we ACE-FEDERAL REPORTERS, INC.
- ~
202-347-3700 Nationwide Coverase 2 336 4646 _~
L 25279.0- 257 L 1 are going to have to fall back on a deterministic treatment 2 of these things. 3 MR. MICHELSON: Sure. But that's what I'd like 4 to read. The paragraph that explains: Here are the things 5 we can't deal with and it could have a significant effect 6 on bottom line. Without that people read: Gee, fire is no 7 problem, flooding is no problem, because they only see a 8 piece of the problem. 9 MR. MARCHESE: I think we have to do a better 10 job on giving you the limitations on the quantitative 11 treatment of these things. N 12 MR. CRAMMOND: I'm not sure that we haven't - (V 13 discussed some of that in fair depth in chapter 3, and 14 certainly in the appendices. But it's been some time since 15 I' read them, and I don't remember, but I feel quite sure 16 that they have looksd at breaks. Every room where there's 17 a possibility of something failing they looked at the pipes 18 in that room and looked at a break or inadvertent operation 19 of something or test and maintenance problem, a valve open 20 or something of that sort, that would lead to internal 21 . flooding'or spray. But not necessarily two rooms at one 22 time. 23 MR. MARCHESE: Okay. Let's leave it. We'll get 24 back to you on that. O) (_ 25 (Slide.) ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
25279.0 258 1 MR. CRAMMOND: This is the viewgraph showing the 2 internal flood analysis where, effectively, it was the 3 firewater-header in the service water pump room that was 4 the culprit and, by increasing a shield barrier, it was 5 made to be negligible. 6 (Slide.)
-7 Now, when you take all the internal 8 vulnerabilities and the external, or special emergency 9 vulnerabilities, it turns out that we had 31 identified 10 vulnerabilities distributed as shown here, and 16 of these 11' were considered in the impact analysis that -- that wasn't g-)- 12- necessarily the top 16, but it was top 16 based on those V
13 that we could do something about. Those that we couldn't 14 or didn't know what to do about, didn't get carried into 15 the impact analysis. 16 MR. MICHELSON: Before you leave that slide, 17 could you tell me to what extent you looked at the 18 vulnerability, perhaps, due to loss of room cooling, and 19 particularly in sensitive rooms that might contain solid 20 state control circuitry and how you dealt with that on a 21 probabilistic basis, because that, again, has been showing 22 up in LERs on a couple of occasions where they lost room 23 cooling and, sure enough, funny things do indeed happen. 24 MR. CRAMMOND: As a general principle on all our () 25 analyses we model room cooling if it's deemed to fail ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
25279.0 259 1 anything in~ the room.
'2 MR. MICHELSON: How do you do that, because you 3 have lots of possibility now on solid state control panels 4 as.to failure modes and effects. In fact the experts can't
.5.' tell me, they can't even predict which way circuits are 6 going to flop.
7 MR. CRAMMOND: There are only certain rooms that 8 have cooling. If it's in the open area -- 9 MR. MICHELSON: In dealing with the rooms that 10 have this potential question, how did you incorporate that 11 in your analysis? Or did you not incorporate it? That's a l' 2 - good answer. I just would likp to know, do you deal with 13 that question?
~14 MR. CRAMMOND: If, for example, it's the control 15 circuits to a pump and those control circuits'are in the 16 pump room, for example, and that pump room requires cooling, 17 then in the case that says " pump fails" underneath it says 18 "LOCA faults to the pump, test maintenance" and so forth, 19 .and room cooling. Then you develop room cooling underneath 20 that.
21 MR. MICHELSON: In a solid state control cabinet 22 room -- this is a room which contains half of all plant 23 control systems in some cases, and it has a very. critical 24 ventilation requirement, then. If that were to fail
;() 25 .there's not just pumps involved there, half of everything i
ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800-336 6646 g.
25279.0 260 RT 1 in'the plant is potentially involved. 2 Did you deal'Vith that particular question? 3 MR. ERICS'dN: As an initiating event? No. No. 4 MR. MICHELSON: , Loss of room ventilation, and, 5 if not, is this a, potential' concern? You must have had 6 some basis for not ' dealing ^with the loss of ventilation. 7 MR. CRAMMOND: In* conjunction, assuming that 8 there is a room like that in this plant -- 9 MR. MICHELSON: And there may not be in Point 10 Beach. I don't know.
.11 MR. CRAMMOND: I believe that's true. But it f- y 12 was considered for -- in the model of the systems that it i) 13 applies to we didn't consider it as an initiating event, 14 just as we didn't consider, for example, a service water 15 failure as an initiating event. There's this limitation on 16 how many things we could try. That's a good one.
17 (Slide.) 18 I won't go through all this, but this is a list 19 of the modifications that were proposed for the internal 20 analysis. 21 MR. ERICSON: They need to skip a page or two 22 here or there. 23 MR. CRAMMOND: Yes. You have to skip a page or 24 two over and then come back, but all we are doing here is (") V 25 depicting the fact that we had seven vulnerabilities that ACE-FEDERAL REPORTE4, INC. 202-347-3700 Nationwide Coverage 800 336-6646
25279.0 261 1 we proposed modifications to. Then when we get to the 2- : alternatives, the alternatives will be different 3~ combinations of these modifications in conjunction with 4 external event modifications. I 5 (Slide.) 6 In fact, here is the constituents of the four 7 alternatives. The first alternative was to do all of these, 1 8 including-the dedicated diesel generator startup battery, 9 RHR pump cooler valve improvements, and backup air supply 10 for PORVs -- that's a seismic one -- and so on. 11 And then alternative 2 was all the above plus an
'(>-
%-]
12 . additional condensate storage tank. 13 Alternative 3 is all the above from 2,and 1 plus 14 a -- the diesel generator exhaust, stack -- that's a wind 15 vulnerability, and independent diesel-driven aux feedwater 16 pump and so on. 17' 3/ And then, last, alternative 4 is the add-on 18- chutdown decay heat removal system, all by itself. Nothing 19 elc_e. 20- (Slide.) 21 Now we get to the chart that you have probably-2'2 alresdy come to. This has always caused some difficulty in
. 23 int'erpretation, but essentially what we are saying here is:
24 'iThese aEe the vulnerabilities that we are addressing out of () 25 the total, some of which we can't address.
+ <
^
ACE-FEDERAL REPORTERS, INC.
, 202-347 3*00 Nationwide Coverage 800 336 6646
" * +_____. _ _ _ _ _ _ _ - - - _ _ , _ _ _ _ -- _ _ _ _ , _ - , _ - - . _ _
25279.0 262 RT 1 These are the four alternatives. And 2 alternative 4 is just an add-on shutdown decay heat removal 3 system. That's not a vulnerability. 4 Alternative 1 says if you modify internal 1, 5 which is the failure te switch over to recirculation, and 6 internal 2, 4, and so on -- if you make all those 7 modifications, that combination is alternative 1. 8 Alternative 2, then, is effectively, all the 9 same ones we had before, plus the station blackout 10 modification. 11 So, each of these columns represents the s 12 modifications to the vulnerabilities identified on the left. () 13 And that constitutes the alternative, or a solution to the 14 potential plant problems -- whatever you might determine it. 15 MR. WARD: That's just another way of stating 16 what you had on the previous page. 17 MR. CRAMMOND: That's right. It's a chart that 18 caused some difficulty -- 19 MR. WARD: It does because your alternative 20 doesn't fix anything, is the way it looks. I know it does 21 so that has to to my next question. If you won't put Xs 22 under alternative 4, will you tell me which one of those 23 vulnerabilities you believe the add-on shutdown heat 24 removal system -- 25 MR. CRAMMOND: If you look at it, shutdown ( }) ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
25279.0 A 263 1 capability does something for everything at one , time: wind, 2 seismic, internal flood, everything -- across the board.
~
3 When you get your bottom line number, the add-on 4 probabil,ithjmultipliestimes it. And it's essentially like 5 a piin't 1, approximately. So you improve everything across 6 the board, and that's one of the beauties of the add-on for l ' y 7 tnose people that are add-on advocates. It applies to all 8 the vulnerabilities at once, whereas in this alternative 9 you bnly fix the ones where we have the X, because that's 10 the one where you applied resources, money, to shore up the 11 battery racks or put another pump in er whatever. !b 12 MR. EBERSOLE: Patchwork. O l'3 MR. CRAMMOND: That's one: term. Yes.: 14 MR. ERICSON: You are correct, we could have Xs 15 the whole way down the column, just like we do for 3, and 16 then a big plus sign and say "others." 17 MR. CRAMMOND: This means a specific fix, like 18 adding another pump. :This means another building with its 19 internal systems, totally separate from these, totally 20 - different. 21 ' MR. WARD: One big patch. 22 MR. EBERSOLE: The model is a little barn which 23 is nothing but a bunch of patches that hold it together. 24 MR. CRAMMOND: That's certainly one () 25 interpretation, and when we get to the value impact ACE-FEDERAL REPORTERS, INC.
~. . - - _ _ - - _ - _ _ _ - _ - - _ _ _ -, . - - . - . . . .
25279.0 264 BRT l 1 analysis we are going to look and see just what you get for 2 alternatives 1, 2, 3 versus 4 and see if it's beneficial. 3 It may turn out that some of those are beneficial but they 4 still don't do the whole job but they are worthwhile. 5 MR. EBERSOLE: As you get deeper and deeper into 6 the interdependencies and complexities of them, doesn't it 7 occur that sooner or later you are dealing with an infinite 8 source of problems and sooner or later you have to walk 9 away from this crud, so to speak, and step off into a new 10 environment where things will work. 11 MR. CRAMMOND: You can think of it as a pyramid.
~s 12 Can you address the top one, then top two or three or four,
~'
13 and as the pyramid gets wider and wider, you find you get 14 to the point to make one more incremental improvement you 15 have to fix a whole bunch of things, and then it becomes 16 awkward to do anything. 17 MR. EBERSOLE: And then you pick up with the 18 notion, as we do on an airplane, that is a parachute is 19 what you need, or a lifeboat on a liner, rather than trying 20 to find how many more dozen ways it could f ail. 21 MR. CRAMMOND: Our charter was to look and see 22 what we could do on an individual modification basis and 23 compare that to the add-on. I wouldn't want to defend 24 either one. We are just trying to show you what you get.
/' 25 (Slide.)
ACE-FEDERAL FJPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646 Y.-~.,._.._...~.,--..,..- - -. .
.-.m-____. . . . - . -
_.,.._.-c-..- _
25279.0 265 RT 1 So, if you look at core melt probability before 2 the modifications -- this includes all the external and 3 internal -- 3.6 times 10 to the minus 4, then you look at 4 the various alternatives. The first alternative, 1.4 times 5 10 to the minus 4 all the way to 3.6 times 10 to the minus 6 5, which reflects an order of magnitude improvement for the 7 shutdown decay heat removal system add-on. 8 Now, that's the output from the analytical 9 analysis, the value analysis. 10 The impact analysis, we get cost of equipment 11 and installation, and replacement power costs, annual
,~s 12 operation and maintenance, inspection cost, all in 1985 13 dollars, and the radiation exposure involved in the 14 installation in person-REM. These are going to be the 15 impact measures.
16 Again, here the core melt probabilities -- 3.5 17 times "10 to the minus 4, broken down by internal and the 18 various external -- 19 (Slide.) 20 -- vulnerabilities. This breaks down the 21 modification, showing you what the it was before and after. 22 I say " approximate" because I didn't go through and break 23 out all the cut sets and identify exactly what the 24 improver.ent was. This was an estimate. It's a good () 25 estimate, however. And there's other things that aren't in ace-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646 k ._ . - -- . _ .
._. _ _ _ - s - ,%
25279.0 266 RT 1 here. So if you add those columns up you wouldn't get the 2 number that's here. 3 MR. EBERSOLE: That does not include the 4 environmental dependencies, as you said earlier? You said 5 you did not accommodate such things as environmental 6 capabilities of PORVs or -- 7 MR. CRAMMOND: Or qualifications of components? 8 That's right. 9 MR. EBERSOLE: Since those can affect the 10 multiple systems, they are -- tend to be compounded? 11 MR. CRAMMOND: That's right. And we get in the x 12 rut of calling this our core melt frequency. We have to 13 always remember that interfacing systems, large LOCAs, ATWS, 14 things that were specifically included in our scope of work 15 aren't in there, so this is not the total core melt problem 16 event, it's probably something on the order of 40 percent 17 to 60 percent. I don't know. 18 MR. EBERSOLE: It would be well to put an 19 asterisk on that and parenthetic addition in the bottom of 20 the page wherever you use it, to keep it from being read 21 out of context. 22 MR. CRAMMOND: That's a good idea. 23 (Slide.) 24 The impact analysis, I won't spend any time on 25 this. This simply tells you that what they did was, ({ ') ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646 Y , _ _ _ _ _ . ._..
25279.0 267 RT
~1 initial design report, then they show up at the plant 2 prepared with how they plan to do things and then try to be 3 congenial with the plant people because we really aren't 4
saying we are going to rip things out and change things. 5 We are saying we are just doing a study. 6 We have a site inspection plan. We go out and 7 do that the at the plant visit, we pick up local costs, 8 look at congestion in work areas, things of that sort, and 9 then they come back and come up with a final modification 10 -plan and get numbers like these. 11 (Slide.) 12 Now, for the four z.lternatives, these are the U<s . 13 engineering and installation costs, operations an'd
- 14. maintenance, and the installation dose,-occupational dose.
15 As you can see, the add-on.has the greatest 16 . occupational dose becaule they have to get into the 17 containment to do it. If we look at the. summary of the 18 core melt probabilities for the alternatives, starting with 19 the base case -- 20 MR. WARD: Wally, I guess you don't include some 21 ongoing, annual occupational dose due to. additional 22 . equipment inside the -- 23 MR. CRAMMOND: Yes. It's looked at. That's 24~ right. -In fact, the replacement power was also zero 25
-( ) because in this particular analysis -- it wouldn't hold ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336-6646 W- - -- - -
25279.0 263 RT 1 true under other plants necessarily -- they determined it 2 could all be done on routine outages. In this case we are 3 showing the delta core melt probability between the base 4 case in alternative 1, base case in 2, 3, and 4. And this
'S "I" was the improvement factor in core melt probability 6 from the base case. These numbers will be used in the 7 value' impact analysis, as all of these will be used, as EL positive impacts and negative values.
9 (Slide.)
-10 This is simply showing the CRAC output, low 11 bound, base case, and upper bound, for each of the 12 alternatives. If you take the difference between Q.
G~ .13 alternative 1 in the base case you'll get the averted 14 off-site dose for alternative 1, and so on. Lower bound, 15 central, and. upper bound. Those are --
. 16 MR. ERICSON: Wally, you might point out that 17 -these three source terms, again, are arbitrary for terms of
'18 our study. Because there's an ongoing discussion about 19 what.are the source terms, we simply said let's take the 20 reactor safety study and two variations thereof, 21 arbitrarily, and display them. We are not arguing that one 22 is better than the other. There they are.
23 Then, the averted, that's the product of the 24 remaining life, times the annual, times the core melt and 25 everything cranked into it.
"(O~N l.
ACE-FEDERAL REPORTERS, INC. 202-34'1-3700 Nationwide Coverage 80 4 336-6646
. - - . . , ~ . , .. . , - , , . . . , , - . - - - - . . . . - . . . . . .. . . . . . -
25279.0 269 1 MR. CRAMMOND: Not yet. 2 MR. ERICSON: That's not the annual, is it? 3' MR. CRAMMOND: That's not the annual. That's 4 just for the accident, and then it's all cranked in. 5 MR. WARD: For the life of the plant? Isn't 6' that what those numbers are? 7 MR. CRAMMOND: No. I think this is per accident. 8 MR. MARCHESE: Per reactor year. 9 MR. CRAMMOND: You multiply it, per accident, 10 and take into consideration the remaining plant life and 11 then that comes out in the impact analysis. This is just eg_ 12- the' raw data that goes into-it, just like impact analysis ,
. ( _/
13 raw data and core melt probability raw data. We have set 14 the stage with all the raw data that comes into the 15 analysis and then I will jump over all the details, rather 16 than get bogged down in all the formulation of how we 17- ~ arrive at the value impact analysis. This is written up in 18 great detail in the report. 19 (Slide.) 20 I think it's in chapter 8, and also in appendix L. 21 MR. WARD: Wally, I'm sorry, could you go back 22 to the previous one, please? 23 ' MR. CRAMMOND: Sure. 24 (Slide.) 25 MR. WARD: If I look at the bottom section there, l
- [s.s) l ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336 4646
- ., - . - . - . -, ~ . . . - -
25279.0 270-l 1 alternative 1, 100.8. That's the averted off-site dose in 2 person-rem per what? 3 MR. CRAMMOND: Per accident.. 4 MR. WARD: For what accident? 5 MR. CRAMMOND: Dave? Enlighten me. 6 MR. ERICSON: The expected value of the 7 population dose, the core. melt probability is cranked into 8 that; right? 9 MR. WARD: But it has to be per accident or per 10 reactor year or lifetime or --
'll MR. ERICSON: Expected values per year.
12 MR. CRAMMOND: I did the cardinal sin here, I 13 didn't put the units on. 14 MR. WARD: It looks looks reasonable for an 15 accidental probable annual averted off-site dose for all 16 accidents or something like that. It looks like a 17 reasonsble number to use for that, but it has got to have 18 some other definition than what you've shown here. 19 MR..CRAMMOND: Well, perhaps we shouldn't take 20 the time here. Without 1 coking further, I think it's the 21 averted dose per incident. 22 MR. WARD: For what kind of accident, though? l 23 MR. CRAMMOND: The core melt. And then you have 24 to multiply that by the probability of the core melt and I <~ (,s) 25 then you look at the remaining life of the plant to come up ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
25279.0 271 1 with the actual measure of dose averted and then come up 2 with a value or 'mpact, whether it's a negative value or 3 positive impact or whatever. 4 I could have viewgraphs showing all the 5 equations, but we didn't think that that would be of 6 interest. 7 MR. MINNERS: Wally, if you multiply that times 8 10 to the minus 5 and 100 reactor years -- nowhere. 9 MR. WARD: That's a pretty small number. 10 MR. MINNERS: That's 1/10th of a man-rem. 11 MR. HATCH: It has the core melt probability in s 12 there already. That's the averted dose per reactor year.
)
13 It has the core melt probability already cranked in, and 14 you have to multiply it by the remaining lifetime to get 15 the lifetime dose. 16 MR. CRAMMOND: It has one of the factors in, not 17 the other. I couldn't remember that. You get so bogged 18 down in doing the arithmetic, sometimes you don't remember 19 the details. 20 Thank you, Steve. 21 MR. CATTON: In this report where do I find a 22 description of how you arrived at your costs for various 23 alternatives? 24 MR. CRAMMOND: I think it's appendix J. ( ]'1 25 MR. CATTON: In the impact analysis? ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336-6646
25279.0 272 RT 1 MR. WARD: Yes. 2 MR. CRAMMOND: It's also described or summarized 3 in chapter 7, I think. 4 MR. ERICSON: 6. 5 MR. CRAMMOND: 6. 6- MR. DAVIS: Wally, when you say .1 of RSS , you 7 don't apply the .11 to the noble gases, do you? 8 , MR. CRAMMOND: No. 9 MR. DAVIS: So it's not really 10 percent of the 10 their entire release spectrum of radionuclides? 11 MR. CRAMMOND: Steve, do you want to say rx 12 something ,about that? () 13 MR. HATCH: We didn't change ^the noble gas 14 release, but all the other isotopes were manipulated one 15 way or the other and the central estimate. The upper bound 16 is essentially WASH-1400, and then .3 and .1 of the 17 isotopes of other than noble gas. 18 MR. DAVIS: There was also one issue you said 19 made a big difference, that was the discussion with the 20 Staff about the CRAC analysis. 21 MR. ERICSON: That's coming. 22 (Recess.) 23 MR. WARD: Mr. Crammond, if you will continue, 24 please? {) 25 MR. CRAMMOND: I was just about at the punch ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
x 25279.0 273 RT 1 line. _This-is, for. the alterr.atives, this chart shows the
-2 change in core melt probability which we showed you before 3 an averted off-site dose and on down to the value impact 4 ratio and net benefit in dollars per man-rem, based on 5 off-site cost only and off-site and on-site costs.
6 In particular, if you look at the dollars per 7 -man-rem, you find if it's_just based on off-site costs, it 8- varies from S2- to $13,000; and in the case of based on 9 off-site and on-site costs, S833 up through S12,000. So it 10 becomes more beneficial, essentially. A 11 (Slide.)
~ '12 Obviously, add-on decay heat removal system is
- V 13 much more expensive in terms of dollars per man-rem than 14- are the internal, or the other alternatives. However, the 15 improvement in core melt probability may not be what you
-16 want with the other alternatives. This chart simply gives 17 you the comparison between the alternatives and what you 18 . buy with each of the alternatives.
19 All the equations from which these were 20 developed are in the report, with the appropriate plant 21 lifetime, and we have positive and negative values and 22 positive and negative impacts, depending on how the value 23 impact handbook prescribed. , 24 We can come back to this, but I want to show you j } 25 a couple of other things first. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
~_ . -- -
'25279.0 274 1 (Slide.) 2 Before I show you what we did -- I have to put 3 the microphone on here -- before I show you what the 4 results change to when you don't consider interdiction and'
-5 contamination in the CRAC model, I want to show you the 6 sensitivity analysis that we did, very briefly.
- 7. This is just those sequences that apply to the 8 particular thing we are talking about. There's other 9, aequences and you add those sequences to this and you get 9
10 ' the total core melt probability in the context of what we 11 ' studied without ATWS and large LOCAs, for example, but just eS_ 12' the feed and bleed sequences were 3 times 10 to the minus 5 kJ 13 with feed and bleed. That was our base case. If you take 14 the feed and bleed out those sequences, increase by 2 times 15 10 to the minus 4. 16 If we look at the secondary blowdown we find -- 17 this is with secondary blowdown, which was our base case, 18 and without secondary blowdown, it increases 5 times 10 to 19 the minus 5. It's not nearly as significant as the feed 20 and bleed. 21 And then, if you look -- this, incidentally, 22 although I haven't said it here, was with recovery. I 23 didn't apply recovery to this one because it didn't look as 24 though it was going to make a big difference. 25 (f If we look at the feed and bleed with the PORVs ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
25279.0 275 RTl 1 blocked, that was our base case. If you say what happens 2 if they were unblocked, if that's the way the plant were 3 running when the incident occurred, then you would have an 4 improvement of l times 10 to the minus 5. 5 MR. WARD: I don't understand that. With the 6- PORVs blocked, in that case base -- that number is 7 different from the other. Tell me what you mean by that. 8 MR. CRAMMOND: That's because I only considereS 9 those cut sets, or sequences where this made a difference, 10 just the sequences where it made a difference. So this is 11 a different set of sequences than this and a different set g-) 1:2 than this. 'So those are not the same sequences. 13 I didn't go back and do the total core melt 14 probability. I just looked at those sequences where it 15 made a difference. 16 MR. WARD: Okay. I understand that. When you 17 say "PORV blocked," that means just opened and left open? 18 MR.-CRAMMOND: No. The block valves are closed.
-19 They run with the block valves closed.
20 MR. WARD: Okay. With the normally closed block 21 valves. 22 MR. CRAMMOND: Normally blocked, so that the PORVs 23 are essentially negeted for relieving pressure in the short 24 time situation. () 25 MR. EBERSOLE: Could you comment on the -- ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-33H646 m .. ., . . - _ _
- 20279.0 276 BRT fy V
1 MR. WARD: But have you credited -- let's see -- 2- where do you'-- the existence of a PORV brings with it a 3 certain amount of bad news and potential for a LOCA due to 4 PORV leakage. Is that factored in here in any way? 5 MR. CRAMMOND: No. Just the effect on feed and 6 bleed. What improvement would you have on feed and bleed 7 with the PORVs unblocked? Because, if you have to open 8 them, then there's a likelihood that they won't open. 9 MR. WARD: Right. 10 MR. CRAMMOND: The blocked valves, that is. 11 MR. EBERSOLE: Could you explain the merits of 12 secondary blowdown in the context of pressurizing and 7 _
']
13 cooling the primary as well as enabling water from a 14 variety of sources being put into the secondary? 15 MR. CRAMMOND: Where we use that -- 16 (Slide.) 17 -- is in this branch here. We are saying that 18 given that aux feedwater system is operable, if your high 19 pressure injection system fails, then, in order to inject, 20 if you can depressurize -- and that's the secondary 21 blowdown; event X we call it -- if you are able to 22 depressurize by opening up the PORVs, or anything else, 23 using the aux sprays in the pressurizer, then you can get 24 to the p'ressure where you can use the low pressure system. 25 So that becomes a backup system.
-{ }
ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-6646
25279.0 277 1 If you can do that, then you have that 2 possibility and that's -- 3 MR. EBERSOLE: You said "using the aux sprays in 4 the pressurizer." What if you don't have them; because 5 they are derived from the main coolant pump output, aren't 6 they? 7 MR. CRAMMOND: No. The aux sprays come from the 8 CCV CS. If you don't have them then it's cranked into the 9 model for X. We model both the main and aux sprays and CVCS, 10 which contributes to the aux sprays, and the PORVs.- It's a 11 -combination; the criteria for success depends on so much of
- 12 this and that.
~'
13 MR. EBERSOLE: We learned from the Palo Verde 14 discussions that if you don't have the sprays, it's -- 15 depressurization of.the secondary doesn't produce a 16 substantial depressurization of the primary for quite a 17 long time. 18 MR. ERICSON: That's correct. That's true. And 19 the Los Alamos studies on the other PWRs bear that out. A 20 long, slow process. 21 MR. CRAMMOND: This is depressurization of the 22 primary we are talking about. 23 MR. EBERSOLC: What it means is that you would 24 continue to have leakage over that long interval, if that (') 25 was the problem? ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide coverase soo.33rss646
25279.0 278 1 MR. ERICSON: Yes. If you can't depressurize 2 the primary with your sprays or with PORV, that's right. 3 You are still going to have cross leakage. 4 MR. CRAMMOND: You lead to core melt, then. 5 There's no alternative. If you don't have X, or it fails, 6 you go to core melt. 7 MR. EBERSOLE: What is the time factor you 8 looked at here? Given the fact you don't lose 9 pressurization in the primary no matter how much you bleed 10 the secondary, except for, you know, after 12, 15 hours -- 11 MR. CRAMMOND: We are bleeding the primary. cs 12 MR. EBERSOLE: No. If you are bleeding the 13 secondary to bring it to low temp and pressure, it doesn't 14 reduce the primary for many hours so it would continue to 15 leak at high pressure. 16 MR. CRAMMOND: That's right. If you can't apply 17 a primary blowdown, then it's straight to core melt. 18 MR. EBERSOLE: I see. Okay. 19 MR. CRAMMOND: That would be the case where you 20 don't have secondary blowdown. Then, if the high pressure 21 injection system fails, then even though the aux feedwater 22 system is operating then you are losing inventory and you 23 lead to core uncovering. 24 MR. EBERSOLE: Are any of these studies valuable () 25 from the point of view of "what else"? You have the case ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-6646
m 525279.0 279 RT-
'l' of failure:in-the PORVs, analogous to having no PORVs at 2: all, which11s Palo Verde. One can-deduce from your studies 3- what state ~Palo Verde.is in. They-don't have any.PORVs.
4 MR. CRAMMOND: Not precisely. 5 MR.' ERICSON:. You have to be careful with that 6 statement. We are showing on .all of the PWRs the~ very
'7 . thing that-he just had up. Go back to your other one.
8 (Slide.) 9 MR. ERICSON: You can see the benefit in the
- 107 risk space of the capability to feed and bleed. Again, 111- -this comes back to the~ question we-talked about before, y/-a - 12 that this is in. risk space only, this doesn't address a lot d 13' of the environmental issues. If I'm opening PORVs and 14 dumping _into' containment, you have all those problems to 15 contend with. But this says in this particular situation.
6 the benefit is about 2: times 10 to the lainus 4 to be able 17 to feed and-bleed, and Point Beach has the procedures and
- 18 said they would if they had to.
19 MR. WARD: Wait a minute. Jesse's concern about-20 the-environmental -- what problem is in the risk space. 21 He's worried about that because he thinks you may not'have-22 an accurate estimate of the risk if you don't -- 23 MR. ERICSON: Right now I don't know how to 12 4 : quantify the degradation in PORVs in risk space. Don't say (f :25 because it's 2. times 10 to-the minus 4 in this study, i' ACE-FEDERAL REPORTERS, INC. < 202 347-3700 Nationwide Coverage 80(k33tHi646
..- _~ _. , ,_. . , . . - _ . . ._ _ . _ _ _ . . - . , . . _ _ . - , _ _ , _ -
25279.0 280 T l' particular limited study for Point Beach, it's the same 2 number for Palo Verde, because there are other systems that 3 have to be considered as you are doing that analysis. But, 4 clearly, here there is a benefit. No question. 5 MR. MARCHESE: I think if'the final resolution 6 of A-45 says that bleed and feed is the option to go 7 forward with, I think we would have to address the 8 environmental qualification issue as part of that.
'9 MR. EBERSOLE: Yes.
10 MR. MARCHESE: I think my viewpoint on that is 11 that we would have to insist that the valves have the (m 12 capability to operate in the environment that that's V) 13 created. , 14 MR. EBERSOLE: And-probably not operated on 15 sustained low voltage, but merely go to a position and stay
'16 there and be recoverable to another by motor-operated valve.
17 Get rid of these funny things called PORVs. 18 (Slide.) 19 MR. CRAMMOND: Let me refresh your memory. This 20 is the base case result where, from off-site to on-site, 21- where you run from less than $1000 per man-rem to $12,000.
-22 If we now consider the case where we do not have 23 interdiction and decontamination and we do the CRAC2 code 24 and go through all the calculations, all the same charts
() 25 that we had, that I didn't show you, leading up to that ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-33Hi646
25279.0- I 1 final chart I just took off, we find -- 2 MR. WARD: Is this table 9.9? Or what? 3 MR. CRAMMOND: This is table 9.9. That's
'4 correct.
5 MR. ERICSON: Last page. 6 MR. CRAMMOND: I'm going to wipe off the 1, 2, 3, 7 4, on the other side, 1, 2, 3, 4 -- now we see, whether 8 it's off-site costs or on-site costs, the dollar figures 9~ cre, for dollars per man-rem, are much, much lower. That 10 is simply if you don't have interdiction and 11 decontamination, then everything looks much more beneficial. ('i 12 Now, whether that's a reasonable assumption,
%/
13 that's another issue. But we did what we were asked to do 14 and looked at what it would be like:if we didn't have 15 interdiction and decontamination. 16 MR. ERICSON: But everything else is done the 17- exact same way. 18 MR. CRAMMOND: Installation costs, et cetera, 19 same delta core melts for each alternative. So that's a 20 big difference. 21 MR. WARD: Okay. Wait a minute. Go a little 22 slower here. Let's compare some numbers. 23 MR. CRAMMOND: Okay. Let's see if we can put 24 those on the same time. I) 25 MR. ERICSON: Lock at 9.6 and 9.9. Slide up a ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverase 800-33 4 646
25279.0 282 1 little further, Wally. 2 (Slide.) 3 MR. CRAMMOND: Higher? If we look at' the 4 central values, which I tried to put in orange here for 5 off-site and on-site combined, we have $23 per man-rem 6 versus 833; 64 versus 2335; 111 versus 3990; and 308 versus 7 12,104. 8 The reason is that you are finding that you have 9 much, much larger man-rem. In fact, I have those charts. 10 MR. WARD: They are dramatically different. But 11 is there any reason, I mean, to consider no interdiction or r- 12 decontamination as an alternative? What's the (_3/. 13- reasonableness of that assumption? 14 .MR. CRAMMOND: Andy, do you want to address that? 15- MR. MARCHESE: I'll try. It's going to be 16 . difficult. 17 Some members of the Staff have highlighted this 18 to management in terms of making sure that everyone 19 understands how the of f-site releases are calculated, via 20 the CRAC code. 21 The CRAC code presently has hard-wired into the 22 code, models on interdiction and decontamination. As you 23 know, these are off-site mitigation measures. 24 It has been questioned as to whether or not, ()- 25 really, it is appropriate in a program such as A-45, that's ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 804336 4646
25279.0 283 RT 1 examining changes internal to the plant, that is J 2 preventative measures internal to the plant, as to whether 3 or not it is appropriate to be mixing the value of those 4 changes in with the value that you get from off-site 5 mitigative responses. 6 What is happening is when you include 7 interdiction and decontamination into the calculations, it 8 is distorting the true value of a change, internal to the 9 plant. And what you find is that you have a real 10 insensitivity to internal changes at the plant. 11 That is, you look at the base case -- okay? -- 12 and then you make a modification to the plant. You add a
~~'
13 pump or system or whatever, and you calculate the value of 14 that change, that is reduction in core melt frequency, 15 reduction in off-site releases. Well, that delta is very 16 much influenced by whether or not you have interdiction and 17 decontamination in your calculation. And what some people 18 are saying is that you are really distorting the true value 19 of an internal change to the plant, that is a preventative 20 measure. 21 Other people would argue that in the final 22 analysis, there is really going to be some off-site 23 response. The question is: What is it going to be and is 24 the CRAC code modeling really realistic? 25 So, for the time being we have highlighted this ACE-FEDERAL REPORTERS, INC, 202-347-3700 Nationwide Coverage 800-336 6646 i
25279.0' 284 1 to management. We have told them there is a significant 2 effect on the results, in terr 3 of value impact; and for 3 the present what we are doing here is calculating it both' 4 ways until we get some guidance on how we should go forward 5 on this program. 6~ MR. WARD: Okay. Presumably, the CRAC code 7 includes some cost for the interdiction and decontamination 8 activities, as well as the reduction in person-rems. So 9 this is coming out with sort of a net cost or net benefit 10 or whatever? But the question is: How well is that done? 11 It's so important it shouldn't be just buried in the (' x.,} 12 results. 13 MR. MARCHESE: Right. 14 MR. EBERSOLE: Does the cost to interdiction and 15 decontamination appear to be cost beneficial? 16 MR. MARCHESE:. It's very cost beneficial. 17 MR. WARD: Better than another pump? 18 MR. ERICSON: If the base numbers are correct. 19 MR. MARCHESE: But when you talk about 20 interdiction and decontamination, you talk about other 21 people getting involved making decisions other than the NRC. 22 You have state and local officials and the government 23 involved. Whether or not the CRAC code is really modeling 24 what is actually going to happen is questionable, too. m () 25 MR. MINNERS: I would like to make one i ACE-FEDERAL REPORTERS, INC. l 202 347 3700 Nationwide Coverage 800-33H646
25279.0 285 RT 1 clarification. I don't know whether you meant that, 2 Mr. Ward. 3 The numbers, the dollars that he's showing up 4 there do not include, directly, off-site property damage. 5 Interdiction -- in any of those costs. He's taking 6 basically the on-site costs for making these changes or 7 whatever economic benefit you get from these changes and 8 then he's multiplying man-rems, he's comparing that ratio 9 to $1000 per man-rem. So nowhere in those numbers do you 10 see dollar values for off-site interdiction, evacuation, 11 those kind of costs. 12 MR. CRAMMOND: Off-site population dose and the 13 on-site costs averted become a negative impact. 14 MR. MINNERS: That's the impact. 15 i MR. EBERSOLE: Aren't off-site decontamination 16 and any -- well, I find it a little troublesome to talk 17 about interdiction. Aren't they fraught with all sorts of 18 legal claims and a host of costs that are not envisioned in 19 the current code? 20 MR. CRAMMOND: I'm sure that's true. 21 MR. EBERSOLE: It certainly is not going to be 22 less than the CRAC code -- 23 MR. CRAMMOND: I can't speak with any authority 24 on the CRAC code. [) 25 MR. MINNERS: But you are not using the CRAC ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Co.erage 800-336 6646
25279.0 286
'RT 1 code estimate. You are using S1000 per man-rem. You are 2 using that as a surrogate for health costs and property 3 damage and all of that kind of stuff.
4 MR. ERICSON: Except in the original -- in the 5 CRAC2 estimate, we are computing the dose which cra'nks in 6 areas set out for interdiction, area decontaminated. 7 MR. WARD: Okay. So there really isn't any 8 expense of decontamination factored into this; you are just 9 crediting what could be done as far as reducing off-site 10 dose. 11 MR. CRAMMOND: Just keep the people out of there
~'
12 for a certain amount of time. 13 MR. WARD: There's certainly an economic penalty 14 associated with that. 15 MR. ERICSON: You can go back to the CRAC code 16 and get those numbers out, if you choose that output. It 17 will give you the modeling they have done. 18 MR. CRAMMOND: We were asked just to use this 19 one measure, the population dose out to 50 miles as our 20 measure. 21 MR. WARD: Okay. So even this -- this isn't 22 really a fair display of the real cost / benefit of 23 interdiction then, either, because this isn't explicitly 24 showing the economic penalty associated with 1 25 decontamination or interdiction. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 804336-66 4
25279.0 287 1 MR. CRAMMOND: That's true. 2 MR. MINNERS: The criterion that's used in the 3 CRAC code, if you get a 30-year dose of 25 rem or greater, 4 you get interdiction or whatever -- that's somewhere near 5 the optimum for economics, if you value life more in the 6 S100,000 range rather than in the $1 million, like $1000 7 per man-rem. 8 MR. EBERSOLE: I would think you would find 9 numerous productive enterprises that have to be shut down 10 which would bring about additional costs. 11 MR. CRAMMOND: There certainly are other
.S 12 measures. If one wanted a full impact or economic impact i /
13 you'd have to consider the property damage. 14 MR. MINNERS: You can go look at the report and 15 that will give you those kind of calculated off-site costs, 16 okay? If you are interested in that. These numbers do not 17 explicitly include those costs, only indirectly through the 18 S1000 man-rem surrogate are those costs included. If you 19 want to know what those costs are, go look at the strip 20 calculation for Point Beach. I don't remember what the 21 numbera are. I'll give you the current estimate -- 22 MR. EBERSOLE: It would be very much 23 site-dependent, wouldn't it? 24 MR. CRAMMOND: Yes. Very much. We wouldn't I 25 contend that this is a complete analysis. It's what we ACE-FEDERAL PEPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-6646
'25279.0 288
~
11 were-asked tio do as a measure, and there are many measures i 2 .you can use. [ 3 MR. DAVIS: It seems, though, like the NRC has 4 got to make some kind of decision on whether modifications
~
S- to the plant are warranted. It's not clear to me on what-6 basis that decision 'is going to be.made. 7' MR.-MINNERS: It's not clear to anybody else, 8 "either. 9 MR. DAVIS: I guess that answers the question. L10 MR. EBERSOLE: If the plant is in the middle of i 11' nowhere and the employees can run off and leave it -- '
-12 MR. WARD: You are suggesting, for starters, you
'N '
13- think a more comprehensive look at off-site costs might be- ! 14' appropriate? Is that what you are saying? 15 MR. DAVIS . It seems to me that the picture is 16 incomplete on off-site cost and on the cost / benefit 17 evaluations that need to be made. 18 MR. WARD: Apparently those data, at least t 19 insofar.-- I don't know what CRAC calculates. Warren is j 20 saying there are more numbers available from the CRAC 21 output.
-22 MR. CRAMMOND: Yes.
23 MR. MINNERS: If you take those numbers, j 24 Mr. Ward, if you took the CRAC code and used the off-site () 25- costs that they calculate -- okay? -- you'll get lower ACE FEDERAL REPORTERS, INC. i 202-347 3700 Nationwide Coverage 800 336 6646
b 25279.0 289 RT .'
,1 dollar values than multiplying the dose by $1000 per
'; 2 man-rem, by a factor of 10 or more.
3 MR. WARDS, Well, not if you just add it to what 4 yod are' doing. 5 MR. MINNL'RS: Then that's double-counting, 6 because the S1000 per man-rem is supposed to take that into 7 account. You can either do it with the $1000 per man-rem 8 surrogate, or you can actually go and make the calculation 9 of what the off-site costs are.
~
10 - MR. WARD: How many hours do you put on man-rem 11 .then?
~-s 12 '
MR. MINNERS: You don't. People don't. They go 13 out -- 14 - MR. WARD: There's some health penalty 15 associated with substantial doses -- 16 MR. MINNERS: They calculate hospitalization 17 cost, they calculate cost of death and stuff like that. 18 But if you look at those health costs, I'm telling you, is they are in the $100,000 per death range. 20 MR. WARD: You are saying that $1000 per man-rem 21 is truly exaggerated? 22 - MR. MINNERS: S1000 per man-rem is valuing life
~
23 at millions of dollars per death. 24 MR. WARD: One other thing I guess I'm surprised
~'
25 at is the off-site benefits in the inclusion of the averted
- t
,, , y-T ACE-FEDERAL REPORTERS, INC.
o, 202 347 3700 / Nationwide Coverage 800 336-6646
/
25279.0 290 RT 1 on-site costs are not larger. 2 What is included in averted on-site costs? 3 Replacement power, for example? 4 MR. CRAMMOND: Replacement power, cleanup and 5 decontamination and replacing the equipment. 6 MR. MARCHESE: Capital investment. 7 MR. CRAMMOND: Yes. Capital investment. And, 8 as I recall, probably the man-rems, occupational man-rems 9 to the personnel in the cleanup. 10 MR. WARD: Okay. How did the assumptions there, 11 that are used there, compare with the experience of TMI-2?
-S 12 MR. ERICSON: Some of the generic numbers for 13 on-site costs come out of that experience. We estimate $1.2 14 billion on-site? That's the rule of thumb is on-site cleanup, 15 S1.22 billion.
16 MR. WARD: Okay. That's what I was trying to 17 get at. 18 MR. ERICSON: Those kinds of numbers are in 19 there. 20 MR. WARD: In table 9.6, the fourth column over, 21 total averted dose ratio, where the bottom number is .94. 22 MR. CRAMMOND: Yes? 23 MR. WARD: I'm not sure how that is defined. 24 What does .94 mean? [O 25 MR. ERICSON: You avert 94 percent of what you ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 80 4 33 % 646
25279.0 291 O' 1 would have had, had you had the accident, without the add-on 2 being there. l 3 MR. WARD: Oh, just those accidents that are -- 4 I mean the 40 percent or something? 5 MR. ERICSON: Yes. Of our numbers. 94 percent 6 of the dose that would have occurred under our analysis is
+
7 averted by putting the add-on on. 8 MR. WARD: Okay. That includes the spectrum of 9 accidents, which you said earlier might be about half. It 10 includes everything but ATWS, LOCA and -- 11 ; MR.,ERICSON: That's what we looked at, so it 12 could be larger. You can't eliminate more than all of it. [} 13 You mi'ght not affect ATWS. You might affect ATWS. . 14 MR. WARD: If it doesn't affect all those then 15 it's really half of that or something. 16 MR. CRAMMOND: That concludes my presentation, 17' if there's no further questions. 18 MR. WARD: One more. Some of the results here 19 are substantially changed from the draft report in May. 20 MR. CRAMMOND: That's true. 21 ' MR. WARD: Some of those are because of, I don't 22 know, additions or something that the Staff suggested. I' 23 don't mean to be critical with this question, but, I mean, 24 any analysis, as we proceed in time, we sort of approach () 25 what might be a true result as an asymptote of that. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
i
'25279.0 292 RT 1 How close are you to that? If you had another 2 year to spend on this, would you expect the results to 3 change substantially? Or do you think it's a mature 4 analysis now? What's your judgment on that?
5 MR. CRAMMOND: My personal judgment is that with 6 all things, we did to respond to the NRC concerns, that we 7 probably went from, maybe 70 percent accuracy to 90 percent, 8 or something. 9 There's still some things that we haven't done 10 that perhaps we should have done that might contribute, but 11 that, unless you wanted to go to great detail, which may or 12 may not buy you anything, like some of the PRAs are doing s ! 13 in the sense that they are looking at new territory, I 14 don't think there would be at lot of changes. Just by 15 going in and modeling things in more detail, I think we 16 have skimmed off the cream already. 17 Whether or not you uncover some peculiar thing 18 that you never would have noticed any other way is 19 problematical. Maybe you would and maybe you wouldn't. I 20 know there are some analyses where they found something 21 very, very peculiar that you never would have found if you 22 have hadn't gone into all the details, but I don't think 23 any of those have been big. 24 So I would say that it would be diminishing I 25 returns at this point to do anything in more detail than ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33M686
25279.0 293 I what we've done. Our goal now is trying to finish all the 2 ones that we have done, to try to bring the whole project 3 to some sort of a good conclusion and do what we are tasked 4 to do: Help NRC make decisions. I think. 5 MR. WARD: Any more questions on Point Beach? 6 MR. DAVIS: I had a couple of quick ones. 7 In your report and also in your flyers there's 8 two numbers noted for the core melt probability. 9- MR. CRAMMOND: That's right. I can explain that
!10 very easily. The first one is the estimate that we used
.11 when we were developing the alternatives, and which 12 combinations of modifications we would have. So those were
{ }. 13 estimates. 14 The second number that you saw was the number 15 that was computed using the sets code and a full 16 quantification. So there is the difference. 17 MR. DAVIS: In your slides, in the fourth slice 18 you have.added another sequence, long-term station blackcut,. l 19 which doesn't occur on the second slide. i, l 20 MR. CRAMMOND: That's right. Because -- I don't 21 know why I didn't have it there. It wasn't one of the. 22 accident sequences that was performed by the computer. The 23 long-term station blackout, we took cut sets from accident 24 sequences and then manipulated them by virtue of what would ( 25 happen in the case of a long-term station blackout where l-ACE-FEDERAL REPORTERS, INC. '
. 202-347 3700 Nationwide Coverage 800-33H646
25279.0 294
, ,BRT v)
( 1 the batteries depleted or the CST depleted. Then we came
- 2. up with that, essentially by hand, because it was easy to 3 do it by hand. So that's why it wasn't in the computer 4 list, where it was done separately.
5 MR. DAVIS: One final question: Do you recall 6 what you used for your auxiliary feedwater reliability 7 number? 8 MR. CRAMMOND: No. And I.can say that very 9 clearly because we didn't ccme up with a number for aux 10 feedwater. What we did is we had numbers for all the 11 components in the aux feedwater system and the LOCA faults, 12' test and maintenance, common mode, et cetera -- and on any g)
\- -13 given sequence, if the aux feedwater, whether it's success 14 or failure, if it appeared in that sequence it was simply 15 done in Boolean, so you never would know. In fact, we 16 don't have system unavailabilities for any system. They 17 were never run separately.
18 MR. DAVIS: I noticed that. I never noticed any 19 of those. 20 In appendix B you defined the aux feedwater 21 failure as failure of one of three trains. 22 MR. CRAMMOND: Yes. 23 MR. DAVIS: Does that mean that two out of three 24 trains causes a core melt?
/~% 25 MR. CRAMMOND: Succession is one out of three so V
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-o646 y --- - . . - - - - . - - - . - - - ~ ~ - - - - - - - - - ~ ~ ~ - - . -
. . . - . _ - . _ . . . - ~ _ . . . -
25279.0 295 1- you have to fail all three. 2 MR. DAVIS: Your definition then isn't correct. 3 You say failure is one out of three trains in appendix B. 4 MR. CRAMMOND: That's a shortcut. Yes, that's a 5 bit of jargon there. We say it's failure of success of one 6 of three, is what that means. 7 MR. DAVIS: Okay. 8 MR. CRAMMOND: I say that's a bit of local 9 jargon. I'm sorry about that. 10 MR. DAVIS: Thank you. That's all I have. 11 MR. WARD: Anyone else? Okay. Thank you, Wally. 12 Our next -- we've run a little ahead of time. I [ }_ 13 propose that we not take our lunch break but we go to the 14 next speaker, Mr. Hatch. 15 (Slide.) 16 MR. HATCH: I'm Steve Hatch and I'll be talking 17 about the Quad Cities results. Primarily I was in charge t 18 of the BWR analyses for the program. L 19 (Slide.) 20 I'll just mention a couple of things about the BWR 21 analyses, and point out some of the differences between the 22 BWR and PWR stuff. Primarily all the techniques that Wally 23 discussed were applied on the BWR work. There were some 24 differences, and, again, I'll point those out as I go along. jm., ( ,) 25 This is the Quad Cities transient event tree. ACE-FEDERAL REPORTERS, INC. 3)2-347 3700 Nationwide Coverage 800 334Wi646
s
~
\
C
;25279.0 296
^
T, s 1 I'm .not going to show you the .small LOCA event tree 12 although it's in your packet. It's essentially the same 3 type of tree. 4 .First of all, this tree is consistent with the
'S event trees developed in the risk methods'and integration ,
6 program, RMIEP program, the full. scope on La Salle being 7 done.at Sandia. Not as full as the RMIEP, but it's 8' consistent with those trees. 9 I have both an.early emergency coolant injection 11 0 Land a.. ate emergency coolant injection event. This is to 11 take'into account failures of~the front-lying ECCS systems,
. 112 ~ that cogld occur downstream after an initiating event, 13 three or four hours into the transient.
~
14' In the PWR. analyses they looked at the station
;15 - blackout question as a separate issue. Outside of the 16 event tree and fault tree analyses, looked at that' issue.
I : l'7 For the BWRs, I have modeled those sequences implicitly in t 18' the fault. tree and event tree analyses. 19 So, the early ECI failure is essentially those-20 Lwhere all of your systems fail, at time zero or within 21 'about 30 minutes into the accident. 22 The late ECI is where you initially succeed 23 ! emergency coolant injection. You have water going into the 24 vessel but at some point-downstream, in two or three hours, ()
~
25 you fail due.to a number of causes. Not only station l ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage M336 6646
25279.0 297 1 blackout, but it could be things like room cooling failures 2 that cause equipment to fail downstream, battery depletion -- 3 say you are in an AC blackout and the batteries deplete 4 after about three or four hours -- those type of things are 5 incorporated in this late term failure. 6 MR. EBERSOLE: May I ask, does this include the 7 lifting of this -- the SARs, as they call them? And the 8 interruption of water from a variety of sources? 9 MR. HATCH: Could you expand on that, Jesse? 10 MR. EBERSOLE: If you lost the capacity to 11 inject water at pressure, the line of recovery at the 71 12 boiler is depressurized using the semi-automatic release, xJ 13 primarily automatically, and then accomplishing 14 depressurization to quite low pressures and then inject 15 water, let it heat up the water, subsequently, under duress, 16 even to release it to atmosphere. 17 MR. HATCH: Is your question whether or not the 18 late-term failure would be failure to -- 19 MR. EBERSOLE: Does it include that mode of 20 operation? 21 MR. HATCH: Venting the torus. late term to allow 22 for -- 23 MR. EBERSOLE: Yes. 24 MR. HATCH: Yes. () 25 MR. EBERSOLE: Then it automatically includes ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
25279.0 298 T 1 the low feedwater introduction from even domestic waters? 2 Low pressure introduction - - 3 MR. HATCH: Let me qualify that. We did not 4 model many of the secondary water sources in our fault 5 trees, but we have considered those, where appropriate, as 6 recovery actions. Those would be manual-type things the 7 operator would have to perform, operations to establish 8 late term in the accident. 9 MR. EBERSOLE: I think most boilers have a
'10 method of introduction of cooling, using open pool pieces 11 which'can permit condenser water straight from the river to (3 -12 be put into the core, once you reduce the pressure; and V
13 therefore, you have a recovery method which is extremely 14 simple, simple open boiling to the containment atmosphere, 15 subsequent translation through the, even -- and then even 16 commonly said, a determination to open the containment. 17 MR. HATCH: We have given credit for those 18 secondary water sources where appropriate. For Quad Cities 19 the primary alternate shutdown method, it's called a safe-
-20 shutdown pump, a single pump, like RCIC capacity, it's an AC 21 pump, as an appendix R addition, and that has been the 22 primary secondary coolant source that we have looked at.
23 Although you are correct that most plants do have a service 24 water tie-in of some sort. ( )- - 25 MR. EBERSOLE: Right. The Quad Cities includes ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
25279.0 299 1 this terminal method of cooling? 2 MR. HATCH: Yes. 3 MR. EBERSOLE: Okay. 4 MR. HATCH: And many of the accident sequences 5 that were dominant, though, you didn't have electrical 6 power, so even some of these secondary sources which don't 7 have a dedicated power supply -- 8 MR. EBERSOLE: The fire water, was that used? 9 MR. HATCH: Pardon? 10 MR. EBERSOLE: The fire water? 11 MR. HATCH: They didn't have a tie-in from the (~) v.,) 12 diesel-driven firewater pump to the service, although that 13 was in the planning stages. They are planning to put that 14 in upstream somewhere. 15 MR. EBERSOLE: I see. So that will eliminate 16 the AC problem. 17 MR. HATCH: It should help it. 18 MR. EBERSOLE: To a degree. 19 MR. HATCH: Yes. 20 I'm going to skip over the next couple of 21 viewgraphs. I'm not going to show you the small LOCA tree 22 or the event tree definitions. 23 (Slide.) 24 I'll quickly go over some of the reasons why we () 25 selected Quad cities as a plant to study. Two main reasons: ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
i L25279.~0 300 1 One, Commonwealth Edison was one of the utilities that 2 expressed an-interest in allowing us to do a study on one - 3 of their plants; and then, as you might' remember, we went
,4 through a lengthy screening procedure to try to select 5 those older plants with potential vulnerabilities to study.
. 6 This is why you don't see many of the new plants
.7 popping up in our analyses, because those plants,.for the 8- 'most part,' conform to newer requirements and kind of 9 dropped out of our screening. The older plants, with --
L10 which did not -- do not conform to some of the newer.
- 11 requirements, surfaced and were the subject of our 12 particular plant studies.
13 These are some of the insights of.that~ initial
-14 screening that kind of pointed us into Quad Cities' 15 direction. These insights that could affect the decay heat
- 16. removal capability and' focused us towards Quad Cities.
,- Some were important in the . final : analysis, some were not. ' 18 I'll point out the ones that were more important. 19 The-fact they did not have an integrated shutdown panel was 20- important in the fire analysis. It's a two-unit plant, and 21 right now'they only have one 125 and one 250-volt batteries 22 for each unit and.they have cross ties between the two 23 divisions.But essentially it's only one DC train per unit. 24: That was a very important plant aspect to the study. 25 ( ). The cable separation was fairly important in the i ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 33Mi646
25279.0 301 hf I fire analysis. 2 Vent stack collapse was.something I talked to 13 you about back in March, I believe. That looked like it 4 was going to be an important wind problem; they have.this 5 big 350-foot stack that was thought to be a problem 6 dropping over on the turbine hall. That finally went away. 7~ The probability of that event occurring, initiating event 8 probability was such that that was not important in our 9 final analysis. 10 MR. EBERSOLE: The very first line seems to me 11- to have a substantial safety impact; you can't close the
/~l-V 12 safety relief valve when you want to and you can't open it 13 when you want to. Is that true?
14 MR. HATCH: That is something we did not pursue 15 in detail. 16 MR. EBERSOLE: That's one of the most critical 17 aspects of a boiler, can you depressurize it if you want to. 18 MR. HATCH: Is-the point you are bringing up at 19 what pressure it cannot open? The differential pressure? 20 MR. EBERSOLE: If I have no high pressure water, 21 and:the automatic aspect of safety relief does not function, 22 I want to know if I can forcibly open the safety release
- 23 valves.
24 MR. HATCH: I think this insight had to do with p As)- 25 -the ability to open up the SRBs, outside the control room. ACE. FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
25279.0 302 RT 1 And there were just limited sequences. 2 MR. EBERSOLE: Oh, that's only outside the 3 control room. 4 MR. HATCH: Remote outside the control room. 5 MR. EBERSOLE: They didn't put that in the 6 remote shut-in panel or anywhere else? 7 MR. HATCH: You can open them from the control 8 room. 9 MR. EBERSOLE: It doesn't say that. 10 MR. HATCH: That's perhaps a little misleading. 11 MR. WARD: That's what " remote" means. 12 MR. EBERSOLE: Well, yes, but normally you refer a 13 to remote manual as even in the control room. 14 MR. HATCH: That's perhaps a little misleading. 15 (Slide.) 16 MR. MICHELSON: Could we stop for just a moment 17 to pursue the events. Did you consider at all the 18 possibility of loss of coolant accidents outside of 19 containment? 20 MR. HATCH: We did not go through a 21 plant-specific search of small breaks; no. 22 MR. MICHELSON: That's not quite what I said, 23 identify the size. Did you consider, for instance, failure 24 in the reactor water cleanup system? I don't know if it's 25 even seismic at Quad Cities. I don't know if it's Q-8, at ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646 1
c. 25279.0 303 G L)T
~
'l Quad Cities. Full temperature and pressure constitutes a 2 full blown, almost about a medium break rupture. You know, 3 must isolate, there's further requirements and further 4 probabilities of failure to isolate. You didn't consider 5 any of that in your analysis? Is there some reason why
, 6 that one was left'out?
L7 . MR . HATCH: I wouldn't say it's left out. 8 MR. MICHELSON: You said you didn't include it. 9 MR. HATCH: We did not specifically look at 10 particular break locations. I think part of that is done 11 because of time and resources. That's a very lengthy type 12 of analyste, going and looking at particular places where a
- f'3 p
13 break could occur. We also didn't have -- 14 MR. MICHELSON: A pretty big system, though. 15- This is not a few feet of pipe. It's a-large sy' stem at 16 very high energy. It's reactor coolant. It's a loss of 17 coolant accident until isolated. Is there some reason why 18 these aren't included? 19 MR. HATCH: At the beginning of the program, 20 ~ medium and large LOCAs were excluded from the scope of our 21 study. It might fall in that category. But even the small 22 LOCAs, we did not go and -- whether or not it falls in that 23 range, we did not go and look for a particular break 24- location. We tried to generically come' up with, using past O T) m .25 risk assessment results, what the frequencies might have ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
i I J 25279.0 304 T.
- 1. been. .And,;indeed --
2- MR. MICHELSON: Let me ask, has anybody else who
~3 looked at-it, used asca basis to say, well, it looks so low i 4
as to'use as a basis not to consider it? 5 I.have yet to see the first analysis of reactor 6' cleanup:-- first reasonable analysis. Is there one 7- existing .that can be used as a basis to say it's a "no, 8 never mind"? 9 MR. DAVIS: There is one done for Millstone Unit 10 l recently released by the utility. 11 MR.-MICHELSON: What did it'show? y-$' 12 MR. DAVIS: The conclusion is it's not a
-Q) 13 dominant contributor, but one of the reasons is the system
- 14 ' has flow limitor in the line which are passed, and ends.up 15 being a small-break which can be handled in the normal way.
16 MR. MICHELSON: I'd have to see the analysis, 17 see how they chase-the water and steam -- 18 MR. DAVIS: It's presumably qualified. 19- MR .- EBERSOLE: More often than not those E20 conclusions are reached on the basis of the dose level 21 'obtained by this leaking system. And it doesinot include 22 the' degenerate-aspects of-continued discharge of fluids 23 into sensitive equipment areas. 24 Boilers have three lines which ought to be [ )l 25 looked at explicitly: the steam line supply to the RCIC; ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 2 336-6646
25279.0 . 305 T , 1 the steam line supply, if one, to the HPCI, most of which 2 are at full pressure right up to the throttle valve; and 3 then the line Carl was talking about. 4 These presumably have a competence to distribute 5 almost plant-wide a discharge of steam whicn may or may not 6 be curtailed by valves which have little if any record of 7 ' testing under duress. I think these critical three break 8 areas r.hould be explicitly pulled out and looked at, 9 irrespective of all those inside the containment. 10 MR. HATCH: I would think those particular type 11 of accidents are being assessed in the RMIEP assessment of (~3 12 La'Salle, which is going into environmental questions,. u/ y 13 environmental failures of components, et cetera. 14 MR. EBERSOLE: They may be. I don't know. 15 MR. MINNERS: Now that you mention the RCIC 16 . system, Jesse -- we took your suggestion and redid the 17- prioritization to redo the reactor water cleanup system so 18 there has been an analysis. 19 MR. MICHELSON: It's in the generic issue to be 20 resolved, but I was wondering, has the analysis been done 21 yet? 22 MR. MINNERS: There are scme numbers in there. 23 MR. MICHELSON: But you didn't begin to consider 24 the. problems of releasing the water and so forth and got-to I) 25 be very careful not to do one plant to do this whole ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-3364646
n, 252'79.0 306
- >' RT kJ l analysis and draw a conclusion. It depends on where the 2 system has been put. You'll find it's in various locations 3 in various plants.
4 MR. MINNERS: But that is a generic issue being
.- 5 resolved; presumably if we do it right we'll include all
-6 this.
7 MR. EBERSOLE: Why will you not use the classic 8- valve reliabilities when you examine it? They are'not 9 worth anything because they are not obtained from valves 10 operating 'in duress?
.1 1. MR. MINNERS: I don't know what we'll use.
'12 MR. EBERSOLE: I don't know what you'll use 13' either.
14 MR. MICHELSON: I guess what you are saying is 15 you'll go back and do some kind of comparable study to this 16 for.the BWRs as a part of the solution of that generic 17 issue? Is that what_you are saying?' 18 MR. MINNERS: Yes. 19 MR. MICHELSON: A PRA on that system would kind
-20 of.take care of it.
21- MR. MINNERS: I would presume people working 22 through the issue, at the end they'll have to work out the
- 23 impact analysis with a PRA in it which is going to make 24 some sort of estimate of the risk.
() 1 25- MR. MICHELSON: The solution is so far to keep
. ACE-FEDERAL RCPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336 6646
25279.0 307 T: 1 the valves closed, but of course that doesn't work for 2 reactor water coolant. 3 MR. MINNERS: But that's a good comment to bring 4 .up at the meetings, are you including the environmental 5 qualification effects in these analysis.
- 6. MR. CHELLIAH: I would like to add, including 7 Limerick PRA review, you brought up this point, Limerick as 8 well as Shoreham, in the PRA context, at least those two 9 plants we looked into that, the full range of frequency 10 resulting from that particular safety concern is somewhere
~11 around the minus O range, E minus 5 range.
12 MR. MICHELSON: You really have to go back and 13 .look hard, though, go back and check and see if they really 14 chased the water releases and'so forth and see what 15 assumptions they made concerning how long before they would 16 be able to obtain secondary isolation and so forth. 17- I find the things terminate a little too soon 18 and don't really chase the environment. They do a fair-job 19 of chasing the steam because it happens to have a blowout 20 panel for the steam, but the water doesn't go out the 21- blowout panel. It goes down to-the floor and down to the 22 lower. entrance of the ceilings. 23 MR. CHELLIAH: The frequencies are dependent on 24 the plan -- in Limerick it wasn't significant, but I think ()' 25 in Shoreham it was significant. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33 4 646
25279.0 308 RT' 1 MR. MICHELSON: In Limerick it shouldn't have l 2 been significant because they took a lot of special 3 precautions to vent both the water and steam out the 4 building. In Shoreham I'do not see such provisions at all, 5 and in the basement it's a very bad arrangement. I have 6 .yet to see the analysis that will show what it looks like 7 at Shoreham. 8- M R .~ EBERSOLE: Was the Quad cities plant
-9 vulnerable to floods other than the dam failure?
11 0 MR.-HATCH: Floods both internal and external
- 11 were not.found to be insignificant contributors.
gs c -12 ' MR. EBERSOLE: In the external area you
'%]'
13- mentioned the dam failure, which is the sudden flood, but 14 what about the long-term flood problem? Did it have one of
.15 those? I was under the impression it did.
16 MR. HATCH: I mentioned Cooper a little bit-17 earlier as having actually a dominant upstream dam failure 18 contributor. 19 MR. EBERSOLE: I'm not talking about.that kind. 20 I'm talking about water that comes up slow enough to take -- 21- MR. HATCH: That's what happened at Quad Cities 22 and that's why it.was not deemed to be a problem. Our 23 contractors looked at that long-term flood based on the 24 Army Corps of Engineer data in that area; and for the (') 25 worst-case flood where it rains upstream or whatever, ACE-FEDERAL REPORTERS, INC. - l 202-347-3700 Nationwide Coverage 800-336-6646
~
F l 1 25279.0 309 T 1 there's like a seven-day period where the waters slowly 2 come up, and the maximum is still lower than the top of the 3- doors of the reactor building. 4 But Quad Cities has a very interesting procedure 5 to handle that. What they do when they know that flood is 6 coming, essentially they shut down the plant. They shut 7 down all the equipment on the lower levels, rack it all out, 8 they take the top of the vessel"off, potentially get 9 natural circulation cooling going and they open up the 10 front doors and let the flood come in. They essentially 11 shut down the plant to a degree where flooding in the lower 12 elevation of the plant -- {'} 13 MR. EBERSOLE: Do they then use evaporate 14 cooling with just feedwater into the plant? 15 MR. HATCH: I believe they set up small pumps 16 and upper elevations and get the water in and just 17 evaporate cooling. 18 MR. EBERSOLE: This is the first example of open 19 evaporation to the environment.
'20 MR. HATCH: I'm not sure if it was the first or 21 not. But one of the problems was that the outer walls of 22 the reactor building were deemed to not be strong enough to 23- withstand, you know -- they could probably sandbag it so it 24 wouldn't get in, but then the wall loading would be such
()- 25' they couldn't guarantee they wouldn't have structural ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
r 25279.0 310 i f- RT l C[l 1 failures on the walls. 2 MR. EBERSOLE: Does that culminate in open pool 3 boiling to atmosphere? 4 MR. HATCH: I believe so. I'm not sure it goes 5 all the way to the atmosphere. I'm not sure where they are 6 eventually venting it off, whether it goes to the gas 7 system treatment or not, but I would guess it eventually 8 does. Some retention period and eventually released. 9 MR. MICHELSON: Just for the completeness of 10 your story, what did they do to prevent the torus from 11 lifting when they flooded the building with water? r~'S 12 MR. HATCH: They have flooded the torus. V 13 MR. MICHELSON: They weren't using open pool i 14 boiling of the torus, they were just opening up the vessel . 15 and letting the rest of it fill. 16 MR. EBERSOLE: Did they drain the oil tanks to 17 prevent displacement and subsequent fires? 18 MR. HATCH: I'm not sure what steps they took in 19 that direction. It was a rather lengthy preparatory list, 20 but they claimed they could do it in a day or two, once 21 they had the planning. 22 While we implement going from the first drafts 23 to the more finished drafts, I'll mention a couple of 24 things'plus some peculiar to Quad Cities.
<w l '(_) 25 We add pump and valve common mode failures. I l
l ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646 i t-
r ., o 4
- 25279.0 311 l- might add, that wasn't quite as important on the PWR as it 2 was'for BWR. We have more pumps, we don't have as many 3 redundant pumps in the same place, we don't have as many 4 valve combinations Lssociated with common mode failures, so 5 that really didn't add as much to the BWR analysis as it 6 did to the PWR analysis.
7 We increased the control circuit failure a 8 little bit. 19 We added sapport system initiating event. 10 Rather-a limited look at the support system initiates, one 11 DC and one AC bus failures as initiates. We felt from past (3
. %/
12 experience those were the most important of those types. 13 We reevaluated other data such as diesel LOCA faults, et
- 14 cetera.
15 One of the particular ones for Quad Cities was -- 16 one of the post-TMI modifications.being made at BWRs, is to 17 adjust the ADS logic, such that if you have a transient 18- with a stuck open relieve valve or many other transients, 19 .you will get automatic blowdown, even though you don't have 20 high drywall pressure. 21' In the initial version of Quad Cities, we did 22- not give them any credit for that. That is something they 23 are'doing. Either they have begun in this last outage or 24 will be doing in the next outage. So it was decided, since (). 25 we were going back and addressing other things, to go ahead ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
g
~25279.0: -312
, T 1- and giv'e them credit for that post-TMI modification.
2 M R .' EBERSOLE: So they put water on the core if 3 it.gets' low no matter what'other conditions exist.
'4- MR. HATCH: Essentially low water level; yes.
5 MR. EBERSOLE: That long has been evident that 6 that should be done. 7- MR. DAVIS: Just-a second on that one. In the 18 Millstone PRA, the argument is made that that-fix could 9 create problems because for some transients the water level
;10 .goes below the-low load level and these transients-would
- -11. normally would be recovered at high pressure. If this
}
12 .modif'ication is changed then you would blow the system down 13 and'cause lots of problems'and turn it into.an accident
~
, 14' more ' serious than :it would otherwise be. And they are 15 resisting this modification change. I don't know whether
-16 that's going.to cause any reconsideration'at Quad Cities or'
-17 not.
18 MR. HATCH: That's one of the reasons they'had l 19 the high drywall pressure signal in there. They want to
- 20 make sure you don't get an inadvertent blowdown. There is 21 a 30-second or two-minute -- there's a couple-of-minute
= 22 timer' involved with this which gives the operator an j 23 opportunity to assess the situation and determine whether v
24 it really is a case where we want to blow down. So they + () .25. haven't taken out all the safety precautions to try to ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646 E _- _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
25279.0 313 1 avoid a preliminary blowdown. 2 MR. EBERSOLE: But you can have a main steam 3 leak LOCA through the main steam valves and you are looking 4 at a vacuum through the reactor core which will suck it dry 5 in no time and you haven't got any containment pressure at 6 all in that case. You know, failure of the main steam 7 isolation valves are closed and subsequently looking into 8 bypass and other blowdown vessels. 9 MR. HATCH: The bottom line to changes, there 10 was a depletion in the core melt probability by about a 11 factor of 2. So it wasn't a real dramatic change. I think 'x 12 probably the biggest thing was the support system initiates. x._) 13 That was probably the bulk of the increase. 14 (Slide.) 15 MR. EBERSOLE: Did you look at common mode 16 failures in the dump mode design? Did you look at the 17 reactor control system? 18 MR. HATCH: No. And most ATWS systems were 19 deemed out of the scope of the analysis. We did not look 20 at that particular system. 21 Real briefly, the overall results. 22 (Slide.) 23 We had 21 internal event sequences that were 24 dominant prior to application of recovery factors. (h 25 " Dominant" in this case is with probabilities greater than ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
25279.0 314 1 10 to the minus 7. 2 We had 16 sequences dominant after recovery. 3 And, in general, the application of recovery actions was 4 like a factor of 4. Internal event core melt probability 5 went down by a factor of 4 after recovery was given.- 6 Seismic events and fires dominated the special
- 7. emergencies, and I'll talk more about them in a minute.
8 (Slide.) 9 There were three types of accidents that we 10 really looked at. We looked at accidents where you have 11 immediate emergency coolant injection failure, where the (% 12. ECCS~ systems fail, 1, 2, 3, right. within 30 minutes.
'd 13 The second type of accidents involved situations 14 where emergency coolant-injections succeeded internally but 15 then within about four hours failed due to a variety of
.16 reasons, room. cooling, battery depletion during an AC 17 blackout, temperature shock ongoing-from the condensate 18 shock to a hot' pool -- a whole variety of things were 19 looked at in terms what could fail the system long term.
20 The third type of accidents were transients 21 where you had a cooling for the bulk of the time and -- but 22 you don't have any cooling to the pool. And this is the 23 case where the suppression pool slowly heats up and in 24 about 30 hours was estimated to fail structurally and which (') 25 would lead to failure of ECCS. So we have kind of a ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 4 646
I
)
25279.0 315 1 30-minute sequence, about a four-hour sequence, and this 2 long-term 40-hour sequence. 3 MR. EBERSOLE: Those numbers at the right are 4 probabilities of core melt as a result of these? 5 MR. HATCH: Yes. I'm going to explain these. 6 The ones on your handouts are the total core melt 7 probability, of the things we've looked at, internal and 8 external, which follow in the three categories. So this 9 number includes the seismic-related accidents that led to 10 this type of scenario, and so forth. The fire-related 11 events that led to this type of scenario.
< 12 The numbers that I have above here are strictly 13 the internal event results which fall into each of the 14 three categories. So you can see that internal events 15 really dominated this first category, and those of that is 16 station blackout-related and electrical power-related. And 17 you can see that, in general, internal events dominated 18 that whole accident category. That accident category made 19 up about 60 percent of the total core melt probability.
20 Internal events only made up about half of the
' 21 transients, where you immediately fail ECCS; and special 22 emergencies, such as seismic and fire made up the other 23 half. Most of the special emergencies fell in this 24 category in terms of their contribution to the core melt
(_) 25 probability. Most of them, if you have a fire, it's going ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nanonwide Coverage 800-336-6646
f 25279.0 316 1 to probably get you immediately rather than -- you are 2 probably not going to get to the four-hour and 30-hour case. 3 Then you can see the relative contribution frem 4 internal events, and then total for the long term PW, going 5 back to the WASH-1400 terminology, those sequences only 6 contributed about 6 percent to the core melt probability. 7 I might note this number. This is the internal 8 event results for that long-term suppression pool heatup 9 accident. 10 This agrees very well with the NSAC study, DHR l'1 study on Brunswick. I believe they only looked at internal 12 events and they only looked at this class of accidents,
)
13 where essentially it's RHR failing, long term. And that's 14 their results of the whole study. I think their number was 15 40 to the minus six or something like that, for that type 16 of accident. 17 So, I think for a comparison of what NSAC looked 18 at and what we've got, it was very close, with less than a 19 factor of 2 difference. 20 MR. EBERSOLE: Which of these in both these 21 methods of depressurization and low pressure coolant 22 injection that you mentioned -- all of them? 23 MR. HATCH: No. Primarily this one. We did not 24 any credit for that mode of cooling for the real quick I[ ) 25 accidents. For this set of events you have cooling for the ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
y .
.: ew f ..25279.0 317 r
- 1
~
bulk of.30 hours. It's primarily where your systems fail,
-2 initially succeed but fail down line somewhere, you have
,N,w t f,g . 3 time to get the pecple.down,. manned at the stations and m
f , 4 Taligned and spool pieces in-if necessary. So most of.that j5L .particularsrecovery action'was that four-hour time sequence. _T :
-. 6 MR. EBERSOLE: Even though this was a diverse A ,
7 system you still didn't get a substantial -- 1, 2, 3, 4 --
- 8. MR. HATCH:.-This is the base case, internal. We
~~
, 3 -
'9. haven't, looked at modifications yet.
r M ~10
% MR. EBERSOLE: .Oh, you haven't?
+ - -11 MR. HATCH: The total core melt probability of f 12 thfs' class, which includes -- the difference between this
- q ,1 13 :num,ber and this number is not the application of a
;p 14 ,- mo'dification.
It's -- this'is internal, this is internal 15 plus seismic and fire and flood and all the others. MR. EBERSOLE: I see.. 17 MR. HATCH:. So we have internal event,
- / /
/ . -
-18 probability,-and then the total probabilities for each of
'19 these types of accidents.
. ~
.20 This slide is a little bit mislabeled.
21 MR. MICHELSOE: Could you refresh my memory
~
22 ag'ain as to what LOCAs you-actually did even look at in au ; t 23 t'his whole busin ss?' I notice there, there's only a~66' 24 percent -- there's practically nothing in your third item. 1 h 25 So y'ou better clarify for me what you mean by "LOCAs"? l
. ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 800-336 6646 c' -- __-m
25279.0 318 BRT O 1 MR. HATCH: In general, LOCAs were not important 2 at the depth we looked at them. Again, we did not look at 3 specific break location LOCAs. We essentially, we looked 4 at small LOCAs, less than, oh -- less than a square foot, 5 easily. I'm not sure. We didn't really have a set range. 6 MR. MICHELSON: Inside a containment or cutside 7 a containment? 8 MR. HATCH: We didn't distinguish between inside 9 and outside. 10 MR. MICHELSON: So a small LOCA in the RCIC 11 steam line, for example -- well, it should be a "no, never 12 mind" if properly isolated but there is a finite
)
13 probability of f ailure to isolate, but I guess you just 14 came out and ignored them? 15 MR. HATCH: I wouldn't say we went to the depth 16 where, we didn't look at particular depth locations -- 17 MR. MICHELSON: I don't know how you look at 18 LOCAs if you don't look at what is breaking. 19 MR. HATCH: We looked at it in terms of how it 20 would change the success criteria of the front line systems. 21 We looked at some of General Electric's break calculations, 1 22 which did include steam breaks, and looked at how the front 23 line' systems had to respond to those range of breaks. Some 24 were small, some were a little bit bigger. And based on ( ; 25 our analysis and success criteria on some of those General v ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-336-6646
i a 25279.0 319 1 Electric calculations. I 2 Essentially what G.E. has said is you have to 3 get a pretty big break in order to get a change in what 4 equipment needs tc' respond to a small break, including the 5 steam breaks. Essentially those requirements are the same 6 as a stuck open valve or even accidents without any loss of 7 coolant at all. The transient criteria -- 8 MR. MICHELSON: Except when the breaks are in an 9 area where there's some question about the environmental 10 qualification of this equipment. If you read LERs, you'll 11 see this equipment doesn't really seem to handle st'eam and 12 water too well in the basement, because there'have been 13 small leaks in the basement and lo and behold, the RCIC 14 doesn't want to work or whatever. 15 , I have a great deal of difficulty grasping these 16 numbers and then looking at reality, although mine are 17 apparently data points and not a big picture. But I just 18 -don't have a feeling of confidence that we could just 19 i'gnore these kinds of LOCAs because they are down in the 20 trash.
-21k; MR. HATCH: .C wouldn't say we are ignoring them.
I 3
~
22 At this stage it's difficult -- 23 MR. MICHELSON: People focus on your slide and 24 say we better worry about the transients. Don't worry
.() 25 about the small LOCAs down in the basement because that's ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
T
- 25279.0 320 l /5T x).
1 50-percent of our showing. 2~ MR. CHELLIAH: The third role does not include 3 the; frequency contribution from LOCAs outside the
~
- 4 containment. -
5 MR. MICHELSON: It's only-inside the containment j6- LOCAs? Okay. That's my answer. I thought I got both, ' 7 both-inside-and~outside, that you didn't differentiate,-but 8; I'must have misunderstood. 9- MR. HATCH: It's LOCAs primarily inside 10 containment, butLwe didn't make a decision about>where the
, 11 break was.
12 MR. MICHELSON: As long as it's inside the 13 containment you simply have to say this doesn't deal with 14 stuff outside the containment,'and if you want to.know .
- 15. about . stuff like ' that, we'll have to look. We don't know.
'16- MR.' HATCH: That's one of the reasons we didn't 17~ go'into more detail. -That is a-very difficult topic and.
18 one which, for a variety of reasons, we have excluded from 19 the, scope-of this study. 20 MR.-EBERSOLE: What do you think would have 21 happened if they had firewater injections, which requires 22 only diesels to. maintain a feedwater flow at low pressure? f, 23- I'm' talking about, you know, a fairly simple modification. E 24- MR. HATCH: One that was easily started up
.,h -25 .within a short time period?
ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-33H646 L , _ . . . _ _ . . _ _ _ _ _
25279.0 321 BRT i ss' 1 MR. EBERSOLE: Sure. 2 MR. HATCH: It would primarily affect this. 3 Well, I guess it would have affected both these numbers. 4 By how much, I don't know. 5 MR. ERICSON: But it's simply a recovery action. 6 It's not part of the automatic action. 7 MR. HATCH: If it was made part of the normal 8 EOPs, or emergency op's -- 9 MR. ERICSON: But it's still a recovery action, 10 not the plant in its normal operating mode, which is what 11 is depicted here. No recovery in these numbers, just what 7~ 12 the plant normally is in its normal operation mode. 13 MR. HATCH: There is -- 14 MR. EBERSOLE: I thought this included a 15 recovery where you bumped water from the SRBs and water 16 from wherever you had it, but you need electric power to do 17 that so you didn't incorporate the LaSalle availability on 18 the power system. 19 MR. HATCH: If Quad Cities had diesel-driven 20 fire pump that had good connections and they said they knew 21 how to do it and showed us how to do it, as they do at 22 Cooper. We have given some credit. Given the fact that we 23 haven't done a detailed human factors analysis, we don't 24 know a whole lot about some of the other mechanisms I; 25 involved -- we haven't necessarily given a lot of credit ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80 4 336-6646 ). . _ . - .
25279.0 322
-1 for.those systems,.but we have given some credit where 2 appropriate. -And if Quad Cities had had that type of setup 3 outside of this safe shutdown pump, the numbers here would
'4 have been somewhat different.
5- MR. EBERSOLE: It's like the credit you give to 6 .the fire department for coming to put out a house fire. 4 7 Better than that, I hope. 8 MR. HATCH: I'll just briefly go over the 9 internal event sequences.
- 10. (Slide.)
11 The first couple, the YZ sequences are-those (~% 12- long-term sequences where the torus heats up. You can see
%)
13 those probabilities were pretty low. These are the 14 sequences that match up pretty well with what happened at 15 Brunswick. The YZE sequences are the dominant ones, this
.16 is the dominant one here, which is primarily an AC 17 blackout-related sequence. ,
18 These TD sequences are the case where you feel
'19 immediately. And the ones down here, the TAC and TDC 20 sequences are the support system initiates, and they
'21 contributed about 20 percent to the overall internal event-6 22 frequency of core melt.
2l3 MR. DAVIS: Excuse me. One of the conclusions of the Brunswick study was that service water failure was fh 25 A,) .quite important and they had a probability over 10 to the [ ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33H646
25279.0 323 1 minus 5 for service water failure contribution. 2 Did you find a similar problem in Quad Cities? 3 MR. HATCH: Yes. 4 MR. DAVIS: Is it on the list there? 5 MR. HATCH: Well, service water failures, 6 secondary cooling of the RHR heat exchangers, those type of 7 f.ilures were important. They are not drawn out explicitly 8 but they are inherent in these YZC sequences. They are 9 part of those sequences. We did not go out and pull out 10 individual contributors in terms of system contribution or 11 component contributions, but I think we would agree that
-s 12 it's the support systems, and in particular, cooling water,
\.. ; 13 that drive a lot of it. 14 MR. MICHELSON: Does your answer mean that you 15 started out with the assumption that you suddenly lost all 16 service water and went through -- you didn't go through 17 that so you don't know the answer in terms of service water 18 losses and initiating an event and how important it is. 19 MR. HATCH: We didn't look at service water as 20 an initiating event, but as a subsequent random event after 21 another initiator, we have included it. We thought about 22 including the surface water initiator but decided not to 23 for a couple of reasons: One, those events are much more 24 difficult to find in the course of the plant study. The ( ) 25 couple we picked out were the AC and DC bus initiates. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336 4 686
[25279.0 324 RT. 1 Those, from past studies were found to be over, again and 2 again, dominant contributors. They are very easy to 3- identify. It's usually.just a couple of buses you have to 4 look at. 5 The service water-type failures, again, are
~6 difficult to find, and they are always -- usually long term 7 lln developing.- You lose some component in service water, 8 things slowly heat up. It builds up over a period of hours, 9 usually. And that was another reason we decided not to 10 pursue that support system initiator because you have a lot 11 of time. There's more possibility for an operator to 12 recover from that type of failure than, say, a bus failure.
(s) 13 And from past PRA experience, it's the bus failures that 14 should dominate. 15 MR. EBERSOLE: But what about the probability? 16 Bus f ailure probability is very low and service water 17 probability, do you find that to be low as well? You find 18 it drawn out, I guess. 19 MR. HATCH: It depends on the type of failure 20 you are talking about. In the ANO study, one of the 21 initiates was a valve failure, that was a 10 to the minus 3 22 event. And they are talking here in the 10 to the minus 3, 23 .4 or 5 times 10 to the minus 3 probability ~of initiating 24 event that we use. It's not that small. You have an 25 initiating event and you have half of your systems gone at ()N ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646 L . , _ - . - . . . . - - - - . , . . - ~ - - . . -. - ~ - .
Y f25279.0 325 ic 1 the fir'st-instant. So even though it's a very small 2 initiating event probability, what you. start out with is 3 'much less than the other initiates. 4 MR. EBERSOLE: Tell me, considering all these, 5 isn't.the lesson to be' learned here the main problem is 6 . loss of feedwater supply due to loss of electric power? 17 MR. HATCH: It's primarily electric. power.
'8 Diesel and diesel support systems.
.9 MR. EBERSOLE: -So then one looks for a source of
-10 water wherever he can find it or not driven to the
~11- electrical network.
t r*N - 12 -MR. HATCH: Or looking for'another network
. Qj '
13 sometimes. ., il4 L MR. EBERSOLE: Well, that's the-patchwork
- 15. approach. Looking for some way to get power.
16 (Slide.) 17 Most of the internal event vulnerabilities, 18 again, are related to electrical power. The first 19 vulnerability was LOCA faults of two diesels. -Quad Cities 20 is a three-unit plant and they have two diesels for each
- 21 unit, so each unit has a diesel and a half available.
22 There's a swing diesel that can flip out. - 23 There's local faults dedicated to diesels and r 24^ then the swing diesel is important. Failure of the field
- f. 25' flashing was found important.
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 4646
25279.0 326 T. 1 .The diesel-cooling water failure.was interesting. 2 Each has a one-pump standby system dedicated to provide
.3 cooling, but also they take cooling out of one of the pumps; NL . one of the diesel pumps supplies room cooling for the whole 5 plant -- whole unit. So these cooling water failures, not
.6. only diesel but room cooling that's associated with this.
7 That was kind.of an interesting system-type failure. And 8 then DC control power failure to the -- some of-the pump 9 breakers and valve breakers and to the ECCS logic was 10 -another vulnerability identified.
.11 MR. EBERSOLE: Do you tell me now that one of jewg 12 the diesels supplies cooling work to the room coolers. .You
-U 13- have'said for the whole-plant, does that mean both diesel 14 room?
'15 MR. HATCH: No. Each diesel has a one-train
'16 system that supplies cooling water to the diesel and room 17~ cooling for that diesel. But off one of those diesel
.18 cooling lines it goes to the RCIC rooms, the basement 19- corner rooms, the RHR lube oil pumps --
20 MR. EBERSOLE: And there's no cross-tie to the 21- other diesel? L -22 MR. HATCH: There's a way to manually cross-tie 23 between one plan, Wally, one diesel to the other. ii 24 MR. EBERSOLE: But it's a low rising problem. , e (,m) 25 MR. HATCH: We did if it's appropriate. We went l l ACE-FEDERAL REPORTERS, INC. ( 202-347-3700 Nationwide Coverage 800-336 6646
6 a d A 25279.0- 327. I blf ' '
'l' down andfgave: credit for operators going down and opening 2- valves --
3 MR. EBERSOLE: This suggests there's a single , l :4' line that' runs to all these cooling pumps.that is somewhere
.5 .off in the building; is.that correct? .
;6 MR. HATCH: Some pumps, some have self-cooling 7 abilities such as core spray. But in general, you are
'8 'right, and the RCIC --
~
9 MR. EBERSOLE: Hcw does that meet the single 110- . piping failures criteria? c . 11: MR. HATCH: I'm not sure. exactly what-the-
-4
"(5L) 1 12 . criteria"is i for this type of system, but indeed this is one
-13 of th'e-things we expected at some of the older plants, and 14 we want to evaluate:the plants for DHR vulnerabilities. ,
15 MR. EBERSOLE: If I have trouble'with this 3 4 16 cooling line, would it? 17i MR. HATCH: Not immediate. There are four hours. r - 18 MR. EBERSOLE: There's some patch-up time. 19- MR. HATCH: Yes. 20= MR. MICHELSON: The water is hot -- t 21 MR. HATCH: Core spray has internal lube oil
- 22. -cooling, but it.still has some cooling --
23 MR. MICHELSON:- It:has to all be from some
, 24. external water source, obviously. I'm a little surprised
? f~\ .
(,) .25 there's only a single pass system' supplying cooling for all e v J ACE-FEDERAL REPORTERS, INC. i 202-3'47-3700' Nationwide Coverage 800 336-6646
25279.0 328 d"' I the pumps, which is what you are saying? 2 MR. HATCH: Yes. That's exactly right. With 3 some manual intertying to the other -- 4 MR. MICHELSON: I guess, although you didn't -- 5 this is one of the deficiencies in this analysis, as I 6 questioned earlier for the others, I assume you didn't just 7 go through and take such pipe breaks as a part of your 8 analysis? 9 MR. HATCH: These systems are -- it's a standby 10 system so it's not normally operating. 11 MR. MICHELSON: No. But the pipe break may lead es 12 to a need for operation, depending on where in the building J 13 the pipe break is located. 14 MR. HATCH: That's certainly true of many pipes. 15 I wouldn't necessarily pick on this pipe. 16 MR. MICHELSON: You don't even go into that kind 17 of analysis. 18 MR. EBERSOLE: So it's normally not in operation, 19 it's called up in emergency? This? 20 MR. HATCH: Yes. Room cooling to the corner 21 rooms is not needed until you get a situation where the 22 torus is heating up and you need the pumps. , 23 MR. EBERSOLE: Is it designed to eliminate water 24 hammer when it starts, on emergency? () 25 MR. MICHELSON: I think you'll find it's ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80 4336-6646
m 25379.0 329 1 designed to start on shutdown. You have to take that motor 2 heat'out of the compartments or areas.
, 3 MR. HATCH: It's not the motor heat so much.
4 MR. MICHELSON: It's a heck of a lot of motor 5 heat.from a 600 horsepower motor. 6 MR. HATCH: The RHR rooms at Quad Cities are 7 pretty much wide open and extend up about three stories. 8 Talking about Oak Ridge on what the room cooling 9 requirements are, were they felt the only time room cooling 10 requirements would be needed for the corner rooms is where 11 you don't have suppression cooling and you have radiated
'}-
/
12 heat coming into those rooms, and that's where we modeled 13 it. For those cases where you have suppression pool 14 cooling, that torus doesn't get hot enough to require room 15- cooling, and so it's only special cases where we decided 16 room cooling was needed, and.this was based on some 17 discussion with Oak Ridge personnel. We're taking a good 18 look at that. 19 MR. EBERSOLE: Those pumps, as I recall, are 20 2000 horsepower. A percentage of that is raw heat into the 21 . room and the rooms are pretty big. 22 MR. HATCH: The rooms are fairly spacious as 23 compared to some other plants and, again, they have quite a 24 bit of thermal mass in the concrete, but they also have A ( ). 25 this extended -- this stairway goes up; esrientially you can ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 4646
25279.0 330 BRT O 1 look up three stories. There's quite a bit of volume for 2 the heat to go into, and it's only in that particular case 3 where the cooling fails where it was felt to be critical. 4 Except for RCIC. It's often a different beats, often a 5 little room with a very small cooler. We modeled RCIC as 6 needing cooling all times, when it's running, anyway. So 7 RCIC is the special case for room cooling. 8 (Slide.) 9 MR. HATCH: The internal event mo<lifications 10 tried to address the vulnerabilities we identified. The 11 first one is to add a fourth diesel. And let me remind you, ,s 12 you have three diesels for two plants, so really we are 13 coming up to two diesels for Unit 1, two dedicated diesels 14 for our unit. 15 We looked at adding a dedicated battery on one 16 of the diesels. It was felt that you didn't really need to 17 add dedicated batteries to both in order to get a good core 18 melt reduction, and we were trying to minimize the cost 19 wherever we could and so we only looked at, put a battery 20 on one of the diesels. 21 We looked at installing an additional DG cooling 22 water pump, which would supply water to the diesels and 23 this room cooling line we spoke of; and we also looked at 24 automatic transfer of critical DC loads of one bus to / 25 another in the event you lose a battery or bus or et cetera.
}
ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6646 _. -. . . . . _ . _ . . .......... ..._ ,._. ..,._ ,.,m _., ,.,,.. .. . _ _ _ . . . . . , . . . ..
i 25279.0 331 1 MR. EBERSOLE: That last item is extremely 2 controversial in that it invites cascade failures if you 3 connect into a load which is in fact on a failure? Do you 4 do that on a supervised basis, and what do you think about 5 this? I thought that had been virtually disallowed as a 6 generic approach for diesel, the transfer to a bus and thus 7 inviting the cascade failure of both sources? Has this 8 been worked up in some detail? Is that under consideration 9 now? Is the electrical department of the Staff here in 10 agreement with this general proposition to transfer DC 11 loads?
~x 12 MR. MINNERS: Jesse, if I may be a little w) 13 impertinent, I think you've gone beyond the scope of our 14 presentation. We are just trying to tell you what we did 15 and you are trying to get us to say whether we agree or 16 disagree and this is what we are going to do or not. We 17 are a long, long way frcm deciding what we are going to do.
18 MR. EBERSOLE: Well, okay. I'll just say that's 19 a proposition fraught with a lot of argument. 20 MR. HATCH: At Quad Cities they already had 21 automatic transfers of RCIC DC loads. They had some of 22 these equipment in already. We were just proposing to put 23 similar equipment in in a similar fashion. 24 MR. EBERSOLE: It's always a controversy. Never g (, 25 escape it. I think the question is: Did you examine the ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80433M646
25279.0 332 RT 1 down side when you did this? Did you get a probability -- 2 MR. WARD: For any of the fixes. It's really a
.3 methodology question.
4 MR. MICHELSON: I'm going to watch this closely 5 and start talking about five times the SSC, because this is 6 probably really enforcement, too. It gets real touchy 7 worrying about the for the SSEC. 8 MR. HATCH: I'm sure the contractor used the 9 best engineering they could to avoid getting yourself into 10 a worse case with the fix than not having the fix. But I 11 don't think we can rule out possibilities of that nature. gm 12 MR. WARD: Okay. But there wasn't any explicit
~
i 13 reevaluation of a risk with this system, for introduction 14 of new risks perhaps? 15 MR. HATCH: No. We did not interject negative 16 performance in our modification. 17 I'll quickly go through the fire analysis. 18 There were two vulnerabilities identified in the fire 19 analysis. Essentially what they did was went around and 20 looked at single locations where, if you had a fire, of a 21 certain magnitude, could take out enough systems to get the 22 plant in serious trouble. And, of course I think the 23 common ones that have been found in other studies: Control 24 room fires and then the main cable spreading room, which is 25 shared by both units. (' ') ACE-FEDERAL REPORTERS, INC. 2@34' 17N Nationwide Coverage 800-336-6646
25279.0: 333 i
- 1 .MR. EBERSOLE: Are they in compliance with 2- appendix R at this time?
.3 MR.' HATCH: I believe.they have initiated their
- 4. safe' shutdown pump modification the way they intended it to 5' - be. :I don' t know what the result is.
6 MR. EBERSOLE: It's kind of a catch-all, isn't
~
-7 it?
- 8. MR. HATCH: I-believe so.
9.' So we had about a 10 to the minus 5 per-reactor
.10. year vulnerability to fire. The modification is - -
11' . involves the-safe. shutdown pump. > Essentially, the people Q
~
' .12 .w ho did"the fire analysis determined that there was a
- 13. - significant probability that the safe shutdown pump would 14 not be aligned to the proper. unit, or when the unit that 15 .needed it had thatLneed. And they felt that adjustments to
-16 the operating procedures for that pump could alleviate the 17 vulnerability to or reduce the vulnerability to fire.
18 So essentially what the modification was was i to 19 modify the procedure for the safe shutdown pump operation
.20 to. assure that this pump gets aligned to the proper unit
, 21 when it's'needed, and'the results were, oh, a factor of --
- 22 whatever. Factor of 4 reduction.
23 MR. EBERSOLE: This is the pump now that' provides low pressure feedwater in the open SRV motor
'25' coolant; right?
ACE. FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
125279.'0 334 1 MR.-HATCH: I'm not sure it's low pressure. I'm 2 trying to remember what the discharge is. It might be 3 higher than a LPSC. 4 MR. EBERSOLE: Well, if it's high pressure it 5_ can also be low pressure. 6 MR. HATCH: About RCIC. 7 MR. EBERSOLE: Well, did the Staff then temper 8- their requirements of appendix R? 9 MR. HATCH: Pardon?
'10 - MR. EBERSOLE: Once this was agreed to as a 11 countermeasure, it presumably would permit a good deal of , (-] 12 tempering of requirements on fire protection elsewhere.
v 13 MR. HATCH: I'm not sure what the results were 14 on that. 15 MR. ERICSON: We are not in the appendix R 16 business. 17 MR. MINNERS: Yet. 18 MR. DAVIS: One thing appendix R can do is 19 increase the problem of pump room cooling if you have to 2:0 install three-hour fire walls around all the pumps, that 21 can, you know, change the convection cooling availability 22 to the pump and change the requirements of pump room 2:3 cooling. 24 MR. EBERSOLE: That's where the shutdown pump ex ( 25 might be a better source.
)
ACE-FEDERAL REPORTERS, INC, 202-347 3700 Nationwide Coserage 800 336-6646
'25279.0 335 ; T-1 (Slide.)
2L MR. HETCH: What I'd like to point out about the 3 seismic analysis is Quad Cities does have a fairly 4 significant shutdown earthquake. It's about twice that of 5 Point Beach. That has some interesting results, I think. 6 If you were to look at the frequencies of earthquakes for 7 the-different ranges of multipliers of the SSE, you see 8 that most of the frequency, initiating event frequency does 9 fall in the range that the p.'. ant was designed-for. In fact, 10 we found -- we felt that Quad Cities was well designed for 11 the SSE. And, you know, if you looked at this number you'd {} 12 13 say, yes, that is what you should design for. It was kind
-of interesting the way the core molt frequencies panned out, 14 though.
15 MR. EBERSOLE: Pardon me, before you go away 16 from there, you have T-1 and T-2. Isn't T-2. automatically 17 a consequence of T-l? 18 MR. HATCH: Yes, that's true. There's not a 19 whole lot of distinction between the two; although if you 20 know you have lost power, that affects other systems beside 21 feedwater.
- 22. MR. EBERSOLE: Okay.
23 (Slide.) 24 MR. HATCH: The interesting thing then, if you
. () 25 look at the seismic core melt probabilities, most of the ACE FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800 336-6646 o
25279.0 336 1 seismic contribution falls in ranges above the SSE. So, 2 while the plant is well designed for the SSE, we did find a 3 fairly hefty seismic contribution to core melt. And the 4 reason is, is that much of the contribution falls above 5 where the plant was designed. And the total seismic 6 contribution being about 8.3 E to the minus 5. 7 MR. EBERSOLE: Is the safety shutdown pump going 8 to offer any recourse? 9 MR. HATCH: The safe shutdown pump was looked at 10 as a recovery action. For some of the sequences -- 11 MR. EBERSOLE: Is it seismic? 12 MR. HATCH: I don't believe it's seismic. I I 13 Strictly appendix R. It wasn't a whole lot -- 14 MR. EBERSOLE: It won't then accommodate loss of 15 DC power? 16 MR. HATCH: No. It's dependent on diesel or 17 off-site power, which, quite likely, won't be available. 18 MR. EBERSOLE: Okay. 19 MR. DAVIS: On this seismic issue, maybe you can 20 clarify something for me. On page 10, section 10, you make 21 the statement that "Although the core melt probability was 22 found to be significant, it was determined that the plant 23 was well designed for such events." 24 What does that mean, exactly?
) 25 MR. HATCH: I think we determined that the plant ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-33H646
'25279.0 337 f)';T h,
1 .is well designed for what it was supposed to be designed, a 2 safe shutdown earthquake. And we were not able to find a 3 _ whole lot of individual fixes or modifications that would 4- really do much to reduce that-probability. What you would 5' have to do at that stage goes beyond patches or whatever. 6- It gets into a place where you'd have to replace whole 7 components of valves or breakers or whatever -- 8 modifications that we weren't prepared to take L look at; 9 but just from the systems, the physical systems, and to the 10 degree we looked at it, there wasn't a whole lot we could
-11 do, and I'll get to the couple of modifications we did come a' .12 up within a minute. ' q% J .
13 MR. DAVIS: I_ understand it now, but it is 14 confusing the way it's written. It looks like it's "well 15 designed for core melts." 16 Anyway, the.second sentence is "The reason for , 17' the large core melt probability is the result of the high 18 SSE level." 19 I have trouble with that, too. You mean if they 20 had design of an SSE level they would have had lower core 21 melt -- 22 MR. HATCH: The other way around. 23 MR. ERICSON: There SSE is a quarter of a G. 24 That's high. So when you get to two times that, you are () 25 getting up there and you are really starting to shake ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 2 336-6646
25279.0 338 RT 1 things pret".y hard. That's the intent of the comment. In 2 other words, they are designed to a quarter of a G, so if 3 you go to two times that you are up to a half a G or more. 4 MR. DAVIS: Probability should fall off rapidly 5 for accelerations that high. 6 MR. ERICSON: If we go to twice the SSE we are
, 7 really shaking the equipment hard, and so we get a 8 contribution. That's what we are trying to say.
9 We obviously did not communicate our thought 10 correctly. 11 MR. WARD: Pete, they've a different hazards n 12 curver right? , i 13 MR. DAVIS: Yes. 14 MR. HATCH: This is Quad Cities.
'15 MR. DAVIS: But it says the reason the core melt 16 is high is because the SSE design basis is high.
17 MR. CATTON: They need to rewrite the paragraph. 18 MR. ERICSON: We didn't communicate what we 19 meant to say. 20 MR. DAVIS: That means if you redesign to a high 21 level the core melt would be higher. 22 MR. ERICSON: If you designed a half a G and 2 23 SSE would be really a G, you'd really be tearing stuff up. 24 MR. DAVIS: But the probability has to fall all I ) 25 the way down for a G. ACE-FEDERAL REPORTERS, INC. 202-147-3700 Nationwide Coverage 800 336 6646
+%
525279.0: 339 11 MR. WARD: If you are at a site where you have 2 to design to' half a G, then what they've said follows. 3' MR. ERICSON: You arc right, Pete. What we said-
;4- 'was poorly worded and we need to take a hard look at that 15 with the seismic guys. We did not communicate the intent..
6: MIR. : HATCH: And-the frequency of seismic events 7 is dominant in the SSE range. But if you look at the core 8 : melt _ probability, it is falling in the later -- higher
.9
~
ranges, so you might argue that the plant is well. designed 10 for earthquakes that -- of the SSE -- in the SSE range. 11 It's'not designed for these higher ones, and unfortunately,
- 12. that's where some;of the core melt contribution is falling, 13 in"the higher ranges.
14 MR. DAVIS: ~That's always been found to be the
-15 case for seismic analysis.
16' MR. HATCH: If the SSE had been lower,-these 17' numbers would not have been as high. More of it would have
- 18. been shifted down here.
19 MR. DAVIS: Right, and the total CMP would have
- 20 been_ higher-for seismic.
21- MR. EBERSOLE: You said the safe shutdown pump i 22 was.not seismic?
' 23 MR. HATCH: I don't believe it's seismic.
i. p 24- MR. EBERSOLE: If that's the case, why was it (f 25' ,p referred to put in a safe shutdown pump rather than a ACE-FEDERAL REPORTERS, INC. 202-347 3700 . Nationwide Coverage 800 336 4646
25279.0 340 RT l- cross-tie into the five system?
-2 MR. HATCH: That's an appendix R modification.
3 I don't think they had seismic in mind when they put that 4 in. 5 MR. EBERSOLE: So it's simply a quick fix for 6 that, _for fire? And doesn't incorporate the generic 7_ quickness of electric power failure? 8 MR. HATCH: Don't think they were addressing 9 electric power -- 10 MR. EBERSOLE: The cross-tie to the fire system 11 would not have been that. 7s 12 MR. HATCH: The two vulnerabilities that we did L) 13 come up with outside of global modifications, their DC 14 battery racks were wooden. Those were felt to be fairly 15 fragile, so we did look at installing metal . battery racks 16 to beef up those battery supports. And the plant also has 17 a number of vital' buses sitting out in this big main 18 turbine hall for both units at main levels, and they are 19 not anchored. They are vertical cabinets and they are not 20 anchored at the top, and the seismic people found that some 21 of the vibration could come up and get those things 22 wobbling, and they were found to fail under low 23 accelerations. 24 MR. EBERSOLE: Were they vulnerable to turbine () 25 explosion? ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nation *He Coverage 800 336-6646
25279.0 341 1 MR. HATCH: I think they were shielded from the 2 turbine, but I didn't do that. 3 MR. EBERSOLE: You said they were in the turbire 4
- hall.
5 MR. HATCH: There are some walls, not full walls 6 but -- between them and the turbine. I don't think we went 7 into great detail on missiles. 8 MR. MICHELSON: Are these safety-related 9 cabinets and they are in the turbine hall? 10 MR. HATCH: Yes. Which is controlled. 11 MR. MICHELSON: What do you mean, "which is y 12 controlled." From a seismic viewpoint, is the , turbine hall 13 seismically qualified? 14 MR. ERICSON: We don't know the answer to that 15 question. Our people analyzed what those would do. 16 MR. HATCH: The seismic' people don't care what's 17 seismically qualified and isn't. They go in and look at 18 what the the component looks like and whether it's bolted 19 down; they look at what it is, rather than what it's 20 supposed to be. 21 MR. MICHELSON: I assume they also looked at the 22 building. 23 MR. ERICSON: We did not look at walls falling 24 down in an earthquake. I j 25 MR. MICHELSON: So you don't know what the ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
'25279.0 342 RT ,
1 source of vulnerabilities might be, be it the crane falling 2 down or the walls coming down -- 3 MR. ERICSON: We looked at component fragilities
-4 and how the component would fail. The building falling 5'- down on it, we didn't examine.
6 MR. MICHELSON: I think it's a little less of a 7 ' concern in most cases, but other components falling on it 8 would be potentially a source of major concern. So you
.9 have to look around.
10 MR. ERICSON: I think they did that. Pieces, 11 the ability for something to fall on it was looked at, but (m 12 the building itself falling down on it was not looked at. L) 13 MR. HATCH: Okay. 14 (Slide.) 15 And then the seismic modification was just to 16 anchor the tops of these cabinets. 17 I might mention that these were found to be very 18 cheap, inexpensive modifications. All of the -- in general 19 the.special emergency modifications tended to be very cheap 20 compared to some of the internal event modifications. 21 Usually it's tie down something or putting in some new 22 racks, which, in terms of labor and materials, is pretty 2
- 3 cheap.
24 MR. EBERSOLE: You mean all these years they 25 have just had wooden shelves for batteries? [) ACE. FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 3.16-6646
. . _ . - ~ . . - _ . --_ __ - -- .
1 25279.0 343 l 1 MR. HATCH: Well, they have sides on them and 2 all that like a big crib, but, yes, they are made out of 3 wood. I think they have committed to change that in the 4 next year or two. 5 MR. MINNERS: They corrode a lot more slowly.
.6 MR. MICHELSON: From the acid --
7 MR. HATCH: Real quickly, this table shows how 8 the core melt frequencies changed. 9 (Slide.) 10 After the application of both these 11 modifications. We go from 8 E to the minus 5 to about 3 E 12 to the minus 5.
<-)
13 (Slide.) 14 Now, discussed all the individual modifications 15 and will now discuss how these were grouped into 16 alternatives. 17 Again, what we tried to do was'take all the 18 individual modifications and somehow put them togethet in 19 packages with two things in mind. I think we were looking 20 at trying to get a variety of costs and a variety of 1. 21 benefit, and try to somehow spread the range and pick some 22 things that seem to be reasonable. 23 The first alternative that was selected was 24 adding the extra diesel, which brings Unit I up to two 25 (v~) dedicated diesels, and then transferring the DC loads from ACE FEDERAL REPORTERS, INC. j 202-347-3700 Nationwide Coverase 800 336 6646 t
25279.0 344 RT 1 a failed bus to a good bus. 2 The second one was adding a dedicated battery to 3 diesel 1 for a fuel flashing and startup; adding a third 4 cooling water pump, which would give you more redundancy in 5 diesel cooling, and the room cooling issue and also the DC 6 transfer. 7 So, 1 and 2 were strictly internal event 8 alternatives. 9 Alternative 3 takes alternative 1 and adds the 10 fire and seismic modifications. 11 MR. EBERSOLE: May I ask you to consider looking 12 at the fire pump transfer? 13 MR. HATCH: Pardon? 14 MR. EBERSOLE: The fire pump transfer in lieu of 15 the new shutdown. 16 I want to do it because it may encapsulate a 17 defense against a number of things rather than just a 18 single point patch which it is now for -- what was it, fire? 19 It may, you know, cover a number of other options. If 20 you'll make it -- well -- just have a hard look at it. 21 MR. HATCH: We can take a look at it. As of now 22 they don't have the ability to patch it in. But as I 23 mentioned, that's something they are going to be 24 implementing over the next few years. I don't think they 1
' ') 25 had it planned to do yet, but --
ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 80G33&6646
ifh3 _,
,, n
//Z
- y 25279.0. 345 1 :MR. EBERSOLE: You know, that's not a lot of
~
2 money sin .a Jdiesel'-driven-- f ire - pump. Generally they are
,, . .i 3- J diesel-driven pumps,-truck pumps of some sort, modest 7 #
4' 4* source' power, probably competent to withstand earthquake, _5 if ou can put them on a foundation it can stand that. R 6L . MR.,CATTON: *Quite cheap. 7 MR. HATCH: They already have a diesel pump 1 8 . on-site. '"j It 's a matter of getting it tied-in and deciding
- 9. -where youiwant to put the piping --
10^ MR. EBERSOLE: I would guess it would have the 11 versatility-to cope with a lot of things. p%J 12 MR. MINNERS: I hope you didn't hear Mr. Hatch 13 say t$at he would look into that. J 14 , MR. EBERSOLE: I thought I did.
~
15 'AR. MINNERS: I think the Staff would have to 16 say I don't know whether it's worthwhile -- 1, ,
^
17 , MR. EDERSOLE: I would like to hear the Staff
.18 " respond to that. request.
'r. >i - >
I9 MR. MINNERS: I guess we'd like to have you 20 .think about this, Jesas -- 21 MR. EBERSOLEY Sure. I would not want to have 22 you do his unthinkingly. It ought to take one minute. 23- MR. MINNERS: We can do more studies. The 24 purpose of these studies is to address whether we need a 25 dedicated heat removal system and not whether to evaluate [
. ACE-FEDERAL REPORTERS, INC.
I 202 347 3700 Nationwide Coverase 800 336 6646 ;
25279.0 346 gT 1 whether there are some things that could be done to plants -- 2 MR. EBERSOLE: In a way you are right, Warren, 3 it's a patch. I would rather go along without it. 4 MR. MINNERS: I couldn't promise we'd do that 5 study for you. 6 MR. CATTON:'10n the other hand, your other 7 approach, the cost, $84 million or something, that almost 8 precludes it out of hand. Whereas the fire pump approach 9 would be quite cheap and would accomplish the same thing. 10 MR. MINNERS: I think you can draw that 11 conclusion without doing the study that Mr. Ebersole (~)
%/
12 suggests. 13 MR. EBERSOLE: Well, it borders on an approach 14 to what the GSA R-2 is going to use, called the -- what is 15 it -- UPPS system. 16 MR. MINNERS: I don't think that Sandia is 17 asserting that these fixes that they have are the best 18 possible fixes that you can have for this. We are doing, 19 basically, an example of what you can do in patch fixes 20 versus what you can do with a dedicated system, to 21 illustrate, more on a generic basis, we are going to 22 extrapolate this to a generic situation. Doing more and 23 more studies I'm sure you can fix up these plants, but 24 that's not the point of the studies. (3
\m) 25 MR. EBERSOLE: The boiler has a unique advantage, ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 33H646
, , 7-s, ,
25279.0 347 1 if you open the valves it's just a pot with fuel and< water 2 ;in-it. It.'s just a pot and all you h' ave to d$ i,s keep it 3 full, and I think one ought to take advantage of that 4 simple concept. 5 , MR. MINNERS: That might be true, but I would 6
' say that there's probably enough information now to make
.7f the decision of whether we would go ahead and require 1 8 ' dedicated heat removal system or whether we will make 9 people go back'and do more reliability,-value impact.
10 studies and get rid of the vulnerabilities, whatever they
.11 are, in the best, the cheapest way.
12 If the. licensee thinks that the fire pump is a 13 better way of doing it than what we suggested, that would 14 be a study that.would come on the implementation phase. I 15 -- think to do that kind of stuff now is just going to drag 16- .this thing on and on. 17 MR. EBERSOLE: Sure. It's just a good patch. 18 MR. MARCHESE . Before we -- 19- MR.:MINNERS: We'll note that as a repair it's a l l 20 go'od patch. 21 MR. MARCHESE: Someone made the comment you 22 could achieve the same thing by adding a diesel-driven fire 23- pump'as-you:can by adding a dedicated system. I'd have to 24- .strongly' disagree with you.
.25 ' l MR. EBERSOLE:. I agree with you. I would, too.
~ ~
't
. ] ACE-FEDERAL REPORTERS, INC.
' 202-347-3700 Nationwide Coverage 2 336 4646
] ,
I l 25279.0 348 ,
. 1 It's just a good patch.
2 MR. MARCHESE: There's a whole host of things 3 that come up that's difficult to quantify that the 4 dedicated system prevents and mitigates. 5 MR. EBERSOLE: You are talking about boilers? 6 The boiler is just a pot full of fuel. 7 MR. CATTON: Andy, I didn't mean to imply you 8 use the same pump, but a diesel-driven fire pump is cheap. 9 Put one in that's dedicated to this other purpose rather 10 than the system you have here for $84 million. 11 MR. EBERSOLE: Andy, to no stretch of the 73 12 imagination should you put PWRs and boiler in the same box. ' O 13 This looks entirely different if you try to do it with a 14 PWR. , 15 MR. HATCH: Okay. Alternative 4. 16 (Slide.) 17 This was essentially the same as alternative 2 18 which was the diesel cooling water pump and dedicated 19 diesel and DC tre09fer. Again, ncW plus the special 20 emergency modifications. 21 Alternative 5 was the add-on system. I'll say a 22 couple of words about how this system differs somewhat than 23 the PWR one and its intent. 24 We decided early in the program to use NUREG (} 25 2883, decay heat removal study done at Sandia a few years t l ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 336-6646
'~
25279.0 349 RT 1 ago, use the designs from that study as our baseline DHR 2 system. For the PWR this included a dedicated injection 3 and secondary cooling water trains. 4 In that report, though, for the BWR, they ended 5 up with something less than " independent" for their add-on 6 system for the boiler. 7 Essentially what we have is a low pressure 8 injection system with heat removal capability but a system 9 that is dependent on RCIC operation for about two hours. 10 What happened was the contractor who did the 11 cost estimate decided, along with Sandia, that, one, c, 12 putting in a high pressure system to take care of accidents 13 where you remain at high pressure was very, very expensive. 14 So they decided to go to a low pressure system with 15 automatic blowdown to get to your low pressure pump. 16 When they decided -- after they decided on that, 17 they found that to size the pump and heat exchanger and the 18 diesel for that system, that you could have -- those things 19 would have to be huge. You'd have to have a heat exchanger 20 300 feet tall or something, very big system, in order to 21 allow a pressure system to successfully remove decay heat 22 at the time of an event -- 23 MR. EBERSOLE: That's the ancient isolation 24 condenser, isn't it? /~D 25 MR. HATCH: It isn't what it looks like, but v ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
25279.0 350 BRT O 1 perhaps it would serve the same purpose. Based on this, 2 the fact that they'd have to have a very big diesel, pump 3 and heat exchanger to remove decay heat at time zero, they 4 decided to go with a system that would initially depend on 5 RCIC operation for two years. RCIC would come on, get the 6 pressure and temperature down a little ways, and then at 7 two hours, they could then design a much smaller system, a 8 system that would be more easily -- more easily fit into 9 the plant that they were looking at, and they went that 10 route. 11 MR. EBERSOLE: So this is not open cycle boiling
- s. 12 to atmosphere? This is closed? ,
( s 13 MR. HATCH: This is closed. 14 Since we went with NUREG 2883 as our baseline 15 design we have this dependency in our alternative 5, add-on 16 for Quad Cities. 17 MR. EBERSOLE: May I ask, was the basis for 18 having that the basis that you had core damage prior to 19 invoking this mode of operation? It must have been, 20 because there's no radioactivity in the primary coolant to 21 begin with. 22 MR. HATCH: They were specifically asked to look 23 at add-on systems of a totally independent nature. 24 MR. EBERSOLE: Without going to any root
/ ; 25 examination of why?
ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-6M6
...___m..,__.m, ,. -. _ _ _,
25279.0 351 RT 1 MR. HATCH: I'm not familiar with -- 2 MR. EBERSOLE: The primary coolant in the boiler 3 is clean to begin with, and if you keep it that way, you 4 don't need to confine it. 5 MR. HATCH: I'm not sure why they didn't address 6 a different type of system. 7 MR. EBERSOLE: This bothers me, not to have the 8 whys, to see the blind, ferocious approach to a design. ~I 9 don't understand that. 10 MR. ERICSON: The rationale of the original was 11 closed cycle cooling. Very simple. 3 12 MR. EBERSOLE: Somebody handed it to you and you
- gs 7J 13 had to go.
14 MR. ERICSON: Nobody has told me yet that I can 15 open the pot to the atmosphere and let it boil. 16 MR. EBERSOLE: Along at GSA R-2. 17 MR. ERICSON: I know what they are proposing. I 18 don't see anything in the regulations that let me do that 19- today. 20 MR. EBERSOLE: You can because that's been 21- invoked on Limerick,. Grand Gulf, on a host of other plants 22 as a terminal operation prior to cooldown.
.23 MR. MARCHESE: Well, I think we've pointed out l
24 in the past that there's several -- we've looked at this () 25 concept and there's several flaws associated with it that I ACE-FEDERAI, REPORTERS, INC. 202 347-3700 Nationwide Cove age 800 336-6646
. . . . . .~ . . .
25279.0 352 BRT 1 don't think the Staff has finally blessed one way or 2 another. As you know blowing down into a saturated 3 suppression pool has some pool dynamics associated with it-4 that have not been thoroughly examined. 5 In addition, there is environmental conditions 6 generating containment, and it's questionable for the older 7 plants, at least, whether or not the valves that you are 8 going to be venting containment with can operate in that 9 environment. 10 And, third we have the problem, is the Staff 11 going to allow containment venting for decay heat removal,
- 12 in which case you would be. venting containment considerably
~
13- below the ul'timate f ailure pressure of containment. Event 14- containment to prevent catastrophic failure is one thing, 15- but for decay heat removal I think it's much lower in 16 pressure, and it's questionable whether or not the Staff is 17 going to allow that. 18 MR. EBERSOLE: I think what needs to be done is
-19 get the Staff to integrate on the matter of Grand Gulf, 20 Limerick, and about four or five others in their big PWRs 21 that haven't have this'in their procedures.
22 MR. MARCHESE: Its failure to prevent ultimate 23 failure of containment -- but for decay heat is another 24 matter.
^T 25 MR. WARD: Do you want to resort to this --
(V ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 336 W4 ...-.._.__.y.-.-...._-. , m , . m. m-- ..m , _- , - , - , . _ . - . ._. , - ,,
25279.0 353 RT 1 MR. EBERSOLE: Prior to core damage? 2 MR. WARD: Not prior to core damage. In a more 3 probable situation. 4 MR. EBERSOLE: It's not chosen, it's a terminal 5 method to avoid core damage. It's the last stage. But the 6 question is, what price do you pay? And it turns out it's 7 not much, apparently. Anyway, I think there's a 8 considerable integration problem in the Staff to come to 9 grips with it. 10 MR. MARCHESE: There's probably not a price 11 penalty for the newer plants, but to backfit this to the
,~ 3 12 older plants I think there would be. In terms of what it N) 13 buys you in reduction of core melt frequency and risk, we 14 are not sure.
15 MR. EBERSOLE: That's right. I agree. But I 16 just suggest it is worth some inquiry. 17 MR. WARD: Steve, I lost track now of the 18 definition of what you mean by the add-on system. Is that 19 a system that picks up at two hours? 20 MR. HATCH: Yes. 21 MR. WARD: That size system. 22 MR. HATCH: When you are looking at some of the 23 tables in the back, alternative 5 is an add-on, two-hour 24 add-on with a RCIC. In section 10, I believe, and some of l (j 25 the final summary tables, we talk about alternative 6, ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
25279.0~ 354 RT 1 which is the add-on, completely independent add-on, which 2 does not have that RCIC dependency. And we've given some 3 . nutruers to try and see how much an independent system would 4 buy you as opposed to what is in cur base case. 5 MR. WARD: Question? 6 MR. MICHELSON: This add-on system is really. 7 dependent on a system which already is somewhat vulnerable 8 to fire, flood, et cetera? We haven't done a thing to the 9 vulnerability of RCIC, and without RCIC you don't have an 10 add-on system. It won't work because it may catch up with 11 you too fast. - m 12 (Discussion-off the record.) U 13 MR. MICHELSON: This is probably not the best 14 time. Let's have a break for lunch, come back at 1:10 15 please. 16 (Whereupon, at 12:10 p.m., the meeting was 17 recessed, to reconvene at 1:10 p.m., this same day.) 18 19 20 21
- ~22 23 24 g- 25 ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwidc Coverage 800-336-6646 6_ - -- - _ _ _ _ 2.__ . _ _ _ _ . _ _ . . . _ _ . . _ _ . _ _ . . - . ~ _ . . _ ~ . , . . ..,-..mm,g., . . . . , , ...r.- . - -
25279.0 355 BRT O 1 AFTERNOON SESSION (1:10 p.m.) 2 MR. WARD: Mr. Hatch, we interrupted you in 3 mid-syllable, so if you would just pick up where you left 4 off. 5 MR. HATCH: I believe I had just finished 6 defining what went into the alternatives for Quad Cities. 7 (Slide.) 8 This is essentially the same type of figure that 9 Wally had. 10 MR. WARD: Except you put the Xs in -- 11 MR. HATCH: Yes. It shows how the alternatives 7 12 address the vulnerabilities. The internal vulnerabilities, ( 13 again, were the diesel local faults, diesel fuel flashing, 14 diesel cooling water pumps, and/or DC loads, circuit 15 breaker loads and DC CS -- BC to the ECCS logic, fire, and 16 then seismic was the battery racks and the AC cabinets were 17 found to be vulnerable. 18 As you see, the first couple of modifications 19 were strictly internal event-oriented; 3 and 4 we have both 20 internal and special emergency modifications; and 5 is the 21 add-on. Although it's really not quite correct. 22 As I discussed, the add-on has this two-hour 23 RCIC dependency, which limits -- that's what I needed to do, 24 I needed to talk more about the dependency -- the baseline ('l 25 add-on that we looked at has a two-hour RCIC dependency. s_/ ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
'25279.0 356
- O' BRT l- Therefore, the add-on as we have analyzed it, really does 2 not affect fire.
'*3 All the fire scenarios occur immediately so.that,
' 4- really, while an add-on has potential to address the fire 5 vulnerability and some of the seismic ones, in our base 6- case We have that extra, really should not be there. And 7- in some.of the seismic sequences, the add-on also has not 8' ~e ffective.
9 We'll see this lack of effectiveness for a
-10 dependent add-on in one of my later slides.
11' In some of the tables in the back and in the 12 report we talk'about an alternative 6. All that is is the V ' add-on without this RCIC dependency. 13 We have included some 14 sensitivity calculations to see what it. buys yo,u to get rid 15 of this RCIC dependency, and I believe.it's a factor of 30 16~ or 50 percent change in some of the' cost / benefit 17- information when you remove that two our dependency. 18 MR._ WARD: Change in the cost / benefit 19 information? You mean the cost goes up by 50 percent?
~02 MR. HATCH: No. The alternative looks 50 ,
1 21 percent better-if that dependency is removed. The dose -- 22 -do'llars per man-rem-goes down by a factor of about half-or 23 so. 24- MR. WARD: I'see. It's 'in the table. I have it.
'7
/ -25 (Slide.)
c O ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nanonwide Coverage 800-336-6646
%__--___..~.__,, , _ .. m . . . .. ,.._,~--_,......~,~.,.m_~__
25279.0 357 RT 1 MR. HATCH: The next slide gives you a broad 2 brush of the results. It splits out the core melt 1 3 frequencies for the internal and special emergencies, for 1 4 the base case and then what occurs for each of the 5 alternatives, 1 through 5. 6 For the internal floods and external floods, 7 lightning and wind, those special emergencies were not 8 found to be large enough or of a nature which warranted 9 modification, at least in our analysis. So essentially for s 10 most of them the base case number is essentially the same 11 number, all the way through except for some of the
,s 12 alternative 5 -- the add-on system does affect some of the I )
13 special emergencies, those where you do not melt 14 immediately. 15 One of you mentioned earlier that with this 16 dependency, the add-on, the effectiveness of the add-on 17 would be reduced, and that is entirely correct. For those 18 sequences where you don't have RCIC for the two-hour period 19 the add-on is not effective at all. 20 For sequences where you do not -- you initially 21 have core cooling successful and you have that RCIC, or 22 perhaps another system to initially depressurize or remove 23 heat, then the add-on is effective. And you can see the 24 core melt frequencies after applications of the 25 modifications, we've gotten some reduction but we've never ( }) I ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 804336-6tA6 _. _ ~ .- - - - - -
.~ - _ - - - -
25279.0, 358 RT 4 1 .been able to get below -- we have not, with these
-2 modifications, gotten below the 10 to the minus 4 barrier.
'3 I think perhaps one of the reasons for this is that we are 4 working with an older plant and some of the modifications 5: kind of bring it up to the same level as some of the plants, 6' some of the newer plants.
7 We are.getting to the point with some of the-8 modifications that we-have two diesels for the unit, more 9 reliability on the DC electrical side, giving them -- 10 looking at things that some of the newer plants already
- 11 -have. That's perhaps one of the reasons why the core melt 7_.
.12 frequencies don't change more than that. ,
V 13 ' MR. MARCHESE: Steve, you should probably point 14 out that if you were to put' alternative 6 results up there, 15 for example, the total you show on number 5, 1.1 times 10
-16 to the minus 4, if you were to show the.same total for
- 17. ' alternate 6, the number goes to 2 times 10 to the.minus 5.
18- That gets you below the 10 to the minus 4 barrier. 19 The total for alternative 6 is 2 times 10 to the 20 minus 5. 21 MR. HATCH: Yes. So for that case we can-get an 22 order of magnitude reduction, just about. 23 MR. WARD:- Is this -- the Point Beach numbers 24 that we were told this is not total core melt but it was {]} 25 only about half of it. Is that approximately the same ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-336-6646
._. . .-_-~,m_. , _..,~ -._. . ... . ~ . . . _ . , . . . _ .
. _7,. - .. . . _ . . , _ . .
25279.0 359 1 story for'Ouad Cities? 2 MR. HATCH: I don't know what the peccentage is 3 but certainly, leaving out ATWS -- ATWS would probably be a 4 20'or 30 percent increase -- increase this by 20 or 30 5 percent. Certainly.some of the other things we have 6 ' excluded from this analysis would up these numbers. 7- MR. DAVIS: Excuse me, Mr. Hatch, I don't want 8' to be picky, but one of your conclusions is that the plant 9 has decay heat removal vulnerabilities, and that these 10 vulnerabilities can be counted. Now, the reduction in core 11 . melt probability doesn't seem to justify that conclusion.
~s 12 You haven't been able to counter them below the proposed V
' 13 safety goal level and none of the al ernatives do more than 14 'about a factor of 2 to 2-1/2 reduction in the core melt 15 probability, which is well within the error bounds of the 16 original number.
17- I would suggest you take another look at that 18 conclusion. 19 MR.-HATCH: The other way to look at this,~and. 20 we'll get-to it, is in the cost / benefit framework. Indeed, 21 there we had marginal cost-effectiveness -- marginal, if
- 22. not cost-effectiveness for most of the alternatives, 23 although for one case we did_get down below the $1000 per 24 man-rem.
25 It depends on how you want to look at it.
- . ACE-FEDERAL REPORTERS, INC. 1 202-347-3700 Nationwide Coverage 800-336-6646
- -..=..,.n.. _ . - -. . . . ~ . . , . - .
_ ._ _.-: __ : n =_w -
25279.0 360 BRT O 1 Certainly if you start looking at uncertainty ranges that 2 can fuzz up a lot of this, which we have not done to a lot 3 of extent. 4 MR. MARCHESE: You have to remember the baseline 5 core melt frequencies are very high. If you had a delta of 6 1 times 10 to the minus 4 core melt frequency, that is very 7 significant. Even though the number is still high, a delta 8 of 1.4 times 10 to the minus 4 core melt frequency is very 9 significant. 10 MR. DAVIS: Perhaps. Your uncertainty is plus 11 or minus a factor of 5. None of these alternatives even 7-- 12 approach that amount of uncertainty.
~'
13 MR. MINNERS: I don't like that way of looking 14 at uncertainties. That's saying that when you get out, if 15 it's a factor of 5 higher or a factor of 5 lower, the 16 probability that the answer down there is the chue as the 17 central estimate, which is not true. There is a wide 18 uncertainty, I agree. I understand what you are saying, 19 but I mean, we all have to play with uncertainties. 20 I think the way the game is best played is to go 21 along with the estimates, always recognizing in the back of 22 your mind that there are large uncertainties associated 23 with it. You can't go on making your decision saying hey, 24 we'll take the high side of the uncertainty, because you i 25 are more likely to be wrong there than you are with the (^J ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
25279.0 361 RT 1 central estimate. 2 MR. bBERSOLE: Yesterday we had a lecture of the 3 potential contribution of common mode failure as a discrete 4 feature in its own right. Did these studies have the 5 benefit of that, which we heard was as much as a factor of 6 107 7 MR. HATCH: We did include common pumps and 8 common valves, and in the initial study we had diesel 9 common modes and battery common modes -- 10 MR. EBERSOLE: When you get to these plants in 11 essence they are composed of pumps and valves and that's
, 12 about it, and something to make tho water move. The valves, 13 we don' t have any reliability for how the valves operate 14 under duress, and what sort of adjustment factor do you put 15 into the valve reliability to counter this, if anything?
16 MR. HATCH: We have a valve common mode number. 17 MR. EBERSOLE: Not necessarily common mode, but 18 the intrinsic weakness of valve, and the fact that they are 19 never tested in their full duress load condition and then 20 over the life of the plant they are swung back and forth on 21 zero load and said to be still operable when, in fact, you 22 don't know whether they are or not. 23 MR. HATCH: We don't have any mechanism to 24 incorporate that in the data at this point. ('~3 25 MR. EBERSOLE: The degrading torque output of is/ ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80 4 336 6646
l 25279.0 362 i BRT ' O 1 the motor or gear train is not accommodated in the lifetime 2 test of 20 to 30 years. Nobody ever knows if the valve 3 will do what it was supposed to do, even if it ever did 4 originally, in the context of closing against a dynamic 5 load. 6 The new valves may. 7 MR. MICHELSON: There's a lot more to common 8 mode than just whether the valve was designed to do the job. 9 The next question is, is it maintained to do the job, is it 10 properly adjusted and so forth. That's part of common mode. 11 I don't know how you incorporate it. 12 MR. HATCH: I guess we try to incorporate that ( )
~
13 by -- we have a test and maintenance number developed. 14 MR. MICHELSON: Where did you get the data? I 15 have never talked to anybody that's developed the data base 16 yet so I don't know how you get your numbers. Have you 17 developed a data base to incorporate things like torque 18 switch, and switch adjustments -- 19 MR. ERICSON: One more time: We are using IREP 20 and ASEP data; we didn't generate any on our own, Mr. 21 Michelson, at all. 22 MR. MICHELSON: They didn't address this issue. 23 MR. ERICSON: We are using the generic data. 24 MR. MICHELSON: So the probability of some of {} 25 these valves may be unity under load on conditions -- you ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationside Coverage 800-336-6646 w ~ ,wn . _ . . . - - -
c:0 ; l 25279.0 363 JRT . 1 -just don't know. ;
-2 MR. HATCH: -As with the PWR, we had our 0 3 contractor, United Engineers and Constructors, take a look n -
- 4. at these alternatives and come up with preliminary. designs 5 and estimates of the costs to install.
0; 6- (Slide.) 7- This is the results for the five alternatives. 8 They' range from.about $6 million to over $80 million for
.3 ;
9 the add-on. These are one-time costs for the installation. 5 10 ~We then also have an associated yearly 11- operations and maintenance cost which would be incurred 12' yearly throughout the lifetime of the plant, and then a 13 occupational exposure, we also cranked ~in at the $1000 per 14 man-rem, evaluated it an $1000 per man-rem.- And many of 15 the alternati.ves did not have any. occupational exposure at [ 16 all. They are done outs ide the reactor building or in q. 17 areas'that.aren't any type of radiation' field.
,.. - 18 ' MR. WARD: Could I askiyou a-question, Steve? I
! ;l ' 19 guess it-probably won't.make.much difference, but it seems '1i ir (L. 20 to me a little strange.- You use $1000 per man-rem for .7
- 21: occupational dose,-but on the other' hand, for off-site 22 consequences, we've heard that;$1000 per man-rem is kind of
, . 12 3 ' .an extravagant surrogate for:all. sorts of things that might 24 ' happen off-site. It seems extremely inconsistent. h i! 25 MR. MINNERS: It's probably a low number for , c - i. L u y, ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 80 4336-6646 u _ , . _ . w. _ _ _ _ . . _ .
.._m -- _ _ _ _ ___ , . _ . , sm y ,. 7 .,. , s _
25279.0 364 BRT-y
\_/
1 occupational exposure. I think most utilities will tell 2- you that occupational exposure to them is about S2000 or 3 S3000 per man-rem. 4 MR. WARD: So they take economic penalties to 5 avoid at about that rate, you say? 6 MR. MINNERS: It's all economics. 7 MR. HATCH: I will mention that the occupational 8 exposure for the add-on is something that limits the overall 9 effectiveness. The value -- in the value impact realm, 10 costs are impacts. The costs of the modifications or 11 alternatives are positive impacts. The cost of
.,.. 12 occupational dose is a positive impact.
(_)- 13 Averted costs such as averted power replacement, 14 clean up, investment, averted costs are' treated as negative 15 impacts and are subtracted down-the load, which I'll get to. 16 Values are equated with dose,-and averted doses. 17 I'll try to explain that a little more in my upcoming 18 slides. 19 MR. CATTON: Your alternative 5 seems to be 20 awfully expensive. 21- MR. HATCH: Yes. It's kind of interesting. One 22 of the interesting results out of that analysis is what 23 that cost -- 24 MR. CATTON: What did? (3 25 MR. HATCH: Primarily it's not the amount of V ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646 , - ,- _ m . _ - -. _ ...
- _ _._. - . ~ ._ _ - - . . . - , - - - . - - - - . , - -
-w
a: l L25279.0 365 1- pumps or va'lves. ~ It's the amount of piping that has to be 2- laid'and how close you can situate this dedicated building 3 to the' unit it's going to serve, tended to drive the cost. 4 For Quad Cities, the building had to be offset 5 from the. plant site somewhat -- I don't know how many 6 hundreds of feet of piping it was, but that more than-7 anything, drove the cost up. 8- To kind of compare it, for the Cooper plant, i9 -which is a very similar type of BWR as Quad Cities, they
-10 were able to locate the add-on building much closer, and
-11 even though they added an extra high pressure pump to do 12- away with this RCIC-dependency, the add-on cost for Cooper ,
~
' 13 was about $60 million.
14 MR. CATTON: Even $60 million seems awfully high. 15 Is-this safety grade?
?
.16 MR. HATCH: Yes.
s 17 MR. .CATTON: Why?
'18 - MR. HATCH: Seismic 1 building and a lot.of the
= 19 cost.is land -- labor of laying that pipe and digging, 20 putting the pipe runs underground or.whatever they are
~21 going to do. -That's a major expense. Perhaps when-Frank 22 talks later on this afternoon -- I think one of our UEC guys 23 is going to tm here -- he can elaborate on that. But it's 24- not so much the components that go into the system, it's
() 25 how much piping has to be routed around. That seemed to be
- ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336 6646
25279.0 366 BRT
/~ % .
N)
~
1- the major sensitive feature to the cost. . I 2 MR. CATTON: It still seems awfully high. 3 (Slide.) ; 4 MR. ERICSON: I might point out, Ivan, that that 5 number is consistent with everything we heard in Belgium, . 6 Germany, Netherlands; an add-on of any kind is going to 7 cost you anywhere from S750,000 to $100 million. That's 8 been their experience also.
-9 MR. MICHELSON: Does that include the loss of 10 the down time and so forth?
11 MR. ERICSON: In our case it does. As Steve has 12 pointed out, in most instances, United was able to do the n-13 engineering such that they could-do it in normal outages, 14 double shifting, that sort of thing. 15 MR. HATCH: It was assumed modifications 16 wouldn't be mandated to be done at a certain time so they 17 could spread it 'ver o several outages. 18_ I'll summarize the impacts, then. The total 19 impact is really, essentially the engineering cost plus the 20 present worth of the operations and maintenance costs. So 21 present worth over all the years that you'd be maintaining 22 this modification, present worth back in today's dollars. 23 The averted costs are essentially the present 24 worth of the averted power replacement, loss of investment, 25 and cleanup costs. The costs that you would avert by { ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
, _ - - , , - - - - - - - -.--- _ . .- -., ..- .,n------- - - - -
--3.-___ , : 2 :.-- - = ;_
4 25279.0 367 T 1 installing a particular alternative.
- 2. Present worth, again, from the time of the 3 accident back into today's dollars.
4 The net impact, then, is the total impact minus 5 these averted costs, and both the total impact and net 6 impact are then used in our cost / benefit calculations. 7 I'll show you those in a minute. 8 (Slide.) 9 This table gives a summary of values. All the 10 numbers I'll be presenting on here are central estimates. 11 I believe in the back of your packet you~have some tables f~) 12 that give the variation based'on source term, et cetera. v , 13 The averted on-site dose is based on about a 5
-14 50',000 person-rem exposure to those people on-site. That 15 dose is multiplied by the probability of having a core melt
-16 and then averaged over the lifetime of the plant.
17 The averted on-site dose are the values 18 calculated out of the CRAC code and then weighted by 19 probability, et cetera, to come up with the -- this is the 20 dose over the lifetime, now. Wally was showing earlier on 21 one'of his slides the dose per reactor year. This is the 22 , dose times the remaining plant lifetime. You can then sum 23 those doses and.come up with a present worth of total 24 averted dose, and this is present worth and based on $1000 25 per man-rem. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
25279.0 368 RT 1 MR. WARD: What's your discount factor? 2 MR. HATCH: 5 percent. 3 MR. MINNERS: I would just like to make a 4 comment about the present wcrthing of doses. Usually we 5 avoid that by just taking a ratio in person-rem which is 6 not discounted over the dollars. When you monetize it, I 7 notice sometimes people present worth it. That's got some 8 philosophical questions which I don't know whether it's 9 good, bad or indifferent but we tend to avoid it. 10 MR. WARD: Well, I think avoiding it doesn't do 11 away with the philosophical question. You've got a 7s 12 philosophical question if you avoid it. G 13 MR. MINNERS: It's a question that -- it's not 14 accepted that you present worth health effects. 15 MR. WARD: I guess I'm still troubled by this, 16 the $1000 and $3000 or something that was the occupational 17 dose for the cost of installation. I think Mr. Minners 18 said there's a number -- oh, you are using $1000 for that. 19 MR. HATCH: All the doses are valid at $1000 per 20 man-rem. This is dose, now, not dollars. 21 MR. WARD: I'm going back a little ways. My 22 complaint was about the $1000 you were using to cost 23 occupational dose. 24 MR. HATCH: Both occupational dose and off-site () 25 dose is valid at the same amount. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationaide Coverage 800-336-6646
~25279.0 369 T
1 MR. WARD: My complaint was that the off-site 2 dose, S15000'used for-that is intended to be a surrogate 13 for other off-site costs. So I thought you were 4 ~ extravagant for using $1000 for occupational dose. And 5- Mr. Minners has pointed out' that in practice the utilities 6 actually spend more than than to avoid occupational dose. 7- But I have a philosophical problem there. If that's 8 utility practice, that might be of their own choice, but ' S9 there's scme question about whether the NRC should credit 10 suchEa large number. 11 I mean, if the CRAC code uses, I don't.know -- 12 what does it use for fatality? A couple of 100,000 dollars
-{
13 for. fatality? 14' .MR. HATCH: I CRAC code I don't think puts a
-15 'value on the fatalities. But if you look at the strip 16 report it's $100,000 per latent fatality'and $1 million for
? 17 early fatality. 18 MR. WARD: Why shouldn't you cost these doses at-19 tha same rate? Workers aren't any more tender than-20 neighbors. 21 MR. CAVE: Cave, C-a-v-e, UCLA. Traditionally 22' _the figure $1000 per man-rem has been used by NRC for 23' occupational _ exposure. The extension to statistical deaths, 24 as it.were, arising from accident conditions, is, as () 25 Mr. Minners has said, a convenient surrogate. But where ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33H646 u
I 25279.0- 370 RT'
~
1 you.can see the dose delivered to a specific person you 2 should, perhaps, differentiate between that and the
'3 statistical chance of a very small dose spread over a large 4- population where you'll never be able to say whether a 5 particular person has or has not received that dose.
6 MR. WARD: Well, Lesley, I find th'at a little 7 hard to -- I mean if you are using the central estimate and 8 you really believe the numbers you are making -- you 9 brought up two points. There probably is some de minimus 10 level, but in the off-site dose that's taken into account 11 to some extent by having a 50-mile or some radius. I don't
- m 12 know to what extent de minimis is taken into account for 13 the occupational dose. I suspect it isn't.
14 The other thing is you are saying a delivered 15 dose - is worse than a speculative dose. I don't know.
~
16 Maybe you can discount it due to uncertainty or something, 17 but.we are talking about probabilistic risk assessments 18 .that are given as purported central estimates. So a 19 speculative dose can be pretty real, also. 20 MR. CAVE: If we go back to a sort of basic 21 situation, we are saying, if we put a price of $1000 per 22 man rem, that we are prepared to spend $10 million to avert 23 a latenticancer. That, I think, is high compared with the 24 sort of compensation that would be awarded for cause, if a () -25 person were involved in a radiation accident. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
l 25279.0- 371 ('3T V 1 If you go back to the case of those in Nevada 2 . exposed to weapons testing, for example, they had to settle 3 for rather less than that. Eut if you have got a work 4 force that you know you are going to expose, then as a good 5 employer, maybe you give a much bigger margin. That's the 6 way I would look at it. 7 MR. WARD: Well, I guess you could make the 8 argument that it's just the cost of doing business, that in 9 order to keep the work force you have to compensate at a 10 rate of $3000 per man-rem, or something equivalent to that. 11 .Maybe that's hard to do. 12 MR. MINNERS:
'{ } I don't understand what your 13 question is, John. What was your original question? That 1-4 you shouldn't treat occupational exposure the same as 15 public exposure?
16 MR. WARD: No. I guess I'm suggesting that you 17 should treat' it. For public exposure you are saying that 18 the total cost related to.off-site damage is expressed by 19 charging $1000 per man-rem on off-site exposure and that
-20 covers putting farms out of business, and this and that.
21 You are using that same number for occupational exposure 22 . where you are not putting any farms out.of business. 23 MR. MINNERS: When you look at an off-site 24 exposure to the public and off-site property damage, and --
. (~.
( 25 you will see if you monetize public exposure at $1000 per ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33(H5646
25279.0 372 RT 1 man-rem, that.off-site property damage for large, high-risk 2 accidents is usually a small fraction -- I forget the 3 numbers, 10 percent or something like that. 4 When you get to low-risk accidents which are 5 usually more probable accidents but with much lower -- 6 smaller releases and therefore much less exposure per 7 accident, in those cases then the off-site property damages 8 will become, or can become approximately equal to the 9 public exposure times S1000 per man-rem. But in most cases 10 the risk is driven by the low probability /high exposure 11 accidents, in which off-site property damage is a small
>gm- 12 fraction of what the $1000 man-rem exposure would be.
'w) '
13 So, surrogate, you could leave out off-site 14 property damage and not change the number very much.
-15 MR. WARD: That doesn't really deal with the 16 question but --
17 MR. MINNERS: I'm trying to say that we are 18 equating public exposure at about $1000 per man-rem, $990 p 19 per man-rem and we are doing on-site exposure at $1000 per 20 man-rem. That's no difference. 21 MR. WARD: I haven't gotten my point across but 22 it's one of several inconsistencies that we aren't going to 23 'be able to -- 24 MR. MINNERS: I don't understand your point. () 25 MR. WARD: Let's go ahead and not take any more ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-66M
~az -
~25279.0- 373 T
1 time. 2- MR. MARCHdSE: ILet me mention, when you pull 3, this all together for value exact analysis I think the s , 4- results are going to be presented in such a format that one
"( ,
S' can play with the numbers to do his or her own sensitivity 6 studies. If you wanted to use a higher or lower number, 7 the_results will-be ih such a form that you can do that and
~
8 draw..your own conclusions. 9 MR.. WARD: Thank you. 10 MR. HATCH: This summarizes the value-impact 11 analvsis"for Quad Cities. q 12 . (Slide.) v - J13 j The information is the change in core melt
'l4
~
frequency for each alternative, the net benefit of the 15 alternative based on off-site costs and dose alone, and 16< then the dollars per man-rem based on just off-site cost, 17 then the net benefit based on off-site plus on-site factors, 18 ' and then 'the dolla'rs' per man-rem based on that total. 19 The net benefit for off-site costs is the 20 present worth-of the averted dose minus the total impact of 21 the alternative. The total impact' is . the impact that 22 doesn't include the averted on-site costs such as cleanup 23 and power replacement. o 24 The net benefit for the total cost is the t 5
,)
25 present worth of the' averted dose minus the net impact, f ACE-FEDERAL REPORTERS, INC. 202-347-1700 ' NationwideCoverage 800-336-6646
25279.0 374 1 where we have included those avertable on-site costs. 2 I will just point out that for all of the 3 modifications, and here we are showing alternative 6, which 4 is that add-on system without the RCIC dependency. In all 5 of the cases the net benefit to the total costs is negative, 6 which essentially means the -- you are going to be paying 7 more in that you have the potential to avert in costs by 8 implementing the modification. 9 However, if you look at the dollars per man-rem, 10 we did find one of the modifications to fall below the 11 nominal $1000 per man-rem criteria, so it really depends on 12 (3 what combinations of measures you want to look at. If it's s ..) 13 only $1000 per man-rem then this says maybe there are 14 things out there that you can do that are cost-effective, 15 if you don't look at other measures such as the net benefit. 16 In general, the modifications that seem to do 17 well and kind of make sense, the ones that don't add big 18 ticket fixes, the ones which don't include any diesels or 19 pumps, et cetera -- we found for Quad Cities and some of 20 the other studies, if you have fixes that are fairly labor 21 -- not labor-intensive and don't include big stretches of 22 piping and pumps, in some of these cases you can get down 23 to fairly low values based on the $1000 per man-rem. 24 MR. WARD: It's interesting that 6 actually [) v 25 gives you a lower number than 5 does. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
{^ ;
- , -l
-25279.0 \ 375
'l '
dif' 1 MR. HATCH: Remember, here we have -- we do not 2 have the RCIC dependency, so, in terms of -- we drop _about 3 a factor of 2 in the dollars per man-rem so.that's what 4 you'd expect. It gets better. 5 MR. WARD: Did you give us the total cost to 6 that system? MR. HATCH:
~
7 I'm not sure if I have it on a slide. 8 It wasn't very much. In alternative 6 all we've done is 9 added a high pressure pump so the system is.available at 10 time equals zero, it's essentially a RCIC pump, AC RCIC 11 pump that's powered by the dedicated diesel to the system r"Y 12 ( which can provide high pressure water from time zero. ~ %-) 13 As I talked about earlier, the addition of one J4 pump or valve or other cittponents, I think it was about "h 15, Sl.5 million betwea9 <
/ 3 6 for the installation plant, it 16 isn't really a whole lot of money in terms of adding out or i
17 adding more piping or putting the y building a little further 18 away. It wasn't very much an' incremental addition. 19 MR. WARD: Why have you downplayed 6, then? 20 MR. HATCH: I haven't meant to downplay it. N
-21 4It's just that in our baseline case we were taken in our 22 ban,eline what was defined in NUREG ,2883, one of the heat '
23 ' decay studies done at Sandia a few years ago. In that 24 . baseline they used a RCIC dependent add-on system and we g n' ( ,) . 25 .reaUized in the study that's perhaps not the optimum way to l 5 ACE-FEDERAL REPORTERS, INC. ; l k- 202-347-3700 Nationwide Coverage 800-336-6646 _ =%
I 1 I . 25279.0 376 i 1 go. If you are adding an add-on it's probably better to l 2- have a totally independent system. 3 MR. ERICSON: Mr. Ward, when we started this 4 study we agreed we would not spend a lot of money 5- redesigning such, particularly the add-on systems; we would 6 use what was out there.- And it was when we got into the 7 analysis that we realized buried in the fine print of this 8 2883 was this two-hour dependency that we did not cost-9 earlier on. 10 MR. WARD: I thought you said the opposite, 2883 11 included the nondependent system? s 12 MR. ERICSON: No, it's not. 13 MR. WARD: I understand. 14 MR. HATCH: Certainly from the cost perspective
~
15 it's not very much more cost 1y to go ahead and add that 16 high pressure capability and the other BWR analysis we have 17 been doing, we have done that. We have included that high 18 pressure pump-in the baseline case. 19 MR. EBERSOLE: If you had looked at an old plant
- 20. with an isolation condenser before the MBAs had gotten a 21 hold of the business, how would it have looked as you
'22 analyzed it for this condition?
23 MR. HATCH: I'm not really sure how it would, 24 Jesse. They had advantages in one area but d,isadvantages () 25 in others. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336-6646
. . ._m.._ . _ - 1 c ..
i 15279.0.' 377 a f'
- 1 1- (Slide.)
2 .As with the PWR, we took'a look at an alternate
.3 _ approach t'o the consequence calculations where we ran-the
-4 'CRAC code without interdiction or decontamination, 5- essentially. _There's an evacuation of the people around
[6f the plant.but everybody is' sort of left-where they are. ~ 7; These~are: raw populations, dose conditional on 8 an accident occurring'so they do'not have the probability
-9 t of c' ore melt: cranked .in at this point so their dose is
, 101 -given in an accident. You can see the second lines are-the 11 cases without' interdiction. -I have given'it for the-upper , 12' bound, central and lower bound estimate and you--have-an 13 idea of what kind of increases we-are talking about.
'14 - Factors of 20 or more for some-of the dose increases.
15 (Slide.) 16 Tnis slides shows how it. translates into the 17- cost / benefit. If-you-carry that assumption through you can
'18 note very quickly_that your net benefit for doing some of 19 this modification.is now positive. You'd be gaining ' more 20 than you'd be putting in by doing some of these -
21- modifications. -Even for the add-on, if you have the ~ 22 totally independent add-on, we are still showing.a total 23 positive. net benefit. And with -- also, the dollars per 24 man-rem.is very favorable when you compare it to the $1000
'5-2 per man-rem criteria and essentially it's because the dose 3- ACE-FEDERAL REPORTERS, INC.
202-347-3700 800 336 6646
. , - . , . . - . . - - - - _ _ . , _ . _ - __ _ . _ ' . _ Nationwide Coverage
l l
'25279.0L 378 l
.l- goes upLby.a factor of 20. The averted dose goes up.
(Slide.) 3- .I'll quickly summarize some of the results from 4 LQuad Cities. 5 AC blackout was the most important sequence,
- 15 . with the batteries giving. out in aboutLfour hours. It 7 contributed about 60 percent to the overall probability of
.8 core melt.
9 On the whole, we found Quad Cities-to have a
'10 pretty: good physical redundancy of DHR systems.. They have
- 11. a lottof pumps and methods of getting water in. It just 112 tends-to be the electrical. things that get you into trouble.
).
-13 LThe fact that you .only' have .three diesels for two units,
-14 .you have only two diesel trains for two units -- those type-
.15 'of considerations dominated the internal event analyses-and 16 .the'overall results.
17 Seismic-events dominated the special emergencies, 18 and I would-like to mention a couple of things about the. 19 special emergencies.
'20 Again, I think I said that on the whole the 21- special emergency modifications were pretty cheap. .We 22 -found that adding an anchor here or support there or 23 wrapping some cables to be pretty cheap modifications, as
~24 opposed to some of the internal event modifications.
$()I 25 I also point out that the special emergencies ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
l l l l l 25279.0 379 T-1 were really of a different accident class than the internal 2 : events. The internal event sequences were dominated by 3 . station blackout, again, where you initially succeed with 4 some equipment and you fail four or five hours down the 25- line. Much of the special emergencies tend to be of a 6: nature where everything fails right at once and w'ithin 30 7 minutes or so you are having core melting. 8 So, the special emergencies really were of a
- 9. different class of accidents than the internal events.
- 10 MR. EBERSOLE: I notice a line there that says 11 " effective venting containment minimal." I would like to 93 12- examine that. -It's minimal because your makeup system is
-(J' '
13 still electrical power dependent. You have a heat sink 14 here that you can't use because you don't have a water 15 . source, won't do you any good.- So each of these lines is l'6 interdependent on what you do somewhere else and the 17 statement doesn't convey that information to you. 18 Containment venting would be valuable if you had source
-19 feedwater.
20 MR. HATCH: I was going to caveat that. Most of 21 the credit for that was given for the long, 30-hour 22 sequences where you have time to do lots of other thinas. 23 'For the earlier time sequences you need both venting and a 24 water supply to take care of you and if you don't have one, 25 the other doesn't help you. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 336 4646
25279.0 380 P'RT
- \)
1: MR. DAVIS: I have another problem with that 2 conclusion. In section 10 you say containment venting was-3 found-to have very little effect on core melt probability 4 or risk. But in that same paragraph you said that, you say 5 that you don't consider the possibility of venting after 6 coal melting because'of the uncertainties of radio nuclide
.7 scrubbing in the suppression pool. My understanding is, is 8 after melting, after core melt has occurred to prevent the 9 gross releases from arising between the failure. You said 10 you didn't consider that scenario but you do conclude it 11 does affect risk. Am I missing something in that r- 12 conclusion?
L]s 13 MR. HATCH: I guess we were hampered somewhat,
- 14. and I guess.maybe we should. change some of those statements.
15 At the-time we did Quad Cities they had not submitted their 16 procedural guideline package for General Electric's EPGs, 17 . essentially-how they are going to implement venting at. 18- their plant, which entails a number of the specific' things 19 you need to know in order to assess what venting can or
-20 cannot do for you, and, in particular, what pressure are 21~ you going to vent at.
22 In the beginning it seemed all the plants were 23 going to vent at'twice designed. Now I think minimum RCIC
- 24 is going to vent at lower than twice designed for a variety
)- 25 of. reasons. We didn't have the specific information for ACE-FEDERAL REPORTERS, INC.
' 202-347-3700 Nationwidc Coverage 80 4 336-6646
w 25279.~d 381 r)3T L Quad Cities to really determine how venting may or may not 2 help you, and so our interpretation of how venting can help ' 3 you perhaps is limited. Perhaps some of those statements 4 in section 10 are a little strong. 5 MR. DAVIS: I will agree with you it may not 16 affect core melt, but I think I won't agree that it may not
. L-- 7 _ affect risk. There are now some good data on suppression
,8' cool scrubbing, even down to the temperature. I n6tice 9 that that' conclusion doesn't appear on all handputs so I 10 washopingmaybeyouhadpickeditupandchgdgedyourmind.
- 11. MR. EBERSOLE: I disagree, it gd s affect core 12 - melt-because it offers the critical thirig we need, a heat
{} 13 _ sink, which is the thing we have. It'gives us an exit for 14 thermal release. All I have to add for that is a source-of 15 water to that to make up our loss, in which case I'm done. 16 MR. HATCH: That's the problem we face because 17 in the earlier sequences you don't have that source of 18- water. 19 MR. EBERSOLE: Right. But it's. easy to get. 20 MR. HATCH: The last thing I would like to 21 mention, I didn't go into detail in this talk but I do have 22 a discussion in the report, is Quad Cities as a two-unit 23 plan. While we did not look at Unit 2 specifically, we did 24 have to model parts of Unit 2, such as the second DG train () 25 and the swing diesel which can supply either unit. One ACE-FEDERAL REPORTERS, INC. 202-347-3700 ' Nationwide Coverage 800-336 66 4
9 25279.0 382 1 observation we came up with is that there are a lot of 2 scenarios, that we have only looked at how it's affecting 3 Unit 1, but in actuality both units could be in dire 4 jeopardy given those combinations of failures. Ideally it 5 would have been nice to look at both units simultaneously 6 and assess the plant site core melt frequency given some of 7 the accidents we've gone through. I found that to be kind 8 of an insight. For the earlier plants that really have 9 those shared dependencies between units I think it's 10 something that should be kept in mind. 11 MR. EBERSOLE: When you looked at a multi-unit w 12 plant like that, did you conclude that if one unit went to G 13 a full degraded state, the management could stay and work 14 the other one? 15 MR. HATCH: Yes. That's a good question. 16 MR. EBERSOLE: On what basis? The control room 17 may have been designed on that basis, just the leakage and 18 direct radiation exposure has not been designed on a full 19 integrated condition. 20 MR. HATCH: It's a good question but we didn't 21 look into it. 22 MR. EBERSOLE: Some of those gases will reside 23 in the auxiliary building in which they will accumulate 24 rather than being dispersed, and the people will be exposed r's ( ,) 25 who have to stay rather than leaving the house. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6646
F. . 25279-~0'
. 383 T
.1 MR. HERNAN: That's what is supposed to be a TMI 2 fix, Jesse.
3 MR..EBERSOLE: Did they do it for the concept -- 4: the loading up the. aux building before releasing it to 5 atmosphere? You know the products of core melt? 6 MR. HERNAN: Probably not but I don't really 7 know that'much about it. 8 MR. HATCH: Are there any other questions? 9 Thank you. 10' MR. HERNAN: I wanted to say that the procedures 11 'in place are for containment -- for gross containment (~1
- V, 12 failure, which I think is a little different than what 13 Jesse said.
14- MR. EBERSOLE: They are not based on core melt 15 at the moment? 16 MR. HERNAN: At the moment. That's correct. 17 MR. MICHELSON: They'd have to reach high 18 pressures before you'd consider venting? 19 MR. HERNAN: 200 to 250 times. 20 MR. MICHELSON: Can they even open the vent 21 valves before they reach those kind of containment 22 pressures? 23 MR. HERNAN: I don't know. 24 MR. MICHELSON: How can you get enough torque to C) (_ 25 open the vent valves? ACE-FEDERAL REPORTERS, INC.
-202-347-3700 Nationwide Coverage 800-336 6646
25279.0 384 1 MR. EBERSOLE: Can I get a documented defense of 2 that position? Because I need it in DSAR 2. 3 MR. HERNAN: We can provide you with the BWR 4 owner's group emergency procedures guide. 5 MR. WARD: Jesse, let me remind you that -- was 6 it last month we designated a subcommittee to look into 7 containment venting procedures? 8 MR. EBERSOLE: I had forgotten that. That's 9 correct. 10 MR. WARD: Now that you remember that -- as far 11 as I know the subcommittee hasn't met and gotten organized.
- 12 MR. EBERSOLE: I can't believe ,this will persist t
13 as a position. 14 MR. WARD: But that's the place to look at it. 15 I'm informed we are honored by having 16 Mr. Mulligan rather than Mr. Cook as a speaker. 17 MR. MULLIGAN: I'll make a few introductory 18 remarks and then get Frank up here. I'm Jack Mulligan and 19 I'm the project manager for the work that United Engineers 20 is doing in support of Sandia's work in the A-45 area. 21 Our primary role is to take the modifications 22 that are being proposed or identified by Sandia, going out 23 into the field and walking those systems down to find ways 24 of installing them in the plants, bringing that data back ,' ; 25 to United, and then providing cost estimates plus a design ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
[25279.0 385 lllf ' 1 description of each one of the modifications, so that for 2 all those things that Steve and Wally and those people have . 3- been presenting here, there is a design document that backs 4 up each one of those-modifications and a detailed cost 5 estimate. And to answer the question we had here before 6f with-regard to the cost-estimates, let me just take a 7 minute here, if I might, Dave, to explain what the basis 8 for those costs are. 9 (Slide.) 10 United Engineers maintains a data base for DOE, 11 we call it the energy economic data base. Some of you may (~v- 12 be. familiar with it. It is considerably -- a large base of
%].-
13- data on new construction plants. We maintain them for all 14- kinds of technologies, including nuclear. Of course, , 15 because of the rise in.the cost of nuclear plants over the 16 years, we have been tasked by DOE to make many studies of 17 -the products in the field, particularly the productivity 18 problem, and we have.a data base associated with just the 19 individual units of production in the field for things like 20- piping and concrete and rebar, in quite a bit of detail. 21 That's the very -- the base for these estimates. 22 We are using what we call the median plant, which is the 23 average -- average plant in the United States. So it is 24 real data that we have collecced from all the plants in the (). 25 United States. We are using the average plant as the basis ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 80033(HE46
l. 1 25279.0 386 RTj 1- forLnew construction type of work. 2' If you were to take the add-on system, for 3' example, the add-on system, the building is. built outside 4 ~ the ' security fence and is more like new construction work .
-5 than~anything else, any of the other modifications.. That-6 portion 1of it is costed just like new cons..ruction for the 7 average plantiin the Unite'd States.
8- The remainder of it, where we have to go in, 9- within the security fence and into the plant and actually 10 make tie-ins, we use'an approach which is our standard 11- approach-in our backfit work that we do, which.is to take j3 11 2 the base' rates and based on health. restrictions, radiation
\J.
13- levels, congestion, accessibility, and the data that we i 14 have' collected in.the field -- and we have done this, as I 15 say, by sending our engineers and cost estimators to the 16 field, walking ~down'the systems, taking Polaroid ^ 17 photographs and recording any data that was necessary, 18 talking to~the people in the plant, and come back and we do
-19 the estimate and we mark up those base numbers, new 20 construction numbers, with a series of enactors depending 21 on the complexity of the particular job.
22 If we are tying into the primary system inside 23 the containment in a high radiation area, you may find the 24- base productivity numbers marked up by a number up to a () 25 factor of 6 over new construction. And that tags right i ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 33(Hi646
- - _ _ _ _ - , _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ . . _ _ . . . . , _ _ . _ _ _ _ _ - _ _ _ ~ _ _ - _ _ _ _
4 7 - e "125279.0 i 387 W 1 falong with actual experience in the field.
- 2 'After_we~ complete the estimates for each plant, V3' what we do is we go -- these are detailed estimates. We i
4 will' roll up the costs into commodities. For example,.all
'S . piping ~1ess than 2;5 inches, all piping greaterithan 2.5 15 : < inches,-safety class,.nonsafety class and what have you..
7 And we.go back and check those with roll-ups for the same 8 kinds'of. commodities in other data bases for other plants 9 -that we have in-house,-just to make sure that the final
'10 -numbers thatcwe come up with, that we can rationalize those
-11: .with respect to what is actually happening out there in 12: plants.in the field on backfit work.
]J t
13 MR. EBERSOLE: Can I ask--a question? You say 14- :the multiplier is something like 6? 15 MR. MULLIGAN: It could be anywhere from a
'16 factor of-1 up to 6, depending on the complexity of the ,
17 particular area. You don't take the whole estimate. We go 18 through line by line. In a particular piping installation 19= in a particular: area where you have got a high radiation
- 2:0 area, - it's hard to get the pipe down into that space and 21 people : have to_ do a lot of welding in odd positions, things 22- like that,-you could find up to a factor of 6. But it 23 could be down, you know -- for example, on these add-on
.24 buildings, most'of the cost is in the new construction i
(]I :25 Lportion of it which is outside the fence. So most of it is l ACE-FEDERAL REPORTERS, INC. 202-347-3700 ' Nationwide Coverage 800-336-6646
25279.0- 388 T. 1 marked up at new construction rates. 2 MR..EBERSOLE: Do you have a corresponding 3 multiplier.for backfit construction where you have to rip 4 .out what was there and put in something different? 5 MR. MULLIGAN: What we put into the cost 6 estimate was line items for rip-out. 7 In addition, when you go back and look at the 8 cost estimate there's a section associated with rip-out and 9 then another section associated with all the installation, 10 so that's included in it. 11 MR. EBERSOLE: I figured this would be a factor (~g 12 of about 30 to rip out and replace with something similar; (_/ 13 is that about right? 14 MR. MULLIGAN: I haven't handled them but we 15 don't handle rip-out by factoring up the new construction 16 cost. We actually put the rip-out cost in there and then 17 mark up the new construction numbers. 18 MR. CATTON: How do those new construction 19 . numbers compare with, say, fossil fuel numbers? 20 MR. MULLIGAN: Much higher. 21 MR. CATTON: Like what? 2? 3? 4? 22 MR. MULLIGAN: The productivities? I don't know 23 offhand. I'd rather not say. I don't do much work in the 24 fossil end and I'd rather not compare. () 25 MR. EBERSOLE: Is that due to the OA record ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 33H646
125279.0. 389 1- ; keeping constraints and so forth? 2 MR. MULLIGAN: Yes. And the fact that safety 3- class work'is: harder to do. JYou'need'to re people.- 4 There's a lot._of problems where you've got to go back and c ,5' 1 rip' out something that1you've already done and replace it. 6 As I say, these are actual ~ numbers based on the 7 average' plant intthe United-States, and looking at it in a-8~ very[ fine basis, you know? For various -- for rebar,'for l9 pipe under 2.5 inches, pipe graded-2.5 inches, safety glass,~
'10 not safety glass,. category 1 concrete, non-category 1 11 1 . concrete, pumps, valves --
L 's 12 4tR. EBERSOLE: If you could diminish these [13 requirements by an.aoequate heat. removal system would you - 14 ' expect.an overall-cost reduction? Significant? 15 MR. MULLIGAN: On new plant? L16 MR. EBERSOLE: Yes. 17 MR. MULLIGAN: My own opinion is if'you could 18 get everything massed inside containment you'd reduce the
- 19' plant costs considerably.- I think that's still to be 20 : proven but there are certainly a lot of people out there
'21 Lworking'on modular systems.
_22 MR. EBERSOLE: We are working on high quality 23 equipment all over the place rather than a recovery system 24 that will tolerate a degree of practice less critical than tf ) ', 25 :we presently use. l l l ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
25279.0 390 1 MR. CATTON: How much would the cost be reduced 2 if this system was built just to good industrial quality 3 rather than safety grade? 4 MR. MULLIGAN: You are really asking me the same 5 question another way. I really don't know the answer. 6 MR. CATTON: That's right. 7 MR. MULLIGAN: If you would like I can get you 8 that information. 9 MR. CATTON: I would like to have that 10 information. 11 MR. MULLIGAN: I'll send some of that 12 information down to Dave. 13 MR. EBERSOLE: It 'omes up -- would the heat 14 removal pay its way by savings in design? 15 MR. MULLIGAN: It's t matter of opinion but I 16 personally think, maybe if we concentrated on more of that 17 stuff instead of putting all that stuff outside containment 18 instead, with all its seismic supports and all that sort of 19 stuff, we'd probably have done better, but that's still to 20 be proven. That's just my own personal opinion. I 21 wouldn't want to make that the company's opinion or anybody 22 else's. 23 MR. WARD: Do the costs for this' program include 24 financing costs?
,' ) 25 MR. MULLIGAN: The total evaluated cost numbers ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800 336-6646
25279.0 391 T' 1 that you saw include the direct cost, field cost, plus the 2 engineering, the field supervision, the quality assurance, 3 a 25 percent contingency, 10 percent owner's cost, and then 4 I think we considered it was to be about an 18-month 5 construction schedule and there was an escalation and 6 allowance of funds used during construction on top of that-7 also. 8 If you look at new plants today, I think they 9 are running about equal, the financing-cost about equals
-10 the base capital cost on new construction plants. I'm not 11 sure -- do you remember any of the percentages, Frank?
fy 12 -What those are, from direct to evaluated?
)
- 13. Once again, those numbers are numbers based on 14 what's actually happening out there based on a median plant 15 experience.- There's a vast array. There's some fairly 16 cheap plants and some extremely expensive plants. We 17 didn't feel it was fair to take either end so we used the 18 median and also our backfit experience in the areas that 19 were truly backfit type of work.
20 Okay. I think I've covered the first bullet,
-21 there.
22 One of -- just from a terminology standpoint I 23 introduce this add-on' decay heat removal system which you 24 have all been talking about, the add-on system, the () 25. bunkered system which other people have been talking about ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646 , , . , . - . . ~ . . . , . - . . - . . - ~ . - . . . . . . . . . . .-
t 25279.0 392 RT 1 before which is a single frame with a feed and water, 2 primary break-up size for a LOCA, dedicated power and water
-3 supplies, and all in a dedicated building.
'4 MR. MICHELSON: Excuse me, the word bunkered 5 isn't necessarily synonymous with dedicated.
6 MR.-MULLIGAN: Did I use bunkered? I'm sorry, I 7 didn't intend to. It is not bunkered. 8 MR. ERICSON:- He didn't mean to. 9 MR. MICHELSON: This is not bunkered? 10 MR. MULLIGAN: No, it's just a seismic category
- 11. 1 building.
12 MR. MICHELSON: Thank you. 13 MR. MULLIGAN: The other thing we have-been 14 asked to do in this program is evaluate other credible 15 ~ approaches to resolving A-45, and really that's why I'm 16' here today. Mr. Glen Reed has proposed what he's chosen to 17 call a primary blowdown decay heat removal system, and we 18 were asked to take a very quick look at how that might be 19 done and give some rough idea of what that would cost. R20 We felt that we ought to look at an alternative 21 system to that also, which is a high pressure RHR system. 22 We went and developed a conceptual cost estimate 23 for both of them. The cost estimating techniques and data 24 base and everything else that is being used in the base () 25 program were used to put these together, although we ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-6646
, . . . . . .. ~ ,. . . .- - - - . - - - - - . - ..
# 1 25279.0- 393 T.
r -
- 1. 'obviously-haven't had enough time to put as much t-i2 engineering effort into these designs:as we have in some of
~3 the other work.
4 The -idea here was to compare this primary 5' ' blowdown . decay heat. removal system, the high pressure RHR, 6 and then this baseline ADHR.
- 7 .We also, through.another alternative that we'are 8 Tlooking at, in the base program, which is the add-on ADHR, ,
l 9 with RHR' capability so you can come to cold shutdown using !. 10 that system that's out in the add-on building. i' 11 What we tried to do here, once again, was to
- 12- pick some: numbers'that seemed reasonable, look at the costs,-
. 13 look at what problems there were associated with the
- 14 designs. We have~not had the time to optimize.them. I'm 15- .sure we are going to - want to get into some discussion about
- 16 that but they certainly are not optimum systems, but we are
; 117 trying to get some idea of where these systems would lie 18 with respect to where others systems lie in the base 19- . program.
20 Mr. Frank-Cook will present.some basis' designs 21 and.I also have with me the manager of our fluid analysis i 22 group, in case we get into any real tough questions. J .. 23 MR. REED: I might point out the high pressure 4 24 RHR system as an alternative has certain drawbacks and 1: x) tR 25 doesn't'have features of coverage that blowdown does. In ' ACE-FEDERAL REPORTERS, INC. 202-347 3700 . Nationwide Coverase 800 336 6646
25279.0 394 7
- i
'w' 1 .other words, it can't depressurize. And right away it is 2 not as versatile in the total decay heat removal activity.
3 MR. MULLIGAN: We'll get into that. We just 4 thought it was worthwhile. We are not recommending any 5 particular system. I think there's a lot.that has to be 6 done in terms of the risk analysis and what have you, 7 before any decision would be made to go one way or the 8 other. 9 We were asked to develop the systems and provide 10 some costs associated with them and that's it. 11 MR. MICHELSON: Is this being developed from the
~
12 viewpoint of a new plant addition or is it from the
'(~}
v 13 viewpoint of alteration of present plant? i 14' MR. MULLIGAN: Backfit. 15 MR. MICHELSON: Thank you. 16 MR. WARDt Let's see, when comparing it to the 17 l ADHR, maybe we ought to go back to Wally, I guess. Did 18 that have -- does what you were calling the ADHR have small 19 break LOCA capability? It does have? 20 MR. ERICSON: Yes. 21 MR. WARD: Okay. Because as I recall, most of 22 the European systems don't really have a small break LOCA 23 capability. They have capability to make up -- compensate 24 for shrinkage but not really small break LOCA. So this is () 25 really maybe a little bit more. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800-336-6646
.~. .-. . - _ . , --- -. ..
25279.0 395 1 MR. MARCHESE: That's right. They are designed 2 to N plus 2 criteria, so they have three trains of high 3 pressure makeup. 4 MR. COOK: As Jack mentioned, we did develop 5 some conceptual designs for the various options, and in 6 order to make sure we were consistent we started out and-7 established a set of design criteria which we would use for 8 all alternatives. We started with the Point Beach plan and 9 the baseline at ADHR system that was used and evaluated 10 earlier. 11 (Slide.) 12 For the purposes of size of the system we took 13 the heat load at one hour after shutdown. We said we would 14 cool to low pressure, RHR intro conditions in three hours 15 following initiation, total four hours after shutdown. 16 We put an arbitrary service water temperature 17 rise limit of 20 degrees on. I can't emphasize enough that 18 that is arbitrary. When we get further into it you'll see 19 that has a major impact on cost, and it's possible with 20 optimumization, with specific sites, with different 21 arrangements we could certainly let this temperature rise 22 much higher from a technical point of view and reap due 23 costs. 24 And, of course, in keeping with the add-on ( l 25 facility we wanted a minimum dependence on or interface ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
l i 25279.0 396 1 'with the existing facilities. 2 MR. REED: I certainly agree with you on that 20 3 degrees, that's really a crazy temperature rise limit. It 4 also seems to me the three hours is rather fast. I would 5 think that you'd be using something more like five to eight 6 hours. 7 MR. COOK: We just did this, Glen, for the 8 purposes of the evaluation, and they are constant all the 9 way through. 10 MR. MULLIGAN: You can get on low pressure RHR 11 at that point in time and that's really why we did it. 12 MR. EBERSOLE: Did you by any chance look at 13 service water evaporation as an alternate rather than just 14' cooling it? 15 MR. COOK: Not in this study, no. 16 MR. MICHELSON: In this study you are going to 17 use these criteria for looking at this high pressure RHR
~ 18 system as well, I guess?
19 MR. COOK: Yes. 20 MR. MICHELSON: In the case of what Glen is 21 proposing there's less question in my mind, at least, about d 22 fire, sabotage, et cetera, some of the things that are 23 presumably taken care of. Where are your criteria now as 24 to what kind of events you are going to take care of? Is ( )- 25 this going to be free of fire in any other part of the ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 33H646
-25279.0 397 T
1 building? Floods, et cetera? 2 MR. COOK: Yes, as is the base case. 3 MR. MICHELSON: This added DHR, for instance, 4 would be functional, irrespective of where the other fires
'5 were located?
6 -MR. ERICSON: As long as you are still tied to 7 the systems inside containment. 8 MR. MICHELSON: Oh. I understand, yes. 9 MR. COOK: We developed these with independent 10 power source, water supplies -- 11 MR. MICHELSON: So the pipe chasers and fT 12 everything are somehow protected against external instances? U 13 MR. COOK: The chases coming in from the 14 facility are underground, concrete. 15 MR. MICHELSCN: Thank you. I just wanted to 16 make sure I understood what your objectives are. 17 MR. EBERSOLE: May I ask Glen a question? Glen, 18 was there a strict -- well, was the matter of the reduction 19 of AC power and service water, volume need and so forth, 20 considered to come to the conclusion you wanted a 20 degree 21 rise rather than just straight evaporation to atmosphere? 22 MR. REEP: In the thing John and I worked to we 23 were talking about an 80 or 100 degree rise on service 24 water. () 25 MR. EBERSOLE: Was the aspect of simply boiling ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646 ..=
25279.0 398 W l secondary water to atmosphere considered? You know the 2_ grade reduction in water flow requirements, house power, 3 electric power and everything -- 4 MR. COOK: Yes, sir. It was not considered and 5 it does have a great impact on the total dollars we are 6 looking at here. 7 MR. EBERSOLE: It might be tougher to get to a 8 lower temperature that quick. 9 MR. COOK: Yes. 10 (Slide.) 11 This is essentially the primary blowdown system (~} 12 as Glen proposed it. We did make some minor changes which U 13 I'll discuss as we go through it, but essentially we were 14 going to blow down from the high point of each hot leg, to 15 2 flash tank, cooling the bottoms and reinjecting into the 16 cold leg. The ' flash tank will be vented back to 17 containment. 18 One of the changes we made, just as an 19 afterthought almost, was to take that vent line back to 20 reactor coolant drain tank. If in fact it's operable, you 21 don'E mess up the containment, and if it's not oprable or 22 blows out the safety desk you haven't lost anything anyway. 23 At this valve on the flash tank which is another 24 change from, I believe, the way Glen originally intended; () 25 rather than taking the drop back here we felt we could ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336 4646 _ _ - _ . - ~_ -
25279.0 399 T 1 control'it better here and minimize water and steam hammer 2 in the piping between containment and the f acility. 3 As you see.later when I put up the plot plan 4 that was used to develop these costs this run of pipe here 5 in the pipe tunnel is very significant, and in fact in some 6 plants it's much longer than it is here. But we would take 7 the pressure drop here at the valve. We had a transfer 8 pump here for makeup tied to a dedicated water source for 9 small LOCA cooling, the'20 degree rise. You'll see the 10 result of that is over 20,000 gpm, in 24-inch pipe. 11 MR. EBERSOLE: This is what would drive you to ,r~g 12 evaporator cooling. (_) 13 MR. COOK: It would drive most people crazy to 14 put something in this bi7 I agree -- with the two heat 15 exchangers here. 16 When we get into the facilities themselves, of 17 course to power all this .,e have a dedicated diesel with 18 its own starting air, batteries, dedicated ventilation 19 systems, et cetera, in the facility. And we put all those 20 components in with the flash tank and make up pumps, 21 whereas on Glen's original sketch he had them in the 22 service water pump-out. The reason we did that, the 23 baseline plant, that's whare they were located and the 24 service water pump house for the BWRs was strictly a pump A () 25 house, so to keep it apples and apples we put everything up ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide coverase 800-336-6646 l
25279.0 400 1 in the one facility. 2 MR. REED: The reason I separated was, was part 3 can be radioactive and part will not be. And I thought 4 part seems to be close to the source of water and the other 5 part needs to be close to the container. So, separated 6 seemed to be important. 7 MR. COCK: I'm not trying to make any technical 8 judgments on it. We did it simply from a cost basis to 9 make life simpler for us on a cost -- 10 MR. MULLIGAN: And we were using, the base 11 design, one of the ground rules was that work had been done <~' 12 on the add-on decay heat removal system before and we were
)
13 going to use the same design philosophy here as was used on 14 the ADHR. Just so we had an apples and apples basis for 15 the design. That's all. 16 MR. EBERSOLE: Earlier on, in talking about the 17 venting system for the boiler, it was said that Staff would 18 not invoke that until you had experienced core melt which, 19 as far as I'm concerned, negates the value of it. 20 In this case you would do it before core melt or 21 after core melt -- either? Right? Certainly you would do 22 it before core melt? Which gives it a far higher value 23 than if you wait until afterwards. 24 MR. MULLIGAN: It's sized for a steam generator
,m.
( l 25 dryout. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 800 33M646
25279.0 401 1 MR. COOK: This is meant to be as soon as you 2 discover you have no aux feedwater you may start using this. 3 I think those are the real major changes we made 4 just for the purpose of this evaluation. To size the flows 5 and heads, et cetera, we took the basic criteria and just 6 followed through with it to come up with the flows and 7 equipment sizing. 8 MR. WARD: Are your costs going to be based on 9 it being seismic design? 10 MR. COOK: Yes, sir. As Jack -- the add-on 11 facility is a separate seismic category 1 reinforced
-w 12 concrete structure'.
13 MR. WARD: What about the single failure 14 criteria? 15 MR. COOK: Okay. Single failure criteria was 16 not applied on this system except to isolate the primary 17 coolant pressure vent. 13 MR. WARD: Okay.
' 19 MR. COOK: That's consistent with the baseline 20 add-on facility that we evaluated earlier, which was a 21 single train, and we've kept that approach here.
22 MR. WARD: Right. 23 MR. EBERSOLE: Since you have a water throughput 24 here, you are going to be flashing highly borated water. j 25 What will happen? ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800 336-6646
-25279.0:
402
- 1 ,
MR. COOK:- We are going to probably concentrate 2 it in the flash tank. 3- MR. EBERSOLE: How are you going to get rid of 4' that problem? 5 MR. REED: Goes back'with the pumping. 6- MR. COOK: We do have dilution capability here 7 and it continues to recircle. It will be taken to-the 8 condensor box and hopefully not venting too much steam back 9- to containment, although there will be some. 10' MR. EBERSOLE: Is the flashing process 11 . compatible with borated water? 12 MR. COOK I would think so. I didn't really r3 V 13 look at'that.
- 14. MR. EBERSOLE: Would the wells load up?
15 MR. REED: I'd say there would be no problem, 16 completely soluble at these temperatures and concentrations 17 and you'd be pumping the borated bottoms with very little 18- change. 19 MR. MULLIGAN: You have 65 percent of the mass 20 of-the water is coming back through the flash tank, even 21' though, like, you know, 85 percent of the energy goes 22 through the.seem through the condenser, that's only 35 23 percent of the mass. So you've got a lot of water coming 24 back through the flash tank.
.()' 25 MR. REED: I might note that one change that you ACE-FEDERAL REPORTERS, INC.
202 347 3700 Nationwide Coverage 800 336-6646
( 25279.0 403 RT 1 have made which in subsequent issues of our document, which 2 I'll give to Mr. Marchese and Dr. Ericson today, the latest 3 version, you have added a pressurizer spray injection. We 44 just add pressurizer blowdewn for blowing down from the top 5 of the pressurizer as well as the hot legs, and there's a a6 debate going on as to whether there should be a blowdown 7 off the reactor vessel head. I don't think so, but -- 8 MR. COOK: That's right. We did put a spray in 9 to help take down the steam space. 10 So then essentially we followed the same 11 criteria to develop a closed loop high pressure RHR. n 12 (Slide.) l 13 This facility here, essentially, is based on the 14 same kind of temperature change, approximately the same 15 sized flows. The reason for the two heat exchangers in 16 parallel is'this little program we used to optimize and 17 take a rough cut at heat exchanger sizes indicated that s 18 because,it's such a large surface water flow compared to a 19 small primary side flow we need to put two in parallel. We 20 sort of run out of time trying to optimize time, but again
<21 if we relieve that 20 degree delta T --
22 MR. EBERSOLE: Does requirement -- large motors,' 23 valves, pipes all over the place -- doesn't this sort of 24 drive you to an automatic consideration of open boiling on ( ,
) 25 the secondary as an alternate?
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80 4 336 6646
25279.0 404 BRT
/
Q ,, 1 MR. WARD: Hasn't he already answered that, 2 Jesse? 3 MR. EBERSOLE: Well, I know. I don't know 4 really what we are talking about. In pump horsepower or 5 whatever, can you give me kind of a qualitative number as 6 to the nature of the support system that goes with it. 7 MR. COOK: I'll get into that but this is a 8 relatively low head. You don't drop pressure. You are 9 just making up line losses here. 10 MR. MULLIGAN: But in the other system it is a 11 big pump, 12 stages. 12 MR. COOK: I have a slide here to compare the (' 13 various equipment. 14 MR. EBERSOLE: Okay. Fine. 15 MR. MICHELSON: As I understand it, this is for 16 a solid operation. There's no possibility of this working l 17 with two-phase, is there? 18 MR. COOK: It was sized with that in mind to 19 keep it solid. 20 MR. MICHELSON: If you should be losing liquid 21 during this event for whatever reason, losing inventory, 22 sooner or later this would cease to be functional. 23 MR. COOK: Or we'd have to take injection from 24 the alternate tank. r~ . 25 MR. MULLIGAN: If you have noncondensables u.) ACE-FEDERAL REPORTERS, INC.
! 202-347-3700 Nationwide Coverage 800-336 4646
25279.0 405 1 coming out of solution you'll have to vent them. 2 MR. MICHELSON: You may have two-phase coming 3 off. The other system looks like it worked equally well, 4 whether you were blowing down steam off the hot legs or 5 two-phase or liquid. This one doesn't work so well if it's 6' two-phase, or steam. At least unless you design very 7 carefully for steam. 8 MR. COOK: Oh, these are conceptual designs, 9 please, I can't emphasize that enough either. 10 MR. REED: I wish to repeat again that this 11 system has limited application compared to the blowdown 12 system. It can deal with very small breaks where pressure w/ 13 can be kept up, but it can't deal with such things as 14 equalizing pressure on tube rupture and many other 15 situations. 16 1 also should point out that this was the system 17 used on the original PWR, Sl-W, the Navy version. And 18 closed heat exchanger, high pressure systems for decay heat 19 removal had their problems in those days and I predict they 20 will have the same -- similar problems today. 21 MR. MULLIGAN: The Navy still uses them on all 22 their plants. 23 MR. REED: Well, I hope they are better than 24 they were in the early days when they didn't work I) 25 whatsoever at all. And they do involve lots of thermal ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-664o
25279.0 406 1 shock problems. It's hard to make heat exchangers that 2 will stand up to this kind of 80 degree standby situation 3 and then you are going to hit them. 4 MR. EBERSOLE: This is a system which eliminates -- 5 a system, I think, which could eliminate lots of times, the 6 component cooling interface loop direct side to side. It 7 would have to be stainless because in the primary loop 8 chemistry, on the secondary side undoubtedly the water is 9 going to be loaded with borides. It has to be laid up or 10 somewhere or other you have to argue that you can both test 11 it and preserve it against stress cracking. How are you 12 going to do that? , 13 MR. COOK: At this stage in the design we 14 haven't come up with any real details on something like 15 that. We could come up with a bypass around the heat 16 exchangers, induce an artificial pressure water drop, test 17 the operability and then put a flooding wet lay-up type of 18 system in there -- flood water -- 19 MR. EBERSOLE: Yes, just sterile water. 20 MR. COOK: As you noted, the heat exchangers are 21 stainless and they were priced that way as well. 22 MR. EBERSOLE: Right. 23 MR. COOK: The next step after we came up with 24 the conceptual system here was to put it in the facility j 25 and come up with the design. ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 80 4 336-6646
i 25279.0 407 RT. 1 (Slide.) 2 Just as a matter of history here, I'll back up a 3 little bit, this is the baseline add-on facility with a 4 single train aux feedwater and high pressure makeup which 5 we have up in this particular plant. The building is about 6 85 by 55. We've got a diesel with all its auxiliaries for 7 starting, fuel oil supply, the high pressure makeup pump, 8 the aux feedwater pump, electrical capabilities, its own 9- separate dedicated ventilation facility; and it is, as I 10 mentioned, reinforced concrete seismic category 1. 11 In back of that, again you'll see later on the 12 plot plan, is the facility that houses the two separate
)
- -(mx 13 dedicated storage tanks, the condensate storage tank for 14 aux feedwater and borated water -- again, reinforced 15 concrete, seismic category 1. This is the building that we 16 evaluated for-cost in Point Beach.
17 So, what we did to evaluate the other facilities 18 is start _with this and massage it to try to get the same 19 components in, as best we could. 20 What you'll notice in the previous section or in 21 the.following sections, and drawings, plans, is I'm only 22 showing two floors. What essentially we've done is taken 23 this upper Plenumsome area and turned it into a ventilation 24 floor and put the ventilation up on top. The cost, even 25 o( ) though the dimensions of the plant haven't changed, has ACE-FEDERAL REPORTERS, INC.
.T2-347-3700 Nationwide Coverage 800 33H646
25279.0 408 1 risen because the height has risen. But to simplify things 2 here we didn't bother to make that third p,lan. 3 I'll start here. This is that same facility, 4 then, as we modified it to house aux feedwater, primary 5 makeup, and RHR, to get to a cold shutdown. 6 (Slide.) 7 The lower elevation now is strictly equipment 8 with the fluid equipment being separated from the 9 electrical equipment, everything draining back into the 10 pipe chase. Your grade elevation here would then be your 11 diesel and auxiliaries, the control room and offices. As I 12 mentioned the third floor, again, would be ventilation. 13 And this building stayed about the same size. 14 Now, we have made a little effort here to 15 optimize and come up with some good arrangements here 16 without changing building size. 17 For the last two that we just undertook, 18 recently, we went and did a very quick layout that we 19 really couldn't take the time to fully optimize. 20 (Slide.) 21 We felt the components weren't fully optimized 22 either. This being the primary flow down, flash, pumps at 23 the lower level, with the diesel, control room, and offices 24 on top. Again, the building size didn't change, but for
) 25 the purposes of cost we felt we didn't really look at this ACE-FEDERAL REPORTERS, INC.
202 347-3700 Nationwide Coverage 80433H646
25279.0 409 gfT
-1 as close as we could have. There's extra volume up here.
2 There's volume here that's not really used. So from the 3- standpoint of cost we figured that we could reduce the cost 4 of the facility by 25 percent. Another, you are looking at 5 one building size here, the cost numbers actually reflect a 6 20 percent reduction, to try to be fair to these other 7- systems here that came on a littic later in the game. 8 MR. EBERSOLE: The diesel generator, I presume, 9 runs the service water pump system? 10 MR. COOK: Yes, sir. It powers everything in 11 this building plus the pump house. r^j ~12 , MR. EBERSOLE: So you still have the pump house C' 13 as a service system? 14 MR. COOK: Yes, sir. There's a pump house. 15 When I get to the plan -- 16 MR. REED: You mean the. pump house as an intake 17 point, but the pump house pumps -- service water pumps in 18 _the pump house? 19 MR. COOK: Yes. We put the pumps out there. We 20 put the service water pumps out there, cooling, service 21 water pumps. 22 MR. REED: Out in the normal -- 23 MR. COOK: Not the normal, an additional, 24 separate service water pump house. .p). (_ 25 MR. REED: Along the concept I had proposed in a ACE-FEDERAL REPORTERS, INC. 3)2-347-3700 Nationwide Coverage 800-336-6646
25279.0 410 1 separate pillbox? 2 MR. COOK: Right. 3 MR. REED: But rather than putting the diesel 4 out there you put it here? 5 MR. COOK: Yes. But like you said, too, Glen, 6 it's very site-dependent. 7 MR. MULLIGAN: From an arrangement standpoint 8 you have to understand there was a certain design 9 philosophy done to put the auxiliary decay heat removal 10 design together. We couldn't go back and redesign that, 11 and we felt in order to make a fair comparison we'd make c
-12 some changes on what you proposed to make an equal design V) 13 basis.
14 MR. REED: So you really haven't optimized at 15 all the primary concept? You just fitted it in to do the 16 cost analysis? 17 MR. MULLIGAN: We figured whatever we could do 18 to your system we could probably go back and do the same 19 things to the other systems anyway. There are some 20 fundamental technical differences where you can't do those 21 kinds of things, we agree, but we haven't done a lot of 22 engineering work to try to optimize each one of these 23 systems or try to get order of magnitude cost. 24 MR. EBERSOLE: So there are some long large () 25 pipes as well as another building in this picture? ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33m
25279.0' 411 T 1 MR. MULLIGAN: Yes. 2 MR. COOK: Yes. Right here, I'm sorry, I'm just 3 looking really at the base add-on facility. -There are 4- three facilities involved: There's the base, the tank 5 facility,.and the pump house. 6 MR. EBERSOLE: Yes. 7 MR. COOK: And the tank facility here would be 8 half of what the baseline is because you don't need 9 condensate storage tank. You are just talking about 10 borated water. But you still have a pump house and tunnels 11 from the pump house to these facilities, and from these 7s 12 facilities into the plant. This is conttarst for all
- ()
13' designs, as we evaluated here. 14' (Slide.) 15 And then we did a similar thing with the high 16 precsure RHR. 17 MR. MICHELSON: Where did ycu intend to 18 discharge the water that you had to pump into the facility? 19 MR. COOK: Back out through combination intake
'20 discharge structure.
21 MR. MICHELSON: That's a combination structure? 22 You'll discharge at roughly the same point where you intake? 23 MR. COOK: Yes. 24 MR. MICHELSON: Thank you, sir. () 25 MR.-COOK: Again, this is the high pressure ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 80433(H5646
25279.0 412 4"' 1 facility which we felt would shrink some. And then in the 2 cost estimates, again, we have shrunken it another 20 3 percent from the standpoint of cost. But both halves are 4 independent facilities. 5 (Slide.) 6 Everything fits together as we started out, at 7 Point Beach, where we had the main facility with the pumps, 8 the diesels, ventilation equipment. This is the baseline 9 tank facility with both condensate and borated water 10 storage. 11 Then the baseline plant doesn't have a pump 12 house, and we tunnel here and enter it into the aux 13 building and into the centainment. 14 As we grew and changed equipment facility, we 15 still kept-the location the same here. The tank facility, 16 now, for blowdown or high pressure RHR, is half of this, 17 because we've -- we don't need the condensate. But now we 18 have the pump house that will be located up here in its own 19 facility. 20 All facilities, reinforced concrete, seismic 21 category 1. All tunnels are below-grade concrete enclosed 22 until we get into here where we have to step up and get 23 into the building, at which point we come above-grade, 24 again in a seismic category 1 structure. () 25 All piping and components are supported in here, ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
25279.0 413 RT 1 costed out on the basis of a seismic category 1 support and 2 analysis system. 3 As Steve alluded to earlier, and as we found in 4 visiting the various plants during the baseline of the 5 study, is that the location of these facilities relative to 6 the existing facilities is a substantial cost factor. 7 These tunnels are worth a lot of money when you 8 go to put them in, between the excavation, the concrete 9 work, and hanging the pipe. 10 MR. EBERSOLE: Is this a' shared facility? 11 MR. COOK: This is just for one unit.
-12 MR. EBERSOLE: Then you could on a per unit .b,~
- 13. basis split the costs.
14' MR. COOK:, You could realire some cost savings. 15 We did all these -- all these estimates were based on a 16 single unit,-of a twin unit site.
. l'7 MR.-MICHELSON: Is that tunnel necessary as opposed to burying it in the ground?
19 MR. COOK: It depends, I guess, on design 20 criteria that would be imposed on the piping. Whether it's 21 the criteria of. today -- 22' MR. MICHELSON: There's no requirement even 23 there, is there, that it be in a tunnel? 24 MR. MULLIGAN: No, it doesn't have to be.
-(} 25 MR. COOK: Not specifically.
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336 6646
25279.0 414 RT 1 MR. MICHELSON: I wonder why it's in a tunnel. 2 MR. MULLIGAN: Because that's the way the 3 original design was done. It's not the concrete that costs 4 the money. It's the pipe. 5 MR. MICHELSON: I thought you meant the tunnel 6 cost a bundle. 7 MR. MULLIGAN: It's the long runs of safety 8 class pipe. 9 MR. COOK: You'd still have the excavation and 10 piping. 11 MR. MULLIGAN: It's going to be a design c' 12 decision based on the particular site, what 6.he groundwater s / 13 conditions are and all those kind of things, how well you 14 can protect the pipe, what the seismic loads are and 15 whether you can handle them in the underground scenario as 16 opposed to the tunnel. All those things come into play. 17 This is what the base design was and that's what we used. 18 MR. EBERSOLE: May I ask the Staff a question? 19 What is the current position, if any, on a multi-unit plant 20 degrading to the stage and having to look at, for example, 21 all AC powered units? All three units or two units in this 22 case? Is there any -- 23 MR. MINNERS: Are you talking about A-44? 24 MR. EBERSOLE: When you look at more than one () 25 unit at a time, now. ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-336-6646
, }
J 25279.0 415 T-
~
MR. MINNERS:- l I don't know what our position is.. 2 I guess the A-44 position is that. the plants should be able 3 to' survive loss of all AC-power for two to four hours.
-4 MR. EBERSOLE: 'There's no particular credit 5- .given'to self-help from the other auxiliary units, those 6 other units?. Contiguous units?
7 MR. MARCHESE: I think they are treated on an EF individual basis.
- 9' MR. MINNERS: I guess the reason I pause is if 10 'somebody ~ came in and credited the second unit as providing
- 11 power, I guess we-would listen to them. !
{} 112' 13 MR. EBERSOLE: But there's no orderly
. consideration of this at the time?
14 MR. MINNERS: I think-there is an orderly 15 consideration because the A-44 solution is going to require 16- people to go out and make a demonstration, which is they'd 17 have to make some kind of engineering study with maybe some
~18: PRA attached to it which shows that,.yes, for a certain 19 length of time they can provide without an AC -- '
'20 MR. EBERSOLE: The reason I ask this is I was :
- 21. wondering why you put it at this station.
. 22 MR. MICHELSON: In the old designs they have
- 23 ~ ,done a great deal of designing a control room and single
~
- 24. spreading units and -that sort of thing to the point where I
, n (,) 25 expect you have to have separate decay heat removal for ACE-FEDERAL REPORTERS, INC. . . 202-347-3700 Nationwide Coverage 800-336 6646
25279.0 416 1 each unit because they could be in a two unit situation. 2 New units you could do it. 3 MR. MARCHESE: I would just like to make a 4 comment on the separation aspect. 5 You can correct me if I'm wrong but I think 6 those European countries who have decided to backfit a 7 dedicated facility have purposely insisted on criterion 8 that calls for separation of that facility from the rest of 9 the plant structures, because it's the separation in itself 10 that buys you added protection against such events as 11 aircraft crash, gas cloud explosions and third party rw 12 intervention. t . Q,/ 13 If you have the dedicated facility separated 14 from the other structures, one of those events cannot 15 disable all means to cool the plant down. 16 So the separation is expensive but it does buy 17 you added protection. 18 MR. EBERSOLE: Is it not a German practice to 19 cross feed aux feedwater when you have two contiguous units? 20 I know one plant that does it. 21 MR. WARD: I don't know what the newest plants 22 do. 23 MR. MARCHESE: The plants that we looked at were 24 single units. () 25 MR. WARD: For what it's worth, the French, with ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
l 25279.'O 417 1 their very standard design, the latest plants have gone to 2 having each unit entirely separate from the others, except 3 for,-at Palo Alto, for example, it's four units, entirely- _4 separate except for a common intake structure. 5 MR. EBERSOLE: No cross feeds?
- 6. MR. WARD: Where earlier plants had some cross --
7 MR. EBERSOLE: It's a neverending argument. 8- MR. MICHELSON: In doing the design for this 9 system, of course one must be assured, now, that the 10 ' primary loop is in a proper conditicn to be cooled, if, for
-11 instance, this is a high pressure decay heat removal. =r3 12 What are you going to do now on' the design of tg 13 the balance of the rest of the plant now in terms of being 14 assured that you are not blowing down the reactor at the 4
15 same time due to the fire or whatever? Since this is in 16 there, in part, I think, for fire protection -- as an 17 answer to fires. How are you preventing the reactor from l 18' being blown down during the fire so that your high pressure 19 decay heat removal may not work? 20 MR. COOK: The basic philosophy behind the 21 add-on system is to take total control of the plant. 22- MR. MICHELSON: That's right. How are you going 23 to. assure taking total control? Where are the disconnects 24 that prevent the relief valves from opening? That sort of y,~ 25 thing. ACE-FEDERAL REPORTERS, INC. ll02-347-3700 Nationwide Coverage 800 336-6646 l
25279.0 418 RT 1 MR. COOK: I believe they are within the 2 existing facility. 3 MR. MICHELSON: Well, no, the fire is there. 4 MR. EBERSOLE: They have not been asked to go 5 into that detail yet. That's detail design -- 6 MR. MICHELSON: I was asking what the philosophy 7 was. In other words, I think your answer was going to be -- 8 what is your answer? How are you going to handle it? Are 9 you going to go back now and review -- would you go back 10 and review the design and make some provision to asture 11 that for a postulated fire at any location that you can
- , 12 disconnect the relief valves and certain other critical '~
13 items? Is that the approach? 14 MR, MINNERS: I don't think we are going to go 15 back and do that. That's not the purpose of the study. 16 The purpose was trying to get a scoping study of how much 17 it might cost to put in a primary blowdown system or high 18 pressure system. Okay? 19 You have got to realize it's a scoping study and 20 it may need some other pieces of equipment. No, I don't 21 think that we intended at this point to take the scoping -- 22 MR. MICHLLSON: I only meant to ask what it is 23 going to cost to do this other part. It has to be a part 24 of this scoping study, if it were significant. 25 Maybe it's an insignificant cost, to take care ( ACE-FEDERAL REPORTERS, INC.
- 02-347-3700 Nationwidc Coverage 800-336-6646
l l 25279.0 419 1 of this-question. If it'is, then you don't need it in your 2 ~ estimate. Otherwise you need it in your estimate because 3 your system doesn't work without a proper isolation from 4 the event that you are worried about. 5: MR. WARD: Wait a minute. Maybe it's not an 6 insignificant cost but it would be common for each of the 7 four proposals.here, wouldn't it? 8 MR. MICHELSON: It may be considerably more for 9 the high pressure heat removal, which I think depends on 10 single phase removal, than for Glen's case which I don't 11 think it makes a difference. I don't know. He may be able 3 12 to tolerate a blowdown going on and,it will, if anything, 8
)
13 it may help. I haven't thought it through but it n.ay help. 14 In the' case of high pressure decay heat removal I think it 15; has to be prevented. 16 MR. WARD: That will have a small break LOCA 17 - capability. 18 MR. MICHELSON: Yes. But not big break, not the 19 size of several relief valves opening, depending on the 20 number on the system. That's generally one or two relief 21 valves. In each of 10 -- E22 MR. REED: I agree with what Carl is saying. 23 I'm more concerned about -- there's so much difference 24 between an auxiliary feed add-on on a primary blowdown (). 25 system and an a high pressure heat removal system from the ACE-FEDERAL REPORTERS, INC. l 202-347-3700 Nationwide Coverage 800 33Md46
?j5279.0 420 RT' 1 primary. There is so much difference in those systems-that 2 it bothers me.
3- I understand the scope of your work and the fact 4 that you are trying to add something in at a very low cost. 5 I believe th ( are so unique and different that even their 6 layouts, that is in the area, will be lots dif ferent. - 7 For instance, a primary blowdown system should 8' be snuggled real close, and then you use a buried service 9- ' water pipe. You don't have to have tunnels. The burying 10 would be wrapped in buried pipe and the cost differences 11 from these aspects can change radically, based on which of gs 12_ the three you choose. (). :13! I believe you have to go more to optimization to 14- decide on costs. 15- Now,.I understand this is a rough shot, a rough , 16 shot. The other thing that Carl is talking about --
; 1'7 MR. WARD: A rough shot. -It's the only shot we 18 ~have. We have been hearing about these systems. We are 19 trying. to begin to get some kind of engineering comparison..
20- We are getting it. I suggest we hear the rest of it. 21- Would you go on? 1 l 22 MR. COOK: Well, basically the basis for the ' 23 cost, we sized the major components and the piping 24 structures, j 25 (Slide.) I i l ACE-FEDERAL REPORTERS, INC. 3)2-347-3700 Nationwide Coverage 800-33H646
25279.0 421 RT 1 The facilities, the structures with the base 2 case only having the primary 2, the equipment building and 3 tank facility, and if you add RHR, then you have to go to a 4 pump house; and with primary blowdown and high pressure RHR, 5 you end up with all three as well: service water, tanks, 6 - and the equipment, although your equipment facilities are 7 slightly reduced and so is your tank building. 8 . Tanks being another major item. The base RHR 9 had 220,000 tanks. When you add RHR you have added on any 10 tanks. Of course primary blowdown we had the RWST, BWST, 11 which of course we reduce because we had the inventory at _ gm, 12 the. time and then you have the base case with the high V. 13 pressure RHR again. 14 As far as pumps with the base ADHR, you have 15 your aux feedwater pump and primary injection pump, which -- 16 and then, go to RHR we add the low pressure RHR pump. The 17 primary blowdown, the 1 -- let's see -- the makeup pump.
-18 This is the primary injection pump. There is another small 19 makeup pump but we didn't add one on this table. And then 20 the service water pumps at which you start with your normal 21 RHR and flow, goes up considerably. But the primary 22 blowdown and the high pressure RHR. (Indicating.) Again, 23 primary because of the that arbitrary limit. But-any cost 24 penalties here.would be constant between the two.
() ss 25 MR. MULLIGAN: The other thing is the heat load ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-33H646
25279.0 422 RT 1 for that RHR is much lower because that comes on much lower 2 in the game. 3 MR. COOK: Oh, yes. The diesels, you can see 4 the main injection pump is the primary driver in the overall 5 diesel sizing, which affects the primary blowdown a little 6 more than some of the other options, the smallest one still 7 being the baseline case and the high pressure RHR. And 8 then with the addition of low pressure RHR, or the 9 primarily blowdown. 10 Heat exchangers, the base case has none since we 11 are using secondary side evaporation. The RUR heat ,- 12 exchange here is sized about 2000 for the condensor and 13 bottoms cooler for the blowndown. We have two, for a total 14 of about 5400 square feet and 3000 square feet for high 15 pressure RHR. 16 The big difference in the costs, as we'll see in 17 a minute, comes down here in the piping. 18 Whereas, with the base case and the base case RHR, 19 all the piping is 3, 6, 8-inch schedule 40 lightweight wall 20 standard pipe, service water was 18 inches. 21 When we get into these two -- and one of these, 22 by the way, is the change that we made and put control to 23 flash tank which makes all your piping high energy piping, 24 heavy wall pipe. It's sized about the same size for both (~ ' , 25 systems. O ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-66s6
l 25279.0 423 1 If we can resolve it and design to take the 2 control back into containment, we'd see some additional 3 here. The service water piping, we are paying the penalty 4 here again because of the high flows. These two later 5 systems do have the higher, larger pipe sizes. 6 (Slide.) 7 As a result, if we take the direct costs, break 8 them out into some of these major categories, the 9 structures -- it changes a little bit but if you look at 10 structures and piping together, piping is the overwhelming 11 driver. You know, it's anywhere between 25, 30 -- when we
- rx 12 -get into these high energy systems, we are up to alxost 50 i ,)
s 13 percent of cost, over 50 percent in some cases. The 14 problem there being the heat treatment. It's not only
-15 welding its heat treatment. When you are back on the 16 standard wall piping it's just basic welding. That is a 17 significant driver. Instrumentation conduit,-cable are 18 small drivers, but they do change slightly.
19 MR. EBERSOLE: Would you elaborate.on what you 20 mean by " heat treatment"? 21 MR. COOK: When wall thickness gets up-in this 22' range, schedule 1630, 180 pipe got to pre-weld and 23 post-weld heat treatment to relieve stresses and that, in 24 addition to time, takes a lot of man-hours. im (,). 25 If we could drop the pressure down earlier and C:,
. ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6646
25279.0 424 BRT O 1 sooner, and minimize those pipe runs or minimize the design 2 pressure, get back to a schedule 80 pipe, anything below a 3 half-inch wall -- 4 MR. EBERSOLE: You mean right back at the 5 reactor piping someplace? 6 MR. COOK: If we could do it somehow. With the 7 limited time we couldn't agree on how to do it there. 8 MR. MULLIGAN: You have some serious problems in 9 that you are running 65 percent liquid and 35 percent steam, 10 and you are blowing big slugs of liquid down that line, 11 MR. EBERSOLE: Couldn't you do it with some good s 12 ball valves right on the pipe? ~ 13 MR. COCK: The piping down on the stream is the 14 problem. ~ 15 MR. EBERSOLE: I m?an at the primary piping with 16 some good ball valves -- 17 MR. MULLIGAN: You still run into the problem 18 where you are throwing big slugs of water down the pipes, 19 and restraining those pipes and keeping them together 20 during this whole thing is a major design problem if you 21 can do it at all. 22 MR. MICHELSON: You don't know what the upstream 23 fluid condition is, necessarily, either, in the case of 24 Glen's proposal. When you are blowing down you don't know
- ) 25 what you are blowing down. It could be any one of either ACE-FEDERAL REPORTERS, INC.
M.347 3700 Nationwide Coverage 800-336-6646
25279.0 425 RT-1 phase or mixed. 2 MR. EBERSOLE: Well, let me get shot out of the 3 saddle here by saying you depress the primary coolant down 4 in the reactor and then you are carrying steam out. 5 MR. MULLIGAN: Well -- 6 MR. COOK: Yes. We sized this to take over at 7 one hour. 8 MR. MULLIGAN: That's in the core, isn't it? 9 MR. EBERSOLE: No, it's above the core. We've 10 got some new devices now that will tell us where we are. 11 MR. MULLIGAN: I understand. ,\ 12 MR. EBERSOLE: Not ambiguous, shall I say. 13 MR. REED: To clarify in the concept I was 14 dealing with, you would flash inside containment and you 15 would have water and steam and we recognized that. We 16 recognized there would have to be lots of restraints and 17 the flash tank, that little half moon you have in there is 18 quite an impact structure into flash tank. 19 That's why I didn't -- somebody suggested why 20 don't you go underwater to discharge. No way could you 21 ever hold that. You are going to have to come into the top 22 of the flash tank, above water and hit an impact plate. I 23 really think that we should be able to design that kind of 24 a blowdown and that makes your high pressure line situation ( ) 25 go down. ACE-FEDERAL REPORTERS, INC. 202-347-3700 ' Nationwide Coverage 800-336-6M6
25279.0 426 1 MR. MULLIGAN: The other thing it gives you now 2 is you've got some pretty big pipe that you have to figure 3 out how to get out of containment. If you have to put in 4 additional penetrations of that size, what did we figure, 5 20-inch? 6- MR. COOK: I don't think it was that. 10, 12 -- 7 MR. MULLIGAN: You have to be able to find the 8 penetrations to do it. If you don't find the penetrations
-9 to do it -- because if you can't you are in a major problem; 10 -putting another penetration in the replacement power costs 11 far exceed anything we have on this.
12 MR. REED: I recognize that and I know what some (~~)T
\_
13 of the spares are and the sizes are. I recognize that. 11 4 Maybe you are going to be just outside or just inside. I'm 15 not sure. 16 MR. COOK: Essentially all the costs do is sum 17 up from the baseline facility. These last two at this 18 stage of design probably should get a little more work
- 19. before they are used anywhere to confirm this pipe.
20 Because the piping has such a heavy burden on the overall 21 ~. d i re ct . 22- MR. MULLIGAN: I think the only thing you can 23 conclude is that, you know, the primary blowdown and the 24 high pressure RHR are considerably more expensive than the
) 25 add-on decay heat removal systam.
i l ACE-FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 2 336 4 646
- 25279.0- 427 RT 1 You know, I think there's enough uncertainty in 2 what we've done to date that you can't say whether one or 3 the other is more expensive. From the overall control 4 standpoint the high pressure RHR looks like it is quite a 5 bit simpler than the primary blowdown. You are dealing 6- with, you know, for the most cases that we were considering, 7 single phase or the other issue of multiple phase that was 8 brought here.
-9 But you are talking about a couple of pumps and 10 valves and you don't have the problem with trying to 11 balance this whole system. We still haven't really come up c- 12 with a control system that we thought would work on the
-~ %)
'13 primary blowdown. After looking at several different kinds 14 of control systems we are still up in the air as to what it 15 would really look like in order to be able to control the 16 whole operation, to balance the whole operation without 17 slamming down the pressurizer or draining the pressurizer 18 - during the process. And then the delicate balance of 19 bringina the thing down to RHR conditions, to put it on RHR.
L 20 There's a lot -- it's a lot more complex to 21 control the whole operation. As I say, to this-point we 22 haven't figured out how we would control it.
.23 MR. WARD: But it may offer advantages, as Glen 24 points out, in dealing with a wider spectrum of accident
(} 25 conditions. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646 w - Q
i l 1 1 25279.0 428 BRT Y_] l 1 1 MR. MULLIGAN: We are not in a position to say 2 anything about that. I think that really has to be 3 analyzed in the context of the PRA, or whatever, to decide 4 whether that really is worth the extra money. 5 MR. WARD: Right. 6 MR. CATTON: Did you look at the loft blowdown 7 tank when you were thinking of this design? 8 MR. MULLIGAN: No. 9 MR. COOK: No. 10 MR. CATTON: It might be helpful. 11 MR. WARD: How that's designed, you mean? p 12 MR. CATTON: Yes. They have been running it for i ) 13 a long time, for large break LOCA studies. 14 MR. REED: We didn't, but we will. 15 j (Slide.) 16 MR. COOK: As a matter of comparison we have 17 developed what we call the total evaluated cost, the number 18 that Steve was showing, discussing with the direct costs, 19 the indirect costs such as construction management, 20 engineering, owners' costs -- I'm sorry -- construction, 21 management, on-site materials, temporary f acilities, to 22 come up with a total. Then we do have a 25 percent 23 contingency in here as well. It's clear, across the board, 24 10 percent owners' costs, escalation and AFUDC, based on an
; 25 18-month to two-year period to come up with a total number.
ACE-FEDERAL REPORTERS, INC. 202 347-3700 Nationwide Coverage 80 4336-6646
-25279.0 429 RT 1 At this point the directs are just as good as 2 looking at everything else here.
3 Questions? 4 MR. ERICSON: My comment -- Mr. Reed made the v 5 comment earlier, that their original thoughts were that the 6 flash tank, that sort of thing was " snuggled up" I think 7 was your comment, to the containment. We might come back 8 and mention some of the concerns we had about trying to do 9 construction right adjacent to or as part of containment. 10 MR. MULLIGAN: Yes. There is considerable 11 concern trying to dig a hole right next to the containment, r~3 12 and if you have any problems at all that shut you down it's V 13 going to cost you a quarter .of a million dollars a day. 14 You very rapidly '- you are going to shorten up some of r 15 that more expensive pipe. 16 Out of that $20 million, even though you said 17 it's big, it's maybe $2 million worth of pipe you could 18 save. You could very rapidly lose all that advantage by 19 doing the construction right next to the containment and 20 running into operational problems as a result of the 21 construction. 22 s Now, you can say well, that's hypothetical, but 23 I think it's a real concern. I think on the concern of the 24 Point Beach people they wouldn't think of building anything
)) 25 in right snuggled up to the containment. They want it ' I, ACE-FEDERAL REPORTERS, INC.
l 202-347-3700 Nationwide Coverage 800-336 6646 l
25279.0 430 RT 1 right out there where they could put it outside the fence, 2 build that whole thing and tie it in when they could tie it 3- in. 4 As I say, it's worth -- you are right. But you 5 know you've got to balance off, you are losing something by 6 reducing the length of those -- that piping, but on the 7 other hand you could very well wind up -- for example, if 8 you-looked at this, you put this thing down here you wind 9 up making the service water piping longer. So you don't 10 get the full advantage of doing that, you are kind of 11 moving a building in between some expensive piping and not
. s 12 qu'ite so expensive piping, and you don't get the full 13 ~ benefit of it and you could very rapidly lose that 14 advantage if you run into operational problems while you 15 are working close to the containment.
16 MR. REED: Recognizing that, my definition of 17 snuggling is different from yours. My snuggling is at 65 18 years of age, and yours is somewhat less. 19 MR. EBERSOLE: I thought you were going to say 20 at 65 feet.
- 21. MR. REED: Okay. I want to drive home again if 22 I can, and I certainly ask Dr. Ericson and company to.
23 reread the final papers on this idea. Because the primary 24 blowdown system and the closed system has been around for
'(} 25' years and it has its pressurized therm 3458~ shock problems ACE-FEDERAL REPORTERS, INC.
202-347-3700 Nationwide Coverage 800-336-6646 K_ .= _ __
, _ m .,,m,__ _ , . . ,
i J 25279.0 ' 431 RT E 1 and all that kind of thing -- I would like to find out that 2 this system of primary blowdown does a multitdde of things 3 and helps in a lot of areas. 4 Small break LOCA; pressurized ther'al m shock; 5 steam generator tube rupture; ste'am ejenerator overfill; 6 loss of AC, outside AC; loss of' primary natural circulation; l 7 and that can't be discounted. 8 Secondary system: rupture problems, auxiliary 9 feed, lack of diversity. 10 The other thing is, keep in mind when you build
,, 11 on the auxiliary feed system you still must move the heat 12 through the steam generators.
,3 !
~
13 MR. MULLIGAN: You look at the high pressure RHR, 14 okay? It should handle everything except a large break as 15 .long as you can vent the high points in the system. 16 For example, on the B&W plants they are venting 17 i and can decay over to the head and venting the head. As 18 long as you can restore -- vent the high points in the 19 system you should be able to make that -- 20- MR. REED: How do you deal with high pressure 21 thermal shock and tube rupture and steam generator overfill? 22 How does:it. help those cases, I mean? 23 MR. MULLICAN: I haven't looked at those cases. 1 24 Were those listed on his list?
) 25 MR. COOK: I'm trying to remember.
m_.- l ACE-FEDERAL' REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
1 25279.0 432 RT 1 MR. REED: Primary blowdown helps many of the 2 problems that we've had and we have been patching for years.
~3 MR. MICHELSON: What pressure did you have in 4 mind-for the high pressure RHR?
5~ MR. MULLIG AN : It operated at normal system 6 pressure. 7 MR. COOK: Normal pressure. 2250. 8 MR. MICHELSON: 2200 pounds? That's a relief. 9' Thank you. 10 MR. WARD: Any other questions? l'1 Well, thank you. It has been a very interesting j- 12 look and useful comparison of the concepts. (/ 13 Let's see, next on our agenda -- I guess we are 14 back to-item G7 Let's take a break for -- about a 15 15 minute break and come back. 16 (Recess.) 17 MR. WARD: Let's reconvene and our next speaker 18 is Mr. Marchese on the CE/PORV issue. 19 MR. MARCHESE: One of your favorite topics. You 20 specifically asked us for a discussion on how information
'21 developed in-the A-45 program will support development of a 22 staff position on the Combustion Engineering PORV issue, 23 which, as you know, has been a longstanding item.
- 24. (Slide.)
1('V 25 This slide lists the A-45 input that we feel
%)
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 8 & 336-6646 _ . . _ . . .....,.1 .
1 l 25279.0 433 RT 1 will assist us in reaching a decision on this issue. 2 I believe -- I think I alluded to this when we 3 had the meeting last month on the Palo Verde Unit 2 4 licensing. 5 In the A-45 program, all 5 PWRs, including the 6 Saint Lucie, which is a Combustion Engineering unit, are 7 being analyzed with and without bleed and feed capability, 8 therefore we can calculate the change in core melt 9 frequency and also risk of adding bleed and feed capability. 10 MR. MICHELSON: Excuse me, is there going to be 11 a hand-out on this? r--- 12 MR. BOEHNERT: You have it. It's in the initial 13 one Andy put out. First packet. 14 MR. MARCHESE: It's the fifth slide in the first 15 packet. 16 In addition to changing core melt frequency and 17 risk' as associated with adding bleed and feeu capability, 18 we are also going to determine the cost and impacts of 19 addi'ng said capability. 20 The bleed and feed capability is one, as you 21 kn'o'w7~of several alternatives being examined in the A-45 22 program. Based on the value impact analyses of all of 23 these alternatives, a bleed and feed will be compared and 24 we feel that this will provide the kind of input we need to l i
; 25 determine whether or not we select bleed and feed, versus ACE-FEDERAL REPORTERS, INC.
i 202-347-3700 Nationwide Coverage 800-336-6646
i 25279.0- 434 T~ l some other alternative. 2' The next slide, which you really, I think, have 3 seen in the past, but just to give you a feel of-the other .; 4 alternatives that are being looked at'in the program, and
- 5. they include such things as improved procedures, just
- 6. adding PORVs, going to dedicated bleed and feed' systems 7 such as the one Mr. Reed has proposed, high pressure RHR, 8 and the other ones that are listed there. In other words, 9 we will examine this in value impact space, and make a 10 recommendation.
11 Now, in terms of specifically what we are doing
~s 12 on bleed and feed -- again this has been presented several U 13 times in'the past, specifically at the meeting we had on 14 Palo Verde -- we are doing a lot of work on bleed and feed, 15 not onlyflooking at system concepts that were proposed --
16 that were discussed here-earlier, but we are also doing a 17 -lot of work in the thermal hydraulics area to assess the 18 viability of bleed and feed. We are' doing this for a range 19 of PWRs. 20 (Slide.) 21 We are extending this to other plant types, 22 getting a handle on the key parameters such as valve size,
-23 RCIC pump shut-off head, operator action time to bleed and 24 feed, and, again, assessing the impacts associated with I; 25 adding such capability.
v ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80433H646
25279.0 435 1 MR. REED: You talked to the alternative systems 2 and high pressure and primary blowdown system. 3 As far as I'm concerned, and I-have always felt 4 this way, bleed and feed is really a primary blowdown 5 activity. 6 Now, the extent to which you polish that and 7 make it be.tter and less challenging to inside container 8 equipment gets you, perhaps, to the Cadillac, and the 9 Cadillac in the system is a primary i lowdown or a bleed and 10 feed to a flash tank outside which keeps from messing up 11 the inside of the containment. gz 12 So, you can go it one of three ways. U 13 Bleed and feed dirty, messing up the inside of 14 the containment, or you can have what we are calling a 15 containment-supported primary blowdown of bleed and feed, 16 or you can go to the Cadillac. 17 You are costing out the -- what we'll call the 18 messy; that is assured, but it is not very expensive. 19 MR. MARCHESE: I think we'll have cost 20 information on all three different ways of doing it. 21 Okay. And then finally as I mentioned, we are 22 going to compare all of these alternatives in a value 23 impact analysis and make some recommendations. 24 Now, Brian Sharon is not here, but he asked me [ ') 25 to remind you that when we did the CE/PORV study, it was NRR ACE-FEDERAL REPORTERS, INC. ! 202-347-3700 Nationwide Coverage 800-336-6646 l
25279.0 436 RT 1 management's position that adding PORVs to those Combustion 2 Engineering plants that didn't have them, that that was a
=3 minimum. requirement, that it was determined that it was 4 necessary. It was the Staff's opinion that those values 5 did buy you a reduction in risk and was worth implementing, 6 but it was decided to hold off on that pending a decision 7 -on A-45. Once this study is completed that decision will 8 be made.
9 MR. WARD: That decision will be confirmed or 10 reversed. But I think you said, rightly, that there really 11 was a decision made to require PORVs. But then it was
-~ le agreed it would be -- implementation would be held in 13 abeyance until after A-45.
14 MR. MARCHESE: Okay. That's all I have on the 15 combustion. 16 MR. REED: Well, you've got me all excited here. 17 Where do you think we are coming out with requiring PORVs, 18 or bleed and feed, or the dirty primary blowdown? 19 MR. MARCHESE: I think with respect to those 20 Combustion Engineering plants that don't have PORVs, it's 21 NRR's position that that ought to be a minimum requirement 22 that comes out of A-45. A-45 may come out with something 23 more comprehensive than just adding PORVs. 24 For example, we may recommend a dedicated bleed () 25 and feed capability which is more extensive than just ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646 h
25279.0 437 T-1 adding PORVs. 2 .MR. REED: But there's something along the PORV 3 line that's going to happen; is that what you are 4' predicting? 5 MR. MARCHESE: That was NRR's position. That 6 was prior to going to CRGR, but -- 7 MR. MICHELSON: What is roughly the schedule for 8 getting the resolution? 9 MR. MARCHESE: That was my final viewgraph which 10 I was going to come back to later on to wrap things up with. 11 MR. EBERSOLE: Andy, at the risk of intercepting j-] 12 what could- be a better dedicated decay heat removal system L.) 13 and to introduce my own best patch, I think you better look 14 at a low pressure, low flow, 2- to 300 pumping system at 15 .about 200. psi, with diverse pumping water systems to the 16 boiler, and then along at open rating with blocked open PORSs 17 and boiling to atmosphere on a pre-core melt basis, not 18 with this, as I regard assinine position of the owner's 19 group to wait until the core is melting and then do it. I 20 think that is absolutely without logical foundation. I'd 21 just like that to go on the record. 22 Any comments? 23 MR. MARCHESE: With respect to the boilers, we 24 are -- we have taken a cursory look at that system. () 25 MR. EBERSOLE: I know. You said that. We've ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800-336-6646
I 25279.0 438 RT 1 got to do more than cursory.
- 2. MR. MARCHESE: I think it will be ranked along 3 with these others as best we can. But as Warren pointed 14 out, we've got to get on, in terms of getting a resolution 5 put together.- You know, we keep doing studies and studios.
6- MR. EBERSOLE: I invite you to consider in the 7 course of doing this, as I always say, we should.do the 8 relative complications of this versus this thing that we 9 just looked at in terms of cost and complications and 10 liability. It's like an order of 1 to 50; in complexity 11 and cost. 7 sg 12 MR. MINNERS: Are you suggesting, Jesse, that 8',) 13 when we finally get out with a proposed resolution, that we-14 specify a particular system that licensees should install? 15 MR. EBERSOLE: Yes. It's prescriptive, in caps. 16 FROM THE FLOOR: That's against the -- 17 MR. EBERSOLE: I understand that and that's one 18 of my major problems with this regulatory process.
- 19. MR. MINNERS: Okay. That's clear.
20 MR. MARCHESE: Okay. We are now clear to 21 discuss the sabotage analysis. Do you want me to do that 22 now? , 23- MR. MINNERS: Why don't you do that in the open 2:4 session, get rid of it in t:'e open session?
'i 25 MR. WARD: All right. Good.
("!
~^
ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800 3364646
- - .. ,- - . .. ~ - . ~ . - . . . . . - . . . . . . - - . - . . . - - - . . . ..
u
'25279.0 439 RT--
1 MR. MINNERS: So we are not hiding anything. 2~ MR. MARCHESE: Maybe some people want to go home. 3 Should I do it now? 4 MR. WARD: Go ahead. 5 MR. MARCHESE: Okay. Schedule. 6 (Slide.) 7 The main ingredient of the schedule right now is 8 to complete the plant analysis, the seven-plant analysis. 9 Dave Ericson talked about the plant analysis 10 status. Basically, I-think what he said is we can expect 11- 'four plants to be complete by Christmas, and the_ following e- 12 three to be done in about the February-March time frame.
%.j -
13 So that's driving -- that's one of the things driving the 14 schedule. 15 Now, in terms of the main -- the critical 16 . milestones.- 17 (Slide.) 18: This is my best_ estimate on what it is going to 19 take to complete those. We are estimating that we can 20 produce a summary report on all seven plants in about the
- 21 May -- springtime frame. In addition, a regulatory 22 analysis'will proceed in parallel with that effort.
23 As you know the two main documents we need to go 24 forward to CRGR are a technical findings report and a ()- 25 regulatory analysis. We are anticipating that we can get ACE FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336-6646
- . . - . . . _ . . ~ _ . . . . - - . . - . . . . - . . . _ -
25279.0. 440
-BRT 1 those out for Staff comments next summer.
2 Then that starts the internal process of 3 bringir.g this through the approval chain, and the 4 milestor.es are shown here, including getting division 5 concurrence, going to the director of NRR and getting his 6 concurrence, and then finally submitting that to the CRGR, 7 and they'll do their review and provide their comments to 8 us. 9 That comes back. We resolve the comments and 10 then it goes out for public comment, and we are projecting 11 that this thing would go out for public comment about x 12 February of '87. 13 That's all I have on scheduling. Any questions? 14 MR. WARD: Is the schedule constrained by 15 limitations on resources within the Staff -- your ability 16 to contract resources? 17 MR. MARCHESE: The schedule problems that we 18 have been having are partly attributed to, once we provide 19 the documentation to other parts of the Staff, we kind of 20 lose control in terms of the time it takes for them to do fl their thing. That has been a continuing problem with this 22 program. 23 Our contractor resources right now we feel are 24 adequate. The funding that we have available, we feel, is 25 consistent with doing this job. But once it leaves our
')
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 8043364M6
25279.0 441 1 hands and goes into other divisions within NRR, it has been 2:
~
la problem in terms of getting a timely review because they 3 are busy with other things.
'4 MR. WARD: But yet that review seems to have 5 been very important, a lot of important input.
6 MR.-MARCHESE: I think it is very important. 7 MR. WARD: Okay. 8 MR. DAVIS: Excuse me, Andy, I had a quick 9 question. This third item from the bottom, the package-to 10 CRGR, is that going to include your recommendations on 11 which plants.need to be fixed and how they should be fixed? es 12- Or is this just the complete analysis of the plants?
~'
13 MR. MARCHESE: No. We'll have to have some 14 specific recommendations.
~15 MR. DAVIS: Okay.
16 MR. MARCHESE: Just what the resolution is going 17 to-be, we haven't developed that yet. It could be a number 18 of different things. 19 MR. WARD: But rather than what plants have to 20 be-fixed, it might be a set of criteria against which 21 plants will have to judge themselves or something like that, 22 I guess. 23 MR. MARCHESE: Right. Right. 24 MR. DAVIS: That's what I was trying to decide, 25 is what' exactly would be in it in terms of recommendations.
-( } '
ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 80 4336 4646 e- -, --- . . _ . - . - . . .. . . . _ . . - - . - - - -.
i 25279.0- 442 1 MR. MINNERS: I don't think we know that, Pete. 2 .I guess there's lots of questions in what form of 3 resolution. Jesse says ir. can't be in the general form of 4 -do good, that's not going to be enough for him. He says 5 put high pressure system on plant A, and -- 6 MR. DAVIS: The source problem will not be 7 resolved by that time in all likelihood and neither will lh , 8 the safety goals, likely. I just wonder what criteria you
-9 are going-to use to make the decision. But I guess we'll 110 just.have to wait and see.
11 MR. MINNERS: The safety goal rule'isn't that 12 big a problem. Maybe I speak out of turn but I think
'7,). ,
13 people are generally accepting the proposed safety goal 14 criteria that were laid down. 15 All day today we talked about $1000 per man-rem, 16 and things like that. I think there's kind of a consensus 17 in the technical part of the Staff. That's what people 18 compare things to all the time, But it's not that simple. 19 There are lots of other factors to be considered 20 than just the number crunching of cost / benefit analysis, 21 and the next session on sabotage is probably one of the 22 bigger issues that has to be considered. 23 All of the things that Glen brought up, this 24 system can fix a multiplicity of problems, and maybe we {} 25 should include that defense in depth. Should we prevent ACE FEDERAL REPORTERS, INC. 202 347 3700 Nationwide Coverage 800-336 4646
L25279.0 443 RT' 1 core melts as well as prevent public risks. I guess that's 2 the process the Staff has to go through. _That's what the 3 Cooper package is going to provide. It's' going to give 4 some kind of technical resolution, the form of which I 5, don't see what that is at the moment, and then it's going 61 to give a value impact analysis which is more than just 7 number crunching. It is going'to give the rationale of why 8 we think that resolution is the proper resolution for the 9 Commission to adopt. 10 It's going to try to weight all of the
-11 quantitative factors,_all the qualitative factors. It's 12- .not an easy process. We have to figure out how to handle 7-)
G 13 occupation exposure versus public exposure and all those 14 kind of things. 15 MR. EBERSOLE: Warren, would it make sense to do 16 a study based on some fluctuations in the source term as we 17- -anticipate they might come? 18 MR. MINNERS: We already presented some of those 19- results. 1.3 in RSS, okay? 20 MR. EBERSOLE: If it goes down by a factor of 10, 21 things get a lot better, don't they? 22 PIR. MINNERS: Well, yes, as the numbers show. 23 MR. DAVIS: Your sensitivity study shows they
- 24. are sensitive. Now we have to decide what we are going to O 25 u e-4 ACE-FEDERAL REPORTERS, INC.
. _ . 202 347-3700 Nationwide Coverage 800-336-6646 L
25279.0 444 T 1 MR. MINNERS: Public risk is directly 2 proportional to the source term. 3 MR.. DAVIS: It's even more than directly 4- proportional. 5 MR. MINNERS: Actually not, because of the 25 6 rem-interdiction criteria. You won't see that big a 7 difference-in criteria on the public risk because people 8 are now being moved out when they get 25 rem, and with the '1,
! 9 new source term they are going to get out when they get 25 10 rem. If the source term is so small that nobody gets 25 11 rem, then you'll see some difference. But it really isn't 12 that sensitive to source term. I misspoke when I said 13 proportionate. With the interdiction it is not. With the 14 full potential it would be almost directly proportional.
15 What it will. affect-on the noble gases -- 16 MR. EBERSOLE: I hope there will be a lot of. 17 work done on the relative aspects of using these systems in 18 the context of mitigating an entire core melt versus what I 19 regard the better use of them, to prevent a core melt, and 20- then have them in reverse for use subsequent to core melt 21 if you have to. 22 MR. MINNERS: Well, there's economics in there. 23 That to me is largely an economic decision, whether you 24 want to prevent core melts or whether you want to mitigate ()- 25 them. ACE-FEDERAL REPORTERS, INC. 202-347-3700 Nationwide Coverage 800 336 6646
-25279.0 445-T l' MR. EBERSOLE: I guess I can't quite get rid of 2 the economic picture.
3 MR. MINNERS: That should be considered but that
'4 kind of stuff is handled -- if you include on-site damages 5 you automatically include whether you should prevent-or 6 mitigate. If the numbers say it's cheaper to prevent, well, 7 that's the way you go.
'8- MR. EBERSOLE: That's where the big money is, 9 building these facilities.
10 MR. MINNERS: That's right. If you have high 11 core melt frequencies then the on-site costs become very n
. \,1
~12 significant.
13 MR. WARD: I see the United Engineers people are 14 gone but I should have asked one more question. Those four 15 designs, let's see, none of those had any redundant
- 16 capability, as I recall.
17 MR. MARCHESE: No. But we are doing some 18- variations of those systems that will give us a delta on 19 costs of adding redundancy as designed to single failure 20 criteria. 21 Bunkering has been brought up several times. 22 There's been a lot of confusion on what a bunker really is. 23 We also will have a delta in there, on what the cost is of 24 going from a seismic category 1 structure to a bunkered
~(m) q 25 structure.
r ACE-FEDERAL REPORTERS, INC. 202-347 3700 Nationwide Coverage 800-33H646
25279.0 446 1 As you know, bunkered structure is more 2 extensive in terms of the amount of structural hardening 3 than a seismic category 1 structure, so we'll be able to 4 display the costs of doing that. 5 MR. WARD: But as far as single failure 6 resistance and multiple trains, there'll be a delta cost 7 available? 8 MR. MARCHESE: Yes. 9 MR. WARD: Do you have any idea what that is? 10 MR. COOK: I do, but it's not with me. We do 11 have the numbers, but it's not with me. y 12 MR. WARD: Unless there are any other questions, i 13 we'll go into the closed session for the meeting. Let's 14 take a 5 minute break and clear the room. And we'll end 15 the record right here. 16 (Whereupor, at 3:50 p.m., tha meeting proceeded 17 in closed session.) 18 19 20 21 22 23 24
') 25 ACE-FEDERAL REPORTERS, INC.
202-347 3700 Nationwide Coverage 800-336-6686
a CERTIFICATE OF OFFICIAL REPORTER f v.
.i 1
This is to--certify 'that- the ' attached proceedings before e the LUNITED STATES NUCLEAR REGULATORY COMMISSION in- the matter of: NAME OF PROCEEDING: l ADVISORY COMMITTEE ON REACTOR SAFEGUARDS SUBCOMMITTEE ON~ DECAY HEAT REMOVAL SYSTEMS' b i
, DOCKET:NO.:-
-PLACE:
, Washington, D. C.
-_ Q -
J
*"Y'~ "" ' '
DATE: were held as herein' appears, _and that this is the original transcript thereof for the file o'f the United States Nuclear Regulatory Comnission. (sigt) (TYPED Official. Reporter Reporter's Affiliation
t i TASK ACTION PLAN A-45 RESULTS FOR QUAD CITIES ACRS SUBCOMMITTEE MEETING ; i DECEMBER 3,1985 l STEVEN W. HATCH
~
SANDIA NATIONAL LABORATORIES l l O G G
- QUAD CITIES TRANSIENT EVENT TREE INIT REnCT StVS SRYS wfPoR EMLY Mnthf gyg &
En?NT SNTDMb OfEN CLOSE $UPP ECT COND Ecr stowace2nueuence c enR T K M P V D Y 2- E #. l T og 2 TY CK 3 TfE CD 4 TV2 CV s Trat a ; 6 TO CD 1 TP OK r TN OK s TPYE co la Tft'E CV II Tff2E CD at TPO CD IS Try cv sv TM cv is TK NA : e l 9
1 QUAD CITIES SMALL LOCA EVENT TREE INET REACT Ynfot SK YS SR VS EnnV LATE EVENT ~ $NTDMV sufP ortN ClasE ECT RUR ECT ym yg S If V M P D 2 E
# 1M i S o<
Z SE CD , 3 SE CV V SEE CD 5 SD CD 4 SP OK , 7 SPE CD ; t SF2 CV 1 SPEE C9 to SPD CD I! S/A CV l 12 SV CV
--o SK NA :
# 9 se 4
l, i , i QUAD CmES EVENT TREE DEFNTIONS T1 - LOSS OF OFFSITE PONER TRANSIENT T2 - LOSS OF FEEIMATER TRANSIENT' ! j T3 - TRANS1ENT WITH FEEDWATER-INITIALLY AVAILABLE ! T-AC - AC BUS FAILURE INITIATOR ! T-DC - DC BUS FAILURE INITIATOR i S - SMALL LOSS OF COOLANT ACCIDENT K - REACTOR SHUTDONN FAILURE M - SAFETY / RELIEF VALVES FAIL TO OPEN P - SAFETY /REL1EF VALVE. FAILS TO RECLOSE i V - FAILURE OF VAPOR SUPPRESSION i ) D - FAILURE OF EARLY EIERGENCY COOLANT INJECTION j Y - FAILURE OF THE MAIN CONDENSER i
- Z -
FAILURE OF THE RESIDUAL HEAT RElWOVAL SYSTEM E - FAILURE OF LATE EIERGENCY COOLANT INJECTION l l l I l O , O 0
i I - I ! CHANGES SINCE PREUMINARY DRAFT i ADDED PLAWP AND VALVE COWMON IWODE FAILURES l 1NCREASED CONTROL ClRCUIT FA! LURE ADDED SUPPORT SYSTEM INITIATING EVENTS l REEVALUATED OTHER DATA l l RESULTS ! CORE lELT PROBAB 1 L I TY I NCREASED BY A i FACTOR OF 2 i l l O O O
i
. i .
t l j NTIAL QUAD CmES DGGHTS { NO REM 3TE MANUAL CONTROL OF SAFETY / RELIEF VALVE 8 : NO REM)TE CONTROL FOR RECIRCULATION PL2#8 i NO INTEGIATED RElWOTE SHUTDCMN PANEL ONLY ONE 125 AND 250 VDC BATTERY PER UNIT ' PLANT APPEARS TO BE VULNERABLE TO UPSTREAM DAM FAILURES { , . j BOTH SAFETY DIVISIONS ROUTED TlftOUGH ONE CABLE 1 SPREAD 1NG ROCN BUILT WITH TlftEE FOOT VERTICAL CABLE SEPARATION VENT STACK COLLAPSE COULD IIWPACT 4160 VAC SWITCHGEAR CONDENSATE STORAGE TANK 18 NOT SEIEMIC CATEGORY I G ,
# 0
QUAD CmES RESULTS - l ! 21 INTERNAL EVENT SEQUENCES IXNINANT PRIOR TO ' l RECOVERY ! 16 INTERNAL EVENT SEQUENCES DCNINANT AFTER RECOVERY i SEISMIC EVENTS Als FIRES DCNINATE THE SPECI AL i EIWERGENC1 ES e I i 1 9 O G
I i i . HTERNAL EVENT SEQUENCES PROBAB1LITY
- 1. TRANSIENTS WITH LATE EC1 FAILURE (STATION 1.6E-4 (57%)
1 4 BLACKOUTS, TYZE) I
- 2. TRANSIENTS WITH EARLY 1.0E-4 (36%)
ECl FAILURE (TD)
- 3. TRANSIENTS AND LOCAS l WITH NO POOL COOLING 1.8E-5 (6%)
- (SZ, TYZ) a e *
! DCNINANT QUAD C1 TIES ACCl DENT SEQUENCES t ! SEQUENCE IDENTIFIER PROBABILITY WITH ' l RECOVERY l l T1YZ 1.3E-06 i T2YZ 3.4E-06 l T3YZ 4.1E-07 l T1YZE 8.7E-05 t T1PYZE 7.0E-07
- T2YZE 1.6E-08 l T3YZE 4.0E-08 i
T1D 4.3E-05 i T1PD 3.5E-07 ! T2D 2.7E-07 l T3D 1.6E-06 ! SZ 1.3E-07 1 T-AC-YZE 3.7E-05 i T-AC-D 2.4E-06 T-DC-YZ 1.1E-07 T-DC-D 1.7E-07 Note: AlI probabiiities are per reactor-year. l . I O O O
4 i - l ; l i
~
I 1 MOST SIGNIFICANT VULNERABL!TES - INTERNAL
- 1. LOCAL:;-FAULTS OF TWO blESELS l .
! 2. FAILURE OF DIESEL FIELD FLASHING I -3. Di ESEL COOL ING WATER FAI LURE
- 4. FAILURE OF DC CONTROL PONER TO BREAKERS AND ECCS LOGIC l .
e O G
NTERNAL EVENT MODIFICATIONS
, 1. ADDITION OF A FOURTH DIESEL ,
i l 2. DEDICATED BATTERY FOR D1ESEL 1 i i l
- 3. INSTALL ADDITIONAL DG COOLING WATER PLAP
- 4. PROVIDE AUTO TRANSFER OF DC LOADS i
l : l l l e ~ 9 O t
5
^ ~
- FIRE ANALYSIS _
~
VULNERABILITIES CONTROL ROOM FIRES 6.7E-6 CABLE SPREADING ROOM FIRES 5.8E-6
~
1.3E-5 PER R-YR MODIFICATION
. REVISE PROCEDURE FOR SAFE SHUTDOM4 PLR,P OPERATION -
- CONTROL ROQWI FIRES 1.5E-6 CABLE SPREADING ROGA FIRES 1.3E-6
- 2.8E-5 PER R-YR
SEISMIC ANALYSIS SAFE SHUTDOWN EARTHQUAKE .24 G EARTHQUAKE FREQUENCIES
.5 -
1 SSE 1.0E-3 ' l 1 - 2 SSE 2.3E-4 i 2 - 3 SSE 1.4E-5 ! 3 - 4 SSE 5.3E-6 ! 4 - 5 SSE 1.5E-6 U SEISMIC INITIATED EVENTS i S - SMALL LOCA-l T1 - LOSS OF OFFSITE POWER i T2 - LOSS OF MAIN FEEDWATER i i
=
p % 't U- 't i t- _
- h ,
- l i
i SBSMIC CORE MELT PROBABILMES !! LEVEL S T1 T2 TOTAL
.5 -
1 SSE 4.6E-9 2.6E-6 9.3E-7 3.5E-6 ' ]' 1 - 2 SSE 1.0E-5 4.1E-5 5~.0E-7 5.2E-5 ..! i 2 - 3 SSE 3.8E-6 1.5E-5 5.1E-7 1.9E-5 y '
- )
3 - 4 SSE 1.6E-6 3.6E-6 4.1E-9 5.2E-6 4
>4 SSE 1.9E-6 9.6E-7 1.1E-9 2.8E-6 TOTAL 8.3E-5
-l J
e G #
i i SESIMIC-RELATED VULNERABILITIES FAILURE OF BATTERY RACKS FAILURE OF 4160 VAC BUS CABINETS 1 SEISMIC MODIFICATIONS INSTALL METAL BATTERY RACKS ANCHOR TOPS OF ' 4100 VAC BUS CAB 1 NETS k 4 e G G
- \
d q SEISMIC CORE MELT PROBABILITIES WrrH MODS l: LEVEL S T1 T2 TOTAL U h
.5 -
1 SSE O 2.0E-7 6'2E-7 8.2E-7 '
- 1 -
2 SSE 2.9E-6 1.4E-5 0 1.7E-5 i i 2 - 3 SSE 2.5E-6 3.1E-6 2.2E-8 5.6E-6 ! [ 3 - 4 SSE 1.0E-6 1.4E-6 4.1E-9 2.4E-6 i
>4 SSE 1.4E-6 8.0E-7 0 2.2E-6 4 ,
j 2.8E-5 4
~
L e . e
- 1 i
i l f i. l ll QUAD CfTES RESULTS FOR ALTERNATIVE 8 1 Alternative y Base Case 1 2 3 4 5 a h 4 Internal 1.8E-04 7.4E-05 1.3E-04 7.4E-05 1.3E-04 5.5E-05 Se l eni c 8.3E-05 8.3E-05 8.3E-05 2'.8E-05 2.8E-05 4.3E-06 Fire 1.3E-05 1.3E-05 1.3E-05 2.8E-06 2,8E-08 1.3E-05 L Internal Flood negilgible - - - - - - i, External Flood 9.8E-08 9.8E-08 9.8E-08 9.8E-08 9.8E-08 9.8E-08 Extrane Wind 1.4E-07 1.4E-07 1.4E-07 1.4E-07 1.4E-07 1.4E-07 Lightning 1.7E-08 1.7E-08 1.7E-08 1.7E-08 1.7E-08 1.7E-08 2.8E-04 1.7E-04 2.3E-04 1.1E-04 1.8E-04 1.1E-04
- Change in core melt 1.1E-04 5.4E-05 1.7E-04 1.2E-04 1.7E-04 Note: all values are per reactor-year of operation e '
O #
es--4, -L -m + L e II L DEFINmON OF DHR ALTERNATIVES ALTERNATIVE 1 ' FOURTH D1ESEL GENERATOR AUTO TRANSFER DC LOADS . . ALTERNATIVE 2 - DEDICATED BATTERY TO DG 1 TH IRD DG COOL 1 NG WATER PLAVP AUTO TRANSFER DC LOADS ALTERNATIVE 3 SAME AS ALT 1 PLUS - ENHANCE PROCS FOR SSP - NEW BATTERY RACKS , ANCHOR AC BUS CABINETS
- l 1l 1 DEFINmON OF DHR ALTERNATIVES
- o ALTERNATIVE 4 L
SAME AS ALT 2 PLUS l: ' ENHANCE PROCS FOR SSP NEW BATTERY RACKS ANCHOR AC BUS CABINETS ALTERNATIVE 5 ADD-ON DHR SYSTEM ' O i O O G
I E 4 9 [ QUAD CITES DHR ALTERNATIVES I i ALTERNATIVES : i i [ VULNERABiL1TY 1 2 3 4 5 i [ INTERNAL 1 X X X l INTERNAL 2 X X X INTERNAL 3 X X X , INTERNAL 4 X X X X X r FIRE X X X i SEISMIC 1 X X X ) SEISMIC 2 X X X i i i l 9 . 8 8
) ? COST OF ALTERNATIVES ENGR + 4 . ALTERNATIVE INSTALL 0+M . OCC DOSE
- 1 $15.2E6 9.7E4 0
, 2 5.8E8 4.0E3 1 0 - i 3 15.3E6 9.7E4 0 4 5.9E6 4.0E3 0 5 82.4E6 5.3E5
- 1200 -
1 - l
'*y
.. . ~
I
SUMMARY
OF IMPACIS (CENTRAL ES1) L TOTAL TOTAL AVERT NET
- IMPACT COSTS IMPACT ALT ($XE-6) ($XE-6) ($XE-6) 1s 16.5 3.3 13.3 -
2 5.9 1.6 4.3 3 16.6 5.2 11.4 4 5.9 3.5 2.4 ' 5 89.3 5.0 84.3 O O O
1- 'l .
SUMMARY
OF VAI UES (CENTRAL EST) ,! AVERT AVERT PW OF. TOT !! ONSITE OFFSITE AVERT DOSE l! ALT DOSE DOSE ($XE-6) l' 1 123 3053 1.9 i l 2 61 1486 0.9 3 196 4831 3.0 ) 4 133 3263 2.0 .t 5 190 4705 2.9
,j .
- f. .
-1 li.
4 ! >\ *
SUMMARY
OF VALUE-lMPACT ANALYSIS i i DELTA NET BEN $ NET BEN $ i CORE OFFSITE PER TOTAL PER ALT IWELT ($XE-6) P-REM ($XE-6) P-REM i '. I. \ 1 1.1E-4 -14.7. 5400 -11.4 4200
- i. 2 5.4E-5 -5.0 4000 -3.3 2800 l
l 3 1.7E-4 -13.7 3400 -8.4 2300 4 1.2E-4 -4.0 1800 -0.4 700 5 1.7E-4 -86.5 2E4 -82.6 2E4 7 l 6 2.6E-4 -86.3 1E4 -82.4 1E4 e - O O
ALTERNATE APPROACH TO CONSEQUENCE CALCS 50 M1LE POPULATION DOSE RELEASE UPPER CENTRAL LOWER j CATEGORY BOUND ESTIMATE BOUND QC1 1.6E6 1.5E6 8.6E5
*1.3E8 4.7E7 1.3E7 QC2 2.1E6 1.7E6 1.1E6
*1.8E8 5.4E7 1.8E7
- QC3 1.7E6 1.2E6 6.9E5
- *5.5E7 1.7E7 5.5E6 QC4 3.3E5 1.7ES. 8.3E4
*2.OE6 6.OE5 2.1E5 CALCULATIONS WITH NO INTERD1CTION 4
iI il- '
~
- i -
SUMMARY
OF VALUE-lMPACT ANALYSIS
- (NO INTERDICTION).
! DELTA NET BEN $ NET BEN $ . !: CORE OFFSITE PER TOTAL PER ji ALT IVE!LT ($XE-6) P-REM ($XE-6) P-REM I! 1 1.1E-4 +20.4 270 +23.7 210 2 5.4E-5 +11.5 200 +13.2 150 l j 3 1.7E-4 +41.1 170 +46.4 120 4 4 1.2E-4 +32.2 90 +35.8 40 5 1.7E-4 -32.5 940 -28.5 900 6 2.6E-4 -3.4 620 +3.5 570 b e ' O O
,l' i
l I l
SUMMARY
OF QUAD CITES RESULTS >i . i !} STATION BLACKOUT SEQUENCES IMST llWPORTANT ii Il ' QUAD C1 TIES HAS GOOD PHYSICAL REDUNDANCY IN Die SYSTEMS ,, ELECTRICAL FAILURES DCMINATE INTERNAL EVENT ANALYSIS SEISMIC EVENTS DCNINATE SPEC 1AL eke!RGENCIES IE)DS VWERE FOUle VWHICH Mi!ET THE $1000/P-REM CRITERIA MANY SCENARIOS WILL AFFECT BOTH UNITS f . G .
.. e
'l [
- i
.l SHUTDOWN DECAY HEAT REMOVAL ANALYSIS POINT BEACH CASE STUDY ACRS SUBCOMMITTEE MEETING i DECEMBER 3, 1985 WALLIS R. CRAMOND
< SANDIA NATIONAL LABORATORIES O ,
9 O
POINT BEACH SDHR SYSTEM EVENT TREES TRANSIENT EVENT TREE ' LOCA EVENT TREE Ie M L P Q E H. IF M L Di Pi % Ot Hi Hs c . I g
' y,#
I T3 , sten i _ _ I7 _ _ _ _ _ _ { ._ _ + ,
#"O
( sec. oany D
>-n,
" l ssH y stowo. e.
n.o# > E'M
>G n f ! N' y teg I l- - - - - - + I wm
! 6ia , e.o revo h=es >"'"
T,n , i ,
>Len I i
>ec= l l
- -_ _ - 8
>Ech l l I I I __J l TM .
l amo ,
, >ecM l
-____J ,
>GcM e .
G G
1 .I i a 4 CONTAINMENT SYSTEMS EVENT TREE convAnwMeNT fos 7 Acespe#T CM seq g l p/ oveA9Aessone p a. ors <.Tso 68 (same Activity asHovAL l' 6 svaess s S uwFm.W P (R o M e a c H S enRLYFAH*** Tft.A N T l eNT" on. LocA E pg,a n,g $ C0G.2 hGLT '; Aec: 9 iF NT- F L ATE F*" *"* l TeqvsN c s' g g g,Ly p Att u a. E p E = Y, (ci + F)( Cs 4- %.) S S e G , 9 O
~
POINT BEACH CORE MELT ACCIDENT SEQUENCES DOMINANT PROBABILITY PROBABILITY ACCIDENT BEFORE AFTER SEQUENCES RECOVERY RECOVERY SMH{Hj 5.96E-5 5.96E-S 2 T30Hy2 H 2.95E-5 2.95E-5 T MLE 1.13E-5 9.69E-6 5 S MDy2 D 8.50E-5 8.98E-6 2 TyMLE 2.39E-5 6.66E-6 T2MLE 9.64E-6 6.61E-6 T4MLE 7.55E-6 6.20E-S T30Dy2 D 4.17E-5 4.80E-6 TMQHjHj 4.17E-6 4.17E-6 2 T2MLH y 2.08E-6 2.03E-6 T MQDy2 D 5.95E-6 6.86E-7 2 ~ S MXD y 1.79E-6 6.51E-7 2 2.82E-4 1.40E-4 O , 9 O
CHANGES FROM ORIGINAL POINT BEACH ANALYSIS TO CURRENT ANALYSIS
- 1. BASE OPERATOR FAILURE PROBABILITY CHANGED FROM 3E-3 TO 1E-3.
- 2. LONG TERM STATION BLACK 0UT DUE TO BATTERY OR CST DEPLETION CONSIDERED.
- 3. i CONTROL CIRCUIT FAILURES INCLUDED IN PUMP AND VALVE LOCAL FAULTS.
- 4. COMMON MODE OF APPLICABLE PUMPS AND MOVS CONSIDERED.
- 5. TEST AND MAINTENANCE DOWN TIME INCLUDED.
- 6. AC AND DC BUS INITIATING EVENTS CONSIDERED.
- 7. BATTERY FAILURE PROBABILITY CALCULATION IMPROVED.
- 8. ADDITIONAL SENSITIVITY ANALYSIS ON P,0RV BLOCKING INCLUDED.
- 9. ADDITIONAL SENSITIVITY ANALYSIS ON CRAC2 MODEL INCLUDED.
I 10. CHANGES WERE MADE TO THE SECONDARY BLOWDOWN MODEL.
- 11. LOSP INITIATOR FREQUENCY CHANGED FROM 0.100 TO 0.086.
- 12. EXTREME WIND DG EXHAUST STACK VULNERABILITY MOD INCLUDED.
- 13. MODIFICATIONS TO SEISMIC LOGIC MODEL ESPECIALLY FOR THE RWST.
i G , 9 O
'i COMPARISON OF DAS AND TOTAL CM PROBABILITY BETWEEN ORIGINAL AND ; CURRENT POINT BEACH ANALYSIS
- ORIGINAL ANALYSIS CURRENT ANALYSIS DOMINANT PROBABILITY PROBABILITY DOMINANT PROBABILITY PROBABILITY 4
ACCIDENT BEFORE !.FTER ACCIDENT BEFORE AFTER SEQUENCES , RECOVERY RECOVERY SEQUENCES RECOVERY RECOVERY SMHjHf 2 6.4E-5 6.4E-5 S2MH'1H j 5.96E-5 5.96E-5 TyMLE 9.9E-5 3.9E-5 TQHjHj 3 2.95E-5 2.95E-5 TQHjH2 3 3.2E-5 3.2E-5 T MLE S 1.13E-5 9.69E-6 TMQHiHj 2 4.5E-6 4.5E-6 S2MD12 D 8.50E-5 8.98E-6 S2MDy2 D 2.8E-5 3.4E-6 TyMLE 2.39E-5 6.66E-6 T30Dy2 D 1.4E-5 2.2E-6 T MLE 2 9.64E-6 6.61E-6
- TyMQLD y 1.4E-6 1.1E-6 TgMLE 7.55E-6 6.20E-6 T2MLE 1.9E-6 6.4E-7 T30Dy2 D 4.17E-5 4.80E-6 TyMQHy2 H 4.7E-7 4.2E-7 TMQHjHj 2 4.17E-6 4.17E-6 T3MQHy2 H 3.2E-7 3.2E-7 T2MLH y 2.08E-6 2.03E-6 T2MQDy2 D 2.0E-6 3.1E-7 T2MQDy2 D 5.95E-6 6.86E-7 TyMQD12 D 4.3E-7 1.9E-7 S MXD y 2
1.79E-6 6.51E-7 LTSB 1.98E-4 3.56E-5 TOTAL P(CM) 2.55E-4 1.49E-4 4.87E-4 1.82E-4 e O - O
CORE MELT PROBABILITY RESULTS - INTERNAL
. SEQUENCE GROUPS BY TYPE OF FAILUWE i
SMALL LOCA AND FAILURE OF ECC RECIRCULATION SMHjHf P = 6.0E-5 2 4 TRANSIENT WITH LOSS OF AUXILIARY FEEDWATER AND FAILURE OF BLEED AND FEED TIMLE T MLE 2 TgMLE T MLE P = 2.9E-5 5 TRANSIENT INDUCED LOCA AND FAILURE OF ECC RECIRCULATION TMQH{Hj 2 T0H{Hj P = 3.4E-5 3 SMALL LOCA AND FAILURE OF ECC INJECTION S2MDy2D P = 9.0E-6 - G G
d i SEQUENCE GRGUPS BY TYPE OF FAILURE (CONT.) l TRANSIENT INDUCED LOCA AND FAILURE OF ECC INJECTION T MQDy2D f 2 T30Dy2 D P = 5.5E-6 4 TRANSIENT WITH LOSS OF AUXILIARY FEEDWATER AND i FAILURE OF ECC RECIRCULATION 1
! T MLH 1 P = 2.0E-6 2
- - SMALL LOCA WITH FAILURE OF ECC INJECTION AND FAILURE TO ACHIEVE SECONDARY BLOWDOWN
,f S MXU P = 6.5E-7 2 1 1
# # O.
D s MOST SIGNIFICANT POINT BEACH VULNERABILITIES - INTERNAL INTERNAL VULNERABILITY 1: FAILURE TO SWITCHOVER FROM EMERGENCY
! CORE INJECTION TO RECIRCULATION SMH{Hj 2.00E-5 2
TMQHjHj 1.40E-6 2 T0HjHj 9.90E-6 3 _ 3.13E-5 INTERNAL VULNERABILITY 2: STATION BLACKOUT DUE TO BATTERY l FAILURE TyMLE 4.89E-6 INTERNAL VULNERABILITY 3: STATION BLACKOUT DUE TO DIESEL GENERATOR FAILURES TyMLE 5.49E-7 l . O - 9 O
~
MOST SIGNIFICANT POINT BEACH VULNERABILITIES (CONT.) INTERNAL VULNERABILITY 4: FAILURE OF ECC RECIRCULATION DUE TO RHR PUMP COOLING FAILURE CAUSED BY A VALVE FAILURE SMHjHj 9.60E-6 2 T30Hy2 H 4.75E-6 TMQHjHj 6.72E-7 2 1.50E-5 INTERNAL VULNERABILITY 5: FAILURE OF ECC INJECTION DUE TO CCWS FAILURE CAUSED BY LOSS OF COOLING FROM THE SWS THROUGH THE CCW HEAT EXCHANGER i S2MDy2 D 3.80E-7 T30Dy2 D 5.78E ,7 T2MQDy2D 8.26E-8 1.04E-6 O O O
MOST SIGNIFICANT POINT BEACH VULNERABILITIES (CONT.) ~ INTERNAL VULNERABILITY 6: COMMON MODE FAILURE OF SAFETY SYSTEMS PUMPS-SMH{Hj 2 4.00E-6 S MDy2. 2 D 4.60E-6 S MXD y 2 1.20E-7 - TyMLE 2.18E-7 g T2MLH 1 2.03E-6 j T2MLE 6.61E-6 TMQH{Hf 2 1.80E-7 T MQDy2 2 D 3.22E-7 T30Hy2 H 1.98E-6 T3QDy2 D 2.25E-6 T4MLE 1.18E-6 T MLE 5 5.80E-7 i 2.42E-5 J O . O O
c - _ _ - _ MOST SIGNIFICANT POINT BEACH VULNERABILITIES (CONT.) INTERNAL VULNERABILITY 7: COMMON MODE FAILURES OF SAFETY SYSTEM VALVES i SMHjHj 8.00E-6 2 TMQHjHj 5.60E-7 t 2 T0H{Hj 3.96E-6 3 1.25E-5 !j INTERNAL VULNERABILITY 8: FAILURE OF THE LOW PRESSURE INJECTION SYSTEM IN THE RECIRCULATION MODE
-l SMH{Hj 1.43E-5
- l 2 TMQHjHj 9.99E-7 2
T0HHj 7.07E-6 3 y 2.24E-5 INTERNAL VULNERABILITY 9: FAILURE OF THE AFWS TURBINE DRIVEN PUMP TyMLE 4.35E-7 T4MLE 4.25E-6 T5MLE 5.71E-6 1.04E-5 h &~
5 a
.-f MOST SIGNIFICANT POINT BEACH VULNERABILITIES (CONT.)
~
INTERNAL VULNERABILITY 10: FAILURE OF THE CCW PUMPS S2MD12 D 1;23E-6 l' i T30D12 D 6.03E-7 1.83E-6 INTERNAL VULNERABILITY 11: LONG TERM STATION BLACK 0UT CAUSED BY DEPLETION OF THE STATION BATTERIES OR THE CONDENSATE.. STORAGE TANK T1M * (FAILURE OF DGS) * (BATTERY DEPLETION AND/0R CST DEPLETION) 3.56E-5 e
*g G ,
O G
1 SEISMIC ANALYSIS
- SAFE SHUTDOWN EARTHQUAKE 0.12G
- EARTilGUAKE FREQUENCIES
\
1-2 SSE 1.57E-3 4 2-3 SSE 1.34E-4 3-4 SSE 1.49E-5 4-5 SSE 3.18E-6
- SEISMIC INITIATED EVENTS S - SMALL LOCA EQ1 - TRANSIENT W/PCS INITIALLY AVAILABLE EQ2 - TRANSIENT W/PCS FAILURE DUE TO INITIATING EVENT (LOSP DOMINANT EVENT) e O O
SEISMIC CORE MELT PROBABILITIES LEVEL (SSE) Sj! EQ1 EQ2 TOTAL I-2 6.8E-6 6.3E-6 1.2E-5 2.5E-5 2-3 1.1E-5 2.5E-6 8.8E-6 2.3E-5 3-4 4.5E-6 2.8E-7 3.7E-6 8.4E-6 4-5 2.3E-6 3.0E-8 1.0E-6 3.4E-6 6.1E-5 SEISMIC RELATED VULNERABILITIES
- 1. REFUELING WATER STORAGE TANK
- 2. ANCHORAGE OF ELECTRICAL BUSES, TRANSFORMERS, INVERTERS, AND BATTERY CHARGERS
- 3. BATTERY RACKS INADEQUATELY ANCHORED
- 4. INSTRUMENT AIR SYSTEM FOR PORVS 9 ,
G 9
'l.
! MODIFICATIONS BASED ON SEISMIC ANALYSIS l 1. PUMPS AND PIPING TO MAKE SPENT FUEL POOL BACK-UP WATER SOURCE FOR RWST
- 2. PROVIDE ADDITIONAL ANCHORAGE FOR BUSES, INVERTERS, AND CHARGERS l 3. REPLACE WOODEN BATTERY RACKS WITH METAL RACKS AND PROVIDE ADDITIONAL ANCHORS
- 4. INSTALL SAFETY CLASS NITROGEN BOTTLE AND ASSOCIATED EQUIPMENT SEISMIC CORE MELT PROBABILITIES WITH MODS' LEVEL S2 EQ1 E02 TOTAL 1-2 9.6E-9 1.1E-7 1.9E-7 3.0E-7 2-3 5.1E-7 3.7E-8 2.2E-7 7.6E-7 3-4 1.5E-6 1.3E-8 3.3E-7 1.8E-6 4-5 1.8E-6 6.7E-8 4.9E-7 2.2E-6 5.0E-6 O . O O
t FIRE ANALYSIS VULNERABILITIES: AFW PUMP ROOM - CABLE RUNS
- SW PUMPS MD AFW PUMPS ALL SI AND CCW PUMPS PROB CM 1.3E-5 4160V SWITCHGEAR ROOM SAFETY RELATED CABLING U PROB CM 2.0E-5 S
MODIFICATIONS: AFW PUMP ROOM - AUTO SUPPRESSION SYSTEM (WATER-DRY PIPE-PREACTION) PROB CM 5.0E-7 li160 SWGR ROOM - RELOCATE DC BUS / CHARGER ~ PROB CM 6.9E-7 5 9 9
4 t i INTERNAL FLOOD ANALYSIS (SPRAY) VULNERABILITY INTAKE STRUCTURE - MAIN FIREWATER HEADER SW PUMPS PROB CM 7.7E-5 MODIFICATION INCREASE BARRIER HEIGHT TO SHIELD PUMPS l PROB CM 7.7E-8 G
# O e
_ m mu4
.h a
e ALTERNATIVE SELECTION AND INTEGRATION THIRTY-ONE POTENTIAL VULNERABILITIES IDENTIFIED INTERNAL ELEVEN SEISMIC EIGHT SPRAY ONE FIRE TWO WIND AND MISSILE SIX FLOOD TWO LIGHTNING ONE SUPPORT SYSTEMS IMPORTANT ONLY SIXTEEN MODIFICATIONS CARRIED INTO IMPACT ANALYSIS 9 O O
- COMPOSITION OF ALTERNATIVES EACH ALTERNATIVE IS THE COMBINATION OF THE MODIFICATIONS LISTED IN THE COLUMN BELOW ALTERNATIVE VULNERABILITIES 1 2 3 _
4 1 INTERNAL 1 X X X INTERNAL 2 X X X INTERNAL 4 X X X INTERNAL 8 X INTERNAL 9 X INTERNAL 11 X X SEISMIC 1-6 X X X SPRAY X X X X X X FIRE 1 i X X X FJRE 2 i
. WIND 6 X ADD-ON SDHR O
~
O O
i EST.IMATED CORE MELT PROBABILITY WITH MODS TOTAL CM PROBABILITY BEFORE M13S 3.6E-4 ALTERNATIVE 1 1.4E-4 ALTERNATIVE 2 1.1E-4 ALTERNATIVE 3 7.5E-5' ALTERNATIVE 4 3.6E-5 0UTPUT FROM IMPACT ANALYSIS COST OF EQUIPMENT AND INSTALLATION IN 1985 DOLLARS COST OF REQUIRED REPLACEMENT POWER IN 1985 DOLLARS COST OF ANNUAL OPERATION, MAINTENANCE AND INSPECTION IN 1985 DOLLARS RADIATION EXPOSURE INVOLVED WITH INSTALLATION IN PERSON-REM l O O O
MODIFICATIONS BASED ON INTERNAL ANALYSIS
- 1. ADD A MORE PROMINENT ALARM WARNING THAT SWITCHOVER FROM INJECTION TO l RECIRCULATION-IS NECESSARY AND IMMINENT
- 2. INSTALL DEDICATED STARTU'P BATTERIES TO EACH DIESEL GENERATOR TO ELIMINATE THE DEPENDENCE OF THE DIESEL GENERATORS ON THE STATION
! BATTERIES
- 3. INSTALL A TURBINE DRIVEN GENERATOR TO SUPPLY VITAL AC AND DC LOADS
- 4. INSTALL PARALLEL MANUAL VALVE TO VALVE XV30 IN THE RHR PUMP COMPONENT COOLING WATER LINE AND CHECK THE VALVE POSITIONS ONCE PER SHIFT
- 5. INSTALL A THIRD INDEPENDENT LOW' PRESSURE TRAIN WITH ADDITIONAL SUCTION LINE FROM THE SUMP AND CROSSTIES TO ORIGINAL TWO TRAINS 1
- 6. INSTALL INDEPENDENT DIESEL DRIVEN AFWS PUMP IN PARALLEL TO THE TURBINE DRIVEN PUMP BUT SHARING THE SUCTION AND DISCHARGE LINES
- 7. ADDITIONAL 770,000 GAL. CST SHARED BETWEEN UNITS 4
O , O O
-~
CONTENT OF ALTERNATIVES i ALTERNATIVE 1: RWST LEVEL ALARM IMPROVEMENTS DEDICATED DIESEL GENERATOR STARTUP BATTERIES REDUNDANT RHR PUMP COOLER OUTLET VALVES RWST SEISMIC IMPROVEMENTS IMPROVED SEISMIC ANCHORAGE OF ELECTRICAL EQUIPMENT BACKUP AIR SUPPLY FOR PORVS i INTAKE STRUCTURE SHIELD WALL EXTENSION AFW PUMP ROOM FIRE PROTECTION SEPARATION OF DC EMERGENCY POWER SUPPLIES ALTERNATIVE 2: ALL ABOVE PLUS ADDITIONAL CST ALTERNATIVE 3: ALL ABOVE PLUS
- THIRD INDEPENDENT LOW PRESSURE TRAIN INDEPENDENT DIESEL DRIVEN AFWS PUMP DIESEL GENERATOR EXHAUST STACK ANCHORAGE IMPROVEMENTS ALTERNATIVE 4: ADD-ON SDHR SYSTEM 9 .
CORE MELT PROBABILITIES INTERNAL 1.8E-4 SEISMIC 6.1E-5 SPRAY 7.7E-5 FIRE 3.2E-5 WIND AND MISSILES 4.0E-6 . EXTERNAL FLOOD 1.9E-8 LIGHTNIN6 5.8E-8 3.56E-4 BASE CORE MELT PROBABILITY APPROXIMATE CORE MELT MODIFICATION OF BASE CORE MELT PROBABILITY AFTER VULNERABILITY PROBABILITY MODIFICATION INTERNAL 1 3.1E-5 3.1E-6 2 4.8E-6 1.0E-7 4 1.5E-5 3.3E-6 8 2.2E-5 6.3E-6 9 1.0E-5 4.3E-7 11 3.5N;5 1.3E-6 SEISMIC 1-6 6.1E-5 5.0E-6 SPRAY 7.7E-5 7.7E-8 FIRE 1 1.3E-5 5.0E-7 2 2.0E-5 6.9E-7 WIND 6 4.0E-6 1.0E-7 O O O
IMPACT ANALYSIS APPROACH INITIAL DESIGN REPORT 4 INTERFACES DESCRIBED SCOPE OF WORK BILL OF MATERIALS j RIP-0UT 4 MAJOR CONSTRUCTION SITE INSPECTION PLAN AREAS OF INTEREST LOCAL PRACTICE QUESTIONS I PLANT VISIT
! LOCAL COSTS CONGESTION ACCESS FEASIBILITY OF LOCATION RADIATION LEVELS FINAL DESIGN USING STANDARD INDUSTRY PRACTICES O .
O O
POINT BEACH IMPACT ANALYSIS RESULTS ENGINEERING AND INSTALLATION OCC DOSE ALTERNATIVE COSTS 5 OEM COSTS 5 .P-REM 1 7.419E6 1.1E4 17 2 14.376E6 3.7E4 17 3 22.497E6 1.71ES 27 4 59.047E6 3.79ES 486
SUMMARY
OF CORE MELT PROBABILITIES WITH ALTERNATIVES DELTA PROB. IMPROVEMENT ALTERNATIVE PROB. OF CM OF CM FACTOR BASE CASE 3.6E-4 1 1.4E-4 2.1E-4 2.5 2 1.1E-4 2.5E-4 3.3 3 6.9E-5 2.9E-4 5.1 4 2.4E-5 3.3E-4 15.0 0 0 0
POINT BEACH - EXPECTED VALUE POPULATION DOSE (50 MILE STANDARD CRAC2) LB SOURCE BASE SOURCE UB SOURCE 0.1 RSS 0.3 RSS RSS 4 j 4 BASE CASE 160.1 221.5 253.5 I ALT. 1 59.3 83.1 97.5 ALT. 2 41.9 59.0 69.7 ALT. 3 24.5 34.2 39.2 ALT. 4 9.6 13.4 15.5 1 POINT BEACil - AVERTED OFFSITE DOSE (P-REM) ALT. I 100.8 138.4 156.0 ALT. 2 118.2 162.5 183.8 ALT. 3 135.6 187.3 214.3 ALT. 4 150.5 203.1 238.0 i O O O
TABLE 9.3 POINT SEACH - St99 TART OF INFACTS (Based Upon 52 placount) NFCATIVF IMPACTS DUE TO AVFRTABLE POSITIVF IMPACTS ASSOCIATFD WITil HODIFICATIONS (Present Worths) ONSITE COSTS (Present Worthe) tit tlity costs Change in Installa- Operations peplacement Power costo Peplace- Imse of Total Cnre Melt tion and and Mainten- In Service TOTAL ment Invest- Site Avert- NET Alternative Probability Engineer- ance Costs Instal- (PW) POSITIVE Power ment Cleanup able IMPACT No. [ Central ing Costs (PW) lation IMPACT Costs Costs Costs Costs valuel ($ ul0' ) ($ 310) ($ alo-6) ($ m10~0) ($ ulo'0) ($ ul0' ) ($ ul0' ) ($ u 10' ) ($ al0' ) ($ ul0' ) ltr. C. & t. Values) A, P Ig 13.5 I, 1) 13.5 I 4 TI I{g 152 I53 I5 NI trot I. 0.17 0.09 0.69 0.95 6.618 1 2.13E-4 7.419 0.149 ' O.0 Available 7.568 0 0.84 0.43 3.44 4.72 2.848 Probably II 4.19 2.17 17.25 23.61 -16.042 Negligthle
- t. 0.19 0.10 0.80 1.09 13.786 2 2.48E-4 14.376 0.500 0.0 -
14.876 0 0.97 0.51 4.02 5.50 9.376 II 4.87 2.53 20.09 27.49 -12.614
- 1. 0.23 0.12 ,
0.93 1.28 23.526 3 2.87E-4 22.497 2.309 0.0 *
- 27. 806 C 1.13 0.54 4.6% 6.37 18.436 II 5.64 2.93 23.25 31.82 - 7.014
- f. 0.76 0.14 1.08 1.48 62.684 4 3.32E-4 59.047 5.117 0.0 "
64.164 C 1.30 0.68 5.38 7.36 56.804 U 6.52 3.39 26.89 36.80 27.364 TI =11 + 13.5 I2+I3 + 13.5 14 I5' = 151' + I52' + 353' NI = TI-I 5' Notes L. C, and U represent the lower, central, and upper bound estimate for core melt probahtlity e .
- e. O
i , . TAal.R 9.4-a POInry aEACH =
SUMMARY
OF VAIAFES (Based on Population Dose to 50 pelles, 5% Discount Rate) POSITIVR VAItWS Onsite Offsite total Change in Present Worth Averted Present Worth Averted Present Alternatlee Core Helt Avarted . of Averted Averted Dose I of Averted Avert ed Dose I Worth of No. Probability Dose pose O Dose Base case pose S Dose Base case Averted pose (Central (p-rem) $1000/p-res (p-res) pose $1000/p-res (p-rem) Dose S $1000/p-rem Valuel ($ ul0' ) ($ s10' ) ($ s10' ) (U. C. & f. Values l lU, C, & I. Values l A, P Vg V g V 2 ADR, V 2 12 n 2 L 2118 63 1.360 2570 65 1.50s 1 2.13E-4 252 0.148 c 3183 62 1.868 3435 64 2.016 li 3588 62 2.106 3840 .63 2.254
- f. 2719 74 1.596 3013 76 1.768 2 2.48E-4 294 0.172 C 3738 73 2.194 4032 75 2.366 is 4227 .72 2.481 4521 74 2.653 L 3119 .R5 1.831 3459 86 2.031 3 2.87E-4 340 0.200 c 430R 85 2.529 4648 06 2.729 U 4929 85 2.893 5269 85 3.093 L 3462 94 2.012 3855 95 2.263 4 3.32E-4 393 0.231 c 47a6 94 2.a09 5179 94 3.040 U 5474 94 3.213 5867 94 3.444 Vg = 51500 m Ap. x 23 Vg* = 51500 m Ap m $1000 x 13.5 V2 = Averted Dose a 23 V2' = Averted Dose z $1000 x 13.5 ADRn =V2 I (Basecame Dose x 23)
V12 *VI+V2 ADpa =V12 I (Basecame Dose N 23 + V3 ) V12* *V'+V' 1 2 Notes Le c, end U stand for lower, central, and sapper bound source term estimate 4
TABLE 9.4-b POIN? SEACH - SUlet4RY OF VALUFS (Based on Population pose to 50 Mile, 5s Discount pate) NEGATIVE VAiJES Installation operation "otal Net Value Change In Present In-service Present Installa- Present North Present Alternative Core Nolt Is. stall- North of opera- North of tion and of Install. & Ave rted North of No. Probability atton Installa- tional In-Service operation- oper. Dose Dose Averted Dose (Central Dose tion pose O Dose oper. pose al pose 8 $1000/p-rom (p-rea) 0 $100d/p-res valuel (p-ren) $1000/p-rem (p-rem) ($ u 10' ) ($ al0' ) ($ x10' ) lu, C, & L Values) A, P V 3 Vj V 4 Vj V3+V4 Vj + Vj NV WV' L 2553 1.491 1 2.132-4 17 017 Hegligible 17 017 C 3418 1.999 U 3823 2.237 L 2996 1.751 2 2.482-4 17 017 Negligible 17 017 e 4015 2.349 u 4504 2.636 9 L 3432 2.004 3 2.87E-4 27 027 Negligible 27 027 C 4621 2.702 II 5242 3.066 L 3369 1.777 4 3.32r-4 486 486 Negligible 486 .486 C 4693 2.554 U 5381 2.950 V' 3 =V3 m $1000 V' 4 = (V4 1 23) x $1000 x 13.5 NV =V1+V3=V3-V4 NV' = Vg' + V 2' *V'-V' 3 4 Notes L, C, and U stand for lover, central, and upper bound source term estimate e ,
-G G
TAaLa 9.5 POIN? BEACH - SUPMARY OF VAIJFF-INPAC'* ANALYSIS (Based on Pogwalation ense to 50 N11ess 54 ptscount pate) V-I ANAF.YSIS BASFD ON OFFSI C (*OSTS V-I ANM.YSIS BASFD ON OFFSITF AND ONSITR CO3TS Present North Neasures of V-I Present North Nessures of V-I Total Averted Total of Averted Net of Averted Alter- Change in Offsite Averted pose impact pose 8$ 1000/ V-T Net pallars impact Dose # $1000/ V-1 Net Dollars native core Melt Ratio penefit per Probability pose i Base (Centret p-rem patto penefit per (Central p-rem No. [U,C,EL Valueel (Centrat (p-res) case pose cost) lu,c,sL Valueel p-rem coet! p-tem ($ m10-0) ($ ul0~0) ($ m10-6, valuel lu,t,6I. Valueel ($ x10" ) ($ u 10" ) ($ z 10-6) NI NV8 VID N8V DPR A p, V 3 ADM, TI Vj VIR, N8V, DP R, 65 1.360 0.18 -6.208 3265 1.491 0.524 -1.357 1116 L 2318 7.568 1.868 0.25 -5.700 2378 2.848 1.999 0.702 -0.849 833 1 2.13E-4 C 3183 64 0.786 -0.611 745 U 3588 63 2.106 0.27 -5.462 2109 2.237 76 1.596 0.11 -13.280 5471 1.751 0.187 -7.652 3130 L 2719 C 3738 75 14.876 2.194 0.15 -12.682 3980 9.376 2.349 0.251 -7.027 2335 2 2.48E-4 0.281 -6.740 2082 U 4227 74 2.481 0.17 -12.395 3519 2.636 86 1.831 r.074 -22.975 7953 2.004 0.109 -16.432 5372 L 3119 -15.731 3990 86 24.806 2.529 0.102 -22.277 5750 18.436 2.702 0.147 3 2.87E-4 c 4308 0.166 -15.372 3517 U 4929 65 2.893 0.117 -21.913 5033 3.066 95 2.032 0.032 -62.132 19534 1.777 0.031 -55.027 16861 L 3462 -54.250 12 04 94 64.164 2.809 0.044 -61.355 13407 56.804 2.554 0.045 4 3.322-4 c 4786 0.052 -53.846 10556 o 5474 94 3.213 0.0%C -60.951 11722 2.958 VIpo a V'2 I TI VIPn = NV' I NI NBVo=V3' - TI NV8n*NV'-NI DPpo a T1 I V2 DPRn " NI I NV Notes L, C, and U stand for lower, central, and upper bound source term estimate O O O
TABLE 9.6 POINT BEACH - StJHMARY OF VAIRE-IMPACT MFASURES (Central Value) V-I Analysis Based on Of fsite Change Offsite Total 7-I Analysis of Offsite Costs and Onsite Costs in Core Averted Averted Alternative Melt Dose pose V-I Net Dollars V-I Net pollars No. Probability (p-ren) Ratio Ratio Benefit per p-ren Ratio Benefit per p-res ($ slo' ) ($ sto' ) A, P V 2 ADP n vI9 0 "Bvo PPR , VIR, NBV DPR, 1 2.13E-4 3183 0.64 0.25 -5.700 2378 0.702 -0.849 833 2 2.48E-4 3738 0.75 0.15 -12.682 3980 0.251 -7.027 2335 3 2.87E-4 4368 0.86 0.102 -22.277 5758 0.147 -15.734 3990 4 3.32R-4 4786 0.94 0.044 -61.355 13407 0.045 -54.250 12104 O . O O f
TABt.E 9.6 POINT BFACH -
SUMMARY
OF VAIEE-IMPACT MEASURES (Central Value) V-I Analysis Based on Of fsite Change Offsite Tota! V-I Analysis of Offsite Costs and Onsite Costs in Core Averted Awarted Alternative Melt Dose Dose V-! Net Dollars V-I Net Dollars No. Probab!!!ty (p-res) Ratio Ratio Benefit per p-rem patio penefit per p-ren ($ s10~ p ($ sto' ) Ap, V U"" ""# 2 AU"n #I"o """o o #I"n n U""n 1 2.132-4 3183 0.64 0.25 -5.700 2378 0.702 -0.849 833 2 2.48E-4 3738 0.75 0.15 -12.682 3980 0.251 -7.027 2335 2 2.872-4 4308 0.86 0.102 -22.277 5758 0.147 -15.734 3990 4 3.32E-4 4786 0.94 0.044 -61.355 13407 0.045 -54.250 12104 c.n 9 G G
+O POINT BEACH SENSITIVITY ANALYSES WITHOUT FEED & BLEED DIFFERENCE WITH FEED & BLEED 2.57E-4 +2.23E-4 3.43E-5 WITH SECONDARY WITHOUT SECONDARY BLOWDOWN BLOWDOWN i
1.89E-4 +5.20E-5 1.37E-4 FEED & BLEED WITH FEED & BLEED WITH PORVS BLOCKED PORVs UNBLOCKED 4.58E-5 -1.12E-5 ! 5.60E-5 IN EACH CASE ABOVE ONLY THE ACCIDENT SEQUENCES AFFECTED ARE INCLUDED HERE. THESE ARE NOT TOTAL CM PROBABILITIES.
# 9 e
./}
4 i CRAC2 Conditional Population Dose (50 mile radius for Point Beach
\_ / ^
Standard CRAC2 Estimate Release Upper
- Central Lower Category Bound Estimate Bound 1 7.9E+5 6.6E+5 4.4E+5 2 8.0E+5 7.5E+5 5.6E+5
'3 9.6E+5 6.2E+5 3.7E+5 4 5.4E+5 2.7E+5 1.3E+5 5 2.2E+5 1.0E+5 4.8E+4 6 5.3E+4 2.2E44 9.8E+3 7 2.6E+3 1.7E+3 1.4E+3 Revised CRAC2 Estimate (no interdiction, no decontamination) g (h ; Release Upper
- Central Lower
\s_,/ Category Bound Estimate Bound 1 6.3E+7 1.9E+7 6.3E+6 --
2 1.1E+8 3.2E+7 1.1E+7 3 3.3E+7 1.0E+7 3.3E+6 4 7.3E+6 2.2E+6 7.3C+5 5 1.7E+6 5.0E+5 1.7E+E 6 1.5E+5 4.4E+4 1.6E+4 7 3.1E+3 1.8E+3 1.4E+3
- Upper Bound, Central Estimate and Lower Bound are based upon Reactor Safety Study releases and represent 1xRSS, 0.3xRSS, and 0.1xRSS levels respectively. The 0.3xRSS has been used as the central estimate for the TAP A-45 analyses.
O o V O b
- - - , - - --,-w w,- --+v , ~ , - ,
p
-s s Expected Population Dose in Person-rem Per
. \ Reactor Year (50 Mile Endius) for Point Beach ( Standard CRAC2 Estimate Expected Dose - Release Release Category Upper Central
- Lower Category Probability Bound Estimate Bound 1 1.2E-6 1.0 0.8 0.5 2 2.5E-4 200.0 187.5 140.0 3 5.1E-5 49.0 31.6 18.9 4 2.3E-6 1.2 0.6 0.3 5 1.3E-7 0.0 0.0 0.C 6 4.3E-5 2.3 1.0 0.4 7 5. 7 E- 6 __0 0 0.0 0.0 253.5 221.5 160.1 TOTAL F')
\ \~' Revised CRAC2 Estimate (no interdiction and no decontamination)
Expected Dose Release Release Category Upper Central Lower Category Probability Bound Estimate Bound 1 1.2E-6 75.6 22.8 7.6 2 2.5E-4 27,500.0 8,000.0 2750.0 3 5.1E-5 1,683.0 510.0 168.3 4 2.3E-6 16.8 5.1 1.7 5 1.3E-7 0.2 0.1 0.0 6 4.3E-5 6.5 1.9 0.7 7 5.7E-6 0.0 0.0 0.0 TOTAL 29,282.1 8,539.9 2,928.3 O Ov W
- -w-,. -,
. . s e TA8LE 9.9 POINT BEACH - VALUE-1MPACT RESULTS - No Interdiction or Decontamination V-! Analysis of Offsite Costs V-I Analysis Based on Offsite and Onsite Costs Reduction Offsite in Core Averted Averted Alternative Melt Dose Dose V-I Met Dollars V-I Net Dollars Probability p-ren Ratio Ratio Benefit per p-rem Ratio Benefit per p-ren A Pm ($ x10-6, gg ,gg -6 3 j Ap, Vy ADR, VIR, NBV, DPR, VIR, NBV, DPR, L 43194 0.64 3.35 17.785 175 8.948 22.636 66 1 2.13E-04 C 125856 0.64 9.76 66.304 60 25.984 71.155 23 U 431848 0.64 33.49 245.908 18 89.047 250.759 7 L 50485 0.75 1.99 14.757 295 3.177 20.412 185 2 2.48E-04 C 147131 0.75 5.81 71.484 10 1 9.227 77.139 64 U 504781 0.75 19.92 281.409 29 31.617 289.064 19 57063 0.85 1.35 8.638 435 1.826 15.231 321 L 3.94 72.853 149 5.307 79.396 111 3 2.87E-04 C 166382 0.85 0 570538 0.85 13.50 310.075 41 18.174 316.618 32 63296 0.94 0.58 -27.012 1014 0.650 -19.907 899 L 1.69 44.174 340 1.903 51.279 308 4 3.32E-04 C 184573 0.94 86 U 632914 0.94 5.79 307.329 lol 6.535 314.434 Notes L, C, and U represent estimates based on lower, central, and upper source term values. 9 - O . O
l NRR STAFF PRESENTATION TO THE o o ACRS l l l l
SUBJECT:
STATUS OF USI A-45 RESOLUTI0fl EFFORT l DATE: DECE!1BER 3,1985 l PRESENTER: A. R. MARCHESE O PRESENTER'S TITLE / BRANCH /DIV: TASK MAtlAGEP/ REACTOR SAFETY ISSUES BRAflCH/ DIVISION OF SAFETY REVIEk' AtlD OVERSIGHT PRESENTER'S NRC TEL. NO.: (301) 492-4712 - SUBCOMMITTEE: DECAY HEAT REMOVAL SYSTE!!S SUBCOVIITTEE. 4 0 O
i O O O . PRESENTATION OUTLINE j PLANT ANALYSIS REVISION ACTIVITIES AND SCHEDULE.
.o
SUMMARY
OF REVISED POINT BEACH PLANT ANALYSIS RESULTS. j , I *
SUMMARY
OF QUAD CITIES PLANT ANALYSIS RESULTS. ) PRELIMINARY ANALYSIS RESULTS OF A DEDICATED PRIMARY BLOWDOWN SYSTEM. i CONSIDERATION OF HIGH-PRESSURE RHR SYSTEM. j l CE/PORV ISSUE. i i SAB0TAGE ANALYSES. PROGRAM SCriEDULE. i a l 99 - 49 . 99
\ o O O . l - l l l INTRODUCTORY REMARKS l l ! ROUGH DRAFT PLANT REPORTS ISSUED IN MAY. l REVIEWED WITH ACRS SUBCOMMITTEE ON MAY 30. i DETAILED STAFF COMMENT RECEIVED IN JUNE. l COMMENTS REQUIRED SIGNIFICANI ADDITIONAL ANALYSES BY SANDIA. l REVISED PLANT REPORTS RECEIVED IN NOVEMBER.
~
- SIGNIFICANT CHANGES IN RESULTS.
i i i i OO OO OO .
o o o c. l . l .
.: s. .
I ALTERNATIVE MEASURES
' .'lMPROVED. OPE.. RAT. IONS (E.G., PROC 5DURES., TRAINING,INFbRM
- USE EXISTING SYSTEMS IN ATYPICAL,, MODES , ,
1 .
- ADD PORVS ONLY'
- DEDICATED FEED AND. BLEED SYSTEM -
?. ,.
1 ,
\
- HIGil PRESSURE RHR SYSTEM
- ADD-0N EMERGENCY FEEDWATER SYSTEM .
l'
~
'* ADD-0N SUPPRESSION P00L COOLING AND LOW PRESSURE INJECTION SYSTEM S
l . 4 i, '* BUNK.ERED AND DEDICATED SHUTDOWN C0OLING SYSTEM - i. l . i 3 . . j' . :: i ,
- . y_
g
~
~
O . O -
.. O .
ROLE OF FEED AND BLEED IN A-45 PROGRAM ASSESSVIABILITYOFFEEDANDBLE5D - DETERMINE UNDER WHAT CONDITIONS PLANTS (ONE FOR EACH PWR VENDOR) CAN.SUCCESSF e e
- i FEED AND BLEED .
EXTEND 2ND'ITEMTOARANGEOFPLANTTYPE5 -
. g i
DETERMINE IMPORTANT PARAMETERS (I.E., VALVE ' SIZE, HPI PUMP SHUT 0FF HEAD, OPERATOR ~ F
- i A.CTION TIME) TO SUCCESSFULLY FEED AND BLEED..
r DETERMINE REQUIRED PLANT MODIFICATIONS TO SUCCESSFULLY FEED AND BLEED' , , ; ASSESS FEASIBILITY OF MODI:lCATIONS IN TERMS OF BACKFITT.ING TO.0PERATING PLANTS DETERMINE IMPACTS (E.G., COSTS, DOWNTIME) 0F BACKFIT . ASSESS VALUE OR BENEFITS OF FEED AND BLEED f COMPARE VALUE/ IMPACT OF FEED AND BLEED TO OTHER ALTERNATIVE. MEASURES TO IM , OVERALL DHR RELIABILITY f.. b O
PLANT ANALYSI STATUS - QUAD CITIES ANALYSIS COMPLETED DRAFT REPORT RESUBMITTED - NOVEMBER 1 POINT BEACH ANALYSIS COMPLETED DRAFT REPORT TO BE RESUBMITTED - NOVEMBER 15 COOPER ANALYSIS NEAR COMPLETION DRAFT REPORT IN PREPARATION DRAFT REPORT SUBMIlTAL DUE - NOVEMBER 29 TURKEY POINT FAULT TREES BEING EDITED / CHECKED INTERNAL SEQUENCES IDENTIFIED SPECIAL EMERGENCY ANALYSIS WELL ALONG DRAFT REPORT SUBMITTAL DUE - DECEMBER 13 TORJAN FAULT TREES BEING EDITED / CHECKED INTERNAL SEQUENCES AND SPECIAL EMERGENCY .. ANALYSIS IN PROCESS DRAFT REPORT SUBMITTAL DUE - JANUARY 15 ANO-1 FAULT TREES BEING EDITED / CHECKED INTERNAL SEQUENCES AND SPECIAL EMERGENCY ANALYSIS IN PROCESS EXCEPT VULNERABILITIES NOT IDENTIFIED AE PLANS PLANT VISIT IN DECEMBER DRAFT REPORT SUBMITTAL DUE - FEBRUARY 15 ST, LUCIE ANALYSIS TO START JANUARY 10 AE PLANS PLANT VISIT IN MID-JANUARY DRAFT REPORT SUBMITTAL DUE - MARCH 17
O O O . l MAJOR SCHEDULE MILESTONES FOR USI A-45 l CONTRACTOR FINAL
SUMMARY
REPORT ON SEVEN PLANTS MAY 1986 , l l DST DRAFT REGULATORY ANALYSIS AND TECHNICAL FINDINGS JULY 1986 > l REPORT ISSUED FOR STAFF COMMENTS STAFF COMMENTS TO DST AUGUST 1986 1 COMPLETED PACKAGE TO DIRECTOR, NRR SEPTEMBER 1986 PACKAGE TO CRGR OCTOBER 1986 CRGR REVIEW COMPLETE DECEMBER 1986 ISSUED FOR PUBLIC COMMENT FEBRUARY 1987 9 Og GG
3 !e ANALYSIS ACTIVITIES AND SCHEDULES SCOPE INTERNAL ANALYSIS REDONE TO INCLUDE: BROADER TREATMENT OF COMMON MODE ADDITIONAL TREATMENT OF ACTUATION LONG TERM BLACKOUT RE-EXAM OF HUMAN FACTORS / RECOVERY 1 SPECIAL EMERGENCIES - REVIEWED AND MODIFIED AS REQUIRED > li SABOTAGE " MODEST" CONDITIONAL DUANTIFICATION ATTEMPTED i e . 9 O
t PLANT ANALYSIS STATUS / SCHEDULE QUAD CITIES DRAFT SUBMITTED SANDIA MANAGEMENT APPROVED SENT TO CECO FOR COMMENT / INPUT PUBLICATION IN JANUARY POINT BEACH DRAFT SUBMITTED IN SANDIA APPROVAL PROCESS SENT TO WECO FOR COMMENT / INPUT PUBLICATION IN FEBRUARY COOPER DRAFT SUBMITTED IN SANDIA REVIEW PROCESS TURKEY POINT DRAFT IMMINENT . ST LUCIE INTERNAL ANALYSIS AT FIRST DUANT SPECIAL EMERGENCY IN PROGRESS TARGET DATE FEBRUARY FOR DRAFT ANO 1 INTERNALANAbYSIS IN PROGRESS SPECIAL EMERGENCY IN PROGRESS TARGET DATE FEBRUARY FOR DRAFT J TROJAN TEMPORARILY ON HOLD G ,
s l t M l SPECIAL TOPICS j
SUMMARY
/ GENERIC CONCLUSIONS TO DATE - IN PROCRESS 1ST DRAFT JANUARY .
VALUE-IMPACT CONSIDERATIONS FOR MORTORIA - IN PROGRESS REVISED DRAFT FEB STRUCTURE OF GENERIC V-I TO SUPPORT REGULATORY ANALYSIS - IN PROGRESS DRAFT JAN
SUMMARY
OF VALUE MEASURE AND V-I STtJRCTURES - IN PROGRESS DRAFT ASAP e . O 4}}