ML20196K132
| ML20196K132 | |
| Person / Time | |
|---|---|
| Issue date: | 03/07/1988 |
| From: | Advisory Committee on Reactor Safeguards |
| To: | |
| References | |
| ACRS-T-1648, NUDOCS 8803150186 | |
| Download: ML20196K132 (196) | |
Text
,
,- R 2 \lA _
UNITED STATES O NUCLEAR REGULATORY COMMISSION IN THE MATTER OF: )
)
)
)
)
)
RELIABILITY ASSURANCE )
SUBCOMMITTEE )
(OPEN MEETING) 1 LOCATION: Washington, D.C. PAGES: 1 through 161 I i
DATE: March 7, 1988
--............--....-===-======..======.=============r'==
HERITAGE REPORTING CORPORATION OficialReporters
, s.w., so..
O '22. t s, Washington, D.C. 2 5 1 (2 2) 628-4888 D\
8803150186 370307 PDR ACRi T-1648 DCD
1 PUBLIC NOTICE BY THE t
()
2 UNITED STATES NUCLEAR REGULATORY COMMISSION'S 3
ADVISORY COMMITTEE ON REACTOR SAFEGUARDS 4
5 i 6
7 The contents of this stenographic transcript of the 8 proceedings of the United States Nuclear Regulatory 9 Commission's Advisory Committee on Reactor Safeguards (ACRS),
10 as reported herein, is an uncorrected record of the discussions 11 recorded at the meeting held on the above date.
12 No member of the ACRS Staff and no participant at '
)
13 this meeting accepts any responsibility for errors or 14 inaccuracles of statement or data contained in this transcript.
15 16 !
17 l
18 19 ,
20 21 '
22 1
23 24 '
25 Heritage Reporting Corporation O (202) 628-4888 -
i.
9
1 j( ) 1 UNITED STATES NUCLEAR REGULATORY COMMISSION 2 ADVISORY COMMITTEE ON REACTOR SAFETY 3 SUBCOMMITTEE ON RELIABILITY ASSURANCE (VALVES) 4
)
5 In the Matter of: )
)
6 )
)
7 GENERAL MEETING )
)
8 Monday, 9 March 7, 1988 10 Room 1046 1717 H Street, N.W.
11 Washington, D.C. 20555 12 The above-entitled matter came on for hearing, 13 pursuant to notice, at 1:00 p.m.
(
~
14 BEFORE: MR. CARLYLE MICHELSON 15 16 ACRS MEMBERS PRESENT:
17 MR. JESSE C. EBERSOLE Retired Head Nuclear Engineer 18 Division of Engineering Design Tennessee Valley Authority 19 Knoxville, Tennessee 20 ACRS COGNIZANT STAFF MEMBER:
21 RICHARD K.' MAJOR .
22 23 24 25 O Heritage Reporting Corporation (202) 628-4888
2
/^) 1 PROCEEDINGS
%)
2 MR. MICHELSON: The meeting will now come to order.
3 This is a meeting of the ACRS Subcommittee on Reliability and 4 Assurance. I am Carl Michelson, temporary chairman for this 5 subcommittee, since Charlie Wylie was unable to be here today.
6 The other ACRS members in attendance are Jesse Ebersole, and 7 Chet Siess will be here tomorrow but not today.
8 The subcommittee will review current industry and 9 NRC staff efforts relating to valve reliability. This meeting 10 is intended to be complementary to previous meetings held in 11 calendar year 1985 on this subject.
12 Richard Major is the cognizant ACRS staff member for 13 today's meeting, and he sits on my right.
O,
(,) 14 The rules for participation in today's meeting have 15 been announced as a part of the notice of this meeting that was 16 published in the Federal Register on February 23, 1938. This 17 meeting is being conducted in accordance with the provisions of 18 the Federal Advisory Committee Act and the government in the 19 Sunshine Act.
20 It is requested that each speaker first identify 21 himself or herself, and speak with sufficient clarity and 22 volume so that he or she can be readily heard.
23 Jesse, do you have any comments before we get 24 started?
25 MR. EBERSOLE: I have none to start.
Heritage Reporting Corporation (202) 628-4888
3
('T 1 MR. MICHELSON: Since I am not the normal
\_/
2 subcommittee chairman, I do not know that I have any particular 3 planned comments. Charlie might have. But this meeting is a 4 follow-up, as I mentioned here, from the previous meeting.
5 Quite a few things have happened in-the valve business in the 6 last year, and we wanted to become better informed. And we 7 also thought that it was an opportunity for the industry and 8 nuclear community to have a chance to hear what was going on 9 also.
10 There may be occasions during the next day and a half.
11 in which we might want to solicit opinions or comments from 12 members of the community, if they so wish and are available.
13 So the meeting will be a little bit informal._ We will keep it
() 14 formal enough hopefully to stay on schedule somewhat, and that 15 is the main thing.
16 MR. EBERSOLE: Carl, I might have something to say in l 17 retrospect here. As you all know, I am sure, our business has !
~
18 gotten into the probablistic risk assessment area, and is !
l 19 heavily involved in that. And when one stops to think about a 20 nuclear plant, in the context of where trouble is, it is mostly 21 in fluid systems, and the active parts of those are pumps and I 22 valves. And valves are a main source of interesting 23 difficulties.
24 One of the things that causes our PRAs to be 25 suspicious in their conclusions is that they use a great deal O Heritage Reporting Corporation (202) 628-4888 2
i
_. ___ __ _ l
'_ ~
l N
4
\
(} 1 of valve reliability based on data collected over what they 2 call a successful repetitious operition' of valves when in f act 3 the valves are not loaded in any particular way at all, 4 sometimes not even at full normal flow, just going back and 5 forth closed.
6 Yet the valves when called upon to do their critical 7 thing and save the plant by isolation or sectionalizing systems 8 will be operating under mass flows a great deal higher than the 9 normal conditions. And I hope that these new measurement 10 techniques will at least give us a closer approach to what the 11 real reliability of valves is when they are working under 12 severe duress rather than just normal conditione. We will be 13 looking for some sort of help in that direction.
() 14 MR. MICHELSON: There will be some discussion of that 15 tomorrow on what the state of the art might be.
16 Is Mr. Encleball here today from NUMARC?
17 MR. COLVIN: He will be here in a moment. i 18 MR. MICHELSON: Okay.
19 MR. COLVIN: Do you want me to go ahead and start? ,
I 20 MR. MICHELSON: Are you representing hin? j I
21 MR. COLVIN: Yes, I am. i 1
22 MR. MICHELSON: Okay, yes. If you will represent 23 them then, I would appreciate it.
24 MR. COLVIN: All right, sir.
25 MR. MICHELSON: Be sure and give your name and b'
'- Heritage Reporting Corpo. ration (202) 628-4888
S affiliation, and so'forth.
.(} 1 Each ripeaker will have to do that 2 to keep the record looking good.
3 'MR. COLVIN: My name is Joe Colvin, and I am the 4 chief operating officer and executive vice president of 5 HUMARC. And my comments are going to be very brief and kind of 6 just. set.out an introduction for theLspeakers who wil1~come'up ,
7 and describe some of the industry initiatives that we have had l j
8 underway.
9 We thank the subcommittee fer the opportunity and the 10 interest to provide some overview or industry efforts in valve
, 11 reliability improvements. We as an industry and the utility 12 industry have been very active in establishing initiatives to I 1 ;- reduce plant problems caused by failures of valves. And these
() 14 initiatives really have been in support primarily of 15 improvements that will increase safety, reliability, and 16 capacity of our units. i n
17 Of primary interest to us all have been the problems l 18 with motor operated valves and check values. And.ao a result i 19 of that, I think that our comments will focus primarily on ,
j ;
20 thone areas and initiatives.
i !
j 21 NUMARC's role in motor operated valves and in valve l 1 22 initiatives has been one of more of a facilitator trying to {
I s
{ 23 bring to, .er th. 'ocus of industry resources and the efforts, i t
l
! 24 and t- to aca: -
'ome effective implementation and i
! l l 25 corro . - , ,
'1roughout the industry. l l
i 1
, . age Reporting Corporation l (202) 628-4888 l
- 1 l
. . . . . , , , -, ,-- - a
i l l
6 l
[} 1 2
NUMARC relies primarily on INPO, EPRI, and the Owners Groups, and the utilities themselves to develop and implement l 3 solutions to the problem. I think that you will see today as 4 we proceed that there are lots of cooperative efforts in 5 progress, and there is a lot of overlap between INPO efforts, 6 Owners Group efforts, EPRI efforts, and those of the industry 7 at large. And hopefully, that will come out as a part of the 8 presentation today.
9 I might go back in history just a Alttle bit from 10 NUMARC's standpoint. NUMARC really became active in 11 initiatives in valves with,the problems that were illustrated 12 in the area of check valves. And it was through Vic Stello's !
13 comments to NUMARC leadership in the late 1985 time frame that
() 14 NUMARC became active.
15 And as a result of the problems in check valves, the ;
16 four Owners Groups formed an ad hoc working group that ended up 17 being chaired by Hal Tucker, then chairman of the B&W Owners 18 Group, and they led the industry initiatives in check values in I f
19 cooperation with EPRI and INPO. l I
20 And NUMARC played a I hate to say silent role, t
21 because that is not an accurate representation, but NUMARC j 22 plcyed a monitoring role to ensure that those initiatives were I 23 being focused properly, and would help the industry solve some 24 of the problems with check valves. And you will hear a lot ;
i 25 more about that today, and 7. will not dwell on some of the ,
- i l
Heritage Reporting Corporation (202) 628-4888 l
l 1
7 1 details.
[
2 The other major initiative that we have undertaken in i
3 valves has to do with motor operated valves. And that again 4 came at the urging of the Nuclear Regulatory Commission in the 5 letter that Vic Stello sent to Warren Owen,-then chairman of 6 the NUMARC steering committee. That was NUMARC as a committee 7 and not a permanent organization as we are staffed today.
8 But we became involved in that as a result of the
- 3 AEOD report on motor operated values, and agreed with a number 10 of conclusions in that report, and agreed with a number of the ,
11 recommendations for corrective action. And as an industry, we 12 have implemented a numbsr of initiatives trying to correct some 13 of those problems. And we will identify those, and speak to .
() 14 you today in more detail.
i 15 We did brief the staff, the NRC staff, on 16 September 1st of 1987 on a number of these initiatives, and we 17 will provide a little update on that as we proceed today.
18 I would just like to say that I think that the focus 19 of our remarks will be on a number of problems related to 20 valves, both motor operated valves and check valves. And again j 21 we are focusing on the problems. I think through INPO's 22 evaluations and through the work that EPRI has done, there are 23 a lot of utilities who are maintaining and have designed their j 24 systems well, and they are maintaining their valves, and are
- 25 not having many problems. And I think that is true across the 4
l
() Heritage Reporting Corporation (202) 628-4888 4
i 1
8
- 1 board,
[}
i 2- But again,.we are focusing on the negative side on .
3 some of the problem areas. And as a result of that, I think j j
4 that you will'see that we do have some initiatives that will l
- 5 help correct these across the board. I guess that I would like 6 to go ahead with your permission. !
7 MR. MICHELSON: I have a couple of questions first. i 8 It is probably premature today, but I would like to make a, .i a
I 9 statement on it anyway. Motor operated valves have received a 10 great deal of attention from the community. And many good
- 11 techniques have been developed for a good understanding of how 12 they operate and how to verify that they are operating l 1
13 properly.
! () 14 So they have been a focus of a lot of attention. But
} 15 not much attention, at least that we have heard about, seems to 6
16 be focused on air operated valves, for instance, arid [
17 hydraulically operated valves which are far fewer in number, i
18 but still performs very sensitive functions. And of course, i i
! 19 the check valve, we have heard some about. And I think that it e l
l 20 is getting attention, but we have not heard about the same rate (
- 21 of progress that I have heard about in the case of motor l
$ l 22 operated valves. !
23 I wonder if you might want to comment just briefly [
! 24 about initiatives beyond the MOV that we have all focuse d on l 25 now for the last two or three years, and that I think are !
Heritage Reporting Corporation i
' (202) 628-4888 :
i l
t i
a
. . ~ . - ~ . . _ . . - - - - . .- - . .
i 9
t 1 making extremely comforting progress.
[ }-
2 MR. MICHELSON: Yes, sir, I would like to. I can
{
3 give you a coupis of comments on that. And hopefully, our j 4 speakers who are presenting for both EPRI and INPO can give you-5 any insights that they might have on that.
6 From the NUMARC standpoint, we have no initiatives 7 specifically underway to address air operated valves or 8 hydraulic valves. We do have initiatives underway to address 9 the loss of air system generic issue that is being studied by 10 the staff, of which the air operated valve failure is a part. i 11 EPRI also has work underway on the air system issues that I am 12 sure deals with some of the problems of air operated valves.
] 13 In the initiatives that we have addressed on l
i
() 14 motor operated valves to date, the first area that we focused 15 on happens to be Limitorque operators and I think the SMB 16 triple lot, because that is where we have seen a lot of the l 17 problems.
l 18 The initiatives that we have-planned as an industry 3 19 would go further to address or will go further to address other l l
20 operators, and then expand into other types of design and I l
. 21 manufacture. So I think that the focus of our attention to i l 22 date has been on those areas where we could get the biggest j
23 bang for our buck and make the most impact.
9 24 MR. MICHELSON: Do you have an approximate I 25 appreciation for the percentage of motor operated valves that i l
Heritage Reporting Corporation l (202) 628-4888 i
i
- =
10 1 are Limitorque in the nuclear plants?
[}
2 MR. COLVIN: I do not. Perhaps one of the other 3 speakers could give you some insight into that as we proceed.
4 MR. MICHELSON: Who has the next largest population?
5 I suspect that it is Rotorque, but I do not know.
6 MR. COLVIN: That is correct.
7 MR. MARSTON: It is my understanding that you are 8 probably dealing with between 80 and 90 percent Limitorque, j 9 probably closer to 90 percent. And Rotorque is the next.
I 10 MR. EBERSOLE: I would like to get into perspective 11 the importance of valves by a few very I think graphic examples 12 of what can happen if they do not work right. When this )
13 committee was established many years ago, the popular thing was !
() 14 a loss of a pipe, you know, in the main LOCA. And over the 15 years, it turned out to be such a low probability that it is 16 well handled anyway by the containment, and it took 25 years to 17 realize that that is not where all of the trouble was.
18 The trouble is outside of the containment. And I 19 will pick a particular example, and I will identify about three 20 kinds of pipe out there that can create havoc. The one that I 21 can think of 20 years ago was the ten inch main steam line on a 22 boiler at the Browns Ferry plant.
23 If one invches the type of pipe break outside that 24 value, the exterior valve, you have an interesting 25 configuration. The immediate condition might be that the
'T (k / Heritage Reporting Corporation (202) 628-4888
E 11 1 environment out there would preclude the operation of the
(' }
2 external valve, because it was not designed for a hostile 3 environment, and yet there was one there now.
4 So whether it worked at all or not would be in 5 question. And then the main thing is there are two valves in 6 ceries, one is AC and one is DC. If they have environmental 7 failure at the first and then just a random at the second, you 8 have continuity flow from the standing steam line into the 9 equipment areas of the plant. And in short order, it will 10 almost certainly kill the operating equipment which restores 11 the water, the water boiling out of the core out into the 12 buildings. And.the plant degenerates as far as I am concerned 13 to a total loss of all three units. Why? Because of valve
() 14 failures.
15 So valves get to be right at the top of the list 16 along with control rods to keep the plant safe. And I just 17 want to enhance your view of how important valves are. You 18 know, and to tell you that you are under some heat to make them 19 work.
20 MR. COLVIN: Yes, sir. I think that we do recognize 21 that. And I think that when Ted Marston from EPRI will make 22 his presentation that he has r.ome interesting statistics on the l l
23 valves and the percent that relate to safety related problems ;
24 and so on. I do not want to steal any of Ted's thunder. l l
25 MR. MICHELSON: I think that we can proceed. j O Herttage Reporting Corporation j (202) 628-4800 l
l
-p-
12 ,
. 1 MR. COLVIN: The next speaker would be Ted !!arston, l 2 director of engineering and operations at Electric Power 3 Research Institute. Ted.
4 MR. MARSTON: Thank you. Ladies and gentlemen, good 5 afternoon. As Joe had mentioned, I am Ted Marston, director of j 6 engineering and operations, a department within the nuclear 7 power division of EPRI. And I certainly welcome the 8 opportunity to come and discuss our R&D program on valves with 9 the ACRS. I believe that Boyd Brooks was here last year and .
l 10 made a presentation, a detailed presentation. He is the 11 principal manager of our valve work.
12 (Slides shown.)
13 MR. MARSTON: That is who I am, and what I hope to !
14 cover. What I would like to do this afternoon is to put in 15 perspective our research and development in support of the 16 valve issue in the industry. I would like to address several l 17 perceptions on the issue itself. And certainly, in support of [
18 what you said, Jesse. !
19 But in addition to the safety related, it certainly 20 is a significant economic consequence to the utility. So there 21 is tremendous ir.Lentive for the industry to work on the valve i 22 issue.
)
23 After that, I would like to give you a very brief l
24 sort of philosophical statement on how we structure our R&D, l l
25 what kind of objectives we hope to reach with doing that, and j i
l O Heritage Reporting Corporation ]
(202) 628-4888
13
(} 1 then how we implement that R&D in terms of determining what is 2 to be done, going into the plants to ensure that it is 3 feasible, practical, and economical, and then finally in the 4 technology transfor section which is always one of the more 5 difficult processes in the RAD industry.
6 And I would like to end up with a discussion of what 7 we believe to be some of the root causes of the valve issues.
8 And it is a very interesting perspective on the root causes.
9 Finally, I would like to give you a listing of the 10 available reports that are related to valves, and highlight 11 just a few of those to show how the ,rocess in our R&D has in 12 fact been started and completed and the technology transferred.
13 So without further adieu, in valve problems, I would i
(
(3) 14 like to talk about contribution and impact. Every two years, 15 we conduct a ste.-ly conducted by the S.M. Stoller Corporation to 16 look at what sources of unavailability and lost capacity factor 17 has the industry experienced in the last two years.
18 And the latest report which covers the 1985-1986 time 19 frame indicates that valve malfunctioning in all categories 20 results in approximately a 2 percent loss in U.S. capacity 21 factor. Now this is total valves, and th'at translates to about 22 a $340 to $350 million loss in capacity for the industrv. So 23 as I mentioned before, the industry has a great incentive to 24 work on their valve problems.
25 MR. EBERSOLE: What about the coal and oil burners, b\~
4 Heritage Reporting Corporation (202) 628-4888
i ;
14 !
(} 1 do they not lose some valves there, too?
MR. MARSTON: Absolutely. And unfortunately, our j 2 3 divisional discussion does not cover the coal combustion 4 4 system. We have a lot of internal. But our systems are so 5 different and so much more complicated than theirs are, that a 6 lot of these do not apply.
7 What is an interesting statistic is that
}
8 approximately 10 percent of the in plant-valve populations 9 contributes to this loss. It is not all valves that are 1
10 contributing. So it is 10 percent.
11 And I guess that the message that I would like to l 12 point out here which is true for us supporting the industry as l 13 well as both in the safety and also the capacity side, that we !
() 14 ought to focus our resources on those valves that are causing ;
15 the problems. I 16 MR. MICHELSON: That 10 percent, is it approximately [
1 i 17 the same valves in each plant that are causing the problem, or 4
i 18 how do I interpret 10 percent, or are you going to tell us 19 later? j 20 MR. MARSTON: I am not going into great detail, but I 4 i
- 21 can certainly supply the report on this. !
i ;
j 22 MR. MICHELSON: Is it a particular group of valves !
i ,
l 23 that is giving most of the trouble., or does not that group
{
j 24 represent 10 percent of the population?
25 MR. MARSTON: Yem. But unfortunately, you cannot say
(
Heritage Reporting Corporation (202) 628-488n .;
i
15
{} 1 2
that it is all motor operated valves.
valves in certain applications.
It is motor operated And it depends on the 3 supplier, whether you have a PWR or a BWR, and also to a 4 certain extent to the architect-engineer who did the original 5 design work.
6 MR. MICHELSON: The 10 percent then applies to air 7 and motor operated?
8 MR. MARSTON: This is all valves.
9 MR. MICHELSON: All valves?
10 MR. MARSTON: All valves.
11 MR. MICHELSON: Okay.
12 MR. MARSTON: Now of that 10 percent which 13 contributes to this 2 percent loss in capacity factor,
() 14 approximately one-third of those are in safety related systems.
15 So again, we are talking about over 3 or maybe 3.5 percent as 16 being safety related applications. That is not to say that it i 17 is an insignificant amount, but it is an amount that the effort :
18 should be focused ..
19 And as was stated earlier, tha major events are often 20 low probability, but have significant consequences. So again, 1 21 the conclusion here is that the industry, and the regulators, 22 and everyone else really should focus on those valves that are 23 contributing as they affect safety and capacity.
24 To give you an overview of the approach that we have 25 taken since the company started in 1973 on valve related i
Horitage Reporting Corporation j (202) 628-4868 ;
l l
16
() 1 research, I would like to give you a discussion on the 2 objectives and the potential benefits to the industry.
3 It certainly is to improve plant safety. There are 4 those applications of valves that do contribute to plant safety 5 or Catract possibly from plant safety. But also important is 6 improved plant availability. Because a safe plant is a plant 7 that is reliable, available, and economical. Those all go 8 together, and tt.ey are inseparably tied.
9 Reduced radiation exposure of personnel. As you well 10 know, valves, if you ask a maintenance individual, particularly 11 mechanical, he will tell you that valves are their number one 12 maintenance headache. And it contributes in certainly those 13 areas on the primary system to the additional radiation 14 exposure.
15 We have done studies on looking at the contribution 16 on valve maintenance. It is a significant part of that. But 17 also, valves are a significant contributor to the activation 18 within a system. If you look within a boiling water reactor, 19 either valve maintenance or wear of the hard face in the cobalt 20 bearing stellit is a major contributor to the activation within 21 the system. And that has brought about stimulus within the 22 industry to look for cobalt free hard facing replacements.
23 So again, the radiation exposure comes from two 24 sides. One, the maintenance on the valves, and also the 25 contribution to the activity of the system.
(')
~'
Heritage Reporting Corporation (202) 628-4888
1
... 1 I
I 17 l
() 1 And of course, reducing the O&M expenses for the i 2 plants, the margin over other forms of generatien that has been 1
3 enjoyed by the nuclear business is eroding. And in 1987, the j i
4 variable cost differential between coal and nuclear is really l
5 insignificant, almost the same. l 6 So there is strong incentive to reduce operation and 7 meintenance costs. The o&M costs have risen over the last five f
8 years about 60 percent. And that is not to say to sacrifice
) 9 safety, but just to make sure that the operation and 10 maintenance resources are targeted to the maximum amount. ,
1 11 MR. MICHELSON: Just so I can understand where it 1 l 12 might fit in. ;
lI i 13 I am wondering which is'the organization in this
]
i
() 14 nuclear community that is most closely coupled tn international
, 15 experience in valves as opposed to domestic experience, are you 8
j 16 looking at the international aspects at all? !
17 MR. MARSTON: Well, certainly, INPO has extensive f
18 international relationships. j
! 19 MR. MICHELSON: Should I ask my questions about l 4 i j 20 foreign experience and address it to them then? !
l l 21 MR. MARSTON: If you deal with the operators, the j l 22 operator utilities. Certainly, INPO has the principal i
j 23 responsibility there. We deal with mainly the international I i
! 24 research organizations. And of course, we would deal with CEGB i i i
- j 25 research and EDF research. ,
i I
( Heritage Reporting Corporation i 1
(202) 628-4888 ;
l 1
t 18 ;
MR. MICHELSON: You are getting input from foreign
(]) 1 2 experience as it might relate to the problems that you are !
3 looking at? !
4 MR. MARSTON: Absolutely.
5 MR. MICHELSON: And in what manner do you get that l l
6 input, and how do you get it?
7 MR. MARSTON: Through active exchanges, technical
! 8 exchanges, formal exchanges, informal exchanges. !
l i 9 MR. MICHELSON: Are you working closely with them? ;
l 10 MR. MARSTON: Very closely. In fact, the manager of ;
i 11 the maintenance department from EDF is on loan in my department !
l .
i 12 for the next two years. And I spent a year with EDF four years j 13 ago. l
( 14 MR. MICHELSON: The French, for instance, have been 15 doing a fair amount of work on particular air operated valves, l
16 for instance.
J 17 MR. MARSTON: That is a particular problens for them, l l
J 18 that is correct.
19 MR. EBERSOLE: There is a particular subset of these 20 valves that I wanted you to sort set aside and put particular 21 attention on, and that is those that handle the relief problem 22 and closure problem of the primary system.
23 MR. MARSTON: Yes.
24 MR. EBERSOLE: On PWRs, it is PORVs. And on boilers, 25 it is the extreme SRVs, the latter being valves that are l
O Heritage Reporting Corporation (202) 628-4888 .
l l
I i
_ , .--_ -. . _- - ~ . -. - . . . .
l l,
19 i 1 dependent both on hot circuit functioning sometimes assisted by
-(])
2 cir or whatever. In other words, there are interdependencies ,
I i 3 in there that lead you to wonder if they ought not to be !
l
- 4 simplified and made more positive in their function, since j i
5 blow-down on a boiler is absolutely critical if you are going l 6 to get water in it at high pressure.
7 The PORVs, as you know, are backed by block valves,- ,
4 8 and they are backed by safety valves. The SRVs on the boilers l 9 therefore are particularly significant, because of their very i l
- 10 vital role in a fundamental safety function. Blow down the i !
4 11 boiler, so that you can get water in it.
t j 12 MR. COLVIN: Yes, I understand. As you may know, j i ,
l 13 Boyd Brooks on our staff was one of the principal GE valve f 1
() 14 engineers. So he gioes me that message on a daily basis. That l i
) 15 is certainly true.
16 What I wou'd like to do now is to switch from the l 17 goals and objectives of the work to show you how we implement !
I '
j 18 that. The first, of course, and I will discuss some of the l 19 characterization work that we done in the pest, is to 4
J i l 20 prioritize and address the problem. The problems are more :
21 extensive than the resources available. ;
i j 22 Of course, those issues, and I stress the issues q l
)
23 rather than problems here, some of those blue chips, red chips, '
d 24 and white chips. And the key thing is to identify those high j 1
- 25 priority items to apply the resources. i 1 -
1 1
- Heritage Reporting Corporation !
- (292) 628-4888 j j !
t ]
1 i
I 20
(} 1 So after the prioritization effort, the next step is 2 to develop actions and guidelines that would address those 3 issues. The next step in that is to take those actions or 4 guidelinus, maybe it would be hardware procedures or just 5 guidelines, and take them out into the plant and test them.
6 Probably there would be an intermediate stage here 7 where we do extensive laboratory testing to make sure that if 8 we did ask a utility to try to say improve packing or something 9 like that, that we have a very high degree of confidence that 10 it in fact will work.
11 And then finally, the third stage is to in fact go 12 into the plant and try out some of the hardware procedures or 13 guidelines before we take it before the industry in general.
() 14 And I think that some key examples here would be the 15 main steam line plug, and 1 will discuss that in a moment. But j 16 when we did the evaluation of main steam isolation valves, we 17 found that one or the contributing factors to lost availability 18 and lost capacity factor was the inability to test for leak i
19 tightness by the very design of the pressurization scheme. So 20 that led to the development of this main steam plug.
21 Valve steam packing improvements. This is a problem 22 that has been identified with the valve I guess ever since they 23 have been valves. And so we have been actively involved here.
24 There was previous discussion on the application guide for 25 check valves.
- Heritage Reporting Corporation 1
(202) 628-4888
21
() 1 This was an effort that we did jointly with the 2 Owners Groups, all four of them. They provided the initial 3 funding, and we provided the program management and any 4 additional funds tt c..oplete the project.
5 MR. EBERSOLE: Now tell me, all four of those bullets 6 there discuss matters that are active trouble areas, your 7 leakage for instance. What I do not see up there is 8 investigation of the fundamental intrinsic capacity of the 9 valve to do its thing under duress.
10 MR. COLVIN: If I could possibly defer that question 11 for just a moment.
12 MR. EBERSOLE: Okay.
13 MR. COLVIN: I am not sure that I can give you a 14 satisfactory answer, but I will try to answer it on the basis 15 of what we have done and what we plan on doing.
16 And then probably the most important once we have 17 succeeded this step, the most difficult step of all is really 18 getting the technology transferred. And we use workshops, 19 seminars, and plant demonstrations. We have close interaction 20 with INPO and with the Owners Groups to effect as best as we 1 21 can technology transfer. I believe that either Ed or Gary 22 later will discuss some of these joint efforts with INPO.
23 But that is basically the approach that we have used ,
24 on all of our efforts. Before I get into the specific projects l i
25 which I hopefully will not spend too much time en, I would like O Heritage Reporting Corporation ,
- (202) 628-4888 )
4
F 22
{} 1 2
to give you the perception of my staff on the root causes of the valve problems.
3 And this is represented maybe not all-inclusive. But 4 I think that one of these, and this applies to many of these, a 5 failure to treat as an entity the valve, its operator, and its 6 specific application. And this is clearly for the motor 7 operated valve, whether it be electric motor, or air operated, 8 or hydraulic. It is to look at not only the application in the 9 system, but actually the installation in the plant, whether it 10 be vertical, horizontal, upside down, or whatever, and 11 beginning to reelize that this affects valve performance in 12 many ways.
13 And the next two items really have to do with an
() 14 exchange of information that is available. Failure to focus 15 existing knowledge, whether it be from this industry or others, 16 and problem solving on a case by case basis approach.
17 Hany of these problems have been experienced by 4
18 plants, and a lot of them have been solved very successfully.
l 19 The key job that I think that INPO has and we in a support 20 function is to get those successful applications transferred 21 from one utility to the next. And also the failures that l
t 22 people have had, to get that information out. And I think that 3
23 INPO is doing an outstanding job in that case.
24 And then finally, I think that we tend to look overly 25 critical at the design and performance requirements of these l (~T l k/ Heritage Reporting Corporation (202) 628-4888
23
(~) 1 valves. We are asking valves that were designed to isolate in L s-2 many cases to be leak tight. We may be asking a sery robust 3 valve to perform a fairly precise function.
4 And the job that we have at hand that these plants 5 were designed and built for is to go back with advanced 6 techniques to ensure that the valves as they are installed 7 really meet the functionality requirement of the design. We go 8 back with best estimate calculations rather than some very 9 conservative licensing type calculations.
10 And that covers the design, but the performance 11 requirements in a similar manner tend to be very restrictive.
12 Whether a valve opens in ten seconds versus twelve seconds, i i
13 does that really affect the performance of thi.s system in a 1
14 needed situation. And I think as an industry that we need to ,
t 15 go back and look more objectively at what we are asking the 16 valves to do.
I i 17 MR. EBERSOLE: A lot of the specs were grabbed from 18 thin air, because they were thought that they could be met, not 19 because they were needed.
20 MR. COLVIN: That is correct.
21 MR. EBERSOLE: So are you doing anything about that?
22 MR. MICHELCON: Yes. And I will identify several
- 23 projects that we have. And I know that the Owners Groups are 1
24 very active in that also.
- 25 But this is a key point. We have situations where we !
T d
i Heritage Reporting Corporation (202) 628-4888
, I t
24
() 1 xnow that the installed equipment probsbly does not meet nor 2 does it need to meet the sps-ification. So we need to go back 3 and reconcile that.
4 MR. EBERSOLE: I think that you are right.
5 MR. MARSTO11: That is more or less a philosophical 6 statement of our work. And I would like to go back and 7 highlight some of the valve related projects.
8 (Continued on next page.)
9 10 11 12 ,
13 14 15 16 )
l 17 j 18 1
19 20 j 21 22 l
1 i 23 1 24 l 25 Heritage Reporting Corporation (202) 628-4888
s i
l
- 25
{ (} 1 MR. MARSTON: In 1976 we asked EPRI to put together
- 2 what I consider a really -- an archival document where they i
, 3 assessed industry valve problems and this was published in ;
4 November of 1976. And this is excellent reading for anybody, '
4 i
- 5 because I think what it does say, it identified a number of !
d i
6 valve problems and issues that the industry has successfully j 1-7 addressed; and I think that is important. We have come a long i
8 way in many areas since 1976.
i 9 I would like to skip down to this. I had mentioned j I 10 this earlier that we looked at the effect of valve performance :
i I j 11 on unit capacity of BWRs and the leak rate testing of the l a >
12 tandem MSIVs was a contributing problem. !
- 13 So we got into initial work on isolation valves !
,1
() 14 looking at what were the different varieties available in the 15 industry, and eventually that ended up with the development of !
16 this main steam line plug to assist in the testing of the j 17 redundant MSIVs. l 1
! 18 Pointed out clearly in '82 was a valve stem packing ,
i 19 study; this is no surprise to anyone. Packing leaks have been 20 a problem around since valves have been in existence.
21 Safety and relief valve test program, this was the
) 22 Owners Group that was supported after TMI to look at the 1
i 23 performance of safety and relief valve; and a lot of the work 1
j 24 that was done was in support of the requirements came out of 1
- 25 TMI-2. I think that was a successful effort at that time. We 1
i j Heritage Reporting Corporation (202) 628-4888 i
S l
26
{} 1 are continuing that.
2 I had mentioned the cobalt contribution. In 1983 we 3 came out with a report that really identified the contribution 4 of cellite to the activities in the plants; it is significant.
5 We need to make every effort to get cobalt bearing materials 6 out of the primary systems, not only in valves but in the fuel 7 and control rods and anywhere else we can. We're really 8 successful.
9 This project with Westinghouse addressed the question 10 that was raised earlier. When we look at the real needs of the 11 plants versus the design needs of the plant, this was an effort 12 to go back and look at overpressure protection at the installed 13 equipment to see if we really needed to have these valves meet
() 14 these very, very stringent requirements; and the answer is, no.
15 That work is continuing. It is not a small effort by I 16 any means. But I think every time we look at this we see many 17 applications of over restrictive requirements on these valves 18 which forces us into undo maintenance, undo testing and undo 19 surveillance; and takes our efforts away from where we really 20 ought to out it in maintaining the valves for their 21 functionality.
22 In 1984 we looked at key valve prioritization, and 23 what we found out there was that we needed to work on motor 24 sperated valves; increase the level of effort on valve packing.
25 The next level was containment isolation valves, and then down O Heritage Reporting Corporation (202) 628-4888 l
N= , 4.rd i
27 *
() I the line. And air operator valves are clearly within the top 2 root thers.
3 These are simply projects that have come out looking 4 at set point testing of safety valves; interim studies on motor 5 operated valves; packing, et cetera.
6 What I would like to focus on now are the recent and 7 current projects addressing key valve issues. Valve motor 8 operator improvements, Doyd Brooks was here last year and I 9 think he discussed the work we did in that area in some detail; 10 and I won't review that. That is being incorporated in our 11 current studies which I will discuss in more detail.
12 Valve stem packing improvements I think was a 13 significant progress. We worked with Chesterton on the >
0 14 development of a dye form wet shaped graphite packing which we 15 found in the laboratory and will soon find in the plants. So 4 16 it gives you about 90 percent of the benefits of live loading, 17 without going to live loading. As you know in many 18 applications it is difficult to put live loading on some of the 19 smaller valves.
20 This effort started, not only because of the packing 21 leak problem but the desire to get asbestos out of the plants.
l 22 So it focused on the improvements of going to graphite. It 23 also gives guidelines on, if you want to live load your 24 packing, here are some guidelines; and there are a number of 25 commercial organizations that are providing these kind of Heritage Reporting Corporation j (202) 628-4888
28 l
/~)
V 1 services.
2 I would like to sk:.p down to here, application 3 guidelines for check valves. This is the work that was 4 supported by five organizationo the four Owners Groups and 5 ourselves. It was closely coordinated with the INPO work, as 6 Ed Moore will discuss in a moment. That application guide has 7 been reviewed. It was out in draft form in September of last 8 year and it is out for publication now. And if you all would 9 desire a copy of it, we can certainly make that available.
10 MR. EBERSOLE: I would like to see that and ask you, 11 does it have in it the criterion design features for coping 12 with the impulse loading when you have extremely rapid 13 closurec. -
() 14 MR. MARSTON: I'm going to have to give you, I think 15 so, but I am not certain, 16 MR. EBERSOLE: You don't know whether the valve will 17 keep on going when it shuts.
18 MR. MARSTON: It has been a while since I looked at 19 this guide, but my --
20 MR. EBERSOLE: The particular case is valves for BWR 21 boiling water main feedlines. If you go to hypothesize on 22 outboard failure.
l 23 MR. MARSTON: Between the time that I take my seat 4
24 and the next speaker is up, I will call back and get an answer 25 for you.
Heritage Reporting Corporation (202) 628-4888 l
29
() 1 MR. EBERSOLE: Well, it applies anywhere, whether you 2 are stopping forward or reverse.
3 MR. MARSTON: That's correct.
4 MR. EBERSOLE: Do they cope with the dynamic loadings 5 including shop loadings? It is incompatible with stalite j 6 ' seeding as you well know.
7 MR. MICHELSON: You do understand the question?
8 MR. MARSTON: Yes, I do. And I say, I don't know the 9 answer, but I soon will. !
10 What we have here is an item that is on your handout, 11 but was not on the viewgraph and it's a motor operator valve 12 repair guide that we are doing under the auspices of the 13 Nuclear Maintenance Assistance Center which is a new initiative O 14 we started up in September of 1987. And this is an effort to
- 15 collect the experience -- collected experience in the industry; 16 the successes, the failures, the problems, et cetera.
17 And included in that repair guide, let me just give 18 you an idea of what is included. As I had mentioned the 19 expa :lence --
20 MR. MICHELSON: May I interrupt.
21 MR. MARSTON: Certainly.
22 MR. MICHELSON: Is an RP report something like a 23 repair -- how does that differentiate from NP?
24 MR. MARSTON: My apologies. What this stands for in 25 EPRI lingo is research project. 2233 which is a O Heritage Reporting Corporation (201) 628-4888
. .. .- - . ~ . - - .. ._ . - = - . . _ . _ _ -
l 30 1 chronologically assigned number, so the lower numbers are older i
(])
i 2 than the new ones.
3 MR. MICHELSON: What does NP stand for then, which is 4 all the rest of your reports?
5 MR. MARSTON: Let me look at products versus process.
6 The research project produces reports. The reports are 7 identified by NP. These are the mother projects that produce l 8 the reports.
9 MR. MICHELSON: These are the procure --
10 MR. MARSTON: These indicate the reports are not on 11 the street and when they come out there will be an NP assigned 12 to that.
, 13 MR. MICHELSON: So we will expect to see these
( 14 eventually as NP reports with the same title, is that the idea?
15 MR. MARSTON: That's correct. For example, this is -
- 16 an NP, 5,067 on good bolting which really supports a lot of the 17 body to valve leaks. And this came out under RP whatever.
j 18 So an NP indicates a research report; an RP indicates ;
l 19 research project. ;
l 2
20 HR. MICHELSON: So it is still in a preliminary state l l
21 such as it is maybe available to some of the utilities or other i
j 22 interested workers, but not official by EPRI; ic it that the
'I 23 sort of thing?
i 24 MR. MARSTON: Well, for example, the application 25 guidelines for check valve is in fact out. I saw a copy of it,
]
I O Heritage Reporting Corporation
., (202) 628-4888 I
1
31 >
() 1 but it has just been out within the last couple of weeks.
4 2 MR. MICHELSON: As an NP report? !
i 3 MR. MARSTON: As an NP report.
4 MR. MICHELSON: Same number, just NP?
5 MR. MARSTON: No, it will have a different number, l 6 because we are in the 6,000 now. Those were assigned ,
7 chronologically, also. The MOV repair guide is drafted, it is 8 out for a second review and we expect to publish this in April i 9 of 1988. So we do not have a number on it.
10 MR. MICHELSON: If you can see such of these reports j 11 as you feel free to send us. We have, I think, probably all 12 the NP reports, but the RPs, at least to me, it's now. So we [
] .
13 would like to get them at such time as you are comfortable in
) 14 sending us those.
J !
15 MR. MARSTON: Certainly. Let me explain why this is 16 not an NP. There's an NP interim report which documents the 17 laboratory tests, the development work. What this final report
]
1 i 18 will document are the in-plant test. So that is why it is not 1 19 an NP. And the same with a valve motor operator. We have the
] 20 diagnostic control system going into Comanche Peak and Point i
21 Beach right now.
l
- 22 MR. MICHELSON
- My next question was going to be, is 23 that first document the micro-processor thing you were working 24 on or is it something else?
i 25 MR. MARSTON: There is an interim report that 1
j i () Heritage Reporting Corporation !
(202) 628-4888 l l d
i 1
32
'( ) 1 describes that. l 2 MR. MICHELSON: But this report is a couple years 3 old.
I 4 MR. MARSTON: That's correct.
5 MR. MICHELSON: That is the last time -- have you --
6 is somebody here today or tomorrow going to tell us what 7 happened to it?
8 MR. MARSTON: Well, I guess I'm going to tell you 9 because we went out, really, promoting this we wanted to make 10 certain that it would work in the plants.
11 MR. MICHELSON: Well, the last we heard you were 12 going to go to Duke and try it out.
13 MR. MARSTON: It is not easy to get use. I'm O 14 certainly not faulting Duke on this, they have been extremely i
15 cooperative, but we do have Comanche Peak and Point Beach.
16 MR. MICHELSON: Instead of Duke.
17 MR. MARSTON: That are installing -- and these are in 18 non-safety systems so we can make sure that we watch these like l 19 a hawk and they don't cause any adverse effect.
20 MR. MICHELSON: How many are you going to put in?
21 MR. MARSTON: I believe we are putting in four.
22 MR. MICHELSON: Four in each plant or total of four?
23 MR. MARSTON: Total of'four. And they are different 24 size valves, different size actuator 3. :
l 25 MR. MICHELSON: The same basic processor, though. l O Heritage Reporting Corporation 1 (202) 628-4888 !
33
() 1 MR. MARSTON: That's correct.
2 MR. MICHELSON: Now that will be described -- is that 3 2233 going to describe a little more about what you are doing?
4 Will I expect, perhaps, to see it in there if I had a copy?
5 MR. MARSTON: Yes. Now, what we have in this project 6 is the interim report. We will write this report when we 7 finish the in-plant test.
8 MR. MICHELSON: Now there has been a great deal of 9 activity in the industry, of course, trying to use various 10 devices, MOVATs and one of them you are well familiar with I'm 11 sure; and to what extent have you been monitoring that activity 12 and to what extent might that be described, whatever you are 13 doing in one of these reports?
14 MR. MARSTON: If I can skip down to this one, which I 15 think is probably the most significant in terms of our work in 16 MOVs, let me describe to you the kind of contents that are in 17 here.
18 We have a section on experience, the root causes and 19 the fixes. We have tear down repair, rebuild procedures.
20 Detailed set of procedures that have been reviewed by seven 21 different utilities that have various ways of diagnosing MOV 22 problems, where it be with commercially available equipment, 23 their own techniques, whatever.
24 We have a section that gives guidance on maintaining 25 equipment qualification for those operators that require it.
- Heritage Reporting Corporation (202) 628-4888
34
('T v
1 Guidelines on purchasing and stocking and inspecting spare 2 parts to make certain that the parts you want to install in 3 your operators are in fact what you get. There are post-4 maintenance testing guidelines.
5 So I would say that -- and it is heavily illustrated.
6 A lot of -- I won't say, JPAs, but it is a very -- for a person 7 like me it is a great help.
8 I think in that document you will capture 9 recommendations on diagnostic testing, post-maintenance testing 10 of valves.
11 MR. MICHELSON: Have you done any coraparative or let 12 us say, recent comparative evaluation of the various techniques 13 that are available for studying torque switches, and so forth,
() 14 because there is more than one commercial supplier of such 15 techniques. And have you done a comparative evaluation of 16 those, and is it in one of these reports?
17 MR. MARSTON: The answer is, no. We did an 18 evaluation early on that you are aware with MOVAT. And since 19 that time we have not done a head to head evaluation. That is 20 really not our --
21 MR. MICHELSON: I know it is not necessarily your 22 thing. I just wondered if you had done it, though.
23 MR. MARSTON: The repair guide is to give plants 24 guidance, and also, flexibility. If they wanted to pursue 25 diagnostic testing with brand X or brand Y cr whatever, the O Heritage Reporting Corporation (202) 628-4888 i
r
'l i l 35 l t
t
' () 1 idea is to make sure these are the attributes you' desire from I i 2 your diagnostic test. And to make sure that when you sign up j 3 with a company you buy the equipment that will meet these
)_ 4 requirements.
5 So the requirements are embedded in there, but it !
. 6 doesn't say that you will buy brand X because it is better than i ;
~
) 7 brand Y. We carefully avoid getting into that.
j' i
8 MR. MICHELSON: My other question is, we are going to 9 hear from TVA today or tomorrow, I guess it is probably 10 tomorrow -- today, on some problems that they had uncovered 11 because of the -- the problem of identification. You can have 12 SMV 000 and it can have many, many different kinds of j
]
i l 13 components within it without necessarily being fully aware of i
() 14 that from just looking at even the name plate unless you do
, 15 some chasing of the serial numbers and whatever. l
, 16 Is some of that sort of problems discussed in the !
J i 17 repair guide? j 3 18 MR. MARSTON: In fact, much of the document is ,
I
' l 19 discussed on that. And within the experienced-based there are l 4
j 20 cautionary statements that are highlighted in black that said,
- l I
21 you know, these parts can vary from one operator to the next. 1
. 22 MR. MICHELSON: So you try to inform the industry of l.
23 which ones to watch for in particular that could be
) 24 interchangeable and the ones you have to check.
i l 25 MR. MARSTON: Let me just discuss for a moment the l
j
() Heritage Reporting Corporation l j (202) 628-4888 ;
i I I
i
_ _ . . m . _ _ _ ,-._.-._..l
36
() 1 review group on this. We have hired a contractor to assemble 2 the information.
3 MR. MICHELSON: Which information are you referring 4 to?
5 MR. MARSTON: For this repair guide, for the MOVs.
6 MR MICHELSON: Thank you.
7 MR. MARSTON: On the review group we have seven 8 utilities and we picked them very carefully because they have 9 very active MOV programs. Some use brar.d X; some use brand Y; 10 some use their-own brands, to make sure that we really capture l 11 the whole industry.
12 We also have representation from Limitorge. We have 13 representation from the B&W training group, I think it is power C:) 14 safety, goes out and trains people on operators. We have two 15 valve manufacturers participating. So we think we have 16 captured the whole user population and they have all 17 contributed greatly to this effort. And you need that kind of 18 contribution, you can't work in a vacuum on this because of 19 these fine detailed differerces between one operator ard the 20 next.
21 MR. MICHELSON: Now, that repair guide is going to be 22 available in April; is it going to be possible for us to see a 23 copy at such time as you are comfortable sending one to us.
24 MR. MARSTON: Yes, sir.
25 MR. MICHELSON: Hopefully not beyond April.
O Heritage Reporting Corporation (202) 628-4888
37
() 1 MR. MARSTON: Yes.
2 MR. MICHELSON: We would appreciate receiving it.
3 MR. MARSTON: That is undergoing review right now and 4 we hope to make whatever changes are appropriate by the end of 5 March.
6 MR. MICHELSON: Thank you.
7 MR. EBERSOLE: In this area of repair and monitoring, 8 I remember many years ago identifying non-modulating valves as 9 really just a by-stable device, it's either opened or shut like 10 a chip, off or on. And there was never any evidence, and today 11 little if any, as to what the popular term is, margin of force, 12 with which you successfully executed a function. You not t
13 merely caused it to go to red or green, but you knew you could ,
( 14 do it with extra force to take care of further degradation of
, 15 the gear teeth or whatever, i
16 Are you getting into the realization that you need 17 evidence of margin of force as they did in the Salem case where 18 the breakers didn't open and consequently the plant didn't 19 scram. That's where that term came from, by the way.
20 MR. MARSTON: Yes. When this work started we thought 21 we could satisfy the needs in MOV by simply coming out with a 22 repair guide. It became very obvious that a certain portion, 23 whether it be 30 or 30 some odd percent, this related to the 24 initial design; the matching of the operator to the valve. And i
25 that is why we had this complimentary effort which will be out ,
l l
Heritage Reporting Corporation (202) 628-4888 i
38 1 in September to really look at the requirement side, not
(]}
2 necessarily how do you maintain the operator, but in fact, are 3 you maintaining the right operator.
4 MR. EBERSOLE: Well, it is nice to know when it a
5 closed that was the closure it could have ever made.
6 MR. MARSTON: Yes. That's correct. So I think we
! 7 have a balanced approach here.
8 That really ends my presentation.
9 MR. MICHELSON: Thank you very much, Ted.
10 Are there any other questions?
11 (No response.)
12 MR. MICHELSON: I think then we will proceed with
- 13 your next speaker.
( 14 MR. MARSTON: I would like to introduce from the 15 Institute of Nuclear Power Operations, Mr. Ed Moore.
16 (Slides being shown.)
i 17 MR. MOORE: Good afternoon, my name is Ed Moore, I'm 18 the Director of the Plant Support Division at INPO. This 19 afternoon Gary Peterson who is the Manager of our Maintenance l 20 Department and I will talk to you a little bit about some of l
i 21 the maintenance activities at INPO that deal directly with the ,
l 22 valve area that we are discussing.
23 My purpose here is to give you a little bit of 24 background, and then I intend to turn it over to Gary Peterson 25 who will give you a more detailed assessments in the area.
] Heritage Reporting Corporation (202) 628-4888 )
1
39
() 1 You will recall that we really got involved in motor 2 operator valves and in check valve area because of some very 3 significant industry events. Those events involved both motor 4 operator valves and check valves. And those events were 5 responded to by the NRC-in various different ways. And they 6 have been responded to by INPO through experience, 7 dissemination, through our SER, SOER, O&MR and also through a 8 lesser thing called significant even notification.
9 Just as a refresher type thing, often time when one 30 event occurs we will issue a significant event notification; 11 then that gets followed up with a significant event report. l 12 As the information develops we recognize that as a j 13 more significant event, we issue a significant operating O 14 experience report.
15 And just to refresh your memory a little bit on those 16 SOERs, those provide specific recommendations for the industry 17 to follow. We expect the industry to take those actions; and 18 then we at INPO follow up during our evaluations to make sure 19 that those actions have been implemented.
20 MR. MICHELSON: Do you follow-up on what the utility 21 is doing when you have issued an SER, but not an SOER?
22 MR. MOORE: We follow-up from the point'of view of 23 how the industry reviews and screens those SERs.
24 MR. MICHELSON: Do you look at the process by which 25 they were handled, but not necessarily how a particular one was O Heritage Reporting Corporation (202) 628-4888
. . . ~ - .. . - - - ._
l i
i t t
f 40
() I handled?
2 MR. MOORE: That's correct. ;
4 j 3 MR. MICHELSON: On the SOER, though, you really are 1 I
4 following up that particular SOER.
l+
5 MR. MOORE: We are following up that particular SOER. f 4
6 And typically, since an SOER is a follow-on from SERs, I mean, ,
)
i 7 typically when we review the SOER we find the action has ;
i
! 8 already been taken as a result of the SER, both utilities. We I 1' !
9 will see if the SER is appl'icable to them, and if it is, we'll I 10 proceed right away and take that action. That's what we l 11 normally find when we go out in the field. l 12 MR. MICHELSON: Now, the O&MRs, again, you just I 13 monitor the process by which the utility handles them; is that ;
]
() 14 right? !
i
~
i 15 MR. MOORE: I don't think we track those up as much. i
+
16 Those are really, you might say, lesson learned that we have 17 learned some lesson from a particular problem in the industry j 18 and it is just a quick mechanism to get that information out to 1
19 the industry. {
l j 20 So in terms of tracking to see if that action has 21 been taken, it is not that kind of a document.
22 I'll give you just a little more in this area. l l
23 Again, the other part of this is the evaluation. During .
^
i l 24 evaluations, as I mentioned, we do follow-up on SOER !
i !
- 9. 25 recommendations; and particularly, you might recall we l t (2)
] Heritage Reporting Corporation !
(202) 628-4888 !
i i
{
41
^T identified some 82 recommendations from the earlier SOERs that (J 1 2 we follow over a three evaluation cycle to mnke sure those are 3 followed up.
4 We are also following performance in -- part of that 5 is specifically the area of MOVs and check valves; we are 6 following up on those recommendations.
7 And, of course, again, as Ted Marston indicated 8 earlier, we do have a strong interaction and coordination with 9 EPRI and with NUMARC on these kind of problems.
10 MR. MICHELSON: Are you going to tell us a little bit 11 about your thinking, if any, at this time on the air operated 12 valves and hydraulically operated valves?
13 MR. MOORE: Well, I could tell you that quickly
() 14 because we really haven't pursued those as a generic type 15 issue. Now, we will pursue those if, in terms of a specific 16 problem.
17 MR. MICHELSON: That is the next question, have 18 you -- another part of INPO is, I guess, watching operating 19 experience, of course, and have they ever highlighted air 20 operated valve problems as a particular problem as opposed to 21 having highlighted MOVs early in the game? Maybe somebody --
22 is there anybody here --
23 MR. MOORE: I don't really know the answer to that 24 question. But certainly, not to the same degree that we are 25 looking at MOVs at this particular time. j i
Heritage Reporting Corporation (202) 628-4888 i
'1 i
1 42 3 4
() 1 MR. MICHELSON: So, as far as you know, at the 2 present time you are not attuned to it.
3 MR. MOORE: Whether we have had an SOER or SER in the 4
4 air operated valves?
5 MR. COLVIN: To my knowledge there has not been an 6 SOER specific to air operated valves. There have been several 7 SERs and SOERs written on air system problems, yet also relate 8 to valve problems. But I don't recall one specifically, but we 9 can check into that.
10 MR. MICHELSON: I just wondered what the feeling of 11 the industry was about, is there really a problem even out 12 there on air operated valves. The valves in many cases are the 13 same valves, it's the ope ator only that is different. And O 14 certainly, many of the same kind of questions can be asked ;
15 about over torquing and stem cacking, and things of this sort.
16 And in using an air operator is there something magic that it 17 just doesn't have problems like motors have or, you know, what i
18 is the experience telling us, if anything, on this area?
19 I can't find many people that seems to feel that they 20 have got any experience that says there is a problem. But it 21 is intuitive that there could be similar problems in certain
- 22 aspects of valves. I don't see them jumping out.
23 MR. MOORE: I guess I would say this in that point of 24 view and that is, in out v.uclear plant reliability data system j 25 the information is reported on safety related equipment,
, Heritage Reporting Corporation (202) 628-4888
43 m
(_) 1 important to safety equipment. We do track failure rates. We 2 have developed a component of failure analysis system that lags 3 failures based on rate.
1 4 If we were seeing significant failures on air 5 operated valves, those would jump out as a problem and then we 6 would take a solid look at it.
7 MR. MICHELSON: And to your knowledge, you are not 8 aware that they have been jumping out as a problem?
9 MR. MOORE: That is correct.
10 MR. MICHELSON: Thank you.
11 MR. MOORE: Today's review in our area we are going 12 to talk specifically about MOV, initiative and progress at INPO
- 13 and check valve activities today at INPO.
() 14 Let me just show a couple of slides that begin to go 15 back to the kind of experience. And I do this because, as you 16 recognize operating experience as one of the cornerstone 17 programs at INPO. So there has been a lot of activity over the 18 years dealing with motor operator valves.
19 I guess one of the points of interest here is, you 20 know, kind of the ratio of significant event reports to 21 significant operating event reports; so a much more significant 4
22 report when these come out. You kind of get a feel for that.
23 Again, a similar slide for check valves which shows, 24 again, we have been working to make sure that the information 25 that we have available to us is getting out to the industry so Heritage Reporting Corporation (202) 628-4888
44
() I they can take appropriate action on it.
2 Let me just --
3 MR. MICHELSON: One of the problems with check 4 valves, of course, is you don't necessarily even know that they 2
5 are a problem until something happens as a result of them being i
I 6 out in the open position.
7 MR. MOORE: That's true.
8 MR. MICHELSON: In other words, during normal plant 9 operation you may not even be aware, and you don't have good 10 diagnostic methods, that we have heard about yet, maybe we will 11 hear some before we are finished tomorrow. But check valves 12 can't be comforting. Maybe they are a more serious problem 13 than you realize.
Q
' 14 Have you got any thoughts on that aspect of it?
15 MR. MOORE: Well, we will talk a little bit about i
16 what we are doing in the check valve area as we get into the ,
17 problem. Gary Peterson is going to be talking about that.
18 I guess I can say, we are looking at that area now,
< 19 and really working to make sure that check valves will go to 20 work when they need to.
21 I just wanted to take a minute here to say that we 22 are talking specifically about valves and we tend to talk only 23 about motor operator valves or check valves or some other kind 24 of valve. But there is a lot of things going on at INPO that 25 are there to promote good performance in the maintenance area.
Heritage Reporting Corporation (202) 628-4888
i 45 (j 1 And consequently, since they are there to make, 2 let's say, a better performance in maintenance, they also have
- 3 a direct bearing on having better valve performance because we 4 do our maintenance better, j 5 The NPRDS system, nuclear plant reliability data 6 system which I mentioned just before, that data is collected 7 and analyzed in many different ways by us and by the NRC and by 8 others to identify generic type failures.
9 One of the ways that we do it is a component failure 10 analysis system which we call CFAS; that system applies seven 11 failure rate tests to components to raise a flag if we have 12 significant failure rate.
13 And then that information is picked up by our
' O 14 evaluators that go out in the field to see if they can get to l
15 the bottom of why those failures are occurring.
1 1
16 MR. MICHELSON: Does that keep screening the total l
17 NPRDS data base on valves to spot one when something -- when a 18 trend is starting to develop; is that the idea?
j 19 MR. MOORE: Yes.
20 MR. MICHELSON: Do you do double data base?
i 21 MR. MOORE: That is not a continuous operation. It 22 is done on demand.
23 MR. MICHELSON: I mean, it doesn't turn through the j 24 entire historical data base on a particular subject?
25 MR. MOORE: I'm not sure if its the whole data base O Heritage Reporting Corporation (202) 628-4888 i
l 46
() 1 or it's more frequent -- it's more frequent information that we 2 have in the system.
3 MR. MICHELSON: I assume it is looking at the total 4 population.
5 MR. MOORE: I think that is one of the tests, but 6 also, there's other tests which I think really focuses on it 7 more.
8 MR. MICHELSON: I wasn't aware of that. Is that 9 written up anywhere of how that works, in the NPRDS manuals?
10 MR. MOORE: I don't think it is.
11 MR. MICHELSON: I don't recall seeing it.
12 MR. MOORE: We are just about to have that in an 13 interactive mode so the utilities can call that information up.
14 MR. MICHELSON: It's under development, then, you are 15 saying?
16 MR. MOORE: Yes, it is.
- 17 MR. MICHELSON
- Thank you.
18 MR. MOORE: Our evaluators take that information and 19 go out in the field and see if that is real. Some times those 20 failures are based on data reporting. As an example, because 21 of our emphasis on motor operator valves, people going in and 22 doing a lot of testing, you see a lot more failures reported in 23 motor operator valves because of the maintenance we're 24 performing on them, they identify problems.
25 MR. MICHELSON: Well, they're identifying valves that
(
Heritage Reporting Corporation (202) 628-4888
l 47 l
() 1 really have failed; you just weren't aware that they had j 2 failed. You know, we are in a state such as they may not be 3 able to perform the function required.
4 MR. MOORE: You may find valves that have same 5 problems, not necessarily that they wouldn't perform their 6 function. .
7 (Continued on next page.)
j 8 ,
j 9
} 10 11 l
12 1 13 i
() 14 ,
i 15 16
} '
- 17 l j 18 i 19 i
, 20 l
j 21 l
22 i I i
23 !
1 i
24 I
25 l
([) lieritage Reporting Corporation (202) 628-4888 i
a l
i 48 MR. MOORE: Let me move on. You are probably aware
(]) 1 l 2 of the industry self-assessment initiative and maintenance that 3 we initiated last March a year ago which involved asking each i 4 utility to do a self-assessment against the maintenance 5 guidelines. Those assessments are well down the path at this 6 point in time.
7 And part of that initiative was the establishment of 8 maintenance assistance review teams which enabled us to take a i 9 focused team out and work with the utilities, the utility staff 10 being part of that team, to focus on parts of the maintenance 11 area identified, the base line, the maintenance performance, 12 the problems that needed to be corrected, and to develop option 13 plans to fix those problems, i
() 14 MR. MICHELSON: Are there teams that just go out on 15 MOV problems, for instance?
16 MR. MOORE: Gary will get into that more. But these 17 teams look at motor operated valves on their two week visit on 18 site. And we also have individuals that go with the evaluation l
19 team to look at motor operated valves, and our evaluation teams !
i 20 themselves look at motor operated valves. So we are looking at 21 it in several different ways.
22 And I only mention this because this has been a very 23 effective initiative in terms of being able to work with the l 24 plant and identify what the problems are, and to develop action 2 .". plans to fix those problems.
() Heritage Reporting Corporation l
(202) 628-4888 l I
l I
49
() 1 Moving on down, accreditation, accreditation in the 2 maintenance area. Most programs, maintenance programs, are 1'
3 accredited at many of the utilities at this point in time. But 4 you recognize that that applies initially to new employees 5 coming into the company and continuing training.
6 And so there is a time constant before the effect of 7 that accreditation and the upgrading part of that will affect 8 the entire utility operation.
9 MR. MICHELSON: Let me give you an example of 10 something that I have been concerned about, and then you can 11 tell me how you can handle it. Some of the diagnostic 12 equipment that has been used to look at motor operated valve 13 performance requires a pretty fair amount of skill and judgment
) 14 on the part of the person who is doing the monitoring of the j 15 output in order to properly interpret what it is saying.
i 16 How does the industry assure itself that adequately
]
- 17 prepared individuals are interpreting this data, so that it is .
i t l 18 properly utilized as opposed to having some incompetent i 19 individual who might misinterpret it and make things worse than 1
20 they already are, how is that handled?
1 21 MR. MOORE: Are you going to address that, Gary?
22 MR. PETERSON: Yes.
j 23 MR. MOORE: We take a look at the training program, j 24 and Gary will get into that in more detail. But we have some
, 25 of our individuals trained at INPO which gives us a better warm l
I i ($) Heritage Reporting Corporation *
(202) 628-4888
(
l 50 i
1 feeling I guess on how well those people know their jobs.
i
({}
I 2 But accreditation again, some of the utilities are
, 3 also bringing their own maintenance people into the program ,
i 4 with the new people. So they are expediting bringing those ;
I 5 people up to speed in terms of the accreditation effort.
6 Evaluations, there are a number of things in
! 7 evaluation that I have already mentioned. Operating experience 8 review of specific recommendations. We look at preventive 9 maintenance, and we look at machinery history. We look at l 10 post-maintenance testing.
i 11 And I also mentioned here root cause analysis.
12 Because in about the last year, we at INPO have been focusing i
13 more on making sure that the utility does not just put a l () 14 bandaid on a problem, but they really dig down to find out what ,
! 15 the causes of the problem are.
I 16 MR. EBERSOLE: Maybe that is the point where I should ;
j 17 say what I need to say. I have found that INPO like all ;
4 i i 18 operating organizations has a kind of religion that precludes 19 it from being critical of design. In other words, they will 20 operate anything that is handed to them to operate and be proud 21 of the fact that it can operate it even as how lousy it is. ;
22 Are you going to ever, and you certainly were not a 23 year or two ago, begin a program of criticizing the designs and 24 putting heat on the design suppliers?
l 25 MR. MOORE: I am not quite sure how to answer that, '
- i 5
1
() Heritage Reporting Corporation l (202) 628-4888 I i
),
e f
4 l 51 !
() I sir.
! 2 MR. EBERSOLE: It is a root cause problem. I am in t
3 that area. And certainly, many of your problems, you are going 1
j 4 to find that the root cause is that it was not made right in ,
- l j 5 the first place. ,
~
6 MR. COLVIN: I was going to try to get Ed off the !
7 hook. He is with INPO, and I am not. But one of the things 8 that INPO does in its review of operating experience is that 9 they do look closely at whether design was a contributor. And ;
10 in those areas where there is a root cause analysis that i
11 indicates a design contributor to the risk or to the event, 12 they identify that. And their recommendations in the SOERs do 13 require a look at the design and appropriate modifications ,
i
() 14 being taken, ,
15 Now in the area, INPO is also looking in other !
16 initiatives outside of the engineering area, to look at design ;
) 17 created problems. And they are looking at the NPRDS data base, 18 and operating experience, and other inputs, and working with i J ,
i l 19 the vendors, triple S suppliers and others, to try to identify
. 20 those areas where design is a contributor, and look for ways of :
i I 21 improving.
) 22 But I think that the original intent at INPO was to 23 look at the area that was within the plant's ability to 1
! 24 control, and that typically did not include the original design 25 and design basis. Maybe that helps the explanation.
l i Heritage Reporting Corporation (20') 628-4888 e
i
i f
i 1 52 MR. EBERSOLE: I need to understand why they could
({) 1 l 2 not loudly complain about the design and do so.
lI 3 MR. MOORE: I certainly acknowledge your comment. i i
4 And that is we as operators take a great deal of pride in being i
] 5 able to make it operate. ,
! 6 MR. EBERSOLE: As a matter of fact, the harder it is i
1 7 to operate, the more pride you take in operating successfully. [
i i
! 8 MR. MOORE: But the biggest specific exemple in the i 9 motor operated valve situation in terms of a hydraulic lock, we 10 have been talking with EPRI and working to try to get to the ;
11 root of that. And unless I am mistaken, and maybe Ted can 12 correct me, but EPRI is now passing that problem to Limitorque l
- 13 saying you as the designer need to get on this problem and fix !
( 14 it.
j 15 MR. MICHELSON: You mean the grease problem in the
[
16 springtime, is that the one that you mean? j l
17 MR. MOORE: Yes. We cannot even get agreement on i
! 18 whether it is some people who have the problem and some do not, !
19 because we need to get agreement on whether it is a real i
) 20 problem or not, or whether it is something else.
1 21 But we are working much harder on this. This is kind 22 of a new area for us, but we are working much harder to squeak 23 away those layers and get down to do that. And one of the I
24 things that this particular group has been very effective in l I
d 25 doing is to dig down to the bottom and find out what is really i !
l l
j
() Heritage Reporting Corporation 3 (202) 628-4888 j i
l
?
53 !
()
l 1 causing some of those maintenance problems. So we are gaining i
- 2 some expertise in that, but we are not where we want to get ;
l 1 3 yet. ;
1 t i 4 We have started a new look at material condition {
J !
5 assessment. And let me say that we have been looking at i
! 6 material condition for a long time. But the normal things that
- 7 you would find, water on the floor, or steam coming out here, ;
8 or leaks, those are pretty much getting fixed at this point in !
~
- l 9 time.
10 And so we are looking at some of the more important i
l 11 equipment to see what is its performance record, is it really ;
I I
I 12 ready to operate. And so we are starting to take a much harder l i 13 look at those areas. !
I 14 In the area of technology transfer that Ted talked f
, I
! 15 about, we are moving toward and we have been werking very hard l I 16 in the last year or two to make sure that that information gets l 17 out to the utility down at the working level where the people !
18 can use it. !
a
! 19 As you know, many times we will send information to 20 the executive point of contact and it may never find itself .
l j 21 down to the maintenance manager down at the working levels.
I i 22 But we are working at the maintenance manager level and below
! 23 to make sure that they understand what information is available 1
l 24 and how to get it. As a matter of fact, we provide them copies i
j 25 many times.
l I i
C) I 1 Heritage Reporting Corporation l l (202) 628-4888 l J
l i
?
54 O 1 oerias eveteettoa , e worx aera to erovide
- 2 assistance to people while we are on site. And I guess that we i
i i 3 think that we are doing a lot better at that. And I guess that s i l 4 you know that we are holding a lot of workshops which we bring l 5 all of the people together and do a lot of information exchange j 6 including the technology exchange. ;
l 7 MR. MICilELSON: The area of material condition j
! 8 assessment, did that include the problem of say corrosion of 9 the bonnet bolting from boric acid and that sort of thing, are 10 you looking into that type of problem as well?
11 MR. MOORE: Yes, we are. If you look through our 12 reports, you will find that we frequently have concerns in our 13 evaluation reports abcut boric acid corrosion.
14 MR. MICllELSON: Yes, because we lose too many bonnet 1
15 bolts not being fully away. You might lose a bonnet, which ;
.i 16 would be kind of an interesting event.
]
17 MR. MOORE: Sometimes it is hard to tell whether 18 somebody has como around the day before we arrive and cleaned j 19 everything up.
20 MR. MICHELSON: That also can be a problem, but you
! 21 are fully attuned to the importance of the bolting corrosion
] 22 question?
23 MR. MOORE: Yes, sir.
- 24 MR. EBERSOLE
- What is the corrosion rate that I I 25 heard, Carl, is it a half inch per year or what?
j IO Heritage Reporting Corporation (202) 628-4888
)
55
() 1 MR. MICHELSON: Well, no. On carbon steel, depending 2 on the bolting. Of course, what it gets around to is that we 3 have to use different bolting materials. And carbon steel is a 4 nice bolting material, but it is not so good.
5 Do you have a comment?
6 MR. KARSTON: Yes. I think that the approach would 7 be to stop the leaks as much as possible. And then if we 8 cannot, to resort to a different bolting material.
9 NR. MICHELSON: In terms of stopping leaks, are you 10 looking then at gasket materials as well?
11 MR. MARSTON: Gasket packing.
12 MR. MICHELSON: Because that bolting corrosion could 13 be potentially a very dangerous situation. Because we have not 14 designed these plants necessarily for the loss of bonnets as 15 missiles. It just has not been a design basis. He just worry 16 about nice solid pipes breaking, but we are not worried about 17 bolting nails coming off and bonnets flying through the air, 18 which could be very serious. Also the leak before break does 19 not work too well on bonnets. They may not ever have a leak 20 before they give.
21 MR. MOORE: That is all that I was going to present.
22 Doctor, if you have no further questions, I will introduce 23 Gary Peterson.
24 MR. MICHELSON: Jesse, do you have any other 25 questions on this material?
O Heritage Reporting Corporation (202) 628-4888
56 MR. EBERSOLE:
(]) 1 I have none.
2 MR. MICHELSON: Okay. Then Gary will be next.
3 MR. MOORE: Gary Peterson, who is the manager of the 4 maintenance department at INPO.
5 MR. MICHELSON: Ed, before you leave though, could 6 you tell me, if there were to be a focal point for valve 7 problems at INPO, if there was one, would you be it, do you 8 think that you are the focal point for valve problems, or is it 9 kind of spread around and there are many depending vn what 10 problem I name, that maybe there are different people?
11 MR. MOORE: It it spread around a little bit, because 12 it gets involved with us in the maintenance area, and it gets 13 involved with our technical support area, and it gets involved
() 14 in our operating experience atin.
15 MR. MICHELSON: How does INPO try to integrate; 16 basically, if there is a valve problem out there, how do they 17 Integrate all of their activities to focus on valve problems, 18 is there any attempt to set up a project leader or something on 19 valves, a lead engineer or something on valves, as opposed to 20 just letting each area handle its own aspect?
21 MR. MOORE: It is interesting that you ask, because 22 we have developed about a year ago a list of what we consider 23 to be some of the more significant equipment problems. And in 24 that listing, then we identified the lead department at INPO 25 and the individual responsible for promoting that kind of O Heritage Reporting Corporation (202) 628-4888
57
() I coordinaticn.
2 MR. MICHELSON: How did valves come out when you did 3 that?
4 MR. MOORE: Well, we had motor operated valves and
, 5 check valves on that list, for example.
6 MR. MICHELSON: Who was named the lead department?
7 MR. MOORE: Well, the maintenance department has the 8 responsibility for motor op -*ed valves.
9 MR. MICHELSON: .
.ou the focal point then?
10 MR. PETERSON: F, nOVs.
11 MR. MICHELSON: For MOVs. If I have a MOV problem .
12 and I wondered what INPO was doing on it, you we~ld be the name 13 who would tell me? '
14 MR. PETERSON: That is right.
15 MR. MICHELSON: You do not do it all, I realize, but 16 you are sort of the focal point?
17 MR. PETERSON: That is correct.
18 MR. MICHELSON: Okay. Well, good. Thank you.
19 MR. PETERSON: Thank you, Ed.
20 (Slides shown.)
21 MR. PETERSON: I would like to get a little more 22 specific on what INPO is doing in the motor operated valve area 23 and in the check valve related activities this afternoon. This 24 is just a briel summary. I would like to go over the INPO 25 initiatives and. the rotor operated valves initiative.
O Heritage Repo rting Corporation (202) 628-4888
_ . __ . ~. __ . . _ _ . ..- _ .
58
( )- 1 I would like to go over the INPO initiatives,-the 2 initiative results that we have seen to date, and our future 3 plans in this area.. And then in the check valve area related j 4 activities, I would like to discuss briefly again the reports 5 of the industry that we are putting out, our evaluation and
~
6 field activities in this area, and also industry interactions !
7 related to check valves. !
\
8 As you gentlemen all know, the motor operated valve 9 problems have been surfacing over the past several years. The 10 two most significant events, at least that I am aware of,-la 11 the Davis-Besse lost of auxiliary feed water, which was the 12 failure of the valves to open after closure, the containment I
13 isolation valves.
14 And then in San Onofre Unit 3, the rapid cool down 15 initiated by check valve failure, but motor operated valves 16 contributed to this, and the shutdown ~ cooling valves'were not 17 fully closed. And they were leaking,-although they indicated .,
18 closed. l 19 We have been addressing, the industry has, addressing 20 MOV problems over the past several years through INPO reports 21 that Ed touched on there for a minute through our evaluations 22 and assistance visits. And then, of course, the NRC AEOD 23 review in December 1986 again highlighted the problems with the 24 motor operated valves.
25 After that-AEOD review, we again looked at our data
/
s
()' Heritage Reporting Corporation i
l (202) 628-4888 f 4
f e
, b... . . , - - . ,.--,7_ _ _ . . . /,,,_- , . - _ , ,
1 59
() I through NPRDS and our look of MOV performance confirmed that 2 there were still problems in the industry in this area.
3 We'therefore embarked on an issue last year to 4 improve the performance. The plan contained several elements I l
5 in its performance in conjunction with EPRI and NUMARC. We 6 have been coordinating especially closely wich EPRI on this. )
7 The action plan contained several elements. We l u
8 started with a letter to senior utility management assigned by 9 the president of INPO to highlight the MOV problems, and the 10 importance of resolution of these problems; this letter was 11 mailed in September of 1987.
12 It basically asked the industry to look at those i
13 three SOERs Ed previously mentioned and to implement those 14 recommendations.
15 It also asked industry to perform post-maintenance 16 testing after any maintenance that could affect operability.
17 And it asked them to implement diagnostic testing to determine 18 the root cause of valve operability problems.
19 Finally, it asked them to periodically conduct 20 assessments of their own MOV performance. ,
1 21 In addition to that letter we did additional follow-22 up more enhanced looks during evaluations, more time spent in 23 both the evaluations and also these maintenance review and 24 assistance teams that Ed mentioned.
25 In preparation for that we had a utility that we felt O Heritage Reporting Corporation (202) 628-4888
60
() I had a very good MOV program come down and give training to many 2 evaluators at INPO, not only the maintenance department, to get 3 a good understanding. I think between 30 and 60 people f
4 attended that. They actually shipped.in some valves as large 5 as 500 pounds on carts. We disassembled them. Went over-piece 6 by piece, the typical industry problems that were associated 7 with these components.
8 MR. MICHELSON: This was done in Atlanta?
9 MR. PETERSON: Atlanta, that is correct.
10 We have also put on many of the evaluations an extra 11 evaluator to look at the MOV program for two to three days.
12 And the extra evaluator we have been using is two individuals 13 we have, are fortunate of having right now, has extensive MOV Cs 14 experience in the industry, five years or more. They are very 15 knowledgeable of the inner workings of the valve.
16 During these evaluations they review the SOER 17 implementation in-depth; look at their program, that sort of 18 thing.
19 MR. MICHELSON: Just as a slight side light, there 20 are certain aspects of the design and functioning of the motor 21 control center associated with a given valve that are quite 22 integral to how the valve itself may properly function.
23 Do you people look at motor control center problems l 4
24 as a part of an integrated view of motor operator valves? l 25 MR. PETERSON: I guess I personally am not aware of O Heritage Reporting Corporation (202) 628-4888 l
l 61 the particular problems you are talking about.
({} 1 2 MR. MICHELSON: Well, a simple example, the therma 3 overload device within the motor control center has been the 4 focus of past concerns about the proper functioning under 5 accident conditions and so forth. You don't want it -- since 6 it wasn't, perhaps, a sophisticated device as one would like, 7 there was suspicion that maybe -- there was suspect, and 8 therefore it was bypassed during ECCS signals.
9 Have you gone back, for instance, as a part of your 10 looking because we have had some motor operator failures as a 11 result of not having such therma bypass devices available.
12 Have you done any kind of rethinking of whether that was really 13 a good idea and if you felt it was a bad idea, then you would
() 14 probably start some kind of a dialogue with the NRC as to why 15 it ought to be changed? Do-you go in then to the motor control .
16 center in that way to make sure that we haven't missed 17 something along the way?
18 MR. PETERSON: As I remember, that incident came up 19 one time and it was whether they were bypassing therma 20 overloads during that particular condition. The details of 21 that I am going to have say are a little fuzzy at this point.
22 I can get them for you if you need it, but right now I don't
. 23 have them off the tip of my tongue.
24 MR. MICHELSON: I'm thinking more in a programmatic ,
l 25 sense. l O Heritage Reporting Corporation (202) 628-4888 l
.i I
62
() 1 MR. PETERSON: On a programmatic -- since we looked 2 at it, we concentrate more at the valve and the diagnostic 3 testing of the valve. If, incidentally, to the evaluation 4 those types of things come up, and they do look at prints 5 specifically -- things we want to know in particular is whether 6 the limit switches also are wired separately from the torque t
7 bypass switch. And as a result of that they sometimes get into 8 the motor control center.
9 MR. MICHELSON: I was thinking more from the 10 viewpoint of reliability of valve operation. An integral part 11 of that reliability question has to be, is it getting power 12 under the right circumstance.
13 MR. PETERSON: We have addressed the issue, not more ;
14 from that particular aspect.
15 To go on we have continued support of the EPRI 16 nuclear maintenance assistance center which Ted previously 17 mentioned. And in this we are on the Techn_ cal Advisory Group 18 that is reviewing this repair guide. Ed Moore is also on the 19 Steering Committee which is the overall management committee 20 overlooking the NMAC center.
21 We have also -- we are on the committee to review 22 this guideline and have attended several meetings on it.
23 We have also, in conjunction with our action plan on :
24 MOVs, have visited plants that have good MOV performance to ,
25 gather information. This was completed in February of 1988 and
[
- Heritage Reporting Corporation ;
(202) 628-4888 l t :
. , , , -_ -. . _ . , . - . ._ - , . -- = ..
63 we are presently compiling the results of those two visits
{}) 1 2 along with the problems that we have found at other places.
3 We are communicating these problems to the industry 4 through work shops, and we have got one planned in October of 5 '88, maintenance superintendents workshop, maintenance managers 6 workshop. Also, discussions with industry representatives; and 7 again, that is EPRI, but it is also the maintenance peer 8 program.
9 Nearly every evaluation we go on we have a 10 maintenance manager or maintenance supervisor from another 11 station accompany us. And with us looking at MOVs 12 specifically, he gets the opportunity to look at that program.
13 MR. MICHELSON: How close do you work with the people 14 at Limitorge who are the most familiar perhaps with lubrication 15 requirements and lubrication affects on wiring and all these 16 other little things that from time to time have given a 17 problem? Do you have real close communication with Limitorge 18 or are you somewhat distant from them?
19 MR. PETERSON: Our communication there is more 20 through EPRI.
21 MR. MICHELSON: Well, you know, there are a lot of 22 interesting problems that you are seeing in the field that are 23 perhaps related to -- ;
1 24 MR. PETERSON: Well, for instance, on grease lock up.
l 25 MR. MICHELSON: Yes.
1
(~3
'/ Heritage Reporting Corporation (202) 628-4888 l l
64 MR. PETERSON: That's right.
(]) 1 I 2 MR. MICHELSON: I'm wondering even if whether we, in 3 some cases, have chosen the right lubricants and so forth.
4 MR. PETERSON: That's correct. For instance, on 5 that grease lock up, we attended the same meeting -- there was 6 a meeting in Washington on that, right here in fact, and I 7 believe Limitorge was there, certainly EPRI. Ted can help me 8 on that.
9 MR. MICHELSON: But really, that's the Limitorqe 10 problem. And I just wondered how, when you feel you have got a 11 manufacturer's problem in front of you what route you go to get 12 back to the manufacturer with the problem; and you are saying 13 you go through EPRI.
( 14 MR. PETERSON: Through EPRI or we call them directly 15 in some cases.
16 MR. MICHELSON: Oh, I'm sure you do.
]
17 MR. PETERSON: That's right.
18 MR. MICHELSON: But the real working relationship is 19 between Limitorge and EPRI, then; is that correct?
20 MR. PETERSON: That's correct.
21 MR. MICHELSON: Thank you.
22 MR. PETERSON: The last point I would like to make on 23 there is the sharing of EPRI products. And again, we have made 24 much more concerted effort in the past year or two to 25 distribute those things and make maintenance managers and i Heritage Reporting Corporation
- (202) 628-4888
65 ,
1 supervisors aware of what available information is out there to
({} 1 2 help them, particularly in these areas. l
.l 3 Other activities related to motor operated valves, in :
4 January --
5 MR. MICHELSON: On the same subject that just bears 6 the question on one other aspect that you run into and that is, 7 from time to time I'm sure from your experience with the 8 utilities you are finding that maybe maintenance manuals t
9 haven't been either written correctly or are such that it is 10 very difficult to -- very easy to do things wrong, easy to 11 misinterpret what perhaps the manual is saying.
12 When you have problems, just like simple things like 13 with the manual, do you then go directly back to Limitorge
() 14 telling them what you think the problem is and seeing what can 15 be done or how do you sclve such problems? Do you rewrite the i 16 manual yourself?
17 MR. PETERSON: No. I guess I need to have you 18 rephrase the question; I'm not quite sure I understand. Are 19 you talking a vendor manual?
20 MR. MICHELSON: Yes. These are the Limitorge !
21 supplied manuals, for instance, that tell the user now how to 22 do certain operations on the valve operator itself. What1 kind 23 of lubricants to use. How to remove this device or that 24 device. They are, in some cases, rather detailed manuals. If 25 you have problems with the procedural aspects, do you go to l Heritage Reporting Corporation j (202) 628-4888 i
r _ . _ __ . _ . . . , , - . . . , . . . _ . _
66
~
Limitorge directly or do you still go to EPRI and tell them
'({} 1 2 that the procedures aren't any good?
3 MR. PETERSON: We would really probably do both. A 4 related subject we have been working on is batteries, and we 5 have gone to the IEEE subcommittee on that. And we have called 6 the vendors when necessary. We also work through EPRI, so it 7 is a multi-pronged approach, I think.
8 I would say the primary means is through EPRI.
9 MR. MICHELSON: But in a case where 90 percent or 10 approaching 90 percent of all operators are made by one .
11 company, it sure seems that one would want to just go to that 12 company; there's no use to going to valve committees or that ,
13 sort of thing, you go back to the real source of the problem,
() 14 if there is a problem.
15 And I would think you would do it by the most direct 16 means possible.
17 MR. PETERSON: Which may be directly Limitorge.
18 MR. MICHELSON: Yes, without going through EPRI or
} 19 whatever, unless it's a research and development kind of a 20 problem as opposed to a problem with procedures or where to 21 place the lubricants or whatever.
22 MR. PETERSON: I understand what you are saying, but 23 the main reason we are going through EPRI is the repair guide 24 and application guide, and we want to make sure they are in the 25 communication chain. If they got left out there's a chance i
j O Heritage Reporting Corporation (202) 628-4888 1
() 1 that manual wouldn't be properly updated. 1 2 MR. MICHELSON: To pick up a number of these items in 3 the process..
4 MR. PETERSON: That's right.
5 MR. MICHELSON: Thank you.
6 MR. PETERSON: Sure.
7 As I started to say, other activities, in January a 8 representative of AEOD accompanied one of our evaluation team 9 and we were looking specifically at motor operator valves. He 10 is also out in the next -- during this two week period he is 11 also at another -- on another evaluation doing the same thing, J
12 so we can look at a little broader perspective.
13 In addition, one of the questions you previously :
) 14 mentioned is, how do we make sure that things come out the same 15 at INPO. And on anything related to motor operated valves.
16 There is a gentleman that works in my department that o
17 everything through the Institute, operating experience comes-18 through him, so that we have one unified approach. And again, 19 he is also the gentleman that is on the Technical Advisory 20 Committee through EPRI.
21 MR. MICHELSON: Would you care to name that 22 gentleman?
- 23 MR. PETERSON: I don't see any problems. His name is 24 Jim Tills.
25 MR. MICHELSON: Does he work for you?
O Heritage Reporting Corporation (202) 628-4888 j
l 68
() 1 MR. PETERSON: Yes, sir.
2 Some of the sample issues that are identified at 3 plants that we have seen with motor operated valves, and these 4 are not necessarily -- are not generic problems, they just 5 occurred at some of the plants that we have seen. You see 2
6 problems with preventive maintenance, with post-maintenance 7 testing after work on the valve. i 8 You see problems with their procedures, where the 1
9 procedures didn't necessarily accompany operating experience to l 10 date; with root cause analysis of various failures and why l
11 those failures are occurring. Then, things that I'm sure you 12 have heard about which are the torque switch set point l 13 controls.
O 14 MR. EBERSOLE: I want to ask a question before you 15 leave that. You talk about failures which, of course, are 16 realized, you really have them.
17 MR. PETERSON: I'm sorry.
18 MR. EBERSOLE: My concern is failures you haven't had j 19 yet when you were in real trouble and you couldn't afford to l l
20 have failures.
l 21 I saw not long ago a daily report, I believe, of some l 22 sort of analysis. Carl, you may remember it, where a part of l 23 it was they confirmed that the -- you know a vent V, don't you, 24 where you have the valves open between the high pressure and 25 low pressure side of the circuits and you feed either 1100 O Heritage Reporting Corporation (202) 628-4888 1
i
i 69 1
pounds or 2200 pounds into a 400 pound system, which is out in
(]} 1 2 the receiving room, and you do this when the valve is open and l 1
l 3 blow up the low pressure system and then degrade the service 4 facilities.
l 5 This particular event cited the case where they now 6 knew that the valve operator was small enough co it could not 7 open the valves when there was an excess differential. And 8 that therefore, protected against event. And that follows the 9 European practice, I think, of deliberately being sure you 10 can't open a valve if there is a pressure differential in-11 excess of that which would damage the low pressure system. ,
- 12 But as all things, when you do a good thing you do a ,
13 bad thing; and I immediately recall, well, ncw how do I know
() 14 that the valve will open against the pressurized when the pump 15 starts up instantaneously and develops a dead head pressure 16 against the gauge with the low pressure on the trouble side.
17 And I couldn't, you know, it's -- the dynamic picture 18 is very puzzling. It's a design problem, I know, but it never 19 seems to be very clear what margins you have got to do your l
20 thing.
21 MR. PETERSON: Well, one of those SOER 22 recommendations specifically addresses that, and that was -- we 1 23 are asking them to diagnose -- can't get the word out --
a 24 implement diagnostic testing to determine the root causes of 25 failures, but also that they need to do baselines so that they Heritage Reporting Corporation ,
(202) 628-4888 l l
i l
70 l
,, I V 1 know that they have got the proper --
2 MR. EBERSOLE: The capability.
3 MR. PETERSON: -- the capability to do that.
4 MR. EBERSOLE: Yes.
5 MR. PETERSON: That is part of that.
6 MR. EBERSOLE: Not that you do the daily thing, but 7 do the critical thing.
8 MR. PETERSON: So we are asking them to do that and 9 we are looking.
10 MR. MICHELSON: As the utility does baselining of a 11 particular component, they don't necessarily send you the data 12 of it, how are you even necessarily aware that they have done 13 it, in case you might be interested in getting the data, which f-14 I'm sure they would probably give you if you ask for it, at 15 least I hope they would. But how do you -- how good is your 16 communication to know what is really going on out at the 17 various utilities in terms of testing valves, for instance, 18 particularly for the viewpoint of full load testing where 19 possible.
20 MR. EBERSOLE: I think the operator needs to know 21 what the point, not only can they do, what it might have to do.
22 MR. PETERSON: I think I understand the question, we 23 are following up on motor operated valves probably even 24 stronger than we are on other things. But we do -- part of the 25 process that we use to follow-up is to many times watch field l
- ( ,
Heritage Reporting Corporation '
(202) 628-4838
, e . - - , -- - - ---.,c,- - - , , . - , - ---= - - - , , , ,--n, ,, - - - -
71
() 1 activities. We interview. We'll go out and watch a motor 2 operated valve maintenance job being performed.
3 MR. MICHELSON: I'm going to give you a more specific 4 case and then you can tell me how you handle it. A number of 5 utilities, I understand now, have gone back and done MOVATs 6 testing with near full load conditions on valves as they can 7 get by running a lot of water through it and so forth. And 8 they are apparently getting some rather interesting results, 9 some surprising and some not. But all very interesting.
10 I just wondered, are you aware of each and every one 11 of these cases or most of the cases or just a few of the cases 12 where this is going on; and how do you even become aware? -
13 MR. PETERSON: We are aware of what is going on in ;
) 14 the industry. We sample -- we look at their program from an 15 overall sense and see that they -- they know what they're 16 suppose to do.
17 MR. MICHELSON: Do you do that once a year roughly or 3 18 18 months or whatever.
19 MR. PETERSON: That's right.
20 MR. MICHELSON: That isn't going to help you much if 21 the data is already 18 months old. There has got to be some 22 good way of knowing close to the time that the work was done as
. 23 to what results they were getting in case they might be able to 24 factor in near programs to tell all the utilities. Or maybe
, 25 the individual utility spreads the word all over the industry, O
i Heritage Reporting Corporation
+
(202) 628-4888
72
() 1 I don't know, I haven't heard that is the case necessarily.
2 You are the instrument by which information, I think, j 3 is disseminated to many other utilities; and you're not aware 4 that the tests were done and what the results were, then how 5 would other utilities find out?
6 MR. PETERSON: The specific -- I'm trying to 7 understand the question totally. Let's just say there is 600 8 valves on a plant, and they are going to test all of these. >
9 The results of those, you know, if they have a generic problem 10 with failures, they would send that out on nuclear network for 11 other utilities to see. l 12 They would also report it on safety systems or some 13 systems through NPRDS.
O 14 MR. MICHELSON: These are not failures, j
15 MR. PETERSON: All right. That's tne question I'm 16 puzzling over. I don't quite understand, 17 MR. MICHELSON: Well, many of the utilities are l 18 testing all their major safety related valves using MOVATs or i 1
19 similar equipment. And there is nothing in the process that 20 says you have to put them under their most adverse loading )
l
- 21 condition when doing this test, but a couple of utilities, '
22 several in fact, have been, as I understand it, doing just 23 that, trying to put in very large loads, very high flows'and so 24 forth to get some good comparable data through these dynamic 25 positions.
O Heritage Reporting Corporation (202) 628-4888 i l
l l
l 1
l 73 1 I just wondered if you were aware of that, and how
(])
2 you become aware, and how you factor that information into your 3 perusal of the valve problem?
4 4 MR. PETERSON: The actual specific results they get i
5 on every test, we wouldn't know about them. We are asking them 6 to do that, however, as part of the SOER follow-up 7 recommendations and part of this letter that'went out. So we 8 are looking specifically when we go out, and we are follow-up 9 on the letter. We obviously can't, with our resources, go to 10 every place every month.
11 MR. MICHELSON: No, I wouldn't expect that. I would 12 expect, though, that you would somehow inform the utility l 13 industry that when they are doing such unusual tests that you j
() 14 would certainly like to be aware of it and get the data from it 15 to see how it might factor into a bigger picture. But I 16 thought maybe you are already doing that, but I'm not sure I'm l 17 hearing that is the case.
18 And you might become aware of it either through the i l
19 grapevine and I don't think they would normally put it on !
20 nuclear network just because they ran a test unless they ran l 21 into some really serious problems as a result of the test, then 22 it might get on the network.
I j 23 MR. EBERSOLE: To give you a little more specific 1
24 example, and Carl will remember this because he was the 25 detective that found it. The big 42 inch valves that control O Heritage Reporting Corporation (202) 628-4888 l
74 l
() 1 containment ventilation opening and closing, early in this game 2 it was found out that it wat.T't even in the procurement specs a 3 statement about these valves having operators with a dynamic 4 functional capability to close in the advent there was a 5 substantial flow through them.
6 MR. PETERSON: I remember that.
7 MR. EBERSOLE: There's a lot of interesting aspects 8 of that.
9 MR. PETERSON: That's right.
10 MR. EBERSOLE: So we stial have that problem. And 11 interestingly enough, I think the most predominate solution to 12 that was close them all or just crack them.
13 MR. PETERSON: Unfortunately, yes.
O 14 MR. EBERSOLE: Which illustrated the problem very
- 15 well, and I don't know what the distribution is now. [
f 4
16 MR. MICHELSON: There's still a large number of them j l 17 partly opened varying degrees; the results that we discussed at 18 this meeting about a year ago, it is time now for the NRC to 19 tell us how they are assured that they really know those valves 7 l
20 could still be closed.
21 MR. EBERSOLE: That's my next thing and my next 22 question. How do you assure yourself as INPO that these valves i 23 will close even a little bit, if they have to close.
24 MR. PETERSON: Even what, I'm sorry.
25 MR. EBERSOLE: A little -- the few degrees they are
(:)
Heritage Reporting Corporation !
(202) 628-4888 :
I !
75
() 1 now left to close, do they test them? _
Is it by analysis? Is 2 it satisfactory?
3 MR. PETERSON: Have we moved over to the containment '
4 purge valves now?
5 MR. EBERSOLE: Yes.
6 MR. PETERSON: Frankly, I'm not prepared to address 7 that issue today.
8 MR. EBERSOLE: This is the valve I guess that we are 9 talking about.
10 MR. MICHELSON: No, I don't think so, Jesse.
11 MR. EBERSOLE: You mean these are special valves?
12 MR. MICHELSON: I think when you talk motor operator 13 valves, rarely are they motor operated, but some of them are.
14 MR. EBERSOLE: Some of them are.
15 MR. MICHELSON: They are generally air pistons. Some {
16 of them are not. !
17 MR. EBERSOLE: Well, they are air -- NRC required 18 that they be leak tight if they could get them closed, but they ,
1 19 never did require that they close them in the first place. ;
! 20 MR. MICHELSON: When we get into air operated valve 21 discussions, then we will bring up those containment purge 22 valves and focus on what we know about a butterfly valve
- 23 operating under dynamic conditions. l 24 MR. PETERSON
- Anyway, I understand your comment, and i t
. 25 all I can say is, we will look at it.
() Heritage Reporting Corporation ;
(202) 628-4888 j 1
76
() 1 Our future plans in the motor operated valve area is 2 to continue the review of the MOV programs at plants during 3 evaluations and the maintenance review visits. And also, to 4 bring up these lessons learned at the maintenance managers 5 workshop to communicate specifically what we have. learned in >
6 this last several months. -
7 There is another thing that is not on here is that, q c
8 again, with the peer program going very strongly at INPO l 9 there's a lot of communications, it's a little informal, that 10 really is sometimes much more effective.
1 11 In check valve area there has been three significant l 12 operating experience reports; one in '82, one in '83, there in 13 October of '86. The first two dealt primarily of the aux 0 14 feedwater system and check valves were contributing to that, 15 but recommendations didn't specifically call out to look at all 16 the check valves.
17 However, 8403 in October of '86 certainly did that.
18 There has been another of INPO reports, as Ed previously 19 mentioned. In addition -- I'm sorry.
! l 20 MR. MICHELSON: How are you approaching the check r 21 valve problem as a program in INPO, keeping in mind that there 22 is maintenance, simple maintenance activity, but there is also 23 some kind of fundamental questions involved on why do we find
. 24 parts of valve weighing down the pipeline and so forth. Is 25 EPRI the one that goes back and tries to piece together those Heritage Reporting Corporation (202) 628-4888
l 77 i
1 stories to determine what happened or is that part of your !
('})
2 program, too, or just what? l l
3 MR. PETERSON: Well, piecing together what happened i 4 is the job of these experienced reports; and that comes out of 5 our operating experience department at INPO.
6 MR. MICHELSON: Do those people have to deal with the 7 LAR and then they talk to the utility, of course, but the 8 utility is not necessarily a good root cause seeker either, 9 particularly, when it might even involve metallurgic 10 examination of the remains and that sort of thing.
11 We do a lot of fixing of what we see, but I'm not 12 necessarily convinced we understand too well what even 13 happened, you know, the whys, the root causes of what happened.
() 14 And until we do we are just going to have to do some more 15 fixing again when we find the parts again weighing down, 16 because what we thought was the problem is not the problem, 17 it's some other ones.
18 So I'm just wondering who is trying to dig into this 19 enough to hopefully cure the problem instead of just keep 20 fixing?
21 MR. PETERSON: Are you on that, Ted, at all, on the 22 metallurgical?
23 MR. MICHELSON: Is that EPRI then that is doing it.
24 MR. MARSTON: This application guide that was 25 mentioned previously has really looked in, I would say, C1 Heritage Reporting Corporation (202) 628-4888
78
() 1 sufficient detail to determine the root cause of these 2 failures. And a lot of them are design related, where the 3 initial design was to get reduce the pressure problem as much 4 as possible. And in fact, that's really caused, in some cases, 5 the check valve has sort of beat itself to death.
6 In terms of metallurgical analysis, I think the 7 industry has some very good metallurgic. I used to work in 8 this field and the evaluations I have seen really look fairly 9 complete.
10 MR. MICHELSON: Well, I'm not at all questioning the 11 competence of the industry in their metallurgical area.
12 MR. MARSTON: Well, even in the utilities --
13 MR. MICHELSON: I just try to understand while the
( 14 utilities are not quite as good, but pretty good, but there's 15 fluid hydraulic problems and perhaps a little more than you 16 might ordinarily expect the utility to be wandering through.
17 But does EPRI -- is EPRI the focal point in the check 18 valve business for the metallurgical and the fluid hydraulic 19 problems and so forth?
20 MR. MARSTON: To give you some background on the 21 application guide, it was developed -- half of the contract was 22 with a person that's an expert in therma hydraulics, did a lot 23 of model flows with clear models to look at, what is actually 24 going on in some of these check valves -- check systems.
25 MR. MICHELSON: Did he actually do experimental work?
w Heritage Reporting Corporation (202) 628-4888
\
I t
l i
79 r i (3) 1 MR. MARSTON: Yes. And in fact, I just called Boyd 2 Brooks and a copy of that guide is on its way, so you should 3 have it soon.
4 MR. MICHELSON: Good. So EPRI actually has underway 5 then some amount of laboratory investigation of the fluid 6 hydraulic behavior of check valves.
7 MR. MARSTON: Principally to look at the effects of 8 nor:nal operation on the induced degradation of check valves, 9 which is really the concern. And I would say that all the 10 participants are contributing to this, the Owners Groups, and 11 the utilities.
12 MR. MICHELSON: Does EPRI also look at what happens 13 when you put a check valve from a vertical run instead of a 73 O 14 horizontal run, that sort of thing?
15 MR. MARSTON: Yes, that's covered. That's covered.
16 MR. MICHELSON: That could make a significant 17 difference.
18 MR. EBERSOLE: Is there any knowledge -- I think that 19 San Onofre's event was -- wasn't the root cause of it, they 20 reduced the flow down to something like 70 percent and started 21 a wearing action. Are there any critical periods flow rates 22 that involves go into a motion other than by two little flow.
23 Can we go up to some critical flow condition and start a 24 reciprocating function -- are the dynamics of check valves 25 floating in streams getting pretty well understood?
O Heritage Reporting Corporation (202) 628-4888
I >
80 I) 1 MR. MARSTON: I don't know how well it's understood, 2 but there are certain design to check valve that are more ,
3 susceptible to that than others, and a lot of it has to do with 4 pressure drop than check valve. And I assume there are flow 5 rates in which you cause an instability in the --
6 MR. EBERSOLE: Even high flow rates.
7 MR. MARSTON: Well, what I was going to say, I don't !
8 know the exact level because I think it has to be independent.
9 But clearly, the high flow rates, the thing has got to be ;
10 completely free.
11 MR. PETERSON: In addition, we have sent several l 12 letters to the industry highlighting the importance of SOER 13 8603 again to the executive level personnel, March of 1987. l O 14 And then we asked them to give us the results of their :
i j 15 implementation progress in June. l 4
16 In December of '87 the EPRI check valve guideline we i 17 recommended they use that for their implementation of SOER :
1 1 18 8603, even though they should have already started work on l
) !
19 that.
20 During evaluation or evaluations we look at the '
21 recommendations of SOER 8603 concerning check valves with our l 22 tech support department and operating experienced department. i 23 To give you an idea of what we are finding, in the 24 December time frame, December, January time frame, we looked at i 25 both recommendations. Recommendation one which essentially i y
() Heritage Reporting Corporation ;
3 1
(202) 628-4888 l l
j !
f i l
-- _ - ~ __ _ _
81 l
(} 1 establishes a preventive maintenance, asked them to establish a i 2 preventive maintenance program for check valves.
1 3 And recommendation two which is to perform a design
]
I 4 review. On the design review portion all seven had either 5 started or completed this review.
6 On the preventive maintenance program six of the j 7 seven had either started it or had completed the program that 8 was in progress. One had not had any failures, but had planned l
9 to, in the near future, initiate a preventive maintenance 10 program.
11 So we feel pretty good that the industry is 12 addressing what we have told them to address at this point.
I 13 In addition, we have sent event review teams out to i
() 14 these -- for these two events, the check valve failures, to see 15 what as INPO as a separate set of eyes could learn. And again, !
16 we wrote SERs concerning those events from our perspective. t 17 To kind of close out here, the other industry l 18 interactions rciated to check valves, there was a check valve L
19 workshop in October of 1986. We are in liaison with the Owners ,
20 Group task force for check valves through EPRI.
21 As I indicated we reviewed the check valve l
22 application guide. The draft was issued in September and the 1
23 final version was just out in January. l 0
?
l 24 Again, at INPO we have one individual where all the i l
25 information flows through on check valves. And we are ;
i Heritage Reporting Corporation (202) 628-4888 i i
s
i 82
( 1 participating or have participated in their check valve 2 workshop.
'3 MR. MICHELSON: The large check valves I'm sure are 3 i
4 getting a considerable attention because they are the ones that 5 have been involved in some of the more interesting events. But 6 there is a large population of small check valves in a nuclear 7 plant that may not even be reported to NPRDS. I don't recall 8 just offhand where they put the lower size reporting limit on 9 them, but thought it wan like four inch or something, isn't it?
10 What is the smallest check valve you even know about in NPhDS?
11 MR. PETERSON: I cannot answer the size, but I can
'f 12 tell you that the following systems we are looking at: main 13 steam; service water; aux feedwater; diesel air start, so
() 14 there's some small valves, that's probably about two inch.
1 15 Suppression pool support for BWRs; CBCS for PWRs; main i
16 feedwater; RHR; and low pressure safety injection.
17 MR. EBERSOLE: In the main feed check valve area -- (
18 MR. MICHELSON: Let me finish the question, though. I
) 19 There clearly are a number of, could be potentially very ,
3
! 20 troublesome, there's a check valves that might involve lines 1 ;
21 two inches and under nuclear plants, particularly if the line l
22 is coming from a high pressure system inside a containment out ,
23 into an area where it's the failure of the check valves to
. 24 withstand the pressure can result in blowing down through a two j
25 inch line which is a pretty serious blow down if it's into a S
i
() Heritage Reporting Corporation (202) 628-4888
- D
83 i
O 1 confined comgertment conte 1ning eome tempereture eeneitive 2 equipment.
i 3 I think we aren't looking too hard, except in the 4 case of air start motors you're-looking at for a different 5 reason. There has been a lot of problems with the air start l 6 system on diesels, and so the check valvec in there are getting 7 focused on. But how about all the other check valves we h6ve 8 around the plant, whose integrity is being depended upon. For i
! 9 instance, the excess flow checks on boiling water reactor l
10 instrument lines, some of them are up to the neighborhood of l
] 11 one and a half inch. We think they work. We don't have any
]
I 1 12 necessarily good ways without a lot of effort to make sure i
1 13 they're working.
O 14 But if you fail the instrument line some day and the
! l 15 excess flow check is not preventing a reverse flow, you might t I
j 16 have an exciting event. l 17 Do we look at smaller check valves? In other words, j 18 are we focusing on the big ones that have been involved with !
i
, 19 the San Onofre kind of events? .
1
- 20 MR. PETERSOt1
- As I was reviewing a number of these 21 SERs in the last several days, I remember several evente in f
22 there where they were talking like the mini flow or mini recerc j 23 lines on pumps which are in the range of one inch.
24 MR. MICHELSOtl: flo , they're a little bigger than 25 that, I think. Those are safety injection pumps, they're about j O Heritage Reporting Corporation l
I (202) 628-4888 i
i 6
i \
l 84 r
() l' three inch lines.
2 MR. PETERSON: Some of them -- l 3 MR. MICHELSON: You don't get much flow through a one !
i 4 4 inch line for a minimum flow. l 5 MR. PETERSON: When you're talking high pressure, 6 they're not very large. f i
7 MR. MICHELSON: But we aren't losing sight of the .
8 fact that we could have small check valves that could give ,
J 9 serious safety problems. You know, I'm thinking of monetary i
] 10 safety problems to the industry as well; not necessarily, they l
] f 11 won't necessarily lead the core melt down, but they could lead 12' to an amount of sufficient excitement that the plant might be 13 shut down for awhile while they are thinking about it and so ,
() 14 forth. And you might have trouble thereafter.
15 Check valves are something that we know little about j
- 16 in small sizes. We don't have good ways of monitoring. They i i
- 17 don't nave any requirements to see if they're even working in !
18 many cases. And the diagnostics-that I have seen that might !
i 19 even work on check valves, if any of them work, just aren't l l 20 amendable to one and a half inch check valves; it just couldn't j 21 do it. ,
- 22 Do we worry about them? Well, we could say, if we 1
- 23 don't get any reports. The reason we don't get any reports is ,
24 because you don't know they have failed.
. l 25 MR. EBERSOLE: Because the plates don't -- ;
l Heritage Reporting Corporation i s (202) 628-4888 !
l \
)
! t
.. -. - _ - - - -.- .. -- . . - - _ = . - _ . _ _ . . .. - - ..
_ . _ _ . = . . - . -
i 85 !
( 1 MR. MICHELSON: The report you-finally hear about, 2 in fact in many cases they aren't reported also because even 3 when they take the thing apart and find it's bad, it's not a 4 reportable item because it isn't a big enough valve or i
5 whatever. So we may not -- I just wanted to throw it out on 6 the table. i 7 I just wonder if we have small check valve problems 8 as well, and aren't fully aware of it.
9 MR. EBERSOLE: It's protected by the integrity of the 10 pipe.
11 MR. MICHELSON: I think a little thought has to be 12 given to, well, which ones do I really worry about. And I 13 don't know off the top of my head which ones, but I suspect O 14 with a little bit of effort we can find a number of them that 15 we ought to really be wtcried about; and those we ought to have 16 some means of knowing that they are.
17 MR. EBERSOLEt I ask you this question ab7ut check 18 valves and large ones, now, I'm going to get out of the smaller 19 category. I found it fascinating that somehow or other GE was 20 imposed on or imposed on itself a requirement that their check 21 valves be exercisable, that is, they got light duty action 22 swinging devices that will test them through the traverse that 23 they have to go, which can be overridden by the main control 24 pumps if they have to close.
25 Yet, interestingly enough, none of the PWRs have to O < Heritage Reporting Corporation (202) 628-4888 a
- s. : yv
\ f 6'
w 86-O 1 ao thet,' e"a ce1a12'v-kunderstoodanybasisforthatother '
% .. , q 1 2 than political. Ano dor.ht. understand that at all. ,
s ,- 1 3 The second as,,q,I w, think I've seen in such critical big 4 vhhEs as ma.i.n feed kheck v ' ves which are some checks, both
\ ( .,
1 ,
g 1 5 the free W odtinj dtbk 3that t).aps or doesn't flap depending on i
6 the flow; and dhes whic Qwould have an extended shaft and a p ,.. , ', 44 7 ' damper t'c 1pdibit too rapidioigenro.'s b 6 s3
\
8 In . Inspection of b;,ne f amily of these large check '
valvon heve you found this disphity in design?
9 .U ,.
N .V 3
S 3, . .
1-i.4 ,
10 hRy.FETERSON: It'I could addpYs thefirsfquesglon. p j (1 , MR . 3BERSOLE: The expreiseref ; \
12 FR. PETERSON: -I don' i s 2 l ,
lelleeweareawaref'fitor i 4
.o , ,
13 have been look'ng at it, t 14 MR. EBERSOLE: Well, y, u will find it's standard H.s
. . $ T . .
15 There !c no characteristics that bo.ilers '
h,av,e ex., orcis ers . . s
,\
( , s ,
. 16 reaon F.< :. I have vrer ceen agle,' to find -- ( q
- 1. L MR. PETFR$0N: All e way fully opened and f ly ,
i i!s ,
18 closed?4 41 /, ,
, o s lo \ MR. EBERSOLE: Yes. r3 x s e 20 < .i ' MR. PETERSON: The, plant I was from and I semember,
\
, 21 it wds'a PWR.
d.5 l 4
22 MR. EBERSdLE:
\
You den't have it s' .[
d . .
t )
y '23 MR. PETERSON: We did. .n i <
i ' 1,f
{ 24 '
MR. F,'BERSOLE : Oh, you did, i 1,? -(
4 x , ( ! ;
25 '
,, MR. PETERSON: Itcycle}themalit/lebit. ;
!O i
' Heritage Reporting C rporation ,,
l -
(202) 628-486 '
7 !
_s i
_x, s
f N,s' ' 1
\i 3 w.
Y^ g \j , ,
-,.. ,;+ .c T -_ v . - , , , -.
_ ._m . _ _ _ . _ _ _ _. _ . . _. _ . _ _ . . __
87 [
V
() 1 MR. EBERSOLE: Yes.
MR. MICHELSON:
j 2 Which valve was on the PWR that you i
- 3 had --
4 MR. PETERSON: Main steam installation valves. [
t 5 MR. MICHELSON: But that's a 10 percent cycle, j 4
6 -MR. PETERSON: Right. l 4 7 MR. MICHELSON: That is not really what Jesse is
'8 talking about. These are full. cycles, full swing exercisable 9 checks.
10 MR. PETERSON: They're in operation.
11 MR. MICHELSON: There's normally nothing~ flowing 12 through them, so you can do the check. l 13 MR. PETERSON: Not during operation. i
-i C:) i 14 MR. MICHELSON: Yes, during operation, sure. The :
15 exercise, there is nothing flowing through them. !
- 16 MR. EBERSOLE: If there is nothing flowing, yes, j 17 MR. PETERSONt Well, then, the turbine is out of j
! 18 line, f
19 MR. MICHELSON: These are not on the main feedwater i I
20 of a boiling water -- !
21 MR. EBERSOLE: No, no. !
t 22 MR. MICHELSON: -- these are on the RHR.
i
, 23 MR. PETERSON: I see.
I j 24 MR. EBERSOLE: But in the main feed lines have you
) 25 found both the hydraulic dampers as well as the free flapping 9
( I Heritage Reporting Corporation l (202) 628-4888 j l
i , - . __ . _ _ _ . . ,___ _ _
- - _- . _ -. _ , ___ _ __ _ _ . _ . _-_ _ _ _ J
i l 88 t
j
(} 1 kind?
i 2 MR. PETERSON: Honestly, we haven't looked for it or 3 haven't specifically looked for that issue.
4 MR. EBERSOLE: See, that tells me you haven't looked !
5 at the dynamic problem. ,
1 6 -MR. PETERSON: Well, that-may be. ;
7 MR. EBERSOLE: And that's I'think a critical' aspect 8 of these valve functions, and has been historically totally 9 ignored aspect of check valve function because only in this 10 business have we postulated pipe failure.
11 MR. MICHELSON: The classic one I think Jesse has in I
d 12 mind, I'm sure you're familiar with, and that is some utilities -
l 13 who would reduce check valves on the main steam line for ;
. () 14 reverse flow of checking. And, of course, there they are kept 15 cocked open solenoid or whatever; and when they are released l 1 16 and there is a larger risk of flow base that would like to :
1 17 close fast, and do they close so fast that they self-destruct.
! 18 MR. PETERSON: Understand.
. 19 Just to finish up here. We intend to look even a 4
1 20 little harder in the upcoming months at check valves. Over the j 21 next 10 evaluations we are going to do an enhanced look. We l 22 are still in tne preliminary performing stages of what we are '
l 23 going to do. And then after that we are going to review the i 24 results similar to what we have done in the MOV area.
j 25 MR. MICHELSON: Are you looking at the diagnostic 1 l 1
j Heritage Reporting Corporation
].
(202) 628-4888 J
-- --- - . - . ~ . . _ . _ _ _ . , _ _ _ _ _ ~ ._ _
,.___i
i 89
() I techniques that are being developed or check valves or is that 2 EPRI? Do you depend upon EPRI to finally tell you, we think 3 we've got one that works and here is what you ought to now be 4 promoting or do you do some of your own looking in diagnostics?
5 MR. PETERSON: We're using what, not only EPRI does, 6 but what somt other utilities may be doing. And if, for 7 instance, we -- during the evaluation we look at check valves 1
8 or MOVs, we find something that is done well, we write a good 9 practice on it. That good practice then goes out on network, 10 gives them a full number for them to call us and we can give 11 more details on it.
12 Specifically, from what you're asking on check 4 13 valves, what of our recommendations is, and I will just kind of !
1 %
14 paraphrase what it says, the program should include a 15 combination of following elements to ensure check valve 16 reliability; one, periodic testing to demonstrate the check 17 valves will operate.
18 Two, surveillance monitoring, i.e., acoustic and ,
19 vibration analysis to identify leakage and other developing i 20 problems.
I l
21 And third, disassembly and inspection on a sampling :
22 basis to look at check valve internals. .
23 So that is the thrust we are headed for in check 24 valves. f 25 MR. MICHELSON: The first item, though, you can in O Heritage Reporting Corporation (202) 628-4888 ;
i 90
() 1 some cases through inference tell if a check valve is 2 functioning; in other cases it is difficult to determine.
3 You are saying there ought to be some means, I guess, 4 but not necessarily saying there is a means of doing it. And I 5 would ask -- like to ask, are you aware now of any means of, 6 you know, short of having an extended operator or extended 7 shaft rather, is there any means of telling whether a check 8 valve is functioning properly or not without intrusion or 9 without inference by saying, gee, the flow looks like it goes 10 up high enough it must be opening all the way or how do you do
- 11 it?
12 MR. PETERSON: I'm going to have to answer, I'm going 13 to have to get you the answer, because check valves again are 14 not in my particular area of responsibility at INPO and if I 15 said, no, I may be saying something --
16 MR. MICHELSON: Maybe EPRI is going to tell us later 17 if they have got anything underway on check valve diagnostics. :
18 MR. MARSTON: I suspect you will hear quite a bit 19 tomorrow. ,
20 MR. MICHELSON: Okay.
21 MR. PETERSON: Any other questions.
22 (No response) i i
23 MR. PETERSON: Thank you.
24 MR. MICHELSON: Well, thank you very much. This '. 8 a 25 good appropriate time, I think, to take our break. So va will :
(
()
, Heritage Reporting Corporation i (202) 628-4888 l
1
. . . . _ - . - _ _ . _ - . . . - . _ . - . . _ = . - . . . - _ - _ - . . - . . . . . . . . . . - . _ _ _ _ -
1 i
i 91 h 1 be back at 3:15. )
l 2 (Recess.) l 1
3 (Continued on next page.) i i
4 5
l 6 ,
1 7 I 1
i 3 8 l
l 9 !
j.
10 !
t j 11 l 12 ;
j 13 14 l !
i 4
15 :
i l 16 i i !
18 i
i l
! 19
}
20 J
- 21 .
I !
I i
! 22 l I i i 23 J 1
i i 24 i 25 f; i
(
e lieritage Reporting Corporation i
(202) 628-4888 l I I
- - _ _ _ - . . . . - _ . _ , , _ . , _ . - - . _ _ _ _ _ _ _ . _ . _ , _ _ _ . _ , _ _ , . . , . _ . _ _ , , _ _ _ _ _ _ . _ _ _ _ , _ . _ _ _ _ _ . _ , _ _ _ , . _ _ _ . _ _ _ , , __ (
t 92 l
()- 1- MR. MICHELSON: We will get started again now so i that we can finish up fairly early today, or at least on time.
2 [
> r l 3 (Pause) ;
4 MR. MICHELSON: You are going to introduce yourself, i
5 I guess you are going to talk about the hydrogen detonation 6 that was observed, right?
a e 1 7 MR. HAMMER: Yes.
8 MR. MICHELSON: Good. Introduce yourself, then, and 9 we will hear about it.
10 (Slide) ;
) ,
11 MR. HAMMER: I am Gary Hammer. I have Chris
]
12 Parczewski with me here today. We are both with NRR, and we l l !
13 are both in the division of Engineering and Safety Technology. j l 14 As you said, Carl, we are here to talk about the hydrogen i
- 15 detonation which took place in a German reactor.
16 That event took place last May at a facility called l
]
l 17 the Gundremmingen -- if I pronounced that correctly -- Unit C, 1 18 which is a BWR in West Germany. They had a hydrogen detonation ;
l ,
l 19 actually inside the internals of the main steam safety relief 20 valves. There were three of them that were pretty severely ;
, i i
21 damaged. They stuck open. It took some number of hours to get l 22 them all closed and the system completely isolated again.
23 MR. MICHELSON: Are you going to tell us whether j 24 these were headed together? How close a proximity were they to j 1 !
j 25 each other? I notice you have a detailed drawing of the valve, }
1 4
() Heritage Reporting Corporation l i (202) 628-4888 ;
i !
r
.r
93
() 1 but not of the valve arrangement.
2 MR. HAMMER: Right. I'm afraid I don't have the kind 3 of systematic information that you're talking about. The 4 information which we had was actually given to us by the ACRS S back in November when we were asked to look at it. What it was 6 was a foreign incident reporting system report. It had some 7 sketches of the valve itself and some brief description of how 8 the pilot valve was connected to the header, which was also 9 connected to the valve.
10 MR. MICHELSON: But you never heard in the process of 11 looking into it how close the valves were to each other?
12 MR. HAMMER No.
13 MR. MICHELSON: Okay, thank you.
m 14 MR. HAMMER: In looking at it and reviewing the 15 report, we found out that the hydrogen was created by 16 radiolytic decomposition of the water. It was able to collect 17 in the valve because it was in a high, subcooled and stagnant 18 pocket. Because the valves are located relatively high in the 19 system and they are slightly subcooled, the steam itself of 20 course saturated. So if you cool it off a little bit, it is 21 going to be subcooled.
22 It is basically stagnant up in the internals of the 23 valves. You don't have a lot of circulation or anything like 24 that. It basically requires these conditions.
25 MR. MICHELSON: Did you have a chemist look at this O Heritage Reporting Corporation (202) 628-4888
I i l 1
1
- 94 ;
() 1 situation to see if it is even theoretically possible?
1 2 MR. HAMMER: Yes, we verified that. !
3 MR. PARCZEWSKI It is possible, yes. l 1 .
4 MR. MICHELSON: The vapor pressure is going to be 5 very high in that region because you have hot water not far l
! 6 away -- or steam, I should say. How can you get a mixture in !
s 7 that area? What was the concentrating mechanism that was
! 8 defying the vapor pressure source, which is infinite?
?
- 9 MR. PARCZEWSKI Well, the mixture of steam and 10 hydrogen and oxygen was accumulating. But since this j 11 particular location was cold, this then condensed. l t
12 MR. MICHELSON: Yes, but there is a new source j i :
] 13 constantly coming in. I O 14 MR. PARCZEW3KI: Yes. This is a mixture of steam and l
l
! 15 gas. It was condensed, and the water was draining. -
4 :
16 MR. MICHELSON: But if you'-- !
t 17 MR. HAMMER: It was basically stagnant, Carl, because j 18 it was in a highpoint. The non-condensibles would tend to just !
l' I 19 stay up there and sort of float. l l
. 20 MR. MICHELSON: But somewhere there was an interface
]
I
- 21 with an infinite source of steam.
I 22 MR. HAMMER: Yes, that's true.
I j 23 MR. MICHELSON: And that interface doesn't defy i
l 24 diffusion if the steam is constantly mov!ng up into this 25 region. You are going to have to condense it a lot faster than i
d
() Heritage Reporting Corporation (202) 628-4888 i
2
95
() 1 it is coming in for this to happen.
2 MR. PARCZEWSKI: Yes, but the mixture of hydrogen and 3 oxygen is lighter than the steam is. So you have a chance to 4 condense at the top and steam at the bottom. There is a 5 separation mechanism.
6 MR. MICHELSON: But it has to be very fast-acting to 7 overcome the rate at which the steam is entering into this area 8 of equilibrium that you are developing this detonateable 9 mixture in.
10 MR. HAMMER: I think maybe looking at the pictures of 11 the valve may help with this a little bit.
12 MR. MICHELSON: Maybe it will become apparent.
13 MR. HAMMER: Yes.
14 MR. MICHELSON: But you're satisfied that this didn't 15 defy the laws of physics at all?
16 MR. PARCZEWSKI: No.
17 MR. HAMMER: We believe that it is a critical 18 phenomenon, yes.
19 MR. MICHELSON: Okay. If it is critical, the next 20 question is going to be what is going to happen in our boiling 21 water reactors.
22 MR. HAMMER: Yes. I will address that.
23 We talked to the gentleman in Germany with the GRS, 24 which is the German Society for Reactor Safety, or something 25 like that -- Dr. Lindauer. He explained some of these things O Heritage Reporting Corporation (202) 628-4888
I l
1 96 l
() 1 to us in a little greater depth than what we had merely read 2 from the report.
3 okay. Once the stoichiometric combination of oxygen 4 and hydrogen does collect in the pockets inside the valve, the 5 detonation mechanism was caused by a fast compression of the 6 gas when the pilot valve was actuated. It's basically such a 7 fast compression that it is basically adiabatic. If you look 8 at your ideal gas relationships, if you raise the pressure very ,
9 suddenly, you also raise the temperature very suddenly. So it 10 was enough to actually detonate the mixture, and that's what 11 caused the explosion.
12 If I could go to the next slide, I'll come back to 13 that.
14 (Slide) 15 MR. MICHELSON: You had to have a mixture of three i
- 16 constituents, hydrogen, oxygen, and water vapor. You can't 17 tell me that there was no water vapor there. So what kind of 18 7.ercentage of hydrogen and oxygen, in stoichiometric i 19 combinations, was there? What degree of dilution do you have 20 from the water vapor?
21 MR. MAMMER: Well basically we figure that because l 22 the water is able to condense out, the water is being 23 displaced by the non-condensibles. l 24 MR. MICHELSON: Yes, but the water is also constantly i
25 entering into this equilibrium volume, i l
($) l Heritage Reporting Corporation (202) 628-4888 i
J
i 97 ;
i
() 1 MR. HAMMER: Okay. That is the point you were l 2 bringing up earlier. ,
1
- 3 MR. MICHELSON
- That's where the difficulty -- l 4 MR. PARCZEWSKI: Even if you have thirty percent of l 1
5 steem, the ignition temperature for spontaneous detonation is ,
l 6 about 1,000 degrees Fahrenheit. The temperature obtained by l ;
I 7 sudden compression, mostly adiabatic, would be over 2,000 8 degrees.
9 MR. MICHELSON: But you can only get detonation if 10 you have either a very low concentration of diluent -- well, 11 that's the only way. You have to get the concentration of
{
j 12 diluent way down in order to get a detonation to even occur.
i 13 MR. HAMMER: Right.
14 MR. MICHELSON: Even under ideal conditions. So 6
2 15 somehow you or somebody over there convinces themselves that ,
i !
j 16 there just wasn't enough steam entering the' system compared to j l 17 the rate at which it was being removed from the system. l 1 i 18 MR. HAMMER: Well, this was quite subtle to the j 19 German researchers also. They originally postulated maybe a
{ 20 water hammer. But they found a water hammer itself didn't have 21 sufficient energy to do this kind of damage. So then they l
22 started looking at other things and found this.
23 This is a sketch -- and I did this myself, because I 1
] 24 didn't want to put the proprietary information up here. But i
j 25 this shows you basically how the valve works.
Heritage Reporting Corporation j (202) 628-4888 i
i
L I
98
- O 1 Ma. M1CaELsOn Thie 1e non-erogrietary here.
- 2 (Laughter) l 3 MR. HAMMER
- This is non-proprietary.
4 I figured without some kind of sketch you would miss 5 some of these subtleties. But this is not really the way this 6 looks. I have shown the valve disc in two different positions.
a 7 This is the open position; this is the closed position. You
-8 can see where the main disc seats right here.
9 The pilot valve supp]ies steam into this tube up into 10- this chamber here and pushes it open here. Normally the valve 11 is seated by the differential pressure that is created above 12 and below. The inlet is here, and the steam discharge is j 13 through here. This is all the stationary part of the valve.
! 14 MR. MICHELSON: How large a valve is that? q l
l 15 MR. HAMMER: I think it was about eight inches, if I I I
16 remember correctly. ;
{
17 And what we have is a pilot valve back here which is 18 normally closed, of course. These valves have dual pilots, one l j 19 magnetically operated -- a solenoid -- and another spring l 1
l j 20 actuated -- which is self-actuating due to pressure. But they !
i 21 both operate very quickly and cause this pressure wave to l 22 propagate up here and detonate the hydrogen.
23 Now, as to the mechanism of it collecting, this is l 24 what we have reasoned has happened. That is what they think,
- 25 too. The pilot valve leaks a little bit. If it is perfectly i
3 O Heritage Reporting Corporation j (202) 628-4888 l
l
_ ._ _ _ . . ._. _. . _ __ _ _. _m . _ _ ._ . _ - _ . . _ .
1 '
l l l 99 ;
2 l () I tight, you don't have this problem. It leaks a little bit and -
! 2 introduces a mixture of steam and non-condensibles.
.. 3 MR. MICHELSON: Wait a minute. A mixture of hydrogen f I r 4 and non-condensibles -- oh, okay. I'm sorry, i 4
5 MR. HAMMER: Steam and non-condensibles. Whatever is !
6 in there just goes on up in there.
- 7 MR. MICHELSON
- It is a very small concentration of i
8 hydrogen, of course?
I ,
j 9 MR. HAMMER: That's true.
10 MR. MICHELSON: Very, very small.
j 11 MR. HAMMER: Right. Well, hydrogen, oxygen, and i 12 steam. And because this is high pressure and this is basically
{
., 13 vented, and none of these surface fits here are very tight, it l
j 4
14 is able to physically displace the material which may have been i a
j 15 in here, which could have been air or steam from the previous i i !
- 16 actuation, or whatever. You get this stuff pushing its way up ,
f i 17 in here. !
j 18 Now, the steam, when it gets up into this area, there l
20 are going to fill up with the steam and non-condensibles. We j 21 feel like the steam is going to be able to run out this tube, i
22 or the water-condensate is going to be able to run out. You 23 are probably going to have a lot of oxygen and hydrogen up 24 here. ,
25 But the way this valve operates is that this valve 4
) '
Heritage Reporting Corporation (202) 628-4888 l I
J l
l 101
(} 1 pops open and decompresses this chamber.
differential pressure forces to open the valve.
This allows these So there is no 2
3 compression taking place.
4 MR. EBERSOLE: Does the German design have the air 5 operator that you have there?
6 MR. RAMMER: Oh -- yes. Up here on this design there 7 is an air diaphragm up here, and you can remotely actuate this 8 with a solenoid valve.
9 MR. EBERSOLE: But the SRVs have two functions. They 10 are to relieve excess pressure or to deliberately blow down.
11 The second is just as important as the first.
12 MR. RAMMER: Right. On their design, they actually 13 have two pilot valves. One is electrically operated, and one
() 14 has a spring set point. So if they want to actuate it at below 15 set pressure of the other one, they can just electrically 16 actuate it.
17 MR. EBERSOLE: They actually raise the pressure 18 inside the cylinder?
19 MR. HAMMER: That's right.
20 MR. EBERSOLE: Well, that was an unfortunate design, 21 wasn't it?
22 MR. HAMMER: Yes, as it turns out.
23 MR. EBERSOLE: Was the Target Rock deliberately -- we 24 just had a happy --
25 MR. HAMMER: Well, I don't know. I can't speak for O Heritage Reporting Corporation (202) 628-4888
102
(} 1 them -- whether we just left out our not. I am sure that the 2 German design has its good points.
3 MR. MICHELSON: There has never been any effort, 4 though, to make those spark-resistant and so forth -- the 5 internals of the power valve. If that was really 6 stoichiometric -- hydrogen and oxygen -- you would have to be 7 careful on material selection, or you could ignite it with a 8 spark --
9 MR. HAMMER: Yes --
10 MR. MICHELSON: -- in the process of rubbing the 11 metals together. Frictional heat can also do it.
12 MR. EBERSOLE: They have been operational for a long 13 time, haven't they?
() 14 MR. HAMMER: I would have thought it would have 15 happened before last May.
- 16 (Slide) 17 Let's see. We looked at another design that was very 18 similar. This is the Target Rock three-stage, a little older 19 design. I am afraid my picture is oriented a little wrong. I 20 am going to have to turn it this way, because that is actually 21 how it sits. The inlet comes in the bottom, and the outlet 22 comes out the side.
23 So you can get the gasses collecting in here, we 24 feel, and up in this area of the valve as well, this is a 25 little different design. When you actuate the thing, the Heritage Reporting Corporation (202) 628-4888 l
l l
1 103
(} 1 pressure in this bellows here is what does it. The pressure 2 goes up here and acts against this set spring and pulls this
, 3 pilot disc open, which allows pressure to get into this cavity.
4 So it was originally a question to us is we could have an 5 explosion in this cavity.
6 The answer to that was that it didn't look like that 7 was credible because if this valve was leaking more than you 8 were bypassing around the sides of this fit here on this 9 cylinder, then it's going to actuate this valve if you were set 10 up to have this problem. So in effect, this valve is actually i 11 designed to have some bypass leakage around here.
12 So what you get when this valve leaks is some 13 constant flushing out the discharge tube. Then these cavities
() 14 over here, when this thing actuates, get decompressed rather
- 15 th a compressed. So it doesn't look like you have any similar 16 phenomena there to the German design either.
) 17 (Slide) 18 There was one other design, a Dresser Electromatic i
l 19 design, which is on some of the really old BWRs like Dresden.
20 Here again we have a design where these cavities are normally 21 pressurized to seat the valve tight. Here is the valve seat.
22 Here is the inlet; here is the outlet. This is the solenoid i 23 actuator here which pushes down on this pilot disc to open it.
} 24 What it does when it opens is that it allows this to 25 depressurize throegh this tube and out this port here.
) Heritage Reporting Corporation (202) 628-4888 l
t
~
t 104 l f
i 1 Basically you don't have any compression phenomena taking place
- 2 in this valve, either. [
3 3 Now, what the Germans did -- and I didn't put this on j
4 the slide -- I should have. There were two things that they ;
- I j 5 did to fix this problem -- or two potential things they were r l
4 6 talking about when we talked to the gentlemen over there. They j i ,
j 7 were going to put a bypass line around the pilot valve back in [
r 8 this area so that you would have a continuous flushing to keep [
9 the gasses from building up in this region so that it would [
l f 10 never migrate up here. f I
- 11 But what that was going to do was increase the ,
l I 4
12 suppression pool temperatures, and they didn't want to do that l i 13 by having this thing flushing all the time. The second thing l i
O 14 they were more eerioue1r thinking about wes using eome i I
j 15 catalytic materials inside the pilot valve back in this area to i I i i 16 cause radiolytic recombination. It would promote the 1
17 recombination of the hydrogen and oxygen to reform water. They i 18 thought that looked like a better possibility.
f r
l 19 MR. MICHELSON: If we really believe this, and I will l l i
.j 20 assume for the moment that we really do, then I think there is 21 a far more serious problem than that related to the safety i l
l 22 valve. That is, why don't we believe that we can have leaky :
l 23 safety valves that are filling the tail pipes on BWRs? They {
24 are not pressurizing, because the condensib1e is condensing up f
i l 25 with the hydrogen and the oxygen. It sits there an builds up. ;
t t I !
O Heritage Reporting Corporation i l
l (202) 628-4888 t i
j 4
j
- l. .. - __ -
f 1
7 105 1
1 The first thing you know, all our tail pipes are full of 3 2 hydrogen. l t
3 Then you open a relief valve, and that steam [
c t j 4 compression wave runs those pressures way up in a hurry. But I 5 don't know if it would detonate it. You would have to look. i l 6 Why don't we worry about hydrogen and oxygen in the tail pipes
]
7 instead?
8 MR. HAMMER: Well, I don't think you could develop i 9 enough -- maybe somebody can help me with.this. ,
i 10 MR. MICHELSON: Well, the shock waves are pretty high !
E 11 pressures when you first open that relief valve. There is a l
! l j 12 shock wave probably getting on the tail pipe that is quite l i !
- 13 extensive. Whether it could detonate hydrogen, I don't know.
l
!O 4
14 eut you m1 2ht eek, If 1 hed stoichiometric hydrogen end exveen !
I i 15 in the tail pipe, could I ignite it by opening the safety l
- 16 valve?" l I
j 17 MR. PARCZEWSKI I think this might be very specific :
4 18 geometry. So you have this condensation of steam and l 4
] 19 separation of --
l 20 MR. MICHELSON: Well, I can do it even easier on the l
1 21 tail pipes. Water is sealed at the bottom -- purposely. And l l
i 22 it is leaking in constantly at the top at some rate. The steam l i ,
j 23 is disappearing because it is nice and cold there. But the ,
24 hydrogen and oxygen has no place to go, so it just builds up.
j
[1 25 Finally the pipe is full of hydrogen and oxygen. When it gets l I l
! t i O Heritage Reporting Corporation l
l (202) 628-4888 i l !
t
)
106
() I too full, and the pressure would slowly start rising, then you
, 2 would suspect something.
3 But we have never seen this, as far as I know. We 4 have never detonated one. But I might point out that when you 6 open those release valves, you don't know what is really 6 happening. Unless you bust the pipe with the detonation, you 7 wouldn't know it. It is quite an interesting event all by 8 itself.
9 I just wondered, if we really believe that this can 10 happen, why don't we worry about that problem?
11 MR. PARCZEWSKI: This is the only credible mechanism.
12 There is no other source for such a high energy.
13 MR. MICHELSON: Oh, yes. It doesn't seem like there
(
14 is a set-up for it here.
i 15 MR. PARCZEWSKI There is enough energy from --
16 MR. MICHELSON: Have you looked at the tail pipe 17 question at all --
3 18 MR. PARCZEWSKI: We haven't --
19 MR. MICHELSON: -- or any other place where we could 1 -
1 20 have leaks into areas that are trapped and that condense out 21 the steam but can't get rid of the hydrogen and oxygen?
t
]
22 That would be the first one I would name, although I
, 23 don't know that I have an ignition mechanism ii the opening of 24 the relief valve and the compression wave that goes down.
25 MR. EBERSOLE: When you look at the tail pipe, you l
() Heritage Reporting Corporation (202) 628-4888 i
i
i 107 !
1
[} 1 2
instantly then dig up the old problem of suppression bypass.
You could break the tail pipe, and traverse the suppression i 3 pool. They you are going to go into the unsuppressed 4 condenser.
5 MR.'MICHELSON: You don't want it to bust, that's for 6 sure. I just wondered that if we really believe all of this, 7 why don't we believe that we have hydrogen and oxygen in our 8 tail pipes?
9 MR. EBERSOLE: How much resistance can you build up?
10 MR MICHELSON: That's what you have to look at.
11 It's at atmospheric, but it's still stoichiomotric hydrogen and 12 oxygen.
13 MR. EBERSOLE: But that requires some temperature
() 14 rise.
1 15 MR. MICHELSON: You have to compress it -- shock 1
16 waves can detonate -- i a 17 MR. EBERSOLE: But again you have steam acceleration 18 and a strange configuration -- it isn't that simple, 19 MR. MICHELSON: I know it isn't that simple. If it i 20 was simple, I would probably be able to tell you. I just 21 wondered whether we were worried about -- :
22 MR. EBERSOLE: Well, the pressure surges in these 23 pipes was enough to make the donuts jump off of the ground.
1 24 MR. MICHELSON: Well, that is just pure -- the ,
! 25 compression -- the pressure sensors you put on tail pipes l
() Heritage Reporting Corporation !
- (202) 628-4888 f r
l
4 i
I
! 108 i register pressures like 150-200 pounds. I don't know what the I
({} 1 ;
l 2 shock wave pressures are doing in what would be this gas ;
i '
- 3 volume. I don't know.
3 4 MR. EBERSOLE: Whether you can set off hydrogen that !
i !
j 5 way, I don't know. ,
a :
i 6 MR. MICHELSON: Well, anyway, the ultimate 7 consequence of interest would be suppression bypass.
l 8 MR. EBERSOLE: Well, that's another issue.
9 MR. MICHEL3ON: It would be serious if you broke the
]
10 pipe, clearly. '
l 11 MR. EBERSOLE: At the same time you destroyed the I
l 12 cuts of the valve -- that's the classic combination, isn't it?
1
! 13 MR. MICHELSON: That would be the worst possible
( 14 case. Now you go for about four or five minutes and you blow j 15 the containment. That is a big hole.
1 16 MR. RAMMER: The last thing I wanted to mention, 1
1 17 which is sort of an aside, is that you really don't have any 1
- 19 lot of excess hydrogen. So you don't get this problem up in 20 the PORVs, and that sort of thing. i
! 21 MR. MICHELSON: What that shows is that you can l
22 design an oxygen concentrator by metering in steam carrying the l l
- 23 hydrogen and oxygen. You can concentrate the gas if you have a 24 way of condensing the vapor. l 25 MR. PARCZEWSKI
- Doing rpecial geometry -- with the j
}
! Heritage Reporting Corporation
! (202) 628-4888 i
109 4
(} 1 2
proper geometry.
MR. MICHELSON: So I would think that it would pay to .
4 3 look around to just see if there is any place else where this
- 4 is going to be happening before it does that. It's a cute i
! 5 explanation -- lacking a bottor one, it sounds good.
) 6 So your view is that it doesn't scom credible for the l
7 kind of relief valves that we use, although it appeared to you 8 to be credible --
9 MR. PARCZEWSKI: That's right.
10 MR. MICHELSON: I think that takes care of it. Do 11 you have a question, Jesse?
12 R. EBERSOLE: No, that's it.
I 13 MR. utru7 g-n, y.,u will at least think about that
() 14 tail pipe question, though, won't you?
J 15 MR. HAMMER: Yes sir.
16 MR. MICHELSON: . think it's unlikely, but I don't 17 know enough about it.
18 MR. PARCZEWSKI: Ne might have to do an analysis, you 19 know -- a site analysis.
20 MR. MICHELSON: ell, I don't know if you want to 1
l 21 make a project out of it. But it certainly is an obvious 22 question, I think, as an eetension of what happened.
1
! 23 (Continued on r xt page) l 24
- 25
( Heritage Reporting Corperation (202) 628-4888 1
I 110 0 1 MR. ZIEGLER: I'm Joe Ziegler, and I'm out of TVA 2 Licensing, Chattanooga. Buzz Beltz to my right, Assistant 3 Grants Chief and Counsel Engineering Branch. Next to me on my
. 4 left is John Tyrrel, Mechanical Engineer at Watts Bar. And ,
j 5 Darrell Murphy next to him, Senior Electrical Engineer at 6 Sequoia on the operations side of the house. The speaker is 7 going to be Darrell Cox who is the Senior Mechanical Engineer 8 of our Knoxville Design Staff. I think he can go over a 9 discussion pretty good of what you asked us to go over.
l 10 MR. MICHELSON: It sounds like you've come well i 11 prepared.
12 (Slides Shown)
O u MR. Cox Good afternoon. This afternoon we wou1d j 14 like to speak on the interchangeability of valve actuator J
- 15 parts, specifically addressing the 5055 E Notice that was 4
1 filed for our Watts Bar nuclear site.
l 16 17 This afternoon we'd like to look at four particular 18 areas. The background as to what actually happened and what
)
19 got us to where we are right now; the investigative process j 20 that we've gone throught some of the thingc that we found in 21 the status of the work so far.
4 22 As a matter of background, the maintenance personnel 23 at our Watts Bar plant were performing valve maintenance and 24 inspections for actually a common industry problem, that being 25 lubrication problems. They found during the inspection of the O l l Heritage Reporting Corporation I (202) 628-4888
111
(
1 actuator lubricant, the grease, that we had two apparent 2 problems. One was separation of the lubricant and the other 3 was mixing. Mixing of lubricants with two dissimilar soap 4 bases. The problem the industry has seen most recently is the 5 mixing of the calcium base with the lithium based greases.
6 MR. MICHELSON: When you're talking about separation 7 and mixing, are you talking about out in the work space, or 8 are you talking about within the valve operator?
9 MR. COX: Within the valve actuator. The separation 10 that I'm speaking of is the actual separation from the oil and 11 the soap base. The mixing is the actual mixing inside the 12 actuator of the two greases.
} 13 HR. MICHELSON: Where are these two greases used in 14 the actuator, and why would they have a mixing potential?
15 You're talking about they were first of all lubricated with 16 one and then somebody came in and put the other type in? Is 17 that the type of mixing?
18 MR. COX: That is correct.
19 MR. MICHELSON: There is only one lubricant within 20 the operator, is that correct?
21 MR. ZIEGLER: In the main gear I don't see how there 22 could be more than one.
23 MR. EBERSOLE: Was the separation caused by two 24 dissimilar lubricants being put in there?
25 MR. COX: The actual separation problem was due to O
Heritage Reporting Corporation (202) 628-4888
I 112
)
() l 1 the fact that at the Watts Bar site we'vu gone longer than we t 2 expected to actually be operational, and a lot of the valves ;
3 have actually been idle for anywhere from 10-14 years.
i 4 MR. EBERSOL2: Separated by gravity. !
J l 5 MR. COX: Yes, they just sat, being idle, and we had
~
6 the separation problem.
]
7 The resolution was to completely disassemble the 8 actuator and relubricate. Now in the disassembly the actuator 9 was actually removed from the valve, taken to the shop, 10 stripped down to the sub-components, and gone through a 11 solvent process, cleaned, and then reiubricated. '
12 MR. MICHELSON: What does Limitorqo say, when you !
( 13 receive Limitorge operators, what instructions if any do they l l
- 14 give you concerning how soon, or what you should do about the ,
i 15 lubricant that was shipped with the operator? Typically what i
}
l 16 do they tell you to do? Is there any requirement that after X l 17 number of years that you have to relubricate, or do they just 18 assume the user knows all those things somehow?
1 19 MR. COX: Actually through Limitorge what we find is I 20 that you replace lubricant more often because of a j 21 contamination than the lubrica.it wearing out. You end up with 22 dirt or grit or something being introduced to the operator, 23 but not that it wears out and would require a periodic 24 replacing.
25 MR. MICHELSON: And there was no warning or reason O 1 Heritage Reporting Corporation (202) 628-4888 c
->n,
i 2'
c) 1 to believe from the supplier at least that the grease that it 2 came with wouldn't be good even if you took ten years before
- 3 you decided to ins' tall the valve?
l 4 MR. MURPHY: That has contributed to our requirement 5 for periodic cycling of the actuators. L 6 MR. MICHELSON; What do you mean periodic cycling?
7 You mean just to operate it, to connect it up and temporarily l
\
8 operate it?
9 MR. MURPHY: To contribute to -- I i
10 MR. MICHELSON: That's supposed to keep the grease
' 11 mixed up well enough?
I 12 MR. MURPHY: That's the intent.
( 13 MR. MICHELSON: Is the lubric3nt thought to have 4
14 been a lifotine lubricant originally? Is that the idea?
15 MR. MURPHY: I believe so.
16 MR. HICHELSON: From what you were told, at least 17 from your maintenance nanuals, were you led to believe that it
\
18 was supposed to last indefinitely?
19 MR. CoX: During the disassembly parts discrepancies i
i 20 were identified.
i
} 21 MR. EBERSOLI: That was just an incidental thing, 22 wasn't it?
)
- 23 MR. COX
- At that time, yes it was. The parts that i 24 were noticed immediately were the spring packs inside the i
l 25 actuator that are used for the torque seno.ing. In the
(
Heritage Reporting Corporation 1 (202) 628-4888 i
- l
\ \ j l
u
, j%,[i s
{
i 1
l
' 114 i
i O s ' '
1 actuators th are especially the vintage,at our 'atts W Bar )
I /
2 plant or the TVA vintage, there is a tag on the 3mit switch )
3 3 which shovW v spr.bgpackshouldyeinthea eptor as
\ (r n a 4 well'ta U}e to que switc3 settings.b phat was w ict,ci ,
s (
r 5 immeci.ately watt that t)hre seemed to %e a differed n
).sumber of o 1 G springs.in some of the sprind Jpacks thartwere showd on the 7 tagsthatwereinsidetheactutAbr. .
f / (
8 Thisledusintdtopthingr[. One, the initiation of 9 the 5058i ENoticethatweha?'N potential' \
problems (withe 1of s <
10 our valve actuators, lnd the o'h'ei- s a more detailed '
11 investigative proces[. 'c' '
j 5
\
12 MR. MICHEIlpN: Was the spring p,acd usually ( g.
13 specified by the numi er oNwashers it shoul contain or 7
. 1 4 3 ,
14 '
somsthing of thi.t sort? ' ,- '
y a 2 f
.5 MR. COX: The rgikbed and chickness N of y1ur trashefs.
T C A ,. ' 5- >
36 MR. MICHELSON: b/ght. Thenumberandk_hlgkness.
I ~ l l'/ And you noticed) thah it wappt the sawp ) number or thickness I gj z
18 guess, one or the other. <> it il 19 MR. EBERSOLE: Is this padt the impact device?
20 These spring packs? '
i 21 MR. MURPHY: No. 'i [
22 g
k,\ \
MR. "OX:
The investigative process that we were led 1
3 3 1 23 into cal. led for us to dida$semble the actuatbr a.c' we said ,' '
g 24 before; identify,all of thd sub-components.
This involved 25 going back to tce biA. of material that was used to '
O -
8 Heritage Reportirh Corporation (262) 7'62u-4888),4
, 4 .
' i s .h l
[ ) l
IA 115 Yf 1 manufacture the actuator at Limitorge's facility. What we did 2 was compare the configuration of the actuator as shipped from 3 Limitorge to the configuration of the actuator as it was u
4 sitting out in our plant.
5 MR. MICHELSON: What drawing, or where did you get 6 the document to do this comparison? Did Limitorqe send you a 7 new one?
8 MR. COX: The new material came from Limitorge.
9 MR. MICHELSON: You're not talking about the old 10 bill cf material that was shipped with the valve at the time 11 you bought it. This is a new bill of material, what Limitorqe 12 said should have been in there.
( 13 MR. COX: That is correct.
14 MR. MICHELSON: Okay.
15 MR. COX: As a result of the comparison, we enter 16 into a decision process.
17 MR. MICHELSON: Let me ask another thing. Did you 18 go and count the number of teeth on the gears and all that U
19 sort of thing? What did you do, measure the diameter of the 20 gear?
21 MR. TYRREL: We counted the teeth of the gears.
22 MR. MICHELSON: They gave you a lot of detail then.
23 MR. TYRREL: Enough detail for what we thought we 24 needed.
25 MR. MICHELSON: The pitch on the worm and that sort Heritage Reporting Corporation (202) 628-4888
l
() I 1 of thing they gave you then.
2 MR. TYRREL: The worm ratio. l 3 MR. COX: Yes.
l 4 MR. MICHELSON: That was a tedious job then, wasn't i 5 it?
6 MR. TYRREL: Yes.
7 MR. COX: If the configuration of the actuator as it 8 was installed in the plant was the same as it war shipped from 9 the factory, then the actuator would be reassembled and we 10 would proceed on to the diagnostic testing. At TVA we're 11 using the MOVAT system.
12 MR. EBERSOLE: Who got into it to make it 13 different?"
14 MR. COX: If I could defer for one more slide, I 15 think that's where we're headed.
16 MR. MICHELSON: Okay.
17 MR. COX: If we get into the actuator and we find 18 out that the configurations are different, what was shipped 19 does not seem to be what we had. The first question is, was 20 the actuator changed at the valve vendor's or the valve !
i 21 supplier's facility? There is a possibility that a gearing 22 change may have been made to correct a thrust requirement; 23 more often than not a stroke time requirement. There would be 24 a changing of the motor helical set, the motor pinion and the 25 worm shaft clutch gear.
Heritage Reporting Corporation (202) 628-4888 1
l
117 O l 1 MR. MICHELSON: With no paper record of that?
2 MR. COX: The paperwork didn't appear on Limitorge's 3 bill of material, but vben we went back to the valve supplier, 4 the valve supplier could document that yes indeed, that change 5 was made in the supplier's facility.
6 MR. MICHELSON: The valve supplier is the retailer, 7 or what?
8 MR. COX: Whoever the supplier would be.
9 MR. MICHELSON: Okay. The valve.
10 MR. COX: Not the actuator. )
11 MR. MICHELSON: Okay. !
12 MR. EBERSOLE: It's the ancient problem that the j 13 valve is in two pieces. The driver and the valve. So the 14 valve vendor did it when he assembled the operator to a valve.
15 MR. COX: Yes. Once the actuator was shipped from 16 Limitorqe it could have gone to the vendor's facility and 1 17 modified, then shipped to TVA. I 18 MR. EBERSOLE: Who's responsible for the valve 19 function in the final analysis? It's the valve vendor?
20 MR. COX: Yes.
j 21 MR. EBERSOLE: And that includes his adaptation of 22 the Limitorge?
23 MR. COX: That's correct.
24 MR. EBERSOLE: So it's his face that should be red, 25 isn't it?
Heritage Reporting Corporation (202) 628-4888
118 O
1 MR. MICHELSON: We're not looking for whose face is 2 red.
3 MR. EBERSOLE: But I want to go to the roots of this i 1
4 happens.
5 MR. ZIEGLER: I think what we came out with was that 1
6 the valve vendor did an appropriate design change on the valve l I
7 and documented it through their design change process. That i 1
8 paper work was supplied to TVA. I don't think it was readily 9 evident when we first identified the discrepancy that that had 10 taken place.
11 MR. MICHELSON: The bill of materials that you were 12 checking against was a Limitorqe bill of materials?
() 13 MR. ZIEGLER: That was the first. This process here 14 was identified after we had identified some discrepancies, but 15 that was the first place we were checking. We were checking 16 against Limitorge. I think the great majority of the time.
17 MR. MICHELSON: In cases where there were j 18 discrepancies perhaps the valve vendor had made a change and l 19 hadn't modified the bill of material from Limitorge to l 20 indicate the change. Is that what you're saying?
21 MR. ZIEGLER: I don't think they had modified the 22 bill of material, but they had supplied it in other 23 documentation they had supplied.
24 MR. TYRREL: They had not gone back to Limitorge and 25 told Limitorqe that they had made that change.
Heritage Reporting Corporation ,
(202) 628-4888 l l
1
1 119 O
1 MR. EBERSOLE: So they have the prerogative of 2 overriding Limitorge's, whatever they gave them, and changing 3 it. And that's accepted practice?
4 MR. MICHELSON: I assume changes that are made to 5 the Limitorqe operator, it's the responsibility of the valve 6 vendor to check with the manufacturer of the operator to make 7 sure it's an acceptable change, isn't it? I mean a QA program 8 would require that.
9 MR. ZIEGLER: Who was the supplier of the valve?
10 MR. TYRREL: In most they were changed under 11 Westinghouse FCN's.
12 MR. MICHELSON: Does this normally occur with the
) 13 Westinghouse supplied valves, or did you see it with others?
14 MR. TYRREL: We have only inspected the Westinghouse 15 valves to this point. We're only about a third of the way 16 through inspecting all Limitorge operators at the Watts Bar 17 facility. We started with the safety valves first, which 18 include mainly all Westinghouse valves.
19 MR. MICHELSON: Have you found cases yet where 20 nobody claims that it should have been that way, but it is?
21 In other words, changes that have been made that nobody would 22 agree should have been made?
23 MR. TYRREL: Not with the gearing of the operators.
24 MR. MICHELSON: It's always explainable, in other 25 words, after you've gone through the work of taking it apart Heritage Reporting Corporation (202) 628-4888
l 120 0
1 and finding the differences?
2 MR. TYRREL: We have found some differences, though, !
3 with the spring packs, but not with the gearing of the 4 operator itself.
1 5 MR. MICHELSON: I see. And you're going to tell us 6 later about that.
7 MR. TYRREL: In a slide, yes.
8 MR. COX: If the valve vendor was not cognizant of j 1
9 any change, then we would have to initiate corrective action l
10 in accordance with out internal policy, that indeed something l 11 is wrong with this and that would go into what we call the 12 condition adverse to quality report which would require a 13 generic review by all of our plans. So if Watts Bar goes 14 through and they find that they have an actuator that has been 15 changed, the valve vendor was not cognizant of it, we would 16 initiate a CAQR to kick off the corrective action, and also 17 require an analysis by each of the other sites as to whether i 18 or not they have the problem. I 19 The actuator at that point would then be reassembled 20 to the original bill of material from Limitorge, the last 21 engineering document that would define the configuration of I
22 that actuator. l 23 Now all of these will result in one thing, whether 24 it was correct as found; correct as changed by the valve 25 supplier; or incorrect but reworked to the original bill of O
Heritage Reporting Corporation (202) 628-4888
l 121 1 material. All of them end in diagnostic tests. No matter how 2 they were found, we test them all. That's to give us a 3 reasonable assurance that what we have works. We're not going l l
4 to rely on just the paperwork and the fact that the bill of I l
5 material matches what's out there. We also go back and test 6 them.
7 MR. MICHELSON: But your tests are at best nominal 8 load tests on the valve itself, so you can't tell how the 9 valve would perform under adverse loading conditions in the 10 real world.
11 MR. ZIEGLER: Darrell has to address that. I don't 12 think that's exactly true.
13 MR. MURPHY: Our design bases maximums are part of 14 the formula that determines the requirements.
15 MR. MICHELSON: Do you do some kind of an artificial 16 loading of the operator then?
17 MR. MURPHY: Absolutely. We actuate the stem 18 applications, we actuate the stem into a load cell.
19 MR. MICHELSON: Do you have a breaking arrangement 20 or something to artificially load the operator?
21 MR. MURPHY: The stem is operated into a stationary 22 load cell and this allows us to perform a calibration of the 23 operator.
24 MR. MICHELSON: You're getting a torque calibration.
25 MR. MURPHY: It's actually a thrust.
O Heritage Reporting Corporation (202) 628-4888
122
(
1 MR. MICHELSON: A thrust calibration.
2 MR. MURPHY: Then you're looking back to the 3 calculations to show that that's an adequate thrust.
4 MR. EBERSOLE: Is that adequate thrust consistent 5 with the operation of the valve under its highest loading 6 condition such as with run out discharge?
7 MR. TYRREL: That thrust loading would be what the 8 valve vendors told us what is required for the design base.
9 MR. MICHELSON: The best that can be calculated, 10 which is a problem, of course. Without experimental evidence 11 it's hard to do some of the calculations.
12 MR. EBERSOLE: Where you have document change up 13 there, does that mean that the Limitorqe bill of material, 14 have you then invalidated the Limitorge bill of material and 15 put in a new one?
16 MR. COX: What we would do is change our engineering 17 documents that we use to control the further configuration of 18 the actuator.
19 MR. EBERSOLE: And Limitorge's bill of material is 20 no longer applicable?
21 MR. COX: I wouldn't say it's not applicable.
22 MR. EBERSOLE: But there's a wrong gear in there, 23 isn't there, or something? Is there a wrong gear or a wrong 24 spring pack? Where you've made a change, it's not entirely 25 correctly, is it?
O Heritage Reporting Corporation (202) 628-4888
123 O
1 MR. ZIEGLER: Let me try it. Correct me if I'm 2 wrong, but what I think happened is in these particular valves 3 where Westinghouse had made a change through their design 4 change process, arid it was a full quality type change, is what 5 we were making sure is in our engineering documents that we 6 documented that change so if we did have to change that part 7 then we would put the proper parts back in. What we found is 8 those design changes were proper changes.
9 MR. EBERSOLE: What I'm getting at, did you destroy 10 the document that might have led you to put in the wrong part 11 later on down the years?
12 MR. ZIEGLER: We didn't destroy the Limitorge 13 document, but what we did is upgraded our own document such 14 that we would put the correct part in. !
l 15 MR. EBERSOLE: You didn't go back and void the wrong l l
16 gear.
l l
17 MR. MICHELSON: Did you review the bill of materials )
18 to reflect the correct component?
19 MR. EBERSOLE: The Westinghouse change.
20 MR. TYRREL: The maintenance procedures we have 21 instituted dictate that the maintenance crews when they take 22 out a part, they will put in a like part. So if they take out 23 a gear with 21 teeth they will put back in a gear with 21 24 teeth.
25 MR. MICHELSON: The question was, did you go back Heritage Reporting Corporation (202) 628-4888
i l
124 1 and change the documents now that the fellow is going to work l
2 from to make sure, maybe he pulled out with 21 teeth and the 3 bill of materials he's working off of says it should be an 18 4 tooth gear and he puts an 18 tooth gear in.
1 5 MR. COX: I believe the answer to your question is !
6 that we're not going back to the Limitorge bill of material l 7 for the future work. What we're going to are the engineering l
8 documents that we have in-house. We document it in our l 9 records that we use. ;
10 MR. MICHELSON : Are you going to go and burn the old 11 bill of material so nobody accidentally uses it? i 12 MR. EBERSOLE: So someone doesn't pick up the wrong 13 piece of paper.
14 MR. COX: We're changing the documents that we use.
15 MR. TYRREL: They won't be picking up the wrong 16 piece of paper.
17 MR. MICHELSON: It's not conceivable.
18 MR. EBERSOLE: You say he puts in a gear with the 19 same number of teeth. I believe I heard you also have to put 20 it in the same face forward instead of that way, don't you, 21 because of gear meshing.
22 MR. TYRREL: That's correct.
23 MR. EBERSOLE: There's a couple of cases where you 24 can put things in backwards on those operators.
25 I don't think it's been fixed except by instruction, Heritage Reporting Corporation (202) 628-4888
125 P\
(/
1 has it?
2 MR. TYRREL: Yeah.
3 MR. ZIEGLER: And training, which is required.
4 MR. TYRREL: Very mandatory training.
5 MR. MICHELSON: Are the teeth indexed relative to 6 each other with a quick punch or anything in any case? Like a 7 timing gear.
8 MR. MURPHY: No.
9 MR. ZIEGLER: They're not critical in that aspect.
10 MR. COX: After testing, the test results are 11 analyzed, the test results are acceptable. You move on to 12 place your valve in service. If the test results are not 13 acceptable in that the testing that is done, the signature 14 analysis testing tells us that we are not getting the thrust 15 out of the valve actuator that would be required to stroke the 16 valve under its design conditions, then we have to initiate a 17 design change. We have to go back, make a modification.
18 After that modification is made, go back through the loop I 19 again, and retest. This loop will be in effect until we get 20 to a yes, that we are actually producing the thrust that 21 should come out of that actuator to stroke the valve. Then we 22 would proceed to place the valve actuator in service.
23 I believe this is where we get to the question that 24 you had asked earlier about the potential sources. What we l 25 identify are four potential sources for thn parts O
Heritage Reporting Corporation (202) 628-4888 l
/
l l
l 126 l
() 1 1 discrepancies. One, we can't rule out the fact that it could 2 have occurred in Limitorge's shop. Someone just put the wrong 3 part in. It could have occurred in the valve vendor shop 4 after the actuator left Limitorge. There could have been a 5 mistake by our maintenance personnel. And there could have 6 been a mistake that we call actuator switching, where someone 7 looks at say an SMB-OOO and they figure a 000 is a OOO and we l 8 can move it from one vale to the other. Any of these could 9 lead to these problems.
i 10 MR. MICHELSON: What prevents that problem, or what I
11 used to prevent it at least? Anything in TVA that used to j i
12 prevent people thinking that all OOO's are alike? Is there l O 13 some mechanism? Had there been an existing mechanism?
l l
14 MR. TYRREL: There had been a mechanism when they l l
15 swapped operators to get Design approval before they swapped l 16 it and Design gave their approval.
17 MR. MICHELSON: What did they look at when they gave 18 their approval? 1 4
19 MR. TYRREL: That was prior to Design realizing that 20 the internals were different. Part of the CAQR, part of the j 21 5055 report identifies one of the problems was that Design was 22 not aware that the internals were different.
23 MR. MICHELSON: Now are they using a serial number 24 or something right on the valve as a control mechanism?
25 MR. TYRREL: Serial number on the operator.
O Heritage Reporting Corporation (202) 628-4888
l
()
1 MR. MICHELSON: I mean on the operator.
2 MR. COX: Yes.
3 MR. MICHELSON: Now you go back to the operator 4 serial number and you can pull a pack of paper on that serial 5 number alone and it tells you whether it's comparable to what 6 you were trying to achieve, 7 MR. EBERSOLE: But you may have many for that same 8 serial number?
9 MR. COX: No. Only one. Serial numbers are unique.
10 MR. EBERSOLE: You don't communize a given serial 11 number so you get multiple use of it?
12 MR. COX: No.
13 MR. EBERSOLE: Unique valve numbers.
14 MR. ZIEGLER: Unique actuator numbers.
15 MR. MICHELSON: That tells the day it was 16 manufactured and all that sort of thing.
17 MR. ZIEGLER: Right. And then from that we can tell 18 what internal parts are within that actuator, even though it's 19 the same model number as several others.
20 MR. EBERSOLE: When you go to the shelf to pick out 21 another valve for one that's malfunctioned, you've got to go 22 get the number?
23 MR. ZIEGLER: You have to not only check the model 24 number of the actuator, but you have to check the internal 25 configuration.
Heritage Reporting Corporation (202) 628-4888
1
(
1 1 MR. MICHELSON: You do that by sending Design the l
2 serial number and asking does this look alike. Hopefully the 3 vendor now of the valve has not fiddled around with what you ,
i 4 thought was in that serial numbered operator.
5 MR. EBERSOLE: The valve vendor was responsible for 6 the integral actuation of the whole thing, wasn't he?
l 7 MR. ZIEGLER: Uh huh.
8 MR. EBERSOLE: And what you found was a deviance l 9 from that, am I correct? Or was it only when you opened it up 10 and found out there were different gears?
11 MR. ZIEGLER: Westinghouse had supplied design 12 documentation. They just had not supplied, had not changed --
13 MR. EBERSOLE: From a functionality standpoint you 14 found it was all right in the initial condition, or did you?
15 MR. TYRREL: What was the question?
16 MR. EBERSOLE: From a functionality standpoint, did 17 you find the entire package of valve and actuator to be all 18 right?
19 MR. TYRREL: We've tested the valves and the ones we 20 have tested do work in the configuration they were.
21 MR. EBERSOLE: So you've yet to find a mismatch, the 22 wrong parts?
23 MR. TYRREL: As I stated earlier, we found some 24 spring packs that were not what the valve vendor, or what 25 Limitorqe said should have been in there.
Heritage Reporting Corporation f (202) 628-4888 l l
4 i
1
l I
l
' 129 i C) 1 MR. EBERSOLE: But you don't know where that 2 occurred. You don't know where it was, where it happened?
3 MR. ZIEGLER: We have four possibilities.
4 MR. COX: One of the things that should be noted is 5 this is an ongoing process. We're still in the middle of it.
6 Trying to first off, finish our investigation.
7 MR. MICHELSON: Are you going to talk about valve ;
1 8 maintenance on your slide there? ;
l I
9 MR. COX: Yes I am. This goes in with the findings.
10 MR. MICHELSON: Okay.
11 MR. COX: In what we have done to eliminate the 12 potential sources of any parts discrepancies. One is that TVA
() 13 now requires all maintenance personnel that are involved with i 14 the Limitorge actuators to attend a three day class, a three I 15 day hands-on course that deals with the construction and 16 maintenance of the valve actuators. This course is not just a 17 site-specific course. This is taught out of the Power 18 operations Training Center. All maintenance crews are 19 required to go through this training if they are to be working 20 with the valve actuators. This course goes through and 21 details the internal construction of the actuator, the 22 configuration, and the fact that parts are not necessarily 23 interchangeable. What has to be done to determine whether a 24 parts change can be made, and what happens when you have to 25 replace parts.
Heritage Reporting Corporation (202) 628-4888
r 130 U~
1 One of the other things that we have done to 2 alleviate the potential for a problem is that we have assigned 3 crews for the valve actuators. You don't send all of your 4 electricians to Limitorqe training and then send any 5 electrician out to work on the valve actuator. We've 1 6 established valve actuator crews. There are certain people l 1
7 within the maintenance organization now that Limitorges are 8 their responsibility. And we have come up with a combined 9 electrical and mechanical crew that that is all they do, the 10 valve actuators. So their training has been increased. ,
1 11 What we've also done is gone back and revised the I 12 maintenance procedures that establish control points that call 13 for one, replacement of a like component with an exact like 14 component. If you pull out a 21 tooth gear, then you will put 15 back a 21 tooth gear unless you have an engineering evaluation 16 where it's gone from the operation side over to the 17 engineering side. All the principal parties are involved, the 18 valve vendor, the actuator manufacturer, and it goes thrcugh 19 an actual design change process before you change a part.
20 There will not be any modifications without an analysis.
21 So we feel that by changing the procedures for our 22 maintenance people and educating our maintenance people as 23 well as our engineering people, that we greatly reduce the 24 potential for this problem to come back out of that area.
25 MR. EBERSOLE: What about the craft mixing problem?
Heritage Reporting Corporation (202) 628-4888
I 4
131 1 Do you have to have an electrical and a mechanical man on a {
2 valve in the crafts?
3 MR. TYRREL: The crew is made up of equally )
i 4 mechanical and electrical. So you have both crafts there.
5 MR. EBERSOLE: Both crafts attend these classes?
l 6 MR. TYRREL: Yes, they both have to attend those l 7 classes.
l 8 MR. EBERSOLE: So it takes two men on a valve.
9 MR. TYRREL: At least. !
l 10 MR. EBERSOLE: If not 14. l 11 MR. MICHELSON: I think at the time you reported the ,
1 12 deficiency you indicated there were perhaps about 19 l 0 13 interchangeable parts. Is that still kind of your feeling?
l l
14 or did you find more before you got done, or is 19 the magic 15 number?
l 16 MR. TYRREL: Nineteen is what we used based on a 17 parts catalog supply of the torque. Nin^ teen is a gross 18 number which can be pared down to a smaller number when you 19 look at some of the parts. If you use a worm gear that's got 20 a 33 to 1 gear reduction, the mating gear has to be geared the 21 same way in order to match. In their parts catalog they call 22 that two parts, and for the purposes of report we call it two 1
23 parts. But when you look at the actual operation it's really 24 only one part.
25 MR. MICHELSON: Have you thought through in a Heritage Reporting Corporation (202) 628-4888 l
l t
132
()
1 similar fashion then, and could you give me a list of probably 2 the six or more parts that you think are the ones that one has 3 to watch for? I never did see a listing anywhere of what 4 these parts might be. Could you supply such a listing? You 5 must have had such a listing because when you went through all 6 these inspections you certainly had to tell the fellows the 7 ones to particularly watch for where this condition could 8 exist.
9 MR. TYRREL: The 19 parts we've pared down to five 10 basic components.
11 MR. MICHELSON: The motor is interchangeable of 12 course, is what you're saying.
(O
_/ 13 MR. TYRREL: That's correct. The motor is 14 interchangeable. You could put a five foot pound motor or a 15 ten foot pound motor, the same size actuator.
16 MR. MICHELSON: And they would fit with the same l i
17 mounting bolts and everything like you wouldn't know the 1 l
18 difference if you didn't watch.
19 MR. TYRREL: The motor pinion gear, the worm shaft 20 clutch gear, the worm gear, and the torque spring pack 21 assembly.
22 MR. MICHELSON: In the case of the torque switch 23 itself, do all SMB OOO's for example have only a particular 24 model torque switch that could be in there? There could be j 25 several different torque switches in there, I believe. You Heritage Reporting Corporation (202) 628-4888 l
1
1 133
) l (J
1 didn't list the torque switch. Did you mean to say that 2 wasn't one you worried about?
3 MR. COX: The switch itself?
4 MR. MICHELSON: Have you got the right torque switch 5 in so when you set it at three and know that's going to mean l 6 something. I don't knew what the variations in the models can 7 be.
8 MR. MURPHY: There is a difference in the torque 9 switches as far as a nuclear grade torque switch that is 10 qualified for 5049 application.
11 MR. MICHELSON: I was thinking more in terms of if ;
I 12 it's possible to have more than one particular torque switch ;
O, 13 on a SMB 000.
14 MR. MURPHY: What you actually bolt in there is 15 going to do the same thing.
16 MR. MICHELSON: They all do the same, but do they l 17 all have the same calibration scale on them and so forth?
18 MR. MURPHY: They aren't calibrated.
19 MR. MICHELSON: I know they aren't, but once 20 somebody decides somewhere else that they should be set at 21 three, three ought to mean the same thing. I'm not that 22 familiar with it.
23 MR. ZIEGLER: Let me explain how we set the torque 24 switch.
25 MR. MURPHY: The torque switch setting is only Heritage Reporting Corporation (202) 628-4888
1 l
134
( l 1 appropriate if it results in the proper thrust. The numerical 2 number is irrelevant.
3 MR. MICHELSON: Let me ask you if you break a torque 4 switch and you have to replace it, how do you decide that the 5 torque switch you're putting back in is okay to use in that 6 application? Or is there only one torque switch that would 7 even fit? If only one fits, I have no problem. But if two or ,
l 8 three different kinds would fit.
9 MR. MURPHY: There is one type of torque switch that l
10 varies only in its material qualification. It is one model 11 for a 000.
12 MR. MICHELSON: How about the case of the spring O 13 pack then? There's where you have to count the washers and j 14 measure their thickness so you've got a way of knowing what i
15 you've got there.
16 MR. MURPHY: Several.
17 MR. MICHELSON: So these are the only components 18 then you've found that really could potentially be 19 disadvantageously interchanged. I 20 MR. ZIEGLER: Let me add something hero that I think 21 needs to be added. My technical man can correct me if I'm 1
22 wrong. Whenever we do any type of work on the actuator we, of 23 course we're not operational at Watts Bar, but at Sequoia 24 where we're further along, is we require it to be re-MOVATS.
25 I think we reset the switches based on the test. So we make O .
Heritage Reporting Corporation (202) 628-4888
l 135 l
( l 1
1 sure that the switch is going to perform its appropriate 2 function and deliver the thrust at the right settings, not 3 based on a pre-set number on a switch.
4 MR. MICHELSON: You go back and recalibrate the stem i l
5 thrust and the whole bit when you do this. l 6 MR. ZIEGLER: Every time. l l
7 MR. COX: In each parts change which could affect 8 the output thrust of the actuator we'll require a retest. I '
9 won't say a parts change, but any maintenance that could 10 result in the output thrust, that would go back to adjustment i
11 of stem packing because that could increase the drag on the 12 actuator. So procedurally any maintenance that could O 13 potentially result in a change in the thrust requirement would 14 require a retest. I 15 MR. MICHELSON: Do you own your own MOVATS 16 equipment?
17 MR. COX: At each site.
18 MR. MICHELSON: What do you have, two or three units 19 in TVA, or a dozen of them?
20 MR. MURPHY: Two sets at Brown's Ferry, one set at 21 Sequoia, and one set at Watts Bar, and on occasion we lease 22 additional equipment.
23 MR. MICHELSON: How about your operators that use l l
24 it? They're your own people, I assume, and you are confident 25 they've been appropriately trained to interpret the results l
Heritage Reporting Corporation (202) 628-4888 l
l 136 i
(')
\~J 1
j 1 and so forth? !
2 MR. MURPHY: It's an ongoing process. On occasion 3 we, as our testing demands dictate, we can bring in 4 technicians on a contract basis.
5 MR. MICHELSON: But you're basically doing it in-6 house.
7 MR. MURPHY: That's our long term objective. We 8 feel that's in our best interest to develop skills in-house in 9 the long run.
10 MR. EBERSOLE: Let me ask you. You are very anxious l 11 that the valve actuator has sufficient strength to execute its 12 function under whatever hydraulic condition exists and in the !
() 13 face of wear and degradation over some period of time until 14 you look at it again, so that's why you have a big enough '
I 15 operator.
16 Now you'll remember the case when at Brown's Ferry I 17 was advised that one of the valves, I think it was an RHR j 18 valve, sheared the three inch stem because the torque switch 19 short circuited. In any of your tests do you verify whether i
20 or not you have self destructive capabilities within the valve j 21 such that you might rip the yoke off and create havoc?
22 MR. MURPHY: Every test we verify that. We have 23 minimum requirements. In addition to minimum design 24 requirements we, as often as we can, try to achieve a safety I
25 margin above those minimum requirements to encompass test '
Heritage Reporting Corporation !
(202) 628-4888 l
137 f(h) 1 equipment accuracy and give us more conservatism than the 2 original design calculation. In addition to that we have 3 maximum ratings of the Limitorge operator that we have to keep ,
4 a handle on, both at the torque switch trip point and at the 5 total thrust at the stall, the inertial level. And in 6 addition to that there is another parameter which is the 7 weakest link of the valve assembly itself. Whatever the ;
I 8 weakest component is, it plays a part. ,
9 MR. MICHELSON: Who supplies those numbers? l 10 MR. MURPHY: The valve vendor. The valve vendor ;
11 supplies us information about the Nalves aled the Limitorge 12 information is -- ;
C:) 13 MR. MICHELSON: The kind of information you've 14 identified is more than used to be at least supplied by valve ,
15 vendors, although they may have been fully knowledgeable of 16 these numbers, ;rou didn't have them. Are you now going back 17 to valve vendors and telling them here's a list of the things 1 18 we need to know?
19 MR. MURPHY: That's exactly right. The process was i
20 extensive.
21 MR. MICHELSON: So now you're saying that not only I l
22 do you know what the thrust requirements are for i 23 functionality, but also you know what the limits are to make ,
l 24 sure it doesn't self destruct. '
25 MR. MURPHY: That's exactly correct. .
Heritage Reporting Corporation (202) 628-4888 l
()
1 MR. EBERSOLE: Will this apply to Brown's Ferry and 2 Sequoia?
3 MR. MURPHY: Brown's Ferry is not as far along as we 4 are.
5 MR. EBERSOLE: I mean eventually.
6 MR. MURPHY: Eventually they will have to.
7 MR. TYRREL: This problem was identified as part of 8 our corrective action. All the plants of TVA will --
9 MR. MICHELSON: Clearly you're depending upon the 10 capability of the vendor then to tell you for instance that 11 under conditions on a ten inch steam line that he knows how to 12 figure out what the thrust requirement is as the valve goes to
(,)
(_ 13 closure. The same way at Brown's Ferry, your reactor water 14 clean up line. If that line breaks, it's a six or an eight 15 inch line where the valves are, and when that line breaks and 16 the system is blowing down under two phase flow conditions and 17 so forth, that valve vendor, and it's a gate valve, he somehow 18 knows how gata valves behave as you're planning to blow that 19 kind of a mixture through them.
20 You're depending on that because you can't be 21 expected to do any more than use the numbers you're finally 22 supplied, but you certainly appreciate that you might like to 23 ask them some time how does he even get these numbers.
24 MR. MURPHY: The valve vendor, it must be 25 understood, is not a specialist on our plant systems and we Heritage Reporting Corporation (202) 628-4888 i
139 1 can't expect him to be.
2 MR. MICHELSON: But you do go back and talk about 3 what the functionality of the valve has to be, what it's there l 4 for.
5 MR. MURPHY: That is what I'm getting to. The I 6 system designer, it's up to the system designer to know the 1
7 maximum possible conditions that that particular valve in that 8 location of the system could be subjected to. That's the 9 information that's given to the valve vendor when he makes his 10 count.
11 MR. MICHELSON: The way you're doing it nowadays, 12 you're going back and saying this is what the purpose of this O 13 valve is, these are the bonding conditions on its operation, 14 and then he is supposed to tell you what the thrust 15 requirements are to satisfy it.
16 MR. TYRREL: Worst case condition.
17 MR. EBERSOLE: Have you found, for instance, like at i 18 Brown's Ferry, that those valves, pick the two particular 19 ones, the ten inch main steam line, were they found to be 20 adequate yet, or do you know?
21 MR. TYRREL: As he said, Brown's Ferry it, not as far i
22 along.
23 MR. MICHELSON: We'll ask that when we come to j l
24 Brown's Ferry. '
l rT 25 MR. EBERSOLE: All right. One of the darker corners u {
Heritage Reporting Corporation (202) 628-4888
140 0 of that is that if you do stffer this hypothetical, and I'd be 1
1 2 the first to agree that it's probably a hypothetical, for this 3 main steam feeder line to break, there will be a nearly 4 instantaneous environmental change in the region of the 5 outboard valve. The normal configuration of valves in those 6 years was that was considered to be a normal valve in a normal 7 environment, so it would be subjected to the blast environment 8 and steam and condensation from the hypothetical pipe rupture 9 and could hypothetically be said not to work anyway. That 10 would lead to a legal decision type of problem which is that 11 would destroy your single failure concept and you would have 12 to depend on the internal valve which was an AC vs.ihe to O 13 effect the c1osure and seve the three units. Do yow:fo11o.
14 me? e 15 I'm saying an external steam line blow will change 16 environmental conditions for that external valve.
17 MR. MURPHY: Our Design Group is responsible for ;
18 establishing environmental requirements. ,
19 MR. EBERSOLE: I don't know what tihat is.
20 MR. MICHELSON: That is another whole issue. It's 21 beyond what we can handle here.
I i 22 When you go back to your valve vendor with your i 23 bounding design requirement, do you tell him about the reduced 24 voltage that it must operate under at that particular point 25 and that sort of thing?
Heritage Reporting Corporation (202) 628-4888 1
- - . - . - , - . - , . -m -
i 1 I
?
141
(
1 MR. MURPHY: That's part of the specification.
2 MR. MICHELSON: Soyoufvetoldhimeverythingyou 3 think.you know that establishes the bounding requirement and 4 he comes back and tells you then what the thrust is.
5 Does he supply to you a document that represents the 6 calculations and so forth?
7 MR. MURPHY: He sure does. The end results.
8 MR. MICHELSON: Does he give you the basis for his
- i.
9 calculation including the assumptions he made?
10 MR. MURPHY: His data sheet, it's about a two page 11 document for each unique valve.
, 12 MR. MICHELSON: I can assure you that won't be what
\- 13 I would be looking for. In other words, do you concern 14 yourself on an audited basis at least with the validity of his 15 calculation? How do you know it's any good? What's the basis 16 for his calculational process? There are a lot of ways to 17 calculate,these things. Some of them are good and some of i 18 them aren't so good. Some of them are pretty flaky.
19 /~ MR. MURPHY: This is why in addition to achieving 20 the minimum that he says is acceptable, we add in a safety 21 factor.
22 MR. MICHELEON: He may be off a factor of ten. You 23 aren't going to put factors of ten on this, I'm sure, because 24 if you did then you might worry about putting too big an
{ 25 operator on the device or something. Do you at all chaX1onge Heritage Reporting Corporation (202) 628-4888
/
1 .
I
.. _ _ ______._____f_ _
\
,, h~
i
/ y 5 g,
\ >
') '
' , 142
']{,11 T T 4 1 or at all inquire on an auctitN basis as ta -yhUther the !
g i %! ,
l 2 calculations are any gthd/ , ,<.
)
> 'l s 3 MR. MURPHY: these cre bhooretical quettions.
> .s !
4 MR. MILUFLSN': ThatLSne is not al theoretical -
, i 5 question. That'L a QA que,stion.
6 MR. MURPHY:, I wourd just have to refer to the-7 Design Group.
l 8 MR. MICHELSON: See,.Our concern is, and we'll find 9 out tomorrow what the present r. tats of the art is on 10 determining how you even tell, calculate it. l 11 MR. MURPHY: We ars always open to better 12 calculations, by all meanFl.
O 13 na. M1CHEtSOn: 1 thimx the process you're fe11owing 14 is not an unreasonable one.
15 MR. MURPHY: Compared to what we had before where we l 16 had an extremely high occurrence of excess t.nrust or under 17 thrust, it's not comparable.
18 MR. MICHELSON: It sounds like "it. is much more 19 organizek.
20 .
MR. EBERSOLE: I'll give you something to l 21 contasaplate that the valve vendor doesn't know anything about i 22 ar.d vou do, if you broke that ten inch steam line and if you 23 didn't close the steam line, it won't take you long, wh4t will 24 happen to the plant because the steam will permeate all of one 25 unit and stop all replenishing systems for water. You'll bave Heritage neporting Corporation
] (*J 02 ) 628-4888 J
'[ ;
143 l 1 a direct hele to the outside world to that pipe. You can l 4
-2 imagine the proportions of consequence. The valve vendor, he ,
3 might think that's a two bit job or calculation, and it isn't. l i 4 MR. COX: The answer to your original question Lbout !
- 5 cbtaining this information, Mr. Murphy.would be the end user 4
] 6 in'the maintenance organization. Actually the information is ,
- 7. purchased through engineering and it is routed through the ;
8 procurement quality assurance branch. It is monitored as any
- , 9 QA contract as far as auditability. It is run through PQAV ;
1
- 10 and also through the vendor quality assurance program. Those l
- 11 questions are asked up front prior to the contract being let ,
i i i 12 and the vendor proceeding with the calculations.
() 13 MR. MICHELSON: There is one further aspect of this ;
F i
! 14 business. What we've done is we have compartmentalized the !
l 15 design process, what the vendor is doing at the valve shop and [i n
b 16 what you're doing in the design shop and so forth. You always [
! 17 worry about things getting lost in the cracks. An example, i
! 18 there is a real electrical problem associated with this whole business, particularly if you're under the reduced voltage 19 l 20 condition of a diesel engine starting at the time. So you
[ 21 give the valve vendor some theoretical number like 80 percent 22 voltage and whatever. He somehow is supposed to convert that
{. ,
f, 2? into part of this thrust requirement, but you can't do any t
! 24 testing unless you set up a variable voltage source which I i I i 25 doubt you have, but you could. You don't ever test these j j
i
( Heritage Reporting Corporation j
- (202) 628-48ss l t
l :
. f
1 things other than under rated voltage conditions or very near 2 rated voltage conditions. It gets real questionable as to 3 what the true thrust characteristics of that motor are, unless 4 they've done some good laboratory work on that particular l 5 model motor and so forth, and have got good results, and the ,
t 6 valve vendor is aware of these and knows how to do the 7 calculation. It's not just a fluid hydraulic problem, it's an f
8 electrical problem as well.
MR. COX:
l 9 That goes back to the actual sizing by the 10 actuator manufacturer. Back into Limitorge's shop when the 11 actuator is sized. It's based on the design conditions that f 12 Darrell alluded to, and the reduced voltage at the time of 13 siding.
14 MR. MICHELSON: But the problem was Limitorge was f 15 never told that that actuator was going to go on this t
16 particular valve that had to isolate under these particular ;
- 17 conditions. It's an integration job. No one person was ever 18 told and could focus on the design in terms of knowing all 19 about what this thing was going to do. Limitorqe was selling [
20 operators and the valve vendor was supposed to make sure he f
21 picked the right operators. .
4 22 MR. MURPHY: From copies of the old vendor data f 23 sheets that I've seen from Limitorqe the Delta P's were ,
f 24 specified.
25 MR. MICHELSON: That's almost meaningless. They I O !I Heritage Reporting Corporation i (202) 638-4888 1 1 .
145
)
1 don't know what they ever did with that in a couple of static 2 calculations.
3 MR. BELTZ: Typically what we do now and what we've 4 done in the past five or ten years when we order a valve, we 5 use a valve data sheet. Of course we list the Delta P's on 6 that valve data sheet. We also list the harsh environmental 7 conditions, we list the upset conditions, and the reduced 8 voltage. So the valve vendor has all that information.
9 Through our QA process we say okay, you have a process that 10 controls your design.
11 MR. MICHELSON: Delta P is only a static thing.
12 Dynamic Delta P's if you could calculate them, that would be O 13 useful. But all the Delta P is good for is to tell you what 14 it takes to activate those under a differential pressure 15 condition. That's one important part of it, no doubt. But it 16 turns out that depending on who you believe on tests that have 17 been run, it may be much more rigorous to close against 18 flowing fluid than it is to open against full Delta P. If you 19 could ever get the answer that says if you can open a gate 20 against full Delta P you've covered it all, that would be 21 nice, but I don't get people to tell me that's really the 22 case. Maybe the dynamic loadings under the chattering and 23 everything is a more rigorous requirement than opening against 24 full Delta P.
25 MR. MURPHY: In fact it varies from valve to valve.
Heritage Reporting Corporation l (202) 628-4888 )
I i
146
]
1 These data sheets we get from our valve vendor include both P
2 the opening requirement as well as the closing requirement.
3 for a valve that opens on limit entirely, you have up to the i
4 full output of the operator to get the valve open. For a 5 torque seated valve we have to use the torque switch in the 6 last few percentage of valve travel to get our tight seat on ;
i 7 it. It would be in a repeatable, controllable configuration.
8 That is set up specifically for the closing requirement.
9 MR. MICHELSON: Do you back ce*.t any valves on ;
10 torque? l j 11 MR. MURPHY: No.
12 MR. MICHELSON: Do you use torque for protection
() 13 during mid travel on --
14 MR. MURPHY: For non-safety related applications.
15 MR. MICHELSON: But none of the safety valves use l l
16 torque in mid travel? !
l 17 MR. MURPHY: There are some safety related valves j 18 that do have the torque switches, but they're not critical 19 actuators for accident mitigation.
20 MR. COX: The last item on the list here is the j i
21 actuator switching. What we've done is implement engineering 22 design documents that tie particular actuators through a 23 unique serial number to a particular valve. So that at this 24 point in time any change between, any switching of actuators 25 from one valve to the other is going to require an engineering I
Heritage Reporting Corporation (202) 628-4888
_ , ~ -
147 O
1 analysis to be sure that the valve that the actuator is going 2 to is within the limits of the actuator that we intend to use.
3 The bottom line is before we change anything now we 4 go back and we analyze. After we analyze, we go back and we 5 test.
6 MR. MICHELSON: It does make it hard to change one 7 out in a hurry though, doesn't it? Do you use your on-site 8 people for this analysis or do you ship it back to Knoxville?
9 MR. BELTZ: In most cases if we want to switch an 10 operator, we would go back to the valve manufacturer first and 11 say look, we want to switch the operator that we bought on 12 that particular data sheet with these design conditions, and O 13 now we want to switch it to another valve with these design 14 conditions.
15 MR. MICHELSON: Do your on-site engineering people 16 do this or do the people in Knoxville do this?
17 MR. BELTZ: The on-site people do that. They have 18 the capability to do that.
19 MR. MICHELSON: So each site has set themselves up 20 to be sufficiently knowledgeable in this area to make the 21 judgments.
22 MR. BELTZ: That's correct.
23 MR. COX: That leaves us with several things. One 24 is that now we have a controlled actuator configuration. The 25 actuator has been stripped down, it's been inspected, the Heritage Reporting Corporation (202) 628-4888
.t i
i ;
i i 148
- O 1 internals have been documented. We know what's there. We 2 also have the controls installed to prevent parts switching. !
- 3 You don't go out and change anything indiscriminately. l 4 They've also gone through the signature analysis testing, l 5 through for us it would be MOVATS testing, to determine !
^
6 whether ce not this actuator will actually put out the thrust !
I 7 that we would require of it# j
! 8 We have~gone back on the third item, enhanced the j 1
9 maintenance procedures, and enhanced the training that our !
10 maintenance people get now, so that the people that operate on l 11 this equipment are knowledgeable in what they're doing and the 12 limitations and restrictions in their activities. They know l' t
O 12 how far they can go prior to going through a design change.
! 14 Generic applicability, though, is the last item. !
j 15 It's been documented for Watts Bar. We've got four other j t -
j 16 sates, though. As part of, and if you'd like you can flip l
17 back to the actual flow chart, once a condition is identified l 18 as being potentially wrong and if we flip back to the i
) 19 investigative process you'll see it right here. -If we do come j j !
! 20 up with something that is questionable, we have to do the '
i 21 generic evaluation for all sites. So that once a problem is 22 identified at any one site, it becomes evaluated for all sites li 23 at each site. The site organization will go through and l
24 determine whether or not this problem actually exists for 25 their site. So as soon as Watts Bar initiates the CAQR, I Heritage Reporting Corporation
] (202) 628-4888
?
l
,--..-......-...---,---_.-.--,.,-,._,..---.---.,,,-...-----n-- -
149 .
O I 1 Sequoia goes through and analyzes; Brown's Ferry has responded i 2 and Brown's Ferry knows that this action has to be done, 3 they're not as far along; and the same thing for the 4 Bellefonte nuclear site. But all of them are tied together in 5 that a problem at one kicks off a search at all of them.
6 MR. MICHELSON: On a relate
- question, these 7 operators have in some cases been qua. 2d for the harsh 8 environment inside a containment, for other courses they are Q 9 listed items but mild environment application, outside r 10 containment. Has somehow the difference between a harsh 11 environment qualification, whatever that operator looks like 12 that's any different. Well let me ask it differently.
13 Are the two operators identical? Whether it's 14 qualified for harsh or non-harsh?
15 MR. COX: They will be very similar. They're not 16 identical. .
17 MR. MICHELSON: In the ways in which they are non-18 identical, have those been identified to workers to be sure to 19 protect, in both cases in fact, how do you know what 20 particular feature that operator has to be preserved to 21 maintain its environmental qualifications?
22 MR. MURPHY: There are work documents that identify l
23 whether or not the device is a 5049 device. And there are !
24 steps that are flagged in obvious places in the procedures for f
25 maintenance and corrective action. !
O '
Heritage Reporting Corporation ,
(202) 628-4888 t
i 150 0 .
1 MR. MICHELSON: Maybe there's a particular seal that i
2 has to be regreased when you close it or has to be replaced 3 when you close it or whatever. That kind of thing is in the !
4 procedure route.
i 5 MR. ZIEGLER: It's also in the EQ binders under the 6 quality maintenance data sheets and they're very specific as
- 7 to any extraordinary measures you need to take.
8 MR. MICHELSON: This would make it non-9 interchangeable if it's a mile environment operator, but 10 identical in all other respects. You couldn't put it inside a 11 containment unit? t 12 MR. MURPHY: You could install it if everything else I
() 13 was equal, but if there is some uniqueness about its 14 components you would have to go through that procedure as a I 15 final check off and verify that the requirements are met.
1
! 16 MR. MICHELSON: You don't happen to recall off hand ;
l 17 what is unique about what's done on a Limitorqe operator to I 18 make it environmentally qualified for say a harsh environment?
! 19 MR. MURPHY: There are some plastics that are unique 1
1 20 to harsh environment. There are motors that require two l 21 drains on the low point, that sort of thing. Main gear case l 22 grease relief valve.
23 MR. MICHELSON: The lubricants are even different? ll 24 MR. COX: We can take a step to alleviate that l l 25 problem by settling on one lubricant. 1
)
1 Heritage Reporting Corporation ,
- (202) 628-4888 I i
()
1 MR. MURPHY: It's good enough for everywhere. ;
2 MR. MICHELSON: The other problems with Limitorges 3 in the past has been a small one, and that is some of the wire 4 wasn't exactly insulated the way it should have been. That's 5 all been worked out and all replaced or whatever?
6 MR. MURPHY: Massive change out.
7 MR. MICHELSON: You've gone through all of that.
8 MR. ZIEGLER: At Sequoia we're finished with it. !
l 9 MR. MICHELSON: You had to go through every single i
10 operator and look inside to see what was there?
11 MR. ZIEGLER: I believe what we did was just '
12 arbitrarily decided to change the wiring out inside of those 13 things. We didn't go look, we just rewired them all with 5049 14 applications. :
1 15 MR. COX: We went back and settled on one type of 16 internal wiring, one type of grease and one type of seal.
17 MR. MICHELSON: And that was all good for harsh 18 environment and therefore more than good enough for mild. But 19 that would not necessarily make the operator yet qualified for 20 harsh environment because some things you didn't do. You ;
i 21 might not have replaced certain plastic.
) 22 MR. MURPHY: Not in itself, no.
23 MR. MICHELSON: The wiring you replaced throughout, 24 but it doesn't mean you've rebuilt the operator to make them 25 all qualified for harsh, though.
1 Heritage Reporting Corporation (202) 628-4888 l
l
152 A
V 1 MR. MURPHY: Not just changing the wire out, no.
2 There are some plastics that are acceptable outside 3 containment. Of course the inside containment plastics are 4 acceptable anywhere. The long term philosophy is to 5 eventually only have the parts available so we can use them ,
6 anywhere. It's an expensive effort but it has a safeguard 7 advantage.
8 MR. ZIEGLER: We may not ever get there, either.
9 The other thing, since I've been through a few procurement 10 inspections here lately, is that we have segregated storage in 11 our materials controls right now, and we've gone back and 12 looked at past installations of maintenance items to make sure 13 we didr,'t have a problem. We don't ever want to get to where 14 we have to do that again. It was a massive effort. So we
- 15 segregate out our 5049 materials, and they're tagged such.
1 16 It's difficult to make a mistake. If it's not 5049 it's not l
17 going to be handed out for that application.
i 18 MR. MICHELSON: Things have changed, haven't they? l 19 MR. EBERSOLE: I would like to give you some history 20 so you might want to go back and have a look at it. An aspect 21 of that outboard valve not working, if I recall correctly, 22 some consideration was given to adding a third valve in the 23 line, I think in a piece of straight steam line inside the 24 containment, wherein such a valve could be put. But it was 25 never put.
. Heritage Reporting Corporation i (202) 628-4888 i
- l
153 '
1 Another argument that still stands, those steam 2 lines are kept open because of a hypothetical need for rapid 3 acceleration of the HIPC NRCIC, and you don't need it. So you 4 can invoke the question of why are they open in the first l
5 place? If they were closed, you wouldn't have the standing 6 head. That's something else.
7 MR. MICHELSON: I don't know, though, that they ever 8 established whether or not you can really operate with HIPC 9 closed and then opening it, whether you could tolerate it. I 10 don't know. That's what you'd have to do. If you could show 11 that HIPC doesn't need to come on quickly, you can keep the 12 lines closed, so if you press the pipe the most you have is a 13 little pipe --
14 MR. EBERSOLE: If you've got a hole that big, Carl, 15 you're going to have a large blow out.
16 MR. MURPHY: There was one valve of a critical I l
17 nature on the supply side that they did put a small bypass l 18 valve around so they could be assured of equalizing the i
19 differential so the valve would be assured to open.
20 MR. EBERSOLE: You may not remember that during the j j 21 early '74, I think, there was a tornado at Brown's Ferry and 22 they had the valve malaligned and had collected a slug of l 23 condensate and they nearly knocked the HIPC off its foundation 24 straightening the pipe out. When the valve on automatic 25 signal recalled itself to function from a non-aligned J
Heritage Reporting Corporation 1
(202) 628-4888 1
2 i
i 154 i O '
1 condition which was not in the design basis, that was an 2 interesting revelation, that it should not have come on-line ,
3 from a mal-aligned condition. I don't know what they ever did 4 about coordinating that signal again, but it was a fascinating l 5 possibility right there. It could have blown that pipe. And 6 of course if you don't close it, Brown's Ferry is in big .
l 7 trouble.
8 MR. MURPHY: That may be the same valve I remember, 9 them having made the nod to put a bypass around it so they 10 could leave it closed.
11 MR. MICHELSON: I think you leave your steam lines 12 open yet at Brown's Ferry, don't you? !
13 MR. EBERSOLE: It would be well to examine why you f l
14 do that. I would offhand say that's the most risky position }
15 to have it in, by far. You have a standing pressurized line j 16 that you absolutely must close if anything got into that line. !
i 17 MR. MICHELSON: Let's proceed.
18 MR. COX: To sum it up, what we've done is {
19 identified that there was a potential for problems, and went
]
20 back in through procedure and education and tried to alleviate i l
21 potential sources for people making mistakes. Actuators that 1 l
- 22 are in the plant now are being reworked so that we know what l 1 . I j 23 we have, and we're placing these into controlled programs, and l 24 that new actuators going through the installation process are l
i l
25 also examined. That's the one thing we hadn't gone over O
Heritage Reporting Corporation (202) 628-4888 i I
155 O
1 before. We've covered our end of the house. If there's a 2 mistake in the configuration of the actuator prior to it 3 getting to us, then it still shows up in your testing because 4 the testing is going to show whether or not the thing should 5 be functional. So we have taken steps to cover not only 6 ourselves, but anything coming into our organization.
7 MR. MICHELSON: Your problem was, I have not heard 8 any reason to believe that it was unique to TVA. Other people 9 must have had similar situations develop. To your knowledge 10 has anybody else claimed that this has been a problem for 11 them? Have they gone through the kind of thing you had to go 12 through to straighten it out? Or is it some reason to believe 13 that it's unique to TVA?
14 MR. ZIEGLER: I don't think there is any reason to 15 believe that what we found is unique to TVA. But when we were 16 trying to identify a root cause we came up with four 17 possibilities.
18 MR. MICHELSON: INPO is aware of your difficulties 19 that you think you found, is that a correct statement?
20 MR. ZIEGLER: I think that's fair.
21 MR. MICHELSON: To your knowledge is INPO aware of 22 it? Or do you just think they ought to be aware of it?
23 MR. ZIEGLER: I know they're aware of the SSE 24 because they have to --
25 MR. MICHELSON: They must have read it. I even Heritage Reporting Corporation (202) 628-4888
156 1 became aware by reading it.
I 2 MR. ZIEGLER: By definition in the experience review l 3 programs, everybody has to incorporate those.
4 MR. MICHELSON: Do you know if they have done 5 anything, or have they ever communicated with you to' find out' 6 the details of the problem and see if it was something they i 7 ought to appraise other utility members about?
8 MR. EBERSOLE: How about NRC, Carl?
9 MR. MICHELSON: NRC can read too, just like I read i 10 it.
11 MR. EBERSOLE: But they've got to pick up their ears [
12 quickly.
O 13 MR. MICHEtSON: we need to asx NRC seperete1y whee 14 they're doing about it, which we will do. but I just j 15 wondered, it puzzles me a little bit that you people are doing 16 a good job of cleaning it up, but how about others? Do they 17 not have the problem or are they just not going to clean it 18 up? i i
19 MR. ZIEGLER: I guess we can't speak for them.
20 I'm not sure we know what everyone else is doing.
i 21 MR. MICHELSON: I realize that. I just wanted to '
22 see if to your knowledge you had --
23 MR. ZIEGLER: As far as we know, what we were )
l 24 finding through the work done at Watts Bar was that there were Heritage Reporting Corporation (202) 628-4888 i
I 157 l
(^) l 1 some differences in a fair number of the spring packs, 2 MR. MICHELSON: Did you find any changes, did you 3 have to make any changes as a result of this program? In 4 other words, did you find operators that actually were, that 5 they should have been 21 tooth gears and there were 18 of them 6 in there.
7 MR. MURPHY: No, not on the gears.
8 MR. MICHELSON: I mean on any aspect, a gear only 9 being an example. Did you have to make any changes to your 10 operators? l 11 MR. TYRREL: We made changes to the spring packs 12 that we found that we couldn't document where the change was '
() 13 made.
14 MR. MICHELSON: What I'm driving at is simply, is !
15 this just a paper cleanup or is this an equipment cleanup as 16 well?
17 MR. MURPHY: It's not just a paper cleanup.
18 MR. MICHELSON: If it's an equipment cleanup, then
)
i 19 maybe other people need to clean their equipment up too. '
20 MR. MURPHY: The final word for whether the 21 equipment is functional has to dictate the configuration of 1
22 the equipment. No matter what the paperwork says, if your 23 spring pack is too sof t for you to get your design thrust ,
1 l
24 requirement, you're going to have to put a different spring 25 pack in there because you have to go stiffer. If your gearing ,
(
Heritage Reporting Corporation (202) 628-4888 ,
i l
~ 158 i
1 is too fast for you to get your design thrust requirement, 2 you've got to slow your gearing down as long as you can stay 3 within your tech spec requirements for closure time. This 4 has been done. We have to go through a design procesey and we 5 have been through a design process. >
6 MR. MICHELSON: Part of your equipment cleanup may l 7 be due to the fact that the vendor is a lot smarter than he 8 was 20 years ago when he sized the operator. Now he says you 9 do have to change the spring pack in it or whatever.
r 10 MR. MURPHY: This is very important that the !
11 original thrust calculations were not adequately conservative.
12 When we went through in 1986 and got fresh recales, some of
- O 13 the o1d numbers were obvious 1r 1ew, we got some new numbers.
l l 14 Where the installed equipment may havut been adequate for the l 1 t 4 15 old design values, now their formulas are better.
I
- 16 MR. MICHELSON
- But again that same problem can l 17 exist in other places. It needs to be fixed.
18 MR. ZIEGLER: I know INPO has identified to 19 everybody the possibility of spring packs losing their ,
j 20 springs, so to speak.
21 MR. MICHELSON: Have we identified the fact that i.
j 22 maybe people need to go back and recheck the calculations that t
23 were done 20 years ago to see if the stem thrust was really 1 24 correctly specified to begin with?
25 MR. ZIEGLER: I'm not sure that Limitorqe has done O
, Heritage Reporting Corporation (202) 628-4885 l,
l
() ,
P 4
1 that. I think through the bulletin 8503 and the testing 1
2 required under that, I think pretty much points out that if 1 3 you want to be sure of what you got out in the plant, then you 4 need to test it and one test is not good enough. You need to 5 test the entire --
6 MR. MICHELSON: You think that should be a part of 7 the answer to the hulletin as to what the utilities program is 8 to assure that they really do have functional valves. ,
9 MR. ZIEGLER: I don't see any way around it. That's 10 what we're all trying to do, to make sure they work.
11 MR. MURPHY: If we want a comprehensive program and ,
12 know that the equipment is definitely going to perform at 13 least up to the level the valve designers have postulated.
14 You cannot achieve those levels by taking a torque switch 15 setting recommendation and going out and setting all your
- 16 switches to two or two and a half or three. It won't do it.
l
+
17 MR. EBERSOLI: It was once thought that's all you i
I 18 had to do.
19 MR. MURPHY: We were all in the same boat. We all 20 blindly believed that you take the recommended number, you set >
1 r 21 them to those numbers, and it will all be okay. We know a lot i
]
i 1
22 better now, those that are resisting this knowledge and have ;
J 23 got a ways to go. l j
4 24 MR. MICHELSON: I believe we've probably interrupted 25 our speaker on a number of occasions. You've been very polite Heritage Reporting Corporation i
(202) 628-4888
160
()
< 1 to not get angry at us, but we are just trying to learn and j 2 appreciate how all these situations develop.
) 3 MR. COX: We appreciate the help from the rest of 1
2 4 the guys. But unless you have any other questions, that does 0 5 wrap up the presentation.
6 MR. MICHELSON: Thank you very much. I think you're 7 doing an admirable job. I have to say that 20 years ago I 8 tried to get the folks down there to take an old switch hard 9 type facility.
- 10 MR. EBERSOLE
- I didn't have any luck.
11 MR. MICHELSON:
More than 20 years ago we took over 12 100 fast high speed tracings on motor current at Brown's
() 13 Ferry, but it seems that all those charts have disappeared. ,
14 It's only an archival thing. It was to be taken until we 15 figured it out. We had a suspicion that we could do something 16 with them, but we weren't smart enough to know what. If we l l;
17 had them, that would be helpful.
18 MR. MURPHY: I'll tell you something. About a year i 19 before we started using MOVATS at Brown's Ferry I was doing 20 exactly the same thing. I had the responsibility of 21 answering an LER in 1984 on a valve failure. I essentially
- 22 set up some of these same experiments, taking high speed motor 23 current traces, and high speed voltage readings, and I 24 thought, I knew more than I did before I did those tests, but l
25 I was so far from knowing the real important aspects that were i
Heritage Reporting Corporation j
(202) 628-4388 1
161 0 occurring that even I didn't, you'd get a clue, but you didn't 1
2 have much more.
3 MR. MICHELSON: Gentlemen, we thank you very much 4 for coming in such numbers to help us out. We certainly 5 appreciate the information. Thank you again.
6 I also would like to thank the other people, some of ,
7 whom had to leave, but we really, I think it's been a very 8 fruitful afternoon in terms of learning, and hopefully 9 tomorrow we'll learn even more because tomorrow we get into a 10 little more exotic aspects of testing itself from where we're 11 at now on testing. I think it's going to be educational to 12 see what some of the various vendors are doing now, I
() 13 particularly in terms of check valves and so forth which is 14 another subject which you weren't here to discuss, but I'm ,
15 sure you're very much interested in it.
l 16 So thank you again, and we'll see those of you who ;
l 17 maintain and persevere, we'll see you tomorrow.
18 (Whereupon, at 5:08 p.m. the hearing was adjourned.)
3
} l l
4 f i
.}
1 1
(
- Heritage Reporting Corporation
] (202) 628-4888 l
1
_ - - . _ ~ _ - _ _ _ . _ _ . . _ _ - . _
3 ,
j 1 CERT 1TICATE 2
lO 3 This is to certify that the attached proceedings before the i
j i
4 4 United States Nuclear Regulatory Commission in the matter of: j
! 5 Names ACRS: BOARD PANEL-- Reliability Assuranco Subcommittee I
i 6 l 1 ,
7 Docket Number:
Uashington, D.C. !
l 8 Place:
l
- 9 Date
- 11arrch 7, 1938 l>
1 10 were held as herein appears, and that this is the original f 11 transcript thereof for the file of the United States Nuclear
) 12 Regulatory Commission taken stenographically by me and, i
i 13 thereafter reduced to typewriting by me or under the direction l 14 of the court reporting company, and that the transcript is a !
O 15 tree "e cc=r t
/S/
rec re or th ch A t r 9 1#9 ar c at=9 - !
16 1:
j 17 (Signature typed): Joan Rose 18 official Reporter !
19 Heritage Reporting Corporation )
20 21 22 l
i 23 i
{ 24 1
1 25 i
\
i j O ne sts,< :1mporting Corgeration
- (202) 626-4888 1
I
.I
. . ..~ ws t PRESENTATION TO THE A.C.R.S.
O VALVE RELATED RESEARCH & DEVELOPMENT Conducted by the Electric Power Research Institute i T. U. Marston Director Engineering and Operations Department EPRI Nuclear Power Division EPRI O
i -
1 OVERVIEW l EPRI VALVE RELATED RESEARCH 03JECTIVES/ BENEFITS
. IMPROVED PLANT SAFETY
. IMPROVED PLANT AVAILABILITY
. REDUCED RADIATION EXPOSURE OF PERSONNEL
. REDUCED PLANT OPERATICN AND M AINTENANCE COST l
1 4
EPRI O
Presented by : Ted Marston, Director )
i Nuclear Engineering & Operations l EPRI l 1
PRESENTATION TO THE A.C.R.S.
O ;
wCtrarmammrarmarS,mmmt QVERVIEW_
EPRI VALVE RELATED RESEARCH i METHODOLOGY '
. PRIORITIZE AND ADDRESS PROBLEMS
. DEVELOP / TEST / DEMONSTRATE GENERIC CORRECTIVE :
ACTIONS AND GUIDELINES
. MAIN STEAM LINE PLUG VALVE STEM PACKING IMPROVEMENTS
. APPLICATION GUIDELINES FOR CHECK VALVES (WITH NSSS OGs)
TECHNICAL REPAIR GUIDELINES FOR MOV (NMAC)
. TECHNOLOGY TRANSFER OF RESULTS THROUGH WORKSHOPS / ,
SEMINARS 1H PLANT DEMONSTRATIONS EPRI :
O NUCLEAR ENGINEERING AND OPERATIONS DEPARTMENT ROOT CAUSES OF "VALVE PROBLEMS" l
! . FAILURE TO TREAT AS AN ENTITY THE VALVE,ITS OPERATOR, AND ,
ITS SPECIFIC APPLICATION l
. FAILURE TO FOCUS EXISTING KNOWLEDGE l
. PROBLEM SOLVING ON A CASE BY CASE BASIS APPROACH
. OVE.'" Y CRITICAL DESIGN AND PERFORM ANCE RiiOUIREMENTS EPR1 O
Presented by : Ted Marston, Director Nuclear Engineering & Operations EPRI
e e
PRESENTATION TO THE A.C.R.S.
O wmmmmmmmmrmre VALVE PROBLEMS CONTRIBUTORS ANDIMPACT
. VALVE M ALFUNCTIONS ( ALL CATEGORIES) RESULT IN APPROXIM ATELY A 2% LOSS OF US LWR CAPACITY
. APPROXIM ATELY 10% OF THE IN PLANT VALVE POPULATION CONTRIBUTE TO THIS LOSS.
. APPROXIM ATELY ONE THIRD OF THIS CAPACITY LOSS IS ASSOCI ATED WITH S AFETY RELATED VALVES
. M AJOR EVENTS ARE OFTEN LOW PROBABILITY / SEVERE CONSEQUENCE l
EPRI O
EPRI VALVE RESEARCH i
PUBLISHED REPORTS
. NP 241 ASSESSMENT OF INDUSTRY VALVE PROBLEMS (NOVEMBER 1976)
. NP 1136 LIMITING FACTORS ANALYSIS OF NUCLEAR PLANTS l NP 1139 ( AUGUST 1979) l
\ \
. NP 1473 EFFECTS OF VALVE PERFORM ANCE ON BWR UNIT C APACITY
( AUGUST 1980)
. NP 2381 & BWR M AIN STEAM NP 2454 ISOLATION VALVES (JULY 1982)
. NP 2560 VALVE STEM PACKING STUDY (AUGUST 1982)
. NP 2628 SAFETY AND RELIEF VALVE TEST REPORT (DECEM BER 1982)
. EPRI l l
O Presented by : Ted Marston, Director
- Nuclear Engineering & Operations
- EPRI l
2
a f
.1 l
PRESENTATION TO THE A.C.R.S. !
O !
I l
EPRI VALVE RESEARCH !
PUBLISHED REPORTS '
. NP 3111 ON LINE LEAKSEALING METHODS (M AY 1933)
. NP 3220 COBALT CONTAMINATION RESULTING FROM VALVE M AINTENANCE (AUGUST 1983)
. NP 3291 HONING TOOL FOR MSIV SEATS (NOVEMBER 1983)
. NP 36d6 FIELD TESTING OF SM ALL HAND OPERATED VALVES (JULY 1984)
. NP 3734 PWR ADVANCED OVER PRESSURE PROTECTION (OCTOBER 1984) ;
EPRI O l 4
I l
NUCLEAR ENGINEERING AND OPERATIONS DEPARTMENT EPRI VALVE RESEARCH PUBLISHED REPORTS
. NP 3E49 BWR MAIN STEAM LINE PLUG (SEPTEMBER 1984)
. NP 3611 KEY VALVE PRIORITIZATION STUDY (OCTOBER 1984)
. NP-3926 IN SITU APPUCATION OF VALVE SEAT HAPDFACING (M ARCH 1986)
. NP-4236 SET POINT TESTING OF SAFETY VALVES USING ALTERNADVE TEST METHODS (SEPTEMBER 1985)
. N P-4 254 IMPROVEME NTS IN MOTOR-OPER ATED VALVES 1 (INTERIM REPORT. NOVEMBER 1985)' l
. NP-425$ VALVE STEM PACKING IMPROVEMENTS (INTERIM REPORT !
F EB RU ARY 1956)*
- NP5479 APPUCATION GutDELINES FOR CHECK VALVES (J ANUARY 1988) I
. _ . . _ . . _ . . EPRI O i l
Presented by : Ted Marston, Director l Nuclear Engineering & OperatlOns l EPRI
PRESENTATION TO THE A.C.R.S. !
O TMc[IlRTsoih6 RECENT AND CURRENT EPRI PROJECTS ADDRESSING l "KEY VALVES" -
ISSUES ,.
j
. RP2233 2 "VALVE MOTOR OPERATORIMPROVEMENTS" ,
! . RP2233 3 "VALVE STEM PACKING IMPROVEMENTS" )
. RP2233 5 "APPLICATION GUIDELINES FOR VALVES"
. RP2233 20 "APPUCATION GUIDELINES FOR CHECK VALVES"
.NMAC MOTORIZED VALVE REPAIR GUIDE l
.NMAC MOTORIZED VALVE APPUCATION GUDE [
EPRI i _
O t i
i i
.i ,
i l
l l i i
-i l
l l l
l.
O l Presented by : Ted Marston, Director 1
Nuclear Engineering & Operations l EPHI l
i
? Lc>w 2 l
a l'O I
i
,1 !
l MOTOR-0PERATED VALVE j
4 j -----0----- l I
i CHECK VALVE !
4 I l
ACTIVITIES !
3 O !r ED MOORE l GARY PETERSON !
t t I i INSTITUTE OF NUCLEAR POWER OPERATIONS l l !
MARCH 7, 1988 .
I l
I l i
i O
- _ _ . . . - _ - .__..__-___..-_____..-________.____-______--__..__-___-._._I
'O 0VERVIEW AND BACKGROUND INDUSTRY EVENTS o MOTOR-OPERATED VALVE FAILURES o CHECK VALVE FAILURES AND DEGRADATION RESPONSES BY NRC TO EVENTS o NRC IE BULLETIN 85-03 o AEOD REVIEWS OF MOV PERFORMANCE ,
O o NUREG REVIEW 0F CHECK VALVES
+
0
()
0VERVIEW AND BACKGROUND (CONTINUED)
INP0 ACTIONS FOR MOTOR-0PERATED VALVE AND CHECK VALVE EVENTS o EXPERIENCE DISSEMINATION (SERs, S0ERs, 0&MRs) o EVALUATIONS / REVIEWS o INTERACTION / COORDINATION WITH EPRI, NUMARC l
TODAY'S REVIEW' o M0V INITIATIVF- IN PROGRESS o CHECK VALVE ACTIVITIES TO DATE 1
2 m _n __-_ _ _ __ . _ _ _ . _ _ . _
O MOTOR-0PERATED VALVES j l
l o 3 SIGNIFICANT OPERATING l EXPERIENCE REPORTS (SOER) l o 20 SIGNIFICANT EVENT REPORTS l (SER) s o 11 OPERATIONS & MAINTENANCE l REMINDERS (0&MR)
SIGNIFICANT EVENT O NOTIFICATION (SEN) 4 1
l i
1 O
l
.g,,,w,. , , ,-, - - - , -- ,,--.-,-g-m, ---,-,-r -- - -- --v,w- ,----,m
,-----,,.,-,,,,---r-,- + .-r, -r--,--, r----,---,-, ,
()
CHECK VALVES o 3 SIGNIFICANT OPERATING EXPERIENCE REPORTS (SOER) o 24 SIGNIFICANT EVENT REPORTS (SER) o 6 OPERATIONS & MAINTENANCE REMINDERS (0&MR) o 1 SIGNIFICANT EVENT NOTIFICATION (SEN)
O l
1 O
1 C)
RELATED INPO ACTIVITIES NPRDS COMPONENT FAILURE ANALYSIS SYSTEM (CFAS)
INDUSTRY SELF-ASSESSMENT INITIATIVES o MAINTENANCE ASSISTANCE AND REVIEW TEAMS ACCREDITATION PROGRAM EVALUATIONS o OE REVIEW o PM
() o MACHINERY HISTORY o POST-MAINTENANCE TESTING o ROOT CAUSE ANALYSIS o MATERIAL CONDITIONS ASSESSMENT TECHNOLOGY TRANSFER o WORKSHOPS o EVALUATIONS o ASSISTANCES i
0 O
(BEGIN GARY PETERSON SPEECH)
O i
O I
.,,__,_,-,,-_,,--_,,,,,_s ,. _,_ _,, _ _ , . . -,
O MOTOR-0PERATED VALVE INITIATIVE INP0 INITIATIVES INITIATIVE RESULTS FUTURE PLANS CHECK VALVE RELATED ACTIVITIES REPORTS TO INDUSTRY EVALUATION / FIELD l ACTIVITIES INTERACTIONS
() -
INDUSTRY O
1 1
() l
- l MOTOR-0PERATED VALVE INITIATIVE 0 M0V PROBLEMS OVER THE PAST YEARS ,
l DAVIS-BESSE LOSS OF AUX FEEDWATER l (JUNE 1985)
SAN ON0FRE UNIT 3 RAPID C00LDOWN (DECEMBEP 1985) 0 M0V PROBLEMS ADDRESSED TO INDUSTRY
() FOR ACTION INP0 REPORTS (SERs SOERs, 0&MRs)
INP0 EVALUATIONS / VISITS NRC AE0D REVIEW (DECEMBER 1986) 0 INP0 INITIATIVE TO IMPROVE MOV PERFORMANCE IN INDUSTRY PLAN CONTAINS SEVERAL ELEMENTS IN CONJUNCTION WITH EPRI AND NUMARC l O
MOTOR-0PERATED VALVES ACTION PLAN ELEMENTS o LETTER TO SENIOR UTILITY MANAGEMENT TO HIGHLIGHT MOV PROBLEMS AND IMPORTANCE OF RESOLUTION LETTER MAILED IN SEPTEMBER 1987 o DETAILED REVIEW AND ANALYSIS OF NPRDS FOR MOV FAILURES
-- EVALUATIONS AND MART o INCREASED EMPHASIS ON MOV MAINTENANCE DURING INP0 EVALUA m HS AND ASSIST VISITS O
HANDS-0N TRAINING FOR EVALUATIONS COMPLETED EXTRA EVALUATOR ON SELECTED EVALUATIONS AND MART VISITS TO REVIEW M0V MAINTENANCE REVIEW 0F SOER IMPLEMENTATION O
l I
b) MOTOR-0PERATED VALVE ACTION PLAN ELEMENTS (CONTINUED) 0 SUPPORT EPRI NUCLEAR MAINTENANCE ASSISTANCE CENTER (NMAC) IN ADDRESSING MOV TECHNICAL REPAIR ISSUES TECHNICAL ADVISORY GROUP STEERING COMMITTEE MOV TECHNICAL NEPAIR GUIDELINE REVIEW o VISITS TO PLANTS THAT HAVE GOOD MOV PERFORMANCE TO l GATHER INFORMATION ,
COMPLETED FEBRUARY 1988
() o COMMUNICATE MOV PROBLEMS AND SOLUTIONS TO THE INDUSTRY :
WORKSHOP SUBJECTS DISCUSSIONS WITH INDUSTRY REPRESENTATIVES SHARING 0F EPRI PRODUCTS l 1
i
()
l
MOTOR-0PERATED VALVE INITIATIVE (CONTINUED)
OTHER ACTIVITIES I o NRC-AE0D REPRESENTATIVE ACCOMPANYING INP0 TEAM o REVIEW 0F MOV RELATED REPORTS BY MOV EXPERT AT INP0 SAMPLE ISSUES IDENTIFIED AT SOME PLANTS o PREVENTIVE MAINTENANCE o POST-MAINTENANCE TESTING o MAINTENANCE, REPAIR, AND TROUBLESHOOTING PROCEDURES o ROOT CAUSE ANALYSIS T RQUE SWITCH SETPOINT CONTROLS O
l l
i O
O MOTOR-0PERATED VALVE INITIATIVE (CONTINUED) l FUTURE PLANS o CONTINUED REVIEW 0F M0V :
PROGRAMS AT PLANTS EVALUATIONS MART VISITS o MAINTENANCE MANAGERS WORKSHOP (OCTOBER 1988) h I
i o !
i
l I
() INP0 CHECK VALVE ACTIVITIES REPORTS TO INDUSTRY o 3 SIGNIFICANT OPERATING EXPERIENCE REPORTS 82-03 APRIL 1982 84-03 APRIL 1983 86-03 OCTOBER 1986 o 24 SIGNIFICANT EVENT REPORTS o 6 OPERATIONS & MAINTENANCE REMINDERS O LETTERS TO INDUSTRY o MARCH 1987 -
EMPHASIZE SOER 86-3 RECOMMENDATIONS o JUNE 1987 - SURVEY OF SOER 86-3 PROGRESS l
o DECEMBER 1987 - USE OF EPRI CHECK !
VALVE GUIDELINE FOR SOER 86-3 ACTIVITIES 1 O
() INP0 CHECK VALVE ACTIVITIES I
(CONTINUED)
EVALUATION ACTIVITIES o REVIEW SOER 86-03 RECOMMENDATIONS TECHNICAL SUPPORT OPERATING EXPERIENCE o EVENT REVIEW TEAMS SAN ON0FRE UNIT 1 (DECEMBER 1985)
V0GTLE VALVE FAILURE
() (JUNE 1987) i l
l
l'..
- l
() INP0 CHECK VALVE ACTIVITIES (CONTINUED)
INDUSTRY INTERACTIONS o CHECK VALVE WORKSHOP (0CTOBER 1986) o LIAIS0N WITH OWNER'S GROUP TASK FORCE FOR CHECK VALVES o REVIEWED EPRI CHECK VALVE APPLICATION GUIDE (ISSUED SEPTEMBER 1987) o CHECK VALVE REPORTS REVIEWED BY DESIGNATED INDIVIDUAL o PARTICIPATION IN EPRI CHECK VALVE WORKSHOPS ACTIVE INP0 REVIEWS o PROVIDE ENHANCED LOOK AT CHECK VALVES DURING NEXT 10 EVA.UATIONS o REVIEW RESULTS O
b't 3 NRR STAFF PRESENTATION TO THE O ACRS
SUBJECT:
HYDR 0GEt1 DET0llAT10f1 AND DAMAGE TO SAFETY RELIEF VALVES Ili FOREIGN BOILIfiG WATER REACTOR DATE: March 7, 1988 PRESENTER: Gary Hammer Krysztof Parczewski PRESENTER'S TITLE / BRANCH /DIV:
Gary Hammer, Mechanical Engineer /EMEB/ DEST Krysztof Parczewski, Sr. Chemical Engineer /ECEB/ DEST PRESENTER'S NRC TEL. NO.:
Gary Hammer 492-0919 Krystof Parczewski 492-3252 SUBCOMMITTEE: Reliability Assurance (Valves)
O
O ps HYDROGEfl DET0 flat 10fl APD DAMAGE TO SAFETY RELIEF VALVE If! FOREIGN BOILillG WATEP REACTOR MAY 6,1987 EVENT AT GUf1DREMMif1GEtl PLAf1T, UtilT C ( A BVR) Ill FRG
- HYDR 0 GEN DET0 flat 10fl If1 SIDE THE MAlfl STEAM SAFETY RELIEF VALVES (SRVs).
HYDROGEfl DET0 flat 10fl SEVERELY DAMAGED SRV IllTERNALS
- CAUSED 3 SRVs TO STICK OPEfl.
RADIOLYTIC HYDROGEfi Af!D OXYGEil C0f1CENTRATE OVER FEW WEEKS If' HIGil, SUBC00 LED, STAGNAf!T P0CKETS, DET0f!ATION CAUSED BY FAST COMPRESS 10ll 0F GAS If1 SIDE PILOT O Lit'ES AflD INSIDE SRV WHEN PILOT VALVE OPEf1S TO ACTUATE SRV, REVIEW OF US SRV DES!GNS IflDICATES GASES CAf! COLLECT, HOWEVER, ACTUAT10ft OF VALVES CAUSES A DECOMPRESS 10fl 0F GASES, THEREFORE, f10 EXPLOS10fl.
f1GT A PROBLEM Ifl PWRs BECAUSE EXCESS HYDROGEN CAUSES REC 0MBillATION OF HYDR 0GEf1 AND OXYGEfl iflTO WATER, O
.--- --- , -_ ---_ _-- _ - _ . . . - - . . .J
I t O -
\k 'x
\s ll i l
!!x 3,1 3
\ \ 'N N
)
N h
?, 4
?)
-\ s -
h '
s N N j \
g ;
\ N." '
.)
N "
x :
- \ N \
'}
INLET s N s N -
g N N N s N \
N \ N \
\ s A .
N
\ N \
x N \
\ \
\ \ htoT Nxx- VAtvs N
N u
Pl5 CHARGE O
SKETCH OF FOREIGN SRV
l O
DIAPHRAGM TYPE PNEUMATIC ACTUATOR a Oa
<> j,
~4 ]
< Jb gh, 3 %
.} O O
O " "
O ..
~'
O
?";
O O O -
O E O 9
Es;EJ
- i f E &
O \ '
DISCHARGE q "_ _ _ _ _ _ _- "ZT' ____; __~7
\
j j 1I r-sJ ll L * * * *,
- Ii
{- MAIN PISTON j
ri ,
g I O
I I
I BODY / i I
I l
Ib l y .
I A
INLET O Target Rock Two Stage Pilot Actuated, Safety / Relief Valve
1 j
1
)
(G ABUTMENT -* +-
GAP RMMM BELLOWS ACTUATOR BONNET
![ '
! BELLOWS h \ LEAKAGE N )'
\ L T w.
~J
' y ALARM PORT PRELOAD
[L N\ SPACER
///////W/
PISTO SECOND STAGE m N y
^
'Nh& mh_ - 6' p
f PILOT PRE LOAD AND SET POINT ADJUSTMENT l
PRELOAD SPRING -_.
,0 SPRING SECOND STAGE m [
DISC (CLOSED) b? \ PILOT STEM
'rlJY / / // /
O MAIN VALVE \ \l !
h PRELOAD SPRING '
"N l / YOKE PORTION
}
N
%: x l' OF PILOT VALVE s . -
DISC
$ p k N'_. . Il e PILOT VALVE I M,/
DISC (CLOSED)
P Of OR FICE # \
MAIN VALVE k
\ S$N RiEM Nw DISC (CLOSED) m .
\
\ '
N c INLET i
N k
q kv N. .
nn"'"""'
OUTLET (N y U
Target Rock 3 - Stage Safety / Relief Valve
O O O s
SOLEN 0ID ASSEMBLY M AIN V ALVE DISC CHAMBER 8 I ~
l Outl g
\ '
h --
SOLEN 0ID PLUNGER H[ AD m :x w.
O RATING W R SP NC ;
k CHAMBER C o yG ql '
bM -
PILOT VALVE SPRING OtsC Cul0E 1 ,x - (N' ::4 -< -
K \_ <'
[ ('/ \ \<
\ )l0 h^
-# 'l h
' . -; p h A U .
]
' =
M ' ._ - P LOT VALVE DISC N CHAMBLR E
'e r 1 CHAMBERD 12$$ " 'N L : 2[/ 4 inlet \
CHAMBER A l RELIEF VALVE PILOT VALNE ssa 6
DIUTSER ELEUITOWFIC POWER ACTUATED RELIEF VALVE
. . Le C
l Wa':ts Bar Nuclear Plant Valve Acu':uator Parts In':erchangeabi:.ity 0
l O
o Presentation I. Background ll. Investigation Ill. Findings o IV. Status o
g I. Background Actuator Lubricant Separation ,
Mixing
! Resolution ,
o l Disassemble l Relubricate Identified Part Discrepancies
- l O
- - - - - , - - - - - - - - - _ . - - - n----n-------------,-e--r---'
o ll. Investigative Process A "$ H SUBCOMPONFJTS e J 1'
IN3 AL 9"liiiPJP" ,
CONFIGURATION l 1,
h!N $
e >
YES YE3 r , r ,
r ,7, 3 ,L, ,, l
'S"8 $fEtx55' o *!$c!uSE '
ti m ,E e > e _i e > :
1
( 3 l l
%!53rW' .
e >
TEST AChABLE DE3 NGE i
( J '
YES PLACE E
l e >
I O
Ill. Findings o
Potential Sources for l Parts Discrepancies Actuator Vendor Valve Vendor Valve Maintenance I o !
Valve Actuator Switching l l
O l
i l
o IV. Status l
- 1. Controlled Actuator Configuration
- 2. Functional o .
- 3. Enhanced Maintenance
- 4. Generic Applicability 1
o
..