ML20199D366
| ML20199D366 | |
| Person / Time | |
|---|---|
| Site: | San Onofre  |
| Issue date: | 03/18/1986 |
| From: | NRC COMMISSION (OCM) |
| To: | |
| References | |
| REF-10CFR9.7 NUDOCS 8603240037 | |
| Download: ML20199D366 (126) | |
Text
~
~ ~
5 ORIGINAL ..
g, ,
UNITED STATES OF AMERICA -
(
. NUCLEAR REGULATORY COMMISSION In the matter of:
COMMISSION MEETING Briefing by Southern California Edison ,
Comparty on San Onofre 1 .
(Public Meeting) ,
Docket No.
(.v %.4. 5:
s l
l Location: Washington, D. C.
Pages: 1 - 84 i Date: Tuesday, March 18, 1986 1
l l
l l
1 j
ANN RILEY & ASSOCIATES Court Reporters l[ 1625 I St., N.W.
! \_. -
I Suite 921 Washington, D.C. 20006 i
! 8603240037 860318 (202) 293-3950 1
1 D-l SCLA I MER 2
3 4
4 5
6 This is an unofficial transcript of a meeting of the 7 United States Nuclear Regulatory Commission held on S 3/18/86 . In the Commission's office at 1717 H Street, 9 N.W., Washington, D.C. The meeting was open to public 10 attendance and observation. This transcript has not been 11 reviewed, corrected, or edited, and it may contain
[2 inaccuracies. '
13 The transcript is intended solely for general
~
14 informational purposes. As provided by 10 CFR 9.10S, it is 15 not part of the formal or informal record of decision of the 16 matters discussed. Express' ions of cpinion in this transcript 17 do not necessarily reff'ect final determination or beliefs. No-18 pleading or other paper may be filed with the Commission in 19 any proceeding as the result of or addressed to any statement' 20 or argument contained herein, except as the Commission may 21 authorire.
22 l 23 i
. 24 25
1
.1 UNITED STATES OF AMERICA 2 NUCLEAR REGULATORY COMMISSION 3
4 ------
5 BRIEFING BY SOUTHERN CALIFORNIA 6 EDISON COMPANY ON SAN ONOFRE.1 7 -----
8 9 Public Meeting 10 11 Tuesday, March 18, 1986 12 1717 H Street, N.W.
13 Room 1130 -
14 Washington, D.C.
15 16 The Commission met.in open session, pursuant to 17 notice, at 10:09 a.m., the Honorable Nunzio J. Palladino, 18 Chairman of the Commission, presiding.
19 COMMISSIONERS PRESENT:
20 Nunzio J. Palladino, Chairman 21 James K. Asselstine, Commissioner 22 Frederick M. Bernthal, Commissioner 23 Thomas.M. Roberts, Commissioner 24 Lando W. Zech, Jr., Commissioner 25
i
. 2 l 1 STAFF AND PRESENTERS SEATED AT COMMISSION TABLE:
2 S. Chilk, SECY 3 H. Plaine, OGC 4
5 R..Krieger 6 M. Medford 7 K. Baskin 8 D. Fogarty 9 L. Papay 10 J. Rainsberry 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 ,
a
- , , + , - . - , - - - - - - , re,- - - . . , - - - ---- -,,n
O 3
1 PROCEEDIN.GS 2 CHAIRMAN PALLADINO: . Good morning, ladies and 3 gentlemen.
4 On November 21, 1985, Southern California Edison, 5 San Onofre Unit 1 Plant, experienced a total loss of all 6- in-plant AC power, as well as a severe incidence of water 7' hammer in the feedwater system, which caused a leak,' damaged 8 plant equipment, and challenged the integrity of their plant's 9 heat sink.
10 Although there were other malfunctions and human 11 errors, plant operators were successful in bringing the plant 12 to a stable shutdown and in preventing an abnormal release of 13 radioactivity. -
14 However, I and, I believe, most of my colleagues 15 were greatly dismayed while reading through the incident 16 investigation team report, NUREG-1190, to learn of the many 17 plant deficiencies, procedural shortcomings and apparent poor 18 Licensee approach toward root cause identification disclosed 19 by the investigation of the event.
20 On January 22, 1986, the Commission heard the 21 incident investigation team's description of the event, and 22 the principal findings and conclusions.
(
23 At the end of the meeting SCE was afforded an 24 opportunity to make a few brief remarks. Today we will hear 25 further from the Licensee.
i l
I l
4 1
1 I understand that representatives from the Staff are 2 in the audience, and that representatives from Regions IV and 3 V will-be listening.in by telephonti.
4 Iut me ask my fellow Commissioners-if they have any 5 additional opening remarks at this time.
6 COMMISSIONER ASSELSTINE: No.
7 COMMISSIONER ZECH: No.
8 CHAIRMAN .PALIADINO: All right. Then I will turn 9 the meeting over to Mr. David Fogarty, Executive Vice 10 President of San Onofre-California Edison Company, and I will 11 ask him to introduce his other colleagues.
12 MR. FOGARTY: Thank you very much, Mr. Commissioner. ,
13 [ Slide.]
14 The agenda we have for today's meeting is on the
- 15 board here, and in addition to Mr. Baskin, Mr. Medford and 16 Mr. Krieger, we are also joined by Dr. Larry Papay at the l 17 front table,' Jack Rainsberry, also at the table, and Mike l 18 Werton, our Deputy Station Manager,'as well as two 19 representatives of San Diego Gas & Electric, Jim Holcomb and i
20 Steve Allman of San Diego is our partner in San Onofre Unit 21 No. 1.
22' Next slide, please.
23 (Slide.]
24 In the preface or overview, I'd like to just make a 25 few comments before I turn the meeting over to Mr. Baskin.
i
-- 5
.1 First of all, the water hammer event on November 2 21st was very significant. The event itself was an 3 unfortunate occurrence, but the event and the surrounding 4 issues are very significant.
5 The second point we want to make here today is that 6 the San Onofre unit age requires additional care, diligence 7 .and attention over that of a newer-plant. The San Onofre 8 Plant is 18 years old, and it is about the same age as the 9 average steam unit on our system.
10 For that reason, it requires additional care, 11 diligence and attention'than would be required by a newer 12 plant. We must operate and maintain San Onofra 1 unit to an 13 especially high standard, and I think the events that occurred 14 November 21st indicate that we were not doing as good a job in 15 this regard as we should have been.
16 Some of the things that we have learned indicate that not only did we have problems at San Onofre, but in some
~
17 18 . cases the San Onofre Plant, because of the differentiation we 19 had made in setting out the separate operating organization 20 with a separate vice president, produced many benefits. There 21 were also some-problems that came about, problems of things, 22 for example, in the switchyard equipment.
23 The switchyard equipment at San Onofre was not i
24 installed and operated to the same standards that we used in i
25 our other steam plants. So it is part of the differentiation 4
- - - , -w,--- -- - - . - - - - -
.w- , - _ - - - - - - - . - - . . - , - - - - - - . .
6 1 of San Onofre into a separate nuclear facility. We had lost 2 some of the meat and potatoes, if you will,-part of the 3 business in things like the switchyard.
4 The IIT report, we believe, was professionally done 5 and raised very valid concerns. The IIT report identified to 6 us areas requiring additional attention and we are giving 7 these items the attention they deserve. Although there are 8 some small differences of opinion in the report, we accept the 9 report as it was made to you and as we have received it.
10 And the final preface comment is that we are 11 committed to address all of the NRC concerns raised by the 12 event.. We are thoroughly investigating the incident. We have 13 retained outside expertise to assist in our investigation, and 14 . we will be maintaining continued contact with the NRC Staff 15 an'd with the Region.
, 16 Now I would like to turn the presentat' ion over to 17 Mr. Baskin, who will pick up on the next agenda item.
18 MR. BASKIN: Thank you, Dave.
19 [ Slide.]
20 Mr. Chairman and Commissioners, first I would like 21 to review briefly the agenda with.you and let you know what we 22 are going to be talking about today. The presentation we have 23 planned runs about 40 minutes, so I wo'11d hope that would 24 allow enough time for questions and answers as we go along i
25 and/or at.the end.
7 1 Mr. Fogarty presented the preface. I'm going to 2 give a little introduction. Dr. Mark Medford, who is our 3 licensing manager, is going to discuss the design-related 4 issues in regard to this event. Mr. Russ Krieger, who is the 5 superintendent of operations for the three San Onofre units, 6 will discuss the program-related issues, and then I will 7 conclude with some closing remarks.
8 In Mr. Medford's discussion, he is going to cover in 9 a broad category the feedwater system, the electrical system, 10 include some discussion on maintenance, and will end his 11 presentation with a general discussion of plant modifications 12' that have arisen out of this event.
13 -
In Mr. Krieger's discussion, he is going to cover 14 the operator STA performance and what we call the material-15 condition of the plant.
16 Throughout their discussion they will be talking 17 about various causes and proposed changes as a result of the 18 causes. However, in my conclusion, I will be discussing what 19 we consider the basic root causes of the event, and how they 4
20 relate to what we've done and what we're doing.
21 (Slide.]
- 22 Just again to summarize very briefly what R23 Mr. Fogarty said, the event clearly was something that was 24 very significant, something that we are learning from, and I 25 think everybody will have an opportunity to learn from.
.. 8 1 The most significant aspect of the event was.the 2 failure of the five feedwater check valvas. This will be 3 discussed in some detail in a moment.
4 Mr. Fogarty indicated that we have.given a great '
5 deal of attention and study to this matter, not only in-house, 6 but with the assistance of outside consultants.
7 Feedwater check valves will be replaced, and you 8 will hear more about that, and what we call material condition
- 9. assurance programs are being implemented. These are two
- 10 programs that address, among other things, the unique aspects 11 of the plant, namely its age and how we are going to handle 12 that aspect.
13 .
Before we get into more details, I would like to 14 review just very briefly the incident.
15 (Slide.]
16 You have all heard and had an opportunity to read 17 the IIT report, so I am not going to go through it in detail.
j 18 Refreshing our minds, San Onofre Unit 1 is a 450 megawatt, 19 three-loop Westinghouse plant, went on the line in 1967, has 20 three loops, three steam generators, a rather' unique safety 21 injection system which, if necessary, we can talk about later.
22 Prior to-the event, activities were in process to 23 locate a fault on the electrical system. We then paralleled 24 some buses and we will get into this in a little more detail
! 25 in terms of exactly what was done. You needn't worry about
- . - . , - - - - - . - , . , - , - , , - . . , , . , - , . - , . , . ~ , - - - - - - - - - - - ~ . . , , , .. - - . . . - , , , - - . , - . . - - . - ,-- - - - - - - ,
1 l
9 l
1 that now. l 2 The auxiliary transformer, that was the supply of 3 power from the switchyard to the plant relay. At that point 4 in time diesel generator No. 2 automatically started, as it 5 was designed to do. It did not load, also as it was designed,.
6 and again we will hear a little more about that.
7 Because of the parallel bus operation, when the 8 transformer relayed, the unit'did not trip, it was manually 9 tripped 20 seconds later, and at the same time the second 10 diesel generator started, diesel generator No. 1, as planned.
11 Approximately one minute after the trip, the main 12 feedwater lines voided -- in other'words, emptied themselves of water -- due to the failure of the check valve, and they 13 14 voided through a rupture further upstream in the condensate 15 system, namely the flash evaporator.
16 The auxiliary feedwater pump went through its 17 warm-up cycle, began delivering auxiliary.feedwater 18- approximately four minutes after the trip. At about the same 19 time, electrical power was restored from an offsite source, 20 and the auxiliary feedwater motor-driven pump then began 21 operation roughly a minute later, and was pumping feedwater.
22 Because of a more rapid than desired cooldown, 23 . auxiliary feedwater flow was reduced roughly nine minutes into 24 the event down to zero for a very, very short period of time, 25 in an effort to control the more rapid than desirable
~
10 l 1 cooldown, and then auxiliary feedwater flow was increased to 2- 25 to 40 gpm range about ten minutes into the event.
3 Approximately ten minutes later the conditions existed, the 4 water hammer occurred. This~ water hammer ruptured a feedwater 5' bypass valve -- Mark will point that out later -- which leaked
- 6. steam, in effect, to'the secondary portion of the plant, the 7 turbine building. That was brought under control and the 8 plant cooldown was completed' roughly 11 hours1.273148e-4 days <br />0.00306 hours <br />1.818783e-5 weeks <br />4.1855e-6 months <br /> after the trip.
9 That is a very summary overview. Some of these 10 things we will touch on as'we go through the rest of the 11 presentation.
12 If there are no questions on this at this point, we 13 will proceed.
14 COMMISSIONER ZECH: The only question I have is on l
i 15 the check valves. You say you're replacing them all. .Do we 16 know what happened to the check valves that failed and are you 17 replacing with valves'you have confidence in that will not 18 fall? Have you looked at some root causes?
19 MR. BASKIN: We have looked at root causes. We 20 believe we know why they failed, and we are replacing them 21 with valves which should not be subject to the same problem.
22 Dr. Medford is going to discuss this in some detail in a 23 couple minutes, with pictures of the valves themselves.
24 COMMISSIONER ZECH: And tell us why he thinks the 25 new valves will not fail, and why the old ones did fail?
11 1 MR. BASKIN: That is correct.
2 COMMISSIONER ZECH: Thank you.
3 . CHAIRMAN PALLADINO: We will be interested in that
- 4. because, as I understand, you have identified that the check 5 valve failed due to the reduced flow, and I'm not sure that is 6 the only mode that they might have failed under.
7 MR. BASKIN:- As of today -- well, our later 8 evaluation and based on what we know today, that was a factor, 9 but there are other factors which we will discuss also, which 10 contributed significantly to the failure.
11 CHAIRMAN PALLADINO: Okay. We will also be 12 interested in whether the other valves that did not fail had 13 been tested further.
14 MR. BASKIN: We are prepared and we will discuss 15 that also.
16 CHAIRMAN PALLADINO: Okay. Thank you.
17 COMMISSIONER ASSELSTINE: Ken, as you go through the 18 individual items, too, if you could highlight where you are 19 now in the process in each one, that would be useful.
20 MR. BASKIN: Okay, we can do that.
21 Mr. Medford.
22 [ Slide.]
23 MR. MEDFORD: There will be three sections of my 24 discussion. The first section will be the feedwater system.
25 I will discuss some of the key design aspects of the system.
12 1 I will also discuss event-related aspects of the 2 feedwater system, and then I will discuss the corrective 3 actions associated with the feedwater system.
4 There will be a similar discussion for the 5 electrical power system. In this area I will also discuss 6 some of the maintenance aspects of the event.
7 I will conclude with a discussion or a summary of 8 the plant modifications which are being made as a result of 9 the event.
10 We consider that the IIT report was quite useful in 11 doing our study. We used it as a resource in our own 12 evaluation, and as we go through here, I think you will see 13 many of the issues that were raised in the IIT report.
14 The first area will be the feedwater system.
15 (Slide.]
16 This diagram looks rather busy and complicated.
17 However, the functioning of the feedwater system is basically l 18 quite simple. It takes flow from the. condensate system, it 19 divides it properly into the proper flow rates, and provides 20 that flow rate to each of the three steam generators which Ken 21 mentioned earlier.
i 22 [ Slide.]
j 23 on this slide I would like to go over some of the i
l 24 major features of the system.
25 Here we have the two main feedwater pumps. As you
. ~ _ _ _ _ _ _ _ _ . . . _ _ . _. - . .
~
13 1 may know, on San Onofre Unit 1, these are dual purpose pumps.
2 They serve not only as feed pumps, but also as safety 3 injection pumps.
4 Here we have the two pump discharge check valves. I 5 will- be talking more about those later. The system then 6 manifolds from two pumps down to a single line, and then 7 manifolds again back into three feed lines, one for each of 8 the steam generators.
9 This complex of valves here looks rather 10 complicated. I will go over it. They are essentially 11 identical. I will go over one set.
12 This top line on the diagram is the bypass line
'13 around the primary flow control valve. That bypass line is 14 used during startup when reduced flow is required.
15 There is the flow control valve itself, a discharge 16 check valve from the flow control valve, and a manual valve.
17 on the main line, which is used during normal 18 operation -- during normal operation, by the way, the bypass 19 line would be closed. On the main line there are four valves 20 here. There is a motor-operated valve, flow control valve, 21 feedwater system check valve, and finally, a manual valve.
22 This check valve I will also be mentioning more 23 about later.
24 I should point out to you the five valves which were 25 in a failed condition prior to the occurrence of the event.
o 14 1 Those five were the two pump discharge check valves and the 2 three main feedwater line check valves.
3 CHAIRMAN PALLADINO: One on each of the steam 4 generator loops?
5 MR. MEDFORD: Yes.
6 I would also like to point out to you that as a 7 result of the event, as a result of the water hammer, which 8 Ken mentioned earlier that occurred in feedwater line B, this 9 valve failed as a result of the water hammer. When I say 10 failed, it developed a leak.
11 CHAIRMAN PALLADINO: And then the line that ruptured 12 was off to the right, was it?
13 MR. MEDFORD: Early in the event, what happened is 14 the cause of the electrical alignment that existed when the 15 loss of power occurred, for the first 20 seconds of the event, 16 this feedwater pump was energized. This feedwater pump was 17 not. Because of the failed condition of this check valve, 18 flow came through this line and back-flowed through this pump, 19 and the rupture occurred in the condensate system.
20 CHAIRMAN PALLADINO: But there was also a rupture in
- 21. the steam line, wasn't there, off to the right?
22 MR. MEDFORD: Effectively the rupture in the steam 23 line was the failure of this check valve.
24 (Slide.)
25 This slide summarizes the key event-related aspects
15 1 for the feedwater system. As we mentioned earlier, we had 2 five failed check valves. During the event, as a result of 3- the pre-existing failure of those check valves, all three 4 feedwater lines voided.
5 During the event, during auxiliary'feedwater 6 refilling of these lines, there was a water hammer which 7 occurred in feedwater line B.
8 Now the reason why the water hammer occurred in B 9 and only in B is that at the time that auxiliary feedwater 10 flow was reinitiated at the lower level, lower flow rate, 11 lines A and C had already filled with water. So they were not 12 vulnerable at that point to water hammer.
13 However, it is worth noting that in a different set 14 of events, had they not been filled when aux feedwater was i 15 reinitiated, lines A and C would also be vulnerable to water 16 hammer.
17 So what we have done for line B, with one minor 18 exception we have done for the other lines as well, and I will 19 get to that later. -
20 (Slide.]
21 This slide and the next one address the primary 22 feedwater system-related failure. Again that is the failure 23 of the check valve.
24 Chairman Palladino, -you asked earlier about the 25 causes of failure. The two primary causes are turbulence in
16 1 the flow stream adjacent to the valve -- in the case of the 2 pump discharge check valve, that source is the pumps 3 themselves. In the case of the'feedline check valves, the 4 source is the flow control valve, which is very near to those 5 valves.
6 The other factor involved was the fact that the 7 orifices of these check valves was oversized for this 8 application. That resulted in flows through the valve that 9 are too low, and allowed the check valve disk to dangle in the 10 flow stream.
11 With the checb valve disk dangling in the flow 12 stream --
13 CHAIRMAN PALLADINO: If the orifice is too large, 14 wouldn't that allow more flow, or did I misunderstand you?
15 MR. MEDFORD: The velocities were too low. It 16 expands, the flow path expands, and the flow velocities are 17 too low.
18 As a result of.the disc dangling in the flow stream 19 in combination with the turbulence which provides excitation 20 forces on the disc, we believe that we had damage ~ occurring 21 over a period of time. These are listed as the primary 22 causes.
23 Now, you had mentioned earlier that reduced flow 24 operation was a factor. We consider it was a factor, but 25 frankly, the valves were not well designed for even 100
17 1 percent power operation. They were marginal. Because of this 2 undersizing of the orifices, they were marginal, even at 100 3 percent flow.
4 The effect that the 90 percent flow operation had on 5 this was that it revealed that weakness, and accelerated the 6 degradation of the valves associated with that weakness.
7 Another contributing factor is that the valves we 8 had used in this application had a two-part disc. I'll show 9 you on a later slide the two major parts of the valve disc 10 were the disc itself and the hinge arm.
11 What had occurred in these valves was disa'ssembly of-12 those two part5i.
13
~
CHAIRMAN PALLADINO: Why wasn't this noticed before 14 the incident? Is there no way of knowing that the valves are 15 deteriorating, or have?
16 MR. MEDFORD: As I'll discuss later, we had a way of 17 knowing, at least in one case -- one of the five valves. We 18 failed to properly recognize a precursor event that could have 19 allowed us to detect damage in that case.
20 Now, the IST program is not designed to detect 21 incipient check valve failure. The IST program normally 22 wouldn't tell you that you had this kind of degradation. What 23 it does is to tell you whether or not the valve is functioning 24 at the moment.
25 It doesn't -- it assumes, the IST program assumes
18 1 that_the valve is properly designed to begin with. I'll get 2 into that a bit more later.
3 [ Slide.]
4 This slide identifies the corrective actions for 5 this failure. As Ken indicated earlier, we're installing 6 prior to startup, valves of a new design. I'll discuss that 7 new design somewhat more later.
8 In addition, as a back-up to the valves that failed 9 during this event, we're going to -- Jack is pointing out to 10 you on this chart over, which is identical to the earlier 11 diagram -- is pointing out to you the flow control valves.
12 That's the flow control valve for line C, B, and A.
- 13. We are modifying the flow control valve to go 14 completely closed on a combined signal of loss of main feed 15 and aux feedwater injection. The purpose of that is that the 16 closed flow control valve acts as a back-up to the check 17 valves. The previous. design of the flow control valves would 18 have had them go to only 5' percent closed at this time.
19 Another modification which is also a back-up to the 20 failed check valves is this third one on the. slide. And that 21 is to add redundant feedline check valves.
22 In this case, I can't use the diagram to show you 23 the location. The check valves would be downstream of the 24 right-hand side of the diagram. They'd be very c]cse to the 25 steam generators themselves.
. .l 19 l 1- I ment'ioned earlier that there's one relatively 2 minor modification that'we're making to the B feedline only. '
3 And this is that one. We are putting a gradual slope in the B 14 .feedline, which has the effect'of reducing the length of the B 5 feedline.- The B feedline is the longest of the three by a 6 factor of two.
7 By providing this sloping, we effectively cut its 8 length in half, and thus reduce the probability in that line 9 to no greater than either of the other two.
10 COMMISSIONER ZECH: Have these modifications been 2
11 reviewed and approved by the Staff, our Staff?
'12 MR. MEDFORD: They have been presented to the 13 Staff. The NRC Staff has not completed their review.
14 COMMISSIONER ZECH: . All right. -
, 15 CHAIRMAN PALLADINO: What do-you mean, that you i
16 effectively cut the length in half? I don't follow.
17 MR. MEDFORD: The susceptibility to water hammer is I
18 determined by the length of the steam space in the line. By 19 providing a gradual -- the problem with a horizontal line as
- 20. far as feedwater goes,.is that the line tends to fill from the 21 bottom. And as you get closer and closer to the. top of the 22 line, that small steam space is what becomes vulnerable to 23 water hammer.
24 In tilting the line, we've provided for cutting that 25 space in half. At any one time that space would be half the '
~
20 1 length it was during the event.
2 CHAIRMAN PALLADINO: Are you sure this is the right 3- solution? I'm not saying it isn't, but you had a water hammer 4 problem before. You had another one. I'm anxious to make 5 sure that we don't see a third one at San Oncfre.
6 MR. MEDFORD: Let me say, we think this sloping is a 7 benefit. We don't claim that -- and reviewed any negative 8 aspects and are convinced there are no negative design aspects 9 of the sloping. We don't think it's the be all and end all of 10 eliminating water hammer in those lines.
11
~
The key to eliminating water hammer is to ensure 12 that they remain filled with water during the event. Our 13 focus is on the check valves themselves. This is something 14 which we think enhances the water hammer resistance of line B, 15 but it's not the sole solution by any means.
16 CHAIRMAN PALLADINO: Are you coming back to check 17 valves?
18 MR. MEDFORD: Yes, I am.
19 In addition, we will be providing for the check 20 valves outside containment. Essentially the five failed check 21 valves identified earlier. We will be providing as part of 22 startup, a test to determine what the position of the disc is 23 during high flow operation, 70, 80 and 90 percent flow 24 operation.
25 We have not fully developed this test yet. One of
o 21 1 the things we're considering is acoustic monitoring as a t' col 2 to determine the position of the disc.
3 COMMISSIONER ZECH: Do you intend to do that test i
4 just after the initial startup, or after every startup from 5 now on?
6 MR. MEDFORD: It depends on the results of-the test, 7 frankly. It depends on how effective the test is. If the 8 . test is effe:tive, it might well become part of our standard 9 startup procram. We haven't gotten that far in our planning.
10 COMMISSIONER ZECH: All right. But you're ,
11 considering it?
i 12 MR. MEDFORD: Yes.
13 CHAIRMAN PALLADINO: Are you planning to test the 14 other valves that didn't fail, the other check valves?
15 MR. MEDFORD: Yes. In fact, this program down here,
- 16 the bottom' bullet, that entails disassembly and inspection of 17 all of the Pacific valves. That's the valve design that 18 failed in this case.
19 We're also doing a design review of all swing ch'eck 20 ' valves to determine what position the disc is in in relatively 21 high flow operation, 70 and 80 and 90 percent flow operation, .
22 to determine if we've got the same problem we experienced here 23 in other valves.
24 CHAIRMAN PALLADINO: Do you plan to test the 25 non-Pacific valves?
1 e.-r-.-w_ _-. ,.,w_m ._ - ,-,-.m.__._ -,,_ . , _ , _ , , _ , , , , . _ , . _ _ _ , . . - _ _ , . , , ,,,_,m__.,, ,,
o 22 1 MR. MEDFORD: We haven't -- in terms of the acoustic 2 monitoring that I talked about? We don't have that in our 3 planning right now. ~ We don't think .tJust it's necessary. . We 4 think we can determine analytically whether or not the valves 5 are susceptible to this type of failure.
6 CHAIRMAN PALLADINO: What makes you' convinced you -
7 can do that? I'm just curious to why a test program wouldn't i
'8 include those that didn't fail, as well as those that failed, 9 because, the reason you didn't identify those that failed was 10 because there was no test before they failed. And here we ,
11 have some that didn't fail.
i 12 MR. BASKIN: Let me see if I can clarify this. One 13 of the problems that we're facing is finding a good way to 14 test these check valves effectively in service, when there's i
15 flow. We're looking at acoustic, various sonic devices.
16 We're looking at energy balance type things, and are going to 17 apply whatever we can to the key check valves.
18 If we find a way that we think provides good 19 results, we will use that same method in selected valves, at
! 20 least, for other parts of the system.
21 COMMISSIONER ASSELSTINE: How many of the check 22 valves when you start up would be new check valves of the new 23 design? And how many would be older check valves that have i
24 been in the plant for some time? !
25 MR. MEDFORD: We have approximately 150 swing check
.--.-._.--.y7_ , _ _ - ,,,r _,,,_,,___..-.c . . , , , , , , , _ . _ ,m,.,.m __,., -.. . - . . , , . . -
d a
23 1 valves. And we are replacing -- well, we're adding three new 2 check valves, which I'll mention later. And we're replacing 3 eight valves. So approximately eight of the 150 existing at 4 the time of the event will be replaced.
5 Next slide, please.
6 (Slide.]
- 7 CHAIRMAN PALLADINO: I don't -- I'm sorry, go ahead.
8 COMMISSIONER ASSELSTINE: Well, I guess my follow-up 9 question was, what precisely are you doing for the other 142 10 before you want to start the plant up again to assure 11 yourselves of the condition of those check valves before 12 resuming oeprations?
13 MR. MEDFORD: Okay, we are analyzing the va'lves to 14 . determine what the position of the disc is during high flow 15 operation. And where there's any indication as a result of 16 that analysis that there's a problem, we're going to go in and 17 inspect the valves.
18 Now I mentioned we're also inspecting all of the 19 Pacific valves, and I don't have the number, but that's 20 substantially more than the eight here.
21 COMMISSIONER ASSELSTINE: Have you looked at those 22 conditions that you feel are the principal contributors to the 23 failure of these specific -- the five check valves -- to 24 determine if those kinds of flow conditions, sizing conditions 25 exist in these other 142?
. s 24 1 MR. MEDFORD: That's~ precisely the nature of the 2 analysis.
3 COMMISSIONER ASSELSTINE: Okay. And how far along 4 are you? Do you have an idea for how many are susceptible to 5 the same kinds of design vulnerabilities that the five that 6 failed were?
7 MR. MEDFORD: At this point I don't have an answer 8 to that question. By the way, that program will be completed 9 prior to startup.
10 COMMISSIONER ASSELSTINE: If you find those 11 conditions at however many of these other 142 there are, do 12 you know specifically what you're going to be doing in terms 13 of looking at those valves? Are you going to tear them 14 apart? Or what are you going to do to make sure that the 15 degradation hasn't gotten to the. point where you might see a 16 failure?
17 MR. MEDFORD: Any of those valves that indicate 18 they're susceptible to the same mode of failure we experienced 19 here would be disassembled. And that disassembly would be 20 completed prior to restart.
- 21 MR. BASKIN
- Commissioner Asselstine, just so you're 22 up to date, we about 90 percent complete with the analysis and 23 inspection of the.other valves. We've found some problems.
24 I think it's worth noting,_too, that when we talk
! 25 about 150 check valves -- I don't have the exact number -- but
, 25 1 certainly a lot of them'are not safety-related.. And a lot of 2 them are what you'd consider miscellaneous type valves in 3 little drain liner, and that sort of thing.
4 So ther'a's probably, off the top of my head, half 5 that number that are really significant to operation.
6 However, we're do!.ng.it for all.
7 COMMISSIGNER ASSELSTINE: Okay. Were these five
. 8 valves categorized as safety-related, the five that failed or 9 not?
. 10 MR. MEDFORD: Yes, they were.
11 COMMISSIONER ASSELSTINE: They were all 12 safety-related?
13 MR. MEDFORD: Yes, they were.
4 14 MR. BASKIN: At the time they failed, they were. At 4
15 the time the plant was initially built, they weren't. And it i
16 was through'the transition process that they ultimately were.-
17 CHAIRMAN PALLADINO: Isn't it possible to tell 1
18 whether a check valve is functioning by an appropriate 19 hydrostatic test?
20 MR. MEDFORD: Yes. And IST does that. It 21 determines whether or not the check valve is functioning.
22 The problem we had here, is we have a degradation of 23 the valve over time. And if you're, you know, part way along 24 in that degradation process normal IST of check valves will 25 not detect that failure.
I
_ . _ _ . _ _ _ . . _ _ _ , , _ . , . , ..,___,,..__m._, , _ . . . - , _ , _ . , , . . , , , - _ _ . . . . _ , . _ _ , , ,-_.,,_..r__._ _ _ . , . - .
26 ,
i 1 -
CHAIRMAN PALLADINO: They will not what? !
2 MR. MEDFORD: -Normal in-service testing of check 3 valves would not detect a failure, unless you'd already 4 reached the point in the degradation where the valve had 5 completely failed.
h
!T 6 M' R . BASKIN: Why don't you move to the picture, 7 Mark, and I think that will become clearer.
8 CHAIRMAN PALLADINO: Well, in response to 9 . Commissioner Zech, he asked, are these tests going to be done 10 annually. And I don't remember what you said.
11 MR. MEDFORD: What I indicated is, we have not fully '
12 developed this test yet. We're looking at a number of things; 13 acoustic monitoring is one of them. Depending on how 14 successful we are in identifying a good test, that may well i 15 become part of our procedures for starting up from a 16 refueling.
t.
17 CHAIRMAN PALLADINO: I think he and I, and the 18 others may also share the same concern, that if you got to 19 this by not testing over a period of time, if we don't test on 20 .some periodic basis, how will we know that we're not getting
! 21 degradation again?
22 MR. MEDFORD: I think that's a good point.
t 23 MR. BASKIN: Again, let me make sure we're clear 24 here. The normal in-service testing program is done on all 25 valves on a set periodic schedule. That detects valve L
1
.s 27
~
1 tightness, but doesn't detect incipient failure.
2 If we're able to come up with a reasonable, accurate 3 o
-- reas'nably accurate way to test these sorts of valves at 4 their open, or.whatever position they're in, we will use it 5 not only on these valves, but on other valves periodically, 6 not just for startup' testing. And that's only common sense 7 from our standpoint aside from anything else.
8 So it will be done, if we can find a good way to do 9 it. The problem is, it's-hard to do it.
10 CHAIRMAN PALLADINO: Well, hard though it may be, we 11 got here by not knowing the condition of the valve. So I 12 think there's a very strong need to find out some way of 13 telling whether they're degrading or not.
14 MR. BASKIN: And if we don't find a good way o'f~
15
~
testing them, the other thing we can do and will do, is 16 physically take them apart and inspect.them.
17 CHAIRMAN PALLADINO: Periodically.
18 MR. BASKIN: Yes.
19 CHAIRMAN PALLADINO: I think we share a common 20 concern. You know, when you get five check valves failing, 21 you know, our single failure criterion goes out the window.
22 And this concept of defense in depth is impaired.
23 So I think these are very important things to give 24 attention to. And I'know you said that, but we're pressing to 25 make sure that it's on a periodic basis.
o
,- 28 1 . MR. BASKIN: Yes.
2 MR. MEDFORD: I should point out, one of the 3 corrective actions we have listed on the previous slide is 4 that we will have a valve inspection at the next refueling 5 outage as part of this program.
6 We. share your concern about the need to ensure on a 7 continuing basis the operability of these valves. Some i 8 combination of inspection and testing will do that.
9 CHAIRMAN PALIADINO: Okay, thank you.
10 COMMISSIONER ASSELSTINE: Do you know how extensive 11 that inspection program will be at the next outage? r
. 12 MR. MEDFORD: It's not been fully defined yet. I 13 would say, a minimal program would be disassembly of one of
. 14 each application of the valves. That is, for example, one of 15 the two pump discharge check valves would be inspected. One i
! 16 of the feedline check valves would be inspected. One of the 17 redundant valves would be inspected.
18 But frankly, we aren't far enough along in our 19 thinking to define that precisely. We may decide to inspect 20 all valves.
21 COMMISSIONER ASSELSTINE: Okay. How are you going 22 to verify the validity of the testing approaches that you're 23 looking at? How are you going to know whether you've got a 24 reliable test or not?
25 Again, presumably you don't want to wait for
,..,-------r---- -, e, , - - - - - - - - - - - - - - - , - . ,---,p.---
i
+*
29 l 1 1 failures to assure yourself that'the testing program is or is 2 not working.
3 MR. MEDFORD: That's correct. I think, you know, 4 the nature of the.information -- assuming, for example, we do 4
5 an acoustic test -- I think the nature of the information that
, 6 we derive'from the test, it will either be ambiguous, or we 4
7 will be able to determine clearly that we know the position of
, 8 the valve disc.
9 MR. BASKIN: I think maybe there's two aspects of
- 10 this. And maybe we're not making it clear.
. 11 There's the aspect of making sure the valve disc is 12 out of the flow path and isn't being subjected to the J 13 turbulence and vibration. And that, of necessity, has to be 14 done by some external device, acoustics or something like 4
15 that.
1 4
16 COMMISSIONER ASSELSTINE: Right.'
17 MR. BASKIN:- And we're looking for and will find the i
18 best way to.do it.
f 19 In addition, you want to verify that whatever c 20 position a valve is in, it's not being degraded. At least t
21 initially, that will be done by valve inspection. And j 22 depending on the results of that,.we'll set our future 23 programs for more or less inspections as the years go by.
1 l 24 COMMISSIONER ~ASSELSTINE: Okay.
25 CHAIRMAN PALLADINO: Are you looking at the Nuclear r
- , , . . - - - . , - . - - - . . .,,..,,,-.-,_,.,,,.--,.-,_,..n-, , , . - - , - . . , . , - - , _ . . - - . ,
o 30 1 Plant Reliability Data System information to see if any of 2 that'information gives you a clue as to problems with check 3 valves?
4 MR. MEDFORD: Let me tell you how we used NPRDS in 5 the selection of-these valves.
6 -
We had identified a certain mode of failure and 7 certain application problems associated with these valves.
8 What we did first was to identify valves which would not be 9 susceptible to this same mode of failure.
i
- 10 Having done that, we then used NPRDS as a check to 11 determine whether or not there's any information in NPRDS 12 which would tell us that use of these valves is. inappropriate.
4 13 We looked at NPRDS for the replacement, the 14 Atwood-Morrill valves. And for the specific type of 15 Atwood-Morrill valve that we're using here, we found that the 16 failure rate was substantially less than the industry average..
17 We've also looked at the Atwood-Morrill valve data 18 on a gross basis. On a gross basis, the Atwood-Morrill valves 19 don't look so good, frankly. But we were more interested in 20 the. specific valves we're using in this application, j 21 COMMISSIONER ASSELSTINE: Did NPRDS tell you 22 anything about the Pacific valves?
23 MR. MEDFORD: NPRDS data indicates that the Pacific
[ 24 valves have a very low failure rate.
i 25 But again -- one thing, I don't want to spend a lot 4
4
- , , - - - -- m-n. - - - , . - - , - , - ,,.,,--n--.-n.. . , , - - , . . - ---
31 1 of time on this slide. And let me mention a couple of things 2 about it. One is, that we're not trying to say that Pacific 3 valves are bad and Atwood-Morrill valves are good.
4 Rather, what this slide is intending to say is, that 5' Pacific valves -- or the Pacific valves we had at least in 6 this application -- were not appropriate for the application.
7 The Atwood-Morrill valves that we're using as a replacement 8 are appropriate for the application.
9 COMMISSIONER ASSELSTINE: That's interesting 10 though. Here you've got NPRDS will give you sort of gross 11 information on the performance of certain valve types. In 12 this case, it says Pacific valves in general are good; 13 Atwood-Morrill valves have a higher failure rate. And yet, 14 the specific application is what really controls.
15 MR. MEDFORD: That's correct. I think that's one 16 thing that we've learned, or at least relearned as a result of 17 this event. That you've got to be very careful.
18 We think NPRDS is useful. But you cannot use it in
- 19 a cavalier manner. You have to look very carefully at the l 20 data.
i
! 21 COMMISSIONER BERNTHAL: Let me ask a dumb question.
22 I'm not a valve expert, but how many of these things, this 23 type of valve are in the plant total?
i 24 MR. MEDFORD: The total swing check valves, about i
l 25 150.
l t
I
" 32 1 COMMISSIONER BERNTHAL: In this particular line 2 there were five. Is it a fact that any one of those five 3 could have prevented the problem had it functioned properly?
'4 Or does it require several of all of them to --
5 MR. MEDFORD: I'm sorry. No, it requires failure of 6 two. It' requires failure of either one of those two valves, 7 and one of those three.
8 COMMISSIONER BERNTHAL: Okay, I see the diagram. I 9 guess my question is -- and maybe there's no place where you 10 have a straight line sequence, or at least parallel sequence 11 like this. If these are really such dicey valves -- and maybe 12 with this new one there isn't that question of performance --
., 13 why don't you just stick another valve in the line somewhere?
14 MR. MEDFORD: That's one of the corrective actions.
15 COMMISSIONER BERNTHAL: You're going to do that?
16 MR. MEDFORD: Yes, we are. In fact, those are ones 17 that I mentioned. They're off the diagram. Unfortunately, we 18 can't show you where they are, but they are -- those lines go
- 19 to the steam generators. And as you get very close to the i
l 20 steam generators, that will be the location of the new check 21 valves.
22 (Slide.]
23 I'll quickly cover this slide. There are two 24 key points again about -- not the valves themselves, but the i
25 application of the valves for this service. The old valves
sw J a - - -
i'-
33 1 were oversized relative to the application. They have, as 2 I'll show on a future slide, a two piece disc hinge assembly,
~
3 which is more vulnerable to vibration induced failure than a 4 single piece assembly is.
4 5 The new valves are sized to remainLfully open 6 down to 70 percent flow. And they consist of a. single piece 7 disc hinge assembly.
8 COMMISSIONER ZECH: You know, they're oversized --
4 9 did you know they were oversized, or did we have to learn it 10 the hard way?
3 11- MR. MEDFORD: We learned it the hard way.
~
12 COMMISSIONER ZECH: 'You simply didn't realize they l
13- were oversized beforehand?
14 MR. MEDFORD: That's correct. And we're
-15 investigating, as is the NRC, the design process.that led to 16 that. I can't give you a definitive answer today on exactly 17 how that happened to be.
18 COMMISSIONER ZECH: Well, I hope we're looking to i
19 see whether other plants -- it's applicable to other plants, 20 too, around the country.
21 MR. MEDFORD: I believe that the NRC staff is 22 looking at the design practices of all AE vendors.
! 23 COMMISSIONER ZECH: I understand they're doing 24 that. Thank you.
25 COMMISSIONER ASSELSTINE: That's a part of the AE f .
I
--. .- .__..__.___.,__,m.- , . . . . . _ _ . . _ . . - _ - _ . . _ _ _ _ _ - - - _ . _ _ _ - _ - . - _ _ _ - ~ . .
34 1 design effort; is that correct?
2- MR. MEDFORD: Yes, that's correct.
- 3
[ Slide.]
4 I've shown on this slide a typical Pacific valve -- '
5 and again, that's the valve that failed -- disc hinge 6 assembly. The two major parts are the hinge arm itself and 7 the disc. They're connected by'a threaded fastener. I'll
- 8 mention a bit about these design aspects later.
9 These are anti-rotation lugs. And this was a 10 modification on San Onofre Unit 1. Their purpose is to 11 prevent rotation of the' disc around the threaded fastener and 4
12 disassembly in that manner.
13 [ Slide.]
14 This slide compares the old Pacific valves with the 15 new Atwood-Morrill valves pictorially. I'd like to point out 16 on the Pacific valve that when the valve go the fully opened 17 position, the impact point is the threaded fastener. The 18 i
threaded fastener which is keeping the two parts together.
19 i [ Commissioner Roberts left the room.]
20 MR. MEDFORD:
1 The threaded fastener which failed in 21 this case.
22 On the Atwood-Morrill valve --
23 CHAIRMAN PALLADINO: When you say the threaded
! 24 fastener failed, you mean the bolt broke?
25 MR. MEDFORD: There were varying -- in the five I
i l
t - - _ _ _
35 1 valves there were varying mechanisms of failure, but that was 2 one of the mechanisms that was exhibited. In all cases, the 3 threaded fastener showed signs of distress.
4 MR. FOGARTY: In some cases it actually unscrewed.
5 COMMISSIONER ASSELSTINE: The nut unscrewed?
6 MR. FOGARTY: Yes, or stripped.
7 CHAIRMAN PALLADINO: Don't they use.what we used to 8 call Class A locking devices so that they can't do that?
9 MR.-MEDFORD: The three feedline check valves, in 10 this case, had a pin-threaded fastener. In some cases, those 11 had failed as well.
12 COMMISSIONER BERNTHAL: Why is it such a --
1 '13 CHAIRMAN PALLADINO: Class A also captures them.
14 But that's okay, go ahead.
15 COMMISSIONER BERNTHAL: Why is it such a major 16 operation to pull that top plate off and just look at the 17 crazy thing during a maintenance period?
18 MR. MEDFORD: It depends on the design of the 19 specific valve. Unfortunately, one of the unfortunate' aspects 20 of the Atwood-Morrill design is that it's a relatively 21 expensive process to remove this. A lot depends on the nature 22 of the gasketing design.
- 23 The Pacific valves are somewhat easier to get at the 24 internals.
25 COMMISSIONER BERNTHAL: I guess it's just hard to k
I
- - . , ._ _ . _ . , - - - _ . , _ _ _ . _ _ _.,m , .,, ,_-,,____ -- _,..____, , ,_, ,,.-_- ,--_ _ _ . _ . . . _ _ _ - _ _ _ . . ,
~
36 1 understand why that wasn't done occasionally.
2 MR. MEDFORD: In retrospect, we wish that we had.
3 Had we known of the problem --
4 COMMISSIONER BERNTHAL: Does this look easier than 5 it is? I mean, it looks like you remove some bolts, pull it 6 off, repack it and put it back on.
7 MR. BASKIN: It was done, Commissioner Bernthal.
8 And I'm getting a little bit ahead of the presentation, but 9 when we saw distress in some of these valves roughly in the 10 mid '70s, we made some changes to them -- the lugs Mark talked 11 about -- some of the pins.
12 And then for some period of time after that, one 13 refueling or two, we then did pull the cover and inspect 14 them. So it's something that you can do. It's not simple, but 15 it's not tremendously difficult.
16 I think part of the problem is, you know, at that 17 time, without having a specific concern about these valves, 18 it's not something you routinely do other than on some 19 relatively infrequent basis. But we did do some of that.
20 COMMISSIONER BERNTHAL: When was the last time you 21 pulled one of them?
22 MR. BASKIN: That would have been late '70s, early 23 '80s, as I recall. Sometime around 1980, plus or minus.
24 COMMISSIONER BERNTHAL: Five years, roughly.
25 CHAIRMAN PALLADINO: Or longer.
37 1 MR. PAPAY: But the plant was out of operation.
1 2 MR. BASKIN: Yes, we didn't have much operation, 3 unfortunately in that period of time. '
4 COMMISSIONER BERNTHAL: Of course, that doesn't 5 necessarily help. That can even be worse.
6 (Commissioner Roberts reentered the room.)
7 ~ MR.'BASKIN: Yes.
8 CHAIRMAN PALLADINO: Wera you going to describe the
- 9 Atwood-Morrill valve design? I didn't understand the picture j 10 in the handout.
11 MR. MEDFORD: Let's go back. Okay, the advantage of 12 the Atwood-Morrill valve is that it has a single piece disc.
13 And if you compare the two pictures, the disc itself on the 14 Atwood-Morrill valve design is more substantial than the 15 Pacific valve.
1 16 Also, the impact point for the valve when it goes to 17 the fully open position is taken up on the single large, 18 massive single piece itself, compared again with the threaded -
i 19 fastener for the Pacific valve.
20 CHAIRMAN PALLADINO: And what am I supposed to get 21 from the right-hand view?
22 MR. FOGARTY: That's down the line.
23 CHAIRMAN PALLADINO: That's what?
24 COMMISSIONER ZECH: That's looking down the line.
25 MR. MEDFORD: Looking down the line.
-- - y=w. - - - - ,_, . - . - . - ,_,,,-.,_.m . . . _ _ _ . --- - - .,r _ , _ . - _ - - , _ -
38 l
- 1 CMAIRMAN PALLADINO: Yes, but what I am -- _,
l 2 MR. MEDFORD: It shows the valve in the fully closed 3 position.
4 MR. BASKIN: It shows that it's a single piece.
5 That the stem and the valve cover is a single piece.-
6 CHAIRMAN PALLADINO: Thank you.-
7 [ Slide.)
8 MR.-MEDFORD: A secondary failure in the feedwater 4
9 system area, again, related to the check valves themselves, 10 was our failure to recognize precursor events. In June of l
11 1985 a noise was heard in the B feedline. And we now believe 12 that that noise was associated with the progressive failure of
, 13 the valve. Probably the va'lve had not failed at that point in 14 time.
15 CHAIRMAN PALLADINO: Where is that, in one of these 16 lines over here?
i 17 MR. MEDFORD: Yes, that line right there 18 '[ indicating.]
19 CHAIRMAN PALLADINO: I see.
20 MR. MEDFORD: We assessed the safety significance of l 21 that noise. We tried to diagnose what the cause of the noise 22 was. Unfortunately, we misdiagnosed it. We believed that the
~
23 failure was not in the check valve, but rather was in the 24 manual valve, because that's where the noise appeared to be 25 coming from.
i
, . , _ - . _ . _ _-- ,,_--,e,._ m. ..m.,_-.._,.,_m.__. ,_,mm - . _ - - _ - - . _ _ , , , . , _ , _ _ _ .--,_-_-_- _
.. l 39 1 However, in retrospect, given the maintenance 2 history of these valves, Mr. Baskin mentioned earlier that 3 we've had several problems associated with these valves during 4 the 1970s. They'd been observed during the application of the 5 maintenance. We had made modifications to the valves. I i
6 mentioned the lugs earlier, and the pins. We thought we had 7 corrected that problem.
8 Frankly, we didn't consider that information as 9 thoroughly-as we should have.
t 10 CHAIRMAN PALLADINO: When was this noise heard?
11 COMMISSIONER ASSELSTINE: June of '85.
12 MR. FOGARTY: June of '85.
13 COMMISSIONER ASSELSTINE: How far downstream is that 14 manual valve from the check valve?
15 MR. MEDFORD: It's very close. On the order of a 16 foot or two.
17 COMMISSIONER ASSELSTINE: Okay.
18 CHAIRMAN PALLADINO: When you heard the noise, did 19 you consider taking off the top and looking at it?
20 MR. MEDFORD: Yes, we did. We did radiograph it.
21 Didn't get any indication from the radiograph that there was a
- 22 problem. Although frankly, the radiograph was not a great use 23 as a test in this case.
24 COMMISSIONER ASSELSTINE: I take it the plant was 25 operating during that time period?
l
40 1 MR. MEDFORD: Yes, it was.
2 COMMISSIONER ASSELSTINE: Which would have required 3 a shutdown to take the valves apart?
4 MR. MEDFORD: Yes, it would have.
5 MR. FOGARTY: Just to elaborate a little further on 6 that. It's clear that of all the-events that took place, this 7 is the one where we have done more second-guessing of 8 ourselves in-house than any other area.
9 The plant was in distress. There was a' problem 10 there. We determined there was a noise in the line. We made
~ 11 a very thorough analysis of it, without shutting the plant 12 down. There were reviews by the on-site review committee and 13 other people in-house. The conclusion was that.it was in the 14 block valve. And the concern was more to be sure that no part 15 of it carried over into the steam generator to cause 16 mechanica3 damage there.
17 But if there's'one thing, above all others, in terms 18 of 20/20 hindsight, it's this one. The plant should have been 19 shut down in June. The valves should have been opened up, ara 20 we would have found the difficulty.
21 So I just want to be clear on that, that this is a 22 major problem. We have a recommendation that we hope will 23 address that point that Mr. Baskin will make later on. But if 24 we had to do it over again, that is where we clearly. goofed in 25 our review of what was going on in the plant.
.. ' 41.
1 CHAIRMAN PALLADINO: All right, thank you.
2 MR. MEDFORD: One of the primary causes of this 3 failure was the lack of readily available maintenance 4 information. I should point out that we, as a result of this 5 . event, reviewed the adequacy of our maintenance programs 6 today. Not only the performance of maintenance, but record 7- retention and trending.
1 8 We think the current program is very good. The 9 information that existed during the period of primary interest 10 here, during the mid and late '70s, was not as good.
I i 11 CHAIRMAN PALLADINO: What kind of maintenance do you
- 12 do to a check valve?
13 MR. MEDFORD: They're periodically disassembled and 14 inspected and refurbished; replacement of pins.
1 15 CHAIRMAN PALLADINO: How frequently? If I 1
16 understood --
! 17 COMMISSIONER ASSELSTINE: Apparently not for five 18 years.
i 19 CHAIRMAN PALLADINO: .Yes, '79 or '80. That's six or 20 seven years ago.
21 MR. BASKIN: Without any indications of problems, i
22 it's not done frequently. It's done over a multi-year period.
23 COMMISSIONER ASSELSTINE: Are you going back to look J
24 at the whole in-service testing and preventive and predictive
- 25 maintenance program to try and identify those areas where you
j* 42 1 may have some questions or uncertainties about your ability to 2 identify this kind of a degradation, whether it's in valves or
- 3 other areas?
3 4 (Commissioner Bernthal left the room.]
4 5 COMMISSIONER ASSELSTINE: And where you've got t
6 questions about your ability to do that when the plant is 1
7 operating, supplementing that with maybe a more frequent 8 periodic maintenance program like the Japanese do. I gather, 9 where you tear things down, see what the actual physical J
10 condition is, then put them back together again. Does this 11 raise systematic questions that you need to look at?
12 MR. BASKIN: Yes, we are doing that in light of this #
13 to, as you say, Commissioner Asselstine, to look at those 14 aspects where detecting stress -- if.I can use it that way --
15 is a little bit more difficult.
i 1 16 And I should point out -- Mr. Krieger just refreshed l
17 my memory -- the preventive maintenance program as it exists 18 today has us looking at all check valves at least every five 19 years. That's the maximum time. In some cases it's shorter.
1 20 COMMISSIONER ASSELSTINE: Okay. It does seem'to me l 21 that one of the evolving areas is this question of your 22 ability while the plant's operating to detect that kind of' 23 degradation.
24 In some areas, we're making pretty strong advances.
25 But in others, where the technology, I think, is still i
i
- - - , .. . - . , - , - - . - , , , - . , - - - - . - , - - . . , - . - - . . - - - . - , - , , - - - . - - - - . - - - . . - - . ~ . - . . . .
.+ 43 1 questionable, you still have to back that up with a pretty 2 aggressive physical inspection program.
3 MR. BASKIN: That's true. And again, what we're 4 doing.today in maintenance in general in terms of trending and 5 tracking is worlds apart from what we did in the early '70s.
6 I mean, you know,'it's just a different ballgame today.
7 COMMISSIONER ASSELSTINE: yes.
8 CHAIRMAN PALLADINO: _Did I understand you to say 9 that you're going to, roughly every five years, open up every 10 one of these check valves and take a look at them? Or did I 11 misunderstand?
12 MR. BASKIN: No, that's correct. The normal 13 maintenance program that exists today has us opening check 14 valves at least once every five years. That would the minimum 15 amount for all valves.
16 Now obviously, for these valves that we're talking 17 about today, certainly initially that would happen much more 18 often.
19 CHAIRMAN PALLADINO: Do you do them all at once?
20 MR. BASKIN: No, it's periodically spread out.
21 CHAIRMAN PALLADINO: Okay.
22 [ Slide.)
23 MR. MEDFORD: There are two corrective actions for 24 this failure. One, is that we're going to incorporate into 25 our review of abnormal condition, a more systematic search of
f
- 44 1 the plant history relative to that condition. ,
2 Secondly, we're going to enhance our on-site safety I 3 committee review process, by the addition of a member from
-4 outside the company. Now, the kind of person we're looking 5 for is somebody with senior operating experience from the 6 utility industry. ,
7 The purpose of this, again, is to broaden our 8 operating experience base, not so much to provide specific 9 technical expertise.
10 COMMISSIONER ASSELSTINE: On that first bullet, 11 apart from the noise in June of '85, do you think the early 12- history was sufficient so that there was enough information
'13 that looking back you can say, gee, we should have been tipped 14 off by this, that we needed to do more with these valves.
15 I guess, what I'm wondering about is, when you do 16 that historical look, is it going to give you enough 17 information so that you can spot the next one. It probably 18 isn't going to be check valves, it's going to be something .
19 else. Is that going to give you enough warning so that you 20 can spot the next one?
21 MR. MEDFORD: The combination of the noise 22 observation in June with that early history, I think, would do 23 that.
24 The history itself, we were of the opinion, firm 25 technical opinion, that we had solved the problem. So the-
f 45 1 history by itself wouldn't have d;ne it. But I thInk the even 2 combined with that would have.
3 Next slide, please.
4 (Slide.]
5 MR. MEDFORD: Now I'm going to move on to my second 6 topic which is the electrical power system. I won't belabor 7 this slide. This slide deals with several of the major 8 features of the electrical power system for the plant.
9 Let me focus on just one aspect of the slide. And 10 that is, the diesel generator capability. The plant has two 11 six-megawatt diesels. They're sized to provide power 12 sufficient for all engineered safety features. Their design 13 provides for automatic loading upon an emergency core cooling 14 requirement.
15 For other transients to plant safety, analysis 16 demonstrates that we have sufficient time for manual loading.
17 And for those transients, the design provides for manual 18 loading.
19 CHAIRMAN PALLADINO: But why wasn't automatic 20 loading considered?
21 MR. MEDFORD: It had been considered. The 22 preference in the plant design had been for restoration of 23 off-site power. And upon loss of the primary immediately 24 available source of off-site power, the preference would be to 25 go to the second source of off-site power, which is a
f
- 46 1 combination of automatic and manual actions, aqd then to go to 2 the diesels.
3 The diesels, by the way, do automatically start.
4 4 They don't automatically -- they automatically start on loss 5 of off-site power. They don't automatically load.
6 As a result of this event, we are considering ,
7 providing for automatic loading of the diesels.
~
8 CHAIRMAN PALLADINO: Would this situation have been i
9 helped by automatic loading?
10 MR. MEDFORD: No. ,
11 COMMISSIONER ASSELSTINE: No.
12 CHAIRMAN PALLADINO: It would not?
13 MR. MEDFORD: No. It would have changed the time of 14 occurrence of certain things. But the failure itself, would i
15 still have occurred, and the consequences would have been just 16 as severe.
. 17 COMMISSIONER ASSELSTINE: Is that just a difference 18 in design philosophy at the time that the plant was designed?
19 For example, I assume Units 2 and 3 are automatically loaded.
20 MR. MEDFORD: That's correct. And your earlier 21 statement is correct,-too. It's just a difference in design 22 philosophy.
23 CHAIRMAN PALLADINO: You say you are reconsidering 24 automatic loading?
25 MR. MEDFORD: Yes, we are. It's not a simple
47 1 modification, frankly. And it involves some other safety 2 ramifications. We want to proceed down that path in a very 3 thoug'htful manner. We have not come -- decided one way or the 4 other which way we're going to go.
5 COMMISSIONER ASSELSTINE: What were the reasons why, 6 at least at the time, acving to the delayed off-site source 7 was viewed as the favorable approach to simply going to the 8 diesels?
9 MR. MEDFORD: By the way, I think it's fair to say, 10 that's not just an Edison position. It's an industry position 11 and an NRC position.
12 For one thing, the off-site power is the permanent, 13 long term source of power available. In the specific case of 14 Edison, we've got an extremely reliable grid.
15 A third reason is,'that once you go on the diesels, 16 you have to go through certain manipulations to get back off 17 the diesels and on to on-site power.
18 MR. FOGARTY: There was one other factor, just going 19 back in history. At the time of the license for Unit number 20 1, a great deal of reliance was placed on the fact that on our 21 system we could isolate all of our transmission lines in a 22 manner such that San onofre could have been fed directly from 23 the Hoover generators, which were considered a very reliable 24 source of power.
25 So again, that's just part of the rationale that the
,. - l 4
. 48 l 1 off-site power was extremely reliable and would be~available 2 all of the time, with this line dropping system we have.
3 MR. MEDFORD: This line addresses the primary 4 electrical failure. That was the failure of the four KV 5 feeder cable from aux transformer C. The primary cause of 6 that failure was age-related degradation of the cable.
7 Now there was a contributing factor. And that is, 8 that there was a localized heat source. Specifically, an 9 uninsulated feedwater system flange in the vicinity of the 10 cable failure.
11 But again, we think the primary cause was 12 age-related degradation.
13 (Slide.)
14 MR. MEDFORD: This slide depicts the corrective 15 action for that failure. First, prior to startup, we're going 16 to test all four KV cabling. And I can't quantify the 17 experience to date, but frankly, we've had a mixed i 18 experience. We've seen some other cabling which did not pass 19 a high potential test.
20 Second, we're replacing, independent of test 21 experience, we're replacing all of the four KV feeder cabling 22 from the transformers to their buses.
23 Third, we've got a program to protect four KV 24 cabling from localized heat tources.
25 Mr. Baskin mentioned earlier that we're working with
.- 49 1 Idaho National Engineering Laboratory on a state-of-the-art i.
2 system, which would provide for a continuing monitoring 3 program for all cabling, not just the four KV' cabling.
4 Finally, this age-related cable failure calls to 5 question, in our mind, the material condition of the remainder 6 of the plant. We think the plant is in good condition. We 7 . don't think this is representative of what we'll find in the 8 remainder of the plant. But we do think it indicates that a 9 look is required.
10 Mr. Krieger is going to address two separate 11 programs later on. One is addressed at reaffirming the 12 baseline condition of the plant. That will be conducted prior 13 to startup.
14 The second program is designed at assuring, on a 15 continuing basis, the material adequacy of the plant.
16 COMMISSIONER ASSELSTINE: Did this come as a 17 surprise to you, the failure of the cable? Or was this 18 something that was assumed, that the cabling would last 19 throughout the life of the plant without requiring 20 replacement?
21 MR. MEDFORD: Yes. The capability of the cabling 22 was assessed during the initial design of the plant, and it 23 was believed to be adequate for the life of the plant.
24 COMMISSIONER ASSELSTINE: So this is really kind of 25 a plant aging kind of question. Probably of somewhat generic
.- 50 1- applicdtion,,not just limited to you.
2 MR. FOGARTY: Yes. Commissioner Asselstine, to 3- elaborate on that a little bit more. We did test this cable 4 periodically at eight KV and 12 KV. But the failure we found 5 on some of the other pieces of cable didn't occur until we got 6 to 28 KV.
7 So it indicates to us that even with our high pot 8 testing of cables, the voltage used was not high enough, and 9 we had to revise that procedure to isolate the piece of 10 equipment from the cable, and just test the cable itself to a 11 higher voltage.
12 So this, we think, has generic implications.
13 And it's something we',ve learned here and plan to use, not 14 only at San Onofre, but our other fossil plants.
15 COMMISSIONER ASSELSTINE: Okay.
16 COMMISSIONER ZECH: We should also make sure that 17 our Staff takes that aboard and gets that word out as lessons 18 learned for others.
19 COMMISSIONER ASSELSTINE: Absolutely.
20 COMMISSIONER ZECH: Other facilities, too.
21 COMMISSIONER ASSELSTINE: Let me just -- in a way, 22 what we need to think about, I suppose, is a preventive and 23 predictive maintenance program, even for equipment that you 24 assumed at the outset and the initial analyses showed were 25 going to last the lifetime of the plant, to make sure that
51 1 that's really comin'g true.
2 MR. FOGARTY: And even where you're following it up 3 with routine testing. Here, the testing wasn't adequate to 4 predict the failure that we did have on the cable.
5 MR. PAPAY: An excellent example of plant aging.
6 COMMISSIONER ASSELSTINE: Yes.
7 (Slide.]
8 MR. MEDFORD: This slide discusses a secondary 9 failure in the electrical area. The failure was the failure lo of the sequencer which helps the operator provide the 11 alignment of the second source of off-site power; the failure 12 of that sequencer to complete its cycle.
13 If we could, in the interest of time, I'd like to go 14 on beyond this slide.
15 CHAIRMAN PALLADINO: On this aging question, isn't 16 that part of an~ environmental qualification of the equipment?
17 MR. FOGARTY: This isn't safety-related equipment, 18 as such. It was a piece of cable in the non-nuclear part of 19 the plant. But the failure of that cable led to the ground, 20 and led to the other things.
21 So it led us into a safety challenge, but it's not 22 initially part of the safety system.
23 CHAIRMAN PALLADINO: I see.
24 (Slide.)
25 MR. MEDFORD: Finally, the final portion of my
.. \
52 1 1 presentation summarizes the plant redifications. I've already 2 discussed the feedwater system modifications, I'll pass on 3 that.
4 (Slide.]
5 MR. MEDFORD: This slide depicts the electrical 6 system plant modifications, if we could go on beyond that.
7 MR..BASKIN: Excuse me, Mark. 'Back up to the bottom 8 one there. .I think you need to mention that.
9 MR. MEDFORD: Okay. Because I skipped over that 10 previous slide, I did not address thio ~ point.
11 We've decided, as a result of this event, in order 12 to simplify the operator actions following a loss of the 13 primary source of off-site power, we're going to provide at 14 the next refueling outage, a second immediately available 15 source of off-site power.
16 MR. BASKIN: Let me just comment in answer to your 17 question earlier, Commissioner Asselstine. On this list, 18 those items that have an asterisk will be completed prior to 19 restart from this outage. Those items that are not asterisked 20 would be for the next outage.
21 COMMISSIONER ASSELSTINE: Okay.
22 CHAIRMAN PALLADINO: And so far they're all 23 astorisked, aren't they?
24 COMMISSIONER ZECH: Except the last one.
25 MR. FOGARTY: Except one.
I 53
, 1 . CHAIRMAN PALLADINO: Which one.
2 MR. PAPAY: A second source of off-site.
3 MR. BASKIN: On the second page of the listing --
4 CHAIRMAN PALLADINO: I see. I'm sorry.
5 COMMISSIONER ASSELSTINE: They do that at the next 6 outage.
7 CHAIRMAN PALLADINO: I didn't turn the page. All 8 right, thank you.
9 MR. MEDFORD: Next slide.
10 (Slide.]
11 MR. MEDFORD: Let me quickly go over these. We have 12 eliminated the cause of the spurious ringing of the red phone.
13 CHAIRMAN PALLADINO: Did you find out why? Or how 14 do you eliminate the spurious ringing unless you know how it
. 15 happened?
16 MR. MEDFORD: Okay. The problem was, that although 17 the phones had an uninterruptable source of power, there was a 18 momentary -- approximate two-tenths of a second --
19 interruption in power to the phones. Telephones operate with 20 a current, or a constant current to them. When they lose that 21 current, they ring. That's what happened here.
22 CHAIRMAN PALLADINO: Okay.
23 COMMISSIONER ASSELSTINE: And how have you fixed it?
24 MR. MEDFORD: We provided for an uninterruptable 25- source of power that is the primary source of power, backed up
. 1 54 1 by normal power. So in a loss of power, the current would be 2 continuous.
3 We're going to provide an uninterruptable source of 4 power to the tech support center computer to ensure its 5 availability,. We're going to add a DC control room clock.
6 We're providing for automatic isolation of steam 7 generator blowdown. We're providing for control room 8 indication of steam generator blowdown. And finally, we're 9 going to eliminate the spurious safety injection signal which 10 occurred during the event.
11 That concludes my presentation. And I'll turn it 12 over to Mr. Krieger, who will discuss the program related 13 issues.
14 MR. KRIEGER: Thank you, Mark.
15 (Slide.)
16 MR. KRIEGER: Mr. Chairman, other Commissioners, I'm 17 pleased to have the opportunity to speak here today, to share 18 with you some results of the assessment that we have done in 19 really two areas, which are depicted on this slide. Namely, 20 operator /STA performance and material condition of the plant.
21 In discussing the operator /STA performance, it is i
22 convenient to divide the discussion into two distinct 23 periods. That associated with the troubleshooting for the 24 four KV ground, and that associated with the response to, and i
25 recovery from the loss of off-site power, and the water hammer
55 1 avant.
2 As I go through here, I'll refer to both of those I
3 are pre-trip performance and post-trip performance for '
4 convenience.
i 5 The assessment of operator performance was extremely j 6 -difficult for me as an operator -- operating manager, excuse 7 me -- in terms of root causes. You deal quite a bit more here
-8 with subjective issues than you do in some of the other areas 9 which Dr. Medford referred to.
10 However, in performing an assessment of the 11 activities during this period, we did identify, after some 12 amount of assessment, a root cause. And that root cause is 13 associated with the fact that the ground which had occurred 14 was not as quickly isolated as it might have been.
15 In assessing the effect of that, I believe that it 16 is a contributing factor to the primary cable failure, which 17 Dr. Medford identified earlier.
- 18 Our assessment in no way indicated that the 19 operators did not perform their actions as intended, or not in l 20 accordance with procedures.
4 21 However, again in asking the question, could there 22 have been something done differently, I believe the answer to l- 23 that is yes. And it is to assure an effective isolation of 24 electrical grounds.
j 25 Next slide, please.
56 1 (Slide.]
2 COMMISSIONER ASSELSTINE: Is this something that the
~
3 plant personnel have to do on a fairly frequent basis? And 4 how much training did they have in being able to identify and !
5 isolate grounds like that?
- 6 MR. KRIEGER
- No, sir. That was -- I was going to 7 respond to this as part of our corrective actions.
i 8 COMMISSIONER ASSELSTINE: Okay, fine.
9 MR. KRIEGER: But the answer is that it's not 10 experienced very-frequently. They don't happen very 11 frequently, and therefore, you don't have an opportunity to --
12 MR. FOGARTY: It might be over a period of several 13 years between events of this type.
14 COMMISSIONER ASSELSTINE: Okay.
15 MR. KRIEGER: We, in our assessment, identified 16 really two causal factors which were associated with the slow
,. 17 isolation. The first one, here is noted as a need to increase 18 the operator experience. Clearly, if you provide that i
19 experience frequently, you can make decisions appropriately.
20 The remaining three items here are associated with l
21 what I've termed a lack of a sense of urgency, and that the 22 ground isolation needed to proceed very quickly.
23 And very briefly, the STA, which was in the control i 24 room during this period, while he participated in l 25 understanding the actions of the operators, his technical and i
i
. ~ , . - - , ,~m-w-y ..-,-.v.---., . , . . ~ , . . . . , , - ..cm,% . , , - - . . ~ , , < - - , - - . - . - - - - - - - < - , . . - - . - - - - -
I 57 1 expert safety analysis expertise during a situation that was 2 not an emergency -- namely, it was an abnormal non-emergency 3 -- did not serve us very well.
4 Therefore, he did not input to the operators the 5 need to quickly isolate the ground.
6 The next item deals with a design of our ground 7 alarm system. Currently, that system is installed on the 8 electrical buses. There is not one on the C transformer. -
9 Therefore, when the breaker was opened, isolating 10 the ground, the alarm reset. And that tended to reduce the 11 urgency that had been placed on searching for the ground.
12 We will improve this system by installing a ground 13 indication directly off the transformer, this outage.
14 And the last item here is the implementing procedure 15 used during the troubleshooting. It is currently a normal 16 operating instruction. It is not --
17 COMMISSIONER ASSELSTINE: On that first -- go ahead.
18 MR. KRIEGER: It does.not invoke a sense of urgency.
19 COMMISSIONER ASSELSTINE: On that first item, 20 operator experience, are you going to set up a training 21 program that will actually give these people hands-on 22 experience in going out and' finding these grounds and getting.
23 them fixed? Or how are you going to go about doing that?
24 MR. KRIEGER: Although we haven't finalized that 25 program, the factors that we are looking at include those.
.~
58 1 COMMISSIONER ASSELSTINE: Okay.
2 MR. KRIEGER: Namely, on-the-job training, and --
3 enhanced on-the-job training -- actual man'ipulations, and also 4 enhanced classroom training.
5 COMMISSIONER ASSELSTINE: Okay..
6 CHAIRMAN PALIADINO: And knowledge about the 1
7 procedures also.
8 MR. KRIEGER: Yes. Again, I didn't' finish here. We
, 9 will elevate the level of the instruction used for searching 10 for grounds to an abnormal operating instruction.
11 COMMISSIONER ZECH: If you don't put it in your 12 training program or emphasize it that, you know, then you're 13 never sure what's going to happen.
14 I certainly agree that it should be in the training 15 program. In fact, I think the operators, as you well know, 16 rather see those kind of realistic training evolutions as 17 something that is really worthwhile and beneficial. And they 18 rather get enthused about it.
19 Because, if you look at a print and, you know, you 20 can think of a what-if type thing. This is the way you do it, 21 as you well know. You say, okay, what if this happens and 22 this happens. And a good instructor can, you know, look.at 23 the print and decide what's going to happen if you have a 24 ground, or if you have an electrical problem somewhere, and 25 give the set of conditions. And.then let the trainee, your
e 59 1 operators figure out exactly where the failure was.
2 And when you're tracing a blueprint and going 3 through that kind of a drill, it's been my experience, that 4 - the operators really do get enthused about that. Because it's 5 a matter of, you know, how quickly they can identify it and 6 gives them a lot of satisfaction about really understanding 7 their plant.
8 And those kind of things can be applicable to 9 lots of other -- you know, you can't nail down every little 10 thing that's going to happen. But if you get your people 11 looking at the prints, your electrical prints, and familiarize.
12 themselves so that, you know, they can pull out any print 13 you've got -- and you've got lots of them. They're in the 14 control room. And I've seen them and usually ask the 15 operators when I visit to pick out a certain print for me, 16 just to see how familiar they are with those prints.
17 But if they really are familiar with them, they can 18 -- you know, they.can diagnose these problems very well. But 19 if you don't include it in the formal training program, it 20 just doesn't happen.
21 So it's something that, I certainly, would commend, 22 not only to you but to other training programs, so that we 23 emphasize this. It's a blueprint reading type thing, and it's 24 very easy to do. And perhaps, needs emphasis. But it's 25 something that the operators, I think, rather enjoy doing
s' 60 1 because it gives them a lot of satisfaction as to how well 2 they know their plant.
3 And then when something comes up like that, they're 4 a little bit more comfortable looking at that print and 5 tracing things down. And perhaps, could solve the problem a 6 little bit sooner.
7 So I think it's an important lesson to be learned.
8 (Commissioner Bernthal reentered the room.]
9 COMMISSIONER ZECH: And I hope also, that the Staff 10 can discuss it with INPO and others to apply it in a more 11 generic way.
12 COMMISSIONER ASSELSTINE: Russ, on your training 13 program, how much of your training program is oriented towards 14 this kind of thing? Nuts and bolts, here's what can go 15 wrong. Let's test and train your ability to understand what 16 the problem is. Finding an electrical ~ ground or fault and 17 getting it corrected. As opposed to, say, the more serious 18 accident kind of situations and responses to those kinds of 19 occurrences? Is it a mix?
20 MR. KRIEGER: Yes, it is a mix. However, I believe 12 1 the majority deal with abnormal operating conditions, as 22 opposed to evolutions and troubleshooting.
23 COMMISSIONER ASSELSTINE: Do you think we've gone 24 too far in striking the balance?
25 MR.-KRIEGER: I'm sorry, I don't understand that t
- - 61 ,
1 question.
l2 COMMISSIONER ASSELSTINE: Well,'are we spending too 3 much time worrying about the more serious, responses to the 4 more serious accident situations and not enough time on the 5 kind of day-to-day activities that help you avoid getting into 6 one of those kinds of situations to start with?
! 7 MR. KRIEGER: I don't believe. In this particular 8 area of troubleshooting for San Onofre, we're -- it is 9 covered. But we need to enhance the depth of that training.
10 COMMISSIONER ASSELSTINE: Okay.
11 MR. KRIEGER: And that's what we will do.
12- MR. FOGARTY: And I think we need to do another 13 thing, too. Because with 42 steam units on the system, there 14 are lots of lessons to be learned in those other facilities.
15 And we clearly have not been transferring those kinds of 16 experiences over into the San Onofre operator training 17 program.
18 Because if something happens in an oil and gas-fired 19 unit, in finding a ground, for example -- exactly the same 20 electrical problem -- they may learn something about doing 21 it. And we haven't done as well as we should in incorporating 22 that into the San Onofre training program. We will do that.
23 MR. KRIEGER: Relative to post-trip operator /STA 24 performance, our assessment indicates that the operators and 25 the STA understood the behavior and response of the plant and
. .. .. = - -
i .* 62 1 were able to maintain the plant in a controlled condition 2 utilizing a formality, attention to detail, consistency and 3 team effort and training.
4 I'd like to conclude my discussion of post-trip by 5 mentioning that our assessment of the procedural and training 4
6 guidance here indicates that, as you might expect, for any new 7 event.-- and I suppose for old ones as well -- you can always 8 enhance procedures and training. And in this particular case 9 it's no different.
10 We have identified a number of enhancements in the 11 procedural and training guidance that we are implementing. As f
12 an example, you probably recall the difficulty in-initiating 13 residual heat removal.
14 Now our assessment of that event indicates that 15 while the operators understood through procedures and training 16 that the set point was 400 pounds, and that there was a dead 17 band associated with that, the details of that dead band, the 18 depth was not included. The operators did not expect it to be l 19 as large a. dead band as it turned out to be.
20 And we are going to -- we believe that for actions 21 that need to be taken by.the operator, with respect to 22 instrumentation set points, should be specific in the area of 23 dead band. And we are proceeding to review our procedures on 24 a generic basis and assure that that is accomplished.
25 I won't go into the details of the other areas.
r
.~
.- 63 1 But we are pursuing that as well.
2 COMMISSIONER ZECH: I'd just like to say very 3 quickly that I thought -- I agree with your assessment of the 4 operators. I thought they did a goed job. And also, I think 5 your STA, you know, looking over all the STAS I've seen in my 6 plant visits around the country, I thought your STA performed, 7 . frankly, in a very commendable manner, and perhaps above 8 average from what I've seen. I think he did make a 9 contribution.
10 And I thought he did quite a good job. In what 11 I would expect the STAS to do, I thought he did quite a job,.
12 and perhaps above average or maybe better than that.
13 MR. KRIEGER: We agree.
14 CHAIRMAN PALLADINO: I'd like to ask a question 15 somewhat related to this training issue. In the NRC's team 16 report, they indicated that they didn't see enough of an 17 inquiring attitude on the part of the people at the plant.
18 That is, the team said they identified a number of problems 19 that would not have otherwise been found.
20 Do you think that your modified training program 21 covers improvement of attitude, at least so far as finding 22 causes of problems?
12 3 MR. KRIEGER: I don't have any specific information i 24 on that.
25 MR. BASKIN: Maybe I can address that in this sense.
.. 64 1 CHAIRMAN PALLADINO: 'You need some assurance that t
2 when things happen that the questions are asked, and that the 3 technical effort is being made to answer them.
4 MR. BASKIN: The training program helps in that 5 . respect. I think more importantly, however, is a couple other 6 factors. [
7 One, some of the lessons we've learned -- and I 8 don't mean the specific lessons -- but the overall context of l
9 this event, and the fact that it made it clear to us that we 10 do need to reinforce at all levels of management, as well as 11 operators, this attitude. And that is being done, will 12 continue to be done.
13 The addition of some outside help on our safety 14 review committee is a step, albeit small. But to help bring 15 in other viewpoints in this regard.
16 So we have the message. We understand the need.
17 And believe me, we're going to try as hard as we can to 18 fulfill it.
19 CHAIRMAN PALLADINO: Yes, it's a big challenge to 20 get people to ask questions.
21 MR. BASKIN: That's right.
22 CHAIRMAN PALLADINO: But it's something that can be 23 instilled. And actually, I think, training can help. ;
24 MR. BASKIN: Training very definitely does help.
25 CHAIRMAN PALLADINO: If they're called to identify
.. 65 1 causes of events in their training program.
2 MR. BASKIN: That's correct.
3 MR. PAPAY: I think the bringing in of the STA in 4 the abnormal conditions before the event occurs is a way we 5 can use the on-site expertise in a more meaningful manner. I 6' think that's important.
7 COMMISSIONER BERNTHAL: What plants are there in 8 this-country that are -- which ones are the closest analogs of 9 your plant?
10 MR. BASKIN: Probably Connecticut Yankee. That 11 would be by far the closest analog. Then your next one newer 12 would be Ginna, and older would be Yankee Rowe. But 13 Connecticut Yankee is the closest.
14 COMMISSIONER BERNTHAL: Yes, Yankee Rowe is not 15 probably a very close analog. Connecticut Yankee, I haven't 16 been there, but.it may be.
17 MR. BASKIN: There were built within about -- went 18 in service within, as I recall, about six to nine months of
, 19 each other.
20 COMMISSIONER BERNTHAL: Have you ever been to 21 Connecticut Yankee, any of you?
22 MR. BASKIN: I have.
23 _ COMMISSIONER BERNTHAL: To inspect their 24 operations. Now there's a plant that has had an 25 extraordinarily good track record. They must be doing
.* 66 1 something that maybe we can learn from.
1 2 MR. BASKIN: Yes.
I 3 MR. KRIEGER: If there are no other questions, that 4 concludes my discussion of operator /STA performance.
5 COMMISSIONER ASSELSTINE: One other question on 6 operators. To what extent are your operators people who have 7 been -- this plant is an older plant -- have been with the 8 plant for a long number of years? And to what extent do you 9 have newer people as well that are serving on your shift 10 complement?
11 MR. KRIEGER: While we at San Onofre have a number 12 of newer operators, we have a significant number of older.
13 For~ example, the shift -- excuse me, the plant superintendent I
14 has 20 years of experience on San Onofre. The assistant shift 15 superintendent also has comparable experience. The shift 16 superintendents have lesser than that, but it's on the order 17 of five, six years. And the operators, control operators 18 also.
! 19 So I think at San Onofre Unit 1 we have a fairly 20 good experience level.
21 MR. BASKIN: Before we proceed, Russ, let me ask.
22 Mr. Chairman, we've reached the end of our allotted time. How 23 much more time do you think you' gentlemen --
< 24 -CHAIRMAN PALLADINO: Let me ask the Commission. I 25 know Commissioner Zech has to leave in a few minutes.
l
- 67 l 1 COMMISSIONER ROBERTS: So do I. i 2 COMMISSIONER BERNTHAL: I'm going to have to leave 3 fairly promptly here at 15, 20 minutes, too.
4 CHAIRMAN PALLADINO: At 20 minutes to?
5 COMMISSIONER BERNTHAL: No, by 10 of 12:00 or so.
6 CHAIRMAN PALLADINO: Okay, why don't we plan to go i 7 another 20 minutes.
8 MR. BASKIN: Okay, we will --
9 CHAIRMAN PALLADINO: I have a couple of general 10 questions. I want to have at least two or three minutes to --
11 MR. BASKIN: Okay, Russ, why don't you review very 12 quickly the last part of your presentation in no more than 13 five minutes. And then we can get questions and pick'up 14 details as we need.
$ 15 MR. KRIEGER: As Dr. Medford mentio.nd, there are 16 two -- in the area of material condition of the plant, there 17 are two programs that we're implementingt material condition
-18 review program, and an area monitoring program. And I'll very 19 quickly summarize those programs.
20 The material review, condition review program has 21 an objective of assure reliability of system operation upon 22 return'to. service. The scope of that program has as elements, 23 an identification of items to be included which are those 24 systems which would create a challenge to safety systems.
25 -There will be -- by the way, just for reference --
.. 68 1 there are 23 safety systems being considered, and 138 2 different components within those systems being considered.
3 With respect to the determination of material 4 condition of these items and the determination of material 5 history of these items, there are currently 19 teams of five 6 members each, including operations, technical and maintenance, 7 and additional expertise from within and outside of SCE.
8 They will, in terms of material condition, will
'9 perform an actual inspection in the field. It will include 10 consideration of performing tests, both non-destructive and 11 destructive. And will include an assessment of the preventive 12 maintenance schedules and practices.
13 An example of the testing, I've mentioned. I'll 14 simply refer.you back to the discussion that Dr. Medford had 15 on the four KV cables.
16 The material history, as also Dr. Medford mentioned, 17 we consider this extremely important and will perform a 18 research of our historical files to develop this material. We
( 19 will also search the NPRDS database.
20 Following the generation of this~information, it 21 will be evaluated by a committee, which will be composed of 22 material and apparatus experts both within and outside of
! 23 SCE. They will review this material to assure that there is l 24 adequacy for operational readiness in terms of reliability.
l 25 That the objective I mentioned earlier is met.
1 i
,. 69 1 That will include, if appropriate, identification of 2 any special testing or performance testing, prior to return to 3 service.
4 COMMISSIONER ASSELSTINE: Is this program limited 5 only to safety-related systems and components?
6 MR. KRIEGER: No, sir. It is related to both 7 nonsafety-related and safety-related systems which could 8' challenge --
9 COMMISSIONER ASSELSTINE: Good. It seems to me --
10 MR. BASKIN: In his earlier comment, he said 23 11 safety systems. I think he misspoke. It's basically divided 12 into 23 systems, both safety and nonsafety. It covers aspects 13 of the whole plant.
14 COMMISSIONER ASSELSTINE: Good. It seems to me one 15 of the lessons from this, and I think some other events, is 16 the one that Joe pointed to earlier, and that is, you've got 17 to look at the plant as a whole.
18 MR. BASKIN: Absolutely.
i 19 COMMISSIONER ASSELSTINE: There are a lot of these 20 nonsafety-related systems and components that can really cause 21 .you trouble.
22 CHAIRMAN PALLADINO: I'm not sure I know what a 23 nonsafety-related system is anymore.
24 COMMISSIONER ASSELSTINE: I'm not either.
i 25 MR. FOGARTY: They quickly become safety-related.
l I
l i .- -
,. 70 1 We've certainly learned that.
2 COMMISSIONER ZECH: Well, the point is, the balance 3 of plant systems simply have got to be emphasized more than we 4 perhaps have in'the past. They can cause all kinds of 5 troubles, and certainly have.
6 COMMISSIONER ASSELSTINE: That's right.
7 COMMISSIONER ZECH: And so, I agree that the 8 emphasis should be on the whole plant and not just the 9 so-called safety-related systems. Balance of plant can be 10 very, very important and should be looked at carefully,
- 11. MR. BASKIN: And this program covers the whole 12 plant, no question.
13 COMMISSIONER ASSELSTINE: Good.
14 MR. KRIEGER: Again, to summarize, this program is 15 currently being implemented to assure reliability of systems 16 upon return to service.
17 Once we do return to service, the area monitoring 18 program will be the program to maintain a high material 19 condition standard in operation, to provide an appropriate 20 physical context for all work.
21 Before I go on to the scope of this program, let me 22 simple mention that in the fall of last year we recognized the 23 need to implement a program which would look at the physical 24 context in which people do work at the station, to assure that
.25 an environment is created the symbolizes the importance of
l
,. 71 1 work quality. So'we began development of this program at that 2 point in time, and have in January of this year, began 3 implementation of the program.
4 This program has as its scope the following 5 elements. We've divided the plant into 11 distinct areas, 6 based on size, system orientation, equalization of work, et 7 cetera. We've assigned an area team composed of operations, 8 ' technical, and maintenance, who have ownership of these !
9 distinct areas.
10 Those teams will perform on a weekly basis, as a 11 minimum, a joint inspection of the plant in their areas to
.12 monitor and evaluate the effectiveness of the existing work 13~ processes to maintain the~ material condition standard. .They 14 will be looking at the actual material condition of the 15 components and systems, the housekeeping conditions, to as'sure 16 that those are proper.
17 (Commissioner Roberts left the room.]
18 MR. KRIEGER: And will also be evaluating the 19 findustrial. safety standards in the area.
20 And this information, the results of their 21 inspections and monitoring will be provided through a monthly 22 summary report ~to two other teams that have been established, 23 composed of supervisory and management level peoplei who also 24 perform periodic inspections in the plant of their own.
25 ~ Assess the results of the area team evaluations, and determine
,. 72 1 whether or not revisions to the work processes are necessary 2 to assure continued high level of material condition.
~3 COMMISSIONER ASSELSTINE: Could you give a couple of
.4 examples of the distinct areas? Are they~ systems, for 5 example, where you get basically.this sort of ownership 6 concept by some of your people?
7 MR. KRIEGER: Yes. To the extent we can, we've 8 broken it down to cover a minimal amount of systems in each 9 area, not to overburden the teams. We've included, to the 10 extent we can, a division of areas involving radiation access.
11 COMMISSIONSR ASSELSTINE: Okay. How about a couple, 12 like examples? Like,'for example, would the emergency --
13 MR. KRIEGER: Containment, the turbine building, the 14 diesel generator building, the intake structure. ,
15 COMMISSIONER ASSELSTINE: Okay. Are these two 16 programs, both the material condition program and the area 17 monitoring program sort of new concepts in terms of -- or.are 18 they concepts that other utilities have used in the past? And 19 with what kind of success? To what extent are you charting 20 new ground with these kinds of initiatives?
21 MR. KRIEGER: I don't have any specific information, 22 but my opinion is that these are fairly unique to San Onofre.
23 COMMISSIONER ASSELSTINE: Okay.
24 MR. FOGARTY: We haven't found anyone else doing '
25 anything like this. We think we are plowing new ground
,. 73 1 COMMISSIONER ASSELSTINE: It sounds like good 2 concepts, both of them.
3- To what extent -- or I guess, how far along are both 4 of these items? And to what extent have they been able to 5 identify, so far, problem areas, areas that need attention 6 that weren't identified specifically by the event? The 7 feedwater check valve failure.
8 MR. ZRIEGER: Let me address those in the two 9 parts. With respect to the material condition review program, 10 we are just now getting that underway, and do not have any 11 results -- significant results, other than the four KV cable 12 testing which Dr. Medford mentioned.
13 COMMISSIONER ASSELSTINE: Okay.
14 MR. KRIEGER: With respect to the area monitoring 15 program, beginning in January, the operators performed 16
~
inspections of all areas to establish whether there are 17 additional deficiencies. Those have been input to the work 18 processes.
19 And just beginning in February, the team concept is 20 underway. And at this point in time, the results haven't yet
- 21 been generated.
22 COMMISSIONER ASSELSTINE: Okay.
23 MR. BASKIN: If I could have the last slide.
24 (Slide.]
25 MR. BASKIN: I'll summarize here very briefly. Our
.. 74 1- judgment, and based on all the analysis we'.ve done, is there's 2 two basic root causes of the event. The first one was the 3 check valve failures resulting from turbulent flow combined 4 with inappropriate application. This we believe, or items 5 similar-to this will be precluded in the future for a number 6 of reasons.
7 Number one, the inspection program for valves we've
-8 talked about, at San Onofre.
9 Number two, derugn processes today as opposed in 10 1965 certainly make the inappropriate application much more 11 unlikely today than they were in 1965. The program we have 12 for any abnormal occurrence, to review.past history in a more
'13 disciplined manner than we've done in the past-should minimize
~14 the possibility of that.
15 And lastly, the addition to our capabilities, our j i
16 on-site review committee of outside expertise should help 17 provide a little bit more -- a little wider perspective.
18 .The auxiliary transformer failure, the result of 19 aging, we believe will be -- is adequately addressed by the 20 material condition review program. That will assure we're in 21 fine shape everywhere before we start up. And the area.
22 monitoring program, that should provide assurance that we
.23 maintain that.
24 Two other quick comments. We've been working 25 with the Staff over the past months and will continue to do
,, 75 1 so. We've had several meetings with them; expect to have 2 more. We will submit to the Staff at the end of this month 3 what we're. calling our return to service report, which will 4 document everything we've talked about today, plus tte results 5 of other areas of investigation that we haven't necessarily 6 gone'into today.
7 And that,.Mr. Chairman, concludes what we have to 8 say, and we'll certainly entertain any questions that you or 9 anybody else might have.
10 CHAIRMAN PALLADINo: All right, thank you, 11 Mr. Baskin.
I must say I am heartened by the corrective 12 actions you're at least planning, and will look forward to 13 confirmation by the Staff that, A, they are satisfied, and, B,
<14 that they are properly implemented.
15 I have three broad questions I'd like to ask. One 16 comes about from the configuration of San Onofre. Unit 1 -- and 17- I think in Unit 3 also -- the steam generators are 18 cross-tied. Therefore, a main steamline break could cause the 19 blowdown of all three steam generators in'the system.
20 Containment design usually calls for absorbing the energy from 21 one such steam generator blowdown.
22 What is the design basis for containment energy 23 mitigation in this plant?
24 MR. BASKIN: In this plant, the design basis 25 includes absorption from all steam generators.
.. 76 1 CHAIRMAN PALLADINO: All three?
l
, 2 MR. BASKIN: Yes. In other words, the simple is, 3 although we don't have main steam isolation valves, we meet 4 all the criteria that we would have to meet if we did have 5 main steam isolation valves; containment absorption of 6 energy, cooldown requirements, all of that.
7 I should comment -- and we have looked over the 8 years, Mr. Chairman, at the installation of main steam 9 isolation valves, in spite of the fact we meet all 10 requirements. It is a vezy significant undertaking, were we 11 to do such a thing.
12 CHAIRMAN'PALLADINO: I presume the Staff has 13 reviewed your containment arrangement and is satisfied. I 14 won't put them on the spot now, but I am interested in making i
15 sure that they concur.
16 MR. BASKIN: It's my understanding that that's the 17 case.
18 CHAIRMAN PALLADINO: Okay. Let me go to two other 19 areas. There was a major fire at SONGS in 1968, analogous in
, 20 serious' ness to the Browns Ferry fire in 1975. I note, and I 21 must say with some dismay, that Appendix R fire protection-has 22 not been completed at San Onofre yet, at least in this unit.
23 Can you explain why not and when you expect to see 24 some completion?
25 MR .- BASKIN: Yes, sir. The primary reason it has
.,.. 77 1 not been completed is -- I think as we're aware, during the 2 period from roughly 1981 to end of 1984, the plant operated, 3 .unfortunately, a very small amount of time. And during that 4 period of time, because of the seismic reevaluation program, 5 we had to assess the continued viability of operation of the 6 unit, which we did and found in a favorable light.
7 In that time period,'little, if any work was-done on 8 this or some other backfit work that is required. The 9 remaining work for Appendix R will be completed during this 10 outage before we return to service.
11 CHAIRMAN PALLADINO: Before you restart?
12 MR. BASKIN: Yes.
13 CHAIRMAN PALLADINO: You're going to have Appendix R-14 completed?
15- MR. BASKIN: Completed, yes, sir.
, 16 CHAIRMAN PALLADINO: I don't know if the Staff's 17 going to inspect it prior to startup. Do you plan'to?
18 MR. MARTIN: Yes, that will be done.
19 CHAIRMAN PALLADINO: Okay. Another area. I note l 20, again that SONGS does not have a safety-grade control room 21 ventilation system. And again, it's with somewhat dismay.
22 And again, let.me ask, why not, and when might we expect it?
4 23 MR. BASKIN: Again, part of the reason it doesn't 24 exist is from what I stated earlier in regard to the hiatus t
25 that occurred during the time when we were reassessing the 1.
, 78 1 viability. .
2 There's a significant amount of work here. Some 31 work will be done before restart in this outage. And the 4 remainder of the work will be complete by the and of the next 5 refueling outage.
6 Is that correct, Mark? '
- 7 MR. MEDFORD
- Yes.
8 CHAIRMAN PALLADINO: This has to all be done during 9 an outage?
10 MR. BASKIN: Well, we are doing the work that we can 11' within an cutage. But much of it has to be done during an 12 outage because you have to shut off the system, and you get in 13 the question of heat loads, and you don't want equipment 14 running and that sort of thing.
15 CHAIRMAN PALLADINO: Well, I think it's an item that 16 deserves some attention. And I would be interested in seeing 17 it done just as soon as it.can. I'll leave it to the Staff to 18 follow up on that.
, 19 MR. BASKIN: I should point ou't, Mr. Medford just 20 made the comment that, the work that will -- well, the work 21 that is remaining and will be remaining, is to install a 22 second redundant safety-related train. We do currently have 23 in effect one safety-related train. It's the redundant train 24 that won't be complete until the next outage.
25 CHAIRMAN PALLADINO: Okay. The ventilation system
- 79 1 .has a lot of ramifications'that at first sometimes don't i
2 appear. It's not only comfort for individuals, but also 3 making sure that the instruments are operating in their proper 4 temperature range, and therefore, can be relied on.
5 How are you doing on the seismic upgrade? I know 6 it's not quite related to --
7 MR. BASKIN: I think we're doing great. Again, 8 we're going to be complete this outage, which I'm sure will 9 not only make you gentlemen very pleased, but believe me, it
~10 will make me very pleased. It seems like it's been my career 11 almost. But yes, all the work will be complete this outage.
12 The plant will be fully upgraded to the new design value.
13 COMMISSIONER ASSELSTINE: Before it starts this 14 time?
15 MR. BASKIN: Before it restarts, yes.
, 16 CHAIRMAN PALLADINO: All right, other questions?
17 COMMISSIONER ASSELSTINE: I was just going to ask 18 about the seismic. upgrade, but that basically covers that one.
19 The other point I would make, I think, is I think 20 the concepts of the material condition program and the area 21 monitoring program to assure that you maintain that material 22 condition from now on really are interesting concepts.
23 I would urge you, and I think, maybe our Staff can 24 help in this too, in sort of spreading the experience that you 25 get from that program among other utilities as well. This'is
80 1 not the only plant of this vintage, and I suspect that some of 2 these vulnerabilities are out there in other plants as well.
3 And what we ought to learn from this experience is to find 4 those vulnerabilities and fix them before we have operating 5 events.
6 MR. BASKIN: And we expect that they will be very b 7 useful, not only addressing the sorts of problems we've been 8 -discussing. But we expect to see a pay-off in reliability.
9 COMMISSIONER ASSELSTINE: Absolutely.
10 MR. BASKIN: And much more than just safety, if you 11 want to separate the two. We expect that they will more than 12 pay for themselves.
i 13 .MR. FOGARTY: And Commissioner Asselstine, we do 14 intend to publicize this and present papers wherever it's 15 appropriate.
1 16 COMMISSIONER ASSELSTINE: Good.
17 COMMISSIONER BERNTHAL: Just to make a brief 18 comment. I'm encouraged by the actions that you appear to be 19 taking here, and would just want to emphasize again, the 20 thought that I think I've expressed privately to one or two of 21 you, that unlike some other cases where utilities, perhaps 22 with a single older plant have major' events and find 23 themselves at loose ends institutionally and 24 organizationally. For them it's a major problem to bring the 25 plant up to standards.
. 81 1 Not only shouldn't it be a major problem for you.
2 You also don't have any excuse. I mean, you're a large 3 capable organization. I' visited your plants. Perhaps you 4- were a bit too distracted by these two large,.new plants and 5 the old one didn't get all the attention it needed.
6 But I expect -- and I think your people in 7 California have a right to expect -- that you'can address this 8 problem quickly and bring this plant up to the kind of 9 standards now that you expect and all of us expect. And I 10 would urge you to do that.
11 I'm also very encouraged by the direct commitment 12 and involvement of the chief executive officer in this effort, 13 .Howard Allen. I know that -- or at least, I've been told that 14 he is taking personal. interest in walking through 15 occasionally. - And I trust he will continue to do that, and 16 make sure that this work is carried out expeditiously.
17 MR. FOGARTY: He certainly intends to do that, yes, 18 sir.
19 COMMISSIONER ZECH: Just a final comment, too. Your
, 20 program does sound like a reasonable program. I would only 21 commend you to continue your management involvement, which is 22 apparent. But also, with attention to detail and making sure 23 that your program, which sounds good, is executed properly ,
24 with discipline and follow-through.
25 The big thing is to follow through. The program r
m ------ e w ,,ew- w-m-rw.-m- -
?<m -4 ,_,,,-w -n -w--y-m- , - - - gm, m p-n4,.-. wm ,e -vvv,y , ----w---,------,v-
, .. 82 1 sounds good, but if you don't follow -through, nothing's going' 3 2 to happen. You've got to follow through. And you've got to 3 watch the details, and you've got to be -- and management has 4 got to be involved. And you can't just let it go without 5 doing that.
6 And I commend your efforts and your energies J
7 towards that end. 'And I think that's important, as far as a restoring ~your plant to a condition of operational safety and 9 reliability.
10 COMMISSIONER ASSELSTINE: One last point. Maybe, 11 Dave, you could highlight a bit about what you have in mind in 12 terms of greater management involvement. I gather there's
. 13 some efforts to see more detailed involvement by you, Ken and 14 Larry in the day-to-day operations.
15 .MR. FOGARTY: I think -- well, one commitment we've
, 16 made is, first of all, of course, to spend more time at the 17 plant instead of at Rosemead.. And that's already started, and 18 it will continue.
19 Another thing we're doing, and again, maybe not i
20 directly applicable, but Harold Ray, our site vice president 21 and myself are going to visit some of the Japanese plants in i 22 early May.
i f
p 23 COMMISSIONER ASSELSTINE: Good idea.
I 24 MR. FOGARTY: We've heard quite a bit about these 25 plants and we think it's time we went over on a tour of our
,. 83 1 own to see how their doing it. Because they do seem to have a 2 better history in some of these things, particularly in the 3 operation and maintenance area.
4 CHAIRMAN PALLADINO: Particularly in maintenance, 5 yes.
6 MR. FOGARTY: So we have a lot to learn. We're 7 going to be doing that. I think continued meetings with site 8 personnel and a discussion of events. For example, our chief 9 executive officer, Mr. Allen, met with the station management 10 people when we had a problem with the gauge glass on the 11 auxiliary feedwater pump several months ago.
12 I think we're going to have to do more of that. To 13 indicate, first of all, to the people in the plant that the 14 CEO or_the highest levels of management are aware of these 15 problems, and to have candid, open discussions.
16 So I think all of those things are things we're 17 going to be doing.
18 CHAIRMAN PALLADINO: All right. .Let me make one 19 final comment. We look forward to your fulfilling the goals l 20 that you set for yourself. I would urge you to work closely 21 with the Staff to make sure that they're knowledgeable about-22 what you're doing and they're giving-any necessary approvals 23 along the way. And look to confirmation on their part when 24 the plant is ready.
25 So let me thank you. I think it was a worthwhile
.* 84 1 presentation. We encourage you to go forward with your plans.
2 MR. BASKIN: Thank you very much.
3 CHAIRMAN PALLADINO: The meeting stands adjourned.
4 [Whereupon, at 11:56 a.m., the commission meeting 5 was concluded.]
6 7
8 9
10 11 12 13 14 15
-16 17 18 19 20 21 22 23 24 25
i
.1 CERTIFICATE OF OFFICIAL REPORTER 2
3 4
5 This is to certify that the attached proceedings 6 before the United States Nuclear Regulatory Commission ~in the
.7 matter of. COMMISSION MEETING e
9 Name of Proceeding: Briefing by Southern California Edison Company on San Onofre 1 (Public Meeting) 10 11 Docket No.
12 P l ace Washington, D. C.
13 Date: Tuesday, March 18, 1986 14 15 were held as herein appears and that this is the original 16 transcript thereof for the file of the United States Nuclear 17 Regulatory Commission.
19 (Signature) g[ ['
(Typed Name of Reporter)Page d j briggle 21 22 23 Ann Riley & Associates, Ltd.
24 25
SCHEDULING NOTES ~
IITLE: 3RIEFING BY SCUTHERN CALIFORNIA EDISCN CCMPANY (SCE) ON SAN ONOFRE-1 SCHEDULED: 10:00 A.M., TUE5 CAY, ,4*ADCH 18, 1986 (OPEN)
DURATICN: A 'P D :0X l-1/2 WP3 SPEAKERS: I, PRE:ArE (TENTATIVE) DAVID J FCGEETY EXECUTIVE VICE P4ESIDENT, SCE '
il, INT ODUCTION KENNETH P. BASKIN VICE PRESIDENT NUCLEAR ENGINEECING SAFETY AND'llCENSING, SCE III. DESIGN RELATED-ISSUES A, 'FEEDwATER SYSTEM B. ELECTP! CAL POWEP SYSTEM C. OTHE8 DESIGN ISSUES MARK 0. MEDFORD, MANAGER NUCLEAR LICENSING, SCE IV, P nGAAM RELATED ISSUES A. MATERIAL CCNDITION OF THE PLANT B. OPERATICNAL ISSUES RUSS W. KRIEGER, MANAGER OPERATIONS, SCE
- v. CcNCtustcNs/ClostNG CCMMENTS KENNETH P. 3ASKIN VICE P:ESIDENT NUCLEAp _ENG:r:EERING SAFETY AND LICENSING, SCE INE FOLLCWING PERSCNS ALSO WILL EE SEATED AT THE COMMISSION TABLE:
DR. LAWRENCE PA:AY SENICR VICE PRESIDENT SCE JACK RAINSEE:RY, SUPERVISCR SAN CNCFCE UN:T 1 LICENS NG
i l
l MARCH 18,1986 .
l i
PRESENTATION TO THE l NRC COMMISSIONERS J
l NOVEMBER 21, 1985 LOSS OF POWER /
WATER HAMMER EVENT SAN ONOFRE UNIT 1
i i
PREFACE -
- Water Hammer Event Was Very significant l
!
- San Onofre Age Requires Additional Care.
l Diligence and Attention Over that of a '
Newer Plant l
l
- IIT Report Professionally Done and Raises
! Valid Concerns j e SCE Committed to Address All NRC Concerns l Raised by the Event l
l i I !
i e
l 4
j ..
i l
AGENDA i
Preface David J. Fogarty
- Introduction Kenneth P. Baskin.
! Design Related issues Mark O. Medford Program Related issues Russell W. Krieger i
- Conclusions / Closing Comments Kenneth P. Baskin
- 1
l 4
I INTRODUCTION
- Event Was Very significant l
- Most Significant Aspect of Event is the Failure of Five Feedwater Check Valves e Careful Attention. Study and Corrective Action Has and Will Be Taken
- Feedwater Check Valves Will Be Replaced With improved Valves e Materia! Condition Assurance Programs Are Being implemented
_lp
l i
i SEQUENCE OF EVENTS 1
J j e Activities to Locate Ground I
i
- Bus 1 A/1C Parallel Operation l
j e Auxiliary Transformer "C" Relay Operation
- Diesel Generator No. 2 Automatically Starts i
i
-
- Unit is Tripped 20 Seconds into Event
- e Diesel Generator No.1 Automatically Starts i 20 Seconds into Event I
i .
t I
l .
i SEQUENCE OF EVENTS (Continued) -
l l
1
- e Main Feedwater Lines Voided 1 Minute into
- Event l
!
- AFW Flow from Steam-Driven Pump 4 Minutes into Event .
e Electrical Power Restored 4 Minutes into Event
- AFW Flow from Motor-Driven Pump 5 Minutes into Event t
1 l
l SEQUENCE OF EVENTS i
(Continued) -
l l
- AFW Flow Reduced 9 Minutes into Event i
l
- AFW Flow increased to 25-40 GPM 10 l
Minutes into Event I
i e Water Hammer Occurred 20 Minutes into Event I
- Feedwater Bypass Valve Ruptured 20 Minutes
- into Event
- Plant Cooldown Completed 11 Hours After Trip i
i 4
l l
i 1
i DESIGN RELATED ISSUES i
l
- Feedwater System
)
- Electrical Power System 1
- Plant Modifications l
l l
l ,
Main Feedwater System San Onofre Unit 1 l FE CV 143 FWS FWS 458 417 415 L To SG M Ak V 1C P
MOV FCV FWS FWS g 22 458 398 396 s
456 1st PT HP HIR R FE 8 S MS ggm 378 376 jL r 18 PUMP 3y TE M p To Safety Fr m Safety injection Condensate MOV FCV FWS FWS inject' n k 20 System 457 346 342 i
.. P 3y f 4h CV-142 MS mS To SG 8528 9 , g HV 439 p p 8M0 jg r IA W MFW PUMP r' r' MOV FCV FWS FWS From Auxilwy 21 456 345 343 Feedwater System L
l .
i i
FEEDWATER LINE WATER HAMMER EVENT i.
)
l
- Feedwater Lines Voided Due to Failed Check Valves
- Auxiliary Feedwater Refilled Feed Lines Resulting in a Water Hammer in Feed Line "B" i
!
- Feed Line "B" Has the Longest Horizontal
- Piping Run to the Steam Generator
i i
i l
i PRIMARY FEEDWATER i
i SYSTEM FAILURE -
i l
- Failure of Five Feedwater System Check Valves
- Primary Causes Turbulence of Flow Stream Oversized Check Valves
- Contributing Factors j - Two Piece Check Valve Disc Design Reduced Flow Operation i
1 i
i 1
i i
i PRIMARY FEEDWATER SYSTEM FAILURE
- (Continued)
- Corrective Actions New Check Valve Design Prior to Startup Flow Control Valve Closure Modification l Prior to Startup l
Add Redundant Feed Line Check Valves i Prior to Startup Install New Sloping "Feedwater Piping to ,
{ Steam Generator B Prior to Startup New Check Valve Demonstration Startup l Test Inspect Check Valves Next Refueling Outage Check Valve Application Review Program l Prior to Startup !
1
l COMPARISON OF CHECK VALVE DESIGNS j e Old (Pacific Valves) Design j -
Oversized for Application Two Piece Disc / Hinge Assembly ,
l Susceptible to Vibration Induced '
Failure l
- New ( Atwood-Morrill) Design 1
Valve Properly Sized to Remain Full Open Down to 70% Flow One Piece Disc / Hinge Assembly l Precludes Observed Failure '
i Mechanism l
i
_ _ a _ _
'l i
t )I i l
} l
- 1 e e
'l I *
'f
.I P'
i -u.
e 4 :
=
l
'!)
J,yl Y
\ ILB i
i i
4 i
--,m---v,- se-wn-w---_,-_ e,-- .__.-r-- - - - - - , - - -
i i
j ---, ,
>s , >
Yi g -
i h - -
vb -
i Pacific Valve Design i
i i
. a- . ..
l1 #"%') t v @"t +
i 7 ' . ' s . ,
l ig( 3f
. k [ ~J Atwood-Morrill Valve Design
i l
SECONDARY FAILURE I
j
- Failure to Recognize Precursor Events e Primary Causes Lack of Readily Available Information
~
- Concerning History of Check Valve Performance l
\
i l
l .. __
SECONDARY FAILURE '
(Continued)
.
- Corrective Actions Incorporate Systematic Search of Early Plant History into Evaluation of All ,
Future Abnormal Conditions Add Outside Expertise to the Onsite !
Safety Committee
6 i
i l
J l .
DESIGN RELATED ISSUES i
i i
j e Feedwater System 1
I j e Electrical Power System i
e Plant Modifications l
l 4 i
I_ __. _ __ _ _ . _ . -
ELECTRICAL POWER SYSTEM e One immediate Access Offsite Source e One Delayed Access Offsite Source e Redundant Onsite 6 MW Diesel Generators Power for All Engineered Safety Features Automatic Loading Provided when Emergency Core Cooling Required Manual Loading Provided for Other Transients
- Core Cooling and Containment Integrity Are Assured 1
J 1
~
1 4
PRIMARY ELECTRICAL FAILURE .
i
!
- Failure of 4-KV Feeder Cable from Auxiliary Transformer "C" I
- Primary Cause Aging of Cable i
i
- Contributing Factor l
Local Heat Source i
PRIMARY ELECTRICAL FAILURE (Continued) e Corrective Actions Test All 4-KV Cable Prior to Startup Replace 4-KV Feeder Cable from All Transformers Prior to Startup Protect 4-KV Cables from Heat Sources Prior to Startup Implement Programs to Assure the integrity of All Plant Cabling' Implement Programs to Address Plant Material Condition
i l
i j
SECONDARY ELECTRICAL FAILURE -
3 i
i e Failure of Automatic Breaker Sequencer
) to Complete Cycle
)
- Primary Cause Design of Output Signal from the Generator Overspeed Protection Device
!
- Corrective Actions i
j Correct Generator Overspeed Protection j . Device Design inadequacy Prior to Startup Provide Second immediate Access Offsite Power Source During Next Refueling Outage
. _r_
l l
l s
l DESIGN RELATED ISSUES i
i j
- Feedwater System
- Electrical Power System i e Plant Modifications l
1 i
i f
l I.
, PLANT MODIFICATIONS l
l
- Feedwater System Modifications New Check Valves' i
Flow Control Valve Closure ;
l Modification' Add Redundant Feed Line Check i Valves' Install New Sloping Feedwater
! Piping to Steam Generator "B"'
i
' Complete Prior to Startup i
i.
- 2 14 -
i
PLANT MODIFICATIONS (Continued )
l
- Electrical System Modifications Replace 4-KV Feeder Cable ~from All Transformers' Protect 4-KV Cabling from Heat Sources' Correct Generator Overspeed Trip Design inadequacy' Provide a Second immediate Access Offsite Power Source l
l ' Complete Prior to Startup
i PLANT MODIFICATIONS (Continued)
- Other Modifications l
Eliminate Spurious " Red Phone" Ringing on LOP' l
Correct Technical Support Center Computer ;
Unavailability During LOP DC Control Room Clock' Install Steam Generator Blowdown isolation' l
Install Steam Generator Blowdown Indication in Control Room' :
Eliminate Spurious Safety injection Signal on LOP'
' Complete Prior to Startup 1 -
l ,
4 I
i 1
i 4
i j
PROGRAM RELATED. ISSUES 4
- Operator /STA Performance 1
- Material Condition of the Plant i
l l PRE-TRIP l OPERATOR /STA PERFORMANCE Electrical Ground Developed into a Fault l
i
- Primary Cause '
Ground Not Quickly isolated
i PRE-TRIP OPERATOR /STA PERFORMANCE (Continued)
- Corrective Actions Increase Operator Experience in Troubleshooting and Manipulation of Plant Electrical Systems Increase Effectiveness of STA in Responding to Abnormal. Non-Emergency. Plant Conditions Improve Ground Alarm System to Provide Indication of Transformer "C" Ground Condition Upgrade Status of 4 KV Ground Isolation Procedure to Abnormal Operating Instruction
l I
l POST-TRIP .
! OPERATOR /STA PERFORMANCE
- Operators Understood Plant Behavior and Response and Maintained Plant in Controlled Condition Formality. Attention to Detail. Consistency
- and Team Effort Training Utilized to Properly Respond
PROGRAM RELATED ISSUES
- Operator / STA Performance
- Material Condition of the Plant l
l
~
l J-4 MATERIAL CONDITION OF THE PLANT i
e Material Condition Review Program
- Area Monitoring Program I
l t
i i
MATERIAL CONDITION
- REVIEW PROGRAM I
I
- Objective Assure Reliability of system Operation Upon Return to Service
<me 9
m
_33_
! MATERIAL CONDITION l REVIEW PROGRAM .
i (Continued) e Scope l
Identify items to be included 1
Detemine Material Condition of items l
Determine Material History of items Evaluate Acceptability of Material Condition of Plant and Adequacy of Operational R'eadiness. Including i Any Special System Testing or Performance Testing
- 3 11 -
1 AREA MONITORING PROGRAM i
l
- Objective Maintain a High Material Condition _
- Standard in Operation to Provide an -
Appropriate Physical Context for All Work i
AREA MONITORING PROGRAM .
l (Continued) e Scope Divide Plant into Distinct Areas Monitor and Evaluate Effectiveness of Existing Work Processes to Maintain Material Condition Standard 3
Revise Existing Work Processes as -
Necessary
l I
i i
t l
I
- ROOT CAUSES OF EVENT .
4 1
- Check Valve Failure Due to Turbulent Flow Combined With inappropriate Application i
l
- Auxiliary Transformer "C" Feeder Cable Failed Due to Aging i
i l
G kkkkkkkkkhhhhhhhhhh)h G 9/35 g
'G TRANSMITIAL TO: [
/ Documnt Cbntrol Desk, 016 Phillips s:
k ADVANCED CDPY TO: / / The Public Docment Ibczn l
$ p.g .
alxla SECY OPS BRAtOI cc: C&R w/attachs.
g ,
(w/o SECY g papers?
Attached are copies of a Oormission rneeting transcript (s) and related meeting docunent(s) . They are being forwarded for entry on the Daily Amession List and placernent in the Public Eccument Focxn. No other distribution is requested or .requirai. Existing DCS identification nurters are listed on the individual
- documents wherever known.
Meeting TitleMr Oh hu MM L$n-A E. M m
. 1
)
h
- it:; 5 w-0+ Kw Mre I M Meeting Date: 3llR f 8(, @ )( Closed s
% DCS Copies (1 of each checkai)
Iten
Description:
Copies Advanced Original May Duplicate To PDR , Document be Dup
- Co=v*
- 1. TPANSCRIPr 1 1
.. hhen checked, DCS should send a ,
copy of this, transcript to the .
LPDR for: Q \)p L ss (cu s
o, k s s. .
Sch , MLA
- 2. .
$e
=
- > 3 A
- 4. .
(PDR is advanced one copy of each document, *
- Verify if in DCS, and two of each SELY paper.)
- Change to "PDR Available."
kkYbYbYbYbbY$bYbYbYbb YlYlYlYb bYl I bbYb iYbYlYY