ML20153D977
| ML20153D977 | |
| Person / Time | |
|---|---|
| Issue date: | 05/03/1988 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1664, NUDOCS 8805090345 | |
| Download: ML20153D977 (450) | |
Text
- .-.
ACN37~- /44f 9 A!
f v 1 \ ) i, -
l O UNrrED STATES NUCLEAR REGULATORY COMMISSION '
.. -=======================================================
In the Matter of:
BADCOCK & WILCOX )
)
REACTOR PLANTS )
O Pages: 1 through 317 i Place: Washington, D.C.
Date: ftay 3, 1988
.==...====............................................m===,
- HERITAGE REPORTING CORPORATION
' OpicialReportes I 1220 L Street, N.W., Sedte 600
-l g[oSo9034508050 7- 664 S (2@ 6 M
, PDR
1 PUBLIC NOTICE BY THE i 2 UNITED STATES NUCLEAR REGULATORY COMMISSION'S 3 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
5 ;
6 7 The contents of this stenographic transcript of the 8 proceedings of the United States Nuclear Regulatory 9 Commission's Advisory Committee on Reactor Safeguards (ACRS), .
10 as reported herein, is an uncorrected record of the discussions 11 recorded at the meeting held on the above date.
12 No member of the ACRS Staff and no participant at
! 13 this meeting accepts any responsibility for errors or i
14 inaccuracies of statement or data contained in this transcript.
O 15 16 -
i !
4 17
~
18 i !
19 20 1
21 22 i 23 f 24 25 g Heritage Reporting Corporation (202) 628-4888 d
W G
1
-m---- ,, v.-.--. 7 4 r- , ,-.---w--.w,- . - . i.-., ,-,,m,w m-..,, _
. , , . , - ,.3,. ,,-.,.-.,r,,-,,,, ,.,,,=>_n , ,- - - - - , -m
1 0 1 UNITED STATES NUCLEAR REGULATORY COMMISSION 2 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 3
)
4 In the Matter of: )
)
5 )
SUBCOMMITTEE ON B&W REACTOR )
6 PLANTS )
)
7 Tuesday, 8 May 3, 1988 9 Room 1046 1717 H Street, N.W.
10 Washington, D.C. 20555 11 The above-entitled matter came on for hearing, 12 pursuant to notice, at 8:30 a.m.
13 BEFORE: MR. CHARLES J. WYLIE k'_' Retired Chief Engineer 14 Electrical Division Duke Power Company 15 Charlotte, North Carolina 16 ACRS MEMBERS PRESENT:
17 DR. WILLIAM KERR Professor of Nuclear Engineering 18 and Director of the office of Energy Research University of Michigan 19 Ann Arbor, Michigan 20 MR. CARLYLE MICHELSON Retired Principal Nuclear Engineer 21 Tennessee Valley Authority Knoxville, Tennessee 22 and Retired Director, Office for Analysis and Evaluation of Operational Data i 23 U.S. Nuclear Regulatory Commission i Washington, D.C.
24 l
(
~
HERITAGE REPORTING CORPORATION -- (202)628-4888 l l
2 l
i 1 HR. DAVID A. WARD Research Manager on Special Assignment 2 E.I. du Pont de Nemours & Company Savannah River Laboratory 3 Aiken, South Carolina 4 ACRS COGNIZANT STAFF MEMBER:
5 Richard Major 6 NRC STAFF PRESENTERS,:
7 B. Jones B. Siegel 8 J. Calvo R. Kendall 9
Consultants:
10 G. Reed 11 I. Catton P. Davis 12 H. Etherington
() 14 15 16 17 18 19 20 21 22 23 24 25 O
HERITAGE REPORTING CORPORATION -- (202)628-4888
3 1 PROCEEDINGS.
n
(_ 2 CHAIRMAN WYLIE: The meeting will come to order.
3 This is a meeting of the ACRS Subcommittee on Babcock and 4 Wilcox reactor plants. I am Charlie Wylie, Chairman of the 5 ACRS Subcommittee on the B&W reactor plants. The other ACRS 6 members in attendance are Dave Ward, Carlyle Michelson, and we 7 are expecting Dr. Lewis and Dr. Kerr later today. Consultants 8 present are Harold Etherington, Peter Davis, Ivan Catton and 9 Glenn Raad. Richard Hajor is the cognizant ACRS staff member 10 for today's meeting.
11 The rules for participation in today's meeting have 12 been announced as part of the notice of this meeting that was
() 13 published in the Federal Register on April 22nd, 1988. This 14 meeting is being conducted in accordance with the provisions 15 of the Federal Advisory Committee Act and the government in 16 the Sunshine Act.
17 We have received no written or oral statements from 18 members of the public.
19 It is requested that each specker first identify 20 himself or herself and speak with sufficient clarity and 21 volume so that he or she can be heard.
22 The purpose, as I mentioned, the purpose is to 23 review and discuss the B&W owners group plant reassessment
(} 24 program and the staff's evaluation.
25 The ACRS has been requested by the Commission to HERITAGE REPORTING CORPORATION -- (202)628-4888
4 1 give advice on this activity as well as to consider specific l
() 2 concerns by Mr. Demetrius Basdekas of the NRC staff regarding 3 .the consideration of specific technical safety issues as 4 pertaining to the reassessment.
5 The ACRS reviewed in 1986 the initial activity in 6 this reassessment and wrote a letter in which the ACRS 7 expressed concerns and made recommendations. The time has 8 been identified on our agenda to address the disposition of 9 those concerns.
10 I will ask the Subcommittee members and the t 11 consultants to keep in mind that as to what we should as the 12 meeting progresses, what we should bring to the Full Committee ,
i 13 on Friday morning. There is three hours allotted on the 14 agenda on Friday morning for the full ACRS.
15 Also any recommendations t:4nd comments that the 16 Subcommittee or the consultants would suggest the ACRS put in 17 a letter to the Commission in regard to this reassessment 18 program; we will discuss those later at the conclusion of our 19 meeting.
20 I will ask the Subcommittee members and consultants ;
21 if they have any comments at this time?
l 22 Hearing none, we will proceed with the me9 ting, and l
23 I believe Neil Rutherford, are you going to be the spokesman? l l
24 MR. RUTHERFORD: Is the staff going to have any I 25 presentation, Charles?
s HERITAGE REPORTING CORPORATION -- (202)628-4?',8
5 1 CHAIRMAN WYLIE: Okay.
O)
(. 2 HR. SIEGEL: We are just going to follow up.
3 HR. c' VO: My name is Jose Calvo. I am the 4 designated manager for the review of, the staff review of the 5 B&W owners group reassessment effort, and what I would like to 6 do, I would like to start the presentations by introducing the 7 topics that we are going to discuss as well au the staff is 8 going to present them.
9 I would like to give you a little background. Byron 10 Siegel, who is the project manager for this effort, is the 11 next one in line, and Bob Jones, who in addition to his work, 12 he was the overall technical coordinator for the review. We
() 13 have got Rick Kendall, who is representing the instrumentation 14 and control systems; Bill LeFave, the mechanical, from the 15 Plant Systems Branch who will be introducing those systems on 16 a particular area. You have got Gary Hammer, who will be--the 17 Hechanical Branch, who will be talking about the task force, 18 and from the area of human factors, we are going to have a 19 consultant, Joseph DeBor, who is from the Science Application 20 International, and he will be supported by the staff of, the 21 the staff representative for that particular topic will be Ann 22 Smith representing the Human Factors Assessment Branch.
23 Finally, we have got Mark Rubin, who is from the
(} 24 Division of System Engineering and Systems Technology, who 25 will be representing the risk assessment with the assistance HERITAGE REPORTING CORPORATION -- (202)628-4888
6 1 of the Brookhaven National Laboratory.
2 That presents the staff. That's all I am supposed 3 to say, and if you have any questions, I would like to bring 4 it back to Byron Siegel, who is going to give you an overview 5 of the B&W prograa.
6 HR. SIEGEL: As Jose mentioned, my name is Byron 7 Siegel, and I was responsible for the, I was the lead 8 emergency manager as far as coordinating the whole efforts on 9 the B&W program. I started last August or July about the time 10 that the owners group sent this, their final revision to their 11 report, the B&W 1919, which is the safety performance and 12 improvement program for B&W plants.
() 13 I didn't realize, I guess I thought that the owners 14 group was going to start off which something else so there is 15 a slide that sort of is a back-up slide that I have that I am 16 going to use that isn't on the handout that I have up there.
17 Just so that this doesn't get you cold, 18 originally--this is just a little history--originally Mr.
19 Stello, who is Chairman of the EDO, sent a letter to Tucker 20 who was chairman of the owners group, in January of 1986, and 21 essentially he identified some concerns related to the 1
22 menpitivity of the B&W design plants to operational transients 23 of the Davis-Besse and Rancho Seco events that happened in 24 1985, and he also said that the staff was going to reassess
( })
25 the overall safety of the B&W plants to determine that the HERITAGE REPORTING CORPORATION -- (202)628-4888
7 1 present requirements were appropriate for long-term and lead
() 2 to level of safety comparative to other PWRs. I'm sorry.
3 MR. REED: I seem to recall more in the challenge to 4 B&W and the owners group with respect to what they were to 5 look into, and let me remind you that the public announcement 6 No. 86-88 had words in that that went something like this--the 7 reassessment will include thermal hydraulic design aspects.
8 Also talked about basic design requirements, and basic design.
9 That seems to be more operationally requirements from the 10 challenge that was issued, what you have there.
11 MR. SIEGEL: Well, these were exactly out of the, 12 those are words almost out of the letter per se. There were
/'S 13 other things that were included.
V 14 MR. REED: So are my words exactly out of the public 15 announcement.
16 MR. SIEGEL: I think most of the areas were 17 primarily operational that were looked at.
18 MR. REED: That's the thinking that bothers me 19 because it seems to me that the real issue here is design of 20 the B&W plants, and say the two B&W plants--
21 MR. SIEGEL: We looked at--I think the owners group 22 and the staff will address those aspects as we go along in the 23 program. This is really just a basic overview.
24 MR. REED: I wanted to point out it seems to me some 25 words got dropped.
HERITAGE REPORTING CORPORATION -- (202)628-4888
8 1 MR. SIEGEL: What I just tried to glean out is what
() 2 I thought were the key aspects that he had, that he had, that 3 Mr. Stello keyed on.
4 In a letter from Tucker to Stello in February 1986, 5 the owners group committed to take the lead in this effort.
6 During the time that we, that the owners group took the lead, 7 the staff interacted with the owners group to 8 determine--during the development of the program and after the 9 original program was outlined, the staff then maintained its 10 normal role of assessing transmittals and limited independent j l
11 analysis. As a result of this effort, the owners group 12 developed their report called safety and performance 13 improvement program. B&W 1919, there were five revisions to 14 it, the last one was in July of 1987. I 15 In that report, basically there are, the report has 16 recommendations to improve the performance of B&W plants to l
17 reduce the number of complex transients and the number of i
18 trips that occurs at B&W plants, and there are approximately l 19 when all is said and done, about 275 recommendations that came i
20 out of that program.
21 The staff in its review reviewed all those 22 recommendations. The reports themselves may not specifically 23 identify each and every one, but there are, we looked at each l
24 and every one of them, and some of the statements in the 25 report combine, more than their general statements that HERITAGE REPORTING CORPORATION -- (202)628-4888
9 1 combine more than one of the recommendations.
( 2 We had a meeting with the Commission in August of 3 1987. I believe was August 5th, and in addition to what, in 4 addition to what was already in the-report, the Commission I
5 through a letter from, memo from Chilik asked you to verify 6 the safety of the B&W plants, comment on the adequacy of the 7 once through steam generators, the U tube steam generators, 8 comment on the acceptability of the emergency feedwater 9 initiation and control systems for B&W plants.
10 The staff completed its review just in March 1988.
l 11 There has been a report that was issued in November of 1987 12 which you all have I believe, and a supplement report in March
() 13 of 1988.
14 (Slide) l 15 MR. SIEGEL: Basically like I mentioned on previous R 16 l slides, this was a continuous interaction on the scope of the ;
l 17 program. We reviewed, evaluated the results and then we did 18 some independent work in certain areas, and a risk evaluation, l
19 we had a contract with a consultant, Brookhaven, consulting in 20 human factors, SAI. There was a limited thermalhydraulic 21 analysis which addresses part of your concern and limited 22 operating experience review.
23 Basically now that the, now that the, we have
() 24 evaluated the report, we see our role as that of oversight 25 role. We are, we have evaluated the recommendation approval HERITAGE REPORTING CORPORATION -- (202)628-4888
10 1 process and I will go into this, on the implementation part, I 2 will go into more detail on these. We are going to evaluate 3 the implementation programs at the utilities, what programs 4 they have in place, evaluate the implementation process,.
5 verify that they have implemented the recommendation, and 6 track the implementation process, and again, this is just an 7 overview. The owner groups and the staff will go into more 8 detail those aspects as we get into the program today.
9 MR. MICHELSON: In an overview sense, I noted the 10 fairly careful consideration of the human factor input to the 11 problem, but I also noted that practically no mention of the 12 external events aspects of these plants and whether or not ;
() 13 they are much more susceptible to possible core melt from 14 external events, for example, which may be by far the 15 overriding consideration.
16 MR. SIEGEL: When you say external events like loss 17 of off-site power?
18 HR. MICHELSON: No. External events like fire, like 19 pipe breaks outside of containment, things of this sort; these 20 are not, the pipe break aspect is almost treated in the PRAs, 21 treated rather superficially. The real flooding aspects are 22 not treated in the PRA. The fire is not treated in the PRA, 23 and this plant may be very susceptible to relatively modest
(} 24 fires in certain part of the plant. I don't know. And you 25 really haven't looked into it.
HERITAGE REPORTING CORPORATION -- (202)628-4888
i f 11 1 MR. SIEGEL: I, I think maybe the owners group can 2 answer a little better, but you know for the best, just l 3 speaking for myself, and I will hand it over to the Dick to 6 4 4 address and Bob perhaps--
4 5 MR. MICHELSON: The PRA experts tell me that i )
6 external events-may be the true out outliers, not the internal L
7 events.
t
> 8 MR. JONES: Right up front, we have made a decision 1
1 9 not to look at external events. We did concentrate on the l 10 basic design of the system and whether there were problems ,
11 with the basic designs, and the operational problems that we i 12 have seen at the plant. We did not expand the scope beyond l
i
() 13 those types of operational problems such as external events.
i i 14 Those are to a large extent being looked at on the seismic ;
15 programs. I forget the number of it, but USI on seismic, et !
16 cetera. We would feel those would be-the appropriate place to i 17 take care of it as opposed to this, this reassessment effort.
18 MR. MICHELSON: That should have been made clear in 19 the SER, that you are assuming that test will be taken care l
! 20 of.
i
- 21 MR. SIEGEL: It was not identified. l 1 I
, 22 MR. MICHELSON: May not be there. I didn't find it. I
! i l
23 MR. SIEGEL: Appendix R would cover the concerns i i
~
(} 24 that you had on fire protection.
l 1
25 MR. MICHELSON: That's a very simple answer to give
! )
HERITAGE REPORTING CORPORATION -- (202)628-4888 l
12-1 which is I'think basically not correct. It's, if you are 2 going to dig into and make great claims with the low 3 probabilities or show how you have corrected it, whsre the 4 probabilities are higher and then totally ignore which many 5 PRA analysts claim to be the taajor contributors to risk, we 6 haven't got the picture then. You know it is just not there.
7 MR. JONES: If I could just again in one sense 8 comment on it, when we did the risk assessment, what we were r
l 9 trying to answer to a large extent was number one, does the 10 operating experience demonstrate that the B&W plants have much 11 greater risks than other PWRs? That was the question, .first 12 question we try to look at.
13 The second question then was does the PRAs that have
, 14 been done which seem to indicate that B&W plant risks are not 15 out of line with the other PWRs, did they appropriately i
16 account for the operating experience? I mean were the l
17 assumptions valid in light of the operating experience that we I l
18 had seen over last few years?
19 And it was from that perspective we looked at the l 20 risk of these plants. We did not'again open them up to the 21 external events. We are really looking at operational history 22 and whether they were challenging our perception of the B&W 23 plant risk.
24 MR. MICHELSON: The key question in my mind at least 25 is whether or not these plants cre any more susceptible to HERITAGE REPORTING' CORPORATION -- (202)628-4888
13 1 external events than are other plants. I won't find that 2 answer in any of this material?
3 HR. JONES: No, you tron' t.
4 MR. MICHELSON: Although I find hints it might be, 5 because air systems and so forth of chese plants were easily 6 susceptible to damage of the air system and that's the easily 7 damaged system in case of fire because it is non-safety, not 8 protected. The same in case of earthquake, not protected; the 9 sarue and so forth. So I don't want to spend any more time on 10 it, but I really would like to make the point that I'm not 11 sure whether these plants are less safe or equally safe to the 12 others, or perhaps more safe.
() 13 MR. RUTHERFORD: Let me add a comment to what Mr.
14 Jones said.
15 We have not seen any evidence, although we have not 4
16 done any systematic review, but we certainly haven't seen any 17 evidence that the B&W plants are any more susceptible to 18 external events consequences than any other vender type.
19 MR. MICHELSON: You don't see it if you don't look l
20 for it in the PRAs. j 21 MR. RUTHERFORD: We have done PRAs on some of the 22 plants.
4 23 HR. MICHELSON: External event PRAs?
() 24 HR. RUTHERFORD: They have included external events 25 and certainly external events can drive the results, but where HERITAGE REPORTING CORPORATION -- (202)628-4888
14 1 we have seen problems with external events, it has been a -
4 O V 2 plant-specific issue, not a vender type issue.
a 3 MR. MICHELSON: Did Oconee do a fire PRA?
l j 4 MR. DAVIS: They did do a full scope PRA.
5 MR. MICHELSON: Including fire?
6 MR. DAVIS: Including fires; they did find' 7 vulnerability to flooding at that plant which was a !
8 plant-specific design situation that was subsequently 9 corrected.
10 MR. RUTHERFORD: That is correct.
- 11 MR. DAVIS
- I think we do have some evidence that 1
12 PRA plant capability for external events.
() 13 MR. CATTON: Just out of curiosity,could you put f 14 that last slide up? I found no reference to the MIST program !
15 anywhere. And your limited thermalhydraulic analysis was done i
16 by INEL using RELAP 5. It is my understanding that research -
17 is using TRAC to do the analysis of MIST. Isn't that a little 4
18 strange, or don't you use the results of the research, or they 19 are not applicable or something?
20 MR. JONES: Let me try to duck that last part of the 21 question, because that's not a characterization I would use 22 anyway.
23 No, we did not use MIST, not really in assessing the
() 24 overall program.
I 25 MR. CATTON: It shows I believe there are transients HERITAGE REPORTING CORPORATION -- (202)628-4888
15 1 that have been found.
{} 2 HR. JONES: That is correct. I can't remember 3 specifics. I think at least two of them certainly were on 4 HIST. When we did the evaluations which was with--the area 5 you are talking about is thermalhydraulic analyses, and the 6 sensitivity analysis stuff, that was done by NPR and your 7 on-site role, of those results. We were primarily trying to 8 ascertain one, the reasonableness of the results that NPR came 9 up with and their conclusions and also to look for any 10 inherent sensitivity which should be addressed via some of the 11 other programs in step such as the system reviews and to tell !
12 us something, that we have to concentrate a lot of effort in a i
13 specific system review activity to better control the steam ;
14 generator or something like that. That's what we primarily 15 were looking for--Were there any sense sensitivity insights 16 that we needed to deal with?
17 And from that extent, we really didn't use MIST 18 directly in the computer calculations to come up with and say 19 this given time is exactly right or you know off by 20 percent 20 or whatever, wasn't the overall thermalhydraulic response 1
21 indicative of the problem? I 22 There is a Basian reference to MIST and the issue of j i
23 interpretation of natural circulation where we do mention the i
24 HIST results as recognizing that interruption of natural !
25 circulation does not lead to unsafe condition.
I HERITAGE REPORTING CORFORATION -- (202)628-4888
16 1 HR. DAVIS: You did not use the code? You tuned 2 against the facility? The only reason I brought this up is 3 when I read the recommendations at the end of the INEL report, 4 I believe there were ten, of the ten odd recommendations, at 5 least eight nad the word simulator in them. To me that means 6 some sort of a reproduction of something with fidelity, and 7 here you have one of the major programs within NRC that is, 8 supposedly is achieving that for you.
9 Anyway, I just raise, wanted to bring that out for 10 some of the people who were here.
11 HR. JONES: Why Jon't we hit it again when we go 12 through the sensitivity discussion? But I know where you are
() 13 coming from on it.
14 HR. SIEGEL: Basically this is a summary of what wo 15 have done in the SER and SER supplement, neither one of them, 16 either one which conveys any new positions and none of them 17 contain any new requirements--only recommendation offered for 18 consideration.
19 The reason for that is that most of the areas that 20 we reviewed in this program were balance of plant type systems 21 where the staff does not have any requirements on it, and as a 21 result, any of the recommendations that we have in their 23 numerous recommendations throughout the report are couched in
() 24 the terms of offered for consideration. We encouraged the !
25 owners group to, or we would desire I should say the owners i HERITAGE REPORTING CORPORATION -- (202)628-4888
17 1 group to take our recommendations into consideration, and we
{} 2 would like them to go back and look at what they have done ir.
3 the past to see if they are applicable, but as of, but.in t
4 addition to that, anything that they do in the future we were I
hoping that they would, our intent is for them to use those 5 ,
i 6 recommendations when they are looking in the future for any r
7 events or transients that occur or any evaluations when. They 8 do their plant-specific, the utilities do their own 9 plant-specific implementation, the intent is that they will [
10 look at this report and incorporate the recommendations we 11 have into their evaluation when they implement the 12 recommendation.
13 The SER there were, I think the owners group is 14 going to go through the eleven major projects, and I have said 15 that the SERaddressed nine of the eleven major projects. It 16 really only addressed eight. There is one item which is 17 called review of emergency procedures that wasn't addressed 18 and it isn't addressed in this report because the owners group 19 hasn't completed their evaluation yet.
20 We addressed the owners group programmatic and 21 management actions impleme. station. We addressed additional 22 staff recommendations, recommendations and concerns identified 23 by staff member. The supplement addressed the integrated 24 control system, non-nuclear instrumentation, reactor trip O 25 initiating events review, acceptability of the emergency HERITAGE REPORTING CORPORATION -- (202)628-4888
18 1 feedwater initiation control type systems. That was one of
') 2 the items that the Commission asked us to address, a
3 We in addition went through, and again the owners 4 group has in their report of recommendation, they have 5 identified what they call high priority recommendations, or 6 key recommendations, and we will get into that later. In 7 addition to the ones they have identified as key 8 recommendations, the staff looked over the list and determined 9 some additional recommendations that we thought were high 10 priority.
11 We also discussed the staff impact assessment 12 proposed owners group recommendations. There were some open 13 items in the SER that we closed out in the supplement and 14 again we had additional staff recommendations results from our 15 review of the issues in the supplement.
16 We did not add- ' all these concerns identified by
, 17 the staff member in the Snx so we completed those in the 18 supplement, and there were two other Commission items 19 identified. I am trying to remember what they were. Oh, the 20 ones through, this is the U tube steam generator was an item 21 they asked us to address to verify the safety of the B&W 22 plants.
23 That was all I was going to say as far as 24 iLtroductory remarks, and I believe--I don't know if it is 25 Neil or Dick that is going to do it first. Is there any ;
HERITAGE REPORTING CORPORATION -- (2021628-4888
19 l 1 comments before?
n
. 2 MR. WARD: Yes. Byron, what is the, just in 3 summary, what is the owners group reaction to the 4 recommendations?
i j 5 MR. SIEGEL: Well, the biggest area that I guess
, 6 that the only comments we have gotten back were on the SER, 7 and the biggest issue in the SER was the human factors area.
8 We felt they ought to go back and look at,.use human factors 9 expertise to look at that, i i I j 10 The owners group sent us a letter in December of l 11 1987 saying that they didn't feel that that was necessary in i l .
] 12 the future and in transients they would look at human factors
() 13 aspects for any event that is, occured in the future in more 14 detail, but they didn't commit to or their opinion was it 15 wasn't necessary to go back and incorporate the
- 16 recommendations we had in the human factors area.
17 I am not sure this is the right time to discuss the ;
18 details of that. It would be better off later on in the, j 19 after they go through their whole program and we do, and in :
I 20 closing remarks to probably address--
21 MR. WARD: I wanted a summary.
i 22 MR. SIEGEL: That was basically the main issue in l
23 the SER. There were some additional recommendations. If you l
() 24 look at the SER, and the summary conclusions, we specifically 25 didn't identify with the exception of human factors--there may -
HERITAGE REPORTING CORPORATION -- (202)628-4888: l
20 I
1 be a few other areas--any of the recommendations that we
() 2 addressed throughout the report, primarily because taken by 3 themselves, we didn't think they were important enough to 4 warrant being addressed on a line-by-line item in the summary 5 and conclusions.
6 They were things we said it would be nice for them 7 to do, but we didn't feel strongly enough that they 8 necess /ily, the program was, that the program was--we were 9 concerned about the program because they hadn't done this, so 10 the intent was, the hope was that they would take them at face 11 value, and use them in future programs and also perhaps go 12 back like I mentioned before and look at what they have done 13 in the past and incorporate those, but we didn't feel that 14 strongly that we felt that it was necessary to make an issue 15 of it because you know, you are talking about meny 16 recommendations. Okay. -
17 MR. RUTHERFORD: Good morning. My name is Neil i
18 Rutherford. I'm from Duke Power, and I'm currently Chairman 19 of the B&W owners group Steering Committee. We certainly 20 welcome the opportunity this morning to meet with you and (
21 discuss the results of our safety and performance improvement 22 program. I think we are rightfully proud of this effort and 23 we feel like it will contribute to make the B&W plants top 3
24 performers in the industry. ;
25 Before we get into the main part of the agenda, I HERITAGE REPORTING CORPORATION -- (202)628-4888
21 1 would like to introduce the presenters that you will be seeing
( 2 today.
3 In addition to myself, we have Dick Skillmanfrom VPU 4 who was chairman of our SPIP management team. Dick is over 5 here.
6 We have Stuart Rose from Duke Power who served as 7 the project manager for the sensitivity study. Stuart is 8 sitting back there.
9 And Larry Stalter from Toledo Edison who is Chairman 30 of our INC Committee, and led the review effort for the ICS 11 and NNI system.
12 I would like to offer apologize for our Executive
() 13 Committee Chairman and Vice Chairman Wally Wilcus who is 14 Chairman of the Executive Committee, is entertaining 15 Commissioner Zech at Florida Power today and he could not be 16 here, and Dick Wilson also had a conflict that prohibited his 17 presence.
18 I would like to briefly go over an outline of the 19 presentation that you will be seeing today. First of all, we 20 are going to have a general overview of the program which will 21 include the definition of the issue, the information gathering 22 process that we went through during the program, and a summary 1
~
23 of our primary findings, and also we will mention the open
(} 24 items that still remain to complete the program.
25 When we met with the ACRS back in '86 originally,
- HERITAGE REPORTING CORPORATION -- (202)628-4888
22 1 the chairman of the subcommittee wrote Vic Stello a letter in 2 which he expressed some concerns about the program, and we 3 would like to respond to those items and make sure that those 4 are appropriately covered also.
5 Then we will get into a more detailed review of the 6 findings in the various systems reviews, the sensitivity 7 study, and the ICS/NNI and the risk review, and at that point 8 in time, we will also get into a discussion on the findings in 9 the SER related to ICS/NNI and the human factors area.
10 Once we go through the program itself, then we will 11 address recommendation follow-through and implementation.
12 (Slide)
() 13 HR. RUTHERFORD: As you are all aware, this is 14 certainly not the first time that we have been here to talk 15 about the SPIP program in the last two years. We have had 16 numerous interactions with the ACRS and the commissioners over 17 the past two years concerning the program. I am not going to 18 go through each one of those, but suffice it to say there have 19 been a number of interactions, the most recent of which was 20 the presentation that we made to the commissioners back in 21 August of '87.
22 (Slide) 1 l
23 HR. RUTHERFORD: The objective of today's meeting is
() 24 to make sure we provide you all the necessary information so 25 that the ACRS can conclude that the owners have been HERITAGE REPORTING CORPORATION -- (202)628-4888
- 1 1
23 1 responsive to the issues raised in Vic Stello's letter and 2 that we have adequately addressed the complex transient and 3 trip related safety concerns on the B&W plants.
4 HR. REED: Did you in the owners group interpret the 5 public notice and early papers that passed as asking for 6 thermalhydraulic evaluation and reassessment?
7 HR. RUTHERFORD: That was part of it. The 8 responsiveness of the B&W machine if you will was a part of 9 the concerns that we wanted to address.
10 HR. REED: Well, the words up there--
11 HR. RUTHERFORD: Certainly.
12 HR. REED: Of the objective don't seem to reflect
() 13 that, either.
14 MR. RUTHERFORD: Well, that all has to do with the 15 complex transients I think. A part of the complex transient 16 transient question is the design of the machine itself.
17 MR. REED: The trouble that I have is when I think 18 of basic design requirements, basic design, I think about such 19 things as the placement of the steam generators, high or low, 20 whether the primary system has natural circulation potential 21 or does it tend to defeat it, in fact I would have looked at 22 why are there two B&W basic design differences? Why does 23 Davis-Besse have high set steam generators, and why does the
() 24 other seven have low set steam generators? What are the 25 impacts on containment height and these, kind of those are HERITAGE REPORTING CORPORATION -- (202)628-4888
24 1 what I think of as basic design, not what was evaluated in the
[} 2 documents that I have.
3 MR. RUTHERFORD: Well, we didn't look at that 4 particular aspect of it because I don't think there has been 5 any questions or concerns raised about whether the B&W plants 6 have natural circulation capability, et cetera. That's been 7 proved out. The concerns were--
8 HR. REED: There haven't been any questions raised?
9 Not even from MIST?
10 HR. RUTHERFORD: No. I think HIST proved that 11 indeed natural circulation capability exists just like we 12 have--
13 MR. REED: You realize with the low set steam 14 generator you have a 30-foot plus riser that rises up to the 15 cold leg entry into the vessel? You realize that if you slug 16 that with dense water, you can block natural circulation?
1 17 MR, SKILLMAN: Can I make a comment, please? I am 18 Dick Skillman. Is this on? We proved, Glenn, we proved 19 natural circulation at THI-2. It worked and it worked 20 perfectly. It is not a matter whether it will or won't. TMI, 21 we removed decay heat at THI-2 for a year in natural 22 circulation even with the difficulties we had in the 23 instrumentation after THI-2. In Toledo, the raised loop 24 design will enhance that, but the lower loop 177--
0 25 MR. KEED: I would agree. I would like to see this HERITAGE REPORTING CORPORATION -- (202)628-4888
_ _ _ . _ . _ . - ._ = ..
25 2
-1 1 kind of comparison in the work that you did, that the raised 2 loop won't have natural circulation problems and that the 3 lowered loop can aet into natural circulation blockage, and I l I
4 think that this kind of thing, this basic design issue, has l
)
5 not been put to the forefront forthrightly, and I might remind j 6 you that the first pressurized water reactor that was ever 7 built had blockage of natural circulation from a riser, a l 8 vertical riser in the cold leg.
9 HR. SKILLMAN: THI had quite a bit of blockage. ,
1 10 MR. REED: I look at data we got down when it was i 11 down at 90 degrees Farenheit. You were in a period of i 12 blockage in TMI-2. Yes, f l
13 MR. SKILLMAN: From the stoppage of the reactor :
(:) 14 coolant pumps until we ceased to have enough thermal driving 15 head.
16 MR. REED: You melted the core, and there was a l
17 period when it was 190 degrees Farenheit in the cold leg !
l 18 riser, and the rest of the inside the vessel you are up at 104 19 degrees.
20 MR. SKILLMAN: Let me try to answer for Neil.
21 MR. RUTHERFORD: Let me respond here. I think those 22 types of issues were beyond the scope of this program, and 23 that we had not seen any evidence that we have any weaknesses 24 in these areas versus any other plant type. We were driven by O 25 the issues coming out of the complex transients that had HERITAGE REPORTING CORPORATION -- (202)628-4888
26 1 occurred on the B&W pisnts, and the design issues raised by 2 those complex transients.
3 Natural circulation, while important phenomenon, I 4 think was adequately proved out by HIST, the plant tests that 5 we have done, the THI experience, and it was beyond the scope 6 of this program.
7 HR. REED: The sacred challenge in this business is 8 to always and forever assure the removal of heat from the 9 core, be it power heat or decay heat. ,
10 Natural circulation because of loss of outside AC e 11 incidents, is a very, very important part of that as long as 12 you depend on the secondary system or the steam generators as
() 13 the path for that heat removal.
14 Now I quite frankly think there are some 15 thermalhydraulics aspects with respect to the low set steam '
16 generators that say that the path through the steam generators I 17 for decay heat removal is not well enough assured, and 18 therefore, the path and ultimate path must be, must be in 19 back-up reliable state. We will get to that later.
I 20 HR. RUTHERFORD: Well, I guess we have seen no 21 evidence to support that, and that's--
22 HR. REED: I guess that's where we disagree. I say l
23 the evident was with Three Mile Island 2 ano che, and the
() 24 evidence was at HIST. All right.
25 HR. RUTHERFORD: Suffice it to say our program did l
l l
HERITAGE REPORTING CORPORATION -- (202)628-4888
_ ___--,~ _ ._ __. . _ ._ . . _ . -
27 i
- 1 not address that, and it was s re-review of the B&W plants for
() 2 specific areas, not a general complete re-review'of the total i 3 design. I don't disagree with your comments about the ;
r 4 importance of the natural circulation capability, but that was 5 beyond this program.
- 6 MR. CATTON
- Natural circulation was beyond this 7 program did you say? !
8 MR. RUTHERFORD: Looking at natural circulation l 1'
9 capability, other than in what I would say a very high-level, 1
10 was beyond the scope of this program.
11 MR. JONES: If I could comment from the staff-- l 12 MR. CATTON: Could you tell me where the cut is when !
l ;
13 you say high-levol?
14 MR. JONES: The owners group set forth a program in 15 B&W 1919 which had a very specific objective to it. They 16 didn't, their program in and of itself was not sufficient for i 17 the staff. We did additional looking. We looked deeper in >
- 18 the reactor. We looked ourselves at previous NUREGs to assure ,
l i 19 that previous staff concerns were being addressed. We looked 7 i i j 20 a little further than they looked.
I 21 They looked at their specific problems, systems, and l
\
j 22 went into it, and as you will note, well, in general, we think :
) -
23 that the job they did was favorable. We went beyond that. [
4 24 The specific issue of natural circulation raised by Mr. Reed, l (:) 25 l you will not find a mention of it in the report. That's HERITAGE REPORTING CORPORATION -- (202)628-4888 i
28 1 clear. I mean it is not there. Not that it was excluded; it O
\_/ 2 wasn't given a comment in the report itself.
3 The reason it is not there is we don't see the 1 4 problem that Mr. Reed sees. We had, we discussed this issue 5 with the Subcommittee, the Decay Heat Removal Subcommittee
! 6 two, three months ago I believe. It was related to natural j i
- 7 circulation and thermal block in the B&W designs. I thought 8 we satisfied the Subcommittee at the time that there was no 9 decay heat removal problem related to natural circulation from J 1
10 the specific lower loop design of the thermal block, and you ,
11 know, it is just not in this program directly because previous 4
12 staff concerns, yes, it was raised, the issue about natural A 13 circulation and how you got HIST in two phase natural
(_)
i 14 circulation, concerns and questions. They were felt to be l
15 adequately addressed already or in the process of MIST- and we 1
] 16 haven't seen anything that raised a safety concern such that i
j 17 it needed to be directly considered within this overall 18 effort. Not that it was excluded per se--we didn't see s 19 safety problem that we fell that we had to address it in this I i
20 report.
21 HR. CATTON: But every one of your reports is i 3
22 initiated by discussing Three Mile Island and following Three 4 23 Mile Island, that was a major concern. It seems to me that '
1
() 24 you ought to put it to rest if that's your, what should have l 25 to happen to it right at the outset within your report; also HERITAGE REPORTING CORPORATION -- (202)628-4888
29 1 you are not addressing the so-called lessons learned in Three 2 Hile Island, i
- 3 HR. JONES
- It was put to bed I thought. .
I 4 HR. CATTON: Not in your report.
5 HR. JONES: Not in the report; ;
6 HR. CATTON: It should be in this report because you i
7 start your report by talking about Three Hile Island, and that i
8 brings all of these things back to mind. They should be put i 9 to rest there by reference or by something.
10 HR. JONES: Recognize this report to a large extent i
11 is an evaluation of the owners group program with transients
! 12 as the staff felt necessary to assure that there were no
() 13 safety problems.
j 14 HR. CATTON: I hear you, but if I took-- ,
i l 4
15 HR. JONES: The issue of natural circulation, you 16 are correct, there is no mention of it in the report. It was 1
l 17 done in one of the earlier NUREGs which came out after THI and i !
18 we did not repeat it.
19 HR. CATTON: If you take the lessons learned c
1 20 document adjacent to your SER, people are going to be very l 21 unsettled about what is left out.
i I 22 HR. JONES: We are willing to discuss any specific J 23 issue you have and try to address it here, if necessary. :
I
() 24 Maybe in this report it is not as complete as you would have ,
t 25 liked to have seen it, and there is nothing I can do about i 1
4 HERITAGE REPORTING CORPORATION -- (202)628-4888
30
- 1 that now, but all I, what I can say, to the best of my 2 knowledge, we have picked up things like natural circulation J
4 3 issue, and made sure there was no problem and then went on. f 4 If there are specific issues, we will be glad to discuss them 5 with the appropriate people on the given section of the report 6 or whatever and we will try to address it at that time to the I j
j 7 best of our ability.
i 1 8 HR. HICHELSON: The thing I found lacking in your !
l i j 9 safety evaluation, though, was not just the question of 10 natural circulation. I realize that's being put to bed in a 11 different program. That part didn't bother me. 1 i
t 12 What did bother me was a discussion of natural
() 13 circulation in terms of the uniqueness or additional j 14 complexity that might be involved in properly addressing that l
i l 15 circulation for the B&W plant versus say a Westinghouse plant. ,
16 I didn't find a feeling in here as to whethcr i
17 natural circulation was so natural that it didn't require any 18 operator intervention or operator intervening inappropriately.
19 it would not necessarily disrupt it. It was no different than i
- 20 any other PWR, and I didn't find a discussion whether there is 1
21 any difference in maintaining the natural circulation for this l 1
I 22 plant versus other plants. If there is, it ought to be 23 considered as a human factor.
i 1
(} 24 MR. JONES: Actually it is discussed at least with 25 reference to the small break LOCA transient within Section 5 f i
l HERITAGE REPORTING CORPORATION -- (202)628-4888 t
31 ;
1 on the sensitivity evaluation. We note very specifically that
() 2 the B&W plants, on line, the U tube plants, will not, can ;
I '
3 interrupt r.atural circulation long term and it may be more ,
4 difficult to accomplish orderly shutdown of the plant and we l a
J 5 looked at procedural aspects say that there is guidance !
J 6 provided to the operator, there wasn't specific safety concern 7 with respect to core cooling, and so we recognized the ,
8 uniqueness and neted that uniqueness, but said there wasn't a 9 specific problem that we could tell associated with it.
10 MR. MICHELSON: What I found lacking was a 11 discussion that first of all referenced me back into whatever 4
12 was coming out of MIST or wherever, where the requirements for ,
(} 13 natural circulation were established, including the
] 14 operational steps, if any, so that I could go back to see how ,
15 that worked. I realize the statement was in there but no ,
i l 16 connection back to the MIST work or some other work that )
l 17 established that connection, j l
j 18 MR. JONES: As I said, I am willing on any given i
I 1
19 item and the other staff members in their specific area I'm l 20 sure can address the types of concerns you are l 21 addressing--maybe not to your satisfaction, but to the best of 22 our ability.
j 23 MR. MICHELSON: As a generalized statement, is it 24 more difficult then to obtain, are there more, does operator 1
j 25 training level have to be higher, and understanding have to be
! HERITAGE REPORTING CORPORATION -- (202)628-4888
32 1 higher and so forth for a B&W versus say a Westinghouse?
2 MR. JONES: Not normally initiate natural
(~)T
\-
3 circulation.
4 HR. REED: I would like to take exception to that 5 and disagree with what you said, Carlyle. The documents that F I received that I think covered the waterfront, don't clearly--they point out all kinds of comparisons between B&W ad Oconee and Calvert Cliffs and Robertson but they avoid 9 the very basic design issue of decay heat removal and natural 10 circulation of each systems on loss of AC. They don't clarify 11 it.
12 Now what ought to have been in there is the fact g3 13 that the U tube steam generator plants get their driving head G!
14 from the core delta P, from water heating rising up through 15 the core, and that the B&W system gets no, in both their 16 designs, gets no driving head from the core. Delta P and 17 temperature rise.
I 18 They all ought to point out that the B&W very
]
l 19 clearly hits its only driving head for natural circulation 1
20 from the external component known as the steam generator. 1 l
l 21 These are basic design issues rather than a lot of little nuts 22 and bolts. I would like to see diagrams in a clear discussion i 1
23 of these differences. I don't think the operators in these 24 plants understand it, that where the driving head comes from, 25 why they must safaguard the actual level of the water in the l l
I HERITAGE REPORTING CORPORATION -- (202)628-4888
33 1 B&W system in the steam generator, very, very carefully, and
~
2 in fact, I am not so sure they can always safeguard that level 3 at that point.
4 MR. RUTHERFORD: I would like to take exception to 5' that. I think the operators do understand that. We have 6 enhanced our training over the last few years to point out
. q 7 some of the items that Mr. Jones has alluded to as far as 8 interruption, and what the operator might see on these ;
9 machines.
10 I have got just to take exception to those comments. j 11 I don't believe that is a correct presentation of the present 12 state of operator understanding on the B&W plants.
13 One other comment, too, as far as decay heat removal O 14 and in the largest sense, and we can touch on it later, too, 15 but it is true there are differences between a Westinghouse 16 plant with a U tube generators and a B&W plant with once 3 17 throughs, different characteristics. It is going react 18 differently, but there are positives and negatives in that ,
19 reaction.
20 If you look at decay heat removal in total, 21 certair.ly the B&W plants have better capabilities in terms of 22 feed and bleed. That is obviously one of the ways to remove 23 decay heat if natural circulation fails. As we said, we don't ;
24 think we have any problem whatsoever with natural circulation, O 25 but that is in the total risk.
1 HERITAGE REPORTING CORPORATION -- (202)628-4888
34 1 MR. REED: We will get to that feed and bleed l
~- 2 capability later with your single internal pilot operated ;
l 3 relief valve which probably won't operate, but I don't think 4 you are here for that purpose. You are here for sort of an I l
l 5 overview and we will get that feed and bleed capability that l I
6 you claim. l 1
7 HR. RUTHERFORD: It was specific to Davis-Besse as ,
1 8 far as feed and bleed capability.
9 HR. REED: What I read is there is a single PORV in 10 each plant.
l 11 MR. RUTHERFORD: The high pressure injection pumps 12 are capable of pumping against--
D)
(, 13 HR. REED: They run the pressure up and up at a time 14 you ought to be running it down a little bit.
15 MR. RUTHERTORD: Well, let's move on with the l 16 presentation.
17 (Slide) l 18 HR. RUTHERFORD: Before I turn the program over to 19 Dick Skillman, I would like to just say a few words about the 20 SPIP program itself.
21 When the Stello letter was issued back in early 22 1986, the first thing we wrestled with was somehow 23 quantitatively defining the issues. There were a number of
() 24 qualitative issues in the letter, but we wanted to get to 25 something that we could measure, something that we could set HERITAGE REPORTING CORPORATION -- (202)628-4888
35 1 improvement goals on and measure our improvement over the
() 2 years.
3 Also throughout this process, there has been a 4 strong executive level review commitment to the program and to 5 the goals that we initially set. This is not only in terms of 6 monetary support, but also in the support of our, each of our 7 organizations to making sure the program was manned 8 adequately, and supported by the technical people in each of 9 our organizations.
10 The bottom line here is we wanted a program that had 11 some tangible results for us in terms of improving the plant 12 safety and performance.
13 In addition, we wanted to make sure that not only we
(')T 14 were satisfied by the scope of the program, but the scope also 15 adequately addressed the issues that had been raised by the 16 staff.
17 While the program certainly does have some economic 18 benefits to the owners, safety was the primary orientation 19 throughout execution of this program, and I think you can see l l
20 as we go through the systems reviews that we have done, and l
l 21 the other type reviews that we have done, that that indeed was )
l 22 our primary goal. l 23 To execute the program, we not only looked at our l 24 historical data base, but we also looked at potential plant i
25 performance that heretofore had not been exhibited in natural i
HERITAGE REPORTING CORPORATION -- (202)628-4888
36 1 events. We were fortunate in this area that we have got.a 2 very good data base for historical purposes from our TAP 3 program where we have been systematically collecting data over 4 the last few years, and that did provide a sound basis for 5 proceeding with the program.
6 The next point I would like to touch on is our use 7 of independent support and technical overview in the program.
8 Certainly the owners individually and collectively provided a 9 lot of manpower and support to the program. B&W did a lot of 10 the technical work and support, but we also felt the need to 11 have some independent review and oversight in the program, and 12 we will cover exactly the mechanics of how we did that later,
() 13 but not only the outside support from the contractors that we 14 used, but the NRC staff, ACRS comments, et cetera, were also 15 helpful in sharpening our focus as we went through the 16 program.
17 As I mentioned, we are very proud of the results of 18 the program, and we already see instances where plant 19 improvement has resulted from implementing recommendations, 20 and we are confident that as we continue implementation of the 21 recommendations, that we will see continued improvement in 22 safety and performance.
23 And certainly in closing, SPIP, the SPIP program
() 24 complements and reinforces other industry safety and 25 performance improvement, improvement initiatives that are HERITAGE REPORTING CORPORATION -- (202)628-4888
37 1 ongoing in the industry today. I think as you look across the f%
f,
) 2 industry, we are seeing improvements in terms of trip 3 reduction, et cetera, but we think this program is going to 4 enable the owners to be in the forefront of that improvement.
5 Okay. At this point in time I would like to turn 6 the program over to Dick Skillman, who is going to offer a 7 summary of the program, and a summary of the results.
8 Any additional questions at this time?
9 HR. SKILLMAN: Good morning. I am Dick Skillman 10 from GPP Nuclear Corporation, and I was the SPIP chairman. I 11 was head of the management team comprised of the B&W owners l l
12 personnel who did the review of the Babcock reactors. l
{} 13 (Slide) 14 MR. SKILLMAN: I would add to one comment that Neil 15 made regarding the program. It was a large program. It was 16 comprehensive. It is continuing today, as we are here tode',
17 there are people at each of the plant sites working on 18 recommendations that now number well over 200. The owners 19 have invested in the neighborhood of $10 million in this 20 program. We have spent over a hundred man-years, so by any 21 measuring stick, this program was substantial. It is a 22 one-of-a-kind program and we believe that we have interrogated 23 these plants to a level that other plants have never been 24 interrogated.
25 What was the issue which we were dealing with?
HERITAGE REPORTING CORPORATION -- (202)628-4888
38 1 Several issues have been spoken at the table here. We were
2 not trying to determine whether or not a Babcock' plant is 3 better than or worse than Combustion plant or Westinghouse 4 plant. We weren't trying to do a basic performance comparison 5 between the plants, and this was not a commercial review.
6 Two events preceded the Vic Stello letter.of January 7 24 of 1986. Those two events were the June 9 Davis-Besse 8 incident, and the December 26 Rancho Seco. In both of those 9 cases, the appearance was the plant was too hard to handle.
10 There was too much operator energy involved in recoviring from 11 those transients, and as you think now Rancho Seco is just 12 coming back. Davis-Besse has been on line for sometime, but 13 those two incidents coupled with the perception of THI-2, !
O 14 lessons learned, operator action, core melt, caused Mr. Stello 15 to say something should be done about these plants, we want 16 these plants reassessed, and that's the backdrop'for SPIP, two 17 mid-late '85 incidents involving a lot of operator energy, 18 extreme plant conditions, and the perception that the Babcock 19 plants behaved inadequately, so the issue that we tried to get 20 our arms around is the issue of transient complexity. )
l 21 Not only were the post-trip behaviors of the plants 22 complex, but it appeared as if the plants were tripping too l
23 often. Consequently we said we wanted to define these complex 24 transients, using some parameters or some measuring sticks O 25 that would let us understand the transients in detail. We i
1 HERITAGE REPORTING CORPORATION -- (202)628-4888 l
39 1 used six parameters. Those are basically the ATOGI
) 2 parameters. This let us quantify the issue so that we could 3 determine where we should concentrate our review efforts, and 4 in pasring we would comment that interactions with the NRC 5 staff, with the ACRS, with the commissioners, with other l
6 people really let us focus in on where we wanted to go with l
7 the safety and performance improvement program.
8' Some definitions--trips, a trip is a situation where 9 you have an unplanned de-energization of the control rod drive .
I i
10 control system power breakers, thus resulting in dropping the l
11 rods into the core. The salient word is unplanned, and power
- I l
12 and transients, the aggregate dynamic plant behavior following
() 13 a trip. What we are trying to focus on there is the entire 14 range and spectrum of activities that the operators are .
15 obligated to attend to after the plant is tripped. It is 16 combination of energizing systems, making post-trip systems 17 function, and getting the plant back to where the plant is 18 supposed to be in a near and far-term timeframes.
19 MR. MICHELSON: Before you leave that slide, it 20 appears that the emphasis of this study is on transients, not '
21 necessarily on accident responses.
22 So that part is fine. I understand. Now when you I
23 are dealing with transients, though, you have to be quite 24 concerned about the state of the plant at the time of the 25 transient because the transient may be initiated by a number i
HERITAGE REPORTING CORPORATION -- (202)628-4888 j
40 1 of different kinds of events outside of the, or inside the 2 primary containment which are not of themselves what we 3 classify as accidents, but if the transient is not adequately 4 responded to, it results in a core melt anyway, so I get quite 5 interested in how you can initiate plant transient outside of 6 containment, and then I get back to the question of for 7 instance, have you considered pipe break outside of 8 containment as a transient initiator? It is a, some people 9 call it internal event, some people call it external. I can't 10 get uniformity of opinion.
11 Are you including pipe break as an internal event, 12 outside of containment, included as a transient initiator?
() 13 MR. SKILLMAN: We included anything that would trip 14 the plant as a plant trip initiator.
15 MR. MICHELSON: Pipe break will trip plants, a lot 16 of external events which you have already I think told me you 17 didn't include. I think I already got the answer to that. I 18 am just asking, though, is pipe break an internal event or 19 external event for the purposes of this study?
20 HR. SKILLMAN: Regardless of whether it is internal 21 or external, Carl, the issue is what does the plant do after 22 the plant trips?
23 MR. HICHELSON: In view'of that initiator existing ;
() 24 as well.
25 HR. SKILLMAN: We did not specifically address HERITAGE REPORTING CORPORATION -- (202)628-4888
._ . _ _ --._ _ .- . _,- . _ . . . . . , . . _ . . , . . _ ~
- . _ _ ~ . _
._. __. . _ . ~ - _ - . _ .._ _
41-i 1- initiator of' Class A versus initiator of Class B versus 2 initiator of Class C.
1 3 MR. MICHELSON: Effects of.the pipe break, if indeed i 4 it is transient initiator all right, but you didn't chase down
., 5 what-else the pipe break was causing while-you'were trying.to 6 mitigate this transient?
7 MR. SKILLMAN: We did not for the reasons that Bob 8 pointed out.
9 MR. MICHELSON: -Pipe break is not in your study 10 really either in that respect?
11 MR. SKILLMAN: That is correct.
12 MR. MICHELSON: I want to make sure because some 13 people don't consider those external events. They call them 14 internal events--clarifying the scope. ;
15 MR. DAVIS: Holp me out a little built with your 16 definitions. On transients, there is three categories listed 17 in your report 1919--A, B and C. Which of those are you 18 calling complex transients?
19 MR. SKILLMAN: The Cs and the B and the significant 20 Bs, and I will get to that if you would like in a few minutes.
21 I can explain that in detail.
22 The reason I put this up here is we had a youngster 4
23 in one of these sessions where the definition of transient was 24 thought to be a vagrant who slept beyond the cooling tower.
.O 25 What we talk about when we talk about transients is -
HERITAGE REPORTING CORPORATION -- (202)628-4888
42 1 the aggregate dynamic behavior, the interaction of ICS, c
(_) 2 emergency feed secondary plant relief, all of the systems that 3 are called upon to behave properly to make the plant settle.
4 We are trying to unravel whether or not the post-trip 5 behavior, that is, the complex transient that was experienced 6 at Davis-Besse, the complex transient that was experienced at 7 Rancho Seco, have common elements, if all Babcock plants have 8 problems like we are, those that were experienced in those 9 particular incidences, and what do we need to do about them?
10 MR. CATTON: A question about the transients--it 11 seems to me that if you are going to look at transient 12 behavior in your plants, you have to understand them.
() 13 MR. SKILLMAN: Yes, sir.
14 HR. CATTON: Yet B&W designed the MIST facility, and 15 it is my understanding that they can't reproduce some of those 16 transients in the plants on the MIST facility. That leads me 17 to the conclusion that gee, you don't know how to, you don't 18 understand these transients well enough to reproduce them.
19 MR. SKILLMAN: I don't think that reproducing the 20 transient is a necessity.
21 MR. CATTON: Hold it. You mentioned a number of 22 things that are supposedly built into this facility.
23 MR. SKILLMAN: Into which facility?
24 MR. CATTON: The MIST facility; it was built for a V(~T 25 purpose, was built to reproduce these kinds of things that you HERITAGE REPORTING CORPORATION -- (202)628-4888
~
43 1 want to study. If that effort is somewhat unsuccessful, that
() 2 leads me to believe that you really don't understand your 3 plant that well.
4 Could you maybe help me with this? Where am I 5 running wrong in my thinking?
l 6 HR. SKILLMAN: Neil, would you like to make a l 7 comment?
8 MR. RUTHERFORD: One comment, certainly, that the 9 MIST facility was not, the design of the MIST facility was not 10 intended to exactly duplicate what actual plant performance 11 might be. It was meant to examine certain phenomenon that 12 might occur in the B&W plants, and to verify that we can
{} 13 adequately benchmark and mode 1~those phenomenon in a real 14 plant, so we have never I guess characterized the MIST 15 facility as being a simulator as such.
16 MR. CATTON: Certainly not a simulator, but I, it is 17 my understanding that the counterpart tests within MIST 18 weren't all that people expected them to be. Can you explain 19 this? If you can, then that puts it together as you suggest.
20 MR. SKILLMAN: I cannot explain it, but I would like 21 to try to take a different tack in answering your question.
22 Clearly you have some concerns regarding the thermalhydraulic 23 comparisons between MIST and the other activities regarding 24 Babcock plants.
25 HR. CATTON: Somehow the whole thing is brought HERITAGE REPORTING CORPORATION -- (202)628-4888
44 1 together with the code somewhere. .You guys are developing.
2 RELAP 5. The NRC is tuning TRAC against the MIST facility, 3 and then the staff is using RELAP 5 to look, look at the 4 thermalhydraulics and how it interacts with other things. You 5 either understand it or you don't. If you don't, I am not 6 sure what good all this code study does you, which is the 7 basis of tracking the transient.
8 MR. SKILLMAN: Let me say this.
9 MR. RUTHERFORD: One of the big purposes of MIST was 10 to benchmark our codes so that we could understand, make sure 11 that we did understand what might actually occur in a real -
12 plant.
13 MR. CATTON: But benchmarking codes in modern jargon .
(
14 is tuning codes. You still have to make that big step from j l
15 the facility to the plant, and if you can't explain why the '
16 counterpart test didn't turn out as expected, somewhere you 17 are lost. Your tool is flawed. I'm sure we are going to get 18 into this more later. Why don't you--
i 19 MR. SKILLMAN: I would like to make this comment. 1 20 The Stello letter of January 24 made some assertions regarding 21 the sensitivity of the B&W owners group plants and made some l
22 comments regarding the behavior of the plants.
23 Stuart Rose will talk later about the sensitivity 24 study which was the portion of the program performed for us by )
(:) 25 NPR, steered by Stuart as a project manager, to address the i
l l
HERITAGE REPORTING CORPORATION -- (202)628-4888 1
45 1 thermalhydraulic sensitivity issues, a number of them.
, (f 2 MR. CATTON: That's the place for me to raise these 3 questions.
4 MR. SKILLMAN: The issues pertaining to plant 5 behavior, and we went to some ordinary, ordinate detail to 6 understand the behavior, is the subject of a large portion of 7 this presentation today. What we found is what, what most 8 people perceive may not really be the. case in terms of Babcock 9 plant behavior and in so far as decay heat removal, ability of 10 the operator to get a hold of the plant, the way in which the 11 plant maneuvers, ramps and so on.
r 12 Transient, the aggregate dynamic plant behavior 13 following a trip, that's an important point because we do
[}
f 14 classify those transients so as to come to some understanding 15 of the sensitivity of the Babcock plants. I i
~
16 (Slide) 17 HR. SKILLMAN: The objective of this program was to ,
i 18 improve the safety of the B&W owners group plants, to reduce :
19 the number of trips and complex transients which occur, and 20 ensure appropriate response for those trips and transients 21 that do occur, and our goal at the-beginning of.tne program 22 was to get the average per plant trip frequency to less than 23 two per year. We are currently at about 1.7 per year, and the 24 new goal as of last September is 1.4 per year.
25 Those numbers might not seem like great changes, but i
HERITAGE REPORTING CORPORATION -- (202)628-4888
46 1 when the program was started, we were tripping at a frequency 2 of about four trips per plant per year, so this goal was a 3 result of having of the then existing trip frequency and we
~
4 are doing much better today and we expect to do better still.
5 We are about 1.7 today.
6 MR. DAVIS: When you compute those trip frequencies, 7 do you use only operating units or do you use all B&W units I 8 regardless of the--
9 HR. SKILLMAN: Operating units, and we subtract out, l
10 out of that the units that are not operating.
11 MR. DAVIS: Thank you.
j 12 HR. SKILLMAN: Am I correct? By the end of 1990,
() 13 the number of complex transients classified as parameters 14 category C will be less than .1 per plant per year or one in 15 ten plants per year.
16 When the goals were established,.we were running at 17 a complex transient frequency of .3 per year, so it is a 18 three-fold reduction in that complex transient frequency rate.
19 We are approximately .24 per plant per year at the current i 20 time. Our goal is still .1.
21 MR. ETHERINGTON: Is frequency of complex transients 22 faster than the frequency of the decreasing frequency of trip?
23 MR. SKILLMAN: Say again, please?
() 24 MR. ETHERINGTON: You have the increasing in the 25 trip frequency, and yoit have the decrease in the complex 4
HERITAGE REPORTING CORPORATION -- (202)628-4888
47 1 transients.
{J N
N 2 Are they the same ratio on a--
3 MR. SKILLMAN: No, sir. The trip frequency has 4 dropped much more significantly in the same time period than 5 the complex transient frequency.
6 MR. ETHERINGTON: So you could say that you haven't 7 really improved on the complex transient situation?
8 MR. SKILLMAN: We have improved except for a single 9 category C transient experienced at about Christmastime. We 10 have had ten C transients. We had ten C transients go through 11 Thanksgiving of last year beginning January 1 of 1980, and we 12 had the lith category C at about Christmastime. When you
() 13 factor that lith complex transient based on a three-year 14 average to the number of plants then operating, the complex l
15 transient frequency is at about .22 to .24 whereas the trip 16 frequency has dropped from 4 to 1.7.
1 l
17 The real issue is a single complex transient gets l l
18 magnified because it stays in our data base for three years, )
i l
19 so we are running at about .24 now waiting for some of the
)
20 previous complex transients to fall off. l l
21 (Slide) '
l 22 HR. SKILLMAN: This is the SPIP program. We have an l l
23 information gathering phase well behind us now; integration i
/~) 24 phase, also well behind us now; and the implementation phase, l
\/
l 25 which we are deeply involved in. What I would like to do is j i
HERITAGE REPORTING CORPORATION -- (202)628-4888 l
48 1 to talk about the information gathering phase. I will keep 2 the slide out so you can see our program through this 3 presentation.
4 (Slide) 5 HR. SKILLHAN: What we were trying to do in the 6 information gathering phase is to determine the areas in the 7 B&W owners plants that were appropriate for improvement. This.
8 related to systems. It related to thermalhydraulic 9 sensitivities. It related to relative core risk, and in some 10 questions of design for the sensitivity study, and so what we 11 did is we performed a broad and comprehensive search for 12 problems both in the NSS and the BOP. That's important
() 13 because what we found as we dug more and more deeply into our 14 data is that the balance of plant, its relationship to the 15 core, to the reactor coolant system, particularly in the near 16 term timeframe post-trip, was a portion of the plant that we i
17 really hadn't spent as much time looking at, making sure that #
18 it was integrated with the other portions of the plant as we -
19 should have. And you will see that as we go on.
20 What we did is we reviewed our transient assessment 21 program data. This is a program that the B&W owners have 22 whereby we monitor every trip and every transient, each one i 23 that has occurred since January 1 of 1980, so that we
() 24 understand how these plants are behaving and why they are 25 behaving the way they are behaving. )
i i I HERITAGE REPORTING CORPORATION'-- (202)628-4888 i
49 1 We reviewed our systems and some components within
_/ 2 the systems. We interviewed our operation and maintenance 3 personnel, and we reviewed other pertinent data related to the 4 B&W plant performance history.
5 We employed an outside consultant to perform the 6 sensitivity study which is a relative sensitivity assessment, 7 b&W type plants, 177 lower loop, versus Combustion plants, 8 Westinghouse plants, Westinghouse plant, singular, and we 9 performed a review of the PRAs germane to the Babcock plants.
10 We didn't do a PRA per se. We reviewed existing PRAs as Bob 11 Jones said earlier to determine whether or not the category C 12 transients that we had had were properly reflected in those
() 13 fault trees.
14 HR. REED: Why don't we go back to the complex 15 transients a minutes, And the reply you made to Harold 16 Etherington, and go back in history a little bit to the first 17 program that the B&W owners group brought in which was called 18 I believe the stop trip program, which the ACRS rejected, and 19 other people did as that wasn't what was intended and what we 20 needed to know about the B&W systems.
21 Now to me as I looked at the stop trip program, that 22 didn't mean anything, because the B&W reactors had I believe 23 at the time like 1986 or '85, prior to that, had less trips
(} 24 than the other reactors, so that that's easy to say we have 25 less trips.
HERITAGE REPORTING CORPORATION -- (202)628-4888
50 1 But now we get to complex trips. To me, complex g- 2 trips are either undercooling or overcooling events, correct? l NS1 l 3 MR. SKILLMAN: Yes, sir.
4 MR. REED: Those are certainly the biggies.
5 Undercool, overfill, overfeed which can be unusual in 6 themselves through those basic three categories.
7 Now let me try to bring this around that circle. I 8 am going to read from a report by B&W owners group of May 1987 9 called final report, page 3, 3-6. It says here following the 10 loss of all main feedwater flow, either with or without forced 11 circulation in the reactor coolant system, that is loss of AC, 12 the steam generator level should be controlled to the desired 13 heights before the lower tube sheet to provide an adequate O 14 heat transfer surface for water-to-water heat removal and to 15 promote adequate natural circulation flow rates."
16 Now those words to me say there is some problem in 17 promoting adequate natural circulation flow rates, and there 18 has got to be something done in that single driving head 19 location of the steam generator. There has got to be 20 something very touchy done to make, start and to cause to 21 endure natural circulation on loss of AC. Now that to me is 22 certainly a complex transient to think about.
23 MR. SKILLMAN: You interpret that as being, 24 identifying a problem. We identified that as being words
(')
\/ 25 that, that communicated requirement for the emergency
{
HERITAGE REPORTING CORPORATION -- (202)628-4888
51 1 feedwater system.
2 MR. REED: In other words, if that emergency i 3 feedwater system control-thing which is very' sensitive and 4 gets into problems--
5 MR. SKILLMAN: It is safety grade.
6 MR. REED: But look at history. Look at history of 7 your, of your feed system and its control and all the 8 problems, and the oysters and other things that can happen !
9 down the line and the check valves that can happen and all the 10 complexities of providing feedwater to a desired level from 11 the steam generator side, and there are two levels, depending 12 on which accident is there. It sounds to me to be a very l 13 complex situation which I think should have been reviewed in
- 14 great depth.
15 MR. SKILLMAN: It was. It was reviewed in our 16 emergency feedwater review, and we will talk about that in a 2 l 17 few minutes. I 18 Information gathering, a major piece of the program ;
19 to get if I can say it this way navitation points so that we '
i 20 knew where to go in terms of the rest of the program. <
l 21 (Slide) j l
22 MR. SKILLMAN: Let me give you.some detail on )
l 23 information gathering. We assessed the, we reviewed the TAP 4
24 data. Sir, this is the question that you asked, what about O 25 these complex transients?
HERITAGE REPORTING CORPORATION -- (202)628-4888
52 1 We reviewed and sorted the TAP data transient 2 assessment performance data to better focus on the areas where 3 improvement is needed. As I said, every time a Babcock plant 4 has tripped since January of 1980, our teams or the utility 5 teams or the Babcock teams have gone in to describe what 6 happened, to find out what happened prior to the trip and what 7 happened after the trip, and with that data base, there have 8 been 258 trips since then. We have a tremendous amount of 9 information that says when this happens, and this happens, 10 this result occurs, and by weaving our way through that data, 11 we were able to understand the areas that needed improvement. ,
12 We took the TAP data then, 220 trips when we did M
Qp 13 this in early '86, and we defined specific measurable 14 parameters to grade the complexity of transients. You will 1
15 see which ones. I will tell you in a minute, and then we--
16 MR. CATTON: I was going to say something else.
17 Have you ensured that your calculational tools will reproduce 18 these transients and that they are properly put on to the 19 simulator for training?
20 MR. SKILLMAN: Yes, sir, and Stuart Rose can talk 21 about that in terms of the sensi:ivity study, in terms of 22 analysis, and the computer, the simulators throughout the 23 Babcock units do in fact fairly reflect this.
() 24 MR. CATTON: You have demonstrated your simulators t
25 indeed have the proper fidelity?
r HERITAGE REPORTING CORPORATION -- (202)628-4888
53 1 MR. SKILLMAN: The short answer is yes, sir. Yes, 2 sir.
3 MR. CATTON: I will pursue that a little more 4 whenever you have, what you pointed to is up here.
5 HR. SKILLHAN: Using the graded transient TAP 6 reports, then we were able to identify the specific areas to 7 go to further work on.
8 MR. MICHELSON: Of the over 200 events that you 9 analyzed, did you sit down and make out a table of what was 10 the, shall we say the root cause of initiator of that 11 particular transient, be it human error, external 12 considerations or internal failure? Those kind of categories?
() 13 Going to show us that later?
14 MR. SKILLMAN: I would like to amplify this 15 question, Carl, that did we sit down and really go through the 16 root cause to determine whether it was a piece of equipment or 17 an operator, and the answer is yes, sir, we did, in detail.
18 MR. MICHELSON: Are you going to show us a little 19 simplified graph showing the distribution of those initiators?
20 HR. SKILLMAN: I can. I hadn't planned to, but I 21 will.
22 MR. MICHELSON: Either that or tell us in words. It 23 is probably buried in the report, although I had trouble
() 24 finding from time to time exactly where, but it is fairly 25 important consideration to see what is initiating these HERITAGE REPORTING CORPORATION -- (202)628-4888
~54 1 transients.
2 MR. SKILLMAN: Which categorization parameters did O, ,
3 we use to determine which transients are more significant? We 4 used six parameters--reactivity, reactor coolant system 5 pressure, temperature and inventory, those three, and i
6 secondary side, pressure and inventory; those six.
7 reactivity, three. And primary, two on secondary.
8 You might say, well, okay, what is an A? An A-l 9 response, everything does what it is supposed to, kind of no '
-i 10 never mind. Plant behaves properly after the plant trips. No I 11 passion; nobody running around; everything cool in the control 12 room.
13 B, the plant response slightly exceed the expected O 14 range in one or more parameters, but it doesn't overshoot. In ,
15 other words, it is a more energetic response, but it isn't one 16 where there is a quandry in the control room.
17 And the C 1s where you reach limits which require 18 safety system and extensive operator action to mitigate the I
19 transient.
b 20 Does anybody need to understand what the parameters 21 are? Let's go on, t
22 MR. DAVIS: Is it true all trips are in fact !
23 transients? ,
l 24 MR. SKILLMAN: Generally, yes. The post-transient I
\
() 25 response, the behavior, of an, in an A category, is the. plant !
i l
HERITAGE REPORTING CORPORATION -- (202)628-4888 !
._,..-.,-,._-,,c.,-,_=
,,m,,---, w,---, v rr , - . - - , - = v vr- v--, *--e--------- --=-vw-- v~
55 1 just cools down without perturbationi. All the parameters r
( )'/
2 stay where they should, so yes, a transient always follows a 3 trip.
4 How many have we had? Through May of 1976, had 258 5 transients, trips and transients. No more response; 247 6 abnormal; with the C level. 11.
7 (Slide) 8 MR. SKILLMAN: Carl, you asked did we do a root 9 cause? We did. this is back-up it is not in your package.
10 As Glenn pointed out, transients fall in three categories--
11 overcooling, undercooling, and overfill, overfeed. We broke 12 them down further into category C, quasi-C, which is a single
() 13 or a, maybe a second parameter exceeds the C range, the plant 14 does not trip, and a significant B. Significant B is a B 15 transient. We haven't gotten into the C category, but where 16 there is a fairly more strenuous response than an A.
17 At the time of this writing, we had 237 trips.
18 Those that--this is pre-trip behavior. And what we did is we 19 said all right, what system or systems were involved in the 20 pre-trip plant character? And we are saying the emergency 21 feedwater was the tripper in two out of 237. Main feedwater 22 was the main tripper in 56 out of the 237, so if one were to 23 take a look across the bottom part of the chart, you would see
() 24 a signature for a fingerprint of the Babcock pre-trip history 25 or the Babcock tripper history, and what it says is we want to HERITAGE REPORTING CORPORATION -- (202)628-4888
56 1 knock out trip, do something about your turbo generator.
() 2 MR. MICHELSON: The thing that I asked was not this, 3 although this is important, too. What I asked is how many of 4 those trips are caused by somebody jarring an instrument, for 5 instance? And you will identify the system in which the 6 instrument was jarred out. I would like to know whether or 7 not that was a physical jarring of the instrument or 8 inadvertent actuation of fire protection, or it just a failure 9 of component within the control system itself?
10 HR. SKILLMAN: It is something like 11 out of 127.
11 These are bona fide system trips.
12 MR. MICHELSON: You mean it was nothing external to 13 the system that caused it?
[}
14 HR. SKILLMAN: To the best of our ability to unravel 15 it.
16 HR. MICHELSON: Not less the air--total failure? !
l 17 That is a separate?
1 18 MR. SKILLMAN: This is probably a person. The 1 l
19 answer is-- l 20 MR. MICHELSON: All the others are strictly internal i
21 equipment failures, not prompted by external influences, be it 22 human or otherwise?
23 HR. SKILLMAN: As you broaden the question, I want 24 to say yes, but in general.
25 HR. MICHELSON: That is just what I call external HERITAGE REPORTING CORPORATION -- (202)628-(888
r
}
57 1 events and human factors combined. I am broadening the 2 question, trying to find out which one of these are external 3 initiated and which ones are human factor initiated and which 4 ones are internal.
5 MR. SKILLMAN: Generally speaking, the eleven are 6 the ones that you are concerned about. The more imporcant i
7 piece of this, the more important piece of this slide I think i
8 is out of 237, there were 76 misbehaviors, and this is where I l
9 think that the real value of SPIP begins to come forward, and 10 it begins to lean on some of the issues that you have raised, 11 Glenn.
12 MR. JONES: Mr. Michelson, if I could also refer you 13 to page 4-13 in the original SER, there is a table that talks
() 14 about transients versus reactor trip signals, and you will see p 15 there is three columns dealing with human errors, operation, 16 maintenance, surveillance and testing type errors which have 17 initiated transients in that, and that is a summary as I 18 remember from one of the specific B&W reports.
19 MR. MICHELSON: Was it a summary of those 237 or 20 whatever?
21 MR. JONES: I think this is from the earlier report l
22 which is 204, but it is from one of the first, this came from i l
23 what, the owners group so-called quick sort of the TAP data. i 24 MR. MICHELSON: That this, for instance, says it was l O 25 never caused by inadvertent actuation of fire protection in l
HERITAGE REPORTING CORPORATION -- (202)628-4888
58 1 any B&W plant because that doesn't show up as any of these 2 items as a for instance. Is that what it says?
3 HR. JONES: That possibly would be covered under 1
surveillance and testing or operation, but that was not clear. I 4
i 5 I can't say there was and there wasn't.
6 MR. MICHELSON: Yes.
7 HR. JONES: But it would have been a small number.
8 MR. SKILLMAN: Please let me make my point before I 9 stop. Out of 237 trips, 76 misbehaviors, keep that, that 10 title in mind, because what we are saying is out of 76 11 misbehaviors, for the overcooling, the undercooling and the 12 overfill, overfeed transients, of which there are about 50
() 13 significant transients, the emergency feedwater and the 14 secondary system pressure controlc that the main steam safety 15 valves, atmospheric dump valves and turbine bypass valves, 16 compose the majority of the MIST misbehaviors, followed by 17 ICS/NNI followed by main feed, with electrical, PORV, and 18 instrument error showing up as also rans.
19 The real message for most of us in the B&W owners ,
20 group is when you back away from it, the trip history, and say 21 what is the misbehavior pattern, you do not find the reactor 22 coolant pumps or the pressurizer or the steam generators or 23 the hardy components of reactor coolant system misbehaving.
() 24 What you do find is after the trip breakers have opened, the 25 post-trip, that there is a dominance of misbehavior in r
HERITAGE REPORTING CORPORATION -- (202)628-4888
/ )
59 1 secondary plant relief and in emergency feedwater, and I would propose the question what, do these two systems really control J
) 2 3 post-trip?
4 Decay heat; it is the balance of decay heat 5 post-trip that has been giving the Babcock plants the complex 6 transient misbehavior title. This is where the problems are 7 coming from in a dominant fashion.
8 HR. MICHELSON: Does your chart include the manual 9 trips, in addition to the automatic trips?
10 HR. SKILLHAN: Yes, sir.
11 HR. MICHELSON: You treated those the same? Either 12 one was considered--
13 HR. SKILLMAN: There is n--
0 14 HR. MICHELSON: Sometimes you have the trip before 15 the automatic would have tripped.
16 HR. SKILLHAN: I think what we cet back to very l
17 quickly is the root of your question, sir, regarding decay l 18 heat, but it is, it isn't as if there is an inadequacy of 19 capability to remove decay heat, but clearly there is a l 20 mismatch between these two, and we will get into that a little 21 later.
22 We have ICS/NNI also becoming a part of the 23 misbehavior pattern, and Larry Stalter from Toledo Edison will 24 talk about that a little later, but when you look down this ,
25 list you find what is misesing from the list. There are a lot i
HERITAGE REPORTING CORPORATION -- (202)628-4888
60 1 of things that are missing, but the things that are on that 2 list are balance of plant,and they are significant. And 3 that's where we really zeroed in because when we get into the ;
4 undercooling, we are concerned about the PORV. The i
5 overcooling are the dominant types of behaviors that we have 6 and we find the emergency feedwater and secondary plant' relief i
7 really leading the pack in terms of overcooling, and while 8 overcooling certainly removes heat, they set the plant up for ;
f 9 a lot of, for a lot of hustle in~the control room, and we see 10 ourselves clearly at risk because of the frequency of those i 11 overcooling transients. !
l 12 MR. REED: 'I am very interested in the chart, andi la 13 says to me beyond what you are saying, it says well, the O 14 emergency feedwater system isn't very predictable, reliable or 15 dependable, and it says that the main steam safety valves, I 16 think it was what SSPC is, aren't very predictable, dependable 17 or reliable, and yet the issue that we are'trying to make sure 18 we take care of here, the key issue, the sacred issue is decay 19 heat removal. Right? ,
i 20 MR. SKILLMAN: Yes, sir, post-trip, k I
21 MR. REED: Okay. We will reflect on that throughout l 22 the meeting.
23 MR. SKILLMAN: hhat I would offer in response is l t
l 3 24 yes, sir, that is true, and that's why so much of our energy LO 25 has gone to attending to these in our recommendations, and we l 4 1
, )
i HERITAGE REPORTING CORPORATION -- (202)628-4888 j
. ,-..-.-. - , . . - _ . - . . - . - - - . , - - , _ . - - - _ , , - - . .-_-.--,.-.--.--a
61 1 believe that we have zeroed in on the areas that have given us
) 2 this misbehavior pattern. We know what this is now, and we 3 have gone to work on it.
4 MR. REED: Just one little point I would like to 5 make is that emergency feedwater really means steam generator 6 secondary side level. That's water level. That's what it 7 really means, and you could put that in brackets under 8 emergency feedwater. We will reflect on that throughout the 9 meeting.
10 HR. SKILLMAN: Okay. Thank you.
11 HR. RUTHERFORD: Host of the problems we have 12 encountered have been from removing too much decay heat, not
() 13 too little.
14 MR. REED: Okay. Matter of control of the level in 15 the secondary side of the steam generator.
16 HR. RUTHERFORD: Not so much level control.
17 MR. REED: Well then, flow.
18 HR. RUTHERFORD: Pressure; pressure goal has been --
19 MR. SKILLMAN: Which is sonic to primary, different 20 way to look at it, but it was what it is.
21 MR. ETHERINGTON: To pump natural circulation, you 22 want to raise the water level. Then you imply overcooling is, i 1
23 this a tricky adjustment between the two.
,/') 24 HR. SKILLHAN: Again, sir--
(/
, 25 HR. ETHERINGTON: To promote natural recirculation, HERITAGE REPORTING CORPORATION -- (202)628-4888
, 62 [
1 you wanted to raise the level of the water in the steam ,
4
(:) 2 generator, is that right?
7 3 MR. SKILLMAN: Yes, sir.
. t 4 MR. ETHERINGTON: In doing so, if you overcool it, 5 then you invite an overcooling, is that right also?
6 MR. SKILLMAN: You can, yes, sir, you can.
7 MR. ETHERINGTO!(: Is this, has it been the cause of 8 overcooling?
9 MR. SKILLMAN: Yes, sir, it has been.
i 10 MR. ETHERINGTON: That is not really a critical 1
11 adjustment? Is that what you are telling me? ,
i 0 J 12 MR. SKILLMAN: It has been a, an issue of complexity i
(/ 13 for the operators in the past, and we have, we have looked at 14 that in detail and offered recommendations to reduce the ,
i 15 complexity of balancing emergency feedwater flow rate against i ,
16 the then existing decay heat removal requirement, and here is 17 the real, the real goal on that. In the Babcock designs, the 18 emergency feedwater is injected very high in the generator.
i ,
d 19 It has a long length of heat transfer. For every pound of
- i j 20 water that goes in there, thousand BTUs are removed. ;
21 The real trick is we have designed the emergency 22 feedwater flow rates for extreme conditions when in reality a ,
, i j 23 fraction of that water is truly needed to remove the decay ;
I ,
()
24 heat that exists at that point in time, so getting that
- 25 balance control has previously been a, a challenge for the
'l 4
HERITAGE REPORTING CORPORATION -- (202)628-4888
.,-.m. . . - - - ,
63 1 operators, and through SPIP we have identified limitation of
( 2 emergency feedwater flow rate, settling, increasing the time 3 for run-up for the turbines that produce emergency feedwater 4 is key, recommendations to get the balance between the 5 emergency feedwater injection flow rate, and decay heat 6 removal requirements back in balance, but you are reading it 7 right. That level, and the over injection of emergency feed, 8 has been part of overcooling signature that we have had.
9 MR. ETHERINGTON: So apparently you have fairly wide 10 latitude, is that right?
11 MR. SKILLMAN: Say again.
12 MR. ETHERINGTON: You have a wide latitude? It is
() 13 not a physical problem?
14 MR. SKILLMAN: We do have a wide latitude because of 15 ability to remove copious decay heat with a little bit of 16 emergency feedwater, yes, sir.
17 MR. REED: Harold--
18 MR. MICHELSON: The thrust of his question--maybe I 19 should let him interpret it himself, l 20 MR. REED: I would like to clarify something for 21 Harold because he perhaps took something I said at face value 22 and there was more to it.
23 The level in the steam generator secondary side is ;
i /"T 24 very important, and also the fact that water is sprayed at a V
25 certain low rate up at the top of the steam generators.
HERITAGE REPORTING CORPORATION -- (202)628-4888
I l i
64 )
1 Now way back in ancient history, when B&W created 2 this system, they injected the emergency or auxiliary i
3 feedwater at the bottom of the steam generators. After they )
l 4 had delivered the steam generators to Oconee, somebody figured ]
5 out hey, that wasn't going to work for natural circulation 6 because they could get the thermal block, so they put a spray v
7 ring, they added the spray ring up at the top of the steam 8 generator in order to get that dense water up higher so they r
9 would have enough driving head to push up the hill, the 10 30-foot hill on the big cold leg pipe. Very complex; that's 11 why I wonder if it isn't too complex to not have a very f 12 reliable ultimate system for decay heat removal.
4 13 MR. SKILLMAN: What we have found through the '
O 14 sensitivity study is that the pressure, the pressure in the 15 secondary side is really the tattletale that you are looking l t
16 for for post-trip decay heat removal using secondary side.
17 The first reaction is to say you need a lot of water
{
18 level in order to have the differential temperature to drive
[
19 the decay heat across the tubes in the generator, but what the-20 sensitivity study has shown us is that the lower inventory of i
21 the Babcock design compared with R$G designs is not a 22 disadvantage, but thermalhydraulically it is a secondary side ;
23 pressure that lets the decay heat be removed, and so if we 24 focus in on that aspect of it, ensuring that emergency feed in 25 delivered at the right flow rate against secondary pressure, i
HERITAGE REPORTING CORPORATION -- (202)628-4888
i 65 l l
1 we can have smooth control of post-trip secondary pressure and l l
fT
\l 2 decay heat removal.
3 MR. REED: Will somebody speak--I don't understand 4 how pressure is a reliable indicator, to my knowledge. What 5 about the fact that tubes can fail and main steam lines can 6 fail and some B&W plants do not have intercept or stop, main 7 steam stop valves?
8 MR. SKILLMAN: The systems that are TAP data review, 9 that is the review of all of the transients, told us we should 10 be looking in more detail at the integrated control system, 11 ICSNNI system, main feedwater system, the emergency feedwater 12 system, also known as auxiliary feedwater system, the
() 13 secondary plant relief system, and the instrument air systems.
14 HR. MICHELSON: Would you clear up one little point 15 for me because I think I hear it both ways? I'm not sure. If 16 I had no water level on the secondary side, but had my spray 17 fully functional, would I have to have any water level to 18 remove the decay heat?
19 HR. SKILLMAN: Emergency feedwater spray, yes, sir.
20 MR. MICHELSON: Level has got nothing to do with the 21 base requirement? It only has to do with some of the 22 analytical processes to go through, but you don't need level 23 to remove heat? You just need spray?
() 24 MR. SKILLMAN: You need emergency feedwater 25 delivered at several hundred gallons a minute, r l
3 HERITAGE REPORTING CORPORATION -- (202)628-4888
66 1 MR. MICHELSON: You keep telling me about
() 2 controlling levels and I think that is--
3 HR. SKILLMAN: I don't believe I said that. I 4 believe Glenn said that.
5 MR. MICHELSON: Spray is--I think I heard it both 6 ways.
7 MR. SKILLMAN: I believe it was Glenn that said 8 that. I said the pressure.
9 MR. MICHELSON: Your intention was to talk about i 10 controlling the spray rate, and you do that through pressure.
11 (Slide) 12 MR. SKILLMAN: Very important piece of our 13 information gathering was the interview of the operations and 14 maintenance personnel. We used INPO developed interview 15 process. We did focused interviews with operations and 16 maintenance personnel at each of the plant sites. We looked 17 for transients which may not have resulted in reactor trips, 18 but were considered to be significant by those people.
19 We looked for areas of improvement where we can 20 reduce the need for post-trip operator action, and we tracked 21 down and defined concerns with procedures, with issues of 23 maintenance, and issues of operator burden.
23 There was other information available in the a
1 1
24 industry that would help us focus on our problem areas. We j
( l 25 looked at, for instance, INPO information, NRC type ;
I HERITAGE REPORTING CORPORATION -- (202)628-4888 1
67 1 information, our own internal event reports, other operational 2 histories such as that was available from INPO, from the NRC,
}
3 and we conducted these reviews. This was a thoroughness 4 activity.
5 To be covered independently from my presentation 6 this morning and this afternoon will be a presentation on the 7 sensitivity study by Stuart Rose from Duke Power, and I will 8 be talking later about the risk assessment that we did.
9 (Slide) 10 MR. CATTON: At some point later will somebody get 11 at these questions that Glenn was raising?
12 MR. SKILLMAN: About the design of decay heat?
13 MR. CATTON: Well, the emergency feedwater, and why 14 you feel that controlling the pressures is the key variable?
15 MR. WARD: I think they just did. Did you hear 16 Carl's discussion?
17 MR. CATTON: Yes.
18 MR. SKILLMAN: What Carl said is do you really need 19 level, yes or no? I am quoting Carl. From memory he said, 20 I'm getting the sense that what you really need is spray, high 21 in the generator, as opposed to level, and my response is yes, 22 sir, that's exactly right. What we need is the emergency 23 feedwater flow rate whose removal, heat removal rate mctches 24 the then existing decay heat generation rate and the secondary
(:) 25 plant system removal rate so that you end up with an energy i HERITAGE REPORTING CORPORATION -- (202)628-4888
68 l l
l 1 balance to hold the generator at pressure. j (O/ 2 HR. CATTON: I understand that. I am just wondering i
3 how do you get it? How do you get this relationship?
4 HR. SKILLMAN: By maintaining secondary pressure and 5 by ensuring that the emergency feedwater pumps are delivering 6 the appropriate amount for that pressure. Let me--
7 MR. CATTON: You have a little feedback loop.
8 HR. SKILLMAN: Oh, we do. For_ instance, at all of 9 the 820 plants we had emergency feedwater injection and 10 control, for all plants except Davis-Besse. Davis-Besse has 11 its own steam and feed rupture control system. Oconee 1, 2 12 and 3 each have emergency feedwater with their own control
() 13 system by Duke Power, and THI-1 has the HSPS, the heat sink 14 protection system, but Duke's emergency feedwater control 15 system, THI-l's HSPS, the other eight Babcock plants with EFIC 16 and Davis-Besse with SFRCS each have that control system as a 17 safety grade system in the plant.
18 HR. CATTON: As the pressure in steam generator 19 rises and falls you raise or lower the emergency feedwater 20 flow rate?
21 HR. SKILLHAN: Alter emergency feedwater flow rate, 22 yes, sir.
23 HR. CATTON: Where did that relationship come from? l 24 Is it just calculated?
[}
25 HR. SKILLHAN: Yes, sir, through the EFIC HERITAGE REPORTING CORPORATION -- (202)628-4888
69 1 activities, through the emergency feedwater design and 2 analysis activities by each of the utilities.
3 HR HICHELSON: This is based on the principle that 4 you can't spray but a small portion of the tubes because it is ,
5 a bundle, spraying from the outside by the separators I guess 6 are spreading the water around, wet all the tubes out.
7 Are you accounting for the fact that some tubes 8 never get cool? MIST I don't think simulated that aspect of 9 the configuration at all? That's why I asked you the question 10 because I'm not sure you gave the right answer, but I think it 11 is the answer that we can accept right now, but I am not sure 12 your test facilities have really demonstrated if I had no
() 13 water in the steam generator, and only sprayed it, that you 14 know, there is, part of the tubing isn't going to get sprayed 15 in the real generator. A large portion will not get sprayed, 16 may not even get wet, depending on how good the separator 17 plates are spreading the water around, and I don't know that 18 we have ever really tested that. I am accepted it. I don't 19 think we will be in a no water situation in the generator 20 anyway but it is not an entirely frivolous question. l l 21 HR. CATTON: I believe your question is one of the i
22 items in the research program to look at the--
f 23 HR. HICHELSON: It should be. >
24 HR. CATTON: As near as I can tell, you don't have
~}
25 any kind of demonstration that you can do those calculations l l
1 I HERITAGE REPORTING CORPORATION -- (202)628-4888
70 1 properly at this point in time? HIST certainly did not give ,
( 2 it.
3 HR. MICHELSON: MIST couldn't do it.
4 HR. CATTON: Your earlier steam generator tests, 5 near as I can recollect, were inconclusive. The data had too 6 much scatter. There were too many problems with it to really 7 be able to use it in any meaningful way.
8 How important is it that you are able to calculate 9 the thermalhydraulics of the process? Who is monitoring the '
10 pressure enough?
11 HR. RUTERFORD: It is true we are looking at these 12 kinds of questions in potential research program, and I think
() 13 what we are trying to do in pursuing some additional tests 14 that we might run is characterize the importance of being able 15 to precisely calculate what is going on. It might be, it 16 might be from a scientific standpoint interesting, but the 17 control of the plant standpoint, not very important. As long 18 as you maintain flow into the generator, maintain the level, 19 then the penetration and the flow of what kind of spread you 20 get in the tube bundle is a bit more of an academic question.
l 21 HR. HICHELSON: What do you do as the level is I l
22 dropping and the pressure is where you want it? For control l 23 purposes you ignore the level or do you have a rule that says i l
i 24 you must maintain both level and pressure?
25 HR. RUTHERFORD: Your control system is going to i
)
I i
l i HERITAGE REPORTING CORPORATION -- (202)628-4888
71 1 respond to that dropping level, but if your level is
() 2 dropping--
3 MR. MICHELSON: Maybe I misunderstood. I thought 4 the control system was monitoring pressure.
5 HR. CATTON: It is the pressure low.
6 HR. MICHELSON: And level in some magic way.
7 HR. CATTON: Separate system that is monitoring 8 level? You said that the emergency feedwater is controlled by 9 pressure.
10 MR. SKILLMAN: The key parameter, the, this 11 discussion started with Glenn saying you must have water 12 level.
() 13 HR. CATTON: Must have heat transfer; that's the i
14 translation, what I understood Glenn to say.
15 MR. REED: Let's say you have got to have the level 16 at certain levels not to stop certain events or to cause 17 certain events to happen. In a small break, for instance, you 18 must have the water level as high as you can get it in the 19 steam generator and in a normal transient, a normal loss of i
20 AC, you must have the water level as low and get your spray as ;
21 high as you can get it. There are all kinds of things that 22 interact. Right? 1 1
23 HR. SKILLMAN: Each plant has its own emergency l
24 feedwater control system. Each ' emergency f eedwater control 25 system goes by a different name at each of the sites. Some l
HERITAGE REPORTING CORPORATION -- (202)628-4888
72 1 look at water level. Some look at pressure. One has ,
() 2 cavitating venturis on emergency feedwater injection at the "
3 generator, so they are all a custom--for each of the different 1
4 plants, they have different types of equipment. They have {
l 5 different failure modes.
1 ,
6 MR. CATTON: Let's be very simple. All I need to
~
7 know is flow' rate, and a surface area of the tubes that are 8 covered, and the steam generator pressure and I can do a 9 pretty good calculation of the heat transfer.
l 10 What I can't do-now is relate those variables that I 11 need to--what you are telling me, because on the one hand I 12 hear level, level is not very important to those three 4
() 13 parameters if it is way down on the bottom somewhere, so if f i 14 you control based on pressure, you have a different control i 15 system that is doing something with level that may interfere, ;
1 l
16 I really think that I would like to see some sort of a written 4
j 17 description of the-process, what kind of calculations you have :
, 18 done and how do you get to opening or closing the valve?
- 19 MR. REED
- I don't get a lot of comfort, either, 20 from having you say that each plant has its own emergency j
~
21 feedwater control system. It seems t.1 me that part of the 22 industry's success will come from some kind of 23 standardization, and its success in safety, and to think that 1 !
i 24 each owner is doing some kind of of thing, calculation, I j 25 calibration, and set up of its own system bothers me. !
1 i
I HERITAGE REPORTING CORPORATION -- (202)628-4888
i 73 1 MR. JONES: If I could at the risk of just confusing
() 2 the discussicn further, we have discussed this matter as I 3 mentioned previously, about two, three months ago, decay heat j 4 removal, some Committee meeting, and I made that presentation.
5 Let me try to summarize it again, i 6 There are indeed three control levels for the lower, 7 the B&W plants, for the emergency feedwater system. You do 8 need level in the steam generator _because you do not maintain 9 a continuous emergency feedwater flow. Once you fill up the 10 control level, the emergency feedwater is shut off so the 11 question becomes why do I need two, three levels? What is 12 your sensitivity to those levels is I think what the question
{) 13 is.
14 Now the three levels you have there is approximately .
15 a 30-inch level, which can be maintained by either main i 16 feedwater or emergency feedwater depending on the plant '
i 17 design, which is used when the reactor coolant pumps are 18 running that gives you an adequate heat transfer surface for l
19 heat removal to remove decay heat loads.
20 If you lose the reactor coolant pumps, and partly in !
21 response to this issue of the long cold leg on the B&W plants, 22 you do indeed shift to a higher level which is referenced
]
1 23 again on the lowered loop plants as 50 percent of the operate j 1
24 range, which is about a little less than half full in the i
25 generator. It is about 20 feet. ;
1 t
HERITAGE REPORTING CORPORATION -- (202)628-4889
74 1 Now you go up with the aux spray or in the case of 2 Oconee they can do it by redirecting the main feedwater system 3 through the emergency feedwater nozzles. They will raise that 4 level with some continuous flow rate, to approximately this l 5 20-foot level in the steam generators. At 20 feet, that 6 stable level by calculation, and by test, actual plant 7 experience has been shown to provide adequate natural 8 circulation for transients.
9 Now the actual sensitivity of that level I can't 10 tell you exactly other than my memory says at THI, that THI 11 did some tests where they looked at how low can that level be 12 without the reactor coolant pumps, in order to keep natural 13 circulation going? These were some otart-up tests following O 14 the long, following the accident, and my memory as I remember 15 it is about 35 percent in the operate range is what is the 16 minimum level they must achieve to maintain stable flow. This 17 is without the continuous aux feed for heat removal.
18 For the small break LOCA, the issue came down to how 19 do I maintain the condensing surface in the steam generator?
20 This arose out of THI experience where it was recognized that 21 20 feet in the steam generator would not ensure a surface area 22 above the lip of the reactor coolant pumps, so I have, could 23 get guaranteed condensation heat removal in the long-term in 24 the boiler condenser mode in boiling in the core, and 25 condensing the steam generator.
HERITAGE REPORTING CORPORA'1 ION -- (202)628-4868
- . .. _ . ~ .. . . . . - - - .. .
75 7
. e 1- 1 The operating procedures were constructed such-that. '
2 if they lose natural circulation, even if they are at the 20 3
3 foot level or whatever level it is, should they not be getting i 4 the natural circulation they are supposed to be getting, that ,
5 there are indications they are not getting adequate heat :
l 6 removal from the core, the temperature is, hot and cold leg ,
i 7 temperatures are continuously diverging, then the operator is 8 instructed to raise the level to 95 percent of the operate 9 range. That is a manual action. It is also done, saturated 10 the primary system because the small break LOCA concern.. i j 11 The sensitivity on that number is about 80 percent I
12 on the operate range. A sure 75, 80 percent assuren'that you 13 have the condensing surface above the loop seal, above the lip
( ,
14 of the reactor coolant pumps so I could get a continuous ,
1 i 15 natural circulation flow in a boiler condenser mode if the i
16 primary system levels should drain down.
17 And that's the basis for the numbers that are in ;
1 18 there and that's the relative sensitivity. The staff has 19 looked at it. They have looked at it following THI, said l
1 2
20 THI-1 did its test. Looking at the lower levels, the MIST is 1
21 confirming the adequacy of the 95 percent level, the MIST 1 1
1 22 test. And at this point, looking at the procedures, the 23 levels that are there, we don't see a problem with natural
- 24 circulation, but that gives you a range of the numbers you are
, 25 dealing with.
.i I
j HERITAGE REPORTING CORPORATION -- (202)628-4888
76 1 Obviously the raised loop design has different 2 numbers because you do not need to overcome this loop seal.
[}
- 3 They also have a very low-level in reactor coolant pumps 4 operating. In the emergency situation, where they are not 5 getting. adequate natural circulation flow, I believe they i 6 raise to about a ten-foot condenser level in the stoam ;
J I 7 generator.
! 8 MR. ETHERINGTON: Level would have to be higher?
I 9 MR. JONES: Yes, sir. It is about, as I said, it is ;
} !
10 about I think the 95 percent level is about 30 foot of level j
l 11 above the lower tube sheet in the lower loop plant. I am not I 1
12 sure of the exact numbers because the way they run the TAPS, 13 but I believe it is about 30 feet, 30, 35 feet, that the lower [
4 14 looped plants it raised--they do not have this warm loop Leal l t
15 in the inlet to the coolant pump to overcome, to guarantee l
) 16 condensing surface, so they only have to raise it about ten ,
i l 17 feet. t 18 MR. ETHERINGTON: The same distance between-- ,
l '
! 19 MR. JONES: That is correct. !
\
20 MR. MICHELSON: I think the reason for the i
21 discussion is this introduction of the idea of controlling on I 22 pressure which they didn't mention at all as a part of the i
23 procert. It was kind of news to me. I didn't go to the last i 24 meeting in whatever.
( 25 MR. JONES: There is--you try to control the steam '
! l t
i HERITAGE REPORTING CORPORATION - .(202)628-4888 e
77 1 generator pressure. There are operator guidelines how you 2 control steam generator pressure. There has been, as was 3 previously mentioned, there has been operating history which 4 indicates too much feedwater will depressure the steam 5 generator, causing overcooling, which is not desirable. There 6 is a specific SPIP recommendation to put in automatic means to 7 limit the flow rates to the steam generators so they don't 8 overcool while still achieving an adequate fill rate of the 9 steam generator, and inadequate heat transfer surface to 10 continue the process.
11 MR. MICHELSON: The control system is monitoring 12 level as I understand correctly what you were telling me?
13 MR. JONES: The control systems are, there are O 14 diversity in the control system. There are, there is in the 15 EFIC system I believe you will cut back flow in the start te 16 depressurize the steam generator because that is an indication 17 you are putting the more, for the reason you need to 18 accomplish the heat removal.
19 MR. MICHELSON: I guess we have gone--I understood 20 your other explanation real well. Let's not get into this. I 21 assume that.
22 MR. JONES: I hope that was helpful.
23 MR MICHELSON: We covered this, which this would be 24 a good time for a break I think.
O 25 MR. SKILLMAN: Yes, sir.
HERITAGE REPORTING CORPORATION -- (202)628-4888
78 1 CHAIRMAN WYLIE: We will take a break, come back at
() 2 a quarter until eleven.
3 (A brief recess was taken.)
4 CHAIRMAN WYLIE:
5 HR. REED: Can we hit that pressure just one more 6 second? I am hearing something new and I am a bit confused.
7 CHAIRMAN WYLIE: Go ahead and ask your questions.
8 MR. REED: All right. I won't read, entirely, 9 reread entirely this statement on page 3-6 of your final 10 report, but it says here that the steam generator level should 11 be controlled to the desired heights above the lower tube 12 sheet to provide adequate natural circulation.
(} 13 Now you are telling me you control on a pressure.
14 Now what I think you are doing is maybe you are using some 15 form of three element control. Pressure ma be is the key
- 16 control in fact, but level is still in there I would hope, and 17 flow is in there from the spray, 18 Now the things that bother me about pressure being i
19 so important in this dry element control, if I am reading this 20 correctly, is that main steam, in every B&W trip, I read in ;
21 your report, now days, main steam safety valves off, okay.
22 Main steam safety valving, and your chart said this are 23 involved in a lot of these complex transients, and they are 24 highly unreliable on receipt, and if a main steam safety valve 25 opens, what does this do to trying to use pressure control as HERITAGE REPORTING CORPORATION -- (202)628-4888
79 1 the key parameter?
2 MR. SKILLHAN: I have made an error in overplaying 3 the importance of pressure. Level is the key element in !
4 emergency feedwater initiation and control systems HSPS and i 5 emergency feedwater. The relationship between pressure and 6 the heat removal is, what we have found through the 1
7 sensitivity study is that what is really providing the heat -
8 removal is the relationship between TSAT and the generator, 9 and T subcooled primary. I 10 The point I was making in response to your question 11 earlier is the level isn't the most important parameter is 12 what we have found through the sensitivity study, is it is
() 13 that, it is the differential temperature is telegraphed by the 14 secondary side pressure which is affected in turn by secondary 15 plant relief as well, point out the condition of the main 16 steam safety valves and the emergency feedwater injection flow i 17 rate.
i 18 I made an error in leaving unchallenged the thought 3
19 that level is not an important parameter. It is the key 20 parameter in controlling decay heat removal via--
l 21 MR. REED: I can't keep from thrusting this dagger 22 into you. I'm sorry about it, but I go back a long usy, and 23 PWRs.
24 Now if you made an erro tvolainf<a to .s. and 25 you had to go back and check it Other type person, 1
i HERITAGE REPORTING CORPORATIOt - ,uj ..J-4848
l 80 1 I wonder if the operators really today understand all this and 2 if the simulator is correct, and if this complex control is l 3 still, is possibly in error too.ay?
! i 4 MR. SKILLMAN: I didn't say I made an error l 5 technically. I said I made an error in leaving unchallenged 6 the thought that the level is the key parameter. Level is the ,
7 key parameter, but what we found, I will say it again, through
! 8 the sensitivity study, is that it is really the secondary side 9 pressure that is, since it is a saturated system, it ,
i 10 telegraphs secondary side temperature, which then is the 11 driving head to primary side compressed temperature. That's 12 the point I am making. I would like to go on from there.
13 MR. REED: It is three element control? Is that
]
14 what you are saying, three element control?
I j 15 MR. SKILLMAN: I am not making any statement i
16 regarding the design of the emergency feedwater control 17 systems.
18 Having talked through generally what we were doing ;
l l
i j 19 on information gathering, I wanted to say to you here is what 20 is in the safety and performance improvement program. We have 1 1
21 the system reviews I&C, main feed and feed emergency feed and !
22 instrument operating experience of that is the TAP data l
23 review, the review of the other transient information, the 24 operations and maintenance personnel interviews sensitivity !
l O 25 study, risk assessment. We carry as a major task the duty of l l
l l
I l
HERITAGE REPORTING CORPORATION -- (2021628-4888 1 l
81 1 interacting with the NRC staff to get a trip initiator review.
2 We are doing operator burden, completed operator burden ,
3 review. Bob pointed out the emergency procedures review is 4 ongoing. We had as our advisors an independent advisory 5 Board, and we convened from within the B&W owners a safety and 6 performance recommendation integration group that was a team l
7 of very experienced operations and design people to look at 8 recommendation package 200 plus recommendations, with the 9 intent of identifying those that were most beneficial and 10 increasing the level of safety of the Babcock plants.
11 I would like to speak for a minute about our 12 independent advisory bosed. We wanted the oversight by people
() 13 who were not commercially related, by people who were not 14 commercially related to the B&W owners or to B&W, and we 15 retained for about a year and a half Bill Lehman from EPRI, 16 Saul Levy from Saul Levy Associates, Neil Toderay from MIT AND 1
17 Bob Brodsky with Beta. Their involvement was an every two ,
I 18 month review of the program in detail. Their mission was to 1
19 evaluate the comprehensiveness and the safety emphasis of the 20 safety performance improvement program, and basically evaluate 21 the ability of the program to achieve its objectives. And I 22 will talk later about their conclusions.
23 (Slide)
() 24 HR. SKILLHAN: From within our own ranks, that is 25 from within the B&W owners, we convened the safety and i I
l HERITAGE REPORTING CORPORATION -- (202)628-4888
82 l 1 performance recommendation integration group, these folks 1
() 2 representing each utility in some significant depth. Had a 3 formal charter, multi-discipline, that is, they were 4 operations, design, maintenance, and so on, and they, they 5 provided a concentrated effort of review after essentially all 6 of the data became available for the recommendations were 7 available.
8 What they were really doing is trying to identify 9 those recommendations the most important and beneficial to
> 10 safety performance improvement, and they were reviewing the j 11 recommendations from the perspectives of operation, safety ;
12 design and maintenance.
() 13 So our position is that we did a multi-system 14 review. We had an independent consultant perform a thorough i 15 and exhaustive sensitivity study. We re-reviewed six years of 16 operating data. We went into detail on the issues of operator 17 burden of our operations and maintenance people. We performed 18 a relative risk assessment looking at the category transient C ,
19 transients against the PRAs that were available. We went to l
20 extend those PRAs to plants that did not have PRA. We have 21 used a great deal of input from the staff in terms of 4
22 sharpening the focus of the program, and we had our IAB and 1
23 SPRIG input significantly to the program to spread them to its 24 conclusions.
] 25 (Slide)
HERITAGE REPORTING CORPORATION -- (202)628-4888
^
l 83 1 MR. SKILLMAN: I would like to move over into
() 2 findings and conclusions. Our conclusions are in four basic 3 areas. They have to do with plant safety and design, have to 4 do with the finding on complex transients, and trip 5 initiators. Please keep those separate in your mind. And we 6 have some conclusions regarding the value of SPIP that we 7 think are germane to this discussion.
8 (Slide) 9 HR. SKILLMAN: Based on our risk assessment, the 10 work that we did with SAIC regarding relative risk, the risk i
11 of core damage is comparable to other PWR designs. We believe 12 it was a very important conclusion because the real thesis of ;
13 the Stello letter in January 1986 was determine whether or not 14 these plants needing special design requirements are in fact 15 less safe than other plants.
16 The answer is, in our judgment, is no. They are 17 comparable in terms of core damage.
l 18 CHAIRMAN WYLIE: How many of the B&W units have 19 PRAs? i 20 MR. SKILLMAN:
Started with two I believe. Now 21 three, oconee, and Florida each--now THI-1 has submitted its !
1 22 PRA so that there are three now, and I don't know the status .
23 of the other, the other five plants, i :
) 24 CHAIRMAN WYLIE: Are they--you don't know the (2) 25 status? ,
HERITAGE REPORTING CORPORATION -- (202)628-4888
84 1 HR. SKILLHAN: I do not, no, sir.
F
\_e 2 CHAIRMAN WYLIE: All right.
3 HR. SKILLHAN: The areas of sensitivity, 4 thermalhydraulic sensitivities of the Babcock plants are 5 different than the areas of sensitivity for non-Babcock PWRs.
6 A key concern oft repeated situation is the 7 pressurizers on Babcock plants are too small. That is 8 unfounded. Our pressurizers are adequate in size based on ;
9 whichever parameters you choose, whether it is outsurge l
10 following the trip, or any other parameter you might use. ,
11 HR. CATTON: I recollect following THI statements ,
12 made by PWR operators that they often lost the level following
() 13 a turbine trip. This is no longer the case?
14 HR. SKILLHAN: There is still plenty of water in the 15 pressurizer. The problem is the location of the lower TAP. '
16 HR. CATTON: I assume that that way the sense of it 4
17 is you essentially make your operator blind?
18 HR. SKILLHAN: It can be. Yes, sir.
19 HR. CATTON: He has to run the plant and he doesn't l
20 know what is going on, so I don't think that is--
l 21 HR. SKILLMAN: We will get into this.
22 HR. CATTON: It is adequate from the 23 thermalhydraulic point of view, but it certainly doesn't seem i
(} 24 to me it is adequate from an operation point of view.
25 HR. SKILLHAN: Okay.
l HERITAGE REPORTING CORPORATION -- (202)628-4888
85 1 MR. REED: Is it true, and I think it was true back l f
() 2 before some modifications that may have been made, that if you ^
3 trip B&W unit above 30 percent power level the water level in j l
4 i
4 the pressurizers in the first few seconds or milliseconds ;
i 5 would go up to a point where it would lift the power operator 6 relief, and then in the next, as spray hit the steam generator 7 tubes, the water level in the pressurizer would go down out of i 8 site? Are those reliable statements from two years ago? You ;
i 9 go up, lift up the PORV in the first few seconds or !
10 milliseconds then the water level shrinks in the pressurizer 11 to out of sight? !
12 MR. SKILLMAN: I would like to suggest that that
(} 13 question come up when Stuart is talking sensitivity because 14 that is probably the more appropriate place to address the way 15 in which the plant responds. Stuart? Is that, can I defer [
16 that one to you? ;
17 MR. ROSE: To some extent. I 18 MR. SKILLMAN: Operator burden, another item that is i 19 pointed to as being extremely important, at the Babcock plant i i
20 we believe the operator burden is acceptable. Integrated :
21 functions of the integrated control system are a benefit. We ;
22 also understand that when the integrated system, when the l 1 23 integrated control system is the system that has failed, that i 1
24 the operator burden is too great and consequently we are 0 25 putting a great deal of resources in modifying, considering I
i HERITAGE REPORTING CORPORATION -- (202)628-4888
I 86 1 modifications to the integrated control system to reduce that
' 2 burden for incidences when power is lost to ICS/NNI or input 3 failures occur.
4 We retain the basic note that the integration of 5 control functions in the ICS is appropriate. We acknowledge 6 that the plant is more responsive to secondary side changes, 7 and we believe that the recommendation which covers each of 8 these will further improve the safety level of the Babcock 9 plants.
10 HR. DAVIS: I have a question. Maybe you would want 11 to defer this. I am not challenging your fourth conclusion, 12 the steam generator inventory, but I was concerned about the 13 statements made in B&W 1919 giving the implication that the 14 inventory is, in terms of accident response, is similar to 15 other PWRs, and I recall reviewing calculations on Station 16 Blackout, for example, where as I recall the numbers, there is 17 about an hour available in a B&W plant before core uncovery 18 starts versus three hours for a Westinghouse plant, for 19 example, and as I recall that difference was attributed to the 20 inventory difference between the two plants.
21 And I am wondering now if you have some other 22 calculations indicating that these times are comparable, or 23 whether the statements in the report maybe are not valid?
24 HR. SKILLHAN: I am going to defer that to Stuart
( 25 because that's a, very clearly a topic of the sensitivity HERITAGE REPORTING CORPORATION -- (202)628-4888
, e r
87 1- study.'
2 HR. DAVIS: Thank you.
3 HR. REED: Of course, I don't agree with a number of
, 4 those, and just to hit briefly the top one, just to make a 5 statement, the top one, of course, refers probably to the root 6 issue decay heat removal.
, 7 So what you are saying, the risk, the decay heat 8 removal in the B&Ws is as good as the other PWRs? And I have 9 a great deal of problem with that because decay heat removal 10 is for the most part coming from your steam generator f
11 complexity.
12 Now you could have put in your comparison comparing
() 13 your high set steam generator with the low set, and whether i 14 the risk of core damage is better with the high set or the low i t
15 set. I always wonder why did B&W change to a high set steam 1
j 16 generator on the last one or two plants they were going to L
17 build or built? Like Davis-Besse and the one out in this, the
! c l 18 West therr., WPPSS? Why did they change to the high set steam 19 generator? It costs a hell of a lot more to build it that way 20 so why did you do it?
I 21 HR. SKILLMAN: The question that you have asked I i
22 won't attempt to answer, but I am going to make my comment. '
t 23 The fact, this fact rests on non-secondary plant i I
() 24 considerations, so your thesis that the secondary plant level 25 is somehow inappropriate and therefore this sentence is fiswed l
l HERITAGE REPORTING CORPORATION -- (202)628-4888 ,
88 l l
1 in this my' judgment doesn't wash. The reason that that i l
2 statement is true is because the PRA, the thing that saves the 3 PRA on a Babcock plant is an operating makeup pump that does '
4 not have to answer a start command. Say it_ began, the things 5 that saves the Babcock plants is the fact that a makeup pump 6 does not have to-answer a start command.
7 At TMI-2, the pump that was secured it, the. pump-that-8 should have been left operating. It was an operating error to 9 shut down the makeup pump, but that's what melted the core.
10 All plants, that makeup pump not answering the start command 11 makes these PRAs, makes these Babcock plants particularly 12 resisttant to core damage. If you trace through the PRA fault 13 trees you will find that this is the truth.
) 14 HR. REED: Dave Okrent had a lot of faith in a PRA. '
I 15 He was an operating type person. I have'never had any faith 16 in them.
l l
17 What you are telling me is if the' primary makeup I 18 pumps are centrifugals and one is always running, and now you .
19 get into the situation of whether it is better to have a I
20 running component to provide safety, for safety systems, or a-21 stopped component--
22 MR. SKILLMAN: Operating component does not have to 23 obey.
24 MR. REED: Path of decay heat removal has the
\ 25 pressurizer loop sealed and low set in the containment and I l
1 HERITAGE REPORTING CORPORATION -- (202)628-4888 !
89 1 1 am not so sure that is a good path decay. heat removal.
() 2 MR. SKILLMAN: Regarding the complex review of the
- 3. transients,'if you recall when'I showed the table, the trip 1
4 complexity pattern versus the plant trip pattern, we have 5 conclusions in those two categories regarding the complexity ,
6 of transients. Complex transients are principally the result'-
7 of mismatched heat balance between primary and secondary.
8 Again back to the overheating, underheating, ,
9 overfill, overfeed, the three different categories of 10 significant transients, the overheating events are of greater 11 safety significance, but they are indeed infrequent. The 12 overcooling events may be precursors to overheating because of 13 the way in which the operators interact with the plant. The 14 overcooling events are dominated by misbehavior of the 15 emergency feedwater and the secondary plant relief systems, 16 and we offer that other system misbehaviors is important'and r
17 therefore must be acted upon, the behavior'the integrated 10 control system and the behavior of the main feedwater system.
19 (Slide) 20 MR. SKILLMAN: If you ask what is the pattern of 21 post-trip misbehavior, what you find is that the bulk of a
22 misbehavior comes from the secondary plant relief. That's not 23 just main steam safety valves. That's main steam safety I 24 valves, atmospheric dump valves, and then turbine bypass 25 valves, and when you dig more deeply into that area you find t
HERITAGE REPORTING CORPORATION -- (202)628-4888
l 90 ;
1 1 that the bulk of the problems here are maintenance, BOP type
(~}
k- 2 care and feeding of those systems type problems as opposed to l
3 fundamental design. In the Glenn's thesis, main steam safety 4 valves are unreliable. We find that their failure rate is 5 about one in a thousand.
6 Emergency feedwater, approximately the same in terms 7 of misbehavior, leading up to the conclusion that these two 8 systems normally act hand in hand post-trip. The ICS, main 9 feed, electrical also contributing to our post-trip complexity 10 of behavior.
'L1 (Slide) 12 HR. SKILLMAN: Regarding frequency of overcooling
() 13 events versus overheating events, of course these are the ones 14 that is we sense are bona fide safety significant. About 90 15 percent of our complex transients are overcooling, about 8 16 percent are overheating, and several percent of the overfill, 17 overfeed. Clearly these, these attract our attention. These 18 are the ones that we are very concerned about.
19 HR. CATTON: Overfill doesn't lead to problems?
20 MR. SKILLMAN: Yes, sir, it can. The Rancho Seco 21 event was an overfill event, and that is what we ascribed to 22 them. The importance, that the ones that have our attention 23 are the ones that are, the overheating event as it turns out,
(} 24 the fixes that go with overoverheating type scenarios will fix 25 the overfill, overfed events as well.
HERITAGE REPORTING CORPORATION -- (202)628-4888
-91 1 Regarding trip initiation,.this is the second r
2 pattern I have shown as the trip complexity pattern. Now the 3 trip frequency pattern, the majority of trips by BOP' systems 4 or BOP components are people, and turbine-generator was the 5 dominant trip initiator before 1985. Fixes are in place.to 6 address the turbine generator type problems that we have had.- .
7 The current trip initiator, the dominant trip initiator is the 8 main feedwater system, with ICS/NNI and electrical also 9 important.
10 DR. KERR: In the listing you did not distinguish 11 between trips that might be caused by testing while the plant 12 is in operation? Have you separated those out?
13 MR. SKILLMAN: Yes, sir. These are run planned O 14 trips. The de-energization of the control rod drive, control 15 rod breakers, these, at power these are bona fide trip as 16 opposed to testing scenarios or start-up and test type 17 functions. These are at power, unplanned trips.
18 DR. KERR: Well, isn't a trip that is caused by ;
l 19 testing or surveillance during full power also an unplanned
)
20 trip?
21 MEMBER OF THE AUDIENCE: .They are included.
22 HR. RUTHERFORD: The technician--
23 DR. KERR: 'I am just curious about what fraction of 24 the trip may be caused by testing and surveillance as carr'ed 25 on during full. power operation, so I asked if you separated 1
i l
HERITAGE REPORTING CORPORATION -- (202)628-4888 4 l
,- ,- ..,~ n., ,-- ,-.. -.--n,,-,, -.-,..,n. ,-..-.--,.n,,-.~,-,,-.e , - , - - . , - , , , . , , - . . - , . , . . , - - - -
92 1 those out.
() 2 MR. SKILLMAN: These are included in our trip 3 pattern. That would be, we would classify that as a human 4 error.
5 DR. KERR: Do you have an estimate of what fraction 1
6 of these are caused by testing and surveillance during full 7 power?
8 MR. SKILLMAN: Yes, sir. The number is between 2 9 and 3 percent. Eleven out of 235 is the number that I would 10 offer right offhand.
11 DR. KERR: Thank you.
12 (Slide) 13 MR. SKILLMAN: Conclusions regarding SPIP value, we
(]}
14 found the attention that this program forced us to concentrate 15 in the BOP areas to be extremely valuable because we, we found 16 ourselves instead of looking at classical reactor coolant 17 system type problems, looking rather at plant performance and 18 behavior problems which led us further and further and further 19 into the BOP functions, and the BOP design. A great deal of 20 attention on BOP; we thought SPIP was valuable because it 21 reinforced the value of having this TAP data base,,but we have 22 found this data base to be, to be tremendously helpful in 23 being able to go back later, contruct patterns of events that 24 can happen in other plant sites in addition to the one where 25 the incident just occurred.
HERITAGE REPORTING CORPORATION -- (202)628-4888
93 1 We found value from the point of view of focusing A
'\/ 2 our own people on a common problem. We are implementing the 3 safety and performance improvement program recommendations.
4 Consequently the safety performance of these plants will 5 continue to improve. We have had diverse input, external to 6 the operators group that was valuable. We have been able to 7 gather points of view from our consultants, from our IAB, from 8 contractors that work for us and we have had a very strong 9 involvement by our executives in the ordering of the. program, 10 and the running of the program.
31 I would have anticipated a question what did your !
12 advisors tell you? Our advisors started with a basis, with
() 13 the concern that the program not reduce itself to the lowest 14 common denominator, that in any case where we were developing 15 recommendations or trying to determine where the program ought 16 to go, we had not to do the easy thing. Rather we ought to do 17 the right thing and the hard thing if that was what was needed 18 to get the job done properly.
19 Following their advice, we ended up with-a program 20 which they have reviewed in detail, which they have 21 communicated by letter to our executives, and these are points l l
22 from that letter; identifying areas for improvement. It l
23 examined all. factors involved in past complex transients. As
(} 24 of this writing, there were 13 complex transients, 10 from 25 January of '80 to mid '86 and in addition, the Rancho Seco I HERITAGE REPORTING CORPORATION -- (202)628-4888
94 1 event in 1978--excuse me--the Davis-Besse event in '78, Rancho
{} 2 Seco in the year before, the year later, and THI-2, so they 3 were involved, when they say past complex transients, they 4 were involved in unrave34.ng the 13 complex transients, not 5 just the 10. I 6 They found also that the TAP data base had been 7 valuable. They let them understand what the problems had 8 been. Their belief is the action that the operators proposed 9 taking would reduce transient frequency and improve plant 10 safety.
11 The IAB concurred with our process for prioritizing 12 the recommendations. They accepted that the SPIP goals are s 13 achievable. They felt that the schedule as we had proposed 14 was ambitious, and they reinforced the need for persistent 15 management attention by the B&W owners group, executives.
16 They had a concern regarding the specificity of some of the 17 recommendations, that is, the recommendations were cast too 18 generally, and they thought those recommendations should have 19 been cast more specifically.
20 DR. KERR: Speaking of recommendations that lacked 21 specificity, what is meant by management attention required?
22 MR. SKILLMAN: The tone of that comment was hey, 23 execs, if you really want to achieve an increase in the level l
24 of safety of your plant, you cannot simply take the ;
(
25 recommendation package, throw it into your site or throw it j l
l HERITAGE REPORTING CORPORATION -- (202)628-4888
95 1 into your engineering organization, and say here, do something
[)\
u 2 with this. Your execs need to get involved, your people that 3 you specifically designate to shepherd the recommendations 4 through and cut through all of the fog at the plant sites to 5 make sure that you really do achieve the behavior changes in 6 the plant that you desire with the recommendations. This 7 can't be defaulted to others. You have got to get involved 8 yourself.
9 IAB told the owners you need to assure that there is 10 a level of quality in the action regarding these 11 recommendations that ensures that the recommendations get done 12 properly, and the issues of the integrated control system and r^3 13 feodaater reliability deserve special attention by the execs.
V 14 The recommendations pertaining to those two systems need to be 1
15 earnestly considered. Again, if you were to have asked what i
1 16 did your advisors tell you, what did your advisors tell the 17 executives, this is a synopsis of what the four advisors i
18 Brodsky, Bradley, Lehman and Levy, told us. !
l 19 DR. KERR: Are you going to get to a more specific l l
20 comment on the feedwater reliability? For example, did you l l
21 get comments that indicated that one had to make drastic I 22 changes in the existing system, or was it simply a matter of 23 more attention or finetuning or these kinds of things?
24 Perhaps you were going to get to that later.
25 MR. SKILLMAN: I am going to talk about specific HERITAGE REPORTING CORPORATION -- (202)628-4888
96 1 recommendations for main feed later, yes, sir.
{} 2 DR. KERR: Thank you.
3 MR. SKILLMAN: All right. Finally, SPIP open areas--
4 need to talk for a few minutes here about open areas per the 5 agenda.
6 Let me start first by saying where we are in 7 substantial agreement with the staff. In specific, in 1
8 specific, with regard to the project, we are in agreement on 9 main feed, emergency feed secondary plant relief, 10 instrumentation, operating experience review, other transient 11 information, the operation and maintenance personnel 12 interviews. The sensitivity study risk assessment, we feel we 13 are together with the NRC on interacting with them.
14 MR. REED: Where were the operations and maintenance 15 interviewed conducted at Oconee?
16 MR. SKILLMAN: They would be conducted at each plant 17 site, the level of experience by operations and maintenance, 18 both types of people, at each plant site.
19 MR. REED: I notice you are doing some comparison at 20 Robertson and I wondered if operation and maintenance reviews 21 were done at Three Mile Island, Crystal River, SMUD?
22 MR. SKILLMAN: Yes. The operations and maintenance 23 personnel interviewe task is different from the sensitivity l 24 study task. This is comparative evaluation of O
V 25 thermalhydraulics, babcock versus CE versus Westinghouse.
HERITAGE REPORTING CORPORATION -- (202)628-4888
97 l
1 This is a different task. This is where a team of engineering l
() 2 people went.to each plant site, spent several days with 1
3 operations and maintenance personnel at'that site, and with 4 a--real goal was, was to find whether or not there were some 5 transients that were significant that did not result in plant 6 trip.
7 The real heartbeat of this task was hey, people, we 8 are here to find out if there are things that have happened at 9 this site that aren't somehow communicated elsewhere in the 10 data base, or are there other items here'that influence where 11 we are going in terms of the safety and performance 12 improvement program?
() 13 MR. REED: And the people that were interviewed were 14 objectively picked out, selected, in that did the management 15 of the plant say we will interview this guy, this guy and that 16 guy? Or did you take all the names in a hat and take the name 17 out?
18 MR. SKILLMAN: I don't know. Let me ask if there is 19 someone from the owners who can answer that question? The 20 randomness with which the interviewees were selected for the 21 operations of maintenance personnel interviews?
22 MR. GANTHER: I think in general they were 23 designated by the plant side, but they were like all the
[} 24 operators at this shift, all the maintenance people on that 25 shift the convenience of the interview times, but they would HERITAGE REPORTING CORPORATION -- (202)628-4888 r
90 1 do all the experienced operators on a particular shift. There 2 is some randomness there.
3 MR. SKILLMAN: Again we are in agreement with the j 4 staff regarding our findings, and the recommendations coming 5 from the operations and maintenance personnel interviews. We i 6 are together in our finding on the trip initiator review, the 7 operator burden review, and the integration activities.
8 Probably we are together with the staff. We are in 9 substantial agreement with the staff on the.overall program 10 scope and breadth, on the process that we used in the program, 11 on the content of the recommendations, on the program 12 management and the administration, and on the findings
- () 13 regarding plant safety.
14 MR. CATT0N: Where did you combine'the sensitivity 15 study which was basically thermalhydraulics and operation?
16 Because it seems to me your arguments--put the, put the other 17 one back on for a moment. When you, from what-I gather from 18 what you have said, when you say sensitivity study, this was 19 strictly thermalhydraulic. Was there enough water? Aren't 20 there sensitivities that are a result of the thermalhydraulics 2
21 that enter into operation like shortened up times and having 22 to do things quicker and so forth?
4 23 MR. SKILLMAN: Yes, sir.
(} 24 MR. CATTON: Where do they come together?
25 MR. SKILLMAN: Right here on operator burden review.
HERITAGE REPORTING CORPORATION -- (202)628-4888 t__ _ _ _ _ __ .. _ _ _ _ - _ . _ _ _ . _ _ - -- _ _,__ _ _ . _ . _ . _ _ _ , _ _ . - _ , . _
.. .. . .- .. - - . .- -~ ..- ~ . . . ,
99 1 MR. CATTON: Did you do a comparative study there 2 between the three plants?
3 MR. SKILLMAN: We did not. ,
4 MR. CATTON: Seems to'me you missed a key issue 5 then.
6 MR. SKILLMAN: I don't think so. We considered, we 7 considered doing a relative burden review, babcock versus 8 Combustion versus Westinghouse, and we started down thatLpath 9 in early '86, and we stopped and we stopped on purpose because 10 what we said is that what we are going to do is we are-going 11 to get into an arm wrestling contest on who has to do-how many 12 actions more quickly for what type of transient on which type
() 13 of plant, and we said to ourselves that's the wrong question.
14 The right question is do the Babcock and Wilcox
[
i 15 design PWR operators have what they need to do the job that 16 they need to do, regardless of what somebody else does in some i
17 other plant? And having unraveled that with the structured 18 process using the human factors professional in our own 19 people, we believe we came to answers that make sense. We did 20 find some things that need to be corrected.
21 HR. CATTON: I guess I just don't understand this 22 human factor business. To me four minutes is a hell of a lot 23 different than having 20 minutes or 60 minutes to do
(} 24 something.
25 MR. SKILLMAN: Let's get to that on the sensitivity t
HERITAGE REPORTING CORPORATION -- (202)628-4888
100 l l
1 study. Let's don't--
) 2 MR. CATTON: The sensitivity study is clear--four l l
3 minutes, 20 minimums, 60 minutes. It states to dry out, it 4 states it clearly. If you consider that not very sensitive, 5 or not very different, then I just disagree with you.
6 MR. SKILLMAN: Isn't the real issue what the 7 operator thinks about that?
8 MR. CATTON: That's right. But you have taken those 9 things apart, and have not addressed that question.
10 MR. SKILLMAN: In the operator burden review, is we 11 went into each control room with a select number of scenarios 12 and walked each of the teams through those scenarios.
(') 13 MR. CATTON: But did you ask the operator gee, would 14 you rather have 60 minutes to decide what to do than four 15 minutes?
i 16 MR. SKILLMAN: No. i 17 MR. CATTON: Did you ask him that question?
l 18 HR. SKILLis;.N : No, sir.
19 MR. CATTON: You didn't do a sensitivity study.
I 20 MR. SKILLMAN: Not that way; the sensitivity study l 21 was for a different purpose. We had to do tnis that way was 22 our thesis.
23 MR. CATTON: The question early on in all of this 24 was or at least the statement that used to be made was that 25 gee, a B&W reactor is sort of like owning a Ferrare. It is a l 1
l I
l HERITAGE REPORTING CORPORATION -- (202)628-4888
101 1 fine piece of machinery but it is damned tough to run and keep
() 2 running. One of the reasons was, is this difference in times-3 that you have available to do things. . I mean I can spend all'
- 4 day keeping my Ford running and it doesn't matter. You didn't ,
5 address that.
6 MR. SKILLMAN: We did.
7 MR. CATTON: You just told ~me you didn't.
8 MR. SKILLMAN: I told you we did.
9 MR. RUTHERFORD: Let me make comment we did address i
10 that in the sensitivity study, and one point to make here, .and 11 I don't want to dwell on it extensively, is that dry up time '
12 in and of itself is not the critical parameter you are 13 interested in.
14 MR. CATTON: That's true. It is whether or not you ,
15 have to, have to do something within that time.
16 MR. RUTHERFORD: You don't'have to. We will get 17 into that more in the sensitivity study. That is not the j 18 critical point in time where you have got to take action.
19 Okay, and the times you do have to take action are comparable 20 among the venders.
1 21 MR. SKILLMAN: Okay. In the past two slides, I have 22 tried to define where we are in substantial agreement with the 23 staff, where are we in disagreement with the staff.
24 We are in disagreement with the staff in terms of
, 25 content in the review of the ICS/NNI system review, and we are 1
3 HERITAGE REPORTING CORPORATION -- (202)628-4888
102 1 in disagreement with the staff regarding the use of human
\_) 2 factors expertise in the process of going through SPIP. Those 3 are the two areas of all the areas where we have disagreement.
4 (Slide) 5 MR. SKILLMAN: And we will talk about those later.
6 Regarding open areas, areas that need to be attended to, we 7 have an open area of implementation of plans and schedules.
8 Each site scheduled for implementation of the recommendations, 9 those are due this summer.
10 Both Bob and I touched on the issue of emergency 11 operating procedures review. This is where we wanted to go 12 through each of the emergency operating procedures and make
() 13 sure that they tie in appropriately to the other procedures 14 that are necessary for post-trip response, particularly in the 15 emergency situations. That's due a year from this summer.
16 NRC staff has requested the TAP annual report. We 17 are going to make that available. It will be available this 18 summer. The valve task force report will be due late this 19 summer. We will make that available when it is completed.
20 And the staff would like the applicable portions of the 21 recommendation tracking system that pertained to the trip 22 reduction recommendation to the SPIP recommendation. We will 23 make those updates available to the staff as they become
~~T 24 available.
(V 25 MR. REED: On item 4, valve task force report, for HERITAGE REPORTING CORPORATION -- (202)628-4888
103 1 come time when I was a member of ACRS I tried to get an 2 industry-wide action going to investigate valves and valves 3 that should not be applied to certain duties on PWRs and BWRs 4 and so on and so forth, and I don't know whether that ever got 5 going or will ever get going.
6 Is there going to be an industry-wide valve task 7 force? The answer is no? ;
8 CHAIRMAN WYLIE: Don't know.
9 MR. REED: I am just wondering, this task force 10 report may become very important because one of the things I 11 would like to see is a resolution of whether or not internal j 12 pilot operator relief valves on PWRs with boron and hydrogen
() 13 are acceptable. You can look forward to questions like that.
14 HR. SKILLMAN: With that, I am prepared to turn the 15 podium back to Neil Rutherford, or back to the staff. I'm 16 corry.
17 MR. JONES: I am Bob Jones from the Reactor Systems 18 Branch, and I had the overall technical coordination of the 19 reassessment review. I am going to go through two items since 20 the owners group has been through both, kind of the overview 21 or summary and conclusions from their perspective, and the 22 information gathering efforts that they went through, so I am 23 going to hit both of those activities one right'after the
() 24 other. First I want to start with the summary and 25 conclusions, i
HERITAGE REPORTING CORPORATION -- (202)628-4888
l 104 1 (Slide) i 2 MR. JONES: And something that has been a matter.of 3 discussion over'the last few hours has been the overall scope 4 of the program. This slide addresses how the staff sees the 5 scope of *he program from what was done, but I-think a lot of
'6 the questions that were raised earlier had to do with the 7 breadth or the depth of the overall programs, so let me first 8 start out dealing with'that activity.
9 Back in January of '86 when Stello wrote his letter, 10 said we were going to go reassess these plants, and there are ,
l 11 phrases that have been quoted about the thermalhydraulic ]
I 12 designs, basic design requirements, the fact that we have made t i
13 modifications since THI and yet we still have these complex '
) 14 transients at these plants, and staff was a bit unhappy about i 10 it, but wa struggled for a' couple of months in trying to >
16 define a scope for this reassessment effort somewhat 17 independently from the operators group which we, when the 18 owners group volunteered to take the lead of the program, we 19 provided our program plan so to speak to the owners group for 20 incorporation into their overall program, but to give an 21 impression of how we were going about this, there really are i 22 three, I don't want to call them phases--parts, parts to the 23 scope of the program.
24 First off, why were we concerned? We were concerned 25 not because we had anything we can put our finger on that says
]
4 HERITAGE REPORTING CORPORATION -- (202)628-4888
105 1 gee, these plants are unsafe. We were concerned because we f)\
(, 2 were seeing a lot of transients that we were not very happy 3 with. So right there we made a decision that one of the major !
4 factors in doing this review had to be the operating 5 experience of these plants. What was going on? What are the 6 systems involved? And then to take a good hard look at those 7 systems.
8 Part of that included looking back at some of our 9 concerns raised after TMI and all the related NUREGs that came 10 out at that point in time, and to try to define what issues 11 still appeared to be around. Thus we did not relook at 12 natural circulation because we felt those were adequately
{} 13 being addressed or had been addressed.
14 We also said that's still somewhat narrow given the 15 directive which we were given by Mr. Stello, so we ought to 16 take a look at or revisit the issue of the sensitivity of the 17 B&W design, and that essentially also came from the operating 18 experience from THI, so we said we ought to have some 19 comparative analysis, look at, try to understand what actions 20 were required by the operator, what kind of operator burden 21 exists, and whether anything needed to be done in that area.
22 Were there basic design characteristics that need to be dealt 23 with in some manner, shape or form?
]
l w 24 Thirdly, risk--as I mentioned earlier, there was a !
l 1
25 risk assessment done by the owners group. We did our own l HERITAGE REPORTING CORPORATION -- (202)628-4088 1
106 1 independent risk work. The risk expert will.be here~ tomorrow, 2 I believe it is along with his consultant, so I may undersell 3 or oversell in his words what was'done, but in essence we 4 wanted to make sure that what we had previously concluded so 5 to speak was that the plant, B&W plants do not appear to have 6 as much, have about the same level of risk as the other PWRs, 7 that indeed that was still true in light of'the operating 8 experience and that was a major focus of the risk assessment 9 effort, along with trying to look at what is the risk 10 assessment telling us is where the program ought to be 11 focusing? What systems should be fixed? What are the issues?-
12 Where could they get the most bang for the buck so to speak?
() 13 DR. KERR: Excuse me. I thought I understood you to 14 say that you looked for plant risk primarily using PRAs, and 15 you concluded that the risk of B&W plants was about the saine 16 as that of other plants. Is that--
17 MR. JONES: That was talking to the risk guy at the 18 time, looking at the risk assessments that were available in 19 say 1986 timeframe. That's what they appeared to 20 indicate--ten to the minus 5, 6, 21 DR. KERR: I just want to make sure I understood it.
22 Then you said something which I thought I understood you to L
23 say this was true in spite of the operating experience?
(} 24 MR. JONES: What we wanted to do was make sure that 25 what we were seeing from the operating experience, that was i
I l
HERITAGE REPORTING CORPORATION -- (202)628-4888.
i 107 ,
1 being properly reflected in the PRAs, that the PRAs had an
(} 2 appropriate basis, that the operating experience didn't say i 3 that the PRAs were invalid.
4 DR. KERR: How would you tell whether operating !
5 experience was reflected in the PRA or not?
6 MR. JONES: At that point in time I have got to step i
7 back. I-would need the risk expert to standard up here and 8 tell you what was done and why and how.
9 MR. WARD: Are we going hear that tomorrow?
10 MR. JONES: You should be hearing that tomorrow. ;
, i 11 Somebody will be on the phone to tell him that for sure, I >
12 hope. You should be hearing that' tomorrow.
i 13 MR. CATTON: That is an important connection. l
[
14 MR. JONES: Yes. That was one of the focuses of our :
i :
) 15 effort along with evaluate what the owners group did,. which '
I 16 we, we had disagreements with how they did things, ended up 1
17 doing our own independent assessment as a result of that. I i
18 MR. CATTON: Did you try to track through some of 19 the various events that took place using PRA to look at the 20 situatiors?
21 MR. JONES: In fact that is exactly what we did was 22 we took the category C events, developed a humongous event .
, 23 tree of I think it was over 200 events. There is, the risk 3
d
. 24 consultant is in now--about 200 events in the event tree, to 25 make sure we could follow those and track them and try to bid 1
HERITAGE REPORTING CORPORATION -- (202)628-4888 i
108 1 all these things and get a relative probability for these 2 things. What it means from a percent.
3 HR. CATTON: If the risk is the same, does that mean 4 most of those events were uneventful?
5 MR. JONES: Let me hold off on drawing a conclusion, 6 a specific conclusion on that, and let the risks people do i 7 that. Okay, but that is one of the subjects or main thrusts 8 of the evaluation that they did.
4 9 MR. REED: I still have a problem as I started out .
10 the meeting with this program scope shift. It seemed to me 11 what has happened has been a good job of, as represented by 12 that chart, of finding out what caused the trips, what caused ;
() 13 the complex transients and so on and so forth, and you narrow 14 it down then to a system or the main steam safety valves. .
] 15 Now it seems to me that you dropped from this :
1 16 program scope, this issue of basic, the words basic design j 4
17 requirements, and thermalhydraulic details, comparison and 10 details. <
19 HR. JONES: The thermalhydraulics came from the 20 sensitivity work. ;
i 21 MR. REED: Well, to me it isn't, it doesn't go far )
l 22 enough. The program scope has got to be more operational than !
i 23 statistical and they have got to be design basics.
l
[}
24 MR. JONES: This is how we interpreted the Stello 25 letter within roughly two months after it was done, and with l
l HERITAGE REPORTING CORPORATION -- (202)628-4888
109 1 the blessing of management that was an appropriate scope for p) s_ 2 the program.
3 MR. REED: So there was a change somehow.
4 MR. CATTON: Are you going to--maybe I should ask 5 you, I asked earlier about the MIST counterpart tests, 6 counterpart to Davis-Besse and Rancho Seco. In fact they 7 really couldn't reproduce them. How much confidence should I 8 place in the sensitivity study, if I don't know enough to run 9 the test on MIST or even get close?
10 MR. JONES: First off, MIST was never constructed to 11 examine those transients. Okay. I want to make that clear.
12 It was not constructed for that purpose. It was constructed
(}
t 13 to examine natural circulation, scoall break LOCA type 14 behavior. That was the basic premise upon which MIST was 15 founded.
16 Now when we looked at putting together this Phase 2 17 or Phase 4, depending on what numerical scheme you want to use 18 in dividing up the work, but to follow on this program that's 19 essentially completed now, we asked what tests could we do?
20 And we decided we would try to look at some of these 21 counterpart tests and see how well MIST could produce that 22 behavior.
23 We recognized that there are scaling compromises 24 made in MIST, and part of the problem I think had to do with 25 those scaling compromises which had to be made. It was not HERITAGE REPORTING CORPORATION -- (202)628-4888
I i
110'
)
1 constructed to be a very detailed steam generator simulat' ion. i 2 As far as what I think would be the significant .
3 issues you saw from the SMUD test and the Crystal River test, 4 and that's why we have ongoing a~ discussion with the B&W 5 owners group about the ability of the computer codes, to, to 6 look at steam generator behavior, and we are discussing the 7 need for follow-on testing related to the steam generators, 8 and that's the so-called TAG 2 process, which is currently l 9 underway and should be wrapped up sometime this summer or 10 early fall. That program is underway.
11 MR. CATTON: Why did Idaho use RELAP 5 rather than 12 TRAC which is tuned to B&W?
() 13 MR. JONES: If I could hit that when we do the 14 sensitivity, I would appreciate it. ,
15 MR. CATTON: That's fine.
16 MR. JONES: Okay. I would like to make another 17 comment on the response to some, some of the things that Mr.
18 Skillaan talked about.
19 The owners group set a set of goals for reactor 20 trips, complex transients. They also came up with a 21 categorization scheme for the transients. ;
22 The staff has never bought that. We have not bought 23 those goals as being appropriate. In fact, what we said was
(} 24 we thought that to some extent, they were not ambitious 25 enough, certainly the goals, and that it is stated in the SER i
i HERITAGE REPORTING CORPORATION -- (202)628-4888
111 1 clearly.
p
(.) 2 The owners group is using that as a measure upon, a 3 means upon which to measure whether they are achieving 4 progress by implementing these recommendations and whether 5 they are indeed fixing the problems they have seen in this 6 plant, and they are probably fine and appropriate for that, 7 but as goals, specifically safety type goals, we haven't 8 bought that. I want to make that clear.
9 CHAIRMAN KERR: What would you buy?
10 HR. JONES: We never wrestled with it. We didn't 11 try.
12 DR. KERR: How did you know it was not a good goal
() 13 if you don't know what goal you would accept?
14 HR. JONES: I will give you the area that we 15 basically had our biggest problem with was the category C 16 transients.
17 DR. KERR: I'm sorry. I am asking a question, and--
18 HR. JONES: I am going to try to address it. The 19 one we had the most difficulty with--you know, you can't 20 prevent all reactor trips, and a goal of two per year is 21 probably not an unreasonable goal, but the complex transients 22 we scratched our head about.
23 First off, we had trouble with their 24 categorizations, and in all honesty if I put ten people in the 25 room, I would get ten different categorization schemes.
HERITAGE REPORTING CORPORATION -- (202)628-4888
-112- )
! 1 DR '. KERR: I am not pretending it is an easy thing l() 2 to do. I am simply saying you said that you did not accept 3 their goals. When I asked you what you would accept, you said 4 you didn't know.
5 MR. JONES: That's right. And then you asked me 6 why.
7 DR. KERR: No, I didn't ask you. I said how could ,
8 you decide you wouldn't accept their goals when you didn't s
9 know what you would accept? ;
10 MR. JONES: Let me give you the--category, the 11 categorization or the, the goals they set for the category C 12 transients in our opinion were not very aggressive, and the 13 reason we came up with that conclusion is what we did was we i O ,
14 said--
1 15 DR. KERR: What does aggressive mean?
16 MR. JONES: Let me keep explaining and then if I 17 haven't answered your question, I will try again.
18 If you took, at the time this program came up and
- 19 the time they put the categories together, if you took the
~
20 ratio of number of category C events which had occurred, and 21 you took the reactor trip frequency that had occurred, you 22 would come up.with a number that is very much like the .1 of 23 category C events per reactor trip per year, so that 24 effectively if they would achieve their two reactor trips per i
25 year, and do nothing else to the plant, you would get the .1 f
HERITAGE REPORTING CORPORATION -- (202)628-4888 l
113 1 almost by default. :
'(} 2 DR. KERR: What does by default mean?
3 HR. JONES: The ratio would come down to .1. What 4 happened was the ratio of .1 to the two reactor trips comes 5 out to whatever number, .005. If you go back and look at what 6 things were experienced, that's exactly what you had, so if 7 you just reduced the number.of reactor trips with the numbers 8 of transients, you are still getting the same percentage of 9 complex transients.
i 10 And we said gee, that doesn't seem like a very, that l
11 doesn't seem like you are trying to attack your complex 12 transients other than trying to reduce the number of risk 13 trips you are getting. If you wanted to make sure, we thought 14 there ought to be a low goal for complex transients to assure 15 they were aggressively seeking fixes to the causes of these 16 complex transients. Let me also-- .
17 DR. KERR: Suppose you didn't have any trips and 18 therefore had no complex transients? Would you still insist f
19 on the same ratio?
20 HR. JONES: That's an undefined, but-- I 21 DR. KERR: To me, it seems to me the important thing <
22 is how many complex transients you want to, events or how many l i
23 trips, and this seems to me to be relevant. It seems to me 24 '
the ratio is relevant.
}
25 HR. JONES: What we were concerned about when these j HERITAGE REPORTING CORPORATION -- (202)628-4888 ;
114 1 goals were constructed was that if your reference were placed
() 2 only on reducing the number of reactor trips, you still 3 haven't really taken care of maybe some of the fundamental 4 design characteristics that are leading to these complex 5 transients.
} 6~ Now I would also want to back up and say while that l
7 is their goal, okay, the way they went about the program is 8 they did look at what caused these complex trips so had the
! 9 goals almost to a large extent become irrelevant, but I did l i 10 not want to leave the impression that we bought the goals as !
l i 11 appropriate levels of safety. We never tried to tackle the i l
l l 12 issue. We didn't try to tackle the issue.
I 13 What we tried to do was look at are they looking at 14 the right systems? Are they looking at the behavior problems 15 in these plants? Are they fixing the right things and are l
16 they doing a, a fairly broad scope effort in looking at those 17 systems? That's what we concentrated on, not whether these 18 goals made sense, or whether they would achieve the goals.
19 DR. KERR: You didn't care about how many trips or 20 how many complex transients you had, as long as they went 21 about it in the right way?
22 HR. JONES: That's right. That's right. Rather 23 than knock your number, let's go look--
24 DR. KERR: We are more interested in procedure than 25 results? Is that a fair statement?
HERITAGE REPORTING CORPORATION -- (202)628-4888
115 1 MR. JONES: No. No. We are interested that they
() 2 get what--we are not looking at a number which says this is 3 what is acceptable, but rather what we were looking at is are ,
4 you fixing the things that are causing you problems?
5 Now that will give you some number, and the owners 6 group has stated that they will continue to try to improve 7 these plant performance, and that their real goal is to get 8 them down to zero ultimately, but under that base, we said 9 okay, let's not argue numbers, let's argue what you are doing 10 in the program.
11 DR. KERR: But if you don't look at numbers, how 12 will you judge that the program is accomplishing anything?
{} 13 MR. JONES: That's what we set the goals, as a way
[
14 of seeing whether you are making progress and whether you are, 15 whether you are really making improvements in the plant.
16 This type of approach makes sense. What we did not 17 want to do was argue specific numbers. That's the point. The l
l 18 goals are fine as having something to kind of look at, make j i
19 sure you are going in the right direction, have some way of l 20 measuring you are going in the right direction. We think that I
21 is a good thing to do.
22 MR. WARD: That's what goals are usually for, isn't 23 it?
24 MR. JONES: What we didn't want to do is start 25 arguing is that the right number? Because you know, it is is l
l i
HERITAGE REPORTING CORPORATION -- (202)628-4888
116 1 .1 complex transients up front the right number to shoot for?
/~T
(_) 2 We hadn't done the risk assessment work to say that is the 3 right number to shoot for. We hadn't figured out from a risk 4 assessment up front whether the complex transients as defined 5 made any sense from a plant safety standpoint.
6 DR. KERR: Let me hypothesize something. You said 7 that zero was not an appropriate goal. You said--
8 MR. JONES: It is a good goal. You are never going 9 to achieve it.
10 DR. KERR: You said two was not an appropriate goal 11 because it is too big.
12 MR. JONES: I said two is probably a reasonable goal
() 13 for trips.
14 DP. KERR: That leaves only one number it seems to 15 me.
16 MR. JONES: We said two is reasonable probably for 17 trips, and it is consistent with the industry kind of goal 18 standard so to speak. We thought that was okay, was a complex 19 transient that we didr't want to really argue that much about, 20 partly because then we had to get into the definition scheme.
21 That was the other thing I wanted to get into, how they 22 categorized these things, A, B, C.
23 There were tons of comments as to whether those were 24 the appropri(te ways to break it off. We did not try to 25 resolve whether A, B, C was appropriate, because the owners HERITAGE REPORTING CORPORATION -- (202)628-4888 l
117 1 group, we asked the owners group does that measure it? They
() 2 have said no. We said fine. You are not going te use as 3 measure of safety ARTS. That's fine. And what we were 4 concerned about was gee, are you looking at all these 5 transients which indicate off normal behavior? Don't restrict 6 your efforts to just the category Cs, and they didn't.
7 When they did look at the, where the problems were 8 happening, they looked at the Bs and the Cs. We said given 9 they are looking at all events that progress beyond what would 10 be a normal expected transient, let's not argue with it. They 11 are looking at everything that we want them to look at, so 12 let's not quibble over the details. And that's what we did,
(} 13 but I cidn't want to leave the impression we bought that.
14 Now just going back to how this slide is put
- 15 together, it has kind of already been covered. We did--as 16 noted already, there was a lot of emphasis placed on balance 17 of plant systems within the program. We did interact with the 18 owners group continuously throughout the entire program.
19 These were in working level meetings and formal.
20 communications, letters, questions, back and forth. We 21 believe given the scope of this, the interaction, with the 22 ataff, that the overall program was comprehensive.
23 Looked at thermalhydraulics, tried to 3 .
at the 24 other, it tried to look at the systems tl.st wer) :rablematical 25 at the plants. From that standpoint we thought the program HERITAGE REPORTING CORPORATION -- (202)628-4888
118 1 was comprehensive and we think the recommendations that came 2 out of it'will-improve the safsty of B&W plants. We will get 3 into a few specifics in a few minutes.
[ 4 HR. CATTON: When you did this, you didn't find that I
- 5 it was a little df.mconcerting that the pace was much quicker 6 with the B&W reactor?
j 7 MR. JONES: We will talk about that in the j l 8 sensitivity.
l 9 MR. CATTON: I am just asking for a yes or no.
10 DR. KERR: What was your question? j l ?
l 11 MR. JONES: The answer to that is to a large extent, l 12 no.
13 MR. CATTON: We have been discussing the sensitivity O 14 issue off and on. It is focused very tightly on 15 thermalhydraulics, and that's all it looked at, and the fact 16 when it dries out in four minutes rather than 20 or 60 17 minutes, doesn't seem to ring any bells with anybody, so I 18 just keep trying to get them to focus and t. ell me the fact the 19 operators can do things quicker doesn't matter. It does 20 matter, and I have not been able to get an answer.
21 MR. JONES: The answer is it does matter, to some 22 extent, and it was reflected in the PRAs, and from that 23 standpoint, although they did have some things they have to do 24 quicker, it is not a problem, a large problem. I mean you 25 will find from the PRAs that one of the things is, improvement HERITAGE REPORTING CORPORATION -- (202)628-4888
119 1 of the operator performance is one of the areas that would
() 2 improve the risk of these plants.
3 MR. CATTON: I have found in the Idaho report--I am 4 not sure which volume it was where they discussed the fact 5 that there was, that one had to incorporate a factor of ten to 6 40 on the operator error rate.
7 It seems to me a factor of ten to 40 on operator 8 error rate would make a big difference between plants and 9 somehow this quicker pace certainly should impact on that and 10 that ought to be a, a number that you could track right out 11 the end on the PRA.
12 HR. JONES: Yes. I think the PRA did that, and he g 13 will address it. Okay.
~)
14 Now one of the things, while we were in a lot of 15 agreement with the owners group as their chart indicates, what 16 their chart does not indicate, however, is we had a lot of 17 comments that they did not incorporate.
18 DR. KERR: Wouldn't you have expected them to 19 incorporate all the comments?
20 MR. JONES: No, not all of them, but we had them 21 looking in the ICS--
22 DR. KERR: They were comments that should be, have 23 been incorporated, or throw-away type comments?
24 MR. JONES: There were some comments which arguably O 25 they should have incorporated.
t HERITAGE REPORTING CORPORATION -- (202)628-4888
120 1 DR. KERR: The fact they didn't incorporate all of
() 2 them is not the issue. The issue is that they didn't !
3 incorporate some that you felt they should have?
4 MR. JONES: That's right.
5 MR. WARD: Did they know which were which? Did they 6 have a way of knowing which were which?
7 MR. JONES: Well, I think, I think in the area of 8 the ICS, in all honesty, I think almost every single one of 9 thors comments in our opinion were very valid comments that 10 should have been incorporated.
11 I think there were some of the earlier comments, the 12 very early comments associated with the negotiation of the
() 13 scope of the program, which where they told us we are not 14 going to do something, it didn't bother us a hell of a lot, 15 okay, at that, you know, recognizing we are in the negotiation 16 phase almost at that point in time, so we were pushing the 17 boundaries of their program as far as we could push it to try 18 to look at as much as we can, and get as comprehensive a 19 review as we could.
20 So either that was kind of, you know, almost 21 negotiating room. The more detailed comments that came 22 through the meetings, through the interactions, the letters, 23 those were comments we believe should have been incorporated, 24 all of them.
25 They, we think because of that, we are going to see HERITAGE REPORTING CORPORATION -- (202)628-4888
121 1 some problems in these same systems that are causing problems 2 today that we have known about, and seen. -
{
3 CHAIRMAN KERR: You are convinced all of these t 4 comments would have improved the system had they been .
5 incorporated? ;
6 HR. JONES: We think if they were incorporated into .
7 the original program scope--these were more scope questions to 8 a large extent. It would have improved the system.
9 DR. KERR I thought you were talking about changing 10 or something that would be made to the reactor?
11 MR. JONES: No. Most of our comments were not of i
12 the changed nature. Most of them were of the nature of in i
13 your ICS review, ensure that you rolook at your responses co "
14 INE bulletin 79-27 and assure that you have proper fixes for ;
15 that. And we will get into that that's one of the so-called i
. 16 significant issues. We think because of that some of these I
- 17 concerns may pop up again which we have known about which we '
l 18 have expressed.
19 DR. KERR: It isn't what is finally done to the 20 reactor, to the operation of the reactor, that you have 21 concerns about? It is the description of the program that you 22 are concerned about?
i 23 MR. JONES: No. If you don't define your program, 24 you can't--if it is not included in your program, you can't i 25 fix the issue necessarily. Okay. ,
- i b '
HERITAGE REPORTING CORPORATION -- (202)628-4888
122 1 You will see that when we talk ICS. You are going
() 2 to see a series of concerns that we have expressed to try to 3 see whether they were resolved, and you are going to see the 4 answer is a mixed bag. All the old concerns we have 5 identified in the ICS/NNI, not all of them are going to be 6 addressed.
7 DR. KERR: See--
8 HR. JONES: And therefore, those concerns are not 9 going to be resolved, and you are going to see or it is 10 likely, there is some additional likelihood because the 11 program didn't look at it, that the fixes you have come up 12 with aren't going to fix some of those issues and you are 13 going to see those very same types of transients that you have f}
14 seen before.
15 DR. KERR: It is entirely possible it seems to me in 16 my own experience in reading various generic letters and 17 issues, that over the course of a few years, perspectives 18 change, perspectives change and the perspectives of the staff, 19 and to ask that one take into account everything that has been 20 written over a period of four or five years may involve taking 21 into account even contradictions, and I am trying to find out 22 whether--
23 HR. JONES: I don't think that was the case.
24 DR. KERR: Trying to make certain they satisfied a 25 number of publications or whether they attacked the problem i
HERITAGE REPORTING CORPORATION -- (202)628-4888
123-1 appropriately, and it saems to me the two are not necessarily 2 parallel.
3 MR. JONES: When you hear the ICS/NNI discussion, I 4 will know--I haven't looked through the owners slide, but I 5 know the approach we use, and I know the approach.they 6 generally use. Their first things that they did was to come 1
7 up with a set of design requirements for that system. We have 8 a set of concerns that have, we have-seen with experience of 9 problems with the ICS. We would think that those concerns 10 should, resolution of some of those concerns should be 11 embodied in this design requirement.
12 The design requirements document is then where they 13 took the plant designs and compared to, said this is what we .
4 O 14 are missing at this given plant. Therefore, these are the -
15 types of things we have got to fix. ,
16 Now if you haven't assured up front these concerns i
17 were at least examined in the process, then your design 18 requirements are somewhat inadequate, and then we are saying 19 that you have not done as comprehensive a job of looking at 20 the ICS, for example, as you should have done, 21 DR. KERR: You feel that either the staff
- 22 understands the system better than the owners group, and i-23 therefore the owners group is not capable of defining lt 24 requirements, or else they have deliberately left out i'
O 25 something? .
t HERITAGE REFORTING CORPORATION -- (202)628-4888 ;
1 l
124 l
1 MR. JONES: I will leave that to both sets of
() 2 experts.
3 HR. MICHELSON: Don't leave the slide yet.
4 MR. JONES: I think the very specific issues I am 5 going to back down, but I think yes to some extent, that's 6 where we have some disagreement. As noted on the content of 7 the ICS/NNI review, there are substantial issues--I don't want 8 to say substantial--there are issues on content.
9 HR. MICHELSON: I have a real problem with program 10 scope. I read the Stello letter that talked about a need for 11 a broad evaluation of the B&W design. It addressed the staff 12 by saying the staff will reassess overall safety of B&W
(} 13 plants, so overall safety to me maybe means a little 14 differently than it does to you. I read it as being 15 transients and accidents including external and internal 16 initiators.
17 If I really want to talk overall safety, that's l 18 beside the point a little bit, and then I go to page 9-86, 19 your staff evaluation. You indeed point out some shortcomings 20 of the, this whole busit. ass. When you read the analysis, 21 where I am addressing only sequences similar to category C 22 ovents, that have already occurred at B&W plants. "Results 23 therefore do not reflect the contradictions of other important ,
1 rs 24 initiators such as the loss of off-site power, loss of service
(' >
25 water, or external events."
HERITAGE REPORTING CORPORATION -- (202)628-4888
125 1 Then I go on to page 9-9, and it points out a (O
_/ 2 general finding. It says, "Of significance is the finding the 3 B&W event trees are not fully representative of the observed 4 category C events." Let me finish. That is, there does not 5 exist on the event trees a clearly identifiable sequence which 6 adequately represents the observed scenario for each of the 7 category C events.
8 Now these are bits of information which lead me to 9 believe that you really haven't assessed the overall safety.
j 10 It is extremely important areas that you have not addressed 11 even to the extent of saying why you didn't address them. You 12 just put it in as one place, buried and forgotten.
() 13 You know, what, what happens? Am I to believe that 14 you are really assessing overall safety or am I to believe you 15 have only looked at certain transients that to determine what 16 their safety contribution might be? And besides that, on the 17 previous page you say these weren't all that important anyway, 18 because it says category C events are not likely to be
, 19 significant contributors to core damage frequencies for 20 Oconee, Crystal River, ENO, Rancho Seco, particularly when 21 compared with other potential contributors such as Station 22 Blackout, loss of service water, or external events. You have 23 looked at a minor part. You admit you have looked at the 24 minor part, and you haven't really addressed what you think is 25 the major part of this problem, and that's assessing overall HERITAGE REPORTING CORPORATION -- (202)628-4888
126 1 safety, and I say no, I just can't, can't buy it.
() 2 HR. JONES: I gave you the scope. We used six and--
3 HR. HICHELSON: You didn't justify why you didn't do !
l 4 what Stello told to you do. !
5 HR. JONES: An I said, we put that program together 6 in 1986. His assistant was involved in it, and we never got 7 any comments as to the scope of that program was wrong. l 4 8 And in fact, one time we got told that some places ,
9 we were going to look we shouldn't look, so I think to some 10 extent it was looked at, and felt that the scope was 11 appropriate, and I would have to assume after the Stello 12 letter, or close to the Stello letter, it was deemed l I
l
(} 13 appropriate scope, 14 With respect to your very specific comments, I would j
1 l
1
! 15 like you to if you want addreas those to the risk assessment i
, 16 person tomorrow. He will be there, but I do want to note that 17 a lot of the deficiency you are noting which were with respect
- 18 to what the owners group did, we have at least partially i
19 compensated for by doing our own detailed independent risk 3
l 20 assessment. That's one of the reasons we did it.
21 HR. HICHELSON: I don't dispute that. I don't find j 22 in here you went back and looked at yours, for example, as 23 external events. I think you said earlier you didn't, you i 24 didn't cover that one.
J l
1 25 HR. JONES: No.
1 i
d HERITAGE REPORTING CORPORATION -- (202)628-4888
127 1 HR. MICHELSON: How about the case of pipe breaks
() 2 outside of containment? Did you do an analysis of those?
3 HR. JONES: No.
4 HR. MICHELSON: This to me is what overall safety is 5 about. It isn't just about transients being the only 6 contributorn to risk. It is one of the contributors maybe, 7 only a minor one. You kind of admit on page 9-7, you said 8 really these are the minor ones.
9 HR. JONES: We looked at why are we in this program 10 in the first place? We are in here because of the operating 11 experience. Therefore, our primary thrust was the operating 12 experience.
(} 13 HR. MICHELSON: Then I think Stello ought to, well, 14 when we write a note, I think the Committee ought to point out 15 whether or not they think you reassessed the overall safety or 16 not. Overall safety to me means a lot more than transients.
17 MR. JONES: That's fine. I just wanted to give 18 where we came from and make sure it was clear that is the 19 scope of our review, and that's what it was limited to.
20 MR. DAVIS: Isn't it true, though, that you do have 21 as part of this exercise, some PRAs for B&W plants which have 22 looked at all of the initiators, and those PRAs are not 23 showing significant safety problems?
24 HR. MICHELSON: You better go back and look 25 carefully at how well they model external events before you HERITAGE REPORTING CORPORATION -- (202)628-4888
128 1 say they looked at external events.
() 2 HR. DAVIS: The Oconee PRA is considered a state of 3 the art PRA and did look at external events, and did in fact 4 find initiator that was important and fixed.
5 HR. JONES: We did look at the existing PRAs. Okay.
6 Now as to the extent of those, and the details of those PRAs, 7 hold your question for the risk expert, please. Okay.
I' 8 HR. SIEGEL: Can I make a suggestion? I-understand 9 the frustration apparently of the Subcommittee on these issues 10 because since we have started this today, Mr. Reed and Mr.
l 11 Catton, and Dr. Michelson have expressed a concern in their 12 particular areas. l 13 But it would, I think a major point we understand, I 14 think you ought to make it your letter. I think you really j l
15 ought to try to assess the program on the merits that it has 16 and not try to make it what it isn't, and you should make that 17 point when you send your letter to Stello. l l
18 HR. JONES: Now just to give you a flavor of the l l
19 review, and that's the only purpose of this chart, these are l l
l 20 the systems that were looked at, as part of the overall 21 program--turbine, main feed, ICS, electrical valve, et cetera.
22 These are the approved, number of approved recommenaations in ;
1 23 the B&W recommendation tracking system.
24 These are the number of key recommendations which O 25 were identified by the owners group. The high priority HERITAGE REPORTING CORPORATION -- (202)628-4888
l 129 l
1 recommendations are additional recommendations and the. staff l
() 2 looked at them, that we thought.ought to be given a high 3 priority attention in their implementation phase, and those j i
4 were outlineed in the report. l l
5 We had some disagreement with some of the I l
6 recommendations which are noted in, in the SER, and we have i 1
7 provided some additional guidance, comments, on the number of l 8 recommendations, roughly along here, and again you can see the 9 one that b the, well, we have had the most difficulty with 10 was the ICS/NNI. The next five slides kind of go through some 11 examples of what are key recommendations, what are high 12 priority types of disagreements and guidance we have.
13 I am not going to throw those up. They are simply 14 there to provide you an overview to the flavor of the 15 evaluation and the recommendations that came out, and you will 16 hear more about that I am sure as you go through the detail, 17 specific comments of the review and can use that as a hit list 18 if you wish.
19 (Slide) 20 HR. JONES: For any you wish to understand further; 21 I would like to drop, however, to put a perspective on the 22 program, and where we come out with this thing.
23 This is somewhat different perspective than is 24 provided in the summary inclusion section of the SER, but it O 25 is not inconsistent with it by no means.
m HERITAGE REPORTING CORPORATION -- (202)628-4888
130 1 What were the significant improvements that came out
() 2 of this program? We saw a couple of things. Number one, a 3 lot of the most difficult transients that have been 4 encountered on the B&W plants have generally been in response 5 to ICS/NNI failures, difficult from the operational sense, and 6 causes a lot of plant and equipment to go to some strange 7 states.
8 There have been, there are two recommendations.
9 CHAIRMAN WYLIE: I think they are very key, 10 associated with improving the response to ICS/NNI failures.
11 One of them is the concept of known safe state. This is if 12 the ICS loses power, you will specify failure states of 13 equipment, so that the transient is basically benign, so you
[}
14 should not have things like the Rancho seco event of 1985.
15 Recovery of ICS/NNI power also can cause equipment 16 to move to some strange states, require some manual actions, 17 rapid manual action by the operator to put it back in manual, 18 and take control of the plant. It involves, said that the 19 plant will stay in known safe state when you put the ICS back 20 if you lost power and start to put back components. We think l
21 those are significant. l l
22 HR. MICHELSON: On the known safe state, one of the J
23 problems we have observed in LERs is the equipment doesn't 24 necessarily see a total loss of air pressure, for instance, or 25 total loss, but rather sees a degradation of the air supply or HERITAGE REPORTING CORPORATION -- (202)628-4888
131 1 the holding supply or whatever, and those lead to yet other
() 2 states which are not necessarily safe and are not necessarily 3 included in the analysis.
4 HR. JONES: I understand that, and we have some 5 comments on known safe state. We think the concept, we think 6 the concept is very good where there are a lot of things l
7 embodied in that concept such as the operator shouldn't have 8 to take actions that are abnormal from what he normally takes.
9 The response for reactor trip; we are not sure all plants can 10 achieve this right now, and we would like them to pursue that 11 and make sure that that is what happens.
12 Again, if you want to get at level detail on the 13 concept of what it covers and doesn't cover, I think the 14 ICS/NNI expert is here, but we have had comments on known safe 15 state. We think the concept and what has been done are vast 16 improvements. SHUD implemented this, and we think, I think we 17 are generally happy with it.
18 HR. MICHELSON: As I recall, that the Crystal River 19 event, loss of voltage, kind of caused serious control 20 instruments to fail. Some fail upscale. Some fail downscale.
21 HR. JONES: That's the question that's embodied in 22 this known safe state.
23 HR. MICHELSON: I think what you are saying is that 24 they have gone through and reanalyzed all of these situations 25 so that they know better just what their state is on this kind HERITAGE REPORTING CORPORATION -- (202)628-4888
132 1 of condition if it develops?
() 2 HR. JONES: I would rather have the ICS/NNI expert 3 answer that.
4 I am not sure tnat is totally a true statement. I 5 think it is, this is the recommendation is to make the 6 modifications necessary to achieve the known safe state. They 7 haven't defined it, or what those specifics are, and there are 8 plant to plant variations that would have to be considered )
9 anyway.
10 HR. ETHERINGTON: These are necessary 11 recommendations at the time. Do you think some of them should 12 become requirements? ,
13 MR. JONES: These the owners group recommendations.
14 MR. ETHERINGTON: Yes, but do you think any of them 15 should be requirements?
16 MR. JONES: This program was, was initiated as a 17 cooperative program. We did not and have not decided to make 18 any of these requirements at this point in time.
I 19 However, we have full intention of following up,
- 20 monitoring their program and assuring that these are 21 implemented as Mr. Siegel will talk about, including going out 22 to the plants to audit the fixes that are out there, and if we 1 23 find that they are not doing an appropriate job, we will take 24 whatever necessary regulatory action is required at that time.
25 HR. SIEGEL: In answer to your question, the HERITAGE REPORTING CORPORATION -- (202)628-4888
1 133 l
1 operators group or the. utilities through the owners' group has 2 committed to implement all the applicable recommendations for 3 their plants.
4 MR. ETHERINGTON: The individual--
5 MR. SIEGEL: Yes, through the owners group; the 6 owners group is monitoring their, them and how they are 7 implementing these recommendations. They will probably, we 8 will get it in more detail later, but they have that 9 commitment. We don't have any requirements, but so far, from 10 what we have seen, they are following through on those 11 commitments. We may disagree on the times, but that is 12 another issue.
13 MR. JONES: Another area we think they have made O 14 progress, this integrates several of the actions they are 15 taking, is in the area of the heat sink reliability. There 16 has been a lot of discussion this morning about the heat sink.
17 In the B&W machine, there are basically three means 18 of decay heat removal available. Initially it is the main 19 feedwater system. If that reactor trips, they will run back 20 main feedwater, attempt to use the main feedwater system. If 21 the main feedwater system is unavailable, then they would go 22 on to the emergency feedwater. If the emergency feedwater 23 system is unavailable, then you would end up in the feed and 24 bleed type mode.
25 How they have, there are numerous recommendations.
HERITAGE REPORTING CORPORATION -- (202)628-4888
134 1 None of them are listed here as far as the detail, but there
() 2 are numerous recommendations to try to keep the main feedwater l
3 system on line, and that includes the control aspects from_the f 4 ICS controller valves in the main feedwater system, to ensure
)
5 proper runback of one feed pump trips the other, handle the l
! 6 load without the plant tripping, for example. I l
l l 7 There are EFW reliability improvements. There are 1
8 several issues associated with the ensuring that the secondary l 9 pressure control is maintained where you would like it to be.
1 10 Ar.d then there is, there are, have been problems with motor 11 operated valves, which there is an INE bulletin on. I can't i 12 remember the number, but there are actions being taken by the
(} 13 owners group with respect to valves which will have impact on 14 the HPIS system, makeup system, et cetera. These will enhance 15 the heat sink availability.
16 HR. REED: But I believe in that chart of 17 equivalency with other PWRs that Mr. Skillman showed with 18 about six or eight bullets, and the bullet number one was, 19 said that the B&W system was more adequate on, was, had less 20 risk on core damage than the other PRAs, or implied that.
21 Now I would like to find out if the staff agrees 22 because it is a very important statement made by tir. Skillman.
23 I challenged his statement us because of the complexity of 24 emergency feedwater on the secondary side. He said we didn't 25 use that. What we based this adequacy on was the primary HERITAGE REPORTING CORPORATION -- (202)628-4888
135 1 plant bleed and feed system, and then he said the primary
(} 2 plant bleed and feed system gets its water from the makeup 3 pumps, the primary makeup pumps, charging pumps which run all 4 the time. Only one is used and therefore, this always running 5 system, pumping system, is more reliable than the other PWRs 6 who have them shut down and swing them into service.
7 Now that was a very, very key statement, and I think 8 the industry from my some 35 years in it has struggled with 9 whether or not a system at rest is more reliable than a system 10 operating or pumps and equipment operating. .
11 Do you agree with Mr. Skillman that a system 12 operating is a more reliable safety system than one at rest?
13 HR. JONES: If you would hold that to the risk 14 assessment expert, but I will say I don't think the staff is 15 saying that we think the B&W plants have less risk than the 16 other types. I think what we are really saying is they are i
17 comparable at best.
la MR. SKILLHAN: That's what the slide said, too.
l 19 HR. JONES: Okay. That's where we are at. j 20 HR. DAVIS: Hay I ask a question about a specific l
21 transient? I
! 22 HR. JONES: Sure.
23 HR. DAVIS: Under Station Blackout conditions which l
24 are generally found to be fairly significant in terms of risk, O 25 you're left with really only one way, one means of removing i
HERITAGE REPORTING CORPORATION -- (202)628-4888
l 136 1 heat, and that's the emergency feedwater, is that correct?
() 2 You can't operate high pressure injection under loss of all 3 AC? ,
4 MR. JONES: That's right.
5 MR. DAVIS: Now do you know if the B&W plants go 6 into a cooldown procedure upon Station Blackout conditions?
7 MR. JONES: We did not examine Station Blackout in 8 this thing, in this program, and that was by choice. The 2 9 reason for it is it is an ongoing generic issue, and 10 therefore, we didn't see the need to incorporate--we would 11 have to accelerate to make a decision on this plant as opposed 12 to anybody else's plant. We did not see any particular reason
[} 13 to accelerate it for the B&W plants. Therefore, we let it go 14 as part of normal resolution in the, on the Station Blackout ;
r <
15 issue.
16 HR. DAVIS: You don't know what the precedures are
]
17 for Station Blackout?
1 18 HR. JONES: No. 1 l
19 MR. REED: I have got to get something straight.
20 Did you say that the high pressure safety injection pumps i l
21 would not have power on loss of AC and does that mean that i 22 these charging pumps on the B&W system, these primary makeup 9
23 pumps would not have AC power on loss of outside AC? ;
- I 24 MR. JONES
- No, no. He was talking about Station i O 25 Blackout outside, all AC. That's what I was answering. Okay. !
l i
i HERITAGE REPORTING CORPORATION -- (202)628-4888
. = --
137 ',
i 1 The last one was they are making, there have been (
() 2 numerous recommendations associated with the instrument air 3 system which we think are significant improvements to the,
! 4 improvements in the plant, but there are other improvements 1
5 which we think are important, but we don't want to, we didn't ,
I 6 put them in a significant improvements category based on j
7 judgment.
8 There are, again there are numerous actions defined i
9 in the program to reduce reactor trips, to input signal l 3
10 failures in the ICS by hardening signals, improving l 11 maintenance, and tuning of the ICS. Run through some of the i
12 more benign transients, main feedwater system improvements so 1
(} 13 they can run back and handle loss of single main feed pump and
! 14 turbine system which has been a cause in the past of numerous ;
15 reactor trips.
l 16 We think that we are going to reduce challenges to 17 the safety systems by improving the main steam pressure <
i i
- 18 control. You won't depressurize the steam generator, less J
I l 19 likelihood of challenging. HPI system, again reducing ;
J l 20 overcooling, associated with main feedwater in EFW; again it l
j 21 is to keep you down off the, off the high pressure injection 22 system and in fact, part of these, in reducing reactor trips, F
I 23 will keep on the main feedwater system; also reduce challenges 24 to the emergency feedwater system.
O 25 Improvement in response to ICCS are going to start i
l l
HERITAGE REPORTING CORPORATION -- (202)628-4888 .
g 1.s_-. -..u . . u, a.--u .= x . . a -. _w a . x s . -. . - - -
1 l l
138 ;
a
! I challenging the safety systems in the plant. There are again-o I
j
{} 2 operator burden issues and operator performance issues where 3 there have been a bunch of recommendations made which we also i
4 think.will improve, will decrease reliance on the operator by I 4
i l 5 decreasing the number of events. .;
! 6 Finally, we think'that it is also going to reduce
^
7 PTS risk. The largest problem with these B&W machines is they i r
- 8 are really prone to overcooling. A lot of these actions will l
- j. i 9 help mitigate that or keep it under control. j i
10 (Slide) !
j !
l 11 MR. JONES: That's the good news. What is the bad l 1 i 12 news so to speak? What do we see as the significant issues l I i 13 remaining from this program? Implementation--progress may be j
(:) 14 fine and it may have all the right stuff in it, but if it i
i 3
,3 !
15 doesn't make the hardware changes and software changes, you ;
- 16 haven't got anything except a pile of paper. That's the l
! l
! 17 key--Implementation, i r
! 18 And what we intend to do is to monitor the plans and j
j 19 progress, monitor the recommendations, track system reports i
20 that will be provided by the owners group, and we are going to l 21 go out and do some plant-specific audits, do some inspection.
l.
j 22 Human factors issues, there has been some discussion 23 on that earlier. In the area of human factors, there have iJ
- 24 been some significant human factors improvements. Okay. We 1
(:) 25 are not saying that what they have done isn't good. We are 1
) HERITAGE REPORTING CORPORATION -- (202)628-4888
l
' 139 l 1 just not sure it is far enough. They didn't have a human
() 2 factors expert involved in their program, and as a result, we l 3 are not sure how deep they could have gotten into the overall 4 issue of the operator. We think they could have done more, l
j 5 and we have suggested they go back and look at what they have l
6 done with the human factors expert and the data they have l
7 gathered as part of their various tasks under that issue, we 8 have rejected it at this point.
l
! 9 (Slide) t 10 MR. JONES: Finally, the one I mentioned earlier, 11 INE BULLETIN 79-27; this bulletin came out after the Oconee 12 event. As I understand it, there was supplement that came out 13 after the Crystal River event in 1980. This issue is
[
l 14 basically the assuring that the plant responds reasonably and 15 you have adequate instrumentation and such to respond to a 16 loss of non-class 1A instrumentation and power system buses 1 17 during operation. The Rancho Seco event clearly indicated l
\
l 18 that modifications that have been made at the plant may not be l l
! 19 adequate. l 20 Now how widespread that is, we weren't sure. We l l 21 have looked at INE Bulletin 79-27, wrote it off back in 1981, 1 22 '82, in that timeframe, but given Rancho Seco which was also l )
I 23 written off, we said, gee where do the other guys stand with l l !
j 24 relation to this bulletin? So we requested them to 25 incorporate that as part of their overall ICS/NNI system l f i
HERITAGE REPORTING CORPORATION -- (202)628-4888 l
140 1 review.
() 2 They didn't do it. Their claim, and to some extent 3 it is a correct claim, it is plant-specific, and the 4 modifications were plant-specific, but we wanted to have some 5 assurance that this issue has been resolved. We think that if 6 you resolve 79-27, and fully incorporate this issue of known 7 safe state, the recommendation known safe state, you have 8 essentially resolved your ICS/NNI problems, the major ones.
9 Given that the owners are saying essentially that 10 79-27 is resolved, from their perspective, what we are going 11 to do is we are going to allow, ao part of the audit process, 12 we are going to look at how they resolved 79-27 and take
{} 13 whatever action is appropriate to assure that it has been 14 implemented properly.
15 We think that will go a long way to assure that the 16 ICS/NNI will be resolved. That summary 17 conclusions--information gathering--this is real short. The 18 owners have described earlier their activities that they used 19 in gathering information to determine where abould you go with 20 this program?
21 The staff, the staff's activities in this area 22 consisted of two things. Number one, we looked at what they 23 did. They produced some reports. They went through in detail 24 the TAP report and made some recommendations. We looked at 25 that.
l 1
HERITAGE REPORTING CORPORATION -- (202)628-4888 ;
I
141 I
1 In addition, to provide ourselves some assurance I
() 2 that what they were looking at was appropriate, we went in and 3 we audited about 40 of the 200 TAP reports. We also looked at l
4 the yearly summaries to get overall perspective of the type of 5 problems that we were seeing at the project. We looked at our !
6 previous concerns from NUREG documents that were produced in i l
7 response to these various events.
l 8 We asked the regions what do you think is wrong with l l
9 these plants? Some of those we incorporated as part of the
+ :
10 review effort. Some of those we passed to the owners group as 11 comments, as staff comments. We looked at whether or not 12 there was any specific actions that haven't been done similar 13 to Rancho seco and not having EFIC installed as of 1985, see
}
i 14 whether there was anything outstanding we thought we needed to >
) 15 accelerate, and we looked at the Davis-Besse decay heat 16 reliability improvement program which they did as part of 4
17 their restart effort, to see whether there was anything in 18 there that we ought, that ought to be generically considered l 19 by the owners group, and we found there were some and there 20 were--and some of the other were very plant-specific j 21 modifications, but a number of them were indeed already beingg
+
22 developed on the part of the owners group program.
l 23 DR. KERR: So far you have described this activity !
- 24 as if you sort of lumped all the plants in the same basket. ,
l 25 Is that a fair assessment? You sort of considered all of l
HERITAGE REPORTING CORPORATION -- (202)628-4888
142 1 these B&W plants as being about the same level of performance?-
O an aon=s:
2 1 ao e #t to Due ta wora we two eat 3 they were about the same level of performance because that is 4 not what we did.
5 What I would rather say is there may have been 6 experiences at other B&W plants which may be just as valid on 7 another B&W plant and therefore ought to be looked at for 8 generic implications. That's what we did.
9 DR. KERR: Well, emphasis seems to be on the plants.
10 HR. JONES: Yes 11 DR. KERR: Rather than on the operating 12 organizations.
13 HR. JONES: That is correct.
14 DR. KERR: And it occurs to me that both can have is significant influence on plant operation and yet the emphasis 16 seems to have been almost entirely on the plant.
17 HR. JONES: We will be discussing that in I guess a 18 few minutes when we talk about the concerns, the previous 19 staff concerns. That was one of your issues that you raised, l
l 20 but that was not part of SPIP, not part of SPIP or the, our l
l 21 overall evaluation. The owners group did not do that, nor did 22 we. I tell you that we didn't do it. We will talk about it 23 in a minute.
24 HR. ETHERINGTON: I think this didn't apply to--
25 HR. JONES: Excuse me?
HERITAGE REPORTING CORPORATION -- (202)628-4888
143 1 HR. WARD: How did we meet that? I l
() 2 HR. ETHERINGTON: I can understand that the owners i 3 group working together would not want to be critical of each 4 other, and taking into account the plant operation, I think 5 the staff should--
6 HR. JONES: We will talk about that in a few 7 minutes, but we did not. I am not hiding anything. We didn't l
8 do it.
I 9 DR. KERR: The fact you did not must imply that you 10 don't consider that important in plant safety?
11 HR. JONES: Boy, I wouldn't want to touch that 12 statement with a ten-foot pole.
( 13 DR. KERR: You have got to touch it--
14 HR. JONES: I don't want to agree with that 15 statement is what I am saying, but it was beyond the scope of "
f 16 this program I will tell you, but I agree. I would totally 17 agree that the operating organization does have bearing on the l 18 safety of a plant, but we did not look at it as part of this i 19 program. We did look at plant, the design characteristics of l 20 the plant, and that was the decision made up front. We I
21 thought that we already had ongoing mechanisms. SALPs, et '
22 cetera, which we could monitor the plants and the operating 23 organization. This effort was focused on the system.
I 24 HR. ETHERINGTON: Maybe the question is whether 25 there should be an extension. ,
t HERITAGE REPORTING CORPORATION -- (202)628-4888
144 1 MR. RUTHERFORD: If you look at.exparience with SMUD !
2 and Davis-Besse, the staff has been heavily involved in the O 3 revamping of the operating organizations for those two i
l l
4 particular plants, and I think that's the appropriate way to 5 handle it. It is a plant-specific issue if you will versus an 6 owners group issue. We do some things in the owners group as-7 far as looking at each other, and I will talk about that 8 later, but I think it was handled appropriately by the staff 9 in those two instances.
10 MR. JONES: Now in light of what we looked at from ,
l 11 the information gathering, what did we come up with? It 12 really is fairly simple. Given all the words that are up ;
13 there, what the bottom line is, is the operating experience O 14 was, did provide, or the looking at all this information, did
]
15 provide us the assurance thet the overall direction.of the I 1
16 program, the systems being chosen, the type of things that 17 were being looked at, were the appropriate th;ngs to look at.
18 That's the bottom line. And that's essentially what we did.
19 We didn't find anything glaring that was missing from the l 20 standpoint of the operating experience, and it was a 21 broad-based thing that looked at various things.
22 MR. REED: But in your comparative studies, you 23 picked out three plants--one CE. one B&W and one Westinghouse, 24 and I think I can say that certainly the B&W plant picked was 25 not, not a Davis-Besse--
HER.tTAGE REPORTING CORPORATION -- (202)628-4888
[. 145 1 DR. KERR: I can't hear you.
2 MR. REED: Was, the plant that was picked'was not a 3 Davis-Besse or a SMUD. You picked Oconee for your. comparative 4 PWR. In lay opinion, the Oconee plant,.the oldest, the most 5 seasoned operating organization, with their best-history 6 perhaps isn't the one that should have been picked for the 7 comparative conclusions.
8 MR. JONES: When you say from the standpoint of 9 sensitivity or the risk, Glenn, because I am not, if you talk 10 about it from the sensitivity, that's not, that doesn't mean 11 anything to us because we weren't looking at the operating 12 organization nor the operating history.
13 We'are trying to find out thermalhydraulically their O 14 significant design difference in response, and the, to be '
15 honest with you, I don't see tremendous differences between 16 the Oconee or SMUD or Crystal River, and to a large extent, 17 Davis-Besse. !
1 18 M* REED: There is even a design difference that l l
19 Oconee has a dedicated independent afterthought decay heat '
20 removal system.
21 MR. JONES: That was not credited for, in looking at 22 the thern alhydraulic, thermalhydraulic design characteristics. ;
23 MR. SKILLMAN: Could I please make a comment? We 24 did not only consider Oconee in the sensitivity study. We 25 included Davis Besne, THI-1, and Arkansas, in addition to HERITAGE REPORTING CORPORATION (202)628
_ _ ...__ _ ..- _ _,_._...~_ _ _ .. .~.-._ _. ._ _- _ ~..-..__ _-4888_ _ _ _ _ . _
146 1 Oconee, and so we have represented the raised 'oop 177, 2 modified 177, THI-1, Oconee, and original 177, and Arkansas, 3 classical 820, 177. I think the sensitivity study did a fair 4 job of looking at a spectrum of current design Babcock 177s.
5 MR. REED: You had design aspects in here, and you 6 did you certain comparisens, and you had people, operating 7 organizations aspects in here, and you did certain 8 comparisons.
9 Were those separated out clearly in your final l
10 conclusions?
11 HR. JONES: When you said we looked at people, you 12 mean from the operator burden perspective. That was covered
() 13 under the operator burden str and we will be discussing 14 that in more detail later, how those were separated.
15 We did look at a global operator performance aspect 16 as part of the sensitivity study where we looked at time to do 17 actions, et cetera, and whether it indicated anything with 18 respect to an operator, and that was also translated into part i
19 of the overall human factors assessment which will be 20 discussed later.
21 With respect to risk, we did attempt to come up with !
l 22 some quantification of the differences between the plant, but 23 that is quite difficult given that we did not have full PRAs l
{} 24 for the plant, but it does provide some additional insight, so 25 we did try to account for specific plant-to-plant design HERITAGE REPORTING CORPORATION -- (202)628-4888
147 1 differences as best we could, and the risk person will talk
(} 2 about the details of that as you deem appropriate, or would 3 like to pursue further, 4 That's it for me.
5 CHAIRMAN WYLIE: Okay.
6 MR. REED: Lunch.
7 CHAIRMAN WYLIE: We had one item before lunch. I s.
8 will leave it up to the Subcommittee. Do you want to pick up
. 9 anything, or want to go to lunch?
10 MR. WARD: Let's go to lunch.
11 CHAIRMAN WYLIE: Go to lunch. We will be back at, 12 in an hour.
13 MR. REED: One-thirty.
14 (Whereupon, at 12:35 p.m., the meeting was 15 adjourned, to reconvene at 1:30 p.m. .the same day.)
16 17 i
18 l
i 19 l 20 21 22 23 24 O 25 HERITAGE REPORTING CORPORATION -- (202)628-4888
- - - , . . . . . . , . . . . _ _ - - . _ . . ~_-.-___ . - , _ _ . . , . _ , _ _ - . _ _ , ,. . _ _. _ -__ _ - - _ . . _ . . . - _ -
148 1 AETERNOON 1EE11ON 1:35 p.m.
2 CHAIRMAN WYLIE: Let's resume the' meeting, and we
~
3 are going to pick up with item 4 on the agenda, disposition of ,
4 previous ACRS concerns from the July 16, 1986 letter, and I 5 call on Mr. Rutherford.
6 MR. RUTHERFORD: Okay. After our previous meeting 7 with the ACRS back in 1986, the Subcommittee wrote a letter to 8 Vic Stello, and in that letter they expressed three concerns.
9 First of all, they expressed a concern about the 10 effects of the operating organization on the system 11 performance, and we touched on this a little bit this morning 12 during the discussion.
() 13 Seccnd of all, they expressed concerns about how the 14 unique features of the B&W plants were going to be 15 characterized, and then noted the attention that we ought to !
16 be paying to decay heat removal.
l 1
17 (Slide) l
. 18 HR. RUTHERFORD: First of all, for the question 19 about management, certainly we agree, we agree that management l
20 plays a key role in the safe operation of any facility. I l 21 think it is, like we said this morning, it is something that, l
22 though, that was beyond the scope of our particular program as I I
I 23 far as plant performance. We have dealt with that issue at i a
l 24 the Executive Committee level in that each executive 25 recognizes the interdependence that we have and the effects l
l i
HERITAGE REPORTING CORPORATION -- (202)628-4888
-_ _ _ , - , . _ _ _ . _ . . _ _ . _ . . _ . _ _ . _ . . _ _ _ . . ~ . _ . _ . ~ , _ _ . _ _ _ _ _
149 1 that an event at one plant can have on all.of us in terms of 2 increased NRC scrutiny,oet cetera, and this has certainly-3 prompted us to initiate some of the oversight and audit 4 functions that we have installed in the program and that I 5 will be touching on a little bit later in the presentation.
i 6 Thirdly, in response to the letter that was written l
~
l 7 by the Subcommittee, Carol Denton responded to that by noting l l
8 that those type concerns should be handled'on a plant specific 9 basis, and we basically agree with th'at. As I mentioned 10 earlier this morning, the handling of the SHUD event and the l
11 Davis-Besse event I think were. appropriately handled by the l 12 staff in that they looked at those individual utilities. It 1
() 13 is not reslly a vender specific problea if you will.
14 Plant-specific or management type concerns can arise at any 15 particular plant in the country.
16 Here again, the bottom lino is that we do look at 17 each other. We do evaluate how we are implementir.g {
18 recommendations, how our plants are operating. The steps that 19 we can take when we see an outlier are somewhat, are somewhat 20 limited, although peer pressure I think has been very j 21 effective in spurring each of us on to better performance, and 22 I think we approach .;at issue more on the positive side of 23 the coin if you will versus the negative side, and that we
(} 24 offer encouragement to other members. We offer assistance in l
25 a lot of cases to other members in solving problems, and I :
I l
1 1
HERITAGE REPORTING CORPORATION -- (202)628-4888
150 1 think through our group efforts, we ought.to. elevate us all in l( ) 2 terms of our overall management capabilities of the plant.
3 MR. WARD: Can I comment on that? I'm not sure.
4 that, well, perhaps neither you nor Denton understood what we 5 were trying to say in the letter, but this whole issue arose 6 from something that operating experience was saying, that 7 there were too many transients and too many'of a certain type 8 of transient in some class of plants. That was called the B&W 9 plants.
10 However, when you looked at the, at the data, there 11 really seemed to be two different populations. There were the 12 three Oconee plants, and there-were all the others. And there 13 just, you know, looking simply at those data, there seemed to
)
14 be no more reason to single out lets say the three Oconee 15 units for some sort of special attention under a program like 16 this than there were any other three plants in, out of the 17 hundred in the country, and so we are raising the question of 18 whether, you know, the right. things were being looked at.
19 It is--is there something in the operating 20 procedures at Oconee? I mean, when you did, you try to l
21 look--we didn't want this to be, we weren't just making a.
22 motherhood statement about good management is wonderful. We 23 all know that. It is not particularly useful to-say that 24 again, but is there something very specific in the way those 25 three Oconee units are operated that made them a different HERITAGE REPORTING CORPORATICN -- (202)628-4888 .
- y v.e...-- 9 g.-r- m. m --.-w.,r e s. ,v-->.e -i ,.e-, ,---,-y--w.*e+,---e.--wc----y--g,-.-f,w,-- py4+ . -=,e *v-~89+***-te***-** *---=1*T==w***-7-e= "-e r '-vme-*~ev~- 7--=-
151 1 population? Did you happen to look at that at all?
() 2 MR, RUTHERFORD: We didn't really look at that 3 specifically. Obviously that's hard to separate out, what are 4 management related problems versus what are design 5 considerations that may play a role in complex transients.
6 And when you look at management, looking at our particular 7 experience and view, I don't think you can point to any one 8 thing and say there is the key, we can pass this on to the 9 other plants.
10 It is hundreds and thousands of little things that 11 you do I think in a plant that end up in having a good 12 performer. And I mentioned through our group action we do f~x 13 share a lot of things and a lot of those things are passed on,
\-) l 14 but we haven't necessarily gone out and systematically looked 1
15 to define well, what are the good attributes and procedures or i 1
1 16 preventive maintenance programs, et cetera?
17 A lot of those things do come out in our, in our l 18 committee activities and those kinds of efforts, but I can't 19 say that it is totally systematic. l l
20 MR. TAYLOR: I think there is one other point, too. l l
21 This addresses Dave Ward's question, and that is I think there l
22 was a recognition on the part of both the Executive Committee l
23 of the B&W owners, as well as on the part of the NRC, that I
24 there were a lot of other efforts in that area going on. They
( l 25 were evolving then and they are still evolving today in terms '
l l
HERITAGE REPORTING CORPORATION -- (202)628-4888 l
l 152 l
l 1 of things like INPO's corporate evaluation process, trying to 1
I r^3 2 find out management problems, the NRC SALP evaluations which C
3 do get into management problems, so I think there was a 4 recognition that this was an important area, big area, but 5 there were some other very significant things going on that 6 would make it inappropriate to try to make a structure of it.
7 MR. RUTHERFORD: Certainly that is a more systematic 8 process of trying to find good practices and passing those on 9 to other members in the industry. Okay.
10 (Slide) 11 MR. RUTHERFORD: Okay. Moving on to the next point 12 regarding the safety significar.ce of the unique B&W plant 13 characteristics, we spent a great deal of the agenda this ;
O 14 morning talking about that, but basically that was the major l
15 thrust of our sensitivity evaluation, is to try to offer some l
16 perspective on that whole issue, that oh, the B&W plants are 1
17 different, more responsive, Ferrares versus Fords or whatever, l
i 18 however you choose to characterize the differences, and after l
19 this presentation, we will be moving on to the sensitivity ,
l 20 study and we can, of course, address that in more detail.
21 Thirdly, the issue of decay heat removal--we, of 22 course, have already made presentation to the full ACRS on 23 decay heat removal in which we described the three lines of 24 defense being maintenance, main feedwater, auxiliary 25 feedwater, and feed and bleed.
HERITAGE REPORTING CORPORATION -- (202)628-4888
- . . ~ - ~ . - . .
153 1 Now we kind of made a leap here over into the feed-() 2 and bleed issue, and you have got to-keep in perspective here 3 that this is really the third line of-defense if you will.
4 Once you lose main feedwater, once you lose your 5 saf'ety-related source of emergency feedwater,.then you have 6 got the feed and bleed capability as back-up, and we are 7 certainly confident on the B&W plants that that is a very 8 viable back-up source of decay heat removal.
9 MR. DAVIS: It has been a while since I looked at 10 this bleed and feed procedure, but the last time that I had 11 occasion to look at it, there was some concern expressed that 12 by the time the operator decided that this. option was the only 13 one he had left, it would likely be too late for it to be 14 effective if it were implemented, and there was also some 15 concern expressed about the fact that he would, it'would be 16 difficult for him to balance feeding and bleeding and maintain 17 adequate core cooling.
18 Have those problems been resolved and does the staff 19 agree with thtt conclusion D up there?
20 MR. RUTHERFORD: I don't think we ever saw it as a 21 problem. There were certainly questions about feed and bleed 22 capability and being able to take credit for that after the 23 THI accident, and we, of course, have addressed that question 24 in the sensitivity study, but.this relates to the comment that 25 I made earlier about steam generator. Really.that's not HERITAGE REPORTING CORPORATION -- (202)628-4888
154 1 critical time point.
( 2 The critical time point is on out in time before the 3 operator has to take actions to restore emergency feedwater or 4 to initiate feed and bleed, and that time is--I forget the 5 exact time, and Stuart will get into that, but it is 6 comparable to the, to the other PWRs, not as long as some, but 7 it is significantly out there to allow the operator time to 8 initiate those type actions.
9 DR. KERR: Does your answer imply that you have not 10 heard of this problem Mr. Davis mentioned, or is that the 11 problem has been solved?
12 MR. RUTHERFORD: Here again, we didn't perceive a 13 problem. I think the staff had questions about feed and bleed l l
14 capability. Some of those questions were resolved via the !
15 l MIST facility, and I think we are on pretty firm ground, both -
16 us and the staff now, saying that feed and bleed is a viable 1 l
l 17 mechanism. l 18 Bob, do you want to address from the staff I 19 viewpoint? ;
I 20 MR. JONES: The issue of initiating feed and bleed, 21 it has been a generic issue in-house, being evaluated now, but 22 the operators group program did identify as one of the 23 recommendations looking at drastic action, things such as l l
24 initiating feed and bleed, ensure appropriate guidance is O 25 given. That is one of the recommendations in the program. We HERITAGE REPORTING CORPORATION -- (202)628-4888
155 1 have think that is a risk.significant recommendation, will
() 2 mention later about the risks affecting that.
3 Witn respect to controlling the feed ~and bleed 4 process, in the B&W plant, control of the. feed and bleed 5 process can be as simple as turn on the HPI and walk away.
6 You don't have to do anymore than that.
7 Ideally it is dene-Inr turning on the PORV and 8 running the HPI, just running it. On Davis-Besse_it is'a 9 little different because they have got a lowhead HPI pump, did 10 depressurize down, but that's--
11 MR. ETHERINGTON: Would you include the Crystal 12 River as feed and bleed?
(} 13 DR. KERR: The question was you would include the 14 Crystal River episode as feed and bleed?
15 MR. JONES: The Crystal River event kind of 16 exercised so to speak the mechanism, the opening of the PORV, 17 the actuating of the HPI water going out the valve. They 18 really didn't have to rely on it at that point in time. It 19 was only used for very short period of time anyway, as I C
20 reismber it.
21 HR. ETHERINGTON: There was feed and bleed--control 22 room, pumping--
23 HR. JONES: It is in the, a controlled operation.
24 You do not control on level. You do not control on any 25 parameter. About the only thing you may do is in the long HERITAGE REPORTING CORPORATION -- (202)628-4888
. .. - - - - - . = .
156
~
1 term. I don't have a timeframe, exact timeframe for you, but 2 certainly on'the order lof hours, you could get into.a, where 3 you would be throttling the HPI to.just maintain the 4 subcooling margin, at a reasonable level to depressurize, 5 continue to depressurize the plant, but that would be after 6 the initial actuation, getting the core returned to the 7 subcooled condition, then a long-term you' throttle your HPI on 8 your hundred degree subcooling line, throttle the HPI and you 9 are going to come down to pressure. Your outflow is there.
10 HR. ETHERINGTON: Thank you.
11 MR. RUTHERFORD: Okay. And of course, the bottom 12 line is, as we presented earlier today to the ACRS, is we feel
() 13 like we have abundant capability here as far as decay heat 14 removal, and that we don't see any need for any basic changes 15 in that capability, although we have a number of 16 recommendations in the SPIP program that relate to improving 17 the reliability of that process.
18 That's all I had on that. Was the staff going to 19 respond?
20 MR. SIEGEL: Yes. I don't really have a prepared 21 response because I didn't think we were going to discuss this 22 specific unit, but--item, but with regard to man'agement, I ,
i 23 would just like to elaborate on what Jim said.
[} 24 Basically we feel that what we do, first of all, 25 this was an owners group review. The individual plant l
HERITAGE REPORTING CORPORATION - .(202)628-4888
_ . . . . _ - ~ _ _ _ _
l 157 1 management--but we feel it is more than adequately addressed 2 by the INPO, but what the staff does on their SALP evaluation, j }
3 for instance, we have, every project directorate has a meeting-
-4 .with Merley every year, to brief him.on the status of the 5 plants. The reg' ion is involved in these_ conference calls.- We 6 sit down and one of the areas that is discussed.is management t
7 both from a corporate level and also from a plant operational 8 level.
9 And in addition, when we are identifying or 10 discussing which plants we consider a troubled plant, like.the l 11 staff is going to have a management meeting in June, one of 12 the areas that we are, we review is plant management and 13 corporate management, and specifically I think the management O 14 per se not only in this area but in every area we look.at-15 plants.that the management is good, it reflects in all the l 16 areas. If it is bad, it generally. reflects in all the areas, 17 too, so we think that that area is more than adequately 18 covered.
19 With regard to the other two items, the significance 20 of B&W units are more responsive, Bob Jones is going to 21 address this during his sensitivity evaluation, and the
)
22 ability to trip the reactor and remove decay heat, Bob has i I
23 already discussed that this morning, so we really don't have l 24 any further comments.
( 25 DR. KERR: Let me go back to the-first comment to l I
l HERITAGE REPORTING CORPORATION - (202)628-4888 l
. .~. - .. . . _ . . . . . . - _-.- _ _ - _ - _ .
158 1 see if I understand.
2 This investigation was initiated because of 3 operating events, perspective on the part of someone-that 4 .these operating events raised some questions about the safety 5 of the B&W plants.
6 Now it is my memory most of the' operating events did 7 not take place at the Oconee plants.
8 MR. SIEGEL: That is correct, of the category C .
f 9 type, yes.
10 DR. KERR: Hence it seems to me one might have 11 wanted to look at what was unique about the Oconee plants.
12 Clearly it is not that they have B&W reactors because the ,
l 13 other three have, B&W 4 whatever, have B&W reactors and yet !
i O 14 what one is looking at is B&W reactors as if somehow the fact ;
15 that Oconee 1, 2 and 3 exist has been forgotten in looking at i
16 those transients, and it was for this reason that we raised 1
17 the question. Maybe it isn't something about the reactor.
18 Maybe it is something else. Maybe the reactors are okay.
19 MR. JONES: If I could add one comment-on that, Neil 20 may remember it, but my memory says that there was an 21 ACRS--not ACRS--a Commission briefing in November probably '06 22 where we looked at the operating history of the plants, and my -i l
23 memory says while indeed Duke didn't have any of the' category l i
24 l C events, they did have a lot of the category Bs, and 1 1
(f 25 significant Bs. I mean they were not, they are really not HERITAGE REPORTING CORPORATION -- (202)628-4888
159 1 that far out of line on an overall operating experience except
-) 2 they haven't seen any Cs. They do have or have experienced 3 some significant Bs, and not an outstandingly low level of 4 them in comparison to the others.
5 MR. ETHERINGTON: Those include instruments, 6 instrumentation on the bottom of the reactor?
l 7 MR. JONES: No, we are not including those. These l l
8 are the events from the 1980 to '85, '86 timeframe is what it 9 is based on, and they did have seueral significant category Bs 10 in some of the Oconee plants which you know, tended to 11 indicate that really while there is some diversity in the C 12 transients, there is clearly still some overall design issues
() 13 that affect across the entire B&W product line.
14 DR. KERR: But they weren't the thing that called 15 the problem to your attention, were they? It was those severe 16 transients?
17 MR. JONES: Yes. It was the Cs that essentially 18 have raised, what started this, but some of those Bs don't 19 miss by a heck of a lot.
20 DR. KERR: And that's, we are getting to--these 21 questions appear to indicate that you don't understand what it 22 is we are talking about, and what I was talking about at least 23 was how one first identified the problem I thought was on the
{} 24 basis of severe transients, and again, I thought that the 25 severe transients didn't happen at these three plants, and yet HERITAGE REPORTING CORPORATION -- (202)628-4888
160 i
1 what has been focused on is the whole population of plants.
() 2 Now I will say no more, but--and I'say this only 3 because it appears to me that people to whom we.are addressing 4 this question don't understand our question or the basis for.
5 it.
6 HR. JONES: Clearly we interpreted your, back in '86 7 when we wrote the response, that answer was really~looking at 8 did we need to consider a management across the board?- We 9 weren't looking at it as just narrowly the way I think you are !
10 now talking about--is there something in Duke particularly 11 beneficial that they can learn fron? And really we judged it 12 that we have, we were looking at it from correcting bad i
13 performers as opposed to looking for the good and finding it i (I) 14 back, and in that way, what we already were looking at.was the t
15 fact that we have existing mechanisms in place that'we thought 16 would do an adequate job.
17 DR. KERR: My point was--I absolutely will say no i
4 '
18 more about it--it is a hypothesis, given these severe 19 transients, one.might have concluded hey, there is not 2 20 anything wrong with these plants. It is the'way they are.
21 being operated that is the problem. Now you have looked at 22 it, apparently concluded that's not the case, but if one 23 simply took the evidence of severe transients, at least 24 hypothetically, that might have been a conclusion that one i
1
(:) 25 would have reached, l I
HERITAGE REPORTING CORPORATION -- (202)628-4888
161 1 MR. SIEGEL: I think directly getting at the problem 2 by, through the recommendations, even though perhaps a lot of 3 these stem from the fact that management perhaps'didn't, in 4 some plants didn't identify some of the problems they had, 5 through the recommendations, these, some of these problems are 6 being corrected.
7 The management per se is a more broader-based one 8 and it is more long-term one that the staff is continually 9 looking at, and I'm not sure, I think we are looking at that 10 area, and when we do see areas that weak management, we have 11 pointed out to the utilities like I mentioned before either 12 through the SALP or the review process that the staff goes
() 13 through on a yearly basis to evaluate plants, but I think you 14 are probably right. Some of them were due to the fact that L
15 perhaps there was a lack of management attention in some areas 16 that there should have been. I think those responses were 17 corrected through some of the recommendations. The owners 18 group might want to comment on that.
19 DR. KERR: Serendipity is going to save us?
20 MR. SIEGEL: I think we are--at management where 21 there are weaknesses, we have identified them on numerous 22 plants by the mechanisms I mentioned.
23 MR. WARD: Can I pursue this a minute? We didn't
{} 24 say anything about management in our-comment, although we did 25 say something about operating organization which may have HERITAGE REPORTING CORPORATION -- (202)628-4888
162 1 distorted it a little bit, but you seem to be saying that the, I'
\ -) 2 you know, it is the balance of plant features that tend to be 3 driving some of these, everybody seems to be saying that some 4 of these incidents. Clearly the three Oconee units have the 5 same balance of plant design.
6 Are you sure that's not the key? Are there some key 7 difference between the balance of-plant at Oconee and the i i
1 8 others? I mean I am just, we are just stuck with this sort of )
9 simple minded observation that there seem to be two 10 populations, and instead of jumping to the conclusion that you 11 already know of the reason for a problem, why not try look at 12 it more objectively? Maybe you have done that, but tell us if
() 13 you have.
14 MR. JONES: I think that's why we have taken the 15 approach, first off, while the two transients, Davis-Besse and 16 Rancho Seco, had a big impetus, a push to go forward, we have 17 had lots of, nagging problems with these plants. Those were 18 kind of the straws that broke the camel's back. If you go and 19 look at the abnormal transients, and you can quibble how many 20 of these were abnormal and normal, the owners group put them 21 up as normal, at one time they were discussed as abnormal, I 22 don't know how to exactly characterize them, but there are at 23 least a subclass of the Bs which are quote, significant 24 category Bs, and as I said, those have shown up across the
{
25 board; they do not, Duke does not stand out as having none of HERITAGE REPORTING CORPORATION -- (202)628-4888
. =. .. .- . . - .- _. .. ., .- . -. .-
l 163 1 those. They do have a fair number of those, so I think it is 2 to a large extent a design, you.are seeing the design effects 3 across the board, yet there are differences in balance of
- 4. plant which would be difficult to fix within the existing i
5 frameworks of the system.
6 I mean Duke doesn't have isolation valves, MSIVs. ,
7 Other plants do. That is going to have certain impacts on 8 some transients which are good, and bad on other transients, 9 so there is a balance across the boar, d and we had to deal 10 with each system and look at them almost collectively to.see ;
11 whether it was experience that applied across the board,'and i
12 that's how the recommendation came out, but I don't think ;
13 there is a simple-minded Duke has everything right because !
O 14 they have experienced some significant transients. :
15 MR. RUTHERFORD: Okay. Let me respond a little bit i 16 to that. Now I don't basically agree, disagree with what Bob i
17 has said. There are a number of reasons I think when you look l 18 at Duke that are experiences different from the other 19 utilities in some respects.
I 20 First of all, it is true that we have had comparable 21 frequency of category B type transients, but I think if you 22 look at the root cause of those transients, they.were fairly l 23 self-limiting versus some of the category C events that we got I i
24 into that were not self-limiting.
25 And I think the differences there are attributable j l
HERITAGE REPORTING CORPORATION -- (202)628-4888'
164 1 to a number of factors, management being one of them, and if I s
2 had to put a finger on one particular item, it would be 3 finding the root cause of events and learning from industry 4 experience early in the game and making appropriate changes.
5 I think perhaps we have done a better job of that, 6 blowing our own horn, than perhaps some other plants, and we 7 concentrated on that aspect of things in the SPIP program to 8 make sure that we all have good root cause programs and that 9 we do indeed thoroughly analyze events and go in and take 10 appropriate corrective actions.
11 Secondly, there are design differences between 12 Oconee and other plants that do influence the fact that we
() 13 have avoided some category C events, and those relate to the 14 complexity of some of the control systems that are utilized in 15 the plants, but I think that's a, while some degree of 16 contributor, not the overriding contributor to the differences 17 that you see.
18 MR. WARD: There is a non-complexity in the Oconee 19 design? Is that what you are saying?
20 MR. RUTHERFORD: I think in some case we have got 21 somewhat simpler control systems. 1 22 MR. WARD: But you haven't been able to zero in on 23 that as--
/~ 24 HR. RUTHERFORD: You look at Davis-Besse and the k_') l 25 contribution that the SFRCS system made to that event, we HERITAGE REPORTING CORPORATION -- (202)628-4888
165 1 don't have that complex control system at Oconee. Of course,
) 2 control systems can cut both ways. They can help you or they 3 can hurt you, depending on failure modes, et cetera, so in 4 that respect, I think that Oconee design is a bit more 5 forgiving, but overall, I think that it is probably a minor 6 part of this overall puzzle.
7 Getting back to the main thesis here, I think we 8 have picked out some things that do relate to quote, 9 management of the plant, that we have attempted to pass on to 10 all the owners and ensure that we are all doing a good job in 11 this area of reacting, identifying problems and reacting to 12 those in a timely manner.
(} 13 Any other questions?
14 CHAIRMAN WYLIE: Any other questions on this? Let's 15 move on.
16 HR. RUTHERFORD: If not, I will turn it over to 17 Stuart Rose, and he is going to go into the sensitivity study.
18 MR. ROSE: As Neil said, with no small amount of 19 fanfare, we get to the sensitivity study this afternoon.
20 I would like to accomplish two things this 21 afternoon. I would like to review with the Subcommittee the 22 purpose, scope, and methods of the sensitivity study, the work 23 that was done, and take the opportunity to remind you that 24 this material was discussed with you on July of 1986, at which 25 time we went through the analysis matrix in some amount of HERITAGE REPORTING CORPORATION -- (202)628-4888 ,
166 1 detail, and talked about'what we would be looking at and what 2 we' would not be looking at under the umbrella of the 3 sensitivity' study.
4 The other things I want to cover are the results, (
5 the conclusions, the recommendation, results from the study, ;
6 and based upon the work that was performed.
7 (Slide)
) 8 MR. ROSE: As you know, the owners group employed ,
9 NPR Associates to assess the differences in thermalhydraulic 10 design arising from differences in design characteristics 11 between the three PWR venders.
12 These analyses were based upon the issues that were I 13 on the table at that time, were bounded by the scope of the -
0 14 study that we talked about about a year and a half ago. Our i
[
15 purpose was to try and define what sensitivity was, and then i a
16 to quantify thought differences in sensitivity.
17 NPR performed a comparison of the typical B&W units 18 against representative Combustion and Westinghouse reactors, t
19 and you can see the list of units in the B&W category that 20 were analyzed--older vintage Combustion plant, a more recent 21 vintage and a recent Westinghouse plant with incorporating the 22 pre-heater, feedwater pre-heating section in the steam 23 generator.
1 24 Again, I would like to emphasize that the purpose of 25 the study was to quantify relative differences in HERITAGE REPORTING CORPORATION -- (202)628-4888
167 1 thermalhydraulic behavior within the scope of issues that were
( ,/ 2 before us at the time that the study was put together.
3 These results were submitted to the NRC on April the 4 4th of last year, 1987.
5 (Slide) 6 MR. ROSE: In terms of scope, the sensitivity study 7 included thermodynamic response of the reactor and steam 8 system during normal plant operation. These were 9 characterized as bare plant analyses in that they were run 10 without control system and protection systems in place. They 11 were run using step changes or impulses in reactivity change, 12 feedwater flow, steam pressure, steam flow, and an item to
{} 13 understand the inherent behavior of these units that is 14 based--there was a function of the design differences. The 15 study quantified these differences in the control of key plant 16 variables.
17 Secondly, the study analyzed the plant thermodynamic 18 response for anticipated operating occurrence, things such as 19 reactor trip, turbine trip, runback, loss of one feedwater 20 pump, those types of events.
21 MR. CATTON: What did you use to do the analysis?
22 HR. ROSE: All right. We used, NPR employed some 23 simplified dynamic codes that were simplified by NPR, 24 simplified in the sense they were not the RETRAN, RELAP type 25 codes, but simplified in terms of simpler model of each of the f
HERITAGE REPORTING CORPORATION -- (202)628-4888
168 1 primary steam generator and another components that were then 2 validated against actual plant data,.or against other existing 3 validated computer runs like RETRAN and RELAP and others.
4 HR. CATTON: RELAP doesn't have the differences 5 between the plants well'modeled, and that's why we have MIST.
6 What did you do about that? l 7 MR. ROSE: Well, as I said,.part of the validation J
8 of the NPR model was used on actual plant data. There.was--
l 9 MR. CATTON: That doesn't get you into some of the 10 more narrow aspect of the transients you are looking at.
11 MR. ROSE: I think in some cases it does because the 12 transients that they used for validation was a, a turbine trip l l
() 13 and THI back in 1974, with a artifically suppressed. reactor 14 trip function; the 1985--
15 MR. CATTON: What were the transients? You didn't 16 list them. Do you have a list of the design differences?
17 MR. ROSE: The desigr. differences, no, I don't. I
- 18 don't have a list of transients. They I think included
. 19 turbine trip at THI, 1985 Davis-Besse event, a turbine trip 20 and reactor trip on Westinghouse unit, and also a transient on 21 a Combustion unit.
22 HR. CATTON: And what were the thermalhydraulic
, 23 design differences that you were trying to consider?
l [} 24 MR. ROSE: We were looking-for differences--I have a 25 slide that will get to that in just a second--looking at key l
c HERITAGE REPORTING CORPORATION -- (202)628-4888
169 1 plant variables.
/~'S 2 MR. CATTON: Okay, b
3 MR. ROSE: Let's see. I believe in the third 4 bullet, the study also looked at thermodynamic response 5 comparison for selecting transients and accidents, things that 6 are, some of the things that are analyzed in the FSAR such as 7 rod withdrawal, and loss of feedwater events. Some of those 8 analyses were based upon a comparison of the analyses 9 presented in the FSARs. That is, they were not detailed 10 computer codes. Computer analyses run within the scope of 11 this sensitivity study--
12 MR. CATTON: You are really staying pretty close to 13 normal operation, not getting way far away.
14 MR. ROSE: I don't think that is true. i l
15 MR. CATTON: Are you getting into circumstances l l
16 where you have two phase in the primary system? j 17 MR. ROSE: No, no two phase, but it is getting away l
18 from normal operation in the sense other failures, steam I 19 relief f ailt re, that is the sort of thing. I i
20 MR. CAT 70N: Suto. 1 l
21 MR. ROSE: The study looked at the protection and l i
22 control systems, and assessed the design differences affecting j 23 the complexity of hardware, looking at the number of modules l 24 and components that P. ave to work properly to achieve the l l
5 25 desired control function, comparing those counts with the B&W !
- HERITAGE REPORTING CORPORATION -- (202)628-4888
170 1 integrated control system, the Westinghouse and.the Combustion
() 2 design, and did some assessment of operatinc experience 3 actuations and off normal occurrences to see if that, that 4 relative level of complexity was reflected in actual-5 experience.
6 And finally, NPR performed assessment of operator 7 responsibility, the things that the operator must accomplish
, 8 following a reactor trip in order to properly control the, 9 control the event, and they looked at both normal and off 10 normal conditions in that assessment. They were looking at :
11 things like properly controlling makeup and let down, and 12 steam generator level as actions that the operator has to, to 13 assure are carried out.
l 14 (Slide) j 15 HR. ROSE: This is an indication of how we defined f 16 sensitivity and how we attempted to quantify it. We are !
! 17 looking at sensitivity in terms of margin time and frequency, 18 margin with respect to how close the plant response, the 19 thermalhydraulic response approaches a limit, a safety limit 20 record, more particularly, an operational constraint of some
! 21 sort such as reactor protection system setting, or flooding of !
I i 22 the aspirator ports on once through steam generator. ;
\
23 We also looked at the response in terms of time.
l 4
24 How quickly does that transient proceed? Are the rates, the l
- 25 slope of the curve different between the three types, and also .
HERITAGE REPORTING CORPORATION -- (202)628-4888
. .-- .. ---. - ~~ ..
171 1 in terms of frequency, are there aspects of the plant design 2 that would cause a particular transient to be experienced more 3 frequently than on the other design?
4 Some of the safety parameters that the study looked 5 at were Escondary design pressure, primary design pressure and-6 temperature, saturation margin, minimum, kilowatts per foot.
7 limit, PTS limits. More particularly the study focused on 8 these, what we will call operational limits and criteria, as
~
- 9 being more pertinent to what the operator sees in controlling 10 the plant, and these are the quantification of the sensitivity 11 that the study looked at.
i
. 12 And analyzing the plant response, NPR looked at how 13 quickly the steam generator would dry out. For example, it 14 looked at how quickly the aspirator ports, the level would -
15 till up above the aspirator port on a once through, or to the f 16 moisture separator on a circulating generator, how quickly 17 pressurizer level indication might be lost. This is the 18 method with which sensitivi ty was quantified. :
1 19 (Slide) !
J l
l 20 MR. ROSE: That then is the basis of the sensitivity
+
21 study, what was looked at, and by omission what was not looked 22 at. Now I want to get into a review or a presentation of the
-i 23 conclusions and results from the sensitivity study.
24 First of all, a couple of general conclusions that
( 25 were reached as a result of this study show that in several i I l
HERITAGE REPORTING CORPORATION -- (202)628-4888
172 1 aspects of the B&W design, which had been perceived as causing
() 2 greater sensitivity, in fact did not. It was a conclusion of 3 the study that the pressurizers on the B&W units are as large 4 as or larger than those on other units.
5 MR. CATTON: Why is it that you have a loss of level 6 on the turbine trip and the others do not?
7 MR. ROSE: We don't have loss of level on a normal 8 turbine trip. If the rest of the plant operates within the, 9 within its limits, we don't.
10 MR. CATTON: What happened at Three Mile Island?
11 MR. ROSE: When the--
12 MR. CATTON: The accident; didn't they lose sight of 13 the level following the turbine trip? My recollection is that 14 that was a common occurrence on B&W plants, that they, i
15 following the turbine trip, you get the shrink and then they i 1
16 have to jump in and maybe--
l 17 MR. SKILLMAN: The trip, the level remains in the l 1
l 18 pressurizer. On an extreme cooldown, like when the emercency i
19 feedwater system and the secondary plant system get out of l l
l 1 20 kilter, you can experience such extre. cooling you have a l l
21 super shrink and you go below the lower TAP, but that is not a l I
22 common occurrence, the lower level TAP on the pressurizer. At l I I 23 TMI, what happened is, was steam formation, transfer of level l l
24 from the reactor vessel to the pressurizer, and with the stuck l 25 open PORV, water and pressurizer exhausted to the effluent !
l l
l HERIT.'.GE REPORTING CORPORATION -- (202)628-4888 !
173 1 condensing tank, leaving empty core with a concurrent
( 2 pressurizer level--completely different scenario.
3 MR. CATTON: I didn't want to get that far into the 4 scenario. I am referring to right at the front end. It was 5 my understanding that it was common relative to the others.
6 MR. SKILLMAN: No, sir.
7 MR. CATTON: Lose sight of the level in the 8 pressurizer.
10 MR. CATTON: It says that in the lessons learned.
11 Is that document incorrect? nt least I think it does.
12 MR. ROSE: I don't recall where it says that, that
(} 13 there have been a couple of changes in the history, spend a i
14 minute or two talking about that. j 15 Prior to the TMI accident, on the B&W design, the 16 PORV lift set point, 50 pounds I believe the high pressure I 17 trip set point, as currently on a number of Westinghouse i
18 designs, so that any time the reactor was experiencing a high j 19 pressure transient, the PORV would open. Not s function of l
20 the pressurizer being so tightly compressed; it is where the
)
1 21 set point was established.
22 Secondly, once the reactor tripped, again with 23 nominal steam pressure control and steam generator inventory 24 control, pressurizer level would remain on scale.
25 One or two of the B&W plants with a compressed HERITAGE REPORTING CORPORATION -- (202)628-4888
174 1 indication range experienced a loss of level, but they 2 corrected that. These are the two B&W pressurizers, one of f-}
V 3 which we call the--
4 HR. RUTHERFORD: Stuart, let me offer some 5 information here on, to document the experience on pressurizer 6 levels.
7 Since January 1, 1980, we have had no off-site low 8 pressurizer, off scale low pressurizer indications at Crystal 9 River, Davis-Besse, Oconee 1, 2 and 3, and THI, and ANO. No 10 one, has had no off scale low events since they raised their 11 pressurizer level, their normal operating pressurizer level in 12 mid-1983, and Rancho Seco had one off scale low event in 1985, t
13 the overcooling transient.
14 MR. ROSE: Keep in mind again the difference in 15 level spans the full range. Wide range indication is 400 16 inches on some of the B&W units. The same pressurizer has 320 17 inch span. I believe Arkansas is in that category, and they 18 took the step of raising their nominal operating level to give 19 themselves a little more cushion. As Neil said, since 1980, 20 there has been the one at Rancho Seco.
21 MR. WARD: Could you adjust that or maybe read us 22 off those volume numbers across the--
23 HR. ROSE: 1520 cubic feet for the two B&W designs, 24 1517 for this Combustion unit, 1220 for the second unit, 1435
( 25 for this Westinghouse unit, and I believe that is 1835 for HERITAGE REPORTING CORPORATION -- (202)620-4888
175 1 the--
() 2 HR. SKILLMAN: Let's read the cubic feet per 3 megawatt, which is a very important index in terms of--
4 HR. WARD: Why did it take--these are just facts.
5 You say the study revealed this. I guess I am not so much 6 questioning the--but l' mean I have been coming to meetings f or 7 years and people have been telling me that the B&W 8 pressurizers are smaller than the other FWR pressurizers. I 9 guess I was too dumb to know where to look for facts, but why 10 has this been such a mystery all these years?
11 MR. TAYLOR: I think part of it goes back to the 12 perception that Ivan talked about. They go back prior to THI, 13 losing the pressurizer level was a frequent thing. It 14 happened more often than it should have, and that's when 15 people perceived boy, you know, the pressurizer is small, and 16 really what was happening was that they were talking about the l
17 visible range of the pressurizer.
l 18 MR. CATTON: We were told it was too small.
19 MR. TAYLOR: That is what people were saying. It 20 wasn't until this study when they made a close examination of l l
21 some plants that were supposedly comparable and look, not only j i
22 is pressurizer level not going off scale low like it used to j l
23 seven or eight years ago or prior to seven or eight years l l
1 24 ago--here are the actual volumes. I think it is partly 7_
D 25 perception.
l HERITAGE REPORTING CORPORATION -- (202)628-4888 I l
176 1 MR. DAVIS: It is an important relationship, the 2 sort of pressurtzer relative to the volume of the primary x
3 system, because you can have shrinkage and expansion during 4 these transients.
5 How does B&W compare on that base because I don't 6 see that number up there?
7 HR. ROSE: That was not the method that NPR used in 8 assessing pressurizer volume. ' hey looked at volume over 9 megawatts.
10 MR. SKILLHAN: I think they also looked at volume 11 over shrink volume. That wou3d take into consideration the 12 changing density for a larger reactor cooling system volume.
13 On that base, the Babcock pressurizers compare about O 14 equivalent. There are different things among them, but in 15 general, the allowable, if you will the volume change is a 16 function of shrink mass equated to volume in the pressurizer, 17 compares about the same across the line of pressurizers for 18 Babcock, Combustion and Westinghouse.
19 HR. REED: Let me add a little bit to what Mr. Davis 20 just said. In my opinion, pressurizer sizing, it is a, size 21 should relate somewhat to not only the pritnary cooling system 22 volume, but the post-coupled secondary side inventory steam 23 generator volume. That's close coupled. You can't shut it.
24 It is going to act fast in controlling shrink or controlling ;
( 25 volume expansion, so they hr.ve not addressed this in the l
HERITAGE REPORTING CORPORATION -- (202)628-4888 l
177 -
1 documents I have seen here, that the close coupled, secondary
() 2 volume is a factor, and qttite frankly, transients are always 3 interesting, the kind of transients you can get, and in fact 4 pressurizer sizing is very complicated, pressurizer sizing, 5 but I would not ignore the total couple inventory in sizing, 6 and therefore, I always come out with even those though might 7 look big volume-wise, they are not. They have got a much 8 smaller coupled inventory to temper the change.
9 MR. ROSE: I believe that NPR did a sensitivity 4 10 study, touched on some of the issues that you just raised.
11 Looking at the time, the response of transients; especially 12 high pressure transient was looked at as part of the study.
13 MR. REED: I saw the pressurizer ratio thing, and I 14 didn't see that in there. l i
l 35 MR. ETHERINGTON: I think what we are leading into 1 1
16 is a little bit superficial. If you take one BTU out of cubic 17 foot and get about the same shrinkage in, you take one BTU out 18 of two off the--should be somewhat independent of the volume.
19 MR. CATTON: So the steam generators wetted area and 20 volume would be an important--
21 MR. WARD: He means the primary system volume.
22 MR. CATTON: That was what I am referring to. You 23 need to know the heat removal rate if you are going to have to 24 calculate the shrinkage for a given volume.
25 MR. WARD: Yes.
1 HERITAGE REPORTING 00RPORATION -- (202)628-4888
! i 178 1_ MR.-CATTON: Once you-know that, you take some kind .
2 of ratio to the pressurizer side and see if you can handle it. !
3 I don't see that anywhere, either.- It seems to me it is i
! 4 relatively simple concept. I am saying the same thing-you 5 are.
3 F
! 6 .MR. DAVIS: Since 1980 you have had 270 some I t t I L i 7 transients, and perhaps only one case did the level in the 8 pressurizer go off scale, and I presume in none of them did ,
9 you overfill the pressurizer.
10 MR. ROSE: I believe that is correct.
11 MR. DAVIS: To me that says something about the i
12 adequacy of the volume of the pressurizer, something positive. ;
h 13 MR. ROSE: I think it does. Okay.
14 (Slide) l 15 MR. ROSE: Second general conclusion, B&W units are 16 less sensitive to steam demand upsets than are the other l
17 designs. That has to do with a,'the variable heat transfer 18 area and the response to a step change in steam pressure and l 19 steam flow.
l l
- 2) Thirdly, B&W units do not impose greater burdens on I
l l 21 the ortrators following most--that is normal reactor trips, i
22 These positive conclusions--the study's recommendations for
! 23 improvement, we want to make sure that they do clear the air 24 on a couple of different points. .l 25 MR. DAVIS: One of the conclusions in your report on l
l \
t
! l l HERITAGE REPORTING CORPORATION -- (202)628-4888 l E .-
179 1 the basis of the NPR study is that the steam generator volumes s 2 are not small in terms of the time available to initiate
[v 3 feedwater before dryout, and this also goes against what at 4 least I perceived as a difference in the design of the B&W 5 plant.
6 Can you elaborate a little bit on that?
7 MR. ROSE: Yes, I can.
8 MR. WARD: Are you going to get to that?
9 MR. ROSE: I am coming to that.
10 MR. REED: Just before, on the slide you just had 11 up, I am still having trouble with the pressurizer size as 12 large as the other recirculating steam generator, PWRs, 13 because you don't have this coupled large secondary side 14 inventory to help cushion the expansion and contraction and 15 these fast transients that trip and up and down.
16 Now I think the proof of the pudding would be to go 17 to the record there of these plants and ask B&W, Westinghouses 18 and CEs what the fluctuation in level in most of your normal, 19 in your normal transient; what is the fluctuation level? I 20 think you will find if you look at that fluctuation level that 21 B&Ws are, have substantially more, which says something about 22 pressurizer sizing.
23 MR. WARD: If it doesn't fall out either end, what 24 is the difference?
() 25 MR. REED: Fine. I am not bothered by this problem i
HERITAGE REPORTING CORPORATION -- (202)628-4888
180 1 whether the pressurizer is too big or too small. I am 2 bothered about bigger problems.
3 MR. ROSE: I will do specific conclusions and we can i
4 get to the other question then--relative to other PWRs, the l I
5 sensitivity study concluded that B&W units are not more 6 sensitive to reactivity upsets, concluded that B&W units are 1
7 not more sensitive to reactor coolant flow upsets. Response I 8 to coolant flow upsets are more a function of core designs 9 than differences between the three PWR venders. The study 10 concluded that the B&W units are less likely on average to 11 experience a leak leading to a net loss of coolant. This is 12 tied back to the high-head makeup capacity at most of the B&W 13 units, and the ability to compensate, in large compensate for A
(_) 14 a larger leak size at operating pressure.
15 The study concluded that the B&W units are somewhat 16 less sensitive to steam demand upsets such as low rejections 17 and turbine trips. The reactor trip on turbine trip is not 18 required to ensure plant safety in NPR's opinion.
19 NPR also concluded that the units are not more 20 likely to overcool failure in a reactor trip. This is an 21 outshoot of the the steam pressure control reliability and the 22 observed reliability of code safeties versus atmospheric dump 23 valve and turbine bypass valves.
24 They concluded that the B&W units are more sensitive
() 25 in their response to main feedwater upsets, although the HERITAGE REPORTING CORPORATION -- (202)628-4888
i 181 1 frequency of such upsets is not as great as in other PWRs.
2 HR. WARD: You said not as great. It says not 3 greater than.
4 HR. ROSE: Is not greater than other PWRs--excuse 5 me.
6 Again, due to the variable heat transfer areas, and 7 the smaller initial inventory--
8 HR. REED: I am having trouble with all of these, 9 but suffice it to say that if Westinghouse is doing the study 10 on their reactor, they would come in with identification of 11 things about their reactor as you subjectively have done it 12 about your reactor.
(} 13 HR. ROSE: I think there is a great deal of 14 objectivity to this.
15 HR. SKILLMAN: I take real strong exception to that.
16 That was not a subjective bunch of conclusions. These are 17 uenclusions that you can read for yourself in the appendices 18 of B&W 1919. The models are there. Differential equations 19 are there. These models have been tested against actual plant 20 responte. This information has been given to the Combustion 21 Engineering owncrs group, Transient Assessment Committee, and 22 they in many cases are saying yes, that's true. Now that we 23 see that presented the way it is being presented, we agree 24 that your plant does that and ours doesn't.
25 So this is not just a B&W owners group and NPR
- i HERITAGE REPORTING CORPORATION -- ( 20P,) 62 8-4 8 8 8
l l 182 1 making unfair conclusions. These are objective conclusions
(} 2 based on hard data which you have access to in your copy of
! 3 the B&W 1919.
i i
4 MR. REED: I stand by my statement, and let's take a 5 bullet. Are not more likely to overcool following a reactor l
> I 6 trip--and yet we have got all these overcooling events that l l l 7 were there on the chart.
l 8 HR. ROSE: This statement is based on a nominal--
l i
9 MR. REED: Twenty. '
l l 10 MR. ROSE: Response of the steam pressure control l
l 11 system; what you see on the chart are failures of that system.
l 12 MR. SKILLMAN: The chart shows the complex i
13 transients, not the responses. As I pointed out on that l ( I l 14 chart, I was trying to break up the dif f erence between trip l
15 and complex transient response. Whet we showed you quite l
16 frankly is our dirty laundry and what makes the laundry dirty, l
i l
17 and the bulk of those are overcooling transients. In fact, we 18 have had 50 significant transients out of 250 trips, and if 19 you look at the 50 significant transients, about 90 percent of l 20 those are overcooling transients, but what I didn't talk about l
l 21 are the 200 trips that are normal in t;hich case there was no
( 22 overcooling. I talked about the complex transients. '
l
( 23 MR. REED: Well, to try to rebut again, you are, l l I 24 well, what we will call throwing off certain information and .
( 25 putting on best case listings which are not let's say too, i l
I l
l HERITAGE REPORTING CORPORATION -- (202)628-4888
1 183 1 well, you can call them normal, but they have lots of cases l
[%-) 2 that aren't the normal. And quite frankly, I don't believe !
3 Westinghouse or CE are going to rebut to your list because why 4 stir up a hornet's nest?
5 HR. WARD: I guess I didn't quite understand Mr.
6 Skillman's response. If I look at the fifth bullet there, if 7 these are only the, if you are only talking about situations 8 where the, all the systems respond normally, then there is no 9 likelihood at all for it to overcool, ight? I mean if 10 everything works, it is never, it is not-going to overcool? I 11 mean what we are talking about here is the tendency to 12 overcool given something going wrong.
() 13 MR. ROSE: The specific point of information that 14 NPR reached this on could include, had to do with the observed 15 failure in code safeties versus observed failures in turbine 16 bypass valves, atmospheric dump valves. The rate of failure 17 to reseat code safety valves is significantly below the 18 cthers, and the overcooling transients that have been seen in 19 the past by and large have not involved the code safety 20 valves, but have involved the interaction between the ICS and l
21 the turbine bypass valves and high emergency feedwater flow j 22 rate. Those two things coupled together have led to those 23 overcooling transients.
{} 24 HR. REED: Atmospheric dump valves are sized to pass 25 a max of about 10 or 15 percent of the heat flow rate, and I
HERITAGE REPORTING CORPORATION -- (202)628-4888
l 184 1 main steam safety valves are probably going to pass 20, 40 2 percent of the heat rate.
3 MR. ROSE: Not individually; at 5 to 20 valves per.
4 MR. REED: Atmosphere dump valves, some are only 5 sized for 5 percent, 10 percent, and you are talking about 6 opening a big valve versus a little valve, and no one has put 7 a lot of tension on atmosphere dumps on the plants because 8 they don't involve a big safety problem.
9 Again, I have got to disagree with you.
10 MR. WARD: If I understand what you just said, the 11 fifth bullet really means are not more likely to overcool 12 following a reactor trip which the code safeties don't reseat?
13 Is that what you held I mean therein?
O 14 ,
MR. ROSE: Let me see if I can put it a different 15 way. The B&W units' use of the code safeties for steam 16 pressure control post-trip is not more likely to cause an 17 overcooling transient than on other units. The code safeties 18 are reliable.
19 MR. ETHERINGTON: What temperature constitutes 20 overcool?
21 MR. ROSE: What temperature constitutes overcooling?
22 In our definition, below 500 degrees was a significant 23 overcooling.
24 MR. CATTON: Are there similar qualifiers on all the 25 other bullets?
HERITAGE REPORTING CORPORATION -- (202)628-4888
- -. . . - ~ . . . . _.. .. - -
185-1 1 MR. ROSE: Certainly the~ details behind all of these i
~ () 2 bullets are laid out in 1919. This is just a summation.
3 MR. CATTON: I understand, but that fifth bullet, it. ,
4 4 turns out that the qualifier was rather-significant.
5 HR. DAVIS: If you are going to say something about
~
6 the feedwater control, the statement might say that they are 7 more likely to experience overcooling-from malfunctions of i
j 8 feedwater.
i 9 HR. ROSE: Next slide.
10 MR. DAVIS: Also No. 3 is a *.ittle bit confusing to 11 me. Are you talking about leaks in steam generator tubes and 12 safety valves? I certainly don't think there is much
(} 13 difference in the loss of coolant accident frequency among the 14 plants?
l 15 MR. ROSE: We are talking about primary coolant )
16 system leaks.
17 MR. DAVIS: Including pipe ruptures or just--
18 MR. ROSE: No; below that level. On this side of -l ,
i 19 the chart is makeup capacity, 100 to 800 gallons per minute.
20 On this side is the leak size in square inches, and what this 21 data shows is for 2200 psi operating pressure, two makeup 4
22 pumps in operation in B&W units span this range of' makeup l 4
1 23 capacity which corresponds to this range of leak size that l 24 they can accommodate without a net loss of coolant, reactor 25 coolant.
i HERITAGE REPORTING CORPORATION -- (202)628-4888
e -
186 1 The CE units span this range, and the Westingbouse 2 units span this range.
3 MR. DAVIS: It is really not the frequency of the 4 leak? It is the ability to make up leakage?
5 MR. ROSE: Yes, sir. I 6 MR. DAVIS: Thank you. Incidentally, I don't think 7 we were provided with the appendices to this report.
8 HR. ROSE: All of these figures are in 1919.
9 (Slide) 10 MR. ROSE: The second slide, specific conclusions 11 continued--the B&W units are in some but not all of the plants 12 subject to greater cooldown rates from overfeeding of
() 13 emergency feedwater--the point you made just a second ago.
14 That fact is acknowledged, is a part of the NPR conclusions, 15 in some but not all plants because some of the plants already 16 have limitations on the emergency feedwater flow rate.
17 MR. WARD: Those are the plants. What do they have?
18 MR. ROSE: Cavitating venturies: the B&W units are 19 equivalent to many other PWRs in terms of time available to l 20 use alternative means of decay heat removal on complete loss 21 of emergency feedwater or feedwater, j 22 MR. CATTON: When you say equivalent, what do you 23 mean?
24 HR. ROSE: I will show you. This is also in 1919.
25 This is based on Los Alamos study in NUREG CR 4471, and this i
k HERITAGE REPORTING CORPORATION -- (202)628-4888
187 1 is a comparison on a consistent time line of the time
[} 2 available, of the milestones and end points involved in a 3 total loss of feedwater accident, and what it shows for the
- 4 B&W units is that yes, the steam generator will dry out more 5 quickly than it does on the other units, but that is not the ,
6 end point. That is only an intermediary milestone.
7 The more important milestone or end point is the 8 time with which the operator can take action and prevent core 9 heat-up and damage, and what the Los Alamos study shows is 10 that the time available in the B&W units is comparable to that 11 in the CE unit. It is not, this is not the key point that we 12 need to focus on. It is more this point out here.
13 For the CE units, the study concluded that they must
}
14 begin feed and bleed cooling before steam generator dryout or 15 else they will be unable to avoid core heat-up in the absence l 16 of emergency feedwater in all cases, so what this study did 17 was try to put these things an a consistent time scale and l 18 compare not the time to steam generator dryout which is that y l
19 intermediate point, but the total time available to the 20 operator to take action to avoid the undesirable end point.
21 That's the basis.
22 MR. MICHELSON: Risk analysis, did you put in there 23 any factor to account for the reflooding of a dried-out 24 generator which would be the B&W case versus in the other 25 case, not having to reflood a dried-out steam generator?
HERITAGE REPORTING CORPORATION --- (202)628-4888
188 1 Certainly the risk has got to go up a little bit.
(} 2 Tube rupture, differential stretch, all that sort of thing has 3 got to be greater if you dry out the generator and sit there 4 for 20, 30 minutes before you wet it down again.
5 MR. ROSE: To my knowledge, that was not a part of l 6 the Los Alamos study.
7 MR. MICHELSON: That would increase the risk.
8 MR. ROSE: Time and again that is a point that is !
9 brought up, the time element.
10 MR. MICHELSON: It is very undesirable to dry out a 11 generator and then reflood it.
12 MR. REED: What Carl is saying is very, very true, i 13 and I am trying to figure out why you are comparing apples and O 14 orangess or what is here, because the real issues are when 15 does core heat-up begin? You notice it is out substantially.
16 The other another real issue is for all your great 17 injection that you have, you say you have got capable more 18 injection. More injection means higher pressures on the 19 primary side. The dryout of the steam generator means more 20 stress and strain and concern on the secondary side.
21 Now you have maximized the differential on the l l
22 tubes, which to me means more opportunity for tube leak, and l
23 if you put all these things together, I always come out with I
24 the fact that you have got sensitivity here that is of great 1 1 O 25 concern.
I HERITAGE REPORTING CORPORATION -- (202)628-4888
- = . .-
189 1 Now I don't think it is a good idea necessarily to
() 2 run pressures up when you are in a transient. A complex 3 transient, and drying out the steam generatoEs, up the 4 pressures up to a max of 2500 plus. It only increases the 5 opportunity for rupture.
6 MR. RUTHERFORD: I thought an intentional revolution 7 would be I lost the back-up to the backup system. If you lose 8 your safety grade auxiliary feedwater system, obviously you 9 are not going to sit there without putting water in the 10 cooler.
]
11 MR. REED: This morning we were told that you are 12 weighing this core damage issue on the bleed and feed and *
(} 13 ignoring the fidelity of the secondary feedwater emergency 14 system.
15 MR. RUTHERFORD: That is an incorrect 16 interpretation.
17 MR. REED: Well, that's what I heard.
18 MR. RUTHERFORD: What was said this morning, all 19 PWRs would go through a similar evolution if you will of 1 20 maintaining the most viable source of feedwater, main 21 feedwater hopefully first, if not that, emergency feedwater, 22 and then ultimately, if you have none of those sources, going l
i ,
23 to feed and bleed. I l
24 MR. REED: I agree that certainly feed and bleed or j O i 25 bleed and feed should be a back-up thing. It is dirty because i
HERITAGE REPORTING CORPORATION -- (202)628-4888
_. _ - - . - . - _ _ _ . _ _ _ _ - - ~_ _ - ..,._-.. -.. ._
190 1 it squirts in the containment. You don't want to squirt 2 inside containment any more than you have to because you have 3 got to clean it up.
4 I agree with that, but on the other hand--you use 5 that as the basis of your equivalency, and on another hand, 6 you want to say well, we don't want to use bleed and feed, use 7 the other things which in my opinion don't have equivalency, 8 namely time for dryout, and emergency feed capacity that you 9 have had in the past.
10 MR. ROSE: The only thing we were getting at in this 11 part of the sensitivity study was time available, and that was L 12 largely in response to the Subcommittee's concerns after our
() 13 last meeting in July of '86. This was in addition to the 14 original scope.
15 MR. RUTHERFORD: I don't think there are any 16 differences in the global sense between reliability of main i
17 feedwater on a B&W plant versus a Westinghouse plant, and i
18 likewise for the auxiliary feedwater systems. l l
19 MR. REED: Except for the big thing--inventory. '
\
l 20 Sure, I wil: agree with you, but you have got to figure that I l
21 inventory. l l
22 MR. RUTHERFORD: Well, the problems that we have ]
i 23 had, as we mentioned before, have been excessive inventory l
24 versus failure to deliver inventory, which is a different 25 problem if you will. j l
i HERITAGE REPORTING CORPORATION -- (202)628-4888 l
191 1 MR. REED: Am I to understand--I believe it was in 2 the record that Davis-Besse saved themselves strictly by time.
f}
3 If they hadn't gotten something done in so many minutes as 4 they did, that is the bringing in of isolated makeup pump, 5 they would have been in real serious core damage trouble, so 6 time and inventory are very important.
7 MR. RUTHERFORD: And what Stuart is saying in his 8 presentation is that the time available at different vender 9 plants would be comparable. The Westinghouse plant is better. :
10 It would be comparable to the heat plant in terms of total 11 time to take operator action.
12 MR. DAVIS: I got the impression your main feed 13 might be more reliable because you had or were going to 14 institute main feed runback procedures and would not trip main '
15 feed on some transients that for example, the Westinghouse 16 plants would trip.
17 MR. RUTHERFORD: Well, we have that capability now.
18 I think the recommendation was to wait to improving the 19 reliability in some cases. Yes, we do have that capability at 20 present.
21 HR. WARD: You don't consider that an advantage? ;
22 MR. RUTHERFORD: Oh, certainly; if you look at i
23 Westinghouse plants, they are a lot more likely to end up on 24 emergency feedwater than the B&W plants.
)
C) 25 MR. WARD: They absolutely depend on it as I i l
1 l
I HERITAGE REPORTING CORPORATION -- (202)628-4888
. _ . -. , -- _ _ _ _ _ = _ - . .- -. .
'l i
192 i i
1 understand, given a reactor trip, a turbine trip.
2 MR. RUTHERFORD: Right.
l 3 MR. DAVIS: Your success hasn't been real good in l 4 controlling the runback; as I recall, about one out of four or 5 five times that it has not worked. Is that a wrong number 6 or--
7 MR. RUTHERFORD: I, I can't dispute your numbers 8 there, that in some cases we haven't been able to, but there 9 have been a lot of cases where we have had success.
10 MR. TAYLOR: Let me make a general response to 11 Pete's question. There were, there have been quite a few 12 efforts in the last three years to improve main feedwater
)
() 13 control systems. There have been very significant r 14 improvements in the Arkansas plant. There have been very 15 significant improvements in Crystal River, and the frequency 16 of loss of feedwater events has dropped off significantly.
17 In the last year, there have been three cases where l
18 trips were avoided which would have occurred prior to that ;
- i 19 period of time because of these changes, and there have been 20 very few trips on the B&W plants in the last year, but because 21 of these main feedwater control system improvements, there 22 have been three cases where one feed pump has tripped, plant 23 has not tripped. The plant ran back. If it had happened a i
24 year prior to that, it would have gone down.
25 MR. DAVIS: Is your main feed--
]
I EERITAGE REPORTING CORPORATION -- (202)628-4888
l 193 j 1 HR. TAYLOR: Steam, all steam.
2 MR. MICHELSON: Help me out for just a minutes You (
3 showed on your previous draft that you dried out much sooner 4 in the B&W case but that--I think it was about 30 minutes 5 before you started to have difficulty, had to respond by that 6 time. And you said that's also the same for I guess the 7 Westinghouse. I can't read it that well from here.
8 MR. ROSE: For the Combustion unit; westinghouse is 9 on the bottom.
10 MR. MICHELSON: Okay. Now I am trying to determine 11 whether the, is the unique differences because the secondary 12 side of the Combustion unit certainly has a lot more water
() 13 then the B&W, doesn't it?
14 MR. ROSE: Yes.
15 HR. MICHELSON: And so first of all, I got to boil 16 that water away before I start really heating up severely the 17 primary side, because my heat sink is a reservoir of water.
18 You already said it is going to be much bigger for Combustion 19 ones having to go to the primary side.
20 Now I have got to start going off, relieving 21 something to take the heat out, but don't I have just as much 22 water on the primary side of the Combustion as I do on the 23 B&W, or at least approximately?
24 MR. ROSE: The way the study was conducted, what NPR 25 looked at, the Los Alamos study looked at emergency operating HERITAGE REPORTING CORPORATION -- (202)628-4888
194
-1 procedures that the Combustion units have to go through, and i 2 as has been noted, they do not have the high-head charging 3 pumps. First have to open the PORV, allow the primary system 3
4 to blow down before they can get charging, significant charged 5 water into the RCS. I believe that it is that additional i i
6 time-- 1 7 HR. HICHELSON: That's not the criterion I was I i
8 particularly interested in. You are getting pretty darned 9 excited on a PWR plant long before 30 minutes because you, i
, 10 particularly a plant like Davis-Besse, you are going to have 11 to do blowing down or something to it or it is going to blow 12 itself down through the safety relief valves long before the 13 Combustion unit is going to below itself down, long before.
14 Combustion has got so much secondary side water to take the 15 heat out first, so I don't know quite what we are comparing, 16 apples and bananas or just what.
17 HR. ROSE: I believe it is a valid comparison i
18 because--
19 HR. MICHELSON: Maybe operating procedures look ,
20 good, but what is happening to the plant may be quite l
t 21 different than the two cases, and the level of e::citement of i
22 the operator might be quite different.
23 HR. ROSE: The point we are trying to make here is
, 24 that up until this time, you are talking about steam generator
( 25 dryout. This was considered to be the end point. Steam HERITAGE REPORTING CORPORATION -- (202)628-4888
195 1 generator dryout, dryout, dryout, and there was no more O- 2 comparison.
! 3 MR. MICHELSON: B&W, only about 12 minutes or ;
i 4 something?
5 MR. REED: Four minutes.
6 MR. MICHELSON: Depending on the condition at the 7 time, could be as short as 2 minutes. ,
8 MR. ROSE: The point we are trying to make is that '
9 that is not the end of the picture. You have to also, when 10 you are trying to assess the time available for the operator r 11 on the board to take action to avoid core heat-up and damage, 12 that's not the end of the analysis. !
1
() 13 MR. MICHELSON: And the B&W case is seeing a lot I
14 different condition, up to 30 minutes, than it is--the CE case i
15 is a lot different. You already have a lot of primary side i 16 inventory I think by the end of 30 minutes, unless you keep :
17 making up. !
18 MR. WARD: All he is saying, if you look at the time 19 of core heat-up, it is a difference of only a factor of two j 20 instead of a factor of four with steam dryout, steam generator 21 dry.
l l 22 HR. ROSE: What we are saying-- i i
l 23 HR. WARD: It shows there is a big difference. '
l 24 MR. MICHELSON: Four minutes versus 30 minutes I
}
t i 25 thought was the time. i
~
)
l !
l HERITAGE REPORTING CORPORATION -- (202)628-4888
196 1 HR. WARD: Steam generator dryout, but so okay. So '
2 that's a factor of seven.
3 HR. MICHELSON: Not occurring until you get down 4 into the reactor vesse. That just happens a hell of a lot 5 sooner on a B&W plant.
6 HR. ROSE: Given the high makeup capacity in the B&W 7 units, all of this time is available to begin feed and bleed 8 cooling. If the operator initiates feed and bleed at any 9 point up until this time, they will be successful. For the
, 10 combustion unit, it must begin up until this point. If they 1
11 begin out here, it will not be successful, and so we are l
12 trying to put the time scale on a comparable basis. '
13 HR. SKILLHAN: What the arrow on the bottom shows is 14 if I can use the term, the drop dead time by which time those !
15 components must be in operation to prevent there from being 4
- 16 core damage. The top line shows that on the Babcock plant, 17 using two out of three 'fI and one PORV, on the time scale you 18 have got to make that decision and do it at about 28 minutes.
19 If you are going to go with two, three pumps, no PORV, you 20 have until about 38 minutes and you can stave off core damage.
21 Those pumps deliver so much flow you can cover the core right 22 now. Next line: Combustion Engineering, if you fail to get 1
23 three of three charging pumps plus the PORV, by about 22 l
2 24 minutes, you have lost it. You will never catch up to it.
() 25 CR . MICHELSON: Because of the capacity?
HERITAGE REPORTING CORPORATION -- (202)628-4888
197 l 1 KR. SKILLMAN: Other thermalhydraulic conditions in 2 the cooling system at that time. Drop down to the next line.
3 MR. MICHELSON: Plant is getting a little more 4 excited early on, but you are saying you have got big enough '
i 5 pumps?
\
6 MR. SKILLMAN: To recover; you can wait longer and 7 recover sooner on a Babcock plant because of I think the 8 inventory condition and some other, but you get down to the 9 bottom line on the Westinghouse plant, you have about 50 10 minutes to make the decision using one out 60 HPI pumps, two 1
11 PORVs also or about 130 or so with both HPIs. They have got a [
12 lot of secondary side.
() 13 MR. DAVIS: B&W line indicates that is for the 14 Oconee one. Is that, would that curve be identical for all I 15 the B&W plants?
16 MR. ROSE: All but Davis-Besse.
17 MR. MICHELSON: Large capacity high pressure makeup?
18 Large capacity.
i 19 HR. REED: Just so the B&W folks will feel better, 20 we have criticized quite a bit the CE valve 80 design because '
21 of its inability to depressurize rapidly, so you know that we I 22 criticize other people, and that's what you are talking about, i 23 MR. REED: Handed pumps depressurization through 24 PORVs.
25 CHAIRMAN WYLIE: On your earlier slide, on your j l
i HERITAGE REPORTING CORPORATION -- (202)628-4888
198 1 studies regarding the B&W units being less sensitive to steam T
{m/ 2 upset, do I conclude that, do I conclude from that that the 3 B&W units can consistently take full load rejection without 4 tripping the reactor?
5 MR. ROSE: No, sir.
6 CHAIRMAN WYLIE: I thought that is what that said.
7 MR. ROSE: This conclusion--I lost it--is based upon 8 the analysis that shows that given a step decrease in steam 9 flow, or a large increase in steam pressure, leading to a 10 decrease in steam flow, that the nature of the once through 11 generator will tend to try and return naturally to 12 equilibrium. As the steam pressure goes up, steam flow goes
() 13 down. The feedwater flow rate remains essentially the same.
14 The level starts to rise of the once through generator, and 15 eventually you tend to recover, come back to a lower power 16 level, but come back to an equilibrium. Because the U tube 17 generators don't have that variable heat transfer area, they 18 don't respond that way.
19 This is a plot of steam flow as a function of time 20 for a step change in turbine control valve position closure.
21 CHAIRMAN WYLIE: Small step.
22 MR. ROSE: Approximately 10 percent or so; this is 23 part of the bare plant analysis.
24 CHAIRMAN WYLIE: What happens when you get full load 25 rejection? They are designed to run back?
HERITAGE REPORTING CORPORATION -- (202)628-4888
199 1 MR. ROSE: They are designed to run back, and if 2 everything responds the way it is designed to, they can.
3 CHAIRMAN WYLIE: Have they?
l l 4 HR. ROSE: I believe they have.
5 CHAIRMAN WYLIE: Consistently?
! 6 HR. SKILLHAN: Not consistently, but they have done l c
! 7 it. It is not normal.
l 8 HR. ROSE: And since THI, of course, the point--
l l 9 CHAIRMAN WYLIE: What prevents them? ;
10 !
j HR. SKILLMAN: The high pressure set points has been 11 set down, and the PORV set point has been changed. Prior to l_
l 12 those changes, the plants were designed to be able to take a 13 hundred percent load rejection. With those changes, you are 0 14 not able to absorb that quick energy lift. Consequently, you 15 get yourself in the HP trip set point about 2300 psig. Prior i
16 to those changes in the plant, the plants were able to take, l 17 theoretically able to take runback from 100 percent, and the 1
l 18 plants were, they were designed to do that.
19 CHAIRMAN WYLIE: Did they ever do it?
l 20 HR. SKILLMAN: Yes, sir. THI-2 did it. ,
21 MR. RUTHERFORD: We weren't able to.
22 CHAIRMAN WYLIE: I just drew the wrong conclusion l
23 from that bullet I guess.
24 HR. ROSE: Final two bullets on this page, again for I
() 25 most reactor trips, B&W units do not impose greater control HERITAGE REPORTING CORPORATION -- (202)628-4888 ;
i
200 1 burdens on plant operators, but they do impose greater burdens
( 2 on the operators in diagnosing and respoinding to failures of 3 the automatic control systems. Such failures are more likely 4 to lead to complex transients in B&W units.
5 MR. MICHELSCN: I guess what you are saying is if 6 averything works right, it is no greater burden than other 7 types of plants, but maybe there are more things that have to 8 work right and quickly in order to keep the burden off the 9 operator? I think t st's just a fast response system, and if i 10 it works right, then it is no different than others, but if it 11 doesn't work right, the operator really has got to scurry in a 12 hurry.
() 13 HR. ROSE: In some respects, it is less of a burden 14 on the operators under the normal case two versus four loops, l
15 that they have to keep taps with, and automatic feedwater [
16 control on the B&W design versus manual control on other some
? of the other designs, but again, that's for the nominal case.
18 MR. MICHELSON: Automatic feedwater control, all of 19 them have automatic feedwater control.
20 MR. ROSE: Post-reactor trip; some of the other ,
21 6esigns, when they isolate or trip main feedwater, go to 22 emergency feedwater, that's manually controlled. The 23 operators have to step in ant control it.
24 MR. MICHELSON. Control the auxiliary feedwater; I i 25 thought you said main feedwater control. I misunderstood.
I i
HERITAGE REPORTING CORPORATION -- (202)628-4888-
201 1 1
l 1- Sorry.
() 2 MR. ROSE: Recommendations--which there were five, 1 3 although some of them had several' parts to them. )
4 NPR recommended modifications to the ICS as being i
5 essential for a substantial. reduction in the frequency of 6 feedwater upset, and to the elimination of complex transients.
7 Types of things they were recommending be done is a 8 separation of the feedwater pump speed control throu st the.
9 feedwater, regulating feedwater control valve position.
10 An independence between trains of the main feedwater 11 system, the feedwater pump speed control, and position of its 12 corresponding regulating valves be made independent between ,;
{} 13 the two trains.
14 HR. MICHELSON: These eve all steam-driven main ;
15 feedwaters, all the plants?
3 16 MR. ROSE: Yes. That NPR reflected that control of L
17 RCSTs and secondary side pressure, those two key variables, be 18 assigned to separate individual sub-systems, and not >
19 integrated into the rest of the control system.
t 20 MR. REED: First two bullets, has-this separation i
21 taken place yet? Has it been done anywhere?
22 MR. ROSE: No, sir. It is being considered as part 23 of the larger issue of ICS.
24 MR. REED:. I would be bothered having the two things-O 25 independent because these two independent things then would HERITAGE REPORTING _ CORPORATION -- (202)628-4888
.a ,_ . u ..,_-__u.__..._.. . . . _ . _ . . _ _ _ . , _ _ _ , _ . . . _ . _ , . . . _ . . _ _ . _ , .
. - - . - . .~ . .._ . - . . . . . .- . . . - . . .
202 ;
l 1 both be trying to do the same thing, namely, control the flow 1
.i O- 2 of feedwater into the steam generators. I would worry about 3 bucking reactions between these two sets of controllers, 4 trying to do the same thing independently,.but you haven't 5- done it yet.
6 MR. ROSE: I think we have similar concern. That-is r 7 why we are wanting to look at this in a total. context.
8 NPR recommended that we consider use of redundant ,
9 detector with auctioneering to select output, and we consider 10 use of pneumatic accumulators to reduce the likelihood of i
11 losing key positioners, feedwater regulating valves.
12 They also carefully pointed out that we should
() 13 retain certain positive features of the integrated control 14 system, that these were very desirable and beneficial in terms I 15 of controlling the plant; the coordinated reduction reactor 16 power and steam demand on loss of one feedwater pump, which ,
17 the ICS provides; limiting feedwater flow to each steam 18 generator based on a number of reactor coolant pumps th..t are 19 supplying that generator, and over, having overriding control I 20 to maintain a minimum water level in the generators--these 21 were positive features that we need to ensure that we retain.
22 MR. MICHELSON: Is there going to be an. attempt.now 23 to make dual trains or something of-that sort ,on ICS, in other 24 words, so single failure doesn't put it in such a fancy
, 25 complex transient?
r HERITAGE REPORTING CORPORATION -- (202)628-4888 l
_ _ _ , , - _ _ , - _ ~ . _ _ _ . _ . . - . . _ . _ _ - . _ _ . ~ , _ _ _ _ . ,
203 1 MR. ROSE: If I could defer that question to Larry
() 2 Stalter when he comes up to talk about the ICS, I think he 3 will tell you what we are looking at.
4 MR. DAVIS: I had a question on the main feed. You 5 require AC power to operate main feed?
6 HR. ROSE: I believe so; yes.
7 MR. DAVIS: Even though they are steam driven?
9 MR. DAVIS: Yes.
10 MR. RUTEERFORD: You have got booster pumps in some 11 desions and back-up--
12 MR. DAVIS: Thank you.
- 13 (Slide) 14 MR. ROSE: The second recommendation from NPR was 15 that we should reduce by order of magnitude the frequency with 16 which main feedwater upsets occur.
17 Thirdly, that we limit the maximum flow rate 18 delivered by the emergency feedwater system or reduce the 19 likelihood of persistent demand for high flow limitation on 20 steam generator fill rate. We are pursuing the first part of 21 that.
22 Number 4, perform analysis for and cbtain licensing 23 approvals to remove the anticipatory trip on turbine trip.
24 This trip increases the plant's susceptibility to overcooling O 25 without markedly increasing overall plant safety. We are not l
HERITAGE REPORTING CORPORATION -- (202)628-4888
204 1 pursuing that recommendation.
A k) m 2 And No. 5, employ the turbine governor valves to 3 relieve surplus energy following reactor trip in lieu of 4 atmospheric dump valves, turbine bypass valves or main steam 5 safety valves. We are not pursuing that at this time.
6 The last two recommendations were couched as an 7 option on NPR's part, not related to conclusions of theirs 8 within the context of the sensitivity study.
9 MR. MICHELSON: What is your thought on item 4 10 concerning the reliability of turbine trip, the thought that 11 it is very reliable?
12 MR. ROSE: The anticipatory reactor trip on turbine 13 trip, I think has been very reliable. I am not aware of
{
14 instances where it has failed.
15 MR. MICHELSON: I am thinking more of the case you 16 should be getting turbine trip and you aren't. Better have 17 reactor trip on some other signal. I notice lately there have 18 been a lot of LERs on turbines that have failed to trip 19 because of the plants in the turbine control system. That 20 worries me. I thought that was one of the most reliable of 21 the arrangements we had in our entire nuclear plant was the 22 turbine trip, not turbine trip on the lower side, but turbine 23 trip on lead. If the turbine trip is not so reliable as I 24 thought, then it ought to be anticipatory reactor trip on 25 other than this turbine trip.
HERITAGE REPORTING CORPORATION -- (202)628-4888
205 1 MR. JONES: Crystal River.
(%
MR. MICHELSON: I guess that was it. That was,
(_) 2 3 really kind of concerned me. I read not too long before that 4 another person discovered that the reactor wouldn't have, the 5 turbine wouldn't have tripped except the heating tripped it, 6 but it is a different circumstance, a second LER, different 7 one. That led me to believe turbine trips aren't as good as I 8 thought they were.
9 MR. ROSE: I believe those concerns are somewhat 10 different from what NPR was getting at here.
11 MR. MICHELSON: I guess it must be. Yes. Thank 12 you.
13 (Slide)
(m) 14 MR. MICHELSON: Why won't you want to trip the 15 reactor, trip the turbine?
16 MR. ROSE: To recain the capability to ride it out, 17 ride out the transient, as originally designed.
18 MR. REED: But that's a utility desire that doesn't 19 necessarily relate to safety. I think safety should get 20 precedence, and not a desire to maintain the reactor rods up.
21 MR. MICHELSON: The idea is you would like to try 22 the bypasses first, that sort of thing, to see if that will 23 carry you through. That's a full power trip.
g, 24 MR. ROSE: As I said, we are not pursuing this V
25 n'rticular recommendation.
HERITAGE REPORTING CORPORATION -- (202)628-4888
-206 1 The sensitivity study also incorporated a peer I) 2 review process. The peer review-group was established to 3 provide analytical input and review. The group included 4 utility, vender, and NRC thermalhydraulic analysts. These 5 people were experienced in the development of codes and 6 models, and were.very valuable. Therefore, input was very 7 beneficial during this phase of the sensitivity study.
8 The peer review group concluded that the analysis 9 and resc .s are technically sound and appropriate for the 10 study's comparative scope, that the results of the sensitivity 11 in and of themselves are not sufficient to draw conclusions on 12 design changes, and that subsequent evaluation of ICS design 13 changes should caretally consider the total impact cn1 plant 14 response and other transient aspects. I think several of the ,
i 15 concerns that you expressed, Mr. Reed, on main feedwater i
16 control, you want to look at that in its total context. .
i 17 And it is up to the owners group to confirm that the !
i 18 proposed reconfiguration can be achieved while preserving the .
19 current positive capabilities of the ICS as NPR pointed out.
20 That's all the material I had to cover. If there are no other i 21 questions? Bob, are you next?
i 22 MR. JONES: I think we are going to get L oreak.
23 MR. WARD: Oh, please, Mr. Chairman. :
1 24 CHAIRMAN WYLIE: Okay. Let's take a break. We will 25 take a 15-minute break.
HERITAGE REPORTING CORPORATION -- (202)628-4888
- _ , - - _ _ _ . . _ . , _ , _ .. . , . _ , ~ . . _ . . _ . _ - . _ - . _ . _ . _ _ . . _ _ _ , _ , . _.- . _ , . _ . - . - . , . . _ ._, _. - --
207 1 (A brief recess was taken'.)
1
(} 2 CHAIRMAN WYLIE: We will resume our meeting.
3 MR. JONES: The staff's view of the sensitivity 4 evaluation the NUREG document--whatI would like to cover first 1 5 is what did we do?
6 The basic activities that the staff performed was we 7 reviewed the specific MPR evaluation, both model adequacy and 8 the conclusions and recommendations. This was done using some 9 contractor assistance for some calculations which were done as 10 part of another research program.
11 Stuart Rose mentioned that NRC was a member of the 12 peer review group. I want to take a little bit of exception 13 to that. We, we participated in the peer review process in 14 order to audit it, but we were not a formal member of the peer 15 review. We didn't feel that would be an appropriate thing for 16 us to do, so their peer review report was done without us, to 17 clear it up. l l
18 We also went and looked at the issue of pressurizer l l
l 19 size. Stuart mentioned that the owners group looked at it and l 1
20 NPR. The issue of pressurizer size is an old issue that we, l l
21 that was raised by the staff following the TMI accident. l l
22 One of the questions that we asked ourselves was 23 what, what rules do we have out there where the B&W plant 24 design may have an impact on the response of the plant, I 25 whether or not rules that we put out appropriately considered HERITAGE REP'RTING O CORPORATION -- (202)628-4888 '
. . - . . . . -. .. - . ~. - .-
208 1 the specific or unique design characteristics of the PRA 2 plant.
3 We looked at two specific rules, the ATWS and PTS 4 rule, to assure ourselves that any specific design 5 characteristics of the PRA plant were appropriately accounted 6 for, and per the Commission's request, we looked at'the' 7 adequacy of the OTNG.
8 (Slide) 9 MR. JONES: What did we find? In general, you saw a 10 large list from Stuart of the, what NPR concluded with respect
~
11 to sensitivity, and we generally agreed with what they came up 12 with.
() 13 We looked at the responses that were calculated by 14 NPR and some independent stuff done by, by our contractor, and 15 basically what we were finding was the model was indeed 16 reasonable and was coming up with reasonable answers and we 17 had no reason to doubt the conclusions that were being drawn.
18 We did have some a small caveat on some of them 19 which I will try to outline as I go through it. One of the 20 issues that always seems lo come up is the issue of 21 ' sensitivity of B&W plants to feedwater upsets. Clearly the 22 B&W plants are more sensitive. The NPR study just further 23 confirmed the previous findings.
24 We did agree that B&W plants are generally not more 25 sensitive to other design upsets, including the upset in steam HERITAGE REPORTING CORPORATION -- (202)628-4888
209 1 flow. Now there has been a lot of discussion on that today.
O
%/ 2 Basically what happens is for a steam flow upset, you get a 3 change in the surface area in the steam generator which tends 4 to mitigate the upset, so if you have a step change in feed 5 flow, you will, your level changes a little bit to 6 accommodate, give you a little bit extra heat transfer area, l
7 and then the pressure goes up and you get about the same heat i 8 re:- cval , you get a balanced situation. l l
9 NPR also looked at the response of the plants to a 10 total loss, I mean a turbine trip without a reactor trip in 11 the system, and you can again see clearly the B&W plant does 12 not reach as high a system pressure, for that transient
() 13 tending to level off. While there is, it is less sensitive, 14 it is not, we don't buy the global statement that it is much 15 less sensitive. We think it is somewhat. It is not a big 16 deal because all the plants will generally trip in response to 17 a turbine trip with the revised set points at the B&W plants 28 today for reactor trip, et cetera, and the adversion of the 19 PORV.
20 In the area of operator burden, the NPR came up two 21 conclusions.
22 MR. WARD: You skipped the third.
23 MR. JONES: Oh. EFW--we conclude--again, this is, 24 this is not a surprising conclusion--that B&W plants do place 25 a greater reliance on emergency feedwater delivery, including HERITAGE REPORTING CORPORATION -- (202)628-4888
210 1 the proper regulation of emergency feedwater flow. What does 1
.() 2 does that mean?
I l
3 MR. WARD: That surprised me. I thought we just i
4 heard a little while ago that there is greater capability of 5 running deen on main feedwater.
6 MR. JONES: Right. And so the question is what does ;
7 this mean? First off, the issue of flow I think is 8 straightforward. That's they have got enough, more than 9 enough feedwater flow capability, that they had it with full 10 food flow. They depressurize the steam generator-and initiate ;
11 overcooling transient, so when you do deliver EFW, if you want 12 proper plant control, you have to control your emergency 13 feedwater flow within reasonable values.
14 The issue of greater reliance comes from the loss of i 15
~
feedwater scenarios, total loss of feedwater scenarios. As ,
i 16 shown earlier, there is only about two minutes in a B&W plant 17 to, before dryout, two to four minutes, while in a l
18 Westinghouse, CE plant, those are 20 minutes or 60 minutes. I 19 Depends on the plant. So that from the standpoint of l
20 maintaining the heat sink for, with assuring the heat sink is l l
21 there, you do require your emergency feedwater system sooner 22 in the B&W plant.
23 MR. WARD: What that really means is given loss of 24 main feedwater, you have got to turn on auxiliary feedwater
(:) 25 faster?
l l
)
i HERITAGE REPORTING CORPORATION -- (202)628-4888 i
l 1
211 l l
1 MR. JONES: Yes. I i
() 2 MR. WARD: It doesn't say that.
3 MR. JONES: Okay. The other issue is, for example, .
4 the repressurization of.the plant to the PORV, given a loss of
~
5 feedwater, and in a B&W plant, .that will occur in about seven, 6 eight minuten, so you would like to have your emergency 7 feedwater flow on before you open up another hole in ths 8 primary system, or open hole in the primary system. ;
9 And for those reasons, we think that they place a 10 greater reliance on response to a loss of feedwater event.
r 11 With respect to operator burden, for a quote normal 12 reactor trip, basically if most of the control systems 13 function properly, the B&W plant and the other plants, really 14 there isn't a great big deal of difference between them in 15 what action the operator is required to take.
- 16 The NPR study, however, was limited. It did not 17 look at transients which got into actuation of high pressure 18 injection and SFRCS, et cetera. The operating experience 19 clearly shovs that when you get to one of these category C 20 events, there is e lot of operator actions that are, that take 21 place at those points in time, so while the NPR conclusion is 22 valid, it is also limited in its applicability.
23 We clearly agree that following a failure of the ,
24 ICS/NNI, you definitely have a much greater operator burden.in 25 B&W plants than the other plants.
HERITAGE REPORTING CORPORATION -- (202)628-4888-
212 1 DR. KiRR: Is that a disadvantage or an advantage?
(')/
(_, 2 MR. JONES: Which?
3 DR. KERR: To place a greater burden on the others?
4 I ar. not being facetious.
5 MR. JONES: In, for example, loss of the ICS/NNI I 6 think it is a disadvantage. Part of that disadvantage, and 7 again I am going to answer from the standpoint of the, before 8 the study to some extent because I think that's, that's the 9 timeframe we have to look at--you ha c a lot of problems with 10 the loss of ICS/NNI, what instruments are valid. You don't 11 know the state of the plant or it's difficult to determine the 12 state of plant. You have got to diagnose which instruments
{} 13 are valid, et cetera. Then you have got to take the proper 14 action, and that's what makes it much more difficult.
15 DR. KERR: More than placing a greater burden, the 16 operator doesn't know what is going on. Is that what you are 17 saying?
18 MR. JONES: He may have difficulty knowing what is 19 going on or what action to take. That's correct, although 20 there were some actions taken back in response to like the 21 79-27s and the Crystal River events to improve some of that.
l 22 There are a lot of actions which are specified or developed as 23 a result of SPIP to further enhance the operator's performance i
24 in this area.
g-)
(/
25 We generally agree with the recoinmendations of NPR HERITAGE REPORTING CORPORATION -- (202)628-4888
213 1 which.are listed, although the ICS recommendations is going to 2 be discussed more by Rick Kendall when we talk to the' specific
'3 ICS/NNI review.
-4 We disagreed with the removal of the ARTS on turbine 5 trip. fWe see it has some safety advantages. They.are not 6 pursuing it anyway. It is a moot point at this' point.
7- We also looked at the previous action that staff has 8' taken as a result primarily of the TMI event and whether they 9 were effective in at least responding to some of the inherent 10 B&W plant design characteristics, action such as raising the 11 PORV set point, lowering the high pressure trip set point, and 12 the addition of the anticipatory reactor trip systems.
() 13 Looking at them collectively, we think they have been an asset 14 to safety.
15 The addition of the ARTS, especially the main i
16 feedwater, ARTS maintains inventory in the secondary system, 17 and is, really eliminates the initial pressure overshoot for a 18 loss of feedwater transient. It. won't do it in every exact 19 case, but if I have lost--depending on the signal employed at 20 the plant. For many losses i main feedwater situation where 21 they lose all main feedwater, you will get a fairly prompt 22 reactor trip and nice orderly shutdown of the plant.
23 Raising the core obviously decreased the possibility
[} 24 of a, of a small break LOCA, so we think that-the actions we 25 have taken in the past are appropriate and indeed-were shown t
HERITAGE REPORTING CORPORATION -- (202)628-4888 ,
214 1 effective by operating experience.
One other point I want to make there--I didn't have
[w '/I 2 3 this on the slide, but the icsue of pressurizer size, there 4 was some discussion on that.
5 DR. KERR: You reached that conclusion based on just 6 good engineering judgment? You didn't do--
7 MR. JONES: Looking at NPR results; they had 8 analyses where they did not have the ARTS, had the ARTS.
9 DR. KERR: Didn't do a risk analysis or anything?
10 MR. JONES: No, we did not. That was engineering 11 judgment.
12 One of the issues was the pressurizer sizing issue j
g- 13 that we did look at. We looked at various sources where we V
14 have compared pressurizer sizes, and there were ecrly reports j 15 which irdicated B&W pressurizers were undersized. There were 16 other reports such as the ATWS report that you indicated B&W I
17 pressurizers were bigger than other guys on some relative i i
18 term. We looked at it in a slightly different manner which j l
19 was what was the concern associated with the pressurizer size? !
l 20 The primary issue we came up with was losses of j 21 pressurizer indication. Pre-TMI there had been several l
22 incidences where pressurizer level indications was lost. The !
l 23 post-THI history, especially since the owners group had or all '
24 of the B&W plants have apparently raised their normal 25 operating level a couple of feet in the pressurizer, has kept HERITAGE REPORTING CORPORATION -- (202)628-4888
.. ._ ~ _ . . . - - .- . . . .- -
215 1 the pressurizer level on scale for all but I think it was one 2 event.
3 Since about 1982, there was an earlier event.at 4 Arkansas before they raised their set point where they lost-l 5 pressurizer level indication, but since then, there have been
-6 only the one. That was Rancho Seco, which is not_much of a l l
7 surprise.
8 So if they, the operating history has preved that 9 while the pressurizer maybe could have possibly been larger, 10 there is no need for it to be larger from the safety-11 standpoint. In most transients, it will stay on scale, 12 We looked at the inherent characteristics of the
() 13 plant with respect to the PTS, ATFS rules. We relooked at 14 some of the calculations done for PTS. Clearly what the PTS 15 rule does, the PTS rule clearly.shows that what drives PTS for 16 a B&W plant is different from the other guys. It is
- 17 overcocling, secondary side events that drive it for B&W as 18 opposed to LOCAs for the Westinghouse and CE type plants, so
~
19 in looking at it, going _through some of the history for the f 20 rule, it is clear that the inherent design characteristics of 'l 21 the B&W plant was considered. i l
22 Similarly ATWS, there was a bunch of evaluations of j 23 the basic designs, and clearly the ATWS rule which has l 24 different requirements for B&W plant, as opposed to 25 Westinghouse plant, clearly those design characteristics HERITAGE REPORTING CORPORATION -- (202)628-4888
216 ,
1 differences were accounted for.
2 Once through steam generator design--a lot of
(( )
3 questions about the responsiveness of the design and how if 4 you feed back into the primary system, so we tried to take a 5 look at is it a good design, is it a bad design? What can we 6 say about it?
7 What we found in looking at it is, and again this is 8 not really that new of a conclusion, we arrived at this back 9 in 0667, we just confirmed it with today's operating history, 10 there are certain operational advantages to the OTSG, 11 primarily the superheat advantage and tube integrity 12 advantages. The tube integrity for the B&W plants has been g3 13 better for than the Westinghouse and CE designs, so cleari, b]
14 there is something about the design of the OTSG from that 15 aspect which is an advantage.
16 HR. REED: I would like to get more information on ;
1 17 tube integrity. l The tubes on the B&W probably fail near the ,
18 top I would guess, whereas the tubes on the recires fail near 19 the bottom, or below the, many feet below the water level.
20 Now certainly from an atmospheric release point of 21 view, tube failure near the top in the steam phase is much 22 more of a radioactivity concern than a tube failure down in 23 the bottom, although you may if you apply differences, 24 calculate the differences to that, h
25 Now isn't it the case they are failing up near the 1
HERITAGE REPORTING CORPORATION -- (202)628-4888
217 1 top?
(} 2 MR. JONES: First off, how do we come up with the 3 issue of tube integrity? The failures that have been seen 4 have been up top. Now failures as, is the question. What do 5 you mean by failure? One of the things is we didn't find any 6 tube ruptures for any B&W plant. There have been leakers, but 7 not many ruptures. There have been ruptures on the 8 Westinghouse, CE designs.
9 MR. REED: There was no sudden rupture incidents at 10 all?
11 MR. JONES: There have been leaks. The histories of 12 plugging the tubes indicative of some sort of damage mechanism 13 for the B&W tubes, well, for any steam generator tube, it is 4 14 less--B&W has had to plug a lot less steam generator tubes l I
15 than the other PWR venders. All those indicate that the, )
l 16 there is something about the design, and I'm not the f
17 engineering type for the design aspects, but clearly the 18 operating history indicates that they are not as susceptible 19 to the damage mechanisms as the other PWRs are experiencing.
20 They are not having the same type of--they are having the same f 21 types of problems, but not of the same magnitude.
l 22 MR. REED: Aren't they gatting much more tube 23 fouling?
24 MR. JONES: They have fouling in the tube sheets or O 25 tube support plates.
HERITAGE REPORTING CORPORATION -- (202)628-4888
218 1 MR. REED: I am talking by that loss of power
(.
\ 2 capability because of scale formation on the outside of the 3 tubes on the secondary side.
4 MR. JONES: I think the fouling that has been seen 5 has been seen at the tube support plates.
6 MR. REED: I don't think so.
7 MR. GANTHER: Yes.
8 MR. JONES: The broached holes, there has been crud 9 build-up in those.
10 MR. REED: What has this--I read recently some, some 11 reactors had to do a cleaning operation on the secondary side 12 of the tubes because of tube scaling found and then couldn't 13 get up to full power?
(( })
14 MR. JONES: My understanding of that is crud 15 build-up in the broached holes in tube support plates.
16 MR. REED: Just in the little holes?
17 MR. JONES: Increasing the resistance which 18 increases the water level in the generator, which causes them 19 problems in operating because they can't operate above a 20 certain level and the steam generators have to cut back power 21 to assure --
22 MR. RUTHERFORD: Oconee units 1 and 2 chemically 23 cleaned their steam generators in the last outages, refueling 24 outages for those units.
25 MR. REED: It wasn't because of gross scale at the HERITAGE REPORTING CORPORATION -- (202)628-4888
219 1 water level and above because of, or near the water level -
2 where residue was bound to be left on the tube, or you think 3 would be? Well, it is just in the holes. All right.
4 MR. RUTHERFORD: Primarily the holes; support 5 plates.
6 MR. REED: As I understand it, once through steam 7 generator, the tubes are essentially in, well, except for the 8 pressure differential, are essentially in zero stress 9 condition at operating temperatures. In other words, the 10 tubes are more stressed when the steam generators are cooled 11 down than they are at the operating condition. It was 12 designed to be that way, am I correct?
13 MR. RUTHERFORD: Correct.
[
$ 14 MR. REED: Which tells me that overcooling events 15 will increase the mechanical longitudinal natural stresses, 16 which bothers me about your statement of better integrity. I 17 am wondering if you need some more overcooling events to see 18 whether you can pop a tube circumferentially.
19 MR. JONES: You have design limits on thermal cycles 20 for the tubes. You have design limits on indications in the 21 tube for when you have to plug the tubes and take them out of 22 service, and they account for these transients cycles.
23 What really, what the history is te] ling you is you j 24 are not seeing the degradation down to the minimum thicknesses j 25 where you need to worry about or really start to worry very l
HERITAGE REPORTING CORPORATION -- (202)628-4888
220 1 badly about the cracking or the failure of the tube due to
( 2 these transient conditions. It is not being experienced as 3 rapidly on a B&W plant as it is on the other designs.
4 MR. REED: Well, I might say that I never considered 5 the recirculating steam generators were designed at all before 6 Yankee-Rowe, Haddam days, or some of the others, that only in 7 the last few years with steam generator replacements have 8 there been the kind of design look as that there should have I 9 been when they just went and ran and picked up an old i
10 feedwater heater, turned it en, so that may be a wash in ten l
11 years out.
l l 12 MR. JONES: It may be. The other, the other thing 1
() 13 that clearly seems to pop up, well, clearly the smaller 14 inventory and the steam generator offsets some of those 15 operational advantageous because you can repressurize quicker.
16 You can open PORVs quick even if you have a delay in your 17 burdens.
18 The other thing that seems to pop up from the NPR I
19 study and looking at the operating history is not the design l 20 that is a problem. Not really the once through steam i 21 generator; what you are; seeing is its response to failure to l
l 22 properly control those systems that interacts with the steam 23 generator--the emergency feedwater system, the main feed l
l 24 system, ICS/NNI failures which propagate through those systems 25 and affect it. And one way to assure proper post-trip control l
l HERITAGE REPORTING CORPORATION -- (202)628-4888
221 1 of these plants and keep, to keep the response in balance is l
(\ /~) 2 not to go out necessarily and replace the steam 3 generator--just fix those systems that connect to the steam 4 generator _and assure they are properly controlled, maintained, 5 tuned, which is the objective of the SPIP, which is the 6 objective of many of the SPIP recommendations, so from that 1
7 standpoint, we come to the bottom line which is looking at the 8 sensitivity issue as a whole, we don't see a specific design 9 or safety problem associated strictly from the sensitivity of 10 the B&W plant, and we think that SPIP is addressing those 11 systems which the B&W plant seem to show are the most 12 sensitive, or reflect itself in the sensitivity.
() 13 That's it.
14 CHAIRMAN WYLIE: Okay.
15 MR. REED: I certainly agree with you. If you can 16 decide what the spray flow should be on the steam generator, 17 what the level should be and what the accident is that you 18 have, and you can get a control system to take all those 19 things into recognition, and do things appropriately, it is 20 complex.
21 MR. JONES: We know what the levels need to go to.
22 I don't think there is a problem with the levels, and specific 23 flow rates I think we can get a handle on.
24 MR. CATTON: Can you do that without knowing about 25 the internal thermalhydraulics?
HERITAGE REPORTING CORPORATION -- (202)628-4888
222 l 1 MR. JONES: Within reason; I think we can come up .
l
("')
(/ 2 with a fair estimate, but if you try to pinpoint it within a :
3 couple of percent, no.
4 MR. CATTON: Who cares about a couple of percent?
5 But well within the uncertainty?
6 HR. JONES: Yes, I think we could estimate it.
7 MR. CATTON: You really don't need to wait for the 8 experimental program? Or maybe you don't even need it.
9 MR. JONES: Depends on how far the licensees propose 10 to cut it back. The finer you want to cut it to keep your 11 overcooling potential down, the more detailed understanding 12 you need to know with the steam generator, of its
() 13 thermalhydraulics, the wetting characteristic, the heat 14 transfer, the flooding, et cetera. The more you cut it down, ,
15 the better you need understand it. The values they have
'6 gotten no1, I don't think you have got a problem, but the 17 values are too high and you are seeing that in operating l
18 experience, so what is the proper value is a difficult o
19 decision to make.
i 20 MR. CATTON: Sounds to me like you do need it. j l
21 MR. JONES: That's our opinion. That's why we are l 22 discussing it.
23 CHAIRMAN WYLIE: Any other questions? Okay. Well, I
24 thank you. And let's move on to the integrated control 25 system.
1 HERITAGE REPORTING CORPORATION -- (202)628-4888
223 1 MR. STALTER: I am Larry Stalter_with Toledo Edison
() 2 Company. I am chairman of the B&W owners group INC Committee.
3 MR. MICHELSON: With which company?
4 . HR . STALTER: Toledo Edison; I am going to discuss 5 the ICS/NNI evaluation-that was performed by the INC C Committee, and first couple of slides _here is going to discuss 7 the overall process, and then I will get into some of the 8 specifics of the program.
9 This process began by establishing first of all some l 10 updated design requirements, and to do that, we chose that we 11 would consider only the existing system as it was in the 12 plant, given we had that technology, what requirements will, j i
13 would apply to that today, and we developed from that a set of 14 design requirements. We used original design requirements 15 that existed. !
16 We also utilized previous failure modes and effects 17 analysis that were performed, in this case B&W 1564, which was 18 done following the TMI event. We reviewed industry and NRC t
19 recommendations, and we did, we did utilize INE Bulletin 20 79-27. We did look at NSAC 3 and did look at NUREG 0667. I i
i 21 We also took in developing our requirements previous 22 industry studies, and there was a study that was ongoing 23 before this one began which, which was to develop i 24 recommendations for ICS imp'.ovements, and that was done by O 25 another committee on the B&W owners group. We utilized that.
4 HERITAGE REPORTING CORPORATION -- (202)628-4888 :
224 1 We looked at the operating philosophy at each plant.
{} 2 We reviewed the operating history and we looked at the FSAR 3 assumptions that were used relating to the ICS/NNI, and that 4 gave us a base, updated set of ICS/NNI design requirements 5 from which we could work from.
6 We then looked at the established as-built 7 configuration at each of the plants. We reviewed the 8 drawings. Wo looked at the modifications that had been 9 performed, and those modifications, many of them came from 10 previous recommendations and previous studies, and had been 11 implemented in plants. And we looked at the programmatic 12 practices at each of the plants, and with that, we established 13 the as-built configuration and how the ICS operated at each of
( 14 the plants.
15 The next step in that process then was to do a 16 comparison between those and that, in that comparison, utilize 17 it to develop a set of problems and potential solutions.
18 Now the problems, potential solutions existed as 19 supplementary documents in B&W 1919 and the ;wo in particular 20 are the problems of in the past the solutions that we 21 developed, and there is also configuration matrices for both 22 the ICS and the NNI, which also have in them cases where one 23 particular plant would have a feature and another one may not.
24 Now at the same time that this was going on, we 25 tried, we tried to incorporate in our, in our program HERITAGE REPORTING CORPORATION -- (202)628-4888
225 1 technological improvements, and our initial look at b
s/ 2 technological improvements, we tried to see how they could fit 3 in the existing design, and we didn't find anything in the 4 industry that we could utilize, and this was early on in the 5 evaluation.
6 Now I point this out because later on in the 7 evaluation, some of our key recommendations depend on 8 technological improvements, and that that's where there was a 9 development effort to look at smart signal selection, and a 10 development effort to, initiated by us, to develop an 11 indicator that would assist the operator in giving him an 12 unambiguous status of his indication if we lost ICS/NNI power.
() 13 We then looked at this list of problems, and 14 potential solutions, and we utilized that to establish some of 15 our initial recommendations. We also utilized in our 16 recommendations the failure modes and effects analysis that 17 was performed as part of this program 18 Now to do the failure modes and effects analysis, we 19 took previous failure modes and effects analysis that was done 20 by B&W, There was a previous analysis that was done on 21 Davis-Besse specifically, and that was back in 1980, and we 22 updated that Davis-Besse analysis and established the 23 methodology from which we could then proceed to do the failure 24 modes and effects analysis for the other plants.
25 MR. MICHELSON: Question--on your FMEA, did they HERITAGE REPORTING CORPORATION -- (202)628-4888
226 1 reflect the observations that we have had on a couple of
() 2 occasions that when certain kinds of instrument power supplies 3 fail, some of the instruments go upscale, some go downscale, 4 some stay where it is, some start drifting? Is that scrt of 5 thing incorporated into a FMEA.so that we understand how 6 equipment behaves when say the power supply failure?
7 HR. STALTER: Yes, sir. What we did was we looked 8 at the failure modes and effects that would occur on failures, 9 of various kinds of failures, input, various failures of the 10 outputs, those things that we drove including indicators, 11 recorders, et cetera, and that encompassed failures that 12 occurred within the system.
13 HR. MICHELSON: How do you know--did you have to go 14 back to the particular vender like B&W instrument people and 15 find out how equipment behaves, or did you do some tests or 16 how do you know what the failure effects are?
17 HR. STALTER: We, we had previously developed a FMEA 18 that was performed by Science Applications, Incorporated, and 19 to do that, we utilized the original instruction manuals that 20 were part of the system.
21 Coupled with that there was a report that was done 22 by ORNL on the Oconee units which covered--I guess I should 23 point out here there are two different kinds or two different 24 ICS/NNI configurations. One that is, one is called a Bailey 25 820 system, the other, Bailey 721 system. Those HERITAGC PEPORTING CORPORATION -- (202)628-4888
l 227 1 differences--the ORNL report covered the Bailey 821 system. J 2 The SEIC efforts that we performed utilized the 820 plants 3 equipment.
l 4 MR. MICHELSON: Do those analyses reflect what
]
5 happens when the failure mode of the power supply is such as 6 to cause it to output voltage to increase above normal or to 7 decrease below normal, but not necessarily to go to zero?
8 MR. STALTER: Yes. j 9 MR. MICHELSON: Degraded voltage conditions, in 10 other words, are they reflected in the FMEA? And if so, how 11 do they get the data to know what the the degraded condition 12 causes?
13 MR. STALTER: One thing that occurs in the B&W pltat 14 820 system is that when the voltage starts to degrade, and 1 15 am talking about the DC voltage which is driving voltage, when 16 it starts to degrade, the power supply monitor which is there 17 to prevent degradation voltage takes over, and de-energizes 18 the system.
19 MR. MICHELSON: Is it possible to have single 20 failures that of one power supply such as it doesn't, it is 21 unable to take over and do that because it is a part of the 22 failure?
23 MR. STALTER: Single failures of the power supply 24 upon--
25 HR. MICHELSON: Within the power supply, such that HERITAGE REPORTING CORPORATION -- (202)628-4888
228 1 the monitor doesn't do'its thing? Is that possible?
W"T ;
(_/ 2 MP. . STALTER: The configuration of the power supply 3 s, there are--let me think about it a minute. Now in each 4 plant now, there are two DC power supplies that supply each, 5 each individual section of the ICS or the NNI. Now those two 6 power scpplies are grouped together using isolation diodes 7 such that if one power supply were to fail, the other one 8 automatically takes over. The power supply mor'toring 9 function is to look at the bus to see if there is a 10 degradstion f not just one, but more th+.1 one power supply.
11 'fR. MICHELSON Auctioneering tb7 two buses.
12 MR. STl.LTER : Auctioneering the two power supplies- ,
the other function of the power rapply monitor is to serve as
(]) 1. 3 14 cn alarming condition to tell you that one power supply may 15 aave failed, so it looks at the power supplies and it looks at 16 the buses. The power supplies fail, sets off alarm, lets the 4 7 operator know that. I think if the bus were to fail, it 18 de-energizes the bus so you don't get inadvertent controlled 19 action. You then go to a known condition, and that is 1
20 de-energized.
21 F7 che IP3A, we, then we then added additional i
22 re commen*. os, an:. ^- t .ok the FMEA a little bit.further and ;
23 did som>- i , "< *
- edictions that could result from ,
tailures c .
m;' of this system, and from that add this
{
25 ;; >'titional te., .nendations.
HERITAGE REPORTING CORPORATION -- (202)628-4888
229 1 MR. MICHELSON: You did not consider anything like
() 2 earthquake response on the part of your study?
3 MR. STALTER: We did not consider suismic events.
4 MR. MICHELSON: External events like the elevated 5 temperature in the cabinet, was that considered?
6 MR. STALTER: Elevated temperature in the cabinet .
7 was, was one of the design requirements, yes.
8 MR. MICHELSON: I am talking about beyond design.
9 MR. STALTER: Beyond design, if you get a fire or 10 something like that--
11 HR. MICHELSON: No, no. Like if poor circulation 12 fails or the natural circulation, or if there are certain
- 13 components that are not getting adequately cooled tor whatever 14 reason, in other words, it causes more than one failure at a 15 time. Did you consider elevated temperature? Because it may 16 be that the room has lost its cooling and now the cabinet sees 17 an elevated temperature.
18 Did you consider that sort of '.hing?
19 MR. STALTER: If I recall the FMEA--Phil, m11be you 20 can help me. I think the FMEA did not consider that.
21 MR. LIDDLE: That did not include that, but at the 22 environment where these cabinets are located in plants, 23 typically is the same where the safety cabinets are.
2 '4 MR. MICHELSON: Typically that's a chilled w ter 20 system keeping the room cool.
HERITAGE REPORTING CORPORATION -- (202)628-4888
230
'l MR. LIDDLE: That's the case. The RPS is--
2 MR. MICHELSON: The problem is what'happens when 3 this one cabinet goes? What kind of unwanted action. running 4 around the plant are you going to get?
5 MR. lit"LE; J There'are alarms for things that 6 indicate when the failure is, but that was not considered.. !
7 MR. MICHELSON: FMEA does not consider thi change in r
8 the environment of that' sort'of thing. Okay. Thank you.
9 MR. STALTER: in addition to that, our recommended l
10 solutions also had inputs from the sensitivity study, and 11 those were previously mentioned, and we will talk about them 12 again. And they, we looked at equipment obsolescence. We 13 looked at aging. We looked at what kind of support was out 14 there that was available for this equipment that we.have in I
15 the plant, and what kind of availability will it be in the i J
16 near future? And those all went into our recommendations, i 17 recommended solutions.
18 HR. MICHELSON: You said support. What did you l i
19 mean? Maybe I misunderstood. What kind of support?
j 20 HR. STALTER: The vender will be there if we need ;
i 21 suppcet. .
i j 22 MR. MICHELSON: How about the support systems 23 required for this equipment to function properl2 such as room i
1 24 cooling? Was that 2ooked into at all?
() 25 MR. STALTER: We did not get into room cooling.
HERITAGE REPORTING CORPORATION -- (202)628-4888
231 1 CHAIRMAN WYLIE: What is the rating of the equipment g3 2 as far as temperature goes? Are these forced cooled cabinets?
G 3 MR. STALTER: Forced cooled cabinets; there is fans 4 in there. They are designed to operate'I believe at 120, 5 forty degrees 0; 122 degrees is the design.
6 CHAIRMAN WYLIE: That's their design.
7 MR. MICHELSON: Is that the air in the room, that 8 120?
9 MR. STALTER: Ambient conditions.
10 MR. MICHELSON: All right. And you have got forced 11 circulation in the cabinet besides?
12 MR. LIDDLE: Yes.
13 MR. STALTER: That was the input that we had. We 14 had some follow-on activities that are ongeing now. The-15 output of that, we needed some additional look at our known 16 safe state that we, that I will get into that in a little bit.
17 We needed some additional work that we outlined in B&W 1919 on 18 the definition of how the ICS/NNI was used in FSAR 19 assumptions. That works on normal.
20 HR. MICHEL?ON: On your FMEA, these, these systems l 21 are powered off the vital bus I guess, four different vital l
22 buses, is that correct? l 23 HR. STALTER: I don't know what, your term vital i
24 bus. They are uninterrupted power sources, typically battery
( 25 backed.
l HERITAGE REPORTING CORPORATION -- (202)628-4888
232 1 MR. MICHELSON: Running off a converter probably.
( 2 MR. STALTER: Thr* is correct.
3 MR. MICHELSOf t: Now if the invsrter goes to full off 4 voltage, was that considered in the FMEA? Your power supply 5 voltage, one single failure of the power supply would cause 6 that.
7 MR. LIDDLE: Individual power supply failures were 7 8 considered.
9 HR. MICHELSON: Considered in the FMEA?
10 MR. LIDDLE: Losses of power supply were. I'm not 11 sure if increased voltage was.
22 MR. MICHELSON: Loss, I don't have too much problem
(} 13 where it is increase or degradation of the voltage condition, 14 that to some extent the equipment can accommodate to it, but 15 not, not necessarily if it goes say to 130 when it is 125.
16 You are not sure then on that?
17 MR. STALTER: In the case of Bailey 820 system 18 design, if you have individual inverter, it would go to very 19 high voltage. Let's say it would cause a condition where that 4
20 power supply cannot function, then the other power supply will 21 take over and perforin.
22 MR. MICHELSON: This is not in the Bailey equipment?
23 This is DC system in the building. Now in terms--I asrume you 24 are getting 125 DC as the feed?
25 MR. LIDDLE: No--118 ACR.
HERITAGE REPORTING CORPORATION -- (202)628-4886 m
~ .
233 1 MR. MICHELSON: Excuse me. And that is what goes up 2 to a 130, forty, and that has to be somehow accommodated by 3 the Bailey equipment.
4 MR. LIDDLE: The Bailey system has requirements on 5 what we can allow the AC to vary. That's put on the plant 6 requirements.
7 MR. MICHELSON: That is a range. If it quits 8 completely, it gets out of range, does it just--
9 MR. LIDDLE: The power supplies will shut down.
10 MR. MICHELSON: Okay.
11 MR. STALTER: We saw in that evaluation the need to 12 look into, because of equipment obsolescence, and because of
() 13 design features we would like to see, we saw a need to look at 14 advanced control, e.nd that's why advanced control system was a 15 follow-on effort and that is an ongoing effort and I will get 16 into that in u little while.
17 MR. MICHELSON: Present ICS non-service?
18 HR. STALTER: That is correct. It is a non-safety
- 19 system. We visualize that any replacement system that we 20 would have f or the ICS would also be a non-saf ety system.
21 HR. MICHELSON: Not necessarily single crack in 4
22 terms of power supplies to it and that sort of thing, is that 23 right, or will you have a single power supply to all?
]
24 MR. LIDDLE: ICS presently doer not have single J 25 power supply. Has redundant AC and redundant DC, and i
l HERITAGE REPORTING CORPORATION -- (202)628-4388
234 1 replacement system would be as good or better. !
() 2 MR. MICHELSON: You are feeding it oft four 3 different trains, aren't you?
4 MR. LIDDLE: On the back; on the back, yes.
5 MR. STALTER: I will get into some of the features 6 of the ACS toward the end of the discussion.
7 (Slide) 8 MR, STALTER: So to summarize a little bit here, we 9 utilized previous work. We brought the documents up to date.
10 We established the requirements and those requirements are 11 with the given technology that we had installed in the plant.
12 We made comparisons between the requirements and what our 13 actual plant configuration was, formulated the recommendation, 14 and the people involved in this process were utility INO 15 engineers, operation personnel. We wanted to be sure to have 16 operator input into particularly some of the recommendations.
17 Maintenance personnel, whau it came to maintaining it, those 13 people were the ones that provided us the best input on those, 19 those kinds of recommendations.
20 We had B&W's controls engineers and specialists with 21 us, and we had some industry experts and particularly 20 informed, we took a lot of advice from the independent j l
23 advisory board, the SPIP, and we also had EPRI and different i 24 consultants.
25 (Slide)
HERITAGE REPORTING CORPORATION -- (202)628-4888
235 i
1 MR. STALTER: The magnitude of the evaluation, and I
() 2 guess I put this slide up here to indicate that this was a 3 major project because it had a direct expenditure of over 4 $500,000. The indirect was more than that, and it spanned a 5 period of fourteen months following, well, it actually began 6 in early '86.
7 It included elet'en I&C committee meetings, and these 8 were working meetings, four working level meetings at the NRC 9 where we updated them on the progreso, and presentation to six 10 independent advisory board meetings.
11 The output was published in six major documents, 26 12 supplementary documents, and they are now Appendix R at 1919.
13 Key here was that there was over 450 different ideas
~
14 and suggestions and references or recommendations that we 15 looked at. This was a large, large number for us to deal 16 with, and each suggestion or reference had to be 17 dispositioned, and the effort to do that was a big effort. We 18 had to 'dentify each and every one of them. We developed a 19 matrix t- make sure we didn't miss anything. We went back and 20 re-reviewed the documents, and in the end when we established 23 our recommendations we feel that we were very, very complete.
22 MR. MICHELSON: Let me ask you something on the 23 history, indicate why I am a little bothered.
24 Back when the light bulb occurred at Rancho Seco, l
, 25 that was a surprise. People hadn't thought through this HERITAGd REPORTING CORPORATION -- (202)628-4888
236 1 situation so well, but they learned a lot and they presumably 2 fixed the plant, yet when the next Rancho Seco event occurred, 3 there was more surprises again that they didn't fix things the 4 first time or learn more or just what happened?
5 See, that 1985 event, the whole lot, I had just i 6 assumed in my mind everything was fixed and I saw a lot of 7 what looked very familiar in 1985, and what happened? Why was 8 the thing so familiar? Was it not yet fixed or not really 9 learned the first time or--
10 MR. RUTHERFORD: I think for the rost part Rancho 11 Seco had not installed systems that they intended to install, 12 that would have fixed--
13 MR MICHELSON: They understood their shortcomings O 14 but hadn't got around to, had not fixed them? '
15 MR. RUTHERFORD: I won't say they understood 16 everything, but the systems that they intended to install 17 would have taken care of some of the proolems that cropped up.
18 MR. MICHELSON: They were going to do it later or 19 just never intended--
t 20 MR. RUTHERFORD: They were going to install the 21 equipment.
22 MR. MICHELSON: That was quite a long time in 23 history between the two events, but apparently it moved along 24 slowly.
() 25 MR. RUTHERFORD: That's true. I HERITAGE REPORTING CORPORATION -- (202)628-4888 2
i 237 1 MR. MICHELSON: So it is not like I should really 2 believe that you can understand these things. That led me to l 1
3 believe we didn't understand the first event, just because I 4 didn't know the histories, what happened.
5 MR. STALTER: Dealing with these many, many ideas l
6 and references or dispositioning those, the task became a 7 great one, and we decided to do it in this manner. We chose--
8 MR. MICHELSON: One other question on this old 9 Rancho Seco event, the light bulb one, and that is that you 10 are not looking at axternal events, but one of the things that <
11 concerned me because I have seen more than one LER relating to 12 it, that is if you get some water in those backlighted
() 13 switches, the kind that have the bulb in, you can also get i 14 some exciting things happen because it shorts out some power 15 supplies and so forth. !
16 Is there something done to make sure that they don't l
17 drink coffee over the bench boards in the control room, that !
i 18 sort of thing? That's how you get the slight dropping--the ,
19 light bulb, you drop a cup of coffee on the lighted switches, 20 and things go into scintilating modes. Is that part of FMEA 21 then, or did you not consider water because that is an I 22 external event? '
\
23 HR. STALTER: I can't address that specifically, but 24 I can tell you there is one of our recomrendations that would 25 help along those lines, and that is that we have a !
i 6
HERITAGE REPORTING CORPORATION -- (202)628-4888
238 1 recommendation that asked that we fuse all signals leaving the [
() 2 cabinet.
3 MR. MICHELSON: That would help.
4 MR. STALTER: If we fuse those and we get a short in 5 one of those, all we do is lose that individual recorder, 6 indicator, or auto station, whatever it might be, and not the 7 whole system.
] 8 MR. MICHELSON: As far as you know, in B&W plants do 9 you allow coffee over the bench boards in the control room? $
10 Any prohibition against drinking coffee while on duty? )
1 1
11 MR. RUTHERFORD: I can't speak for them.
12 MR. MICHELSON: There is a standard going through we l l
13 heard about the other day about not going to allow them to O 34 eat.
]
i I
I 15 MR. DAVIS: You have got the problem cf sleeping. l 4
16 I MR. MICHELSON: That's another problem. That coffee i i
17 cup is a real bad thing with the B&W design. Not many other i 1
1
! 18 people use those switches big this system uses them l I
19 exteasively. l l
20 MR. RUTHERFORD: The FMEA doesn't necessarily care l l
21 about where the source is coming from.
22 MR. MICHELSON: It does it one at a time. If coffee l 1 l 23 hits several switches at once, I don't think the FMEA goes in 24 and faults several things at once. I think it is going i O 25 through one at a time.
HERITAGE REPORTING CORPORATION -- (202)628-4688
239 1 MR. LIDDLE: Single input and power distribution.
2 MR. MICHELSON: Enough said. At any rate, watch the 3 coffee cup. It really is an interesting source on some of 4 those backlighted switches.
5 MR. STALTER: We categorized these recommendations, 6 classified them into three levels, and those three levels are 7 stated in this slide, the first being those items which will 8 provide immediate improvements to the operation availability 9 and reliability of the ICS/NNI systems.
10 Typically now the RTS handles, contains 51 11 recommendations, and I think at one time somebody said 53, but 12 I believe there are a couple that have superseded that are 13 related to the ICS/NNI, and these will in fall into this level
[}
14 one category.
15 Now there are three key areas that these provide 16 improvement to. We improved the reliability of the power 17 sources. We look at hardening the inputs to reduce the 18 probability of a trip or transient, but what I mean by harden I 19 is that we want to make sure that that input, if it fails, we l
. 20 won't have a trip or transient.
21 And the third one being if we lose power to these ,
l 22 systems, we want to be able to achieve a known safe state, and 23 the next, the next slide, I am just going to move on to the j 24 next slide for a minute because it talks about all of the 25 level one recommendations, or not all of them, but typically J
HERITAGE REPORTING CORPORATION -- (202)628-4888
j 240 1 example of the level one recommendations that apply to those )
O 2 three categories.
1 3 And you asked about failure of the light bulk, and 4 we said we are going to fuse an auto station. We are going to 5 do things, install current and voltage meters so we can 6 monitor the power supplies. There is a case where they don't ,
7 exf.st. In the case of the 721 design, we are going to supply 8 the hand power which is one sort on the auto power circuits 9 from separate panels where we are now. We found in this 10 review that they are supplied by one; 182 you here the AB, 11 automatic bus transfer for main feed pump controllers. We are 12 suggesting there that gee, we can protect the ICS. Here is a
() 13 vulnerable case where the device we are trying to control, if 14 we lose the power that is Gupplying the ICS, it can knock out 15 all of our feed pumps.
16 Fuse all power leaving the cabinet, that these two !
17 really kind of go together. We wanted to, we have one case 18 where the NNI has a wide DC power where only one power supply ]
19 is supplying that, and we are, and this recommendation is 20 saying let's put in the second DC power supply in that area.
21 The power supply monitor is the result of the SMUD l 22 event. The power supply monitor we have found was monitoring ;
23 power in the wrong place. It was not in accordance with the 24 drawings at most of the 80 plants, and this recommendation )
25 says make sure that it is installed in accordance with the HERITAGE REPORTING CORPORATION -- (202)628-4888
m 241 1 drawings.
2 MR. MICHELSON: Wnen you did this FMEA, did you find 3 that there were any single failures which would cause the flow 4 control valves, main feedwater flow control to go wide open, 5 for instance, and overfeed the generator, suggesting failure d that did that in the present ICS system?
7 MR. STALTER: I am going to--Phil, I believe we did.
i 8 MR. LIDDLE: There are some things that could cause 9 that to happen.
10 MR. MICHELSON: I wouldn't be surprised. So you 11 haven't done anything about them yet, the overfill question?
12 MR. STALTER: We are going, we are addressing that 13 in the known safe state iusue on loss of power.
14 MR. MICHELSON: Okay. That's where you are going to 15 handle it.
16 MR, LIDDLE: Plus the input side, we are handling 17 that, so it--
18 AR. MI dELSON: Is there some kind of role in your 19 mind that says that no aingle failure should cause steam 20 generators to start overfilling, or who you write a set of 21 rules like that, how do you decide what fixes to make, how 22 much vulnerability to tolerate, what criteria, in other words, 23 to use?
24 MR. STALTER: I'm sorry?
() 25 MR. MICHELSON: What criteria are you using to HERITAGE REPORTING CORPORATION -- (202)628-4888
242 1 decido whether or not maybe something needs to be fixed? ;
(} 2 MR. STALTER: This was primarily engineering 3 judgment in looking at what we came up with in the 4 recommendation area. We are saying these are the 'reas that 5 we need to take care of, and it was a judgment of the, those 6 people that were involved in evaluation. '
7 MR. MICHELSON: Maybe I missed it then, or maybe it 8 just isn't a ecmplete list, but for instance--
9 MR. STALTER: This is an example. This is not a 10 complete list.
11 MR. MICHELSON: You did something to i:. prove the 12 overfill of steam generators?
13 HR. LIDDLE: In terms of what we get from the ICS, 14 yes.
15 HR. MICHELSON: ICS is one of the sources? There 4
16 are other sources of overfill causes, but the ICS is one of l
17 them?
18 MR. LIDDLE: Aight.
19 M*. DAVIS: Are these improvements primarily for
, 2C safety considerations or reliability considerations?
21 MR. STALTER: These are for reliability, 22 availability considerations. These do not, these are, do not 23 affect the ability of the plant safety systems to perform 24 their functions.
25 MR. DAVIS: Don't they decrease challenges to the HERITAGE REPORTING CORFORATION -- (202)628-4888
243 1 safety systems, thereby--
2 MR. STALTER: Certainly they do.
3 MR. DAVIS: That would improve safety?
4 MR. STALTER: Yes, it would in that sense.
5 MR. MICHELSON: Steam generator overfill on a B&W is 6 net presently safety grade, is it? Or is it?
7 MR. LIDDLE: No.
8 MR. MICHELSON: I don't think you have any 9 provisions for overfill presently, although that is an 4 '
10 engineering issue we recently reached a tentative resolution 11 on. I think right now you, ICS is not, it is non-safety, but 12 it could cause plant to get into some potential safety 13 difficulties like overfilling the generator and starting to go O 14 into main steam lines.
l 15 MR. REED: On the reactor coolant flow, is that ;
[
16 replacing the flow because it is too slow acting, or-- ;
17 MR. STALTER: I was going to get into this, into l
18 this particular area right in here. This is the area where we i 19 are talking about hardening the inputa, and we had some very 20 early recommendations in the program. As a matter vf fact, J
j 21 these are recommendations that preceded SPIP where we tried to 22 improve the ICS, and some of these are--not all of them, but 23 we said that, we looked at the failures that occurred of the 24 reactor coolant flow signal going to the ICS. Now thia O
is 25 resulted because the signal going to the ICS came out of one l
l HERITAGE REPORTING CORPORATION -- (202)628-4888
244 1 single RPS cabinet, and if for any reason the power was lost
() 2 to a single RPS cabinet, we lost the flow signal that went to 3 the ICS, so we said let's reduce the probability of having 4 that failure, cause a plant transient, so we said all right, 4
5 we will do that by eliminating the flow signal and replacing 6 that with a pump status circuit that corrects the delta TC '
7 circuits in this case, in the event that you lose one reacter 8 coolant pump, so it has one signal that is in there for normal 9 operation, four pump operation, and transfers in a signal on 10 the reactor coolant pump.
1 i 11 MR. REED: It seems to me there has always been sort 12 of argument against using a simple input like voltage and i
13 amperage or pump status I guess come from those two 14 parameters, not just voltage.
15 MR. STALTER: It is analogue signal, combination of 16 analogue and contacts.
17 MR. REED: Because flow is really the parameter. I
?
i 18 mean if I broke a shaft, the pump will have some changes in l
l 19 volts and amps I expect or particularly amperage, not, it will 20 not really be directly coupled to flow, so really what--you {
21 think that's a good signal? The status of the pump versus the
, 22 actual flow?
23 MR. STALTER: Yes. We do feel that that Jr, a good I 24 signal to use.
O- 25 CHAIRMAN WYLIE: Is the natural contact off the i
HERITAGE REPORTING CORPORATION -- (202)628-4888
245 1 breaker or--
( 2 MR. STALTER: Breaker contact.
3 CHAIRMAN WYLIE: Okay. Just says it is running.
4 MR. REED: It says the breaker is closed.
5 CHAIRMAN WYLIE: Yes.
6 MR. REED: And the shaft is broken.
7 CHAIRMAN WYLIE: Analogue signal, is that the power 8 or the current signal or--I believe you mentioned analogue 9 signal.
10 MR. STALTER: RC flow? RC flow was an analogue 11 signal. We relayed that.
12 CHAIRMAN WYLIE: Just a motor breaker contact?
(} 13 MR. STALTER: With the contact, yes, plus a voltage 14 because we ncno t59 voltage for the analogue control within 15 the ICS.
16 MR. LIDDLE: Just a signal generator representing 17 the flow.
18 MR. STALTER: Represents the flow for that.
19 MR. LIDDLE: Or some plants actually--
20 MR. STALTER: There is another way to take care of 21 this problem. This was, like I say, this was an early 22 recommendation. At the time that this recommendation was 23 made, there was no smart analogue signal selection capability 24 designed or available to it for use, so we couldn't really at 25 that time say okay, let's look at these signals and see if ws HERITAGE REPORTING CORPORATION -- (202)628-4888
246 1 can really make a good selection of signals.
() 2 Now further down the road, as we got into thet 3 evaluation, there became available from a number of venders, 1
4 from two venders in particular, a means to do smart analogue 5 signal selection or smart automatic signal selection. That 6 could be substituted for this, and it would work fine. We 7 could use the RC flow signal. In a case if we got failure, it 8 would automatically transfer to the one that was good, 9 provided it was implemented properly, and would prevent any 30 plant transients of that single transmitter failure.
11 CHAIRMAN WYLIE: What was wrong with the RC flow 12 signal?
{} 13 HR. STALTER: Primarily we were experiencing two 14 things. One was a single failure of transnitter that was j 15 selected, and these were hard wire selected, would cause the ;
16 ICS to initiate a plant transient if that signal failed, and )
i 17 the second part of that was if the RPS channel from which that l l
4 18 signal came, if 4ts power supply were de-enargized for some l
\
19 reason, it would also cause a plant transient through the ICS, l
l 20 so that the thrust of this was to prevent those plant l I
21 transients. l l l 22 HR. MICHELCON: Was that elbow tap that you were 1
23 using for flow? '
l 24 HR. STALTER: No. This, there is a flow element O 25 that is in the hot leg.
l l
i HERITAGE REPORTING CORPORATION -- (202)628-4888
247 1 MR. MICHELSON: Actually you have to take it right
( 2 off the flow element. That is much more positive, of course, 3 pump status--could be rather ambiguous. ,
4 MR. STALTER: We agree that it is more positive and 5 that's the reason it was part of the original design. ,
t 6 However, the effect of using pump status in aur view is 7 satisfactory.
8 MR. LIDDLE: For the uses of what this RC flow i
9 signaling is used for, the pump status is equally valid.
10 HR. MICHELSON: Good enough; it isn't quite the l l
11 same? j
- 12 MR. LIDDLE: That is correct.
(} 13 MR. MICHELSON: Okay. Thank you.
14 MR. STALTER: The thrust of each of these others 15 was, was a single failure would cause the ICS to initiate a 16 transient of some sort. Those were early recommendations, and 17 really they could all be encompassed into what we now have as 18 this one, and that's to do auto selection of valid inputs.
19 And we now have a technological improvements available to us ,
20 for, available to us for our use in affecting that. It is 21 being installed, has been installed at one plant, one unit, is
- 22 being installed in a number of others at this time.
23 MR. CATTON: Are you going to go back and take r
24 another look at one?
O- .
25 MR. STALTER: As a part of the installation of this, I
HERITAGE REPORTING CORPORATION -- (202)628-4888
- ~_ . - . - _
248 1 we will, all of these will be considered. We may elect to
( 2 leave the pump status in place, for instance, because of its 3 perfectly performing its function, there is no reason to 4 change it.
5 MR. CATTON: But somehow it seems like a step in the 6 wrong di.*ection. The flow is the primary variable.
7 CHAIRMAN WYLIE: How do you detect a broken shaft?
8 MR. LIDDLE: ICS doesn't care under those 9 conditions.
10 CHAIRMAN WYLIE: I know, but how do you detect it? ,
! 11 MR. LIDDLE: RCS trip; it has happened in Crystal 12 River.
(} 13 MR. STALTER: Get a flux to flow trip in the ICS or 14 in the RCP. Excuse me.
15 CHAIRMAN WYLIE: Oh, okay.
t 16 HR. STALTER: The last area that we categorized 1
17 generally these recommendations into was known as safe state.
18 We have one recommendation that is known safe state, and that 19 is that we achieve a known safe state, balance, heat balance 20 the plant, normal operator actions following a loss of either 21 NNI power or ICS power.
22 And now this could be different things at different 23 plants. If we implement auto selection of the valid inputs 24 properly, when we get down to this known safe state effort, we 25 want to be able to lose a section of the NNI, lose that power, HERITAGE REPORTING CORPORATION -- (202)628-4888
249 1 and continue to operate. If we lose the 7.CS, we will end up 2 in, may end up in a trip condition, but if we do end up in a 3 trip condition, we come to a state where the heat balance is 4 maintained by the operator, and we don't get into severe 5 overcooling, overheating transients.
6 These other recommendations all help in that regard.
7 One other thing to point out here is one of the other 8 technological areas that we made some improvements was to 9 provide thic unambiguous status of indicators to the 10 operators. We took upon a project to develop a digital 11 indicator that could replace the existing indicator and 12 control panel that would provide the operator with an
() 13 immediate status so he could lose power that was supplying 14 that indicator. It would, if you lost the AC power, it would 15 be black, totally black. You would know it right away. You 16 wouldn't try to read it. If the signal were lost to that 17 indicator and went to mid-scale, there would be an indication 18 on it to let him know that was at mid-scale, that it was a, 19 may not be a valid reading, he shouldn't rely on that.
20 MR. MICHELSON: I notice you had pressurized heater 21 interlocks down there.
22 Are the heaters such that if you, if you were to 23 lose level and fail to interrupt the heaters, that when they 24 melt down, which I guess they will do next, won't they?
}
25 M '< . STALTER: The problem here was on the HERITAGE REPORTING CORPORATION -- (202)628-4888
250 1 pressurizer interlocks, that if we lost power to the cabinet--
1
() 2 HR. MICHELSON: The control. ,
3 HR. STALTER: NNI in particular, that we lost the 4 power to--there is an interlock in that cabinet that is l
5 utilized to prevent you from energizing the heaters. If you ,
6 uncover them, then that interlock actuates the way where it !
7 won't let's you use the heaters even though they are not l 8 uncovered. ;
9 HR. MICHELSON: You have, in essence you lose the 10 ability to apply heat to pressurizer?
11 HR. STALTER: Right.
12 HR. MICHELSON: How about the other?
13 HR. LIDDLE: Just certain banks of heaters, to make
{
14 it clear.
j 15 HR. MICHELSON: Yes. Each NNI has certain number of
.i 16 heaters on it. Each has the--
. 17 HR. LIDDLE: It is actually separate controls from ;
18 the NNI.
19 HR. HICHELSON: What prevents the pressurizer i 20 heaters presently from being powered when the level has 21 dropped below this cabinet? Will a single failure cause the ,
22 heater to burn out then?
23 HR. LIDDLE: it would prevent the interlock from
- 24 occurring, that is correct.
25 HR. MICHELSON: The heater would burn out in that i I
i 1
HERITAGE REPORTING CORPORATION -- (202)628-4888
l 251 1 case. .
() 2 Are your failure modes of your heaters such that 3 don't .Tose the pressure boundary when you melt the heater out? l l
4 Is that a design basis, that you should not lose the primary ,
4 i
. l 5 pressure boundary when the heater burns out? It will melt the i 6 sheet. You have lost the boundary on the heater, but you are j 1
7 using that type of rods that won't blow out.
i 8 MR. GANTHER: It is the care predicts--
9 MR. MICHELSON: You don't really want to blow the 10 heaters out. That's a primary LOCA.
11 MR. TAYLOR: The heater is quite a distance away l
12 from the pressure.
13 MR. MICHELSON: It is a design tasis, though, 14 requirement that you melt the sheet without blowing out the 15 heater. '
l 16 MR. GANTHER: That's the basis. I don't know if it .
I J
l 17 is design basis or not.
18 MR. MICHELSON: Thank you.
1 19 MR. CATTON: Is the ICS computer based?
20 MR. STALTER: The ICS is an analogue computer. It i
21 is not a digital computer.
22 MR. CATTON: Do you know what happens to it if it 23 starts to wara up?
24 MR. STALTER: It starts to warm up, and you mean the 25 temperature of the components heat up?
HERITAGE REPORTING CORPORATION -- (202)628-4888
252 1 MR. CATTON: Yes.
2 MR, STALTER: We know what happens to it up to the 3 design temperature. Beyond that, we have never taken it to a 3
4 failure in the sense that we have gone beyond the design 5 requirements for that.
6 That was the essence of our level 1 recommendations.
. 7 We had two other categories, level 2 and level 3, level 2 ;
8 recommendations being those which potentially involve major 9 modifications to the existing equipment, and further in-depth t i
10 evaluation needs to be accomplished before we approve these, 11 and we felt that at this point the continuing evaluation of i 12 these would require some long-term commitment to the existing O 13 a ra re-14 The level 3 recommendations are those items which 15 consist of replacement of the existing system. They would be 16 much more comprehensive, and that a new, new system based on l 17 modern digital technology, technology would be utilized, and l
13 an advanced control system task force was organized to 19 establish, and is currently pursuing the development of an 20 advanced control system. l 21 MR. MICHELSoli: Is that for replacement of the l
22 present ones or-- I 23 MR. STALTER: The intent would be it would replace ,
24 the present ICS/NNI, yes.
25 MR. .4ICHELSON: At some point down the road.
)
5 HERITAGE REPORTING CORPORATION -- (2021628-4888
253 1 MR. STALTER: Intention was to have it available.
2 If a utility chose to put that system in, we would have that 3 available for use. I have got a slide further or down the 4 road that talks about what, some of the details of that.
5 (Slide) f 6 MR. STALTER: These are recommendations that we i
t 7 classified in level 2 and level 3, and if you look at these 1
8 recommendations, many of these are ones that came out of the j 9 sensitivity study, and that is to segregate the system, l
f 10 separate the sub-systems, control main feed pump on delta P, l
! l '. separate the controls from the main feed pumps and main 12 feedwater valves.
13 The elimination of the mid-scale failure was O 14 classified as level 2 and certainly the elimination of 15 mid-scale failure would be taken care of in a level 3. Now
]
16 mid-scale failures occur because we have a bipolar design. It 17 is a plus to minus ten volts system, and in order to take care
{
18 of that problem, it would be a very extensive modification to 19 the existing system.
20 MR. MICHELSON: Do all of them fail on voltage to a 21 fixed position or can they drift around?
22 MR. STALTER: If you lose voltage to them, we know 23 where they fail to.
24 MR. MICHELSON: To fixed system?
- 25 MR. STALTER: Yes.
HERITAGE REPORTING CORPORATION -- (202)628-4888
254 1 MR. MICHELSON: I got the impression one time that
- 2 the problem was instrument drift, failed to supply voltage.
3 That must have been folklore.
4 MR. STALTER: That was the major thrust of our known 5 safe state is that we knew where that failure occurs and where 6 it goes.
J 7 (Slide) 8 MR, STALTER: That pretty much concluded the look at 9 the evaluation. I now want to get into a little bit about our 10 review of the SER, and we did a detailed review of the 11 Supplement 1, Section 6 on the ICS/NNI, and the NRC generally 12 communicates overall agreement with the ICS/NNI evaluation.
13 They mentioned some 170 individual recommendations, 14 suggestions t.nd ideas, and generally had agreement over 90 15 percent of those.
16 There were some areas of disagreement and I want to 17 go ahead and address those areas, and we generally categorized 18 those in three areas. Those't'st involved the process of the 19 evaluation--
20 DR. KERR: I am not quite sure I know who is doing 21 the reviewing and what is being reviewed according to what is 22 on that slide.
23 MR. STALTER: Okay.
24 DR. KERR: Review of the SER supplement was
() 25 completed by whom?
o HERITAGE REPORTING CORPORATION -- (202)628-4888
255 1 MR. STALTER: This was completed by the IC 2 Committee, B&W owners group IC Committee.
3 DR. KERR: At what point did the NRC communicate 4 overall agreement?
5 MR. STALTER: Within the SER, they generally 6 concurred with our evaluation; and this is--
7 DR. KERR: Over 90 percent of what?
8 MR. STALTER: Over 90 percent of these 170 areas 9 that we defined they agreed with.
10 DR. KERR: Okay. Thank you.
11 MR. MICHELSON: Is the staff going to make a 12 presentation on this a little bit later today? Thank you.
13 MR. STALTER: With regard to the evaluation process, ,
('w/')
14 the, they had some disagreements in the area of our F
15 requirements documents. Generally these involved in whether l 16 or not we were going to maintain them for future, in the 17 future, and we felt that we established these requirements 18 documents for the purposes of this study and we really didn't 19 have any intent to maintain them as future references.
1 20 DR. KERR: What is meant by maintaining documents?
21 MR. STALTER: That means any time that we made a !
22 modification to the system, that we would update this :
23 document, this matrix. We intend to keep our individual plant 24 records up to date as to--
25 DR. KERR: The requirements document says what you ,
h HERITJGE REPORTING CORPORATION -- (202)628-4888
256 1 are going to do. It doesn't say you have done it, or does the 2 staff want to have one column saying we are going to do this
) 3 and another saying we have done it?
4 MR. STALTER: We had configuration matrices also 5 that were part of this process and the configuration matrices 6 are really the ones that would need updating, and that's when 7 we implemented recommendation or made some other change to the 8 system that we deemed necessary.
9 In the area of the failure modes and effects 10 analysis, we limited the scope of that in that we did not 11 consider operational errors. NRC comment was that we should 12 have considered operational error as that occurred in addition
() 13 to the failure modes. We limited it to the system failure 14 modes and effects analysis, and not the operational aspects of 15 it.
16 MR. MICHELSON: If I understand what you told me a 17 little earlier, you did not include the effect of room cooling 18 on this equipment? Was that a correct understanding?
19 MR. STALTER: That is correct. We did not.
20 HR. MICHELSON: What puzzles me is FMEA doesn't 21 consider that when it does consider loss of supply voltage to 22 the equipment and so forth, but not room cooling, which is 23 equally important. In fact, it may leap to more confusion on 24 the part of the operator board than say loss of power.
25 Is there some justification on why you think that HERITAGE REPORTING CORPORATION -- (202)628-4888
257 1 that essential service is not necessary to be considered?
( 2 MR. STALTER: I did not know of any room cooling '
3 events that have occurred that have affected the operation of 4 the ICS.
5 MR. MICHELSON: I can tell you at least one. The
. 6 McGuire event got one whole train of equipment, elevated 7 temperature. All that elevated was what, 90 degrees, Charlie?
4 8 Was that right?
9 MR. LIDDLE: I think the operator is going to know l
10 before it gets beyond the design temperatures. This 11 equipment--that's the thing. The design temperature is 120 12 degrees, they are going to be taking action.
(} 13 MR. MICHELSON: You are sure that's the room 14 temperature and not the--
15 MR. LIDDLE: Not the internal temperature, that is 16 correct.
17 MR. MICHELSON: That gives them a little more 18 leeway, but does he know what to expect or is he just going to 19 shut all the equipment down?
l 40 MR. LIDDLE: That is correct. He would have to shut I
21 it down. '
22 MR. MICHELSON: Not performing any essential 23 service; he would just kill the power to it? ,
l 24 MR. LIDDLE: Well, there is more control hopefully.
O 25 j
i MR. MICHELSON: Better kill the power; the power !
i HERITAGE REPORTING CORPORATION -- (202)628-4888
258
't 1 starts the heat in the cabinet. -
() 2 CHAIRMAN WYLIE: Open the door, put a fan on it.
3 MR. MICHELSON: If you can, yes.
4 MR. STALTER: In the area of IE Bulletin 79-27, 5 which the NRC hit a little hard on us, we did consider 79-27 6 in our evaluation. What we did not do was to specifically 7 document each plant's response to that bulletin and how they 8 dealt with it.
9 When we looked at the up-to-date configuration of 10 each plant, they had included in there many that we had 11 included, many of those areas that were resolved by IE 12 bulletin 79-27, and others in 79-27 could not be resolved on a 13 generic basis, and needed a plant-specific evaluation, and so 14 we suggested at that point that, that they look at the 15 plant-specific responses and not rely on the owners.
16 They were, there were some areas where--
17 DR. KERR: Are they, you convinced they have been 18 satisfied?
19 MR. STALTER: Had we provided--I don't think we 20 could have provided.
! 21 DR. KERR: No. You said yes, that they did look at j
22 plant-specific responses. The implication is that, to me that 23 if they locked at the plant-specific responses, they would be 24 happy.
25 MR. STALTER: No. I think you need to do not only 4
HERITAGE REPORTING CORPORATION -- (202)628-4888
259 1 plant-specific response review, but also you need to go to the 2 plant.and see what was done. I think if you did those two 4
3 things, you could probably satisfy yourself.
4 DR. KERR: Are you convinced they would be 5 satisfied? I mean have the thing--are you convinced that the 6 things that are required by IEB 79-27 have been done? '
7 MR, STALTER: I am convinced that there was a good 8 response by each of the utilities to IE 79-27. ilow whether 9 every single item in there was accomplished, I guess I can't l
] 10 answer that.
1 J 11 DR. KERR: You just have a good feeling about it but l 12 aren't sure whether the staff would be, if they followed your 1 13 advice, would be happy?
( 14 MR. STALTER: That is correct.
15 DR. KERR: Thank you. We had some recommendation t
16 that they deemed that we lacked references to sources and we "
i 17 have gone back in a couple of areas and looked at some of the '
l 18 recommendations, and have even made modifications to the 19 recommendations, including new source references, and I think 20 that was appropriate. !
21 The other area was the previously identified l 22 concerns. There was a list of previously identified concerns !
23 that was provided to us in a response for additional '
J 24 information which we responded to. That included the IE ,
() 25 BULLETIN 79-27 and other references dating clear back to 1978. l HERITAGE REPORTING CORPORATION -- (202)628-4888
260 1 We indicated in that response that there were 0' 2 additional, there was additional items that needed to be 3 closed out plant specifically, and if you looked at some of ;
- 4 the details of those items, one could conclude that that's the 5 only way to close them out.
6 (Slide) 7 HR. STALTER: There were some programmatic areas 8 where there were differences. We had a recommendation in I
9 there that we, that we maximize the dependence on the NNI-X 10 powers or NNI-X area. The thrust of this was that if we lost 11 NNI-X or NNI-Y at some plants, we would cause a transient.
12 It turns out that in most plants the majority of the
() 13 controls that feed the ICS signals--excuse me. They feed the 14 ICS--come from NNI-X so a loss of NNI-X i s going to cause a -
15 transient no matter what. Therefore, we felt that we could i'
1 16 maximiz" the dependence on NNI-X, we would not initiate a '
17 transient if NNI-Y failed, thus reducing the probability of .
1 18 failure.
19 It turns out that this is, this becomes a moot point 20 when we implement the recommendation for hardening of the 21 inputs because we intend to, if we do it properly, we intended 22 to develop enough signals from the X and Y source. That is, 23 we can lose either and still continue to operate and/or assure 24 ourselves that we get to a known safe state.
25 HR. MICHELSON: What did you do about the problem of HERITAGE REPORTING CORPORATION -- (202)628-4888
261 -
1 in some cases losing the instrument power supply but not the 2 ICS power supply that was, that that instrument was feeding, 3 the instrument was feeding a certain ICS panel? And there are 4 two separate power supplies. Instrument is supplied power 5 locally wherever it is located, and as I recall the light bulk i 6 event, they hed several cases where the instruments had lost 7 their power supplies, but the controllers hadn't, and 8 vice-versa.
9 And that was creating--
10 MR. STALTER: The instruments, the instruments that 11 feed that come out of NNI-X, that go to the ICS. Those 12 instruments are fed from the NNI-X cabinet.
13 MR. MICHELSON: Right. There are power supplies in O 14 that cabinet.
15 MR. STALTER: And if likewise the signal comes from I i
16 NNI-Y, that instrument is fed from Y power, so if I lose Y 17 power totally, I have lost both the instrument and I have lost i 18 the ability to feed that to the ICS.
19 MR. MICHELSON: ICS cabinets are supplied by other
. 20 buses, though, aren't they, instead of X and Y?
i 21 MR. STALTER: ICS cabinets have their own source.
22 MR. MICHELSON: The problem was that the ICS cabinet ,
l 23 got the power, but the instrument didn't, or vice versa, j l
a 24 MR. STALTER: That's right, and that's why they are f' l l
j 25 hardening. That's why we are saying let's look at somo l HERITAGE REPORTING CORPORATION - (202)628-4888 I
t 262 l
1 different inputs, and if we lose the power to an X cabinet, we
/')
i
\d 2 want to be able to utilize that signal that is coming through l
l 3 the Y cabinet, and the implementation of smart automatic l
i l 4 signal selection will give us that ability to continue.
l 1
l 5 MR. MICHELSON: I guess you mean auctioneering of i i
l 6 some sort, l l 7 HR. STALTER: It is not just auctioneering. It is )
I l i
8 automatic signal selection, and I can talk about that. I l l
- 9 wasn't prepared--I don't have a slide prepared for that. The 1
- 10 details of how that selection takes place are, get a little ,
1 11 bit complicated, but it looks at high failures, low failures, !
(
1 12 slight drifts, and mid-scale failures.
l
() 13 HR. MICHELSON: That leaves me at ease. Thank you.
14 MR. STALTER: There was a note in there that they 15 suggested that we put a recommendation in to check out and in 16 this case control rod drive control system modules prior to i
17 their institution into the system, and this is a case where l 18 you have a failure in replacing a module. We have felt there f
19 wasn't any need to do that because that's already included in 20 the procedures for, at each of the plants, so we didn't feel l
l 21 that there $s a need forecast a recommendation in that area.
l 22 The frequency of preventive maintenance, we chose a 1
l 23 frequency of every other fueling outage for tuning the ICS.
24 We feel that is a good frequency.
l l
O That does not mean that we 25 don't calibrate the instruments more frequent than that. We 1
HERITAGE REPORTING CORPORATION -- (202)628-4888
u 263 {
1 have calibration schedules that do come around every eighteen 2 months, but for tuning purposes we don't see enough drift in 3 the tuning characteristics to, to tune it more often than 4 every other refueling outage.
5 DR. KE2R: What was the basis of selection? '
6 HR. STALTER: We looked at the past experience that 7 we had in tuning, and the changes that occurred. We didn't 8 see a lot of change in tuning characteristics between the 9 times that we tuned. The problem that we had was we, if you 10 go back from the early Oconee days in 1970 up through the il middle '80s, there was very little tuning done and we didn't (
12 look at--
{} 13 DR. KERR: By tuning, you mean testing it, or what 14 does tuning mean?
15 HR. STALTER: Tuning the ICS means to operate the 16 system, and adjust it so that it, that it, its output 17 patameters stay within a given acceptance criteria. That is, 18 you make a rated change, you make a rated change in the 4
19 reactor, and you want to make sure that all the parameters 20 follow that. When you change the power level, you make sure 21 the feedwater goes where it belongs, power level in the 22 reactor goes where it belongs, turbine goes where it belongs, 23 and then--
24 DR. KERR: Preventive maintenance you do not to 25 repair things that have failed but rather to replace things HERITAGE REPORTING CORPORATION -- (202)628-4888
264 1 that might. fail? Is that what is meant by preventive .
(} 2 maintenance? I see something up there.
3 MR. LIDDLE: Right. It is not just replacement, but 4 also you know. just verification, check and calibration, to 5 identify those things, like you say, that possibly need i 6 replacement.
7 DR. KERR: You are saying that your past experience 8 would indicate that nothing is likely to go wrong? The 9 likelihood of anything going wrong over a two-year period is 10 so small that you don't need to do anything more often than 11 that?
12 HR. LIDDLE: That is correct.
13 DR. KERR: What sort of likelihood of drift or O 14 failure or whatever over a two-year period would you find a
15 acceptable? l l
16 HR. LIDDLE: The module is at--we have experienced j i
17 in the past, they have not had a high failure rate at all nor 18 a drift problem associated with them.
19 DR. KERR: What is high and what is low? Are you l l
20 willing to accept a change of one in ten or one in a thousand 21 or one in two or what?
l 22 MR. LIDDLE: I don't know that we put a number on it !
23 that specified this was exactly what wo had to have, but the l 24 past experience just from our TAP data base of the failures J
l O 25 that have occurred, and our knowledge of what has happened in l
l l
HERITAGE REPORTING CORPORATION -- (202)628-4888 )
. .--. -\
265 1 the plants, that's what we based it on.
2 CHAIRMAN WYLIE: You do this eighteen months to two d
3 years?
4 HR. LIDDLE: I am saying it is every two refueling '
5 cycles, which would be three years.
6 CHAIRMAN WYLIE: Is this true of your power 7 supplies, too? Inverter and what have you?
i 8 MR. LIDDLE: We are talking here of strictly 9 internalized--
10 CHAIRMAN WYLIE: Not power supplies?
11 MR. LIDDLE: Like Larry mentioned, a lot of this 12 hardware is checked every 18 months anyhow, for other reasons.
() 13 MR. STALTER: Power supplies will be checked every 14 eighteen months.
15 MR. LIDDLE: This is specified as a maximum and if a 16 utility chooses to, they can go below that.
17 MR. STALTER: Calibration of the instrument strings 18 would be done every eighteen months.
19 DR. KERR: It may be a good selection, but I haven't 20 heard any very convincing arguments that there is much of a 21 basis for it, other that it seems like a good time to do it.
22 (Slide) 23 MR. STALTER: The last area that we categorized I 24 these recommendations into was where the recommendation itself 25 was not agreed to and/or the implementation process, and there HERITAGE REPORTING CORPORATION -- (202)628-4888
266
. I were a number in this area, and they involved first of all n
() 2 replacement of an RC flott of the pump status.
3 There was one method that we used to do that with, 4 whereby we normally had the pump contact status in there, and )
5 in the event that we had a trip of a reactor coolant pump, we 6 then transferred the actual flow signal to allow a smoother 7 tre.nsition from four pump to three pump operation, and what we 8 accomplished by that was we, we eliminated the possibility of 9 initiating reactor trip in four pump operation which is 10 certainly the majority of the time, and improved the, improved 11 the reliability somewhat.
12 We didn't--it is not a hundred percent solution 13 because obviously during three pump mode, you still could have
{
14 transmitter failure in that case. We are going to take a 15 second look at that when we implement the other recommendation 16 for hardening the input.
17 They did not agree with deleting of feedwater 18 temperature correction circuit. Again, that was one where we 19 felt there was not a need for that, and we wanted to get 20 something out of the ICS where we didn't see there was a need 21 for it.
I 22 Removal of the BTU limits was an early l
23 recommendation that was, that preceded SPIP. We endorsed it j 24 because there again these were, these were circuits that could !
( l 25 fail that could cause the ICS to do something we didn't want l HERITAGE REPORTING CORPORATION -- (202)628-4888
267 1 it to do, and we saw a way of eliminating the BTU limits, and l l
2 we said okay, let's get them out of here, but one key thing in l O 3 this area is that we need to then provide a means to run back i
4 feedwater on a reactor trip, and that, that I believe was 5 added to that recommendation, was it not, Phil?
6 MR. LIDDLE: It was just a recommendation was 7 incorporated into our RTS based open draft report when the 8 final came out that clarified it, and the whole recommendation 9 was incorporated.
10 MR. STALTER: There were a couple other areas that 11 they, that were talked about. These I have provided here as 12 examples. SMUD had their, their implementation of known safe 3 13 states as if we lose NNI power, we are going to trip the ICS, 14 primarily to get the main feedwater out sf the way so that we ,
15 can go to an emergency feedwater control, and that would 16 initiate the emergency feedwater control. We felt that hey, 17 maybe we shouldn't trip the ICS. What we should do is ,
18 actually trip the main feed pumps, initiate main feedwater
- 19 control, and still have ICS control available for other things ;
20 if we needed it.
21 The removal of the Z power is Z power at--SMUD is 22 the only one that has Z power. We felt that that introduces 23 one more failure mode. If we lose Z power, we lose the 24 inability to transfer signals to the ICS.
- 25 Location of the power supply monitor, we indicated HERITAGE REPORTING CORPORATION -- (202)628-4888 !
268 1 that we wanted the power supply monitor installed as it was r's 2 originally designed, that is, that it monitors the voltage on
\) -
3 the bus, and its purpose is to detect a voltage degradation l 4 that could cause the modules to do something that they ,
t 5 shouldn't be doing, and we felt that it was appropriate to put 6 it in that area. j 7 Now we did look at. We utilized the study that SMUD ;
8 had done shortly after their event, and we accepted that study 9 as the way that we should go and that study looked at putting 10 power supply monitors parallel, utilizing a two out of two 11 systems instead of just one. It looked at putting a power 12 supply monitor on the very end of a string of daisy chain so 13 that it detected the voltage at last module down the road, and 14 each of those, each of those was evaluated. We felt that the 15 appropriate location for that was on the bus, and still feel 16 that way.
17 MR. MICHELSON: The Z power on SMUD, I thought that 18 came about because of the light bulb event, is that right?
19 MR. LIDDLE: Yes, it did. [
4 20 HR. MICHELSON: Now you decided it shouldn't be
] 21 there? In that it?
22 MR. LIDDLE: That's right. The failure mode of that f i
i 23 can cause loss of all signals to the ICS.
l 24 MR. MICHELSON: Had not been adequately elevated 25 when it was decided to do it that way?
HERITAGE REPORTING CORPORATION -- (202)628-4888
269 1 MR. LIDDLE: SMUD has, has done several evaluations
( 2 of this since that time, and their final conclusion is that 3 they should remove NNI-Z and replace the signal selection 4 capabilities within the NNI-X in cabinets.
5 MR. MICHELSON: Okay.
6 MR. LIDDLE: That was done for a specific purpose, 7 and it served one purpose.
1 8 MR. MICHELSON: Hadn't been necessarily fully l 9 analyzed at the time?
10 MR. LIDDLE: That is correct.
l l
11 MR. STALTER: We had a recommendation, we had a I 12 condition where it has occurred already, where on start-up
(} 13 particularly the ARTS bistable doesn't get reset and then you 14 reach, reach the ARTS arming point which is typically as you
! I i 15 are going up about 20 percent in power, and it arms that and l 16 it says, gee I have lost feedwater because the ARTS bistable i
17 hasn't reached that so we tripped the reactor, and we felt
[
18 that an automatic reset on that should be instituted so that i
! 19 that's one concern that the operator doesn't have to deal with l
< t i 20 on a normal start-up situation. i t
- 21 In summary, we felt the magnitude of the effort even i
22 though it was large, the methodology we used was sound. We ;
23 also determined that there are vulnerabilities that still I i
24 exist in the ICS/NNI design and those are that we can't 25 eliminate the design of mid-scale failure without really ;
HERITAGE REPORTING CORPORATION -- (202)628-4888
270 l
1 replacing the systems, and that's one of the vulnerabilities 2 that exist. ,
i 3 However, with the implementation in the level 1
{
4 recommendations, we can reduce the effect of those 5 vulnerabilities. We can go make sure that the system does go 6 to known safe state if we lose ICS or NNI power, and in 7 addition to that, we are continuing to pursue follow-on 1
8 activities and the advanced control system design is one of J
9 those follow-on activities that is heavily going on at this 10 time.
11 HR. HICHELSON: What does it mean vulnerability 12 This is a non-safety system, so what kind of '
still exists?
() 13 vulnerabilities are we talking about?
, 14 HR. STALTER: We can still lose the power to the ICS 15 so it is vulnerable to a power loss.
16 HR. MICHELSON: This is not vulnerability in the 17 pense of any kind of safety vulnerability, just in terms of i:
j 18 reliability of the instruments?
i
- 19 HR. STALTER
- Reliability and availability.
4 20 HR. MICHELSON: Ckay.
21 HR. STALTER: What I would like to do is give you a
]
22 brief rundown of where we stand on the advanced control 23 system, what we are doing, where we are headed.
24 This program now is leading the industry literally.
25 We are kind of a jump in front right now at the technology 4
4 HERITAGE REPORTING CORPORATION -- (202)628-4888
271 q 1 that's exists especially in the United States, and there is a ;
L
() 2 couple areas where that is key, and I will point those out in f I
3 a little bit.
i 4 HR. MICHELSON: Before you get to that, have you 1 5 read the ACRS letter on A-47, USI A-47? That's the safety ,
i j 6 implications of control systems in the title. Are you aware
{
7 of that one?
8 MR. STALTER: I am aware of it, yes. I am aware of ;
0 9 the issue. !
i 10 MR. MICHELSON: Did you--in there we remarked, of l l ;
i 11 course, that we felt that the true safety implications hadn't l i
12 been fully searched out because of the failure to look at l 13 external events, and several other things in there, and !
- (
- )
j 14 degradation of voltage and air conditions, that sort of 1 15 things. ;
l 16 Did you take issue with that letter in terms of it l 17 doesn't seem like you are correct. What we complained about :
18
! in there as possible safety implications, even what you (
- 19 presented to us here, has not addressed those kinds of safety i 4
j 20 implications, as near as I can tell. I just wondered if-- !
4
[
21 MR. TAYLOR: Specifically is this a real recent f
22 letter?
. 23 HR. MICHELSON: Was it last meeting we got it out, 4
24 wasn't it? i 25 HR. CALVO: This is Jose Calvo. I think I saw a j l I HERITAGE REPORTING CORPORATION -- (202)628-4888
(
i i
272 ;
1 couple of weeks ago--I think it was too late. We had E l 6 V 2 everything else.
3 HR. MICHELSON: B&W has seen it, or you don't--
4 HR. CALVO: There is a publication of the letter.
I j 5 I'm sure it was available to them. I just read about it.
j 6 HR. HICHELSON: Didn't get into any of--I just !
7 wondered whether or not what we complained of in there has (
4 8 been taken care of by some of the things that were done on !
a ,
9 the--
s
[
] 10 HR. CALVO: Some of the things--not all the way;
- 1 i 11 some of the things. I think you may hear that when Rick 12 Kendall will be talking about the other things, some of those 13 things in there.
(]}
- 14 HR. STALTER
- Things which we are including in the !
I 15 advanced control system program are certainly lessons learned l
j 16 from this evaluation, and previous evaluations. We are 17 including a look at foreign control systems and what they are i t 18 doing. '
i j 19 In particular we looked at the German designs. We l
j 20 have looked at Japanese designs, and we are pursuing a look 21 right now at what the Canadians have done, and the Germans and i
i 22 the Japanese, and even the Canadians have really taken a step 1 j
j 23 in the direction of implementing technology. We feel there is
)
24 a lot to be gained by looking at what they are doing.
- 25 HR. HICHELSON
- Have any of them considered making l
HERITAGE REPORTING CORPORATION -- (202)628-4688 l
f 273 I
~
() 2 so forth, are they still considering them as non-safety 3 systems? Are they designing them as safety grade systems?
! 4 MR. STALTER: I believe that they are still designed ,
4 !
1 5 as non-safety grade systems. Now there may be portions of j 6 what the ICS does that they might classify as safety grade. I
)
7 quess I can't specifically answer that.
t 8 Do you know, Phil?
I I
- 9 MR. LIDDLE
- I can't add any more to what he said. j 10 MR. MICHELSON: ICS doesn't anyway centrol auxiliary !
i 11 feedwater in the plant, does it? ,
- 12 MR. LIDDLE
- No. !
l I j 13 MR. MICHELSON: Just main feedwater? i J 14 MR. LIDDLE: Main feedwater. '
15 HR. STALTER: We are going to include as part of l 16 this AP a detailed human factors study in the ICS/NNI area, j j
17 instrumentation controlled. We have on board an expert
< i l 18 advisory panel and those are, we have a representative from ,
19 NASA, one from EPRI, one from Oak Ridge Natural Laboratories.
I l
) 20 Their input is, has and continues to be, have a very major I
] 21 effect on what the program is doing.
l i 22 The ACS as we envision it now will be both a plant j j i 23 control system, and a monitoring system. We are going to try !
l 24 to include, improve on the control techniques. We are going
(:) '5 to look at, at selective integration. We are going to look at i
I HERITAGE REPORTING CORPORATION -- (202)628-4888 1 1
274 1
t 1 segregation, which is the, the recommendations that came out
{} 2 of the sensitivity study. We are going to see if they apply We have an objective to 1
3 in the advanced control system arena.
4 keep the operator burden down, and we want the plant to ,
5 continue to be responsive to the demands that we put on it, ;
6 but if it loses a pump, we want to be able to respond to that 7 pump loss.
a 8 Increased reliability is a major, major input into ;
9 this program, and I emphasize that because we are going to 4
10 look at it in two respects. One is fault avoidance, and i
11 that's where we are going to look at all the weak points that l
f 12 we can define, and we are going to try and design those out.
l 13 We are going to make every effort to design out the weak '
i 14 points.
j 15 The second thing that we are looking at is fault r j i 16 tolerance, and the fault tolerance says we want to be able to ;
i 17 take any single fault and we want to continue to operate when f
, 3 l
18 we have that singic fault. i 19 Now how we define those single faults is still -
)
J 20 ongoing. I a
1 4 21 CHAIRMAN WYLIE: Is component life selectivity part b l 22 of that program, too?
j 23 MR. STALTER: Component life? l l
1 24 CHAIRMAN WYLIE: Failure. I
\
j 25 MR. LIDDLE: That goes into the fault avoidance 1
HERITAGE REPORTING CORPORATION -- (202)628-4888
L 275 1 issue, high reliable components to be developed as part of i 1 ;
2 this system. (
i !
3 MR. STALTER: First of all, we are going to design !
4 highly reliable components here. We are going to pick the (
i
- 5 best that we can find, and then we are' going to, and then if l
! r 6 failing this, if we, if there is some fault that still occurs I i
j 7 oven though we design fault avoidance, put fault avoidance I 8 into it, we are going to fall back on fault tolerance and say ,
9 okay, we want single failure proof.
10 We are looking at, now at, at such things as triple
- 11 redundancy and triple redundancy of the whole ICS system is, I 12 is right now unheard of in the United States. I don't believe I
() 13 there is anything installed at this time. We are looking at l
1 14 that, and this will all be a digital system. The only way you l'
15 can do it is with the modern digital technology.
16 MR. WARD: Larry, how much increased reliability i i 17 just simple redundancy give you? I mean the risk analysis i
(
i 18 people keep telling us that there is, because of common l 4
i 19 failures, there is kind of limited additional reliability as ;
)
20 you had level of redundancy. This sort of system, what do you i 21 count on? l
}
22 MR. STALTER: There is an optimum point beyond which
- 23 the more you put in, the worse it gets. Our initial, and our !
! i j 24 initial review has indicated that, and I think maybe Phil can '
25 correct me if I'm wrong, but my recollection is that it falls I i
i HERITAGE REPCRTING CORPORATION -- (202)628-4888
276 1 somewhere between double redundancy and triple redundancy. f 2 MR. WARD: What do you mean by double redundancy?
L 3 MR. STALTER: Double redundancy says you have two 4 systems operating. Failure of one--they are sort of looking 4
5 at each other, and failure of one says okay, I am going to 6 quit doing this and you take over and do the operation because 1
I 7 I have failed.
8 Triple redundancy says I have got three systems that i
9 are systems that are operating, all three systems doing the 10 same thing, all three looking at the other three to see which !
11 ones are functioning properly, and when one fails, two of them !
12 say hey, that one has failed, we are not going to let him f t
() 13 control anymore, and it takes over and drops that one off the !
I j 14 system, and the problem is we have to do some other things )
l 15 like let the operator know that it is off the system, and it f 1 ,
t 16 continues to operate. i l
17 MR. DAVIS: What is single redundancy? '
i 18 MR. STALTER: Double redundancy is two.
I 19 MR. LIDDLE: Single system. j j 20 MR. WARD: That is what I was wondering what the
\ l j 21 definitions were. ;
l !
I 22 MR. MICHELSON: how does redundant system decide l l
23 which of its two channels is the misbehaving channel when we
. 1 24 can't auctioneer between the two? Auctioneer three; I don't i 25 see how you handle two because the failure is in the device
- l l
l
- HERITAGE REPORTING CORPORATION -- (202)628-4888 .
1 i - . _ _ , - . ,_..__.._..,_m. . - - - . . _ . . _ _ _ . . . , . . , _ . _ . . _ - _ _ _ _ . _ _ _ . _ . ,. _ _ _ , _ . . , . . _ . , , _ _
277 1 that is deciding which one is good and he picks the wrong one. !
2 MR. LIDDLE: What you rely on is the software that 1
3 has been designed to go through itself and check itself and to :
4 make sure its hardware is performing properly. l a 5 MR. MICHELSON: If you lose power or something you I :
? ,
- 6 may lose that. l i 7 HR. LIDDLE
- Power is, you account for that in l l
8 different means. We are talking about module failures, i 9 MR. MICHELSON: Modules can fail locally. Such 1
10 power in that module--
. 11 MR. LIDDLE: In that case, the other module will i
- 12 identify that counterpart has failed. It has lost power.
1
(} 13 MR. MICHELSON: Third module is doing the ]
i 14 auctioneering between the two then?
I 15 MR. LIDDLE: No. No. !
l
) 16 MR. MICHELSON: Module fails, the auctioneering !
1
! 17 circuit has got to be in one or the other, has got to be on i 18 either train A or train B. j 19 HR. STALTER: I don't want to get hung up on this i ,
20 fault. I want to let you know that we are looking at this in )
i 1 21 the evaluation of fault tolerance, and how we are going to I
! l 22 implement it is critical to the implementation of this l
l 23 advanced control system, and we are going to do the best job I l
l 24 that we can do on this, and we are going to answer those j 25 questions in maybe some future date. If you want to see, if l )
I l
l
, HERITAGE REPORTING CORPORATION -- (202)628-4888
1 1
1 278 1 you want an explanation of how we go'about this study, we will !
2 be glad to come up and talk to you about it.
3 HR. MICHELSON: What are you going to do about the l 4
g 4 environmental question which we have passed off earlier, but 4
5 now advanced system? I hope you go back to review the cooling l
6 of this equipment as an intogral part of the functioning of 1
7 the equipment. And are you going to put new redundant cooling 8 supplies, that sort of thing?
I 9 HR. STALTER: We don't know how we are going to !
1 l 10 solve that issue yet. l l'
11 CHAIRHAN WYLIE: What is your schedule for this?
i 12 HR. STALTER: The schedule is that it is a ;
i <
1 13 three-phase project. The first phase is to develop design !
l i 14 requirements. As part of that, we are updating the current t a i 15 design basis for the current, the ICS, existing ICS, and we 16 are going to establish a design basis for the advanced control
)
J 17 system, detailed design basis.
t l 18 Phase 2 is to look at the algorithms, advanced 19 control algorithms to be implemented in the system, and Phase
]
20 3 is to pick a hardware and implement the hardware.
l '
l 21 The Phase 1 is ongoing right now. Phase 3 has been ;
22 moved up in schedule to where it is beginning right now, and )
l l 23 we hope to have the whole project completed early in 1990.
i
) 24 And that's the current schedule. That schedule can fluctuate, 8
) 25 depending on the need of the owner because it is owners group i I
! I I HERITAGE REPORTING CORPORATION -- (202)628-4888 '
1
279 I
! 1 efforts. If some owner decides that he wants it available l 2 carlier, it might he a little bit difficult, but there, we 3 might be able to adjust the schedule to accommodate that. As
! t
) 4 a matter of fact, we did move Phase 3 up because there werc [
} :
i 5 some interest shown in installing the control system. I i 6 CHAIRMAN WYLIE: Owners group is financing this i l
7 total, in support of this total project?
t 8 MR. STALTER: The operators group instituted the
- f 9 project, established the task group for financing the program. !
10 However, in the process of doing that, when we went out and ,.
11 looked for the advisory panel, we also saw they pointed out I i
12 some areas that said gee, we can, EPRI in particular--I will ;
j
(} 13 point that out--they said there ere some things you are doing i 14 here that are leading the industry that could certainly help 15 Westinghouse and GE and the others and we want to take 16 advantage of those and we want to see if we can help you in 17 some of those areas, so they have taken over a portion of that 18 project anM they have recently got approved about a S450,000 I
i 19 effort to help us out in doing some of the things that we a
l 20 would have had to do otherwise.
j 21 HR. WARD: You are calling it a project?
22 MR. STALTER: Pardon?
i 23 HR. WARD: Excuse me. Go ahead and finish.
l 24 HR. STALTER: Primarily funded by the owners group; I 25 we are going to receive some outside funding for it.
1
& i 1
HERITAGE REPORTING CORPORATION -- (202)628-4888
280 1 HR. WARD: Calling it a project. This is a design (3
il Is that right? I mean what is the product of this 2 project?
3 effort going to be?
4 MR. STALTER: The product of this effort will be a 5 hardware that can be implemented in one of the B&W plants G utilizing the advanced control system technologies.
7 MR. WARD: Your organization is going to build the 8 hardware?
9 HR. LIDDLE: No. , cally it would be a 10 specification to be able to v2rchase the hardware, both from a 11 hardware aspect and from a' software aspect for the algorithms
~
12 that is implemented in that hardware.
() 13 MR. STALTER: What I meant, we are going to specify 14 the specifications. We are going to take those out and 15 utilize those and we are going to select hardware. We are 16 going to say okay now, and the reason, the reason we need to 17 do that is so tha*. we can assure that those things that wo 10 want are implemented in the final stages so we are going to 19 select the hardware before the end of the program.
20 MR. WARD: There is a good bit of software involved.
21 Are you going to design, create the software?
22 MR. STALTER: Yes.
23 MR. WARD: You have got to buy hardware to do it?
'N 24 MR. LIDDLE: We are not in terms of actual what (J
25 would go in that hardware, just in terms of functionality of I
HERITAGE REPORTING CORPORATION -- (202)628-4888
281 l
1 this is the algorithm, this is what the type of control will !
~be, not the indis 'ual programming of that hardware.
(]) 2 .
l
. 3 CHAIRMAN WYLIE: ~Your spec would spell out'the !
4 information for the performance requirements? ,
1 5 MR. LIDDLE: Yes.
6 MR. DAVIS: I have a little bit of a concern that 7 . safety is not mentioned anywhere as an objective for this 8 system. It seems to me one of your objectives ought to be 9 that the imglemencation of this system will not compromise the 10 plant safety, and maybe another objective would be that you 11 might want to look for things in this system that would 12 improve safety. Maybe that's implied.
13 MR. STALTER: This slide is a very--I want_to stress 14 this right now. This slide is a, is minimal, okay. Certainly 15 plant safety is one of the things that is foremost in our 16 minds, and what we intend to do is to make sure that this 17 systen can control the plant within the boundaries, within the I
18 boundaries of the existing safety systems with the intent that 19 we reduce the challenges to the safety systems.
20 MR. DAVIS: Also want to avoid the possibility of
- 21 system interactions and dependencies that can cause loss of '
I 22 all feedwater, for example.
- i 23 MR. STALTER
- Yes.
24 HR. DAVIS: I am sure you have that in mind, but we l
(:) 25 have seen examples where improvements have been made to a i
I i
HERITAGE REPORTING CORPORATION --' (202) 628-4888
282 1 plant at the same time they have introduced dependencies that
() 2 weren't found up front in the design. But that's all I have.
3 MR. LIDDLE: We realize that. Those are definite 4 objectives of the program.
5 MR. WARD: Along those lines, have you thought about 6 how you are going to control the software that is involved?
7 MR. STALTER: We haven't specifically defined how we 8 are going to place the controls on the software. We know that 9 we have to do that.
10 The last aspect I want, that I wanted to point out 11 in this area was the monitoring capability, because this is 12 something that is different than what we have now, and this is
(~
13 a system that we are going to try to incorporate to predict .
l 14 failures before they occur, and certainly to detect failures 15 when they do occur, and to alert the operators in both those 1 1
16 cases. l l
17 Now the prediction of failures can be sue.5 things as 18 the system puts out a signal to a valve that it wants it to 19 move to a certain position, and the evidence says that that l
20 valve didn't move, and maybe you need a 2 percent signal to l
21 make it move or a 1 be percent signal should have made it 1
22 move. This monitoring system is going to look at the l l
23 parameters and try to evaluate what kind of condition exists j
24 there and let the operator know that you are starting to see a 25 sticky valve or maybe a pump didn't move when it should have j l
l HERITAGE REPORTING CORPORATION -- (202)628-4888 j
4 .
283 1 moved, and I think that this is one area where many tic s the
() 2 existing ICS takes the blame for something when actually it is 3 the centrol component that is the problem area, that maybe the 4 ICS did present a signal that asked it to do the right thing 5 but the valve didn't m' e or the pump didn't go, and we have 6 tried look at those areas in existing evaluation where we 7 limited ourselves to the inputs and outputs.
8 HR. MICHELSON: Maybe you could make a real step 9 forward and put in some internal temperature monitoring in the 10 cabinet which most of our cabinets now lack. Never know when 11 they are getting too hot.
12 MR. STALTER: We do have temperature monitoring in 13 the cabinet.
14 MR. MICHELSON: The present ICS?
15 MR. STALTER: Have a monitoring in them, that is, a 16 device to detect the fan has failed.
17 MR. MICHELSON: That's one device. I was thinking 18 of a more direct measurement like what is the actual 19 temperature? A couple more things than the fan itself; 20 somebody might have blocked the end of the bottom of the box 21 or something.
22 You are heading in the right direction a little bit.
23 Sone of them don't even have monitors on the fans.
24 MR. CATTON: This is particularly important I think O 25 if it is digital because they just sort of sneak away a little HERITAGE REPORTING CORPORATION -- (202)628-4888
. _ - ~ .. .- - .- --- - - . - . - - .- -
284 1 bit, and a thing gets' flipped somewhere, and it-is-not a
) 2 . complete failure,-just sort of. runs the program differently.
3 'MR. STALTER: We agree.
4 MR. CATTON: There is not very much-known about i I
i 5 this.
6 HR. MICHELSON: You can build the temperature l
7 monitor right into the cabinet.
8 MR. STALTER. That's all I have. l 9 MR. WARL: Maybe you said this, I missed it, but you l
P 10 have got tiae three levels, and as I understand it, all of the 11 owners are committed to the level 1, is that it?
12 MR. STALTER: Yes.
13 MR. WARD: And someone is going to do 2 and some
[}
14 maybe are going to do 3. What is the status of level 2 and 37 1 15 MR, STALTER: The level, the level 2 and 3 16 recommendations are being considered presently as part of the 17 level 3 efforts for the control system. The level one 18 recommendations are those which we feel will provide immediate )
19 improvements. Those are in the recommendation tracking system b
20 and are being evaluated as part.of that program. )
21 MR. WARD: If you come up with something really good 22 on level 3, everybody is going to ignore level 2, is that the 23 idea? Just go right to level 3?
l 24 MR. LIDDLE: Level 2 is boing evaluated, separate l
25 also.
HERITAGE REPORTING CORPORATION -- (202)628-4888 f
.-- ~ _- . . . .
285 1 MR. MICHELSON: Level 1 will be done? All the
. 2 owners say yes?
3 MR. LIDDLE: Yes.
4 MR. KENDALL: My'name is Rick Kendall. I was 5 involved in the staff review of the instrumentation of, the 6 Instrumentation Control Systems Branch review of the ICS and 7 NNI evaluation that the group performed.
8 In listening to the presentation that Larry Stalter 9 just gave, I jotted down a couple of notes on certain items.
10 The question was raised about lessons learned from 11 the light bulb event, and why some things weren't, apparently 12 may not have been done or lessons n.ay not have been learned or '
(} 13 things implemented.
14 I believe that the light bulb event involved--I 15 probably should not say that I would be speaking for SMUD, but 16 the light bulb event involved a loss of NNI power, and SMUD 17 did do some modifications to the NNI power distribution 4
l 18 system. The event in 1985 involved a loss of integrated 19 control system power. The power distribution systems are very 20 similar between the NNI and the ICS, but I believe that the l 21 focus at the time was strictly on the NNI and things didn't 22 get carried over to the ICS.
23 Another point that was brouqht up was steam i 24 generator overfill protection. I believe that some plants l O !
25 that have the EFIC system need safety grade isolation of main ,
-. ~.-.___ - _._._.__ _ _ ___,_ _ _._ _ ,._ ~ _._ _ _ _- ._(202)628 HERITAGE REPORTING CORPORATION :
_ . _-4888_ _ . _ . _ . _ . _ _
286 1 feedwater to prevent overfill. I believe Rancho Seco has
() 2 that. I think Crystal River has that also, and maybe some 3 other plants have it, too.
4 MR. MICHELSON: It is already installed?
5 MR. KENDALL: At Rancho Seco it is installed. At !
6 Crystal River I believe it is installed. I think perhaps 7 Arkansas has it. I'm not sure all across the board, but some l I
8 plants do have safety grade isolation main feedwater to j 9 prevent overfill, isolation steam generator high level.
i 10 The first item on the agenda was the ICS/NNI systems 11 requirements. I have got a very simplified version of the 12 ICS/NNI evaluation process that are used by the owners group.
13 It is somewhat similar to the slide you saw earlier.
14 The reason for this slide is to show the importance 15 of the systems requirements and the development of the systems 16 requirements in the overall scheme of the ICS/NNI evaluation.
17 There were, the owners group looked at a number of 18 sources of information--NUREGs, FMEAs, owners group studies.
19 They looked at LERs, information from their TAP program, 20 transient assessment program, to try develop an optimum set of 21 systems requirements, and the systems requirements were 22 basically on what the original ICS/NNI designers would have 23 used if they had had the knowledge of the experience to date, 24 operating experience to date, so it was intended to be a very O'
\"'
25 comprehensive set of systems requirements.
HERITAGE REPORTING CORPORATION -- (202)628-4888
287 1 Then the owners group determined the as-built plant O)
(_ 2 configurations for the ICS and NNI. They performed plant 3 walk-downs, and verification of ICS/NNI drawings so that they 4 knew exactly what the installed designs were.
5 Then they compared the installed designs to the l l
I 6 systems requirements they had developed for the ICS, and NNI, l
7 and then through a comparison effort, compared the existing I 8 designs to the design requirements. From that comparison 9 effort came forth the recommendations for improveme,nts to the 10 ICS and the NNI. The comparison efforts we viewed as the most 11 important phase of the ICS/NNI evaluation. We felt that's 12 where the most benefit was going to get gained, that's where
{} 13 the problems would be identified, and the recommendations 14 would be made. Through this comparison process, you see the 15 need to have a very good set of systems requirements. Without 16 a good set of systems requirements, the comparison process 17 might not mean too much.
18 (Slide) 19 MR. KENDALL: The items at the top of the slide are 20 a list of some of the systems requirements that are common to 21 both the ICS and the NNI systems, environmental specifications 22 such as temperature, humidity and radiation, instrument 23 accuracies, power supply design. Under power supply design, 24 there is requirements for redundant AC and DC power supplies, 25 requirements for loss of power alarms, and the next two items HERITAGE REPORTING CORPORATION -- (202)628-4888
288 1 are what we consider to be key items.
() 2 The systems requirements are written such that the 1
3 effect of loss of power should not prevent the plant from l 1
4 being able to achieve the known safe state, and also that upon 5 restoration of power, the plant would remain in a known safe 6 state and that we think that perhaps the most significant 7 concept focused on by the owners group in their review was 8 that of trying to ensure that the plant would achieve a known 9 safe state given any loss of ICS or NNI power.
10 Down below, we have the definitions provided of a 11 known safe state. They are summarized somewhat, but it 12 basically involves maintaining heat transfer balance between 13 the primary and secondary systems, either automatically or 14 manually. Doing it manually allowed credit for operator 15 action for which the operator is normally trained and which 16 can be taken from the control room on a loss of ICS/NNI power.
17 MR. MICHELSON: Did you give any thought to the loss 18 of environment control over the environment of the equipment 19 and what the failure state should then be?
20 In other words, did you need automatically kill 21 power at some temperature points so that you don't go into an 22 unknown failure state?
23 MR. KENDALL: Well, that's something that we, the 24 owners group didn't pursue, and in their review--
O' 25 MR. MICHELSON: Certainly the staff was certainly HERITAGE REPORTING CORPORATION -- (202)628-4888
289 1 aware of the potential difficulties-that the equipment could
() 2 get into. We have had enough experience now to any longer 3 deny the problem, so how was the staff approaching that 4 question in this particular case?
l 5 You are obviously addressing the failure state of ;
6 the equipment and you are addressing voltage. That is a good l 7 idea, but how about loss of environmental cooling? You would l
8 like to know what the failure state of the equipment is. l 9 MR. CALVO: I guess we would consider failure modes 10 effects analysis or was that a recommendation that you made 11 should be offered for consideration for later?
12 MR. KENDALL: I don't believe it was considered in 13 the failure modes and effects analysis. Neither did we O 14 consider it in our review, 15 MR. MICHELSON: It was not, according to the owners 16 group, was not in their failure mode. That doesn't mean you 17 ignore it in your review. You just pointed it out as an area 18 that needs to be done if that were the case or whatever. I 19 would like to know what the staff feels about it in the 20 absence of it in the owners group evaluation.
21 MR. KENDALL: I an not sure how to respond. I'm 22 sure it is obviously a valid concern.
23 MR. MICHELSON: You understand the question I think.
24 MR. CALVO: The fact it was brought up was, brought 25 up as part of insistent requirement, it is one thing to be HERITAGE REPORTING CORPORATION -- (202)628-4888
290 1 considered. I think it is indicative of the the fact that
(} 2 everything was not pursued by the owners group, so it remains 3 up there for the consideration.
I 4 MR. MICHELSON: The staff has kept that 5 consideration open? ,
I l
6 HR. CALVO: This particular subject, if I remember 7 correctly, was discussed in either the safety evaluation 8 report or in the contractor safety evaluation report it 9 defines one of the very important system requirements.
10 MR. MICHELSON: You say the contractor discussed 11 this. Tell me where to read it.
12 MR. CALVO: Maybe--
13 MR. KENDALL: I don't believe we discussed it. I O 14 don't think, I don't think our review even looked at that.
15 MR. CALVO: Okay.
16 MR. MICHELSON: Maybe we ought to go back and think 17 about it at least as to whether it should be included. It is 18 again in line with the A-47 letter.
19 MR. CALVO: That is correct.
20 (Slide) 21 MR. KENDALL: In general, we felt that the systems 22 requirements were fairly good. We felt that if existing plant 23 designs were modified to achieve conformance with the systems 24 requirements, that many of the concerns identified from 25 previous B&W reactor transients involving loss of ICC power HERITAGE REPORTING CORPORATION -- (202)628-4888
291 1 would be resolved. The systems requirements from our point of l
() 2 view as we went through them tended to focus on improvements 3 in two areas.
4 One was to minimize the effe2t of single instrument 5 channel component failures, such as single transmitter 6 failure. The B&W owners group recommendation to provide, to 7 provide redundant channel and then select the valid one of 8 those two if one of them should fail, falls along those lines.
9 The other was to limit the consequences or power 10 losses by choosing equipment failure positions such that 11 unknown safe state would be achieved. That I skipped--let's 12 see here.
- 13 We felt that design basis documents needed to be 14 developed for the ICS and NNI systems. The systems 15 requirements documents dictate some design requirements, 16 dictate some set points for different actions. It is not 17 clear why the set points were chosen or why the requirements 18 were selected. We felt that the design basis documents that 19 provided the base for the systems requirements ought to be 20 developed and ought to be maintained.
21 We felt that the systems requirements documents 22 themselves ought to be kept up to date, revised and maintained 23 as changes were made such that the system designers or the 24 users of t he ICS would have some documents to go to to O 25 understand why changes were made or why the design is-the way HERITAGE REPORTING CORPORATION -- (202)628-4888
292 1 it is. We thought it would be a useful document, useful
() 2 reference document, something that would be used, and that was 3 the basis for our recommendation for design basis documents.
4 We did feel that the ICS system requirements 5 documents ought to be revised to include some areas that were 6 not addressed or at least were not addressed in detail. Some 7 examples of those were the power supply monitor. We felt that j 8 the systems requirements ought to specify the power supply 9 monitor alarm trip set points.
10 Also a loss of instrument air we thought ought to be 11 further addressed. We felt that a similar requirement to that 12 of achieving a known safe state on loss of power ought to be 13 providad for loss of instrument air. Loss of instrument air, 14 the equipment would position itself such that the plant would 15 achieve a known safe state.
16 MR. MICHELSON: By loss, do you mean degradation or 17 do you mean reiduction of the pressure to zero? It can be 18 quite a difference in how equipment is formed.
19 Did you have both in mind, or only the latter?
20 MR. KENDALL: I would like to think we had both in 21 mind. For power, we felt that we clearly had both in mind, 22 both loss of power and degraded power condition. That gets, 23 that starts to involve the power supply monitor which we will 24 talk some more on later. We just--
O 25 MR. MICHELSON: Transmitter error we know can get HERITAGE REPORTING. CORPORATION -- (202)628-4888
293 1 plants into difficulty,. Reactor protection system, boiling O
(. / 2 uater reactor is a good example of that.
3 MR. CALVO: That particular concern was--I think we 4 will talk about that as one of the elements of limitations.
5 MR. MICHELSON: Later?
6 MR. CALVO: Tomorrow.
7 MR. MICHELSON: Thank you.
8 MR. KENDALL: As far as the staff review of the 9 ICS/NNI hevaluation performed by the owners groups, we 10 specifically concentrated our review on trying to determine 11 whether implementation of the recommendations would resolve, 12 resolve concerns identified from the investigations of
(} 13 previous reactor events involving losses of ICS/NNI power.
14 NUREG 1195, which is the staff's report on the 15 December '85 event at Rancho Seco, specifically pointed out 16 that the, that event involved the same types of concerns that 17 occurred during previous events such as the Crystal River 18 event, and even back to the light bulb event, that the 19 problems were known and that in the absence of modifications, 20 that the transient at Rancho Seco should have been expected 21 and should have, not have surprised anybody, so what we did, 22 we went back and we tried to compile a list of those concerns 23 that were common to those events.
24 It turns out that most of those concerns are 25 documented in three documents. Those are IE Bulletin 79-27, HERITAGE REPORTING CORPORATION -- (202)628-4888
294 1 NUREG 0667, which was published after the Crystal River event o' 2 in 1980, and NUREG 1195, which was ppublished after the Rancho i
l 3 Seco event of 1985.
4 We provided a fairly comprehensive list of those 5 concerns in Appendix E to the NUREG. What we have got here is 6 just a, an example of a few of the items that'were common to 7 some of those events. These are examples of items where the 8 staff focused its review--adequacy of control room indication, 9 enunciation, loss of ICS/NNI power, mid-scale failures, either 10 the prevention of mid-scale failures, or provision of 11 mechanism by which failures of indicators were easily 12 recognizable such that operators would not use failed
() 13 instruments to take action.
14 Procedures--inappropriate, undesirable equipment, 15 response to losses of ICS or NNI power, things like 16 atmospheric dump valves failing open on loss of power, turbine 17 bypass valves failing on loss of power, load fusing to ensure 18 proper clearing of faults, loss of remote manual control 19 capability for ICS controlled equipment, and the, what we 20 considered to be the major one, was resolution of IE Bulletin 21 79-27 concerns.
22 Again, we feel that probably the most important 23 concept emphasized by the owners group during their review was f 24 a concept of known safe state, trying to ensure that the plant 25 will attain a known safe state given a loss of power. We feel HERITAGE REPORTING CORPORATION -- (202)628-4888
- ~. - .- -
i 295 l
1 that resolution of the IE Bulletin 79-27 concerns would 2 essentially accomplish the same thing; 79-27 was issued 3 following the loss of ICS and NNI power event at Oconee in 4 November of '79. The intent of the bulletin is basically to-5 verify, to have plants verify that if you lose power to any 6 single bus that is supplying power to instrumentation and :
7 controls, that the plant design and plant procedures be 8 verified to be able to cope with that loss of power to make 9 sure that the plant can attain a safe shutdown condition.
10 MR. MICHELSON: Question--on the next to the last 11 bullet, you talk about loss of remote manual to develop. Are 12 their cawes now where ICS if it were to fail in a certain way
(} 13 would cause the inability to go to the remote control station .
14 say for the valves?
15 MR. KENDALL:
j Not remote; remote shutdown station, a
16 what this refers to is during the Rancho Seco event of '85 17 they lost integrated control system power, and therefore the 18 ability of the ICS to automatically control plant equipment. ;
, 19 With that loss of power they also lost capability to control 20 certain ICS controlled equipment manually from the control j 21 room. They could not close the atmospheric dump valves, 22 turbine bypass valves, had to go into the plant, do it ;
23 locally.
24 MR. MICHELSON: FMEA never identified any device 25 that is, could not be controlled locally as a result of that i
HERITAGE REPORTING CORPORATION -- (202)628-4888 l
296 1 failure, is that correct?
() 2 MR. KENDALL: I'm not--we have Mike Waterman from 1 3 INN who reviewed the--
l 4 MR. MICHELSON: Did FMEA show with certain kinds of 5 failures you might not be able to operate the device locally, I
6 from the local control station or from the control center?
7 MR. WATERMAN: The scope of the--Mike Waterman.
8 FMEA addressed the failures of the equipment, but I don't 9 think it really addressed whether or not equipment could be 10 operated remotely.
11 MR. MICHELSON: Depending how you lash it together, 12 and it is possible you may use remote. Depends on how that 13 circuitry is hooked up.
14 Did the owners group ever find the case of loss of 15 remote manual capability for local stations?
16 HR. LIDDLE: No.
17 MR. MICHELSON: You did look for it. You didn't 18 find it?
19 MR. LIDDLE: That is correct.
20 MR. MICHELSON: Thank you.
l 21 MR. KENDALL: During the Rancho Seco event, they l 22 could have used the remote shutdown station to close the 23 atmospheric dump valves. I 24 MR. MICHELSON: I have seen a number of LERs that
( 25 infer it wasn't for these, these particular control systems, i
{
HERITAGE REPORTING CORPORATION -- (202)628-4888
297 1 but they had lost the remote, manual local capability as well 2 as a result of the failure of the full system.
3 ( A discussicn was he:.d of f the record. )
4 MR. KENDALL: The next slide shows the, it is a 5 summary of the technical action required by IE Bulletin 79-27.
6 The concern being addressed by IE Bulletin 79-27 was basically 7 that you could lose power to a bus that has supplied 8 instrumentation and control circuits, and thereby initiate a 9 plant transient, and you could also lose indications in the 10 control room that the operator would use to take action to 11 mitigate that transient.
12 The actions required the review of all buses in the 13 plant, be it class 1E or non-class 1E, safety-related, not 14 safety-related, that provides power to instrumentation and 15 controls, to review to ensure that adequate indication of bus 16 power was provided to the control room, to evaluate the effect j 17 of loss of power to all bus loads, to see the effects that 10 power loss would have on plant operation and its ability to 19 achieve cold shutdown condition. l l
20 The evaluation of bus loads, he was to include the l 1
21 control room indicators that the operators would use to take !
I 22 action to respond to the bus failure; to review the candidate, 23 which of the procedures used to achieve cold shutdown i
24 following loss of power to each bus--that basically consisted O 25 of making sure that there were proper indications to guide HERITAGE REPORTING CORPORATIOli -- (202)628-4888 l
298 1 operators to p. cedures, and that the procedures directed the (s~-)- 2 proper use of back-up instrumentation and controls that 3 remained available that was unaffected by the power loss to be 4 able to achieve a safe shutdown.
5 We believe that if the ICS/NNI buses were reviewed 6 in accordance with IE Bulletin 79-27, and that if any 7 appropriate acticns were or changes were made to ensure that 8 the bus failure could not cause a transient, and at the same 9 time cause loss of control room indications and complicate the 10 ability to achieve a safe shutdown condition, that most of the 11 ICS/NNI concerns identified from events would be resolved.
12 That's why we have placed so much emphasis on IE BULLETIN ;
(} 13 79-27 throughout our review of the SPIP program. We believe 14 it is probably the single most important document that the NRC 15 has put out regarding this effort, and we think that it 16 is--and we will get to it in the next slide further on--we 17 think that it is something that still needs to be done.
18 HR. WARD: I guess this bulletin has been out for 19 nine years, eight or nine years apparently?
20 HR. KENDALL: Right.
21 MR. WARD: And there hasn't been a lot of response 22 to it or what?
23 HR. KENDALL: No. That's a complicated issue in 24 itself. I will try to summarize it, and refer you to NUREG 25 1195. I believe it is Section 7 of NUREG 1195 provides a real HERITAGE REPORTING CORPORATION -- (202)628-4888
299 1 good discussion of the 79-27 review process both by industry 2 to an extent and by the staff. It turns out that when the 3 staff initially reviewed the response.to the bulletin, many of 4 the responses were found to be inadequate, and in fact I think 5 there were about-70 operating reactors at the time, and the 6 staff concluded that about 69 of the 70 responses were 7 inadequate to determine that the concerns of the bulletin were ;
8 being adequately addressed.
9 Time went on, and the review scope was narrowed 10 somewhat, and then I guess it was narrowed some more somewhat, 11 and eventually the same responses were, that were found to be 12 unacceptable originally, we found that most of the responses 13 were now acceptable, and instead of doing a detailed review of O 14 the responses to the bulletins, we did kind of a cursory 15 review just to see if I guess basically the plants responded 16 to the bulletin. Again, the details are in 1196. We felt 17 that because of the initial review or the lack of a detailed 18 initial review, and the Rancho Seco event which clearly 19 cvidenced at least at the one plant that the concerns of IE 20 Bulletin 79-27 had not been resolved--if they had been, we l i
21 don't feel that event ever would have happened--we felt there l
22 was a real need to go back and at least for B&W plants, try to 23 make sure that the concerns of the bulletin were adequately 24 addressed.
N 25 We did in recent plant licensing reviews, OL reviews i
HERITAGE REPORTING CORPORATION -- (202)628-4388
300 i l
1 of GE and some Westinghouse plants, very detailed reviews to l i
(~%
\_) 2 IE Bulletin 79-27. Essentially we found that there weren't 3 many problems, and in reviewing the event at Rancho Seco, it 4 was very clear that the problems existed there.
5 They did a re-review of IE Bulletin 79-27 as part of 6 their restart efforts. They were I believe somewhat amazed at 7 some of the things that they found and they are making a 8 number of modifications because of their review, and this is 9 not just solely restricted to ICS/NNI buses.
10 We feel that for B&W plants, and this comes in a 11 later slide, but in a nutshell, we feel that combination of 12 implementation of the recommendations that the owners group 13 has put forth, are going to implement, and resolution of the
(])
14 concerns of IE Bulletin 79-27 would essentially lead us to the 15 point where we would feel pretty comfortable with the ICS/NNI 16 designs. I think at this point we would feel that ICS or NNI 17 failures would be such that the plant essentially would 18 achieve a known safe state.
19 MR. WARD: Does this tell you something about the 20 regulatory process? I mean should--this was an IE circular?
21 MR. KENDALL: This was a bulletin.
22 MR. WARD: Should that have been issued as something .
23 with some regulatory force?
24 MR. KENDALL: I don't know. I guess from a--it t
25 looks to me like it was kind of an exceptional case. It was t
HERITAGE REPORTING CORPORATION -- (202)628-4888
301 1 exceptional I think in several aspects.
2 One, I think normally a more detailed review is 3 done.
4 I think secondly, the contents of this bulletin 5 were, I think we would consider it the most significant 6 bulletin probably ever issued in the instrumentation control 7 systems area. That's probcbly a pretty broad title to give 8 it, but--
9 HR. CALVO: Let me say something. At the time when 10 the bulletin was issued, like we accepted the utilities' 11 response to the bulletin at a face value. We do that 12 voluntarily following TMI like the SPDS system, and sometime 13 after that, when we started looking into the implementation, O 14 we found out that the implementation of the bulletin was not 15 the same way the staff interpreted it. I think that was found 16 out at Rancho Seco. "
17 Actually what we do here, we should go back and we 18 look in view of the fact of Rancho Seco--we want to be sure we 19 properly understand all the implications that the actions that 20 was recommended by the bulletin. It has a kind of slow time 21 delay before we got into it. I guess Rancho Seco threw it to 22 the front, whether we should revisit the bulletin to be sure 23 it is understood now it should be properly implemented.
24 That's what we are doing now. We also intend to follow the 25 thing up. We want to be sure we do not miss anything.
HERITAGE REPORTING CORPORATION -- (202)628-4888
302 1 HR. WARD: Okay, but in the meantime, we had an 2 incident which it turned out didn't have any, well, had no 3 public health and safety consequences. If it had, had public 4 health and safety consequences, I should think there might be !
5 questions about whether this important information should have 6 been dealt with in a different way.
l' 7 HR. CALVO: The Rancho Seco, the problems that 8 occurred up there, you had some of the safety systems in i
9 place. The consequences would have been acceptable. This i
10 happened to all the plants, I think would have had the l
11 transient. I think the safety systems confounded the i
l 12 consequences. The question is we don't know. Now the owners l
13 are aware how the Rancho Seco review has been done, how the O 14 issues have been identified. We think we should go back and 15 relook at how the thing was interpreted, see whether they had 16 misinterpreted like they had done at Rancho Seco. l i
l 17 HR. WARD: Okay, i 18 HR. KENDALL: Another interesting sidelight is at 19 one time, the staff had proposed that sentiment to the 20 bulletin, but through the course of activities, whatever, we 21 never issued it.
22 The next slide is just documents, the correspondence 23 that took place from the staff to the owners group. From the
! 24 very beginning, we focused our review efforts on resolution of 25 concerns that seemed to be common to previous ICS/NNI loss of HERITAGE REPORTING CORPORATION -- (202)628-4888 ,
. _ . _ _ _ _ __ ._ _ . _ _ _ _ .- - , _ . , . _ _ _ _ _ _ _ --____ _ _ _J
i 303 1 power incidents. The letters really consist of feedback I 2 letters from meetings. We had about four major meetings with 3 the owners group INC committee. We wrote, we wrote feedback 4 letters following three of those meetings. We also wrote 5 several of tL se letters, requests for information. The 6 requests were geared toward obtaining sufficient information 7 for the staff to determine whether the previous identified 8 concerns--IE 79-27, NUREG 0667, 1195, whether those types of 9 concerns were being resolved through the owners group efforts.
10 As far as the staff's evaluation findings, we found 11 that implementation of the owners group recommendations will 12 help to resolve a number of concerns identified from the
(} 13 investigation of events, things like equipment responses to 14 loss of power, reduction in operator burden, prevention of i
15 single instrument or power supply failures from being able to i 16 initiate transients, accurate indications provided in the 17 control room of losses of power. We felt that implementation 18 of tha recommendations would go a long way to resolving a 19 number of concerns. We put both approved and proposed 20 recommendations there because we felt that some of the 21 proposed recommendations were necessary. I believe the 22 recommendation to eliminate mid-scale failure systems, j 23 proposed recommendation, I don't believe that it is one that 24 has been approved b'f the Steering Committee and put forth as a 25 level 1 recommendation.
HERITAGE REPORTING CORPORATION -- (202)S28-4888
_ _ . . _ _ _ _ . .~_ _ _ . , _ _ _
304 1 We believe that clearly with implementation of the 2 recommendations, with implementation of the recommendations, 3 that the ICS/NNI contribution to reactor trip frequency and 4 transient complexity should be reduced or will be reduced.
5 Future loss of ICS/NNI power events are expected to be less 1
6 severe.
7 Another thing that we felt good about in the owners 8 group effort was that they placed, seemed to place a lot of 9 emphasis on dictating the plant response to lots of ICS power 10 such that the plant would go to a known safe state, placed a 11 lot of emphasis on dictating the failure modes of valves and 12 punps and IRS controlled equipment. Passed efforts were more
() 13 to improve power supply reliability, which they have done, 14 but--and are continuing to do, but despite the efforts to 15 improve power supply reliability, the losses of power still 16 seem to occur now and then, and feel their approach to make 17 sure if they lose power that the plant response will be 18 tolerable is a better approach, and an improvemant in their 19 philosophy.
i 20 MR. MICHELSON: Are there any instruments or 21 controls which upon loss of power supply fail as is?
22 MR. KENDALL: I believe some feedwater valves 23 perhaps. I don't know about fail as is. I believe fail at 50 )
24 percent.
25 MR. MICHELSON: Controls fails in as is position or j l
HERITAGE REPORTING CORPORATION -- (202)628-4888 i i
i l
305 l l
1 they fail-- 1
("] 2 MR. LIDDLE: Components could fail in that position.
%)
3 Yes.
4 MR. MICHELSON: Why don't you make those failsafe?
5 MR. LIDDLE: We work around that such as the valve 6 fails as is, say, you run back to pump, to the condition 7 where--
8 MR. MICHELSON: Instrument controls and components, 9 I know valves have this problem. How about instruments and 10 controls, controllers, instruments, pulling device? Do any of 11 them fail as is? Does it fail?
12 MR. LIDDLE: All the ICCS components would fail 13 mid-scale, on loss of power.
i 14 MR. MICHELSON: There are no--I was thinking of that 15 during some of these funny events, things failed as is. Maybe 16 not.
17 MR. LIDDLE: Recorders might.
18 MR. STALTER: If right at zero volts it failed, it 19 would stay there. l l
20 MR. MICHELSON: You don't, the recorder would lock l
21 in the last position, but none of the other devices always 22 fail in one of those three positions.
23 Thank you.
24 MR. KENDALL: We felt the owners group made a number 25 of significant recommendations. The key one, number one on 1
I HERITAGE REPORTING CORPORATION -- (202)628-4888 i
306 l i
l 1 the list, I am sure the plant goes to known safe state on loss ,
( 2 of power in the ICS or NNI. Again, we feel that that 3 resolution of IE Bulletin 79-27 concerns essentially would 4 accomplish the same thing.
5 Incorporate automatic selection of valid inputs 6 ICCS/NNI, this is where they would have to provide two 7 signals, and upon failure of one signal, ensure that the valid 1
8 signal is provided for use by the ICS. Other ones, provide 9 the operator with an smple state of control room indications i
10 and recorders. i 11 There have been some problems with the operators i 12 taking actions based on failed indications. f
{} 13 Elimination of mid-scale failures--develop 14 unambiguous back-up controls for pressure level and pressure 15 control; right now most of the pressurizer, I think all of the 16 pressurizer level and pressure control instruments are NNI-X.
4 17 That would provide another set of controls such that if you 18 lost NNI-X power you still control pressurizer level and 19 pressure.
20 Provide independent controls for reactor power steam ;
21 and feedwater, and the firs t one on the next slide, provide 22 separate subsystem for reactor coolant temperature steam i 23 pressure control. They are recommendations that would tend to 24 lessen the effects of single failures by lessening the effect l 25 of plant to within the certain portion of the control system i
HERITAGE REPORTING CORPORATION -- (202)628-4888 ;
307 1 versus introducing the failure to the control system in total.
2 Ensure equipment failure modes on losses of power, 3 making sure the turbine bypass valves, dump valves, PORVs, 4 remain closed; having main feedwater valves fail to preferred 5 positions--another significant area, very significant area we 6 think is the recommendations to provide maintenance and 7 surveillance and tuning and testing of the ICS and the NNI.
8 The root cause of at least one past event, loss of 9 NNI power event, was found to be lack of surveillance of power
, 10 supply overvoltage protection set points which drifted down 11 towards the operating voltage of the supply. We felt that 12 this would be an area of significant benefit.
(} 13 HR. MICHELSON: Clarification--the second bullet on 14 ensuring preferred equipment failure modes, is that a part of 15 the Level 1 agreement?
16 HR. KENDALL: All of the recommendations are Level 1 17 recommendations.
i 18 HR. MICHELSON: They are going to go back and fix l ,
i 19 the equipment such that it always fails in a preferred mode. #
20 That is not necessarily a safe mode, I guess just a known i
21 mode. What do you do--
22 HR. KENDALL: I think known and repeatable and 23 predictable so they know where.
24 MR. MICHELSON: Main feedwater, is this going to be 25 set such that ICS fails, that main feedwater will go closed?
HERITAGE REPORTING CORPORATION -- (202)628-4888
308 1 In other words, shut off or go wide open?
(} 2 HR. KENDALL: I am not sure. I believe some valves 3 may fail to 50 percent, but I believe most of the main 4 feedwater pumps either go to minimum speed or eventually to 5 stop running.
6 MR. MICHELSON: How do you decide what the safe mode 7 is? ;
i 8 MR. LIDDLE: All this would be incorporated as 9 essentially as part of the known safe state such that the 10 components would go to a known position, let's say repeatable, 11 but also the safe condition, since that, for feedwater, for r
12 example, would run back or the pumps would be tradeable. l p
13 MR. MICHELSON: Feedwater is safety as opposed to i O 14 maximum? t 15 MR. LIDDLE: That's correct.
, 16 MR. MICHELSON: Decide what is the safer then. You (
l j 17 fix it so it will always go to that--
l 1 ~
j 18 HR. LIDDLE: Go to that direction j l
i 19 MR. KENDALL: Reliance would be on the emergency ;
20 feedwater system to come on and control steam generator flow.
21 MR. MICHELSON: I would hope you have assured 22 isolation of main feedwater. That is probably more a danger i
23 to safety consideration. ,
l 24 MR. CALVO: It varies from plant to plant. Some .
^
i l
l 25 plants may have the capability to go to status quo position. ;
i i S
HERITAGE REPORTING CORPORATION -- (202)628-4888 -
309 1 Others may do other things, depending on'the considerations.
b s/ 2 MR. MICHELSON: Status quo mey not be very safe.
3 You are already full open feedwater, i
4 MR. CALVO: Well, at least I guess they have to be 5 evaluated to be considered.
6 MR. KENDALL: So we felt that the owners group made 7 a number of recommendations that are good recommendations that 8 will lead to increase in plant safety.
9 We did have some concerns. Most of these have 10 already been addressed by the owners group. We will give a 11 slightly different perspective on a couple of them.
12 The first one of the specific concerns is
(} 13 surveillance frequency for the ICS and NNI. equipment.
14 We feel that every other refueling outage could 15 perhaps be as long as up to four years or more. Plants are 16 gcing to extended fuel cycles now. Counting down time, the 17 time for the outage could be what we feel is excessive for en J 18 interval between calibrations of electronic equipment. It 19 just appears to us that every other fuel outage is too 20 excessive, s
21 From the information in 1919, it is not clear to us 22 that's just related to tuning of the integrated control system 23 either. It looks like it is also being, that frequency is 24 also being attached to other surveillances such as power 25 supply monitor set points, power supply overvoltage set HERITAGE REPORTING CORPORATION -- (202)628-4888
310 1 points. Whether it is being done more frequently or not, I 2 suppose it is in a number of cases. Based on ifnormation in 3 1919, we felt that that time looked excessive.
4 As far as maximizing the dependence on NNI-X, it 5 looked to us as if maximizing dependence on NNI-X, although it 6 would make a loss of NNI-Y power transient more tolerable, it 7 looked like it would also make a loss of NNI-X power more 8 severe.
{
9 The owners group indicated that upon providing j 10 hardened inputs or multiple input channels, that this concern 11 would go away because you always, you have a back-up.
12 It is not--it may be true, especially with the implementation l 13 of redundant pressurizer level and prassure control circuits.
i 14 MR. MICHELSON: Is this all going to come in as a 15 package for staff review at some future date? The things that 16 you know the utilities are going to do, are you going to do 17 any more review than you have just done?
18 HR. SIEGEL: Yes. Tomorrow we will discuss it.
19 Implementation, we will tell you what we are going to do, if 20 you want to wait until then.
( 21 MR. MICHELSON: That's fine.
22 MR. KENDALL: Elimination of some of the integrated l 23 control cystem functions--the reason they made the list, l I
l 24 especially the first two, was that they were just not, we did l
l 25 not feel there was sufficient basis provided in 1919 to jusify I
1 1
i l HERITAGE REPORTING CORPORATION -- (202)628-4888
l 311 l 1
1 the deletion of these features, i
{} 2 As far as removal of the BTU limits function, the 3 primary concern there looked like it was failure of the BTU l l
4 limit circuitry due t o invalid input signal, not really i l
5 failure of the BTU limits itself; the actual limits would take 6 given invalid input signal; given their recommendations to 7 harden the input signals such that the failure of an input 8 signal would have an effect.
9 We don't see this as a recommendation, especially 10 because the BTU limits function will run back main feedwater '
11 to help prevent overcooling for certain scenarios.
12 d
(The transcript continues on the following page )
13 O 14 15 16 17 18 19 20 21 l
. 22 )
l 23 )
I 24 25
- HERITAGE REPORTING CORPORATION -- (202)628-4888
, ~ . . - _-_ _. - _ . . - . . _ _ _ . , _ _ -
312 1 The owners group has proposed to provide a rapid 2 feedwater reduction circuit to take the place of the BTU 3 limits function.
4 However, it is not clear to us that the rapid 5 feedwater reduction circuit has a lot of, number of benefits 6 that the BTU limits function has.
7 One is it appears that the rapid feedwater reduction 8 circuit would only be active following a reactor trip, and 9 secondly--which is when you would need it most, but secondly, P
10 that for certain conditions such as a valid control circuit 11 manual, that the rapid feedwater reduction circuit would not 12 be operative.
(} 13 The next one, wiring the power supply monitor 14 directly to the plus and minus 24 volt DC buses, over the 15 years we have received a number of conflicting stories for why 16 the power supply monitor is there, what its design function 17 is. l 18 Following the Crystal River event, we were told that 19 the purpose of power supply monitor was to prevent the modules 20 from, integrated control modules, from operating on degraded 21 voltage, that the modules would not function properly on ,
22 voltages outside of the 24 volt DC with a plus or minus 10
- 23 percent tolerance, that at voltages, low voltages, degraded j 24 voltages, that some modules, you might have relays begin to 25 cycle or equipment component malfunctions.
HERITAGE REPORTING CORPORATION -- (202)628-4888
313 1 Now we have been told that the purpose of the power
(} 2 supply monitors is not to protect against degraded voltage,
. 3 only to protect against hard bus faults, but the immediate i
4 loss of power.
l 5 We find that it is kind of curious that the set :
6 point would be chosen at 22 volts for a, to detect a hard bus l
7 fault. It looks like the power supply monitor itself hau some i 8 questionable design features. We don't believe that 9 connecting it to the bus is necessarily the optimum location 10 to connect the power monitor. It would not detect faults f
i 11 downstream that could affect module, could have modules ,
12 misoperating and never detect it.
13 We think we have provided a fairly extensive O 14 write-up in NUREG 1286 which is the restart SER for the Rancho 15 Seco plant for their events, and we would like to see the 16 owners group address the issues that are listed there.
17 We think there just exist a lot of questions t
18 surrounding the power supply monitor and its operation, and 19 the information provided in 1919 doesn't help us to try to 20 resolve these concerns. FSAR assumptions realy to go ICS and 21 NNI. The ICS and NNI systems are non-safety related. Systems 22 requirements state that they are non-safety-related. They are 23 in no way used to mitigate the conseqences of transients or 24 actuate or shut down the plant. .
25 We found in review of portions of the 1919 that it
)
HERITAGE REPORTING CORPORATION -- (202)628-4888
l l
314 1 appeared that for some accidents or transient scenarios, that i
() 2 the ICCS was assumed to function properly to mitigate the l
3 effects of the scenario. We realized that this particular ,
i 4 portion of the owners group review is preliminary to the ,
5 ongoing effort, still looking at it. Weeel that they should
,; 6 verify that no credit is taken for ICS or NNI performance in 7 the accident analyses.
8 The 79-27 concerns were not specifically addressed 9 by the owners group. Portions of the items addressed by the 10 Bulletin 79-27 did appear in certain areas and were covered by
~
11 recommendations, but the overall scope and focus of Bulletin 12 79-27 was not addressed. We were told that it was a t
(} 13 plant-specific item, not a generic item. It was not within 14 the safety performance improvement program scope. We really 15 believed the need still exists to verify that the-Bulletin 16 79-27 concerns have been resolved at the plants.
17 (Slide) 18 MR. KENDALL: Several more general concerns--we
- i 19 thought the guidance provided in some of the recommendations l l
20 was too general to ensure resolution of the concerns. We felt 21 that sometimes the recommendations did not drive a proper 22 source document. Apparently that has been corrected now.
j 23 Other thing was that we could not conclude based on 24 the information provided in 1919 that all of the staff 25 concerns identified from the investigations of previous events l l
I HERITAGE REPORTING CORPORATION -- (202)628-4888
315 ,
1 would be resolved for implementation of recommendations, i
( 2 Essentially all of these concerns are again summed up in 3 79-27, things like mid-scale failures, inadequacies of back-up 4 instruments, and reference and use of back-up instruments by ;
5 the procedures that 6 have not been addressed in detail.
7 (Slide) 8 MR. KENDALL: The last slide on summary of staff f 9 conclusions--we thought that the B&W owners group review plan 10 for the ICS?NNI evaluation was an excellent, excellent plan.
11 It took advantage of all the information that seems to be 12 available as input for developing system design requirements.
~
(} 13 It appeared to be well thought out and to be appropriate for 14 producing recommendations that would improve plant response to 15 ICS/NNI failures, and that would lead to a reduction in i 16 reactor trip frequency and transient complexity due to ICS/NNI 17 failures, and that's basically what the second bullet said, '
18 that implementation of recommendations will be successful in 4
19 achieving reduction in trip frequency and transient
. 20 complexity.
21 We found that information with the information i
22 provided in B&W 1919 was not sufficient to conclude that some l 23 of the concerns identified from previous events had been
, gs 24 resolved. The staff tried to obtain the information through i
25 feedback letters and requests for information, but we ended :
! )
i HERITAGE REPORTING CORPORATION -- (202)628-4888
l 316 1 up being able to arrive at the positive conclusion for some of l l
() 2 the concerns. We really, based on lack of information, did 3 not know whether certain things had ben resolved such as 79-27 ;
4 concerns. The owners group I believe feels that they have all l
5 addressed the bulletin one time and that they feel confident i l
6 that the concerns have been resolved.
I 7 What we have proposed to do is audit the ;
1 8 implementation or the resolutions, the 79-27 concerns at B&W l
}
9 plants. We are developing an audit plan where we can go and 10 look at the bus structure and bus load and try to make a i 11 determination of whether the plant design, plant procedures 12 are adequate. , i 1
13 MR. ETHERINGTON: There is no additional 14 recommendations you feel strongly about?
15 HR. RANDALL: I would say outside of 79-27, that ,
1 16 that's correct. You mean recommendations that we have in 4
1 17 addition to what the owners group has?
l 18 MR. ETHERINGTON: Yes. ,
19 MR. RANDALL: I think that's correct. It appears ,
4 20 that the owners group effort wille successful in achieving -
21 significant improvements in design and ICS/NNI operation, and 22 we feel that a combination of implementation of this SPIP i
23 recommendations and verification through plant audits, proper !
24 resolution of the 79-27 concerns would in general give the l 25 staff the confidence, give us the confidence that ICS/NNI i
HERITAGE REPORTING CORPORATION -- (202)628-4888
I 317 1 concerns had been reso1ved in B&W p1 ants, that upon 1oss of O 2 1cs or NN1 po.er, the p1 ant .em1d ineeed achieve a xno.n sate ;
)
3 state. i 4 CHAIRMAN WYLIE: Any questions? Let's take a
~ '
5 ten-minute break.
6 (Whereupon, at 6:35 p.m., a recess .as taken, and it ,
i 7 .as decided by the Chairman during that recess to adjourn the 8 meeting.)
9 10 11 12 13 14 15 16 17 18 l
19 l l
i 20 21 22 23 24 O 25 HERITAGE REPORTING CORPORATION -- (202)628-4888 ,
1 CERTIFICATE 2
3 This is to certify that the attached proceedings before the 4 United States Nuclear Regulatory Commission in the matter of: ,
5 Name: ACRS--Subcommittee on B&W Reactor Plants 6
7 Docket Number:
8 Place Washington, D.Cs 9 Date: May 3, 1988 10 were held as herein appears, and that this is the original 11 transcript thereof for the file of the United States Nuclear 12 Regulatory Commission taken stenographically by me and,
~
13 thereafter reduced to typewriting by me or under the direction 14 of the court reporting company, and that the transcript is a
() 15 true and accurate record of the foregoing proceedings.
16 /S/ $I//w' - e, 17 (Signature typed): Catherine S. Boyd 18 Official Reporter 19 Heritage Reporting Corporation 20 21 22 23 24 25 O' Heritage Reporting Corporation (202) 628-4888
I
() B&W OWNERS GROUP SAFETY AND PERFORMANCE IMPROVEMENT. PROGRAM PRESENTATION TO ACRS
. MAY 3, 1988 e
ARKANSAS POWER & LIGHT CO.
C)
DUKE POWER CO.
FLORIDA POWER CORPORATION GPU NUCLEAR CORPORATION
- SACRAMENTO MUNICIPAL UTILITY DISTRICT TENNESSEE VALLEY AUTHORITY TOLEDO EDISON CO.
BABC0CK & WILCOX CO.
O 1
B&WOG PRESENTERS O ,
l N. A. RUTHERFORD ... CHAIRMAN, l B&WOG STEERING COMMITTEE l
G. R. SKILLMAN...... CHAIRMAN, SPIP l S. T. R0SE..........P.M., SENSITIVITY STUDY i
L. C. STALTER....... CHAIRMAN, I&C COMMITTEE 4
0 O
i
! 2
I
() PRESENTATION OUTLINE OBJECTIVE OF MEETING GENERAL OVERVIEW / INTRODUCTION ISSUE DEFINITION INFORMATION GATHERING
SUMMARY
OF PRIMARY FINDINGS SPIP OPEN ITEMS ,
DISPOSITION OF PREVIOUS ACRS CONCERNS SYSTEMS REVIEW SENSITIVITY STUDY AREAS OF SIGNIFICANT NRC COMMENT 1
l 0 ICS/NNI l
, 1 o l HUMAN FACTORS ,
RISK ASSESSMENT REVIEW SPIP ADMINISTRATION FOLLOW-THROUGH AND IMPLEMENTATION O CONCLUDING REMARKS
! 3
i O
I PAST ACRS AND NRC MANAGEMENT INVOLVEMENT IN SPIP 1
ACRS SUBCOMMITTEE MEETING ON B&W PLANTS 6/25/86 B&WOG (H.B. TUCKER) FOLLOW-UP LETTER TO 7/10/86 ACRS (C.J. WYLIE)
ACRS LETTER (D. WARD) TO NRC (V. STELLO) 7/16/86 REGARDING SPIP O
- NRC LETTER (H.R. DENTON) TO ACRS (D. WARD) 8/14/86 REGARDING SPIP ACRS FULL COMMITTEE MEETING ON B&W PLANTS 9/12/86 B&WOG (H.B. TUCKER) FOLLOW-UP LETTER TO 10/20/86 ACRS (C.J. WYLIE)
B&WOG PRESENTATION TO NRC COMMISSIONERS 11/6/86 B&WOG PRESENTATION TO NRC COMMISSIONERS 8/5/87 l l
O
! 4
w,, ,_A a s..a a _ap-+ s- a-e-- ras. 2 .a,. a-., - - ,a ,, a m --g a--- - -- - --_a _---- a--me-+- -- m.sa -- --_ . - - - ,-w ------~- - - - - - - - -
O j
MEETING OBJECTIVE TO PROVIDE SUFFICIENT INFORMATION TO ALLOW THE ACRS TO
- CONCLUDETHATTHEB&WOGSAFETYANDPERFORMANCEIMPROVE-4 O MENT PROGRAM ADEQUATELY ADDRESSED THE COMPLEX TRANSIENT AND TRIP-RELATED SAFETY CONCERNS ON B&W PLANTS.
i 1
l 1
l i
i O
l l 5 t 1
,. - .-- , ---,,.. , . - --.. ,- - . _ .- n.-
- l l l
l l
O l
- i OVERVIEW OF SPIP EXECUTION o ISSUES WERE QUANTITATIVELY DEFINED.
o IMPROVEMENT GOALS WERE SET.
l 0 STRONG EXECUTIVE LEVEL COMMITMENT TO PROGRAM.
o B&WOG WANTED A PROGRAM WITH TANGIBLE RESULTS. '
1 o PROGRAM SCOPE DEVELOPED BY B&WOG AND AGREED TO BY STAFF. ;
Oo PROGRAM HAD STRONG SAFETY ORIENTATION.
o PROGRAM BASED ON HISTORICAL FACTS AND POTENTIAL PLANT PERFORMANCE.
o ENGAGED INDEPF.NDENT SUPPORT AND TECHNICAL OVERVIEW.
o IMPLEMENTATION HAS AND WILL CONTINUE TO IMPROVE SAFETY AND PERFORMANCE.
o COMPLEMENTS AND REINFORCES OTHER INDUSTRY SAFETY AND PERFORMANCE IMPROVEMENT INITIATIVES.
i O 6
l
~
ISSUE DEFINITION
()
ISSUE: COMPLEX TRANSIENTS WERE T00 FREQUENT AND THEIR SAFETY SIGNIFICANCE WAS QUESTIONED.
COMPLEX TRANSIENTS WERE DEFINED IN TERMS OF SIX MEASURABLE PARAMETERS.
(PREVIOUSLY DESCRIBED TO ACRS) THEIR DEVIATION FROM PREFERRED RANGES WAS USED AS AN INDICATION OF COMPLEXITY.
QUANTIFICATION OF THE ISSUE PERMITTED PINPOINT.ING THE FUNCTIONAL AND SYSTEM AREAS OF INTEREST.
INTERACTIONS WITH NRC STAFF, ACRS AND OUR INDEPENDENT ADVISORY BOARD HELPED FOCUS THE ISSUE.
DEFINITIONS: TRIP - THE UNPLANNED DE-ENERGIZING OF THE CONTROL ROD DRIVE CONTROL SYSTEM THUS INSERTING CONTROL RODS INTO THE REACTOR'S CORE.
TRANSIENT - THE AGGREGATE DYNAMIC PLANT BEHAVIOR FOLLOWING A TRIP.
7
O B&W OWNERS GROUP SAFETY AND PERFORMANCE i
IMPROVEMENT PROGRAM OBJECTIVE IMPROVE SAFETY, REDUCE THE NUMBER OF TRIPS AND COMPLEX
- TRANSIENTS ON B&W OWNERS GROUP PLANTS, AND ENSURE ACCEPTABLE PLANT RESPONSE DURING THOSE TRIPS AND '
TRANSIENTS WHICH DO OCCUR.
j GOALS
- 1. BY THE END OF 1990 THE AVERAGE PER PLANT TRIP FREQUENCY WILL BE LESS THAN TWO PER YEAR.
(CATEGORY "C") WILL BE REDUCED TO 0.1 PER PLANT PER YEAR BASED ON A MOVING THREE YEAR AVERAGE.
i 1
, l
! 8 I
! O O O 1 ;
l SPIP Program Process i 1. Information Gathering II. Integration Hl. Implementation i '
I
! l 4
.. l .
- .. 1 l
! I I ERISTI.E S.II.E
.l. l ,
- I _, ,
g i,5, -
i=1aca 1 . .o .
- l ...... 5=g.g .. -
+ -
l ,
.'l1." . .- -
T
_, -. . -. 5 cs -- s, .s.-
_ . ,o. .. .
P.OMC T S Ca.IIIME 2 h l
a l 1
I ses.vstws l I',',g',',f
& II.IP5I.T I
I ,MCIS
! l I .
mestu I
L -
si m .r
, mi-mu m Siss5111TIIT l l i I t, .
l $_ E 15 1 I maatsue.s
-o l i
i
!O I
INFORMATION GATHERING PROCESS o 0BJECTIVE: TO DETERMINE AREAS FOR IMPROVEMENT o PERFORMED BROAD AND COMPREHENSIVE SEARCH FOR PROBLEMS; BOTH IN NSS AND BOP e
! REVIEWED TRANSIENT ASSESSMENT PROGRAM I DATA O -
REVIEWED SYSTEM AND COMPONENT DESIGNS INTERVIEWED OPERATIONS AND MAINTENANCE
- PERSONNEL j -
REVIEWED OTHER PERTINENT DATA i
j o EMPLOYED OUTSIDE CONSULTANT TO ASSESS RELATIVE B&W PLANT SENSITIVITY I o PERFORMED A PROBABILISTIC RISK ASSESSMENT REVIEW O 10
- . ~ . . , . . . , _ _ . - - _ _ _ . - - _ . _ . _ . . , , - _ _ _ . _ . _ - . . - , _ _ _ _ . , , _ . _ _ , , . _ , _ _ _ _ _ - _ _ , _ , , _ - _ _ _
- O i
REVIEWED THE B&WOG TRANSIENT ASSESSMENT PROGRAM (TAP) DATA l' -
REVIEWED AND S0RTED THE TAP DATA TO BETTER FOCUS ON AREAS WHERE IMPROVEMENT IS NEEDED i
1 -
DEFINED SPECIFIC MEASURABLE PARAMETERS TO i O ,' GRADE" THE COMPLO(ITY FOR TRANSIENTS USED THE "GRADED" TRANSIENT TAP REPORTS TO
. IDENTIFY SPECIFIC CONCERNS FOR ACTION
)
i 11 i
1
- O l
1 i
REVIEWED SYSTEM AND COMPONENT DESIGNS A PERFORMANCE AND RELIABILITY REVIEW 0F SYSTEMS /
- COMPONENTS WAS UNDERTAKEN TO IDENTIFY RECOMMENDATIONS FOR IMPROVEMENT.
1 #
IDENTIFIED SYSTEMS:
O 1. ICS/NNI SYSTEM
- 2. MAIN FEEDWATER SYSTEM
- 3. EFW/ AUXILIARY FEEDWATER SYSTEM i 4. SECONDARY PLANT RELIEF SYSTEMS
- 5. INSTRUMENT AIR SYSTEM
^
1 l
l l
!O
! 12 i
I__-_._______-..___.
l !
I i i l
!O l t
p INTERVIEWED OPERATIONS AND MAINTENANCE PERSONNEL USED INPO DEVELOPED FOCUSED INTERVIEW PROCESS PERFORMED FOCUSED INTERVIEWS WITH OPERATIONS 4
AND MAINTENANCE PERSONNEL FROM EACH PLANT.
LOOKED FOR TRANSIENTS WHICH MAY NOT HAVE RESULTED IN REACTOR TRIPS BUT WERE CONSIDERED TO BE O SIGNIFICANT I
LOOKED FOR AREAS WHERE IMPROVEMENTS COULD REDUCE THE NEED FOR POST-TRIP OPERATOR ACTIONS TRACKED DOWN AND DEFINED CONCERNS WITH PROCEDURES, MAINTENANCE, AND OPERATOR BURDEN l l I
.i l
l O u I
l i
O ;
i i
REVIEWED OTHER PERTINENT DATA l IDENTIFIED TRANSIENT OPERATIONAL HISTORY AND i
ANALYTICAL RESULTS FROM DOCUMENTS SUCH AS:
UTILITY INTERNAL EVENT REPORTS NRC INFORMATION INP0 INFORMATION
! O -
PRE-TAP OPERATIONAL DATA OTHER OPERATIONAL HISTORY DEVELOPED A PLAN FOR REVIEW OF THIS INFORMATION i
CONDUCTED THOSE REVIEWS
)
i I
1 O 14 i
I - - -- _ _ - - . _ _ _ _ -_
,A _m _ a - m a ____ ._.... 3 om m-.
l O 1 i
l 1 l l
TO BE COVERED AS SEPARATE' AGENDA ITEMS :
4
- O -
SENSITIVITY STUDY l
l RISK ASSESSMENT REVIEW 4
i 1
l 1
.i 1 4
) O i
15 !
i j
- _ _ _ - _ . , --. ..._- - - - . _ _ - _ ._- _ - ._\
i
- O 1
ELEMENTS COMPRISING THE SPIP
- 1. ICS/NNI SYSTEM REVIEW
- 2. MFW SYSTEM REVIEW
- 3. EFW/AFW SYSTEM REVIEW
- 4. SECONDARY PLANT RELIEF SYSTEM REVIEW 5.
OPERATING EXPERIENCE REVIEW
- 6. REVIEW 0F OTHER TRANSIENT INFORMATION
- 7. OPERATIONS / MAINTENANCE PERSONNEL INTERVIEWS
- 8. SENSITIVITY STUDY O 9. RISK ASSESSMENT
- 10. NRC INTERACTION
! 11. INSTRUMENT AIR SYSTEM REVIEW l 12. TRIP INITIATOR REVIEW j 13. OPERATOR BURDEN REVIEW
- 14. INDEPENDENT ADVISORY BOARD I
- 15. SPRIG (SAFETY AND PERFORMANCE RECOMMENDATION 4
INTEGRATION GROUP) i
!O 16 I
4_.._ _ _ _ _ . _ . - - - _ . _ . _ _ _ _ - - _ _ _ _ . _ _ _ _ _ _ . _ _ _ . - . . _ _ _ _ _ _ _ _ . _ - . __ _ _ _
6
! O l l
1 INDEPENDENT OVERVIEW OF PROGRAM l
0 B&WOG EXECUTIVES ESTABLISHED AN INDEPENDENT !
! ADVISORY BOARD.
) J o MISSION OF ADVISORY BOARD l
l l -
EVALUATE PROGRAM COMPREHENSIVENESS ,
AND SAFETY EMPHASIS EVALUATE ABILITY OF PROGRAM FOR ACHIEVING i OBJECTIVES O
o ADVISORY BOARD MEMBERS:
W. H. LAYMAN (EPRI)
- S. LEVY (S. LEVY ASSOC. , INC.)
- N. E. TODREAS (MIT)
R. S. BRODSKY (BETA) l 0 INVOLVEMENT: SEVEN, 2-3 DAY MEETINGS l -
i 1
i O 17 .
1 l l _- - _ - - - . _ - - -_ _ _ _ _ - . _ - - __
O SAFETY AND PERFORMANCE RECOMMENDATION INTEGRATION GROUP (SPRIG) e o FORMAL CHARTER FROM EXECUTIVE COMMITTEE O o MULTI-DISCIPLINED REPRESENTATIVES FROM EACH B&WOG MEMBER o CONCENTRATED EFFORT AFTER ESSENTIALLY ALL RECOMENDATIONS WERE AVAILABLE O is
O SPRIG FUNCTIONS INTEGRATED EFFECT ON PLANT BEHAVIOR i o REVIEWED EACH EXISTING SPIP RECOMMENDATION FROM SEPARATE SPIP PROJECTS FROM THE STANDPOINT OF OVERALL FUNCTIONAL PLANT BEHAVIOR. ,
l PRIORITIZATION o SELECTED FROM TOTAL RECOMMENDATION PACKAGE THOSE DEEMED MOST IMPORTANT AND BENEFICIAL TO SAFETY AND PERFORMANCE IMPROVEMENT. '
DIV.ERSE PERSPECTIVES o REVIEWED TOTAL RECOMMENDATION PACKAGE FROM PERSPECTIVES OF:
OPERATIONS SAFETY DESIGN MAINTENANCE i
REPORT o PROVIDED REPORT TO EXECUTIVE COMMITTEE.
O 19
O O O B&WOG SAFETY AND PERFORMANCE IMPROVEMENT PROGRAM i
MULTISYS OPERATIONS AND REVIEW l MAINTENANCE INTERVIEWS DY RISK ASSESSMENT l
INDEPENDENT
^
f BROAD
- REASSESSMENT l
1 OF PAST SIX YEARS NRC STAFF INPUT
, OPERATING i
EXPERIENCE
/
i OPERATOR IAB INPUT BURDEN SPRIG INPUT
- 2 - - - - - - - - - - - - - - - - _ ---_ _ -___ - - __ - - - _
O
SUMMARY
OF PRIMARY FINDINGS'AND CONCLUSIONS O
i O ,
i n,,,. , - - ..- ,--y,. , _ _ , , ,, _,vgg.,, ,- ,---y,n-n _ -p -
O FINDINGS AND CONCLUSIONS CATEGORIES o PLANT SAFETY AND DESIGN l
0 COMPLEX TRANSIENTS O
o TRIP INITIATORS 0 SPIP VALUE 22 l
FINDINGS AND CONCLUSIONS REGARDING B&W PLANT SAFETY AND DESIGN
()
o RISK 0F CORE DAMAGE IS COMPARABLE TO OTHER PWR DESIGNS o AREAS OF SENSITIVITY ARE DIFFERENT THAN OTHER PWRs o PRESSURIZER SIZE IS ADEQUATE o OTSG INVENTORY IS ADEQUATE o OPERATOR BURDEN IS ACCEPTABLE o INTEGRATION OF CONTROL FUNCTIONS IS APPROPRIATE o PLANT IS MORE RESPONSIVE TO SECONDARY SIDE
() CHANGES o RECOMiENDATION IMPLEMENTATION WILL FURTHER IMPROVE SAFETY I
() 23 WNC f w T- ' W 9 "'WWW7'F-T
()
FINDINGS AND CONCLUSIONS REGARDING COMPLEXITY OF TRANSIENTS o
COMPLEX. TRANSIENTS ARE PRINCIPALLY THE RESULT OF MISMATCHED HEAT BALANCE BETWEEN PRIMARY AND SECONDARY SYSTEMS o OVERHEATING EVENTS HIGHER SAFETY SIGNIFICANCE THAN OTHER EVENTS INFREQUENT COMPARED TO OVERC00 LING o OVERC00 LING EVENTS MAY BE PRECURSOR TO OVERHEATING DOMINATED BY HISBEHAVIOR OF TWO SYSTEMS SECONDARY PLANT RELIEF EMERGENCY FEEDWATER OTHER SYSTEMS MISBEHAVIOR IMPORTANT ICS/NNI MAIN FEEDWATER
o O O
- Contributors to Post-Trip Complexity DIllERS (8.97.)
IH (2.5%)
ELE (5.1%) EFu (29.1%)
1 N
\ %.
'{%
N }
\ N hg
\\
SSPC ( 31.6% '- ' 'k 11fil (10.1%)
/
N v
\
) -
ICS/NNI (12.7%)
l O ,
m W M
~ H 2 z O W l
(.D > l fJ o\o a - e
-J C
m l Q C J W U w ,
c ,
> l C C O
Om C
C o\*
2 N L.)
I W J J
H H Z b L,j N .
~
m Z
w
- u. co l
1 c as W g l
2
~ s-5 '
- y. -
x >
w W J
1 I C.5 E E O M U H b
=
0 2 W
C 26 I
~
l O l l
l 1
FINDINGS AND CONCLUSIONS REGARDING TRIP INITIATION o MAJORITY OF TRIPS CAUSED BY BALANCE OF PLANT SYSTEMS / COMPONENTS OR PERSONNEL i o TURBINE GENERATOR DOMINANT TRIP INITIATOR BEFORE 1 1985 O ,
o DOMINANT TRIP INITIATOR CURRENTLY MAIN FEEDWATER i o OTHER FREQUENT TRIP INITIATORS t
ICS/NNI INPUTS ELECTRICAL O 27 l
._-..--.-____...._--..____.___._.._-...----1
O FINDINGS AND CONCLUSIONS REGARDING OVERALL SPIP VALUE o FOCUSED GREATER ATTENTION ON B0P o REINFORCED VALUE OF QUALITY TAP DATA BASE FOR OPERATING EXPERIENCE o FURTHER CONFIRMED VALUE OF A GROUP EFFORT ON COMMON CONCERNS O
o SAFETY AND PERFORMANCE HAVE ALREADY BEEN IMPROVED AND WILL CONTINUE TO BE IMPROVED o SPIP VALUE~HAS BEEN ENHANCED BY DIVERSE INPUT EXTERNAL TO B&WOG o SPIP VALUE INCREASED BY STRONG EXECUTIVE INVOLVEMENT O 28
()
FINDINGS OF THE INDEPENDENT ADVISORY BOARD o SPIP WAS EFFECTIVE IN IDENTIFYING AREAS FOR IMPROVEMENT o SPIP EXAMINED ALL FACTORS INVOLVED IN PAST COMPLEX TRANSIENTS o TAP DATABASE EXTENSIVE AND VALUABLE o DEFINED ACTIONS SHOULD REDUCE TRANSIENT FREQUEi1CY AND IMPROVE SAFETY
() o CONCURRED WITH PROCESS AND BASIS OF PRIORITIZATION o SPIP G0ALS ACHIEVABLE, BUT SCHEDULE IS AMBITIOUS; MANAGEMENT ATTENTION REQUIRED o SOME RECOMMENDATIONS LACK SPECIFICITY o MEANS TO ASSURE ACTION QUALITY NEEDED o ICS/NNI AND FEEDWATER RELIABILITY DESERVE SPECIAL ATTENTION -
() 29
1 1
0 SPIP "0 PEN" AREAS O
l
\
O 30
, SPIP ACTIVITIES / PROJECTS ON WHICH B&WOG AND NRC
() STAFF ARE IN SUBSTANTIAL AGREEMENT INDIVIDUAL PROJECTS MFW SYSTEM REVIEW EFW/AFW SYSTEM REVIEW SECONDARY PLANT RELIEF SYSTEM REVIEW OPERATING EXPERIENCE REVIEW REVIEW 0F OTHER TRANSIENT INFORMATION OPERATIONS / MAINTENANCE PERSONNEL INTERVIEWS SENSITIVITY STUDY RISK ASSESSMENT NRC INTERACTION INSTRUMENT AIR SYSTEM REVIEW TRIP INITIATOR REVIEW OPERATOR BURDEN REVIEW INDEPENDENT ADVISORY BOARD SPRIG (SAFETY AND PERFORMANCE RECOMMENDATION INTEGRATION GROUP)
() 31
.,.,,-e--. ---..-._---.-e, , . , - - - - - - - - - , - , - . . , - - - . - - , , _ , . _ . . - . _ _ . , - - - . - _ _ - . , _ _ _ _ , _ - _ _ , , , . . . _ , _ _ - , _ _ . . , . . , _ _ , , . , _ _ _ , , , - , , , , , . - , - . , -
> ." s e SPIP ACTIVITIES / PROJECTS ON WHICH B&WOG AND NRC.
STAFF ARE IN SUBSTANTIAL AGREEMENT OVERALL PROGRAM ,
PROGRAM SCOPE AND BREADTH PROGRAM PROCESS RECOMMENDATION CONTENT PROGRAM MANAGEMENT AND ADMINISTRATION B&W PLANT SAFETY
() 32 t
O SPIP ACTIVITIES / PROJECTS INVOLVING DISAGREEMENT o ICS/NNI (CONTENT)
O o USE OF HUMAN FACTORS EXPERTISE (PROCESS) l O 33
O OPEN AREAS PLANNED CLOSURE
- 1. IMPLEMENTATION PLANS / SCHEDULES JULY 1988
- 2. EMERGENCY OPERATING PROCEDURES AUGUST 1989 REVIEW REPORT 4
- 3. TAP ANNUAL REPORT JULY 1988 O 4. VALVE TASK FORCE REPORT AUGUST 1988
- 5. APPLICABLE PORTIONS OF PERIODIC RECOMMENDATION TRACKING UPDATE SYSTEM REPORT l
l i
34 O
i O ,
l l
l PREVIOUS ACRS CONCERNS (FROM JULY 16, 1986 LETTER)
- 1. EFFECT OF OPERATING ORGANIZATION ON SYSTEM PERFORMANCE ,
g 2. SAFETY SIGNIFICANCE OF UNIQUE B&W PLANT CHARACTERISTICS
- 3. ATTENTION TO DECAY HEAT REMOVAL l
l l
35 O
()
B&WOG DISPOSITION OF ACRS CONCERNS
- 1. OPERATING ORGANIZATION EFFECT A) B&WOG CONSIDERS MANAGEMENT ROLE VERY IMPORTANT TO SAFE OPERATION (FURTHER DISCUSSED IN B&WOG LETTER TO ACRS 10/20/86)
B) B&WOG EXECUTIVE COMMITTEE MEMBERS RECOGNIZED THEIR INTERDEPENDENCE; THIS LED TO NUMEROUS OVERSIGHT AND AUDIT ACTIONS IN SPIP e
C) NRC CONCURS THAT MOST MANAGEMENT MATTERS ARE BEST TREATED ON PLANT-SPECIFIC BASIS -
() H.R. DENTON LETTER TO D. WARD 8/14/86
- 2. SAFETY SIGNIFICANCE OF UNIQUE B&W PLANT CHARACTERISTICS A) MAJOR THRUST OF SENSITIVITY EVALUATION AND RISK ASSESSMENT REVIEW 1
B) SEPARATE PRESENTATIONS WILL ADDRESS THESE MATTERS c)
B&WOG DISPOSITION OF ACRS CONCERHS (CONTINUED)
- 3. ATTENTION TO DECAY HEAT REMOVAL A) IN RESPONSE TO THIS ITEM, FULL ACRS WAS GIVEN A BROAD DHR PRESENTATION B) DESCRIBED THREE LINES OF DEFENSE; MFW AFW/EFW FEED AND BLEED C) SPIP HAS ADDRESSED MFW AND AFW D) FEED AND BLEED CAPABILITY WAS JUDGED ADEQUATE; PROCEDURE RECOMMENDATIONS RELATED TO FEED AND BLEED WERE DEVELOPED
() E) THE FOLLOWING WAS STATED IN THE 9/12/86 PRESENTATION TO ACRS "THE ISSUE OF HEAT PRODUCTION / REMOVAL IMBALANCE AND RELIABLE DECAY HEAT REMOVAL CAPABILITY IS RECOGNIZED AND ACCEPTED AS A DOMINANT ISSUE. A NEED FOR ADDITIONAL DECAY HEAT REMOVAL CAPABILITY IS CONSIDERED UNNECESSARY."
37 C)
- -m-------..------,-..~,.-_.--v,, - - - - - - , -----.-,,---~,-,--m,,,,-, r---v~.,--,,.-,.----c,-... -+-,- - - ----ww--------,,---,---
C)
MFW SYSTEM REVIEW PROCESS FOCUS ON MFW EVENTS OF 1984-85 INFORMATION GATHERED ON EACH PLANT'S OPERATING PROCEDURES AND CHARACTERISTICS DESIGN MAINTENANCE PRACTICES ,
SITE VISIT (ONE WEEK)
() -
INTERVIEWS WITH OPERATIONS AND MAINTENANCE PERSONNEL SYSTEM "WALK DOWN" REVIEW 0F RECENT TRANSIENT DATA AND PROBLEMS COMPILATION OF FINDINGS INTO 42 RECOMMENDATIONS l WITH GENERIC OR PLANT SPECIFIC APPLICABILITY 38 l C)
()
MAIN FEEDWATER (MFW) AND CONDENSATE SYSTEM o IMPROVE RELIABILITY OF MAIN FEEDWATER AND CONDENSATE SYSTEMS TYPICAL KEY RECOMMENDATIONS (1) IMPLEMENT A PROGRAM TO IDENTIFY IMPROVEMENTS NEEDED IN MFW PUMP CONTROL SYSTEMS INCLUDING THE ICS.
EVALUATE THE INTERACTION BETWEEN THESE TWO SYSTEMS (2) CORRECT MFW PUMP CONTROL PROBLEMS (3) ENSURE THAT A SINGLE ELECTRICAL FAILURE IN THE MFW AND CONDENSATE SYSTEMS WILL NOT CAUSE A LOSS OF BOTH FEEDWATER TRAINS (4) ELIMINATE UNNEEDED TRIP FUNCTIONS ON THE MFW PUMPS (5) ELIMINATE AUTOMATIC CONTROL OF THE MFW BLOCK VALVE EXCEPT DURING A REACTOR TRIP (6) PROVIDE CAPABILITY TO OVERRIDE A CLOSE SIGNAL TO THE MFW BLOCK VALVE (7) PROVIDE AUTOMATIC HFW OVERFILL PROTECTION l CAPABILITY (8) INSTALL MONITORING SYSTEM ON MFW PUMPS TO IDENTIFY l CAUSE OF TRIP.
() 39
O EMERGENCY FEEDWATER (EFW) REVIEW PROCES3 o COMPILED FUNCTIONAL DESIGN OBJECTIVES Oo COMPILED TESTING OBJECTIVES o COMPILED A LIST OF MAINTENANCE RECOMMENDATIONS TO IMPROVE RELIABILITY / AVAILABILITY O
40 l
___ - ,,., _._. ,.~ _ _.._____. __
i ()
l l
EMERGENCY FEEDWATER (EFW) o EXCESSIVE EFW FLOW HAS RESULTED IN OVERC00 LING EVENTS TYPICAL KEY RECOMMENDATIONS (1) REMOVE EFW INITIATION AND CONTRROL FROM ICS/NNI (2) LIMIT EFW FLOW RATE OR FILL RATE (3) EXTEND START TIME FOR EFW TURBINE DRIVEN PUMPS (4) ENSURE MAINTENANCE AND TEST PROGRAMS CONFIRM COMPONENTS READY FOR SERVICE (5) REDUCE SPURIOUS EFW ACTUATIONS I
i 41 !
O l
l SECONDARY PLANT PRESSURE CONTROL SYSTEM REVIEW PROCESS REVIEW 0F TRANSIENT ASSESSMENT PROGRAM (TAP) DATA TO IDENTIFY PROBLEMS RELATED TO MAIN STEAM SAFETY VALVES (MSSV)
TURBINE BYPASS AND/0R ATMOSPHERIC DUMP VALVES O_ IDENTIFY AND PERFORM PRELIMINARY ASSESSMENT OF METHODS OF REDUCING THE FREQUENCY OF MSSV LIFTS IDENTIFY METHODS TO INCREASE RELIABILITY OF BOTH MSSV AND TBV l
l O
42
SECONDARY SYSTEM PRESSURE CONTROL
()
l o CONTROL OF POST-TRIP FEED AND STEAM FLOW HAS BEEN A CONTRIBUTOR TO COMPLEX TRANSIENTS TYPICAL KEY RECOMMENDATIONS (1) PROVIDE (IN MAIN CONTROL ROOM) MANUAL CONTROL AND ISOLATION CAPABILITY OF ALL POST-TRIP STEAM AND FEED FLOW PATHS (EXCLUDING SAFETY RELIEF VALVES)
(2) IMPROVE TURBINE BYPASS (TBV) AND ATMOSPHERIC DUMP (ADV) SYSTEMS I PREVENT EXCESSIVE STEAM FLOW ON LOSS OF ICS/NNI POWER CONTROLLABLE FROM CONTROL ROOM ON LOSS OF ICS/NNI POWER ISOLABLE AND CONTROLLABLE FROM CONTROL ROOM ON
() LOSS OF 0FFSITE POWER (3) DEVELOP AND IMPLEMENT STANDARDIZED PROCEDURES AND TECHNIQUES FOR MAINTAINING, SETTING AND TESTING MAIN STEAM SAFETY VALVES (MSSV) TBV, AND ADV (4) DETERMINE CAUSES AND CORRECT ANOMALOUS MSSV PERFORMANCE (5) ELIMINATE OVERLAP OF TBV AND MSSV CONTROL POINTS
~
POST-TRIP (6) IMPLEMENT A PREVENTIVE MAINTENANCE PROGRAM FOR RELIEF VALVES THAT ARE NOT AUTOMATICALLY POST-TRIP ISOLABLE
() 43 i
n-, __ _---._..-_.,_---,.g_,av _- ._c,,--., , , ,_. __ , . , , , , ,,_-,,,_., e,r-.w.,,-,,-,e.wme,mn
INSTRUMENT AIR SYSTEM REVIEW PROCESS i o COMPILED AIR SYSTEM FEATURES AND OPERATING DATA o IDENTIFIED CRITICAL AIR-ACTUATED COMPONENTS o EVALUATED PLANT RESPONSE TO COMPLETE AIRE LOSS o DEVELOPED FUNCTIONAL TARGET CRITERIA o DEVELOPED RECOMMENDATIONS 1
O e
0 44 I l
4
, _ - . . - - _ - - _ _ - _ _ _ _ ___-..__-.______.___..,_.__.-.__,,_________-__-_____.._,--__,-._,___________.,_____,______m _
INSTRUMENT AIR SYSTEM o CORRECT AIR SYSTEM FAILURES AND ENSURE THAT THE PLANT WILL GO TO A KNOWN SAFE STATE ON LOSS OF INSTRUMENT AIR ,
TYPICAL KEY RECOMMENDATIONS ;
(1) COMPARE THE PLANT'S INSTRUMENT AIR SYSTEM WITH THE l
FUNCTIONAL TARGET CRITERIA IN THE SPIP "INSTRUMENT AIR SYSTEM REPORT" TO DETERMINE WHICH UPGRADES ARE NECESSARY.
(2) PERFORM AN EVALUATION TO ENSURE THAT AIR SYSTEM FAILURE WILL NOT AFFECT THE ABILITY TO MAINTAIN THE PLANT IN A KNOWN SAFE STATE. :
l (3) PERFORM OPERABILITY TESTING OF CRITICAL AIR l OPERATED VALVES, COMPARE WITH DESIGN BASIS b) STROKING TIME AND REBUILD AS NECESSARY ,
l I
4
() 45 l
O SENSITIVITY STUDY PURPOSE SCOPE QUANTIFICATION OF SENSITIVITY GENERAL CONCLUSIONS O
SPECIFIC CONCLUSIONS RECOMMENDATIONS PEER REVIEW CONCLUSIONS l 46 O
l l
() i SENSITIVITY STUDY - PURPOSE EMPLOYED MPR ASSOCIATES TO ASSESS DIFFERENCES IN THERMALHYDRAULIC BEHAVIOR ARISING FROM DIFFERENCES IN DESIGN PERFORMED A COMPARIS0N OF TYPICAL B&W UNIT AGAINST REPRESENTATIVE CE AND W REACTORS l
B&W -
DAVIS-BESSE TMI-1 . l OCONEE ANO-1 O
CE -
1 PRE-1975 PLANT 1 POST-1975 PLANT ,
W -
1 POST-1975 PLANT (WITH A STEAM GENERATOR PREHEATING SECTION)
LIMITED ANALYSES OF OTHER PLANTS THE PURi'0SE OF THE STUDY WAS TO QUANTIFY RELATIVE THERMALHYDRAULIC DIFFERENCES BETWEEN DESIGNS RESULTS WERE SUBMITTED TO NRC ON APRIL 4, 1987 I
() 47 l
i . _ . . . _ _ _ . .
SENSITIVITY STUDY - SCOPE
)
- 1. THERMALDYNAMIC RESPONSE OF THE REACTOR AND STEAM SYSTEM DURING NuRMAL PLANT OPERATION o PERFORMED COMPARIS0N STUDIES OF DESIGN AND OPERATIONAL CHARACTERISTICS o QUANTIFIED DIFFERENCES IN THE INHERENT CONTROL
- 0F KEY PLANT VARIABLES
- 2. PLANT THERMODYNAMIC RESPONSE COMPARIS0NS FOR .
ANTICIPATED OPERATING OCCURRENCES l
, 1 o DEVELOPED DEMONSTRABLY VALID AND QUANTITATIVE INDICES OF DIFFERENCES AMONG PLANTS TO l HALFUNCTIONS -
PLANT THERMODYNAMIC RESPONSE COMPARISONS FOR
() 3. SELECTED ACCIDENTS !
- 4. PROTECTION AND CONTROL SYSTEMS l o ASSESSED DESIGN DIFFERENCES AFFECTING COMPLEXITY OF HARDWARE
! o PERFORMED OPERATING EXPERIENCE ASSESSMENT OF ACTUATIONS AND OFF-NORMAL OCCURRENCES.
, 5. PERFORMED OPERATOR RESPONSIBILITY COMPARISON DURING NORMAL AND OFF-NORMAL OPERATIONS i
o REQUIREMENTS FOR MANUAL ACTIONS o ASSESSMENT OF THE DIFFICULTY IN CONTROLLING PLANT VARIABLES OR SYSTEMS INCLUDING TIME RESPONSE REQUIREMENTS.
()
48 l
i
O O O
! QUANTIFICATION OF SENSITIVITY l' OPERATIONAL INDICES SAFETY PARAMETERS LIMITS / CRITERIA o MARGIN o SECONDARY DESIGN PRESSURE o STEAM LINE FLOODING l 0 TIME o PRIMARY DESIGN PRESSURE, o SG OVERFILL o FREQUENCY TEMPERATURE o SG DRYOUT o SATURATION MARGIN o SAFETY VALVE CHALLENGE j o Kw/FT LIMIT o PORY CHALLENGES l 0 MINIMUM DNBR o STEAM / FEED ISOLATION o PTS LIMITS o LOSS OF PRESSURIZER
- LEVEL
! . O RPS TRIP LIMITS '
j 0 SAFETY INJECTION l '
LIMITS l o HEATUP/COOLDOWN RATE l LIMITS I
i i
49 t
C)
SENSITIVITY STUDY - GENERAL CONCLUSIONS THE STUDY SHOWED THAT SEVERAL ASPECTS OF THE B&W DESIGN, WHICH WERE BELIEVED TO CAUSE "GREATER SENSITIVITY," IN FACT, DID NOT.
o PRESSURIZERS ON B&W UNITS ARE AS LARGE AS, OR LARGER THAN, OTHER PWRs'
() o B&W UNITS ARE LESS SENSITIVE TO STEAM DEMAND UPSETS o B&W UNITS D0 HQI IMPOSE GREATER BURDENS ON OPERATORS FOLLOWING MOST (NORMAL) REACTOR TRIPS.
THESE POSITIVE CONCLUSIONS ARE AS SIGNIFICANT AS THE STUDY'S RECOMMENDATIONS FOR IMPROVEMENT l
C) 50
- - - - - _ . . . _ _ . . . , , . _ , _ . - - - . . ~ _ _ - _- _ . . - . . - - . - _ - _ - _ _ _ _ - - .
()
SENSITIVITY STUDY - SPECIFIC CONCLUSIONS RELATIVE TO OTHER PWRs, B&W UNITS:
o ARE NOT MORE SENSITIVE TO REACTIVITY UPSETS; o ARE NOT MORE SENSITIVE TO REACTOR COOLANT FLOW UPSETS; o ARE LESS LIKELY, ON AVERAGE,.T0 EXPERIENCE A LEAK LEADING TO A NET LOSS OF COOLANT;
() o ARE SOMEWHAT LESS SENSITIVE TO STEAM DEMAND UPSETS SUCH AS LOAD REJECTIONS AND TURBINE TRIPS; A REACTOR TRIP ON A TURBINE TRIP IS NOT REQUIRED TO ENSURE PLANT SAFETY; o ARE NOT MORE LIKELY TO OVERC00L FOLLOWING A REACTOR
. TRIP; o ARE MORE SENSITIVE IN THEIR RESPONSE TO MAIN FEED-WATER UPSETS (THOUGH THE FREQUENCY OF SUCH UPSETS IS NOT GREATER THAN IN OTHER PWRs);
i 51 l
(
SENSITIVITY STUDY - SPECIFIC CONCLUSIONS
() (CONTINUED) o ARE, IN SOME BUT NOT ALL PLANTS, SUBJECT TO GREATER C00LDOWN RATES FROM OVERFEEDING OF EMERGENCY FEED-
> WATER; o ARE EQUIVALENT TO MANY OTHER PWRs IN TERMS OF TIME AVAILABLE TO USE ALTERNATIVE MEANS OF DECAY HEAT REMOVAL, ON A COMPLETE LOSS OF ALL FEEDWATER; o FOR MOST REACTOR TRIPS, DO NOT IMPOSE GREATER CONTROL BURDENS ON PLANT OPERATORS; AND o IMPOSE GREATER BURDENS ON PLANT OPERATORS IN DIAGNOSING AND RESPONDING T0' FAILURES OF AUTOMATIC CONTROL SYSTEMS; SUCH FAILURES ARE MORE LIKELY T0 LEAD TO COMPLEX TRANSIENTS IN B&W UNITS.
()
9 J
52 i
i . _ - - . .__ - - _ - _ - - . _ _ _ _ . _ _ _ - _ _ _ _
SENSITIVITY STUDY - RECOMMENDATIONS
- 1. MODIFICATIONS TO ICS ARE ESSENTIAL TO SUBSTANTIVE REDUCTION IN FREQUENCY OF FEEDWATER UPSET AND TO THE ELIMINATION OF THE COMPLEX TRANSIENTS.
o FW PUMP SPEED CONTROL BE MADE INDEPENDENT OF
, THE CONTROL OF MFW REGULATING VALVE POSITION (EXCEPT CR & ANO-1) (PUMP-VALVE SEPARATION) o FW PUMP SPEED CONTROL AND POSITION OF EACH SET OF REGULATING VALVES (I.E., ONE MAIN AND ONE START-UP VALVE) BE INDEPENDENT OF EACH OTHER (I.E., TRAIN INDEPENDENCE) o CONTROL OF RCS TEMPERATURES AND SECONDARY SIDE ,
PRESSURE SHOULD BE ASSIGNED TO SEPARATE i
- INDIVIDUAL SUB-SYSTEMS O o CONSIDER USE OF REDUNDANT DETECTORS WITH l AUCTIONEERING TO SELECT OUTPUT o CONSIDER USE OF PNEUMATIC ACCUMULATORS TO 4
REDUCE LIKELIHOOD OF LOSING KEY POSITIONERS I (I.E., FW REG. VALVES)
RETAIN IN THE CONTROL SYSTEM:
o COORDINATED REDUCTION OF REACTOR POWER AND STEAM DEMAND ON LOSS OF FEED PUMP i
o LIMITING FEEDWATER FLOW TO EACH OTSG BASED ON THE NUMBER OF RC PUMPS SUPPLYING THAT GENERATOR I o OVERRIDING CONTROL TO MAINTAIN MINIMUM WATER .
LEVEL )
l O 53 I l
- 2. REDUCE BY AN ORDER OF MAGNITUDE THE FREQUENCY WITH
() WHICH MAIN FEEDWATER UPSETS OCCUR 1
- 3. LIMIT THE MAXIMUM FLOW RATE DELIVERED BY THE EMERGENCY FEEDWATER SYSTEM OR REDUCE THE LIKELIHOOD OF A PERSISTENT DEMAND FOR HIGH FLOW
- 4. PERFORM ANALYSIS FOR AND OBTAIN LICENSING APPROVALS TO REMOVE THE ANTICIPATORY REACTOR TRIP ON TURBINE TRIP. (THIS TRIP INCREASES THE PLANT'S SUSCEPTIBILITY TO OVERC00 LING WITHOUT MARKEDLY INCREASING OVERALL PLANT SAFETY) .
- 5. EMPLOY THE TURBINE GOVERNOR VALVES TO RELIEVE SURPLUS ENERGY FOLLOWING A REACTOR TRIP IN LIEU 0F EMPLOYING ATMOSPHERIC DUMP VALVES, TURBINE BYPASS VALVES AND MAIN STEAM SAFETY VALVES.
()
l e
54
l I
()
SENSITIVITY STUDY - PEER REVIEW GROUP PEER REVIEW GROUP ESTABLISHED TO PROVIDE ANALYTICAL INPUT AND REVIEW; INCLUDED UTILITY, VENDOR, AND NRC TH ANALYSTS PRG CONCLUDED o ANALYSIS AND RESULTS ARE TECHNICALLY SOUND AND APPROPRIATE FOR THE STUDY'S COMPARATIVE SCOPE
() o RESULTS NOT SUFFICIENT TO DRAW CONCLUSIONS ON DESIGN CHANGES o SUBSEQUENT EVALUATION OF ICS DESIGN CHANGES SHOULD CAREFULLY CONSIDER TOTAL IMPACT ON PLANT RESPONSE AND OPERATOR / TRANSIENT ASPECTS
- UP TO B&WOG TO CONFIRM THAT PROPOSED RECONFIGURATION CAN BE ACHIEVED WHILE PRESERVING CURRENT POSITIVE CAPABILITIES OF ICS -
I 55
l l
l lCS/NNI EVALUATION O BASIC APPROACH UPDATE THE ESTABLISH ICS/NNI DESIGN THE "AS-BUILT" REQUIREMENTS l CONFIGURATION r ,
DETERMINE
, , TECHNOLOGICAL IMPROVMENTS COMPARISON: ' "
"AS-BUILT" DESIGN TO EXISTING DESIGN REQUIREMENTS v COMPARE AND EVALUATE FOR Q q 7 GENERIC IMPLICATIONS LIST PROBLEMS AND ESTABLISH POTENTIAL SOLUTIONS 1 P O
56
ICS/NNI EVALUATION O BASIC APPROACH FMEA 1 f y RECOMMENDED m PLANT SOLUTIONS
RESPONSE
PREDICTIONS ,
c f 3 SENSITIVITY STUDY O EQUIPMENT OBSOLESCENCE FOLLOW-ON EFFORTS:
c -
1 CONTINUATION OF FINAL RECOMMENDATIONS
^ '
i ADVANCED CONTROL SYSTEM j
I O
57
ICS/NNI EVALUATION g
HOW THE EVALUATION WAS PERFORMED UTILIZED PREVIOUS WORK BROUGHT DOCUMENTS UP TO DATE ESTABLISHED REQUIREMENTS MADE COMPARISONS FORMULATED RECOMMENDATIONS i
PEOPLE INVOLVED e
UTILITY I&C ENGINEERS O UTILITY OPERATIONS PERSONNEL UTILITY MAINTENANCE PERSONNEL B&W CONTROLS ENGINEERS & SPECIALISTS INDUSTRY EXPERTS LAB, EPRI, CONSULTANTS 1
j O
58 1
ICS/NNI EVALUATION O
MAGNITUDE OF THE EVALUATION DIRECT EXPENDITURE EXCEEDED $500,000 INDIRECT EXPENDITURE EXCEEDED $650,000 SPANNED FOURTEEN MONTHS INCLUDED ELEVEN I&C COMMITTTEE MEETINGS FOUR WORKING LEVEL MEETINGS WITH NRC O PRESENTATIONS AT SIX IAB MEETINGS O
59
i O
ICS/NNI EVALUATION OUTPUT OF THE EVALUATION PUBLISHED IN SIX MAJOR DOCUMENTS, AND TWENTY SEVEN SUPPLEMENTARY DOCUMENTS IN APPENDIX R OF BAW 1919 CONTAINS OVER 450 IDEAS, SUGGESTIONS, REFERENCES, OR RECOMMENDATIONS. .
EACH IDEA, SUGGESTION, REFERENCE, OR )
Q RECOMMENDATION HAS BEEN EVALUATED, AND l CATEGORIZED FOR DISPOSITION I l
)
l l
I i
O 60
ICS/NNI EVALUATION O ReSutTS Recommendations were classified into three levels:
LEVEL 1 Those items which will provide immediate improvements in the o aeration, availlability, and reliability of the ICS/Nhl.
The RTS now contains 51 recommendations related to the ICS/NNI system.
Among these are:
Improve the reliability of the power sources.
Harden the ICS inputs to reduce the probability of a trip or transient.
Achieve a Known Safe State on loss of ICS/NNI power .
LEVEL 2 Those which potentially involve major modifications O to the existing equipment and require further in-depth evaluation before approval.
Continued evaluation requires long term commitment to the existing hardware.
LEVEL 3 Those items which consist of replacement of the existing system with a new system based on modern digital tec 1nology.
An Advanced Control System Task Force was established and is pursuing the development effort.
O 61 1
ICS/NNI EVALUATION O
LEVEL 1 Power source improvement-013 Protect on loss of AC power 039 Power supply monitor 102 Additional "Y" DC source 103 Fuse power leaving cabinets 182 ABT for MFP controllers 184 Fuse hand auto stations 187 Install current and voltage meters 195 Supply hand and auto circuits form separate panels Harden the inputs-001 Replace RC flow with RC Pump status 002 Detect valid RC temperature to the ICS 004 Valid turbine header pressure Q 005 Take care of the neutron flux 006 Eliminate FW tem aerature correction 104 Auto select valid inputs Known Safe State-012 Instruments to detect loss of power 032 Resotoration of power 036 Turbine valve position on loss of power ;
037 MPF Speed control on loss of power l 154 Unambiguous status of indicators & recorders . I 190 Backup controls for the Pzr Ivl & press 193 Pzr heater Interlock logic 195 Supply hand and auto power from separate panels l 198 Sensor power on loss of hand 178 KSS Maintain heat balance normal operator actions O
62
ICS/NNI EVALUATION O
LEVEL 2 Eliminate mid scale failures Sepatate subsystems for Tavg and steam press control.
Control MFP on Press vs.AP Se Sarate controls for MFPs, and MFW vivs.
Inc ependent controls for Rx Pwr, Sitm demand, & FW LEVEL 3 Technological improvements to improve operability, reliability, & maintainability Resolve problem areas identifled in the current ICS/NNI systems Develop optimal configuration for ACS Provide features to prevent challenge to the protection systems.
O improve controi and man-macnine interface 4
l O
63 i
1 i l l
1
() .
ICS/NNI EVALUATION
SUMMARY
OF THE NRC SER REVIEW A DETAILED REVIEW 0F THE SER (SUPPLEMENT 1) SECTION 6 ON THE ICS/NNI EVALUATION WAS COMPLETED NRC COMMUNICATES OVERALL AGREEMENT WITH THE ICS/NNI EVALUATION
- o NRC MENTIONED 170 INDIVIDUAL RECOMMENDATIONS, SUGGESTIONS, OR IDEAS,
- () o OVER 90% HAD THE SUPPORT / CONCURRENCE OF THE NRC REVIEWERS.
! o THERE WERE SOME AREAS OF DISAGREEMENT 1
THE LATTER WERE FURTHER CATEGORIZED AS FOLLOWS:
I o THE EVALUATION PROCESS o THE PROGRAMMATIC SOLUTIONS 1 o THE RECOMMENDATION AND/0R THE IMPLEMENTATION '
- PROCESS l C) !
! 64 l
l 4
_-_,-_..--__r. . , .-_____ _- . ,, _ . . , . . . _ _ _ _ , , . . - . - - _ _ . . - _ _ , - _ , _ _ _ - _ _ , _ _ . _ . - . . _ _ _ , - , , - y. ,--_,.,.y, ,
.7
ICS/NNI EVALUATION O SER REVIEW AREA OF CONCERN EVALUATION PROCESS REQUIREMENTS DOCUMENTS FAILURE MODES AND EFFECTS ANALYSIS RECOMMENDATION SCOPE LIMITED INCLUDING lEB 79-27 EVALUATION SOURCE DOCUMENTS REFERENCED PREVIOUSLY IDENTIFIED CONCERNS O
l I
.; l 1
O ,
65
ICS/NNI EVALUATION O
SER REVIEW AREA OF CONCERN PROGRAMMATIC SOLUTIONS MAXIMlZE DEPENDENCE ON NNI-X POST-TRIP TROUBLE-SHOOTING I
PREVENTIVE MAINTENANCE GUIDANCE TOO GENERAL CHECKOUT OF REPLACEMENT CRDCS MODULES FREQUENCY OF ICS/NNI PREVE,NTIVE MAINTENANCE O
l l
9 i
O 66
ICS/NNI EVALUATION g SER REVIEW AREA OF CONCERN THE RECOMMENDATION AND/OR IMPLEMENTATION PROCESS REPLACE RC FLOW WITH RC PUMP STATUS DELETING FW TEMPERATURE CORRECTION CIRCUlT REMOVAL OF BTU LIMITS IMPLEMENTATION OF PRESSURIZER LEVEL & PRESSURE l BACKUP !
SMUD's TRIPPING OF ICS POWER REMOVAL OF NNI-Z POWER ,
RELOCATION OF THE POWER SUPPLY MODULES ,
O ARTS BISTABLE AUTOMATIC RESET O
67
. -_-_ - __0
I O
ICS/NNI EVALUATION OVERALL ICS/NNI
SUMMARY
AND CONCLUSIONS MAGNITUDE OF THE ICS/NNI EVALUATION EFFORT WAS LARGE METHODOLOGY USED FOR THE REVIEW WAS SOUND VULNERABILITIES STILL EXIST IN THE ICS/NNI DESIGN AND LOSS OF POWER MAY STILL OCCUR f
IMPLEMENTATION OF LEVEL 1. RECOMMENDATIONS WILL REDUCE TRIPS AND TRANSIENT COMPLEXITY AND COMPENSATE FOR THE VULNERABLITIES O FOLLOW-ON ACTIVITIES ARE BEING ACTIVELY PURSUED ,
)
i O
j 68 l
ADVANCED CONTROL SYSTEM !
O PROGRAM
- LEADING THE INDUSTRY
- INCLUDES ICS/NNI EVALUATION LESSONS LEARNED FOREIGN CONTROL SYSTEMS REVIEW DETAILED HUMAN FACTORS STUDY EXPERT ADVISORY PANEL NASA, ERPI, ORNL THE ACS WILL BE A PLANT CONTROL AND MONITORING SYSTEM PLANT CONTROL ,
- IMPROVED CONTROL TECHNIQUES MORE OPERATIONAL FEATURES O
- INCREASED RELIABILITY FAULT AVOIDENCE FAULT TOLERANCE MONITORING PREDICT FAILURES
- DETECT FAILURES
- ALERT OPERATORS 4
O 69
i
! O B&WOG PERSPECTIVE ON HUMAN FACTORS o MAIN PURPOSES OF SPIP ASSESS RISK SIGNIFICANCE OF COMPLEX TRANSIENTS l
DEFINE ACTIONS TO REDUCE FREQUENCY OF TRIPS
- AND COMPLEXITY OF POST-TRIP TRANSIENTS o HUMAN ACTIONS CAN AFFECT TRANSIENT BEHAVIOR o HUMAN FACTORS CONCERNS WERE CONSIDERED ALONG WITH
! OTHER IMPORTANT MATTERS. HUMAN FACTORS WERE NOT l
O A DOMINANT CONSIDERATION IN SPIP.
0 B&WOG BELIEVES THAT ITS ACTIONS HAVE, OR WILL, CONSIDER HUMAN FACTORS CONCERNS.
i o NRC HAD THREE SPECIFIC HUMAN FACTORS RECOMMENDA-
- TIONS.
1 O
70 j
O
! THREE NRC HUMAN FACTORS RECOMMENDATIONS 1
l
\
i o REVIEW 0F HUMAN FACTOR LOSS OF ICS/NNI POWER EVENTS o USE OF HUMAN FACTORS EXPERTISE TO RE-REVIEW:
OPERATIONS AND MAINTENANCE INTERVIEW DATA OLD TAP REPORTS THE 1985 DAVIS-BESSE EVENT PERATOR BURDEN DATA O
i o HUMAN FACTORS INVOLVEMENT IN RECOMMENDATION IMPLEMENTATION AT EACH UTILITY 1
i j
i i
O 71 ,
l w -
i O
)
i BAROG POSITION ON NRC RECOMMENDATIONS i
LOSS OF ICS/NNI POWER EVENTS i
! o ACTIONS HAVE ALREADY BEEN TAKEN TO AID IN DIAGNOSIS AND RECOVERY.
l
! o CURRENT B&WOG EMPHASES ARE TWO-FOLD:
i O
- 1. PREVENT LOSS OF ICS POWER EVENTS ON LOSS OF A i
SINGLE AC SOURCE 4
- 2. IF THEY OCCUR, GO TO A KNOWN SAFE STATE l
2 o LONGER TERM IMPROVEMENTS ARE BEING CONSIDERED AS j PART OF ADVANCED CONTROL SYSTEM PROJECT.
1 O
72 I.---.__ - . - _ . - . - _ . - - - - _ . _ _ . - _ - . , - . - - .
()
B&WOG POSITION ON NRC RECOMMENDATIONS USE OF HUMAN FACTORS EXPERTISE TO RE-REVIEW OPERATIONS AND MAINTENANCE INTERVIEW DATA B&WOG IDENTIFIED SIX CONCERNS WHICH ARE BEING Q ADDRESSED.
ALL B&WOG MEMBERS ARE NOW SUPPORTING INP0'S HUMAN PERFORMAI4CE EVALUATION SYSTEM (HPES) OR ITS EQUIVALENT.
RE-REVIEW OF INTERVIEW DATA 1-1/2 YEARS LATER WOULD NOT BE OF SIGNIFICANT ADDITIONAL BENEFIT.
()
73
4 O ;
l 1
i B&WOG POSITION ON NRC RECOMMENDATIONS j
- USE OF HUMAN FACTORS EXPERTISE TO RE-REVIEW OLD I TRANSIENT ASSESSMENT PROGRAM REPORTS l TAP REPORTS HAVE BEEN HELPFUL IN SPIP. ;
O i l -
PRIOR TO 1987, COLLECTING HF INFORMATION WAS !
NOT EMPHASIZED IN TAP.
GUIDELINES FOR CONTENT OF FUTURE TAP REPORTS HAVE l
- BEEN UPGRADED TO EXPLICITLY EMPHASIZE HF ISSUES.
I I
l O l 74 4
1
()
l l
B&WOG POSITION ON NRC RECOMMENDATIONS USt s. HUMAN FACTORS EXPERTISE TO RE-REVIEW JUNE 1985 QAVIS-BESSE EVENT CERTAIN GENERIC HF CONCERNS FROM DB EVENT WERE REVIEWED BY EACH UTILITY; E.G., VITAL EQUIPMENT
() ACCESSIBILITY, LOCAL VERSUS REMOTE CONTROL, CLARITY OF INSTRUCTIONS TO TAKE DRASTIC ACTIONS.
B&WOG BELIEVES THAT RE-REVIEW 0F THIS EVENT WITH MORE EMPHASIS ON HF ISSUES WOULD NOT PROVIDE ADDITIONAL BENEFITS.
1 i
75
. , , . - . - . -. ---..--.-.---,----------i-----------,--, ,,--,---,.-w- v,- - - - - , -r-n---,,----v m--mw----*iv
C) l B&WOG POSITION ON NRC RECOMMENDATIONS USE OF HUMAN FACTORS EXPERTISE TO RE-REVIEW OPERATOR BURDEN DATA o PROJECT OBJECTIVES:
IDENTIFY AREAS OF UNNECESSARY BURDEN.
DEFINE ACTIONS TO REDUCE BURDEN AND FREQUENCY OF COMPLEX TRANSIENTS.
o METHODOLOGY WAS DEVELOPED BY HF EXPERT
() o SEVEN BURDEN REDUCTION AREAS IDENTIFIED AND i RECOMMENDATIONS DEVELOPED CONTROL OF STEAM AND FEED FLOW l DRASTIC ACTIONS OVERC00 LING MITIGATION INSTRUMENT AIR ACTIONS OUTSIDE MAIN CONTROL ROOM ANNUNCIATORS EMERGENCY PLAN REQUIREMENTS o RE-REVIEW 0F DATA NOT PROVIDE ADDITIONAL BENEFITS
()
76
O l
B&WOG POSITION ON NRC RECOMMENDATIONS l
USE OF HF EXPERTISE IN RECOMMENDATION IMPLEMENTATION Oo HUMAN FACTORS EXPERTISE WILL BE UTILIZED ON PLANT-SPECIFIC BASIS.
o USE OF HF EXPERTISE WILL VARY DEPENDING ON SUBJECT OF RECOMMENDATION.
1 O
77
()
B&WOG POSITION ON NRC RECOMMENDATIONS j l
SUMMARY
B&WOG HUMAN FACTORS COMMENTS o B&WOG AGREES THAT HUKAN FACTORS CONSIDERATIONS ARE IMPORTANT.
o B&WOG MEMBERS ARE NOW PLACING MORE EMPHASIS ON HUMAN FACTORS THAN WHEN SPIP BEGAN.
o B&WOG BELIEVES THAT RE-REVIEW 0F OLD INFORMATION WOULD HAVE LITTLE ADDITIONAL BENEFIT.
o HF CONSIDERATIONS WERE A PART OF SPIP.
78
l 1
()
PERFORMED A RISK ASSESSMENT REVIEW o ASSESSED THE IMPORTANCE OF B&W HISTORICAL CATEGORY "C" EVENTS TO CORE-DAMAGE RISK. CATEGORY "C" TRANSIENTS ARE THOSE WHERE SYSTEM CONDITIONS REACH LIMITS WHICH REQUIRE SIGNIFICANT SAFETY SYSTEM AND OPERATOR RESPONSE TO MITIGATE. THIRTEEN OF THESE EVENTS WERE REVIEWED.
o COMPARED INITIATING EVENT FREQUENCIES OBTAINED FROM THE TRANSIENT HISTORY OF ALL B&W UNITS TO THE FREQUENCIES USED ON THE EXISTING PRAs.
o EVALUATED THE DOMINANT ACCIDENT SEQUENCES, SYSTEMS AND INITIATORS FROM THE PRAs. COMPARED THESE TO
() THE CATEGORY "C" EVENTS.
o GENERALIZED THE RESULTS OF THE ANALYSES TO ALL OF THE B&W UNITS.
o USED THE: OCONEE PRA -
LEVEL 3: INTERNAL & EXTERNAL i EVENTS CRYSTAL RIVER PRA - .
LE EL 1: INTERNAL EVENT AND LOOP
()
79
O 1
l l
SPIP ADMINISTRATION, F01. LOW THROUGH AN IMPLEMENTATION O
O 80
i l
l l
() l l
SPIP RECOMMENDATION IMPLEMENTATION l l
o IMPLEMENTING RECOMMENDATIONS - CRITICAL FOR SUCCESS 1
o IMPLEMENTATION IS BEING FORMALLY MONITORED BY B&WOG EXECUTIVES AND THE STEERING COMMITTEE FOR BOTH QUALITY AND TIMELINESS:
IMPLEMENTATION QUALITY WILL BE MONITORED FROM A PROGRAMMATIC AND TECHNICAL STANDPOINT
() -
PROPER IMPLEMENTATION WILL REQUIRE CONSIDERA-TION OF PLANT UNIQUE CONFIGURATION, RESOURCES, AND OPERATING SCHEDULE RECOMMENDATION TRACKING SYSTEM REPORT IS MANAGEMENT TOOL o EFFECTIVENESS IS MONITORED BY COMPARISON OF PERFORMANCE AGAINST DEFINED GOALS C) 81
()
B&WOG RECOMMENDATION TRACKING SYSTEM (RTS) o RECOMMENDATIONS COMPILED AND EVALUATED o DECISIONS AT THE VARIOUS SCREENING POINTS DOCUMENTED o RECOMMENDATIONS APPROVED BY THE STEERING COMMITTEE ARE PLACED IN THE RECOMMENDATION TRACKING SYSTEM REPORT o RTS REPORT IS UPDATED QUARTERLY BASED ON INPUT FROM THE UTILITIES REGARDING INDIVIDUAL RECOMMENDATION
() STATUS o B&WOG WILL KNOW THE STATUS OF IMPLEMENTATION AND KEEP AN ACCOUNTABILITY OF EACH REC 0HMENDATION THROUGH IMPLEMENTATION. DOCUMENTATION HAS AND WILL BE PROVIDED TO NRC STAFF.
o STATUS OF SPIP RECOMMENDATIONS NOT IN RTS "PENDING" RECOMMENDATIONS FOR RTS -
9 RECOMMENDATIONS INVOLVING GENERIC - 13 B&WOG STUDIES (NOT FOR RTS)
()
82
- -- 4 a .- + - a ~ - .a a O
I A
N N
v Z g o 2
- ~
H C
Q Z
LJ E ,
E Om C.J m m$
O LJ s 1h
- a.
1C C.n L ^
O m vm C.n z mo H e--.
b %H C C ^
W z O N
M zW /
-r ^
Qr y zO g M WU V .
' Q. W y
- 2 m ]
O >
W W O
W U 7 f.n
- i W
N N W
Q Z
D 83
O O O SEQUENCE OF EXPECTED RECOMMENDATION DISPOSITION RTS E/A E/I I EVALUATING FOR APPLICfBILITY EVALUATING FOR ItPLEtENTATION INPLEHENTING 4
CLOSED / CLOSED / CLOSED /
NOTAPPLICABLE REJECIED 0PERABLE i
C/NA C/R C/0 t
2 _ _ - _ . . . --
O i
8 8
S 9 '
T N
A i
1 0
6 \ 2 2 R n
i h
l L
P G
N I
T A
R ,
E P " i 8 O d'" 8
/
0 1 L [" 4 5
6 2
'N L O A [I J
R SN OO '
FI T A
SD U NE O TN AN O $ s T C o SE S r I 7
8 R
N P $ X 0 6
6 X4 X 7 '
OI X 4 1 V P O I
T S Nen I
N A I /
T N
E M
E ,
L P
M I Yri 7
I t 8
/
E T
A N
Ic
/
/
5 1
4 0
3 2
P t E G e I e S E n R
G G
A -
n 0 0 0 0 0 0 0 0 0 0 0 ,
0 3
1 0
2 1
0 1
1 0
0 1
0 9
0 8
0 7
0 6
0 5
0 4
0 3
y$ 8g ]~$=h -
- w
O m
s g
= f=///
//
7/ s s@$c
=
~
//// bbb 5
s p
E V'///
E :,
m m -
~
/
/s/
7
=
=
_ \
's m5 O!
(v7 20 iill:
=
5 EE 3B i
r C ,
() 8 l
8 l
8 l
8 l
8 l
R l
?
l l l e8 l l 8 8 l
8 l l 8
e n n n n ~ ~ ~~~~
SNOI100N3WWO338 J0 'ON 310038000 86'
\
1
O l
4 l
d B&WOG PLANT PERFORMANCE ,
1 O IN AREAS PERTINENT TO SPIP !
l I
O 87
O O O UNPLANNED AUTOMATIC SCRAt1S IJHILE CRITICAL
- INDUSTRY AVERAGE 8&u0G AVERAGE l V///A M i 8--
8 A!!Ed!E0!!E gs5Su!II jig 1980 1981 1982 1983 1984 1985 1986 1987 b
1988 1989 90 i
(
%.J O O FREQUENCY OF CATEGORY C TRANSIENTS (BASED ON A 3-YEAR MOVING AVERAGE) 0.3 0.28 -
0.26 -
[ 0.24 -
3 8 0.22 -
E D 0.2 -
a
$, 0.18 -
h tij 0.16 -
o m
0.14 -
% s
' O.12 -
B&WOG GOAL: 0.1..(BY 1990)..
- o. 3 _ . . _ . . . . . . _ _ . . . _ . . . .-. _
0.08 -
0.06 -
W f KV y 0.04 .... ............... ..,...........,........... ...... .-,,..
JAN JAN JAN JAN JAN JAN 1983 1984 1985 1986 1987 cn so
l I
O !
l i
PROGRAMMATIC EVALUATION OF SPIP-RECOMMENDATION IMPLEMENTATION l
o EXECUTIVES CONVENED EVALUATION TEAMS TO CONDUCT EVALUATIONS AT EACH UTILITY l TEAM TYPICALLY CONSISTED OF EIGHT INDIVIDUALS EVALUATION PLAN DEVELOPED l EVALUATIONS HAVE BEEN COMPLETED AT ALL !
UTILITIES l REP RT ISSUED TO EACH UTILITY 0
o EXECUTIVE INVOLVEMENT IN EACH EVALUATION i o ONE MANYEAR OF TEAM EFFORT (EXCLUDING TRAVEL) o ONE HUNDRED FORTY OWNERS GROUP PERSONNEL WERE INVOLVED o
SUMMARY
REPORT ENDORSED BY B&WOG EXECUTIVES AND PROVIDED TO NRC ,
- . O 90 !
l l
. . . - - . _ - . . - . . . - . - . __.-1
()
EVALUATION TEAM CONCLUSIONS THE EVALUATION TEAM REACHED THE PRINCIPAL CONCLUSION THAT ALL UTILITIES HAD ADEQUATE PROGRAMS FOR SPIP RECOMMENDATION DISPOSITIONING AT THE TIME OF THE EVALUATION. HOWEVER, IN ALL CASES, SOME IMPROVEMENTS WERE NEEDED TO CREATE A FULLY EFFECTIVE PROGRAM.
THE FOLLOWING ADDITIONAL CONCLUSIONS WERE REACHED BY THE EVALUATION TEAM:
o EXCELLENT "CROSS-FERTILIZATION" OCCURRED THROUGH THE TEAM MEMBERS TO THEIR RESPECTIVE COMPANIES.
UTILITY PROGRAM FEATURES WERE CONTINUALLY BEING IMPROVED THROUGHOUT THE EVALUATION PERIOD AS A C) DIRECT RESULT OF THE EVALUATION PROGRAM.
o THE INVOLVEMENT OF EXECUTIVE COMMITTEE REPRESEN-TATIVES CONTRIBUTED GREATLY TO THE EVALUATION QUALITY, THE EMPHASIS ON THE IMPORTANCE OF SPIP RECOMMENDATION DISPOSITIONING, AND THE INCREASED VISIBILITY OF THE SPIP AND ITS GOALS.
o FUTURE IMPROVEMENTS IN THE QUALITY OF DISPOSI-TIONING, REPORTING. AND DOCUMENTATION, AND l INCREASED PROGRESS IN DISPOSITIONING ARE EXPECTED AS A RESULT OF ACTIONS TAKEN BY THE UTILITIES.
C) 91
l O
l l
i l
l l
TECHNICAL EVALUATION OF SPIP RECOMMENDATION IMPLEMENTATION !
o TECHNICAL EVALUATIONS STARTED
- FOUR-MAN TEAM FORMED TO CONDUCT EVALUATION
- FOUR RECOMMENDATIONS INCLUDING ICS (KNOWN SAFE !
STATE) SELECTED FOR INITIAL EVALUATION
- EVALUATIONS COMPLETED AT GPUN AND DPC 1
- 0THER PLANTS SCHEDULED o DETAILED EVALUATION PLANS AND CHECKLISTS UTILIZED o STEERING COMMITTEE HAS REVIEWED RESULTS OF FIRST TWO EVALUATIONS o TECHNICAL EVALUATIONS OF ADDITIONAL RECOMMENDATIONS WILL CONTINUE O
92
()
CLOSING COMMENTS o SAFETY AND PERFORMANCE HAS BEEN AND WILL CONTINUE TO BE IMPROVED AT B&WOG PLANTS ,
l 0 SIGNIFICANT RESOURCES DEVOTED TO REASSESSMENT l
0 SCOPE OF ASSESSMENT WAS CORRECT AND SUFFICIENT l l
o l DEVELOPED OVER TWO HUNDRED SAFETY AND PERFORMANCE IMPROVEMENT RECOMMENDATIONS o UTILITIES DEVELOPED A TRACKING SYSTEM TO ENSURE FORMAL AND TIMELY DISP 0SITI0'N OF ALL RECOMMENDA-TIONS o IMPLEMENTATION OF SOME RECOMMENDATIONS HAS ALREADY RESULTED IN BENEFIT (TRIP AVOIDANCE) o EXPECT THE IMPLEMENTATION PHASE TO BE CONTINUING FOR SEVERAL YEARS o UTILITY MANAGEMENT WILL MONITOR FUTURE PERFORMANCE TO ENSURE IMPROVEMENT IS ATTAINED o NRC CAN READILY MONITOR PERFORMANCE PROGRESS o B&WOG COMMITMENT TO SPIP HAS BEEN AND WILL CONTINUE TO BE STRONG I
()
93 4
O 1
COMMITMENT TO ACTIM o COMMITTED TO TAKE LEAD AND DIQ l
o COMMITTED TO ALLOCATE RESOURCES AND DlQ; '
($10M AND 100MY) o COMMITTED TO MONITOR OURSELVES AND WE ARE Oo COMMITTED TO OPENLY CONSIDER NRC AND OTHER OUTSIDE .
INPUT AND DlQ o ESTABLISHED TRIP AND TRANSIENT FREQUENCY GOALS AND WE ARE PROGRESSING TOWARD THEM o COMMITTED TO FOLLOW-THROUGH AND PLANT IMPLEMENTA-TION AND IT IS HAPPENING O
94
l l
ACRS SMITIE PE_ TING-BWOG SAFETY AfD ERF0WANCE ItPRm9ENT PROGRAM (SPIP)
O l l
STAFF PRESENTATIONS l
SIEGEL OVERVIEW 0F SPIP J0ES Sttt%fN AND CONCLUSIONS J0ES IWOWATION CATERING J0ES ,
SENSITIVITY EVALUATIONS SYSTB S REVIEW 1
KE?0ALL ICS/?NI LEFAVE MAIN FEEDWATER LEFA\E AFW/EFW LEFAVE SECON.'IPLANTRELIEF ,
LEFAVE INSTRUENT AIR HATTER O VALVE TASK FORCE DEBOR EVALUATION OF HtFAN FACTORS J0ES EVALUATION OF OTER BWOG REPORTS RUBIN RISK ASSESSENT JONES REACTOR TRIP INITIATING EVENTS EVIEW SIEGEL BWOG PROGRAW ATIC Af0 PANAGEW NT ACTIONS SIEGEl.
IPPLEENTATION SIEGEL ADDITIONAL CONCERNS IDENTIFIED BY STAFF
. i O
CALVO 5/3/88 0
~
O O O 1
i B&W PLANT REASSESSMENT
STAFF
SUMMARY
AND CONCLUSIONS 1
'l i
MAY 3-4, 1978 NRR/ DEST /SRXB BWACRS1 ,
~
O O O ;
PROGRAM SCOPE t
OPERATING EXPERIENCE FORMS MAJOR BASIS FOR PROGRAM
- EMPHASIS PLACED ON BALANCE OF PLANT SYSTEMS
- STAFF COMMENTED THROUGHOUT ENTIRE PROGRAM
- BWOG DID NOT INCORPORATE ALL OF STAFF'S COMMENTS
- STAFF BELIEVES SPIP WAS COMPREHENSIVE AND WILL IMPROVE SAFETY OF B&W PLANTS
- SINC 5 ALL STAFF COMMENTS NOT RESOLVED
- LIKELY FUTURE TRANSIENTS WILL BE CAUSED OR COMPLICATED BY SYSTEMS STUDIED IN SPIP ENCOURAGED BWOG TO CONSIDER STAFF CONCERNS AS PART OF CONTINUING BWOG PROGRAMS BWACRS6
- -_-------u
~
O O O
~
i
SUMMARY
OF RECOMMENDATIONS APPROVED KEY HIGH STAFF STAFF SYSTEM /
AREA RECOMMEN D RECOMMEND PRIORITY DISAGREE GUIDANCE 10 4 1 O O TURBINE 47 14 1 O O MFW 53 20 1 6 20 ICS/NNI -
EFW 9 7 0 0 2 IAS 34 3 '4 O 2 11 3 1 0 2 El FCTRICAL 9 7 O O 4 VALVES 10 5 1 1 2 MAIN STEAM OPERATIONS 32 8 2 1 1 BWACRS7
- ~
O O O 2
i BWOG KEY RECOMMENDATIONS I
j ICS/NNI l - ENSURE PLANT GOES TO KNOWN SAFE STATE ON LOSS OF POWER
- INCORPORATE AUTOMATIC SELECTION OF VALID INPUTS
- IMPROVE MAINTENANCE AND TUNING FOR ICS/NNI i - PROVIDE OPERATOR WITH UNAMBIGUOUS STATUS OF INDICATORS IN MAIN CONTROL ROOM FOR LOSS OF ICS POWER MAIN FEEDWATER
- IDENTIFY CAUSES OF MFW CONTROL PROBLEMS
}
i - ENSURE SINGLE ELECTRICAL FAILURE WILL NOT CAUSE LOSS OF l BOTH FEEDWATER TRAINS
- ELIMINATE AUTOMATIC MFW PUMP TRIPS WHERE POSSIBLE j - PROVIDE OVERFILL PROTECTION FOR MFW SYSTEM l
) EFW i - REMOVE AUTO INITIATION & CONTROL OF EFW FROM ICS/NNI
- EXTEND TIME TO ACHIEVE DESIGN EFW FLOW j - REVIEW MAINTENANCE, SURVEILLANCE AND TEST PROCEDURES l
BWACRS8
~
f O O O 4
4 I BWOG KEY RECOMMENDATIONS (CONT'D) 4 MAIN STEAM PRESSURE CONTROL l - CORRECT ANOMALOUS POST-TRIP PERFORMANCE OF MSSV's
- REVISE TURBINE BYPASS VALVE MAINTENANCE PROGRAM j -
MOTOR OPERATED VALVES i - ENSURE PROCEDURES PROPERLY SET TOROUE AND BYPASS LIMIT SWITCHES '
- CHALLENGE VALVES m OPEN/CLOSE UNDER DIFFERENTIAL PRESSURES WHICH SIMULATE OPERATIONAL AND ACCIDENT CONDITIONS i
i
! PLANT ELECTRICAL SYSTEMS f - ESTABLISH . MAINTENANCE PROCEDURES FOR INVERTER AND ELECTRICAL BUSES
! - MODIFY INVERTER OVERCURRENT PROTECTION TO '
ENSURE BREAKER / FUSES
. OPEN PRIOR TO INVERTER FAILURE INSTRUMENT AIR SYSTEM
- OPERABILITY TEST CRITICAL AIR-OPERATED VALVES i - PERFORM PLANT SPECIFIC AIR SYSTEM FAILURE EVALUATION OPERATIONS j - IDENTIFY HIGH PRIORITY TASKS DURING EMERGENCIES FOR TRAINING l - REVIEW EOPs TO ASSURE WHENEVER DRASTIC ACTIONS ARE SPECIFIED f PLANT CONDITIONS REQUIRE THE ACTION 1
} BWACRS9 l
~
O -- --
O O HIGH P RIORITY RECO M M E N DATI ONS DEVELOP BACKUP AUTO AND MANUAL CONTROL FOR PZR LEVEL AND PRESSURE CONTROL FROM ANOTHER POWER SOURCE TEST DIESEL GENERATORS UNDER EXPECTED LOADING CONDITIONS VALIDATE EOPs TO ENSURE ADEOLIATE STAFFING & PRIORITIZTION EXISTS VERIFY INSTRUMENTATION AND DISPLAYS FOR ATOG STABILi'IY PARAMETERS IMPROVE RESPOhlSE OF MODULATING TURBINE BYPASS VALVES SYSTEMATICALL_Y INSPECT THE lAS FOR LEAKS-INSPECT ACCUMULATORS AND THEIR CHECK VALVES IN IAS REVIEW TRAINING AND LOSS OF AIR RESPONSE PROCEDURES ENHANCE RELIABILITY OF MFW AND CONDENSATE SYSTEMS AND CDNTRDLS REVIEW EHC OVERSPEED AND FAST CONTROL AND INTERCEPT VALVE CIRCUlTS BWACRS10 l
l
O O O l STAFF DISAGREEMENTS
- STAFF DOES NOT ENCOURAGE PUSUIT OF RAISING MSSV SETPOINTS
- DISAGREE WITH RECOMMENDATION TO SET SELECTOR SWITCHES FOR MAXIMUM NNI DEPENDENCE
- DISAGREE WITH REMOVAL OF BTU LIMITS FU NCTION
- DISAGREE WITH WIRING POWER SUPPLY MONITOR
~
DIRECTLY TO THE OUTPUT BUS
- DISAGREE WITH REMOVING AUTOMATIC ICS TRIP ON
- NNI SINGLE POWER FAILURE
- DISCOURGE PURSUIT OF DESIGN MODIFICATION FOR AUTOMATIC RESET OF ARTS .
BWACRS11
?.*?f.W '
g.u. ,,g O O STAFF SUGGESTIONS ENCOURAGED PURSUIT OF DAVIS-BESSE MAKEUP SYSTEM UPGRADE CONSIDER RECOMMENDATION FOR GRADUAL LOSS OF AIR TEST
' 'h1. . NSIDER ANALYSIS TO ASSESS LOSS OF INSTRUMENT AIR ON LOOP
'?% 4. ..
s, p. NjsIDER REEXAMINATION OF BWOG HUMAN FACTORS STUDIES BY A J- ; UMAN FACTORS EXPERT
'a r ., . ,9 , ,
' h;MHANCE REFERENCES FOR RECOMMENDATIONS CONSIDER DEVELOPMENT OF ICS/NNI DESIGN SASIS DOCUMENTS CONSIDER PERFORMING ADDITIONAL INVESTIGATION OF THE POWER SUPPLY MONITOR DESIGN AND OPERATION IN ICS ENCOURAGES ACTIONS TO ACHIEVE KNOWN SAFE STATE AS DEFINED 1
IN BAW-1919 h,q*.c 1 RECONSIDER REJECTED RECOMMENDAT!ON:
- TROUBLE SHOOTING PROCEDURE FOR DIAGNOSING ICS l MODULE FAILURES
- CHECKOUT OF REPLACEMENT COMPONEi4TS PRIOR TO USE l ,
BWACRS12 i - . - _ _ _ _ _ - _ _ _ _ - _ _ _ _ _ - _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _
O O O SIGNIFICANT IMPROVEMENTS
- IMPROVED RESPONSE TO ICS/NNI FAILURES KNOWN SAFE STATE RECOVERY OF ICS/NNI POWER IMPROVED HEAT SINK RELIABILITY MFW RELIABILITY IMPROVEMENTS EFW RELIABILITY IMPROVEMENTS ,
IMPROVED SECONDARY SIDE PRESSURE CONTROL IMPROVED MOTOR OPERATOR VALVE
. PERFORMANCE l
IMPROVED INSTRUMENT AIR SYSTEM BWACRS13 ,
Y__---__-________--__-____-____ __ _ . _ _ _ _ _ _ . _ _-_ ____ _ _ _ _ . - - . ___ _ .- ___ _ -. _ _-__
~
O
~
O O OTHER IMPROVEMENTS
- REDUCED REACTOR TRIPS
- ICS INPUT SIGNAL FAILURES .
- ICS MAINTENANCE AND TUNING
- MFW SYSTEM IMPROVEMENTS
-- TURBINE SYSTEM IM PROVEMENTS
- REDUCED CHALLENGES TO SAFETY SYSTEMS lMPROVEO MAIN STEAM PRESSURE CONTROL REDUCED OVERCOOLING BY MFW & EFW l IMPROVED RESPONSE TO ICS/NNI ' POWER LOSSES
- REDUCED OPERATOR BURDEN / ENHANCED OPERATOR PERFORMANCE ICS/NNI MODIFICATIONS IMPROVED MSSV PERFORMANCE ENHANCED PROCEDURES AND TRAINING l
REDUCED PTS RISK BWACRS14
~
r O V SIGNIFICANT ISSUES
..a.- .
lM PLEM ENTATION
- FOR SAFETY GAINS TO BE REALIZED, RECO M M EN DATIO NS
- ,3jg. MUST BE PROPERLY IMPLEMENTED
- STAFF WILL MONITOR UTILITY IMPLEMENTATION PLANS AND PROGRESS
- PLANT SPECIFIC AUDITS ARE NECESSARY TO ENSURE PROPER IMPLEM ENTATION
.. HUMAN FdCTORS ISSUES STAFF POSITION - HUMAN FACTORS EXPERTISE NEEDED 4Af. TO IDENTIFY HUMAN FACTORS CONCERNS BWOG DID NOT USE HUMAN FACTORS PROFESSIONAL THUS, STAFF UNABLE TO CONCLUDE SCOPE OF HUMAN FACTORS ISSUES EXAMINED IN SPIP WAS COMPLETE SUGGESTED BWOG REEXAMINE EFFORTS WITH HUMAN FACTORS PROFESSIONAL (REJECTED)
BWACRS15
~
O
~
O O SIGNIFICANT ISSUES (CONT'D)
IE BULLETIN 79-27 "LOSS OF NON-CLASS lE INSTRUMENTATION AND CONTROL POWER SYSTEM BUS DURING OPERATION"
- RANCHO SECO EVENT INDICATES THAT PLANT MODIFICATIONS TO RESOLVE lEB 79-27 MAY NOT BE ADEQUATE
- BWOG REQUESTED TO, BUT DID NOT, ADDRESS j
IE8 79-27 AS PART OF ICS/NNI REVIEW
- STAFF CONCLUDED THAT PLANT MODIFICATIONS TO RESOLVE lEB 79-27, ALONG WITH KNOWN SAFE STATE RECOMMENDATION WILL RESOLVE ICS CONCERNS STAFF RECOMMENDS VERIFYING IMPLEMENTATION OF lEB 79-27 AS PART OF PLANT SPECIFIC AUDITS i
BWACRS16
O O t
l B&W PLANT REASSESSMENT NUREG 1231 SENSITIVITY EVALUATION l l
1 PRESENTED BY PRESENTED TO ACRS SUBCOMMITTEE R.C. JONES ON B&W REACTORS MAY 3-4, 1978 NRR/ DEST /SRXB BWACRS3
O O o STAFF ACTIVITIES
- REVIEWED MPR'S SENSITIVITY EVALUATION MODEL ADEQUACY
- CONCLUSIONS AND RECOMMENDATIONS
- REEXAMIN$D ADEQUACY OF PRESSURIZER SIZE
- EVALUATED EFFECT OF B&W DESIGN ON OTHER SAFETY ISSUES (ATWS & PTS)
- EVALUATED ADEQUACY OF OTSG l
BWACRS19 i ,
O O e
STAFF FINDINGS I
- CONFIRMED B&W PLANTS MORE SENSITIVE TO FEEDWATER UPSETS i - B&W PLANTS NOT MORE SENSmVE TO OTHER UPSETS INCLUDING j UPSETS IN STEAM FLOW l t
- B&W PLANTS PLACE GREATER RELIANCE ON EFW, INCLUDING PROPER ,
! REGULATION OF EFW FLOW ,
- OPERATOR BURDEN:
j - NOT GREATER FOR NORMAL REACTOR TRIP j - GREATER FOLLOWING FAILURES OF ICS/NNI ll
)
- MPR RECOMMENDATIONS GENERALLY ACCEPTABLE f STAFF DISAGREES WITH REMOVAL OF ARTS ON TURBINE TRIP
- PREVIOUS STAFF ACTIONS EFFECTIVE IN ADDRESSING l INHERENT B&W DESIGN CHARACTERISTICS l - RAISING PORV SETPOINT
- LOWERING HIGH PRESSURE TRIP SETPOINT f
l
- ADDmON OF ARTS BWACRS2O
e- . . . .
O o STAFF FINDINGS (CONTINUED)
- INHERENT CHARACTERISTICS OF B&W PLANTS CONSIDERED DURING DEVELOPMENT OF ATWS & PTS RULES ONCE THROUGH STEAM GENERATOR DESIGN OTSG HAS CERTAIN OPERATIONAL AND TUBE INTEGRITY ADVANTAGES BEN FITS OFFSET BY INHERENT RESPONSIVENESS OF OTSG AND SMALL WATER INVENTORY IF SYSTEMS THAT INTERACT WITH OTSG PROPERLY CONTROLLED (e.g. ICS/NNI AND EFW) POST-TRIP RESPONSE WILL BE ACCEPTABLE SPIP WAS RESPONSIVE TO B&W PLANT SENSITIVITY ISSUES BWACRS21
__ j l
NRR STAFF PRESENTATION TO THE ,
O ACRS l l
l l
SUBJECT:
ICS/NNI SYSTEMS REVIEW l
l l
DATE: May 3, 1988 l
. l 4
PRESENTER: RICK KENDALL l
O l
PRESENTER'S TITLE / BRANCH /DIV: SR. ELECTRICAL ENGINEER !
INSTRUMENTATION & CONTROL SYSTEMS I DIVISION OF ENGINEERING & SYSTEMS TECHNOLOGY PRESENTER'S NRC TEL. NO.: x20778 SUBCOMMITTEE: B&W Reactor Plants I
l O ;
l i
O O O ROLE OF ICS/NNI SYSTEMS REQUIREMENTS IN THE BWOG ICS/NNI EVALUATION NUREGS FMEAs DEVELOP ICS/NNI BWOG /
STUDIES SYSTEMS REQUIREMENTS OPERATING d EXPERIENCE COMPARISON OF DEVELOP LIST IE BULLETINS AS-BUILT ICS/NNI 0F ICS/NNI DESIGNS TO THE & PROBLEMS AND ORIGINAL ICS/NNI DESIGN REQUIREMENTS SYSTEMS RECOMMENDED REQUIREMENTS SOLUTIONS i DETERMINE
! AS-BUILT ICS/NNI CONFIGURATIONS i
ICS/NNI SYSTEMS REQUIREMENTS
- ENVIRONMENTAL SPECIFICATIONS
- INSTRUMENT ACCURACIES
- POWER SUPPLY DESIGN
- REDUNDANCY
- LOSS OF POWER ALARMS
- EFFECTS OF LOSS OF POWER '
- RESTORATION OF POWER [ KNOWN SAFE STATE
+ EASY DETECTION OF FAILED INDICATORS / RECORDERS
- SIGNAL RELIABILITY 9 INTERFACE REQUIREMENTS ,
- SYSTEM TEST / CALIBRATION / MONITORING O
KNOWN SAFE STATE i
CONDITION IN WHICH SECONDARY PRESSURE AND INVENTORY l AND PRIMARY PRESSURE AND INVENTORY ARE CONTROLLED TO l ACHIEVE PRIMARY-TO-SECONDARY HEAT TRANSFER BALANCE EITHER
) AUTOMATICALLY AND/0R MANUALLY (ACTIONS FOR WHICH THE !
OPERATOR IS NORMALLY TRAINED AND CAN BE TAKEN FROM l THE CONTROL ROOM ON LOSS OP ICS/NNI POWER )
MAINTAINING THE REACTOR COOLANT SYSTEM PRESSURE /
OF THE l TEMPERATURE RELATIONSHIP WITHIN THE B0UNDS ABNORMAL TRANSIENT OPERATIONAL GUIDELINES (ATOG)
NORMAL POST-TRIP PRESSURE-TEMPERATURE WINDOW (ALLOWS CREDIT NORMAL OPERATOR ACTIONS FROM THE CONTROL l O ROOM ON FOR LOSS OF ICS/NNI F0WER) l
_-_____-___--___-___Y
O O O ICS/NNI SYSTEMS REQUIREMENTS e MANY PROBLEM AREAS IDENTIFIED FROM THE INVESTIGATIONS OF B&W REACTOR TRANSIENTS INVOLVING THE ICS/NNI WOULD BE RESOLVED IF EXISTING PLANT DESIGNS WERE MODIFIED TO ACHIEVE CONFORMANCE WITH THE ICS/NNI SYSTEM REQUIREENTS
- 1) MINIMIZE THE EFFECTS OF SINGLE INSTRUMENT CHANNEL COMPONENT FAILURES I 2) LIMIT THE CONSEQUENCES OF POWER LOSSES BY CHOOSING EQUIPENT FAILURE POSITIONS TO ACHIEVE A KNOWN SAFE STATE d
e ICS/NNI DESIGN BASIS DOCUMENTS SHOULD BE DEVELOPED e ICS/NNI SYSTEMS REQUIREENTS DOCUENTS SHOULD BE REVISED
- POWER SUPPLY MONITOR
- LOSS OF INSTRUENT AIR i
O O O STAFF REVIEW 0F BWOG ICS/NNI EVALUATION
- CONCENTRATED ON EVALUATING WHETHER IMPLEMENTATION OF RECOMMENDATIONS WOULD RESOLVE CONCERNS IDENTIFIED FROM THE INVESTIGATIONS OF B&W REACTOR TRANSIENTS INVOLVING LOSS OF ICS/NNI POWER 3/20/78 RANCHO SEC0 -
LOSS OF NNI-Y POWER 11/10/79 OCONEE UNIT 3 -
LOSS OF ICS/NNI POWER 2/26/80 CRYSTAL RIVER -
LOSS OF NNI-X POWER 6/24/81 DAVIS-BESSE -
LOSS OF ICS/NNI POWER 1/18/83 DAVIS-BESSE -
LOSS OF NNI-X POWER 3/19/84 RANCHO SEC0 -
LOSS OF NNi-X POWER
.12/25/85 RANCHO SEC0 LOSS OF ICS POWER e DOCUMENTATION OF STAFF CONCERNS RELATING TO THE ICS/NNI 4
- IE BULLETIN 79-27 "LOSS OF NON-CLASS 1E INSTRUMENTATION AND CONTROL POWER SYSTEM BUS DURING OPERATION" (11/30/79)
- NUREG-0667 "TRANSIENT RESPONSE OF BABC0CK & WILC0X - DESIGNED REACTORS" (MAY 1980)
- NUREG-1195 "LOSS OF INTEGRATED CONTROL SYSTEM POWER AND OVERC00 LING TRANSIENT AT RANCHO SECO ON DECEMBER 26, 1985"
- (FEBRUARY 1986)
L ___ _ ________ - _ ____ _ _ _ _ _ _ _ . __
O !
CONCERNS l
'
- ADEQUACY OF CONTROL ROOM INDICATION / ANNUNCIATION OF LOSS OF ICS/NNI POWER
,
- MIDSCALE FAILURES OF CONTROL ROOM INDICATORS / RECORDERS
- ADEQUACY OF BACKUP / ALTERNATE INSTRUMENTATION l
- ADEQUACY OF PROCEDURES FOR LOSS OF ICS/NNI POWER ,
e
- INAPPROPRIATE / UNDESIRABLE EQUIPMENT RESPONSE TO LOSS
. OF ICS/NNI POWER 4
- O .
l J
- ICS/NNI LOAD FUSING l
8 LOSS OF REMOTE MANUAL CONTROL CAPABILITY FOR ICS CONTROLLED EQUIPMENT l
- RESOLUTION OF IE BULLETIN 79-27 CONCERNS
} .
i l
l0
^
~
l l
1E ButLETiN 79-27 O
- CONCERN IF REACTOR CONTROLS AND VITAL INSTRUMENTS ARE POWERED FROM COMMON ELECTRICAL SUPPLIES, THE FAILURE OF THE SUPPLIES MAY RESULT IN AN EVENT REQUIRING OPERATOR ACTION CONCURRENT WITH FAILURE OF INSTRUMENTATION UPON WHICH THE ACTIONS WOULD BE BASED 8 ACTIONS
- REVIEW THE CLASS 1E AND NON-CLASS 1E BUSES SUPPLYING POWER TO SAFETY RELATED AND NON-SAFETY INSTRUMENTATION AND CONTROLS THAI COULD AFFECT THE ABILITY TO ACHIEVE COLD SHUTDOWN USING PROCEDURES
- REVIEW THE INDICATIONS ( ALARMS ) 0F LOSS O 0F POWER
- EVALUATE THE EFFECTS OF LOSS OF POWER TO BUS LOADS
- DETERMINE IF MODIFICATIONS ARE NEEDED
- REVIEW THE ADEQUACY OF PROCEDURES USED TO ;
ACHIEVE COLD SHUTDOWN FOLLOWING LOSS OF POWER l 1
, TO EACH BUS
- REVIEW THE DIAGNOSTICS / ALARMS / INDICATORS /ETC THAT GOVERN USE OF THE PROCEDURES
- CONSIDER THE USE OF ALTERNATE INDICATION AND/0R CONTROL CIRCulTS POWERED FROM SUPPLIES UNAFFECTED BY THE POWER LOSS
- REVIEW METHODS FOR RESTORING POWER TO THE l j O bus
- RE-REVIEW IE CIRCULAR 79-02 "FAILURE OF 120 !
l VOLT VITAL AC POWER SUPPLIES" l
O APPENDIX F NRC LETTERS TO THE BWOG CONCERNING THE ICS/NNI REASSESSMENT
- 1. Letter dated June 24, 1966, from D. M. Crutchfield (NRC) to C. Dayel i (BWOG I&C Committee). This letter provides staff feedback to tivt BWOG concerning the ICS/NNI reassessment portion of the SPIP based or information presented by the BWOG during meetings with the staff on ;
April 29 and May 21, 1986. j l
- 2. Letter dated August 1, 1986, from D. Crutchfield (NRC) to H. B. Tucker l (BWOG Executive Committee). This letter requested clarification of the BWOG reassessment program with regard to retrospective i uues raised by the Incident Investigation Team in its report on the loss of ICS power event at Rancho Seco on December 26, 1985 (NUREG-1195).
- 3. Letter dated September 12, 1986, from D. M. Crutchfield (NRC) to C. Doyel (BWOG I&C Committee). This letter provided' staff feedback to the BWOG concerning the ICS/NNI reassessment based on information provided in preliminary documents developed by the I&C Committee and discussed during a meeting with the I&C Committee on August 26, 1986.
- 4. Letter dated December 17, 1986, from D. M. Crutchfield (NRC) to H. B. Tucker (BWOG). This letter provided a list of concerns identified during the staff's review of the BWOG SPIP, as documented in BAW-1919 and its revi-sions, and requested that the BWOG address these concerns in a future update of BAW-1919.
. 5. Letter dated February 13, 1987, from D. M. Crutchfield (NRC) to L. Stalter 1 (BWOG I&C Committee). This letter provided staff feedback to the BWOG concerning the ICS/NNI reassessment based on information provided in revised documents developed by the I&C Committee and discussed with the I&C Com-mittee during a meeting on December 10, 1986.
l
- 6. Letter dated April 7, 1987, from D. M. Cruchfield (NRC) to G. R. Skillman (BWOG). This letter provided staff comments and questions concerning the ;
BWOG ICS/NNI reassessment and requested that the BWOG address the associated '
concerns in a future update of BAW-1919.
- 7. Letter dated November 24, 1987, from J. A. Calvo (NRC) to R. P. Rogers (BWOG). This letter requested additional information needed for the staff to complete its review of Appendix R. "ICS/NNI Evaluation Final Report," to BAW-1919, "B&W Owners Group Safety and Performance Improvement Program (SPIP)."
O NUREG-1231, Supp. 1 1 Appendix F
O STAFF EVALVATION FINDINGS e IMPLEMENTATION OF THE BWOG RECOMMENDATIONS (APPROVED AND PROPOSED) WILL HELP TO RESOLVE A NUMBER OF THE CONCERNS IDENTIFIED FROM THE INVESTIGATIONS OF B&W REACTOR TRANSIENTS INVOLVING ICS/NNI POWER LOSSES e IMPLEMENTATION OF THE BWOG RECOMMENDATIONS (APPROVED AND PROPOSED) SHOULD HELP TO REDUCE THE ICS/NNI CONTRIBUTION TO REACTOR TRIP FREQUENCY AND TRANSI.ENT COMPLEXITY (FUTURE EVENTS INVOLVING LOSS OF ICS/NNI POWER ARE EXPECTED TO BE LESS SEVERE)
O e MORE EMPHASIS ON DICTATING EQUIPMENT FAILURE MODES (AND HENCE PLANT RESPONSE) TO LOSSES OF ICS/NNI POWER O
i l
O O O SIGNIFICANT BWOG RECOMMENDATIONS l
1 e ENSURE PLANT G0ES TO A KNOWN SAFE STATE ON LOSS OF POWER TO THE ICS/NNI (TR-178-ICS)
, e INCORPORATE AUTOMATIC SELECTION OF VALID INPUTS FOR ICS/NNI (TR-104-ICS) i e PROVIDE OPERATOR WITH UNAMBIGUOUS STATUS OF CONTROL ROOM INDICATORS AND RECORDERS ON THE LOSS OF ICS/NNI POWER OR SIGNAL (TR-154-ICS) e ELIMINATE MID-SCALE FAILURES THAT CAN AFFECT INDICATION OR PLANT CONTROL (C.8) e DEVELOP BACKUP CONTROLS FOR PRESSURIZER LEVEL AND PRESSURE CONTROL (TR-190-ICS) i e PROVIDE INDEPENDENT CONTROLS FOR REACTOR POWER, STEAM DEMAND, AND FEEDWATER (C.1) i i .. .. . . - _ . _ . . . . . _ .
O O O SIGNIFICANT BWOG REC 0ffiENDATIONS e PROVIDE SEPARATE SUBSYSTEMS FOR REACTOR COOLANT TEMPERATURE AND STEAM PRESSURE CONTROL (C.7) e ENSURE PREFERRED EQUIPENT FAILURE MODES ON LOSS OF ICS/NNI POWER (TR-036-ICS, TR-037-ICS, TR-096-MSS, TR-097-ICS, TR-172-PRV) e IMPROVE ANNUNCIATION / INDICATION OF ICS/NNI POWER LOSS (TR-012-ICS, TR'158-ICS, B.2.4, B.2.9)
- DDEt0p AND IMPLEENT A PREVENTIVE MAINTENANCE PROGRAM FOR THE ICS/NNI (TiF038-ICS) e IMPROVED MAINTENANCE AND TUNING OF ICS (TR-107-ICS)
I e IMFROVED SURVEILLANCE AND TEST PROCEDURES / PROGRAM FOR EFW
'TPc163-EFW, TR-164-EFW, TR-165-EFW) h
i i STAFF EVALUATION FINDINGS -
CONCERNS I
i e SURVEILLANCE FREQUENCY FOR ICS/NNI EQUIPMENT i e MAXIMlZE DEPENDENCE ON NNI-X (TR-189-ICS, B.2.1/1) t i
e ELIMINATION OF ICS FUNCTIONS l
- ICS RUNBACK ON ASYMMETRIC R0D CONDITIONS (TR-204-ICS) I t
- DELTA T-COLD FEEDWATER CORRECTION (C 5) 4
- BTU LIMIT FUNCTION (TR-007-ICS) ,
! O ,
e WIRE POWER SUPPLY MONITOR DIRECTLY T0 ICS/NN1 i24 VDC BUSES (TR-039-ICS) '
4 e FSAR ASSUMPTIONS RELATING TO ICS/NNI e IE BULLETIN 79-27 CONCERNS 4
l
- INDICATION OF POWER LOSS l
- PROCEDURES
- BACKUP / ALTERNATE INSTRUMENTS AND CONTROLS
! (INCLUDING SURVEILLANCE) a l
1 _ - ,
O STAFF EVALVATION FINDINGS -
CONCERNS e GUIDANCE PROVIDED IN SOME RECOMMENDATIONS MAY BE T00 GENERAL TO ENSURE RESOLUTION OF CONCERNS e THE INFORMATION PROVIDED IN BAW-1919 IS NOT SUFFICIENT FOR THE STAFF TO CONCLUDE THAT ALL OF THE CONCERNS IDENTIFIED FROM THE INVESTIGATIONS OF B&W REACTOR TRANSIENTS INVOLVING THE ICS/NNI WOULD BE RESOLVED UPON IMPLEMENTATION OF THE RECOMMENDATIONS 4
O 1
i O
l
O
SUMMARY
OF STAFF CONCLUSIONS e BWOG REVIEW PLAN FOR THE ICS/NNI EVALUATION IS CONSIDERED EXCELLENT e IMPLEMENTATION OF BWOG RECOMMENDATIONS RELATING TO THE ICS/NNI SHOULD PROVE SUCCESSFUL IN ACHIEVING A REDUCTION IN REACTOR TRIP FREQUENCY AND TRANSIENT COMPLEXITY e INFORMATION PROVIDED IN BAW-1919 IS NOT SUFFICIENT FOR THE STAFF TO CONCLUDE THAT ALL ICS/NNI CONCERNS WILL BE RESOLVED AS A RESULT OF SPIP O
e ADEQUACY OF B&W PLANT SAFETY SYSTEMS TO MITIGATE THE CONSEQUENCES OF TRANSIENTS INVOLVING ICS/NNI FAILURES IS NOT AN ISSUE e IE BULLETIN 79-27 WAS CONSIDERED TO BE A PLANT SPECIFIC ISSUE, AND THEREFORE, NOT WITHIN THE SPIP SCOPE. THE STAFF HAS RECOMMENDED THAT RESOLUTION OF IE BULLETIN 79-27 CONCERNS BE EVALUATED ON A PLANT SPECIFIC BASIS (AUDIT REVIEWS) e IT APPEARS THAT THE BWOG SPIP EFFORT WILL BE SUCCESSFUL IN ACHIEVING SIGNIFICANT IMPROVEMENTS IN ICS/NNI DESIGNS AND OPERATION O
.1
OO Q
h W
B&W PLANT REASSESSMENT NUREG 1231 INFORMATION GATHERING .
PRESENTED TO PRESENTED BY ACRS SUBCOMMITTEE ON B&W REACTORS R.C. JONES MAY 3-4, 1978 NRR/ DEST /SRXB BWACRS2
p t'
L STAFF ACTIVITIES REVIEW OF SPECIFIC BWOG REPORTS INDEPENDENT ACTIVITIES' ,
- AUDITED 40 TAP REPORTS ,
REVIEWED YEARLY TAP
SUMMARY
REPORTS REVIEWED PREVIOUS NUREGS
- OBTAINED REGIONAL INPUT EXAMINED STATUS OF UTILITY COMPLIANCE TO NRC ACTIONS
- REVIEWED DAVIS-BESSE DECAY HEAT RELIABILITY IMPROVEMENT PROGRAM FOR GENERIC IMPLICATIONS :
I BWACRS17
^
n
- - ~~
~ . . . . ,- . ,
i.
A i
STAFF; FINDINGS l
MAJOR REACTOR TRIP INITIATORS WERE IDENTIFIED
- TURBINE SYSTEMS
- MFW SYSTEM AND CONTROLS
- ICS/NNI INPUTS ,
COMPLEX TRANSIENTS PRIMARILY RESULT OF -
FAILURE TO BALANCE HEAT REMOVAL / PRODUCTION
- MAJORITY OF EVENTS WERE OVERCOOLING (34 OF 46)
- ONLY 4 WERE UNDERCOOLING SPECIFIC RECOMMENDATIONS DEVELOPED WERE APPROPRIATE t
SYSTEMS IDENTIFIED FOR DETAILED REVIEW APPROPRIATE j i
OVERALL THRUST OF SPIP RESPONSIVE TO OPERATING EXPERIENCE BWACRS18
_ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ ___-_m__-.____ __-_ .__ - _ --_m______--. _____-___ __ _ _.-__ _ _ _ _ - - - - _-.__.m__ _ _ _ _ _ w
- NRR STAFF PRESENTATION TO THE !
- l l
l
SUBJECT:
B&W Owners Group Plant Reassessment Program -Overview of B&W Plant Reassessment Program DATE: May 3, 1988 ;
1
)
, p* ,
I PRESENTER: Byron Siegel
~
9 i PRESENTER'S TITLE / BRANCH /DIV: Lead Project Manager, B&W Owners Group Plant Reassessment Program i PRESENTER'S NRC TEL. NO.: 492-3019 SUBCOMMITTEE: BaW Reactor Plants ,
1 l
l 4
STAFF APPROACH (l) ;
l i
CONTINUOUS INTERACTION WITH BWOG ON SCOPE OF SPIP PROJE REVIEW AND EVALUATE SPIP RESULTS (BAW-1919)
INDEPENDENT WORK PERFORMED IN CERTAIN AREAS RISK EVALUATION (BROOKHAVEN)
HUMAN FACTORS (SAI)
LIMITED THERMAL HYDRAULIC ANALYSIS (INEL)
LIMITED OPERATING EXPERIENCE REVIEV
- us e OVERSIGHT ROLE EVALUATE REC 0KMENDAT10t! APPROVAL PROCESS (BWOG) ggg EVALUATE IMPLEMENTATION PROGRAM AT UTILITIES EVALUATE IMPLEMENTATION PROCESS '
VERIFY IMPLEMENTATION TRACK IMPLEMENTATION PROGRESS l
l L
1 SIEGEL 5/3/N4l
SER AND SSER 4
CONTAINS N0 NEW POSITIONS CONTAINS NO REQUIREMENTS, ONLY RECOMMENDATIONS OFFERED FOR CONSIDERATION SER ADDRESSED 9 0F 11 MAJOR PROJECTS BWOG PROGRAMMATIC & MANAGEMENT ACTIONS .
IMPLEMENTATION I
- ADDITIONAL STAFF RECOMMENDATIONS
, CONCERNS IDENTIFIED BY STAFF MEMBER -
f.
SSER ADDRESSED O
- INTEGRATED CONTROL SYSTEM /NON-NUCLEAR INSTRUMENTATIO l-
- REACTOR TRIP INITIATING EVENTS REVIEW i
ACCEPTABILITY OF EFIC TYPE SYSTEMS l STAFF IDENTIFIED HIGH PRIORITY RECOMMENDATIONS
- STAFF IMPACT ASSESSMENT OF PROPOSED BWOG RECOMMENDAT l
MISC, OPEN ITEMS IN SER ADDITIONAL STAFF RECOMMENDATIONS j REMAINING CONCERNS IDENTIFIED BY STAFF MEMBER COMMISSION ISSUES IDENTIFIED IN CHILK MEM0 4
l- '
= ~
.in o .
s.
O SIEGEL 5/3/88
- _ _ _ -