ML20150C043
| ML20150C043 | |
| Person / Time | |
|---|---|
| Issue date: | 03/10/1988 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1650, NUDOCS 8803170290 | |
| Download: ML20150C043 (227) | |
Text
ACRSF /6dro
..il\q'c)o l O UNITED STATES NUCLEAR REGULATORY COMMISSION i
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS In the Matter of:
335th GENERAL MEETING AFTERNOON SCSCION
[ t. u -
~, l . 3. ,
'q c... ,1 ;,; ,,
.G
_ [ },$~;N d . h 5 Y N.'-
Pages: 129 through 279 Place: Washington, D.C.
Date: March 10, 1988
n---- ---------------------======
HERITAGE REPORTING CORPORATION O osawa a-r-1220 L Street, N.W., Suke 640 Washinston, D.C. 20065 (202) 628-4888 w o:-1 ,,i:. . . 3, PDR At Ri T-1/cd)
[u_ p
1 PUBLIC NOTICE BY THE 2 UNITED STATES NUCLEAR REGULATORY COMMISSION'S 3 ADVISCRY 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 14 inaccuracies of statement or data contained in this transcript.
15 16 17 18 19 20 21 22 23 24 t 25 O Heritage Reporting (202) 628-4888 Corporation
- ,---r..., - - -- , . , , , , , . , , . , , . . . , , . , , - . . - .,.,----,-n-,n.-,,n-.,, ,,,,n_,--,------n,_n.. ,,---.,,--,n, -. -- n
i C) 1 UNITED STATES NUCLEAR REGULATORY COMMISSION 2 ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 3
)
4 In the Matter of: )
)
5 )
335TH GENERAL MEETING )
6 l' 7 Thursday, March 10, 1988 8
Room 1046 9 1717 H Street, N.W.
Washington, D.C. 20555 10 The above-entitled matter came on for hearing, 11 pursuant to notica, at 8:30 a.m.
12 BEFORE: DR. WILLIAM KERR ,
13 Chairman
() 14 Professor of Nuclear Engineering Director, Office of Energy Research University of Michigan 15 Ann Arbor, Michigan 16 ACRS MEMBERS PRESENT:
17 DR. FORREST J. REMICK Vice Chairman 18 Associate Vice-President for Research Professor of Nuclear Engineering 19 The Pennsylvania State University University Park, Pennsylvania 20 MR. JESSE C. EBERSOLE 21 Retired Head Nuclear Engineer Division of Engineering Design 22 Tennessee Valley Authority Knoxville, Tennessee 23 '
DR. CHESTER P. SIESS 34 Professor Emeritus of Civil Engineering University of Illinois 25 Urbana, Illinois O
HERITAGE REPORTING CORPORATION -- (2021628-4888
r DR. HAROLD W. LEWIS 2 Professor of Physics Department of Physics 3 University of California Santa Barbara, California 4
MR. CARLYLE MICHELSON 5 Retired Principal Nuclear Engineer Tennessee Valley Authority 6 Knoxville, Tennessee and Retired Director, Office for Analysis 7 and Evaluation of Operational Data U.S. Nuclear Regulatory Commission 8 Washington, D.C.
9 DR. DADE W. MOELLER Professor of Engineering in Environmental Health 10 Associate Dean for Continuing Education School of Public Health 11 Harvard University Boston, Massachusetts 12 DR. PAUL G. SHEWMON 13 Professor, Metallurgical Engineering Department
() 14 Ohio State University Columbus, Ohio 15 DR. CHESTER P. SIESS Professor Emeritus of Civil Engineering 16 Argonna National Laboratory Argonne, Illinois 17 MR. DAVID A. WARD 18 Research Manager on Special Assignment l
E.I. du Pont de Nemours & Company 19 Savannah River Laboratory Aiken, South Carolina l 20 l MR. CHARLES J. WYLIE l 21 Retired Chief Engineer Electrical Division 22 Duke Power Company Charlotte, North Carolina 23 #
ACRS COGNIZANT STAFF MEMBER:
24 Raymond Fraley, Executive Director 25 O
HERITAGE REPORTING CORPORATION -- (2021628-4888
129 I) 1 DR. KERR: Operating events and incidents, Mr.
2 Ebersole.
3 MR. EBERSOLE: Thank you, Bill.
4 In the ecrly part of this period we felt that we 5 really didn't have much operating events of interest for this 6 committee. It was astonishing to find it materialized about 7 the middle of the period events fascinating, they related to 8 the issues we recently talked about on the influence of control 9 system malfunctions on safety systems and a few other areas.
10 In short, I find it fascinating that we are going to 11 hit in this little short session here both ends of the spectrum 12 of ma1 performance of feedwater systems. On the one case we 13 drive a boiler to overfill and the water runs on down the main
() 14 steam lines which is the overflow case, you will recall from 15 the case on the control system failures.
16 What is interesting is the coincidence of this 17 occurrence at such a short time after we noticed the staff 18 action to preclude this occurrence. And they're leaving an 19 escape route for the kind of accident that happened here.
20 The second one is, I think, even more significant 21 incident, how we managed to get a boiler to go dry and be dry 22 for some period of time, I don't understand yet, but I'm sure I I
23 will after this session, at least I hope.
. 24 And then the second is sort of a paired set of events 25 where we had annunciated fires in the control room, which in i
t
}
4 lieritage Reporting Corporation (202) 628-4888
i 130 (Q_/ 1 themselves didn't produce any serious results but at least they 2 were interesting as a possible precursor for serious events.
3 Wayne, it is all yours and you only have an hour, so 4 I am going to be very quick.
5 MR. LANNING: Thank you, Mr. Ebersole.
6 We are going to talk to the three events you just 7 addressed. We plan about a 10 minute presentation on each 8 event, and then hope to have about a 10 minute discussion.
9 Vincent Thomas from Instrumentation Control Branch, 10 NRR, will lead the discussion concerning the fires in the 11 annunciator cabinets.
12 MR. THOMAS: I'm Vince Thomas, Instrumentation 13 Control Section of NRR. I will briefly discuss what we know to
() 14 date on three fires within a two week span and remote cabinets 15 that feed the visual and audible features to the control room 16 annunciator.
17 These cabinets, by the way, explained here were 18 provided by the same manufacturer.
19 The first one occurred at Beaver Valley on January 20 29th. And again, it's a remote cabinet.
They had erratic 21 behavorial in the annunciator fires.
22 DR. KERR: What is the significance of calling it a 23 remote cabinet?
24 MR. THOMAS: Well, the remote cabinet is the one that 25 has the intelligence from the field contacts like the O Heritage Reporting Corporation (202) 628-4888
131
() 1 parameters that's monitoring the parameters throughout the 2 power plant; and the provide the logics that come into the 3 remote cabinets which are outside-the control room. And these 4 are the alarm status cards that says, okay, high pressure.
5 DR. KERR I just wondered what is meant by remote?
6 MR. THOMAS: Remote to the control room. When I 7 mention remote I'm talking from the control room operator.
8 DR. KERR Thank you.
9 MR. EBERSOLE: Could you calibrate the character of 10 this cabinet in the context of safety grade or whatever?
11 MR. THOMAS: It's unquestionable.
12 MR. EBERSOLE: This brings up in itself a continuance 13 of the fact that annunciation, audible and visual, I think,
() 14 prevails very widely as non-qualified information; am I 15 ' correct?
16 MR. THOMAS: That's right.
17 MR. EBERSOLE: It is not seismic. It is not multi 18 channel. It is not anything.
19 MR. Th7 MAS: It is normally open circuits.
20 MR. EBERSOLE: A very widely spread notion that all 21 of this is.
22 MR. THOMAS: It's not c'esigned feel safe.
23 MR. EBERSOLE: I believe curbed requirements has 24 corrected or improved that situation; am I correct?
25 MR. THOMAS: Not ' vet.
O Heritage Reporting Corporation (202) 628-4888
132 (ms,) 1 MR. EBERSOLE: Not as yet. So I keep in mind then 2 that the eyes and ears of the operator, he has no ears because 3 the annunciation is not safety grade, he must look at 4 indicators which is spread around the vertical control board.
5 And his information for these things has to be qualified as 6 possibly just being from within the system itself.
7 MR. LANNING: Well, there are redundant 8 instrumentation and readout available to the operators in the 9 control room.
10 Are you going to talk to that, Vince?
11 MR. THOMAS: Yes. For every alarm station obviously 12 you have some continuance indication in the control room to 13 give you the value. These are not absolute values, they're
() 14 just saying something is in an abnormal condition or clear to 15 assist the operator to take some action if he needs to.
16 Becausa you can see that there was a site alert 17 declared, and the only reason I am focusing on that particular 18 one, two site alarms of the three events were declared, whereas 19 in Rancho Seco it wasn't.
20 Common -- well, I better not jump ahead that fast. I l
21 will just say that the fire was extinguished in less than 10 l
22 minutes, and they were without the control room --
23 MR. M 'HELSON: How did you extinguish the fire?
24 MR. 7'JMAS: Two ways, they used a portable 25 extinguisher the was handy, and Fv throwing the appropriate O _itage Reporting Corporation I (202) 628-4888 ,
1
133
() 1 breakers to isolate the fire, cut off the power source.
2 So we don't repeat this again, this occurred in each 3 fire event, fire fighting.
4 And the long term investigative' plans with respect to 5 Beaver Valley as well as Calvert Cliff, San Rancho Seco, 6 they're looking into the sub-fusing design concept to maybe 7 better get more senstive protection to the circuitry involved.
8 DR. KERR: What is the significance of declaring a 9 site alert a half hour after the fire was extinguished?
10 MR. THOMAS: I think the EPs -- maybe somebody else 11 can address that end of it. I think there is a requirement 12 that says, if you don't have the annunciated -- I know at 13 Calvert Cliffs, if you don't have the annunciator available
() 14 within one hour you-have to declare some kind of a site alert.
15 I think that is something like a -- I think it applied to 16 Beaver Valley, I'm not too up on that, but Rancho Seco was 17 decided by somebody else not to declare it.
18 DR. KERR: You declare a site alert and what effect 19 does that have on something or other?
20 MR. THOMAS: Well, that brings in an awful lot of 21 people.
22 MR. LANNING: I think one of the. things it does is, 23 it requires a notification of several government agencies.
24 DR. KERR: No, I mean, what effect does it have on 25 the operation of the plant in terms of its ability to function O Heritage Reporting Corporation (202) 628-4888
134 q
(/ 1 or not function?
2 MR. LANNING: Normally that would require augmenting 3 the technical support center.
4 DR. KERR: I don't see what effect that has on plant 5 operation 6 MR. THOMAS: It requires people to come immediately 7 to the control room and assist in helping with the situation au 8 hand.
9 DR. KERR: 1 thought you said going to the technical 10 support center; did I misunderstand?
11 MR. LANNING: Well, the additional expertise would 12 come to provide technical support, the man at the technical 13 support center.
() 14 MR. THOMAS: That's what happens.
15 At Calvert Cliff especially they brought in a control 16 room operator, and the reason I'll tell you in a moment when we 17 get to that.
18 The specific cause for the fire obviously was 19 overheated conditions on the circuit board. The reason for it 20 is still unknown, but there is extensive ongoing studies 21 applying to the whole three plants. There's interesting things 22 that come up later on if you want to discuss them, but let me 23 go on to the Calvert Cliffs.
24 If there's any other questions here?
25 MR. MICHELSON: How do they detect the fire, roughly, O Heritage Reporting Corporation (202) 628-4888
i 135
() 1 is smoke pouring out?
j 2 MR. THOMAS: Erratic annunciator behavior, and they 3 were obviously known that wasn't a normal operation and 4 dispatched a --
5 MR. MICHELSON: But there was smoke pouring out.
6 MR. THOMAS: Yes. The fire-itself, after they 7 dispatched somebody down to the fire they went in and saw smoke 8 and flames. <
9 An interesting thing on Beaver Valley, if you want to 10 go into it deeper, that they had a muffin fan which is a 11 circulation to the cabinet itself that the flames were actually 12 coming out from the fan itself when they walked in.
13 MR. MICHELSON: Now, did the fire protection system
() 14 pick this up?
15 MR. THOMAS: The fire --
16 MR. MICHELSON: Protection system.
17 MR. THOMAS: The halon didn't go off here or the 18 suppression system wasn't --
1
! 19 MR. PICHELSON: You mean it was that much of a fire .
20 without --
21 MR. THOMAS: The one you get is a general -- this is 22 a local cabinet.
23 MR. MICHELEON: I understand.
! 24 MR. THOMAS: It did go off at Calvert Cliffs.
5 25 MR. MICHELSON: I have been convinced about how O Heritage Reporting Corporation j (202) 628-4888
136 i
( 1 sensitive this equipment is ,cout very small firea, and I was -
2 just trying to get a data point. ;
i 3 MR. THOMAS: At Calvert Cliffs it wsa sensitive 4 enough to or local enough to actuate the halon.
5 MR. EBERSOLE: The input to the SBVS system might 6 well have come from the same transmissions to feed this system.
7 MR. THOMAS: Exactly. And that's where they had one 8 of the dedicated operators monitoring at the time, during the 9 time that they were without the control room annunciators. t 10 MR. EBERSOLE: Did this SBVS go dead?
11 MR. THOMAS: No. This is only to the annunciating 1 12 system itself.
13 MR. LEWIS: Is there any possibility of sabotage
! () 14 here?
15 MR. THOMAS: No. You're asking me, I don't think so.
16 MR. LEWIS: Because, you know, if you calculate the 17 probably of having three out of the four plants that have this 18 gadget have fires.
19 MR. THOMAS: That is unique and it is unique. That's
- 20 all I can say. There's seven to 15 years, Dr. Lewis, there's .
21 seven to 15 years of installation. It's the first fire they !
l 22 ever experienced at the Calvert Cliffs, for example. They have 23 had precursors to the fire by looking at disc 1 orations on the 1
24 current circuits boards.
l 25 MR. LEWIS: These things have been misbehaving for 10
(
l Heritage Reporting Corporation (202) 628-4888 l
l
137 O- 1 veer -
2 NR. THOMAS: It's a non-safety system, so if you're 3 going to sabotage you will have to go somewhere else.
4 MR. LEWIS: No, I understand that. I was sort of 5 half choking about sabotage. But it is a strange set of 6 coincidence.
7 MR. THOMAS: It is. Everybody thought about it. By 8 the way, while we're on that subject we determined that this 9 particular manufacturer is out of business. I'm leap frogging, 10 not because of the fires, but he is out of business; and he 11 only supplied four plants and would you guess the fourth plant 12 without the fire was Calvert Cliff-1. Calvert Cliff-2 had the 13 fire, its sister plant didn't. So there you are. They had one
() 14 other one at Zimmer, but they are no longer. ,
15 MR. SHEWMON: Is there anything they can'do in 16 Calvert Cliff-1 to lower the chances?
17 MR. THOMAS: Whatever they do at Calvert Cliff-2, and 18 it comes out of the ongoing study. The sub-fusing concept, for 19 example. The poor circulation -- better circulation within the 20 cabinets. The automatic trip off of fans when the high 21 temperature is detected in the cabinets. These are all things 22 they're thinking about. They're thinking about temperature 23 scanning, infrared scanning; things of this nature.
24 MR. EBERSOLE: That high temperature trip off is a 25 common thing, thr' you look at I think in the computer cabinets O Heritage Reporting Corporation (202) 628-4888
4 138
() 1 and all sorts of cabinets.
2 MR. . THOMAS: Right.
3 MR. EBERSOLE: This could be the trigger --
4 MR. THOMAS: Well, the thing is, after they saw the 5 flames coming through the fan they figured that maybe they 6 ought to throw this thing off, because it was propogating.
7 MR. LEWIS: Where there's fire there's smoke or 8 something like that.
9 MR. MICHELSON: How do we know that the fire was 10 detected in 10 minutes, that-was from the time of the 11 malfunction --
12 MR. THOMAS: The sequence of events.
13 MR. MICHELSON: -- until they found the fire.
() 14 MR. MICHELSON: I'm a little disturbed that you have 15 to find the fire by watching the performance of equipment 16 instead of the performance of fire detection methods. I assume 17 they had fire detection in the room. Is that right?
18 MR. THOMAS: Yes, it's a large room in a lot of 19 these, and these are small cabinets. And they are well 20 enclosed. It is comforting to know that it didn't get out that 21 much. They have -- Calvert Cliffs had -- theirs was actuated.
22 They actually automatically suppressed the fire, but 23 when the operator went down and saw the smoke coming out, 24 opened the cabinet to find out what it was -- how much damage, 25 it reignited. It reignited and he had to throw the breakers O Heritage Reporting Corporation (202) 628-4888
139 1 right there.
2 MR. EBERSOLE: Is this being looked at the context 3 that internal fires can occur and severe damage take place.
4 Here it was an unimportant annunciation system. It might be a 5 critical solid state control module.
6 MR. THOMAS: Definitely.
7 MR. EBERSOLE: I think these are long overdue for 8 internal temperature, alarms and trips.
9 MR. THOMAS: This is one aspect they're looking at, I 10 think, the infrared scanning.
11 MR. THOMAS: Calvert Cliffs, remember within two 12 weeks Calvert Cliffs I guess came along on February 1st, again 13 the fire was extinguished in less than 10 minutes. Here, I j
() 14 don't have it up, but they, on the viewgraph, but the automatic 15 halon suppression system was activated.
16 The unit 1 operator got the alarm that the automatic 17 unit -- automatic halon suppression system went off, told the 18 unit 2 operator, I hope I got them right. Unit 2 is where the 19 fire occurred. Unit 1 operator got the information through his 20 annunciating system that the unit 2 automatic suppression 21 system went off and he informed the operator.
22 That, in addition to the erratic behavior of the 23 annunciating system, and they couldn't gat the horn off, the 24 audible horn that silences after you go into an abnormal 25 condition, brought them down to the switch gear room or cable
(_) Reporting Corporation Heritage l
i (202) 628-4888 l
\ _ _ _ - _ ..
1 1
140 l
() 1 spreading room below the control room; and that's where the 2 smoke was coming out. The opened the breaker panel and-it 3 reignited and the operator grabbed the portable extinguisher 4 and put the fire out and threw the breakers. And that's when 5 he lost the whole visual-and audible annunciation. He hadn't 6 lost his visual, but throwing the breakers their power source, 7 they were interrupting the power sources to the complete 8 annunciating system.
9 So really, in putting _the fire out they took the 10 annunciating system away.
11 The visuals were back in operation about two hours, 12 but it took them -- the damage in this particular caso really 13 was in the, what they refer to as the driver card. And the
() 14 driver card is the one that provides the horn, the flashing, 15 and the chimes and-all this stuff that alerts the operator that 16 an abnormal condition is existing somewhere in the plant.
! 17 That's why they had two days to repair the audible 18 portion of it. That audible portion is really the manifold, if 19 you will, or the base for the six or five or 600 alarm stations 20 that goes through that horn.
21 MR. LEWIS: Presumably they are going to replace the 22 boards that inopt, because they're going to have to retail them 23 by somebody else. But the ones that are operable, even though 24 they are probably damaged, will be left in the system because 25 it's a non-safety system?
- i j
() Heritage Reporting Corporation (202) 628-4888
141 1 MR. THOMAS: No, they haven't accepted reject 2 criteria that they're trying to apply. And, you know, they're 3 looking for smoke damage or they have. I'm speaking of the I
4 part tense. They have looked into it and looked for any kind ,
5 of damage.
6 In fact, the Rancho Seco people have run informal 7 tests that determine that the circuit boards themselves are l
8 made of nylon and the runners where the cards go in, they're 9 putting in better flame retardent materini, the nylon was not.
10 MR. SHEWMON: You say they were nylon?
11 MR. THOMAS: The boards, yes.
12 MR. SHEWMON: They were made of nylon?
13 MR. THOMAS: Yes. They are switching to fiberglass
( 14 mater.-ial.
15 MR. LEWIS: The thing I was gropping for is that, 16 they're only looking for visible damage. Presumably these are 17 new enough to the solid state, and solid state components, of 18 course, get damaged even if there isn't visible damage to them.
19 MR. THOMAS: That's right.
20 MR. LEWIS: But they are not going to make a 21 wholesale replacement of all the boards.
22 MR. THOMAS: I think in answer to your question, they 23 noted in their current studies to date that some of the fire 24 resistors that draw a lot of the current, and they have to, you 25 know, dissipate an awful lot of heat. They were mounted right O Heritage Reporting Corporation (202) 628-4888
142
() 1 on the boards, discloration around them.
2 So they are thinking of heat fins, heat sinks to put 3 around these increasing the wattage. To lower the temperature.
4 Another thing they are looking into is the battery voltage.
5 One plant, Beaver Valley, has their particular part 6 of the annunciator system buffered such that it doesn't see any 7 float voltage, the battery charging voltage of 140 volts.
8 These systems are really designed for 125, 130 volts. And as 9 you know, some of the plants I think, and maybe I'm speaking, 10 Ed Wenzinger can talk more direct to the thing, some of these 11 plants operate these battery charge an awful long period of 12 time, two, three weeks. It's an overheating problem.
13 I guess the answer to the question is, they're
() 14 looking at the whole structure of the board itself and they're 25 improving on it. They're dipping the resistors and the 16 compassators with flame retardent insulation.
17 MR. LEWIS: This is to keep it from recurring.
18 MR. THOMAS: And they're also in modeling, in a stand 19 off position, they're making circulation, temperature 20 monitoring.
21 MR. LEWIS: I understand that. I was only gropping 22 for whether boards that had been damaged in an invisible way 23 during the fire are going to be in the cabinet after they are 24 finished with the whole operation.
25 MR. LANNING: They had tested all the other boards O Heritage Reporting Corporation (202) 628-4888
143
,r'3 V 1 that did not display any visible damage. If you go to your 2 next slide on Rancho, for example, you see the number.
3 MR. LEWIS: So the answer is, yes, but they will have !
4 been tested.
5 MR. LANNING: That's my understanding.
6 MR. THOMAS: Every station has been tested in the 7 plant, on all plants to see that it's functioning. That's done 8 each shift, as a matter of fact.
9 MR. LEWIS: I sure wouldn't do that with an important 10 board.
11 MR. MICHELSON: None of these are important, though, 12 or they wouldn't be safety related. None of these are safety 13 related boards.
() 14 MR. EDERSOLE: But they're important to safety.
15 MR. MICHELSON: No, they're not.
i 16 MR. EBERSOLE: Well, at least they are in my view.
17 MR. MICHELSON: Nobody has ever said that.
I 18 MR. EBERSOLE: You keep referring to battery voltage, J
19 are these DC supplied?
20 MR. THOMAS: Yes.
21 MR. EBERSOLE: Are they subject to over-voltage 22 during equalizing period?
23 MR. THOMAS: I was saying before that one part of the i 24 annunciating system at Beaver valley is not. But in answer to 25 your question, across the board, yes, they are subjected to O Heritage Reporting Corporation (202) 628-4888 i
144 that. They see that 140 volts. That is another thing they are
( }) I 2 looking into is isolation, the sub-fusing.
3 February 8th Rancho Seco, this particular event 4 Rancho Seco, again Wayne can correct me if I'm saying 5 something, he was at the same meeting that I was, they had the 6 most extensive damage. They lost 112 boards out of a total of 7 192.
8 They have an extensive repair, restoration plan going 9 on, ongoing studies to date. They are the ones who are 10 committed to revising their emergency operating procedures to 11 include events of this nature to cope with them in the future. l 12 Whereas, Beaver Valley and Calvert Cliffs are considering it.
13 One interesting thing, too, as I mentioned Lefore
() 14 that the manufacturer is now out of business. However, each of 15 these utilities have purchased the engineering art work. Did I 16 just say that before?
17 MR. EBERSOLE: No.
18 MR. THOMAS: And the engineering art work permitted 19 them to go to a third party and replace the components.
20 DR. KERR: What is meant by modifying a plant 21 emergency operating procedures to cope with similar events in 22 the future?
23 MR. THOMAS: That should have been changed.
24 Modifying the current --
25 DR. KERR Modifying a plant emergency operating O Heritage Reporting Corporation (202) 628-4888
i 145 1 something.
i 2 MR. THOMAS: Oh, yes. That's what I was talking to. ;
3 There are operating procedures on the plant and they are l 4 modifying them or revising them to include instrut.tions as to i
5 how to handle this kind of an event. They didn't have it.
6 None of the utilities had procedures to handle this kind of 7 event, the loss of -- complete loss of annunciating system.
8 The one plant had it for the buss that fit it. ;
9 DR. KERR: Did they need emergency operating 10 procedures --
11 MR. THOMAS: I think so.
12 DR. KERR: -- to put the fire out. Have they i
13 significantly changed what they would do next time or are they i 14 just writing procedures to tell them to do what they did?
j 15 MR. THOMAS: I don't know how to answer the question, 16 significant. They are having people -- the operators, I guess l 17 during the event -- I do know at Calvert Cliffs they're gaing 18 to have somebody, two people, they're dedicated in the control j 19 room to monitor the indicators. He knows what to do. That 20 wasn't in the procedures.
DR. KERR: What I was trying to find out is whether 21 I
22 the procedures are going to have them do something that will l 23 make things a lot better next time or whether the procedures 24 are going to say, do what you did this time, that's the 25 procedure?
O Heritage Reporting Corporation (202) 628-4888
146
()' 1 MR. LANNING: Vince, as I recall from our meeting, 2 each of them were evaluating the need for additional --
3 emergency operator to deal with the event that involves loss of 4 annunciation. Realize, of course, that two out of three events 5 occurred while the instrument was shut down. Calvert Cliff was 6 100 percent power. They stayed at 100 percent power during 7 this event.
8 MR. THOMAS: That's an importanc thing, Dr. Kerr, 9 because that's where the procedure -- they stayed there because 10 they didn't want to go into a transient.
- 13. MR. LEWIS: I guess if I understand, did they 12 mishandle these events in any way? Did they do anything wrong?
13 MR.-THOMAS: Not that I know of.
(') 14 MR. LEWIS: Then I really don't understand why you 15 need new EOP. ,
16 MR. LANNING: I think one of the things you can look 17 a ?, is, if the situation were somewhat different, had these 18 events occur, had these events been more extensive in the area, 19 you need contingency I think for dealing with such events. !
20 MP. LEWIS: So the new EOPs are not to cope with 21 similar events, but to cope with more severe events patterned 22 along these.
23 MR. LANNING: I'm not really sure what the scope of 24 the new procedures will be or if there will be new procedures.
25 It's just that each of the licensees are looking at, what O Heritage Reporting Corporation (202) 628-4888
147
() 1 action should be --
2 MR. LEWIS: I'm just trying to understand the 3 compulsion to write EOPs.
4 DR. KERR: The reason I ask the question is because I 5 had experience with situations in which after something happens 6 the reaction of an inspector is, you should have a procedure to 7 deal with that, so you write a procedure. And I wonder if that 8 is what happened in this case or if the procedure was based on 9 the assumption that the situation was handled very poorly and 10 it should have been handled different, and no one knows how to 11 handle it much better and the procedure will tell you to do 12 that, or whether it's just a matter of there was no procedure 13 to deal with this and one ought to have a procedure to deal
() 14 with everything, so one writes a procedure.
15 I can't toll from what I have read so far.
16 MR. THOMAS: I think that's true what you say. I 17 have a feeling on it that -- I think eveeything I knew worked L
i 18 out okay, but simply because there were no procedures.
19 MR. MICHELSON: Could you tell me one thing, did the 20 NRC ask the utility or influence the utility to write nes:
21 procedures or did the utility do it because they thought it was l
l 22 prudent?
23 MR. LANNING: As I understand it, the utility is 24 doing it because they thought it was a prudent thing to 25 evaluate, L Heritage Reporting Corporat'.on 1
(202) 628-4888 t
r-148 MR. MICHELSON: Thank you.
(( ) 1 2 MR. EBERSOLE: Wayne, to get this in prospective, 3 correct me if I'm wrong, this vast array of wonders with 4 audible sounds and visual displays and the SBVS are in 5 themselves non-qualified and thus potentially inaccurate 6 sources of information.
7 MR. LANNING: No, the SBVS it's equipment displayed.
8 MR. EBERSOLE: It is not.
9 MR. LANNING: You're right.
10 MR. THOMAS: Jerry Mauck can address that.
11 MR. EBERSOLE: So what the operator must always do is 12 recognize his source of this information with its noises and 13 sounds, noises and sight that it views. The critical
() 14 information that is displayed on the vertical boards, and he is 15 obligated to have a running status of what is on the vertical 16 boards. He must always say, that's the true story. But it 17 doesn't make any sound. It doesn't call out that something is 18 wrong unless you are looking at it and watching it move.
19 MR. WYLIE: Well, he's got his computer to fall back 1
20 on.
21 MR. EBERSOLE: Yes. But the computer in itself is 22 not safety related.
23 MR. MICHELSON: It may not be running necessarily.
24 MR. EBERSOhE: It may not be even there, and it's on 25 a single source of electricity, Charlie, and its transmissions O Heritage Reporting Corporation (202) 628-4888
149
(; 1 are not qualified.
2 MR. THOMAS: The signal that is providing the 3 indicator and the contact that closes the -- announced that the 4 parameter is out of abnormal is coming from the same 5 transmittor, the same bystables and things of this nature.
6 Only one is going through isolation devices that makes it no 7 longer --
8 MR. EBERSOLE: And it's going to a device --
9 MR. THOMAS: But the alarm is there to tell the man 10 that, hey, something is happening and you better take some 11 action or he knows what to do.
12 MR. EBERSOLE: But that alarm doesn't derive from 13 local equipment at the indicator.
I'd
(/ 14 MR. THOMAS: It derives from the same sensor.
15 MR. EBERS9LE: What I'm saying is, the standard 16 indicators provide no audible safety.
17 MR. THOMAS: He has got to look at something else, 18 yes.
19 MR. EBERSOLE: Which is non-safety grade.
20 DR. KERR: Mr. Siess.
21 MR. SIESS: Are these the first instances in which 22 all the annunciation has been lost at a nuclear powerplant?
23 MR. THOMAS: At the point, the staff's effort is, l 24 maybe Rich Lobel can address that, we have a search going on.
l 25 MR. SIESS: It requires either a yes or no answer, l
f'-'.I Heritage Reporting Corporation (202) 628-4888
150 1 Can somebody say, I don't know.
2 MR. LOBEL: We are doing a search now to find out.
3 We don't know of any instances at this time as part of our 4 investigation into this. We are doing a search of 5072 and 5 LERs to find out.
6 It's our understanding from the meeting that there 7 was a loss of annunciators dealing with fire at a fossel plant.
8 MR. SIESS: I didn't say dealing with fire. Any 9 reason.
10 MR. LOBEL: I guess I would be surprised if there 11 hadn't been, but we don't know of any and we are looking into 12 that.
13 MR. SIESS: I'm not concerned with the fire at all.
'( ) 14 I'm only concerned with loss of annunciation?
15 DR. KERR: Can you answer it?
16 MR. THOMAS: They have lost portions of a board, but l 17 never that I have been in that.
18 MR. LOBEL: I understood the question and I think I 19 answered it, let me say it again. We don't know of any now 20 ofthand, but we are doing a search to find out as part of our 21 investigation in investigating these events.
22 DR. KERR Please continue.
23 (Continued on next page.)
24 25 O Heritage Reporting Corporation (202) 628-4888
151
(,) 1 MR. THOMAS: As I stated before, Electro Devices is 2 no longer in business. However, the ability to fabricate 3 replacement parts as needed is there. And only four nuclear 4 plants utilize that particular system. And the systems from 5 Beaver Valley to Calvert Cliffs to Rancho Seco are not 6 interchangeable simply because each licensee wants his own type 7 of design and they just are not compatible.
8 They tried that by the way. We put out an 9 information notice 8805 and our on-going efforts at the moment 10 are to maybe develop some kind of communications through the 11 PMs-of the effected plants to be apprised of any findings that 12 may come of this on-going study that they have underway.
13 In an AEOD study I guess in the long term if 14 something from the search of Rich LaBellew comes up that
(])
15 .indicaces there's a paccern here. That's about it.
16 MR. LANNIliG: You mentioned three plants. What was 17 the fourth one?
18 MR. THOMAS: Calvert Cliffs 1. There actually was a 19 fifth plant. That was Zimmer but they're no longer here.
20 MR. LANNING: Thank you, Vince.
21 The next event we'll talk to is the dry out of the 22 steam generator at Indian Point Unit 2. And Marilee Slosson, 23 the Project Manager for Unit 2 is the speaker.
24 MS. SLOSSON: Good afternoon.
25 As Wayne said, I'm Marilee Slosson. I'm the Indian O Heritage Reporting Corporation (202) 628-4888
152
(~J~)
N 1 Point #2 project manager at NRR. The event I'm going to 2 discuss today was the steam generator dry out at Indian Point 3 #2.
4 (Slide) 5 This event occurred over the time period of January 6 2nd through January 6th, when they finally completed the refill 7 of the generator.
8 The plant conditions were such that they were taking 9 the unit from cold shutdown to hot shutdown for start-up from 10 the cycle 8-9 refueling outage and they were getting ready to 11 do the start up hydrotest. They had one motor driven aux feed 12 pump and two steam generators available for heat removal.
13 The Indian Point system is designed with two motor 14 driven aux feed pump and one turbine driven aux feed pump. The 15 motor driven aux feed pumps are such that ono feeds the two 16 generators. Aux feed pump 23 which supplied steam generators
- 17 23 and 24 was out for service to replace the pump section flow 18 switch, and they were not aware of it at the time, but the main 19 steam isolation valve 1-23 war. passing about 3,000 pounds per 20 hour.
On January 2 nd, 1988, the operators noticed that the j 21 22 steam generator level was decreasing. The shift watch 23 supervisor was informed but was not concerned about the 24 decrease at that time because he felt that they would get the 25 23 aux feed pump back soon after maintenance and that they'd be O Heritage Reporting Corporation (202) 628-4888 i
153
(} 1 able to start feeding. However, they didn't.
2 And on January 3rd at 8:00 o' clock in the morning, 3 the steam generator 23 boiled dry. At that time, the shift 4 watch supervisors did not appear to be aware that the steam 5 generators were boiling dry. However, all of the reactor 6 operators were aware that the steam generator had boiled dry.
7 On January 4th, 1988, when the shift supervisor came 8 on at 7:00 o' clock in the morning for his shift, he did become 9 aware that the steam generator was dry. He informed the 10 operators that they should attempt to refill the generator but 11 to do it slowly. And then he took off to go to an operations 12 meeting that morning.
13 In the meantime, about 8:00 o' clock, the chemistry 14 manager came down to the control room and was going to attempt
({}
15 to get blowdown samples and found out that he wasn't going to 16 be able to do that because the steam generator was boiled dry.
17 So he went upstairs to the manager of technical support and 1H asked him what was going on.
l They were not aware that the generator was dry. And 19 l
20 they started to move into action when they found out about it.
21 However, about 8:35, the operators did attempt to refill the 22 steam generators using condensate storage tanks. They were i
! 23 going to try to do this slowly, but they got about 200 gallons i
24 per minute, the realized that wasn't slowly, and they ceased j 25 the attempt to refill the generator.
l l
Heritage Reporting Corporation (202) 628-4888 l
154
() 1 About 8:45, the chief plant technical engineer came 2 down and said, please do not refill the generator until we've
~
3 had a chance to talk to Westinghouse and figure out how we 4 should fill a hot dry generator.
5 Following discussions with Westinghouse, they came up 6 with a procedure to sluice from steam generator 21 through a 7 blowdown line. And eventually were able to refill. The filled 8 up to about 70 percent level by 7:00 p.m. on the 5th and 9 continue to fill until about a 95 percent level and completed 10 that about 1:00 p.m. on January 6th.
11 MR. SHEWMON: Was there concern that they would have 12 sudden flashing at this and a pressure surge or that the 13 chemistry would be so bad, they'd damage something else, or
(} 14 what?
15 MS. SLOSSON: I guess it was thermal shock to the 16 vessel. It turned out that they did put some water in. They 17 put about 300, 400 gallons of water in. It did not have an 18 effect on the generator but I guess they were afraid of thermal 19 shock too. The girth weld, it turns out is the most limiting 20 case.
I 21 MR. EBERSOLE: Is there a statement in being that 1
22 says you shall not allow this steam generator to run dry, or is 23 that just taken for granted?
24 MS. SLOSSON: Pardon me?
25 MR. EBERSOLE: Is there an existing statement that O Heritage Reporting Corporation (202) 628-4888 l
155
(-
\> 1 says, don't let the steam generators run dry?
2 liS . SLOSSON: There isn't. The problem was that the 3 operators -- there are a couple of cases -- they just, were 4 coming out of the outage and during the outage, they had filled 5 and dried out the generators several times below 200 degrees, 6 and the operators were not, I guess that they were still in 7 that shutdown mode of thought.
8 MR. EBERSOLE: They were in never never land.
9 MS. SLOSSON: Right. And they kept thinking they 10 were going to get the aux feed pumps back, and they felt that 11 if management had, there are operating procedures to allow them 12 to supply water to the generators, and they have used them 13 before through the condensate pumps, that management would have 14 put, would have you know had them filling it filling the 15 generator supplying water.
16 And so there must have beun an overall scheme that 17 knew what was and overall management that knew what was going 18 on.
19 MR. EBERSOLE: There's an instructional void 20 somewhere.
21 MS. SLOSSON: Right. There's several. If I can go 22 to my next slide, I will talk about what the findings are.
23 (Slide) 24 An AIT was sent to the site and a confirmatory action 25 letter issued to the utility by which a regional administrator O Heritage Reporting Corporation (202.) 628-4888
. . ~.
l l
l 156 l
[} 1 would approve start-up from the outage.
2 As I discussed a little bit earlier, as far as 3 equipment damage, there really was no equipment damage. They 4 used up a couple of mechanical cycles on the steam generators.
5 But the findings from'the AIT were such that operations 6 management and shift supervisors were not adequately aware of 7 plant status, and as I said before, didn't provide adequate 8 contingencies for the operators.
9 The shift supervisors seemed to be a little bit too 10 involved in the hydrotest and involved with all of the 11 maintenance activities going on in the preparations for start-12 up and were not really paying as close attention to plant 13 status as they should have been.
() 14 The operators did not adequately report plant status 15 and activities to the shift supervisors. They did log out they-16 had zero level on the steam generators. The shift eupervisors 17 who were on when it went dry said that either they didn't '
18 notice it when they looked at the logs, or if the SRO8 told 19 them, it didn't register as being a problem.
20 And the operators were pretty complacent with leaving 21 it the way it was without you know emphasizing to their I 22 supervisors that shouldn't we be doing something about that.
23 MR. EBERSOLE: I can't help but think it's a good 24 thing they weren't running a fossil boiler.
25 MS. SLOSSON: The licensee exhibited a willingness to Heritage Reporting Corporation (202) 628-4888
157 1 deviate from approved directives and procedures and to perform 2 evolutions without procedures. In several instances in this, 3 they were operating either outside procedures or were 4 deviating. They made changes to procedures without following 5 the plant means to do that.
6 The operators also misapplied the EOP analysis, which 7 they seemed to be a little bit too familiar with the EOP back-8 up analyses that allows them to fill a hot dry steam generator, 9 but they were not in the initial conditions to apply those ,
10 procedures in this case.
11 MR. LEWIS: I'm probably the only one in the room who 12 doesn't know what it means to perform evolutions.
13 MS. SLOSSON: Good question.
14- MR. EBERSOLE: It me="a to do something.
15 DR. SIESS: Why did the steam generator boil dry?
16 MS. SLOSSON: Because the main steam isolation valves 17 were passing about 3000 pounds per hour, and they sare not
- 18 providing any flow.
19 DR. SIESS: And 200 gallons per minute water is not 20 enough to counteract 3000 pounds per hour of steam? I'm having l 21 trouble with units.
22 MS. SLOSSON: The 200 gallons per minute was -- I'm 23 confused now.
24 DR. SIESS: Well, you said they started the pumps --
25 MS. SLOSSON: No, they started the pump and then they O Heritage Reporting Corporation (202) 628-4888
158
-() 1 stopped the pump --
2 DR. SIESS: 200 gallons per minute wasn't getting 3 them anywhere.
4 MS. SLOSSON: No, no. They started the pump and they 5 got too high a flow. And so they stopped the pump.
6 DR. KERR: 200 gallons a minute is about 1500 pounds 7 a minute.
8 DR. SIESS: Yes, but I didn't know how many pounds of 9 steam.
10 MS. SLOSSON: They were not supplying. That was 11 after they had dried out anyway. That was after they had dried 12 out and they were not feeding the generator.
13 MR. MICHELSON: Can I ask a related question. Were
(~5 14 they on level control at all? In these conditions, they don't
\_)
15 go on level control?
16 MS. SLOS90N: They were on level control and did get 17 a low level alarm.
18 MR. MICHELSON: They weren't on level control or they 19 would have been making up. So they apparently weren't on level 20 control. They just sit there and if things leak out and unless 21 it's detect it by an enunciator or something, and I guess what 22 happened to the enunciators in this case? Weren't they 23 indicating low level? !
24 MS. SLOSSON: Yes, they were.
25 MR. MICHELSON: They were lit and they ignored them.
O. Heritage Reporting Corporation (202) 628-4888
159 j{} 1 MS. SLOSSON: No,.the operators were fully aware 2 that the generator went dry.
-3 MR. MICHELSON: Well, that's inexcusable then.
4 DR. KERR: Carol, I thought almost all of the PWRs 5 went on manual control from about 0 to 15 percent of full 6 power.
7 MR. MICHELSON: I just wondered whether they used 8 manual or automatic at all, verification.
9 MR. WENZINGER: I'm Ed Wenzinger from Region 1. They .
10 had thought all during this event that the steam generator was 11 fully bottled up, there was no feed and as far as they knew, 12 no bleed, so it was going to just simply stay there hot and 13 bottled up. Unfortunately, the MSIV was leaking so it was in 14 fact bleeding off and there was no feed. That's why it went
(]}
15 dry.
16 MR. MICHELSON: And that's standard procedure?
17 MR. WENZINGER: For this kind of an evolution, that's '
18 correct.
19 MR. MICHELSON: Isn't there detection for main steam 20 isolation valves leakage under normal conditions. There's no 21 way to know whether steam's? In an accident condition, what do 22 you do? You detect it by radiation right?
23 MR. LOBEL: You thinking of PWR leakage control c 24 system, their detection system.
25 MR. MICHELSON: That's right. Here, there'd be Heritage Reporting Corporation (202) 628-4888
160 O
(_) 1 nothing, right?
2 MR. LOBEL: There's the drains. If they knew they 3 should be looking, they could look for flow in the drains and 4 that kind of thing.
5 MR. MICHELSON: There's nothing automatic about it, 6 you don't carry the PWR?
7 DR. SIESS: The traps are bleeding it off, and 8 they're not monitored, I don't believe.
9 MR. LOBEL: No. What I meant was, if you knew to 10 look, you could ?.ook. But there isn't any monitors.
11 MS. SLOSSON: Now, I think they knew, they did know 12 that they were getting some steam passed those main steam 13 isolation valves. I think that the real thing here is that the
[) 14 operators were fully aware that this generator was dry. It was 15 not the case where the operators weren't aware for 24 hours2.777778e-4 days <br />0.00667 hours <br />3.968254e-5 weeks <br />9.132e-6 months <br />.
16 It was plant management and operations management and shift 17 supervisors that were not aware.
18 The reactor operators maybe had they made management 19 more aware this wouldn't have continued or as it decreased and 20 it went off of the narrow range scale, maybe operations 21 management would have taken son.e action. But the operators i 22 were aware of what was going on.
23 DR. KERR: Whose findings are these?
t 24 MS. SLOSSON: These are the findings from AIT which i
25 was an augmented inspection team. It was a Region 1 NRR team.
, ( Heritage Reporting Corporation l
(202) 628-4888 l
t .
161
(
ss/ 1 DR. KERR: Now, if I believe the first bullet, the 2 AIT must have concluded that there was no particular reason why 3 the operators should have been concerned. Because it says that 4 the safety significance and potential adverse impact is minor.
5 Is that a valid interpretation?
6 MS. SLOSSON: I think it is a valid. But I don't 7 think you want to do that.
8 MR. WARD: But did the operators have enough 9 information to have made the judgment that it was minor?
10 MS. SLOSSON: I don't believe so. I think that their 11 information was, they may have had information that was 12 indirectly correct. Their information was based on an non-13 applicable EOP back-up analysis, okay. That they thought it 14 would be okay to do this, not necessarily even fill a generator 15 -- they were going to fill it without a precedure even to do 16 this.
17 MR. MICHEL3ON: They knew that the water chemistry 18 was extremely important at all tiraes, including shutdcwn.
19 They knew you aren't supposed to dry these things out and lose ,
20 all control over the environment. They must have known that.
21 MR. EBERSOLE: Did anybody ask the operators how far 22 were you going to go up before you decided you better not fill 23 it.
24 MS. SLOSSON: They were going to actually had they 25 been getting lower flows through the aux feed pump they O Heritage Reporting Corporation (202) 628-4888
162
("' 1 probably would have continued to fill up using the cold water
'\-)
2 if station management had not found out about it and directed 3 them not to.
4 MR. EBERSOLE: This gets into that category of the 5 things that people, actions that they i ske because they're not 6 explicitly prohibited from so doing. You know, the negative 7 set of instructions, you shall not. And that's a classic void 8 in the operational instruction book, you know, the red letters.
9 DR. SIESS: Why should we be worrieri about what they 10 did? What possibly could have led t, something that affected 11 the health and safety of the public eith r during this incident 12 or following it?
13 MS. SLOSSON: I think it's the lack of respect for
(} 14 procedures here. The operators were deviating from procedures 15 and --
16 DR. SIESS: The fact that they didn't take proper 17 actions even under a nonsafety condition, you're afraid it 18 would spill over to a safety condition?
19 MS. SLOSSON: Yes, sir.
20 DR. SIESS: Sloppy, in other words. It seems to me 21 the licenste would have reached some of these conclusions, too.
22 It's his plant.
23 MS. SLOSSON: They did and as a result, there are l 24 several corrective actions, that are my next slide, which is:
l 25 (Slide)
Heritage Reporting Corporation I
(202) 628-4888
163 1 ) 1 There are now mandatory board walks by the shift 2 watch supervisor, the SRO and the STA three times per shift. At 3 Indian Point Two, they're on 12-hour shifts, 4 DR. SIESS: What's the SWS7 5 MS. SLOSSON: Shift watch supervisor.
6 MR. MICHELSON: They didn't need a board walk to know 7 that they were low. They got an enunciator window and alarm 8 long, long into the game they had that. And it didn't go out.
9 They acknowledged it, but it didn't go out.
10 DR. KERR: Well, Carl, this is punishment, you know, 11 like having to walk an extra --
12 MR. MICHELSON: You don't need to walk the boards 13 unless you've got a problem unless your enunciator panel fails.
14 Then you need to walk the boards.
15 MR. WARD: Well, wait a minute, though. Let's see.
16 You said the operators were aware?
17 MS. SLOSSON: Yes.
18 MR. WARD: Did that include the SRO?
19 MS. SLOSSON: Yes, the SRO was now the shift watch 20 supervisor if at Indian Point Two the shift watch supervisor is 21 behind a very heavy door without a window and so they come in 22 but they're not there observing operations through a window or 23 able to see things right you know all that often.
! 24 In this evaluation I guess or in this particular 25 case, the shift watch supervisors were there very involved in O Heritage Reporting Corporation (202) 628-4888 i
164 1 the hydro test very involved with getting finishing up the post
[}
2 maintenance tests so that they could continue to start up. And 3 they may or may not have been doing what they should have been 4 doing as far as keeping aware of plant status.
5 MR. EBERSOLE: I've endorsed continued scanning of 6 the operation in the control room with scanning tv cameras.
7 And I think this would be a good case where it might have 8 intercepted this whole situation.
9 DR. SIESS: Is the shift watch nupervisor at a higher 10 level of competence that he would have recognized the 11 significance of this if he had known about it?
12 MS. SLOSSON: Through interviews with the shift watch 13 supervisors, they indicated that had they known about.it when 14 it was actually drying out, they would have taken action and
(]}
15 would not have allowed it to dry.
16 There's a distinction in the plant itself that the ,
17 shift watch supervisors are non-union; the operrtors are union.
18 So they're management and one is union. I don't know if that 19 is relative particularly not. They now have they used to have ,
20 the STA was a 24-hour shift where the person performed their 21 duties during the day and became the STA at night also.
22 That has been changed. They now have an STA assigned 23 to the shift, and that STA's function has been more 2
24 definitively explained so that they know how much they have to 25 keep aware of plant status. They cannot just go and go to O Heritage Reporting Corporation (202) 628-4888 4
,,-.-,e.-,,.4 , . _ . - . - . _ _ , , . , , , . _ , , . - . , , _ _ . , _ _ ~ , . , . , . - . _ , . - . _ , . , _ - - . . ., , , - , . - . . _ . . , _ _ . . - , , , - - , _ . _ _ _ . . -
165
) 1 sleep during their STA watch.
2 MR. WARD: Is that the result of this? ,
3 MS. SLOSSON: That's the result of this, yes.
4 DR. SIESS: Well, this was at 8:00 o' clock in the 5 morning when it boiled dry. Wasn't the STA there then?
6 MS. SLOSSON: STA was there but STA was not aware.
7 And it was boiling dry. It took about 36 hours4.166667e-4 days <br />0.01 hours <br />5.952381e-5 weeks <br />1.3698e-5 months <br /> for this to go 8 dry.
9 MR. MICHELSON: How many alarms do you get before it 10 boils dry? Is there two or three levels of alarms that come 11 in as it goes down, or only one?
12 MS. SLOSSON: I'm not sure if there's a low and a low 13 low level. I'm not sure. ,
} 14 MR. MICHELSON: I don't know. I don't remember that 15 detail on those generators. It would be interesting to find 16 out how many times they cleared the alarms as the thing came on 17 down. You can go back to the log.
18 MS. SLOSSON: The logs are all circled out and j 19 they're indicating zero level.
20 DR. SIESS: The alarms just say it's dry. They don't 21 say do anything.
22 MR. MICHELSON: It's incomprehensible to me that with-23 the training these people get and the understanding that they 24 have and everything, that they would sit there and allow one of 25 these to boil dry.
4 O Heritage Reporting Corporation (202) 628-4888 t
166 1 MS. SLOSSON: That was incomprehensible to a lot of
({
2 people.
3 MR. MICHELSON: It's as serious as finding out that 4 people are sleeping on duty. Just as serious as that, I think.
5 MR. EBERSOLE: Look at TMI-2.
6 MR. MICHELSON: I can't believe they'd' do that. It's 7 incomprehensible.
8 DR. KERR: Please continue.
9 MS. SLOSSON: The operations management is now making 10 sure that they take more frequent periodic tours of the control 11 room, and question the operators about plant status to make 12 sure everyone's aware.
13 They have training of operators concerning a steam
() 14 generator dry out event, telling them that they shouldn't be 15 allowing a steam generator to go dry.
16 They have increased emphasis on communications and 17 operator assertiveness, that if an operator feels that there's 18 a problem in the plant, that they shouldn't think that 19 management knows all and they should question the fact that the 20 generator was going to dry.
21 MR. WARD: So assertiveness means that they should be 22 pushy?
23 MS. SLOSSON: More vocal.
24 MR. WARD: Speak up.
25 MS. SLOSSON: Probably.
O Heritage Reporting Corporation (202) 628-4888
167 1 MR. MICHELSON: We may find that there's an 2 underlying reason for all this such as animosity between some 3 of the operators and the management or something of this sort.
4 I bet if you start digging very long, you'll find reasons why 5 they didn't communicate as they should have.
6 MR. LEWIS: Besides the fact that they were dumb 7 enough to let it happen.
8 MR. EBERSOLE: Wayne, the books seem interested in 9 this but I think they may be equally interested in the other 10 side of the coin. We're about running out of time.
11 MR. LANNING: Are you finished?
12 MS. SLOSSON: Yes. .
13 MR. LANNING: Good.
14 The following event we'll talk to today is a scram at 15 Nine Mile Point Unit 2 involving overfilling a reactor vessel.
16 Mr. Ed Wenzinger is the Chief of the Projects Branch at Region 17 1, will be our speaker.
18 MR. WENZINGER: Good afternoon, my name is Ed 15 Wenzinger. I'm Branch Chief of Projects in Region One. And 20 I'd like to discuss the event that took place on January 20th 21 at Nine Mile Unit 2. There are basically two phases to this 22 event. One was the loss of feedwater and the decreasing level 23 in the vessel.
24 (Slide) 25 And that's shown on the first slide.
O Heritage Reporting Corporation (202) 628-4888
168
'rg
(,j 1 Can you see that in the back, Dr. Kirk? And you can 2 see it on the handout.
3 The second phase which I'll go into on the next slide 4 was the overfill which is probably a bit more interesting. But 5 let me go through the sequence of events.
6 Time equals zero was the reaptor trip that occurred 7 at 159 inches of level which is where it should occur. At 8 about a minute before that, a technician, a non-licensed 9 equipment operator was working on the air system. And due to 10 an error, he actually isolated both filter trains on the 11 instrument air system, and basically shut off the instrument 12 air system.
13 He looked at the equipment train A and train B, and
() 14 looked at the valves on train B, and by looking at them, 15 concluded that the valves were open, so that it was all right 16 to shut train A. Well, as it turned out, he was incorrect.
17 Train B was in fact closed.
18 As a result, air pressure was lost. That was at about
- 19 one minute before the trip. Now, as a result of losing air 20 pressure, the minimum flow valves for the condensate pumps, the 21 condensate booster pump, and the feed pumps failed open, i 22 Now, to understand what that did, let me show you a 23 diagram of that system. This is a simplified diagram, and as i 24 you can see, the minimum flow valves when the open will starve 25 flow to the other pumps.
O Heritage Reporting Corporation (202) 628-4888
169 bss/ 1 Having done that, the result was that the feed pumps, 2 the one condensate booster pump tripped and they tripped 3 because of Aow suction pressure.
4 At this point, the operators seeing that the level 5 was coming down, took manual control of the flow control valver 1
6 and as it turned out, the flow control valves had already gone 7 automatically to full open. But nevertheless, their training 8 called for taking manual control, since there had been some 9 problems in the past with the feed water control valves.
10 They had some oscillation problems as you may be 11 aware. This plant at that time had not yet reached a hundred 12 percent power and had not yet declared commercial operation.
13 They had had some problems with these feed control valves
() 14 oscillating, so the operators had been trained to take manual 15 control which they did. And then called for them to be fully 16 opened. And since they were already fully opened, what he did
. 1 17 had no effect.
18 In fact, it couldn't possibly have helped because the 19 starvation of water from these pumps is what resulted in the 20 level decreasing and eventually at 159 inches, the plant did in 21 fact trip as it should have at that level. The vessel level 22 continued to decrease, again for the same reason. And HPCS and 23 RCIC automatically turned on at 110 inches, which is where they 24 ought to.
25 As a result of their turning on, the level began to
() Heritage Reporting Corporation (202) 628-4888
. .~ __ . __ _ ._
170 (3 1 turn around and started back up again.
V.
2 Just let me add here as a side note that around that 3 same point, the containment isolation system should have 4 initiated and it did not. And the reason it did not was that 5 their set points had been in error. And I'll come to that 6 again-later.
7 As the HPCS and the RCIC pumped water, the vessel 8 level then began to increase and as it increased and approached 9 level 8 which is where the HPCS and RCIC would be turned off 10 automatically, the operator apparently concerned for the vessel 11 level getting too high did in fact secure manually the HPCS at 12 202 inches. The RCIC injection automatically stopped, again, 13 as it should have.
14 Now the reason that the vessel level was coming up
(])
15 was the pressure had decreased during this event, and since tht-16 condensate and condensate booster pumps were still running, and 17 their condensate booster pump outlet pressure had exceeded the 18 pressure in the vessel, there was then water being injected 19 through that flow path.
20 MR. EBERSOLE: Why did the pressure come down?
21 MR. WENZINGER: The injection of the HPCS and the 22 RCIC cold water was basically the reason.
23 At about somewhere around four or five minutes or so, 24 vessel level reached about 202 inches, and that's basically the 25 end of phase one which was the loss of feedwater.
Heritage Reporting Corporation (202) 628-4888
171
() 'l MR. MICHELSON: Now, you say it dropped to 655 or 625 2 pounds. That's not anywhere near condensate capability.
3 MR. WENZINGER: Condensate booster pumps are capable 4 of putting out over 675 pounds and in fact that's about what 5 they were putting out at the time. It would have to be the 6 booster pumps because the condensate pumps, you're right, can't 7 do it by themselves.
8 MR. MICHELSON: Well, not all boosters can go that 9 high either.
10 MR. WENZINGER: Correct. But again, in this case, 11 they could.
12 MR. EBERSOLE: All this cold water you poured in made 13 the pressure come down, didn't it?
14 MR. WENZINGER: Sir?
15 MR. EBERSOLE: All this cold water brought the
, 16 pressure down.
17 MR. WENZINGER: Yes, that's why it-came down.
18 MR. EBERSOLE: There was no significant thermal 19 training was there?
20 MR. WENZINGER: Not that we are aware of, no. It was 21 reasonably gradual.
22 At about this time, the fellow who had done the dirty 23 deed discovered what he had done and simply reversed *As 24 actions and instrument air pressure was restored. Ane .
25 course, the result of that was that the feed water minimum flow O Heritage Reporting Corporation (202) 628-4888
172
() 1 valves and the other minimum flow valves that had opened simply 2 went back to their normal state which was closed, due to the 3 instrument air pressure coming back.
4 This was about five minutes into the transient after 5 the scram. About a quarter of a minute after that, when the 6 feed water pressure increased fairly rapidly, and this was 7 because this is the point at which the pressure had decreased 8 to the point where the condensate booster pumps were 9 significantly higher outlet pressure, the vessel level began to 10 rise rather rapidly.
11 MR. EBERSOLE: I'm thinking that's about 400 psi, am 12 I right?
13 MR. WENZINGER: What's that?
() 14 MR. EBERSOLE: 'That condensate booster pump pressure.
15 MR. WENZINGER: No, it's 600 and 50, 60 pounds.
16 MR. EBERSOLE: I thought it was lower than that. All 17 right, thanks.
18 MR. WENZINGER: In fact, it was calculated later on 19 that it may have been as high as maybe two million six pounds 20 per hour injection at that time.
21 The operator in an attempt to -- and of course there 22 are level alarms and one of the responses to the level alarm is 23 go to the flow control valves and attempt to shut them. Well, 24 as it turns out, unfortunately for the operator, there was a 25 little gremlin if you will in the flow control valve circuit.
O Heritage Reporting Corporation (202) 628-4888
173
{} 1 The little gremlin, what it amounted to was a circuit design 2 deficiency which caused the valve to lock up, meaning it went 3 part closed and wouldn't go any further closed. The operator, 4 however, on the knob that's on the cor. troller, called for the 5 valve going full closed, and the indication on the controller 6 in fact did call for valve full closed.
7 What he failed to do, however, was to and the classic 8 error was to look at what was actually the position of the 9 valve. He didn't do that. He only looked at the demand 10 signal. The demand signal said full closed. The actual 11 position showed the real position.
12 MR. EBERSOLE: Why didn't he trip the booster pump?
13 MR. WENZINGER: Because he had been taught not to ue
/'\ 14 that. Apparently the design of these booster pumps is such O
15 that if you shut them off when they're running and then attempt 16 to start them, there's a caution first of all that you have to 17 wait a number of minutes before you do that because o.t high in 18 rush currents and typically they won't start right away.
19 And their training is that's one of the last things 20 you do. You don't do that. Your principal means is by means 21 of the flow control valves, and there are other ways which 22 eventually did save the day. But not until the steam lines 23 were flooded, unfortunately.
24 At any rate, things were moving fairly fast because 25 of the large amount of water being pumped in, and about 7.7 Heritage Reporting Corporation (202) 628-4888
174
() 1 minutes or thereabouts -- I have the trace if you'd like to 2 look at it -- the steam lines clearly. began filling and it took 3 maybe a little less than four minutes to fully fill them until 4 the level continued on up.
5 Eventually, one of the assistant operations 6 superintendent found himself in the control room. And looked 7 over the situation and quickly realized what had occurred. He 8 understood the malfunction with the valve, and he also 9 understood, from memory apparently, what were the other 10 alternatives that he had.
11 There are valves that are downstream of the flow 12 control valves. He ordered that they be shut. And then went 13 over and removed the full closed algnal to where it was an open
(]) 14 signal and then gradually' turned it to full closed which then 15 closed the flow control valves. The other valves that are in 16 series with the flow control valves are motor operated valves, 17 and take a little bit longer.
18 The two apparently went shut about the same time at 19 about 13.5 minutes after the scram terminating the event.
- 20 After that, everything was quiet. Now, the steam lines in the 21 vessel is just plain full and the water then just gradually 22 drained out via the reactor water clean-up system and the steam 23 line drains.
l 24
25 and so forth?
O Heritage Reporting Corporation (202) 628-4888
~ _
175
( 1 MR. WENZINGER: As far as we know, it went'down every.
2 line'that is connected to the steam lines.
3 MR. MICHELSON: No high level alarm?
4 MR. WENZINGER: Yes.
5 MR. MICHELSON: And he ignored those?
6 MR. WENZINGER: No. He simply responded to those and
-7 unfortunately did the wrong thing because of the circuit design 8 problem, he got caught.
9 MR. MICHELSON: How did he respond to the high level 10 alarm?
11 MR. WENZINGER: He attempted to close the flow 12 control valves, which failed.
13 MR. MICHELSON: But he could see from his lights that 14 they weren't closed, couldn't he?
15 MR. WENZINGER: He did not look at the valve position-16 indication. He looked only at the demand signal.
17 MR. MICHELSON: Well, he sure was taught to look at 18 lights as well to make sure the valve was closed if there was 19 any doubt.
20 MR. WENZINGER: He knew that it was not closed. He 21 did not know why,
- 22 MR. MICHELSON
- Were any of the hangers damaged or 23 anything like that?
i 24 MR. WENZINGER: They were inspected and no damage was l
25 found.
O Heritage Reporting Corporation (202) 628-4888 l
y .g-e+me.w-.-p., - - - - - . -e - p e ,_--
- m. - - _._ .
- 176
{} 1 2
MR. EBERSOLE:
the lines would carry full?
Had it been validated in advance that 3 MR. WENZINGER: They had to already done that test.
4 The did it again that day.
5 MR. SHEWMON: Sir, this valve that hung up at 80 6 percent open, when the supervisor came in, did you say that he 7 basically stroked that? He opened it all the way and then 8 closed it?
i 9 MR. WENZINGER: He took the controller and adjusted ,
10 the control signal to what would be full open. And he had to 11 do that to clear the electrical lock up. Once he did that, if 12 he moved it slowly towards the closed, the valve in fact would 13 go closed. This was a malfunction that would only occur if you 14 attempted to close the valve quickly.
(~}
15 MR. EBERSOLE: Well, here's a plant that had no 16 trouble. It held the water up and the pipes didn't fall down 17 or anything. Is it common knowledge that all steam lines can
. 18 take that load?
19 MR. WENZINGER: It was common knowledge at Nine Mile 20 simply because they'd already done it.
21 MR. EBERSOLE: I mean, across the board, the other i
22 nine --
23 DR. SIESS: It's a design basis isn't it?
- 24 MR. WENZINGER
- I have asked that question at several 25 of the ballers I've been to, and I've gotten mixed responses j
() Heritage Reporting (202) 628-4888 Corporation t
m
177
() 1 from different people as the same plants.
2 MR. WARD: Many of them are not designed fo it but 3 that doesn't mean they can't take it.
4 MR. WENZINGER: Correct.
5 MR. MICHELSON: Well, is there a requirement in that 6 area?
7 MR. WENZINGER: There are high level trips in HPCI 8 and RICI, yes.
9 MR. MICHELSON: But not of the main feed water?
10 MR. WENZINGER: No. Main feed water pumps keep in 11 mind, weren't running.
12 MR. MICHELSON: No. But that was my next question. I 13 thought they tripped both the pump and closed the feedwater
() 14 control valves.
15 MR. WENZINGER: They tripped the feed pumps that 16 weren't running. What actually caused the injection was the 17 condensate booster pumps.
18 MR. MICHELSON: But the answer to my question is, 19 there were no automatic high level trips, anyway?
20 MR. WENZINGER: Not of the condensate booster pumps.
21 MR. MICHELSON: No, that wasn't the question. The 22 high level trip to isolate the main feed water?
23 MR. WENZINGER: To turn off the feed pumps, yes, but 24 they weren't running.
25 MR. MICHELSON: You're answering it indirectly, but O Heritage Reporting Corporation (202) 628-4888
178 O
(/ 1 it does not close the valve, it just trips the pump?
2 MR. WENZINGER: No. That's correct.
3 MR. EBERSOLE: The one funny thing you're saying, the 4 condensate booster pumps were running. The operator was 5 reluctant to stop them because he would want to start them 6 again?
7 MR. WENZINGER: Yes.
8 MR. EBERSOLE: It's not to my knowledge anyway that 9 there shouldn't be a standing requirement that he could stop 10 any big motor and restart it, but he can't make several starts 11 in succession.
12 Charlie, am I wrong?
13 MR. WYLIE: What's that?
14 MR. EBERSOLE: He wouldn't stop the condensate 15 booster pumps because he was afraid he couldn't start it again.
16 All it was doing was running. When you stop a running motor at 17 normal current, aren't you entitled to initiate a restart?
18 MR. WYLIE: You can, 19 MR. EBERSOLE: You can't make several successive 20 starts?
l 21 MR. WYLIE: You can't make several. You can make one l
l 22 hot, normally two cold.
! 23 MR. MICHELSON: Depends on how you bought the motor.
24 DR. KERR: Let me suggest we can explore this in more 25 detail later on.
1
(
Heritage Reporting Corporation l (202) 628-4888 l
179 1 MR. MICHELSON: One further point though that's
[
2 important and that is on A-47, are those people now aware of 3 this event, because I think this invalidated some of the things 4 I read in A-47. This is the overfill of vessels in steam 5 generators. I was surprised.
6 MR. WENZINGER: The analysis that they did at the end 7 after it was all over with confirmed that all the stresses were 8 acceptable and I guess that wasn't a big surprise to anybody.
9 They did change the design of the control circuit, and that's 10 subsequently been fixed.
11 I'd mention the containment isolation valves and 12 their set points. It turns out that there wasn't a big error.
13 The level actually turned around just before they probably
(} 14 would have gone off.
and the limiting safety system setting.
They were set between the nominal value 15 The reason they were 16 set incorrectly is in going from inches of level to milliamps 17 of current, the nominal value of the span in terms of inches is 18 0 to 205, they thought. It turns out it was really -5 to 205, i
j 19 and that was the difference that they were off.
20 It just turned around just at the right point that it 21 just missed that. It wasn't a very big error. Those have 22 also been fixed.
23 They've also gone through their procedures and I ,
24 guess the most important thing is to make sure that they 25 monitor vessel level and certainly to monitor valve positions,
( Heritage Reporting Corporation (202) 628-4888
_ , - _ - _ ,._ _ . , _ _.___ . __ - - , _ . _ . _ . _ _ _ _ _ _ _ _ _ . _ . . _ . _ _ - . , - -_ ~ _ _ _ , - .
180 1 that is, the real position as opposed to the demand position.
2 DR. KERR Any questions?
3 MR. WYL1E: Well, what regulatory action comes out of 4 this?
5 MR. WENZINGER: It turns out there was one minor 6 violation associated with containment isolation valves of level 7 4, very trivial.
8 This is somewhat similur and that's why the two are 9 interesting together bet'reen the Indian Point 1 and this one.
10 The safety significance here I guess is concentrated 11 primarily in the operators and their cognizance of what's going 12 on in the plant, particularly the position of those flow 13 control valves.
14 DR. KERR But here they did not_ consciously violate 15 any procedures apparently.
16 MR. WENZINGER: Not to my knowledge, no, they did 17 not.
18 MR. MICHELSON: I guess they could have gotten a 19 pressurized thermal shock if they'd kept putting the water in 20 and cooling the system down enough over a period of time. I'm 21 surprised you just can't cut off the feedwater. That's why you 22 need to shut the feedwater valves.
23 MR. WENZINGER: All the cold water that was being 24 injected was stopped either manually or automatically and 25 apparently would have stopped automatically if they hadn't done O Heritage Reporting Corporation (202) 628-4888
181
( 1 it manually.
2 MR. MICHELSON: Well, it would have stopped when he 3- exceeded the head of the pump, of course.
4 MR. EBERSOLE: hre you going to investigate why he 5 couldn't probably best shut off the booster pump. That would 6 have been a direct action that would have stopped the whole 7 process.
8 MR. WENZINGER: Yes. Well, of course, closing the 9 control valves can now be done because that design deficiency 10 is gone.
11 MR. EBERSOLE: I guess I have more confidence in 12 turning the power off.
13 DR. KERR: Any further questions?
(/ 14 Thank you, Mr. Wenzinger.
15 Does that take care of it, Mr. Ebersole?
16 MR. EBERSOLE: Yes, thank you.
17 DR. KERR: We have a session on Rancho Seco scheduled i 18 next, and I am told that there is a request to set up some TV 19 equipment before we get started with Rancho Seco. I'm going to 20 declare a five-minute break for reorganizing the setting.
21 (Brief recess is taken.)
22 DR. KERR: I think Mr. Wylie is the cognizant 23 subcommittee chairman.
24 MR. WYLIE: Gentlemen, this portion of our meeting 25 concerns a briefing by the NRC staff of the restart of Rancho
- (:) Heritage Reporting Corporation (202) 628-4888
182
() 1 Seco nuclear-power plant. The owner and licensee of Rancho 2 Seco is the Sacramento Municipal Utility District.
3 The Plant's A&E-was Bechtel, and the operating 4 license was issued in '74. The nuclear steam supply system is 5' a B&W 916 megawatt electrical PWR with'a once through steam
) 6 generator similar to the Oconee, TMI, Davis, Basie and other i 7 systems supplied by B&W.
8 Rancho Seco has been shut down since December 26,
.9 1985, following a reactor trip and a severe overcooling 10 transient'as a result of the loss of DC electric power to the 11 unit's integrated control system.
12 Rancho Seco had previously experienced problems with 13 the ICS including a previous loss of DC power which also resulted in an overcooling event. The December 1985 transient
(]) 14 15 plus a history of other problems at Rancho Seco led the Staff 16 at NRC to direct that the unit remain shutdown until it had 17 made both physical and managerial upgrades sufficient to assure 18 safe start-up and operation of the plant.
. 19 Since the shutdown, extensive modifications to the 4
20 plant and changes to the plant's management have been made 21 which are directed toward insuring safe operation of the plant.
22 A restart SER was issued in October 1987, NUREG 1286, and it 23 contains open items that must be closed prior to restart. ,
24 The supplement to the SER has not been issued.
25 The restart consideration of Rancho Seco 1s scheduled Heritage Reporting Corporation I (202) 628-4888 4
P
_ m - ,
183 g
(_j 1 to come before the Commission on March 18, 1988. The ACRS has 2 not been directly involved to date with the shutdown and 3 restart considerations of Rancho Seco. However, the ACRS has 4 been following the B&W reassessment program and the staff's 5 review through its B&W Reactor Subcommittee which plans a 6 meeting at the completion of the staff's review. This includes 7 reassessment of the ICS.
8 I would point out that with the p?:esent schedules, 9 the Rancho Seco restart considerations by the commission will 10 precede the staff's completion of its review of the B&W owners 11 group reassessment program.
12 The purpose of bringing this before the ACRS is to 13 apprise the committee of the background and planned NRC restart
() 14 considerations, a branch of SECO and to give the committee the 15 opportunity to take whatever action it pleases.
16 I would remind the committee that it did write a 17 letter in regard to the restart of Davis-Besse, another B&W 18 plant, and also it wrote a letter on the B&W owner group s
19 reassessment program, both of those letters were written July 20 16, 1986.
21 We have Gary Holahan of the NRC Staff for Rancho Seco 22 who will brief us on the restart of Rancho Seco. And so I'll 23 turn it over to Mr. Holahan.
24 MR. HOLAHAN: Thank you.
25 My name is Gary Holahan. I am Assistant Director for n'
'~
Heritage Reporting Corporation (202) 628-4888
184 1 Region 3 and 5 Reactors. I have with me, George Kalman, who is V(~N 2 the project manager and will make the detailed part of the 3 presentation.
4 1.s was just mentioned, Rancho Seco has been shut down 5 for the last two years. The purpose of our presentation today 6 is to keep the ACRS informed of the status of the staff restart 7 review.
8 The presentation this afternoon will cover the events 9 leading up to reactor shutdown, the problems associated with 10 the December 26, 1985 event, the findings of the incident 11 investigation team that identified the problems that need to be 12 addressed.
13 I'll also speak to the licensees' and the staff'-
(} 14 actions addressing those problems, basically the state 15 readiness of the plant. As was mentioned earlier, Start review 16 is not complete. We do expect to issue a supplemental safety 17 evaluation report probably within the next week.
18 As of this point there are a few open items. We will 19 mention the status of those also.
20 I'd like to make one change to the information you 21 were given. At this time, the Commission meeting for restart 22 has been rescheduled for March 22nd. It's been shifted from 23 the 18th.
24 I guess with that introduction, I'd like to turn it 25 over to George Kalman, the project manager.
) Heritage Reporting Corporation (202) 628-4888
185
) 1 DR. MOELLER: The information that has been provided 2 to us says as I recall that even if the NRC staff or commission 3 approved restart, that the utility may not restart.
4 Is that correct?
5 MR. HOLAHAN: The plant is owned by the Sacramento r 6 Municipal Utility District and the Board of Directors has been 7 considering a number of options, one of which is permanent 8 shutdown of the unit. Those public hearings and activities are
, 9 still ongoing and the eventual outcome I think is somewhat 10 uncertain.
11 As far as the Staff is concerned, our review is i 12 continuing and is really not contingent upon those activities 13 at this moment. Of course, if they would decide to shut the 14 plant down, we would change the nature of our review.
15 At this moment, that hasn't been indicated.
16 MR. SHEWMON: Presumably they have led you to believe I
17 that they still want the option of operating the plant.
18 That's why you're continuing? Is that it?
19 MR. HOLAHAN: That's correct.
. 20 MR. REMICK: Charlie, in the case of Davis-Besse, if 21 I recall, we had the benefit of a draft SER which we reviewed, 22 and then we had the staff come in and fill us in on any changes
- 23 from that. I assume there's no draft copy of the SER before us
. 24 at the moment?
25 MR. WYLIE: No. I have a copy of the one that was
, Heritage Reporting Corporation
] (202) 628-4888
186
() 1 issued in October. The supplement has not been issued.
2 MR. REMICK: I was thinking of the supplement. ,
3 MR. WYLIE: No, we have not, sir.
4 MR. HOLAHAN: There is a supplement in some stage of 5 draft form. It's still underg'oing management review. I don't 6 believe it's been made available to the ACRS.
7 MR. REMICK: So what we hear today is basically what 8 that draft supplement says,.is that it?
9 MR. HOLAHAN: That's correct. ;
10 There are a number of items which are being closed in 11 that draft, and those will be fully shared with the Staff and 12 the ACRS, and the items that remain open, as well. !
13 MR. MICHELSON: Charlie, is there a particular action
() 14 item of any sort we're anticipating?
15 MR. WYLIE: Just whatever the committee desires.
16 MR. MICHELSON: Been no request for a letter?
17 MR. WYLIE: No.
18 DR. KERR Please continue.
19 MR. KALMAN: Good afternoon.
J l 20 I'm George Kalman, the project manager for Rancho 21 Seco. I've had that job for the past two and a half years.
22 As mentioned previously Rancho Seco was shut down by 1 23 the NRC in December, 1985. And the plant has remained shut 24 down in compliance with a confirmatory action letter issued by i 25 the Region 5 Administrator.
() Heritage Reporting Corporation 3
(202) 628-4888
187 1 The Commission meeting, as we said earlier, is
"}
2 . scheduled for March 22nd. And this will permit the operators 3 of Rancho Seco, the Sacramento Municipal Utility District, to 4 present a case of why Rancho Seco should be restarted, t
5 In preparation for a favcrable ruling from the 6 Commission, Rancho Seco is prepared to go critical during the 7 week of March 21st.
8 This presentation was prepared specifically for the i 9 ACRS to summarize the shutdown and related events at Rancho 10 Seco. Specifically, it will cover why Rancho Seco was shut 11 down by the NRC, what has been done over the past 27 months to 12 prepare the plant for restart, and what is the current plan 13 status and the current problems.
14 (Slide) 15 First a little background of SMUD and it's relevance 16 in this case. SMUD was formed in 1923 to supply power to a 17 640-656 square mile area of Sacramento County California. And 18 SMUD is governed by a board of elected directors, five people, 19 Initially, SMUD purchased power from various power 20 producers but as time went on, they expanded their operation 21 and began purchasing operating hydroelectric stations, 22 geothermal stations, some solar stations, a gas turbine, and 23 eventually in 1974, they were licensed by the Atomic Energy 24 Commission to operate Rancho Seco.
25 Again, Rancho Seco is a B&W designed nuclear system.
() !!eritage Reporting Corporation (202) 628-4888
188
} 1 It was built by Bechtel. And subsequent to licensing, SMUD 2 relied very heavily on Bechtel for engineering support.
3 The Board of Directors ran a very tight financial 4 ship at Rancho Seco. Staffing as a result was minimal.
y 5 Salaries at the plant were low compared to the rest of the 6 industry. The Board also had a hand in approving expenditures 7 at Rancho Seco even to fairly small amounts.
8 There were some benefits to this mode of operation.
9 The cost per kilowatt hour at Rancho Seco was relatively low ,
10 compared to the rest of the country. They were significantly r 11 lower than the neighboring utilities. However, there were also 12 inherent problems with this type of operation.
13 The tight operating budget eventually began to show 14 in lack of plant maintenance. There was essentially no 15 preventive maintenance at Rancho Seco, and compliance with the 16 regulations was cursory at best, if not outright avoidance.
17 Other features were that the plant availability was 18 relatively low and as a result of the poor operating 19 performance, NRC ratings were low.
20 Problems kept multiplying at Rancho Seco and came to 21 a head during the 1985 refueling outage. In an attempt to 22 restart from the refueling outage, numerous inadvertent reactor 23 trips occurred that eventually led to the December 26, '85 24 shutdown.
25 The trips are listed on the slide. There was a LOCA O Heritage Reporting Corporation (202) 628-4888
189 1 from a high pint vent. Feedwater problems which caused reactor 2- trips. Maintenance of aux feedwater pumps became evident. And 3 the last two trips involved the ICS.
.4 And finally on December 26, 1985, the reactor tripped 5 because of an ICS failure. And an overcooling transient 6 resulted.
7 (Slide) 8 The overcooling transient occurred early in the i
9 morning around 5:00 in the morning. The ICS failed which
, 10 caused a reactor trip and eventually a primary system 11 overcooling. The ICS is the integrated control system which is i2 common to B&W power plants. It's a load following automatic l 13 control of the --
14 DR. KERR: Excuse me, Mr. Kalman. I get the 15 impression from your introduction, and from other things I have 16 heard, is that the problems associated with this plant are not 17 primarily equipment associated. They are management associated 18 support associated.
I 19 If I'm mistaken, tell me. But if I'm not mistaken, 20 then it seems to me that details on equipment malfunction are 21 not going to help us very much in trying to reach a decision sa l 22 to what should be done.
23 It seems to me what we should be hearing about is 24 whether these perceived management and support problems have 25 been corrected.
O Heritage Reporting Corporation
, (202) 628-4888 t
1
.--...m._,_,.-- , . _ . _ . _ , , , _ , , _ , , . _ . , , , , , . , _ _ - , ,
190
() 1 MR. KALMAN: Sir, you've hit on the root cause of the 2 problem, and that's management. And we do get into that and it 3 is the main feature of the restart and performance improvement.
4 I'll go through these equipment problems more rapidly and get 5 to the management for you.
6 The ICS failure at Rancho Seco caused numerous events 7 to occur. Turbine bypass valves opened. Atmospheric dump 8 valves opened. Aux feed water controls open up to fifty 9 percent which was the failure mode of the ICS. Also, the main 10 feed pump ran down to idle speed.
11 The main feed pump run down caused the steam 12 generators to lose level and caused the primary system to 13 overheat. The reactor tripped in 16 seconds.
o
(_) 14 When the main feed water pump ran dcwn to idle speed, 15 the loss of discharge pressure caused the aux feed water pumps 16 to start. These began pumping cold feed water into the steam 17 generators. After the trip, there was only DK heat in the core 18 which was minimal because of the long shutdown. And as a 19 result, the overcooling transient commenced.
20 When the ICS failed, there was no control from the 21 control room of the ICS components. As a result, the control 22 room operators were required to go into the plant to manipulate 23 the manual valves. The immediately went to isolate the turbine 24 bypass valves and the atmospheric dump valves which were still 25 discharging steam out of the steam generators and causing O
V Heritage Reporting Corporation (202) 628-4888
191 (h
N-) 1 additional cool down of the primary system.
2 They were able to isolate those valves within nine 3 minutes. The aux feed water was still pumping water into the 4 steam generators. This was cold water. And even after the 5 steam drain was isolated, the steam generators continued to 6 cool the primary system.
7 operators found that they could not close the aux 8 feedwater control valves manually because of malfunctions.
9 They tried to isolate the aux feed lines with a manual valve.
10 That valve was stuck. And the primary system eventually 11 depressurized to about 1,000 pounds and 400 and some degrees.
12 The pressurizer level went off scale low and subsequent 13 calculations indicated that a bubble actually formed in the 14 reactor vessel head.
15 The RCS transient was rectified in 26 minutes.
16 During that time, the RCS cooled 180 degrees.
17 (Slide) 18 As I've already mentioned, the December 26th 19 transient in and of itself was not all that significant. In 20 fact, it may have been a relatively mundane occurrence at 21 another utility. However, based on the previous record of 22 marginal performance and a poor operating history, the NRC 23 decided to take a closer look at Rancho Seco at this point.
24 The Administrator of Region 5 issued two confirmatory 25 action letters which simply asked Rancho Seco to conduct a root O
a Heritage Reporting Corporation (202) 628-4888
192 1 cause evaluation of the reactor trip to justify why power 2 operations should be resumed, and to preserve the failed 3 equipment that resulted from the transient because NRC decided 4 to send an augmented inspection team to investigate the 5 situation at Rancho Seco.
6 MR. EBERSOLE: Could I ask a question?
7 MR. KALMAN: Yes, sir.
8 MR. EBERSOLE: What's unclear to me is whether these 9 unfortunate evolutions took place because of intrinsic design 10 of the Babcock, Wilcox control system which hadn't been 11 designed to go to a defined state after power loss. Or it was 12 anticipated and known by the operator that it would do this and 13 yet they had in place operator response that would intercept
({) 14 any bad consequences.
15 I can't divide this situation between mal-design and 16 mal-operation.
17 MR. KALMAN: We address that specifically if you 18 could wait two more slides.
19 The augmented inspection team came on site one day 20 after the transient. After two days of investigation, they 21 uncovered enough problems where they felt the investigation 22 should be upgraded and it was.
23 And the NRC initiated an incident investigation team 24 which reported to the site before the end of the year and l
25 stayed on-site for 17 days. And the product of their O Heritage Reporting Corporation (202) 628-4888
193 O 1 investigation was published as NUREG 1195, the report of the 2 incident investigation team.
3 The IIT found that the basic cause of the transient 4 was a faulty wire crimp to a power monitor in the ICS system, 5 The problems alth the aux feedwater control valve that would 6 not close manually was that operators used excessive force.
7 They used an extension bar to close the valve.
9 The valve was somewhat improperly assembled. There 9 was inadequate position indication. The operator kept turning 10 the valve after the valve was already closed, and by using 11 excessive force, it caused the seat to come off the stem and 12 the valve opened wide while he was trying to close it.
- 13 There was poor lighting in the area and there was O 14 evidence ef grevices demeee by use of extension bers in the I
15 past.
16 DR. MOELLER: When you say, the operator, is this 17 done from the control room?
18 MR. KALMAN: No. When power was lost to the ICS, 19 there was no control from the control room of ICS components.
20 Control room operators went to the plant and manually began 21 manipulating the equipment.
22 DR. MOELLER: And that is typically a function of a
, 23 control room operator?
l 24 MR. KALMAN: This was an emergency situation.
25 DR. MOELLER: But they don't call on the people who l
i O Heritage Reporting Corporation 1 (202) 628-4888 I
l l
194 l
() 1 really supposedly know something about the valves?
2 MR. KALMAN: Control room operators are responsible '
3 to know that kind of information. Usually,_they're way up from 4 non-licented personnel in a steam plant to control room 5 operators, and par *, of the training of control room operators 6 is to be able to manipulate valves.
7 MR. WYLIE: I assume somebody relayed to the control 8 room?
9 MR. KALMAN: Yes, sir.
10 HR. REMICK: To make sure that's clear, you said 11 control room operator. That can be confusing. Were they 12 licensed operators or non-licensed operatorc?
13 MR. KALMAN: Everybody responded. There were both
() 14 licensed and unlicensed operators responding.
15 MR. REMICK: A control room operator could be 16 licensed or non-licensed in some things?
17 MR. KALMAN: Yes, sir. They had both types in the 18 steam plant responding.
19 The problem with the aux feedwater isolation valve is 20 that the valve was seized because it turned out that there was 21 no lubrication of this valve most likely since the plant was 22 built.
23 During the recovery from the event, a make-up feed 24 pump was damaged and this was burned out because operators 25 secured the suction to the pump.
O Heritage Reporting Corporation (202) 628-4888
195
() 1 DR. KERR: Mr. Kalman, I hate to keep harping on the 2 same topic, but I do believe -- these are beautiful slides and 3 they have a lot of information on them. But I don't think, 4 from what you said earlier, that we're getting to the problem.
5 MR. KALMAN: The next slide and we're here, sir.
6 (Slide) 7 The IIT conclusions. The IIT looked into the 8 underlying cause of the problems at Rancho Seco. One of the 9 conclusions was that ICS failures in B&W plants are common, 10 And you see on the board, there is a list of similar problems 11 that occurred previously.
12 Other utilities had incorporated comoensating 13 features to deal with these problems. At Rancho Seco, there 14 were no procedures to deal with ICS failure. Numerous operator 15 errors occurred during the recovery from the transient. And it 16 appeared that operators weren't well versed on the consequences 17 of an ICS failure.
18 The event itself in December would have had 19 inconsequential results if SMUD had responded to numerous 20 warnings from the NRC. And those are listed there.
21 The IIT was critical not only of Rancho Seco 22 management for allov ag a plant to get into the condition that 23 Rancho Seco was, it was also critical of the NRC staff for not 24 following up on these various warnings that were issued to 25 Rancho Seco.
Heritage Reporting Corporation (202) 628-4888
1 196
(~) 1 All the warnings required responses. And the typical Ur 2 SMUD response was a very cursory response that came into the 3 NRC, and NRC didn't follow up on the inadequate response from 4 the licensee.
5 DR. MOELLER: In other words, they said that they had 6 not responded? That was their response?
7 MR. KALMAN: Very minimal response. Yes, we'll look 8 into it, or yes, we're looking at this. A dramatic example is 9 Item 5 there. The response to the NRC Order which ordered all 10 utilities to install an EFIC system which is an emergency 11 feedwater control system which is safety grade a hundred 12 percent redundant to the ICS system.
13 Rancho Seco responded to the NRC and said they had
() 14 alternate plans. And NRC didn't realize that their alternate 15 plans basically were that they were not going to install the 16 EFIC system.
17 (Slide) 18 DR. MOELLER: Who's in charge at the NRC? I mean, 19 isn't this an inadequacy on the staff, on the regional staff if 20 they don't keep up with these things?
21 MR. KALMAN: There definitely was inadequacy by the 22 NRC as well as with Rancho Seco.
23 DR. MOELLER: Are those people still working for the 24 Commission?
25 MR. KALMAN: I would imagine there's several that Heritage Reporting Corporation (202) 628-4888
197
() 1 are, yes. This has gone on for a ten-year period of time and 2 many people have been involved.
3 MR. LEWIS: But what you're saying has generic 4 implications because it's part of the whole package of whether 5 there should be external investigations versus internal. And 6 you're absolutely right. This is a classic case in which the 7 staff has trouble blaming itself.
8 MR. WARD: It didn't sound like that.
9 MR. LEWIS: No, he's doing a good job of blaming 10 itself.
11 MR. WARD: Yes. He's blaming the staff. The 12 ,uestion q is whether anything has happened as a result of that, 13 and I guess we still don't know that.
14 MR. LEWIS: Well, I thought the answer was that
(])
15 nothing has happened to the staff members who should have been 16 following this presumably. I thought we did hear that. Am I 17 wrong?
i 18 DR. KERR: I don't think we have a definitive answer 19 on that question because I doubt if he came prepared to answer 20 it. I'm not trying to Say that an answer shouldn't be sought 21 or that one does not exist. But I would be surprised if Mr.
22 Kalman knows the answer to this question completely.
23 MR. LEWIS: Yes. I'm willing to make a small bet 24 with you, if it's legal to make bets in the District of 25 Columbia, i
Heritage Reporting Corporation (202) 628-4888
198
() 1 DR. SIESS: Not in this building.
2 MR. KALMAN: That's not part of my presentation.
3 (Slide) 4 The next slide is the SMUD response to the trip.
5 Following the trip in the transient, SMUD identified the ICS as 6 the root cause of the problem. The fixed the valves. They 7 said that they would fix the make-up pump later and schedule 8 restart for March of '86.
9 This was a typical response that they were able to do 10 in all the past evolutions that had occurred.
11 DR. KERR: Mr. Shewmon, did you have your hand up.
12 MR. SHEWMON: I was just scratching my head or 13 something.
() 14 DR. KERR Let it not be said that I'm unobservant.
15 MR. KALMAN: In view of the IIT report, the SMUD 16 response to restart was unacceptable to the NRC staff. NRC 17 wanted a wider range improvement program at Rancho Seco.
18 The SMUD management which was in place since 1974 19 couldn't understand why the delay. This was the normal way of 20 doing buoiness and their initial response was to make repeated 21 requests of the NRC to get the plant restarted.
22 And when this failed to get the utility any place 23 closer to start-up, the SMUD management was replaced, and the l 24 SMUD Board hired an outside contractor called MAC headquartered 25 in San Diego, to manage plant restart. MAC came in with a O Heritage Reporting Corporation (202) 628-4888
199
() 1 plant manager and twenty some managers who took over all of the 2 critical operations at Rancho Seco.
3 MR. WARD: But the collective board remained?
4 MR. KALMAN: Remained.
5 DR. SIESS: MAC is permanent?
6 MR. KALMAN: No, it wasn't a permanent. It was a 7 temporary technique to restart the plant. It was never meant 8- to be permanent.
9 DR. SIESS: How long are they going to be around?
10 MR. .KALMAN: They're gone.
11 DR. SIESS: Oh, they're gone. They're gone but the 12 plant's not. restarted.
13 MR. KALMAN: That's right.
() 14 DR. MOELLER: Were they hired for their managerial 15 skills or were they also experienced in nuclear power plant 16 management?
17 MR. KALMAN: Both.
18 MAC came into the picture about May, 1986. In July 19 of '86, an action plan for performance improvement was 20 submitted to the NRC. The action plan focused on management, 21 plant hardware maintenance and operations.
22 Immediately under MAC, a nationwide search was made 23 for nuclear plant managers. MAC was there temporary and their 24 charter -- I'm not sure what the contract actually said, but 25 they were there until plant restart.
O Heritage Reporting Corporation (202) 628-4888
200
() 1 To help in the seafch for plant managers, the SMUD 2 board had religion by this time and increased the salaries of 3 the utility personnel. As a result, they were able to attract 4 quality managers from around the nuclear industry in this 5 country.
6 The action plan also described a diagnostic plan to 7 identify additional problems at Rancho Seco and the kind of 8 things they looked at are shown on the slide.
, 9 MR. EBERSOLE: What was going down with B&W all this 10 time?
11 MR. KALMAN: B&W was involved off and on with their 12 engineering expertise. They weren't involved in a managerial 13 capacity.
() 14 MR. WYLIE: Did they replace the plant manager?
15 MR. KALMAN: Yes, sir. MAC plant manager came in and 16 took over.
17 MR. WYLIE: But now they've replaced him with a 18 permanent plant manager, correct?
19 MR. KALMAN: Yes, sir.
20 MR. WYLIE: What is his background?
21 MR. KALMAN: MAC stayed in the picture --
22 MR. WYLIE: I m talking about the present plant 23 manager, the permanent manager.
24 MR. KALMAN: Yes. He was plant manager at Pilgrim.
25 I think he was plant manager also at the Arizona projects. And O Heritage Reporting Corporation (202) 628-4888
201
() 1 served as a consultant to the Board of Directors of SMUD for 2 several years beforehand.
3 MR. HOLAHAN: He is speaking of Mr. Endinini whose 4 title is Chief Executive Officer Nuclear. In other 5 organizations, he's probably closer to what you might call a 6 vice president. In terms of plant manager and he is on-site at 7 the plant.
8 In terms of something a little bit closer to -
9 operations that you'd normally think of as plant manager is Jos 10 Ferlich who was previously plant manager at Palisades plant and 11 has about 12 years of nuclear operating experience.
12 MR. WYLIE: Thank you.
13 MR. KALMAN: This is what actually evolved as the
() 14 performance improvement plan and this is what's evaluated in 15 the NRC safety evaluation, NUREG 1286. First of all, SMUD made 16 a detailed response to the IIT findings.
17 The initiated a program to review past events and 18 corrective actions to those events. They reviewed past audits, 19 both NRC audits, other organizational audits, internal audits.
20 And looked at the thoroughness of their responses to those 21 problems identified.
22 They conducted plant interviews to look for plant 23 hardware problems, maybe managerial problems. They also 24 involved B&W and the B&W owner's reassessment evolved at about 25 the same time. And because the problems at Rancho Seco, the C) Heritage Reporting Corporation (202) 628-4888
u 202
() i 571P program actually did focus on many of the problems at 2 Rancho Seco.
3 So these programs identified problems and led to 4 administrative and hardware fixes that SMUD committed to prior 5 to restart. One of the major areas of improvement was in the 6 maintenance plan. As I mentioned before, there was no 7 preventive maintenance plan at Rancho Seco.
8 They were able to obtain a manager well versed in 9 nuclear plant maintenance and develop both a corrective and a 10 preventive maintenance plan and troubleshooting plan and 11 whatever else is required to run a good maintenance 12 organization.
13 The other type of admini'strative fixes were
() 14 procedural. There was a major technical specifica an upgrade 15 program. The last item on the list is the system review and 16 test program. This was a major restart program which was 17 similar to what was done at Davis-Besse following their 18 transient earlier in the year.
19 However, it was developed to be a much more 20 comprehensive program at Rancho Seco. ..aacho Seco took 33 21 systems that were considered vital to successful plant 22 operations and looked at the design basis of those systems, 23 looked at the design, developed a tast program to test those 24 systems to make sure that the systems actually met the design 25 and went through and did the test evolutions. Some of those O Heritage Reporting Corporation (202) 628-4888
I 203
() 1 evolutions are still in progress.
2 (Slide) 3 DR. KERR: Who did this? Was this done by the 4 managers and new staff that MAC brought in, or was it done by 5 MAC or by some combination thereof?
6 MR. KALMAN: It's a combination. MAC brought in 20 7 people initially as the senior managers. They brought in other 8 expertise to help in various plant areas. And the plan 9 development was done by MAC, the old staff, and with input from 10 the NRC. NRC was going through a step by step review and 11 approval of these processes as they were going on.
12 DR. MOELLER: The 20 MAC people who came in, did they 13 replace people there or they supplemented them?
(q, 14 MR. KALMAN: It kind of worked out both. The program 15 that was set up was for the MAC person to be in charge and the 16 SMUD person who had the job as a helper. The eventual program 17 was to reverse those roles, and during this period when MAC was 18 in charge and SMUD was there, they were also looking for 19 outside input as far as personnel.
20 (Slide) 21 The next viewgraph shows the hardware that was 22 modified during the outage.
23 MR. EBERSOLE: Do they now use MOVATs or MOTES or 24 whatever it is that seems to be an on-coming essential common 25 practice in valve reliability?
p"'
Heritage Reporting Corporation (202) 628-4888
l 204
( -1 MR. KALMAN: They developed a valve maintenance 2 program during the outage which went through the first round of 3 valve maintenance, and I'm not sure --
4 MR. EBERSOLE: But without this typical background to 5 understand what valves are doing, I don't know of any 6 refurbishment program that will work.
7 MR. KALMAN: It was a MOVAT refurbishment program.
8 They hired outside professional to handle the MOVATs.
9 This slide is an example of the more significant 10 hardware projects that were undertaken. Most are completed at 11 this point, though some are still on-going. I'm not going to 12 go through all of them.
13 The EFIC installation is a direct response to the
( 14 December 26th transient. EFIC is the emergency feedwater 15 instrumentation and control. This is the safety grade back up-16 to the ICS that handles all the secondary type of emergency 17 components.
18 MR. EBERSOLE: Well, is it invoked now when the 19 standard equipment gets out of range, and does it send the 20 plant to ,ome stated configuration, probably shutdown?
21 MR. KALMAN: It shuts down. It's not a eystem that 22 complimen.s the cperation during power operation.
23 MR. EBERSOLE: But it takes over if there were a 24 shutdown?
25 MR. KALMAN: That's correct. It handles the O Heritage Reporting Corporation (202) 628-4888
205
() 1 secondary plant.
2- MR. MICHELSON: You said~1t was safety graded?
3 MR. KALMAN: Yes, sir.
4 MR. MICHELSON: How does it isolate the feedwater 5 system, say if it gets too high a level in the generator?
6 MR. KALMAN: The generator would be tripped.
7 MR. MICHELSON: A steam generator I'm talking about 8 now.
9 MR. KALMAN: The EFIC does not come into play until 10 some occurrence --
11 MR. MICHELSON: It won't prevent overfill then?
12 MR. KALMAN: No. It has a high level trip. I'm not 13 sure whether it works during normal operation or during EPIC
() 14 control operations.
15 MR. MICHELSON: Maybe the next time we hear from A-16 47, they'll tell us. I thought it did, but I was just trying 17 to verify it.
18 MR. KALMAN: It's addressed in a supplement that will 19 be coming out.
20 MR. EBERSOLE: Is it air driven?
21 MR. KALMAN: It's electrical. There are air driven 22 valves in the system, and there are back up air bottles to 23 provide --
24 MR. EBERSOLE: And it uses little checks when it 25 leaks so there's no supply of air? They didn't add any new 1
(~)
%/
\
I Heritage Reporting Corporation l (202) 628-4888 I
1 206
() 1 smaller high grade air systems?
2 MR. KALMAN: Yes,, sir, they did.
3 MR. WARD: Mr. Chairman, as you've suggested several 4 times, and I guess what's really of interest here, and I mean, 5 I think Mr. Kalman has done a good job in giving us the flavor 6 of what's going on. He's made some very frank comments which I 7 found-helpful.
8 But you know, this talking any more about the 9 hardware really isn't of particular interest here. I mean, I'd 10 rather hear something about who's on the plant staff now. How 11 are they different from the people who were there before. What 12 are the engineering resources that the utility has put in 13 place.
() 14 Without that sort of information -- I guess I don't 15 know why we're hearing this. Are the Commissioners going to 16 make a decision in a couple of weeks. Are we going to provide 17 them with our opinion on this?
18 If we are, we certainly haven't heard anything about 19 the essentials of the question. We've heard a little bit on 20 the oblique that things have gotten a little better but I 21 certainly don't think we're in a position to have an informeo 22 opinion about it.
23 MR. WYLIE: Well, I think that's true. The reason we 24 brought it before the Committee was to make the Committee aware 25 of the fact that the plan is going to come before the O Heritage Reporting Corporation (202) 628-4888
I l
207 i
()
im 1 Commission to be started up and to give the Committee an 2 opportunity to decide whether it wants to get involved in this 3 or not.
4 Now, as far as hardware is concerned, I'd like to ask 5 one question before we leave that. And that is, you're aware 6 of the B&W owners group plant reassessment program. Have they 7 incorporated in this plant all of the recommendations of that 8 group?
9 MR. KALMAN: They've incorporated more 10 recommendations than any other plant.
11 MR. WYLIE: Well, I meant, they've come up with 12 recommendations that the Staff has or are in the process of 13 evaluating. They have not finished their evaluation, a lot of
() 14 which has to do with the ICS. And my question is whether or 15 not they've incorporated those recommendations?
16 MR. KALMAN: They've agreed to incorporate more of 17 the recorumendations than any other B&W plant. They've divided 18 their proposed list of modifications into pre-restart and post-19 restart modifications. And there is a schedule which has been 20 reduced to --
21 MR. WYLIE: That still doesn't tell me whether or not 22 they've included those recommendations in that program.
23 MR. HOLAHAN: Can I answer the question?
24 Several of the B&W recommendations are in the
[ 25 program. Now, the B&W recommendations themselves are in two I
w d Heritage Reporting Corporation (202) 628-4888 l
208 1
() 1 categories. There are key items and there are non-key items 2 which presumably are less important. In addition to that, for 3 the whole restart process, Rancho Seco has set up their own 4 criteria for what items are sufficiently important that they're 5 required to be done before the restart and which less important 6 items of longer term improvement could be done at some later 7 stage.
8 And the Staff has reviewed the Rancho Seco criteria 9 for what's an appropriate restart item.
10 Now, Rancho has committed to implement all of B&W 11 recommendations which are applicable to Rancho. Not every one 12 of the recommendations is applicable to all of B&W plants. So 13 currently there are 207 recommendations, 150 of which are
() 14 applicable to Rancho Seco.
15 They have implemented all of those items which meet 16 their criteria on what they're calling their restart list. In ,
17 addition, they've implemented a number of items which don't 18 meet their restart criteria but because of convenience, they 19 haven't been gotten to already.
20 1 can read you the numbers associated with how many 21 but the notes really tell you.
22 MR. WYLIE: No. I don't think you ought to get into 23 the numbers.
24 MR. HOLAHAN: Let me leave it to say that the Staff 25 has been satisfied that the nost important among the B&W O Heritage Reporting Corporation (202) 628-4888
209
()- 1 reassessment items are all committed to on Rancho and those 2 which are deemed to be appropriate for restart have been 3 completed.
4 MR. WYLIE: You say the Staff. Does that include the 5 Staff's review of the reassessment program and particularly the 6 ICS?
7 MR. HOLAHAN: Yes.
8 MR. WYLIE: That's outside.
9 MR. HOLAHAN: Yes, but to go back to your comment, 10 that review is not complete and it's always possible that some 11 additional items might turn up. But to the extent we know the 12 recommendations at this moment, they are being implemented.
13 DR. KERR: Does that take care of your question, Mr.
( 14 Wylie?
i 15 MR. WYLIE: Yes, thank you.
16 MR. HOLAHAN: I'd like to make a comment on the 17 general question of reviewing hardware versus trying to 18 understand whether the management's better or not.
l 19 We looked at the hardware really for two reasons.
20 One is, there do appear to be some hardware deficiencies, or at 21 least there were hardware deficiencies that needed to be
(
22 addressed.
l 23 Secondly, it is very difficult to make a judgment as l
l 24 to how well management is performing or will perform when the l So although we've looked at the operating 25 plant is operating.
O Heritage Reporting Corporation (202) 628-4888 l
210 1 experience of these people, how many years of experience and 2 where they've been before, a major part of our understanding of 3 how well management's doing and what's different relating to 4- safety comes in fact from looking at what hardware they've put 5 in the plant, how well they've done it. Have their systems 6 review and test programs worked the way we think they should 7 work.
8 And so what you see listed as hardware items in fact 9 to reflect on management. Their ability to put those systems 10 in, their ability to implement those modifications effectively 11 to run those test programs all reflect on management. And 12 we've been very happy with that so far.
13 DR. KERR: Mr. Holahan, when I commented on hardware A
(-) 14 earlier, I was not talking about what was going to be done new.
15 I was commenting on detailed description of what went wrong 16 with existing equipment. I consider that very important but I 17 was told initially that the Staff did not consider that the ,
18 root cause of things.
19 And so I wanted to get to the root cause. Now 20 whether the root cause is management or not, I don't know, but 21 that's what we were told. And I agree it's hard to judge 22 management, but the Staff I thought had already reached a 23 conclusion and we were told in several ways that this was the 24 chief difficulty.
25 Did I misunderstand?
O Heritage Reporting Corporation (202) 628-4888
211
{} 1. MR. HOLAHAN: Well, I think this like many other root 2 causes was a complex situation. Clearly, management was the 3 root cause of the problem. A part of that root cause was 4 management not making hardware improvements when it should.
5 DR. KERR: Well, it seems to me that if that's a root 6 cause, what we need to focus on is what if anything has been 7 done to change that root cause. And I have not heard a great 8 deal about that other than that MAC came in and brought in a 9 couple of new people, one of whom had been at Pilgrim and one 10 of whom has been at Palisades.
11 And if my memory serves me correctly, both of those 12 plants have had significant difficulties. I don't know whether 13 they had these difficulties after these people left or whether 14 these people left because of those difficulties or none of the
'(])
15 above.
16 MR. WARD: Could I make a comment. I mean; I don't 17 want to -- I agree that making the, as Mr. Holahan said, making 18 judgments about the management directly other than by looking 19 at the hardware as one fruit of the management practice is 20 difficult. One of the reasons it's very difficult for you is I 21 suspect that you and all the rest of the staff looking at this 22 are hardware experts. That your experience and training and 23 interest lie in issues of hardware and not in issues of 24 management organization.
25 Now, maybe I'm incorrect in that, but could I ask you Heritage Reporting Corporation (202) 628-4888
212
.( ) 1 your background, Mr. Kalman? Is your background primarily in 2 would you characterize it as in issues of what we're calling 3 hardware here or in the softer science or art of management 4 organization?
5 MR. KALMAN: It's probably both, I think. I have 6 been a senior reactor operator and a commander in Navy Reserve.
7 A little bit of both 8 MR. EBERSOLE- Well, it seems to me the only 9 yardsticks of the presente of good management is the absence of 10 presence of equipment which is well or poorly operated. That's 11 the end product. I catch mysel.f always looking for the end 12 product. I don't know what kind of management characterization 13 could otherwise be used.
r
())
t 14 MR. WARD: Well, we say that there are a lot of other 15 things that are important in a plant. Things such as the 16 qualification and training of the operators. I don't know how 17 you learn about that by looking at hardware.
10 MR. EBERSOLE: Well, you've got to go further down.
l 19 I said not merely the presence or absence of equipment but how 20 it is operated.
l 21 DR. KERR: Jesse, we know that in the past the l
22 equipment was operated very poorly. That's a given.
i 23 MR. WARD: I don't think it's been demonstrated that l
24 the design or the equipment was inherently poor. I mean, we 25 didn't hear that today. We heard it was poorly operated.
O Heritage Reporting Corporation (202) 628-4888 i
l
213 1 MR. EBERSOLE: I think that a piece of equipment y("T 2 that didn't know where it was going when it lost power was 3 automatically poor design.
4 MR. KALMAN: I would think that's poor design, but 5 that poor design was recognized very early. It was corrected 6 in all other plants except Rancho Seco.
7 MR. EBERSOLE: Well, that was management.
8 MR. WYLIE: I noticed that the plant managers you 9 mentioned do not have B&W experience. Is that correct?
10 MR. HOLAHAN: I believe that's correct.
11 MR. WYLIE: Do they have any key management personnel 12 that have B&W experience?
13 MR. KALMAN: I can't think of specific examples at
- 14. this point.
([])
15' MR. WYLIE: So they fired a lot of key managers and 16 they don't have anybody with any B&W experience?
17 MR. KALMAN: I think it's also a concern of the Staff 18 that they haven't been able to operate a power plant in all 19 this time.
20 MR. WYLIE: I understand that. But if you're going 21 out and beat the bushes for managers, you'd certainly beat the j
l I 22 bushes for some B&W experience, I would think.
23 DR. KERR: Do you want to get to the current problem 24 status, Mr. Kalman.
l 25 MR. KALMAN: Okay.
l Heritage Reporting Corporation I
(202) 628-4888 i
i
214 1 (Slide) 2 This is the last slide which shows what is occurring 3 as of today.
4 The goal was to --
5 DR. KERR: Excuse me. Does this represent the 6 judgment of the people at the plant or the judgment of the NRC, 7 or.some combination thereof?
8 MR. KALMAN: I'll explain each as I get to it.
9 The plant schedule calls for criticality on March 10 20th, based on the SMUD schedule of events that have to take 11 place. The plant pressurized last night for the first time up 12 to operating pressure and no leaks have been uncovered at this 13 point.
() 14 The intent of plant management is to come to the 15 Commission meeting on March 22nd and tell the Commission that 16 they're ready to operate and everything is completed.
17 At the present time, we have an operational readiness 18 inspection at the site and they will be finishing their 19 inspection today. There's also a procurement inspection on 20 site.
21 Some of the problems that have not been resolved as 22 of yet, two additional diesel generators were installed during 23 the outage. The diesel generators developed unacceptable 24 vibrations very late in the game when they were tested at their 25 generating capacity. Those vibrations still have not been O Heritage Reporting Corporation (202) 628-4888
215
( 1 resolved. In fact, SMUD is coming in tomorrow to explain to 2 the.NRC what the status of the vibration problems are and what 3 the resolutions are.
4 Assuming that there is a favorable restart decision, 5 SMUD has-proposed to go through a five month power ascension 6 program. This is done to both test equipment and train 7 operators. They don't plan to be at full operating power until 8 some time in August or September of 1988.
9 The question you had initially was the Quest 10 recommendation to close Rancho Seco. The General Manager of 11 Rancho Seco assembled a group of consultants to evaluate the 12 financial options available to the district. The finding of 13 this group of consultants that was based on the financia]
/~N
(_7 14 pluses and minuses that it would be better for the rate payers 15 to close Rancho Seco at this point and buy power from other 16 utilities in the area.
17 The General Manager took this recommendation to the 18 Board of Directors and the Board of Directors considered it.
4 19 They met last night and proposed an alternative to put on a 20 referendum ballot in June, 1988, to have Rancho Seco operate
! 21 for 18 months on a trial basis to see how successful operations 22 would be. And then to let the voters decide again at the next 23 refueling outage whether they want Rancho Seco to continue 24 operating.
25 There's also a second referendum on the ballot in i
1 O Heritage Reporting Corporation (202) 628-4888
l-216
() 1 June which was proposed by a group in Sacramento opposing 2 nuclear power which says that they would like to close Rancho 3 Seco as soon as possible and decomminsion the plant. So at ,
4 this point, those two referendums will be on the ballot in 5 June. .
6 MR. WARD: So when we talk about root cause, the root 7 cause seems to have been an incompetent Board of Directors 8 that'n been there for 20 years and got in over its head.
9 Didn't have the competence and took many years to admit they 10 didn't.
11 So now you're going back to what seems to be an 12 incompetent electorate who has provided the board of directors 13 to look for a solution to the problem. It's a kind of
() 14 disgusting situation.
15 MR. LEWIS: That's California.
16 DR. SIESS: I'm glad you said that.
17 MR. LEWIS: I was the only one allowed to say it.
( 18 MR. KALMAN: The Commission, by the way, has asked 1
19 the Board of Directors to come to the Commission meeting on 20 March 22nd.
21 MR. WARD: Whose going to lead them.
l 22 DR. KERR: Does that conclude your presentation, Mr.
23 Kalman?
24 MR. KALMAN: Yes, sir.
l L 25 DR. KERR: Are there further questions?
() Heritage Reporting Corporation
- (202) 628-4888 l
i
217
.q
(/ 1 MR. REMICK: I would just have one question just for 2 clarification. My understanding is the Staff is not asking 3 ACRS to do anything. .This is just informative --
4 MR. HOLAHAN: That's correct.
5 MR. REMICK: -- at this point. Okay.
6 MR. LEWIS: Do we want to do something even if we're 7 not asked.
8 DR. KERR: And the Commission has not asked us to do 9 anything.
10 Okay. Well, Committee, what should we do?
11 MR. REMICK: Well, my feeling is though, if we do 12 something, I sure haven't heard enough to know what we'd do.
13 In any situation like this, whether it's a new license or the
() 14 few cases we've been involved in restart, we certainly always 15 hear from the licensee or the proposed licensee and have more 16 thorough information.
17 So if we do something I personally feel that we 18 couldn't do it based on what we've heard today.
l 19 MR. WYLIE: I would agree with that. We haven't i
20 really had the opportunity to dig into the management issues 21 and resolution of the technical issues. And the Staff has not 22 finished their evaluation of the B&W reassessment program. So 23 I don't know where we stand.
24 MR. EBERSOLE: One would tend to guess that buried 25 underneath all this are perhaps even more serious things than O
V Reporting Corporation Heritage (202) 628-4888
218
() 1 have been reviewed in the TVA case, records, proofs, the wholo 2 bit.
3 DR. KERR: Mr. Lewis?
4 MR. LEWIS: I think we should do something. I think.
5 we would be derelict to ignore a situation that has existed for 6 so long and is of such gravity. On the other hand it's 7 obviously so that we have not heard enough today to do'anything 8 in an informed way. And in particular, in fairness, we haven't 9 heard from the much maligned management of the plant.
10 What I would propose, Mr. Chairman, is that we take 11 this seriously. We have a fair amount of time. Get the facts 12 and then decide whether or not to do something. I don't think 13 we should just sit by and let this happen.
(/ 14 DR. KERR: I would propose that perhaps we could get 15 to a result more rapidly if we had a motion to do or not to do 16 something. Would you be willing to move?
17 MR. LEWIS: I would be happy to move that our l
18 distinguished subcommittee try to dig into the management 19 problems and rule hardware issues out. Get a feeling for the 20 facts on both sides of the management question, and then either 21 bring the people in to us or give us a recommendation on what 22 you think according to your judgment.
23 DR. KERR: Is there a second?
( 24 DR. SIESS: I second.
2S DR. KERR: There's now a second.
O Heritage Reporting Corporation (202) 628-4888
219 m
(-) 1 Mr. Moeller?
2 DR. MOELLER: Well, in "terms of discussion, although 3 I agree totally with how I'm concerned if the Commission is 4 meeting March 22nd or whatever it is, our input will never be 5 in time to assist them.
6 MR. LEWIS: I don't think that's quite right.
7 DR. KERR: If the motion is past, we have some 8 options. One would be to say we have looked at this on a 9 preliminary basis. We're concerned enough that we want to 10 investigate it. Let them know what we're doing.
11 There are others but I don't think we ought to decide 12 on the basis of that.
13 MR. LEWIS: I would guess that they would be 14 reluctant to come to a firm decision if we were to do that.
15 DR. KERR: Yes.
16 Mr. Steindler?
17 MR. STEINDLER: I was just go'.ng to remark that no
- 18 action is in fact an affirmative action. ,I agree with Hal we 19 need to do something.
20 DR. SIEES: The Commission's meeting at the end of 21 this month. Is the Staff going to make a recommendation at 22 that time to the Commission for a restart?
23 MR. HOLAHAN: We will make a recommendation.
24 DR. SIESS: Are you going to have information then to 25 justify a restart that you could not present to us today?
i
() Heritage Reporting Corporation (202) 628-4888
s v.
220
{} 1-2 MR. WARD:
been maybe our fault.
Well, they only had an hour and that's 3 DR. SIESS: Well, they spent an awful lot of that 4 time talking about why they shut them down, and I think it's 5 much more important for us to see on what basis the Staff is 6 deciding and recommending to the Commission that they be 7 allowed to start up.
8 MR. WARD: Yes, but an hour wasn't very much.
d9 DR. SIESS: But we spent all our time on the wrong 10 subject.
11 MR. WARD: Well, an hour wasn't very much and in 12 fairness to the Staff, I think they felt they probably had to 13 remind us of the history. I think that was necessary. We
(} 14 didn't give them enough time.
DR. KERR: Mr. Lewis?
15 16 MR. LEWIS: I think that if time were of the essence, 17 it would be appropriate to defeat this motion and then move 18 that we bring in the relevant parties to our next full 19 committee meeting. I think that would be the track if we're 20 concerned about the shortage of time.
21 DR. SIESS: Well, I'm not that concerned about the 22 Commission meeting. The Commission did not ask us for our 23 opinion. They're going to go ahead and make the decision 24 without asking us That is their privilege. But if they make 25 the decision and we think it was a wrong one even if it's
( Heritage Reporting Corporation (202) 628-4888
221 1 April, May or June, we can tell them.
(~)x.
2 It may not have any effect on Rancho seco but we can 3 tell them they did something wrong. That's our privilege as 4 advisors to them.
5 DR. KERR: It's our responsibility.
6 MR. LEWIS: And I think we ought to wait until the 7 Staff has something in writing that people have a chance to 8 look at. Have a subcommittee meeting at which the subcommittee 9 can review at some depth the basis for the restart, what the 10 fixes are, the criteria the Staff used to decide that it's okay 11 to restart, and have the representatives of the licensee there 12 at whatever level is appropriate.
13 And then on that basis, come into the full committee.
14 In the meantime, the whole thing may go away. Let's don't act
(^)T 15 hastily.
16 MR. LEWIS: The problem I have is that I think that 17 the Staff hasn't withheld anything from us today. There was 18 just a shortage of time and therefore they have given us in 19 capsule form to be sure the basis for their decision to 20 recommend restart. So I think that waiting for them to have a 21 more complete picture on the management issues which so concern 22 us is waiting for Godot. I think we have to look into that.
23 MR. WARD: Okay. But that really concerns me then, 24 because if that's all the Staff has to present to the 25 Commission is an expanded version of what we heard here today.
O's> lieritage Reporting Corporation (202) 628-4888
222
() 1 DR. KERR: Look, we will find out what they have to 2 present if the motion passes, because they will make a 3 presentation to the subcommittee.
4 MR. WARD: No, no. I'd hate to see us by default or 5 something let this go on to the point where the Commission 6 might actually make some sort of a decision on March 22nd. And 7 I don't see where they have enough information to make a 8 decision if all the information they have is an expansion of 9 what we've heard today.
10 MR. LEWIS: But I think if this motion passes, Dave, 11 then the Commission will be on notice that we think there are 12 issues which have not been resolved by the Staff and I would 13 expect them to behave in a responsible way. Maybe that's too O) y, 14 much to expect.
15 MR. WARD: No, no. I can accept that.
16 DR. SIESS: I don't see this sending that message 17 particularly. And the reason I said wait until the Staff has 18 finished their supplemental safety evaluation report is simply 19 that if I were on the subcommittee, I don't want to go 20 somewhere and spend eight hours listening to a staff 21 presentation without having a good picture in advance of what 22 they're going to talk about, having formulated my own 23 questions, so I know when to interrupt.
24 I just don't believe in going in and getting it all
, 25 orally with the aid of viewgraphs. We usually have a report in O Heritage Reporting Corporation (202) 628-4888 l
l l
223
() 1 advance. We have for licensing of plants. And I think this is 2 almost the equivalent of what we did at the licensing stage.
'3 And I don't think there's any real hurry.
4 If the Commission's going to do it, they're going to 5 do it.
6 MR. REMICK: At Davis-Besse, we had the same type of 7 thing. We had a draft and we had the licensees-in and we 8 discussed all aspects of it.
9 DR. KERR: Mr. Moeller?
10 DR. MOELLER: Well, on Davis-Besse, what was said is 11 true and apparently the decision was made by the Staff that 12 they would go ahead with this without necessarily having any 13 written response from us.
A more fundamental question is, does the Staff have a
(]) 14 15 policy for restarts of this nature where there's been a long 16 delay on procedural matters. And if they do, in which events 17 are we to be involved and which are we not, and what are the 18 criteria that are used.
19- MR. LEWIS: Well, I don't think the Staff does and we 20 may have to develop one.
21 DR. MOELLER: This is true. It may be our fault.
i 22 MR. LEWIS: Well, I don't think we should only be 23 responsive to whether the Staff wants our input. I think we 24 should make our input to the commission to whom we're advisory 25 -- not the Staff-- when we feel we have something to say.
O Heritage Reporting Corporation (202) 628-4888
224
() 1 DR. MOELLER: And that's why I'm saying, what are 2 their policies? How do they decide apparently that they didn't 3 need cur input on this one.
4 MR. LEWIS: You mean the C7mmission?
5 DR. MOELLER: No, the Staff.
6 MR. LEWIS: But the Staff is not the criterion for 7 whether we make an input.
8 DR. MOELLER: I know that but at least as far as 9 they're concerned, they're not the ones to dictate to us what 10 we do. But in this particular event, they --
11 DR. KERR: I'm going to rule that this is only 12 peripherally related to the motion.
13 Is there further discussion of the motion?
14 DR. SIESS: What's the motion?
[v) 15 DR. KERR: The motion is that we ask the subcommittee 16 that is responsible for this plan to investigate further the 17 issues that bear upon any recommendations we might make toward 18 restart. We think many of them have to do with management but 19 I don't think we should bridle the subcommittee to look just at 20 those if they in their investigation come upon other issues.
21 DR. SIESS: I don't remember it being that long, but 22 DR. KERR: I don't either but I thought I could 23 confuse you if I talked a little.
24 MR. LEWIS: He says more words but shorter ones.
25 DR. KERR: Is there further discussion of the motion?
' Heritage Reporting Corporation l (202) 628-4888
9 I
225
(} 1 Those in favor say, aye.
2 .
(Chorus of ayes.)
3 DR. KERR: Opposed?
4 (No response) 5 DR. KERR: Motion carries.
6 MR. STEINDLER: I have a question.
7 Is the existence of this motion going to be made 8 available to the Commissioners prior to or at the same time 9 they consider this issue.
10 DR. KERR: We will see the Commissioners prior.
11 DR. MOELLER: Could we some time address the more 12 fundamental issue as to a policy on these things.
13 DR. KERR: I think we should. ,
14' DR. SIESS: We should raise it with the Commission.
]}
15 DR. KERR: I believe our agenda calls for a break at 16 this point, if I'm not mistaken.
17 We do thank you and we apologize for the short amount 18 of time available.
19 We will reconvene at 20 of.
20 (Brief recess is taken.)
21 (Continued on following page.)
22 23 24 25
( Heritage Reporting Corporation (202) 628-4888
h 226 DR. KERR: The next item on the agenda is entitled
(]) 1 2 DOE advanced reactor severe accident program. Mr. Ward is 3 going to make some opening remarks. Mr. Ward?
4 MR. WARD: Yes, my remarks will be quite brief 5 because we ask for this presentation just so that the committee 6 can be informed about what's going on in -- what may be kind of 7 an important activity.
8 I think we've heard just a little bit about this, and 9 it seems to -- there seems to be work going on in areas that 10 are important to some of the reactor safety questions that 11 we've been considering over the last couple of years.
12 And we haven't gotten a very good picture of exactly 13 what that work is. So we've asked these gentlemen and ladies
() 14 that come in today to describe it to us. And we appreciate 15 that.
16 I think the committee will then need to decide 17 whether it wants, after this sort of summary briefing, whether 18 it wants to hear more about it -- perhaps even involve itself 19 in some way in the process, if that's an appropriate thing to 20 do, or if we think it's an appropriate thing to do.
21 And then also decide, if that is the case, which of 22 our several subcommittees should take it on. I guess one 23 indication of the fact we didn't know much about the program is 24 we couldn't even figure out which subcommittee was suitable for 25 the assignment.
O Heritage Reporting Corporation (202) 628-4888
227 N
1 So I think we'll want to consider that at the end of 2 the presentation.
3 MR. REMICK: Can I ask a question?
4 MR. WARD: Yes.
5 MR. REMICK: Is this advanced light water reactors, 6 or advanced reactors including light water reactors, or 7 advanced reactors not including advanced light water --
8 MR. WARD: Well, I'm sure we're going to hear that.
9 My understanding is that it's advanced light water reactors.
10 MR. REMICK: I see, okay.
11 MR. WARD: There are some problems with terminology.
12 In the NRC's current lexicon, advanced reactors are those for 13 which you can't use a standard review plan, I guess, which are
( 14 the LMRs and the MHGTR.
15 They're calling the other things the evolutionary, or-16 -- I forgot what they -- or the passive, I guess they would 17 call the small ones that are being worked on.
18 My understanding is this program refers to -- has a 19 relationship to those, not to the advanced LMRs and so forth. ;
20 But we'll hear that.
21 DR. KERR: Any further questions or comments?
22 We had allocated about an hour and a half to this.
23 And although we're behind schedule I think it is important 24 enough that we should continue to allocate at least an hour and 25 a half to the discussion.
O Heritage Reporting Corporation (202) 628-4888
. , , , _ - _ - - . . - . _ , _ , , . - - - . - = . , _ . . - _ _ _ - - - - - .
4 228 f3 (j 1 So I shall proceed on that basis.
2 MR. WARD: That's very good of you.
3 MR. LEWIS: Just to put on the record the fact that 4 I've just agreed to serve on DOE's safety advisory panel, and 5 therefore if you perceive any bias on my side in favor or 6 against DOE that's probably the source of it.
7 (Laughter) 8 DR. KERR: With that important comment before us, do 9 we turn things over now to the spokesmen for the group? ,
10 MR. WARD: Yes, I'd like to do that.
11 (Slides are shown.)
12 MR. GIESSING: My name is Dan Gleasing from the 13 Department of Energy. I only have one view graph in the way of
() 14 introduction, and then I'm going to turn it over to Paul Haas.
15 If you'll remember, I spoke to you last month in 16 regards to this second bullet on this view graph regarding the 17 programs that we have under way on the midsized plant programs.
18 I mentioned at that time that in addition we are 19 . supporting the certification of the large plant advanced PWR 20 and BWR by Combustion Engineering and General Electric.
21 For these two programs we have a support program l
22 dealing with severe accidents. It's called ARSAP, Advanced 23 Reactor Severe Accident Program.
24 And that's the program which is going to be described 25 to you today. One of the key activities la this program does I (
l Heritage Reporting Corporation l (202) 628-4888
229
() 1 deal with the support for the certification program, and 2 especially for the application by combustion engineering.
3 . And so before Paul Haas from IT Corporation comes to 4 explain the ARSAP program, George Davis from Combustion 5 Engineering who is in charge of the CE certification program is 6 going to say a few words also to give you some additional 7 context as they view this program from CE's point of view.
8 George?
9 MR. DAVIS: I just have some very brief comments.
10 DR. KERR: Excuse me. Could you identify yourself?
11 MR. DAVIS: My name is George Davis. I'm the Project ,
12 Manager for the design certification for that Combustion 13 Engineering.
() 14 Combustion Engineering is in the process right now of 15 upgrading our system 80 standard design to first of all reflect 16 the AOWR requirements document that's being prepared by the 17 Electric Power Research Institute, EPRI; and to address the 18 NRC's severe accident policy for AOWRs.
19 A significant important part of that effort is the ,
20 consideration of degraded core issue. And since these are 21 issues that go beyond the current NRC regulations and standard 22 review plans, reg guides, et cetera, we think it's very 23 important to get a clear understanding of what the acceptable 24 methods and acceptance criteria are for being able to close out 25 those issues.
P O Heritage Reporting Corporation (202) 628-4888
230
(} 1 _And therefore we sea the ARSAP program, which you'll 2- hear about this afternoon, as a very important vehicle for 3 being able to get those issues on the table, and agreed to 4 between the ARSAP organization and the NRC.
5 So we thank the ARSAP program board in that respect, 6 and also they are supporting us in the actual implementation of 7 some of those methodology and acceptance criteria, so you'll be 8 hearing a lot m- at ot it on the Ceasar docket over the next 9 few years.
10 With that I'll turn it over to Paul.
11 MR. 11AAS : Good afternoon. I'm Paul Haas from IT 12 Corporation. I appreciate the opportunity to talk to you about 13 the DOE, advance reactor severe accident program.
() 14 Some of the other people who are supporting this 15 program and the presentation -- you've met Dan Giessing; Tony 16 Buhl is the Technical Director; Mario Fontana is the Associate 17 Technical Director.
10 Bob Cope is the Program Manager from EG&G, Idaho.
19 This is a programmatic presentation as was indicated by Dr.
20 Ward.
21 In the future, as the technical information evolves 22 trom the program we will be more than happy, I'm sure, to meet 23 with you again and go into some of the technical results.
24 The purpose of what I propose to do today is to go 25 over the mission and the objectives; who's involved in the O Heritage Reporting Corporation (202) 628-4888
231
() 1 program; what are the interfaces of the program with the 2 various industry participants; something -- go through the work 3 break down structure, the various tasks, and some of the 4 accomplishments that we have gotten out in the early part of 5 the program during the formulation of these interfaces and 6 establishing the management plans.
7 Also, we'll spend a little additional time emphasis 8 in particular towards the end on the proposed process for 9 interacting with NRC staff on resolution of severe accident 10 issues.
11 We have developed a process and croposed it to the 12 staff. It involves presentation of pocket papers, focusing on 13 specific issues -- both severe accident phenomenolcgy, PRA
() 14 issues, and other issues dealing with the severe accident area.
15 Talk a little bit about the schedule of those and 16 some of the expected results out of that process. And of 17 course to obtain frcm feedback from you on the programmatic 18 aspects.
19 The mission of the program is to assist reactor 20 vendors and EPRI in the identification and early resolution or 21 risks significant severe accident issues so that they will not 22 be major obstacles to the certification of evolutionary light 23 water reactors during the 1990s.
24 DR. KERR. Mr. Ward's half-way promised that you were 25 going at some point to tell us what evolutionary advanced light O Heritage Reporting Corporation (202) 628-4888
232
(/ I water reactors are, and I take it that's one of the slides 2 somewhere down the road.
3 MR. HAAS: I don't have anything on the specifics of 4 the design. Generally --
5 DR. KERR: Just sort of a hand-waving description of 6 what an evolutionary light water reactor is.
7 MR. HAAS: It is, as opposed to revolutionary or 8 totally new design concepts, it's an extension of existing 9 concepts recognizing the needs for simplicity; reducing the 10 level of complexity; reducing -- addressing design, major 11 design features that have been identified in past experience to 12 be risk significant design factors.
! There are, across the board, evolutionary movements 13
() 14 in terms of not only safety but constructability, operability, 15 cost -- the term evolutionary is explained --
16 DR. KERR: Would it be fair to say that the reactor 17 concept is not very well formulated at this point?
18 MR. WARD: No -- wait, I think the question is 19 simpler. Are the reactor designs that are currently being 20 proposed for whatever we call it -- advanced design 21 certification by Westinghouse, RSAR 90, is that what they call 22 it? In combustion, in the advance BWR? Are those the designs 23 you're talking about?
24 MR. HAAS: No, the combustion, for example, is a 25 further extension of the System '80. It's referred to as the I
(
Heritage Reporting Corporation (202) 628-4888
233
() l' System '80 Plus.
2 So it is based on the standardized plant designs, but 3 extended, and modified and proved.
4 MR. RUBENSTEIN: May I help? Mr. Rubenstein of the 5 staff.
6 DR. KERR: I would like to hear from you but I would 7 also like to know what the people who are planning to design 8 this thing think, and then what you think, and maybe --
9 MR.-GIESSING: Mr. Chairman, one of the ways to get a 10 differentiation is these plants do not require a demonstration 11 plan. That's one of the major distinctions that we have 12 between the HGGR and the LMR concepts. -
13 These plants that --
1
() 14 DR. KERR: May I suggest that the world of what the 15 plants are not is probably much bigger than the world of what 16 they are.
17 And what I'd like to get is some idea of the world in 18 which they are. Or is that impossible at this point?
19 MR. SHEWMON: He was giving you one.
20 DR. KERR He was telling me what they were not.
21 MR. SHEWMON: Well, but if they're close enough to
- 22 current designs so they don't require a demonstration plant, l
23 that --
24 DR. KERR: Okay, you listen to him and when he gets 25 through you can explain it, too.
l Heritage Reporting Corporation (202) 628-4888 l
234
() 1 MR. WARD: Well, wait, it is the -- I don't know if 2 I'm using the right -- is the RSAR '90 plant --
3 DR. KERR: Just a minute. I interrupted this man 4 somewhat impolitely.
5 MR. WARD: Okay.
6 DR. KERR: And I want to give him a chance to finish.
7 MR. LEWIS: By interrupting him rather impolitely.
8 (Laughter) 9 DR. KERR: Please go ahead.
10 MR. GIESSING: Yes, the combustion engineering System 11 '80 Plus built their large plant that goes one step beyond what 12 the Paliburti design is.
13 The ABWR that the Japanese and GE have signed on to
() 14 instruct in Japan and the Westinghouse plant would fit the 15 definition of the evolutionary LWR.
16 MR. WARD: Does that help? Completely satisfy --
17 DR. KERR: If it goes one step beyond then I 18 understand what that means, yes.
- 19 MR. WARD
- I don't see what's puzzling you so much.
20 We've had presentations recently from GE and Westinghouse.
21 DR. KERR: That's it.
22 MR. WARD: This program is to service that.
23 DR. KERR: What I heard -- you're talking about the 24 advance water reactor GE.
25 MR. WARD: Yes, yes.
O Heritage Reporting Corporation (202) 628-4888 l
235
(' .
1 CR. KERR: And that's what we're hearing about.
2 MR. ERVIN: -Could I ask one further clarifyistg 3 question?
4 Is it true that this program will have to involve 5 consideration of core on the floor which got there either by 6 pressurized or unpressurized discharge, irrespective of the 7 probability of that occurring.
8 MR. HAAS: I don't have any idea how to answer what 9 this --
10 MR. ERVIN: Well, that's my question. I don't caro 11 how you argue that it won't happen. You in the end must 12 consider that as a possible condition.
13 MR. FONTANA: It's ..ot that it's irrespective --
() 14 DR. KERR: Excuse me. Would you identify yourself 15 for the record, please, Marlo?
16 MR. FONTANA: I'm Mario Fontana. I'm with IT 17 Corporation, Associate Technical Director of this program.
18 The core on the floor -- the assumption is not made 19 that the core on the floor will not amount it to be there 20 regardless of probabilities:
21 The safety assessments are based on probablistic 22 approaches of prevention and so on. However, and Paul will get 23 into this in a minute, we're also tied in with the reactor 24 requirements document which is identifying requirements for 25 these plants.
O Heritage Reporting Corporation (202) 628-4888
- _ _ _ _ _ _ _ _ _ _ _ - - _ _ _ _ _ _ _ _ - - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ - - _ _ - _ . _ _ _ _ _ _ _ - _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ - _ _ _ _ _ _ _ _ _ _ _ _ = _ _ _ - - _ _ _ _
236
() -1 And as part of this requirements document we will say 2 that the - -we will identify that the reactor cavity, for 3 example, shall have the capability of cooling debris, and shall 4 have the capability of water draining down the containment so 5 it can keep cool.
6 And it will have a depressurization system that's 7 there for other purposes. But it will be sized in such a way 8 so that it can be pressurized in time so you don't have 9 pressurized ejection, and also have a reactor cavity design 10 that will tend to limit the transport of debris from the cavity 11 up into the-upper containment volume where it could cause rapid 12 heating of a containment without depressurization.
13 So the question is -- to answer the question is no,
() 14 you don't assume that the core is going to be on the floor.
15 But you do those thingst and it's primarily a defense in depth 16 for another approach where you do those things so that if you 17 do get into that situation you have some means to act.
18 MR. ERVIN: And you follow the course of that 19 accident.
20 MR. FONTANA: We analyze the severe accidents from 21 beginning to end. Along with the probability we'll be 22 analyzing the response of the plants of many severe accidents.
23 But the plant will be -- there'll be design features 24 in the plant to tolerate some of the more potentially important 25 --
O Heritage Reporting Corporation (202) 628-4888
237
[ss') 1 MR. ERVIN: It's just like the large LOCA program.
2 Irrespective of the probability you're going to provide -- make 3 good reasons for it.
4 MR. FONTANA: No, I hope it's not like a large LOCA 5 program.
6 MR. ERVIN: Well, you're going to make provisions to 7 cope with it.
8 MR. FONTANA: But they will be provisions that are 9 very easy to make because you're going to have to decide a 10 reactor cavity anyway.
11 You're going to have to have a depressurization 12 system anyway, so you make adaptations ao that it can help you 13 tolerate these events should they occur.
() 14 MR. ERVIN: And that depressurization system will 15 perform at extreme conditions of operation?
16 MR. FONTANA: Pardon?
17 MR. ERVIN: Extreme conditions of operation. It 18 would be capable of performing in extremely high pressures and 19 temperatures well beyond --
20 MR. FONTANA: Yes.
21 MR. ERVIN: Okay. Thank you.
22 MR. WARD: Paul, one more clarification. I almost 23 hate to bring it up. But do you include -- is this program 24 also to assist with the design programs related to what EPRI is 25 calling the 600 mega watts sized plants, the so-called passive, O Heritage Reporting Corporation (202) 628-4888
238 em
!. ) 1 or whatever they call them these days.
2 MR. HAAS: That's in the program scope, brt there's 3 no active element in that now.
4 MR. WARD: Very good, thank you.
5 MR. HAAS: The objectives are to support the severe 6 accident assessment analyses and issue resolution that are 7 being performed by the vendors who are a part of the DOE 8 certification program, and to provide input to, support for, 9 and some of the technical assessments for the EPRI AOWR 10 requirements document.
11 I think you all have heard presentation previously on 12 the EPRI document. Also, as a secondary objective, if while we 13 are addressing the specific vendor certifjaations and the EPRI
() 14 requirements document, our severe accident issue resolution 15 process addresses issues more generically, such as basic severe 16 accident phenomenology then we should make every attempt to 17 make our solutions as generic as possible.
18 Program participants, as you know, is DOE's 19 supporting agency. The program director is EG&G, Idaho; IT 20 Corporation's Technical Director; we have Bowski & Associates 21 as a subcontractor for the severe accident methods development.
22 Combustion is, of course, the vendor -- and we'll 23 talk about our interaction with combustion to deal with the NRC 24 staff.
l 25 We have an independent industry technical advisory l
Heritage Reporting Corporation (202) 628-4888
239 I) 1 group that is particularly active in reviewing our issue 2 papers, and getting some sense of an industry consensus on our 3 positions; and we do have consultants in specialized areas, 4 regulatory areas, PRA areas, and some of the seismic experts.
5 This color slide also gives the same program 6 organization.- The emphasis here is that there are generally 7 three areas of the program. One is the severe accident ,
8 analysis methods, and that is primarily lead by FAI on the 9 methodology -- the severe accident methodology on the PRA.
10 IT is the lead there. The severe accident resolution 11 process -- we and the subcontractors are directing. And direct 12 support to the vendors, GE and CE, and to EPRI organized and' 13 run by us also, and the consultants.
( 14 The kinds of ways that information will come out of 15 the program to NRC, and the interactions are illustrated on the 16 slide, are involved with the issues of methods of improvement 17 of severe accident analysis methods, carrying out some analyses 18 and support analyses for the vendor, and some PRA support 19 feeding directly into combustion, and in particular the severe *
~
20 accident issue topic papers going to NRC.
21 And then we support the EPRI requirements document.
22 We have direct tasks with general electric, and I'll go through 23 those -- what's involved there.
24 We are, with regard to the Westinghouse program, 25 really just monitoring each other's activities. There's no
(
Heritage Reporting Corporation 7 (202) 628-4888
?
240
() 1 direct support for them.
2 And each of those is tied in to the EPRI requirements 3 document, and of course the certification actions will be going 4 directly to the NRC.
5 MR. REMICK: Could you explain why the CE System '80 6 is not relying more strongly, I guess are the words -- I'm not 7 sure the best -- on the requirements document at EPRI.
8 It seems to be going ahead where you indicate that 9 the ABWR and the APWR are going to first have the requirements 10 document.
11 MR. HAAS: That's not intended to be indicated here 12 at all, no. This line is as strong as this line.
13 I don't sense, and I think it would be incorrect to
() 14 characterize CE as not paying strong attention to the EPRI 15 department.
16 MR. REMICK: That's a funny way of making the diagram 17 then, I guess.
18 MR. HAAS: I'll go through the work break down 19 structure, then I'll give a brief run down shortly of some of 20 the things that have happened so far.
I 21 WBS work break down structure 1 is management and
- 22 organization. The issue resolution process WBS-2 is an 23 identification and a categorization in prior organization of 24 severe accident issues that may be applicable to advanced light l 25 water reactors.
O Heritage Reporting Corporation (202) 628-4888
241
() 1 There was a fairly extensive process that went on 2 before I got on the program actually to list a large number of '
3 areas that have evolved in severe accident space over the many 4 preceding years.
5 Assess those to determine some sort of priority to 6 those categorized in groups together. We have all six sets of 7 issues that are generally related, and within each set we 8 propose to bring two to six specific topic papers related to 9 that general category of issues.
10 We will propose this resolution approach in the topic 11 paper, and present that to NRC; support combustion engineering 12 interaction with NRC and the ACRS to gain concurrence on the 13 approach, and then a follow up with support for the
( )' 14 implementation of that technical resolution and the closing of 15 the loop with NRC to concur that indeed we laid out the 16 approach; we said what we were going to do. The approach 17 appears go be acceptable.
18 Then we followed up and conducted the analysis, if
\
19 there's additional analysis that's necessary that the issue is 20 not immediately resolved by design, or whatever follows on it i
1 21 in order to close a discussion with the NRC.
- 22 Also associated with the issue resolution process is 23 a -- labelled WBS-7 -- it's a notebook that is being compiled; 24 and the first edition will be out -- it's in final draft now 25 ready to go for approval to the program manager.
O Heritage Reporting Corporation (202) 628-4888
L 242
() 'l And this notebook has documented the experience base 2 for severe accidents that had existed based on IDCOR, NUREG 3 1150, foreign.research, many sources, and then assembled in a 4 succinct form to identify specific lessons learned about design 5 about severe accident analysis and phenomenology.-
6 Initially this was supposed to be a considerably 7 larger effort, and it was supposed to maintain this breath 8 coverage, and try to be really a national reference source that 9 covered and condensed the results in progress in severe 10 accident nianagement around the world.
11 Funding restraints -- after this first lesson, we --
12 this first_ publication -- we will now just essentially document 13 the results coming out of the RSAP program and this interactive
() 14 process with the NRC staff on issue resolution.
15 DR. KERR: But I have to ask what small W - dot -
i.
16 small R --
17 MR. HAAS: With respect to.
i 18 DR. KERR: Oh, thank you.
19 MR. MICHELSON: What is the schedule for that 20 document?
21 MR. HAAS: It should be out something like a month.
22 MR. MICHELSON: A month.
23 MR. HAAS: It's right now in final editing for the 24 first version. And then there's an annual update that will l
j 25 come out.
O Heritage Reporting Corporation (202) 628-4888
243
() 1 WBS-3 is a methodology development. Really it's 2 basically improvements. Some of them rather modest. But first-3 before the selection or development of the computer modelling, 4 there was'again before I came on the' program a rather extensive 5 assessment of the existing tools, both developed by the NRC 6 contractors -- NRC and the industry -- and a selection process 7 to identify the integrated codes that were most appropriate for 8 ALWR severe accident analysis.
9 It will' involve some modification and improvement of 10 those models that we're now -- through that process there had 11 been a selection made of specific codes -- MAAP and MELCOR.-
12 As integrated tools, we have now focused on MAAP and 13 some improvements there. And there will be an independent V&V, 14 AND internal bench marking by the RSAP program; and there will 15 be a formal documentation in user support.
16 MR. WARD: To verify and validate MAAP seems to be an 17 ambitious undertaking. Can you give us some idea what sort of 18 approach you'll take?
- 19 MR. HAAS: Yes, it would be a huge undertaking for i
20 us, for this program, if it were not for the case that the 21 industry, of course, has invested and is continuing to invest a 22 good deal of effort in that.
23 In particular now, EPRI has a formal independent V&V 24 program going on for MAAP 3B. And we will pick up from MAAP 3B i 25 with our modest additions; and we will do bench-marking of any
)
i Heritage Reporting Corporation l (202) 628-4888 i
i
244
(} 1- n odels that we modify or develop.
2 And then we will go through an independent V&V 3 process. And we're working that out with EPRI now to identify 4 what we will have to do to be consistent with the EPRI process.
5 DR. KERR: Will MAAP continue to be proprietary?
6 MR. HAAS: As far as I know at this time the actual 7 coating, yes, it will be.
8 MR. WARD: Even MAAP-DOE will be?
9 MR. HAAS: That's my understanding, yes. Part of the ,
10 conditions which we had to agree to in order to obtain and use 11 the code was non-transfer.
12 Now on the other hand, a great deal of information --
13 a great deal of information about the modeling, and
-( ) 14 specifically about the models that we do, the bench-marking, 15 will be presented to the NRC in the way of these issue papers.
16 It will be a natural unfolding of the discussion of 17 the issues. By that I mean, for example, some of our proposed 18 approaches to resolution of some of the severe accident issues 19 which, in fact, many of those are related to discussions, 20 obviously, that preceded in the NRC IDCOR interactions and 21 subsequent interactions.
22 Many of our resolution pathways would involve an 23 improvement to the map model. I said many, some -- I should 24 say some.
25 In general, the approach to resolution is improved O Heritage Reporting Corporation (202) 628-4888 1
l I
245
^
/ ^%
(_) 1 design. But in some cases we'll say yes, there was part of the 2 issue, or the issue had evolved between NRC and IDCOR to the 3 point where there was discussion and disagreement on the nature 4 of the models.
5 And we had modified those models. And so in those --
6 in the essence of those models and the bench marking that we 7 do, the experimental comparisons, or comparisons to more 8 detailed codes will be in technical reports and available to 9 everyone -- to the public.
10 MR. WARD: So much of the process of verify and 11 validating will really be by reference to more detailed codes, 12 is that the idea?
13 MR. HAAS: No, some of it will be, p
\' 14 MR. WARD: Some of it will be.
15 MR. HAAS: The majority of the bench-marking will be 16 to available experiments, and to some reactor transients. Some 17 comparison to TMI-2 results will be there. And some will be by 18 comparison to more detailed codes.
19 Similarly, methodology development -- and this is 20 really stretching it to call this development in some cases --
21 but it's providing additional guidance and some tools in 22 technical support to EPRI and vendors in the area of PRA 23 methods.
24 There's a, again, a product that's one of the first 25 products out of this, is -- actually, it's kind of a neat v I
. Heritage Reporting Corporation l (202) 628-4888 i
246 l) 1 assembly of past information also out of PRAs, qualitative 2 insights that have been learned by studying existing PRAs.
3 And we've put it on a computerized system with a D-4 base; and 'one can go in and search for these statements of --
5 again, kind of lessons learned from past experience by way of 6 system, by way of various key words.
7 And this will be an interesting compilation for the 8 designer to begin to use PRA results. We are supporting EPRI's 9 development of what we are calling functional PRAs.
10 And by that we mean we're providing a tool by which 11 EPRI can take their requirements document for a design that 12 does not exist yet in detail, but for which they are specifying 13 design requirements, and functionally analyze it and develop a
() 14 PRA model which wou.d allow them to do some assessment of if 15 this plant were built according to this requirement, making 16 some assumptions about the kinds of equipment based on the 17 preliminary design of information, and what we know about 18 current reactors, what kind of results do we get from the PRA?
19 And that can be fed back into the design process 20 also. So it can begin -- it's moving towards beginning to do 21 things like reliability allocation, and will be a very useful l 22 application of PRA in the design process as opposed to sort of 23 a back unit in crunching the numbers.
24 We are also supporting EPRI's development of their l 25 PRA ground rules and assumptions document. And in the future O Heritage Reporting Corporation (202) 628-4888 I
1
4 247
() 1 -
this is not a current task -- we intend to try to direct some 2 applications of how one could use PRA results in the severe 3 accident management area.
4 Our direct support to vendors and for EPRI. We have 5 put into and will review all of the chapters of the EPR1 6 requirements document with regards to the severe accident 7 areas.
8 We have developed some of their specific requirements 9 and will provide technical support and analysis as part of 10 their basis which they will again present as we indicated to 11 the NRC through their requirements document.
12 And then we will provide some technical support to 13 them as they interact with the NRC. One of the major tasks is
() 14 direct support for the combustion engineering certification 15 effort.
16 And there we will take a more active role in the 17 sense, and probably with any of the other participants, in 18 helping to identify and review the various severe accident 19 issues that we see pertinent particularly to their system as 20 their design evolves; bind some sequences.
21 We'll actually perform some of the severe accident
! 22 analyses and assist them with it -- with theirs. Assist them 23 in the PRA. We'll be doing some of the external events
! 24 analysis; some sensitivity studies for support systems, and 25 then support them again as they go through the interaction on j
, Heritage Reporting Corporation l (202) 628-4888 l
248
() I their certification in the PRA area.
2 At GE we have direct support for them, but it's less 3 of an integrated comprehensive kind of a support. It's more on 4 specific task as they see they need help throughout the 5 program.
6 Right now we're providing some assistance with 7 modification of the BWR version of the MAAP code; some 8 technical support for the determination of fishing product 9 particle size that has to do with suppression pool scrubbing.
10 We are assisting them in the seismic area also, 11 recommending an overall approach and developing some data for 12 their fragility analysis.
13 And we are providing EG&G, using a track code to
() 14 support their atlas analysis. As I said, with Westinghouse we 15 basically just monitor their program and try to keep tabs of 16 what each other's doing.
17 Some of the progress -- as I said, this is coming out 18 of the earlier part of the program which was no small task for 19 the people, again, who were there before I was to even 20 establish these various interfaces with the various 21 participants in the industry and work out the programmatic 22 structure -- and these will all be available.
23 We have the program plan, and the implementation 24 plan. They're annually updated. We've got these interfaces 25 established.
Heritage Reporting Corporation (202) 628-4888 m
249
() 'l We have -- and I'll talk about this as I indicated a 2 little bit more some -- in a process of interactions already 3 started with the NRC staff throughout he CE certification EPRI, 4 and we do have this independent industry technical advisory 5 grcup.
1 6 On the --
l 7 DR. KERR: What is that independent of?
8 MR. HAAS: I should have said industry technical 9 advisory group because in fact it merges two ideas, again that 10 were in an earlier version of a program that was a broader and
> 11 larger program.
12 There was going -- there was an attempt to get a 13 totally independent body of people to review technically the
! () 14 complete program, and virtually every product that came out of 15 it.
16 We had modified that through the program constraints, 17 and what this industry technical advisory group is is i 18 representatives from E'#RI combustion, BMW and CE.
19 And then at the meetings where we -- and we're 20 focusing -- their primary activity is focusing on the severe 21 accident issue topic papers that kind of consolidate the 22 information.
. 23 And they of course bring in whoever they want to f
24 mnsult wir.h them on the special areas. So it's not truly
! 2$ indepa.' dent in the sense that they're somehow involved in the i
i l s@#V Heritage Reporting Corporation
[
l (202) 628-4888 4
_ _ _ _ m -
250
() 1 ALWR program.
2 But they're independent of RSAP and DOE.
3 Some of the products -- I may have skipped a slide 4 here. Probably already said everything.on this one anyway.
5 I've indicated we do have the first version of the lessons 6 learned document.
7 Oh, there was some participation by the program in 8 the Chernobyl activities and a report that was -- an internal 9 report that was generated on that.
10 We've established this resolution process. We do 11 have -- and I'll give you what topics are in there later on, I
12 and they're listed by paper in the appendix that I've given 4
13 you.
14 We have the first set of topic papers. This set 15 deals, essentially with the IDCOR issues -- the last of the 16 1819 IDCOR issues sets 1 and 2 deal with those.
17 And set 1 is the ones that NRC and IDCOR agreed were 18 resolved. And resolved means resolved sufficiently to proceed 19 with the design of reactors.
20 The set 2 focuses on the ones that NRC and IDCOR did 21 not come to a final conclusion that they were resolved. And
(
i 22 set 3 is focusing on PRA methods.
23 Set 2 is finished as far as we're concerned. It's in f 24 final review be CE, and we have the comments now; and we'll be
! 25 making some revisions to those.
r (
Heritage Reporting Corporation (202) 628-4888 i
e 1
i
251
() 1 And they will probably be submitted to the staff 2 within several weeks. This one is into the staff and it's 3 under review now.
4 In the methods development area we have chosen MAAP 5 as the primary analytical integrated code. There has been a --
6 this is an internal code configuration quality control 7 procedure for any modifications we do.
8 Then there are some modifications in bench-marking.
9 We have plans for both of those. The final editing of those 10 plans being done now, but the bench-marking and modeling 11 activity has already started.
12 We are coordinating with the EPRI map users group, as 13 I indicated. We're an observer on that group. And we are --
() 14 we have had several technical exchange meetings with the NRC 15 contractor staff at Sandia and at Brookhaven regarding their 16 codes and methods, and severe accident phenomenology, and some 17 of the things like uncertainty analysis that Brookhaven's 18 doing.
19 In the PRA area I've already indicated the first 20 product for this computerized data base of insights --
21 qualitative insights as already done and will be published 22 soon.
23 We've had considerable input to the PRA ground rules 24 assumptions document developed by EPRI, and will continue to 25 have input; and we have done those high level PWR and BWR PRA O Heritage Reporting Corporation (202) 628-4888
4 252 cm.
(j 1 models. And they also are draft reports that are in final 2 editing now.
3 MR. WARD: Let's see, you say the first version of 4 this lessons learned thing is available in what form?
5 MR. HAAS: That should be through the release process 6 at EG&G within weeks. We're right now putting the final 7 editing touches to it. Everybody's reviewed it and we just 8 have to put it into print.
9 And I don't know exactly how long the formal release 10 will be, but certainly once it's clear at EG&G's process, you 11 know, it could be sent out in draft.
12 MR. WARD: Now you said it was a data base, but is 13 there just going to be a print copy, or is there --
/~T
(/ 14 MR. HAAS: Oh, I'm sorry. No, I'm sorry. There's 15 two items that I've been talking about. I wasn't explicit 16 enough.
17 On this one -- I was confusing that with a lessons 18 learned. I think that was your term. But there's two items.
19 One is the lessons learned notebook. And that's a hard copy 20 notebook.
21 And that covers the area of severe accidents 22 completely. It's phenomenology, PRA, experienced plant design 23 -- those kinds of things.
24 This one is qualitative insights that also has a 25 broad coverage. But it simply statements garnered from PRA n
Heritage Reporting Corporation (202) 628-4888
253
() 1 analysis -- statements about kinds of features that have caused 2 problems in specific systems.
3 They're simple guidance to designers. Pay attention 4 to this because in the past experience when, under these 5 conditions, this has caused -- or this is a risk significant 6 item, either because of -- say of the desi.gn, or operator 7 actions that have occurred, and things that are developed 8 during the detailed analysis of the PRA.
9 These are put together by people who have done PRAs, 10 and who have analyzed a fairly good-sized -- half a dozen, 11 eight, of the typical recent PRAs, about right? Primary basis, 12 plus expertise.
13 It's beginning to embody the expertise of PRA people
( 14 as they go through the design, and what they learn. Not, 15 again, emphasizing so much the numbers that come out, but 16 emphasizing what you learn about the system as you analyze it, 17 and trying to provide that information back to the designer on 18 the front end.
19 We have had input to the Chapter 1 and Chapter 5.
20 Now on the Chapter 1, there is -- we did not have a great deal 21 of inpug on the first version.
22 But to some considerable extent, because of our later 23 input there will be a Chapter 1 roll up, and there will be 24 considerable modifications related to the severe accident area.
25 We have had input on Chapter 5, especially. We've Heritage Reporting Corporation (202) 628-4888
254 We're now looking at Chapter
(]) 1 reviewed these chapters already.
2 6 which is the containment arrangements.
3 Chapter 5 is severe accident areas, or the ones most 4 pertinent to it.
5 And we will develop specific technical papers to 6 support some of the requirements that EPRI haa that are related 7 to severe accidents.
8 One is, you know hydrogen control paper is now -- the 9 whole business of handling the hydrogen generation and the 10 control in the containment is one of the technical papers.
11 I'll just try to skim through these. This is, as I 12 said, is problematic. And a lot of the things that -- saying 13 we've accomplished -- the products are storting to evolve now
() 14 for direct support to the vendors.
15 These activities got started considerably later, 16 towards the end of last fiscal year. So some of the first 17 support for CE, and looking at the severe accident issues; and 18 beginning to look at the sequences that they should analyze; 19 beginning to develop parameter decks.
20 We've developed one in-house that is running along 21 tracking as more and more design information, or design 22 information is available in more detail -- bringing the map 23 code up and beginning to do some preliminary sensitivity runs.
24 The GE support, I indicated that some of those 25 activities that we had said are in progress. Particularly, G
Heritage Reporting Corporation (202) 628-4888 l -- ..
255
() 1 there's some progress on the ATWS; and some initial support on 2 the seismic analysis.
3 But there's no products out of those -- as far as 4 formal publications -- they'll be out of those areas in the 5 next -- certainly not for a few months.
6 By the end of '88 we intend to have all six sets of 7 these topic papers tranumitted to NRC. That may be --
8 actually, our target is to CE, and CE delivers them to NRC very 9 shortly thereafter.
10 The annual update will be done. The modification and 11 the bench-marking that we're going to do on the PWR version --
12 we're emphasizing the PWR again focusing on the CE.
13 The CE plan will be completed. That's not the
() 14 e;ternal independent B&B. But all that we're going to do in-15 house.
16 And we will have some upgrades on these functional 17 PRA models. We're getting a lot of review; a lot of interest 18 from the vendors on those, and continuing to improve those.
19 We will generate a series of technical reports that 20 correspond to the five tasks that we outlined for the System 21 '80.
22 Particularly in the PRA area we started some analysis 23 there already. And we will have the reports out on those 4 24 areas in general for the GE plants.
25 Now I'd like to get a little bit more -- this is, O Heritage Reporting Corporation (202) 628-4888
I 256 again, a still very problematic -- not a great deal of
({} 1 2 technical depth, but a little bit more focused on this 3 interaction process with the NRC.
4 What we have done so far is evolve this process, and 5 it started with the office chiefs and worked through the branch 6 chiefs at the NRC staff proposing a process by which we would 7 bring to the NRC through the combustion engineering the 8 specific sets of issue papers.
9 I've already indicated where we are in that. We 10 followed up with the technical information exchange meeting for 11 several days, going through each of the major taodels and the 12 phenomenology addressed in MELCOR, at.1 then with Brookhaven on 13 things like the core concrete interaction work they're doing
() 14 and some of the uncertainty analysis.
15 And in fact we met yesterday just informally with the 16 NRC staff to identify and discuss some of the key decisions 17 that we see -- just our input on just some of the decisions 18 that we see them needing to make pertinent to the ALWR severe 19 accidents in the near future.
20 This process that we've outlined -- and this is a 21 high level diagram. There's a very detailed flow chart in the 22 appendix that shows all the loops in a lot of it.
23 If it's not approved it goes back here and we rewrite 24 the paper. And it shows how the CE design process and their 25 addressing of the severe accident areas merges with our view of O Heritage Reporting Corporation (202) 628-4888
257
() I the severe accident space into the design.
2 But this is the general flow. And as I said we 3 collectively in the program, and with selected programs, and 4 with an attempt to get some sense of industry consensus, define 5 clearly the issue and propose an approach to technical 6 resolution.
7 We document that in a technical paper. Typically
- 6 these are 10 to 20 page papers on each topic. Review this with I
9 CE, and with the ITAG, again gaining some sense of industry 10 consensus and modified as necessary; and then present it to the 11 NRC through the CE certification process, and try to achieve, 12 through some technical discussion, an agreement, a concurrence, 13 that this issue, ideally, is resolved.
) 14 It's over with. Nobody has to do anything else.
i 15 DR KERR: Is the basis for your approach one which I
16 assumes that the NRC staff has resolved these issues for i
- 17 existing reactors and this is simply an extrapolation, or that i
18 they have not yet resolved it for existing reactors but you're 19 going to go ahead with it on your designs anyway.
i 20 MR. HAAS: It can be either. It's not an assumption.
l 21 It's not an assumption about existing reactors that's directly 22 related to our proposed resolution.
1 23 Obviously, we build on that base. And our first two 24 sets of issues.
25 DR. KERR: On what base are you building? Are you i
(
Heritage Repor;ing Corporation l
(202) 628-4888
258
/')
(_/ 1 building on a basis that the issues have been resolved for 2' existing reactors, or that they have not been?
3 MR. IIAAS: Some have been resolved 3rd some haven't.
4 In particular, the first set are the ones t' c had been 5 declared to be, and for which there are letters written saying 6 they are resolved according to the IDCOR NRC process. And we 7 dealt with those.
8 And what we did there was use those, and look at 9 those and say is there anything different about ALWRs that -- ,
10 DR. KERR If I remember, one way of resolving an 11 issue in that arena was to say we disagree.
12 MR. HAAS: That's correct, I think. Yes, we just 13 can't come to an agreement.
14 DR. KERR: And that'll be a considered resolution as 15 far as you're concerned, and a designer can take tnat and 16 design a reactor with that information.
17 MR. HAAS: No, what we'll say is, this is an issue 18 which, for existing plants, the history says NRC did not agree.
19 Now what was the --
20 DR. F.E 9 F - Okay. I thought you -- what I was asking 21 was whether your resolution -- a possible resolution in your 22 mind was to say we disagree on this -- we and the staff.
23 MR. HAAS: No, no.
24 DR. KERR: Okay.
25 MR. HAAS: We're trying to bring closure to issues.
pJ Heritage Reporting Corporation (202) 628-4888
t 259
() 1 DR. KERR: Okay.
2 MR. HAAS: And to the extent that we don't to some 3 extent bring closure -- and let me be clear now -- at this 4 point, what I mean by interim guidance.here is closure in the 5 sense that there is an agreed upon approach which -- and here 6 comes the caveats -- all things remaining constant, which may 7 or may not. And obviously, the answers aren't in.
8 But a general statement, after interaction with the 9 staff, that this seems like a reasonable approach to resolve 10 this issue ultimately; and if it's carried out, and if we like 11 the way you carried it out, and if we analyze your design and 12 indoad you did resolve it, then things are okay.
13 But the attempt here is to gain, first of all expose
() 14 these issues -- identify them and have somebody at least agree 15 that we've got the issue identified properly. You know, 16 identify the problem before we try to solve it.
17 And then agree or discuss the basis, and look at the 18 design, and say, again, what is different about this design.
19 To the maximum extent possible, the issues will be resolved by 20 the design modifications on the ALWRs.
21 And the thing that they have going for them, of 22 course, is that they're right now on paper instead of in the 23 country; and things can be changed.
24 So that's the first line, or the first approach to 25 resolution.
O Heritage Reporting Corporation (202) 628-4888 i
260
() 1 In some cases the resolution will evolve, as I 2 indicated earlier, the real issue is, in the severe accident 3 phenomenology is how one models some of the phenomena and what 4 is the technical basis for those models, and the experiments 5 and results.
6 And in that case we may be making some, and will be 7 making some modifications to the map code, and doing some bench 8 marking and saying we now think that issue's resolved.
9 So we will get some of these, we hope. We'will go 10 through; and we'll say yes, that's the approach. In fact the 11 approach is, this issue's taken care of by the design.
12 It doesn't exist. It may or may not be an issue for 13 existing reactors, but it's not an issue for 3 LWR because we've 14 changed the design completely -- we've changed the design that-15 effects that issue.
16 And it falls our. And NRC staff says great. That 17 one's resolved. Chalk it off. Everybody agrees. Some number 18 of them will involve and say some additional analysis.
19 NRC says that's nice. I' looks like a good approach.
20 We believe you'll do some analysis and show us that it's true; 21 and then we'1] talk.
22 And so we may go back and do some analysis. And in 23 some cases the preliminary design is not too far along. It may 24 be that some design modifications can be made.
25 So absolutely, the idea ls to bring closure and the
(:) Heritage Reporting Corporation (202) 628-4888 l
- t. .
261
() 1 idea is to bring stability to the certification process.
2 Generally, the issue papers as I indicated now will include an 3 issue definition.
4 Sonie the historical perspective would have been the 5 industry action, previous positions and NRC actions, and then 6 an approach.
7 And this again will be as generic as possible, but in 8 fact focusing on -- since we're going through the CE docket we 9 are, at least at this stage in the program, on this issue 16 resolution process, restricting ourselves to issues that are 11 either really, truly are generic between PWRs and BWRs, or PWRs 12 specific.
13 The six sets of papers the general areas, and the
() 14 dates at which we on our schedule that we have them to 15 Combustion are, as I first said, there were -- actually there 16 were 11 issues that were resolved.
17 One of those was BWR specific. So we focused on 10 l
18 of the PWR issues. And I said analyze those, and said as I l
19 indicated, what's different about ALWRs; which ones of these 20 remain resolved? And in fact our resolve for ALWR is exactly 21 the same way that they were agreed to be resolved.
22 Which ones do we suggest, in fact, have some 23 modification to the issue and to the resolution approach. And 24 we contend they are resolved for ALWRs also.
25 DR. KERR: I got the impression from a comment that O Heritage Reporting Corporation (202) 628-4888
262 Dr. Fontana made earlier that a severely degraded core, or
{ 1 2 severely degraded or not, it might go into containment, would 3 not be treated, at least to your preference, as a design basis 4 accident.
5 MR. HAAS: That 's not - fou want me to answer that?
6 DR. KERR: Unless I'm wrong -- well, go ahead.
7 MR. HAAS: That's not the first time Mario's got me 8 in trouble, by the way.
9 MR. FONTANA: And it won't be the last either.
10 Mario Fontana, as I said before. The approach is to 11 take the design basis access, pretty much as you have now, and 12 design the safety systems and the containment systems to those 13 accidents, complete with codes and standards and analysis and
() 14 _everything else.
15 Now you've got a design. Then you go back and you 16 look at the severe accidents which are beyond the design basis 17 in identifying -- there's a margin between design values and 18 ultimate failure values.
19 For example, say containment ultimate failure. So 20 once you have a design designed to the design basis you go back 21 and analyze how that design would respond to the spectrum of 22 severe accidents that were analyzed.
23 And some will fall within a design basis; some will i
24 fall in a margin. And those that fall beyond a margin have to 25 be treated on how long it would take to approach it, and so on.
O Heritage Reporting Corporation (202) 628-4888
263
( 1 So severe accidents are not design bases, but they 2 are -- the designs are analyzed to the extent to which the 3 response to severe accidents.
4 DR. KERR: That would then conclude containment 5 design.
6 MR. FONTANA: Yes.
7 DR. KERR: You are talking about -- so you would 8 design containment to the conventional DBAs; and you would then 9 see how the containment responds to whatever postulated severe 10 accidents you treat.
11 MR. FONTANA: Yes.
12 DR. KERR: Now, so you have to decide on which 13 sequences you consider. So do those sequences sort of become 14 quasi-design basis accidents -- not in the sense that you go 15 through all the class 1 and so on thdt you do for design basis 16 access.
17 In the sense that you look at margins against that 18 set of sequences.
19 MR. FONTANA: The sequences that we analyze are 20 chosen from a set that is not highly unlikely. In other words, 21 the ones that are potentially risk dominant.
22 DR. KERR: Yes.
23 MR. FONTANA: You chose that set, and you chose some 24 that deliberately test your containments of -- and you may end 25 up with maybe eight or ten of them. And you analyze in depth O Heritage Reporting Corporation (202) 628-4888
264 q 1 the sequences from beginning to end.
).
2 DR. KERR: It occurs -- would occur to both of us, I 3 think, that if you design containments to the conventional 4 design basis accideist at least there is a strong body of 5 opinion at this point.
6 But some of those containments would be much better 7 suited to deal with severe accidents than others. I don't know 8 if that's general or demonstrated.
9 Are you going to take that into account in your 10 treatment of containments?
11 MR. DAVIS: Let me take a stab at that. What we'll 12 be doing is with the probablistic risk assessment that will be 13 done for each of the specific designs -- in*our case, the
('T g,j 14 combustion in the System '80 Plus design.
15 That PRA will identify the probabilities of certain 16 sequences that lead to core melt. And so all of the sequences 17 that are, you know, above some cut-off frequency will then be 18 deterministically analyzed using the map code to see what the l
19 consequences are.
20 And based on those analyses which are pretty much 21 best estimate type analyses we'll be able to look at the 22 designs and see whether there are cost effective improvements 23 that could be made, such as briefly figuring the reactor vessel 24 cavity.
25 We're not talking about major design changes that l
O Heritage Reporting Corporation I
(202) 628-4888
265
() 1 would be held to the same design criterion of Class 1E, -- bla, 2 bla', bla -- that would be done for design basis. !
3 But where you can make cost effect changes. And that 4 would include consideration of containment type. .In our case 5 right now we're going through an evaluation of containment type 6 for System '80 Plus.
7 And that's a part of the consideration.
8 MR. EBERSOLE: Well, isn't it implicit in all of this 9 that there will probably be improved methods of depressurized 10 reactors, especially combustion engineering which has none at 11 all now?
12 (Laughter) 13 MR. EBERSOLE: In the System '80 Plus design we will 14 have a safety pressurization system which will be available not 15 only for design basis accidents but beyond design basis.
16 MR. EBERSOLE: Well, let me reflect you p3ssibly back 17 in such instances as Palare -- in view of the fact that they 18 will need it even worse not having a previous containment.
19 You know that's still an issue, a standing issue as 20 to whether Palare [ph) should have had the RVs.
21 MR. DAVIS: As I recall that issue is awaiting 22 resolution of A-45.
23 MR. EBERSOLE: But now it's a new basis.
24 DR. KERR: May I pursue?
f 25 MR. EBERSOLE: Yes, sure.
i l () Heritage Reporting Corporation (202) 628-4888 l
l l
A 266 )
1 DR. KERR: It seems to'me that what you're doing has
- {}
2 some awkward characteristics; perhaps you have no alternative, l 3 'But by using a design basis approach in its conventional sense )
4 it seems to me that you are sort of ignoring the information 5 that one has for reliability analysis, as contrasted for 6 example with PRA.
7 Inherent in the design basis approach is the use, for 8 example of the single failure criterion in the assumption that 9 beyond the single failure nothing happens.
10 Now we know that that's not the case. And indeed one 11 only gets serious situations if one gets more than a single 12 failure.
13 And so to go through the routine that we went through
() 14 in designing things to the single failure criterion using 15 design basis accidents, and then suddenly to take a plunge, or i whatever, and say okay, now we're going to forget that and 16 17 we've got to look at severe accidents just strikes me as being 18 awkward and not very productive.
19 But you guys have thought about it, and I guess 20 you've decide it isn't awkward, and it is productive.
21 MR. DAVIS: Well, I wouldn't look at those as two 22 disjointed activities that we're going to design design basis, 23 and then go off and look at severe accident issues. Under 24 existing NRC regulations, we have like requirements for 25 designing containment. And --
O Heritage Reporting Corporation (202) 628-4888
267
() 1 DR. KERR: I have -- indeed, it seems to me that's 2 one of the problems with existing NRC regulations. They are 3 disjoint at the point where one goes from design basis 4 accidents to beyond design basis accidents.
5 And I would wish that someone could rem 3ve~some of 6 that disjointedness.
7 MR. DAVIS' Well, in our current approach we plan to 8 address the current design basis requirement with all of the --
9 I guess they call it the bells and whistles needed to satisfy 10 current NRC requirements.
11 But the probablistic risk assessment that we are 12 doing at the design, it's going along at the same time that 13 we're talking about designing the current rogulations.
14 That PRA will identify the multiple failure of 15 sequences that you just mentioned. And those will then be 16 looked at as a part of the evaluation of the design of that 17 system.
18 And if we see that there are improvements we can make 19 that are cost effective because of the results of that PRA then 20 we will make those improvements.
21 The catch is that you then talk about sequences that 22 are extremely low probability. So when you make the 23 improvements you may not put all the bells and whistles on but 24 you would for the design basis events, cause you're looking at l
l 25 nore realistic best estimate type analysis, cause you are O Heritage Reporting Corporation j (202) 628-4888
268 1 looking at multiple failures already.
U'~
2 DR. KERR: Mr. Ward, did you have --
3 MR. WARD: Well, I guess I just want to go on with --
4 I think it's the same line. I guess I'm disappointed that you 5 fellows haven't been able to come up with anything better.
6 I mean, a generation ago there was kind of a default 7 recognizing that, I guess, two facts: that a), a containment 8 was needed to protect against whatever the threats from core .
9 melt downs were -- that's really what you had it for.
10 We don't have them there to protect against LOCA blow 11 down. But that was recognized. And b), it also seemed to be 12 just recognized or accepted at the time that not enough was 13 known aci.entifically about, you know, the basis or what would
() 14 happen in the severe accident to directly design this 15 containment for the severe action effects.
16 So the regulatory staff and the ACRS at the time 17 decided to kind of trick the industry into providing something 18 by this artifact of a LOCA blowdown.
19 And with the hope and expectation that if you 20 designed the containment for event A, which you really didn't 21 care about, then maybe it would do some good for event B which 22 you really cared about.
23 And that's the situation we've been in for a 24 generation. Now we've done a billion dollars worth, or half a 25 billion dollars or something of severe accident research.
O Heritage Reporting Corporation (202) 628-4888
c 269
() 1 We hopefully understand things better. Somebody has 2 to make some sense out of it. The staff -- NRC staff hasn't 3- been able to, 4 I had hoped that you guys were going to be smart 5 enough to. But you don't seem to have been. And you know, how 6 much longer are we going to have to wait till -- I mean, this 7 thing of design basis, and beyond design basis, and designing 8 for beyond design basis is kind of -- presents a problem to me.
9 You know, you go through the procedure of designing 10 for the LOCA blowdown; and then you check it against what you 11 know about severe accident to see whether it passes.
12 And my problem is that I'm afraid there's some kind 13 of -- some wish fulfillment could creep into that checking 14 process, that when we find an attribute of a severe accident 15 which doesn't pass the check we then begin to scurry around and 16 look for excuses, probablistic excuses or other sorts of 17 excuses about why we don't have to consider that.
18 And is it just impossible? I mean, in another l 19 generation are we going to know enough about the severe 20 accidents to design containments explicitly to contain what 21 they're supposed to contain?
22 MR. SHEWMON: Are you going to tell them exactly what i
23 they have to contain?
24 MR. WARD: No, of course not.
25 MR. SHEWMON: Well, that's pu of m problem,
(:) Heritage Reporting Corporation (202) 628-4888 l .
270 though, isn't it?
(]) I 2 MR. WARD: I know it. I'm trying to find out why 3 there hasn't been more progress after half a billion dollars 4 worth of research.
S' I mean, is it just a completely intractable problem?
6 MR. DAVIS: Well, Dave, I think the point is, what we 7 were trying to make, is that we are planing to design these 8 ALWRs to address the severe accident-issues; and trying to 9 design containments to work with severe accidents.
10 The point is that for a design basis events, like a 11 LOCA, you over-design, based on conservatism -- the defense and
- 12 depth argument that if you need a safety decking system, for 13 example, that could mitigate a LOCA, by defense and depth
() 14 you'll put in two trains of a system instead of one; and you'll 15 environmentally qualify the equipment to make sure it's 16 available under all possible conditions.
17 And you'll make sure that an item there has single L 18 value proof; and you'll do this, this and this.
l 19 And so you over design based on defense and depth.
20 Now what we're saying for a severe accident issue is we're 21 going to take our best estimates of what's going to happen l 22 based on the probabilities shown in the PRA, and trying to come
( 23 up with co'st effective solutions.
24 And those solutions may not have all the bells and l
25 whistles that you normally see for DBAs. But they will be l
l O Heritage Reporting Corporation (202) 628-4888 l
271
() 1 solutions that we believe, based on the PRAs, really take care 2 of the problems.
3 We're not ignoring them at all.
4 DR. KERR: But isn't it possible that you started out 5 trying to establish criteria and the accidents will occur with 6 the appropriate probabilities?
7 Rather than with design basis accidents you might 8 find yourself with a better and more cost efficient system.
9 MR. DAVIS: Well, the designs that come out from this 10 will do both -- they will address all of the events that could 11 occur;-and rather than try and take on current NRC regulations 12 and propose changes ~.here, it'll also meet the defense and 13 depth that exists in the current regulations.
14 DR. KERR: I wish you well.
15 Mr. Michelson.
16 MR. MICHELSON: Can you tell me approximately what 17 your thoughts are on fire and its severe accident potential; 18 and how are you going to approach that question?
19 *iR . DAVIS: The fire issue will be considered as one 20 of the external events in the probablistic risk assessment.
! 21 Now one of the activities that our staff is doing for us as 22 experts in that area is to evaluate whether we need to
. 23 quantitatively include that in the PRA or whether there's a 24 qualitative approach.
25 And I believe -- correct me if I'm wrong, but I O Heritage Reporting Corporation (202) 628-4888
272
() 1 believe that things are looking like we will quantitatively 2 include prior --
3 MR. SUMMIT: My name is Rick Summit, and --
4 DR. KERR: Rick, you could use this microphone over 5 here if it'd be more convenient.
6 MR. SUMMIT: Probably would be.
7 My name is Rick Summit, and I'm a Management Project 8 Manager -- Project Manager for Safety Assessments at IT 9 Corporation.
10 And has George het mentioned we are going to be 11 involved in developing a lot of the external analysis, or 12 providing guidance to CE for their PRA.
13, And we are going to be doing a peered approach in
( 14 addressing external events -- with the first idea being design 15 solutions to eliminate the potential for many of the external 16 events.
17 As you probably know, as plants have evolved over the 18 past years many of the external events analyses have indicated 19 that some external events are no longer important because the 20 effects are just no longer present.
21 And in the approach we will be using for CE, we will 22 go through and be looking at each of the external events, and l
l 23 assessing qualitatively first what the implications are of that 24 external event.
25 And once we assess those, the ones with extreme l Heritage Reporting Corporation i (202) 628-4888
l 273
{} 1 criteria -- the ones that do not get through that, or do pass through that, will be quantitatively evaluated.
2 3 Based on the present s*: ate that we have -- the 4 knowledge that we have on many of the industry initiatives 5 having to do with the EPRI requirements document, and on the CE 6 design, we presently think that fire will probably not be a 7 large contributor probablistically.
8 Many of the things that were performed along with 9 Appendix R and separation criteria that is presently being 10 planned will probably reduce the possibility of having multiple 11 training failures.
12 It could be common, say, in present plants.
13 MR. MICHELSON: Well, you're perhaps aware that O(/ 14 Sandia has recently completed now their fire study which 15 they're trying to go back and look at the risk contributions in-16 fire.
17 So you might want to get their results. It's in draft 18 form already. It doesn't quite agree with perhaps what you 19 think the view should be.
20 But I'm interested in one other aspect, and that is, 21 in doing severe accident, I think of those as something beyond 22 what we normally design -- what we normally postulate as a fire 23 design form.
24 Are you going go build bigger fires in the plant, for 25 instance, to look at their effects? Or, you know, what is a O Heritage Reporting Corporation (202) 628-4888
274
( 1 severe fire?
%.3
)
2 Can you define one, or look in the text -- or how are 3 you going to approach it?
4 MR. SUMMIT: Along the same lines that would have 5 been done for present PRAs.
6 MR. MICHELSON: Well, present PRAs don't really 7 define the fire. They define some boundaries that we 8 artificially put on it, like either one hour fire barriers; 9 there hour fire barriers.
10 But you've got -- you might want to look to see what 11 happens when you expose -- you've got a larger exposure fire 12 than you might.have anticipated; and whether it would prevent 13 those.
() 14 Because maybe the fire isn't out in one hour.
, 15 MR. SUMMIT: That's part of the things we look at 16 when we're doing the --
17 MR. MICHELSON: But you will build bigger fires, or 18 will you just assume at the design basis fire burns longer?
19 Which -- you know, which approach, for instance, are you doing?
20 Are you doing both?
21 MR. SUMMIT: At the present moment I'm not sure i
22 exactly which one we'd use on that. The idea would be to
! 23 evaluate what potential sources for a fire would be.
24 MR. MICHELSON: If you do figure out what a design 25 basis for fire is I'd like to know, too, when you build a 9 Heritage Reporting Corporation (202) 628-4888 1
l
275 1 bigger one.
{
2 MR. FONTANA: Part of these questions about PRA are 3 the subject of our brief topic paper on these matters.
4 DR. KERR: Mr. Haas, since we have asked you a lot of 5 questions and have assisted you in running over, I would hope 6 that you could move expeditiously towards summary and 7 conclusions. We would appreciate it.
8 MR. HAAS: These are the 6 sets of topic papers. As 9 I said, there's from 2 to 6 individual papers addressing 10 approaches, methods; and how one of the things you're talking 11 about here is interprutation of the severe accident policy 12 statement and safety goals implementation.
13 And then the last two on severe accident management.
() 14 What we hope out of this issue resolution process, 15 what we expect is, as we indicated, concurrence on the 16 identification of the issues, on the technical approach, first 17 of all; and what it would take an agreement that if that 18 approach were satisfied or followed the issue would be resolved 19 for ALWR; obtain the interim guidance from NRC and document 20 this formally.
21 And that impact is a timely and stable regulatory 22 interpretation for the certification process.
23 MR. HAAS: Summary, conclusion.
24 We gave you the permission, and we are accomplishing 25 this mission by the interaction, with the NRC, through CE 9 Heritage Reporting Corpos;ation (202) 628-4888
276
(} 1 System '80 to identify and resolve severe accident issues as 2 generically as possible.
3 We are modifying and developing methods. And two, 4 for both the terministic and the probablistic severe accident 5 analysis as required by the severe accident policy; and 6 providing direct technical support to the EPRI requirements 7 document; and to the specific vender certification efforts.
8 DR. KERR: Any further comments, Mr. Ward?
9 MR. WARD: No, that's all.
10 DR. KERR: Any further comments from anyone with 11 questions? Did the staff plan to make any comments?
12 MR. RUBENSTEIN: Lester Rubenstein, Staff. You've 13 heard a presentation from the industry regarding the approach
() 14 to severe accident policy.
15 I should remind the committee that the staff has a 16 preparation with the dictation of the seven requirements for 17 severe accident, as required by the severe accident policies 18 when we publish them.
19 It's a very timely subject because within the last 20 few days we had a reactor safety paper prepared, which research 21 was sponsoring, which would lay out the plan or organizing the 22 implementation of severe accident policy on the evolutionary 23 reactors.
24 And of course the RSAP program, the interest in this 25 program, is the decide to facilitate the dialogue of meeting Heritage Reporting Corporation (202) 628-4888
277
(~\
(_j 1 those requirements.
2 So it'll be shortly forthcoming in the next four to 3 five months -- a series of meetings, which research we'll have.
4 It will lay out our intentions as to the policy --
5 both preventive and mitigated dealing with the containment 6 requirements and other characteristics of the policy.
7 And I'm glad you could take it in this appropriate 8 context.
9 DR. KERR: Mr. Ward, I will thank you for a well-10 designed presentation. Appreciate you coming.
11 MR. MICHELSON: Will we have a break? Could I have a 12 question relative to this, but not of these people, and 13 particularly for the committee?
f)' How are we gol.'g to handle the severe accident 14 15 questions as they arise during the discussion of Westinghouse, 16 GE or CE reviews?'
17 DR. KERR: As Mark Train once said, I was glad to be i 18 able to answer that question promptly, and I did. I said I I
19 didn't know.
20 (Laughter) 21 MR. MICHELSON: We're going to discuss it later then, 22 or --
23 DR. KERR: Well, we certainly will have to discuss 24 it, but I have no idea how we're going to deal with it. I l
25 think one of the things that has to occur is for the staff to O Heritage Reporting Corporation I
l (202) 628-4888 i
278 Y 1 reach a position.
(~J
\_
2 And we then try to decide if that is in our view a 3 reasonable.
4 MR. MICHELSON: Yes, but maybe you're missing my 5 point, or I don't understand your. reply. My point is, we have r 6 individual subcommittees on each of these reactor types that 7 are putting the improved light water.
8 Is each subcommittee going to handle this independent 9 of the others, or are we turning it over to severe accident 10 subcommittee, or how?
11 DR. KERR: Well, I would think that except for the 12 GSAR, the staff is going to handle that issue generically and 13 not on the basis of each individual application. ,
() 14 MR. RUBENSTEIN: But each of the vendors looks at the 15 topic paper and interprets it, and designs space slightly 16 differently.
17 I think the example you got into on depressurization 18 is a handy one. It may be that someone goes to a different 19 kind of a system and say we don't have this as a credible 20 sequence.
21 So I think you could wind up doing both in a sense.
22 DR. KERR: I was more interested in our internal 23 operation.
24 MR. MICHELSON: I don't know how to answer it until 25 the staff goes further in its --
O Heritage Reporting Corporation (202) 628-4888
279 ,
l
- ' ..() 1 DR. KERR
- So right.now we won't deal with it. . i
.2 MR. MICHELSON: I don't know how to deal with it,.
3 but I'm open to suggestions.
4 Let's have a five minute break before we begin our 5 next topic.
6 (Off the record.)
7 (Back on the record.)
8 DR. KERR: Mr. Bray --
9 We don't need it.
10 -(Whereupon, at 5:05 p.m., the hearing was concluded.)
11
- 12 P
^
13 I
() 14
^
15 16 i
17 18
. 19 20 ,
21 22 23 24 25 O Heritage Reporting Corporation (202) 628-4888
l' CERTIFICATE 2
0 3 This is to certify that the attachcd proceedings before the 4 United States Nuclear Regulatory Commission in the matter of:
5 Name: ACRS: 335th GENERAL MEETING, Afternoon Session 6
7 Docket Number:
8 Place Washington, D.C.
9 Date: March 10, 1988 10 were held as herein appears, and that this is the original ,
11 transcript thereof for the file of the United States Nuclear 12 Regulatory Commission taken stenographically by me and, 13 thereafter reduced to typewriting by me or under the direction 14 of the court reporting company, and that the transcript is a 15 true and accurate record of the foregoing proceedings.
16 /S/ ocu o l 17 (Signature typed): Joan Rose 18 Official Reporter 19 Heritage Reporting Corporation 20
( 21 l
22 23 24 25 l
(}) Heritage Reporting Corporation (202) 628-4888 l
l
b *t Pm O
ANNUNCIATOR CABINET FIRES o THREE FIRES WITHIN A TWO WEEK SPAN IN ANNUNCIATOR
(]
CABINETS OF SAME MANUFACTURER l
O P
,..,.,..,,.s..- -
O o ON JANUARY 28, 1988 BEAVER VALLEY UNIT 2, IN MODE 5, EXPERIENCED AN ELECTRICAL FIRE IN A REMOTE CONTROL CABINET LOCATED BELOW THE CONTROL ROOM o REMOTE CONTROL CABINET PROVIDES THE VISUAL AND AUDIBLE INFORMATION TO THE CONTROL ROOM ANNUNCIA10R SYSTEM o FIRE WAS EXTINGUISHED IN LESS THAN 10 MINUTES o CONTROL ROOM WAS WITHOUT ANNUNCIATOR SYSTEM FOR 3 1/2 HOURS o SITE ALERT DECLARED 1/2 HOUR AFTER FIRE WAS
[]) EXTINGUISHED o CORRECTIVE ACTIONS INCLUDED REPLACING SIX DAMAGED MULTIPLE INPUT CARDS AND RESTORED COMPLETE ANNUNCIATOR SYSTEM TO SERVICE IN 24 HOURS o LONG TERM INVESTIGATIVE PLANS INCLUDE - SUBFUSING OF CIRCUITRY FOR BETTER EQUIPMENT PROTECTION, INFRA RED TEMPERATURE SCANNING, DEVELOPMENT OF EMERGENCY OPERATING PROCEDURES AND OTHERS o SPECIFIC CAUSE OF OVERHEATED CONDITION WHICH LED TO FIRE !S UNKNOWN AT THIS TIME l
1
O c ON FEBRUARY 1, 1988 CALVERT CLIFFS, UNIT 2, EXPERIENCED A SIMILAR ANNUNCIATOR SYSTEM ELECTRICAL FIRE WHILE OPERATING AT 100% FULL POWER o THE FIRE WAS EXTINGUISHED IN LESS THAN 10 MINUTES o COMPLETE LOSS OF VISUAL AND AUDIBLE ANNUNCIATIONS FOR 2 HOURS o VISUAL FEATURE RESTORED IN ABOUT 2 HOURS, AUDIBLE IN 118 HOURS j
o SITE ALERT DECLARED 1 HOUR AFTER FIRE WAS EXTINGUISHED O o CORRECTIVE ACTIONS: MODIFIED EXISTING CIRCUITRY TO RESTORE COMPLETE AUDIBLE FEATURE TO SERVICE, !NITIATED A FEASIBILITY STUDY FOR SUBFUSING, ESTABLISHED A LONG TERM REPLACEMENT PARTS PROGRAM, AND REVISED EMERGENCY OPERATING PROCEDURES TO COPE WITH SIMILAR EVENTS IS BEING CONSIDERED 1
o CAUSE OF FAILURE: UNXNOWN AT THIS TIME-0N-G0ING INVESTIGATION IS ATTEMPTING TO IDENTIFY CAUSE O
l
[
i O
o ON FEBRUARY 8, 1988 RANCHO SECO IN COLD SHUTDOWN CONDITION, EXPERIENCED A FIRE SIMILAR IN NATURE TO THE EVENTS AT BEAVER VALLEY AND CALVERT CLIFFS o THE FIRE WAS EXTINGUISHED IN LESS THAN 10 MINUTES o REVIEW 0F THE DAMAGED COMPONENTS IN PROXIMITY TO FIRE SHOWS:
THE FIRE COULD HAVE PROPAGATED OUTSIDE THE CONTROL CABINET VIA THE CABLE TRAY ENTERING THE CABINET 112 0F 192 CIRCUlT CARDS WERE UNSALVAGEABLE DUE TO EXTENSIVE HEAT DAMAGE l
O -
THREE PRINTED CIRCUIT BOARDS AND ASSOCIATED NYLON CARD GUIDES WERE DESTROYED o ALL ANNUNCIATOR WINDOWS RESTORED TO OPERATION WITHIN 72 HOURS I o CORRECTIVE ACTIONS f -
REPLACED 25% OF ALL PC BOARDS TO DATE MODIFYING PLANT E0PS TO COPE WITH SIMILAR EVENTS IN FUTURE ESTABLISHED THIRD PARTY FABRICATION PROGRAM FOR REPLACEMENT PARTS SINCE MANUFACTURER 15 NOT LONGER IN BUSINESS EXTENSIVE ON-G0ING STUDIES TO MODIFY CIRCUlT l
l Q COMPONENTS AND CHANCE MAINTENANCE AND OPERATING PROCEDURES REGARDING THE ANNUNCIATOR SYSTEM
O CONCLUSION o ALL THREE FIRE EVENTS INVOLVED EQUIPMENT BUILT BY ELECTRO DEVICES INC, o EQUIPMENT INSTALLATION AGE APPROXIMATELY 7 To 15 YEARS o ELECTRO DEVICES NO-LONGER IN BUSINESS
, o EACH LICENSEE HAS PURCHASED THE "ENGINEERING-ART-WORK"
([) FOR THIRD PARTY FABRICATION OF REPLACEMENT PARTS ;
o ONLY FOUR NUCLEAR POWER PLANTS UTILIZE THE ELECTRO DEVICES ANNUNCIATION SYSTEM o SYSTEMS ARE NOT INTERCHANGEABLE BECAUSE OF END-USERS PERSONAL SIGNATllRE ON THEIR SYSTEM o AN NRC INFORMATION NOTICE 88-05 HAS BEEN ISSUED DESCRIBING TIlE THREE EVENTS o THE STAFF IS MONITORING LICENSEES' STUDY FINDINGS TO DETERMINE IF ADDITIONAL ACTION IS NEEDED O
nom O
RANCHO SECO RESTART _
ACRS BRIEFING BY GEORGE KALMAN NRR PM l O HIGHLIGHTS:
. RANCHO SECO SHUT DOWN-DECEMBER 26,1985
. RESTART PREPARATIONS COMPLETE-MARCH 20, 1988
. COMMISSION RESTART MEETING-MARCH 22,1988
' .ACRS BRIEFING-MARCH 10,1988
- 1. EVENTS LEADING TO SHUT 00WN
- 2. SMUD RESTART EFFORTS
- 3. CURRENT PROBLEWS/ STATUS O
o RANCHO SECO BAC(GROLND IN:0RMAT ON
. SACRAMENTO MUNICIPAL UTILITY DISTRICT FORMED-1923
. RANCH SECO NUCLEAR GENERATING STATION LICENSED-1974 0 . PRE 1985 RANCH SECO OPERATING PERFORMANCE MARGINAL l . RESTART PROBLEMS FOLLOWING 1985 REFUELING 06/23/85 HIGH POINT VENT CRACK (UNIS0LABLE LEAK) 10/02/85, Rx TRIP (FEEDWATER HEATER VENT, OVERC00 LING) 10/07/85, AFW PUMP FAILURE (FAULTY MAINTENANCE) 12/05/85, Rx TRIP (ICS RELATED OPERATOR ERROR) 12/ W 35, Rx TRIP (ICS FAILURE OVERC00UNG)
O i._..,,--.,,w...,---,,-,,m-w-c y.. ,-en- _, _- _ _ _ -.__. _ _ _ _ - - - - - - - -
e JECEMBER 26,1985 EVENT ICS FAILURE, Rx TRIP, OVERC00 LING
. lCS POWER FAILURE 4 TURBINE BYPASS VALVES OPEN 50%
2 ATMOSPHERIC DUMP VALVES OPEN 50%
AFW FLOW CONTROL VALVES OPEN 50X MFW PUMP RUNDOWN TO IDLE
. Rx TRIP WITHIN 16 SECONDS (PRESSURE lI
. AFW PUMPS INITIATED j
. NO C.R. CONTROL OF ICS COMPONENTS
. TBV & ADV MANUALLY ISOLATED IN 9 MIN.
. AFW FAILED TO CLOSE MANUALLY JISO. VALVE STUCKl l
RCS DECREASED TO 1064 PSIG, 464 F, 464 F, PRZ OFF SCALE
. lCS PWR RESTORED IN 26 MINUTES
. RCS COOLED 180 IN 26 MINUTES a___ _ _ _ _ _
4
~
o EVENT FOLLOWUP
. CONFIRMATORY ACTION LETTERS FROM RV ADMINISTRATOR
-CONDUCT ROOT CAUSE OF Rx TRIP
-JUSTIFY WHY POWER OPERATIONS SHOULD BE RESUMED
-PRESERVE FAILED ECUIP FOR AUGMENTED INSPECTION TEAM
':AIT'l
. AIT/ INCIDENT INVESTIGATION TEAM <llT?
DIS 3ATCHE) TO SITE ON DECEMBER 27,1985
.llT REPORT, NUREG-1195, FEBRUARY '986 l
1
l 0 IIT FINDINGS
.lCS FAILURE
-FAULTY WIRE CRIMP TO POWER l MONITOR
-POWER MONITOR TRIPS 24 VOLT i
DC POWER
. AFW CONTROL VALVE FAllt AE
-EXCESSIVE FORCE
-lMPROPER ASSEMBLY O
,, -lNADEQUATE POSITION INDICATION
-POOR AREA LIGHTING
-EVIDENCE OF PREVIOUS DAMAGE
. AFW ISOLATION VALVE
-NO EVIDENCE OF LUBRICATION
-VALVE STUCK OPEN
. MAKEUP PUMP FAILURE
-OPERATOR ERROR, l
I (SECURED SUCTION)
O
l o llT CO\CLLSIONS ;
.B&W ICS/NNI FAILURES COMMON MARCH 20,1978-RANCHO SECO JAN. 5,1979-RANCHO SECO NOV.10,1979-0CONEE FEB. 28,1980-CRYSTAL RIVER MARCH 19,1984-RANCHO SECO
. DEC. 26,1985 TRANSIENT EFFECTS INCONSEQUENTIAL IF SMUD HAD RESPONDED TO:
. NRC BULLETIN 79-27 O
(ICS/NNI POWER FAILURE) 2.1980 NRC LTR RELATIVE TO 50% ADV POSITION
- 3. NUREG-0667, TRANSIENT RESPONSE OF B&W REACTORS
- 4. NUREG-0578, TMI SHORT TERM LESSONS LEARNED, ITEM ll.E.1.2 REQUIRED EFIC BY 1981 5.1983 NRC ORDER: COMPLETE EFIC BY 1984 O
SMUD RESTART EFFORT O
7
.SMUD COMPLETED TRANSIENT RELATED REPAIRS SCHEDULED RESTART FOR MARCH 10,1986
- .NRC STAFF INSISTS ON WIDER SCOPE OF RECOVERY PLAN
.SMUD CONTRACTS MAC TO MANAGE RESTART
.SMUD SUBMITS ACTION PLAN FOR PERFORMANCE IMPROVEMENT-JULY 1986
. ACTION PLAN FOCUSES ON MANAGEMENT AND PLANT PERFORMANCE IMPROVEMENTS O
-lNITIATED NATIONWIDE SEARCH FOR MANAGERS l -ADOPTED COMPREHENSIVE PROGRAMS TO IDENTIFY DEFICIENCIES IN:
HARDWARE DESIGN OPERATIONS PROCEDURES TRAINING MAINTENANCE O
I
ACTION PLAN FOR PERFORMANCE IMPROVEMENT l'< >
(NRC EVALUATION, NUREG-1286)
. DETAILED RESPONSE TO IIT FINDINGS
. REVIEW OF PAST EVENTS, CORRECTIVE i ACTIONS l
. REVIEW OF PAST AUDITS, THOROUGHNESS j OF RESPONSES
. PLANT STAFF INTERVIEWS / RECOMMENDATIONS l
.B&W OWNERS SAFETY AND PERFORMANCE IMPROVEMENT (SPIP)
I >
.MAINTANCE PLAN
-CORRECTIVE MAINTENANCE
-PREVENTIVE MAINTENANCE
-TROUBLESHOOTING, ROOT CAUSE PROGRAM
. EMERGENCY OPERATING PROCEDURE UPGRADE
. TECHNICAL SPECIFICATION UPGRADE
. CABLE ROUTING VERIFICATION
. SYSTEM REVIEW AND TEST PROGRAM
g SELECTEJ HARDWARE 3ROJECTS
.MOV REFURBISHMENT
. MANUAL VALVE REFURBISHMENT l
. EFIC INSTALLATION
. ADDITION OF TDI DIESEL GENERATORS
. LIQUID EFFLUENT SYSTEM l
O UPGRADE
. POST ACCIDENT SAMPLING REFURBISHMENT
.lCS/NNI REFURBISHMENT, RELIABILITY IMPROVEMENTS
. CONTROL ROOM /TSC HVAC REFURBISHMENT
. SAFETY PARAMETER DISPLAY SYSTEM (SPDS)
O
~
CURRENT PROBLEMS / STATUS
. PLANT READY FOR CRITICALITY MARCH 20,1988
.0PERATIONAL READINESS INSPECTION THRU MARCH 10,1988
. PROCUREMENT INSPECTION THRU i I MARCH 10,1988 (COUNTERFEIT BOLTS)
.TDC DIESEL VIBRATION EVALUATION THRU MARCH 11,1988
.5 MONTH POWER ASCENSION PROGRAM
.0 VEST RECOMMENDATION TO CLOSE RANCHO SECO
. JUNE 1988 VOTERS' REFERENDUM t i
l l'
Lr i
e s
i i
i :
i i ANNUNC,IATOR CABINET FIRES 1
5 O THREE FIRES WITHIN A TWO WEEK SPAN IN ANNUNCIATOR s
i CABINETS OF SAME MANUFACTURER 1
1 E
i i
i l
)
i 4
t
)
i O
- I . _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ . _ _ _ _. .. ~
i O
o ON JANUARY 28, 1988 BEAVER VALLEY UNIT 2, IN MODE 5, EXPERIENCED AN ELECTRICAL FIRE IN A REMOTE CONTROL CABINET LOCATED BELOW THE CONTROL ROOM o REMOTE CONTROL CABINET PROV!DFS THE VISUAL AND AUDIBLE INFORMATION TO THE CONTROL ROOM ANNUNCIATOR SYSTEM o FIRE WAS EXTINGUISHED IN LESS THAN 10 MINUTES o CONTROL ROOM WAS WITHOUT ANNUNCIATOR SYSTEM FOR 3 1/2 HOURS Q o SITE ALERT DECLARED 1/2 HOUR AFTER FIRE WAS EXTINGUISHED l
o CORRECTIVE ACTIONS INCLUDED REPLACING SIX DAMAGED MULTIPLE INPUT CARDS AND RESTORED COMPLETE ANNUNCIATOR SYSTEM TO SERVICE IN 24 HOURS l
0 LONG TERM INVESTIGATIVE PLANS INCLUDE - SUBFUSING 0F CIRCUlTRY FOR BETTER EQUIPMENT PROTECTION, INFRA RED TEMPERATURE SCANNING, DEVELOPMENT OF EMERGENCY OPERATING PROCEDURES AND OTHERS l 0 SPECIFIC CAUSE OF OVERHEATED CONDITION WHICH LED TO l
FIRE IS UNKNOWN AT THIS TIME O
O o ON FEBRUARY 1, 1988 CALVERT CLIFFS, UNIT 2, EXPERIENCED A SIMILAR ANNUNCIATOR SYSTEM ELECTRICAL FIRE WHILE OPERATING AT 100% FULL POWER o THE FIRE WAS EXTINGUISHED IN LESS THAN 10 MINUTES o COMPLETE LOSS OF VISUAL AND AUDIBLE ANNUNCIATIONS FOR 2 HOURS o VISUAL FEATURE RESTORED IN ABOUT 2 HOURS, AUDIBLE IN 48 HOURS o SITE ALERT DECLARED 1 HOUR AFTER FIRE WAS EXTINGUISHED O
o CORRECTIVE ACTIONS: MODIFIED EXISTING CIRCUITRY TO RESTORE COMPLETE AUDIBLE FEATURE TO SERVICE, INITIATED A FEASIBILITY STUDY FOR SUBFUSING, ESTABLISHED A LONG TERM REPLACEMENT PARTS PROGRAM, AND REVISED EMERGENCY OPERATING PROCEDURES TO COPE WITH SIMILAR EVENTS IS BEING CONSIDERED
~
o CAUSE OF FAILURE: UNKNOWN AT THIS TIME-0N-G0ING INVESTIGATION IS ATTEMPTING TO IDENTIFY CAUSE l
l l
O
l O
o ON FEBRUARY 8, 1988 RANCHO SECO IN COLD SHUTDOWN CONDITION, EXPERIENCED A FIRE SIMILAR IN NATURE TO THE EVENTS AT BEAVER VALLEY AND CALVERT CLIFFS o THE FIRE WAS EXTINGUISHED IN LESS THAN 10 MINUTES ,
o REVIEW 0F THE DAMAGED COMPONENTS IN PROXIMITY TO FIRE SHOWS:
THE FIRE COULD HAVE PROPAGATED OUTSIDE THE CONTROL .
CABINET VIA THE CABLE TRAY ENTERING THE CABINE'i 112 0F 192 CIRCulT CARDS WERE UNSALVAGEABLE DUE T0 Q EXTENSIVE HEAT DAMAGE THREE PRINTED CIRCUIT BOARDS AND ASSOCIATED NYLON CARD GUIDES WERE DESTROYED o ALL ANNUNCIATOR WINDOWS RESTORED TO OPERATION WITHIN 72 HOURS o CORRECTIVE ACTIONS REPLACED 25% OF ALL PC BOARDS TO DATE MODIFYING PLANT E0PS TO COPE WITH SIMILAR EVENTS IN FUTURE ESTABLISHED THIRD PARTY FABRICATION PROGRAM FOR REPLACEMENT PARTS SINCE MANUFACTURER IS NOT LONGER IN BUSINESS EXTENSIVE ON-G0ING STUDIES TO MODIFY CIRCulT
[]) COMPONENTS AND CHANCE MAINTENANCE AND OPERATING PROCEDURES REGARDING THE ANNUNCIATOR SYSTEM
s O
CONCLUSION o ALL THREE FIRE EVENTS INVOLVED EQUIPMENT BUILT BY ELECTRO DEVICES INC, o EQUIPMENT INSTALLATION AGE APPROXIMATELY 7 To 15 YEARS ,
o ELECTRO DEVICES NO-LONGER IN BUSINESS c EACH LICENSEE HAS PURCHASED THE "ENGINEERING-ART-WORK"
() FOR THIRD PARTY FABRICATION OF REPLACEMENT PARTS o ONLY FOUR NUCLEAR POWER PLANTS UTILIZE THE ELECTRO DEVICES ANNUNCIATION SYSTEM o SYSTEMS ARE NOT INTERCHANGEABLE BECAUSE OF END-USERS PERSONAL SIGNATURE ON THEIR SYSTEM i o AN NRC INFORMATION NOTICE 88-05 HAS BEEN ISP!ED DESCRIBING THE THREE EVENTS o THE STAFF IS MONITORING LICENSEES' STUDY FINDINGS TO DElERMINE IF ADDITIONAL ACTION IS NEEDED l ()
INDI AN POINT 2 STEAM GENERATOR DRY OllT JANilAPY 3, 1988 O EVENT DESCRIPTION INITIAL CONDITIONS, ,
PLAFT WAS BEING HEATED TO 350*F FOR STARTUP HYDRO TEST USING 2 RCS PUPPS 1 F0 TOR-DPTVEN AFW PUMP AND 2 STEAM GENERATORS AVAILABLE FOR HEAT REMOVAL AFW PllMP 23 WHICH SUPPLIES SG 23 AND 24 OllT OF SERVICE MSIV 1-23 PASSING AB0llT 3000 LB/HR SE0llENCF 0F EVEllTS:
()
- 01/02/88 SG ?3 LEVEL DECREASING 01/03/88 0800 SG 23 B0ILED DPY 01/04/88 0700 SHIFT SUPERVISOR INFORMED - INSTPitCTS OPERATORS TO St.0WLY FILL SG 23
. 01/0h/88 0800 STATION MANAGEMENT INFORMED OF DRY OllT BY CHEMISTRY MANAGER 01/04/88 0835 OPERATOPS ATTEMPT TO FILL SG 23 01/04/88 0845 OPERATOPS INSTRllCTED NOT TO FILL SG 23 IINTIL ASSESSMENT COMPLETE AND PROCEDURE k'PITTEN i
01/05/88 0100 SG 23 PEFILL PROCEDURE IMPLEMENTED (TRANSFER 0F WATF.R FROM SG 21 THR0llGH A FLOWDOWN LINE)
([)
01/07/88 CAL ISSt!ED AND AIT DISPATCHED TO SITE
[]) FINDINGS DUF TO INITIAL CONDITIONS, SAFETY SIGNIFICANCE AND POTENTIAL ADVERSE IMPACT TO PLANT EQUIPPENT MINOR IN FATURE OPEPATIONS PAPAGEMENT AND SHIFT SUPERVISORS WERE NOT ADEQUATELY AWARE OF PLANT STATUS AND DID NOT PROVIDE ADE011 ATE CONTINGENCIES FOR OPERATORS OPERATORS DID NOT ADEQUATELY REPOP.T PLANT STATils AND ACTIVITIES TO SHIFT SilPERVISORS l ([) LICENSEE EXHIBITED WILLINGNESS TO DEVIATE FROM APPROVED DIRECTIVES AND PROCEDURES AND TO PEPFORM EVOLUTIONS WITHOUT PP0CEDllRES OPEPATOPS MISAPPLIED E0P ANALYSIS TO N0PMAL PLANT EVOLIITI0FS l
O
LTCENSEE CORRECTIVE ACTIONS
(])
MANDATORY BOARD WALKS BY SWS, SP0 AND STA THREE TIMES PER SHIFT STA ASSIGNED TO SHIFT PERIODIC T0llRS OF CCR BY OPERATIONS MANAGEMENT TO EflSUPE PERSONt!EL AWAPENESS OF PLANT STATUS TRAlt'TNG 0F OPERATORS CONCERNING SG DRY OUT INCREASED EMPHASIS ON COMMllNICATION AND OPERATOR ASSERTIVENESS INCREASED EMPHASIS Ot! FOLLOWit!G PROCEDilRES It! CREASED TRAINING FOR OPERATORS CONCERNING COLD SHUTDOWN,
(]) HOT SHilTDOWN, CONSIDERATIONS AND PRECAtlTIONS PEINF09CFPENT OF WORK SCPEDULE AND OPERATIONAL PLANNINr.
FUNCTIONS TO MORE FilLLY EMPHASIZE PPEPLANt!ING AND EVALHATION OF PLANNED P' ANT EVOLUTIONS Ill ADVANCF. OF IMPLEMENTATI0tl O
N!NE Mil.E POINT llNIT ? REACTOR SCRAM AND VESSEL OVERFil.i.
w JANilAPY 20, 1988 (a,
SEQUENCE OF EVENTS
'lllTIAL CONDITIONS: 42" POWER; BOTil PECIRC Pl!MPS OllT OF SERVICE PUE TO PLANNED NATURAL CIPClllATION TESTING RV LEVEL TIME * (INCHES) DESCRIPT10M PHASE 1 - LOSS OF FEEDWATER:
- 1 MIN, 183 INSTPilMENT AIR SilPPLY LOST,
- 40 SEC. 183 M!NIMilM FLOW VALVES FOR THE CONDENSATE, CONDENSATE B0OSTER, AND FEED PUMPS Fall GPEN,
- 17 SEC. 180 FEED PilMoS AND ONE CONDENSATE B00 STER Pl'MP TRIP, LEVEL MASTER CONTROLLER SWITCHED TO MANilAL MODE, 0 159 REACTOR SCRAM ON LOW VESSEL LEVEL (LEVEL. 3),
.25 MIN, 110 HPCS AND PCIC AllT0VATICALLY lt' JECT (LEVEL 2),
~
4-5 MIN, 20? REACTOR LEVEL AT 202 TNCHES, PRESSURE DECDFASES FROM 655 PSIG TO 6?5 PSIG,
- TIME REFERENCED TO REACT 0P SCRAM AT 9:45 A,M, ON JANUARY 20, 1988,
RV LEVEL DESCRIPTION
([) T I MF.* (INCHFS)
DHASE 2 - VESSEL. OVERFILL:
- - ISOLATED INSTRllMENT ATR TRAIN PEOPENED, 5 MIN, 202 FEEDWATER MINIMUM FLOW VALVES RECLOSr.,
5,?5 MIN, 202 FEEDWATER PRESSl]RE (FPOM THE CONDENSATE B0OSTER PilMPS) RAPIPLY INCREASES, AND FEFDWATER INJECTS, 6,b MIN, 235 OPERATOR ATTEMPTS TO CLOSE TflE LEVEL CONTROL VALVES, VALVES LOCK llP AT 807, OPEN, 7,7 MIN, 250 MAIN STEAM LINES BEGIN FILLING.
11,5 MIN, LEVEL RISES ABOVE MAIN STEAM LINES,
([) ?75 I? 5 MIN, 310 OPERAT09 SHilTS THE FEEDWATEP ISOLATION VALVES,
- 13,5 MIN, 333 LEVEL REACHES A MAXIMUM, i LEVEL DECPEAFES AS WATER DRAINS VI A Rk'CU AND STEAM LIFE DRAINS,
- TIPE REFERENCED TO REACT 0P FCPAM AT 9:45 A.M. ON JAN!!ARY 20, 1980, r
l l
O
CONCLUSIONS AND COPRECTIVE ACTIONS O MA,10R FINDINGS CORPECTED PRIOR TO RESTART STRESS ANALYSES OF PAIN STEAM LINES AND VESSEL N0ZZLES COMPLETED ANALYSES OF P! PING, SUPP0 PTS, AND VESSEL N0ZZLES FOR:
-- THERMAL
-- FECHANICAL
-- DYNAMIC /IlYDRAULIC All. STRESS FOUND TO BE WITHIN ALLOWARLE VALUES, LOCKIIP 0F LEVFl. CONT 00L VALVES IN 80% OPFM POSITION
- REVISED DESIGN AND MODIFIED HARDWARE 70 REMOVE AN EXISTING -
GP0llND LOOP TN CONTROL CIPClllT, (LOOP HAD BIASED THE LOCKilP SETPOINT,)
IMPROPER CONTAINMENT ISOLATION LEVEL. 1 AND LEVEL 2 SETP0INTS
(])
- DETERMINED COMPUTATIONAL ERPOR IN ELECTRICAL REFERENCE VALUE,
- PEVISED PROCEDURE AND RESET SFTP0lHTS,
- REVIEWED ALL OTHER I.EVEL, PRESSURE, AND TEMPERATilRE PEFEPENCE VALUES, REV!EW, PFVISE, AND RETRAIN ON OPERATING PROCEDURES
. pcviENED E0PS AND OPERATING PROCEDURES.
- REVISED INSTRllPENT ATP OPERATING PPOCEDURE TO PROVIDE MORE GUIDANCE.
- REVISED SHUTDCP' PROCEDURE TO CONTINUALI.Y MONITOR VESSEL LEVEL WHEN LEVEL CONTROL IS IN MANUAL MODE, RETRATPED All OPERATOR 9 ON PROCEDURE CHANGES AND EVENT LESSONS, l
O l
I
?. . .
- O DOE ADVANCED REACTOR SEVERE ACCIDENT PROGRAM (ARSAP)
DAN GlESSING TONY BUHL O MARIO FONTANA PAUL HAAS PRESENTED TO THE ADVISORY COMMITTEE ON REACTOR SAFEGUARDS i
o MARCH 10,1988
O THE PURPOSE OF THIS BRIEFING IS TO:
(1) PROVIDE INFORMATION ON THE DOE ARSAP PROGRAM o MISSION / OBJECTIVES o PARTICIPANTS / INTERFACES t o STRUCTURE O
o ACCOMPLISHMENTS TO DATE AND
! EXPECTED FUTURE RESULTS O
1
O THE PURPOSE OF THIS BRIEFING IS TO: (2)
(2) DESCRIBE THE PLANNED CE/ARSAP INTERACTION WITH NRC TO RESOLVE S/A ISSUES FOR ALWRs o PROCESS ,
o TOPIC PAPER CONTENT O o SCHEDULE o EXPECTED RESULTS
~
(3) OBTAIN FEEDBACK FROM ACRS O
,_ _g _ g m - mA A.A.-a.- .uh, ---,M--a-- - - - -u-1 m - ) .s.s mma
, _ _ _. - -sA-- _s. --,-a .s-aas 4*
6 4 l
.- (
0 l
e i
O THE ARSAP PROGRAM l
O
O ;
DOE ARSAP MISSION o ASSIST REACTOR VENDORS AND EPRI IN THE IDENTIFICATION AND EARLY RESOLUTION OF RISK SIGNIFICANT SEVERE ACCIDENT ISSUES SO THAT THEY WILL NOT BE MAJOR OBSTACLES i
TO THE CERTIFICATION OF
. EVOLUTIONARY ADVANCED LIGHT WATER -
O REACTORS DURING THE 1990's h
I
.?
e s
O
O i
i DOE ARSAP OBJECTIVES o SUPPORT SEVERE ACCIDENT ASSESSMENT, ANALYSES, AND ISSUE RESOLUTION BEING PERFORMED BY REACTOR VENDORS WHO ARE PARTICIPATING IN DOE's DESIGN VERIFICATION PROGRAM ,
i o PROVIDE INPUT TO, REVIEW, AND SUPPORT EPRI ALWR REQUIREMENTS
- O DOCUMENT IN SEVERE ACCIDENT AREAS.
l l o ACHIEVE MORE GENERIC RESOLUTION OF SEVERE ACCIDENT ISSUES THAN MAY OCCUR THROUGH DIRECT SUPPORT OF l
SPECIFIC VENDOR CERTIFICATION EFFORTS l
r O
, O ARSAP PARTICIPANTS DOE: SUPPORTING AGENCY EG&G IDAHO: PROGRAM DIRECTOR IT CORPORATION: TECHNICAL DIRECTOR ,
FAl: SUBCONTRACTOR CONTRACTOR FOR S/A ANALYSIS METHODS DEVELOPMENT O
CE: ALWR DESIGN VERIFICATION CONTRACTOR ITAG: INDUSTRY TECHNICAL ADVISORY GROUP CONSULTANTS: ERCl(IEAL), SAROS, FHR, a.R. BENJAMIN, G. KAISER (NUS)
O l.___-_.__-_.-_.-_.-_____________ _ _ - _ __
i O O O -
l l
ARSAP ORGANIZATION i
DOE l ALWR DESIGN -
l VERIFICATION -~~----------------~~~
l PROGRAM
! Combustion l Engineering PROGRAM MANAGER
! EG&GIdaho -
l s
TECHNICAL DIRECTOR IT Corporation INDUSTRY TECHNICAL ADVISORY GROUP ANALYSIS METHODS SEVERE ACCIDENT SUPPORT TO ISSUE RESOLUTION VENDORS AND EPRI S/A Analys.is - FAI IT Corporation IT Corporation PRA - IT Subcontractors Subcontractors
~ Consultants Consultants
O O O -
ARSAP SUPPORT TO INDUSTRY TO ADDRESS SEVERE ACCIDENT ISSUES WITH NRC ARSAP Topic o issues CE System 80 + Papers Methods Certification
^
o S/A Analys,is o PRA Support hk
_ _ _ _ __ JL
.. EPRI .
N i ALWR Requirements -
' R l
Document
! l J L ll C
- i i
, ABWR >
l Certification i >
1 l Westinghouse l = APWR
_____________.. Certification
O ARSAP WORK BREAKDOWN STRUCTURE l S/A ISSUE RESOLUTION W BS-2: IDENTIFY, CATEGORIZE AND PRIORITIZE S/A ISSUES PERTINENT TO ALWRs. ,
DEVELOP, CONDUCT AND DOCUMENT A PROCESS FOR RESOLUTION OF THOSE ISSUES IN SUPPORT OF INDUSTRY ALWR O LICENSING / CERTIFICATION EFFORTS o IDENTIFY, CATEGORIZE, AND PRIORITIZE ISSUES (SIX SETS) l o PROPOSE RESOLUTION APPROACH IN S/A ISSUE TOPIC PAPER 1
o SUPPORT CE INTERACTION WITH NRC/ACRS TO GAIN CONCURRENCE l
o SUPPORT IMPLEMENTATION OF APPROACH AND CLOSURE WITH NRC w - - - _ __ _ __ _ _ - __ _ _ . __ . . .
8 O
ARSAP WORK BREAKDOWN STRUCTURE (2)
S/A ISSUE RESOLUTION (CONT'D)
WBS-7: DOCUMENT THE EXPERIENCE BASE l w.r.t. PREVENTION AND MITIGATION OF SEVERE ACCIDENTS IN ALWR's l .
o INITIAL INDUSTRY-WIDE COLLECTION, ASSESSMENT AND DOCUMENTATION OF O LESSONS-LEARNED o ANNUAL UPDATES - BASED ON ARSAP RESULTS O
1
O ARSAP WORK BREAKDOWN STRUCTURE (3)
METHODOLOGY DEVELOPMENT WBS-3: PRODUCE, EVALUATE AND DOCUMENT A SET OF INTEGRATED SEVERE ACCIDENT ANALYSIS TOOLS THAT CAN BE USED BY THE NUCLEAR INDUSTRY FOR ASSESSMENT OF SEVERE ACCIDENT PHENOMENA IN O ALWRs f o EVALUATE AND SELECT COMPUTER CODE (s)
! o MODIFY & IMPROVE FOR ALWR ANALYSIS j
o VERIFY AND VALIDATE THE INTEGRATED SEVERE ACCIDENT CODE (MAAP-DOE) l o DOCUMENT THE CODE & PROVIDE USER SUPPORT i
O
O O
ARSAP WORK BREAKDOWN STRUCTURE (4)
METHODOLOGY DEVELOPMENT (CONT'D)
WBS-4: PROVIDE GUIDANCE, METHODS, AND TECHNICAL SUPPORT TO EPRI, VENDORS AND UTILITIES FOR THE APPLICATION OF PROBABILISTIC METHODS t
O o PROVIDE GUIDANCE AND APPROACHES FOR APPLICATION OF PROBABILISTIC METHODS IN THE ALWR DESIGN PROCESS I o SUPPORT EPRI's DEVELOPMENT OF l FUNCTIONAL PRAs o SUPPORT EPRI's DEVELOPMENT OF PRA GROUNDRULES AND ASSUMPTIONS 1
o SUPPORT THE USE OF PRA IN S/A MANAGEMENT O
i l
~
O l
ARSAP l WORK BREAKDOWN STRUCTURE (5) t SUPPCRT FOR VENDORS AND EPRI WBS-8: SUPPORT THE EPRI ALWR REQUIREMENTS DOCUMENT IN S/A RELATED AREAS .
o INPUT TO AND REVIEW OF ALL CHAPTERS w.r.t. S/A ISSUES o DEVELOP SPECIFIC S/A RELATED REQUIREMENTS l
o PERFORM ANALYSES AS PART OF TECHNICAL BASIS FOR S/A RELATED REQUIREMENTS o SUPPORT EPRI INTERACTION WITH NRC I
l O l
O ARSAP WORK EHEAKDOWN STRUCTURE (6)
SUPPORT FOR VENDORS AND EPRI (CONT'D)
WBS-10: SUPPORT CE SYSTEM 80+ PLANT CERTIFICATION PER REQUIREMENTS l OF THE S/A POLICY STATEMENT l o IDENTIFY / REVIEW S/A ISSUES RELEVANT TO O SYSTEM 80+
o DEFINE SEQUENCES FOR ANALYSIS l o PERFORM AND ASSIST WITH S/A ANALYSIS o ASSIST IN PRA o ASSESS SEVERE ACCIDENT MITIGATION FEATURES o SUPPORT FDA AND NRC REVIEW PROCESS O
~
O ARSAP WORK BREAKDOWN STRUCTURE (7)
SUPPORT VENDORS AND EPRI (CONT'D)
WBS-11: SUPPORT GE ABWR CERTIFICATION IN SPECIFIC S/A ANALYSIS AND '
PRA AREAS o ASSISTANCE WITH BWR MAAP CODE o TECHNICAL SUPPORT FOR F.P. PARTICLE SIZE DETERMINATION o ASSIST WITH DEVELOPMENT OF OVERALL APPROACH AND SPECIFIC ANALYSIS FOR PRA SEISMIC ANALYSIS o SUPPORT ABWR ATWS ANALYSIS WBS-12: COORDINATE WITH WESTINGHOUSE APWR DESIGN EFFORT O
l
I
\~o ARSAP MAJOR ACCOMPLISHMENTS TO DATE PROGRAM MANAGEMENT AND PLANNING o PROGRAM PLAN AND IMPLEMENTATION PLAN COMPLETED (ANNUAL UPDATES) ,
o INTERFACES WITH EPRI, VENDORS, AND NRC (VIA CE) ESTABLISHED i O l
o INDUSTRY TECHNICAL ADVISORY GROUP (ITAG) CHARTERED AND OPERATING O
O
O 1 ARSAP MAJOR ACCOMPLISHMENTS TO.DATE (2)
RESOLUTION OF S/A ISSUES o IDCOR/NRC RESOLUTION EXPERIENCE AND RELEVANCE TO ALWRs DOCUMENTED, AND l LESSONS LEARNED EXTRACTED ,
o PARTICIPATED IN U.S. INVOLVEMENT IN IAEA CHERNOBYL ACTIVITIES O
o RESOLUTION PROCESS ESTABLISHED o INITIAL VERSION OF LESSONS LEARNED NOTEBOOK COMPLETED (DRAFT TO BE PUBLISHED)
O 1
l I
O ARSAP MAJOR ACCOMPLISHMENTS TO DATE (3)
RESOLUTION OF S/A ISSUES (CONT'D) o S/A ISSUE TOPIC PAPER SET 1 UNDER l
REVIEW BY NRC ,
o S/A ISSUE TOPIC PAPER SET 2 UNDER FINAL REVIEW BY CE O
o S/A ISSUE TOPIC PAPER SET 3 INITIAL DRAFT UNDER REVIEW BY ARSAP O
L - _ . - . _ _ _ _____
O ARSAP MAJOR ACCOMPLISHMENTS TO DATE (4)
SEVERE ACCIDENT ANALYSIS METHODOLOGY o MAAP CHOSEN AS PRIMARY ANALYTICAL INTEGRATED CODE .
o MAAP CODE CONFIGURATION QUALITY CONTROL (CCQC) PROCEDURE O ESTABLISHED o MAAP MODELING MODIFICATIONS AND BENCHMARKING PLANNED AND UNDERWAY o COORDINATION WITH EPRI MAAP USERS' GROUP ESTABLISHED o TECHNICAL EXCHANGE MEETINGJ CONDUCTED WITH NRC CONTRACTORS O
j- .
l O
! ARSAP MAJOR ACCOMPLISHMENTS
( TO DATE (5)
( PRA METHODOLOGY AND APPLICATIONS o GUIDANCE FOR USE OF PRA IN DESIGN BEING DEVELOPED ,
l o PRA GROUNDRULES AND ASSUMPTIONS DEVELOPED FOR EPRI O
o HIGH LEVEL PWR & BWR FUNCTIONAL PRA MODELS DEVELOPED FOR EPRI l
O
O ARSAP MAJOR ACCOMPLISHMENTS TO DATE (6)
SUPPORT TO EPRI o SIGNIFICANT INPUT TO EPRI ALWR REQUIREMENTS DOCUMENT CHAPTER 1 ,
AND 5 o DETAILED REVIEW OF CHAPTERS 1,2, O 3, 4, 3 o INPUT ON CHAPTER 61N PROGRESS o DEVELOPMENT OF TECHNICAL SUPPORT l FOR S/A - RELEVANT REQUIREMENTS PUBLISHED AND IN PROGRESS, e.g.,
H2 CONTROL, VALVE SIZING FOR RCS i DEPRESSURIZATION O
O ARSAP MAJOR ACCOMPLISHMENTS TO DATE (7)
SUPPORT TO VENDORS o CE SUPPORT IN PROGRESS o GE SUPPORT IN PROGRESS o MONITORING WESTINGHOUSE PROGRAM l
O
O ARSAP l PLANNED FY88 MAJOR RESULTS o SIX SETS OF S/A ISSUE TOPIC PAPERS TRANSMITTED TO NRC; INTERIM GUIDANCE RECEIVED FROM NRC ON SETS i 1 AND 2 o ANNUAL UPDATE OF LESSONS LEARNED l NOTEBOOK O
j o MODIFICATIONS AND BENCHMARKING OF l PWR VERSION OF MAAP-DOE COMPLETED o UPGRADED FUNCTIONAL PRA MODELS COMPLETED AND DEMONSTRATED FOR EPRI O
O ARSAP PLANNED FY88 MAJOR RESULTS (2) o TECHNICAL REPORTS IN SUPPORT OF EPRI REQUIREMENTS DOCUMENT AND CE/ARSAP ISSUE RESOLUTION, e.g.,
HYDROGEN CONTROL, DEBRIS
~
COOLABILITY l o TECHNICAL REPORTS IN SUPPORT OF CE O SYSTEM 80+ SEVERE ACCIDENT ANALYSIS o TECHNICAL REPORTS ON SYSTEM 80+ PRA EXTERNAL EVENTS ANALYSIS AND SUPPORT SYSTEM SENSITIVITY o TECHNICAL REPORTS IN SUPPORT OF GE ABWR S/A ANALYSIS, e.g., SEISMIC FRAGILITY DATA, ATWS ANALYSIS O
O POTENTIAL FUTURE INITIATIVES (1989-1990) o SUPPORT EPRI AND VENDORS IN S/A ASSESSMENT FOR PASSIVE PLANT DESIGN AND CERTIFICATION o TECHNICAL SUPPORT FOR ADDRESSING l
SEVERE ACCIDENT MANAGEMENT IN ALWRs O
l
, O 1
l L____.___._____________--_. . - . . . . _ - .--
1 O
l l
l l .
i l SEVERE ACCIDENT ISSUE RESOLUTION O PROCESS O
O DOE /ARSAP/CE ARE WORKING WITH THE NRC TO RESOLVE SEVERE ACCIDENT ISSUES o RESOLUTION PROCESS WAS PRESENTED IN BRIEFING TO BECKJORD, MURLEY, AND MANAGERS (7/16/87 AND 8/3/87) .
! o PROCESS AND SCHEDULE DETAILED WITH NRC NRR STAFF (8/12/87 AND 12/16/87) o FIRST SET OF ISSUE PAPERS SUBMITTED TO NRC (11/24/87) o SECOND SET OF ISSUE PAPERS IN FINAL REVIEW BY CE o INITIAL DRAFT OF THIRD SET OF ISSUE PAPERS COMPLETED 1
l 1
0
O DOE /ARSAP/CE ARE WORKING WITH THE NRC TO RESOLVE SEVERE ACCIDENT ISSUES o TECHNICAL INFORMATION EXCHANGE MEETING ON MAAP AND MELCOR (1/5-7/88) o TECHNICAL INFORMATION EXCHANGE l
MEETING WITH NRC CONTRACTORS AT O BNL (1/20-21/88) o IDENTIFIED AND REVIEWED WITH NRC STAFF MAJOR DECISIONS w.r.t. S/A ISSUES FOR ALWRs (3/9/88)
O l
O DOE /ARSAP/CE ARE INTRODUCING ,
ALWR SEVERE ACCIDENT ISSUES TO NRC FOR RESOLUTION THROUGH THE CE SYSTEM 80+ CERTIFICATION PROCESS O
l l
l
! O
O SEVERE ACCDENT RESOUmON PROCESS DEFINE ISSUE AND PROPOSED RESOLUTIO APPROACH U
PRODUCE ARSAP I_
U REVIEW WITH CE &
ITAG AND MODIFY O TOPIC PAPER t
l NRC REVIEW AND ANAL S/ DESIGN INTERIM GUIDANCE MODS.
t l
ISSUE NO RESOLVED?
YES Y
O FINAL RESOLUTION DOCUMENTED BY NRC
e 4 O
TOPIC PAPER SETS - SCHEDULE FOR SUBMITTAL TO CE SET 1. RESOLVED IDCOR/NRC ISSUES -
APPLICABILITY TO ALWRS (10/87) l SET 2. PLANT RESPONSE UNDER SEVERE ACCIDENT CONDITIONS (2/88)
O SET 3. PROBABILISTIC METHODS (4/88)
SET 4. SEVERE ACCIDENT PERFORMANCE (6/88)
SET 5. SAFETY GOALS EVALUATION (8/88)
SET 6. SEVERE ACCIDENT MANAGEMENT (9/88) 1 O
l
,-,,----------w----, -w -w m we-- -v,--,- ~ ~ - ---,-, ,r- - - ---- -e w- --~w'- '
O EACH PAPER WILL INCLUDE:
\ .
o ISSUE DEFINITION o HISTORICAL PERSPECTIVE l
- INDUSTRY ACTIONS, PREVIOUS ,
POSITION t
- NRC ACTIONS, PREVIOUS POSITION o TECHNICAL APPROACH TO RESOLUTION FOR ALWRs (AS GENERIC AS POSSIBLE) e f
O
~
O l RESULTS EXPECTED FROM THE ARSAP/CE - NRC SEVERE ACCIDENT ISSUE RESOLUTION PROCESS l
l o CONCURRENCE ON IDENTIFICATION OF ISSUES o CONCURRENCE ON TECHNICAL APPROACH AND CRITERIA FOR RESOLUTION O o INTERIM NRC GUIDANCE SO DESIGN l
CERTIFICATION CAN PROCEED o RESOLUTION DOCUMENTATION o TIMELY AND STABLE REGULATORY I INTERPRETATION FOR CERTIFICATION l
l O
l
i i
O O
O
SUMMARY
AND CONCLUSION O
O DOE /ARSAP IS PROVIDING ITS RESOURCES AND-EXTENSIVE EXPERIENCE IN PRACTICAL ASSESSMENT OF SEVERE ACCIDENT ISSUES TO SUPPORT ALWR VENDORS AND EPRIIN THEIR EFFORTS TO OBTAIN EARLY RESOLUTION OF
- SEVERE ACCIDENT ISSUES
\
THIS IS BEING ACCOMPLISHED BY:
O o INTERACTION WITH THE NRC VIA THE CE SYSTEM 80+ CERTIFICATION TO IDENTIFY AND RESOLVE ISEUES AS GENERICALLY AS POSSIBLE o MODIFICATION AND DEVELOPMENT OF METHODS AND GUIDANCE FOR I APPLICATION OF METHODS o DIRECT TECHNICAL SUPPORT TO EPRI ALWR REQUIREMENTS DOCUMENT AND VENDOR CERTIFICATION EFFORTS l
O
a 5 9 e
O APPENDIX ADDITIONAL INFORMATION ON THE DOE /ARSAP/CE ALWR SEVERE ACCIDENT ISSUE RESOLUTION PROCESS O
O i
l l
O Q O -
'e O.RaecCeteges'J PeteeWae seeeee. .
AseessiWe Se=*
A.eBobee tafe Ceedbewowe
> Isoc& hog il of s/A bde.
P6edisc o "W
be 0 P' > P- Se.een Re,.dwelry o t. C.entelarvosat op o. + ,,systone sss oe.fte ---
s,oi .e.
co*e>
IJpic PePe*
, -se, ==eyl._ j tctssaan PaA D
Re.ie.
cg & svag [ -
-- j e.,4 1' e 1'
. 2'**'l Ett*L C -*
U
-'t 1'
M7,[, e level 1 PRA g, AR'9d "A E-Present ,
[ leph Peper I
. J L' L .
p i
- M 8 sesect Deelyn Arte8rne Plant Pleat Aae8rne Itsee, 4 Reedree Doelp > Festare for -> Respease le + Re*P*ae* Redwetlese g,,,,, .,j $ Eeuhescon Severe Ardsects taproved? YES C*et-8***dt et Obtehe
,_ [ _, oc s - "" unc 1
[ q, 1.
O R wtee Co h l g by Ameny.sef g scRS seeeeemeat F=* D**8e**
U Reepeace Oseages d- D Requbed? #**II **I 8 08 0 its
- /
-e ,
R&D
+
R.e
- ~
fleeeheedi -snesyste/
l '
R e,A ed 7 C*
" 1i knplesment its vts Coe8ea o U
_ %*e C ** M c eamccis: Co,ndbee,stery e,, ,o.,, e #.160 5 <
s u Reg (Cert. '
scue. by este '
Detailed Severe Accident Issue Resolution Process and Interaction wit *n Coubustion Engineering I TC-07@
i O
TOPIC PAPER SET 1:
RESOLVED IDCOR/NRC ISSUES -
APPLICABILITYTO ALWRS (SINGLE PAPER) o REACTOR COOLANT SYSTEM NATURAL CIRCULATION (IDCOR ISSUE 2) ,
o IN-VESSEL STEAM EXPLOSIONS AND ALPHA MODE FAILURE (IDCOR ISSUE 7)
O o EX-VESSEL HEAT TRANSFER MODELS FROM MOLTEN CORE TO CONCRETE (IDCOR ISSUE 10) o FISSION PRODUCT RELEASE PRIOR TO VESSEL FAILURE (IDCOR ISSUE 1) l o RELEASE MODEL FOR CONTROL . ROD l MATERIALS (IDCOR ISSUE 3) o FISSION PRODUCT AND AEROSOL DEPOSITION IN RCS AND CONTAINMENT O (iDCOR ISSUES 4 AND 12)
O TOPIC PAPER SET 1:
RESOLVED IDCOR/NRC ISSUES ' -
APPLICABILITY TO ALWRS t,2) o EX-VESSEL FISSION PRODUCT RELEASE (DURING CORE - CONCRETE INTERACTIONS) (IDCOR ISSUE 9) ,
o REVAPORIZATION OF FISSION PRODUCTS (IDCOR ISSUE 11)
O o SECONDARY CONTAINMENT PERFORIVIANCE (IDCOR ISSUE 16) o MODELING OF EMERGENCY RESPONSE (IDCOR ISSUE 14) l 1
O TOPIC PAPER SET 2:
PLANT RESPONSE UNDER SEVERE .
ACCIDENT CONDITIONS o IN-VESSEL HYDROGEN GENERATION (IDCOR ISSUE 5) l o CORE MELT PROGRESSION AND VESSEL FAILURE (IDCOR ISSUE 6) l O o DIRECT CONTAINMENT HEATING BY EJECTED CORE MATERIALS (IDCOR ISSUE 8) o CONTAINMENT PERFORMANCE (IDCOR ISSUE 15) o HYDROGEN IGNITION AND BURNING l
(IDCOR ISSUE 17) o DEERIS COOLABILITY (IDCOR ISSUE 10) 1 l O I
O TOPIC PAPER SET 3:
METHODOLOGY PRA o EXTERNAL EVENTS o SUCCESS CRITERIA o ACCIDENT SEQUENCE SELECTION o PUMAN RELIABILITY ANALYSIS o COMMON CAUSE FAILURE l
l l
l O
. l O
TOPIC PAPER SET 4:
SEVERE ACCIDENT PERFORMANCE o ESSENTIAL EQUIPMENT PERFORMANCE (IDCOR ISSUE 18) o CRITERIA FOR SAFE STABLE STATES ,
O
<O
O TOPIC PAPER SET 5:
SAFETY GOALS EVALUATION o SAFETY GOAL IMPLEMENTATION --
INTERPRETATION OF GOALS AND USAGE OF PRA RESULTS IN COMPARISON WITH GOALS, INCLUDING INTERPRETATION OF ,
UNCERTAINTIES o UNCERTAINTIES IN PLANT RISK O ANALYSIS o MAAP-DOE CODE VALIDATION 1
l l
O -
l
4 O
TOPIC PAPER SET 6:
SEVERE ACCIDENT MANAGEMENT o S/A MANAGEMENT -- PLANNING o S/A MANAGEMENT -- EQUIPMENT CAPABILIT'/ AND OPERATIONAL ,
REQUIREMENTS i
O l
l O
t