ML20153H684
| ML20153H684 | |
| Person / Time | |
|---|---|
| Issue date: | 05/04/1988 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1664, NUDOCS 8805130100 | |
| Download: ML20153H684 (266) | |
Text
. _ .,. __ - ._ . - - . -- - - - - - - .- _ - - -
L, R a \A _
g iUNITED STATES NUCLEAR REGULATORY COMMISSION Ill THE MATTER OF: ) ,
)
BABCOCK & WILCOX )
i
)
REACTOR PLAT 3TS )
- O t
LOC ATIOtJ : Washington, D.C.
s t' i~W f PAGES: 318 through 537 , 7; DATE: May 4, 1988 f' r - i . f
, o l
'e. br =, , u i ~ t.J m [G n:0,-
L ,
,1 4,..
j .............-=== --============================="""""""""
9 HERITAGE REPORTING CORPORATION opewanarm O- 1224 L Street, N.W., Sake 640 WanWagton, D.C. 20065 ,
(202) 628-4888
- geos130too e80504 l F'DR ACR$ DCD T-1664 l
PUBLIC NOTICE BY THE 9 1 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 inaccuracies of statement or data contained in this transcript.
("]
LJ 14 15 16 17 l 18 19 20 21 22 23 24 25 Heritage Reporting Corporation (202) 628-4888
318 1 UNITED STATES NUCLEAR REGULATORY COMMISSION 2 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS l
3 1 4 In the Matter of: )
)
5 )
SUBCOMMITTEE ON BABCOCK & )
WILCOX REACTOR PLANTS )
7 Tuesday, 8 May 4, 1988 9 Room 1046 1717 E 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. CRARLES J. WYLIE Retired Chief Engineer .
14 Electrical Division Duke ?ower Company 15 Charlotte, North Carolina l 16 ACRS MEM3RRS 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 y 1
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
! 2 ?, - U.S. Nuclear Regulatory Commission Washington, D.C.
24 25 -
319 1 NR. 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 F. Davis 12 H. Etherington
. 13 14 15 16 17 18 19 1
20 I !
l 21 22
! 23 24 .
i - ;
i i'
(202)628-4888 l
C) 1 PROCEEDINGS 2 MR. WYLIE: This is a continuation of the 3 subcommittee.on Babcock and Wilcox reactors. The subject is 4 the Babcock Wilcox Owner's Group Plant Reassessment Program.
5 We'll continue from last-evening.
6 I'd ask the subcommittee members and consultants to 7 speak up, and others making presentations to speak up and 8 speak into the microphones in making their presentations so 9 that everyone can hear them.
10 I'd like to begin by asking the subcommittee members 11 and consultants to identify those questions which were 12 deferred from yesterday that they still want answers to so
) 13 that we'll be sure and cover those.
14 MR. REED: I have four. I don't think I got an 15 answer to the PRV issue, or the fact that it's a single valve, 16 its reliability, the kind of valve. And related to that is 17 the issue of how do they reasonably and rapidly depressurize 18 in case they had a tube leak and they wanted to get the 19 pressure down and keep it from further opening.
20 The second deferred question had to do with the 21 claimed advantage of primary makeup for charting pumps running 22 continuously and having a running system, and the advantages 23 of a running system versus a standby system or shutdown 24 system.
25 Along with that one I want to make a point. I don't Heritage Reporting Corporation (202) 628-4888
321 0
1 know that the charging pumps are really a running system. I 2 think they have to shut down in order, if you have a loss of 3 AC , a very simple accident, the pumps have to be transferred 4 from normal outside AC to the diesels or something, and so is 5 it really a running system, and is it really an advantage? I 6 think that's a key and important issue for this whole 7 business, to make that decision.
8 The other thing I'd like to know is the reasons why 9 Babcock and Wilcox in their later sold plants, not entirely 10 sold, but in Bellefonte and WOOPS and Davis-Besse and they 11 went to the raised loop plant. What was the fundamental 12 reason behind that since it obviously contains much much more 13 faulty sale.
14 The fourth thing is the issue of Oconee atd Rancho 15 Seco and Three Mile Island not having main steam stop valves.
16 And going along with that, they have non-return valves. I 17 probably should have looked it up, but do they? And even if 18 they do, it sets up the situation for a steam line break 19 inside containment, which I would like to know if the 20 containment was designed for.
21 MR. WYLIE: Okay. Any others?
22 MR. MICHELSON: I just wanted to make a couple of 23 comments. I think the presentations were very good and gave 24 excellent information. I really am quite concerned though,
{} 25 that this is another example of the same problem we ran into Heritage Reporting Corporation (202) 628-4888
t i
322 l
('JS 1 on A47 and that is the scopes of the safety investigations are 2 being very narrow. We wrote this in our A47 letter. I had 3 asked yesterday whether B&W had any reaction to it. I wasn't 4
~
l 4 sure they'd seen it. If we need more copies we can get you 5 copies of it. I would just like to have them read it and see [
6 if they are covering part of what we seem to be concerned 7 about, or are they leaving the same things out that were left 8 out in the A47 review. I'd like to get their view on that, 9 perhaps before they leave today. [
10 How many copies would you like to have?
11 VOICE: We'd like to have ten copies.
l 12 MR. MICHELSON: We'll just ask you to give us 13 whatever views you have. This is essentially the same 14 problem. It's the control system for feed water and we were 15 concerned about the way it was looked at in the A47 work that i 16 the staff had done, and we'd just like to know if you're ,
17 looking at it any differently or are you leaving out the same 18 things they left out in their investigation?
19 I believe that takes care of my comments for the 20 moment, at least.
21 MR. WYLIE: Any others that anyone wishes to
- 22 identify?
f 23 Let me ask the group when.they'd like to address 24 this?
25 MR. RUTHERFORD: I think the question was first of
[
Heritage Reporting Corporation (202) 628-4888 ,
323 O
1 all on A47, we'll have to take a look at the concerns, later 2 on. I think some of Mr. Reed's questions are clearly beyond 3 the scope of what we've covered, and I would not propose to 4 respond to those today.
5 MR. WYLIE: When would you respond to them?
6 MR. RUTHERFORD: We'll be glad to talk to Mr. Reed 7 any time, to explore that subject, but not in the contents of 8 this meeting.
9 MR. REED: I don't know that that would be an 10 appropriate exchange. I think the Chairman would have to 11 decide whether I should be talked to individually as a 12 consultant. I would prefer that it not be that way, that it 13 be open and on the record.
14 MR. WYLIE: Do you prefer not to answer these 15 questions?
16 MR. RUTHERFORD: Some of them we're not prepared to 17 respond to. And like I say, some are beyond the scope or view 18 of the B&W plants. I don't think they can be addressed in 19 this particular forum.
20 MR. WARD: What about getting some written responses 21 in the near future?
22 MR. %YLIE: Why don't we do that. We'll submit 23 these formally then, as questions and you can answer them from 24 that.
25 MR. RUTHERFORD: That's fine.
Heritage Reporting Corporation (202) 628-4888
324
. Q(_/
1 MR. WYLIE: Mr. Reed, you'll prepare those for us?
2 MR. REED: Yes.
3 MR. WYLIE: Harold?
4 MR. ETHERTON: Maybe I'm out of order now, but I had 5 hoped that Glenn was going to say a little bit more because I 6 wanted to givo him a little support in connection with the, is 7 it all right for me to express what I think you had in mind, 8 Glenn?
9 MR. REED: Sure.
10 MR. ETHERTON: In the event that the natural 11 circulation was lost, you have this cold loop at the bottom 12 which acts as a seal and may make it difficult or impossible 13 to restart natural circulation.
14 There is an analogy for this which I'm sure B&W 15 engineers of 70 years ago were familiar with. If you have a 16 furnace with a deep flue and a chimney, you can't just light 17 up the furnace and let it run. There's a cold loop below the 18 ground that will not clear. It isn't a question of time. You 19 can go on pumping fuel into the furnace forever and it won't 20 start. You'll have flames coming out of the doors and maybe ,
21 through the brickwork. You have to clear the chimney first to 22 get rid of this cold loop, and you can do that by lighting a 23 fire at the bottom of a stack. You can use a steam jet.
24 In this case, Glenn is visualizing a cold loop at
{} 25 the bottom and it may not clear easily. I think it would Heritage Reporting Corporation (202) 628-4888
_ s
l C) 1 clear ultimately. But in doing so you might get quite a high 2 hot leg temperature, and conceivably even boiling. Then of 3 course you would have these pressurized elaborations that :
4 caused all the trouble at Three Mile Island.
5 I don't really think there is a problem, but this l 6 could be resolved in two hours by looking at some steam tables 7 for border densities and I don't think this should go on 8 forever, at least on theory. But I think we should know if we 9 did develop a static condition what it would take to reach out 10 and how hot the hot leg temperature might rise, and whether 1
11 you would in fact get any boiling.
12 Have I more or less expressed your thoughts on that?
, 13 MR. REED: You have expressed something that I
- 14 didn't know about. I would want to say that I don't know that i
15 I could agree with your two hours and some tables and
) 16 references and perhaps a few doctors of thermodynamics 17 studying it, because this thing has been gone through before.
18 Thermoblock occurred in the first PWR ever built in the United 19 States of America. For two years it was a controversy after l 20 that to decide whether or not the caic heat could have been i
l'
! 21 removed and whether or not thermoblock did occur.
! i i
22 I might say that as an operating type person l 23 involved, a measurements person involved, my position at that ,
24 time was that thernoblock had occurred and there was no flow.
25 Sece. two years later one of the doctors of thermodynamics did I Heritage Reporting Corporation !
, (202) 628-4888
326
(
1 call me and say yes, you were right. Thermoblock did occur 2 and there was no flow.
3 So I'm not so sure that two hours will work. I 4 think really this is an issue, if I was involved, I would have 5 to do a number of tests, different types of trips, different 6 types of feeds or lack of feeds, and see if you could span the 7 worst case and you would always get initiation.
8 MR. ETHERTON: That's postulating a block, no matter 9 how you get it. The worst possible case, and then if you 10 find that clears then you don't need to worry much about --
11 MR. REED: Yes, but you mentioned the issue of time.
12 How long to core damage? That's the important issue.
13 MR. SKILLMAN: I would like to respond to this 14 please. The period from the TMI-II accident until January 1, 15 1980, I was head of support engineering for Babcock at THI-II.
16 From January 1 of '80 until late '82 I was head of recovery 17 support engineering at Three Mile Island Nuclear Station, Unit 18 II. Approximately 30 days after the accident TMI-II was 19 secure. The final reactor coolant was -- That was the 20 driver, the daily pump that didn't vibrate. We stopped that 21 pump when we realized we could no longer see a primary coolant 22 level in the pressurizer. We had worked earnestly for the 23 month following the accident to ensure that the primary 24 coolant system was cool and pressurized.
25 The question which we had was Babcock, combustion
[}
Heritage Reporting Corporation (202) 628-4888
327 1 engineering, was wouldn't the TMI-II core naturally circle?
2 Would the system cool itself? Our numbers showed that it 3 would. The combustions number showed that it would; 4 Westinghouse's number showed that it would. We stopped the 5 primary coolant pump, TMI-II's reactor coolant system, and I 6 have to say in hindsight, the core blockage, primary system 7 blockage, ran in smooth natural circulation until October 8 1979. It did so with what we calculate to be about 3000th of 9 a psi, differential pressure, between the core column 10 hydraulic height and the steam generator column height at the j 11 identical elevation. It was simply the differential density 12 between the block of water in the core, and at that same 13 elevation the block of wa'ter in one generator or the other.
14 The best is the next story. When smooth natural 15 circulation terminated, we were concerned that natural l Initially there was a four 16 circulation would not restart.
17 hour time delay between the stoppage of smooth, natural 18 circulation and the automatic restart. The reactor coolant l
l 19 system of TMI-II persisted for the next year, 400 and some 3
20 cycles by itself, stopping natural circulation and then by 21 itself starting natural circulation. So we were in a long 22 period of what we called intermittent natural circulation.
23 Natural circulation finally terminated itself when 24 the DPE generation rate was so low that the reactor vessel, 25 the internals, and the amhiant around the reactor vessel was i Heritage Reporting Corporation (202) 628-4888
-, - - , r-,- - n ,n-. --..--_...--,~,n------,..,-,..n-,-
l 328 l
1 able to absorb the heat. We retained a minor flow to the 2 steam generator during that period in the neighborhood of 15 3 gallons a minute. Clearly we had water in the generator, we 4 had generator water levels available to us.
5 But in response to the concern that's been 6 expressed, I would have to say, being an operator and being 7 involved first hand, the TMI-II experience with massive core 8 blockage with 36 inch ID hot legs, and 26 inch ID cold legs, 9 and 30,000 tubes generally available, natural circulation 10 worked just fine, and not in a test facility, but in an actual 11 facility.
12 MR. REED: I don't think your one month later sub-13 cooled liquid situation is at all relevant to right after 14 trip, transients, and all these other kind of things. I have 15 to say that what you are talking about is an ideal condition 16 by comparison.
17 MR. SKILLMAN: I beg your pardon. That wasn't ideal 18 in any sense .it all.
19 MR. REED: The only difference is pressure drop in 20 the core.
21 MR. SKILLMAN: That's a big difference, Glenn.
22 MR. REED: Well let me say, having gone through the 23 first nuclear PWR, thermobl.ock situation, right after trip, 24 right after power with the varying fluids and conditions and 25 having seen thermoblock occur, my story is exactly 100 percent Heritage Reporting Corporation (202) 628-4888
(:) !
1 different from your story one month later. !
2 The issue is, probah'.Ly the heat's not very much 3 after a month, either, j 4 MR. MICHELSON: I think there have been a couple of 5 real plants that have had to undergo instantaneous !
6 recirculation from the loss of off-site power conditions. -'
7 MR. ETHERTON: That's true. !
8 MR. MICl!ELSON: Those went through fine, as I ;
9 recollect.
10 MR. ETHERTON: Do we know the circulation ever l
?
11 completely stopped?
i ;
12 MR. MICHELSON: No, you don't know that the momentum L i
13 cosine was.
) 14 MR. ETHERTON: If they stop completely, that's the ;
1 15 only time - - f 16 MR. MICHELSON: That makes sense. l 17 MR. ETHERTON: If it was related, did it stop I
18 circulating all together? [
l 19 MR. SKILLMAN: Yes sir, our temperature indication I i r
, 20 would indicate that not only did it stop, it reversed for some j i 21 number of minutes. In other words, it moved forward, stopped,
. 22 regressed, and then automatically began flowing forward again. i I
23 The time period between those, if you will the period, this ;
5 24 began at about four hours and then it continued to expand into f 25 days over the course of approximately a year. But it was i
Heritage Reporting Corporation l (202) 628-4888 1 [
-s 330 1 stopped, we believe based on our reactor coolant system 2 indication which is accurate and timely. It tended to regress 3 and then propel itself forward again.
4 MR. TAYLOR: Mr. Chairman, I'd like to just make a
' 5 comment. I think the dialogue that took place between these 6 folks here really tends to amplify the importance of asking 7 that these questions be put in some kind of a context, because 8 I can see us going back and forth on this thing a couple of 9 times. I would only ask Mr. Reed to try to really clarify the 10 context in which the questions are asked, and particularly if 11 it would be possible to relate them to the complex transient 12 issue which this was really aimed at addressing. Just a 13 request.
14 MR. WYLIE: Well the name of the game is 15 reassessment of the safety aspects of B&W reactors as far as 16 the ACRS subcommittee is concerned. As to whether we're in a 17 small box or a larger box is the real question. The scope and 18 breadth of the program is the question. That's a valid 19 question, regardless of whether you want to put us in a small 20 box or a large one.
21 MR. TAYLOR: The thing I was trying to get at, Mr.
22 Chairman, is that there have been many many tests run, there 23 have been many many reports written about natural circulation.
24 We're kind of curious as to whether these things are outside
{} 25 the scope of these previously submitted documents and Heritage Reporting Corporation i (202) 628-4888
l
/ \
331 l
('_'N) i 1 previously answered questions or whether they're within that l 2 and perhaps they're just not available to the subcommittee.
3 MR. WYLIE: I think the subcommittee has the 4 responsibility to ask whatever questions we need to assess our 5 opinion regarding the adequacy of the reassessment program.
6 MR. TAYLOR: We agree with that, and we just want to 7 make sure we understand the questions.
8 MR. MICHELSON: Mr. Chairman, let me ask for 9 clarification. In reading Stello's letter it appeared that we 10 were looking at the overall safety of the plant.
11 MR. WYLIE: That was the subject.
12 MR. MICHELSON: That appeared in Stello's letter.
13 The staff, it appears, has narrowed the assessment because B&W 14 narrowed the assessment down to these 13 transients of 15 particul.ar interest, if I recall the numbers correctly. So 16 where does that leave us? If we want to look at the overall 17 safety we have to ask is there anything unusual about the 18 accident responses which were not treated by B&W or by the 19 staff. The other question being external event responses, 20 vhich again, were not treated by either B&W or the staff.
21 MR. RUTHERFORD: Let me offer a perspective.
22 Certainly the letter was issued because we had complex 23 transient, or had experience with complex transients on the 24 B&W plants. So I think for us to, Mr. Stello's letter was to 25 look at those aspects of the plants. Otherwise, I think it
}
Heritage Reporting Corporation (202) 628-4888
1 i
rs 332 O
1 would have been appropriate to go out to every vendor and 2 plant in the country and relook at these other issues, tube 3 breaks or small break locals or external events. We're no 4 different, essentially, from other plants in that respect.
5 Anything about our plants where the experiences that we were 6 having. That's what we went after, and I think really that 7 was the proper thrust behind the Stello letter.
8 MR. JONES: If I could offer a perspective from the 9 staff. We did look at the Stello letter when it came out in 10 January and we put together an overall program plan before the 11 owners group was involved and had defined their spit program.
12 What you are seeing here in the last two days and in the 13 specific documents is the staff's assessment of the spit 14 program and the other activities performed by the staff under 15 essentially that program plan.
16 That program plan was modified to incorporate or 17 reflect what the owners were doing, what the staff was going 18 to do. But that plan was developed very early, had management 19 approval as of roughly March of '86 and was transmitted both 20 to the ACRS, the EDO's office, and the owner's group in or 21 around that time frame.
22 So we did not narrow the scope of this program 23 because the only proof that it's spit, we incorporated spit 24 and integrated it as part of the program in order to conserve
/~ 25 our resources and to get a broader view by the people that V}
lieritage Reporting Corporation (202) 628-4888
(' 333 (m]/
1 best understand the plans which are the specific B&W Owners 2 Group Utility Members.
3 MR. WYLIE: Can I conclude from that that you have 4 a larger program in progress?
5 MR. JONES: No, this is the program. This program, 6 to the best of my knowledge, satisfies the Stello letter.
7 That program was, said that copies were sent to the EDO's 8 office, and when we were negotiating in the May/ June time 9 frame, the content of the spit program, when we were trying to 10 get them to do more so we could do less, so to speak, from an 11 original thinking type activity where they would take the 12 brunt of the workload and we would be into a more historical 13 look.
14 Representatives from the EDO's office were at those 15 meetings. As I've said, to the best of my knowledge, we have 16 never heard any complaints from the EDO's office that this 17 program did not satisfy the intent of his letter. In fact we 18 briefed the Commission on this in '87, and as far as I know, 19 Stello was happy with the scope of the program. He understood 20 what we were doing.
- 21 So I don't consider that we narrowed the scope from i
22 what Mr. Stello intended. I think that is an interpretation 23 that the subcommittee is making. As far as we know, we are in 24 concert with the intent of that letter. At least what Mr.
(} 25 Stello's thoughts were.
Heritage Reporting Corporation (202) 628-4888 1
334
\_/
1 MR. MICHELSON: Can we get a clarification as to 2 whether or not the ACRS reviewed that plan before we wrote our 3 letter in June? I don't recall.
4 MR. WYLIE: I don't think we did.
5 MR. MICHELSON: I don't recall ever seeing it, but 6 that could be verified. Maybe Richard could find out in a 7 little while whether we saw it.
8 MR. KERR: Gentlemen, let me suggest that I think 9 this subcommittee and the committee has the responsibility to 10 pursue questions which it considers relevant to reactor safety 11 independently of what Mr. Stello may have written or what the 12 staff may conclude. We may be wrong, but our responsibility
/ 13 is to pursue those things that we consider important.
14 If we are simply repeating what we have done 15 before,that's unfortunate. Sometimes we do that. Sometimes 16 we ask questions from ignorance, and I would hope that we 17 would avoid that insofar as feasible. But I don't think this 18 committee should feel constrained about what somebody else has 19 written in a letter or whatever. We certainly should be 20 responsible, but we should pursue those things that we 21 consider relevant to safety.
22 MR. WYLIE: Thank you, Dr. Kerr.
23 Let me advise you that we are beginning this meeting 24 by asking for identification of outstanding questions from 25 yesterday that were deferred. Various subcommittee members
{
Heritage Reporting Corporation (202) 628-4883
/^\ 335
(_)
1 have identified some, and I don't know whether you had any 2 that you wanted that were deferred from yesterday.
3 MR. KERR: I have none.
4 MR. CATTON: I'd like to ask a question.
5 The scope of the owners group study was limited to 6 transients where there was no two-phased flow. Is there any 7 reason for this?
8 MR. RUTHERFORD: The program was based on the 9 complex transients that we have experienced and the reasons 10 behind those complex transients. Certainly the things we're 11 doing in this program have benefits in regard to a lot of 12 transients beyond what we looked at. That was a scope of our 13 study, what drove us.
14 MR. CATTON: The reason I ask is that I'm not 15 surprised at the conclusions they come to. Most of the 16 differences, at least from my point of view, result when you 17 have two-phased flow. The loss of natural circulation because 18 of the candy cane or something and the vent valves. Things 19 like this really only become important when you have two-20 phased flow in the primary system. I think they ought to have 21 looked at it.
22 MR. RUTHERFORD: Realize that a number of things 23 have been done outside the context of this particular program.
24 We've had the Miss facility and the testing that has been done 25 there. Small break locos received extensive look-sees and Heritage Reporting Corporation (202) 628-4888
336
(
1 modeling etcetera. A wealth of studies in these areas. I'm 2 not saying we haven't done anything at all, but we didn't do 3 anything in the context of this particular program.
4 MR. CATTON: The question is plant sensitivity. I 5 don't know that plant sensitivity should be linited to just 6 what you did. It should be the whole spectrum. That spectrum 7 includes two-phased flow in the primary system if it's 8 important.
9 MR. RUTHERFORD: Those issues to a limited degree 10 were looked at in the sensitivity study.
11 MR. CATTON: I asked the question yesterday and was 12 told no, and it was one of your people who was speaking when I l I's ')
13 asked.
14 MR. RUTHERFORD: From the standpoint of looking at 15 small break locos, that's true, we did not look at the two-16 phased flow.
17 MR. WARD: But Ivan, the reason these plants have 18 been singled out for this look was people got worried, 19 nervous, about the series of events that had occurred. There j 20 is sort of an agenda other people have who have been nervous 21 for some reason about the design of B&W plants for some other 22 reason, but not related to this particular set of experiences.
l 23 But it's that set of experiences which really drove the EDO to i 24 write the letter and not this other agenda coming from l
{} 25 somewhere.
Heritage Reporting Corporation (202) 628-4888 m
( 337 V
1 MR. CATTON: So if sensitivity is limited to this 2 particular cloth, then my comment has no meaning. But it 3 seems to me sensitivity is --
4 MR. WARD: I think there are probably two questions.
5 With regard to this set of experiences which got us into this, 6 raised the question. The sensitivity study provides some 7 answers about that. Now you've got another question, what 8 about sensitivity in accidents that go into two-phased flow.
9 MR. CATTON: What we worry about is core damage. So 10 when you go through this series of transients, you're coming 11 up right to the edge. They catch it every time. Does 12 sensitivity beyond that point matter? If it does, then this 13 study is incomplete. If you want to stop it right at that 14 point, then you're right and this is an irrelevant concern.
15 MR. WARD: All I'm saying is what leads you to the 16 concern that there may be some unusual problem or sensitivity 17 beyond the point of which has been studied.
18 MR. CATTON: We know basically there are three 19 differences in the plant. There's the tall candy cane, 20 there's the vent valves, and the 116 generator. What role do 21 they play between the limit of the transients we've looked at 22 and the degraded core? Do they make the process worse from 23 that point on or is it better or what? Is the system more 24 sensitive because of it, or isn't it?
{} 25 MR. WARD: Mare sensitive than what?
Heritage Reporting Corporation (202) 628-4888
l
/^N 338 I L) .
1 MR. CATTON: More sensitive than the other PWRs.
2 Isn't this a relative sort of study?
3 MR. WARD: I don't know. I think you're suggesting 4 another study, and it may be appropriate. But it's clearly 5 another study.
6 MR. CATTON: That's fine.
7 MR. WARD: Have you concluded that the other PWR's 8 are appropriately insensitive or stable or something in the 9 range you're talking about?
10 MR. CATTON: No, I've made no such conclusions. 'But 11 when I read this, the question was sensitivity relative to the 12 others.
t
] 13 MR. WARD: But it's sensitivity in the range of 14 interest that's been pointed out to us by these events that 15 have occurred, and none of those have involved two-phased 16 flow.
l 17 MR. CATTON: That's true because for some reason or I
l 18 another they were caught soon enough. We wouldn't worry about i 19 them at all unless there was the potential for getting into 20 the two-phased flow because that's where core degradation 21 comes from.
22 MR. WARD: I think the conclusion is that, the way 23 it looks to me, it's bad news for a PWR to get into two-phased 24 flow. They're not designed for two-phased flow, so it's bad l
l
{} 25 news. So the question is does this design of a plant have Heritage Reporting Corporation l (202) 628-4888 l
1
1 1
)
l "T 339 (V
1 unusual sensitivity in the pre-two-phased flow region so it's 2 more likely to get over this threshold into two-phased flow?
3 Now I don't know whether we agree with their study, but the 4 MPR study seems to be saying well, probably not all that much.
5 MR. CATTON: If that's the question, then I think 6 the study was okay.
7 MR. WARD: I think it is certainly a question, and 8 it's a good question. There may be other questions.
9 MR. CATTON: That's right.
10 MR. WARD: Mr. Chairman, I think we've gotten into 11 this problem of scope and disagreement on what the scope 12 should be really because of something that happened a couple 13 of years ago, a year and a half ago. After we had the first 14 meeting I think the committee fully expected to have some 15 follow up meetings to further discuss the scope of the study.
16 Our original concern was whether the staff was going to be 17 able to accommodate this operating reliability study which the 18 B&W owners group had started on, and which was certainly 19 appropriate for them to be doing, whether the staff was going 20 to be able to accommodate that to the staff's concern which 21 was talking about safety rather than operating continuity. I 22 think we wanted to have some follow up meetings at that time 23 to talk about that and review that. For some reason those 24 meetings kept getting postponed for 18 months until this one,
{} 25 and now the chickens are home to roost, I think. The Heritage Reporting Corporation (202) 628-4888
23 340 N) 1 subcommittee, obviously, isn't satisfied that the staff has 2 successfully accommodated and combined these two programs. I 3 don't know what we can do except stumble through it.
4 MR. WYLIE: Thank you.
5 MR. MICHELSON: Let me make one more comment. I 6 just looked briefly at B&W's 1919 document to see if they've 7 made the same overstatement of scope. So far just flipping 8 through, it appears that what they are doing and they say here 9 is "a major expansion of efforts focused on reducing the 10 complexity of transients and frequency of reactor trips."
11 That's what they did. But I read the staff's assessment of 12 this whole thing and it says in here, referring back to the 13 B&W document, it says, "The study compares the overall safety 14 of B&W plants with that of other pressurized water reactors,"
15 which the study clearly did not do.
16 MR. CATTON: That's correct.
17 MR. MICHELSON: Yet the staff keeps pushing because 18 it came from Stello's letter that says you are to "look at the 19 overall safety," and those are the words that Stello said. He 20 talked about overall safety. They just didn't do it.
21 So my concern is that they're overselling what was 22 done. They're going to come back and say we've looked at the 23 overall safety of B&W plants and it's no worse than the 24 others, and I can't agree with that. They haven't looked at
{} 25 it. B&W did a very fine job of looking at these transients Heritage Reporting Corporation (202) 628-4888
l 341
'/
1 and seeing what they can do about them in terms of safety of 2 the plant and I think it was a fine job, but it was not the 3 overall safety they were looking at. They were looking at a 4 restricted aspect of it. That's why I really get a little 5 concern about the way it's being oversold for what it is.
6 MR. WYLIE: Thank you. Any other comments or 7 questions that we want to identify?
8 If not then let's proceed with our meeting and I 9 believe we left off under Systems Review, Item 7. The next 10 subject was Main Feedwater Systems Review, and I believe Mr.
11 Skillman is going to speak on this.
12 MR. SKILLMAN: Good morning, I'm Dick Skillman.
f'\
\- 13 I was the Chairman of the Safety and Performance Improvement 14 Program activity for the B&W owners.
15 I'd like to start by saying who did this effort.
16 This effort was done by the Babcock and Wilcox plant owners, 17 and that includes people from Arkansas, Duke, Florida, GP 18 Nuclear Sacramento, SMUD, TBA, Toledo, and Babcock and W1.1.cox.
19 In addition, we hired contractors. We used SAIC for our risk 20 assessment review; we employed MPR Associates for the 21 sensitivity study; we had our independent advisory board use 22 some human factors experts in our operator burden activity.
23 My only point is, there were a lot of people involved: those 24 who own the plants, those who designed the plants, and
(~'
\
25 consultants that we thought were needed to do a thorough job.
Heritage Reporting Corporation (202) 628-4888
1 342
()
1 Yesterday we talked about information gathering, the 2 integration phase, and the implementation phase. Neil 3 Rutherford will talk about implementation later. In the 4 information gathering stage we did a lot of work in our tap 5 data that pointed to the need to review in detail certain 6 fluid hydraulic systems and control systems. Which systems 7 and why.
8 The data showed that the secondary plant relief, the 9 emergency feedwater, the main feedwater, and the ICS/NNI 10 systems deserved a great deal of attention, of review, and the 11 instrument air. So my goal this morning will be to briefly 12 touch on what we did in the systems review. I will not touch O 13 on ICS/NNI because Larry Stolter covered that yesterday with 14 comments from the staff about that review.
15 What did we do on main feedwater system review?
16 Please bear in mind that the main feedwater system review 4
17 activity was being conducted in late 1985 and in early 1986.
18 The basis of that review was the main feedwater events of the 19 years 1984 and 1985; clearly the Davis-Besse transient on 20 June 9, 1985; and the Rancho Seco event on December 26, 1985 21 played into the need for interrogation of the main feedwater 22 system.
23 What was done was information was gathered from each 24 plant site by approximately a one week visit at each of the
{} 25 six B&W owners group sites for understanding of the procedures i
Heritage Reporting Corporation (202) 628-4888
343 k,3/
1 and the characteristics of the different main feedwater 2 systems, the design of the different main feedwater systems, 3 and the maintenance practices on the main feedwater systems.
4 There were interviews with the operations and maintenance 5 personnel about the operation of the main feedwater systems, a 6 walkdown of each system, and a review of the data, 7 particularly of the problems that had been experienced on 8 those systems.
9 That information was compiled into approximately 40 10 recommendations that had both generic, that is all B&W owners' 11 applicability, and specific, i.e. for that particular plant, 12 applicability.
13 We were looking for root causes of feedwater 14 problems. We were looking for recommendations to lead to 15 improved performance of the main feedwater systems. In short, 16 we were trying to increase the reliability of the main 17 feedwater systems.
18 I would expect you would say, "So what did you find?
19 What did you come up with?" I've listed here just typical 1
1 20 key recommendations. I would assume that you can see from the 21 recommendations the kind of detail the reviews went into.
22 Implement a program to identify improvements in main 23 feedwater pump control systems including the ICS. Evaluate 24 the interaction between these two systems. Clearly there is a
{} 25 relationship between ICS and main feedwater during normal Heritage Reporting Corporation (202) 628-4888
344
[~'
s 1 operations. We needed to understand that because we had had 2 the experience where these two would somehow get out of 3 kilter.
4 Correct the main feedwater pump control problems.
5 What we found is in the governors, the over speeds, the 6 electrical power that supplies the main feedwater pump 7 controllers, the lube oil systems, and so on. We have had 8 problems and we were setting about to correct those problems.
9 Ensure that a single electrical failure in main 10 feedwater and condensate system will not cause loss of both 11 feedwater chains. This came from the recognition that both 12 feedwater pumps might have their control system powered by a 13 single electrical circuit. Hence, loss of that circuit for 14 any reason could take out not only main feed but condensate as 15 well.
16 Eliminate unneeded trip functions on the main feed 17 pumps. That sounds like hokey. That sounds like malarkey.
18 But what we found is there's a monstrous main feedwater pump, 19 an enormous turbine-driven packaged unit, and when it arrived 20 on site those who were responsible for hooking it up read the 21 instruction book and said well we need a low lube oil sump 22 trip, we need a low lube oil pressure trip, we need a low lube 23 oil flow rate trip, we need a high lube oil temperature trip, 24 we need a vibration trip, we need a proximity trip, and so on.
{} 25 What we found is the very heart of our heat removal system was Heritage Reporting Corporation (202) 628-4888
345 n/
(-
{
1 subject to what we would call nickel-dime trips that in fact 2 interrupted the entire thermohydraulic process of an 800 3 megawatt plant.
4 So the real question was what can we get rid of that 5 in no way jeopardizes the safety of that component so as to 6 increase the reliability of the component and of the plant.
7 The short answer is maybe you need an over speed and a failure 8 of lube oil altogether, but you certainly don't need all that 9 clap trap, and we set about to get rid of that stuff.
10 Eliminate automatic control of the main feedwater 11 block valve except during a reactor trip.
12 Provide the capability to override a closed signal 13 to the main feedwater block valve. We had in at least one 14 plant, once the main feedwater block valve was commanded 15 closed, the operator could not interrupt that and prevent the 16 termination of feedwater. We wanted to give him the ability 17 to get bar;k in control thermohydraulically.
18 Provide automatic main feedwater overfill production 19 capability. Install a monitoring system on the pumps to 20 identify the cause of the trip. We had so many of these minor 21 pump trippers we said what is causing the problem. The 22 operators would know that he lost a main feedwater pump, but 23 wonder why, and we wanted to understand how come.
24 My only point is that in reviewing all of the sites' 25 main feedwater systems, the B&W owners came up with a host,
)
Heritage Reporting Corporation (202) 628-4888
(-
\m/
346 1 and these are a fraction of the recommendations, but we came 2 up with recommendations that really get into the way in which 3 those components were installed in the plant. Some of these 4 findings are generic. They apply not only to the Babcock 5 owners, but to anybody that owns PWR.
6 And we found a lot of site-specific issues which 7 when corrected would increase the reliability of the main 8 feedwater system. If you recall the chart that I showed 9 yesterday regarding the dominant plant tripper today, the 10 dominant plant tripping system tode.y for the B&W owners is 11 main feedwater. Take a step back and look at the gray book 12 and say what's tripping PWR's in general, it's main feedwater.
13 I just submit to you that it's not that the pumps are bad, or 14 it's not that the system is bad. There are a lot of small 15 things down in the bowels of the system that are able to 16 interrupt main feedwater flow. We are going after these.
17 Many of these recommendations are key 18 recommendations regarding main feedwater system reliability.
19 Clearly we want to eliminate the overheating transient 20 causers, preferentially over the trip causers. But many of 21 these are plant trippers, even though they're rather minor in 22 the control of the main feedwater.
23 MR. KERR: It certainly seems to me that you have 24 selected a key symptom and have worked to improve the 25 reliability of the people in the system, I think. But one Heritage Reporting Corporation (202) 628-4888
l
\
./T 347 V
1 could get the impression from number four, for example, that 2 to make the system more reliable you should take out those 3 things where you indicated you were having trouble. It seems 4 to me an alternate way to make the system more reliable is to 5 make those things that are causing trouble more reliable. I 6 must be missing something.
7 MR. SKILLMAN: No sir, you're right. An 8 interpretation of that could be just to pull the plug on the 9 things that get you. We're not saying that. Let's make sure 10 we eliminate the things that we earnestly do not need.
11 For instance, we have found orifices where they 12 shouldn't be. Parts in the lube oil systems of main feedwater 13 pumps that are parts adrift, parts that came from some 14 previous outage and were left in the system.
15 The tone of the recommendations causes the utilities
- 16 to look in detail at the types of problems that have been 17 experience and to get rid of or to make changes so that past l
18 efficiencies are corrected. Part of that has to do with the 19 maintenance practices at the units. But at the same time, l 20 clearly, there seems to be a group of trippers at each plant l
21 that earnestly are not needed for pump protection, for 22 personnel safety, and so on. We're saying those things that 23 earnestly are not needed, need to be gotton rid of because 24 there are just too many series trippers that will eliminate
{} 25 main feedwater when they shouldn't eliminate main feedwater.
Heritage Reporting Corporation (202) 628-4888 l
th 348 V
1 For instance, loss of lube oil pressure is not in 2 and of itself an immediate cause to shut off the main 3 feedwater system. It may be that there should be a time relay 4 to let the trip occur only after a backup booster pump has had 5 time to come up to speed. But the way the circuits are 6 currently designed, once the pressure goes down, you've lost 7 main feedwater. We're saying that's not cricket. We should 8 not pull the plug on main feedwater unless you earnestly have 9 to take the system down.
10 MR. WYLIE: With those original trips in there, it 11 seems the designer certainly had in mind his first priority 12 was to protect the pump.
b s/ 13 MR. SKILLMAN: Protect the component, yes sir.
14 MR. WYLIE: That's really not your first priority.
15 You're changing the system to reflect a better balance of 16 priorities.
17 MR. SKILLMAN: What we're really saying is the 18 transients that really are of concern to us are the 19 overheating transients. In a hierarchy of what functions you 20 need, you need heat removal, and therefore main feedwater is 21 vital. Let's don't remove main feedwater unless we really, 22 really have to take it down. Let's make sure that what trips 23 main feed should trip main feed. The flip side is true.
24 Let's make sure that we don't lose main feed for fearless 25 reasons.
Heritage Reporting Corporation (202) 628-4888
349 (q_)
1 But that's the tone of the comments. If you were to 2 look through the executive summary of B&W 1919 and see where 3 the main feedwater items fit, they fit in a hierarchy of the 4 undercooling transient eliminators.
5 Let me move on.
6 The emergency feed. What did we do? We compiled 7 the functional design objectives of all of the six sites' 8 emergency feedwater systems. We compiled testing objectives, 9 and we compiled a list of maintenance recommendations to 10 improve reliability and availability. Woven through the 11 systems review including ICS/NNI yesterday, is a clear 12 recognition of the relationship between smooth, reliable fs k- 13 operation, maintenance, maintenance practices, and those trees 14 of things; hence, you will see that coming up persistently as 15 we talk about these BOP systems.
16 Our real goal here was to improve the responso of 17 the emergency feedwater system, the loss of main feedwater.
18 In the B&W type plants with 30,000 tubes, with emergency 19 feedwater injected high in the secondary side of the steam 20 generator, excessive flow in the emergency feedwater will 21 overcool the primary coolant system. So our goal was to find 22 out how to trim emergency feedwater flow rate, trim emergency 23 feedwater start-up times so there is a thermohydraulic match 24 early on after loss of main feed.
(')
V 25 The types of recommendations that came forward:
Heritage Reporting Corporation (202) 628-4888
i
(~~ 350 I
- \-)
1 Remove emergency feedwater initiation and control from 2 ICS/NNI. That has been done. Emergency feedwater is not 3 controlled by ICS/NNI. Emergency feedwater is controlled by a 4 safety system at each of the six sites. They come in 5 different names, but their goal is to control emergency feed 6 following loss of main feed as a safety grade function.
7 Limit the flow rate or fill rate. This is where we 8 spent some time yesterday. Suffice it to say what we were 9 trying to do as an objective of the emergency feedwater review 10 is to get the post-trip heat balance quickly in balance by 11 trimming the rate at which emergency feed is introduced to the 12 steam generators.
13 Extanded start time for emergency feedwater turbine-14 driven pumps. That might sound like a regression from a hard 15 requirement. What we find is most of these emergency 16 feedwater pump turbines are terry turbines. They are very
- 17 susceptible to the density of the operating media.
18 MR. MICHELSON: Which side? The steam side?
19 MR. SKILLMAN: Yes sir. Consequently, you might 20 have a long, long steam line feeding the terry turbine that is 21 now filled with condensate. When the command is given, the 22 valve opens, high density water, terry turbine winds up 23 instantaneously, and in many cases over-speeds. So there are 24 a couple of problems. Make sure that the steam line that 25 feeds the terry turbine is drained and that there is warm, dry Heritage Reporting Corporation (202) 628-4888
,r'\ 351
\/
1 steam ready to operate the turbine. But the second thing is, 2 don't wind it up so quickly. Give it a chance to come up to 3 speed gently so that it doesn't over-speed, which means extend 4 the start-up time. When you look at the point in time when 5 you need to match the energy balance post-trip, it is not 6 instantaneous. You have some seconds, 100 seconds, 120 7 seconds, 150 seconds. So where before we were trying to light 8 these turbines off in 30 and 40 and 50 seconds, now we're 9 saying let's give them two minutes.
10 MR. KERR: Is that an automatically controlled 11 sequence or a manually controlled sequence?
12 MR. SKILLMAN: That's an automatically controlled
\ ') 13 sequence. We're saying let's give those turbines time to do 14 what they do well, which is run at constant speed. But let's 15 run them up carefully so that we don't trip them on the start.
16 MR. KERR: Does that mean a change in the operators 17 on it?
18 MR. SKILLMAN: Yes sir, that type of thing. Perhaps 19 the stroke time o.' the valve, other things like that.
20 Again, encure maintenance test programs confirm 21 that --
22 MR. WARD: Does that mean that with that longer, I 23 guess it's the speed or supply valve in longer stroke, takes 24 longer, does that mean that the concommity of line full of
{} 25 condensate?
Heritage Reporting Corporation (202) 628-4888
(m 352
- \-
1 MR. SKILLMAN: No. What it really means is take a 2 look at your maintenance and test program, and take a look at 3 -your design. Make sure that when the terry turbine is called 4 upon to operate it has a high chance of success. That means 5 extending the start time, making sure the driving media is 6 what it ought to be, making sure the valves do what they are 7 supposed to do, and so on.
8 MR. MICHELSON: Those turbines were purported to be 9 able to digest large amounts of water. That's what GE used to 10 tell us all about how good they were. Of course experience 11 has shown that they don't do that so well unless they're 12 already up to speed and running smoothly, and then you can 13 shoot the water through it. But they don't start worth a darn 14 unless --
15 MR. SKILLMAN: Our fifth recommendation, and these 16 are typical. There are many more recommendations besides 17 these, but we wanted to make you aware of these.
18 Reduce spurious EFW actuations. That seems kind of 19 diminimus, but that's important. If you look at many of the 20 significant B transients. Bear in mind yesterday we talked 21 about 250 trips. We talked ten category C's which are the 22 most severe; we didn't say much about the 40 significant B's, 23 but we regard those as significant and worthy of a great deal 24 of attention.
25 If you look at the significant B's and find out when
[}
Heritage Reporting Corporation (202) 628-4888
353 U'~'s 1 they occurred and why, many of those are emergency feedwater 2 actuations at very low power. For instance, when the plant is 3 coming back from a refueling outage, the core has virtually no 4 decay heat; the metal tends to be colder than warmer; you're 5 operating the generator at about 15 percent power; the 6 generator levels, water levels, are very low; and any 7 perturbation in main feed that would take the generator water 8 level below approximately 30 inches will light off emergency 9 feed because that's a key signal for getting emergency 10 feedwater rolling.
11 So what we were having was a preponderance of low 12 power emergency feed starts. What did that do? Severe k 13 overcooling. No decay heat. So we said hey, let's see what 14 we can do to knock off those spurious EFW actuations. Clearly 15 those actuations were not needed for plant safety. The types 16 of things we talked about doing are raising the start point to 17 give a command to emergency feedwater.
18 Suffice it to say, we were seeing spurious 19 actuations and we were saying we've got to knock that off.
20 We're overcooling and we're tripping too often. Let's don't 21 do that.
22 MR. MICHELSON: By raising, you meant lowering, it's 23 at a lower level than the generator now?
24 MR. SKILLMAN: Yes sir. We were knocking from 30
/ 25 down to 24 or 30 down to 18.
Heritage Reporting Corporation (202) 628-4888
354 C')s s.
1 MR. MICHELSON: That means the generators dry out 2 even faster under some of these other conditions than they 3 would have with less water.
4 MR. SKILLMAN: The signal would be raised for power 5 operation. It was only for a start up mode.
6 MR. MICHELSON: You're having a two step signal.
7 MR. SKILLMAN: Yes sir. I'm looking for Angelo back 8 there to support me if I've --
9 MR. MERCADO: We're looking for margin difference.
10 We're raising the level of the --
11 MR. WYLIE: I don't think the reporter can hear you.
12 MR. MERCADO: What we're trying to do in this 13 particular case is to raise the margin between the low level 14 in the main feedwater low level signal, and the emergency 15 feedwater initiation signal so we won't have any spurious 16 signals.
17 MR. MICHELSON: At what level do you now initiate 18 emergency feedwater?
19 MR. MERCADO: I'm not sure. It varies from plant to 20 plant.
21 MR. MICHELSON: How does that affect the dry out 22 rate for some of these other events? You talk about dry out 23 in terms of four to twelve minutes depending on the 24 conditions. This means it dries out just a little fast
~) 25 doesn't it? Perhaps.
(J Heritage Reporting Corporation (202) 628-4888
355
('~s)
~
1 MR. HENSON: Where you initiate feedwater you dry 2 out its phase, and are not initiating feedwater.
3 MR. MICHELSON: But unless you put in a variable set 4 point on this thing.
5 MR. HENSON: I'm with Toledo Edison. Your dry out 6 is really dependent upon where you operate the main feedwater.
7 Some plants increased that, and that was a way of decreasing 8 the margin for main feedwater control on EFW actuation.
9 MR. MICHELSON: You raise the normal operating 10 level?
11 MR. HENSON: Yes.
12 MR. MICHELSON: Thank you.
13 MR. DAVIS: I have a question. What did you do to 14 initiate number two?
15 MR. SKILLMAN: On some plants there are cavitating 16 venturies that will choke and therefore slow at a fixed 17 amount. On others is the adjustment to the emergency 18 feedwater injection valve decision so that' when the valve 19 strokes open only so much water can deliver. That can later 20 be overridden. What we're really trying to do is to drop back 21 from the 600 or 700 gallon per minute delivery rate to 22 something less than that that more accurate]y reflects the 23 removal of decay heat at the point in time that this system, 24 the secondary plant releases, can be in balance. So it's 25 either cavitating venturies or throttle valve position.
[}
Heritage Reporting Corporation (202) 628-4888
T 356
)
1 MR. DAVIS: It appears that most of these l 4
2 recommendations were instituted to avoid overcooling 3 transients. Have you concluded that these recommendations 4 would likely improve the reliability of the system when it's 5 needed to avoid overheating problems? Which is a more serious l 6 transient, of course.
7 MR. RUTHERFORD: I think certainly in some cases 8 that's true, that they do help the overheating aspect also.
9 When you look at cavitating venturies, at least in the case of 10 Oconee, it precludes certain runout conditions we might get 11 into in certain events.
12 MR. SKILLMAN: My answer would be very similar to 13 Neil's. I would say those things that make the system more 14 reliable, particularly with regard to the overcooling I
15 transients, are equally as beneficial in terms of system 16 reliability to ensure it's available for undercooling 17 transients. I see a double benefit here. Increasing the 18 reliability in any case, whether the transient is an 19 overcooling transient or an undercooling transient, will 20 render the system available for whatever. The problem we were 21 getting into here was too many failures of the emergency 22 feedwater system; too many complex transient involvements.
23 MR. KERR: The key to that, it seems to me if I 24 understood your earlier comment, is that you're trying to 25 match the water supply to the decay heat.
Heritage Reporting Corporation l
l (202) 628-4888 l
n 357 k_ S 1 MR. SKILLMAN: Yes sir.
2 MR. KERR: It seems to me you ought to do this, 3 whether one is talking about an overcooling or undercooling 4 transient if you can indeed do that.
5 MR. WARD: I think Pete's point is, and it seems 6 apparent to me that at least for some of the things that you 7 can control, you're getting contrary demands put on you. You 8 have to figure out how to balance it.
9 MR. KERR: But Dave, you don't ever want more water 10 than you need to remove the decay heat.
11 MR. WARD: But If I wanted to be conservative I 12 might say yes, let's provide more water.
O\l 13 MR. SKILLMAN: That's the problem. That's exactly 14 the problem.
15 MR. WARD: I know. But you've got a narrow line 16 that you have to walk.
17 MR. DAVIS: That's the side you want to err on.
18 MR. WARD: I don't know which side you want to err 19 on. That's the question you were asking. It's not always 20 clear.
l 21 MR. SKILLMAN: How wide is the walkway.
22 MR. DAVIS: Right. And have you really 23 appropriately balanced it.
24 MR. SKILLMAN: It's the wide walkway that's giving 25 us the trouble, because what we've done in so many cases is
{;
lieritage Reporting Corporation (202) 628-4888
s 358
(~'/
\_
i say we want copious amounts of water, we want lots of margin.
2 In getting copious amounts of water, particularly on emergency 3 feed, we are extremely susceptible to overcooling transients 4 which have been the dominant types of complex transients that 5 we've had, and as we said in our slide regarding conclusions, 6 regarding complex transients, while they give us ccmfort that 7 the core is cool, they lead to a whole host of further 8 operator interactions that causa us to say we want t stop 9 that because those can lead to operator error the .n give us 10 later problems.
11 So the real issue here is let's get this system 12 under control. Let's remove the right amount of decay heat at 13 the right point in time. Iet's make the pumps opcrate l
14 properly. Let's make the control systems work properly.
15 Let's get it disconnected from the ICS/NNI. Let's get a 16 dedicated safety system. Let's make this system do what it's 17 supposed to do.
i 18 MR. WARD: You want to have fewer transients, bat
, 19 since an overheating transient really places the plant more at 20 risk directly than an overcooling transient, maybe you ought 21 to put the narrow walkway in the place where you get five 22 times as many overcooling transients as you do overheating i
23 transients which is about where you are now.
24 MR. SKILLMAN: We're about 10 to i right now.
25 MR. WARD: That's sort of the question. We know you l
! Heritage Reporting Corporation l (202) 628-4888 l
r~T 359 V
1 want to reduce them all, but what's the right ratio?
2 MR. SKILLMAN: In a large number of the cases the 3 overcoolings have led to an undercool. What does the operator 4 do when he sees the plant overcool? The first thing he 5 attempts to do is to crank her back, not overcool. The next 6 thing you know, he's undercool. So what we're saying is 7 instead of driving the people through those hoops, let's make 8 this one perform properly and let's make its analog, which is 9 the secondary plant release system, also operate properly, so 10 the operator is not, in a dominant number of cases, fighting 11 his way from an overcooling to an undercooling to an 12 overcooling to an undercooling.
13 MR. KERR: In any event, it seems to me if you can 14 remove the decay heat you're not going to have an overcooling 15 trend.
16 MR. REED: This is a little point, but on the 17 turbine, I assume that such little things as steam being 36 brought r.ight through on the DC stop valve ahead of the 19 turbine and trapped out so it's always a hot line, that's 20 done. And in fact the valves that are going to initiate and 21 support the turbine start-up on an all DC operated valve, off 22 the -- box. Is that the case?
23 MR. MICHELSON: Glenn, it's a little hard to hear 24 some of what you say, 25 MR. REED: I was asking about two little things. If Heritage Reporting Corporation (202) 628-4888
360 U'~N 1 in fact the steam line coming to the emergency feedwater 2 turbine was hot and trapped right to the DC stop valve for 3 initiation, and if in fact the other valves associated with 4 water supply in the turbine are all DC operated from the vital 5 box.
6 MR. SKILLMAN: The scope of the recommendations from 7 SPIP clearly address ensuring that the driving media is 8 available and that the valve that controls the driving media 9 is also available. Where it's powered from, I can't answer.
10 I just don't know. I'll get you an answer later.
11 MR. REED: It's a funny thing, we've run into the 12 fact that people have designed plants and provided decay heat 13 removal systems and then on a loss of AC they have AC valves 14 in and areund the steam drivers.
15 MR. LeFAVE: Bill LeFave from the staf f. At least 16 one of the pumps is completely battery packed, DC power, in 17 accordance with 2 Ell. That was a requirement for all the 18 plants, to have at least one of their turbine drive pumps 19 completely independent of AC power and that includes 20 ventilation, lube oil, that kind of supporting systems.
21 MR. MICHELSON: And steam supply?
22 MR. LeFAVE: And steam supply, yes. They can be 23 operated independent of the AC power.
24 MR. REED: I thought that would probably be the 25 case.
Heritage Reporting Corporation (202) 628-4888
(~T- 361 1 MR. MICHELSON: When you make these proposed changes 2 to improve essentially the reliability of the sVstem, do you 3 go back and reexamine your safety analysis each time to see if 4 that changed, what effect that has on the overall safety which 5 is more in my interest, at least?
6 MR. SKILLMAN: I'd like to defer that question to 7 Neil Rutherford.
8 MR. MICHELSON: Okay.
9 MR. SKILLMAN: He'll talk about recommendations, 10 implementation, follow up, quality, those types of things.
11 MR. RUTHERFORD: Let me go ahead and respond to 12 that. That's something each utility would have to do when
. ( 13 they implement a modification or change. They would have to 14 do the safety analysis. It's going to vary utility to 15 utility, depending on exactly what the case is.
16 MR. MICHELSON: But that would be a requirement, 17 when the SPIP is made, and you said in some cases these have 18 already been done, and in other cases they'll be done. As 19 they're done, the submittal will go to NRC, or does it go as a i
20 50-59?
! 21 MR. RUTHERFORD: In most cases those would be done l-l 22 under 50-59.
23 MR. MICHELSON: Then the NRC is supposed to be 24 reviewing these by whatever means it reviews such change as 25 that. What would the NRC, at least on an audit basis, look at f-)
l (_/
1 Heritage Reporting Corporation (202) 628-4888
A 362 V
1 these 50-59s? Normally it's just the resident inspector 2 that's required to look at them once a year, but I'm not sure 3 he's the man that should be looking at them.
4 VOICE: I think it is -- that normally screens all 5 the 50-59's. If he feels something there is more significant 6 or complex than he can do it, he'll defer that to the staff.
7 MR. MICHELSON: Clearly this particular class we're 8 looking at here at the moment is a class thac you would 9 certainly want to audit on a little more frequent basis than 10 some of the other things.
11 VOICE: I agree.
12 MR. MICHELSON: Thank you.
v 13 MR. SKILLMAN: I'd like to talk briefly about the 14 secondary plant pressure control system. I sense it's
, 15 worthwhile to give you the opportunity to see again the almost 16 equivalent participation in post trip misbehavior of the 17 3rgency feedwater and secondary plant pressure control 18 syitems. For each misbehavior of this system we seem to get 19 about one of this system leading us to believe, and this was a 20 clear finding of SPIP that was something we had never really 21 understood before or reckoned with before, perhaps a dominant 22 relationship between these two systems, post trip.
23 Again, this is the participation. If you will, the 1 24 number of misbehaviors out of 50 complex transients of these 25 systems. In about 30 percent of the cases you get emergency Heritage Reporting Corporation (202) 628-4888 l
()
1 feed, secondary plant relief, somehow operating in concert 2 with each other.
3 What do we do on secondary plant pressure control?
4 Secondary plant' pressure control is the main steam safety 5 valves, turbine bypass valves, and atmospheric compounds.
6 Those systems that have the prompt energy relief 7 responsibility following a trip.
8 We were trying to identify the problems regarding 9 main steam safety valve performance at the turbine bypass end 10 or atmospheric dump valve performance. We'd had problems out 11 in those areas. What we did is we did a preliminary 12 assessment for the methods of reducing the frequency of main 13 steam safety valve lift, and we're trying to increase the 14 reliability of both the main steam safety valves and the 15 turbine bypass valve, atmospheric dump valve components.
16 MR. KERR: Mr. Skillman?
17 MR. SKILLMAN: Yes sir.
18 MR. KERR: I understood you to say that it was a 19 surprise to you to discover that these two things were making 20 about equal contributions.
21 MR. SKILLMAN: Yes sir.
22 MR. KERR: I would hope that not only in this area 23 but in the area of plant operation generally, someone in some 24 group is looking for these kinds of things so that you know 25 where the problems arise. I know you have so many things you
[}
Heritage Reporting Corporation (202) 628-4888
l'h 364
(_/
1 have to be looking at that perhaps you were not driven to this 2 particular study earlier, but it would seem to me that there f 3 ought to be some watchdog group that looks for this kind of 4 thing on a continuing basis.
5 MR. SKILLMAN: There is. Neil, would you like to 6 speak to that?
7 MR. RUTHERFORD: I think the problem arises that we 8 do that on an individual plant basis. We all have operating 9 experience assessment programs where we go in and look at our 10 own experience. But sometimes you get a little bit different 11 viewpoint when you look at collective experience in a group of 12 plants. I think this is where we picked up some of the things 13 that Dick was alluding to.
14 MR. SKILLMAN: It might be worthwhile for me to tell 15 you how we came up with this. We asked ourselves three 16 questions. If we were in the control room at the time the
- 17 complex transient was occurring, if we were omniscient and if 18 we could understand every proper behavior or misbehavior, and 19 if we were omnipotent, that is we could fix each flaw as it 20 occurred, what would the pattern be like in each of the 50
$ 21 complex transients? Ac we worked our way through there we 22 realized that in many cases there wasn't a single misbehaving 23 system. There were, in fact, clusters or multiples of 1
24 misbehaving systems, and in some cases the emergency feed 25 might do a peculiar thing, and then secondary plant relief l
Heritage Reporting Corporation (202) 628-4888 l
365 J
1 might do a peculiar thing, and then ICS might do a peculiar 2 thing causing still another iteration of emergency feed or 3 secondary plant relief.
4 So instead of trying to globally choose one bad 5 actor we said hey, how many bad actors were there in how many 6 of these complex transients, and what were they? What were 7 the specific things we would have to fix? Hardware fix. That 8- means a piece of paper at the site, a drawing change, 9 something that would be the basis of an engineering change.
10 With that question being asked, having gone through 11 those complex transients, we came up with a completely 12 differer.t picture of what the misbehavior pattern was for all
( 13 of the Babcock plante. From that, came this participation 14 picture.
15 MR. KERR: That's impressive, and I applaud you. My 16 point is, I hope it won't be restricted to this study, because i 17 it sounds to me as if it could be a very valuable, generally 18 useful approach.
19 MR. RUTHERFORD: We do have an ongoing program 20 through our transient assessment committee that has looked and
(
21 will be looking at this type of data now and in the future, i
22 MR. SKILLMAN: What did we find in secondary plant 23 pressure control? The control of post trip feed and bleed, t
24 post trip feed and steam flow is going to contribute to l 25 complex transients. In a nutshell, that is the post trip t
Heritage Reporting Corporation (202) 628-4888 i
f
()
1 energy balance. Providing the main control room manual 2 control and isolation capability of all post trip steam and 3 feed flow paths, excluding the safety relief valves. Go 4 through the complex transients and say where did the plant get 5 away from the operators. It's normally in an area where the 6 operator is not able to take control promptly. The reason he 7 didn't have manual control over it. He had a relief valve 8 opening or he had something going on and he really couldn't 9 get control.
10 Using the term bypass valve and atmospheric dump 11 valve systems prevent excessive steam flow and loss of ICS/NNI 12 power, previously some of these components were on ICS/NNI.
\ 13 MR. MICHELSON: Let me ask you, on item one, did you 14 do some kind of a PRA or something that led you to the 15 conclusion it was better for the operator to be able to 16 intervene to keep the transient more under control than it was 17 for him not to intervene and allow the equipment to respond to 18 what might be really an accident and not even a transient 19 alone? In other words, it's kind of a balancing you've got to 20 do when you talk about manual intervention. It means you can 21 also improperly intervene in a bad situation.
22 MR. SKILLMAN: No sir, we did not do a PRA. What we 23 did do is we went back to the Davis-Besse event and the Rancho 24 Seco event and said where did these people really have a 25 complex time in the middle of those transients. What we found Heritage Reporting Corporation (202) 628-4888
367 O(-
1 was that in those transients there was inordinant operator 2 participating outside the control room to catch up with the 3 transient.
4 MR. MICHELSON: This was moving to a centralized 5 control point more than anything?
6 MR. SKILLMAN: Give the operator what he needs at 7 his place of business.
8 MR. MICHELSON: Of course that means he can use it 9 incorrectly, but that's what you have to think about when you 10 decide this way.
11 MR. SKILLMAN: Develop and implement standardized 12 procedures and techniques for maintaining setting, testing, t
13 main stream safety valves, bypass valves, and atmospheric dump 14 valves, and so on. The typical key recommendations are 15 recommendations that are intended to get the secondary plant 16 pressure control systems promptly settled, and also to give l 17 the operator the ability to control those from where he 18 normally is.
19 Let me make a comment about instrument air. Staff 20 has talked about the program, setting up the program by about 21 March of 1986. Instrument air was not part of our original 22 program. Together we said this one needs to be in there. We i 23 added in a major effort on instrument air that was extremely 24 thorough. The point is, we did not stop on the least common 25 denominator. This program went ahead. We and the staff said
}
' Heritage Reporting Corporation (202) 628-4888 I
L
1 368 b%
1 this is missing, this missing, let's add this in, let's add 2 this in. Consequently, instrument air got added in and the 3 risk review got added in.
4 What do we do? Compile the system features 5 operating data again, from all six of the plant sites. These 6 are plant site visits, people a week or so, trying to collect 7 data and understand what each plant site had. Identified the 8 critical or actuated components. The marching orders we gave 9 to the instrument air review people were marching orders that 10 said the thing you're really looking for is this system's 11 participation in decay heat removal: how it works, what it 12 does, what components are needed to get heat balance under
() 13 control.
14 We looked at the planned response to air loss, 15 developed target criteria, and we developed recommendations.
16 MR. MICHELSON: You didn't look, though, at degraded 17 conditions such as air pressure or dirt in the system 18 throughout, or that sort of thing?
i 19 MR. SKILLMAN: The utilities are doing that.
i 20 MR. MICHELSON: Under what program? Under this l
! 21 program?
l 22 MR. SKILLMAN: Yes sir.
- 23 MR. MICHELSON
- So they are looking at degraded air 1
24 effects?
25 MR. SKILLMAN: Let's go through the recommendations.
Heritage Reporting Corporation (202) 628-4888
).69
' x_-
(~}
1 MR. MICHELSON: All right.
2 MR. SKILLMAN: The heartbeat of the instrument air 3 system review is virtually identical to the ICS/NNI, go to 4 known safe state on loss c1 air. But as you know, this one 5 has a peculiar characteristic in that you can lose subsystems, 6 part systems, you can terminate an inner artery, you can have 7 one side of the plant behaving one way and one behaving 8 another, 9 Compare the plant system and air system with the 10 functional target criteria to determine what upgrades are 11 necessary. To do this, we set out about 24 key items
. 12 regarding instrument air requirements. Of the most important 13 are the ones pertaining to clean drive, oil free, particulate 14 free air. In a dominant number of cases that has been the 15 problem with instrument air systems. That is, the air is 16 dirty.
17 Perform an evaluation to ensure that air system
, 18 failure will not affect the ability to maintain the plant in a 19 known stage state. That recommendation is over into each 20 utility for evaluation and for implementation.
21 MR. MICHELSON: What did you mean by air system 22 failure? Loss of air or degraded conditions? It still isn't 23 clear what you're covering. Yeah, you want to keep the air 24 clean, but what happens if it gets dirty anyway, or water gets 25 into the air system because a cooler fails, something of that Heritage Reporting Corporation (202) 628-4888
b -
370 0 1 sort?
2 MR. RUTHERFORD: The testing that we were talking 3 about there is under total loss of air conditions. We are <
4 still examining the question about effects of partial 5 degradation, -- air supply. We haven't arrived at a final 6 conclusion.
7 MR. MICHELSON: You are thinking about it seriously?
8 MR. RUTHERFORD:- Yes.
9 MR. MICHELSON: Thank you.
10 MR. SKILLMAN: And perform operator testing in 11 critical air operated valves, compare with design basis, 12 giving time to rebuild as necessary. I think it might be
() 13 helpful for me to tell a sea story here.
14 In one of the plants this air control turbine, 15 bypass valve, and atmospheric dump valve, those are key 16 components in post trip energy control. The stroke time on 17 those air valves is three seconds. In the one plant the air 18 valve is prone to open in 45 seconds. We said how come? The 19 answer was, well it probably hasn't done enough preventive l 20 maintenance. So the valves, once open, showed rusted barrels l
21 and filled with sand. Until that point in time those valves 22 were controlled by the integrated control system. The !
23 integrated control system had been given the bum rap of l
j l
24 failing to control properly. When we checked with the l
25 integrated control system, the integrated control system gave Heritage Reporting Corporation (202) 628-4088 l
371 1 a prompt, accurate signal for those valves to open. But the ,
2 valve says no way, I'm stuck.
3 Again, we cleaned, the valves, put the valves back 4 together, and we found that the sand in the valve was really 5 silica jell that had broken through the screen in the dryer.
6 the heaters weren't working. There had been no real 7 preventive maintenance. Once the valves were rebuilt and the 8 signal applied to the valve, the valve stroked open in three 9 seconds like it was supposed to.
10 It sounds like a "so what," but if you look at the 11 post trip energy balance, those valves have the duty of taking 12 the pressure lift off the secondary system which translated 13 over to the primary system gives the high pressure trip point.
14 So was it a maintenance problem? Yes. Was it a design 15 problem? It was an attention to duty problem somewhere along 16 the line. Suffice it to say, let's make sure the valves do 17 what they're supposed to do when they're supposed to do it.
18 If that involves preventive maintenance, design change, 19 putting reservoirs on the valves to ensure that they stay 20 where they're supposed to, all of those are part of the 21 instrument air recommendations.
22 MR. DAVIS: I have a question related to that. Your 23 previous slide also indicated that improving the reliability 24 cf the turbine bypass valve and the atmospheric dump valve was 25 an important consideration. But on page 1-3-21 of BAW 19-19, O
Heritage Reporting Corporation (202) 628-4888
372
( ))
1 item six, and maybe you can clear this up for me. Item six 2 says, "Ensure that relief valves other than main steam stop 3 valves, turbine bypass valves, and atmospheric dump valves, 4 are in a preventive maintenance and test program." Why are 5 you excluding these valves that you just said are extremely 6 important and need preventive maintenance?
7 MR. SKILLMAN: We are assuming that these valves are 8 in fact a major part of the preventive maintenance program.
9 But particularly in the area of moisture separator reheater 10 relief valves, which are about a 14 inch diameter open plug 11 if they fail open, some of those valves because it is so far 12 in the bowels of the balance of the plant, were not
\ 13 preventively maintained. What we were trying to get to is 14 those other valves which normally are not considered so 15 important, in fact get swept into the preventive maintenance 16 program.
17 There are other recommendations in the executive 18 surmary a pertaining to these valves that basically say make 19 sure these valves can and will do what they are supposed to do 20 when they are supposed to do it. These valves are already in 21 a preventive maintenance program. So what is on that paper is 22 not intended to exclude these. In fact, it was intended to 23 incorporate others that might fall through the crack.
24 MR. DAVIS: Just reading it alone, it gives the 25 impression that these are excluded.
(~)T
\~
Heritage Reporting Corporation (202) 628-4888 l
373 1 MR. MICHELSON: On your example, you might want to 2 give some thought, although I realize it's not the scope of 3 this study, but you might give some thought to the seismic 4 qualification of the desiccant in the control air system, 5 you'll find those are extremely fragile in design. In an 6 earthquake you're going to end up with desiccant in all your 7 air systems, the possibility. You might want to think about 8 that a little bit, or maybe not use that particular example, 9 MR. SKILLMAN: Thank you. Any other questions?
10 Please let me end on this note. The BOP systems 11 that were reviewed, and I need to caution because while ICS 12 like BOP is clearly a jugular vein to the success of the 13 plant, those reviews were performed by and the recommendations 14 were built by people who designed, build, operate, and 15 maintain, whose day to day responsibilities are the care and 16 feeding of these systems.
17 The thought I'd like to leave you with is this was 18 done with somewhat of a loving hand. The people who did this 19 really want to be successful and want these plants to operate 20 smoothly. But more importantly, they want the plants to 21 operate safely. I think we have within the B&W owners a 22 renewed perception of what it takes in terms of p' nt safety.
23 What it takes is a tremendous attention to duty on the plant.
24 You might say why didn't you review the safety 25 systems. The reason we did not is because those systems have Heritage Reporting Corporation (202) 628-4888 i
374 O
1 proven themselves capable. They operate when they are 2 supposed to. They are preventively maintained. They are tech 3 spec time clocks. They have all the best of everything. What 4 we found is the systems which haven't been given that 5 attention are systems which are causing trips, and which are 6 integrally related with the complex transients that we've been 7 having. Consequently, when we said how do we settle on the 8 safety issue, we said let's go after those things that are 9 putting the plant in what we consider to be a risk situation.
10 Those are the systems which cause tripping, and if you read 11 the data, one in every five trips results in a complex 12 transient. We said what are the complex transient 13 misbehaviors? These systems are those systems.
14 Thank you.
15 MR. KERR: I think what you have just said is 16 extremely important. If we were starting the process over
- 17 with the accumulated wisdom that has occurred over the years, 18 do you think we should make the sharp demarcation that has 19 been made between safety and so-called balance of plant 20 systems?
3 21 MR. SKILLMAN: Are you asking me that question?
22 MR. KERR: Yes.
23 MR. SKILLMAN: I do not think that such a sharp 74 distinction should have been made in the first place. If you 25 were to ask me how would you do it, I'd say your primary heat Heritage Reporting Corporation (202) 628-4888
fs 375 1 balance systems, starting from the core and extending out to 2 those backups necessary to control, those trip energy heat 3 balance, must be treated about equivalently. That would save 4 reactor coolant pumps, motors, pressurizer heaters, next stage 5 seal injection, co ponent cooling water, next stage main -
6 injection systems, condensate, main feed, circulators, lube 7 oil, in service cooling water, component cooling water, and 8 all those that lie in between are vital to just the smooth 9 heater group.
10 MR. KERR: I would hope you can preach this gospel 11 to your colleagues so that the NRC doesn't have to get ;
12 involved in this.
() 13 MR. SKILLMAN: Thank you, sir.
14 MR. LeFAVE: I'm Bill LeFave. I'm going to talk 15 about basically the same staff review aspects of what Mr.
1C Skillman just talked about, j 17 When we point out that the mechanical aspects of the 18 syttems what we really mean is that it's separate from the 19 ICS/NNI review and the in-depth electrical review of the 20 PFW/AFW auto initiation and control.
i 21 MR. MICHELSON: We didn't hear about those from 22 others, necessarily. The feedwater instrumentation and 23 control we haven't heard about.
l
- 24 MR. LeFAVE
- I can discuss some general aspects of
, 25 that.
Heritage Reporting Corporation (202) 628-4888 t
376
(-)]
1 MR. MICHELSON: I didn't know. I'm just pointing 2 out that wasn't a part of the agenda. But there were B&W 3 owner groups talking about the ICS/NNI.
4 MR. LeFAVE: It was not directly a part of this 5 program, no. But it was somewhat embodied in the AFW stuff.
6 But specifically looking at the adequacy or acceptability of 7 those systems was directed to the staff by the commission.
8 MR. MICHELSON: That was a separate issue then.
9 MR. LeFAVE: Yes.
10 MR. MICHELSON: I just wanted to make sure. Thank 11 you.
12 MR. LeFAVE: I just want to point out some 13 observations that we have made. We agree with the BWOG that 14 implementation of recommendations will definitely reduce the 15 frequency of these trips. The actual reductions of trips at a 16 particular plan can vary depending on the implementation of 17 these recommendations, many of them involve evaluations and 18 analysis, depending on what depth the individual utility is 19 willing to go, fully dependent on how much reduction did they 20 get.
21 The recommendations do address design, maintenance, 22 and testing aspects of the system, and they're aimed at 23 improving raliability. I think that's basically based on root 24 cause determinations from all the studies they performed 25 including the operator or interviews and what have you.
Heritage Reporting Corporation (202) 628-4888
i
()
1 At the time of our review, there were .
2 recommendations that were made from the trip initiating events 3 review, and at that time they were not yet entered into the 4 recommendation tracking system. We recommended that we t
5 basically agree with the recommendations that are in there, 6 that they should indeed enter them into the tracking system.
7 The recommendations made regarding the feedwater 8 system are responses to the MPR sensitivity study that was l t 9 done-for this trip program which is basically the reliability 10 of the main feedwater system and the ability to maintain the 11 two main feedwater pumps running to not have a single failure 12 take you down.
~
13 I think a difficult area for implementation that's 14 going to take some coordination at the utility level is that 15 comments in the operator burden report which is Appendix S to j 16 19-19, should be considered during the implementation phase.
17 How this is going to be done, what kind of oversight is done, 18 is probably one of the most difficult tasks of doing the ,
19 actual implementation of these recommendations. Hopefully i 20 there will be some management oversight in that area. (
21 In addition to the recommendations that were on the 22 key list by B&W, we suggested that they also suggested they .
I 23 add, they already have this recommendation on one non-key ,
I 24 list, but enhancement of reliability of main feedwater ,
25 condensate systems 'o.ntrole uhoulc be nddr- They have i He ing Corporation !
1-4888 ;
F I
L
378 O
1 recommendations already on the key list that are very much
- 2 related to this, and this is a pretty general recommendation,.
3 but we thought it ought to be on the key list since it is the 4 basic cause of most of their trips.
i I 5 Going to the agenda, I'm supposed to give you some
~
6 insights on --
7 MR. KERR: Excuse me. Do you have any reliability 8 goals for the main feedwater condensate systems? You say the 9 reliability should be enhanced. To what point should it be
- 10 enhanced?
l 11 MR. LeFAVE: I don't believe there's any, we didn't t
l 12 do any reliability studies ourselves, and I did not look at f () 13 reliability. I'm just talking about availability in general, 14 There are no numbers involved. We don't have any goals that I
) 15 know of.
16 MR. KERR: So you're just saying let's make them t
i 17 better.
! 18 MR. LeFAVE: Yes. I think that is basically what f 19 B&W should also. We feel they didn't have any reliability
! 20 goals.
l 21 MR. KERR: I'm not being critical. I'm just asking 22 a question.
! 23 MR. LeFAVE: I understand.
i l 24 MR. RUTHERFORD: Other than its influence on reactor 25 trip transients, we didn't set any 'epecific goals for main Heritage Reporting Corporation (202) 628-4888
t (Z) 1 feedwater.
2 MR. LeFAVE: I don't really have any insights into 3 the importance of the balance of plant. I think everybody 4 here is aware that balance of plant is an important 5 contributor to risk since about 70 percent of reactor trips 6 throughout the country are caused by the balance of plant, or 7 somewhere in that area.
8 The major causes for the balance of plant related 9 trips are with the main feedwater and turbine control systems.
10 At B&W since the study began, the leading cause used to be the 11 turbine control problem, and they've already initiated some 12 changes or modifications to tne turbine control such that in 13 the last, this is through '86, that that part of the review 14 was itwolved, the studies of the -- that now the turbine has 15 been reduced to the extent that the main feedwater system now 16 is the major contributor to the balance of plant-related trips 17 at the B&W plant.
18 They have not yet, the recommendations for the main 19 feedwater improvements and reliability I don't think, maybe 20 for the last couple of years they have shown some improvement, 21 but up until 1986 they still have not, even though many 22 modifications had been made. You might want to elaborate if 23 there's been any changes in the last couple of years on any 24 reductions in main feedwater initiated trips.
25 MR. KERR: Mr. LeFave, I get the impression that
{
Heritage Reporting Corporation (202) 628-4888
I 1
C) 1 although you're convinced that the balance of plant has some 2 influence on risk, you aren't quite certain what it is, and
? indeed, perhaps in a subsequent slide there is some indication 4 that you're asking SAIC to perform a study of programs.
5 Again, implications could be that you're asking them to find 6 out how much influence balance of plant has on risk.
7 MR. LeFAVE: Yes, that's the basic thrust. What is 8 the safety significance.
9 MR. KERR: It is my impression that the NRC, another 10 branch at least, has recently spent a lot of time and effort 11 analyzing about five plants to do what is called a rebaseline 12 study of risk. Surely somewhere in that study there must be
() 13 some indication that the balance of plant either does or does 14 not have an influence on risk. If it isn't in there, then one 15 either has to conclude that it doesn't have any influence, or 16 else that the study is not very thorough. I would commend 17 that as an important resource for the staff to look at and 18 draft a new reg 11-50.
19 MR. LeFAVE: Those studies including that will be 20 part of the SIC review. All the programs are NRC programs 21 related to the balance of plant or the baseline study for risk 22 estimates, we'll use all those resources when we do the 23 balance of plant review.
24 MR. KERR: But you can't tell whether the balance of 25 plant has any influence until SAIC looks at it?
O Heritage Reporting Corporation (202) 628-4888
n 381 1 MR. LeFAVE: No, it's definitely a contributor. I 2 think I've already said that. It's got to be considered as I 3 would say a major contributor because it's a major precursor.
4 MR. KERR: I don't think that I understand page 97 5 of the safety assessment for the staff then where it clearly 6 says the B&W group concluded that Category C events are not 7 likely to be significant contributors to core damage 8 frequency.
9 MR. LeFAVE: I say precursors, they are potential l
10 precursors.
11 MR. KERR: Sure, but not significant contributors to 12 risk. Unless this report --
13 MR. LeFAVE: I don't want to say they are not 14 significant, but they are a significant precursor.
15 MR. SIEGEL: If I could comment on that, that 16 statement on section nine in the risk assessment is addressing 17 the historical Category C events that have occurred. You can 18 do a prccursor type study on those. You will find that they 19 are not significant contributors to risk at most plants. I 20 think we've got it broken down into not significant, moderate, 21 etcetera, depending on the specific plant design. But we'll 22 discuss that in more detail when we get up on the risk 23 assessment stuff. But it's put in the perspective of the 24 existing Category C events that have occurred, and primarily 25 the basis for that statement is the fact that they majority of Heritage Reporting Corporation (202) 628-4888
)
1 them were overcooling transients.
2 MR. LeFAVE: And obviously we don't have any plant 3 trips, the risk is reduced. Every time you have a trip,
~
4 that's somewhat of a precursor.
5 Obviously aggressive BOP programs including 6 maintenance, testing, design modifications, programs at 7 particular plants think that plant will have higher 8 availability. These are kind of general things. A lot of this 9 is not specifically related to the B&W, but it's kind of 10 related because what was looked at basically on the B&W was a 11 balance of plant.
12 What we have found is that the economic incentives O 13 for reliable balance of plant operation have a definite 14 positive effect on safety.
15 MR. KERR: What does that bullet mean? Does that 16 mean if you find them --
i 17 MR. LeFAVE: This means what we have found out over 18 the past couple of years, that the utilities that, there have 19 been some utilities paying a lot of attention to the balance 2 20 of plant on their own. They basically from the standpoint of 21 plant availability. Those plants have significantly reduced 22 the number of trips they've had over the past couple of years, 23 and we believe that's a definite improvement in safety.
24 That's really all that means. Even economic incentives can go
(} 25 on hand in hand with safety.
Heritage Reporting Corporation (202) 628-4888
(
1 This basically brings us up to where we are with 2 respect to balance of plant issues. Back in 1986 the policy 3 issue gives a good description of the balance, the NRC or 4 industry balance of plant programs that were in effect at the 5 time for a plant, and gave a pretty good discussion of the 6 statement significance of BOP items. The safety significance 7 basically boiled down to two things, how do balance the plant 8 failures effect the frequency of challenges to safety-related 9 systems, and its direct effect on safety-related systems; and 10 the effects on operator control of the reactor following trips 11 and the ability of the operator to mitigate or control the 12 challenges for the safety of related systems.
13 In the past we basically focused our review on the 14 impact of safety-related systems. This has been the 15 historical way the staff reviewed these things. For instance, 16 floods caused by pipe breaks, environmental effects, and 17 missiles, impingement. Today's focus is basically switching 18 to, we're trying to reduce the frequency of balance of plant 19 challenges rather than mitigating the effects themselves. To 20 try and reduce the frequency of balance of plant challenges 21 due to balance of plant system failures.
22 In this vein, temporary instruction for the balance 23 of plant trial inspection was performed at five plants. Right 24 now those inspections are complete. I was involved in two of 25 them. I'm familiar with the others. Right now that temporary
[
Heritage Reporting Corporation (202) 628-4888
()
1 instruction is being extended while an inspection procedure is 2 being written and being published for comment. So during that 3 time period, we'll have the temporary instruction in place. ,
4 The results of the temporary instruction inspections 5 show that the utilities are concerned with the balance of 6 plant systems, and a lot of them do have aggressive programs.
7 WL can see a definite difference between plants that have just 8 recently got involved with this balance of plant stuff.
9 MR. KERR: Excuse me, but I can't imagine an 10 operating utility that wouldn't be concerned with the balance 11 of plant. I must be missing something, that you had to do 12 inspectiona to discover that.
13 MR. LeFAVE: The degrees of concern and the methods 14 of addressing BOP problems vary significantly from plant to 15 plant.
16 MR. KERR: Well again, surely this doesn't come as a 17 surpris3.
18 MR. LeFAVE: It came as a surprise that the 19 attention that was being paid, it did come as a surprise to 20 most of the inspectors, yes, that --
21 MR. KERR: They didn't realize that people who 22 operated power plants were interested in balance of plant?
23 MR. LeFAVE: We knew they were interested, but the 24 depth of the modifications and, put it this way, the programs
{} 25 they had in place for performing root cause analyses, Heritage Reporting Corporation (202) 628-4888
385 1 determinations, and follow up on identified possible problem 2 areas were, in some cases were identical to those that, they 3 gave them the same priority as safety-related systems. That 4 was surprising.
5 The biggest thing that makes a difference ir. that a 6 lot of times just to keep the plant running, rather than to do 7 a total root cause determination is to just fix the symptom 8 rather than the root cause. That has been done a lot in the 9 past, but we think that's changing.
10 MR. KERR: I guess the inspection was worthwhile 11 then. If staff came to realize that some of the utilities 12 really were concerned about the total plant.
13 MR. LeFAVE: Yes.
14 MR. WARD: Is your conclusion that some utilities 15 were and some weren't, or they all were?
16 MR. LeFAVE: They all were to some degree. Some of 17 them had started on an early time frame, and you could tell 18 the difference in the number of trips that the plants that had 19 just really started getting more in-depth, detailed programs 20 to address BOP, complete engineering departments just focusing 21 on the balance of plant. Those that, they were a couple of 22 years behind, some of them that had really got on board 23 earlier, there's a definite improvement in the number of 24 trips. A significant improvement in the number of trips.
25 Where we're at today is we have a task action plan Heritage Reporting Corporation (202) 628-4888
- i
386
~
1 that it attempting to assess the safety significance of the 2 balance of plant and develop rules, regulations, guidance or 3 whatever is necessary based on the study. This contract 4 apparently just went out, I think they start work last weak or 5 this week with SAIC to, among other things, to perform a study 6 of the industry programs, utility programs, and staff programs 7 that are associated with the balance of plant systems.
8 MR. KERR: You're going to be sure they know about 9 draft 11-50?
10 MR. LeFAVE: I'll make sure. I am involved in this 13 right now.
12 MR. MICHELSON: Are they addressing the full safety C. 13 significance, or are they looking at only the transient and t
! 14 trip aspects? I mean are they looking at the effects of 15 external events on the safety of the plant as it may come 16 through the BOP and so forth?
17 MR. LeFAVE: I don't know that any of that will get 16 to that point.
19 MR. MICHELSON: I want your definition of safety l 20 significance.
l l 21 MR. LeFAVE: Actually I think it's basically going l
l 22 to deal with internal events from the balance of plant. What 23 the reliability, not the reliability of the numbers per se, or 24 maybe they will be studying risk and risk assessment I 25 understand. With regard to external events and that kind of
{" }
Heritage Reporting Corporation (202) 628-4888
,m 387
(.)
1 thing, I dca't know if that really would relate to the balance 2 of plant per se.
3 MR. MICHELSON: That's what you have to do by doing 4 a study. That's what you find out by doing a study is whether 5 it relates or not. Such studies haven't been done in the past 6 and I just wondered if this was finally the one.
7 MR. LeFAVE: The external events you are talking 8 about, that's being investigated in A17. One of the A17 9 systems interactions is making --
10 MR. MICHELSON: I know a little bit about A17.
11 MR. LeFAVE: As broad based as this external event 12 category is and the types of studies you are talking about,
' ) 13 are very broad in themselves.
14 MR. MICHELSON: A47 is really the right number to 15 name. That was supposed to have looked at the safety 16 implications of these balance of plant systems. We already 17 got through the resolution of that, and this is not in the 18 resolution of it even, as you got a letter from us.
19 MR. REED: I worry about what I perceive as the 20 enthusiasm for regulatory involvement in-depth in balance of 21 plant. I might point out that probably utility initiatives 22 for looking at balance of plant is spurred on by the plant 23 that spinning reserves in this country are declining all the 24 time, and the outage becomes ever, unscheduled outage becomes l
{} 25 ever more of a problem. Therefore, there is this natural
[
Heritage Reporting Corporation (202) 628-4888 f
g)
(m 388 1 inclination because of state regulatory authority 2 investigations, for the utilities to try to prevent outages 3 unscheduled.
4 Now I might point out something. In my opinion, 5 reactors that will be successful in the future should have 6 more independence of safety systems from balance of plant.
7 We've neard the words dedicated, separated, decay heat 8 removal. I don't know who it was that ever in the regulatory 9 thought that auxiliary feedwater systems or emergency 10 feedwater systems, secondary site equipment, were not safety 11 systems. But somebody apparently thought that was the case 10 12 or 20 years ago. So utilities in the industry got started off 13 on safety systems that were not declared safety systems.
14 Whenever you're going to use emergency feedwater, 15 which is spread-eagled and spaghetti'd all over and supported 16 all over throughout balance of plant, whenever you're going to 17 use this as your only and perhaps most successful way of decay 18 heat removal, then damn it, it should never have been that 19 way. It should have been a safety system, and then I think it 20 would have been built more separated and more independent.
4 21 End of speech.
22 Quite frankly, NRC may get all wound up on balance 23 of plant oversight, and reactors ought to be much different 24 from that in the future, and yet they had like created the 25 great regulatory empire pursuing balance of plant which I'm
, {~~s}
( Heritage Reporting Corporation (202) 628-4888
~
r l
C) 1 not so sure they'll have the budget and competence to take 2 care of.
3 MR. LeFAVE: You may be right. In fact this new 4 maintenance rule, that's one of the things that will be 5 studied under this, that it's a factual imbalance of plant, 6 and there may be nothing, no rules or regulations, or 7 guidance come out of the balance of plant program based on 8 improvements that have been done just due to maintenance, the 9 new maintenance rule.
10 MR. KERR: You're really serious about a maintenance 11 rule?
12 MR. LeFAVE: Isn't it in the mill now, or has it 13 already been --
14 MR. KERR: I guess so, but I would have thought it 15 might be reversed.
16 MR. LeFAVE: AFW system. Recommendation to ensure 17 that does not conflict with rules, regulations, or guidelines.
18 We found there were no conflicts.
19 With respect to the benefit of the actual 20 recommendations, many of the call for, and this is true of a 21 lot of recommendations throughout the program, call for 22 analysis and evaluation by the utilities. We won't be able to 23 tell the actual benefit until the implementation phase of 24 these recommendations, because they can be taken to different 25 depths by different utilities. This is especially true with
[}
Heritage keporting Corporation (202) 628-4888
(~) 390
\_/
1 the design objectives, the design recommendations. So we will 2 not see any immediate impact on the AFW system except maybe in 3 the area of testing and maintenance where these 4 recommendations can be put into place without a whole lot of 5 analysis and evaluation.
6 At the time of our review the steering committee had 7 not acted on many of the recommendations in Appendix Q related 8 to testing and design objectives. I believe that's because 9 they're related to a lot of the recommendations that are 10 already in the tracking system. I don't know where they stand 11 now, how many of them have been entered. But there are many 12 recommendations already in the tracking system related to the 13 AFW systems. I think it would be prudent to review the ones 14 that are already in there. Some of the design objectives are 15 probably already met by the recommendations that are in there.
16 BWOG did not make an effort to address the addition
- 17 of a third AFW pump. We are pursuing this separately under 18 generic issue 124.
19 MR. KERR: I'm sorry, I didn't understand. What did 20 you say? B&W did not --
21 MR. LeFAVE: They did not address the addition of 22 the third AFW pump with respect to reliability of the AFW 23 system.
25 MR. LeFAVE: Third. Number three. Some of the Heritage Reporting Corporation (202) 628-4888 l
(i 391
\m/
1 plants only have two.
2 We are pursuing this. -- gives you 124. I guess 3 it's related to something that was brought up yesterday 4 regarding the feed and bleed, that the present, the staff 5 objective now and under GI-24 is to make the system or a 6 system that can remove heat through the heat exchange, through 7 the steam generators, meet the reliability criteria of 10 to 8 the minus fourth, 10 to the minus fifth unavailability as 9 defined in the standard review plan section, 10-49 related to 10 the AFW system. So we are not relying at all on the feed and 11 bleed with regard to the generic issue 124. That was, I think 12 proposed. Either the AFW system has to meet the 10 to the 13 minus fourth, 10 to the minus fifth unavailability, or other ,
14 compensating factors would be considered that the system 15 itself didn't meet them, but that would not be feed and bleed. ,
16 So we have two plants now that I believe we are still pursuing 17 this with.
18 MR. KERR: Are you telling me that in effect 10 to 19 the minus 4 is now a regulation?
20 MR. LeFAVE: It's in the standard review plan as an 21 assessment criteria. I wouldn't call it a regulation.
22 MR. KERR: You said a plant would have to meet 23 something.
24 MR. LeFAVE: It's a generic issue. We went out with
(} 25 generic letters, I believe it's a generic letter. I don't Heritage Reporting Corporation (202) 628-4888
fs 392 k,)
1 know if you call it a regulation.
2 MR. CATTON: It's the same effect.
3 MR. KERR: If a plant has to do something, as far as 4 I'm concerned it's a regulation. What I'm trying to find out 5 is whether you're telling me that now plants have to have off 6 speed water systems, they have to have a reliability of 10 to 7 the minus 4?
8 MR. LeFAVE: Only in about six plants, we are 9 looking only at the plants that only have two AFW pumps, and 10 yes, I would say basically they have to have a heat removal 11 system that will meet that criteria.
12 MR. KERR: Why should they have to meet the O 13 criterion and other plants not?
14 MR. LeFAVE: We believe the other ones already do.
, 15 MR. KERR: You believe they do?
16 MR. LeFAVE: Yes.
17 MR. KERR: On what basis?
18 MR. LeFAVE: On the basis of the 2E11 review that we 19 did following TMI-I and the improvements made on the systems, 20 we're pretty sure that they would all pass that criterion. We j
l 21 didn't do a recalculation of them. We did calculations of the l
t 22 individual plants, but we didn't do a recalculation after all 23 the improvements were made. We didn't think it was necessary.
24 MR. KERR: Okay.
If these recommendations are
(} 25 MR. LeFAVE:
Heritage Reporting Corporation (202) 628-4888 l
393 T'i V
1 implemented properly to the degree that they ought to be, then 2 the reliability, availability of the AFW system should be 3 enhanced by varying degrees dependent upon the individual 4 utility.
5 With regard to turbine controls, the turbine pumps, we 6 have a recommendation that they start a research program to 7 possibly determine improvement, to come up with a completely 8 new design for a turbine control system. There are already 9 many recommendations in the tracking system fer improvement in 10 the reliability, and they rejected this recommendation. We 11 just suggest that they keep it in mind pending the 12 implementation of the recommendations already in the system if
() 13 they don't get the reliability improvement in turbine control 14 that they hope that they ought to maybe reconsider this 15 research program.
16 MR. REED: I thought I heard you say, and correct me l
17 if I'm wrong, I thought I heard you say that the auxiliary 18 feed water system is the system that you want to meet 10 to 19 the minus 4, and you do not consider the bleed and feed 20 activity as supportive.
21 MR. LeFAVE: That's only from the generic issue 22 standpoint.
23 MR. REED: I see. I was trying to see how that l 24 jived with B&W's claimed advantage yesterday for bleed and 25 feed compared to PWR's. I guess they can claim that advantage l
IIoritago Reporting Corporation (202) 628-4888 l
F
[b s.-
394 1 for their bleed and feed, even though apparently they're not 2 claiming an advantage for their auxiliary or emergency feed.
3 MR. LeFAVE: Yes.
4 MR. KERR: In this third pump you're recommending, 5 you do not give credit for a pump that is not safety grade.
6 MR. LeFAVE: Yes we do. It must be tech spec'd. It 7 has to be tech spec'd, but we do give credit for -- with 8 respect to unavailability.
9 MR. KERR: And the tech spec simply says it has to 10 be there?
11 MR. LeFAVE: It has to be operable. It may have a 12 down time of seven days or something like that. But it has 13 to be tech spec's.
14 MR. WYLIE: Let me ask the owner's group. On the 15 staff recommendations on the research program turbine 16 generator controls, I know you have a recommendation regarding
! 17 improving the reliability of the turbine generator by doing l
18 certain things to the control system. Was there any other 19 actions you plan in this area?
20 MR. RUTHERFORD: Other than the recommendations that
- 21 we presently have?
i 22 MR. WYLIE: Yes, you have one regarding the drain l 23 tank level controls and the EHC controls, improvements. Those l
24 are the only two, they recommend you list in summary now. Do 25 you have another program to improve the reliability? Since
}
Heritage Reporting Corporation l (202) 628-4888 l
395 r~T
(/
1 that is a major contributor to your trips.
2 MR. RUTHERFORD: This is an auxiliary feedwater 3 system turbine.
4 MR. WYLIE: No , I'm talking about, no this is the 5 turbine generator.
6 MR. RUTHERFORD: No, this is the auxiliary feedwater 7 turbine-driven pumps.
8 MR. WYLIE: Oh, okay.
9 MR. RUTHERFORD: Really to address your question on 10 turbine control, I think we pretty well cured that problem.
11 In fact Oconee was a chief contributor to that in the early 12 '80s and we made some changes at Oconee that pretty well have
() 13 eliminated that as a source of trips at Oconee. So the 14 experience over the last two or three years has been very good 15 in that respect, even though it shows up as a dominant 16 contributor when you look at the experience in 1980.
17 MR. LeFAVE: The secondary plant relief system is 18 basically a main steam pressure control review to reduce the 19 number of safety valve actuations following reactor turbine 20 trips. The slides you have, the last slide in the series of 21 slides is one that Gary will talk about, that are related to 22 this regarding the valve task force.
23 Along with that we talk about the testing and 24 maintenance and performance of safety valves and the raising 25 of the safety valve set points that was one of their Heritage Reporting Corporation (202) 628-4888
396 p
b 1 recommendations from B&W. They basically were talking about a 2 three phase approach to this reduction program. The second 3 phase was raising the set points of the safety valves.
4 AFW recommendations are responsive to the concerns 5 identified during this particular review which, as you recall, 6 the overcooling effects on pressure, adjust of flow, trying to 7 limit the flow rate of the AFW, smooth flow control, raising 8 of the level rather than an on/off situation which some of the 9 controllers had, was giving problems due to overshooting and 10 trying to keep the safety valves closed.
11 The staff basically agrees with the phased approach 12 and that the phase one of the program is involved with the lh) 13 testing / maintenance / performance of the safety valve systems 14 themselves, the safety valves themselves.
15 Phase three is the combination that, BWOG made 16 studies of a combination of different modifications that could 17 be done to the plant which included high capacity, bypass 18 systems, quick acting bypass valves, increasing the MSSV set 19 points, the safety valve set points, and possibly changing the 20 signals that actually actuate the turbine bypass system. And 21 also possibly using the turbine to handle the initial high 22 heat load from the reactor. Made studies of these different 23 combinations of these which they don't plan to complete unless 24 phase one and phase two don't give them the performance they 25 are looking for regarding reduction in the safety valve relief O
(G Heritage Reporting Corporation (202) 628-4888
397 1 actuations.
2 Actually phase three, we recommended that they 3 continue to investigate these possible modifications, kind of 4 in parallel with phase one. According to what we understand 5 they are still doing some of those investigations. I 6 understand at one plant they did substantially reduce these 7 safety valve inadvertent openings, not inadvertent, but 8 excessive openings, by just reducing the turbine bypass, that 9 point where you don't overshoot the reseat, the reseat point 10 from the safety valve reseating does not cause the go below 11 the turbine bypass system at that point. Did I get that 12 right?
13 The turbine bypass set point is set to control, 14 probably trip at a certain level. Sometimes the safety valve 15 resent point drops below that so you don't even get out of the 16 turbine bypass system which causes operator action to try and 17 reduce the, take manual control of the turbine bypass system 18 and reduce it to reseat the valves. This is a major cause of 19 the pressure complexity following the plant trips.
20 The staff feels that proper implementation of these 21 recommendations will result in a definite improvement of the 22 performance of the relief system, and the post trip pressure 23 control, and operator burden and ultimately result in an 24 enhancement of plant safety.
I think that's all we have on secondary relief.
(} 25 IIeritage Reporting Corporation (202) 628-4888
398 (G"g' 1 Gary will again talk to the instrument air system.
2 MR. KERR: At what point will the staff conclude 3 that safety enhancement is sufficient?
4 MR. LeFAVE: We don't see any further 5 recommendations that have to be made. I don't know how to 6 answer that question.
7 MR. SIEGEL: I think what's going to happen is once 8 the audit group implements these recommendations, the staff is 9 going to review the plant performance to see how they operate 10 and see, once they're all implemented it may take several 11 years to make a determination of acceptability of the program.
12 But we're going to be reviewing it at the time or looking at
() 13 their performance at that point in time,, and we should be able 14 to see, or we hope based on all these recommendations, be able 15 to see a significant improvement, and that the owncrs group 16 would exceed the goals they've set for themselves as far as 17 the number of complex transients and reactor trips. So it's 18 not a short term assessment. It's going to take awhile.
19 MR. KERR: It won't be based on adherence to 20 regulations or anything of that sort, it would just be --
! 21 MR. SIEGEL: Essentially surveillance.
1 22 MR. KERR: That the number of transients has been 23 reduced sufficiently.
l 24 MR. SIEGEL: That's right, and that we see a l
i 25 reduction in the number of transients. And also I would hope
(
l Heritage Reporting Corporation (202) 628-4888 l
f% 399
%_/
1 that as a result, as we said, we want to eliminate all the 2 transients, but we would hope that the complex transients 3 would be less severe as a result of these programs, too.
4 MR. LeFAVE: Instrument air system. The instrument 5 air system was included because their failures, they were 6 similar in extent and character to loss of the ICS power in 7 that they increased the complexity following the trip, 8 although they didn't experience as many failures as they did 9 in the ICS.
10 During our review we compared the recommendations 11 made by BWOG with the recommendations and the AEOD report 12 related to the instrument air with very favorable results.
13 The recommandations made by the BWOG go well beyond the scope 14 of the AEOD report, but I guess mainly because they're more 15 specific. I think there are about 50 recommendations 16 regarding the air system.
l 17 We agree with the conclusions, basically we agree l
18 with the conclusions and recommendations made by BWOG with 19 regard to the air system. We think we should see a quicker 20 turn around on the safety impact or the improvement in this i
21 air system related to some of the recommendations made for the 22 other systems in that they do not require an extensive 23 evaluation or analysis before they are implemented. So 24 implementation of these can be done a lot quicker than can be 25 done with some of the other systems where the individual
[}
l l Heritage Reporting Corporation l (202) 628-4888
3 400 (V
1 utilities require their own analysis.
2 The recommendations made address the staff's recent 3 concerns, mainly regarding the cleanliness of the air, 4 emergency procedures and training, testing of the system, and 5 the basic overall performance of the instrument air system.
6 We did make recommendations that BWOG should 7 consider making a recommendation for a gradual loss of air 8 system test which is something that Mr. Michelson was talking 9 about. This was one of the recommendations in the AEOD report P
10 that although the instrument air system study by BWOG did 11 address the fact and did make note that they could have some 12 problems due to gradual air loss, they didn't make a specific
(~)/ 13 recommendation. In any rate, the staff is, I think within the 14 next year, will have a recommendation out with this type of 15 test for all plants, B&W, Westinghouse, GE, what have you.
16 There's a presently ongoing study on that.
l l 17 There's a generic letter about to be issued also, 18 that doesn't include this gradual loss of air test mainly I 19 because this is considered a back fed item and needs a lot l 20 more regulatory analysis to go through a CRGR.
l 21 Another item we recommended, that although the BWOG 22 report said that they didn't consider analysis regarding loss 23 of off-site power and causing a loss of instrument air, but 1
24 they did make note of where these things did occur. We 25 thought they ought to make a recommendation that an assessment Heritage Reporting Corporation 1
(202) 628-4888
0 l
() 401 1 be made regarding the loss of air due to the loss of off-site 2 power. We think each plant ought to do that.
3 MR. KERR: Has that ever been done in a PRA?
4 MR. LeF 7E: I don't believe so. The individual 5 plants may have dare it, but I don't think the staff has done 6 a PRA.
7 MR. KERR: Is that what you had in mind? That they 8 do a PRA?
9 MR. LeFAVE: No, not a PRA. Just to make sure that 10 they know what's going to happen through an analysis, 11 basically, to know that they have looked at the effects of the 12 loss of air following a loss of off-site power. Maybe they 13 might want to consider --
14 MR. KERR: When you say assess the impact on safety, 15 you just mean look and see what would happen?
16 MR. LeFAVE: Yes.
17 MR. MICHELSON: That's with a gradual loss of air 18 most likely.
19 MR. LeFAVE: That's true.
20 MR. MICHELSON: Unless you put in a dump valve it's 21 going to be a gradual loss of air.
22 MR. LeFAVE: That's true.
23 There were four recommendations we thought ought to 24 be added to their list of key recommendations because the
{} 25 first one, inspection for leaks, that is not a difficult Heritage Reporting Corporation (202) 628-4888
i O 402
\
1 recommendation to perform, and such leaks could lead to 2 contamination and may lead to exceeding the capacity of the 3 air system.
4 MR. KERR: Offhand, one would say if you have a 5 gradual loss of air likely your accumulators are going to 6 slowly bleed down with that gradual loss of air, and when you 7 need the accumulator, it's going to be too low a pressure 8 already.
9 MR. MICHELSON: It may only be half full.
10 MR. KERR: Just offhand, you would say gradual loss 11 of air is going to be kind of a tough proposition in some 12 cases.
13 MR. LeFAVE: Keeping line with the accumulator 14 and shut valves, we made the recommendation that that be a key 15 recommendation because all the safety-related valves have 16 accumulators and since this has a definite safety impact 17 because it has to do with performing a safety function, we 18 thought that ought to be on the key list. There was a recent 19 experience at Fort Calhoun where they saw a potential serious 20 problem due to the same effect. They performed a test on the 21 accumulator for, actually an instrument for the -- storage 22 tank, air bubbling instrument, saw that the accumulator didn't 23 have enough in it to bleed down to the point where the ECCS 24 would shift over to reserve before you had enough water in So we think this
{} 25 there to assure enough positive suction --
Heritage Reporting Corporation (202) 628-4888
/~ 403 N_T )
1 is one that definitely ought to be on the key list.
2 MR. KERR: What did you have mind when you talked 3 about inspection of leaks? How does one inspect for leaks?
4 MR. LeFAVE: I think that -- the system or a -- of 5 the system by flow rates, capacities. You can get a feel for 6 the air usage by turning the flow capacities, the actual flow 7 rates, the usage factors of the system.
8 MR. RUTHERFORD: That's difficult to do. It 9 involves walk downs.
10 MR. KERR: You go around and put soap bubbles on 11 everything and see if the bubbles burst or something like 12 that?
I
\/ 13 MR. RUTHERFORD: Go around and listen.
l 14 MR. MICHELSON: That's an inspection.
15 MR. RUTHERFORD: Not necessarily soap bubbles.
16 VOICE: I think you can actually look at how many 17 times the air compressors take to refill the tanks.
18 MR. RUTHERFORD: I disagree. That's not a very easy 19 way, you don't know what air demand is. If you've got that 20 big a leak that it's increasing demands on your compressor, t
21 then you're going to know about that. What you're going into 22 is a lot of very small leaks. That's difficult to sort out in 23 just looking at compressor run time and that sort of thing.
I 24 MR. KERR: What is your view on check valves on N 25 accumulators, if you're getting a sloti degrading of the air l
Heritago Reporting Corporation (202) 628-4888
, 404
(_m) 1 system? What's seating the check valve, or why isn't the 2 accumulator just coming on down with the air system? Unless 3 there are spring-loaded checks or something, and even that, 4 I'm not sure would quite do the trick in this case. It looks 5 to me like degraded air, you've lost on accumulators. If it 6 comes down slowly. We found that on the BWR's. We had to put 7 dump valves in to knock the air pressure off quickly.
8 MR. SKILLMAN: I think the real way out of this mess 9 is to ensure that we're monitoring the pressure at the right 10 point in the system and making it necessary for the operators 11 to take action on certain air pressure --
12 MR. KERR: We're worried about loss of off-site 13 power and that sort of thing. We may end up with dump valves 14 that try to break the air line real quick so that the 15 accumulators can -- and hold what they've got left. I think 16 that's a way out of that mess, but it's not a nice way.
17 MR. SKILLMAN: Right.
18 MR. LeFAVE: Any other questions?
19 MR. MICHELSON: One more point. You're aware now
( 20 that the generic issue is in the process of being established l
21 on air systems with a high priority. How is that going to be 22 fit in, since I'm sure that will come out with a resolution l 23 three or four years from now.
I 24 MR. LeFAVE: That will be planned to be done I think i
l 25 by the end of the year.
V(~N Heritage Reporting Corporation (202) 628-4888 t
L. _ _
405 j))
1 MR. MICHELSON: The resolution?
2 MR. LeFAVE: Yes.
3 MR. MICHELSON: Oh, I haven't had time to read it.
4 I got it in the mail but I didn't have time to read it.
5 MR. LeFAVE: It's actually being done in two phases.
6 The first phase is the generic letter tc 2nsure all the plants 7 actually meet what they have, what they're supposed to have 8 now with regard to testing, maintenance. Even that should 9 have been out three or four months ago, but now we have to go 10 through --
11 MR. MICHELSON: It's going to be factored into the 12 B&W owner's group very quickly so that when they finish --
O k/ 13 MR. LeFAVE: They have pretty much addressed 14 everything we wanted to come up with except for the gradual 15 loss of air. That's something they know about, and hopefully 16 they'll factor that in.
i 17 MR. MICHELSON: The other issue then is going to 18 come to a head very quickly in terms of a resolution.
19 MR. LeFAVE: Yes.
l 20 MR. MICHELSON: Thank you.
I 21 MR. WYLIE: Before we get into that, I think we'll 22 take a break.
23 MR. LeFAVE: This is part of the system. There's 24 one more slide.
25 MR. WYLIE: But we haven't heard B&W owner's group Heritage Reporting Corporation (202) 628-4888
406 1 on this, I don't think.
2 MR. LeFAVE: They don't have a separate valve test 3 program, I don't think, but this is basically related to the 4 secondary plant system review, the steam pressure control 5 review, secondary relief valves.
6 MR. MICHELSON: I would think, Charlie, that we'd 7 want to hear about that program once they issue their report 8 which I gather was August, a valve report. So the next valve 9 meeting might be scheduled --
10 MR. LeFAVE: Let me show you want the slide is and 11 see if you want to, it really has to do with the secondary 12 system pressure control.
13 MR. REED: Is this open for discussion?
14 MR. RUTHERFORD: We don't have any specific 15 presentation on this. This is really part and parcel of the 16 systems reviews that we've already gone over.
i 17 MR. WYLIE: Okay.
I 18 MR. REED: Are you going to consider this now or are 19 you going to take a break?
l 20 MR. WYLIE: Let's take a break. We'll come back at l
l j 21 ten after.
22 (Whereupon, a brief recess was taken.)
l 23 MR. SKILLMAN: I was asked to take a minute, so with 24 the clock running and 59 seconds left, valve task force
[} 25 activities came as a follow up activity to the secondary Heritage Reporting Corporation (202) 628-4888
i r3 407 (J
1 system pressure control review that we did. Please regard 2 this information as preliminary. The areas we were looking at 3 were the areas of main steam safety valve performance, turbine 4 bypass, atmospheric dump valve performance, motor operator 5 reliability with regard to this relief function, and general 6 post trip secondary pressure control.
7 We developed a guideline for MSSV set point testing, 8 looking at these maintenance procedures. We have evaluated 9 and are evaluating efforts to reduce the main stream safety 10 valve lift and we're developing a generic program guideline 11 for the testing.
12 Let me speak a few numbers to put this in the proper
'-) 13 perspective.. How many main steam safety valves are there per 14 plant? How many lift on a trip? So when you look at the 15 trip frequency, and the number of valves that are involved, --
16 you find there is indeed a loss of lift because -- coupled 17 with the number of trips per year at a plant, because there 18 are 18 lifts per turbine --
19 MR. MICHELSON: Are those spring loaded safeties?
20 MR. SKILLMAN: Yes sir. These are the codes.
21 There are 16 to 18 lifts at the highest point.
22 MR. MICHELSON: They have a bad reputation for 23 sticking don't they?
24 MR. SKILLMAN: That's what we want to go after.
25 Dave mentioned earlier that one of the key aspects
}
Heritage Reporting Corporation (202) 628-4888
408 1 of a Babcock trip is for the operator to take manual control 2 of steam header pressure, drop steam header pressure until 3 these valves reseat. That is what has been driving a portion 4 of the owner's group effort in this regard. Let's get the 5 operator out of the loop and let these valves seat when they 6 are supposed to and open when they are supposed to.
7 MR. MICHELSON: They relieve to atmosphere?
8 MR. SKILLMAN: Yes sir. These are the ASME code 9 valves. They are able to take full lift speed pressure to 10 protect the system for ASME section three.
11 MR. MICHELSON: There are no block valves?
12 MR. SKILLMAN: There are no block valves. These go 13 right to the atmosphere.
14 What we wanted is the blow down range to be seven to 15 nine percent. In some cases we have in fact overshot. We 16 have done below that. We should not blow down that far.
17 MR. RUTHERFORD: Excuse me, let me clarify that. We 18 found that the blow down is in the range of seven to nine 19 percent. The original design was three to five percent.
20 That's the reason the operator has had in certain instances to 21 step in and lower turbine pressure so that valves reseat.
22 MR. SKILLMAN: In the final bullet there, we would 23 like to get the main steam safety valve performance and the 24 turbine bypass valve performance corrected to improve the post
{} 25 trip secondary pressure control. This ties into much of our Heritage Reporting Corporation (202) 628-4888
(~S 409 l
(_)
1 discussion yesterday regarding what parameter is being 2 controlled. What we're saying here is we need to get the 3 secondary pressure under control so as to get into a prompt 4 energy balance post trip.
5 My sixty seconds are up.
6 MR. REED: Two points.
7 You recognized that main steam safety valves spring 8 loaded have been around for a long time, and there have been 9 a lot of people that have been annoyed for a long time about 10 their malfunctions. You realized that taking on this 11 effort --
12 MR. SIEGEL: Excuse me, at the point of being rude, O
kJ 13 I don't intend to be, but we have somebody that has to leave 14 at 12:00. If you could defer your questions until after, 15 we'll come back to them. If it's all right with you.
16 MR. REED: Should I put them in writing and send 17 them tomorrow?
18 MR. SIEGEL: No, he's been waiting around he came 19 back this morning with a quick schedule.
20 MR. DeBOR: Good morning. My name is Joseph DeBor.
21 I'm with SAIC. I'm here to discuss the human factorr, 22 evaluation of the reassessment program.
23 SAIC was tasked by NRC to evaluate the human factors 24 adequacy of the B&W reassessment program. We had four tasks.
25 We were tasked to review the operator / maintenance personnel Heritage Reporting Corporation (202) 628-4888
r~g 410 V
1 interview project. We were tasked to review the operating 2 experience reviews, the procedure reviews, and the operator 3 burden and sensitivity studies from a human factors 4 perspective.
5 The review was conducted on behalf of NRC by SAIC, 6 COMEX Corporation,and Carlow Associates.
7 The first task we reviewed was the operator and 8 maintenance personnel interview project. We determined as a 9 result of our review that B&@ had identified 11 very specific 10 hardware problems, and identified concrete recommendations to 11 go along with those hardware problems.
12 We also determined that B&W had identified six human 13 engineering problems, but they didn't identify any concrete 14 recommendations or follow-on actions as a result of their 15 identification of human factors problems.
16 The operators and maintenance people identified, for 17 example, ICS feedwater control of T-Ave is poor. That's a 18 human factors problem. They also stated that it was difficult 19 to tune a secondary system at less than 100 percent power.
20 They also concluded that the delayed subcooling margin 21 instrumentation was potentially confusing to operators during 22 emergency operations.
23 Our concern was when we looked at this operator and 24 maintenance personnel interview project that B&W was very good 25 at identifying concrete solutions to hardware problems, but Heritage Reporting Corporation (202) 628-4888
/
411 1 when it came to man / machine interface problems, they didn't 2 identify specific recommendations associated with them.
3 The second task we were involved in was the review 4 of operating experience which was basically the review of the 5 transient analysis program reports. We determined that B&W 6 conducted a comprehensive review, and that the human factors 7 issues were summarized and characterized appropriately. For 8 example, human interface involved operations and operating 9 procedures, surveillance and testing and maintenance issues, 10 and they also identified problems associated with training and 11 displays.
12 The third task we reviewed was the procedures 13 review. This is basically the review of the ICS/NNI 14 procedure, and we determined that B&W again identified 15 significant human engineering concerns involving components 16 and displays on loss of ICS/NNI. We also determined that the 1*/ B&W recommendations for labeling and component modifications i
18 were appropriate, but they're very ceneral and not plant 19 specific.
20 The fourth task we reviewed, actually it was a 21 combined task of operator burden and sensitivity analysis. We 22 determined that the operator burden study findings are valid 23 human engineering issues, and B&W recommendations were 24 appropriate. We felt the assessment of the human engineering
[} 25 issues and their recommendations were quite valid. The top i
Heritage Reporting Corporation (202) 620-4888
412
(~T
-V 1 human engineering concern identified under the operator burden 2 project was the control of steam and feed flow on loss of 3 automatic control. Then they went down the list through 4 drastic operator actions such as being afraid to turn off the 5 pump because they were afraid it wouldn't restart. The third 6 item in the list was overcooling mitigation strategy. That's 7 an interesting problem where the operators face the problem of 8 whether they should overcool or undercool the plant in an 9 emergency, and then they had a number of other problems and 10 recommendations associated with operator burden.
11 MR. MICHELSON: Could I get a clarification?
12 MR. DeBOR: Yes.
13 MR. MICHELSON: You were looking just at the results 14 of the owners group work on these 11 transients, is that 15 right?
16 MR. DeBOR: That's correct.
17 MR. MICHELSON: You weren't attempting to think or 18 look or wonder about accident situations and so forth. Just 19 the transients.
20 MR. DeBOR: That's correct. SAI only looked at the 21 products produced by B&W as a result of their review of the 22 six transients.
23 MR. MICHELSON: So even looking at those particular 24 transients the operators were worried about during a transient 25 of stopping a pump that they might have to restart, that sort Heritage Reporting Corporation (202) 628-4888
~~'
413 (V
i 1 of thing?
2 MR. DeBOR: Yes sir.
3 MR. MICHELSON: Thank you.
4 MR. DeBOR: The second part of task four was the 5 sensitivity study. The B&W sensitivity study we concluded was 6 comprehensive. We determined that the B&W recommendations 7 imply far reaching man-machine interface changes to the 8 control room. One of the ones that sort of startled me was 9 the elimination of anticipatory reactor trip on turbine trip, 10 which I learned yesterday from Dick that B&W decided to shelve 11 that idea for whatever reason, that they're not going to 12 pursue it. But that's a very significant man / machine O 13 interface problem if they do decide to pursue it. They made a l 14 number of other very genera.1 4wccaranda(4ngs such as reducing
. 15 the probability of overcooling on turbine trip.
16 Our overall assessment of the B&W findings was that 17 the B&W studies did in fact result in valid human engineering 18 concerns. However, because we determined that human factors 19 professionals were not involved in the reassessment effort, 20 the completeness of the effort in identifying human factors 21 concerns is uncertain.
22 The second conclusion is that the proposed 23 corrective actions imply very significant changes in the
( 24 man / machine interface environment in control rooms, but they
(} 25 are very general and not plcnt specific.
Heritage Reporting Corporation (202) 628-4888
~
(
1 The proposed corrective actions also do not havo any 2 specific implementation dates associated with them.
3 MR. MICHELSON: Can you give me an approximately 4 example of a significant change to the man / machine interface 5 just so I can get a feeling for how significant we're talking 6 about?
7 MR. DeBOR: If they decided to eliminate reactor 8 trip on turbine trip, that's a very significant change.
9 MR. MICHELSON: I thought that, the second bullet is 10 the one I'm referring to. I thought that you were talking 11 here about the bench board changes or something like that. Is 12 that not what's being referred to?
O 13 MR. DeBOR: Yes, they are planning --
14 MR. MICHELSON: Besides the anticipatory trip 15 question, give me another example of a very significant change 16 to the man / machine interface.
17 MR. DeBOR: Planning to review the enunciator 18 systems and make changes to the enunciator systems, dividing i
They have recommendations 19 up multi-point enunciator alarms.
20 to reduce the probability of overcooling the plant. In order 21 to do that they have to go through a very significant process, 22 by looking at issues such as the emergency feedwater l
23 initiation control system and determining if that man / machine 1
24 interface is appropriate to the tasks that face the operator
(} 25 on an emergency.
Heritage Reporting Corporation (202) 628-4888
1 MR. MICHELSON: That's a little different than I 2 understood from the bullet. I thought they had already in 3 mind some very significant changes in the man / machine
- 4. Interface itself.
5 MR. DeBOR: A lot of the recommendations are very 6 general. They're interested, they, the owners group, have 7 asked me individual plants, to review issues such as the 8 drastic actions issues, and then take plant specific action 9 based on the plant specific reviews.
10 MR. MICHELSON: Thank you.
11 MR. WYLIE: On your first bullet, can you point to a 12 significant deficiency in the review that warrants the O 13 professionals being involved in the process?
14 MR. DeBOR: The problem with the process as it was 15 undertaken by B&W is they only looked at a small set of 16 transients. They looked at their six Category C transients.
17 We over the years have been working with a number of B&W 18 plants on detailed control room design reviews where each of 19 the plants had human factors staff participating in a multi-20 disciplinary team to look at comprehensive sets of operator 21 tasks during emergency operations.
22 MR. WYLIE: That was the result of a TMI action 23 plan. As far as I know all the B&W units have adhered to l
24 those, have they not? Conducted hur m factors reviews? That
(} 25 has been done.
l l Heritage Reporting Corporation 1
(202) 628-4888 l
/
416 k-)
1 MR. DeBOR: They have conducted them.
2 MR. WYLIE: What I read in what you say is there are 3 significant deficiencies out there that weren't looked at.
4 I'm just asking you the question can you point to one.
5 MR. DeBOR: We can't point to one that is a 6 significant deficiency. As I've said, the human factors 7 deficiencies that were identified are valid. Our only concern 8 is that --
9 MR. WYLIE: You didn't look under every rock.
10 MR. DeBOR: Since they didn't have the same type of 11 people who participated in the control room design room use, 12 were very familiar with the control rooms and issues that r"N
\- 13 operators were concerned about, or a comprehensive set of 14 operator tasks, they really didn't take advantage of those 15 people who had that comprehensive knowledge of the control 16 room and the operator tasks in emergency operations.
l 17 Each of these plants has a group of human factors 18 operations and design engineers who designed the modifications 19 to their control rooms as a result of these detailed control
, 20 room design reviews. Thesa people, at Rancho Seco for l
21 example, looked at systems like the emergency feedwater 22 initiation of control system. When that system was first
! 23 proposed at Rancho there was no manual override for the epic 24 system. It was strictly an automatic system that was being I
t
{} 25 implemented in the control room. The operators and the 1 Heritage Reporting Corporation I
(202) 628-4888 l
l
a 417 1 control room design review human factorn people said we need 2 at least a manual override for this, so consequently they got 3 approved panels and they saw the problems that operators were 4 facing.
5 We were jtst a little disappointed in the fact that 6 these people, the people who have had so much experience with 7 the control room design and the modifications that are being 8 implemented over the next few years weren't included in this 9 study. I think a major recommendation that we made out of 10 this study was for B&W to include human engineering experts 11 such as those who participated in the develcpment and 12 modifications resulting from the recent control room design 13 reviews as members of the modification teams, when these 14 recommendations get to the point where they're actually going 15 to redesign panels and systems, include very experienced human 16 factors people in those designs.
17 MR. MICHELSON: I think, Charlie, they're saying if 18 you don't hire snake hunters to look for snakes you may not 19 find any.
20 MR. WYLIE: They might bite you.
21 MR. MICHELSON: Well, that's the only enes they'll 22 find are the ones that jump out and bite you. But we hope we 23 don't have too many of those cases.
24 MR. WYLIE: I agree.
25 MR. WARD: By the same token, we don't rely on Heritage Reporting Corporation (202) 628-4888
418 O
1 mechanical engineers to review electronic designs, even though 2 they might be able to do it more or less capably by the seat 3 of their panta or something. I think there is a prevalent 4 view that somehow human factors engineering is not a 5 disciplined engineering art, and I think Mr. DeBor is saying 6 that it is and it's problems that human factors engineers have 7 learned to deal with that create much of the residual risk in 8 nuclear power plants. It's kind of foolish not to take 9 ' advantage of the technology that exists. I endorse what 10 you're saying.
11 MR. DAVIS: I have a specific question. One of the 12 important conclusions that the owners group came to, according 13 to their report on page 311, states the following: "Operators 14 may be at times reluctant to take what they consider to be 15 extreme actions, such as feed and bleed cooling initiation."
16 Did you examine this issue? This looks like an 17 important one to me because if credit is going to be taken for 18 feed and bleed, then the operator should be well aware of when 19 he needs to initiate it and not be reluctant to do so when 20 it's required.
21 MR. DeBOR: That falls into the issue of drastic 22 actions. Again, the owners group has stated that yes, there 23 are problems where drastic actions present an operator burden 24 problem. The owners group sent a very general recommendation 25 out to each of the plants to look into the issue of drastic O
Heritage Reporting Corporation (202) 628-4888
419 1 actions. What does the plant actually do when they get this 2 recommendation from B&W on feed and bleed and other issues 3 where operators are really reluctant to take the action 4 required?
5 What I suggest here is that each of the individual 6 plants needs to bring that human factors person back into the 7 loop and review these issues on a plant specific basis and 8 participate as part of the overall team to develop a solution 9 to the problems.
10 MR. DAVIS: I guess what I'm asking is you didn't 11 look specifically at the instructions the operator goes to for 12 feed and bleed cooling to see if they are appropriate and 13 would help him get through this activity?
14 MR. RUTHERFORD: Let me respond to that. Maybe I 15 can clear up that issue a little bit for you.
16 You've really got two parts to this puzzle. First 17 of all, are the drastic actions you've got in your procedures 18 correct? Are they being taken at the right time? Second of I 19 all, the operator reluctance, perhaps, to act in accordance 20 with this procedure. We're addressing both of those concerns, 21 one being a technical concern, and the other being a 22 management / training concern.
l 23 MR. DAVIS: But isn't this beyond the scope of the 24 present study?
25 MR. RUTHERFORD: No.
Heritage Reporting Corporation l
i (202) 628-4888 I
L
7- 420 1 MR. DAVIS: I thought you stopped before you got to 2 two phase in your primary system, which is when you really 3 need to be sure feed and bleed is going to go.
4 MR. RUTHERFORD: As I mentioned, there are a lot of 5 things we're doing that cut across a lot of different events.
6 We didn't say okay, we get into two phased flow here so we're 7 going to stop and we won't go any further. When we saw this 8 as a concern, we --
9 MR. CATTON: From a sensitivity point of view, do 10 you think a dried out steam generator would unsettle an 11 operator more than one that still had some water in it?
12 MR. DeBOR: Obviously a dried out steam generator (r
\_) 13 creates a problem for the operators in B&W plants. For those 14 that have done task analysis on B&W plants, they know that it 15 does, the plant operates a little faster and it can present a 16 problem, yes.
17 MR. RUTHERFORD: Let me offer a comment here, too.
18 If the operator understands his system and the time he has 19 available and the scenario that he's in, obviously he's going 20 to be in a somewhat stressful situation there, no matter what 21 the plan -- feed water. But if he has an understanding 22 through his training, I don't think there's going to be a l 23 material difference.
24 MR. MICHELSON: Does he have any precautions on how (q 25 fast he can reflood that dried out generator from a thermo r (>
Heritage Reporting Corporation (202) 628-4888 l
l L
(~N 421 0
1 shock viewpoint?
2 MR. TAYLOR: I think the only precautions,-Carl, 3 would be related to his normal precautions on tube to shelf 4 Delta T. But you've got to remember that the once-through 5 steam generators are 90 percent dry at full power. When he 6- introduces off-speed water, that's always into the dry part of 7 the generator.
8 MR. MICHELSON: But there are no restrictions on how 9 quickly --
10 MR. TAYLOR: Except for tube to shelf Delta T's.
11 MR. MICHELSON: He is restricted on that?
12 MR. REED: Since this is a good time to throw a barb 13 on an old issue of mine, when you talk of training operators, 14 there are all kinds of personalities. There are nervous 15 personalities, and those that don't hardly react to a tornado 16 in their own backyard. I wonder if in this human factors 17 aspect and in this stressful scene of responding to, taking 18 action for a bleed and feed, if it's important to you in human 19 factors to know if the people have been aptitude tested and 20 personality evaluated? The operators, that is, that are going 21 to take this action.
22 MR. DeBOR: Well, all of the operators have been 23 through a training program.
- 24 MR. REED
- Training doesn't do anything to
(} 25 personality embedded in their characteristics.
l Heritage Reporting Corporation (202) 628-4888
422 0
1 MR. DeBOR: You're asking have they been given-2 psychological tests that --
3 MR. REED: Personality evaluation and aptitude 4 tests.
5 MR. DeBOR: That's an area that I haven't worked in.
6 We're reviewing now, as part of another project, the adequacy 7 of the emergency operating procedures at a number of plants.
8 That gets into nome of the training areas.
9 MR. REED: I might point out in this comparison that 10 I know the Oconee people were aptitude tested and personality 11 evaluated.
12 MR. DeBOR: I'm not familiar if there is a specific
() 13 requirement to do that.
14 MR. REED: You're being made the goat for a l
15 longstanding argument. (Laughter) 16 MR. DeBOR: It is an important issue, and certainly 17 in Navy operations those tests are there and have been since 18 the onset of the submarine nuclear program, but I'm not 19 familiar with the status of those on the land-based ones.
i 20 MS. RAMEY-SMITH: Anne Ramey-Smith. I'd just want 21 to make a general point here. The focus on the part of staff 22 in this regard, you may be well aware, is rather than 23 addressing the aptitude, personality testing, and such as l 24 that, to rather go into the field of human factors where you 25 have proven means of reducing stress and improving O
Heritage Reporting Corporation (202) 628-4888
423 0
1 performance. So rather than before the fact determining how 2 someone will react in a situation, rather we try to present 3 each one of the operators or maintenance personnel or whoever 4 it may be in a plant with a situation that has been proven 5 empirically to reduce stress and improve performance. So 6 therefore we come up with guidelines for improving and 7 upgrading emergency operating procedures, providing a physical 8 environment in the control room that is conducive to improved 9 performance, improving human engineering layout of the control 10 room so it doesn't all look stereotyped, and things of this-11 nature. Just a general comment.
12 MR. RUTHERFORD: One other general comment here, I
() 13 think we all can talk about "the operators" and what we're 14 really talking about here is a team of people that work 15 together to solve an event versus this one lone individual who 16 is hung out to dry, so to speak.
17 MR. DeBOR: If there are no further questions for 18 me, Chairman Wylie and Gentlemen, thank you very much.
19 MR. WYLIE: Is the owner's group going to address 20 this?
21 MR. SKILLMAN: I would like to in the next few 22 minutes point out the owners' point of view. In the area of 23 SPIP review of B&W owners and the staff were in substantial 24 agreement with one another. There were two areas where we had 25 some disagreement. We had some disagreement regarding the Heritage Reporting Corporation (202) 628-4888
rg 424 C
1 process, ICS/NNI review, and the content in terms of the human 2 factors area.
3 I'd like to remind myself of what the purpose for 4 SPIP was. The purpose for SPIP was to assess the risk 5 significance of the complex transients and define actions 6 which would reduce the frequency of trips and reduce or 7 eliminate complex transients altogether. I explained the 8 system by which we graded those transients. In other words, 9 we have a thick magnifying glans to sort out normal 10 performance from unacceptable performance. The unacceptable 11 performance is the C behavior trip, the significant B's are 12 also not wantod. In any case, we want to get rid of or 13 eliminate complex transients, and we want to knock out the 14 trip frequency. We see the plant in the early minutes ,
15 following complex transient response clearly at risk, 16 particularly if it's an overheating event. And because in one i 17 trip out of five we have a complex response, significant B or 18 C, knocking out the trip frequency helps in terms of total 19 risk reduction. ,
20 We accept and acknowledge that human actions can .
21 affect, and I would add significantly affect, transient j 22 behavior because of the interactive exchange that the operator 23 erring, making a transient perhaps more complex, ending up 24 with still a different situation to deal with.
l
[} 25 We did consider human factors concerns obviously Heritage Reporting Corporation (202) 628-4888
425 1 not to the satisfaction of the staff and extent, but we did 2 consider it. But they were not a dominant consideration. We 3 were going after overall plant behavior. We make no apology 4 for that.
5 We believe that our actions have or will consider 6 human factors concerns appropriately, and I'll get to that in 7 a minute. Review of the staff report, SER December '87, 8 chapter 7, identifies three basic concerns, and I'd like to 9 talk about those.
10 These are the concerns in the safety evaluation.
11 The review of human factor activities on loss of ICS/NNI 12 power. We acknowledge that that is significant.
13 The staff would like for us to have used human 14 factor expertise in the operations and maintenance of 15 personnel interviews. They would like us to have used human 16 factors expertise in review of the old TAF reports. They 17 would like for us to have used human factor expertise in 18 review of the 1985 Davis-Besse event. And they would like for 19 us to have used human factor expertise in the review of 4
20 operator burden.
21 A third comment, there should be human factors 22 involvement in the implementation stage of the 23 recommendations implemented by the B&W owners, j 24 Loss of ICS/NNI power is a major event for us. It
(} 25 has been in the past. Four of the thirteen Category C's have Heritage Reporting Corporation (202) 628-4888
426 O
1 involved loss of ICS/NNI power. Clearly the operator 2 significantly pressed a loss of ICS/NNI power. Our position 3 is we have already taken actions to address that. - Probably 4 the most important xecommendation we've made is the known 5 safety state recommendation which has as its functional intent 6 settling the plant promptly following loss of ICS/NNI power, 7 but as important, knock off that event in the first place to
~
8 prevent the ICS from losing its power sources.
9 Longer term actions are being considered with regard 10 to ICS/NNI in accordance with the advanced control systems 11 project that Larry Stolter pointed to yesterday. So our 12 response with regard to the recommendation one of the staff to O 12 de somethine about human factors 1CS/NN1, we de11 eve that we 14 have done that. We believe that we have in fact considered 15 the human factors portion of this as the most important. That 16 is getting the plant to where the operator can control it '
17 promptly, but we did not use human factors expertise to get 18 there. We used the review by our own people in the plants, 19 our own operators, our own ICS/NNI people to steer us in this
, 20 direction.
l 21 With regard to the second grouping of concerns 22 expressed by the staff: operations and maintenance interview
! 23 data. The staff says we should have used human factors l
j 24 expertise to re-review, I want to punctuate, re-review the r 25 operations and maintenance personnel interview data. We've O
Heritage Reporting Corporation (202) 628-4888 l
L
427 l 1 identified six concerns in the operations and maintenance 2 personnel interview data. Those are being concerned. All the 3 owners group members are supporting INPCs HPES or its 4 equivalent. Our position is that re-review of the operations 5 and maintenance personnel data which now is two years old 6 would just not be of significant additional benefit. We have 7 already gleaned from the operations and maintenance personnel 8 f,4terviews that information which has been evasive, exists for 9 recommendations, and we think it's appropriate to move on and 10 simply not review that data.
11 The second piece of recommendation two regarding 12 human factors expertise in the old TAF data reports, our 13 position is that the TAF data reports have been immensely 14 helpful in SPIP. They've given us the data base from which to j 15 steer into the areas that have been of significant plant
)
16 performance and concern. Up until this time, collecting human 17 factors information has not been part of the TAF activities.
18 We're developing guidelines to weave into future TAF 4
19 activities. Items pertaining to human factors.
l 20 Regarding the Davis-Besse event. There were certain 21 generic concerns, human factor concerns from the Davis-Besse 22 event reviemsd by each utility, i.e. Vital equipment 4
- 23 accessibility, local versus remote control, clarity of 24 instructions to take drastic actions. We've touched on a l 25 whole host of these already, but as in so much of SPIP, it's Heritage Reporting Corporation j (202) 628-4888 l
t
428 1 invisible. Human factors things are there. If you go back to 2 the recommendations regarding main feedwater, the bulk of 3 these have to do with elimination of items that are truly 4 operator burdensome. Particularly in the secondary plant 5 relief system. Getting back into the control room, the 6 controls that are approximate to the position of the operators 7 so they can regain manual control post trip.
8 We'd offer to you that for lack of any other name, 9 these are truly human factors items, although we didn't use 10 human factors expertise to get there. If I can argue just 11 instantly with one comment, and it's not a big issue. Mr.
12 DeBor said it's unfortunate that the people who run these
' - My counter to that 13 plants weren't involved in these reviews.
14 is the very people who run the plants, the operators and the 15 people in the control room, our operations and maintenance 16 personnel from each of the plants, were the ones who built the 17 recommendations in the first place. They were the ones who 18 were saying we have this problem, fix this problem for us.
l 19 that's what SPIP is doing. It's addressing those problems of 20 those people who said, for instance at Rancho Seco, "I 21 couldn't control that. I had to go outside in the freezing 22 cold winter air to try to fix that thing," and bango, one man
! 23 collapsed. Part of these recommendations have to do with that l
l 24 incident.
25 So our response is, yes sir, we did not use human
}
Heritage Reporting Corporation (202) 628-4888
429 1 factors professionals. But we used people who understood the 2 problem and who understood what it is like, the operator 3 burden to the point of not being able to manage the post trip 4 plant behavior. We're fixing that.
5 Operator burden data. A very important point we 6 like to make. We did in fact use a human factors expert to 7 develop the methodology for the operator burden project. If 8 you recall yesterday when I talked about operator burden, the 9 point that I made is we did not do a relative burden 10 comparison. We did not say that a Babcock plant with this 11 upset, how many galleries of operator energy go into worrying 12 about the response and taking strong action in the control 13 room of a Babcock plant versus a Combustion plant versus a 14 Westinghouse plant? We didn't do that. We said is the 15 operator able to handle this plant? Is the team that's in the 16 control room able to steer through this event?
17 We did six Category C events and laid them on each 18 control room in the Babcock plants, all six plant sites. We 19 defined actions where the operators perceived that they had 20 problems, and we are pursuing those. The types of things that 21 come out are control of steam and feed flow, the drastic 22 actions issue, overcooling mitigation, concerns regarding 23 instrument air, actions to be taken outside the control room 24 thus leading to bringing controls back to the proximity of the 25 operator, enunciators, how many there are, how much
[}
Heritage Reporting Corporation (202) 628-4888
(~h 430
()
1 information comes in promptly, how much can be seen, how the 2 information is segregated, and what's common? And the 3 emergency plan requirements, notifications and things of that 4 nature.
5 Our belief is that we used human factors expertise 6 right up front on this. We can discuss that with the staff 7 later. We sense that this may not have been recognized.
8 Finally, would re-review of the operator burden 9 information be beneficial or sensitive? Not significantly so.
10 Clearly there might be some small benefit, but we think in the 11 overall scheme of things there is no real substantial benefit 12 to redoing the operator burden piece with human factors 13 professionals. That doesn't suggest that professionals 14 couldn't bring something to it. Wu just think there is not 15 enough additional benefit to warrant it.
16 Finally, human factors involvement, human factors 17 expertise should be involved in recommendation / implementation 18 in each utility. It may be used on a plant-specific basis, 19 and would be used depending on the subject of the 20 recommendation. If the recommendation is a cut and dried, 21 hardware issue, we certainly won't use a human factors expert.
22 But if whatever it is that we're getting into has to do with 23 the control room, has to do with what the operator is going to 24 do versus what he otherwise would have done, involves emergency procedures or re-orienting procedures and their
{} 25 i
Heritage Reporting Corporation (202) 628-4888
O 1 connection to the behavior of the plant, we might. But that's 2 at the call of the utility, if they judge that appropriate.
'3 In closing, we concur that human factors 4 considerations are very important, but we would also bring to 5 your attention, to the staff's attention, the way in which the 6 recommendations knit with each other, if you will, relate with 7 each other, and relate to the operator and to the control 8 room, we feel as if we really put our finger upon this human 9 factors area. We're placing more emphasis on human factors 10 now than when SPIP began. We believe that there is little 11 benefit to re-review of old information. And clearly, human 12 factors, particularly in the operator burden area where we 13 think the greatest benefit cones from, human factors 14 involvement, human factors was involved, was a major part of 15 the safety performance program.
16 That's all I have to say.
17 MR. WYLIE: Any questions?
18 MR. MICHELSON: Can I ask a question of the staff on 19 this subject? The staff seemed to have made a pretty strong 20 point of the importance of human factor involvement in the 21 analysis of these situations and corrective action. I was 22 just wondering, the agency's principal evaluator of such 23 experiences is AEOD. Do they have on their staff so-called 24 human factors professionals? Do you know?
Yes. The answer to that, from my
[} 25 MS. RAMEY-SMITH:
Heritage Reporting Corporation (202) 628-4888
432 O
1 perspective, is that no they don't, and yes they should. Just 2 as other parts of the agency also should have people in 3 operating events, evaluations, and so on.
4 MR. MICHELSON: So really we should be pushing this 5 at home as strongly as we seem to be pushing it outside the 6 agency if we really believe it.
7 MS. RAMEY-SMITH: And that's what we're trying to 8 do.
9 MR. MICHELSON: I didn't think they had some 10 experts, but I'm not always up to date on that.
11 MS. RAMEY-SMITH: As a matter of fact, just recently 12 I was told that the office director for AEOD has become very 13 interested in the whole area of human factors and intends to, 14 I don't know whether increase his staff, but certainly to 15 increase his program to better look at human performance 16 aspects of operators.
I 17 MR. MICHELSON: Is it going to do you much good to l 18 operate in a branch dedicated to human factors? That's not l 19 where the action really is. It's the same way with B&W. I'm i
20 not advocating it one way or the other, but I was kind of 21 curious to see if the agency was following through on what it 22 seems to strongly believe in.
23 MR. WYLIE: Mr. Hammer?
24 MR. HAMMER: Again, my name is Gary Hammer. I'm 25 from NRR. I looked at some of the valve reliability issues of
(}
I Heritago Reporting Corporation (202) 628-4888
433 O
1 this review as sort of a sub-task to Bill LeFave's system 2 review. This is the only slide I have. It's sort of a 3 synopsis of the different areas we looked at that Mr. Skillman 4 also talked about a little bit.
5 Main steam safety valves, multi-operated valves, 6 power operated relief valves and block valves, and check 7 valves. On main steam safety valves, there were several areas 8 of improvement that the B&W owners group came up with. They 9 involved these items: Improving performance of the valves for 10 better setpoint procedures, ring setting adjustment 11 procedures, and better maintenance. All of those areas are 12 something that the staff is interested in in a generic sense,
() 13 a little broader than just B&W concerns.
14 I guess the emphasis for the B&W plants on improving 15 valve reliability for the main steam safety valves was due to 16 the fact that there's a more frequent and greater challenge of 17 those valves.
18 Some of the things they looked at were to try to 19 reduce the challenges that Bill LeFave talked about. One of 20 them was to possibly increase the turbine bypass and 21 atmospheric vent valve performance either by making the valves 22 bigger or improving the control systems associated with them.
23 Along with that they also looked at reducing the 24 challenges by increasing the so-to-speak simmer margin of the 25 valves, by raising the MSSV setpoints. This was not l
Heritage Reporting Corporation (202) 628-4888 I
434 1 encouraged because it would not only violate the ASME 2 requirements of having the safety valves set at the design 3 pressure, and what this would effectively do is increase the 4 design pressure in the system. It would also reduce by some 5 amount the intended system margin, the stress margin. We 6 didn't want to see that as a very active pursuit in 7 accomplishing their goal of improving the safety valves.
8 This did get folded into a valve task force that's 9 to be done later. I think someone mentioned a little earlier 10 that they hope to have the report by August. We did receive a 11 commitment from them that that would be submitted to us and we
, 12 did want to look at it.
13 MR. KERR: Excuse me, would you go over again please
- 14 why it was that you did not find their proposed method of 15 approach appropriate?
16 MR. HAMMER: Concerning the third bullet, I guess, l 17 is what you're talking about. Increasing the setpoints.
18 MR. KERR: Yes.
19 MR. HAMMER: If you increase the setpoint you will 20 increase the simmer margin, so it's called, between the normal 21 operating pressure and the valve actuation setpoint by raising 22 the setpoint at a higher level. So you will reduce the number 23 of times that the valve will actuate. For some of the milder 24 transients where you would be able to take care of the old 25 pressure transients by some of the other mitigating pressures
{' }
Heritage Reporting Corporation (202) 628-4888
1 of the plant such as the turbine bypass system and some of the 2 other things.
3 The problem with that is that it, in raising the 4 pressure of the system you reduce the margin to failure of the 5 system due to over-pressure. The ASME code specifically 6 addresses that.
7 MR. KERR: It addresses what? You mean what they 8 were doing is a violation of the ASME code?
9 MR. HAMMER: By raising the setpoint of the valves 10 above the design prersure of the system, yes. That is 11 specifically addressed by ASME.
12 MR. KERR: I'm surprised that the licensees would 13 recommend something that they knew violated the ASME code.
14 MR. RUTHERFORD: Comment please. This was a 15 suggestion to look at whether we could raise the design 16 pressure. If we did that, we would have to go through the 17 appropriate calculations to show that indeed the design 18 pressure of the system could be raised, and then we could 19 change the setpoints accordingly to be in compliance with the 20 ASME code.
21 MR. KERR: You didn't understand that they were 22 going to do that?
23 MR. HAMMER: We didn't want to encourage them to do 24 it because it looked like something that would raise a whole
{} 25 lot of questions in terms of things like actual material Heritage Reporting Corporation (202) 628-4888
i k
i
(
1 properties of the system, some of the problems we've seen with l 2 erosion / corrosion in the secondary system. In other words, '
i I 3 actually knowing what the material thicknesses are. ;
L 4 MR. KERR: Did you actually do some sort of balance i i
- 5 and said the risk reduction that might be achieved by this is t 6 less than the increased risk that would be caused by raising 7 the design pressure? l 4
L 8 MR. MAMMER: No,'it was more in a general [
9 qualitative way. We felt the other things that they were [
t 10 going to do to improve main steam safety valves would probably f 11 improve the performance of the system enough without getting l
. t l 12 into this. l
() 13 MR. KERR: So you have no particular reason to make r
l 14 this decision except it's sort of a good feeling? f, l 15 MR. HAMMER: Right. !
16 MR. RUTHERFORD: We're not currently pursuing this 17 option.
l i 18 MR. HAMMER: I might say, this whole thing is still ;
i 19 sort of up in the air. They have it wrapped into this valve 7 20 task force. l l
21 MR. KERR: It puzzles me a little as to how l
> i 22 decisions are made in an area where there are obvious t l [
l 23 tradeoffs. I was really seeing if there was anything as a L 1
i 24 basis for the decision other than it looked difficult. So be
(} 25 it. [
Heritage Reporting Corporation i (202) 628-4888 l
,..mm.-,c..,.,_-.,.~.,,,__,_,-
(
)
1 MR. MICHELSON: Are they downstream or upstream of 2 the main steam --
3 MR. MAMMER: I think for all practical --
4 MR. MICHELSON: It's either up or down. It isn't in 5 between.
6 MR. RUTHERFORD: It's near by.
7 MR. MICHELSON: I sit upstream of the valve or 8 downstream of the main steam isolation?
9 MR. RUTHERFORD: It's upstream of that, I believe.
10 MR. MICHELSON: The atmospheric dumps. Are they 11 downstream of the main steam isolation valve, or upstream?
12 MR. RUTHERFORD: They would be upstream, I believe.
() 13 MR. MICHELSON: I don't know. They can be either 14 way. The safety valves are upstream, obviously, they have to 15 be.
16 Now these 16 valves you've talked about were main 17 steam safeties, don't they open only after the atmospheric 18 dumps have already opened and handled the transient?
19 MR. RUTHERFORD: That's the way they should work.
20 MR. MICHELSON: So in addition to opening all the 21 atmospheric dumps you still have to open nearly all the 22 safeties on a trip from full power?
23 MR. RUTHERFORD: It depends on whether the valves 24 are groaning open because they're full of clap trap, or 25 whether they're operating properly.
Heritage Reporting Corporation (202) 628-4888 i
.I l
438 O
1 MR. MICHELSON: But even if the atmospheric dumps !
2 are handling the transient you still have, even though they're 3 there and they all open, you still have to open a bunch of 4 safeties. I used to think that the atmospheric dumps were 5 supposed to preclude the opening of the safeties, but I think 6 on Westinghouse and CE that is the case, I thought.
I 7 Now what this is leading to is another question.
8 Are you allowed the same primary side activity level as other 9 PWR, Westinghouse and CE?
10 MR. RUTHERFORD: Yes.
j 11 MR. MICHELSON: And you're allowed the same amount 12 of steam generator leakage and still operate?
13 MR. RUTHERFORD: Yes.
14 MR. MICHELSON: But yet your leakage may all be up 15 above the water line which in the case of other plants it's 16 always below the water line, and you're just going to have 17 quite a bit more release when you trip from full power because 18 you're going to release a lot of steam to atmosphere on a full 19 power trip. If you're allowed to operate at the same activity 20 level and the same leakage rates. I hadn't thought about 21 that, but that's kind of interesting.
- 22 MR. REED
- Carl is encroaching upon my questions 23 that were deferred. Is this now the opportune time for me to 24 ask my questions even though the speaker is a staff speaker l [} 25 rather than B&W? If so, I will proceed with my main steam i
! Heritage Reporting Corporation (202) 628-4888
O
~J 439 1 safety valve questions and observations.
2 I started out by trying to say that main steam 3 safety valves have been around a long time, spring loaded 4 type, and they've been worried about for a long time, and 5 they've been tuned for a long time, and they've been designed 6 for a long time. A task force on main steam safety valves, if 7 it gets into a closet of the nuclear closet, which many times 8 this has happened in the past, that people have designed it in 9 the nuclear c2oset without reviewing total history, they will 10 probably not have much success. And even if they were the 11 smartest task force in the world and they went outside the
- 12 closet and throughout the industry, I'm not sure that in the 13 final analysis they will make much headway in improving main 14 steam safety valves. End of statement on that.
15 Now, B&W in yesterday's presentations, said that
. 16 inventory wasn't a big problem or a big disadvantage by 17 comparing inventory to the other reactors. Now I'd like to 18 point out that if in fact these safety valves lift, as he 19 said, every higher power level trip, then what's causing them 20 to lift? Is it delayed tripping of the reactor? Is it the l
21 inventory? I sort of suspect it's the inventory, and that 22 claim that non-advantage, no disadvantage is probably not 23 realistic, i
24 Now to get to something Carl was starting to open,
(} 25 there are radioactivity inventories in the secondary side of Heritage Reporting Corporation (202) 628-4888
440 4
0 1 the steam generators. If the release of this radioactivity, 2 however small it may be, it doesn't concern me but it would 3 concern the regulatory staff, if the release is by main steam i 4 safety valves, it will be labeled at uncontrolled and 5 unmonitored, probably, release, and unquantified release.
- 6 so therein, with these valves opening every trip you 7 have a very serious regulatory issue with respect to 8 quantifying and accounting for the radioactivity. Quite 9 frankly, I think it will keep a couple of people busy.
10 MR. HAMMER: Let's see. The next item is the motor l
11 operated valves and other power operated valves. There were 12 several recommendations made by the owners group which 13 enhanced, which if implemented would enhance the performance
(])
14 of motor operated valves and power operated valves. The key I 15 thing there was that they were looking at including all of 4
16 their safety-related valves in a program that would be l 17 consistent with Bulletin 85-03 which for those of you not l 18 familiar with Bulletin 85-03, it required diagnostic type 19 testing and limited torque switch setting improvements in l 20 order to handle high differential pressure loads and other j 21 type loads to make sure that they would operate properly.
22 That bulletin only addressed three systems, though, which were 23 the EFW, and ATW turbine supply, and high pressure injection.
24 What they're recommending here is to include all the
! 25 other safety-related valves in such a program. That's also
}
l Heritage Reporting Corporation (202) 628-4888 1
_._.,___._m. . . _ _ _ . _ _ . _ _ _ _ , , _ , _ . , , _ . . _ _ _ . . , . _.
q V
441 1 getting folded into the valve task force so we'll be seeing
! 2 more about that a little later.
3 The next item is power operated relief valves and 4 block valves. They're recommending an increased maintenance 5 and testing in order to improve their reliability with some 1
6 recommendations. That's going to be consistent with the draft 7 standard OM-13 which really isn't officially on the street 8 yet, but which we, the people involved in that including 9 myself on the staff, are in agreement with in terms of that 10 being a good document.
11 MR. DAVIS: Excuse me, I have a question on that.
12 Do you know if the plants operate with their block valves open O 13 or closed normally?
14 MR. HAMMER: I think you find a mix on that. If you
! 15 look at the tech specs for most plants, they're able to 16 operate with the block valves closed if they remove power from 17 it. That keeps them from exceeding leakage limits.
18 MR. DAVIS: That would affect the reliability of 19 bleed and feed?
! 20 MR. HAMMER: Yes. I should say, GI-70, generic 21 issue 70 which is being pursued in the office of research, is
, 22 looking at reliability of power oparated relief valves and 23 block valves. Improving the technical specifications is one 24 of the things they're looking at in that, and they don't have
(} 25 a complete resolution to that yet, but there are things that Heritage Reporting Corporation (202) 628-4888
(' 442 V) 1 should be done to improve it.
2 MR. DAVIS: Thank you.
3 MR. HAMMER: The last items is check valves. As 4 many of you are aware, the industry is experiencing a number i
5 of check valve problems not only with B&W reactors. There is 6 an industry initiative going on to improve check valve 7 reliability. There was an INPO SOER issued which contains 8 some guidelines, that was number 86-03, which addressed 9 design, operation, and maintenance of the check valves. In 10 terms of improving the stability, wearing characteristics of 11 the check valves, ensuring that they're tested properly and 12 things like that. That's also been folded into the valve task 13 force. If those things are implemented, that would be a l
l 14 significant improvement of the check valves.
15 MR. DAVIS: Excuse me. Are these check valve 16 problems that you referred to leakage, back leakage, fail to 17 seat, or stuck open?
18 MR. HAMMER: It's been a variety of things. Most l 19 notably it came about, one of the big events was not at the 20 B&W plant but at San Onofre 1. They had a sovere water hammer 21 due to failure of many check valves, including high pressure 22 and low pressure interfacing system. It was a very serious
(
j 23 event. But there have been several check valve problems in 24 the industry.
(} 25 MR. JONES: This is Bob Jones of the staff. Two Heritage Reporting Corporation (202) 628-4888
443 O
1 comments were made I'd like to address. Dr. Kerr, on how the 2 staff based their decisions, we said we didn't encourage 3 pursuit of this modification to raise main steam safety 4 valves. What we said was pursue the other options first. ;
5 That by changing the design pressure of the system, possible 6 reliance on actual thicknesses of the pipes, etcetera in order 7 to demonstrate code compliance we didn't think that was the 8 beset way to go. We were saying please don't pursue that 9 first. We'd rather you go look at other ways of improving the 10 overall system performances to make sure the main part of the 11 system is functioning properly before you start eating into 12 design. So it wasn't just quite an arbitrary statement or if O I 13 you wish, decision on our part.
14 While we didn't do a risk study, what we really aid 15 was look at the overall system as a whole and say there are 16 other options available which we think are better to pursue 17 first, and thus we did not want to encourage this at the front 18 end of putting together a valve task force.
19 MR. KERR: Very good. I'm reassured.
l 20 MR. JONES: Mr .. Davis, the comment on reliability of l
21 feed and bleed with respect to use of the PORV, for the l 22 majority of the B&W plants, they do not need to rely on the 23 PORV for feed and bleed. They can rely on just use of the l
24 safety valves. Davis-Besse is the exception on that, however.
1
(} 25 MR. MICHELSON: But they have to stay at full Heritage Reporting Corporation (202) 628-4888
, w- w
>~ - , - - , - _
444 (3
V 1- pressure to use the safety valves.
2 MR. JONES: Yes, they would. And I'm sure in a pure 3 risk sense there would be some impact. But in the initial 4 core-cooled impact, whether or not the valve was in fact 5 reliable or not or whether you had the block valve closed, 6 which I take is your question, would probably have a small 7 impact on the overall reliability.
8 MR. DAVIS: You're saying the charging point shut 9 off head is above the safety relief point?
10 MR. JONES: Yes. On most of the B&W plants they're 11 around 2700 to 2900 PSI.
12 MR. MICHELSON: The picture gets pretty complicated,
() 13 though on the stea' .ibe rupture case where you really want to 14 start getting the pn.ssure down and you may have to go to feed 15 and bleed if the natural circulation isn't doing the trick or
- 16 whatever. Then I don't know quite how you handle that.
l 17 MR. JONES: On the multiple-tube rupture scenario 18 and losses of off-site power and those types of things you may 19 end up using the PORV's under those circumstances..
l 20 MR. REED: Those pressurizer safety valves, are they l 21 loop-sealed safety valves, or are they spring loaded safety 22 valves without loop seals? Just what are those?
23 MR. JONES: Those are spring loaded without loop 24 seals.
25 MR. REED: How would you expect their reliability to Heritage Reporting Corporation (202) 628-4888
445 1 be with respect to the reliability of the main steam valves?
2 Similar?
3 MR. HAMMER: I can address that. There's a separate 4 issue that the staff pursued for the pressurizer safety 5 valves. ~ That's a TMI action. In that, there was a large 6 generic test program to show that the valve discharged 7 correctly, the correct capacity. And the adjusting rings were 8 set properly and things like that.
9 MR. REED: I might point out that for some time I've 10 been concerned by the claims of all the manufacturers and 11 owners that bleed and feed cooling is a really viable way to 12 do it, or is it viable? I'd like to hope that it is, but I do O 13 not believe there has been rigorous evaluation and testing of 15 bleed and feed cooling, and I would expect, my guess would be, 15 that it hasn't been done for PORV's rigorously, and I've asked 16 many times the staff for a test program. It hasn't been done 17 for PORV's rigorously, and it hasn't been done for spring 18 loaded safeties which I think will be more of a problem, much 19 more of a problem, than certain kinds of PORV's.
20 MR. JONES: Let me make one comment with respect to 21 the reliability of feed and bleed. We do not credit feed and 22 bleed in demonstrating compliance with the regulations. We
! 23 view feed and bleed for something that is useful in emergency I
24 procedures, in the event that you should have a loss of all 25 feedwater, to provide a means to keep the core cool while he t Heritage Reporting Corporation l
(202) 628-4888 I + .- -
446 b-v 1 attempts to recover his feedwater system, and get back on the 2 normal cooling mode. We do, however, give credit for it in 3 the risk space, in trying to determine whether or not the fact 4 that certain other safety issues, we do do it there, certain 5 safety fixes. But from a regulatory standpoint there is no 6 credit given to feed and bleed. We looked at feed and bleed 7 in the context of this overall assessment because it had a 8 bearing on the overall risk of the plant. But we do not give 9 credit for the regulatory sense.
10 As Bill LeFave mentioned earlier today, in looking 11 at things like feedwater reliabilities and how to resolve the 12 issue of reliability of two pump AFW plants, we're not giving 13 credit for feed and bleed. We are giving credit for standby 14 pumps which will be useful to get water back to the steam 15 generator, but as a means for decay heat removal for that 16 issue, we're not giving credit for it.
17 MR. REED: It's very unfortunate that you aren't, 18 and it's very unfortunate that we do not move forward to 19 rigorously evaluate the merits of bleed and feed or primary 20 blow down or whatever you want to call it. That's very very 21 unfortunate.
22 MR. WARD: As the years roll on and we keep hearing j 23 the same sort of thing, I'm getting less and less impressed by 24 the distinction between regulatory space and risk space. I 25 don't care two cents for regulations unless they control risk Heritage Reporting Corporation (202) 628-4888
q 447 V
1 in some way. That's really what we're interested in. I agree 2 bureaucratically there's a distinction, but I don't think this 3 committee is all that impressed with --
4 MR. JONES: What I wanted to make of one, there 5 wasn't an impression left that we are, in evaluating the 6 safety of these plants in response to transients, that we're 7 crediting feed and bleed as the fantasy or the cave-all for 8 these plants, because we're not. What we're crediting is 9 reliable feedwater systems. We recognize that a means of 10 decay heat removal is there. What we want to assure is that 11 the feedwater systems are reliable and they take the actions 12 there as a very specific example. I understand the comment
( 13 that it is frustrating at times when you run feed and bleed i 14 into the risk studies, you do run some problems in back 15 fitting.
16 MR. REED: Do you realize you just shot down another 17 bullet of yesterday that B&W, where they claimed that 18 superiority or equivalency because they had a better bleed and 19 feed system and were not relying on the emergency feedwater
- 20 system?
21 MR. JONES: I don't remember the specific bullet 22 they had, but from the sensitivity evaluation which is the 23 area it came out of, what they said is they were equivalent, 24 or at least what the MPR study came up with was they are f
25 equivalent in terms of time available to initiate an Heritage Reporting Corporation l (202) 628-4888 1
rm 448 U I 1 alternative means of heat removal in the event that they lost 2 all feedwater. Or at least that's what they should have been 3 saying. I could find the right quote in here and know what it 4 is. We evaluated that and we generally agree, that is a true 5 statement. If you've lost all feedwater on any PWR where you 6 go to is feed and bleed ultimately. Under those 7 circumstances, B&W has approximately 40 minutes of combustion 8 engineering plants with the low head HPI's on the order of 20 9 and 25 minutes, and the Westinghouse plants are in the 100 10 minute time frame. That's a wide disparity, but it's at least 11 in the middle. Certainly they do not stand out in the event 12 of a loss of all feedwater event, that there is something from q.
k' 13 a time standpoint they have to take drastic action.
14 MR. RUTHERFORD: That's a correct summary. We did 15 not state that we don't depend on emergency feedwater. What 16 we said was that it is a backup in the case where you do lose 17 all sources of feedwater.
18 MR. MICHELSON: Mr. Chairman, I'd like to ask an 19 unrelated question to this, but related to valves.
20 MR. WYLIE: Go ahead.
21 MR. MICHELSON: I'm wondering if B&W could give us 22 maybe about a one minute explanation on why so many safeties 23 seem to open on B&W plants but do not have to open on other 24 types of PWR's? What's unique about B&W that so many safeties
( 25 open on a full power trip?
Heritage Reporting Corporation (202) 628-4888
449 1 MR. RUTHERFORD: It's the inventory advantage.
2 MR. MICHELSON: It's not intuitively clear to me why 3 that-should be the case.
4- MR. KANE: My name is Ed Kane from B&W. The 5 question, as I understand it, is why do safety valves open on 6 B&W plants versus not on other plants. There are several 7 differences. I'm going to have to speak from my assumption on 8 operations of certain parts of the Westinghouse plants or the 9 non-B&W plants, and you'll have to forgive me there.
10 The recirculating steam generators that have full 11 power operate at a larger temperature difference between the 12 normal operating set point or pressure difference in the
.(}
13 secondary side, between the pressure at the 100 percent power 14 on a recirculating steam generator, and the main steam safety 15 valve set point. That's correct, right, on the Westinghouse 16 plants? That pressure difference is on the order of 200-300 17 pounds if I'm not mistaken.
18 MR. MICHELSON: The main steam pressure is about 300 19 pounds higher?
20 MR. KANE: Their pressure drops as they go up. Ours 21 are constant.
22 MR. MICHELSON: Super heat effects?
23 MR. KANE: We operate at a constant 900 pounds 24 pressure in the steam generator over the total load range,
(} 25 where there's drops as they go up a pound.
Heritage Reporting Corporation (202) 628-4888
450 0 1 MR. MICHELSON: That's part of it.
2 MR. KANE: I'm going to continue on. The early 3 designs of the B&W plants operate in the range of, with 4 turbine bypass valve and atmospheric dump valve capacity 5 ranging from somewhat over 15 percent to around 30 percent.
6 Most of the non-B&W plants operate significantly greater than 7 that. The TVA plant, I understand, is about 80 percent 8 turbine bypass atmospheric dump capacity.
9 MR. MICHELSON: That's like the -- turbine bypass.
10 MR. KANE: This is in relation to the question of 11 why there are more. So the TVA plant, for example, the later 12 generation of the B&W plants, would not get the large number
() 13 of reliefs, in fact the BBR plant in Germany, my understanding l 14 is they do not, the main steam safety valves. So you've got a 15 much larger capacity to turn it over even on a later 16 generation B&W plant, so you do not get the lifting of the 17 safety valves.
! 18 The second reason, as I pointed out earlier, is the 19 larger pressure difference between the 100 percent operating 20 set point and the lift point. I think it's about, my numbers 21 are around 250 to 275 pounds on a typical Westinghouse plant; 22 125 on a typical B&W plant. So you have larger capacity for 23 turning around quicker and a larger differential.
24 MR. MICHELSON: You're running a lot closer to your 25 set points in other words.
Heritage Reporting Corporation
( 2 0." i 628-4888
g 451 V
1 MR. KANE: Just as a point of reference, that was 2 one of the reasons why it was suggested we look at potentially 3 increasing the setpoint. We're also looking at other 4 alternatives as pointed out to not do that, but to keep 5 from --
6 MR. MICHELSON: Are there any other significant 7 differences?
8 MR. KANE: Those are the ones that come to mind l
9 right away.
10 MR. MICHELSON: Those all make sense. I understand.
11 MR. REED: I am concerned that you do not mention 12 the secondary site inventory.
l
() 13 MR. KANE: That shouldn't be a problem. That does I
14 not really have a significant impact on the results. As I 15 indicated, our BBR plant which has 100 percent capacity does 16 not let the safety valves, on a normal reactor trip.
l 17 Obviously if you have other things going on you may cause them 18 to let, but the inventories per se on a normal reactor trip, 19 as long as you keep inventory in the generators and you can 20 remove the heat through another means, steam path to the 21 condenser or whatever, will turn it around.
22 MR. REED: Well it's the first time I've heard that 23 inventory doesn't have anything to do with cushioning of 24 transients.
25 MR. KANE: What I'm saying is they're sufficient O
Heritage Reporting Corporation (202) 628-4888
~
452 1 there that if you have other means of getting the steam out of 2 the system, given the amount of heat there, it would do that.
3 MR. MICHELSON: You've got so many BTU's you've got 4 to move out somehow.
5 MR. SKILLMAN: In response to Mr. Reed's comment, 6 the inventory question is important because if they stick open 7 -- safety valve on an RSG type plant, you have immensely more 8 inventory to pump into the environment until you finally 9 settle the plant. In that regard the smaller inventory of the 10 Babcock design --
11 MR. REED: Will give you a faster dry out.
i 12 MR. SKILLMAN: Is a benefit.
13 MR. WARD: I think we should have Westinghouse l 14 comparing pros and cons and benefits.
15 MR. KERR: I'm about to become confused. We need to 16 get rid of these PWR's.
17 MR. REED: If we could ever get a standardized PWR, 18 it is the machine that will carry the world for many years.
l 19 MR. WYLIE: Are there any more questions for the l
l 20 staff on valves?
l 21 If not, let's move on. Let's go into risk 22 assessment.
23 MR. SKILLMAN: In February-March 1986 as our 24 original program was being described and as we were working 25 with the staff regarding the program, the owners did not have,
[}
Heritage Reporting Corporation (202) 628-4888
453 O
1 the B&W owners did not have a risk assessment task. The staff 2 encouraged us to include a risk assessment view. To their 3 credit, we did include a review. It was beneficial for us, 4 but we need to give credit to the staff for encouraging us to 5 pick up this task.
6 In 20 words or less what we did is two phases of PAR 7 review. First, using the Oconee level three, which considered 8 internal and exter.nal events, the Oconee PRA and the Crystal 9 River PRA which is the level one that included loss of off-10 site power. Using the skeletons of those two PRA's we took 11 the Category C events, 13 in number. That included the '77 12 loss of feedwater at Toledo ?dison; '78 light bulb at Rancho l
() 13 Seco; '79 TMI-II; and 10 Category C transients beginning in 14 1980. We evaluated, that is reviewed those 13 events against 15 the two fault trees derived from the Oconee and the Davis-16 Besse PRA. What was our goal? Our goal was to assess the 17 importance of the Category C transients.
18 Let me say that differently. As we entered this 19 task the question we were posing to ourselves was besides that 20 which we had seen in 13 Category C events, are there other 21 outlying events likely to cause a significant transient such 22 as Tuleso Cordenne. Is there a boogey man out there? Is 23 there something we haven't considered that we should be l 24 considering?
l gs 25 MR. MICHELSON: But you did not consider external Heritage Reporting Corporation (202) 628-4888
454 s
1 events in that sense?
2 MR. SKILLMAN: The reason we did not consider 3 external events in that sense is because we were limiting our 4 review to the complex transients that had occurred.
5 JICHELSON: I just wanted to make sure you 6 hadn't -t gears.
7 . SKILLMAN: Bear in mind, a complex transient is 8 one that I.. quires significant safety system operation, or 9 really involves operator burden. A lot of operator 10 responsible mitigation.
11 So we compared the initiating of that frequency 12 obtained from the transient history, Babcock units, that is 13 those 13 Category C's. We evaluated the dollar and accident l 14 sequence systems and initiators, and we compared those to the 15 Category C events, and we generalized the results.
16 The actual B&W trends in history is adequately 17 represented in both of the PRA's. That is the 13 Category C's I
l 18 we had experienced are represented in the Crystal River and 19 the Oconee PRA. The corollary is true. Thirteen Category C l
20 events do not lie outside of those two PRA's.
! 21 Now phase two of this activity, and we paid SAIC as l
22 a subcontractor to do this work for us, phase one was to l 23 review Oconee and Crystal River PRA, look at the 13 C's and 24 make sure they fit on those event charts, if not give a 1
! 25 signal they were found to fit. Then phase two, customize the V(~%
Heritage Reporting Corporation (202) 628-4888
455 O
v 1 event trees so we could review the other plants, so we could 2 have covered the spectrum of the Babcock and Wilcox Owners 3 Group plants.
4 Crystal River III and Oconee PRA's indicated minimal 5 core damage to the category C's of those units. Feed and 6 bleed cooling is an important accident mitigator for Babcock 7 units. Service water and AC power, relatively risk important 8 to both accident initiation and mitigation. Generalization of 9 PRA results indicate Category C events not likely to be 10 considered dominant cc ntributors to core damage risk for most 11 Babcock plants.
12 Davis-Besse and TMI slightly different. Davis-Besse
() 13 because of their unique requirements for HPI cooling because 14 they have the split system with the smaller HPI pumps; and 15 TMI-I due to differences in the PRA analysis.
16 Finally, while We intend to reduce the frequency of 17 Category C events or attempt to eliminate them all together, 18 our review, our results of this review is,that Category C
, 19 events are causing a greater core melt damage concern than 20 these events truly warrant, j
21 That's all I really intended to say. I'll be glad 22 to answer questions.
23 MR. MICHELSON: I believe that's a very significant l
24 slide, of course, because it says that we're really spending !
25 an awful lot of time worrying about a fairly modest O
Heritage Reporting Corporation (202) 628-4888
i 456 p)
I contributor to risk, or very minor, perhaps, contributor to 2 risk, and that the main reason you're devoting this effort is 3 because of reliability, economic considerations, etcetera. Is 4 that a reasonable conclusion or an unreasonable one?
5 MR. RUTHERFORD: Let me respond to that. As I 6 mentioned before, it's true the focus of our program is on 7 prevention of these particular events that have occurred. But 8 if you look at the recommendations you'll see that many many 9 of them cut across all kinds of answers. When you look at 10 motor operated valve reliability, that doesn't just speak to 11 the events that we've had on the B&W plants. That speaks to 12 all plants. Doing good root cause evaluations, all incidents.
13 It cuts across all links. There are numerous things in the 14 programs and the recommendations that are going to have 15 benefit across the board, and not just specific to these 16 particular ones.
17 MR. MICHELSON: That's certainly intuitive 18 observation, but you didn't substantiate that with any kind of 19 study showing that. What are the major contributors for 20 risk, those kinds of changes would help to reduce that risk.
21 That was not apparently done.
22 MR. RUTHERFORD: Not systematically, no.
23 MR. MICHELSON: Of course in cases of external 24 events it may even have a lesser effect, we just don't know.
25 We haven't done an external event analysis.
Heritage Reporting Corporation (202) 628-4888
457
(~)
V 1 MR. RUTHERFORD: Certain of the plants have, but 2 that's the plant's decision.
3 MR. MICHELSON: While we're waiting on our next 4 speaker, maybe B&W could answer this question. Apparently you 5 found that the event trees didn't match too well with Category 6 C events that you actually experienced over the years. Is 7 that correct?
8 MR. RUTHERFORD: -- That's partly why we did our own 9 assessment.
10 MR. MICHELSON: So B&W apparently does believe that 11 you did see in the event trees the Category C events you were 12 experiencing?
13 MR. RUTHERFORD: Yes.
14 MR. MICHELSON: Okay, then you'll pick up on an 15 explanation.
16 MR. RUBIN: My name is Mark Rubin. I'm with the 17 Division of Engineering and Systems Technology.
18 Some questions have been raised earlier in your 19 meeting on the scope of the risk evaluation. I'll be 20 providing just a brief perspective on the staff's goals and 21 objectives, and then Dr. Youngblood from Brookhaven will be 22 providing details on our evaluation of the owners group risk 1
l 23 work and the staff's re-evaluation of that work.
24 I would like to comment that the owners group risk
/~T 25 evaluation was a very useful product. We did feel on V
- Heritage Reporting Corporation l
(202) 628-4888 l
\
l
458 1 evaluation that some areas need to be looked at with some more 2 vigor, and we proceeded to do so.
3 The entire reassessment program, of course, was 4 prompted by a number of operating events that were perceived 5 to be relatively severe, and in a sense in some cases near 6 misses. Yet this raised a lot of questions because the few 7 PRA's that were available on operating B&W plants did not 8 identify any substantially different risk profiles for their 9 plants versus the Westinghouse plants, for instance.
10 Because of this, and in light of the perception that 11 the events were significant, we wanted to rethink the B&W risk 12 profiles in light of the information that was coming out of 13 the operating events and was the reason for encouraging the
([ }
14 owners group working for us, conducting their own.
15 Specifically, we were trying to decide whether the 16 events that had been observed, did they raise any new issues 17 due to B&W generic design features or operational features 18 that would be different from the other PWR's, from the other 19 vendors, and consequently substantiate a difference in risk 20 profiles that had not been reported in the literature in the 21 previous studies. We were trying to find out if there was 22 something that was a very significant dispute on the prior 23 risk conclusions on the B&W plants.
24 Because of this, our focus was restricted to 25 internal events for damage frequencies, looking at the B&W Heritage Reporting Corporation (202) 628-4888
459 1 systems for sequences, for damage scenarios, that had been 2 identified as possibly impacted by B&W design differences such 3 as the RCS, feedwater, secondary site, quick response, 4 possibly operator response constraints. Consequently, we 5 looked at such things as loss of feedwater, transients, 6 overcooling transients, pressurized thermo shock 7 possibilities, small break locos to be initiated by 8 repressurization following a loss of cooling or an overcooling 9 and then bringing on HPI injections and not terminating as the 10 pressure --
11 We were looking for differences that were identified 12 by the operating events. Brookhaven, as I mentioned, will be 13 providing results of the reassessment. I want to emphasize i.
14 that we did not attempt to provide a new complete l
15 requantification of B&W plant risk profiles. As has been 16 mentioned a number of times, external events weren't looked 17 at. The whole long list of plant accident sequences were not l
18 reassessed. Only in those areas we felt there were B&W unique l
19 differences that were worth looking at.
20 That's pretty much what drove us in this study, 21 again, to find out if their previous risk profile conclusions 22 were still legitimate.
23 MR. MICHELSON: Did you look at accident situations l
24 as well to make sure there was nothing unique about B&W or did 25 you just look at these severe transients as did the owners Heritage Reporting Corporation (202) 628-4888
fx 460 U
1 group apparently. Did you go beyond that to see if any 2 accident situations, there was no reason to believe there was 3 a unique problem?
4 MR. RUBIN: Only in a limited sense.
5 MR. MICHELSON: Your statement earlier kind of led 6 me to believe that you did look to see where the unique 7 differences were. I'm not sure now, are you looking only at 8 transients, or were you looking at accidents as well when you 9 made that statement?
10 MR. RUBIN: We looked at the accidents that would be 11 impacted by B&W specific design differences. But it wasn't 12 solely an attempt to recreate the Category C events.
l 13 MR. MICHELSON: But in the SER you didn't mention i
14 that aspect of the work then, I guess. I didn't find a 15 discussion of going back and looking at accidents to see if 16 they were unique situations, such as when you get into two 17 phased flow and so forth. That wasn't discussed in the SER I
18 was it? I realize it's a lot to read and I didn't perhaps 19 read it all, but I never sensed it coming through comparing 20 accidents. you only seemed to be tracking back on what B&W 21 owners group looked at, and you're doing the verification job.
22 MR. RUBIN: No, we went quite a bit further than tho 23 owners group.
24 MR. MICHELSON: You didn't document how much further
{) 25 you went, or if you did, please tell me and I'll read it.
l Heritage Reporting Corporation (202) 628-4888 l
461 1 MR. RUBIN: We can provide some details when 2 Brookhaven presents details of their evaluation, but we're --
3 amount of time.
4 MR. MICHELSON: Okay.
5 MR. WYLIE: Before we hear from them, what's the 6 bottom line? Is the risk profile revealed by experience 7 different?
8 MR. RUBIN: The bottom line largely substantiated 9 the overall conclusions of the owners that the Category C 10 events as a general class are not terribly significant. No 11 one would expect a higher frequency of these type of events on 12 the B&W plants, overcooling events. But the more detailed 13 precursor study and risk work that Brookhaven undertook l
14 identified that the vast majority of these overcooling events 15 especially were not terribly risk significant. They would not 16 be expected to produce a high frequency or damage. Damage 17 events were very much produced by the undercooling scenarios, 18 which is a small number of Category C events. So you would f 19 expect a larger frequency of events of concern at the B&W i
l 20 plants versus the other vendors, but the results didn't l
21 substantiate significantly different risk profiles for the l
22 plants.
23 The evaluation produced by B&L did result in some 24 upward requantification of core damage estimates due to these 25 type of events. In some cases over a decade. But still, the
{' }
Heritage Reporting Corporation (202) 628-4888
i
-~ 462 1 total contribution or damage contributions were only in the 2 small to moderate level, on the order of a few -- for 3 instance.
4 What we were concerned about was the existence of a 5 radically different risk profile, risk dominance, and we 6 couldn't identify it.
7 MR. MICHELSON: From Category C events only?
8 MR. RUBIN: From Category C events only. That's i
9 absolately true.
10 MR. WYLIE: Was the precursor to Three Mile Island a 11 Category C?
12 MR. RUBIN: I believe it was, yes. We thought a l
() 13 traceable precursor study for each of the Category C events 14 was one of the most valuable aspects in the Brookhaven report 15 and we'll be discussing it with you.
16 MR. CATTON: What do you mean by ' a precursor to 17 Three Mile Island?
18 MR. WYLIE: I've been hearing about how Category C 19 events really weren't very meaningful.
20 MR. WARD: What event are you talking about? Davis-21 Besse?
22 MR. WYLIE: No, Three Mile Island. The precursor 23 that led to core damage at Three Mile Island. I just wanted 24 to know if it was Category C, and the answer was yes. What's 25 different about it than the others that lead to nothing very Heritage Reporting Corporation (202) 628-4888 l
463 g
V 1 meaningful?
2 MR. JONES: Operator action. There is also the 3 inversion of the PORV setpoint and the high pressure trip 4 setpoint which is a biggy on decreasing frequency of the 5 potential loco due to a stuck open PORV which is -- That was 6 the big action taken fairly quickly after the TMI event, 7 within a week, if I remember the time frame.
8 MR. RUBIN: The evaluation that we performed was pre 9 any of the proven program fixes, but in some cases after TMI 10 fixes, some of the modifications that came out of the TMI 11 plant. So looking at the precursor results, they were 12 probably accurate reflections at a point in time when the O 13 event occurred. In some cesee, eh-eh improvements ere 14 underway or will be underway that will modify the results of 15 those.
16 MR. MICHELSON: Wait. If you're finished now I have 17 to ask a question. I thought you had more to say. I referred 18 to this question on page 9-9.
19 MR. RUBIN: That's Brookhaven.
l 20 MR. MICHELSON: Okay, that will cover it.
21 MR. YOUNGBLOOD: I'm Bob Youngblood from Brookhaven.
22 Also here today from Brookhaven are Bob Fitzpatrick and 23 Charlie Soo. Accompanying us from Applied Risk Technology is 24 Paul Ameko. The four of us are the Brookhaven people involved i
25 in this project.
l Heritage Reporting Corporation t (202) 628-4888 i
('T 464
\_/
1 It's been gone over enough times this morning that 2 this will be a little repetitious in places, but I'd like to 3 not skip anything in order to make sure that the emphasis that 4 I end up communicating is the emphasis that I mean.
5 Our job was to try to put the observed events in the 6 context of PRA. When we started there was a perception that 7 PRA did not single out Babcock and Wilcox plants as a class as 8 being different in terms of core damage frequency, and yet 9 there are these transients occurring more frequently. So 10 before the project we were essentially being asked should we 11 be doubting PRA results based on the fact that these events 12 are happening? Is PRA missing something important? If it is, en k 13 should we be going with different estimates of frequency?
l 14 Fairly early on in the project we were asked to look 15 at the owners group submittal which was summarized earlier.
16 They looked at the events, characterized the failures that 17 occurred in those events, and looked at the PRA models to see 18 whether those events had been addressed, modeled in the 19 existing PRA's. They pointed out correctly that those events 20 were within scope and part of the existing PRA treatments, and 21 from there concluded that the PRA results could continue to be 22 trusted and went on to argue further that support system 23 faults which adversely affect more than one front line 24 function as still the kind of thing that are going to get you 25 to core damage.
}
Heritage Reporting Corporation (202) 628-4888
465 1 Some of the questions that have come up in the last 2 few minutes, this is not properly identified on the handout.
3 This is actually a figure from the submittal, and just with 4 the Davis-Besse event in mind, the '85 one, just look at the 5 first few headings here where the question is asked whether 6 emergency feedwater actuates, delivers flows. This is a 7 fairly simple event tree in the sense that it's got one branch 8 here, B&W either fails or does come on, and the only other 9 question asked is whether it's controlled. On this tree, the 10 Davis-Besse event where feedwater was lost for awhile and then 11 recovered, counts as a success.
12 That's not wrong. They did eventually cool with the O 12 emereency feedweter erstem, but the event tree doesn't te11 14 that story, and as you'll see in a minute, a large part of 15 what we did was try to build a tree that would tell that 16 story.
17 Also arguing that the failures experienced are a 18 part of the model doesn't go one step further to say that the 19 model also would predict the correct Category C frequencies.
20 Knowing that certain failures are in the trees, you can still 21 ask, if I asked this model what's the frequency of 22 overcooling, what's the frequency of Davis-Besse type events, 23 we felt it would be useful to try to build a model that would 24 provide those results as well, and a model which could do that 25 correctly, a model which could correctly predict complex Heritage Reporting Corporation (202) 628-4888
466 1 transient frequencies, would be more credible in its core 2 damage frequency prediction. So that's pretty much what I'm 3 about to try to summarize. That's the approach we took.
4 This is not as legible as I would like it to be.
5 It's a little more legible on your handout. We don't have to 6 walk through it, but without reading it you can tell that we 7 have many more headings. These things across the top here 8 are event tree headings. We have several which ask whether 9 emergency feedwater came on, if it didn't, how did the 10 operator respond, were they able to get it back, then did they 11 control it, and so on. So there's more structure here which 12 allows for a more narrative description of events like the
() 13 Davis-Besse event. The stair step thing here actually is the 14 path through our vent tree that was taken in the Davis-Besse 15 event. So if you start off here, they eventually lost main 16 feedwater, emergency feedwater did not come on but they did 17 realize that they needed to get it back. Eventually they did 18 get it back. So these first several steps is that part of the 19 story, and eventually they ended up okay.
20 MR. CATTON: Does this mean the probability of a 21 Davis-Besse event was two times ten to the minus ten?
22 MR. YOUNGBLOOD: No. The numbers in each step here 23 are the probability given that you're standing there on that 24 plateau, and without knowing how the next thing is going to go 25 what was your chance of getting core damage. So at worst, the O
Heritage Reporting Corporation (202) 628-4888
m i k
q 467 V
1 dark before the dawn according to these numbers here which are 2 still undergoing a little tuning, is about three and a half 3 percent. That comes from about a .2 chance that they weren't 4 going to get back emergency feedwater given that it was lost, 5 and another .1 that they were not going to succeed with makeup 6 cooling. So .2 times .1, well it's two percent not three 7 percent, but in round numbers that's where it's coming from.
8 So that's what Clark Rubin was alluding to in the precursor 9 part of it.
10 This is some measure, according to one event tree 11 breakdown of how close that one was. This and one other were 12 the only two that were-anywhere like that significance. A 13 corresponding plot of this for the undercooling events has 14 numbers nowhere that large.
15 MR. RUBIN: You mean overcooling.
16 MR. YOUNGBLOOD: Overcooling, yes. I misspoke.
l 17 MR. MICHELSON: What this shows it that given an l
18 event you can go back and draw a tree for it, right?
19 MR. YOUNGBLOOD: That did not need to be shown.
20 MR. MICHELSON: But this does show that once you l 21 know what the event is you can figure out where to put all the 22 branches. How well can you figure out how to put all the 23 branches in before the event occurs, since there are now many 1 24 many possibilities at each step even than you have necessarily 1
{}
25 drawn here?
Heritage Reporting Corporation (202) 628-4888
468 g-)
G 1 MR. YOUNGBLOOD: Yes.
2 MR. MICHELSON: There are a lot more branches than I 3 could have imagined in a given situation might appear, that 4 didn't in this case but might in this case. It's almost an 5 unlimited tree you end up with when you get done so you don't 6 do that too well because it's such a big job. Is that the 7 reason why we don't draw detailed trees?
8 MR. YOUNGBLOOD: No. I believe that most people who 9 don't go to quite this much detail, although I have to say I 10 think increasing use of the kind of event sequence diagram, 11 narrative analysis that was behind this, I believe that's more 12 common now than it used to be. You may not see this level of
() 13 detail in an executive summary, but I think increasingly 14 people are resorting to it in some of the basic work.
15 But remember that part of our goal here was to back 16 out the Category C frequency to see whether the basic model 17 parameters were right. If you don't have that goal then you 18 may believe that a much simpler event tree can capture the 19 core damage picture for you. Of course that's what was in 20 doubt before that was a part of the project. But I have to 21 agree that if a sufficiently wacko event occurs next week, it 22 may not be on this and then if we were to do the project over 23 we'd have to add headings. But nevertheless, the plots of 24 this for the overcooling cases have small numbers because as
. 25 close as they got, too much more was going to have to go Heritage Reporting Corporation (202) 628-4888
p 469
\~)
1 wrong.
2 MR. MICHELSON: That was already concluded even -from 3 the more elementary tree?
4 MR. YOUNGBLOOD: Yes. So to go forward a little, we 5 made a number of plots like this, and again, I don't consider 6 the results here to be definitive, but this is a pie chart 7 that's showing you, and I would like to study your pie chart 8 in more detail now that I've seen yours to see what's going on 9 here, but this is based on, this is trying to say what kind of 10 thing contributes to Category C. This is trying to say what 11 kind of thing contributes to core damage, and there is a 12 rather different importance ranking. Undercooling, this I O
(> 13 think perhaps understates the case a little bit. But l 14 according to this there would be relatively many Category C 15 events are overcooling, relatively few are undercooling. But 16 in the core damage events that's reversed. It's undercooling 17 that gets you into core damage.
l 18 MR. MICHELSON: You're also not considering external 19 events --
20 MR. YOUNGBLOOD: Absolutely. We are not only not 21 only not considering external, we're considering a very 22 restricted class of internal events.
23 MR. MICHELSON: So we don't know even when you make 24 that statement, observing undercooling and overcooling, we 25 don't know yet whether this is really all that important Heritage Reporting Corporation (202) 628-4888
("i 470
(_)
I anyway compared with the real contributors to risk.
2 MR. YOUNGBLOOD: That's right. In a draft report 3 which is currently undergoing review, there are numbers for 4 these sequences, and most of them are ten to the minus six or 5 less -- particularly for the undercooling. And knowing that 6 PRA bottom lines tend to cluster around an order of ten to the 7 minus four or thereabouts, that's not large, but I'd like to 8 correct a'possible misimpression that not all of these events 9 are risk significant. I think we consider the undercooling 10 events risk signAficant, and I believe everybody considers 11 them risk significant. But they're not related to the kind of 12 basic B&W versus the rest of the world comparison that was the I'T
(/ 13 motive of this study.
14 If you look at the Davis-Besse again, a full blown 15 accident sequence would be initiating event, initial failure 16 of the EFW, and then failure either to get EFW back or to get 17 makeup cooling going.
18 This part of the sequence I think has to be 19 considered utterly plant specific. A good pnrt of this part j 20 of the sequence is also utterly plant specific because they 21 have the particular makeup cooling. There are B&W generic 22 things in here having to do with time frame, but even those
- 23 are not overriding. So the fact that we attribute i 24 significance to the undercooling events does not mean that 25 there's a yes answer to the original question of whether this
}
Heritage Reporting Corporation (202) 628-4888 l
471 1 class of plant has been understated in previous PRA's.
2 So these are some of our conclusions. As a class, 3 they cannot presently be said to have core damage frequencies 4 significantly greater. Any plant that has a non-diverse 5 emergency feedwater system will have certain contributors, but 6 that's not a B&W thing.
7 Overcooling events dominate the frequency of 8 Category C events, but in our analysis and in many others, l 9 they're not significant contributors to core damage frequency.
10 That's basically said in another way in the third bullet.
11 of those events which are experienced, of the ,
12 initiator types that we considered, loss of ICS function was
(} 13 the most important, partly because it affects main feedwater 14 and partly because it just generally creates --
15 MR. REED: You're now saying as a class, now, with i
16 modified or changed or whatever in this present reassessment, 17 B&W plants cannot be said to have a core damage frequency 18 significantly higher than others. But certainly in the past 19 that wasn't the case. After all we had an actual and a near 20 actual out of seven or eight units.
l 21 MR. YOUNGBLOOD: That's right.
22 MR. REED: So there's been a tremendouc upgrading in 23 what you're saying, now, because you can make this statement.
I 24 MR. YOUNGBLOOD: Yes, we're saying that, yes. All 25 of the event tree modeling that's shown here absolutely O Heritage Reporting Corporation (202) 628-4888 i
1 1
'I l
l l
rT 472 U
1 reflects a lot of post-TMI thinking. I believe the credit is 2 taken in many important areas for people being less likely to 3 be fooled by it, but they were fooled by it at Three Mile 4 Island.
5 MR. CATTON: So you could actually go back and 6 ferret out the benefit of all of those changes?
7 MR. YOUNGBLOOD: Yes. Maybe not in an afternoon, but 8 in principle, yes.
9 MR. CATTON: I've heard in the past that that was 10 impossible to do.
11 What do you think about the use of safeties for 12 steam dump valves? Is that something that a PRA can tell me 13 anything about?
14 MR. YOUNGBLOOD: I don't understand your question.
15 MR. CATTON: It seems to me that safeties are there 16 for a purpose. They're to protect the steam generator and not 17 to be used as steam dump valves. Yet in essence, with a B&W 18 reactor they are. I was just wondering if that's something 19 that's of safety significance, or maybe we don't need safeties l
20 when we have steam dump valves. I don't know.
21 MR. YOUNGBLOOD: There's a heading here, and 22 collaborators feel free to chime in, there's a heading here 23 which allows, are you asking about them sticking open for 24 example?
Yes, and the fact that you're using
{} 25 MR. CATTON:
Heritage Reporting Corporation (202) 628-4888 l
473 b a 1 them really almost on a continual basis.
2 MR. YOUNGBLOOD: There's a heading on this tree 3 which asks the question did any of those stick open. We have 4 some number for that which is not a terribly small number. We 5- could reexamine it. If they do, then further questions are 6 asked. How long does this go on, does anybody take control, 7 does HPI come on, do you end up making such a mess that you 8 cause a primary site loco by running the HPI too hard, or do 9 you, there are a lot of questions like that are actually on 10 this tree.
11 MR. CATTON: So it really doesn't matter?
12 MR. YOUNGBLOOD: I think we used a not too small
() 13 number and it still didn't come out big, so off the cuff, 14 yeah, I think it does not loom in this as a problem.
15 MR. MICHELSON: Did you look at events then that 16 went beyond the Category C in terms of things that happen, 17 like I don't know if any of the Category C's ever stuck open a 18 safety. I don't recall --
19 MR. YOUNGBLOOD: Yes, I think they did. The tree 20 goes all the way out to where some states are core damage and 21 some states are okay.
22 MR. MICHELSON: Your tree went on out to the end?
23 MR. YOUNGBLOOD: Yes.
24 MR. MICHELSON: Okay.
25 MR. RUBIN: Even with the safeties stuck open you Heritage Reporting Corporation (k02) 628-4888
p 474 ,
1 still need HPI failure, so there are other failures that have 2 to occur. It's not obviously a -- event.
3 MR. CATTON: What about the ASME guidelines for this 4 sort of thing? Do they allcw that? That you use the safeties 5 on a regular basis as steam dump valves?
6 MR. YOUNGBLOOD: I'd have to defer that question.
7 MR. CATTON: What does the code say about that?
8 MR. MICHELSON: You're in the mechanical end there.
9 What does Harold Hetherington say about that?
10 MR. HETHERINGTON: No comment.
11 MR. SKILLMAN: I'd like to register a comment for ,
12 B&W owners. Our intent is to knock out the use of the main 13 steam safety valve -- energy relief on secondary site. The 14 way we intend to do that is to get turbine bypass and -- and 15 get integrated control systems to the point where it does what l 16 it's supposed to when it's supposed to do it. So our goal is I
17 to address the concerns of eventc, and that is to get the post l 18 trip energy balance -- so that we do not call upon main steam j 19 safety valves so frequently and to such an extent.
20 MR. REED: The key word in what you said was the 21 word prompt. You will find that this pressure change is so 22 prompt that I doubt you can get any sensing and equipment 23 reaction to curtail it.
24 MR. MICHELSON: Was that a level one type of fs 25 recommendation, or level two or three? This idea of getting
(
l l Heritago Reporting Corporation l (202) 628-4888 '
l 1
I
475 1 off the safeties?
2 MR. SKILLMAN: -- regarding emergency feedwater 3 control, secondary plant relief.
4 MR. MICHELSON: They're level one.
5 MR. SKILLMAN: They are among the 71 or 73 key 6 recommendations we regard most beneficial. The level one 7 terminology came from ICS/NNI recommendations that fit with 8 this.
9 MR. MICHELSON: I see.
10 MR. WYLIE: Any other questions?
11 Let me ask the B&W owners group, if we break now do 12 you have people here to support the rest of the meeting?
13 (Pause) 14 MR. WYLIE: You'll have somebody here after lunch?
15 MR. RUTHERFORD: Yes.
16 I would like to address the recommendation 17 implementation process, and I'll try to speed this 18 presentation up here a little bit and hopefully break for 19 lunch here before too long.
20 As we said at the beginning of the presentations, we 21 regard implementing the recommendations obviously as key to 22 the success of the overall program. That's going to determine 23 what impacts we have in actual plants. In order to ensure 24 that we are doing a good job in this area, implementation is 25 being formally monitored by the B&W owners group, at both the Heritage Reporting Corporation (202) 628-4888
476 p/
(_
1 executive and the steering committee levels.
2 We're looking at implementation quality both from a 3 programmatic and a technical standpoint, and in some later 4 slides I'll get into more details in exactly what we are doing 5 in those areas.
6 We recognize that proper implementation of the 7 recommendations depends on the unique position the plant is in 8 in terms of resources, other modifications they've got to do, 9 outage schedules, etcetera. But we are looking for each of 10 the owners to implement the recommendations on a reasonable 11 schedule, and we'll be looking for outliers during that 12 process of evaluation.
(} 13 As a tool to use in monitoring this progress we have developed the recommendation tracking system report, and we 14 15 alluded to this report earlier in the presentations. This is 16 this document here that contains all the recommendations and 17 the status for each utility. This is our Bible, if you will, 18 for monitoring implementation progress.
19 I'll briefly go over just how a recommendation goes 20 through the review and approval process and gets into the 21 tracking system. The numeroua studies, committee activities, 22 etcetera, that we've alluded to today have generated the 23 recommendations initially. These have gone through the review 24 and approval process via committee review, through the SPIP 25 management team we set up to manage the SPIP activities, and Heritage Reporting Corporation (202) 628-4888
3 477 O
1 subsequently, these have been elevated to the steering 2 committee for final review and approval. Once the steering 3 committee reviews that recommendation and accepts that 4 recommendation, it's placed into the recommendation tracking 5 system report.
6 This report is updated on a, quarterly basis based on 7 inputs from each of the utilities. By that it means the 8 owners group as a whole will know what the implementation 9 status is and will be able to review progress of each of the 10 individual companies. This documentation has and will 11 continue to be provided to the staff so that they can also 12 monitor implementation progress.
i i 13 We have currently got 226 recommendations in the i 14 tracking system report. There are a small number of items yet 15 to surface and make their way into the report. As you note 16 there on the slide, we do have potentially pending 17 recommendations totalling nine that may make their way to the l 18 recommendation tracking system report.
19 There were also recommendationL from a number of the 20 studies that indicated more studies needed to be done. It
! 21 wasn't a recommendation that was right for an individual i
j 22 utility to handle, and thus we do have 13 that fall in the l
23 category of additional studies, if you will, that may result l
24 in additional recommendations.
l
[} 25 I'll quickly put this slide up. This just gives you Heritage Reporting Corporation (202) 628-4888 1
/^% 478 El 1 a pie chart breakdown of just where those recommendations do 2 stand at this point in time.
3 Once the recommendation is entered into the tracking 4 system it then flows through this process. These are the E categories that we have set up for dispositioning of the 6 recommendation. First of all, it's evaluated for 7 applicability by the individual company, and either closed as 8 being not applicable at that point in time, or moved on for 9 evaluation for implementation. At that decision point the 10 recommendation is either closed as being rejected by the 11 individual utility or it's moved over into the inclementing 12 category. Of course at that point in time it moves to the b
l- 13 closed operable category once its implemented.
14 To give you an idea of the present status, this is a lb stack bar chart of the status. You can see the various 16 categories: the closed is the clear; implementing is cross-17 hatched; and evaluating for implementation and evaluating for 18 applicability categeries.
19 Starting ou with September of '87, originally we 20 were on a faster updatt of the report. We were updating the 21 report every couple of months, but we subsequently moved to a 22 quarterly frequency. It was updated in November, January of 23 this year, and then most recently in March. You can see we've 24 gone from 415 to a total of 601 of the recommendations have 25 been dispositioned by the operating plants. These numbers l
l Heritage Reporting Corporation (202) 628-4888 f
479 O 1 represent the total for all the operating plants. So you can 2 see as of March we're approaching the 50 percent mark as far 3 as dispositioning of the total recommendations.
4 .Likewise, there's a pretty similar pattern up to 5 this point in time for the key recommendations. These are the 6 70 recommendations that we have denoted as being key 7 recommendations in the tracking system report. You can see 8 there also that we are ap oaching the 50 percent mark in
, 9 terms of dispositionir.g of the recommendations.
10 of course the bottom line for this whole thing is 11 how it impacts plant performance. I'd like to briefly go over 12 just where we do stand in that regard right now. We have 13 already implemented some of the recommendations from the study 14 and we've seen where they have had an impact on plant 15 performance. Certainly all of this trend cannot be laid at 16 the feet of the SPIP program and the recommendations that have 17 come from that program, but it has had a contribution to this l
18 overall trend of reducing automatic trips while critical.
19 If you'll recall from the initial presentation on 20 the program, we originally set a goal here in 1990 of two 21 trips per plant per year. Subsequently to that we have l 22 revised that goal downwards to 1.4 trips per plant per year.
l 23 That's based on the trends that we saw here in '85, '86, and l
l 24 '87 where for the last year we got down to a level of 2.1 25 trips per plant per year.
l Heritage Reporting Corporation (202) 628-4888
. ,r) 480
\_/
1 Thus far in 1988 we have only had two plant trips 2 for the year, which projects out to a rate of .75 trips per 3 plant per year for this year if we continue at that experience 4 rate.
5 MR. DAVIS: I have a question on that. There seems 6 to be a perception that by reducing the number of trips you 7 not only improve plant performance but you improve plant 8 safety. My concern is that that is not necessarily a 9 correlation because you can reduce the number of trips by 10 eliminating some of the reactor trip system logic and some of 11 the setpoints and actually increase the probability of an 12 anticipated transient without scrap.
13 MR. RUTHERFORD: Oh sure, but that's not the way 14 we're getting the trip reduction. If that were the case, I'd 25 agree.
16 MR. DAVIS: I hope you're not doing it that way. I 17 must confess, the evidence suggests that you're not doing it 18 that way, but in our enthusiasm to reduce the number of trips, 19 I think there is a risk that we can make the plant less 20 sensitive to upset conditions and actually increase the atlas 21 risk contribution. I just say that as a general concern and 22 urge you to keep that in mind as you go through the trip 23 reduction program.
24 MR. RUTHERFORD: I think the only thing even rg 25 remotely related is something like raising the high pressure
(_/
Heritage Reporting Corporation (202) 628-4888
()
1 trip setpoint. You're not eliminating the trip, you're 2 adjusting the trip setpoint up a little bit. But I don't 3 believe anything we've done would contribute to increased 4 risk.
5 MR. DAVIS: I did see, I thought, some anticipatory 6 trip reductions and some trips associated with turbine trip 7 were being eliminated. That's not the case?
8 MR. RUTHERFORD: No, that was a preliminary 9 recommendation. That was not adopted. We are in that 10 particular case raising the trip setpoint also, but we're not 11 eliminating the trip.
12 MR. DAVIS: We, of course, have a recent example of
() 13 where a trip didn't occur that caused an extremely severe 14 accident because the operators disconnected some safety 15 functions that would have caused a trip. I want to make sure 16 ve're not moving in that direction.
17 MR. RUTHERFORD: I agree with your concern.
18 MR. KERR: You want to have one per year just to 19 make sure the thing works.
20 MR. MICHELSON: That's the only test you've got left 21 in some cases.
22 MR. REED: Now that you're talking about it, I've 23 always maintained that trips are not all bad, and if a new 24 reactor coming on line, starting up of its own unique and 25 specific design, which they all are, doesn't have six or eight O
Heritage Reporting Corporation (202) 628-4888
?
482
')
{J -
1 trips the first year, I don't think you're going to find out 2 whether the design is flawed. Trips are not all bad.
3 MR. RUTHERFORD: That is one way of looking at it.
4 The other part of our goals set for the program 5 relates to the frequency of complex transients, the Category C 6 events. We had some discussion yesterday on the 7 appropriateness of the goals we had set in this area.
8 As you see from this particular graph, our trend 9 does not appear to be as good as it is in the trip category.
10 However, if you look at the points here, you've got to realize 11 that this is based on a three year moving average, so we've 12 got some experiences back here in 1985, four events, Category 13 C events, that are really driving this curve still. If we 14 continue to have good experience over the next few months, by 15 the end of the year this curve will be way below our .1 goal 16 by 1990.
j 17 Along those lines, we're going to look at whether 18 that goal should be lowered, as we did the trip reduction goal 19 that we have already lowered.
20 MR. REED: I'd like to add a little humor here about 21 a very very early event on the first pressurized water 22 reactor. We were having a number of trips almost daily, 23 trying to get it into operation. An order came down from on 24 high, and we'll say the order came from Admiral Rickover, who 25 is even higher now than then, (Laughter). The order was, i
! Heritage Reporting Corporation (202) 628-4888 i
i
483 O
1 "There shall be no more reactor trips." It was a very 2 interesting order.
3 MR. WARD: What did you do about it? Don't leave us 4 hanging. (Laughter) 5 MR. REED: Well you can take your choice. Either 6 not operate or try to operate at all, or try to defeat as many 7 trips as you can think of.
8 MR. WARD: What did you do about it?
9 MR. REED: I don't recall what was done about it.
10 (Laughter) I was not in charge.
11 MR. DAVIS: Is there some reason you use a three 12 year moving average? It sounds like that may distort the data 13 somewhat and obscure any trends that you may want to pick out I
14 from your operating history.
15 MR. RUTHERFORD: It does to some degree. I guess we 16 wanted to make sure that we did capture a long enough period 17 of data that would be significant. But you are correct, it 18 does stay around for a long time if you have a Category C 19 event.
20 MR. SKILLMAN: It might be worthwhile to point out 21 that the last Category C event we had was December 26, 1985.
22 Then we had another one just recently. If we had done this on 23 an annual average we would have gone on a Category C trip 24 frequency of zero. So the point is by making it three years 25 versus one year we get a longer sampling time so we can see a Heritage Reporting Corporation (202) 628-4888
,. 484
_/
1 trend.
2 MR. DAVIS: I think you could argue it either way, 3 and it's probably not worth.
4 MR. RUTHERFORD: When you get to zero we won't have 5 to worry about it one way or the other.
6 When I first started out I mentioned that we were 7 doing a couple of different evaluations in order to monitor 8 each of the owners and make sure that indeed we are properly 9 implementing the recommendations. The first program that we 10 had was a programmatic evaluation. Back late last year the 11 executives formed evaluation teams to go to each utility and 12 look at the program process that we had in vlace with each of p,L_/ 13 our companies, the disposition of each of the recommendations.
14 This team typically consisted of eight individuals. They 15 developed an evaluation plan. All of these evaluations have 16 been completed at this point in time, and reports issued to 17 each of the utilities with findings from those evaluations.
18 We had an executive member participate in each one 19 of these evaluations. Obviously he didn't do his own utility, 20 he went to another utility and participated in that i
21 evaluation. We had over a man year's effort involved in this 22 with the people that were participating in the evaluations, 23 and over 140 owners group personnel involved in interfacing 24 with this evaluation team.
25 We prepared a summary report which was endorsed by lieritage Reporting Corporation (202) 628-4888
485 1 the B&W executives and the summary report of evaluations has 2 been provided to the NRC.
3 Just briefly touching on the conclusions of those 4 evaluation teams, in general they found that all the programs 5 that the utilities had set up were adequate, but in each case 6 they found weaknesses or areas of improvement that should be 7 implemented to in all cases strengthen the programs. A couple 8 of extra observations from those evaluations was that we did 9 get a lot of cross-fertilization between the utilities and the 10 evaluation team members would pick up good ideas from one 11 utility and bring that back to their particular utility. it 12 was also very helpful to have the executive involvement in 4]ll 13 these 0 Valuations. It helped show the flag, if you will; show 14 that indeed there was an executive committee endorsement in 15 support of the program across the board, 16 In addition there were a number of comments related 17 to documentation reporting, etcetera, involved with the 18 recommendation disposition, and we expect to see improvement 19 in those areas as actions are taken to close out those 20 findings.
21 We have also initiated a technical evaluation of 22 selected SPIP recommendations. This is being done under the 23 auspices of the steering committee. We formed a four man team 1
24 to go out and look at four selected recommendations. They 25 went out and did a pilot, if you will, at GPU and Duke Power.
IIeritage Reporting Corporation (202) F28-4888
486 O
V 1 Those have been completed, the results reviewed by the 2 steering committee, and we have subsequently schedule 3 remaining visits at the other utilities to look at some 4 selected recommendations. Once this group is done, we will 5 come back later and select another group and conduct further 6 evaluations.
7 I might make just a few brief closing comments or 8 observations. As I noted, we certainly feel that the program 9 that we undertook in the owners group has and will continue to 10 improve the performance of the B&W plants. We feel like we 11 did devote significant resources to this effort, and that the 12 scope of the assessment was correct and sufficient.
{s) 13 We've had some discussions here today and yesterday l 14 about well was the scope of the program on target? Was it too 15 small, etcetera? I think we were responsive to Mr. Stello's t
16 letter. I think we did address the issues of complex l
17 transients and trips on B&W plants. And in the process, 18 provided benefit across the board in terms of safety 19 performance of the B&W plants.
l 20 Also I think we are in general agreement with the i
21 staff in terms of the effectiveness of the program. The 22 staff's had some comments in certain areas, but I think in 23 total when you look at the general agreement, that we're not 24 that far apart.
25 We developed over 200 recommendations out of this Heritage Reporting Corporation (202) 628-4888
l 487 O
1 program. We do have a tracking system to ensure that we are 2 going to timely disposition all of these recommendations. As 3 I previously mentioned, we've already seen cases where we have 4 implemented recommendations and they have saved us trips and 5 improved plant performance already.
6 This is a process, an implementation phase that will 7 continue over the next few years, and that certain 8 recommendations do involve modifications that have to be 9 evaluated, designed, and then introduced into the plants. But 10 we're going to make sure that progress towards final 11 implementation or disposition in all the recommendations is 12 appropriate, and we're providing sufficient documentation to 13 the staffs, so they can also draw that conclusion.
14 In closing, I'd like to say that our commitment here 15 has and will continue to be strong, and we want to ensure that 16 everything we've gone through does pay off in the bottom line.
17 These recommendations are promptly implemented and 18 appropriately implemented in all the plants. We want to take 19 the lead, if you will, in the industry in ensuring that we've 20 got the best running plants there are.
21 Any questions?
22 MR. WYLIE: For clarification, on your bar charts 23 where you show the implementation, when you say implementing, 24 that's not implemented until it's closed, is that correct?
25 MR. RUTHERFORD: Right.
Heritage Reporting Corporation (202) 628-4888
n 488 O
1 MR. WYLIE: Closed also includes those that you've 2 thrown out, right?
3 MR. RUTHERFORD: That's correct.
4 MR. WYLIE: So you can't really tell from the bar 5 chart how many have been implemented?
6 MR. RUTHERFORD: That'c correct. You've got to go 7 to the tracking system report, and the tracking system report 8 does break that down.
9 MR. WYLIE: But not by utility.
10 MR. RUTHERFORD: Yes.
11 MR. WYLIE: It does by utility.
12 MR. RUTHERFORD: By utility.
w 13 MR. WYLIE: Now you've got utilities that have 14 submitted schedules and utilities that have not submitted 15 schedules for implementation?
16 MR. RUTHERFORD: I think all of us have some 17 deficiencies in terms of having schedules for the items. It's l'
( 18 a goal to have schedules for the particular phase we're in by 19 the time of the June update. That's the commitment we've 20 given the staff.
21 MR. WYLIE: By individual utilities?
22 MR. RUTHERFORD: By individual utilities. That will 23 be a schedule just for the particular phase in evaluation and 24 implementation. We've got a lot of schedule information in 25 there now, but it is not complete at this point in time.
Heritage Reporting Corporation 6 (202) 628-4888 ,
489 1 MR. WYLIE: By June you still won't knod for each 2 utility what their final implementation date is?
3 MR. RUTHERFORD: Not necessarily, because you'll 4 still have some in the evaluation phase, and until you 5 evaluate the recommendation and decide what you've got to do, 6 you won't be able to set an implementation schedule. But 7 certainly, let me put the chart back up.
8 MR. WYLIE: Is there a date when you'll know when 9 all of them have been scheduled? All the implementations.
10 MR. RUTHERFORD: We have not set a drop dead date at 11 this point in time. I think what we're really looking for are 12 outliers, somebody falling outside the trends that we're all 13 marching on. You can see that this evaluation category is 14 steadily shrinking. We're moving these on up. So we would 15 expect by the end of the year to have very few remaining down 16 in these first one or two categories.
17 MR. WYLIE: Any questions?
18 MR. MICHELSON: Is this going to be the end of the 19 B&W presentation?
20 MR. WYLIE: Yes.
21 MR. MICHELSON: I just wondered if the B&W owners 22 group wished to make any comments on the A47 letter.
23 MR. RUTHERFORD: Not at this time.
24 MR. WYLIE: Thank you.
25 MR. DAVIS: Excuse me, Mr. Chairman. I just have Heritage Reporting Corporation (202) 628-4888
490 O
1 cne comment.
2 Your goal of .1 for Category C transients, it seems 3 to me like you're going to have to wait a long time to show 4 that you've achieved that goal. That's like one transient in 5 ten years.
6 MR. RUTHERFORD: No, that's about .7 per year.
7 MR. DAVIS: Oh, for the seven plants?
8 MR. RUTHERFORD: Yes.
9 MR. WYLIE: They could look at their Category B's 10 and see what happens to them.
11 MR. DAVIS: If you use a three year average, though, 12 it's going to be several years before you can really b
% s/ 13 demonstrate that you've achieved that goal, I guess.
14 MR. RUTHERFORD: As I pointed out, we are going to 15 have several drop off the board here very shortly. Once we 16 get to the end of the year these will drop off, and given that 17 we have good performance between now and the end of the year 18 this curve will come way down.
19 MR. DAVIS: Thank you.
20 MR. WYLIE: Any other questions?
21 (No response) 22 MR. WYLIE: Okay. I'd like to call on the staff to 23 comment on the implementation schedule.
24 (Continued on following page)
(
Heritage Reporting Corporation (202) 628-4888
v-I, ,Y
, 491 1 MR. SIEGEL: --
the magnitude of it, the number of-h 2 . people involved in it,,the effo'rt involvsd.both on the part 3 of the Owners Group and,the staff. I don't know if all of 4 you-have seen th'e AW-1919, but it probably is about two and a 5 half or three feet long. It covers the book shelf. And con-
. ,6 sidering that this was started, this program, in roughly 7 January of '86 and the staff just started reviewing.this in l 8 August of '87, I think it's been a considerable effort.
9 And one of the things that's a little more gratify-10 ing about this than some of the other programs is the fact 11 that since it is balance of plant systems primarily, the 12 Owners Group has started implementing the program. So we're not just waiting for the staff to complete their review until 13 34 this is being implemented. j 15 But because we think the program overall is going 16 to improve the reliability and predictability of the way the j 17 plants respond--but that's predicated on the fact that they !
I 18 do have proper implementation of the recommendations and in [
19 a timely manner. And since the program, in all essence, is ,
i 20 completed as far as the staff review goes, there is still we ;
21 feel a need to assure that a proper implementation and imple-22 mentation in a timely manner.
O 23 S me of these numbers may be a little different ,
24 than the Owners Group, because this is only updated through O 25 January and the slide that Neil showed--they have more recent Heritoge Reporting Corporation g m-
y .
s , ,_
492 l 1 information through March. Buti basically the whole program i O 2 from the start,- the recommendations that came.from the sub-3 committees, probably total about 375 recommendations. ,
4 During the_cours.e of, con'solidating all these, they 5 have identified about 75 duplicates.- Some were rejected.
6 Neil said I think there's 226 that are now approved, and this [
7 number is probably a little high. But all told, there is ,
4 8 probably going to be about, I would guess, 250 total recom-l 9 mendations, give or take a few one way or the other.
10 We in the SER addressed all of them. The SPIP ll identified seventy recommendations as key. We've looked at [
12 the process. We looked at what they did. We looked at their
- i 13 SPRIG group and how they categorized the recommendation. Did ;
14 not have any problem with them.. In addition, the staff in the j is report identified eleven additional recommendations that we 16 felt had safety significance. But B&W Owners Group is track- l 17 ing these through the recommendation tracking system. They [
. t j 18 are providing us this report. We are not totally satified l 3, with the report because the report doesn't really give us all 20 we think we need to get a handle on how much progress the i 21 Owners Group is making. >
They have, as they mentioned before, agreed to im- )
22 O 23 plement all the applicable recommendations. The Owners Group t
24 is overseeing the program and the mechanism by which the re- j 2s porting is the recommendation tracking system. And the staff [
Heritoge Reporting Corporotion (set) meses
l "193
-1 .will audit the program. - *
. '[ ')
, e 2 And the next Vu-Graph--
3 DR. LEWISt~ Before you leave that'one, I-just won-
- f] - 4' dered--I really shouldn',t beat.on you,with this,.because it's
~
5 a generic. question. But'you have some things listed there
. 6 as high priority because_they enhance safety. And the' ques- 1 7 tion that came up earlier is does that mean than you feel 8 these plants aren't safe enough? If so, what are the cri-9 teria and-is enhancement of safety a good thing in'itself?- .
p 10 MR. SIEGEL. Yes, I think it is.
11 DR. LEWIS: No matter of safety?
12 MR. SIEGEL: We don't think that the plants are not ,
7 safe. We think that what these are going to do is eliminate--
{
({ ). 13 !
14 prevent them from getting into situations that these complex 15 transients may lead to where they would have a harder time I
16 controlling the plants and recovering frc: them. i 17 DR. LEWIS: No. I'm completely happy with the
, 18 understanding that these things enhance safety. It's just {
19 that the ancient question of how you know when you've met 20 success and can lay off. I'm not saying that these plants ,
l 21 are perfect. But the whole Commission has been fighting n t I 22 this--
23 MR. SIEGEL: We're still groping with it. I don't F
(
l i
() 24 think we've decided. You know, we've expanded the areas 25 that we're looking at. We're considering looking at
- Heritoge Reporting Corporation mm ;
i
^
4 , ,
494
~
I maintenance. We're considering at looking at a lot of dif- !
.I) 2 ferent areas. We've got performance indicators that we're !
q 3 using now. We'.re' much more deeply involved 11n ' tracking how 0
s_/ 4 the plants--how their management is, how the maintenance is, :
5 how they perform, than'we ever'were before. And as-We keep i f
6 getting into this more and more. I don't think we've totally ,
r 7 decided on how far to go on this yet.
I 8 DR. LEWIS: No, I understand. I know your tools. ;
9 It's just that--and I know how you grade the people. I'm ;
i 10 groping for what the passing grade is. I teach for a living, ,
i 11 and students always want to know the first day of class, "What ;
! i 12 - do I need to do to pass this course? Because gosh knows, I
{
- gllh 13 don't want to do any more than that."
14 MR. SIEGEL: I don't know how to respond. I don't .
15 know-- !
16 DR. LEWIS: Okay. I don't want to beat on you.
! 17 But the reason the Commission issued safety goals some while l la back is to try to cope with this type of problem. ,
39 MR. SIEGEL: .That's true. t 20 DR. LEWIS: Please go on.
21 MR. SIEGEL: Okay. The staff intends to verify i
22 implementation of the recommendations by the utilities. As j
()
i 23
- a. matter of fact, in one area the staff evaluated the owners !
ll 24 Group's recommendation process back in October of 1987. We j 25 sent a team of five or six people to B&W. We looked at how Heritoge Reporting Corporation
, m== ,
I .* * ' ' ,
i
?
/
495 1 the recommendation tracking system--not the recommendation (3, 'J 2 but how the process that the Owners Group used to--the recom-3 mendation process they used, how they evaluated the recom-
,~
() _ 4 mendations.
3 The recommendations are generated by subcommittees.
6 They go through the SPIP and then they go to the Steering 7 Committee and then they are approved by the Steering Committee 8 and go out to the individual utilities for implementation.
9 So we are concerned that perhaps--and also involved to in this was the SPRIG Committee that also was involved in
- eliminating duplicates, deciding which were the key recom-12 mendations. Their Advisory Committee got involved in this
) too. So we were concerned that perhaps some of the recommen-
{
, 13 14 dations were being diluted once they came out of the Steer-15 ing Committee. So we went through and we probably reviewed 16 probably three quarters of the total recommendations that 17 came out of the system. And found in essence that what the 1g Owners Group had done in their approval process was adequate 19 essentially. We did not find any place where they tried to 20 dilute the intent of the recommendations in any way. When 21 they combined recomme..dations, it was acceptable. The ones 22 that they rejected, we disagreed with the ones that they re-t-%
\ /
23 3ected. So overall, we were quite satified with the results 24 of that audit and found that they were doing an adequate job.
25 What we intend to do now is implement--audit the Heritage Reporting Corporation cms a
l l
1 496 I implementation of the program. Similar to what the Owners 2 l Group did. The Owners Group just completed their auditing of i
3 the progr.'ms at tne individual utilities. We also intend to 7-U 4 j; audit the programs at the utill ;es to see if they have ade-li 5 h quate programs in place.
f 6
l One thing I should mention is that a lot of recom-l 7 [ mendations are fairly general in nature. Some of them as a 4
8 lresultoftheirgeneralitieshavebeensentbackfromthe 1
9 i Steering Committee back to the subcommittees to provide more 10 detail, but even a lot of the ones that coming out ne fairly 11 lgeneralandtheyhavetobeimplementedattheplant.
i 12 l We are concerned that they are properly interpreted 13 by the individual utilities. We are going to audit the imple-l 14 i mentation process at the individual utilities to assure that 15 they are interpreting the recommendations properly. The 16 j Owners Group has performed a similar function. We'll probably ,
17 ,
when we first go out and audit the process that they have in i
18 l place, the program that they have in place in each utility, 19 we'll probably also pick at several recommendations and look 0
20 l at them and see if they are doing an adequate job to get a 21 ! flavor of whether or nrt we think they are interpreting the i
n 22 ! recommendations properly. And then go back--if wo think they
~
) !
23 are doing an acceptable job, then go back when they've com-(l 24 [ pleted a larger percentage of the recommendations and the ones c) e
.5 [ that we consider the most important ones and audit the h
H H Heritage Reporting Corporation (man =
w 497 1 implementation of some of those.
O-- 2 And finally we arefgoing to'v,erify that any recom-i 3 mendations, that the recommendations have been implemented I O.
4 and are in place. . t 3 So that's what we intend to do to ensure that the
'6 program--that we're satified with-the proper implementation 1
4 7 of the recommendations.
i 8 One of the biggest concerns I think the staff has
! 9 in addition to that is whether or not the Owners Group is to really implementing these in a timely manner. And these
- i. 11 numbers are again back from January and not from March. But 12 the numbers-that Neil gave werc a little misleading in the i
A (lll 13 sense that these are percentages of the total recommendations.
- This is overall utility progress of all the operating plants.
34 j 15 And this is the percentage of the recommendations that are 16 being evaluated, implemented, closed operable, closed not i
i 17 applicable, closed rejected and not started.
! gg This'"closed cperable," "closed not applicable,"
I i
i 1, and "closed rejected," were all lumped together on Neil's 4
- o slide.
4 21 And the point that I was really making is the fact
- 22 that even though it shows that a lot of the recommendations
() 23 have been closed, a lot of them have been closed because they J
were closed not applicable.
- f. 24 25 And if you read the report, the supplemental report, Heritoge Reporting Corporation (ast) 686 aast
i.
498 I you'll see that the progress in the past seven.or eight months
'~'
2 has not been in' closing the recommendations and also in evalu-3 ating and implementing, there has beenfprogress made, but not~
as much as the staff feels is necessary. We would like.these 4
5 -to be done in a timely manne,r so tha't we.are assured that 6- these are in place and we can achieve the goals that have 7 been set for the program'.- hnd'we'll'be~ monitoring.these and 8 deciding whether or not the Owners Group or in particular 9 individual utiliti s are in some way dragging their feet or 10 not.
11 That concludes what I have to say on' implementation.
12 Does the subcommittee have any questions?
LI\ss 13 MR. DAVIS: I have a brief comment. You seem to l
! 14 be somewhat critical of the sluggishness with which these 15 recommendations are being implemented. But I noticed on the 16 slides that the Owners Group presented, they've already a-17 chieved one of their goals of 1.4 transients per year and it 18 appears they will achieve the other goal of' complex transients 19 over the next few months and yet they have only implemented l 20 20 percent of their recommendations.
I Do you have different goals on that that you are--
f 21 l 22 MR. SIEGEL: Well, like Bob said earlier, we haven't l 23 set any goals with regard to number of complex transients.
l 24 MR. DAVIS: What are your goals?
[
l
({ }
25 "R. SIEGEL: I think we are going to--essentially l
i Heritage Reporting Corporation I m ean l
3 = . , . . , ,. - , - - . - - _. - -- - _ . . _ , -
4= .
493 I the goals are to try to get the Owners Group to implement n.
U 2 these recommendations as soon as possible. And'then probably 3 for a period of two or'three years trackiwhat happens and see-(3'
'\/ 4 where we're going. See how many Category C events we get.
~'
5 .See how many'high Category /Bf.
Cornpare that to what was done-6 previously. If'it looks like--I'm just postulating--if it 7 looks like this situation _in^our mind is under control, we 8 will think that the Owners Group has don? an adequate job and 9 that we can put this to rest. If it turns out that there 10 isn't any significant improvement over what it was before we-11 initiated. this program, then obviously we will consider tak-12 ing steps to correct the situation. I'm sure the Owners 13 Group will too.
14 For instance, they've got an advanced ICSNNI system.
15 If that's one of the areas that's causing a problem, maybe 16 at that point in time they may consider implementing that.
l 17 I don't know.
I 18 It's hard--we don't have any specific goals in mind 19 at this point.
20 MR. DAVIS: But the goal you are using is the number 21 of transients per year. You are equating that--
22 MR. SIEGEL: 1 think we would evaluate the tran-l.( sients and see how complex they were if they were involving 23
! /~T 24 systems that hadn't been involved in this program. And V
25 if there is something that we're missing or not. Hopefully--
Heritage Reporting Corporatior im> su.e
, - - - . . .- , - - - - - - - , - . , ,- _ , _ . = . - - , - - . .
so 4
500 1 it's hard to assess it right now--hopefully I would think--
~h-t 2 I wouldn't say hopefully--I would expect that even if you do 3 have Category C event :their severity wouldn't be as severe
,~
- N-) 4 as Ehey were previously,because'you!ve taken_all this cor-5 rective action.- 'So I don't really--you_know, I don't think 6 you can really at this pointiin"time say what~would be a 7 triggering point to say, yes,-this is acceptable or--
8 MR. DAVIS: I don't want to belabor it, but we've 9 already heard that Category C events are not risk significant.
10 So eliminating them isn't going to make the plant much safer.
11 Isn't that true?
12 MR. SIEGEL: From a core melt standpoint, that's lh 13 true. From an operational standpoint--I think the staff 14 feels that anytime you challenge safety systems, anytime 15 you have severe transients where the operator can't predict 16 what's happening to the plant or understand what's happening 17 at the plant, and this is the type of events that some of 18 these Category C were--that's unacceptable. And I think 19 that's what we're concerned with. The fact that you are 20 ' working in a gray area where the operators really don't under-21 stand what's happening in transients because they don't have L 22 the proper controls and instumentation to know where they are.
l (2) MR. DAVIS: Thank you.
23 24 MR. WARD: When you say that, Byron, what you are
(
25 saying is you don't trust PRA.
Heritoge Reporting Corporation (set) p6-mes
4 p a 501 t fir . SIEGEL: That's right.
2 MR. MICHELSON: That's right. Because PRA isn't-3 -included.
(~\
%/ 4 MR.fWARD: I J think.that might be a healthy attitude,
.c _
5 but-- because the .PIUi analysis has come to a rather different; 6 conclusion than= your: operative 2 conclusion'. - _
I'm not qual 1fied to really answer
~
7 MR. SIEGEL:
8 that. I mean, if you look .aii the ' numbers, that's true.
9 DR. LEWIS: This is really a very important point
'10 that Dave is raising. Because, you know, you can't regulate 11 by--you can;t just do it on the basis of feeling good because 12 there's a general princ ple of American law, I guess, or l 13 English law, that whoever is being lawed has to know what he
' f-s-
34, has to do to obcy the law. I mean he wants to, but at least 15 he ought to know what it is. And it's not clear to me that 16 if I were a B&W owner, which praise the Lord I am nut, that I l
f 17 would know what to do to satisfy the staff. I'm not sure the 18 Commission has given you much help on that score, but there's 19 got to be some kind of set of criteria se that regulatee 20 knows what he has to do.
21 MR. SIEGEL: I'll draw a comparison. Would you 22 feel comfortable driving down the road at 65 MPH blindfolded l Ik,s) 23 even though you knew you weren't going to crash into some-() 24 thing?
Well, but you know, even there, thare
! 25 DR. LEWIS:
l t
Heritage Reporting Corporation msu L_
a/.
4
~502 I are a set of fairly specific rules. You know, in fact I
',) 2 shouldn't drive with my vision obstructed, whether I put a 3- drape on the windshields or on my eyes. There's actually a 4 rule that says that I can't do that. So:at least I know it.
5 .It's not just a matter of co$ mon ~ sense,,although it's cer-6 tainly that too.
7 MR. S'IEGEL: In that respect,.you can-take the 8 blindfold off and look. But if you've lost your instrumenta-9 tion, for example, you stil1 have the blindfold on, you can't to do anything about it.
11 I don't know. I'm not the one I think that can 12 answer that question. This program started a long time ago.
llh 13 I think that what the Owners Group is doing personally to 14 eliminate these types--to try to reduce these types of Cate-15 gory C transients is the right approach. And I'think what 16 they've done has gone well beyond what we normally have done 17 in the past and covered areas that we haven't--that we've 18 never done before.
19 They did a much better job than we could have if 20 we initiated the program ourselves because they have the ex-21 pertise and the manpower to do it. We could not have had as
- 22 extensive a program as this if we had done it ourselves.
U 23 DR. LEWIS: Well, doing too much is sometimes worse
() 24 than doing the right amount of stuff. I think--you know, as 25 I sense it, you are adopting a position that anything that Heritage Reporting Corporation m m.=
ng g..
503 1 can be done isca good thing. Now, I think that's a simplifi-O.
~'
2 cation--
3 MR. SIEGEL: I don't'think that all'the recommenda-C.,)
- 4 tions~that they've-proposed, for instance, the staff would 5 have'necessarily considered to be necessary. There are a lot 6- of the recommendations that we feel are necessary. There are some that they are put, ting in'for the'ir.own.uses and purposes 7 .
8 too.
9- DR. LEWIS: ;But you are making-that judgement on 10 the basis of your feelings and your experience, which I'm 11 not quarreling about. But I was delighted to-hear my friend, 12 Mr. Ward, defending PRA. He doesn't do that very often. But- -
(vjg 13 MR. SIEGEL: I didn't hear him defend.it.
14 MR. MICHELSON: PRA says these are no never minds 15 though.
16 DR. LEWIS: That's right.
17 MR. MICHELSON: This is a voluntary program. The i
18 utilities are doing this on their own.
19 DR. LEWIS: I understand. In a ficticious world 20 in which there were no NRC, I bet a nickel they wouldn't be 21 doing this.
l 22 MR. MICHELSON: I think I agree.
i l 23 DR. LEWIS: Unless the economic hazard became--
l Because doing
() 24 MR. MICHELSON: Economic and safety.
25 things you don't need to do to a plant has a negative ef fect i
1 Heritage Reporting Corporation
<mnu l - --+- ,. - r--- . . . _ _ _ _ , , , _ , _ _ _ _ _ _ _ _ _ _ . , _ . . _ _ , _ _ _ _ _ _ _ _ . ___ ,, ,
it,{.
504'
,, l. on safety. And I'm just concerned that there are no standards
't"j '
2 we're.using here that can be expressed. And I'm not blaming 3 it on the staff. I think;the Commission has the responsi-
, (~\
b/ 4 bility here, but we're sort of seeing it--
'S MR. JONES: If I may.-comment a little bit. WeLdo J
~ , . < ~' . . .
6 deal--to some extent you'are'given'a' risk base in the PRA 7 World, but we do also'dealcin the-world of perceived risk.
Rancho. Secos. eve 5t which the risk study
^
8 And clearly the 9 says were not that risk signif'icant.. Have a lot of per-10 ceived risk. A lot of perceived risk, 11 MR. WARD: Where in the atomic energy laws does it 12 say you are supposed to deal with perceived risk? Perceived 13 by whom?
14 DR. LEWIS: Perceived risk is a new concept in 15 safety analysis.
16 MR. JONES: Well, in a sense it--it was perceived 17 risk and you may be right about the atomic energy law, but 18 that transient was an awful nasty transient that got a lot 19 of attention. Now, what we found--now, this was a perceived 20 risk. The transient in and of itself was not a risk signifi-21 cant transient. Put that on t he shelf. What we found from l
22 that transient was a lot of problems with the plant design, l
l
() 23 maintenance, all sorts of management issues, that has kept i'
() 24 that plant down for a long time. That's perceived risk which 25 you cannot pick up in a PRA directly. Because the PRA makes l
Heritoge Reporting Corporation f imnw =
l
505 I certain assumptions about the_ quality of the equipment, the 2 care and maintenance of that equipment to come up with the 3 numbers. And a lot of time a lot of the recommendations re-
\ /.. 4 ceived from the Owners Group comes down to proper care, 5 maintenance, tuning of. equipment. And in short fills the
+
6 . original function it was p,ut inzthere ,for.
7 DR. LEWIS: *I don't need.a speech about the impor-Y - . .
8 tance of precursors.- I know something.about it. But could-9 you tell me where in the NRC regulations or issuances I can 10 find the expression "perce'ived risk"?
11 MR. JONES: You won't, but that's the basis on which 12 we got out on AITs and IITs. To some extent, we judge the h 13 significance of the transient. It's a perception at that-14 time. We don't have a PRA to judge it with. It's a per-15 ceived risk at that time.
16 MR. SIEGEL: At the start of this program we really 17 didn't know. If we were where we are now at the beginning 18 of the program we may have changed it.
19 MR. WARD: Good. Rational comment. Good. Because 20 you are kind of going around in a circle. You say this event 21 occurred at Rancho Secos. The perception was that it was 22 a nasty event and a lot of people shared that. So you went 23 out and the IIT found a lot of things wrong. Well, those things were defined as wrong because they apparently contri-() 24 25 buted to that event, which was believed to be a nasty event. I Heritage Reporting Corporofion o.n wnoe
- [n 506 I I ~ Now you've gone through-a more elaborate-exercise'and if you
(~'h .
Il 2' -believe the analysis, you'have to come to the, conclusion that 3- ~that event wasn't really all that nasty. And.that means that D
k/ 4 maybe those maintenance. problems weren't all that nasty and-5 so forth.
6 MR. JONES: And except.the maintenance problems i' . . ;,
7 with respect to the valves were~found to be widespread in 8 the pla'nt, which said it would:be nasty.
9 See, I"think when you get a nasty transient you 10 have to look at it. You h' ave to look'atiit hard and you have 11 to go out and look at the operation of that plant. That's 12 what I'm going now with under the restructure we went through 13 about a year and a half ago. Is to look at the operational h
14 safety of these plants. And if you are getting a lot of 15 transients, some fairly complex transients, go out and study 16 them, and see whether there are generic problems that exist 17 at those plants. That will not be picked up very easily 18 from a PRA or a risk study.
19 DR. KERR Did Rancho Seco have a resident in-20 spector?
21 MR. JONES: Sure, they all have resident inspectors.
22 DR. KERR: Isn't it possible that a resident 23 inspector might pick up poor maintenance?
24 MR. JONES: Sure. And we've got other programs, 25 other processes--
Heritage Reporting Corporotion m.
g -
a 1\
507
. 1 DR. KERR: No,-wait a minute. You've had to send
- (j -
~3 2 .out an IIT team to discover that there were maintenance pro-3 blems? If that's the case, I d6n't see why we're spending
,~,.
\/ 4 as much money as we're spending on resident inspectors.
5 MR. JONES: We sent out an IIT team to look'behind i - .
6 that event for what-were the causes of that event and what 7 was going on at that utilit'y; And I'think they found things 8 above and beyond what a resident-could ever hava found.
9 MR. S'IEGEL: Yes. If you' looked at'the number of 10 . man hour inspection that'goes on.at these sites, it's impos-11 sible'for the resident inspector to get into the depth and 12 scope in some of these areas--he may have a general feeling-() ,
13 for the' plant and its maintenance may not be as good.as he 14 thinks it should be, but unless you do an in-depth study and 15 you send out enough people so that you can go into it and 16 determine it, it's really--
17 DR. KERR: How many people on an IIT teum?
18 MR. SIEGEL: I don't know. I've never been on one 19 of those.
20 MR. JONES: If I remember, that was a seven-man 21 team.
22 MR. SIEGEL: It was probably a week or so?
23 DR. KERR: A resident inspector spends fifty hours 24 a week, forty hours a week presumably at a plant doing some-(V"T 25 thing or other. Now, did the IIT team spend more than that Heritage Reporting Corporation (392) 626 4000
-k.,. l 1
500
- 1. time?.
'N" 2 MR. SIEGEL:' Absolutely.
3~ MR. JONES: The IIT team was there I think on site
(- 4 for over a month. I can't remember the' exact dates. I wasn't 5 a member of that'IIT either.
6 DR. KERR: So they had to be more than-eight people.
7 MR. JONES: They were o.ut,there, about seven people,
'8 very concentrated;on the very specific issues, and-then ex -
9 panding into trying.to figure out generically:what was going 10 on. The residents have a lot of other--as I understand it, 11 they've got a lot of things 'the' y'do.
12 MR. SIEGEL: They've got fixed types of--
llh 13. DR. KERR: They don't look at maintenance, in 14 other words.
15 MR. SIEGEL: Yes, they do to some degree, yes.
< 16 They do. But they do it sort of broad brush. If they see 17 a problem in a specific area, they'll look into it in detail.
18 But they don't do an overview of the whole maintenance pro-19 gram at a particular site. Probably through their walks 20 through the plants and their observations, they may periodi-21 cally look at them doing maintenance en or.e particular compo-7- 22 nent or they may, walking through, spot something that they l ()
i 23 don't think is being done right.
() 24 DR. KERR: But there had not been any hint from 25 your resident inspector up to that point that maintenance was Heritage Reporting Corporation j (set) 6*ases l
'E 509 1 a problem?-
.(~')
h_/, 2 MR. SIEGEL: I don't know.
.3 MR. JONES: I1wouldn't want to say that because s- 4 as-I remember Rancho was at that~ time probably a low performer 5 and many of these issues were probably.there. There is a lot 6 of' activities that involves the resident inspectors. He's 7 got performance' indicators, et cetera. But trying to get 8 ahead of the game on these poor performers and the enhanced 9 inspection programs performers. -
10 - So I think we've'taken~a lot of action in that
- 11. area.
12 MR. SIEGEL: One thing I forgot when you mentioned lllg '13 about the Category C events and the core melt risk involved 14 with them. There was, if you look at the original SER, 15 there was some core melt risk associated with those. And 16 in the supplement we identified the fact that if they took l 17 corrective acti>>ns, the corrective actions that they are tak-18 ing in the area that they are would reduce that to the point 19 where it was essentially not significant at all. Insignifi-20 cant.
21 MR. WARD: It seems to me that what's at issue 22 here--or one thing that's at issue here is what approach to l
l 23 . perfection is recurred in the operation and the maintenance of nuclear power plants? I mean, sometimes I think we behave r"s 24 l (-) as if our goal is perfection and that's not only impossible, 23 Her!'. age Reporting Corporation
<mnw
r .
h- '
510 I it's a silly aim. It's a wasteful aim. Because the Commis-2 sion has taken a courageous position I think for a. government 3 regulatory body and it has provided a safety goal, which is
/m k- 4 really far, far short of perfection. But it's an attempt 5 to balance, you-know, a number of societal issues in a rea-6 sonable way. And I think it's a good way.
7 But it seems!that when we're faced with a real 8 application of it, and I know you are:not directly comparing 9 it with a safety goal, but in a general sense, this is what 10 we've got here,.is beginning to regulate from a risk perspec-11 tive and a risk perspective means a mature understanding that 12 perfection is not attainable.
llh 13 I mean, an IIT or an ACRS committee can always 14 find things that are wrong. And always think of better ways 15 to do something. But that doesn't mean those things are 16 necessary. And we've'even got a way now of judging whether 17 they're necessary or not. It's an imperfect tool. I mean 18 the PRA and the safety goal are tools for doing that. But I 19 think we ought to be using--I mean, I think--you know, I 20 don't know whether to believe the whole B&L analysis or not.
21 But if we take that as a valid look at this whole issue, I
- 22 think we have to take what it says. What it says has some V
23 pretty profound implications I think.
(} 24 MR. CATTON: But, Dave, on the other hand, it seems l 25 to me that if you periodically put this plant through a state Heritoge Reporting Corporation m su-
A -
511 I of frenzy or a period where you're not really sure what's
. s/
(') 2J going ~on, that has to increase risk.
3 MR. WARD: Well, you are saying it does. And so--
/'T -
5/ 4 I mean what I'd like.to do is to take that opinion of yours 5 and that other people seem to have and may be right, and some-6 how get that ground into these PRAs that we take.
7 MR. CATTON: I. don't know why it doesn't show up.
8 MR. WARD: Maybe.that's--well, one reason is that-9 you're wrong. One possible explanation is that you are wrong.
10 I don't know whether that's theccorrect one, but that's a 11 possible explanation is that you are wrong.
12 DR. LEWIS: .It looks wrong.
ll 13 MR. WARD: Well, it may well be. It may well be, 14 MR. MICHELSON: And if it's wrong it's because it isn't 15 modeled into the PRAs. Another reason is even where they've 16 attempted to model, say, the unit factor, which is one of the 17 big contributors, and if they've attempted to model it in, 18 they haven't got the data with which to run the calculation.
19 And that's just on the human factor alone. And I don't think 20 there's any disagreement about all these Category Cs. The 21 staff and B&W both say these are non-significant from the
- 22 viewpoint of core melt. We're working on them for other rea-m) 23 sons than risk. We have to be. Because they are not risk 24 significant.
.O I think we're working on it because we 25 MR. WARD:
Heritage Reporting Corporation (Mf) 4M ages
f,.
512
~1 don't quite believe that'.
(~ '
^") 2 MR. MICHELSON: Well, both sides have agreed that.it 3 is not risk significant.
<-~
s '4 MR. WARD: Well, that's the only' reason. I mean--
5 MR. MICHELSON: Isn't that right?
6 MR. WARD: -Well, the Owners Group could be working 7 on them because they want to-have better operating plants 8 which.is a perfectly valid reason. I mean the only reason 9 we or the staff should be worried about-it is because we-10 don't believe the PRA results.
11 MR. MICHELSON: Well, I thought the reason--
12 MR. WARD: We don't have. full confidence in them.
(lh 13 And that's okay. That may be necessary.
14 MR. MICHELSON: I was originally of the opinion 15 that the reason to get into all of this is we were trying to 16 get a new reassessment of the overall risk. But it turns 17 out they didn't do the overall risk. They did the narrow 18 risk of those certain set of transients. And, of course, 19 that was the set that'had no risk in it to begin with so we
- 20 came up with a no set.
21 MR. WARD: Well, we didn't know they didn't--no. ;
^
- 22 I don't think a year ago--I mean I sure didn't have that
, a 23 understanding that there was a no risk from those. I mean l
(} 24 that's something new.
25 MR. MICHELSON: I think that's a good conclusion to Heritage Reporting Corporation mm.
513 I reach but'now we ended up--even though we've ended up as being 2 non-significant to risk, we've got some kind of a program 3 here to get these changes made. And I think it's only be- .
4 cause the utilities have agreed to do it.
5 MR. WARD: If we really had full confidence in 6' this PRA conclusion, and I guess--I don' t have enough guts 7 to say we should have--at this point I'd say the NRC ought
- 8. to back out of this whole thing and just let the utility run 9 its own--conclude their operation.4 10 MR. MICHELSON: I do too. I believe the PRA. But
'11 now when you start looking,at'the PRAs'a little more care-
'2 fully, we realize the deficiencies and the modeling of these 13 various factors--
. 14 DR. KERR: Well, first of all with the PRAs, they 15 are no damn good and I disagree with your intuition. .
16 DR. LEWIS: Well, that's what troublen me about 17 the whole thing. Because there was an IIT on this particular 18 transient. The charge to an IIT is to find out what happened 19 and determine the root cause. You know, find out what it is '
20 you need to fix in order to reduce the probability that this 21 will happen again. You can't eliminate the chance that that 22 will happen again, but you can at least reduce the probabi-23 lity. And the IIT report I think didn't say, after it was 24 finished--I don't remember what it said--but I bet a nickel i 25 it didn't have a line that said, "Besides these plants don't Heritage Reporting Corporation
<=nu
N 514 1 look very good to us." It didn't say anything like that.
() 2 So what troubles me is that when this proceeding 3 . passed the point of analysis into a land in which there is
~
~
4 no map, there is no guidance, .and that's why we're flounder-5 .ing. I think-it's wrong to pit PRA as one thing in which 6 my good friend, Mr. Ward, doesn't have as;much confidence as 7 he will someday,'against just wandering around, because PRA 8 is one possible map. But the real issue is whether wandering 9 around without any map-at all is a sensible thing to do.
10 And the only reason we're talking about PRA is that it's'one 11 Possible map. , ,
~
, 12 There 'are ' other possible maps. There's one map llg 13 Provided by the regulations. Do these plants meet the regula-14 tions? If we feel they are not safe even if they do, we 15 should fix the regulations. You've said that yourself many 16 . times. But somehow just wandering around fixing them because 17 they don't look good or because you perceive more risk than 18 was actually found is just not good regulatory procedure in 19 my view, i 20 End of speech.
21 MR. WYLIE: Any other questions?
l 22 We'll adjourn to 3:30 I 23 (Whereupon, there was a recess.)
l 24 MR. WYLIE: We'll resume our meeting.
25 We have several items that the staff is to address,
(
i l
Heritoge Reporting Corporation
< =n .
---_,,,.,-,4 ,___,.m.-,,-~_--.,im , , _ -
3;{
515 I starting with Item 9, evaluation of other B&W Owners' Group o 2 reports, and 11,~ reaction trip initiating event review of the 3 B&W Owners Group Programmatic and Management Actions, and (g
(-) 4 additional concerns identified by the staff. So I'll have 5 the staff address those.
- 6 MR. JONES: It's very short. These will be real 7 quick. They were actually put together assuming that the 8 Owners Group was going to go first. You don't have a lot _of 9 background on it, but I'll try to run through them quick.
10 There's a chapter in the report which discusses 11 other B&W Owners 4 Group reports which were incorporated into 12 SPIP that were not.directly part of the SPIP, Program per se.
llh 13 And also included a section or parts of 1154 which the 1154 14 Task Force report which didn't seem to easily fit within the 15 nice neat categories for the system review . It's kind of i
16 in a section of quote "others."
17 As part of earlier efforts in the Owners Group to 18 decrease the reactor trip--the Owners Group proposed--thay 19 sent in two topical reports dealing with raising the high-20 pressure reactor trip set point back to its original design 21 values of about 2353 psig, and to raise the so called arming 22 threshhold for the anticipatory reactor trip on turbine power.
23 to about 45 percent power, or 40 percent power.
() 24 Those two actions were based on operating history 25 which indicated that there were some transients wherein they Heritage Reporting Corporation imum
516 I could probably have written out the transients without having D>.-
s 2 these anticipatory trips or determine whether they had had 3 several that occurred at about 20 to 25 percent power level'.
O
-\ 4 And the high reactor trip was back within its original design 5 basic for the plant.
6 143 looked at that, not really under SPIP, but as 7 part of just the earlier topical review effort. We said, 8 yeah, they.should reduce reactor trips associated with those 9 by I'think it was about 20 percent of the trips at that time 10 or 10 percent of the trips they'were experienced from these
- 11. -could be avoided by making these changes. And the design
~
12 criteria was basically used n the regulatory criteria was llll 13 do not end up actuating the PORV basically for design basis 14 transients and accidents and they"did a study wherein they 15 showed that previous conclusions about the opening of the 16 PORV as a result of transients or the frequency of opening of 17 them would essentially not be changed.
18 On that basis we endorsed the changes, approved the 19 topical reports and since that time we have in fact approved 20 tech spec changes for all the B&W plants except the Davis 5
21 Besse which we just got recently changing theirs and we're I
f- 22 in the process of reviewing that now.
23 Another report that was included in SPIP was a 24 report which had some recommendations to decrease unnecessary 25 turbine trips. These dealt with a variety of changes to the Heritage Reporting Corporation l
< mn-l
A^
9 .
517 l ~ electric' hydraulic control system and some other miscellaneou.>,
2 items to include the performance of the stock valves, et 3 cetera. Those all appeared to be reasonable to us. They g
(l 4 would basically just improve the performance of the turbine 5 system as opposed to having any big safety implications, so 6 we endorsed that report.
7 The last one was the 1154 Task Force report which 8 was primarily the valve recommendationc or some of-the valve 9 recommendations which Gary Hammer talked about earlier today.
10 And that was it on that one.
11 The next subject I want to~ talk about was the 12 reactor trip initiating events review which was performed by llll
~
13 the Owners Group. The Owners Group had a couple of activities 14 associated with reactor trip initiating events. On the last 15 slide on the previous subject you saw that they were already 16 taking some actions before SPIP to reduce reactor trips.
17 Earlier on in the SPIP program they initiated some 18 tasks to get an idea as to what were the systems that appeared 19 to be major contributors to reactor trip frequency and to take 20 a look at those systems as part of the overall system review.
21 Late in the SPIP program chronologically speaking 22 they decided, gee, maybe we ought to go back and concentrate 23 some efforts on those initial reactor trips, analyzing the 24 root cause and see if there was any additional recommendations
)
25 that could be derived from a detailed review of the Heritage Reporting Corporation non.
~- -
lk-1 518 I initiating events that would minimize or tend to reduce the
'> 2 number of reactor trips.
3 A lot of the early SPIP efforts were devoted towards km/- 4 . reducing post trip complexities.
5 So they went back and reanalyzed--I think the' number 6 at that time was probably 235 I think it was reactor trips 7 from 1980 to 1986. And they came up with recommendations 8 above and beyond those which were already identified by the 9 SPIP p1 u, ram dealing with turbine trip system, main feed sys-10 tem,'the ICS, control rod drive system,.and a series of
.11 recommendations to improve' transient response pre-trip.
12 These dealt with i$ ems such as lo'ering.the w low-pressure reactor coolant 1 system, pressure trip set' point'for the lllh 13 14 reactor protection system on the Davis Besse plant, some 15 actions to improve the effectiveness of the pressurizer 16 . spray to tend to mitigate the initial response to the system 17 for a reduction, for example, of feedwater or any transient 18 which would start you towards an over pressure condition.
19 There are about seven of those.
20 We concentrated our review primarily on whether or 21 not any of the recommendations appeared to have negative 22 safety impact. For example, if you wanted to remove a re-23 actor trip of some sort, would that be a safe thing to do?
() 24 That was kind of the focus of our review.
25 Generally we found that the recommendations they Heritage Reporting Corporation
< = -
l (t 519 L
1 came up with were acceptable. There was one recommendation
(]-
A- 2 which the Owners Grcup developed and rejected dealing with 3 when they replaced components, electrical components, or
((-)/ 4 modules for like the ICS, they'd had experiences where they 5 put in the replacement module and the thing tripped again, 6 because the module itself or the replacement' module was de-7 fective. And they recommended--there was a recommendation 8 to check out these replacement components before you put them 9 into place. They rejected that as not feasible or not very 10 cost effective. We thought that was just a reasonable thing 11 to do and asked them to reconsider.
12 The other item we noted was a lot of those transient lllg 13 improvement recommendations while on the surface appeared 14 very valid, it didn't look like there was much of a safety 15 assessment associated with those,' so we basically said, before 16 you put those in, please assure that you do appropriate 17 safety evaluations before you implement them. Things like 18 lowering the reactor protection pressure trip set point would 19 have to have the safety evaluation done by the plant just to 20 assure that they have no negative safety consequences because 21 the evidence wasn't there in the reports.
22 That's all I have.
23 MR. WYLIE: Any questions? No.
24 Byron Siegel.
25 MR. SIEGEL: We were asked to address two areas.
Heritoge Reporting Corporation
(= -
h) .
520 1 One of them was the B&W Owners. Group Programmatic and Manage-f) And I!m just going to go over this briefly 2 ment Actions.
3 because there's really nothing that's really terribly signi-(~- 4 ficant in it.
5 The Owners Group developed a generic. program to 6 look at the root cause process. It included evaluation 7 methods, initiating conditions for doing the root cause de-8 termination process and guidelines for the evaluation.
9 I believe the utilities in concert with the Owners 10 Group approved the process that was developed ~and the staff 31 believes that the process should improve the way the owners 12 evaluate transients and identify causes of equipment failure.
(")
(-
13 They have a transient assessment program report Several. improvements to'their.' transient assessment 14 upgrades.
15 program have been implemented. In addition, the Owners Group l 16 has stated that in response to som'e of our concerns about 17 human factors, that they would address human factors in future 18 events, for future events. And the staff believes that
[9 these improvements should improve the usefulness of the 20 transient assessment program, f 21 There were two boards that we discussed earlier.
22 One was the Advisory Board and the Safety and Performance O 23 Recommendation Integration Group. The Advisory Board we be-3 4fe provided added assurance that all important aspects
(]) 24 25 addressed in the program would be accomplished and provided Heritoge Reporting Corporation mmm
f.- -
l 521
~1 recommendations for program improvement and-implementation of 2 recommendations.
3 We feel that the processes utilized by SPRIG for 4 prioritizing and grouping the recommendations and assessing, 5 tests and high priority recommendations for completeness is 6 also acceptable.
7 As I previously mentioned, the staff during its 8 audit of B&W Owners Group determined that neither of these 9 groups diluted the effectiveness of the recommendations in 10 any way.
11 -The SPRIG also identified'or had a method or de-12 veloped a method for identifying what they considered key 13 . recommendations, and the staff--we looked at their approach
(}
14 and thought it was acceptable. We didn't go through and try ;
to specifically identify recommendations that we didn't feel 15 16 were key. We just took' essentially on face value based on 17 the criteria they used that the key recommendations were in-18 deed key.
19 But we identified independently eleven recommenda-20 tions as I previously mentioned that the staff also feels 21 have safety significance. And these key recommendations and
- 22 the eleven that the staff has ioentified will probably be the i
! 23 nucleus from which we'll pick the recommendations when we do 24 our audit.
l O 25 This I think we covered before, so I'm not going to
{
l Heritage Reporting Corporation 1
msu
'Y cA 522 I I belabor it. The Owners Group has submitted to the staff a I) 2 letter that identifies their evaluation program and we es-3 sentially concur with the fact that they are having an evalua-
.s
) 4 . tion program and with the scope of it.
5 But we, as I said, are doing our own independent 6 one so we're.not too concerned about it.
7 The other thing I was asked to address was the con-8 cerns identified by one of the staff members, and the only 9 reason that these were flagged was because Demetrius sent a 10 letter, as you are aware, to both the NRC Commission and also 11 to the ACRS. And we were asked to essentially address his 12 concerns. The staff isself had concerns. We've identified
,llgg 13 it during the program' . These were concerns that he specifi-14 cally-identified.
15 The first'one has to do'with the potential lack of ,
16 independence of MDR associates performing a sensitivity study 17 for the BWR Owners Group and related work for a B&W plant 18 owner. And the staff feels that we didn't ask them to do an 19 independent--that that work be independent nor did we believa 20 it was necessary that the study be performed by an independent 21 organization so therefore we didn't have any problem with it.
22 The second one dealt with a potential lack of in-23 dependence at Science Applications International Corporation 24 performing similar work as a subcontractor for the NRC and for 2s the B&W plant owners. This was reviewed by our Office of Heritoge Reporting Corporation (set) utsees
h.
523 1 . General Counsel and they felt there was no conflict primarily 2 because at the time that that work was initiated for the-3 Owners-Group, the-staff program was essentially done. There
(~?D NJ 4 was only about two or three weeks. overlap. It was essentially 5 complete. Oak Ridge, who was a subcontractor, did u; evalua--
6 . tion of their assessment of whether there was any conflict of 7 interest with regard to.the person that was involved in the 8 work and they be.li.eved it wasn't. They have since tightened 9 up their requirements so that this sort of thing won't happen.
10 But they did an investigation and study and determined that 11 there wasn't any-effect of that two to three-week overlap 12 because the work was essentially complete. Our legal depart-llll 13 ment agreed with that. , ,
14 MR. WARD: The NRC work.you're talking about, was 15 that reported on today by Mr. Debor, or whatever his name ,
16 was?
I
! 17 MR.- SIEGEL: Human factors'. 'No, it wasn't, it 18 wasn't.
19 MR. WARD: What work are you talking about?
20 MR. SIEGEL: Good question.
21 MR. JONES: I believe it may have been some FMEA 22 work being done by NRC as part of the A-47, 23 MR. SIEGEL: That's what it was.
(} 24 MR. JONES: I guess SAIC did some of the FMEAs I 23 for the Owners Group also.
Heritoge Reporting Corporation
<=n a .
\. .
g .
524
.1 MR. SIEGEL: Yes, I think you're right.
l)' ' 2 'There was another item. The third item had to do 3" with a premature finding-by the Owners Group of the adequacy
'4 .of the ICS NNI. 'The Owners Group assumed the system can be 5 modified and therefore did not consider replacement at this 6 time. -The staff believes, as was discussed by Rick Kendall' 7 yesterday that the existing system can be modified and that 8 there wasn't a premature finding.
9 The Owners Group in adcition is planning or looking 10 into an advance system, so they haven't totally abandoned 11 looking at new type systems and they are looking at. those as 12 a possible replacement.
The owners Group or utilities did not analyze (llh 13 14 effectively the proposed'SPIP recommendations to determine 4
15 the effects on other.lparts.of the plant.
s s 16 The Owners Group performed some systems interactions 17 studies; however, many of their recommendations are general 18 in nature and the staff requires specific evaluations. Both 19 the Owners Group and the staff intend to evaluate these as-20 pects during the audits. I didn't mention that, but there 21 were two old items that we were going to look at during the 22 audits, l 23 One was which Rick Kendall mentioned yesterday was ,
24 on the 7927. INE Bulletin 7927. As part of our review, we
(])
t 25 are going to, at the same time, look at whether or not they Heritoge Reporting Corporation ,
i
<wn man
~ -. -- - , . _ . . - . _ _ _ _ _ . _ _ - - - - _ , _ - . , _ - - - - . _ . , - - _ . ~ . . . . , , - - _ - - - ,
. ~ .. . . .
y 525 I satisfy the requirements of that bulletin and also determine
() 2 if'any of the modifications that are made, there are any 3' system--interreactions with those systems as part of that 4 effort.
5 MR. MICHELSOh- Is this an A-47 kind of safety 6 implication? This is a non-safety system.
7 MR. SIEGEL: No. It's strictly looking to see if 8 any of the modifications have'any interactions with other 9 systems.
10 MR. MICHELSON: Well, that's what A-47 is all about, 11 safety-implications and control system.
12 MR. SIEGEL: Yes, but it isn't just related to 13 control systems. It's across the board for all the recommenda.-
(
14 tions.
15 MR. MICHELSON: Oh, okay. Some of them may not be 16 control systems.
17 MR. SIEGEL: Not necessarily, no.
18 MR. MICHELSON: Okay.
19 MR. SIEGEL: It would have to do with all the sys-20 tems that are involved.
21 MR. MICHELSON: So that's what you mean by determin-22 ing the effects. You are going to look for the safety impli-0 23 cations of whatever changes they wish to make.
24 MR. SIEGEL: Exactly.
23 MR. MICHELSON: Okay.
Heritoge Reporting Corporation (300) 636-a000
db 526 i LMR. SIEGEL: The SPIP' concern _was the catasttophic- .
G
- = 2 _ reactor pressure vessel failure-resulting from an overcooling-3 transient assuming one control rod stuck out and it returned O 4 to_ criticality. The conclusion on this was the reactor t -
5 ing temperature requires the reactor return to criticality 6 was too low. It was approximately 35 degrees F for it to be ,
considered a probable event, so as a result we didn't consi '
8- der it any further.
9 MR. WARD: I guess I don't understand that about 10 tne rod, but the overcooling transients in general--I mean l' Brookhaven--the reason Brookhaven came to the conclusion they 12 did about their not--that these transients don't introduce ,
a bunch of core melt risks was because the bulk of transients ll h 13 14 Are overcooling transients.'
15 MR. SIEGEL: That's right.
16 MR. WARD: 'Anl,so they must be concluding in there l
l 17 somewhere that these overcouling t_ansients don't lead to 18 PTS problems.
l 19 MR. SIEGEL: That's righ', yes. This one bra to ,
20 do with--this is something they addressed that's so broad V
They didn't, did they, Bob?
21
) 'n their analysis.
2' "3 WYLIE: How did they reach their conclusion? 1 3
kd?
- 0. J,0NES: The PTS conclusion? t r
14 *w. WYLIE: Yes.
(~} '
(- 25 AR. JONES: They did it as part of the risk l
Heritoge Reporting Corporation
- (Bet) H6 400 l
l
.v . ;
??
527 ;
I evaluations. They did look at PTS wnere it's associated :
i- 2 with these transients. They estimated the temperature'you 3 can get to. And using some of the similar type of_ vessel !
4 failure probabilities based on those' temperatures that were 5 under the PTS, used as part of the PTS, and came up with a
- 6 numerical value, more a core damage status associated with 7 PTS risk, like 10-7, 10-8 type numbers'. If I remember right.
8 MR. SIEGEL: This was looked at-without a stuck 9 rod and no bore and injection,-and then with a stuck rod and 10 bore and injection, and the temperatures for both cases were 11 so low that you'd never get there on the cooldown.
12 The sixth item was operator burden study, does not h 13 address operator errors. The operator burden study did include 14 those and those are addressed in our human factor section,
- 15 Section 7 of the report.
16 The sevedth item is B&W plants violate off-site 17 dose limits under the condition of steam generators tube 18 rupture design basis accident.
19 MR. WYLIE: I'm sorry, Byron. But sixth you wrote :
20 off pretty easily.
4 f
21 MR. SIEGEL: We did address i t. . l i
22 MR. WYLIE: Well, I bet I could think of errors of 23 commission that you didn't address. I mean I hate to see you ;
24 write that one off. How do you decide what errors of commis- I 25 sion should be analyzei and what one shouldn't?
Horitoga Reporting Corporation me>m m .
< - - , - - , , ~ - . ---..--_.-.-, -. . , . -
_ - _ _ . , , , _ -, __ _ Y
528 >
-1 MR. SIEGEL: I don't have the human factors-- !
['\
- . 2 MR. WYLIE: Maybe I can't think of them, but Carl 3 can think of them, I'll tell you that.
.(')
\/- 4 MR. SIEGEL: I don't think we addressed all of them.
5 We addressed some of them.
6 MR. WYLIE: And'that's what I'm wondering. How
^
7 was that defined? I think you have to do that in some way 8 .but I--
9 MR. SIEGEL: I guess I can't answer that question.
10 The Human Factors people aren't here and I'm not sure--Bob, 11 do you know thc. answer to that?
12 MR. JONES: No, I don't remember.
P 13 MR. DAVIS: There's a reference to SER Section '/52.
h {
14 MR. JONES: Yes. But what it identifies specifi-
!$ cal,1y--what's in there is. pretty. general in nature. It won't 16 address what you'are^asking. I think it's pretty broat.
17 MR. WARD: Okay.- All'right.
18 MR. SIEGEL: . Item 7, the B&W plants violate on-19 site dose limits under the condition of the steam generator i.
20 tube rupture design basis accident. We didn't address this 21 because it was being addressed by Generic Issue 67.
22 On Item 8, the main steam line break in steam 2f generator tube ruptures when they are created as ene design i
24 basis accident for all PWRs. Again, this is under considera- t 25 tion as part of USIA 3, 4 and 5 regarding steam generator Heritage Reporting Corporation
<= p r
, - - , . . - - . - - . _ _ _ _ _ , , . _ _ ~ , . _ ,
<, l 4 -
.v.
529-I tube integrity,.so we didn't address that item.
2 Item 9 was certain failures in the turbino control. '
3 T o d e l a y t u r b i'n e t r i p a f t e r r e a c t o r t r i p c a n r e s u l t i n a r e -
- 4. ' actor trip .without.' a turbine generator trip. -
5 -This was generic to all'the_PWRs. There-was a memo 6 sont to Research' suggesting that this be proposed as a generic
- _7 issue by the originating staff member. . This in fact-has been
, e
, 8 done and he has sent'it over to the Generic Issues Branch 9 to prioritize it and determine the risk associated'with it. -
10 That's all I've got. Does ar.ybody have any ques-11 tions?
12 DR. REED: In my background I'm just a bit surprised [
g 13 on' Item 9 that that is generic. In my experience I don't know 14 of a trip situation where reactor trip is not accompanied by ;
15 ;
a turbine generator trip unless.- there's a ' failure in the cir- f 16 cuit. It's designed in in all' those that I know of.
17 MR. SIEGEL: ,
There was a study that was made--let-18' me see if I can lead my scribbling--it apparently is valid !
19 for all PWRs at the end of -- or'the end:of the fuel cycle.
20 There was an I&EL study that just came out April of '03 and 21 the study indicates that you can have recriticality both with 22 or without a--
O 23 DR. REED: Well, that's a trip.
~
{
24 MR. SIEGEL: That's a trip. I'm sorry. That's a l
25 trip. You're right.
Heritoge Reporting Corporation mm man
~'
T t *' '"""~~'v'" 4 ' ' " ' v *vi ~ ^ ~ *'*-'V ' ' ^ - '
'"-'"i~~ r = ~ ~ 'w ~'
w=
,.yJ. ;
i .530 3 MR. JONES: 1What's the question again?-
2 DR. REED:- Mell, I was-just wondering--in'my ex-3 perience I didn't. realize that it would be generic because
(~)'
'- -4' I just-don't know that--my reactors--reactors that I've been 5 familiar.with, the turbine always gets trip signals if the ;
7 MR.' JONES: Oh, right. And the question is what-8 happens if it fails. And--
9 DR. REED: You are just saying automatically the to circuitry fails, is that it?
i 11 MR. JONES: Yes. We had an experience just re-12 cently at Crystal River wherein that trip signal failed. .
llll 13 The trip signal got there but the solenoid failed and Crystal l 14 River's turbine didn't trip. It was the consequences of that is failure. Similarly on the Westinghouse plants, there is a ,
16 delay in the turbine Sor:about, thirty seconds.
17 DR. REED: Not on. Westinghouse plants I'm familiar 18 '
with.
19 MR. JONES: Well, there are some that have them. 3 20 Either Sanofre or Diablo, one of those have it.
21 DR. REED: A couple of weirdos, huh?
22 MR. JONES: There are some that do have that system.
- 's 23 MR. SIEGEL
- That study that I was mentioning was 24 related to this. And it was to the return of recriticality 25- because you don't get the turbine trip and you get a cooldown. }
i ;
Heritoge Reporting Corporation (300) W4000 i
o .
. f'E '
531
~
l' -And as a result it indicated that the problem was existent s/ 2 both for' Westinghouse and CEN and B&W plants, and for all r
3 Lthree types of plants. .And with or without--they addressed !
,-' 4' it both with and without a stuck rod.
5 MR. DAVIS: But Item 5 says you have to get clear 6 down to 35F for the' stuck rod to get recriticality, i 7 MR. JONES: Yes, sir. ' nd this is something that i .-
8 .just came in--this INL report'came in--either Monday or Tues-9 day I got it and I saw that they had gotten it and I called 10 up the Research guy--there was a. note on the copy that'says, 11 - "What's going on here?" Are they'using different numbers than I L
12 the staff used when they did the calculations? And one or :
i lll) 13 both of these are in error.
14 The information we had when we wrote the SER with ,
l 15 respect to the recriticality issue was that you had to get !
, i l
16 down to about 35 degrees with a stuck rod--
l 17 MR. DAVIS - That's with no bore on injection in the
.i 18 blind. ,
i .
19 MR. JONES: The 35 degrees with a stuck rod in- !
20 cluded the bore on injection.' Without the bore on injection 21 you had to be, I think it was around 300 degrees. ;
22 MR. DAVIS: Okay. That makes more sense.
i 23 MR. JONES: But because the issue of the catastro- .
phic failure was related to also filling back up the pres-24
(
l 25 surizer, getting cold, and then going back. And it may be ;
- i
! Heritage Reporting Corporation I
<m> mm I
u,, . ..-- - - . . . - - - . , - , - - - . . - - . . .- . . - . . - -
[ .
532 1 the amount of bore on was different between that study that 2 Lwe referenced in our resolution of Item 5 and what's coming 3 up out of this recent study on the turbine trip on the re-r
\ 4 criticality issue, t
-5 So I want to try to focus that anyway.- I-don't 6' want'him to bring it up'because I wasn't sure whether.the d
7 numbers were right, so I wanted to delay it.- I didn't under-t 8 stand where they came up from. .I just. glanced at it and 9 noticed the' inconsistency.
r 10 .MR. SIEGEL: But it may be due, you said before, i
11 to the reactivity margins, the differences they are getting 12 in reactivity margins-- ,
llll 13 MR. JONES: It's just not clear.
14 MR. SIEGEL: Because there's probably several hun-15 dred degrees difference between this number and what this 16 INL report--
+
17 DR. REED: And as'you do'your investigation, recog-p 28 nize that many plants have main steamline stop valves that 19 may also be tied into the turbine' trip circuit breakers. So :
20 if the turbine doesn't trip, the throttle.. wells don't go 21 down or something, the main steamline stops may close.
22 MR. JONES: The INL study includes basically three
() 23 cases. It's only done for the P&W plant. It includes three i
i 24 cases. Basically one where you don't trip the turbine and 25 the line stays wide open. Another case where they modulate [
Healtoge Reporting Corporation
- w. .
+ _ . , , . . . . , . .- -, _ _ _ _ , , _ - - , . . _ . , , _ , , _ . - - _ _ - . _ , _ , . - , . . . _ _ .
- r- --
N Qb:!! . e
/.
533 .
I the control valves to determine shut with the ICS runback I
,O ' 2 at 20 percent a minute. Another one.where we looked at con-l 3 trolling the control valves to maintain pressure in the steam ;
O'
=4 generator.
5- In the ICS functions, these things are not near
- 6 as severe _an event, and so really the' turbine trip or the 7 reactor trip without turbine sequence, it gets the worse 8 temperatures, appear to be 10-7 sequence. 10-7 to 10-8 se-
.t 9 quence, if I remember the numbers correctly, i
10 So they looked at several cases and they tried to 11 account for closure of the MSIBs, also result'in some of the I
12 quick studies.that have been done, j This was just information to be used to help priori-l llll 13 14 tize the issues to determine whether further study is war-15 ranted from a safety. standpoint. j i
16 MR. SIEGEL: Yes, that report was sent over along j n
17 with the request to prioritize--
. 18 MR. WYLIE: Okay. Any other questions?
. 19 MR. JONES: We'd like~to make a couple of closing l
' I 20 remarks.
I 21 MR. WYLIE:Okay. l l
22 MR. JONES: Over the last two days, and they've i
-)
s_/
23 been a long two days for evirybody--there's been a lot of l i
24 comments with respect to the scope and the overall reassess-
} t
. 25 ment program. And some of my comments may be redundant to ;
i Heritoge Reporting Corporation ose)** =
~--~,y=---..,,-,m----, m+-r ,-n a ww- , --,,-n--,- -- -r,w w,y w-,m-- - - - . - - - - * , .<--,+q ~ -w y
+ -
k+,: *
.534 I what.I've . aid earlier, but we would like to offer.the overall 2 _ staff vieJ. of this whole project.
3 First of all, the staff developed its plan in 4 about March of86 in response to the EDO letter to try to 5 address all the issues that Mr. Stello identified. These ,
6 .were-provided both to the EEO staff and the NRR management
~
l 7 and to the best of our knowledge did indeed address all.the 8 issues that we were told to address.
9 In addition, we did discuss the plan with you in b
10 the June '86 dubcommittee meeting which provided some comments 11 on--we incorporated your comments with the exception of the i 12 management-related issue items we discussed earlier today. '
lllg 13 The primary focue of the program was B&W design i 14 specific issues. As such, we didn't review external events. -
15 He think.that is a very plant-specific issue that is not an 16 NSS design specific issue. It's just as valid for any given 17 plant out there and there would be no way to do it on a 18 generic basis for the B&W plants and the NSS design itself >
19 does not have a substantial influence on that.
20 Although I don't think this comment was made 21 carlier, there was some comments made by Mr. Tackie about -
22 whether or not we looked at two-phase design characteristics 23 and transient response to accidents of these plants. We did 24 look at it. It was one of the things we did as part of the 25 program was going back and looking at the old NUREG documents. ;
. Heritoge Reporting Corporation ,
<== -
yf, :
535 g 'And looked at the old discussions on two-phase flow issues y 2 and' decided whether_or not that had to be reresurrected as
-3 part of this overall process.' _And:as is discussed at least j O .. the eection on the sme11 steex tocA, in the Section s of the report, we did look at the two-phase natural circulation and -t 5.
6 identified it as a unique characteristic of the plant, but 7_ did not;see thatlas a safety problem that needed to be ad- .
8 dressed additionally at this time. .
, 9 So that in~ general we looked at it. We didn't think that it had to be resurrected. It wasn't the problem
. 10 that we. received today with the plant. !
33 t
- 12 The emphasis of the program was indeed placed under g 13 the operating history. That's what led to this whole letter.
34 And in light of the operating history since the TMI modes, 15 which we thought it would have affected the behavior of these plants, that it more than appeared that it has. As t 16 37 well as rereviewing sensitivity in light.of some of those 18 TMI modes. And looking at the ri'sk>where we had on board 1
w to' determine to.some extent the overall state of the B&W
?
r g,
l , h l 20 plants versus other PWRs. ,
l j 21 In summary, the conclusion we come up with is the .
22 program was responsive to Mr. Stello's letter. We think the O- program has indeed examined the D&W unique design characteris-23
- - tics. We think in review of the operating history and the 24 1 .
( 25 sensitivity study and the risk work that was performed, we i
Heritoge Reporting Corporation (988) M
,- __ _--,,_._.-m . . _ ~ _ , _ - _. ._,._,,_,,,..m-.-,._,,___-.,m,-_.- . - . . . _ , _ _ _ . , _ . _ _ . , . .
W 536 I believe that the owners Group program and the extensions that f'}.
L 2 the staf f has performed that we felt necessary. have confirmed 3 that the proper systems'were' examined and dealt with as part 4 of the program.
5 . We've also'we beleive have confirmed that the over- ;
6 all-safety of the B&W design at least as related to the NSS-
- 7 specific design characteristic has been reexamined and is not 8 significantly different than that of other PWRs.
i 9 Finally, we believe that the implementation of the
, ' 10 recommendations will indeed result in improved operational 11 performance and safety for the B&W plants.
{
12 We thank you for the review that you've done today g 13 and yesterday and we understand some of your concerns on i
14 the scope.
15 What we would like you to consider when you deli-16 berate any letter or recommendations you make to the full I 17 committee is try to keep them like the operating experience 18 which brought to light this program and whether you think l i
r 19 we have done an adequate job in responding to that operating
{ 20 experience.
21 (Go to next page.) '
i 22
.O i
23
[
i- !
O
25 l
Heritoge Reporting Corporation
<=i u. ,
.,--n ,-- n,, -w.,- ,,--e.,e---- . , . e- - - - - . y -- ,w.- ,,y-, - - , , ,-n v -- -,m y- x-- , .
- -17 537 1 What we would like you to' consider now--is any.
N 2 recommendations'to be made'to the'. full committee and to~try 3 and keep in mind the operating experiences which bro'ught to 4 light,this. program and questions as to whether you thing we 5 have done an adequate job in responding to that operationg 6 experience.
7 MR. h"lLIE : I would like to'thank the staff and i
l 8 the B&W Owners Group for their participation and suppory 9 in this meeting and I think we can close the record at this to time and consider a couple of items.
in The services of the reporter will no longer be
.12 required. I 13 (Whereupon at 5:00 o! clock p.m., the subcommittee 14 record was closed.)
i 15 !
16 i
i 17 18 !
19 i
20 21 22 :
O 23 O
25 l i
Heritoge Reporting Corporation ,
<=o .m. ;
f
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 Babcock 8 Wilcox Reactor Plants 6
7 Docket Number:
8 Place: Washington DC 9 Date: May 4, 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 tranneript is a 15 true and accurate record of the foregoing proceedings.
16 / sin >A>, M/
17 (Signature typed): win toffe ry 18 Official Reporter 19 Heritage Reporting Corporation 20 21 22
! 23 24 I
25 Heritage Reporting Corporation (202) 628-4888
.s
'( _
l O NRR STAFF PRESENTATION TO THE ACRS ,
i i
SUBJECT:
SYSTEMS REVIEW, MECHANICAL ASPECTS OF MW, AW/EFW, SF.CONDARY PLANT RELTEF, INSTRUMENT AIR DATE: May 3, 1988 PRESENTER: witit.,,teF,,,
PRESENTER'S TITLE / BRANCH /DIV:
Senior Reactor Engineer (Nuclear)
Plant Systems Branch Division of Ene,ineerine, and Systems Technology PRESENTER'S NRC TEL. NO.: 492-o862 SUBCOMMITTEE: B&W Reactor Plants O
O ;
PM' SYSTD1 1
o STAFF CONCURS THAT PROPER IWLDENTATION OF ECOWD0ATIONS WILL l PEDUCE FE0 0F FM'/ CONDENSATE INDUED EACTOR TRIPS AND SEVER!W j OF TRANSIENTS o REC 0ffE0AT10fS ADDESS DESIGN, P%INTDANCE AND TESTING ASECTS AlWD AT IWPOVING ELIABILIW o STAFF RECatBOS ENTERING INTO THE RTS TE RD%1NING LNPPOCESSED ECQtB0ATIONS (WW) FRm THE TRIP INITIATING EVENTS o TE BWDG EC0ft00AT10NS AE ESPONSIVE TO TE WR SENSITIVITY SR0Y (APP.P) 0 C0 TENTS IN BWDG OPERATOR BURD91 REPORT (APP,S) SHOULD BE OnNSIDERED l
l DURING IWLDENTATION ,
o RECOME0ATION FOR DHANCING RELIABILITY OF PFW/CONDE? SATE SYSTDS AND CONTRDLS SHOULD BE ADDED TO EYLIST l
O l
- e. .-
l J' ,
O B0P I E RTANCE o POOR B0P PERFORMANCE INCREASES RISK ,
o ffW/ TURBINE CONTROL PPfELEPS ARE MAJOR CAUSE OF BOP ,
REl.ATED-TRIPS /PLAhT TRIPS o LICENSEE's PPACTICES/PERF0WANCE VARY CONSIDEPABLY IN BOP APEA o PLANTS WITH AGPESSIVE B0P PROGRAMS WILL HAVE HIGHER PLANTAVAILABILITY o ECONOMIC INCE}'TIVES F0P PEllABLE B0P OPERATION HAVE l O POS m VE AFFECT W SAFETY t
I I
r i l
O ;
t i
h
4 a
O
B0P EGULATimS o POLICY ISSUE, SECY-86-349, NfN ?l,1986 DESCRIBES B0P PROGRNE/ SAFETY SIGNIFICANCE & POP ITDE
- F E0VENCY OF O MLLENGES
- EFFECTS ON SAFETY ELATED SYSlDE/0WRATOR CONTROL o STAFF REVIEW IN PAST FOCUS m IFPACT ON SAFETY-RELATED SYSTDE
~
o TODAYS FOCUS IS ON FE00ENCY OF BOP GALLENCES/0PERATOR CONTROL o TDPOPARY INSTPUCTION 2515/83 B0P TP.lAL INSPECTION PROGRNiCOPPLETE o TI EXID OED WHILE INSPECTION PROCEDURE ISSUED F0P O cor'nT o RESUI.TS OF Tl SHOED UTILITIES APE CONCERNED WilH BOP o TAP TO ASSESS SAFETY SIGNIFICAN& OF B0P AND DEVELOP RULES, EG, OR GUIDANE IF NEEDED
- CCt@ACT IN PLAE WITH SAIC TO PERF0Pt1 STUDY OF PROGP/PS
- PAINTENANE POLICY / PULE EFFECTS ON B0P O
. ., T~ ~
w O
AFW SYST E 1
o VEPIFICATION THAT REC 0ft9DAT10NS DO NOT CONFLICT WITH NRC RULES, PEGS, OR GUIDEllES o PAW ECftfDTATIONS CALL FOR ANALYSES / EVALUATION BY [fTILITIES, STAFF CAtNOT ASSESS ACTUAL BENEFIT LNTil IPPLENNTATION N0 IttEDIATE ltPACT o STEERING C0WITTEE HAD NOT ACTED (N PAW 0F PEC0ft90AT!(HS ,
IN APP,0 REl.ATED TO DESIM/ TESTING OBJECTIVES o BWDG EFFORT DID NOT ADDPESS THE ADDIT 10fl 0F A THIRD AFY PtFP.
STAFF IS PURSUING 1111S VIA GI-124 o IF PROPERLY IfPLB O TED RELIABILITY / AVAILABILITY OF AFW SYST B l
O SHOULD BE ENHAf!CED TO VARYItE DEGREES :'
i c' STAFF RECat00ED WEP IN Mif0 PEJECTED BWDG PEC0ftBOAT10N REl.ATED TO RESEARCH PROGRAM FOR TUPBIE 00NTPOLS SHOULD DESIPED tFW PERF0PPANCE NOT BE ACHIEVED
< l l
f 1
i O !
r
1 O
SECONDARY PLAf.T PELIEF SYSTEM o CONSISTED OF MAIN STEAM PPESSlJPE CONTROL PEVIEW TO RFRICE MSSV .
ACTUATIONS FOLLfMING MAURB TRIPS o TESTING /MAIFENANCE/PEPF0PMANCE OF MSSVS RELATED TO VALVF TASI' FORCE EVALUATION o RAISING FESV SET POINTS CONSIDERED PART OF VALVE TASK FORCE EVALUATION, tr)T ENCOURAGED BY STAFF o AFW RECnftENDAT10t's ARE RESP 0tSIVE TO CONCERNS IDBff!FIED DURIfG PPE5SURE CONTROL REVIS' o THE STAFF BASICALLY AGREES WITH FFASED APPPOACH Oo SwF PECmDS Bu CarlWE TO MESTIGATE PUNT NFICMS, ,
ESCALATE ANALYSES IF ECESSARY o PROPER ltPLENNTATION WILL IPPfWE PERF0 PRANCE OF RELIEF SYSTEM, POST-TRIP PRESSURE O M ROL AND OPERATOR BURDB1, AND ENHAhE PLANT S/FEIY a
0
O iNSTetren air SYSTEM o AIP SYSTEM INCLUDED BECAUSE FAILURE SIMILAR IN EXTEFT AND CHARACTER TO LOSS OF ICS POER o STAFF C0FPAED BWOG ECfTtENDATICNS WITH AF0D EPORT -
FAVORABLE RESULTS o STAFF AGREES WITH CONCLUSIONS AND ECCft90AT10t's o ECfftD'DAT10NS FOR AIR SYS7DS D0 PDT EQUIE EXTD61VE EVALUATION / ANALYSES BEFORE IPPLFM'NTAT10t o EC0ftDOAT10NS ADDESS ECENT ST/FF CONCERNS
- CLEANLifESS
- EERGENCY PROCEDUES
- TESTitti
- OVERALL IPPORTANCE OF IA SYSTD1 g
V o DOG SHOULD 00NSIDER ECCttD0ATlfti FOR GRADUAL LOSS OF A!R SYSTD1 TEST (AEOD REPORT) o "WDG SHOULD CDNSIDER ANALYSIS TO ASSESS IPPACT ON SAFETY ON LnSS OF IA DLE TO LOOP EV9IT o ADD FOUR OF 1A ECfftDOAT10NS TO KEY LIST
- INSPECTION FOR LEAPS
- TRAINING AND LOSS OF IA PESP0tSE PROCEDURES IfGPECT ACCltULATORS/ CHECK VALVES
- DEVELOP / UPGRADE LOSS OF IA PROCEDRES (3 PLJATS)
O
O NRR STAFF PRESENTATION TO THE ACRS
!MPRV E RE ENDA S DATE: May 3, 1988 PRESENTER: cary Hammer PRESENTER'S TITLE / BRANCH /DIV: Mechanical Engineer, EMEB, DEST PRECENTER'S NRC TEL. NO.: 492-0919 SUBCOMMITTEE: B&W Reactor Plants O
B a W REASSESSMENT - VALVE RELIABILITY IMPROVEMENT RECOMMENDATIONS O
l MAIN STEAM SAFETY VALVES l - IMPROVE PERFORMANCE BY BETTER SETPOINT PROCEDURES, BETTER RING ADJUSTMENT PROCEDURES AND BETTER MAINTENANCE
- MAY REDUCE CHALLENGES BY ENHANCING TURBINE BY-PASS VALVE AND ATMOSPHERIC VENT VALVE PERFORMANCE
- MAY REDUCE CHALLENGES BY INCREASING MSSV SETPONTS -
NOT ENCOURAGED BY THE STAFF
- FURTHER STUDY BY VALVE TASK FORCE (VTF) TO BE DONE -
NO REPORT TO DATE O MOTOR OPERATED VALVES (MOVS) AND OTHER POWER OPERATED VALVES (POVS)
- IMPROVE PERFORMANCE OF ALL SAFETY-RELATED VALVES BY ENHANCED TESTING CONSISTENT WITH BULLETIN 85-03
- FURTHER STUDY BY VTF POWER OPERATED RELIEF VALVES AND BLOCK VALVES
- IMPROVE RELIABILITY OF PORVS AND BLOCK VALVES BY TESTING CONSISTENT WITH DRAFT ASME STANDARD OM-13 CHECK VALVES
- IMPROVE PERFORMANCE OF CHECK VALVES THROUGH USE OF INP0 SOER 86-03 GUIDELINES WHICH ADDRESS DESIGN, OPERATION AND MAINTENANCE
- FURTHER STUDY BY VTF
O O O -
Human Factors Evalustion of B&W Plant Reassessment Program Prepared for:
ACRS Subcommittee on B&W - Reassessment Program May 3,1988 Prepared by:
Joseph DeBor Science Applications International Corporation 1
e .
O O O
! Background i
) o Human Factors-Related Elements of Reassessment Program
! Included:
1
'l I -
Task 1 - Operator / Maintenance Personnel Interview Project 4
Task 2 - Operating Experience Reviews Task 3 - Procedures Review l -
Task 4 - Operator Burden Project / Sensitivity Study i
o Review Conducted on Behalf of NRC by:
SAIC l -
COMEX Corporation Carlow Associates i
l 2 l
i
\ O O O -
l Task 1: Operator / Maintenance Personnel l Interview Project o B&W identified 11 hardware problems and identified concrete
! recommendations. ,
! o B&W also identified 6 human engineering problems, but no l concrete recommendations or follow-up actions were developed.
! Examples of problem neluded: ,
ICS' feedwater control of T-Ave is poor.
Difficult to tune secondary system at less than 100% power.
Delayed subcooling margin potentially confusing.
3 4
O O O -
Task 2: Operating Experience Reviews i
I l o B&W review was comprehensive.
- o Human engineering issues were summarized and characterized:
.i Human Interface i
i, Operations and Operating Procedures Surveillance and Testing i
l Maintenance i
)
Training Displays i
1 4
i
T O 5 s t n n t
el a t
np n o e ph n c o ma p oe c y m.
oc i gb cf i w i nd d c ne i
e l ves ap v os g s e v e r n n t n d R i s da i
l el b p a
s n e r e e r l at o
r c e r n u n w fot u d o e cI sb gN n O c o nN i oe t r i r/
t a a P eS d i
r eC nI n p e o 3
- i gf n o mp mp r
k e ss s n o o a a al c re e
T mn uo l r a s
h s a r n t
na y e o ni t al c p e gc a
i s i f i f i
n gd d Wido i
S a n
&m B
o o O
1 : ;jij j
O O O '
Task 4: dperator Burden i
The operator burden study findings are valid human engineering issues, and the B&W recommendations are appropriate, e.g., ;
o Control of steam and feed flow on loss of automatic control.
o Drastic operator actions.
o Overcooling mitigation strategy.
6
t 0 7 e
i n
h c
a m
- p p
n a
i r
i r
t t m e e y .
g n n e n i i d v i br br u
i s h u u t n c a g
. t t S e e n n y
h e r e o o t r r , p g i
p f
a m
i r n v m yo t i
l o
i t o c
l pr o r o o i
s t c c 0 n s
a i mlo a r
e e w s r e r v S y nn t
y o
d oo it c r f o
- u a e o y 4 t s dnt h t
a t k y p i e n l
i i s t c b a
i v mis i t
n a
T i
i t
s me og a b
o r
n c n e p e
s e a t a e rh n c W
Wc
&c e
i i
m d u
e a l E R B Bf r h
e e h n e
t T Ti - -
o o o
- i!!
]
O O O Overall Assessment of B&W Findings o The B&W studies resulted in valid human engineering concerns.
However, because human factors professionals were not involved in the reassessment effort, the completeness of the effort in identifying human factors concerns is uncertain.
o The proposed corrective actions imply very significant changes in the man-machine interface environment in control rooms; but they l are general, and not plant specific.
o The proposed corrective actions do not include implementation dates.
Recommendation I
j Include human engineering experts, such as those who participated in the
! development of modifications resulting from the recent control room design reviews as members of modification teams.
8 1
NRR STAFF PRESENTATION TO THE O ACRS
SUBJECT:
kl3k b$ "
DATE: May 3, 1988 PRESENTER: R }/ovP& blood '
O-PRESENTER'S TITLE / BRANCH /DIV:
GRovP 4 G AD E R / 8 N L_
PRESENTER'S NRC TEL. NO.:
P TS GGC 2 3 G3 SUBCOMMITTEE: asw Reactor Plants b y j'f O sc6 o sa(p l,
9 0 ,
A , _ - su - 2 - , w . a- .m ..
l O
RISK SIGNIFICANCE OF COMPLEX TRANSIENTS AT B&W PLANTS j
l O PRESENTED BY R. YOUNGBLOOD SAFETY AND RISK EVALUATION DIVISION DEPARTMENT OF NUCLEAR ENERGY BROOKHAVEN NATIONAL LABORATORY l
O
, . . - - , . - - - - - - - - - - - , - - - - , . , a,.,-,-__.-,n--_ ,---n-,---,-- , , , ,----,-- , - - m- - --e--- - --,
o PERCEPTIONS B&W PLANTS ARE NOT ASSESSED TO HAVE SIGNIFICANTLY HIGHER CORE DAMAGE FREQUENCY THAN OTHER PLANT TYPES YET B&W PLANTS SEEM TO EXPERIENCE COMPLEX TRANSIENTS MORE FREQUENTLY O
ISSUES IS PRA MISSING SOMETHING IMPORTANT ?
l IS B&W CORE Dt. MAGE FREQUENCY HIGHER THAN THAT OF OTHER CLASSES OF PLANTS ?
o
O ~
BWOG SUBMITTALS WV.._
l l BWOG SUBMITTAL ARGUES EVENTS OCCURRING @MIIT:'"
IN COMPLEX TRANSIENTS ARE OC '~1 l
WITHIN SCOPE OF EXISTING PRA MODELS !$sCs.
l O PROPERLY TREATED di..Y SO NO REASON TO DOUBT PREVIOUS 3. 3 t PRA CONCLUSIONS 2 l
j .
ALSO ARGUE SUPPORT SYSTEM FAULTS STILL. :, i-MOST LIKELY CAUSE OF CORE DAMAGE .,
O
m.
1t D).. e c
n es us qa -
el
. SC o C
o o o a .
C C a
C c c
o l -
T a .
v gt o n a m o ee LHH .
.X.
a
{ [ -
- - [ {
s _
e _
s .
l o 2 - . . _. = _
V c G.. .
A e .
SR _. . . - .
s .
e .
s e o 1
. e V l 'O. . .
R c .
r O e T
PH .-
t
{ n C e W
-. _. v E
s . .
V n e r H e .
e
. e 0 p t 6' O . a
_., w
- .- d un . e l
8 a e
oS 3 t.
_. F rC c -
d n cp 4t.i.t tn E
~
E. .
. f of cL .
COA P N
-. ~ _
. . .a. ,
. f o
n g
_.d a d i n
-t s
s d el
- u. - o
. e e o .
L el o
_ 4 F BC . . .
_ m i . g d- . . - - .
e e - { 2
_ t l- -. -
s y
l .
3
__ r o .
S eer - e sl t S_ - . r cel n . u et eo .
_. - - g S S HC . -
i "
~
o F e
1 1
o r 2 - -
Wn t E F o EC - .
t W E F
_ E r
f e
t o a
- snd w
T 2 5 sie oa e L MF ua,
o BWOG SUBMITTALS (CON'T )
THIS LINE OF ARGUMENT DOES NOT ADDRESS THE QUESTION OF FREQUENCY OF COMPLEX TRANSIENTS OBWOG EVENT TREE DISCUSSION VERY SIMPLIFIED EFW FAILS TOTALLY OR SUCCEEDS TOTALLY VS EFW FAILS INITIALLY BUT IS RECOVERED
- THERE IS A NEED TO ENSURE CONSISTENCY gF THE CORE DAMAGE MODEL WITH THE OBSERVED EXPERIENCE -
1 1
4 O
BNL APPROACH i
CONSTRUCT AN EVENT TREE MODEL WHICH PREDICTS CORE DAMAGE FREQUENCY FOR l
4 SPECIFIED SEQUENCE TYPES l
O PREDICTS FREQUENCY OF COMPLEX TRANSIENTS FOR SPECIFIED TRANSIENT TYPES
\
l LINKS THESE PREDICTIONS l
O .
. , - - - - - - , --c-- - - - - - ---
4-57 P
'd a
ijjjjj n g i !
._ i1 ..
l ijarfij l.
I ..*li!!
ilhest
- zis as Ejjj l l
= ili:!, , i 1 3; y- I -
i
. J -
- 13:!3 IIllji i
. a. J. 2 3 . 7. ~
J'i!
as;1 i i a g i
1 eiv I:.1 ,
s arij yi.25
~ : 211 l
) '.' I j ai
- c. . . 3
.1:1 lisi bw J gi 1-1
?-
i a,
.. g
%# 3. Y i 'l e 1/Ili* 5 5 J
f M'
ME a 3j)fs;I 4 T ] AD'-1 I,'121! I a
4
=
.. -J n
. .al t t3 g
f 3
g r j
- 2 ;l :
. a ;-
, w
'en*^
) !
~
j a;*1 sst i lM,A%w g
- 4 !
listin' 2 J 6 ,
- )1,!
- :
,I 3. '
= ;
= ,
i e h:.*%
f2.: f._ ---- - J t* w f
I*: 83 i d I **
[ __ __ _3
- - M.swya M 3 ja f.,t. -
- I T y L I1T ." .= ,! P :,
h g N kk NI
- I
. $ h d.
C ijf5} } } ~ l B
's .! -J d@
~
!)il
,, 1 i! 3 i
i
!. I .
.K9.,
i., L_ ____
4 ilj- , i f ,
A'
, = . >
, L --- ' 6 ' -
.a 11 l . 2
G
- ( f)
OCONEE UNIT 3 OCONEE UNIT 3 OISTRIBUTION 0: CATEG01Y 'C' EVENTS BY CATEGORY 'C' TYPE DISTRIOUTION OF CORE OffOGE FREGLO CY BY CRTEGORY 'C' TYPE Ex&SSIVE FV (mVEE)
E IDR. R 0b00Wt1 (MitG D SEID(. RodINN (SEVEE)
_=jih[
nz== si EEESSIVE FV (HitOl) I
- EqMff me f. W :
LiEl ?_
d('--
m Eif g i CESSIE W (HitDD RP + l unmai,o j _- W"' ? M gigg Tg5= ;"7$,i ,
i i; a r- -
T d,
sew. R&xm (mvoe
- ExassivE a (mveE) unmano I
CRYSTAL RIVER 3 CRYSTAL RIVER 3 DISTRIBUTION OF CATEGORY -C" EVENTS t DY CfllEGORY 'C' TYPE DISTRIBUTION OF CORE ORMAGE FREa B CY S BY CATEGORY "C" TYPE EX SSIVE FW (SEVERE) .
E00N. R0bOOW4(MitG) '
uom00aItG _
I V EXCESSIVE FW (SEVmE) ljb E'
g' g --
i -
DCESSIVE FV (HI!OU
=iMEA I h gyr gE f a mi! E! f MLj$ y -- g - -
y iqppjyy SEDi.. R0h00We .'HitDU
~ g,-
t UERGELitG
_71g re 4.8. _
1 l
O !
SIGNIFICANCE OF l OBSERVED EVENTS EACH OF 12 COMPLEX TRANSIENTS EXPERIENCED AT B&W PLANTS WAS ASSESSED TW EVENTS (DAVIS-BESSE - SEPT 1977 AND O
RANCHO SECO- MARCH 1978) ARE MUCH MORE SIGNIFICANT THAN THE REST THESE TWO INVOLVED UNDERCOOLING THE REST INVOLVED OVERCOOLING UNDERCOOLING HAS HIGHER CONDITIONAL PRGBABILITY OF GOING TO CORE DAMAGE lo 9
O CONCLUSIONS AS A CLASS, B&W PLANTS CANNOT PRESENTLY BE SAID TO HAVE A CORE DAMAGE FREQUENCY SIGNIFICANTLY GREATER THAN THAT OF OTHER PLANT TYPES OVERCOOLING EVENTS DOMINATE CATEGORY C FREQUENCY, BUT ARE MINOR CONTRIBUTIONS TO CORE DAMAGE FREQUENCY AT ALL PLANT TYPES i
l OF THOSE CATEGORY C EVENTS WHICH ARE EXPERIENCED, UNDERCOOLING IS THE EVENT MOST LIKELY TO LEAD TO CORE CAMAGE OF THOSE CATEGORY C EVENTS WHICH ARE EXPERIENCED, LOSS OF ICS FUNCTION IS CONSIDERED TO BE THE MOST SIGNIFICANT INITIATING ~ EVENT l
l O
i O Ap RISK ASSESSMENT REVIEW o CATEGORY C SEQUENCES MODELED IN THE CR-3 AND 0C0 EE PRAs. ACTUAL B&W TRANSIENT HISTORY ADEQUATELY REPRESENTED IN BOTH PRAs.
! o CR-3 AND OCONEE PRAs INDICATED HINIMAL CORE DAMAGE l RISK DUE TO CATEGORY C EVENTS AT THESE UNITS.
1 0 FEED-AND-BLEED (HPI) COOLING CAPABILITY IS AN IMPORTANT ACCIDENT MITIGATOR FOR B&W UNITS.
l o SUPPORT SYSTEMS (SERVICE WATER AND AC POWER) WERE l
FOUND TO BE RtLATIVELY RISK-IMPORTANT IN BOTH ACCIDENT INITIATION AND MITIGATION.
O o GENERALIZATION OF THE CR-3 AND OCONEE PRA RESULTS INDICATED THAT CATEGORY C EVENTS ARE NOT LIKELY TO BE CONSIDERED AS DOMINANT CONTRIBUTORS TO CORE DAMAGE RISK FOR MOST B&W UNITS. DAVIS-BESSE AND TMI WERE FOUND TO HAVE A NON-NEGLIGIBLE CONTRIBUTION TO CORE DAMAGE RISK FROM CATEGORY C EVENTS - DAVIS-BESSE, DUE TO DB'S UNIQUE REQUIREMENTS FOR HPI COOLING; AND THI, DUE TO DIFFERENCES IN THE TMI PRA ANALYSIS.
o WHILE THE B&WOG INTENDS TO REDUCE THE FREQUENCY OF OR ELIMINATE ALTH0GETHER THE CATEGORY C EVENTS, FROM A PURE PRA PERSPECTIVE, THE RISK ASSESSMENT REVIEW INDICATES THAT B&W CATEGORY C EVENTS ARE RECEIVING MORE ATTENTION THAN THEIR CORE DAMAGE RISK-IMPORTANCE INDICATES IS WARRANTED.
O
7 b G l B&W PLANT REASSESSMENT .
NUREG 1231 REACTOR TRIP INITIATING i EVENTS REVIEW
- I PRESENTED TO PRESENTED BY i
l ACRS SUBCOMMITTEE
! ON B&W REACTORS R.C. JONES I
MAY 3-4, 1978 NRR/ DEST /SRXB i
BWACRSS
2
~
STAFF FINDINGS .
- BWOG REVIEWED REACTOR TRIPS FOR 1980 - 1986
- ADDITIONAL RECOMMENDATIONS DEVELOPED TO REDUCE REACTOR TRIP FREQUENCY
- TLIRBINE TRIP SYSTEMS (8)
- MAIN FEEDWATER SYSTEMS (11)
- ICS (5)
- CONTROL ROD DRIVE SYSTEM (5)
- IMPROVE TRANSIENT RESPONSE (7)
- STAT-F REVIEW CONCENTRATED ON POSSIBLE NEGATIVE IMPACTS OF RECOMMENDATIONS
- STAFF GENERALLY FINDS RECOMMENDATIONS ACCEPTABLE
- BWOG RECONSIDER RECOMMENDATION TO P ROP ERLY CHECK OUT REPLACEMENT COMPONENTS
- PLANT SPECIFIC SAFETY EVALUATIONS NEEDED TO IMPLEMENT TRANSIENT IMPROVEMENT RECOMMENDATIONS I'
BWACRS23
!~ e O o tj t
I l'
( i.
t l
)
B&W PLANT REASSESSMENT i
i i
! NUREG 1231 i
i I EVALUATION OF OTHER
\
j BWOG REPORTS i
i i
-l PRESENTED TO PRESENTED BY i ACRS SUBCOMMITTEE ON B&W REACTORS R. C. JONES -
l l
MAY 3-4, 1978 NRR/ DEST /SRXB i
i BWACRS4 i
a
t STAFF FINDINGS STAFF PREVIOUSLY APPROVED TOPICALS FOR RAISING HIGH PRESSURE TRIP SETPOINT AND ARMING THRESHOLD FOR ARTS ON TURBINE TRIP
- EXPECTED TO REDUCE REACTOR TRIP FREQUENCY
- DOES NOT SIGNIFICANTLY INCREASE PORV OPENINGS
- PLANT TECH SPEC CHANGES APPROVED OR UNDER REVIEW FOR OPERATING B&W PLANTS
\
- TURBINE SYSTEM RECOMMENDATIONS EXPECTED TO DECREASE UNNECESSARY TURBINE TRIPS 1154 TASK FORCE REPORT RECOMMENDATIONS EXPECTED TO f.
IMPROVE RELIABILITY OF MOVs, PORV, PORV BLOCK VALVE AND CHECK VALVES BVLACRS22
NRR STAFF PRESENTATION TO THE !
O .
ACRS !
l
SUBJECT:
B&W Owners Group Plant Reassessment Program -BWOG Programmatic and .
Managment Actions DATE: May 3, 1988 PRESENTER: Byron siegel PRESENTER'S TITLE / BRANCH /DIV: Lead Project Manager, B&W Owners Group Plant Reassessment Program PRESENTER'S NRC TEL. NO.: 492-3019 SUBCOMMITTEE: B&W Reactor Plants 1
O
a O BWDG PROGRAWATIC AND FANAGBOT ACTIONS ROOT CAUSE 990tus .
GENERIC PROGPAM TO SERVE AS MODEL TO UTILITIES TO DEVELOP TEIR OWN PROGRAFE o EVALUATION ETHODS o INITIATING CONDITIONS o GUIDELIES FOR EVALUATION THIS PROCESS SHOULD IWROVE TE WAY OWNERS EVALUATE TPANSIENTS IDDUlFY CAUSES OF ECUIPtBE FAILUES, 1
TRANSIENT ASSESStEU PPDGRAM REPORT UPGRADES SEVEPAL IMPROVDfNTS TO TAP HAVE BEEN IWLENNTED, IN ADDITION BWOG HAS STATED THAT HUPAN FACTORS CONCERNS WILL BE ADDRESSED FOR FUTURE EVENTS,
! IWP VE} BUS PADE SHOULD IWROVE USEFULESS OF TAP IN ASSESSING TPANSIENTS, O
AIMSORY BOARD R SAFETY AND PEPFORMANCE RECOWENDATION IKIEGRATION GPOLP (SPRIG)
BOARD PF0VIDED ADDED ASSURANCE ALL IWORTANT ASPECTS ADDRESSED TO ACHIEVE PROGRAM OBJECTIVES, PROVIDED RE0[ftDOAT10tG FOR PROGRAM IWROV986 AfD IWLBERATION OF ECatB0ATIONS, i Pf0 CESSES UTillZED BY SPRIG FOR PRIORITIZING, GROUPING EC0ftBOAT10NS AND ASSESSING TEST FOR HIGH PRIORITY PEC0ttD0AT10tG FOR COMPLETBESS ACEPTABLE, l
STAFF DURING ITS AUDIT OF BWDG AT Bth! DETERMIED THAT NEITHER GROUP FUNCTIONED IN A FAftER THAT DILUTED THE EFFECTIVDESS OF THE REC 0ttD0AT10tG PROPOSED BY l THE VARIOUS SUBC0ttillTEES, THE STAFF IDENTIFIED IN TIE SSER 11 ADDITIONAL E00ftD0AT!0NS THAT HAVE l O SUFFICIENT SAFETY SIGNIFICANCE TO BE IDDRIFIED AS HIGH PRIORITY Af0 SK)ULD BE Gl\91 ADDITIONAL ATTBRION DURif$ PLANT AUDITS, SIEGEl. 5/3/88
d
"'-) O BWOG EVALUATION Pf0GPAM Pf0GPM IN PLACE TO EVALUATE EA01 UTILITY, o PROGRAM FOR PANAGING SPIP RECDtOOATIONS o ASSESS ADECUACY OF IPPLDDITATION INTERPRETATION 4 o milTOR SPIP G0ALS STAFF HAS SIMILAR P80GPM TO AUDIT UTILITIES
- O l
{
i
)
i 4
- O SIEGEL 5/3/88
NRR STAFF PRESENTATION TO THE O . ACRS
SUBJECT:
B&W Owners Group Plant Reassess 1 cent Program - Additional Concerns .
Identified by the Staff DATE: May 3, 1988 PRESENTER: Byron stegel
, O~ .
PRESENTER'S TITLE / BRANCH /DIV: Lead Project Manager, B&W Owners Group
' Plant Reassessment Program I
i PRESENTER'S NRC TEL. NO.: 492-3019 SUBCOMMITTEE: B&W Reactor Plants O '
l
\
STAFF MEMBER CONCERNS O
- 1) P0TENTI AL LACK OF INDEPENDENCE IN MPR ASSOCIATES PERFORMING A SENSITIVITY STUDY FOR THE BWOG AND RELRTED WORK FOR A B&W PLANT OWNER (SER, SECTION 5.1),
THE STAFF DID NOT REQUEST, NOR DID IT BELIEVE IT NECESSARY, THAT THE STUDY BE PERFORMED BY AN INDEPENDENT ORGANIZATION,
- 2) POTENTIAL LACK 0F INDEPENDENCE IN SCIENCE APPLICATIONS INTERNATIONAL CORPORATION PERFORMING SIMILAR WORK AS A SUBCONTRACTOR FOR THE NRC AND FOR B&W PLANT 0WNERS, OGC REVIEWED ISSUE AND DETERMINED NO CONFICT SINCE WORK WITH NRC ESSENTIALLY COMPLETE.
- 3) A PREMATURE FINDING BY THE BWOG 0F THE ADEQUACY OF ICS/NNI O (SSER, SECTION 6,1.1),
BWOG ASSUMED SYSTEM CAN BE MODIFIED AND THEREFORE DID NOT CONSIDER REPLACEMENT AT THIS TIME, STAFF BEllEVES EXISTING SYSTEM CAN BE MODIFIED TO RESOLVE CONCERNS,
- 4) THE BWOG OR UTILITIES HAVE NOT ANALYZED EFFECTIVELY THE PROPOSED SPIP RECOMMENDATIONS TO DETERMINE THE EFFECTS ON THE OTHER PARTS OF THE PLANT (SSER, SECTION 11.5).
BWOG HAS PERFORMED SOME SYSTEMS INTERACTIONS STUDIES, H0VEVER, MANY OF THE RECOMPENDATIONS ARE GENERAL IN NATURE AND REQUIRE PLANT SPECIFIC EVALUATIONS. BOTH THE BWOG AND STAFF INTEND TO EVALUATE THIS ASPECT DURING THEIR AUDITS,
- 5) CATASTROPHIC REACTOR PRESSURE VESSEL FAILURE RESULTING FROM AN OVERC00 LING TRANSIENT ASSUMING ONE CONTROL R0D STUCK OUT Q AND A RETURN TO CRITICAllTY (SSER SECTION 5.5.3).
1 SIEGEL 5/3/88
~
~2-O THE REACTOR COOLANT TEMPERATURE REQUIRED FOR REACTOR TO RETURN TO CRITICALITY IS T00 LOW (35'F) FOR THIS TO BE CONSIDERED A PROBABLE EVENT.
- 6) OPERATOR BURDEN STUDY DOES NOT ADDRESS OPERATOR ERRORS OF COMMISSION (SER, SECTION 7.5.2),
OPERATOR BURDEN STUDY DI JPLUDE ERRORS OF COMMISSION,
- 7) THE B&W PLANTS VIOLATE OFFSITE DOSE LIMITS UNDER THE CONDITIONS OF A STEAM GENERATOR TUBE RUPTURE DESIGN-BASIS ACCIDENT, Tills IS BEING ADDRESSED BY GENERIC ISSUE 67, i) THE MAIN STEAM LINE BREAK AND STEAM GENERATOR TUBE RUPTURE i SHOULD BE TREATED AS ONE DESIGN-BASIS ACCIDENT FOR ALL PWRs.
UNDER CONSIDERATION AS PART OF USI A-3, A-4 & A-5 REGARDING STEAM GENERATOR TUBE INTEGRITY,
- 9) CERTAIN FAILURES IN THE TURBINE CONTROL OR TRIP SYSTEMS DUE j . TO A DESIGN FEATURE TO DELAY TURBINE TRIP AFTER THE REACTOR
, TRIP CAN RESULT IN A REACTOR TRIP WITHOUT A TURBINE / GENERATOR TRIP.
j GENERIC TO ALL PWRs - MEMO SENT TO RES SUGGESTING THIS l
ISSUE BE PROPOSED AS A GENERIC ISSUE BY ORIGINATING STAFF MEMBER, J
I S!EGEL 5/3/Pa
a NRR STAFF PRESENTATION TO THE O ..
SUBJECT:
B&W Owners Group Plant Reassessment Program -Implementation -
DATE: May 3, 1988 PRESENTER: Byron siegel O -
Lead Project Manager, B&W Owners Group PRESENTER'S TITLE / BRANCH /DIV: Plant Reassessment Program PRESENTER'S NRC TEL. NO.: 492-3019 SUBCOMMITTEE: saw Reactor Plants O
O
_SPIP RECOMMENDATIONS SPIP PROGRAM GENERATED 375 RECOMMENDATIONS 75 DUPLICATES 20 REJECTED 215 APPROVED BY BWOG TO DATE E5 STILL IN APPROVAL PROCESS SER & SSER ADDRESSES ALL RECOMMENDATIONS IN PROGRAM EXCEP THAT WERE DUPLICATES SPIP HAS IDENTIFIED 70 RECOMMENDATIONS AS KEY - (REDUCE CO TRANSIENTS, TRIPS, OPERATOR BURDEN ETC.) STAFF HAS IDENTIFIED 11 O ADDITIONAL RECOMMENDATIONS AS HIGil PRIORITY (ENHANCE SAFE i
BWOG IS TRACKING PROGRESS IN I!'.PLEMENTATION OF RECOMMEN THRU RECOMMEllDATION TRACKING SYSTEM, WHICil IS UPDATED BI-MONTHLY UTILITIES HAVE AGREED TO IMPLEMENT ALL APPLICABLE RECOMMENDATIONS, l BWOG OVERSEEING PROGRAM TO ASSURE IMPLEMENTATION - STAFF lilTENDS TO MONITOR, AUDIT c INSPECT PROGRAM l
l SIEGEL5/3/q
l O
~
BWOG PLANT REASSESSMENT PROGRAM STAFF VERIFICATION OF IMPLEMENTATION OF BWOG RECOMMENDATIONS BY UTILITIES STAFF EVALUATED BWOG RECOMMENDATION APPROVAL FR0 CESS AT B&W
!N OCTOBER 1987 STAFF WILL AUDIT THE IMPLEMENTATION PROGRAM UTILITIES HAVE IN PLACE TO EVALUATE THE BWOG RECOMMENDATIONS STAFF WILL AUDIT THE IMPLEMENTATIOM PROCESS AT UTILITIEG T0 ASSURE REC 0KMENDAT10NS ARE BEING PROPERLY IMPLEMENTES
(])
STAFF WILL VERIFY RECOMMENDATIONS HAVE BEEN IMPLEMENTED BY UTILITIES .
STAFF WILL TRACK PROGRESS OF IMPLEMENTATION OF REC 0KMENDATIONS BY UTILITIES THROUGH BWOG REC 0KMENDATION TRACKING SYSTEM AND PROJECT MANAGERS INTERACTIONS WITH UTILITIES .
e e
(:) .
SIEGEL 5/3/88
([) ' '
BWOG PLANT REASSESSMENT PROGRAM OVERALL UTILITY PROGRESS IN IMPLEMENTING BWrS APPROVED RECOMMENDATIONS PERCENT EVALUATING 45 -
IMPLEMENTING 12 r
CLOSED OPERABLE 20
([) CLOSED N/A 18 CLOSED REJECTED 1 NOT STARTED 4 1
e 8
(
SIEGEL 5/3/88