ML20248H127

Transcript of 890926 Hearing in Rockville,Md Re Generic Ltr 89-10 Workshop.Pp 1-211

ML20248H127
Person / Time
Issue date:	09/26/1989
From:	Office of Nuclear Reactor Regulation
To:
Shared Package
ML20248H118	List: ML20248H116 ML20248H120 ML20248H123 ML20248H127
References
GL-89-10, NUDOCS 8910110182
Download: ML20248H127 (213)

Text

. - _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ _ . -

j UNITED STATES NUCIJtAR REGULATORY COMMISSION q l

l i

a ma me = mm aa mm am as am as sm am an um am am ma m me sm an am am un am am as sm am me m am mm an sn am an um am m am am am ma m am ma am m am an aa mm e sm est s l In the Matter of: )

)

Generic Letter 89-10 Workshop )

1 )

Pages: 1 through 211  ;

. Place: Rockville, Maryland Date: September 26, 1989

===========================================

HERITAGE REPORTING CORPORATION OJiaist Reportars 1229 L Street, N.W., Suke 446 Washington, D.C. 20005 (202) 628-4888 9910110182 891005 mee eme ~ee% ,l ,

1 UNITED STATES NUCLEAR REGULATORY COMMISSION

)

In the Matter of: )

)

Generic Letter 89-10 Workshop )

)

Tuesday, September 26, 1989 Holiday Inn Crown Plaza 1750 Rockville Pike Rockville, Maryland The above-entitled matter came on for hearing, pursuant to notice, at 10:00 a.m.

APPEARANCES:

OWEN ROTHBERG, P.E.

Office of Research Engineering Issues Branch United States Nuclear Regulatory Commission TAD MARSH, Branch Chief THOMAS G. SCARBROUGH THIERRY ROSS TED SULLIVAN Mechanical Engineering Branch Office of Nuclear Reactor Regulation United States Nuclear Regulatory Commission JEFFREY B. JACOBSON, Inspector Division of Reactor Inspection & Safeguards United States Nuclear Regulatory Commission

. RICHARD J. KIESSEL, Senior Reactor Engineer l Generic Communications Branch )

United States Nuclear Regulatory Commission  !

I EARL J. BROWN, Ph.D. )

i Mechanical Engineer l

Office for AEOD 4 United States Nuclear Regulatory Commission I j

i Heritage Reporting Corporation I (202) 628-4888 1 -

i i

L________.._____.._ ._

2 1 PROCEEDINGS 2 MR. MARSH: Good morning. My name is Tad Marsh.

L 3 I'm the Branch Chief for.the Mechanical Engineering Branch 4 in the Office of Nuclear ~ Reactor Regulation. I'd like to l 5 welcome you here this morning to discuss Generic Letter 89-6 10.

7 I want to take just a minute to introduce the 8 panel of experts we have here. Immediately to my left is 9 Thierry Ross who is the lead project manager for this 10 generic letter. Ted Sullivan is the section leader 11 responsible for in-service testing and for motor-operated 12 valves. Tom Scarbrough is the lead engineer responsible for 13 the generic letter and for motor-operated valves in our 14 branch. Jeff Jacobson is a mechanical engineer with the 15 Division of Reactor Inspection and Safeguards responsible 16 for motor-operated valves. Dick Kiessel, also in the Office 17 of Nuclear Reactor Regulation is in the branch responsible 18 for the evaluation of operating events. Owen Rothberg, to 19 his left, is in the Office of Research. He's responsible 20 for motor-operated valves in the Engineering Issues Branch.

21 And Mr. Earl Brown, from the Office of the Evaluation and 22 Analysis of Operational Data.

23 As you can see, we have quite a number of people 24 here today, all of which have been involved in the 25 development of this generic letter. We also have many Heritage Reporting Corporation (202) 628-4888 i

3 1 people in the audience from the NRC, from various offices, 2 who are here to answer questions which may come up which may 3 pertain to the letter..

4 We're here to describe for you, first, the 5 generic letter the technical basis, the content, and what 6 we believe to be'its implementation. But more importantly, 7 we're here to answer questions from you about the generic 8 letter and about its implementation, and to hear 9 difficulties that you may have in terms of its being put 10 into' place at your plants.

11 So the main purpose of this meeting then is not 12 so much for us to give you presentations, but to hear from 13 you and to enter into a dialogue with you about the generic 14 letter so that we understand your difficulties, and we can 15 explain in an interactive way what the generic letter is all 16 about.

17 I have just a few administrative details before 18 we get really started. First, please sign, I believe the 19 sign-up sheets are on the outside of the meeting room.

20 Please sign those attendance sheets as time permits today if 21 you haven't had a chance to. Those are important. We'd 22 like to know from what organization you come, and also your 23 address if you're not a licensee. If you are a licensee we 24 know your address and you don't have to add your address to 25 that. It's important because we want to send to you the Heritage Reporting Corporation (202) 628-4888

1 4 1- summary of this meeting, and we need to make sure that we 2 can get back to you if questions need to be asked or 3 answered later.

4 We have a set of short presentations this 5 morning. Myself giving a brief overview, Tom Scarbrough 6 will be talking about the generic letter content paragraph 7 by paragraph, how it compares to 85-03 Supplement I, what-8 the basic intent is. Then we're going to go right to the 9 questions and answers.

10 We have received from mostly folka in the 11 audience about 21 sets of questions, a lot of questions, on 12 the order of 150 or so. It's been the experience in the 13 last meetings that we were able to get through them all, but 14 we haven't had this many. So I'm not sure we're going to be 15 able to get all through them today, but we're going to give 16 it our best shot. There's a lot of overlap in these 1

l 17 questions, too. So if we don't get to your particular 18 question we think we're going to answer it in some other 19 way.

20 If we do not get to your question, we will answer 21 it in the meeting minutes in either direct form or by 22 addressing it as part of another questions. So if you asked 23 a question, we will answer it.

24 If there are questions that come up in the course 25 of an answer to another question or as part of this Heritage Reporting Corporation (202) 628-4888 l

5 1 presentation or another presentation, and you'd like to ask 2 it, please write it on a card with your name and your.

3 company and the question, and to whom you'd like the 4 question addressed if you'd like it addressed to someone in 5 particular. If not, we'll answer it as a panel. Try to do 6 that. I know sometimes questions come up pretty quickly.

7 We are taking transcripts of the meeting, but do try to 8 write it down in a card frame. That helps the transcribers, 9 it helps us focus on the question too.

10 As I say, we will be preparing a meeting summary.

11 The summary will have the results from all three workshops,.

12 not from this individual one, so you'll have that available 13 to you. You'll also have transcripts of this meeting as 14 well as the Chicago meeting and the Denver meeting to help 15 you in understanding the questions and the answers as we've 16 addressed them.

17 The schedule for today is pretty loose.

18 Basically we're going to take a lunch break at some time and 19 at some time in the afternoon take an afternoon break, 20 depending on how we're doing, whether we need a break, and 21 where we are. It's been our experience also, that we need a 1

22 break at some point in time. We get talking a lot and we 23 have to take a creak.

24 Let me turn now to some motor-operated valve 25 problems, some overview types of issues that led us to first

, Heritage Reporting Corporation (202) 628-4888 i

fi

1. -

b . 6 0

~

l' '85-03Jand1then to Supplement I, and.then to'this particular.

2 Lgeneric letter.

-3 .ILdon't think motor-operated valve problems arei 4 .new to'you,.as they aren't new to us. .

We have seen a

-5 ' variety of problems, and those problems have been related to 6' a number of various aspects-of motor-operated valve design, 7_ operation,.and. testing. ,

8 . We've seen difficulties with' torque,. torque 9- bypass,;and limit switch settings. We've seen damage due to.

10' wear,-operational as-well as thermal-overload difficulties.

If 11 We've also seen problems due to motor-operated valve design I'

12 and sizing.

13 The causes are probably'as varied as the problems 14 themselves. We've seen difficulties with a lack of

15. coordination between various organizations within the plant.

16- How is this particular valve really to be used as opposed to 17 s stylized safety analysis? How should it be purchased?

18 How should it be tested? How should.it be installed?

19' Communication paths within the organization are extremely 20 important and sometimes we find that hasn't taken place.

21- Again, we have seen in some places inadequate 22 engineering and design of a particular valve, the operator, 23 its associated equipment, as well as the body of the valve. 1 24 In cases we've seen weak maintenance practices,  !

25 training of the maintenance people, and the procedures that Heritage Reporting Corporation (202) 628-4888 i

1

7 1 folks are using to do maintenance, testing, and setup.of 2 valves.

3 And we have seen, and I think.it's fairly well 4 understood, the testing criteria itself used by the Section 5- 11 doesn't go far enough in terms-of demonstrating the 6 operability of motor-operated valves.

7 I don't need to describe to you what the 8 seriousness of MOV problems are. You've seen them. You've 9' seen loss of feedwater events, we've seen steam generator 10 over' fill events, we've seen a variety of operational 11 occurrences that cause us concern, and you concern, about l 12 the safety of systems as well as components when the 13 operability of these particular valves are at issue.

1 l

14 Generic Letter 89-10 then, in a nutshell, and l

l 15 you're going to get it in much more detail from Tom 16 Scarbrough, 17 89-10 basically extends the scope of 85-03 and Supplement I 18 to include all safety-related motor-operated valves in the 19 plant as well as those position changeable valves that 20 impact the safety function of the system.

21 It's our basic belief that it will help to I

22 provide confidence in the motor-operated valves, and the j 23 safety systems of which they're associated will be able to 24 do their function.

l 2b I'd like to take a minute now to talk about some Heritage Reporting Corporation (202) 628-4888 l

l l l

maae__-____-___-------- - - - - - -

n' .

8 j j

c

.1 NRC' activities. -This slide is meant to show you what 1 i L 2 v6rious things are going on within the NRC. It's by no 3 meane a comp'iete list, but it's meant to show you the 1

~4 various things that are going on within the NRC that touch ~

5 upon motor-operated valve design, operation, testing, L 6L qualification. It's meant to show you in an overview sense 7 the various offices that are involved, and it's meant to 8 show you the breadth of concern that we have.

9 I'll start with talking about some of the NRR 10 act.ivities . First, there in Bulletin 85-03, followed by 11 this Generic Letter 89-10, there was also Bulletin 85-03

12. Supplement I.

13 Generic Letter 89-04, another very important 14 generic letter this year, dealt with in-service testing 15 criteria. It had in it 11 technical positions, some of 16 which were associated with motor-operated valves. So we 17 have promulgated some guidance in terms of in-service 18 testing programs already with respect to motor-operated 19 valves.

20 We have another ongoing program, at this stage L l

21 it's an information gathering program, that focuses our <

22 concern on pressure isolation valves, specifically inner-23 system LOCA. We are cbncerned that the design, operation, 24 and testing of reactor coolant system pressure isolation 25 valves may be at fault. j

)

Heritage Reporting Corporation I (202) 628-4888 I

J

_ _ _ _--___ -__ ._ 2 1

9 1 There have been some precursors within this 2- country and in other countries. As a result of that concern 3- we are performing pilot inspections at a few plants around 4 the country to determine the design adequacy, the testing 5 adequacy, whether the operators can mitigate an inner-system 6 LOCA, to determine basically where are we from a safety 7 perspective in terms of pressure isolation valves.

8 In terms of the Office of Research, Generic Issue 9 87 is one that led almost directly to this particular 10 generic letter and to Bulletin 85-03. It deals with the 11 high pre _ssure coolant injection system steam line isolation 12 following a break. Tests have been performed at EG&G, I'm 13 sure you're aware of those tests. If you're not aware of 14 those tests, please become aware of those tests. Those are 15 important tests that demonstrate some problems with motor-16 operated valve switch settings, friction factors, etcetera.

17 There are transcripts of a public meeting that was held on i

18 February 1st here at the Crown Plaza where we had a panel of 19 experts composed of valve designers, diagnostic testing 20 people, research people, regulatory people, to discuss in an 21 open way with a lot of interaction from the audience these l

22 particular tests and what they show.

23 There is a general valve performance program 24 which is performing a limited evaluation of the ability of 25 various indicators to predict valve conditions. There is an Heritage Reporting Corporation (202) 628-4888 i

L_i_____ _ . . _ _ _ i

10 1 ongoing large aging program, I'm sure you're aware of that, 2 to determine failure modes and testing techniques 3 themselves.

4. 'AEOD is, as always, involved in the evaluation of 5 ' operational data, performing engineering evaluations to see 6 whether there is a generic problem with motor-operated 7 valves and other associated issues. There have been not 8 just this SECY paper but other SECY papers that have 9 addressed motor-operated valves.

10 Again,-I want you to take from the slide, please, 11 the sense that the NRC is looking at motor-operated valves 12 from a variety of different perspectives. They are at 13 various points, but it does, I think, demonstrate the 14 breadth of concern and the breadth of the issue that we have 15 about motor-operated valves.

16 I'd like to turn now to some activities that are 17 taking place in the industry to address motor-operated 18 valves. EPRI has published technical repair guidelines for 19 limit torque SMB000 that came out earlier this year. Other 20 repair guidelines are in preparation. SMB00 and SMB0 are 21 being worked on at this point. There are application  !

22 guidelines which are being worked on.

23 IMPO has an~ active program to visit plants to 24 determine what motor-operated valve programs currently exist 25 to ensure valve operability.

i

(

Heritage Reporting Corporation (202) 628-4888 l

)

i u_ _ _ _ _ _ l

, q .

2 11 1 October 4, 1988 there was a document published .

2 that contains key elements of a comprehensive MOV program. jl 3 In our opinion, it forms a framework for an excellent motor-4 operated valve maintenance, testing, diagnostic programs at 5 the plant. If you're not aware of that, again, this is 6 another document that you should become familiar with. It 7 does tell you in a framework sense, not with any great deal, 8 but in a framework sense of how to construct programs and 9 practices at the plant.

10 NUMARC has been involved extensively in this 11 particular Generic Letter 89-10. They have been in touch 12 with us throughout the deliberations for this generic 13 letter. As a result of these meetings we've had numerous 14 discussions. They are very involved.

15 Vendors associated with 16 diagnostic testing are working on signature tracing 17 techniques, ensuring that the tests at EG&G are properly 18 factored into diagnostic testing techniques so if there are 1.9 any changes to those diagnostic tools they properly reflect 20 - the problems that are being experienced at EG&G. And there 21 are motor-operated valve design improvements and there is 22 IEEE maintenance practices which are taking place.

23 The code, for its part, is working on various 24 standards that relate to motor operated valves. They are at 25 various stages of approval too.

Heritage Reporting Corporation (202) 628-4888 E _ _ __ _ __.--_____ _--_ . _ _ ___-------_----_-----_------_--_-_--__----__ ------____ _ - - - - - - _ - - - - - - - - - _ _ - - - -----------------J

s 1

l- ^

12 l

1- OM10 is an approved document that discusses valve 2 testing in general, all valve testing. OM8 is particular to l

l 3 the motor-operated valves and the current versions are even 4 more particular to the operators themselves as opposed to 5 the valves themselves. QME, as I stated, relates to the l

6 mechanical component design.

7 From an overview sense the industry has a number l

8 of activities ongoing on motor-operated valves. What we 9 don't see is coordination among the various groups, and we 10 don't see that the testing programs that are taking place, 11 the plant visits, are being orchestrated in the sense that 12 they can be mutually beneficial, and in my mind there is a 13 need for that.

14 That ends my short presentation on an overview of 15 motor-operated valves. I'd be glad to take any questions 11 16 you like at this point. The next presenter is Tom 17 Scarbrough who is going to go through the generic letter in 18 some detail and we would be glad to take questions any time 19 during that presentation, and then we'll follow by the 20 questions and answers themselves. Thank you.

21 MR. SCARBROUGH Good morning. I'm going to talk 22 briefly about the background of the generic letter and why 23 we're here, the contents of the generic letter, and 24 basically where we're going beyond Generic Letter 89-10.

25 In terms of the background surrounding the Heritage Reporting Corporation (202) 628-4888 1 >

l

I. '

U L

{ -

13 l 1 generic letter, this concern started, to a large extent, l

L 2 following the Davis-Besse event June 9, 1985. There a loss ll l 3 of feedwater occurred. It began where the reactor was 1

4 operating at a high power. The main feedwater pumps were l 5 lost, and the auxiliary feedwater system actuated as it 6 normally should.

7 The operator initiated the steam and feedwater l

8 rupture contro1' system on a low steam pressure rather than l

1 low steam generator level.

9' This resulted in the steam 10 generator isolation valves closing automatically, as they're 11 supposed to in that situation. The operator quickly 12 realized his mistake and went to reset the system, but the 13 valves could not be opened electrically from the control 14 panel. The differential pressure that was built up in that 15 situation was such that the torque bypass switch did not 16 stay engaged long enough for the valve to unseat.

17 The operators were sent down and within 15 18 minutes or so had manually aligned the auxiliary feedwater 19 system and reinitiated flow to the steam generators. But 20 this did indicate that there was a significant concern over 21 the ability to operate motor-operated valves under design 22 basis conditions.

23 This resulted, in part, in Bulletin 85-03 in 24 November 1983. That dealt mostly with the high pressure 25 injection type of systems, emergent feedwater, and reactor Heritage Reporting Corporation (202) 628-4888 i

L-__-_-------_.

14 1 core isolation cooling for BWRs. That bulletin requested 2 that licensees develop a program to perform' design basis 3 reviews to ensure that switch settings on MOVs and those 4- selected systems were selected, set, and maintained in the 5 proper position. That program was given two years to run, 6 and it was more or less a one shot deal.

7 In April of 1988 a supplement to Bulletin 85-03 8 was prepared to clarify two aspects. First, that all 9 safety-related motor-operated valves in those systems should 10 be included in the program; and another was to ensure that a 11 closed MOV could be reopened in the event that it had gotten 12 to that position by some sort of inadvertent operation.

13 Once the results of the programs to implement 14 Bulletin 85-03 began arriving within NRC, it was revealed 15 that many more MOV's than had been expected would not have 16 been able to operate under design basis conditions. This 17 number was approximately eight percent which was much higher 18 than the PRA's had been assuming at the time. Further, 19 other deficiencies were revealed as a part of the Bulletin 20 85-03 programs.

21 As a result, the NRC contracted with Brookhaven 22 to perform a value impact analysis to determine whether it 23 was cost effective to extend the scope of Bulletin 85-03 to 24 include all safety-related MOV's. This included a TRA and 25 had some very interesting results.

l 1

Heritage Reporting Corporation  !

(202) 628-4888 I

i

\- -------_- -- - -b

l 15 1 It found that there was a favorable dose l' 2 reduction in dollar savings from expansion of this program.

l 3 Brookhaven also had some very strong qualitative arguments 4 for the need to expand 85-03.

5 About the same time, even more recently, the 6- results of some full scale blow-down tests performed as part 7 of Generic Issue 87 by Idaho National- Engineering 8 Laboratory, found, among other things, that there was much 9 more thrust required to operate valves under design basis 10 cond'itions than had been assumed. One of the conclusions of 11 that study was that the valve thrust equation needed to be 12 reviewed entirely. Further, sub-cooling and valve design 13 were found to have profound effects on the disk valve 14- friction factor, the disk factor. Further, the stem factor 15 was found to change with load.

16 The results of this initial phase of the Generic 17 Issue 87 testing can be found in an EG&G report that was 18 very recently issued, and it's SSR-85-47 dated July 1989.

19 It should be available in the public document room.

20 Even more recently there have been some plant 21 events which have raised concern for the operability of 22 motor-operated valves. There's an Information Notice 89-61 23 dated August 30, 1989 which you may find very interesting.

24 Motor-operated valves in auxiliary feedwater lines could not 25 fully close under a high differential pressure. Some Heritage Reporting Corporation (202) 628-4888

^

16 1 evaluations by the licensee found that the valve friction 2 factors on some very similar valves were twice as high as 3 what the manufacturer had reported.

4 Even more recently there was an event in February 5 of this year where a safety injection line MOV was not able 6 to shut under a full line flow. There are some other cases, 7 too. These are just a couple of examples. But not only are 8 we seeing this problem in the laboratory, but we're also 9 seeing it in the plants.

10 Let me talk now about the scope of the generic 11 letter and how it differs from 85-03. One of the major 12 changes from 85-03 is that the number of MOV's that are 13 included in the program has increased. 85-03 addressed 25 14 or 30 motor-operated valves. Generic Letter 89-10 will 15 address 100 to 150.

16 The other change relates to some changeable MOV's 17 in the sense that whereas the concept of changeable MOV's as 18 in 85-03, now you have more systems, more positional 19 changeable valves to deal with.

20 But the reasons and the intent of position 21 changeable has not changed from 85-03. That is still there.

22 That's the same as it was in 85-03.

23 This is a short summary of the positions in 89-24 10. They're kind of summarized in this fashion to boil it 25 down to just a meat of each of the positions. You'll notice Heritage Reporting Corporation )'

(202) 628-4888

L l i .

17 I

.)

1 that the first four or so are almost identical out of 4

2 Bulletin 85-03, and they are. They're almost word for word. {

l 3 The first item is review and document the design 4 basis for MOV operation. The difference between 85-03 and 5 89-10 is now in the later portion of the generic letter the 6 concern degraded voltage is spelled out specifically.

7 The second bullet there is establish and l

8 implement program to review and revise methods for selecting 9 and setting switches. Same as 85-03.

10 The third is demonstrate MOV operable by testing 11 a design basis differential pressure aim or flow. Once 12 again, the same as 85-03.

13 And document where design basis testing cannot be 14 performed with a description of the alternate testing.

15 Those are all right out of 85-03.

16 The next item there is prepare procedures to 17 ensure that correct switch settings are determined and 18 maintained. One aspect of 89-10 is that the Section 11 19 stroke time testing is not fully sufficient to ensure that 20 the switch settings are correct. That's one of the emphasis 21 made in 89-10.

22 The next item, analyze and document MOV failures 23 and corrective action. That's new. That's different than 24 85-03. This starts to set up a continuing program which was 25 not proposed in Bulletin 85-03.

Heritage Reporting Corporation (202) 628-4888

-p

,m 18 1 -The next item thereiis periodically examine'MOV

~2' ' data every.two years for' trends. There's a trending program 3 that's"new to 89-10.

'4 So'.those'are the basici fundamental positions.of-

~

I 5 89-10. _ Most of: them you should be quite familiar with.-

'6 The schedule'is outlined ~in Generic Letter 89-10,-

_7 .but I thought it would be good to just briefly. summarize it.

1 l 8 I believe this is-much more clear. Program description and 9 schedule _for operating licensed plants, within one year or 10 one r"efueling_ outage, whichever'is later. Construction 11- permit plants, within'one year _or before the OL'is issued,.

12 whichever'is later.- Accomplishment of.the initial: program

~'

13 for OL's within five years or three reviewing outages, 14 whichever is later. .And for CP plants, withinLfive years or 15 before'the OL is issued, whichever is later.

16 Then a continuing verification of switch settings-17 every five_ years,-or three refueling outages, whichever is 18 later.

19 The reporting requirements in the' generic letter 20 are advise and receipt within six months that the generic

^

21 letter will be met or any modifications to the schedule in 22 the generic letter. Notify the NRC of any changes to

23. commitments made in th'ta six month letter. And notify the .

, 24 NRC within 30 days after initial five year program is 25 completed.

1 Heritage Reporting Corporation (202) 628-4888 .

1 l

19 1 .There will be a variety.of actions taken by~the--

2' -NRC' staff to support' implementation of the generic ~ letter.

3- LThe workshops.begins we've held one'in Chicago and one in 4' Denver and this isLour~ third.. Technical guidance provided 5 to.the'NRC staff. We're working'on-a temporary instruction 6 Lto. provide to the jnspectors both in headquarters and the 7 regions to' help'in reviewing the implementation of the 8 . generic letter. Audits of programs, program descriptions.

9' Wenplan to perform some audits of program descriptions next 10 year. . Audits and inspection of program. implementation..

L 11 Those.will start'once the. programs are underway. And review 12 of industry testing. We've been encouraging the industry to-13 work together as a group to develop some testing schemes to apply to data to individual plants, and where we can be of 15 ' help in reviewing that testing or assistance in that area, 16 we plan to review that testing data.

17 The generic letter goes a long way in resolving

'18 may of the concerns for the operability of motor-operated

19. valves, but it'doesn't go all the way and resolve all the .

20 concerns. There's a rule change that-is being contemplated i 21 at the time which will address 10 CFR 50.55A in-service  !

22, testing of valves and pumps. That will help try to clarify 23 .the need to. prepare programs that demonstrate ~the 24 operability of motor-operated valves. ]

25- The next is the improvement of in-service testing ,

Heritage Reporting Corporation (202) 628-4888 u

1 l

l

20 I criteria. The stroke time testing in Section 11 of the code 2 does not tell you very much about the condition of the 3 valve. That's pretty much understood. We're going to try 4 to work with ASME to develop better criteria on an 5 appropriate frequency that will tell us much more about the 6 health of that valve.

7 Regulatory guides. We hope to develop some 8 regulatory guides on acceptable in-service testing methods 9 and maintenance, training, other areas which we feel could 10 be o'f benefit to the industry.

11 Design and qualification of motor-operated 12 valves. In this area we've been encouraging EPRI through 13 contacts with NUMARC to work in this area, to perform 14 research, to evaluate the valve thrust equation, to learn 15 more about it so that we can properly design and qualify the 16 MOV's so that they're initially set up to ensure that 17 they'll operate under design basis conditions.

18 The last one there is thermal overload 19 protection. That's been a concern for some time. There is a 20 new IEEE standard that is in the draft stages right now l

21 which may be very helpful in this area. It's number is P-22 741. We're looking at that and hopefully that will be 23 useful in resolving the concern for thermal overload 24 protection.

25 That concludes my talk this morning., We're going Heritage Reporting Corporation (202) 628-4888 w __- . _ _ _ _ _ _ _ _

21 1 to go on and talk about some other things from up at the 2 table. If there are any questions I can answer right now, 3 I'd be happy to.

4 Seeing none, we'll begin our talk from the table.

5 MR. MARSH: Thanks, Tom. As you know, this is 6 the third workshop. We've had two others. In going through 7 the questions that we've received at those other workshops 8 as well as questions that have been communicated to us here, 4 9 many of them are repeats, many of them are very significant 10 questions. We wanted to prepare for you some responses 11 which we think will be of mutual interest to you all, even 12 though you may not have asked these questions. These 13 concerns have been expressed at various times, various 14 levels within the NRC.

15 (Turning up volume so those in the back can hear) 16 Let me say it again. This is the third road show 17 of the three. We've gotten a lot of repetitive questions 18 that have been significant and have been far reaching, and 19 it showed sort of a general need for interpretation, a need 20 for clarification in what I call eight key areas. We want 21 to give you the answers to those questions now.

1 .

22 One thing we need to express to you is that this 23 generic letter and all of its processes did go through our 24 backfit process dictated by 50.109. It was specifically 25 reviewed and approved as such. The answers that we're going Heritage Reporting Corporation (202) 628-4888 i '

E_-_____--____ _ _ -

e 22 1 to give you today we believe fall within the intent of the 2 generic letter, the scope of it, as well as the process that l-3 we went through. But we do recognize that there is 4 interpretation, that the words aren't crystal clear. But we 5 do believe, nonetheless, that the interpretations we're 6 giving you are consistent with what we said.

7 Tom is going to go through these eight positions.

8 I want to encourage you to dialogue with us on these issues.

9 Please keep in mind also that we have 20 sets of questions 10 that'we're going to go through for the rest of the day and 11 some of these are going to overlap into those 20 questions.

12 If you need the dialogue now, by all means let's do it, but 13 at some point I think we're going to get into this in more 14 detail in just a few minutes.

15 MR. SCARBROUGH: There are eight questions that 16 keep coming up and up. What I'll try to do is paraphrase 17 the area that is of concern, and then describe the consensus 18 of opinion of how to interpret the generic letter.

19 The first one was the prevention of inadvertent 20 operation of position changeable motor-operated valves.

21 The language in the generic letter gives the 22 means by which inadvertent operation of position changeable 23 MOV's can be prevented'. The intent of the language is that 24 an MOV that does not have a safety function may be removed 25 from the generic letter program if all forms of inadvertent Heritage Reporting Corporation (202) 628-4888

n-

' ? v.

{ '

23 l' operation'of-MOV are prevented.

2: It is;the licensee's responsibility ~to develop a 3 method to prevent those forms of inadvertent operation. . We-

~

4 provided several: examples of methods- that we consider' 5 acceptable to' accomplish that objective. One such. example 1

• 6 was rackingLout. circuit breakers and placing an aluminum

~

t' 7 strip on'the MOV itself. The licensees may use these 8 . methods or develop'their own.

9 That's the first one. Arefthere any questions on.

10' that"?

ill' Okay, we'll go on. If you think of any, write 12 them down'and we'll ask them later.

13 The:second was use of test data from similar-14 motor-operated valves.

15 The generic letter states as a goal that each MOV-16 in the program be tested at des'ign basis conditions. Where 17 this is not practicable for a particular MOV, the generic 18 . letter provides for the testing of another MOV in the plant-19 or as a prototype at design basis conditions. The licensee-20 must justify the use of alternate testing data, regardless 21 of whether the MOV is identica.1 or similar.

22 During the preparation of the generic' letter ~it 23 was assumed that identical valves would perform in an 24 identical manner. Recent plant experience, such as that 25- discussed in Information Notice 89-61 which I mentioned a Heritage Reporting Corporation (202) 628-4888

24 1 few minutes ago, and the research results, such as those as 2 Idaho, have shown that valves assumed to be identical can 3 behave differently. Therefore, the licensee must develop a 4 rationale that' demonstrates that the alternate test data are 5 applicable to the MOV in question.

6 As a follow on, we want to reiterate that we are 7 not requiring programs to perform 100 percent full Delta-P 8 testing of all valves. Where it is possible and 9 practicable, however, full Delta-P tests should be performed 10 as explained in the generic letter.

11 Any questions on that?

12 VOICE: You said that full Delta-P testing may 13 not be required. Only if it's impractical?

14 MR. SCARBROUGH: No, what we're saying is we're 15 not requiring every MOV to be full, 100 percent Delta-P 16 tested, period. Where it's practicable we hope and we 17 expect a full Delta-P test to be performed. But where it is 18 not practicable, there are means described in the generic 19- letter for alternatives to that. That's what we're saying.

- 20 VOICE: So wherever it's practical or 21 practicable, you expect the full Delta-P testing?

22 MR. SCARBROUGH: Correct.

23 MR. MARSH: It is a difficult matter. There are, 24 as the information notice describes, there are valves that 25 are inherently identical. That look the same, that were Heritage Reporting Corporation (202) 628-4888

.w ._

, b f'i' 1 25 l1' purchased to the:same' specification, that had the same w

L J2- pressure rating ~, same operator, same tolerances.- The point 3

3 at which a licensee can determine that these valves are

-4 identical is'different. And to a certain extent,.we don't.

.5' 'know the: answers to these questions as-well, because valves

6. that areLapparentlyEidentical, when tested are found not to 7 be11dentical. -So what we basically look for is full 8 differential pressure testing where possible, where 9' practicable in the plant. When that is not possible, 10 similaritycis an allowed alternative as expressed in the 11' generic. letter. But' demonstrating similarity, showing that 12 that valve-in fact does behave'at some lesser differential-13 pressure at some different conditions, is going to'be a.

14 subject that you're going to have to pursue, you're going to.

15- have to work on. Maybe a program, an industry-wide testing

. 16 effort, there's going to have to be some means of 17 extrapolation.

18 Just by looking at two valves and saying they're 19 the same, same purchase, apparently identical, won't be E20 enough. It won't be enough. There's going to have to be i

21 some type of demonstration by referencing a test analysis, a 4 -

22 -test, some type of prototype testing, something to show that 23 in fact those valves are similar, that they behave the same, 3 I

- 24 that in fact you don't need the full differential pressure 1 25 test on all valves in the plant.

Heritage Reporting Corporation (202) 628-4888

a ,

t  ;

.-n 1

.. l p"- '

1' F 26'

~

1- VOICE: Doesn't that statement invalidate the j u[m

[ l2 prototype testing?- If we can't be sure that.twoLidentical

.' 3 ' valves will act similarly, how are we supposed to identify

~

n ~4- whether they will act similarly or not? And what' 5 justification is there for prototype testing?

I' . 6 MR. JACOBSON: I don't think it invalidates. l

~

7 prototype testing. One'of the ways that we. foresee.maybe

~

8 getting around this problem 11s, for instance, let's say.you-9e had"a . family of valves and you did a prototypo' test on 10, lsevehal of the. valves, not just one, but maybe five or ten 11~ or maybe there's already information out there that five or 12 Lten of these' specific valves have been tested'at various 13 , plants in the country. And for instance, all the data 14 indicates that.for this particular valve the valve factor 5

15: and other parameters that are appropriate are all relatively 16 the same. 1 17 That might be a good basis for'then saying that 18 that data can be applied to the valve in your plant.

19 But let's say you just had one prototype valve 20 and one valve in your plant and there was no other testing- q 21 done on these two valves. You really, even if they're

- 1 22 identical, you really don't have any way of ensuring that -l 23 the parameters on the two are going to be the same. One may l

24 have been aged a little differently, the tolerances might be

25. a little different. There are a lot of things in these Heritage Reporting Corporation (202) 628-4888

. 1

s' 27

+ 1 valves'that:can affect the performance.- A lot of these 2 valves were not madelto very tight specifications, as we're -

3 finding out today. There are' differences.even between.

4 identical valves..

5 So you really have to look at'the. data that's'out 6 there,-and there may need to be some more industry testing 7 in this area to get a better data base so you can make-those 8 kinds of determinations as to when prototype data would be 9 applicableLto:your valve and'when it wouldn't'be.

'10 ' VOICE: How long will we have to develop this 11 data base? It sounds almost like to have a proper prototype-12- established we'd.need tests of a number of identical valves, 13; or what we would consider identical, prior to determining

^14 that that data that *:e have can be attributed to the valve l.

15 that we have in the system. I'm talking strictly about a 16 valve that physically cannot be tested at its maximum delta 17 pressure.

18- MR. JACOBSON: First of all, there already is a-19 wealth of data out there. I'm sure it's not enough data to 20 answer everybody's questions for every valve. The generic 21 letter gives five years as a goal in accomplishing this 22 program. Part of the reason we gave the five year time 23 frame, initially we were looking at a shorter time frame, is 24 that we saw the need that some prototype testing was going 25 to need to be done. <

l Heritage Reporting Corporation (202) 628-4888 1

- - - - - _ _ -___ m

{ t e .

, 28 1

There is1a wealth of data already out there. We 2L expect:that more' data can be generated in'the near future.

3 The generic letter does give a five year time frame as a 4 goal for accomplishing the program..

5' MR. MARSH: One of the eight questions that we're

• 6 going to get to, the last question in fact, or next to the 7 last, deals with schedule. It's been construed by things we 8 have said in the past that'we're extremely rigid.on 9 ' schedule. . We don't want to present'that way. We think that 10 'five years is~enough time, we do. But if that's not true,-

11' if it's just physically not possible tx) get the types of 12- tests done that you need to get done, we want to know that.

13- We:want to hear that.

14- VOICE: The only thing I'm saying is there are 115 literally thousands of different kinds of valves out there, 16 and:the. kind of testing we're looking at would require 17 multiple tests of each type. If an identical valve cannot

-18 be assumed to be the same, how can a similar valve be 19 assumed to be the same?

20 MR. JACOBSON: There is testing, I know, that's 21 already out there that shows even some types of valves, a

22 maybe.not even being identical, do perform the same. They 23 may not have to be idehtical in all cases if there's enough 24 data out there to support the use of a similar valve.

25- Now for identical valves it's a little tougher Heritage Reporting Corporation (202) 628-4888

p, _ _ _ _ _ _ _ _ _ _

s 29 1 .because you're trying to get an identical valve and there 2 may be many different types. We know of some bad actors, 3 some bad valves out there now, the information notice that 4 Tom reference. I know of some other families of valves I 5 guess you could call it, that do not perform as originally f

L 6 assumed in the standard calculations. But there's also a 7 lot of data out there that shows certain types of valves do l 8 perform and do meet the standard calculations. When data 9 supports that type of information, we would look at that.

10 So really, I don't think we see the need for this 11 really vast prototype testing. There's going to need to be 12 'some prototype testing done. I really don't have a good 13 handle on the extent of that problem. I don't know if the 14 industry does either. I think you need to get together and 15 try to start resolving that problem as soon as possible, 16 though.

17 VOICE: Well the thing is, I'm sorry. It just 18 goes with the schedule again. This is the kind of work that 19 should be done before we start our testing program. Because 20 we have to start testing valves now to get our schedule fit  !

21 in. If this is data we need prior to getting our program 22 started, it's not going to do us any good if we get it five 23 years from now.

24 MR. JACOBSON: There are a lot of valves I think 25 you've already got data on. There's a lot of other things Heritage Reporting Corporation (202) 628-4888

30 l' that need to be done in the program such as the design basis 2 review and so forth, that can be started upon immediately in 3 parallel with any prototype testing that needs to be done. i 4 And like Tad said, if it just becomes physically impossible 5 to complete the prototype testing that needs to be done in 6 order to resolve these questions, then we're going to look 7 at possible schedule extensions.

8 So I think we're open on the subject. We really

-9 don't have enough information right now to envelope the

~

10 prob 1em, but our position is that the industry needs'to 11 attack this problem very soon and get working on it so we 12 can get a better handle on it.

13 VOICE: On a typical PWR that has about 200 14 safety MOV's, what percent do you see will have to be 15 tested?

16 MR. JACOBSON: We don't have a specific percent 17 that we foresee have to be tested. We would like and we 18 prefer that a full differential pressure test be done on as 19 many valves as practical because that is the best way of 20 ensuring that the valve is going to perform. Where it's 21 impractical due to various reasons, we're going to have to 22 look at that.

23 So we don't' have in our mind a percent that needs 24 to be done. You may have a plant where 100 of your valves i 25 are identical. In that case, maybe not all 100 of them have Heritage Reporting Corporation l (202) 628-4888 l

l

_m____._.__ ______. .__-

31 1 to be tested. You may have a plant where 200 of them are 2 all different, which may mean you'd want to test a lot more 3 of those valves. It's really on a case by case basis.

4 MR. MARSH: Let me cite one particular PWR, I I 5 won't mention its name, and I can't say whether this is 6 adequate or not. We haven't looked at it to know whether in 7 fact it is adequate. One PWR tested about 20 out of 150 8 valves. Twenty'with full differential pressure testing.

9' That's only a benchmark. I don't know whether that's okay 10 or not, I don't know about extrapolation. But that gives 11 you a feeling for what one plant did.

12 VOICE: Also on 90 degree valves versus rising 13 stem. I just wondered if you had a breakdown on those.

14 MR. JACOBSON: You mean on butterfly valves?

15 VOICE: Right.

16 MR. JACOBSON: On how many should be tested?

17 VOICE: In comparison to testing 100 percent.

18 MR. JACOBSON: Really the same rules and thought 19 process goes into the butterfly valves as does the rising 20 stem valves. You need to verify the same types of things.

21 Depending on how many you've got in your plant, what type 22 they are, whether they're identical or not. That would all 23 come into play as to how many you would need to test.

24 VOICE: To reiterate some of the earlier 25 gentleman's question, you said this data base is going to be Heritage Reporting Corporation i (202) 628-4888 )

1

l \

i l L'

• I i

j 32 l 1- available in the future. I think what his point was trying 2 to get, and what I need to know, there is no data base there j 3 now. So basically what you're telling us is you want us to .

4 full Delta-P test where it's practical and it's possible to 5 do, correct? And to substantiate that you need this data 6 base to do that. I don't think there's anything out there.

7 If there is, who's putting this data base together besides l

8 each individual unit or plant or is it going to be 9 correlated together to be available to us?

10 MR. MARSH: Don't misunderstand. We're not 11 saying, we want to encourage full differential pressure 12 tests. We want to encourage that where you can do it. That l

13 comes across in the generic letter and it comes across here.

I i

14 But when you cannot do that, then there should be 15 an industry-wide effort to form the data that you need to do 16 that. Please don't look to us to develop that data for you.

17 Through our research office we're going to have some data i 18 available to you, but there is data that's available now.

19 It may not be sufficient for you. The timing may be such

! 20 that you may have to differential pressure test valves that l

21 you would prefer not to because you don't have the data base 22 that you need. But we do think that five years is enough 23 time in which to construct the data base.

24 MR. JACOBSON: The industry needs to get the data 25 base going. Your industry groups, there's an MOV user's l

l Heritage Reporting Corporation (202) 628-4888

~

1 33 1 group, there's NUMARC. The industry needs to get together 2 and pool all the data that is taken from the utilities that 3 do do full DP testing. You may do 10 or 20 valves at your 4 plant. Another guy may do 10 or 20 at his plant. All that 5 data is going to be part of the data base.

6 VOICE: But the problem I see with that is that 7 it's like anything else. All this information is out there 8 and it's not being given to the people that are working 9 this. It's on the shelf someplace.

10 MR. JACOBSOM: We understand that. And we're 11 stressing to you that that is the industry's problem. You 12 have this information cut there. You need to come up with a 13 means of getting it together and establishing a data base.

14 VOICE: Do you think there's enough information 15 right now to exempt certain valves on identical, like --

16 MR. JACOBSON: I think so, yes. On certain ones.

17 VOICE: The way I get the feeling, it's construed 18 to me, is that there isn't enough information to 19 substantiate that. That there's more information doubting 20 that than there is --

21 MR. MARSH: You can cite some particular plants i 22 that have done test. The subject of the information notice.

23 s full DP.

They have tested valve', They found some problems, 24 but they tested valves. Do you have any of those valves?

25 You don't know, but the point is, there is data available j l

l Heritage Reporting Corporation (202) 628-4888 i

i i

w__ _ - _- _-_-__ -

34 1 sthat may-be useful to.you. I.can't overemphasize the need-2 for.-coordination amongst the industry group. You've got to 3' do'it. .That's'the way to.make success out of this program 4 is to coordinate'the data base.

5 MR. SCARBROUGH:- Auggie,.why. don't you contact 6 Will Ross at Texas Utilities. He's Chairman of the MOV 7 User's Group, and they're working on.this. area. ' We . spoke tx) 8 him'in' Denver and they're trying to develop'a method to 9- , coordinate a lot of this information. .So:you can be the 10 firs'.-t Why don't you contact him.

11 . VOICE: The other thing, the.other gentleman on-

-12 . quarter turn valves, non-rising stem versus rising ~ stem.

13 Most;of our valves, butterflies if you will, they go out on 14 limit, not on. torque. So that presents a different scope, 15: as far as I'm concerned.

16' There are,.you're right, earlier testsfthat when 17 we were starting the plant up since we were a younger plant 18- we had to do certain full Delta-P tests, but at this time we 19 might not be able to substantiate full Delta-P tests on g 20 certain butterfly valves. I' don't know if that is an  !

4 21 adequate test to prove that the valve works or doesn't work. j 22 MR. JACOBSON: On butterfly valves, although they 23 may not be a torque switch, some utilities do have torque 24 switches no butterfly valves, you need to verify that the 1

25 actuator as it's set up can deliver the required torque to Heritage Reperting Corporation (202) 628-4888 L

I I

p 35L w

l 1: operateithe. valve under the design' basis conditions.

2 '- How you need to do that is first of all, you need' .

~

3 to figure out what that torque is, and you need to validate, h 4 .'iffyou're' going to-use the' equation you need to validate.

5 'that.the' assumptions _you're making are; correct by either a 6- data base or something of that nature.. .Then you would have 7 'to do'calculationsLto.show that your, actuator,'the motor,-

8 the gear, and'so.forth is set up to deliver enough torque P 9 under design ba' sis conditions to achieve that goal.

'10 VOICE: 'Then we're back to,the problem with the

-11 typical MOV that was designed to certain criteria'and we're' 12 finding'out through alternate testing methods that it's.not 13 adequate. Now you're asking us,-there's no testing 14 ' mechanism out there besides basically full' Delta-P test, and 15- very. limited testing equipment to do that.

1:

16 MR. JACOBSON: That's why the industry needs to 17' get'together. We can't expect every plant to. test all of 18 their valves. It: won't work.

19- VOICE: My point'is, I don't physically have a 20 way to test them and get that'information that you're asking 21 me for.

22 MR. JACOBSON: You're saying there's no way to do 23 a full DP' test on a butterfly valve?

24 VOICE: No, a full DP test, but my valves go out l\

l 25 on limit. You know what I'm saying? There are certain l-Heritage Reporting Corporation (202) 628-4888

36 1- valves that go_out on torque.

2 MR. JACOBSON: The ones that go out on limit you 3 still can figure out what the required torque is at full DP.

4 VOICE: Then we're going back to calculation.

5 I'm saying an actual testing mechanism like using MOVATS 6 testing or -- testing, etcetera.

7 MR. JACOBSON: There are several diagnostic 8 systems out there that are available. I'm not exactly sure

9 which ones will do what. There are more deviations from 10 these systems under development now. I don't know what's 11 going to be available in the future.

l 12 You guys need to get together and decide where l .

l 13 the problems are and then decide in what manner you're going 14 to attack'them. If this is one of the problems, then that 15 needs to be known and somebody needs to come up with a 16 method to resolve it.

17 We understand some of the answers may not be 18 available right now, but there are a lot of things you can 19 get working on that already are available and you can start 20 progressing towards the goal of achieving the generic j 21 letter.

22 VOICE: I come from a 1969 vintage plant. Some 23 of the valves in that plant would quite clearly fall into 24 your not able to be practically tested category. We don't 25 have the kind of information on full DP testing you're

, Heritage Reporting Corporation l

l (202) 628-4888 l

i i

37 1 requiring,.and some of our valves are now obsolete. I doubt 2 there are five or ten in the country that could be purchased i 3 for the use of testing.

4 Are you inferring that since the lack of testing 5 data may exist we could be required to replace those valves?

6 MR. MARSH: If valves can't be demonstrated to be 7 operable, if you don't have tests to show that the valves 8 would be able to perform their safety function when needed, 9 I'm not going to suggest that you have to change out valves, 10 but there's some action that has to take place.

11 VOICE: It wouldn't necessarily have to be the 12 test. That's what I heard earlier, is that.you would 13 require some form of testing.

14 MR. MARSH: If you can demonstrate operability of 15 that valve under design basis conditions by performing less 16 than design basis differential pressure tests and show by 17 some technique that in fact those tests are extrapolatable 18 to full differential pressure tests, it may be possible.

19 That doesn't mean it will be possible, it may be possible.

20 I don't know.

21 We've heard this concern before and this is real 22 and we understand it, that there are valves out there that 23 aren't manufactured any longer. There are very few of them.

24 You may not be able to prototype test them.

25 The answer is the same that we gave. If you Heritage Reporting Corporation (202) 628-4888 i

- - _ _ _ - - - - b

lc ,

38 1 can't: full DP test them in the plant,.then see-if there's-a 2 way.that you-can' partial DP: test them at as high.a.

3- differential pressure as you can get to: demonstrate' f 4 operability'of the. valve. .

5~ VOICES- But just to'make.it. clear,'you're not

6 inferring that the;1ack of a physical test :woul'd require a 7 modification?

8 MR. MARSH:- I didn't say that'.

9 VOICE: Thank you.

10 MR. MARSH:- Wait a minute. I did not say that 11 because you can't test the valve you may not have to change l'

l 12 that valve out. You may have to change that' valve out.

!' 13 That's a determination that'you hav'e to make.

1

'14 VOICE: If we make that determination, that's all

.15 I'm trying to get clear, there'is the chance, the 16 . possibility that the valve c'ould.be acceptable even though 17 testing as.you would like to see is not available?

18 MR. JACOBSON: I'see this problem a little 19 similar to I saw EQ in that there were some components that 20- had to be changed out because there wasn't data out there 21 that could qualify them. Although the component may have 22 been okay, you didn't have the data to support it.

23 Now you've got a valve in your plant. Hopefully I

24 you can put.together an analysis that will show that the 25 thing is operable with some partial testing, some data, some Heritage Reporting Corporation j (202) 628-4888 1

, i l

!, g

/ -~ - - ~ ~ - - -,

1 m ,

39

. l' :this , 'some that . If~it's. grossly over-sized welneed to look

}

-2 .at the' margin. There are a lot of variables that'comelinto 3 .this' thing. We're'trying to.make it as simple as-possible,.

4 but there are so'many different case by case examples out y , 5 there'that we're just going to-have to look at them.-

l 6 We're not advocating that you go out and start i

7 changing all your valves just because there.are no identical 8, ones out there to prototype. However, we realize that L.-

-9 that's a real problem and we're going to have to'look'at it

, 10 on a case by case basis.

11 VOICE: Thank you.

12 VOICE: Jeff, you made the statement that there's 13- a large amount of data out there that industry can use to 14 demonstrate that the valve is operable, or they can be used

- 15 to determine,'well that the valve'is operable. What is this 16 large amount'of data that you're talking about?

17: MR. JACOBSON: I don't want to mention any names, l'8 but I know.there are some diagnostic people that have data.

19 There are numerous plants that have already done full DP 20 tests on numerous valves. That data is available. Maybe 21 it's not pooled together right yet in one specific place, 22 but the data is out there amongst the different utilities.

23 There's no reason why you can't use full DP data taken at 24 some other plant.

25 VOICE: But the EG&G study or the report that- ,

I I

Heritage Reporting Corporation 1 (202) 628-4888

L l'

L 1 .

(

l :40 ,

L 1 L 1 just came out that was referenced earlier went and stated in L

'{

1 2 it that diagnostic systems that depend upon constant stem {

l L 3 factor may not yield valid results. The major test system '

l 4 that's been used out there in the entire industry has been a 5 system that depends on that. So doesn't that cast a lot of 1

6 doubt on the data that is out there that you're expecting us 7 to use?

8 MR. JACOBSON: There are some inaccuracies 9 associated with all the data. The parameter you're talking 10 about, stem factor, was shown that it could vary in some 11 cases. I'm not exactly sure how much it varies. Whether it 12 varies 10 percent or 50 percent or 100 percent. The data 13 that's out there may be acceptable. You have to always use 14 tolerances on any of this stuff.

15 For a particular application you need to validate 16 how it can be used in your plant. What was the accuracy of 17 the diagnostic system that was being used that measured this 18 particular parameter? All that has to be taken into 19 account. It's not very simple. There's data out there.

20 There's a lot of things that tend to make the data I

21 inaccurate. And as the diagnostic systems have gotten I 22 better I think over the years, I think that data is becoming j 23 a little more reliable. -)

24 VOICE: That's true, the data is becoming more 25 reliable. However, the data that's out there today, we Heritage Reporting Corporation (202) 628-4888

J 41 1 don't know how inaccurate it is based on what we've seen 2- from'the EG&G study. _It. appears'that's the original testing 3 that was done by:INEL last summer, we're getting the results

'4 -this' summer, and there's further. testing being.donelnow.

5 That eats up.an entire year of our five year schedule.

6 MR. JACOBSON: The IEEL testing was done on one I

7 or two specific types _'of valves under specific conditions. j 6 There's been a wealth of other testing that has been done in 9 the individual plants that may show that,'for instance, -l l

10' changes in stem factors for certain types of valves are not 11 a problem. These tests that were done at INEL were done )

12 under very severe conditions in most cases. You may have j 13 conditions that are a lot less severe where this changing I

14 stem factor problem doesn't come into play.

15 We realize there are still questions out there, 16 'and all the answers are not currently in hand. -However, you 17 need to go forth and start to try to answer some of these 18 questions.

19 VOICE: I don't disagree about goir: tth and 20 starting to answer them. However, if I'm pne -1 to test 21 and demonstrate valves being operable, and later on the 22 answers come back and say it isn't, I'm the one who gets to 23 pay. We're dealing with devices which do not have an 24 unlimited upper bound on the amount of force they gan 25 tolerate.

Heritage Reporting Corporation (202) 628-4888 i

l 42 1 MR. JACOBSON: That's true. You always have to 2 worry about whether you're over-thructing the valve or the 3 actuator as well as under-thrusting. We realize that.

4 But we need to get this information to the best 5 of our ability now because we need to assure yourselves now 6 that these valves are operable. We can't wait and hope that 7 all the answers will be answered eventually, and in the j 8 meantime have valves that we're not sure whether they're 9 operable or not.

10 MR. MARSH: Let me interrupt for just a second.

11 I've got messages for Mr. Eaton, Mr. Berkins, and Bob 12 Borsom. I've got messages for you up here.

13 . VOICE: This is John Hostler from EPRI. I'd just 14 like to mention a couple of things that I think are 15 perceptions of the people up at the front table, and maybe a 16 little different from some of our perceptions back here.

17 We, at EPR1, working with the utilities, 18 particularly MOV User's Group, over the last six or eight 19 months have been working very hard wrestling with the 20 problem of how do we provide the technical basis for 21 responding to this letter. One of the first things that was 22 thought about was hey, there's a bunch of data out there.

23 Why don't we go get this data and put it all in a big data 24 base and if Plant A tested my valve and it worked, I'm okay.

25 That's way too simple.

Heritage Reporting Corporation (202) 628-4888

l 43 1 I wouldn't'like the NRC sitting up there' telling-

~

U .2' us we should do that because it's the wrong technical 3 answer. It's not.the way to do it.

4 There are about 18 or 20 or 25 parameters'on a' 1

15 given~ valve that you have to look at to decide if it's the 6 same, first of all. And only one or two of those, or three 7 .or four of those, are written down when they do a test out 8 there in the plant. So we don't even know on those tests 9' what's the material in that test, what was the temperature 10 of the fluid-going through?- We didn't measure that.

11 So'we're trying to go through this in a rational, 12- technical way to provide the adequate response to the 13- letter. I don't think it's just a simplified data base.

14 And if you tried to put together with the data that's out 15 there, with the problems the data has, not the least of 16 which is the fact that in general that data does not have 17 both spring pack and stem force measurement. It has one.

18 And that only has given us half the picture right now. We 19 know that, and we know we have to resolve that.

l 20 The program that we're developing is one that 21 will understand that problem so we can tell people how to do 22 tests in the future so they're going to know what that 23 effect is on that valve. Hopefully we can put the 24 methodology together with some reasonable amount of 25 verification testing that people can use on valves that j i

Heritage Reporting Corporation (202) 628-4888 i

44 1 haven't been prototypically. tested, but we understand how 2 the parameters change. Those individual 18 or 20 parameters 3 change, to tell them what's going to happen.

4 So the main point I want to make is putting 5 together a big data base with data that's out there now, 6 we've thought about it,. and it just doesn't seem like a good 7 idea. The data isn't technically going to give us what we 8 need.

9 MR. MARSH: We didn't mean to imply that it would 10 be easy to put together data. Just sort of gather it 11 together.

12 VOICE: I'm saying even if we did gather it all 13 together. It's not adequately documented where it is now.

14 MR. MARSH: Okay.

15 MR. JACOBSON: We understand that the data that's 16 out there now in itself is not going to be sufficient to 17 solve this problem. However, unless I'm totally wrong, some 18 of the data that's out there would help some utilities in at 19 least resolving the problems on some of their valves. It's  ;

20 not as simple as just gathering it up and one guy using data 21 because the valve's identical. We understand that. We 22 understand that it's a long, hard effort to try to get this, 23 all the data available'that's going to resolve the problem 24 and there's a lot of work that needs to be done in that 25 area.

Heritage Reporting Corporation (202) 628-4888

"s \ -

L l 45 L

i 1  : VOICE: TI think I'd just like you to recognize 2 .that-kindLof the approach we're~trying to take to help'them 3 'is to'put togetherfan analytical understanding of the valve,.

4 verified by a series of prototypical and separate effects 5- tests:to understand how individual parameters affect the 6 performance of the valve,-and basically give them an 7 extrapolation tool:using that.

. 8 That is different than an approach which takes 9 data which'is out there and saying I'm like him. It 10 basically uses a bunch of data.to verify a methodology, and

~

.11 therefore, you.go in and'you say.I'm this valve and you use 12 that methodology on that valvo' design, looking at all the 13- key parameters of that valve and decide how that valve is 14L going to perform in combination with a' static test for that 15 valve lto determine whether it's going to work.

16 That's the overall approach, and I think if you 17 have any big problem with that, you'd rather have us go out 18 and do a big data base and just go say I'm like him or 19 whatever, that's the kind of dialogue I guess I'd like to 20 hear.

21 MR. JACOBSON: We don't have a problem with that

22. approach. Whatever approach that is going to resolve this 23 problem is the proper approach to take.

24 (Laughter) 25 MR. JACOBSON: Whether the data that's out there Heritage Reporting Corporation (202) 628-4888 a

_ _ _ _ _ _ _ _ _ . _ . _ _ _ _ l

u .

-46

-1: can. help solve.part of,it, whichLI think:it can, whether you 2 decide to use~that orinot is your decision.

3. I know it's not going;to solve the whole problem.

4' I think it can help at least some' people solve some of their 5 problems.

6 VOICE: I'd like to point out, we have'been 7- compiling data in the industry to help verify the 8 " methodology. We've been doing that for the last six months..

z9 So we are using the data that's out there.

10 MR. JACOBSON: Okay. I-think we understand each 11 other.

12 VOICE: Isn't it true that the basis of the NRC's 13 concern in taking a look at the events that have occurred in 14 the attachment to_the generic letter, are driven not by 15 inadequacies in the design of the valve primarily, with the 16 exception of INEL testing, but by. inadequate maintenance in 17 installation and quality control program on the as-installed 18 valve? Isn't it true that the basis of your concern, the 19 reason why you're type testing may be a concern and whether 20 you don't know whether it's applicable or not is due to the 21 fact that your real concern is that we don't know what we've 22 got out there in the field? That we don't know all the 1 23 different components that make up an MOV?

24 If that's so, isn't it true that improving our j

25 maintenance program and improving the knowledge level about Heritage Reporting Corporation (202) 628-4888 1

p a

, .47 1- swhat exists out there.and improving the training of'the 2 people that work on the MOV's would resolve, if not all, 99 11 3 ' percent of the. concerns.you have on MOV's?

4 MR. SCARBROUGH: No. The problem,.if you read 5 Information Notice 89-61 where it was-found that the valve 6 . friction factor was two and a half' times above the value 7 -that.the manufacturer had reported. Therefore, every. valve, 8 you can extrapolate, there's a potential there for every 9 valve supplied by that manufacturer-to'have been given.the l

10 wrong friction' factor.

11 Maintenance is a problem and that's something we 12 hope to address as part.of 89-10 and part of the follow-on l 13 efforts we're going to~ work on. And we hope to have some L 14 improvements in that area, but there's a significant concern 15 for the design of the valve and what utilities are told how 16 that valve will operate, and what its performance 17 characteristics are. That is one of the main goals of 18 Generic Letter 89-10, to assure that what is out there in 19 the plants will operate under design base conditions,.

20 because we're not sure.

21 MR. MARSH: Expanding, certainly maintenance 22 testing and setting up is part of the problem, but it's not 23 all of the problem. There are design difficulties that 24 we've seen too.

25 VOICE: Could you tell me a specific event other Heritage Reporting Corporation (202) 628-4888 l

2

'48 I

1- than a. single or isolated:or small. percentage of the valve 2 case,'is'there a report on this design issue other than the.

3 INEL testing?:

4 MR.~ MARSH: As Tom just mentioned, there's an-5= Information Notice that's recently' published'that discusses-o '6 'valvelfactors that:were found to be significantly higher

.7. than were initially reported by the' manufacturer.

8 MR.-JACOBSON: Most of these valves do not get 9~ challenged under-design basis conditions unless there's a 10 seve'e r event that'is a precursor. So we don't. expect to see 11 a lot.of instances of valves failing out there as a result 12 of this improper design. However, there are instances out 13 there.

-14_ Most of the severe conditions that the valve was 15 designed for it doesn't see in normal operation. That's one 16 of the big problems.

17' Let me add something else to that. A lot of the 18 people that went and did.the testing for Bulletin 85-03 had 19 to raise torque switch settings, had to change out some 20 valves as a result of the fact that the valves required more 21 thrust than what was originally calculated. That's the 22 reason we're extending this generic letter to all safety-23 related valves, because we feel there is a problem in this i- 24 area, in the design area.

1 25 VOICE: Wasn't that driven, though, by the use of Heritage Reporting Corporation (202) 628-4888 e

.T-49 1 diagnostic. techniques and.their newness to the: industry,and' 2' our. lack'of.famiilarity with'.them?'.So because'of'that"we 3 set the valves as low as we could to get the tolerance as k 4 much as we'could on the operator to protect.the valves and y

5L the valve operators?'

61 MR. JACOBSON:' Certainly the inaccuracies

'7' associated with the diagnostic systems may have been part ofi

$ 8. the reason'you needed a greater margin in setting up your

9' valves. But throwing that aside, there were several-10 sinst'nces a where the valves are under sized. In the 11- information notice that Tom refers to, you've got valves 12 with friction-factors in the .5 and .6 range which are 13 ' double than what was assumed by.the original manufacturers.

~

14 :There's a large population of Westinghouse valves out there i

15 that have' friction factors that.are a lot higher than'what 16 was assumed. There's a-lot of valves out there that may-17 'have been improperly designed. There's a large majority 18- that are okay, though. We need to do this generic letter to 19 get a handle on which ones are okay and which ones aren't.

20 If there's just five percent out there that 21 aren't okay, well that's way too many.

22 MR. MARSH: Okay, that's about an hour for

.23 question number two. We've got six more to go. But that's 24 a difficult issue and that probably took a lot of the 25 discussion time. These other ones are, I think, probably Heritage Reporting Corporation (202) 628-4888 i

i

~

50 1 not as difficult to discuss.

2 MR. SCARBROUGH: The next small topic here is 3 dampers. The NRC has determined that licensees should 4 include dampers that are in safety-related systems within 5 the Generic Letter 89-10 program to provide assurance that 6 these components can perform their safety function under 1

7 dcaign basis conditions, j 8 Licensees should indicate in the December 1989 1

9 response called for by the generic letter, whether the 10 schedule can be met for dampers. If not, you propose an 11 alternate schedule.

12 The fourth item. The consideration of 13 mispositioning. The generic letter indicates that MOV's 14 that perform a safety function should be demonstrated of I 15 being capable of opening and closing to the extent that 16 these operations are required by the plant's design basis.

17 According to the generic letter, these safety-18 related MOV's should also be demonstrated of being capable 19 of moving to the correct safety position at the time that 20 the MOV is assumed to be available. The safety-related MOV 21 should be set such that the mispositioning stroke does not 22 damage the MOV, such that it cannot perform its safety l 23 function.

24 With respect to MOV's in safety-related systems 25 that have no specific safety function, but which may affect Heritage Reporting Corporation (202) 628-4888

51 --

' 1' the safety functioning being performed by the system,ythe L

2 -generic' letter indicates that those MOV's'should-be

-3 demonstrated:of being. capable of moving to their proper 4 position:following an' inadvertent operation at any. time in 5 an.FSAR accident scenario.

6- JAny' questions no_that?

7 VOICE: Could you just repeat that?-

-8 (Laughter) 9 MR. SCARBROUGH: Let me' paraphrase it.

' 10 MOV's that perform a safety function should l'

. 11 demonstrate that they're capable of opening or closing to-12 the extent that these operations are within your design 13 basis. _If it's. called upon to function, open or close, it

- 14 needs to be demonstrated to be operable'in that regard.

15 Safety-related MOV's also need to be demonstrated 16 to being capable of moving to the correct safety position at 17 the time the MOV is assumed to beiavailable. At the time 18_ it's assumed to be available you need to be able to put it

19 into its right position in the event that it somehow was

-20 placed in an improper position.

21 For MOV's in safety-related systems that have no 4

22 specific safety function, there may be MOV's in safety-23 related systems which have no specific safety function and l

l 24 they just sit there. If they can affect the safety function l

25 of the system in some way, they're in the direct line of Heritage Reporting Corporation (202) 628-4888

52 1 injection, or they're.in the direct line-that they could 2 disrupt the functioning of that system, the generic letter 3 indicates that these MOV's should be demonstrated of being 4 capable of being placed in the proper position following 5 some inadvertent operation at any time in one of your FSAR 6 accident scenarios.

7 So you look at your scenarios and see when these 8 MOV's could disrupt the operation of that safety system.

9 MR. MARSH: Let me give you an example. Safety 10 injection pump maintenance valves, suction and discharge 11 valves, their bodies are safety-related in that they have to 12 perform pressure boundary integrity function, but the motor 13 on those valves, if in fact they have motors and I'll assume 14 that they do for this example, those motors may not be 1

15 safety-related valves because they're maintenance valves.

16 So they' fall into the definition of a non-safety-related 17 valve in a safety system that do not have a specified 18 function.

19 But if those valves are positioned incorrectly as 20 an initial condition for that system, which means the valve 1

21 is closed, accidentally left closed, then you've got to show 22 that that valve, the motor, the torque switch, limit switch, 23 whatever happens to be'on it, is capable of being opened so 24 that in fact the system can be put into its proper 25 configuration. The same argument for the discharge valve on Heritage Reporting Corporation (202) 628-4888 1

53 1 the pump itself.

2 An example of a safety-related valve which has a 3 specified function is the header discharge valve on the 4 safety injection system. It may be open normally. Normally 5 opened and assumed to be open in the safety analysis so the 6 system can perform its injection function. The valve is not 7 supposed to be closed until some point later on during the 8 transient or accident when you throttle safety injection and 9 go into recirculation. But suppose there were an error and 10 the valve were incorrectly closed? The position says you 11 should be able to reposition that valve to its open position 12- by this generic letter.

13 So there are two examples of a non-safety and a 14 safety function for valves in a safety-related system.

15 MR. JACOBSON: Is that clear? We want to make 16 sure everyone understands this point. We had a lot of 17 questions at the last two workshops on this area. We want 18 to make sure you understand what we mean by safety-related 19 and position changeable and so forth.

20 VOICE: Is mispositioning considered to be a 21 single operator error?

22 MR. JACOBSON: We are saying, are you talking 23 about on safety-related valves or on these other category of 24 valves that we're calling mispositioned?

25 VOICE: Both.

l Heritage Reporting Corporation I (202) 628-4888 i l

L_-----_-- -_

l 54 1 MR. JACOBSON: For safety-related valves we are-l 2 saying you cannot take the single failure as the 3 disposition. In the cases where we said you have to be able 4 to assure it can be put in its safety function, that is not 5 your single failure. You have to do that. We realize 6 _you've got redundant systems and so forth and so on, but we 7 do not want to design a fault into the systems. i 8 VOICE: My question is, is it considered a single 9 operator error to disposition one of those valves?

10 MR. JACOBSON: It is an error to disposition a 11 valve. However, we're saying that is not the error you can 12 take credit for in your single failure analysis for these 13 safety-related systems.

14 MR. MARSH: I think I see what you're saying.

15 We're hearing you saying something different than I think 16 you're saying. We hear you say single failure and we jump 17 right to the single failure criteria. I think you're asking 18 something different.

19 VOICE: I didn't ask about single failure. Our 20 design basis includes single failure and includes a single 21 operator error as a separate consideration.

22 MR. MARSH: You only have to consider one 23 operator error. You don't have to consider in terms of 24 sizing the motor, testing it, whatever else, to operator 25 errors.

Heritage Reporting Corporation (202) 628-4888

l 55 1 VOICE: That's correct.

2 MR. MARSH: You don't have to consider that.

3. . It's only individually.

4 MR. JACOBSON: You have to consider one operator 5 error and then a single failure of a component in another 6 train.

7 VOICE: Is this a single operator error, my 8 question is.

9 MR. JACOBSON: It is an error. But what we're 10 sayihg, this may go beyond your design basis in the fact 11 that we're saying that these valves have to be repositioned.

12 Your design basis may take as a single failure an operator 13 error. You may say I've got another train that can handle 14 it. What we're saying is you cannot take that as your 15 single failure. You cannot take the operator error as a 16 single failure.

17 Now if you can take two operator errors and still 18 be okay, that's another question.

19 MR. MARSH: Let me ask you to flesh out for us 20 your question. What are you actually asking us?

21 VOICE: The design basis considers operator 22 error. It also considers single failure. I'm not addressing 23 the single failure. Single failure means a single failure.

24 Operator error means the operator performs an incorrect 25 action.

I Heritage Reporting Corporation (202) 628-4888

56 1 MR. JACOBSON: Does it consider both at'the same 2 time?

3 VOICE: It may vary depending on the plant or 4 utility. Some plants in the FSAR they specifically state 5 that the design basis considers a single operator error or a 6 worst case single failure.

7 MR.-JACOBSON: Or. What we're saying is you have 8 to take and in this case.

9 VOICE: The question is, straight forward, is is 10 this' considered a single operator error?

11 MR. JACOBSON: Yes.

12 VOICE: Then we will have to individually address 13 the way it affects our design basis?

14 MR. JACOBSON: Yes. This is considered an 15 operator error.

16 VOICE: This is not in addition to another error 17 that an operator would make?

18 MR. MARSH: We think we're answering it as we're 19 hearing it but I want to make sure exactly what it is that 20 I'm answering. So rather than go like this more, and think 21 we understand each other, why don't we stop and let's talk 4

22 with you separately and then give clarification to the 23 audience later about what it is. We haven't heard this type 24 of question before. We'll talk to you on the break and make 25 sure we understand your question so we can give you the best Heritage Reporting Corporation (202) 628-4888

)

a 15 7 1

1 answer.

2 VOICE: Okay.

3 MR. MARSH: .Thank you.

4 VOICE: My question is concerning the power 5 supply on safety-related valves, non-safety-related valves

. 6 installed in safety-related systems. .Most of our PWR valvee 7 have two train power supplies. We have'two different-8- trains, train A and train B.

On.our non-1E motor-operated 9 valves they have an untrain-related power supply. Does1this 10 analysis force us to include looking at the power supply to 11 this non-safety-related valve which is installed in a 12 safety-related system?

13 MR. JACOBSON: You're talking about the valves 14- that'we're saying are position changeable.now? That may be 15 non-safety-related?

16 VOICE: Correct.

17 .MR. JACOBSON: You have to verify that that valve 18 will go to a safe position or that it is blocked out. If 19 you block it out you don't have to worry about the power 20 supply.

21 VOICE: That counds like a good alternative, but e

22 it sounds like I still have a lot of analysis to do on 23 whether or not I need to upgrade that MOV's power supply to L 24 a safety-related power supply.

L 25 MR. MARSH: We are not saying in the course of Heritage Reporting Corporation (202) 628-4888

_. _._ _ _ _ _ . _ _ _ . _ _ , . _ _ . < _ _ _ _ - _ _ _ _ _ _ . _ _ _ _ _ - _ _ __m_ _ _ . _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ . _ _ _

M -

a 58 1 th'is~ bulletin to upgrade your power supplies to safety-2 related for'non-safety-related equipment. We're-not saying.

3 that. -This-letter is. confined to-the motor operator.for 4 'that'non-safety grade device.

? 5 If there's a problem with that non-safety grade 6 power supply,-to that non-safety grade motor, you need to'do 7 some thinking about_how to fix that..

We're not saying go 8 put it on the EBUs,.we're not saying to upgrade it or

~9 whatever else. But it really doesn't make sense to qualify a va've to be repositioned if-the power is not available.

~

10 l 11 MR. JACOBSON: You need to look and see what type

-12 of power it would get in the accident scenario., I don't 13 .knowLthat you have to assure that it's 1E power. But for 14 instance, if you know the non-1E isn't-going to be there for 15 some reason, that's no. good.

16 VOICE: Thank you.

17 MR. MARSH: Is that the basic question? That the 18 1E. power may not be'available to you, or the non-1E power?

19 VOICE: That's the question. Thank you.

20 MR. MARSH: That's a good question. And I just 21 want to reiterate that it was not the intent to upgrade or 22 change power supplies, power distribution systems because of 23 this generic letter. That was not the intent.

24 VOICE: The generic letter on one hand says I've 25 got to consider everything within the design basis when I Heritage Reporting Corporation (202) 628-4888

r r 59

'l- . start reevaluating my-design basis for-MOV's. So'I.look at <

l . .

12 : something slightly outside the design basis, I'm

, v

'3- " contradicting myself'because my design basis doesn't include 2

l45 .this operator error.

5 What's the difference between mispositioning that

6 valveland turning off the pump? Now why. don't I have to-n 7; . consider turning.off my head injection pump?

'8- MR. SCARBROUGH: The concept of mispositioning 9- first arose in Bulletin 85-03. It is be' yond design basis 10 for some plants. Why the generic letter says never outside IL design basis and then'mispositioning is right there in front 12 of you,'I can't explain that. But mispositioning has been 13 addressed, and that's the next point we're going to go over 114 .is mispositioning and why it's in there.

15 .So why don't you just hand on and let me...

16 MR. MARSH: The question about the design basis 17 and' constraining yourself to the design basis for the plant 18 was meant only in terms of new accident scenarios. No new 19 accidents, no new transients, no feedline breaks, no 20 steamline breaks if you have not considered that as part of 21- your stylized accident analysis. We do recognize in 85-03 1

22 Supplement I that by having an operator, having an error of 23 commission either initially or during the course of the 24 event, may be beyond the specific consideration in the i 25 stylized analysis. But it was always an implicit assumption Heritage Reporting CorporeLion (202) 628-4888

60 1 that no single operator error could totally defeat a system.

2 For example, in the case of a safety injection 3 pump. If the pump could not be restarted after being 4 secured, that would be a problem. We're not postulating 5 pumps being turned off at this point though. We're only 6 talking about valves being positioned incorrectly either as 7 initial condition within the course of the event.

8 VOICE: I'm just trying to make a comparison.

9 MR. MARSH: But it is a point we've gotten 10 questions on in the past. Geez, NRC, what are you talking 11 about? You're.saying stay within your safety analysis, but 12 you're telling us at degraded voltage and you're telling us 13 about operator errors and aren't you going beyond?

14 We're trying to say that in the case of 15 mispositioning it may be beyond what you specifically 16 considered in your stylized analysis, but it was approved in 17 that way and that's the clarification we need to give you.

18 Degraded voltage, we're going to discuss that in 19 just a minute, what we specifically mean.

20 VOICE: Thank you.

21 MR. SCARBROUGH: Let's go on to the next one.

22 The next one is specifically the concept of mispositioning.

23 As we said, this began with Bulletin 85-03 so you 24 should be very much aware of it. It's been discussed quite 25 a bit in regard to 85-03.

1 i

Heritage Reporting Corporation (202) 628-4888 1

i

61 1 This revolves around the use of single failure 2 criterion to remove mispositioning from your analysis. A 3 March 13, 1987 letter from Edward Jordan who was then 4 Director of the former NRC Office of Inspection Enforcement, 5 to the BWR Owners Group stated the need to consider 6 mispositioning even though this may be beyond design basis 7 for some plants. This need was reiterated specifically in 8 Supplement I to' Bulletin 85-03 in 1988.

9 In regard to Generic Letter 89-10, the fact that 10 mispositioning may be beyond a plant's design basis has been 11 fully addressed in the NRC review process for the generic 12 letter. Particularly, a backfit analysis was prepared for 13 the generic letter and approved. That included 14 consideration of the provision for mispositioning. So this 15 is not something new that we're bringing up in 89-10. It's 16 something that started back in 1985 and has been hashed over 17 and over.again. It may be beyond design basis, but it's 18 already been dealt with.

19 MR. MARSH: We do recognize that it is a change i 20 from classical, stylized safety analysis assumptions. The 1

21 change began in 85-03 Supplement I and it's continuing in 22 89-10 in terms of extending the scope to other valves that l

! 23 were not specifically considered.

1 l 24 Putting the matter aside, though of stylized 25 safety analysis, it really makes good engineering sense in Heritage Reporting Corporation (202) 628-4888

p o

. 62 I

l 1 my mind to make sure that valves can be positioned 2 correctly, to the right position. Just think in terms of

3. the 85-03 generic letter in the first place, in the very .

4 first place, was the mispositioning event. That was t'he 5 Davis-Besse event. The operator closed valves that should 6 not have.been closed in a safety system and by so doing he 7 defeated the safety function of aux feedwater and was not 8 able to reopen those valves. Both trains, simultaneous.

.9 Both trains.

10 So we began down this path by a mispositioning 11 event. That's the nature of our concern with respect to 12 mispositioning.

13 MR. JACOBSON: The next question concerns what's 14 commonly been termed as degraded voltage. I think that may 15 be part of the reason why we've had so many questions on the 16 subject. Let me define what I think we mean by degraded 17 voltage and what I think everyone expects degraded voltage.

18 What we're after .is that you look et the 19 particular accident scenarios and things that the MOV's are 20 required to work under and just simply, you figeru out ubet J

21 the voltage is going to be at the EOV at that time end you 4 l

22 take that into account. That's what we mean by degraded l 23 voltage. j 24 Now there are many different accident scenarios, j l 25 electrical lineups that may have to be cc nsidered. We would Heritage Reporting Corporation j (202) 628-4888 l

l l

L_- _._ _ _ _ _ _ _ _ _ _ _ - _ . __ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ - _ _ - _ _ - - _ _ ~

I 63 1 expect that you would look at when this valve is going to 2 have to operate and what the voltage is going to-be at that 3 time.

4 Many plants have chosen not-to take that 5' approach. A lot of plants have a spec of say 80 percent 6 voltage that they know envelopes all conditions at all times 7 in their plant and they use that. That's fine. But you may 8 find a particular MOV that would not meet that parameter, so 9 you may have to'do a little more specific analysis on that I

10 one and look at what the voltage at that particular MOV is '

11 really going to be.

12 We've had a lot of questions on whether this is 13 outside the design basis or whether it's inside the design 14 basis. I really don't understand where you're coming from 15 in that regard. The valve should have been designed 16 originally to work under the voltages that it's going to see 17 when it's required to work. Whether that was a specific 18 part of your FSAR, I doubt it, that it was specifically laid 19 out that this Mov is gojng to see thic much voltage. But ic 20 should have been part of the original design for the 21 particular MOV and it should have been considered in the y I

22 pec;urement and so forth of that particular valve and )

)

23 actuatar. l 1

24 MR, 11ARSH: Perhaps the concern rises from you y 1

25 thinking that we're, again, adding a new scenario onto the i i

Heritage Reporting Corporation (202) 628-4888 I

i w b

o 64 1 . plant voltage and power supply study. We're not doing that.

L 2 We're not asking you to do any more detailed' bus voltage 3 ' studies if you will.- We're'only considering that the design 4 basis on that bus, what it needs to be and the power drops l' 5 from the bus to the lugs on the motor. That's what you need 1

1 i 6 to do.

l 7 As Jeff said, if you can envelope that type of 8 calculation with a spec that shows that it is able to be 9 enveloped, that's acceptable.

10 MR. JACOBSON: There was a generic letter that 11 was issued I think, when was that Tom, in '79 which talked 12 about a specific electrical lineup that you also have to 13 make sure that your plant will operate under which I guess l 14 specifically addresses the loss of all on-site power. And l

15 that you have to ensure that the off-site power in 16 conjunction with the on-site distribution system will be 17 acceptable for operating the equipment that's required.

18 It's not just MOV's but pumps, valves, instrumentation, 19 whatever is necessary. That was also issued in 1979 and 20 that would apply as well here, to these particular MOV's.

21 They have to operate too.

20 Is there any question new as to what we meaa 'r,y f i

23 degraced v9lt age or wh'y conecne thinks it 's outside tiaeir

{

1 24 design basis?

25 MR. SCARBROUGH: The next item here is the two Heritage Reporting Corporation (202) 628-4888

h h .

65

'l step approach which is described in the generic letter.

2 It's in Item F, if you want to look for it.

3 The goal of the generic letter is to provide 4 assurance that MOV's and their associated systems can 5 perform their functions by MOV testing at design basis 6 conditions. Where such testing ic not practical for a 7 particular MOV, design basis testing of a prototype or 8 another MOV in the plant may be performed with justified l 9 application of the test data to the MOV in question.

10 If design basis test data is not available, the 11 generic letter describes an acceptable two step approach j 12 where the MOV is set conservatively using analytical means, .

13 using the best current information, until design basis test 14 data becomes available. But even with the two step 15 approach, the NRC staff believes that the five year schedule 16 provided in the generic letter is feasible.

17 Is there any question on the two step approach 39 and what we're talking about?

19 (No response) 20 MR, SCARBROUGH: This one concerns exteracions of 21 the senedule i.n the generic let',ar. We terched en this a 2.? little bit this morning, 23 Across the board extenstens of the five year 24 schedule in the geanric letter are not currently planned.

25 The draft generic letter, and I'm sure many of you saw it, 4

Heritage Reporting Corporation (202) 628-4888 i

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ . _ . . _ _ _ _ _ _ _ _ _ _ _ _ _ . . _ _ _ . . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ _ _ _ . _ _ . _ _ _ _ . _ _ . _ . _ _ _ _ _ _ . _ _ . __ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . _ . _ __ J

66 1 included a.three' year schedule. That was extended in part.

2 'because'of concerns that test data may not be available for

.3 certain-MOV's. Where a licensee-has an aggressive program 4

for performing design' basis testing when practicable, and

.5. for. obtaining design basis test data when in situ.testingfis 6

not. practicable, we will be receptive to individual requests 7 for schedule extensions on a case by case basis.

L 8 . VOICE: I'm sorry, I.did have a question on the 9 last one. That is, if we deem that we've set the operator.

-10 cons'rvatively, e would it be possible-to stay with that 11 ' approach indefinitely if DP testing would never be possible.

12 in'the future?

13 I understand what you're trying to say. You're 14 saying let's set the valve conservatively if we can't do 15 adequate testing on.it to get the right answer-for it. Then 16 .somewhere within the time frame of five years do a DP test 17 on it. But what if after five years we still haven't been 18 able to get a DP test on it or do any kind of further 19 testing?

20 Since we've set the valve conservatively, and 21 we've set it conservatively te "try to covert ourselves" from 22 an operability standpoint, it eeomo reasonable that if we  !

23 set it conservatively we should be able to leave it there.

'24 MR. JACOBGON: We've had that question before. I i

25 think we were currently planning to answer it later, but I

Heritage Reporting Corporation (202) 628-4888 1

1

p.

' .i :

1i 1

67

1. I'll go into that now.

2 Basically you're asking.whether if a valve is set 3 ' conservatively, and we're going to have to talk about what .

4 is conservative, does that mean that we still have to do the 1

p 5 DP testing. The key to it there is what is conservative and' L 6 'what do we know about the valve and where did you set it at.

7 I think on a case by case basis we may accept.very 8 conservatively set actuators. If you have an actuator, and 9^ l've seen some-out there, that is designed to' deliver ten 10 time's the amount of thrust that the valve is normally

~11 calculated.that.it's going to need, certainly I think that-12 in a case;1ike that we would buy off that as acceptable. If 13 it's two times, we don't'know where that number is going to-14 be. But there is probably a number out there that if you

. 15 can show it is very very conservatively set, we would be 16 able to live with the no DP testing. However,.you're still 17 going to have to do a static test to ensure that the switch 18 settings and everything are as you think they should be.

19 MR. MARSH: Just a word of caution, though. As 20 Jeff was saying the line of conservative isn't always that 21 clear. The EG&G tests were done.using an operator that was 22- thought to be sized neveral timea large.t than was needed, 23 yet <sven with that conservative sizing there was still 24 problemo with the valve.

25 MR. JACOBSON: yhe more data that's out there the Heritage Reporting Corporation (202) 628-4888

68 1 better we'll be able to get a handle on what's conservative 2 and what isn't for a specific type of valve.

3 VOICE: Are we talking then about a two step 4 conservatism? One that's conservative for five years and 5 another higher conservatism forever? Or do we have to apply 6 this ten times the amount of thrust required right away when 7 we first set the valve if we can't DP test it?

8 MR. JACOBSON: I just threw the ten times out, l

9 first of all, as an example that we know is acceptable.

10 VOICE: I understand.

11 MR. MARSH: There's a two step approach.

12 MR. JACOBSON: The two step approach is basically 13 aided, is in there as a guide in helping you achieve the 14 five year schedule. The goal of the generic letter is that-

, 15 you will come to some determination after five years that 1

16 the settings are correct.

17 VOICE: The way I see it, there is going to be l 18 the same criteria applied to whatever conservatism is 19 determined to be enough right away from the first day right 20 or. t.hrough from that point forward.

21 MP. JACOBSON: If you have doabts that the valves i

L 12 as set in yout pinnt today are not sized prepe.rly, that's a 1.

I' 23 problem, and it's a prtblem you can't wait live years to l

24 eddress.

25 VOICM: The only reascn I bring this up is it may 1

Heritage Reporting Corporation (202) 628-4888 l

69 1 be a very possible answer to some valves that we have, that 2 it might be better for us to just set them very 3 conservatively right now if they can take it, and leave them 4 alone, than it would be to set them conservatively right 5 now. Let's throw out the ten times. I'm not going to say 6 that's what you want, but say we have a valve and we set it 7 at ten times the thrutt we think it needs based on a valve 8 factor of .3. In my mind once we've set it there, and I 9 think ten times .3 is overkill, and if everybody is happy 10 with'that, why bother doing any more with it?

11 MR. JACOBSON: I think that's what we were saying 12 is in some cases where it's very very conservatively sized 13 and set up that way and the valve can take that thrust and 14 you're sure that everything is okay, that we may be able to 15 live with that for that particular valve and you may not 16 have to go and do DP testing later on on that one.

17 It's always preferable to do the DP testing.

18 That's the point we want to get across.

19 VOICE: But most of the time we do DP testing 20 because we can't live with the ten times .3 valve factor.

21 MR. JACO3 SON: Right. And ten times may not be 22 ten times. It may be lowe2 than that.

23 JOICE: I understand. I'm act going to take you 24 .l iteral] y . Thank you.

25 VOICE: I assume the NRC can ask questions? j Heritage Reporting Corporation (202) 628-4888

70 1 MR. MARSH: If you must.

2 . VOICE: The question is does the degraded voltage 3 need to be considered when you're running your test. The 4 valve test.

5 MR. JACOBSON: Degraded voltage needs to be 6 addressed in the design basis review and in the subsequent 7 testing. That's not to say you have to run a test of 8 degraded voltage. However, if you run the test at full 9 voltage you have to then adjust your results for how that's 10 goin~g to perform at degraded voltage.

11 For instance, if you calculate I've got 80 12 percent degraded voltage on a valve in the plant and you do 13 the test at 100 percent, you have to assure that that 14 actuator will put out enough thrust or torque to cover the 15 time when it's only going to see 80 percent voltage. For an 16 AC motor it's .8 squared, so it would almost have to put out 17 twice as much torque.

l 18 VOICE: So the answer is it needs to be 19 considered.

20 MR. JACOBSON: Degraded voltage always needs to 4

21 be considered wher.e applical.le. 4here way be aowe valves in l o

22 a plant that will never see degracesd salt.ags.. Tney nay g l

23 o! ways see 300 percent voltage. I i-24 VOICE: That will be deternined by the -- j 25 MR. JACOBSON: That needs to be veIificad that

]

Heritage Reporting Corporation (202) 628-4688 q h- _ _ _ _ _ _ _ _ . _ _ - _ _

71 1 that is the case.

2 VOICE: Thank you.

3 MR. MARSH: What I would like to do now is take a 4 break for lunch. It's a little early, but I think it's the 5 right time to... Oops, we have another question.

6 VOICE: Jeff, this one's for you. If you could 7 go back to eatlier where you talked about the prototype 8 testing and the type testing. You referred to a family of 9 valves. I'd like to get your definition of family, if you 10 would.

11 MR. JACOBSON: I don't think I have a specific 12 definition of family. I think what I was referring to was 13 maybe there's a specific manufacturer, a specific style of 14 his valves, although not identical. Maybe it's been shown 15' that in 100 tests that have been run the valve factor hasn't 16 varied more than between .15 and .25. That would be what I 17 would consider a family of valves. They may not be 18 identical, but let's say they're all a certain manufacturer, 19 a certain style, but the size might be a little different 20 from one to the other.

I 21 I could really foresee anything being a family if 22 you had the data that I cou2 d bac k it up. I thin). I was l I.

23 using the word in conjunction with the word identical Schere 34 we expect identica] to be the same manufacturer, same size, t

25 samz style, same waterial, same this, same that. Family Heritage Reporting Corporation (202) 628-4888 l

72 1 could be something a little different.

2 VOICE: The reason I ask is there was some 3 concern raised in some discussions that a six inch carbon 4 steel 600 pound valve manufactured by A and the same valve 5 purchased to the same spec manufactured by B may exhibit 6 totally different results.

7 Okay, type testing of similar, is that a similar 8 type valve? I don't know based on what we've heard. That'c 9 why I would like to get some clarification how you classify 10 a family of valves where you may be able to test one of this 11 or two of this family of 20 and say that the others are 12 probably okay.

13 MR. JACOBSON: That's why we're saying the data 14 has to support calling them a family. If you have one valve 15 of one manufacturer and one of another and they have 16 different valve factors, certainly they're not performing 17 the same.

18 VOICE: How far are you going to have to go 19 before you can classify these in a family, though?

20 MR. JACOBSON: We can't tell you exactly how many 21 valves of a type need to be tested to ensure that they're J 2? cil performing the same. We need to look at that on e case 23 by case basis. But if' you have 20 volves of a specific type i

24 or of two specific types, or 50 valves and they all perform 25 the same based on testing, that might be e pretty good  !

i l

Heritage Reporting Corporatic l (202) 628-4888 l l

l l

i 73 1 reason to conclude that that's a good family. But if you 2 have two valves and they perform differently, even if 3 they're identical, that's not a very good family.

4 VOICE: I understand that. Like I said, I'm just 5 kind of looking for where you want to classify the family.

6 MR. MARSH: We don't have, Warren, I wish we did.

7 WE don't have firm guidelines where the boundaries for 8 families are. We don't have that. We think you ought.to 9 base families basically on performance, a demonstration of 10 performance similarities as opposed to a designation of 11 type, as opposed to size, pressure rating, whatever else.

12 It needs to be based on data.

13 VOICE: I think we need to explore that 14 particular thing further at a later date.

15 Another question I had is you stated that, Tom, I 16 believe you said there was a backfit analysis available on 17 the generic letter. Is that in the PDR? Is that available?

18 MR. MARSH: Yes it is. It's in the PDR,  ;

19 definitely.

20 VOICE: We haven't been able to locate it. If j

21 you have it, we'd like to get a copy from you.

I 22 VOICE: Thank you. l 1

23 VDICE: My Qurtstion has u do with ir, daterv.ining  !

24 the feasitsility of imT cests, How far should we go? There 25 ore certain valvos that I could possibly DP teet, but in l l

Heritage Reporting Corporation

) (702) 628--4888 1

74 1 doing so I may inadvertently over-stress another part of the 2 system. I may damage a safety injection pump in doing so.

3 How far do we go in making that determination? What kind of 4 a safety analysis is acceptable?

5 MR. KIESSEL: It is the intent of the generic 6 letter that you not endanger your plant in any way with this 7 testing. We are not advocating that you go and cut metal to 8 rebuild your plant so you can do the test. We are not 9 advocating that you violate your tech specs.

10 If in performing the test you're going to over-11 pressurize components, then that is a reason for not doing 12 the full DP test in your plant.

13 VOICE: -The main concern was like, for instance, 14 a lot of our pumps, recirculation is usually not the optimum 15 range of operation. If we shut a valve and then all of a 16 sudden force all its flow through recirculation, we may 17 damage it and we may not know we've damaged it.

l 18 MR. MARSH: Let me turn the problem around. If 19 you've got a problem with a piece of equipment, that you 20 could damage a piece of equipment because of somebody L 21 mispositioning a valve or because a valve is doing its 22 uornal jcb and you're damaging equipment, you've got c l-L 23 probier.. That syatem should not be designed in that way 1

l 24 VOICE: You're right.

25 MR. MARSH: So you cant say don't do a test i

Heritage Reportinc Corocration."

~

(202) 62814988^

1

zin; - -

e i? ,

75

.'1 : because:the test damages the equipment because it wasn't 2- designed properly.

3 VOICE: Will the positions and interpretations 4 given'by this panel today be binding on those that are going 5 tol review our program five years from now?

6-  :.MR. MARSH: That's tough.

7. -(Laughter) 8 VOICE: Can we take what you say today and buil'd:

9 a program on it and ensure that we're not going.to get 10 jerked around in five years when someone different takes a 11 'look4 at our-programs?

12' MR. SCARBROUGH: We're in the process of 13 preparing _a temporary. instruction for the, I think I-14 mentioned this, to the headquarters inspectors and the 15 ' regional' inspectors anduthat inspection manual chapter will' 16 be routed around for review and comment.by the regions.

17- That will be discussed with them to come up with a

-18 consensus, implementation procedure for the generic letter-19 and how to review the implementation. That's as far as we 20 go.. I may not be here in five years so_I can't say.

21 VOICE: The question is more et what period in 22 tinae will we get e firm set of criteria to which we will be 73 -Joaged, th6t we can base our progrums on'now? ..

1 24 MR. MARSH: We're writing a TI, as we :said. The 25 TI nny not have in it specific criteria that you have to use Heritage Reporting Corporation

{202) 628-4808 j

76 1 in developing your program. You should not be looking to us 2 to tell us exactly how to construct your program. We've j 3 laid out the programmatic guidelines that we think have to 4 be met, the safety concerns. We've transmitted those to 5 you. The burden is now on you to develop your programs.

6 Now you have to turn to me and say okay, that's 7 great NRC. I'll accept the burden because that's my 8 responsibility. But how are you going to judge me? That's 9 a fair question.

10 We've already said that in a lot of areas we're 11 not sure. We're not sure the extent of the envelope, of the 12 family. We don't know that yet so we're not going to lay 13 out a TI with a family type of criteria unless we know.

14 You're working on that type of guidelines. You're working 15 on testing programs. You yourselves have to develop what 16 the family boundaries are because at this point you don't 17 know, and truthfully, we don't know either. So we can't say 18 what we say today is going to bind the NRC for life. That's 19 not fair.

20 We're giving you our best shot on the intent of 21 the letter, how it was structured, what the bases were, how l, 22 we view it, what you should use in contracting your l 23 program. But as to whether what we say today is going to be l

24 gospel for life...

l 25 'JOICE : I just say that with the realization that Heritega Heporting Corporation.

(202) 628-4888 i

f:

77 1 the clock is ticking on us and it will require programmatic 2 implementation of some rather costly measures. We don't 3 want to implement something to have basically your position 4 change on us in three years.

5 MR. JACOBSON: We don't foresee this as being our 6 last discussion with you on the subject. We encourage that 7 you participate in the industry groups that are being formed 8 in this area, and.I would imagine that we are going to be 9 having discussions with them in the future. If our position 10 radically changes on some subject, they'd be the first 11 people to know about it. We don' foresee our positions 12 changing on the subjects. That's one of the reasons we're 13 staying away from providing specific guidance in many areas.

14 We don't have that guidance to give you and we don't want to 15 have to go back and change our position.

16 Our job is to make sure that we implement our 17 requirements in a uniform manner through our inspectors, 18 through people that are going to be looking at your 19 programs. If you think you're getting jerked around by a 20 particular person and we're treating you differently than 21 we've treated someone else, you let us know about that 22 because we don't want that to be occurring.

23 MR. MARS 11: Let me add one more thing. 'I he 24 minutes that we're going to prepare from this meeting and 75 f rosa the past two neetings are going to be very care. fully Ecritage Reporting Corporation (202) 628-4808

78 1 scrutinized by the NRC because when we write minutes to a 2 meeting of this magnitude, of this import, we want to make 3 sure that we say the right things.

4 That doesn't mean what I'm saying today is going 5 to be changed in-the minutes. It's just going to be one l' 6 more level of management assurance that this is the right 7 track.

8 So-the minutes that you get from this thing are, 9 I don't want to say binding, but I'm saying one more level 10 of assurance that you have that the track we're going on'is 11 going'to stay firm.

12. Okay, if you don't mind, I'd like to take a lunch 13 break. Let's be back, if we can, at 1:00 o' clock please.

14 (Whereupon, at 11:56 a.m. the hearing was 1

15 recessed, to reconvene at 1:00 p.m. this same day, Tuesday, 16 September 26, 1989.)

l l

l l Heritage Reporting Corporation l (202) 628-4888 l

l

._m__m__...a_ - -__.

7 g ju 4 -J -

.=

79. i

'l AFTERNOON SESSION 21 MR.. MARSH: I have a couple of quick. -

3  : announcements before we begin.

4 I'd like to'ask when you stand and ask a question j

5. to please tell us your name and your organization so we can 6 identify it in the transcripts.

'7- Also, I want to clarify. The meeting minutes 8 will be.sent to everyone that's here. They will also be 9 sent to'all licensees. So there will be a double 10 tran'mittal s to your particular utility. And for those 11 licensees that aren't here, they will be receiving the

'12 meeting minutes,Ltoo.

13 Transcripts won't be sent to licensees.

14 Transcripts will be put in the PDR. So if you need those, 15 please make.your own arrangements to get copies.

16. The last thing I want to discuss'is quickly, 17 there has been an evolution of our position. There haven't 18 been significant changes, but we have clarified and made the 19 positions somewhat more concrete. What you're hearing 20 today, since it's the last, is basically the final 21 clarification.

1 22- So for those of you who either were in Chicago-or 4

23 Denver or know of thos'e that were in Chicago or Denver, read e

24 today's transcripts. We're going to base the meeting 25 minutes on basically today's meetings. So you may notice Heritage Reportir.g Corporation (202) 628-4888

.i

-1 I

80 ci some evolution of positions ~,o not significant. We haven't 2 made any major' changes, but some clarifications that are 3 today's.are the ones we want you to use. 1 4 Tom's going to now' start going through the 5 questions that we've received. LAgain, we've'got about 21 6' ' sets. Many of the questions that you're go'ing.to hear we've-7 talked about already in some general terms. Now we're going.

8' to get some very' specific questions.

9 Again, if you have follow on l

.10 questions to our answers, go ahead, chime in, stand up, tell

11. us who you'are, where you're from, and what are your 12' -questions, and we'll answer them as we get'to them.

^

13: Thanks.

14 MR. SCARBROUGH: We would start off with a letter 15- we' received from NUMARC. I don't know if you've received 16' this or not. It's August 14, 1989.

17 The first two questions we've already dealt with 18 previously. Mispositioning and degraded voltage.

19 The third one, and I'll read it for you, 20 "Regarding design basis reviews it is our interpretation 21 that a licensee may perform a design basis reconstitution of 22 an MOV application based on actual design and operational L 23 parameters of the system in which the MOV is located, rather 1.

L 24 than reconstruct the original procurement documents. In 25 many cases the valve vendor may no longer exist or 1

1:

L Heritage Reporting Corporation p (202) 628-4888 L

81 1 procurement documents and specifications have not been 2 retained by the AE or other procurement entities. Thus, 3 reconstructing the original design and procurement 4 information is impossible."

5 Our answer to that is that we prefer that you 6 reconstruct the parameters related to that MOV. Even if you 7 use the procurement documents, they're so old they're going 8 ft o have to be validated in any event. So it may be easier 9 in the long run just to reconstitute the design basis review 10 for that MOV.

11 The next comment --

12 MR. MARSH: Don't misunderstand that, though.

13 That doesn't mean that we're requiring reconstitution of the 14 design basis for cases where you've got the design basis 15 already available to you. It's in cases where you don't 16 have the design basis, you don't know what the design j 17 loadings were on these pieces of equipment. The best way to l

18 do that would be to reconstitute it for yourself, 19 recalculate it again.

20 MR. SCARBROUGH: The next comment in the NUMARC 21 letter was, " Referring to alternatives to design basis 22 testing in Item F, we interpret that what is meant by l 23 ' appropriate design basis test results on other MOV's' l 24 concludes test results could come from tests performed by 25 other utilities or organizations, for example, EPRI, not l 1 l l i 1

i Heritage Reporting Corporation (702) 628-4888 i

i I

m _ - _ - - - _ _ - _ _ _

82 1 just from one's own facility."

2 I think we've talked about this quite a bit 3 already today, is that we're encouraging utilities to 4 cooperate and use data from one another and also to work 5 with EPRI in this regard.

6 MR. JACOBSON: I wanted to add a little bit to 7 that and clarify some of the statements that I made earlier 8 that may have been misinterpreted. One of the things I want 9 to ensure is that if you're using other data than what has 10 been' generated in your facility, you have to ensure that 11 that data is valid and it is appropriate Appendix B type 12 data that can be used, whether it's from your facility, 13 another facility, some data base, or wherever that data 14 comes from.

15 The other point is I want to make sure that you 16 understand that we have not endorsed any specific data base 17 that may already be out there. We have not validated 18 anybody's data base. Any data that you need to use, you 19 need to ensure is valid data.

20 There's a lot of data out there that may be good, 21 and there's a lot of data out there that may not be good, 22 depending on how you want to use it and what application you 23 want to use it in. You have to determine for yourselves and 24 assure yourselves that that is good, valid data.

25 That should clarify that issue, I think.

Heritage Reporting Corporation (202) 628-4888 1

4

l i

l 83 1 MR. SCARBROUGH: The next set of questions came 2 from Virginia Power, particularly the North Anna Station.

3 Is Paul Boulden here? Our North Anna rep? We'll answer the j 4 questions anyway.

5 The first one was, " Generic Letter 89-10 states 6 that 'it is now reevgnized that the Section 11 testing alone 7 is not suf ficir.nt to provide assurance of MOV operability l I

8 under design basis conditions.' What testing does the i 9 Commission consider necessary to assure MOV operability?"

10 I think it's rather clear from the generic letter 11 itself. It discusses the testing, the design basis testing 12 and periodic diagnostic testing to verify switch settings. ,

. I 13 So we're not changing the technical specifications. We're 14 laying on another layer of testing that's needed to ensure 15 that the MOV's will work under design basis conditions.

l 16 VOICE: Do you think a position could be put 17 together that said basically that the new and substantial 18 requirements that are-being promulgated now basically would 19 supersede some of the stroke time requirements that we're 20 doing on some of the same valves for Section 11 in OM-107 21 MR. SCARBROUGH: Long range, down the road, 22 that's a possibility. But right now, until the programs are 23 in place and we have s6me data to indicate that the programs 24 are viable, the stroke time testing still is part of the 25 regulations and have to be complied with.

Heritage Reporting Corporation (202) 628-4888

4. .

84 1 VOICE: One thing I read into the letter, and' 2 . correct me if I'm wrong, is it your feeling that stroke time 3 testing.is a.way of. telling.the degradation in the valve 4 from aLmotor-operated valve? -I read into.that that fou-5 didn't think so.

6 MR. MARSH: There is value to be gained in stroke 7 time testing. We don't mean to say it-is of;no value. It 8 is of.value, but it's a question of necessary and 9 sufficient. Stroke time testing andLthe measuring of it for 10 AC inductor motors won't tell you much about what's-going 11 oon . It will tell you that the valve will work in that 12 condition. It will move. . That's not.an unimportant piece 13 of information to have. So from that sense it's good to 14 have stroke time-testing.

15- But there's-so much that's left unanswered about 16- stroke time testing, that that's why you need more than, You~need more 17 'that's why this generic letter was formed.

18 .than stroke time testing. Where the licensees can put 19 together some kind of program to go beyond or to replace the 20 in-service testing requirements, that is the stroke time 21 testing, I think they should think seriously about that.

4 22 I'm not saying yes, you certainly could do it, but I think 23 there is enough knowledge about what needs to be tested on 24 that an acceptable program could be promulgated.

25 MR. SCARBROUGH: Okay, why don't we go on to the Heritage Reporting Corporation (202) 628-4388

r '

85 1 'next one. Dick Kiessel is going to take th'e next' question.

-2 MR. KIESSEL ' The question reads, " Generic Letter H .3 89-10 states that, 'It'is-not clear that; tests of an-MOV at 4 low or' moderate pressure differentials can be directly 5 extrapolated to correct switch settings at design basis 6 conditionsLusing any type of diagnostic' techniques.' Does 7 the. Commission have any specificd 'ata.which indicates that 8 DP test data cannot be extrapolated?"

9' ,

.I think-the~ generic letter itself mentioned the 10 INEL' tests'that were conducted at Wiley. Again, those tests

. 11 have been reconfirmed at least on the valve'that did fail in 12 the recent tests.now in Germany. The recent LERs from 13- Catawba and Brunswick talk of valve failures in valves-that 14' they thought would be working properly based on their' lower

.15 pressure testing.

16 MR. SCARBROUGH: Owen Rothberg is going to take 17 .the next question.

18 MR. ROTHBERG: Question three reads, " Generic 19 Letter 89-10,'page 5, paragraph 1 states, 'It may become 20 necessary to adjust MOV switch settings because of the 21- 'effect of wear or aging.' Does the Commission have any data 22 to indicate that that outport torque or thrust of a properly _

23 maintained valve operator varied significantly over time?"

24- The key words here I think are properly 25 maintained. We have evidence of one problem that I can Heritage Reporting Corporation (202) 628-4888 a

h___ -__m-____.c..- . . . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ __

h '

V 86 1 think of in the specific is. spring pack deflection over time 2 that!has been recently identified. . 'I think what-happens 13 iwith motor-operated. valves are'that the switches do become L 4 subject to' adjustment over time and the valve may require L

5 .more1 thrust as it gets older.

6 -MR. JACOBSON:- The next~ question, " Generic 7 Letter 89-10 states.that no change to the existing plant 8 design basis is intended and none should be inferred, but 9 the design basis review should include.the effects on MOV 10 performance of design basis: degraded voltage even though E 11- operation of certain position ch'angeable MOV's under reduced 12 voltage may not have been included in the plant' design.

I 11 3 basis. Please clarify."

14 Basically our position on position changeable 15 valves and degraded voltage is that you have to ensure the 16 position changeable valve will go to its safe position. If .,

17 in order to do that it has to work under degraded voltage, l 18 then that needs to be factored into it. We understand the.

19 position changeable scenario may be.outside of your design (

-)

20 basis. We went over that before, and that's what we're 21 saying. That's what we're asking you to do in this case is j

22. that on position changeable valves they have to be able to 23 go in the safe position. You need to consider any factor 24 that may prohibit that from. occurring. Degraded voltage is 25 just one of those factors.

Heritage Reporting Corporation  ;

(202) 628-4888

87 1 VOICE: Bob Paladino, Boston Edison. In your 2 earlier conversation you discussed degraded voltage and I'm 3 not sure if I understood your definition. But is it 4 basically the design voltage for the system under the 5 transient as opposed to any, I guess, prior definition of 6 degraded voltage?

7 MR. JACOBSON: Our definition of degraded voltage 8 in the generic letter is the lowest voltage that that motor 9 operator is expected to see at a given time during an 10 accident scenario.

11 VOICE: The lowest design voltage?

12 MR. JACOBSON: The lowest design voltage. Now 13 for position changeable MOV's, if for instance your voltage 14 is' going to be 90 percent, but for three seconds you've got 15 a pump that comes on-line and the voltage dips down to 80 16 percent, we may look at that on a case by case basis and say 17 you don't have to have that particular one designed for 80  !

1 18 percent on a position changeable one because the chances of 19 that one being positioned at exactly those three seconds 1

20 would be very very slim. I 21 But if the steady state voltage turns out to be 22 90 percent and not 100 percent, or 85 percent, you would 23 have to ensure that it'could work under those conditions.

l l

24 But for instances where there's a very short transient of a 25 few seconds, we would look at that on a case by case basis, j Heritage Reporting Corporation (202) 628-4888 i

L ___m_.- _ __.._

I '

88 1 VOICE: Do you have to assume any single failures 2 in the electrical system when you're postulating what that 3 design voltage would be?

4 MR. JACOBSON: On a position changeable MOV?

5. VOICE: On any of the MOV's.

6 MR. JACOBSON: On the safety-related MOVs you 7 would postulate your normal accident scenarios that 8 everything else is designed for, whether that includes a

.9 single failure as part of your design base.

10 MR. MARSH: Again, we're not creating new single 11 failure scenarios. You take those stylized scenarios that 12 you have which determine bus voltage, determine line drops 13 when you go to the motor lugs themselves, and that's as far 14 as you need go. No new failures affecting the bus voltage 15 or diesel generator operability or whatever.

16 MR. SCARBROUGH: That's all for the North Anna 17 questions. The next set comes from Vogtle. I don't have a 18 name. Are the people from Vogtle here?

19 The first question was "The generic letter 20 states 'INEL's preliminary conclusions indicate that 21 industry sizing equations for MOV's that must perform this 22 type of safe ty-related f unction may not be conservative for 23 all design basis conditions.' Elaborate on the results of 24 this testing and its implications with regard to 89-10."

25 We briefly, I mentioned this befere. I think for i

Heritage Reporting Corporation (202) 628-4888  ;

)

l 1

___--- _ \

89 1 a detailed review you should get the EG&G report. Once 2 again, that's-SSR-85-47. That was July 1989. Some 3 highlights were the thrust was much greater than had been 4 anticipated. The disk factor was affected by sub-cooling 5 and valve design both. And the disk and stem factors were 6 affected by low. So there are several aspects of that INEL 7 testing that should be very informative and interesting for 8 your review. So I would suggest you get the report itself.

9 MR. MARSH: Is there anything in particular you'd 10 like'to ask about those tests? We have people here who are 11 very. familiar with them.

12 MR. SCARBROUGH: The next question was, "The 13 generic letter states, 'INEL has concluded that diagnostic 14 systems that measure both stem thrust and motor torque are 15 best suited for predicting valve motor performance under 16 design basis conditions.' Based on this statement, what is 17 the NRC's pcsition with regard to the acceptability of the 1

18 currently available diagnostic systems to support testing 19 and response to the generic letter?"

{

20 I think this was answered before. The NRC has no l 1

21 particular diagnostic system that is preferred. The 22 licensee should choose the system that provices sufficient j I

23 confidence in the proper switch settings. i 24 The third question Jeff was going to take on.

25 MR. JACOBSON: "The equipment currently available l l i  !

L Ileritage Reporting Corporation (202) 628-4888 i

90 y

1 to perform. diagnostic testing'on butterfly valves is'very.

2 limited. Since these. valves are typically utilized in low.

e3 pressure systems and are limit. switch3 opened and closed, is 'l 4- .it.necessary to perform any-type of diagnostic. testing to 5 identify torque switch' set points'for these valves?"

6 First of all, I.think the: question.makes some 7 assumptions:that are not always true. There are, I-know of 8 instances where utilities have: butterfly valves that do have 9- torque switches in the circuit. In that' case you would 10 -certhinly have to verify that those torque switches are set 11- appropriately.

12 If the torque switch le not in the circuit or is 13 . wired.out of the circuit, certainly we're not concerned 14 with where it's. set at. However, you do need to verify that 15 the-actuator can. deliver the torque required to operate the 16 butter. fly valve-under the design basis conditions,-which 17 means that you're probably going to have to do some 18 diagnostic testing on that valve.

19 MR. SCARBROUGH: The next question was, "The 20 generic letter raises questions as to the feasibility and 21 accuracy of taking static test results and extrapolating up 22 to the design basis conditions. However, the generic letter 23 also states that it is not possible to perform a maximum 24 differential pressure. test on some valves. What would the 25 NRC consider an acceptable alternative to maximum

,c Heritage Reporting Corporation (202) 628-4888 u

1 1

t-91 1 differential pressure testing?"

2 I think the generic letter spells this cut pretty i

3 clearly. Testing of a prototype of in-plant where the test 4 data can be justified as applicable to the MOV in question.

5~ The next question was, "Is the scope of the 6 generic letter restricted to valves and fluid systems, or 7 are certain valves and gas systems such as -- purge also 8 included?"

9 The generic letter says MOV's in any safety-10 related system, and that would include air systems.

11 The next question is, "The scope of the generic 12 letter is defined 'to include all safety-related MOV's as 13 well as position changeable MOV's.' What is the difference 14 in a safety-related MOV and a position changeable MOV 15 located in a safety system?" i 16 MR. MARSH: I think we've probably answered that.

17 I think the definition is clear. We'd be glad to go over it 18 again if you'd like. Is there some aspect y,tu want to talk 19 about?

3 20 Okay.

21 MR. SCARBROUGH: Ti.e next part of that question  !

22 is along the same line, so we'll move on to the next one. >

23 .

The next question concerns mispositioning and 1

24 beyond design basis, and we've addressed that. That's it 25 for the Vogtle questions.

Heritage Reporting Corporation (202) 628-4888 p

1

l L -

. 92 1 Is there anything we can expand on? The ones 2 we've answered so far?

3 The next set of questions is from Vermont Yankee.

4 This letter was dated September 8th.

5 VOICE: Dan Golston, South Carolina Electric and 6 Gas. I was wondering about the motor operators that degrade j 7 over time. I'm not too familiar with the diagnostic l

8 testing. I'm more an in-service testing type person. Will 9 the diagnostics generally show degradation over time? Or

~

10 will recurrent full Delta-P testing be required to find 11 these problems?

12 MR. SCARBROUGH: At the current time in the 13 generic letter it talks about verifying the switch settings 14 and indicating that diagnostics at less than full DP may be l

l 15 an appropriate way of doing that. That's quite a bit down 16 the road. That's our current thinking, is that you should 17 be able to use that. We still have a lot to learn in this 18 whole area.

19 MR. JACOBSON: We don't foresee that you're going 20 to have to be doing repetitive full DP testing in order to 21 verify that degradation of, at least the actuator hasn't --

22 These are the Vermont Yankee questions. The 23 first question, " Items'A and E of the generic letter did not 24 delineate the specific design basis reviews to be performed i

25 on each MOV. What guidance would be used in an NRC MOV '

Heritage Reporting Corporation l (202) 628-4888 1 l

I i

93 1 . program audit? Could a list of specific design basis l d

2 reviews be provided?" l J

3 We talked a little bit earlier about how we're 4 going to go about developing a temporary instruction for our j 1

5 inspectors to do MOV reviews. It hasn't been completely i 6 developed yet, however it is in the works. We're not going i

7 to provide a specific list of things you should look for in .

8 .your design basis review. I can throw out a few.

9 Temperature, pressure, voltage, orientation, things of that 10 nature. Basically anything that would affect the 11 performance of the MOV should be considered in the design 12 basis review. I think this is a good opportunity for you 13 guys to get together and somebody may have some more ideas 14 of things I left out that should be considered. Maybe some 15 standardized design basis review could be generated in the 16 industry.

17 MR. MARSH: You do understand that TI's are 18 public information, though. Temporary instructions. That's 19 what we're going to use to inspect licensees. Those are 20 public, so those will, of course, be available to you.

21 You'll see the framework that we use in judging adequacy of 22 programs.

23 MR. JACOBSON: The second question, "What are the 24 required records for the MOV?"

25 Basically any records necessary to demonstrate Heritage Reporting Corporation (202) 628-4888

C' , .

, , t .

94 p.

p t

1 'that'the MOV will perform itsidesign functio n. You have to-2 .mak'e sure that all your assumptions are validated, that i

3 1 references are included and so forth, and it has to be i 4 auditable data. So it's up to you to ensure'that the 5 required records are' appropriate for what you're trying to j 6 - sh'ow. It's similar to any other thing in the plant.

7- MR. KIESSEL: The~ final question deals with the 8 two step approach which'was mentioned in-Item F. The 9- question is, "Since statistical data bases use data points l: 10 other than a particular design basis condition, is this .

11 statement intended.to preclude'the use of the' statistical ,

~

L 12 data base to verify operability?"

l 13 For the overall question of operability, it 14 probably would unless your design conditions are enveloped 15 by the_ conditions that were in the data base. However, for the first cut, in other words the one where you want to try 17 .and do the conservative approach, they could form a basis 18 upon which you could base an extrapolation. Then later on, 19 testing data comes in to confirm your extrapolation, great.

20 You're safe. If not, you may have to increase your 21 settings.

22 But basically the answer boils down to it depends 23 on when you're trying to use the analysis.

24 MR. SCARBROUGH: Does that answer the questions 25 for Vermont Yankee?

Heritage Reporting Corporation (202) 628-4888 p .!

95 1 MR. MARSH: .Are we giv'ing enough of the question l

2 so that you understand the question and you understand the L 3 response? Okay.

4 VOT.CE: Just a quick clarification on a couple of 5 questions back. It's your opinion that we would only have 6 to full flow DP test the valve once and then we wouldn't 7 have to go back and re-flow test that valve?

8 MR. MARSH: There is guidance provided in the 9 letter. It basically leaves that determination up to you.

10 If there has been maintenance on the valve or there's been 11 something that you believe the full flow tests you did are 12 no longer applicable, then you have to do it again. The 13 generic letter doesn't require repetitive full flow 14 differential press tests. All it does is have you do it 15 once, and then I believe there's language in there that says 16 if you make the call that because of maintenance or because 17 of something you think takes it out of the applicability 18 past test, you're going to have to do it again.

19 Maybe I can read you the words directly.

20 VOICE: I'm just responding to the response I 21 thought I heard prior.  !

22 MR. MARSH: Is that what you thought you heard?

23 VOICE: Yes'. That's pretty much what I thought I 24 heard.

25 MR. MARSH: " Testing of MOV's at design basis Heritage Reporting Corporation (202) 628-4888

f p

-96' 11 . conditions needLnot beLrepeated.unless the MOV is replaced, je ,2. modified, or overhauled to the. extent that the licensee -

'3- considers thatJthe existing test-results are not 4 . representative of,the MOV in its modified. configuration."

5- VOICE:. Thank you.

6- . VOICE:' Paul

Dearforce,

New Hampshire; Yankee; 7 The~ question is are we. required'by the generic 8 letter.to have diagnostic test equipment installed'and 9f monitoring an MOV.during the full flow differential pressure 10 : test'ing?

11 MR. KIESSEL: No, you're not. But without an.

12 indication of what your margin is, how'are you going to-13 apply any degraded voltage conditions? Without adequate 14 instrumentation on the valve when you run.the full DP~ test, L 15 'how are.you going to use it as a basis for seeing that:

16' another' valve, an identical valve, or building this family 17 of similar valves is acceptable? Because you have no 18 margins. All you.have is'a go/no-go test if you don't 19 instrument.

20 VOICE: That's correct. I'm not getting plant

'21 specific, I'm trying to keep this in a generic perspective.

22 It may be possible to do a full. flow differential pressure 23 test at a period in time, depending on plant availability, 24 documenting that test without the benefit of diagnostics, 25 and then going back at a later date in a static condition l

Heritage Reporting Corporation (202) 628-4888

97 1 and finding out what the availables are and then line those 2 two tests results up to see...

3 MR. KIESSEL: The one thing you haven't 4 determined is what is the required thrust.

5 MR. JACOBSON: ,' e said he'd do a full flow.

6 MR. KIESSEL: But you have not determined what 7 the required thrust is. All he's proven is that motor 8 operator at that point in time had adequate capabilities.

9' VOICE: So what you're saying is that it is a 10 requirement that a DP test must have diagnostic proof?

11 MR. KIESSEL: No. If you can do the DP test at 12 degraded voltage, if that is applicable, that valve has been 13 demonstrated to be operable. Now five years from now when 14 you want to demonstrate that it has not changed, you don't 15 have any test data to fall back on. The only thing you've 16 got left is another full DP test.

17 Similarly, if your neighbor down the road happens 18 to have that same valve in a situation where he can't do a 19 full DP test, you can't provide him with any information 20 other than the fact that one test on one valve set up 21 somehow, which you can't demonstrate, passed.

22 VOICE: Thank you. 1 23 MR. JACOBSON: Let me add to that that there may 1

24 be things other than degraded voltage that we're not 25 entirely aware of right now that might also need to be l

4 Heritage Reporting Corporation (202) 628-4888

J, ', .

1 98:

1 considered which would make'the need for diagnostic testing-1' E L2 even greater.

3 -One. factor that hasn't really.been' considered and 4 needs to at least be looked at is the1effect of temperature 5 .on.these operators.- In some-installations they have to work

, 6. at very high temperatures and that's going to affect the

?

.7. motor output. That should be considered.

8 VOICE: Excuse me. Could I' follow up on that 9 previous q'uestion~just a little' bit?

l 10: What I thought I heard was when we.do worst case 11 differential, we'd set the valve up and go out.and test it.

12 but.not with diagnostic equipment, we would.have shown that 13 Lthe valve would. work. We may have a baseline under static 14 conditions with diagnostic. We just wouldn't have the.

l l 15 dynamic conditions.

16 So if we went back.to that baseline, we'd be able 17 to show that the valve was in the same condition under-18 static conditions that it.was at under DP conditions. So we 19 could relate the condition of the valve to the static 20 testing and then do the testing, prove that was okay. Then 21 we'd have a static baseline that we could use in the future.

22 If we met that baseline we're okay.

23 MR. KIESSEU: If your instrumentation is 24 sensitive enough to be able to, that by the time you put all 25 your conservatism in you still had some room to operate on Heritage Reporting Corporation (202) 628-4888

4 99 1- .that baseline, yes. But then also how are you going to show.

.2- .that you:have sufficient margin-to coverLconditions in which 3L you cannot dup 31cate.the. test? I mentioned degraded

'4' voltage. Jeff brought up the temperature effects.

K VOICE: The point'was, we would be' showing that 6 the' valve had the capability to operate under worst case-l: -7 differential pressure conditions, and we'd.have a baseline 8 we could go back to.

l' 9' MR. KIESSEL: And assuming that you've also:

10. adequate covered, such as degraded voltage and the-11 . temperature effects, then yeah, I think you may have a-12 possible way around it.

13 VOICE: I understand. But you're not saying'that 14 you require a diagnostic extrapolation to have an acceptable

15. addressment of those issues.

16 MR. JACOBSON:- No.

17 MR. KIESSEL: That's right.

18 VOICE: Thank you.

19 MR. MARSH: Why would you not want to do 20 diagnostics when you're at high DP testing? High rad areas?

21 You could do it at low DP but you couldn't do it 22 at high DP? You could have diagnostics at low DP?

23 VOICE: One'of the situations that the newer

'24 plants are in is that DP testing was done in the start up 25 phase'of the project at which diagnostic testing didn't Heritage Reporting Corporation (202) 628-4888 1

100 1 exist.

2 MR. MARSH: I see.

3 VOICE: We would like to try to figure out if 4 credit can be taken for that DP testing that was done 5 without benefit of diagnostic testing. Say on this date it 6 was demonstrated that at this torque switch setting the

)

7 valve operated under full flow DP. These were the test ,

8 conditions. Now I'm coming along at a later date, taking 9 diagnostic on a static basis at full switch trip and 10 measuring what those loads are. It's sort of a building on 11 previous efforts that have already been accomplished without 12 ~having to rack back and duplicate those efforts.

13 MR. JACOBSON: If you can account for all the 14 other parameters that need to be considered, that's a 15 perfectly acceptable way of doing it.

16 MR. MARSH: What you miss though, and correct me 17 if I'm misunderstanding, if you did the initial testing 18 phase of the plant, you demonstrated the vaive and 19 performance safety function with a full DP test but you 20 didn't do any diagnostics at that time at the high DP or the 21 zero OP, you didn't have any diagnostics at all. Now you 22 come along in time and you've got some diagnostics on zero 23 DP. What you don't know is how that valve has changed over 24 that time.

25 If you went back at this point and did a full DP Heritage Reporting Corporation (202) 628-4888

101~

1 t'est you, of course, reestablish that the valvo in its m? u 2 ' current condition can do it'and you've now got a current DP g 1

'3 ~ ; test;with-diagnostics. You.can establish a correlation. ]

-4 ButLwhat you're missing is;that length of time from the-5 ' initial. test phase until today when you took your.zero 6 diagnostics.

L 7- _Do you see what I'm saying?

1 8 VOICE: I understand. It's just left for us to.

9 determine if there is some way to justify the position.

it

- 10l MR.. MARSH: Right. The concern would be, though,

- 11 what is:the. current. ability of that valve to perform its

- 12 safety function-currently. I don't know what the length of l

b ' '

13 time'is involved in your-plant, whether it's a short time or L 14. 'a relatively long. time, but the aging effects or any.

15- maintenance that's been'done on the valve or anything.that 16 would affects its ability to do its job and its design DP 17 'would be, I-understand your concern. You've got a high rad

18. concern and your concern is can you take advantage of past 19 DP tests.

l 20 If you can, by all means do that. But you need 21 to be able to correlate to today.

22- VOICE: The unfortunate part of it, I'm sure 23 you're all familiar with this, is that many of the tests 24 that were done during startup testing at any plant cannot be 25 repeated due to configuration or whatever, tech specs.

Heritage Reporting Corporation (202) 628-4888 l;

i l

w __~__._----.-_L-__--.L-___-_-- _ . - _ - - . - - - - . . _ _ _ _ - - _ - - - _ _ . _ - - - _ _ . - - - - ~ . _

4

, 1021

.'l- LMR.' MARSH: Right. _But if there's any.way to 2- take advantage of those tests so you-don't have to do large 3 prototypical tests, see-whether using. good engineering 4 judgment, any. type of solidly based analysis,1you can do

5. that.

6 VOICE: Who would be the proper division of 7 regulatory agency.to check with as far as an approach for 8 plant specific?' would that be our individual contacts at-9' the site? Or would we be going back to --

10 ,MR. MARSH: Always go to your. project manager, 11 through'your project manager and to us, to the Mechanical 12 Engineering Branch. .We're the ones that are going to be 13 doing this program.

14 VOICE: Okay.

15 ' VOICE: Russ Goulding with, Florida' Power and 16 Light. To answer a concern we have, we only have 17 instrumentation to cover one valve. We normally line up six i 18- or eight valves and we'll be filling up the vessel, draining p.

19 it back down, filling it up, draining it back down to I

l 20 instrument all our valves. It's a one time shot is what we 1

p 21 try to do. So that gives us a problem on instrumentation.

L 22 The logistics of getting rid of your water.

23 MR. JACOBSON: You mean diagnostic 24 instrumentation?

25 VOICE: Yes.

Heritage Reporting Corporation (202) 628-4888

103 1 MR. JACOBSON: I guess you could buy some more.

2 VOICE: That brings you to another problem. We 3 have about six different-vendors telling us problems with 4 everyone else's. What do we want to buy becomes'another 5 problem.

6 MR. JACOBSON: We can't tell you what to buy.

7 That's one of the things you have to decide, is what system 8 do you want to go with.

9 VOICE: Assuming that you can do high DP to 10 verify that it's operable and then take a signature at loa 11 or no differential pressure to verify that the valve is 12 operable five or six years down the road, and we've all 13 admitted the effects of aging that a valve that's been 14 opersted or stroked 200 or 300 times is not the same exact 15 valve.

16 Why can't we take credit for identical valves 17 with static signatures? It's the same thing as taking into 18 account for the aging effect. If they're designed close 19 enough and they have the same characteristics and are 20 applied in the same application, then if the static 21 conditions look good for both valves and you only test one 22 valve on the maximum DP to verify operability, why couldn't 23 you use that information to verify identical valves?

24 MR. SCARBROUGH: We're not saying you can't apply 25 the data from identical valves. What we're saying is you l

g Heritage Reporting Corporation l 1 (202) 628-4888 l

l

104 1 need'to justify that application. We're not saying you 2 can't use that data.

3 VOICE: Are you saying that's one of the 4 alternatives then, is to justify to signature analysis?

5 MR. SCARBROUGH: Yeah, you need to justify it in 6 some manner, the applicability of that data to the MOV in 7 question.

8 MR. JACOBSON: We're not telling you how to 9 justify the data at this time, but it needs to be jastified.

10 Data'from one identical valve to the other needs to be 11 justified. Signature, diagnostic techniques, may be a way 12 of doing that, whether it's at low DP or static DP or high 13 DP.

14 MR. SCARBROUGH: Why don't we go on to the next 15 set of questions. This is Tennessee Valley Authority. Are 16 there any TVA people here? There were TVA people at every 17 meeting. i 18 We'll go over your questions again, j 19 The first one is, "Is NRC requiring the licensee 1

20 to address modes of MOV operation outside design basis?"

1 21 An answer is yes, the position changeable that we 1 .

22 have discussed quite a bit earlier today. But events, 23 design basis events stay the same. We're not changing the 1 24 events.

25 The second question, Dick Kiessel was going to i

Heritage Reporting Corporation (202) 628-4888 l

1 l

l l

L______--.

R

~

105 s- O 1 .- take.that one.

2- MR. KIESSEL: '" When a valve.has a torque switch

~

3. bypass to the~1ast two to three. percent-from a design basis J +

4' . standpoint of travel in the closed direction, can'the. valve 5 be~ considered closed if it trips.out on thrust?"

6 I'd say in general, no. .What I.would consider to 7 Ebe the definition of a closed valve is'that the disk is in 8 its seat and that you meet your. leak-tight requirements. l 9 From the~ nature'of the question I cannot conclude either one. j 10 of those is being' met.

1

.11 MR. SCARBROUGH: The next question is, "Does- '

12 Generic Letter 89-10 supersede Bulletin 85-03, and.are the 13 1 commitments for both an 85-03 still monitored by the.NRC7" 14 The generic letter does state that it supercedes

.15 Bulletin 85-03. We still expect you to. live up to the

,16 commitments in 85-03 for those valves because those are the 17 .most.important valves. That's why 85-03 was written at the 18 time. 'We would expect those to be completed as quickly as 19 possible.

20 You should state in your schedule what the 21 remaining 85-03 valves are and discuss it with your PM if 22 you have a problem in completing those as quickly as 23 possible.

24 MR. MARSH: Does that answer your question? Tom, 25 you asked about 85-03 Supplement I commitments and Generic Heritage Reporting Corporation (202) 628-4888 Whm-

l' 106-L 1 Letter 89-10 and whether this supercedes those commitments V

.2 'or not.

-3 VOICE: It answers the question. However, the

4. generic letter says it supercedes 85-03 with no. qualifiers.

5 .That would' lead us-to assume that it also-supercedes your-l 6 commitments, that you have a'new opportunity with'the 7 . generic letter to modify those commitments as appropriate. )

l 8 MR. MARSH: That's true. l

~

9- . VOICE: Those that are not covered by-the generic 10 letter that'were covered in 85-03, fall out.

11' MR. MARSH: To the extent that you have to. change 12 commitments previously made, 89-10 lets you change those 13 commitments .: In~other words, if you'said you were going to 14 perform tests at a certain point on 85-03 Supplement I-15 valves but yet when you look at the 89-10 you're. allowed to 16 change those commitments. In that sense it supercedes it.

17 It supercedes scope, it supercedes 85-03 in terms 18 of schedule. It supercedes it in~ terms of reporting .

19- requirements, too. It does talk-about no longer having to 1 20 send any more information in on 85-03. You can now just 21 obey the reporting requirements that are in this generic 22 letter.

23 I think that's about all I can say in terms of 24 superceding. What it's actually doing, it's laying a bigger 25 umbrella on top of motor-operated valves.

Heritage Reporting Corporation (202) 628-4888

____m___..___.m________._m - _ _ _

107 L li VOICE' .So in your_ opinion there'are no 85-03

~

2 requirements ^that are not addressed by the generic letter.

m 3' ;that'could_ fall'out?i 4 MR. MARSH: 'I think that's.true, o

5 MR. JACOBSON:- That's correct.

6 VOICE: Sue Montgomery, from PP&L. This goes 7 back to the question beforehand. Your definition of closed 8 was that it would also depend upon your leakage. test. What 9 if your valve is not required to be LLRT'd? .How do you know 10 whether it's. fully in the seat or.if you don't have any

^

11 leakage?

12 MR. KIESSEL: What does your design basis assume?

13 Let's assume we're talking about'a pipe break 14 isolation. valve. That valve is assumed to stop. flow.

15 VOICE: Okay.

16- MR. KIESSEL: In my mind whether or not you have

-17 leakage requirements, your tech spec leakage requirements on

1. 18 that valve as a normal leak rate test, you.still have to be 19 'able to meet that design basis criteria upon which you 20 designed the rest of the containment, if nothing else.

21 VOICE: So we should also be looking for zero 22 flow as far as closed?

23 MR. KIESSEL's The nature of the question was can 24 I leave my valve two or three percent open. I'm saying no.

25 In my mind, a closed valve is one that's in the seat and is Heritage Reporting Corporation  !

(202) 628-4888 l

)

l l

IL,

.i 1.N 5 108'

l -able to meet'whatever leak' criteria'that valve ~is there for.

'2J LIf the : valve is there to isolate on. pipe, break, then I would 3 assume that-youre'not meeting:your design basis if that 4: valve does not stop1the flow through it..

5- MR. MARSH: It-really depends upon'the safety 6' . function of the valve..

L

17 . VOICE

I guess that's what I was.asking, because 1

8 we have valves:that we would classify as safety-related,-but'

9. they're not.necessarily. included in our leak. rate program as 10 'foritech specs.

Ell -  ;MR. KIESSEL: I would assume most of your pipe

12. : break mitigation valves fall in there.

'13' VOICE: Those are included.

14' MR. KIESSEL: Those areLin?

15L VOICE: I don't know, whatever would be in the

~ 16 LLRT.

17 MR. KIESSEL: Those are typically for containment

18. purposes as' opposed to pipe break inside containment.

,19 VOICE: Right.

20 MR. MARSH: Again, it just depends on the safety 21 function of that valve. If that valve not going to its 22 closed seat impacts the safety analysis that exists for the 23 plant, then you've got'a problem. If it doesn't impact the 24 safety analysis for the plant, that means the system 25 function is able to do its job. You don't have a problem.

Heritage Reporting Corporation (202) 628-4888 i

)

4 i

I l <

109-

. 1 VOICE: Thank:you.~

I:

k 2 VOICE: -)3. E. Curry from Philadelphia Electric.

1r 3 This is a followup to Sue's question. It occurred to me 4 earlier.

5 In the case.of a parallel sheet gate valve, you-6" don't have any wedging. action. There seems to be no l .7 difference between,a valve'which is closed and held against 8 its downstream seat by differential pressure and a valve L 9- that is a parallel sheet gate valve. Whether it wedges or-10l not, it works the same way. Once the main valve orifice is 11 closed, DP action will go and make your valve leak tight.

12 What is the. difference between a valve that doesn't wedge-13- and a_ parallel seat. gate valve?

14 MR. MARSH: Again, I just want to reiterate what 15- I said. It depends on the safety function of the valve. If 16 the function of that valve can be accomplished by. going two

-17: to three percent closed, then it's performed its function.

18 VOICE: If the function is to go and close off 19 the main orifice, then that satisfies the requirement, 20 whether it wedges or not?

21 MR. MARSH: That may satisfy the requirement, 22 depending upon the safety function of that valve. If in the 23 two to three percent position, if it's a wedged valve, 24 you're causing release to outside a containment from a 25 reactor coolant, that's another matter. If you rely on zero Heritage Reporting Corporation (202) 628-4888

110 1 leakage, and you may not have accomplished that function at L 2 two to three percent. If you don't rely on zero leakage, if 3 some small amount of leakage is an acceptable consequence, 4 ' depending upon the analysis of the plant, maybe it's 5 acceptable. It's the safety function of that valve.

6 MR. SCARBROUGH: The next one was "Concerning the 7 KWU alliance testing in Germany, how will the results of 8 this testing be shared with etilities, industry groups, and 9 the NRC, and will the NRC entertain feedback from this 10 test'"

?

11 I assume this is the phase two testing that's 12 ongoing in Germany right now. That current testing is very 13 ' limited in scope, so I wouldn't wait for it to be released 14 before you start on Generic Letter 89-10. My understanding 15 is there's no formal feedback on the test results being 16 planned. There probably will be a report that's issued, the 17 same way a report on phase one was issued. But it probably 18 won't be sent out for public comment.

19 MR. MARSH: Roy Woods from the Office of Research 20 was going to maybe share some kind of information that's 21 available from the current test results. Roy, have you had 22 a chance to get any more results today?

23 MR. WOODS: I'm Roy Woods. I'm the Test Manager 24 for Generic Issue 87. I haven't talked to the people in 25 Germany today, but I don't think that's really relevant.

I Heritage Reporting Corporation (202) 628-4888 I

l '

111~

l' 'We've so far run four blow down tests. We've run

- -2 ;one blow.down test oncValve A which'is the same. Valve A as 3 we used last time. It's been refurbished, or had been 4 ' refurbished.

5 We setfit up based on our experience. We set'it 6 up so we= thought.it had enough thrust to close and'we gave 7 it a little bit of wedging. What it did is it came, 8 . theoretically closed to the point where the lower circle of-9 the. seat on the gate intercepted the upper part of the 10 circ'le cnr the bottom part of the valve body. In other words, 11 it theoretically just touched, and you'd theoretically uay 12 it isolated the flow. But it did not seat. It'was about 131 five percent open.

14 At the point.where it stopped, and it did torque

15. out at that. point and it stopped. There was considerable-16 acoustic noise. It's done in the range where we couldn't 17' measure the-flow, but it certainly hadn't isolated the flow.

18 There was lots of stuff coming out. That's not a very 19 scientific description, but that's the best I can-give you.

12 0 That was given that we took into account our 21 previous experience with that valve. We took into account 22 the fact that we know there are added losses in the stem nut 23 to a friction factor when you're under high load. We took l

l 24 that into account in our setup. We gave it 1000 pounds on l

That's where it torqued out, exactly where we 25 top of.that.

Heritage Reporting Corporation (202) 628-4888

^

112f x v +

-li intended:for it to,~which~was'not quite adequateito close ,

H. I

'2l ~it .

3: We've also run a'second valve, call it valve'W.

'4 lI guess I'm not supposed to identify.it, but'that's close -

, 5- enough.

6 '(Laughter) 7 It-was a 600 pound class valve instead of a 900-

~

8 pound, which Valve A was. Because of the fact that it was 9 600 pounds, it had a smaller diameter stem and the limit.was-10 ~then' stem buckling, as I, understand it. So we talked to, I 11 think.that'particular manufacturer no longer makes. valves 12 for this purpose,-but our contractor INEL talked to some

13. people that'had been involved in the design of the; valve.and 1

14 they' recommended'what.the maximum thrust should be. They 15 set it'at that,.which was considerably,-. vastly below what we

.E 16 had had available on the Valve A which had not,quite done 17 .the. job. So we couldn't even get. anybody to take: bets with 18 us, we couldn't get them to bet thousands of dollars'versus' 19 a beer that'the valve'would work.

20 Guess what?. It worked. It worked-three times.

21 ~It worked at 1000 pounds, it worked at 1200 pounds gauge, 22 and it worked at 1400 pounds which is beyond the design 23 basis. We were trying'to get three points for a curve.

24 We're now installing Valve V, in fact they may i

25 have run it. I'm expecting a call in nine minutes. So if Heritage Reporting Corporation (202) 628-4888 4

e

_f

113 L 1 anything significant-happens I'll update you.

2 Regarding the availability of this data, we have 3 made available in the public document room, and we sent to I 4' think all of the participants in the February 1st' meeting 5 here, we sent a data report, a massive thing that has 6 basically the data that we took in phase one. We anticipate 7 doing the same sort of thing with phase two, if not actually 8 putting out a NUREG. So we will make it available as soon H 9 as we can after the tests are finished.

10 I'll be here if you want to talk to me privately i

11 or publicly or whatever. I'll be around the rest of the 12 afternoon.

13 VOICE: What kind of diagnostic tests? .

14 MR. WOODS: I can't answer that in detail, but we 15 have instrumentation, our own instrumentation on this set of 16 tests that we didn't have before based on what we wish we 17 had measured before. We have invited all of the diagnostic 18 vendors to participate and most of them are participating.

19 But as far as giving you a list, I can't off the top of my 20 head.

21 VOICE: (Inaudible) 22 MR. WOODS: This is the same contractor. This is 23 INEL who is doing the test, so it's a carry on. The only 24 difference is we took competitive bids for the person who 25 actually had the capability of blowing steam through, or Heritage Reporting Corporation (202) 628-4888

) .

l'14 1 water through valves. This time.it was won by KWU Bechtel, 2 Seaman's Union or whatever alliance in Germany instead of r 3 Wiley Labs. So it's the same tests, same instrumentation 4 plus some.

5 VOICE: I just wanted to say one thing. EPRI is 6 getting full access to the data and we're using what we've 7 got so far. We're getting the data and it will be factored 8 into the program.

9' MR. MARSH: Let me just clarify something I said.

10 Tom, the question that came up that I was 11 responding to earlier dealt with commitment that people have 12 made in the context of 85-03 or 85-03 Supplement I. The "

'13 question was, does 89-10 supersede those commitments. I 14 want to add a clarification to it.

15 For some NTOL plants, Comanche Peak and other 16 plants, 89-10 may not superceded 85-03, 85-03 Supplement I 17 commitments because there is a specific licensing-action 18 that's supposed to take place in the interim. In other 19 words, 89-10 is a longer term project than 85-03 Supplement 20 I and 21 85-03. If a licensee said I'm going to get this done by 22 this date, they're still held to those commitments because 23 there is a licensing action due based on that commitment. I 24 think that's probably true in general.

25 If the NRC said okay, because of your commitment Heritage Reporting Corporation (202) 628-4888

_ _ _ _ _ _ _ _ _ _ - . _ - -___ ____ L

s ..g ,

L115

~

1~ we'11"either do or not do"a.11 censing' action, then contact' jg 2 .your project manager if there's a.-change in your. commitment' g- ,

l; 3: based on 85-03. If'we were going to rely on some action to 4 be taken or some program to be.doveloped, then it's-sort of 5 a different context.

6 MR. SCARBROUGH: The next one Jeff's going to 7 take.

,, 8 MR. JACOBSON: I think we already answered this 9 one. I'll go over it very quickly. Basically.it concerns 10 . waiv'ing DP testing ~for. actuators with no torque switch and 11 the fact that margin would be considered in that waiver.

12 1W e discussed previously that that may be an

-13 acceptable approach if.the margin is great enough. I don't

14. really know what more can be said about that. Does that 15 clarify the question?

16 MR.'SCARBROUGH That was it for the first set of 17 TVA questions. The second set is dated September 12th.

18 The first. question was, "Does NRC have a 19 preferred diagnostic system?"

20 I.think we've answered this before- . We don't 21 have a preferred system.

22 MR. JACOBSON: This is another question that 23 concerns testing and whether or not margin can be implied 24 and how do you apply data from one prototype to another.

25 Basically the answer is that you have to ensure that the Heritage Reporting Corporation (202) 628-4888

E...

L .

l 116 1 data you're using from your prototype is applicable to the l 2- . valve you're trying to establish as being okay in your 3 plant. We said before, we're not going to tell you exactly 4 how that's done, however we expect that you will verify that 5 .that data is applicable, whether the valve is similar or 6 identical or whatever.

I 7 Margin comes into place in many instances 8 whenever you're using a diagnostic system or 9 instrumentation. You always have some inaccuracies 10 associated with that. If you're relying on measurements ,

11 taken on.a prototype to set a valve in the plant, you're 12 always going to have to account for those inaccuracies or 13 add that margin, however you want to call.it.

14 MR. SCARBROUGH: The last TVA question was "What 15 are some examples of when differential pressure or flow 16 testing cannot practically be performed?"

17 Any time there's a concern for plant safety, 18 you're violating some of your technical specifications in 19 terms of reactor safety, and the licensee should make this 20 determination.

21 In that situation you would look for a 22 differential pressure test on a prototype or another valve 23 in the plant that you could perform the test safely.

24 That was it for TVA. Is there anything we can 25 expand on among these?

Heritage Reporting Corporation (202) 628-4888 i

117 1 (No response) 2 Let's go on to the next set.

3 MR. MARSH: What I'm going to ask us to do at 4 this point, I think we're probably about a third of the way 5 through it. Is before we answer a question as a panel, 6 let's look at it in terms of have we answered it before, and 7 let's skip it. With the indulgence of the asker, of the 8 requester, we'll move on to areas where we haven't had 9 questions before. We are starting to get repetitious 10 questions.

11 MR. SCARBROUGH: The next set was from Florida 12 Power and Light, and I know Russ is here.

13 The first one Jeff's going to take.

14 MR. JACOBSON: I think we already answered this 15 one concerning storage of documents. We don't care where 16 they're stored. They should be accessible and auditable.

17 MR. KIESSEL: The second question is "What is 18 -considered major or minor maintenance to an MOV7" 19 Rather than defining what is major or minor, when 20 we put it in the context that this is addressing post-21 maintenance testing and we are leaving that at the 22 discretion of the licensee to do the appropriate testing 23 that is necessary.

l 24 If the licensee feels that the work he has done l

25 no the valve would not in any way affect its performance Heritage Reporting Corporation (202) 628-4888 i

i

/ 118 1 then fine, he doesn't have to do any testing afterwards. If 2 he feels that it is a complete rebuild, he may wind up 3 having to do full DP testing.

4 MR. SCARBROUGH: The next one was "If the plant 5 uses the Westinghouse method of qualified limit switches for 6 MOV operation, does this generic letter apply?"

7 The answer is, yes it does. You're still looking 8 at motor capability.in that situation.

9 The next question has three parts to it. It

^

10 .says, "Concerning the design basis review A, are opening and 11 closing Delta-P required if the safety-related function is 12 only in one direction? "

13 You have to test in the safety direction and you 14 have to ensure that damage will not occur from the MOV in 15 the mispositioning direction.

16 The second part is, "Are normal operation Delta-P 17 values needed if abnormal condition is more severe?"

18 Probably not. You're looking for the most severe 19 condition. But could you expand on that? Was there 20 romething in there that we're missing?

21 VOICE: Not really. I didn't develop all these 22 questions myself.

23 MR. MARSH

What you'd need to be aware of is the 1

24 . differential pressure is only part of the scenario. Voltage 25 is another part of the scenario. All the little pieces that Heritage Reporting Corporation (202) 628-4886

y b.

119 R. , l1 add-together to,the conditions under which that valve is L , ,2- expected tofoperate have to be considered.

3 MR. SCARBROUGH: The third part is, "May.the f.l' -4 design review of cabling be restricted to DC valves?"

1. .

5 No. You'll still-have to consider cable losses

~6 for your AC'MOV's. We've talked about.that before.

7 MR.~JACOBSON: The fifth. question I think we've

'8 already answered most of this, let me go through it real 9  ! quick. The first part concerns adding margin for 101 instrumentation which I just talked about.

11 The:second part is how do'you account for

'12 degraded voltage.if you do the test at 100' percent voltage?

13. The answer to that is you'do it calculationally. You take

14. the. square of the degraded voltage for an AC motor, or

'15 straight value for a DC. That's how much torque it's-16 capable of.

17 "Should degraded voltage tests be performed at 18' maximum Delta-P7" 19 It could be if that's the route you choose to 20 take. However, we're not requiring you do that.

21 " Clarify concerning no concerning design basis 22 testing in Item A. Does this apply to the normal or

23. abnormal condition?"

24 The. design basis testing is meant to apply to the 25 most severe condition, whether that's abnormal or normal.

I Heritage Reporting Corporation I (202) 628-4888 l l

1 f _. _ _ _ _ _ _ _ _ _ _ _ _ _ _ ___ __ . ......._...__.d

~

s 4

120.

H' 1 MR. KIESSEL:- On' the s'ixth question, we weren't

'2. really able to understand it. Perhaps.you could provide'

.3 some clarification. Let me read it..

4 "Concerning-continuing maintenance. A,-provide 5- guidance lon acceptable continued monitoring." Could'you 6' flesh outfthat question a little bit for me?

u 7' VOICE: What was being asksd for was a way to try l: '

l 8- and' sort out what you considered correct monitoring of an 9? MOV. What' items 1do you want to look.at? What parameters?

10 MR. KIESSEL: Unfortunately, the answer is those 11 that you' feel are necessary to continue to assure that.the 12 , valve will operate.

l 13 VOICE: I think it goes back to several of our 14' questions were lined up that way, this minor / major 15 ma'intenance is the same way. You have one group who likes

16. to test and wants to test, and then'you have another group 17- who likes to fix. So I know where you're al1~ coming from.

18 I think we have a couple of questions that are that way, 19 trying to delineate between the guys who want to go out and 20 test and the guys who want to fix. Sorry-.

> 21 MR. KIESSEL: Sorry, we're not going to help you 22 there.

23 VOICE: The' change in stem factor, though, is 24 asked more in an analytical way from our engineers. How do .

25 you go about determining change in stem factors? Is there a.

Heritage Reporting Corporation (202) 628-4888

-l

121 l 1 methodology that's approved or applied that you all like? l 2 MR. KIESSEL: No, there's nothing that we have 3 officially blessed or approved or anything along those 4 lines. But by simple definition, the stem factor is the 5 ratio of the thrust to the torque, or vice versa. I can 6 never remember which one is which. But if you know what the 7 operator is delivering as torque and what thrust is in the 8 stem, then you've computed it.

9 The way the INEL people were approaching that 10 stem' factor, since they were not able to come up with a, or 11 they were not measuring torque in the operator, they were  !

12 pulling it off the motor curves, so therefore what they were 13 doing was they were measuring in essence motor torque versus 14 the thrust in the stem which they had data on. That 15 includes a lot of pieces of mechanical equipment tossed in 16 there. So it could have been varying. Those could have 17 contributed to the variance also.

18 MR. JACOBSON: A lot of the problem that we've 19 run into is we've tried to over-simplify a lot of these 20 factors and terms that are involved in these equations. The l

21 stem factor, for instance, we think of it as getting from 22 the thrust to the torque in the stem nut. However, there 23 may be other things that are involved in that as well. The 24 gearing and rate of loading. There's a lot of different 25 parameters that we may or may not know about yet that are r

Heritage Reporting Corporation (202) 628-4888 l

l l

+ .

l 122 1 being lumped into these valve factors and stem factors.

2 That's one of the problems we have to tackle, is try to 3 figure out what all these things are and somehow 4 analytically or by testing validate whether we're accounting 5 for them or not.

6 VOICE: On our list here, number eight, you 7 already. answered. But if you'd address number seven, a 8 couple of parts on that. I pretty well have an idea what 9 you're going to say.

10 MR. KIESSEL: Which parts do you want me to 11 identify?

12 VOICE: A, B, and C.

13 MR. KIESSEL: Okay, all three of them then.

14 The question starts off, " Alternative for design 15 basis testing. A, comparison with other MOV's."

16 In looking down through this, let me go down each 17 part. "How many of 20 similar MOV's would require testing?"

18 The operative word here is similar. The idea of testing 20 19 valves that I can really show are similar in performance is 20 a great idea and could provide the basis, but I have a 21 concern as to being able to develop that confidence among a 22 family of similar valves.

23 "Can an EPRI or industry data base be used?'

24 Yes, again, assuming that the test data that is 25 in the data base envelopes the conditions that you're Heritage Reporting Corporation (202) 628-4888

-- ; - , o ,

;^.-- , A ' u ,

123 1 talking about:to.be applied to the valve.

'2- -"Can Avilon' gate valve data be applied to an 3- Anchor' Darling gate. valve?"

4 Unfortunately, right now I'd have to!say 5 definitely no. Based on one set of data that I've seen, it 6- didn't_ work out.

'7 B. "How high can a 1000 PSID test be 8- extrapolated? 1500, 2000?"

9 'I" don't think it-can'be extrapolated. The 10 prob'lems'we've been running into with these high 11 differential. pressure tests, particularly when you start 12_ talking about fluid. temperatures that are approaching 13_ saturation,_are chat the valve does not behavs linearly as 14 far-as differential pressure is' concerned. Therefore, I 15 would find it very difficult to do any extrapolation, 16 particularly in the pressure ranges _that you're talking 17 about here.

18 "Can a," and I'm-not sure if that's a 6 or an 8 19 inch " diameter valve data be applied to a 12 inch diameter 20 valve?"

21 VOICE: It doesn't matter.

22 MR. KIESSEL: It doesn't matter, right. Again,.I 23 doubt that you can do this with your data, but if you can 24 demonstrate that this f aniily of valves does behave nice in 25 the size ranges, then yes, you could apply it.

Heritage Reporting Corporation (202) 628-4888 i

a C - ~_--u___.. _ _ _ _ . _ _ _ _ _ _ _ _ _ _ , _ _ _ _ _ _ _ _ _ _ _ _ _ _

124 1 For question 8, you said I don't need to answer 2 it, but I'm not sure this has really been approached. The, 3 question is, "Is the documentation required for this generic 4 letter the same as was required for each valve by Bulletin 5 85-03?" The answer is no. There's a lot more data that's 6 required.

7 VOICE: Thank you. The questions there on the 8 pressures and the extrapolations will help when we set up 9 our test procedures, because we will have people that will 10 want'to try and use a smaller system to qualify our bigger 11 valves. That's why I wanted you to answer them, because I 12 think some no's in there tells us how we have to set up our 13 test procedures. They're going to be longer test procedures 14 at this time it appears.

15 VOICE: Dennis Hassler from TMI. I just want to 16 get a reclarification on that major / minor maintenance and 17 retest criteria, etcetera. If the licensee determines a 18 program or develops a program that is based upon sound 19 judgment and is well thought out on their part, that will be 20 accepted and that's not subject to being --

21 MR. KIESSEL: I'm not guaranteeing that it will 22 be accepted. But right now we do not have any criteria that f 23 i says that if you go and change the paint on the valve that <

24 you've got to do this,'or if you tear the valve down totally l

25 and rebuild it you've got to do something else to it. I v

Heritage Reporting Corporation (202) 628-4888

125 1 think that as you described a well thought out and sound 2 program, probably would be acceptable. There may be some 3 minor tweaking to it. But I can't envision that if, as you 4 said, a well thought out and sound program, that it's going 5 to be totally scrapped and have to start over again.

6 VOICE: But because it's developed by the 7 licensee it's still subject to your review and concurrence.

8 MR. JACOBSON: Everything you do is always 9 subject to our view and audit.

^

10 MR. MARSH: Don't misunderstand, though. There's 11 not a review and approval process embodied in your program.

12 Your program you're required to develop and implement 13 yourselves. That's not something that you need to send in 14_ to us and get approval for it. That's a program that you 15 develop on your own.

16 VOICE: But we were inspected on that.

17 MR. MARSH: You would be inspected on that, yes 18 you would.

19 VOICE: We were already. Under 85-03.

20 MR. KIESSEL: But don't forget, Generic Letter 21 89-10 goes much beyond 85-03 in the post-maintenance and the 22 continuing nature of the program.

23 MR. JACOBSON: Let me bring this out. Specific j I

24 items that may have been looked at in 85-03 may have been 25 deemed acceptable and may not have covered everything that I Heritage Reporting Corporation (202) 628-4888

126

11 '89-10 is requiring. So there'may be.some changes to wnat 2' you:did.in' respect to 85-03 that:have to be' looked at.for-

'3 89-10.

l l 4- If you did 85-03 right, you should have probably l

5 covered everything in 89-10. But there'was a lot of 6- variation.

L 7. VOICE: As a follow up to:your response on how 8 far or whether you.can extrapolate DP test data,.I assume

'5U your. response is based on.what you believe the current-10- indu'try.

s understanding is of the relationship between 11 performance and DP, and the. fact that the INEL tests have l 12 ~given us some problems there. In the context that the 13 industry is now developing, an improved methodology that's 14 going to be validated with test data and once that becomes l

l~ 15 available I assume you would allow us to extrapolate.

16. MR. KIESSEL: If you've got the data to 17 demonstrate that you are capable of cr;vering all of these 18 uncertainties that have been raised by such things as the 19 INEL testing or the Brunswick problems.or the Catawba 20: problems or the recent LER type of things, I can see no 21 reason for not accepting it.

22 MR, JACOBSON: We just feel uncomfortable with 23 that extrapolation at this time based on what we currently

-24 know is out there.

25 VOICE: John Hirsch, Northeast Utilities.

Heritage Reporting Corporation (202) 628-4888 i

.g- .

-127-l' I understand'the. saturation' question as far as 2 extrapolation is concerned,:but has the testing that.you're

'3 doing right.now, has that identified any other things.thati

4. ~ might eliminate the ability to extrapolate from one' delta 5  ! pressure toL another?

6 MR. KIESSEL: Yeah, the Valve A problem was-7 actual mechanical interferences within the valve. There was 8 metal shaved off of the disk as it was riding into the seat.

9- That's the concern. Another concern above the saturated

10 Lfluid type of flow.

11 MR.lJACOBSON: The problem with.'his t --

12 VOICE: That's an improperly designed: valve then, 13 if you ask me.

14 MR. KIESSEL: It depends on where you'run the 15 test. Running those tests down in cold water, the valve 16 functioned beautifully.

17 VOICE: I understand the cold to hot. I can 18 understand the problem that might come there. But they 19 found that when the valve was tested cold it was fine and 20 when the it was hot it started shaving off the seat?

21 MR. KIESSEL: We don't know exactly which one of 22 the' tests started to take the metal off. It may have been 23 done during some of the hot functional tests that it was run 24 through. It's unfortunate that they hadn't disassembled the i

25 valve between each test to determine.

Heritage Reporting Corporation )

(202) 628-4888

9 128 1- But we do know the effect was very pronounced 2 when they did it, what was it 1400. pound test? It was the 3 upper pressure test where they were starting to get what 4 would have computed out to be a valve factor of somewhere 5 around .65.

6 MR. JACOBSON: The problem is we're trying.to 7 lump many things into this valve factor term, and it's not 8 as simple as just that. Some of these things that do affect 9 the thrust requirements only come into play at a higher 10 differential pressure. I think we're using valve factor to 11 encompass all the frictional forces that are in that. valve.

12 There's a lot of things going on. I don't know if you were 13 privy to the presentation from the Germans on some of the 14 work they've done, but they've got a lot of other things j 15 they've discovered that are really going on in the valve 16 that really aren't modeled by the current industry equation.

17 Some of those things only become apparent at the higher l 18 differential pressures.

19 VOICE: Thank you.

20 VOICE: Mike Mancini. Just a response to Mr.

21 Kiessel, or an addition. At the latest INEL testing there 22 is some torque gauges that are being put on the stem itself 23 to get a more accurate number for the stem factor.

24 MR. KIESSEL: Right, and I'm looking forward to 25 seeing how that shakes out then.

Heritage Reporting Corporation (202) 628-4888 i

l l

129 1 VOICE: Just a follow up, Dan Stangor with the 2' Bishop Cook law firm. A follow up to the question by the 3 gentleman from TMI. I'm trying to understand how the 4 responses to the generic letter fit into the licensing 5 context.

6 When the submittal comes in and the licensee 7 identifies alternatives to the testing called for by the 8 generic letter, will there be review, approval, disapproval 9 of the alternatives that the licensee puts forth?-

10 MR. MARSH: There may or may not.be depending 11 upon if a licensee first specifically says that a review is 12 needed because it's an unreviewed safety question. If he 13 makes that call, that an amendment is needed, then a review 14 is owed and will be done.

15 Beyond that, though, the alternate programs need 16 not necessarily be reviewed and approved. They may not 17 prepare an SER for that. We may acknowledge the receipt of 18 them, we may sample them, we may go out to the plants and 19 see how they're done specifically.

20 But I want to emphasize, this is not necessarily 21 a review and approval process. If alternatives are proposed 22 as the generic letter allows, licensees should not wait for 23 a response, generally. j 24 But if a licensee feels like they need it, they )

)

25 feel that they have extended themselves, they're really Heritage Reporting Corporation (202) 628-4888 ws~_-._m----_a-.- - . .- - - - - - - -. - - - - - - , - - - - -

L <

130 1 proposing something for which staff approval is warranted 2 and needed, then we want that to be communicated to us 3 through the project managers.

4 But the first answer, first blush, no. A review 5 and approval process is not the intent of the letter. SER's b 6 are not planned on to be written for the responses. Two 1

7 caveats: unreviewed safety question, yes. If a licensee 8 feels like he needs guidance from the NRC because he feels 9 like there is something that he feels he's extending himself 10 or he broaches on an unreviewed safety question, then you'll 11 get guidance. That may not be an SER. that may be dialogue, 12 something of that sort.

13 VOICE: If there is an area in which the NRC 14 disagrees with the licensee's proposed alternative, how do 15 you think that would be handled?

16 MR. MARSH: First is dialogue. First project 17 managers will be asked to look at the responses to see 18 whether there are exceptions for their plants. Then they'll 19 be given probably some guidelines as to what is an 20 acceptable variation in programs. If they fall outside those 21 guidelines, they'll come to us, the Mechanical Engineering 22 Branch, and say what do you think? We'll look at those and 23 determine whether we need to dialogue with the licensee.

24 If we need to go the next step, which is the more 25 formal letter and RAI type of business, we'll do that.

Heritage Reporting Corporation (202) 628-4888

g .

~ . .s 0 131 1: VOICE:- What about-the inspections down the road?

2- . WillJthose be compliance-oriented, will they be in the' sense 1~_

l' 3 of a verification of commitments that' were tude, or will 4: 'they go beyond the commitments and take a'look at'whether L 5 .the licensee's. alternatives may-not meet what you asked for 1

6 in the' generic letter?

7 MR. MARSH: You said about three things, and it 8 would probably be all three. There will probably be 9  : compliance, they'll probably determine whether the-10 alternative programs they've proposed are acceptable, and 11 they'll probably.see how in fact they've implemented.the 12 programs.- So it is a compliance and it is an enforcement-13 related issue. So those are going to be legitimate audits 14 at some point and inspections at other points.

15 First you need to know that the first stage in 16 this plan is to do audits of plants. Audits mean go find 17 out what is the general state of program development. Those 18 are. planned about four or five this fiscal year. Those are

- 19 information gathering audits.

20 The next step is the inspection step. .That's 21 after the TI is developed and we go and find out in fact how 22 they've been implemented and look for enforcement-related 23 issues.

24 VOICE: What would be the basis for enforcement 25 action? If there's a failure by the licensee to meet a Heritage Reporting Corporation (202) 628-4888

._h_-. --_-_--_---.__m. - _ _

'e ,

132f.

p "1 commitment it made.under the enforcement policy, that'might' r :2 be a deviation,.but is'there'a basis for enforcement action?,

3 MR.. MARSH - I'm way offof-my area.of 4' . qualification and responsibility. I wouldn't even begin to 5 . discuss that at this point.

l 6 MR. JACOBSON: A lot of'these things would be-R p 7 enforceable. .If they.have a valve that is notl operable,t 8 certainly that's.an. enforceable-item. I! don't know.what.

9' particular thing.that would be enforced'against, but we're:

.10 L goin'g to be looking for whether these things are designed ~

11 and whether they're going-to operate under desj , basis l . .

12: conditions. That would be the type of enforcement we'd be 13 looking at.

14 VOICE: Thank you.

I 15 MR. SCARBROUGH We had a question by telephone 16' from Florida Power and Light, a couple of them. One of them 17 we've already answered. The other one was "Could we give an-18 example of a position changeable MOV that is not safety-19' related."

20 I can understand the concern there.. We consider 21 MOV's and safety-related system to be safety-related. But 22 it's my understanding that some plants classify maintenance 23- valves or test valves as non-safety-related, but they're

-- 2 4 still located in safety-related valves. That is where the 25 distinction came out for position changeable and made us Heritage Reporting Corporation (202) 628-4888

U,> , .

133' m

l- call them out specifically.

L That was it for Florida Power-and' Light.

h.

3 The next set.of questions comes from Georgia 4 Power, specifically the Hatch Plant.

5 The first one concerns mispositioning, and we've 6- already addressed that. The second one says, "The. generic 7 letter refers to Section 11 testing and. safety-related 8 MOV's. .Are the valves to be included under the generic 9 letter the same as the MOV's which form a subset of our 10 overall Section 11 ISC program? If not, why not?"

11 The FSAR defines safety-related systems. The 12 motor-operated valves that are in those systems are the 13- valves that are within the scope of Generic Letter 89-10.

14 So.you don't need to.try to correlate the two sets of MOV's- .

15 There will be a lot of overlap, of course. But you 16 shouldn't try to ensure-that'they're overlapping sets or 17 one's a superset of the other. Just look at the FSAR and 18- find out which safety-related systems you have. The MOV's 19 in those systems are.the ones that are covered by Generic 20 Letter 89-10.

21 The third question Jeff was going to take.

22 MR. JACOBSON: I think we've already answered 23 this one. The first p' art concerns whether, if the design 24 differential pressure is low and static loads dominate the 25 required thrust and torque, can they delete a full pressure Heritage Reporting Corporation (202) 628-4888 l

. I 134 l

1 test?" That would depend on a case by case basis as to what  !

2 type of margin and so forth was involved.  ;

3 The second part of the question also concerns an l

4 over-sized MOV and margin. We've already talked about that. i j

5 MR. SCARBROUGH: Ted was going to take the next ,

6 one.  ;

7 MR. SULLIVAN: The next question says, "The 8 generic letter says, 'It is now recognized that Section 11 9 testing alone is not sufficient to provide assurance of MOV 10 operability under design basis conditions.' Is it l l

11 permissible to delete Section 11 stroke time testing for  !

i 12 MOV's, at least those powered by AC, once we have an MOV i 13 testing program acceptable to the NRC in place under Generic 14 Letter 89-10?"

i 15 There are several points that need to be brought l 16 out. One is that we are not necessarily going to review all 17 these programs. This particular question seems to infer 18 that we are. So I'd like to dispel that notion.

19 In terms of deleting the testing requirement, I i

20 think there are a couple of things that are operative that 21 need to be mentioned or c vsidered. One is that the 22 requirement is coming from the ASME code which is required 23 via 50.55AG. So at the very least this would require a 24 relief request that we would have to review. I think that 25 would become the vehicle for review and approval.

Heritage Reporting Corporation (202) 628-4888

(

135 1 'The.second point I'd like to note is that~some-2 valves'are contained in' tech specs, and'although the tech-3 ' spec requirement refers to the code for the testing, the 4 tech spec is going after a different point than the code is.

5 So a tech spec. change may also be needed to accompany _this 6; . type of1 frequency and method change.

7- ~The systems aspects of this test, the 8 reliability, can the valve stroke in the time assumed in'the 9 safety-analysis, on this frequency would have to be reviewed.

10 by a'different set-of people than ourselves in Mechanical 11 Engineering Branch.

12 So I don't mean to make this overly complicated, 13 but there are at least two things that need to be considered 14 and possibly two sets of review to push this kind of change

-15 through the system.

16 MR. MARSH: I want to also point out that section 17 11 testing for motor-operated valves does have, stroke time 18 testing, of course, is the main element. But it's got 19 various time constraints too. For example, for either an AC-20 or a DC motor-operated valve or any kind of motor-operated

'21 valve, any kind of power-operated valve, licensees have to 22 establish a reference value, then they have to determine an 23 alert range, and then a limiting value of full stroke.

24 These tests are, of course, done quarterly. Whether or not 25 you can replace that quarterly test with a test that's to be Heritage Reporting Corporation (202) 628-4888 l

L 136

1. done every five years is'a pretty big leap. It's a pretty 2 big change.

3 As I said, stroke time testing does give some 4 information about valve performance. It is not a null set.

5 It does give you some health information. To support 6 replacing the IST Section 11 requirements with just the 7 Generic Letter 89-10 program is probably not, on the face of 8 it, enough. Because you're giving up quarterly tests with 9 some diagnostics basically on alert ranges and stroke time 10 test's with every five years. That's a pretty large leap, in 11 my mind. There would need to be more thought given to that.

12 MR. SCARBROUGH: That was it for the Georgia 13 Power questions. Is there anything we can expand on?

14 (No response) 15 MR. SCARBROUGH: Thank you.

16 The next set, we had a couple of telephone 17 questions from Dennis Hassler. The first one concerned 18 safety-related systems, and those are defined in the FSAR 19 and that's the ones we're talking about.

20 The second one was the position changeable f l

21 definition, and we've addressed that one, I believe. {

. i 22 MR. MARSH: Would you like to take a break? I 23 Let's take a 15 minute' break, please.

24 (Whereupon, a brief recess was taken) 25 MR. SCARBROUGH: The next set is from Heritage Reporting Corporation (202) 628-4888

137 1 Consolidated Edison, Indian Point Two. Are those 2 representatives here?

3 There are eight questions here. We have a number 4 more to go through, a number more sets. Is there anything 5 in the comment letter that you have that you want us to 6 address in particular, or do you want us to go through all 7 of-them?

8 VOICE: Some of them you've already talked about.

9 10 MR. MARSH: If you want us to go through them...

11 VOICE: We've covered so much stuff I'm not 12 sure...

13 MR. SCARBROUGH: We can try to zip through them 14 real quick.

15 The first one was "What methods other than 16 testing at full design basis differential pressure, if any, 17 are acceptable to prove operability of a given safety-18 related motor-operated valve." Where practicable, where you 19 can't do the testing at full DP in the plant. We've talked 20 about prototypes or another MOV in the plant as long as you k 21 justify its application of that data.

]

22 The second question was "Does NRC need benefit )

i 23 and diagnostic testing?" Absolutely. We've talked about 24 that quite a bit this morning.

25 The third question, Dick was going to take.

Heritage Reporting Corporation (202) 628-4888

o V :s

'138-l' MR. KIESSEL: "What methods for calculating 2 required l thrust,for a given valve 1 application are acceptable 3 to the NRC?"

.4. There seem to be too many variations to 5 determining thrust industry-wide.- The simplest _ answer'is' 6 only those that are justified by full l design basis testing.

7- MR. JACOBSON: The fourth question, "In. cases 8 where differential pressure tests cannot be. performed on a 9' given-valve, what kind of documentation will be required?"

10 We talked about you need to have everything 11 documented and validated and so forth.

12 " Willia 50.59 safety evaluation be required?"

13 I don't think that one's come up before. The 14 answer to that~is that 50.59 evaluations are usually.done at 15 a utility to ensure that there'isn't any unresolved safety 16 question or change of the design basis. The real 17 requirement there is that if there is a change to the design 18 basis or an unresolved safety question that you have to 19 inform us beforehand. There really is no requirement that 20 we have that these 50.59 evaluations be done for any given 21 level of change or modification. You have to assure 22 yourselves that that change or modification is not causing 23 an unreviewed safety gbestion. Different plants have 24 different says about doing that.

25 Normally they're called 50.59 evaluations. You Heritage Reporting Corporation (202) 628-4888

.,____-_a_ ,u.-- - _ _ . . ~ - . _ _ - -

e 139 1 ask a bunch of questions and determine whether these changes 2 are causing a change to your design basis or an unreviewed 3 safety question. So I guess it would depend on your own 4 . plant and your own plant's procedures as to when you would 5 do this particular type of evaluation.

6 MR. KIESSEL: The fifth question we've already 7 addressed briefly. That is that in doing the in situ 8 testing we are not asking that you place the plant in any 9 condition that is not safe or that places you outside your 10 tech' spec.

11 MR. SCARBROUGH: The sixth question was "After 12 initial differential pressure testing or alternate testing, 13 how often should testing be repeated?"

14 The generic letter has two points in this regard.

15 First, verification of switch settings every five years.

16 You're allowed to use diagnostics in that area at something 17 less than full DP if that provides you enough confidence 18 that the switch settings are correct. The other item is 19 testing after maintenance. You'll have to make a decision 20 on how high DP needs to be risen in order to test the valve, 21 to ensure that it's still operable.

22 MR. KIESSEL: Also you should note that we also 23 provide a caveat and an encouragement for doing predictive 24 and preventive maintenance on it, that you can request to 25 base your cycle on your own trending data that has shown Heritage Reporting Corporation (202) 628-4888

Y .

h L 140 1

I 1 that you can go to a greater interval.

2 MR. JACOBSON: The seventh question, "Should 3 design basis review and documentation be performed 4 concurrently, or should all design basis activities be 1

5 completed prior to initiation of-testing activities?"

6 I think our position on this is we really don't 7 care what order you do the steps of the generic letter. The 8 generic letter gives you a five year time frame to complete 9 all the steps. Whatever best would suit the way you're 10 going to attack this problem would be the way you're going 11 to go about it.

12 MR. MARSH: Of course the program has to be 13 developed within a one year time frame, whereas the testing 14 is to be complete within the five year time frame. So in 15 that sense there is a sub-time scale that's laid out.

16 MR. SULLIVAN: The eighth question says, "Should 17 valves which were previously manually operated and have been 18 modified to be motor-operated as a result of TMI lessons 19 learned be tested under full flow pressure conditions solely 20 due to the valve's function, that is safety injunction, pump 21 discharging valves?"

22 If these valves fall under the generic letter, 23 that is they're safety-related or position changeable as 24 defined in the letter, and come under the program defined in 25 the letter, then it doesn't matter what they used to be.

Heritage Reporting Corporation (202) 628-4888 j 1

_ _ _ _ _ _ _ . _ _ _ _ _ . _ _ _ _ . ._ ._ J

141' l

1 They!would be covered by this program.

2- MR. MARSH: In cases like.that valve, that may be l

3 something that you can rack outLpower and put aglocal lock 4 .on or something so you don't have to have it be position ,.

5 changeable, that particular valve.-

6 MR. SCARBROUGH: That was it for the Consolidated \

7 ' Edison questions. Is there anything we can expand onJin-8 those answers?

L" 9 (No response)

. 10 MR.'SCARBROUGH: The'next set came from' Alabama 11 Power, Michael?Edson. Is he here?'

12 VOICE: I'm here. I didn't-write them.

13 MR. SCARBROUGH: The first one concerned position

' 14 changeable, and we've addressed that.

15 MR. MARSH: Before we.go through the questions, 16 can-we just ask you, you've got them there in. front of you',

17 right?

18 VOICE: Yes.

19 MR. MARSH: Is there one you'd like us to go 20 through, or do you feel like we've answered them to your 21 satisfaction?

22 VOICE: Two through five need to be answered.

23 One's been answered.

24 MR. KIESSEL: Question two reads, " Specific to 25 Item H under recommended actions, a proper and responsive Heritage Reporting Corporation (202) 628-4888 L,

____.__._._________._______m.. _ ___._ _ _ I

L ,

142 1 predictive'and preventive maintenance program should include 2' an ongoing process of failure evaluation, evaluation of 3 corrective actions, and an evaluation of any actions taken 4 to prevent subsequent failures. If.thisl process is in 5 place, why then are the licensees required to single out MOV 6 failures for review on a two year or every refueling outage L

7 basis? Does not this create another unnecessary 8 administrative burden?"

9 My answer to that is no,.it'doesn't. That should

~

10' lx3 part of your proper and responsive predictive and 11- preventive maintenance program. How else can you do a 12 proper predictive maintenance' program if you're not looking, 13 for your. failures and. reviewing them?

14 MR. KIESSEL: Question three. "To date, the 15 majority of Mov failures have occurred due to improperly set 16 switches rather than improperly chosen settings. With this 17 in mind, why then are licensees asked to undertake what 18 could be conceived as an R&D process to determine and 19 validate switch settings considering the.large uncertainties 20 that exist in both the factors that affect thrust 21 requirements and the accuracy of the existing diagnostic 22 test equipment systems?"

23 5he basic answer to that is we want to make sure 24 that these motor-operated valves are operable and that they 25 will perform their intended functions.when called upon. We Heritage Reporting Corporation (202) 628-4888 1

143 l' want you to do that to the best of your ability at this 2 time.

3 MR. SCARBROUGH: The next question was, "With 4 respect to Items D and J, is the staff taking the position 5 that'MOV operability must be substantiated by diagnostic l 6 testing and trending rather than Section 11 testing only? j i

7 Please justify your response." q 8 I have an answer to that, and maybe I'll say it, 9 and maybe you can clarify the foundation for the question.

10 Generic Letter 89-10 does apply initial test to ensure that 11 MOV will perform its function under the design basis 12 conditions. The reason for that is what we talked about 13 this morning. The results of the Idaho testing and the 14 plant experience.

15 Is there some other facet to that question that l

16 I'm not perceiving?

17 (No response) 18 MR. KIESSEL: "With respect to Item J, what is 19 the basis for a five year or a three refueling outage 20 surveillance interval?"

21 We picked the five years, three refueling cycles 22 because that's also the initial interval, figuring that it 23 should at least be repeated that frequently. However, as I 24 indicated earlier, we're also encouraging industry to l

l 25 develop adequate trending programs and we would entertain Heritage Reporting Corporation (202) 628-4888 f

. 4?

i 144 ,

b x 1 requests-for changing of-that.scheduleLbased on the data 'l

'2 that you acquire.

l.

l 3 MR.EMARSH: I want to clarify an Indian Point Two 4 response we made a minute ago on 5 0.59_ evaluations. The' 5 -question was, "Will 50.59 safety evaluations be required?"

l 6 I was reminded that there'is language in 50.59 l

7 that says that any new test procedures that you may want to j i

8 do have to be subjected'to a:50~59 evaluation. . So if in the i

9- course of doing 89-10 or'85-03 or anything you.have a new '

10L test 1 procedure and new tests- have to lxa done_ at the plant,

~

11 the regulations would probably require you to do a 50.59 12 evaluation. J 13 So we are not imposing anything beyond the 14 regulations, let me say that. The regulations say when you.

15 have to do a 50.59. And if you have to do one because of 16- -the regulation, then so do that.

17 VOICE: We have a subset of MOV's that are 18 safety-related, that are included in Section 11 testing 19 progra:as . .They're also included in the technical  ;

20 specifications. Those are the MOV's that we have to assess 21 or determine operability by and also make deportability 22 another consideration. 89-10 implies a requirement for 23 determining MOV operability using a criteria, which is 24 really undefined today, that is beyond the tech spec and 25 Section 11 requirements which are normally pretty much the Heritage Reporting Corporation (202) 628-4888 iA-l $ i i ' - . _ _ _ _ _ _

h .

l 145 l-l 1 same.

i.

l-l 2 Given that this methodology for determining and L

3' evaluating the adequacy of MOV switch settings is l 4 established under your program for 89-10, is it expected L 5 'that the MOV operability would then be determined using that 6 criteria?

7 Today I'm using the stroke time. After you pass j 8 the stroke time it's okay.

9 MR. MARSH: You've also got super-imposed on your 10 stroke time testing the general statement about equipment 11 operability in the technical specifications, which means 12 that a component or system has to be able to perform its 13 ' intended safety function. Without looking at the specific 14 surveillance requirements, you've always got that 15 requirement that's implicit in the technical specifications, 16 right? It's there up in the front in terms of definition of 17 operability. That means the system component and all its l 18 associated equipment has to be able to perform its intended 1

19 safety function.

20 VOICE: Right, and by definition, Section 11 l

21 clarifies what that means for motor-operated valves.

22 MR. MARSH: That's certainly one part of it. But 23 there's many processes at the plant that you have to rely 24 upon. For example, qualification programs and other testing 25 programs to show that equipment is operable. If you perform Heritage Reporting Corporation (202) 628-4888 m _ _ _ _ _ __. - - . . _ - - - -

N' ,

146 1 'a. design review of a piece of equipment, that's.not required h 2 by a tech spec per se,.but l'f.you perform.a. design' review 3 and you find'out the' system is not able.to perform its' 4 safety function, it's not. operable by definition. Butiit 5 wasn't an IST. requirement or a Section 11 requirement, or 6 some other particular part of the tech spec that you had to 7 go beyond.

8 Anyway, the tech specs define operability. This 9 generic ~ letter.says here's one way to determine operability 10 of th'e valves themselves. If you do this way and it passes 11 all the tests, then'it's going to be an operable valve. .It 12 doesn't prescribe specific ways that you show that a valve

-13 'is operable or not, but it does give the general guidelines 14 and'the general concerns that'you have to address.

15 MR. BROWN Perhaps this is e. good time to pass 16 on information of an LER that just came inLwhere the plant 17 was voluntarily expanding the Bulletin 85-03 to the rest of 18 safety-related valves. Three. years after they did some work 19- on the valve they found out that valve was inoperable under

20. the. conditions that it would be~ called upon to operate. It 21 tripped out on torque switch-22 percent open.

22 What they had done was a simple thing of changing 23 the_ packing. They ran'a stroke timing test, everything 24 passed. Three years later when they were doing some more 25 work, they drew the conclusion that it was inoperable.

Heritage Reporting Corporation (202) 628-4888

\

4 147 1 This shows why it is so'important to understand 2 why something is inoperable. d 3 VOICE: Then if we determine that this MOV is 4 inoperable using the new criterion, then we would also have 5 to assess it for deportability?

6 MR. MARSH: That's right. Absolutely. The 7 regulation is still in' place. It doesn't remove 50-72 or 8 50-73 or any of those requirements.

9 VOICE: What about the subset of MOV's that are 10 not included in tech specs and not included in the in-11 service test program? Would we then be required to report 12 those failures?

13 MR. MARSH: Let me understand your scope. This 14 is not in IST?

15 VOICE: Right.

16 MR. MARSH: This is not in 89-107 17 VOICE: No, it's in 89-10 because it's safety-18 related or it's position changeable, but it's not covered by 19 a tech spec requirement.

20 MR. MARSH: So you're saying suppose there is a i

21 position changeable valve or a non-safety-related position 22 changeable valve that's not in the technical specifications, 23 it's not in the IST program, and you find out that once you  ;

l l 24 do an evaluation or a test that it was not able to perform

)

25 its function.

Heritage Reporting Corporation I (202) 628-4888

m 148 1 VOICE: Correct.

x

.2 MR. MARSH: Is there a problem in terms of 3 deportability? Operability of the system.

4 Probably so. But I'm not sure. It really.

5  : depends upon the operator action part. If this is something 6_ that an operator could accidentally disposition and in so 7 doing defeat the safety function of the system, I would' tend 8 to say yes, .that's a reportable item under 50-72, but I need 9 to look at 50-72, 50-73 to see what the words are.

10 I'm not-certain of the answer to your question.

11 VOICE: Thank you.

12 MR. SCARBROUGH: That's it for Alabama Power.

13- The next set came from General Physics Corporation. Steve

14. Wochek, is Steve here?

15 Have we answered your questions? A lot of them 16 concern data bases and things of that nature." Do you want

~

17 us to run through them?

18 VOICE: If you can.

19 MR. KIESSEL: The first question, "In response to 20 the Generic Letter 89-10 requirements for DP testing, what.

21 established informational data. bases have been accepted by.

22 the NRC for extrapolating past valve data?"

23 Limiting just to Generic Letter 89-10, none have.

24 No one has approached us with any. But what would we 25 accept? It would be data bases that contain testing Heritage Reporting Corporation (202) 628-4888 L

149 1 conditions that envelope the conditions under which you want 2 to use the valve.

3 MR. JACOBSON: The second question, "What 4 essential parameters would be required within a data base 5 for reliable data extrapolation?"

6 I'm not sure that I understand that question.

7 Could you expand on that?

8 VOICE: Approach the friction factor --

9 MR. JACOBSON: Basically we want to ensure that 10 any ' factor that could affect the performance of the valve is 11 accounted for in the data base, whatever those factors may 12 be.

13 MR. KIESSEL: Third question, "Will the NRC allow 14 the same DP testing methodology used by the Calloway Plant 15 in response to their 85-03 commitments for meeting the 16 intent of Generic Letter 89-10?"

17 The simple, short and sweet answer there is no.

18 We're going to be looking for additional full design basis 19 testing.

20 MR. MARSH: Will you please tell the audience 21 what that test was we found was not acceptable, so they'll 22 know what we're really answering?

23 MR. KIESSEL: You mean what the Calloway Plant 24 was that was okay for 85-03 but not okay for 89-10.

25 MR. MARSH: Yes.

Heritage Reporting Corporation (202) 628-4888  !

l l 150 1 MR. KIESSEL: The Calloway Plant basically had.

2 three criteria by which they could' extend data. The first l

3 one was if they went'and did either an in-plant or l

4 prototypical test at the maximum design conditions. They L 5 could then apply that test data directly to an identical l

6 valve in plant. Right now we're saying we're not quite sure 7 that identical is as identical as we were assuming under 85-8 03.

9' The second criteria said if I did testing on four 10 identical valves and showed that they performed in a similar 11 manner, could I then extrapolate data? In other words I got 12 tests at 200 pounds, 300 pounds, 400 pounds, and 500 pounds 13 out of four different valves. They all trended beautifully.

14 Can I extrapolate that out to 1000 pounds?. The answer is 15 now no, because we're not sure how far you can extrapolate 16 that data. We need more test information at those higher

-17 pressure ratings in order to be able to take and provide for 18 an extrapolation.

19 The third method was can I take 20 similar 20 valves. Now what were similar valves? It was something 21 more than just defining it as gate valves. We were hoping 22 we were getting down into individual manufacturers of 23 approximately the same' size, a four inch and a six inch, 24 that type of thing. But could I take the data from 20 25 eimilar valves and then apply that to the valve in-plant, i

)

Heritage Reporting Corporation (202) 628-4888

..$.. \. .151.

l [ Asf you've' heard,: right .now for the purposes of the l generic.

2 ' letter we're1 backing away from this similar concept unless n .

3. you have sufficient. data to prove.that over the entire 4 . pressure ranges that we're. talking about that'the valves-5 really do function in a similar manner.

6- Question four, "When will the NRC allow'a data 7' base; approach in lieu of actual MOV testing?" .

l 8 Again, I think we've answered this many times 9 coming atLit from many directions. But primarily, we're 10 .look'ing for.that data base to be supported by test date at 11 the design' conditions that envelope the conditions that you R'

12 want to use the valve at. Just'because a valve was not 13 tested at the most severe conditions does not prevent you-14 from using the data for some less severe condition. But we 15 ~want the: test data to at least envelope your design basis 16- conditions.

17 Did I answer your questions?

18 MR. SCARBROUGH: Thank you. The next set of 19 questions came from the Nuclear ~ Utility Backfitting Reform 20 Group, NUBARG. I won't read the whole letter, I'll try to 21 paraphrase the questions.

l 22 The first one has several parts to it. It starts 23 out, "Certain Generic Letter 89-10 recommendations appear to 24 require several licensee actions that may be either one, a 25 change in NRC position; two, involve new plant procedure and Heritage Reporting Corporation (202) 628-4888

i  :

e.

152.

1- . program development requirements; or three, require a-2 modification of the plant licensing. basis more specifically.

-3 '.than Paragraph A."

.4 "The staff has previously approved most 5 licensees' methodology for complying with 10 CFR 50.55AG in-

6. ' service testing requirements. This. methodology utilized 7 ASME Section 11 testing as thefbasis for. ensuring MOV 8 operability under design, basis conditions.

9 "Since the staff has determined in Generic Letter 10 89-ID that Section 11 is now inadequate to verify MOV 11 operability, should this provision of Generic Letter 89 L 12 be considered a backfit? If not, please state the staff's 13 basis for its conclusion."

14 Our. position there is yes, the generic letter 15 could be considered a backfit in certain respects. That is-16 why a backfit analysis was prepared, it was sent to the PDR.

r 17 It seems to.have been lost in the PDR. We'll send another 18 copy down there. This generic letter went through the full

'19 NRC review process. It went to the committee to review 20 generic requirements. It's gone through all the process for 21 reviewing a generic backfit.

22 The second paragraph here, let me see if I can 23 paraphrase it. The preliminary test results indicate that 24 some MOV's may be subjected to mechanisms and loads that 25 were not accounted for previously. Notwithstanding the Heritage Reporting Corporation (202) 628-4888 1

~ _ . - _ _. _ - _ _ _ _ - - _ _ _. _ _ _ _ - __. _ - ____ _ _ _ _ ___ _

153 1 staff's views regarding the present necessity to consider 2 these loads should this change in licensing - , that again 3- falls under the whole generic letter as reviewed through the I

4 backfit process. i 5 Item C is the generic letter strongly recommends I

6 that the scope of the suggested MOV programs include all j 7 safety-related MOV's as'well as position changeable MOV's.

8 The generic letter also states that no change in existing q 9 plant design basis is intended and none should be inferred.

10 Position changeable, that's where this question is going. 1 11 Accordingly, would these facilities be exempt 12 from considering this category of MOV? Those facilities 13 where the FSAR did not address position changeable.

14 They're not exempt from the generic letter.

15 Mispositioning has been specifically considered by the NRC 16 as a potentially beyond design basis and it's within the i 17 scope of the generic letter.

l 18 MR. MARSH: As it was in 85-03 Supplement I, too.

19 That was pretty clear. As 6 matter of fact, the language in 20 85-03 Supplement I is probably a little clearer than 89-10 21 about position changeable. We are going beyond and why 1

22 we're going beyond. And it was proposed, reviewed, and l 23 approved in the same way. Just as an 89-10 position 1

24 changeable was done. It was an acknowledgement that it may 25 not have been a explicit consideration in licensees and we Heritage Reporting Corporation (202) 628-4888 l

L l

w_______ . _ _ - - _ .

p < , .

I' 154--

1: were going.out beyond where we needed to..

l 2- MR. SCARBROUGH: D was, the staff states that as N 31 a minimumLNRC envisions that as part of.a good maintenance 4  : program, other MOV's 'in the balance of plant -should be~

h 5  : considered for inclusion in the program. Commensurate'with 6- the licensee's assessment of their importance to safety.

~

7- -The staff appears to suggest,that BOP components.

8 be included in the MOV program in order for the staff to 9 consider licensee efforts' complete. If the staff-insists on

- 10 inclusion of BOP systems that were not included in the plant 11 licensing basis and which arguably could be outside NRC's 12 regulatory concern and jurisdiction, please explain why this 13 would not be a backfit.

14: MR.' MARSH: That wouldn't only be a backfit, that 15 would.be going beyond the regulations. That's a different 16 beast: when you go beyond the regulations. .Then you cay say, 17 as you pointed out, that you're beyond your authority 18 . currently.

. 19 1 know you've got an answer but I've got to chime 20 in here. We are not requiring the balance of plant 21 equipment to be included in the program. We are stating 22 that we do think it's good maintenance policy to have those 23 . valves in the program, and it is just that. But we're not 24 judging through this generic letter of maintenance programs,

' 25 we're not requiring BOP MOV's to be in the program. And Heritage Reporting Corporation (202) 628-4888 P

y h ,

1 f y '-

i k 155-i h

$ l' please: don't.think we are insisting that these components be d

'2 in the program. We're.not. '_It was only'a suggestion for a d 3 good: maintenance program.

l 4 As you say,.if we did' insist that, we would be 5 beyond the current regulatory basis.

L

~

6 MR. SCARBROUGH:' I'll just add'one little point.

7 The generic letter does not set any boundary on what's 8 within or outside the NRC's regulatory concern or 9 jurisdiction. The regulations do that-themselves. That's

10 -anotiier issue.

11 Paragraph E, the generic letter states that

~12 whether:the switch settings are changed or not, the MOV 13 should be' demonstrated to be' operable by testing a design 14 basis = differential pressure interflow determined in response '

415- to, Item 1. An explanation should be documented for any 16 cases where testing with a design basis differential li pressure-interflow cannot practically be performed.

18' This recommendation, if de facto required by NRC, 19 should be considered a backfit in that it requires 20 additional testing previously not required to satisfy 21' regulations.

22 Once again, this is along the same line. The 23- generic letter has gonb through the review as a generic 24 document provided by the NRC. All portions of it were 25 reviewed for potential backfit.

Heritage Reporting Corporation (202) 628-4888

--_____._--m_____-__ - _ . _ _

+ ..

156 5.x 1 MR. MARSH: 'It is a change in'the way we.have 2L . judged 1the adequacies of switch settings and judged the i: '

i <

3 operability of equipment before. It is a change and it is a

-4 backfit. It is clear.- That's.the reason we prepared, I

.5. think when'it'went through the deliberative kind of process 6 people were-unsure whether this particular part was a 7 backfit,.or this particular part was not a backfit. It.was 8 difficult to draw the line what'part was.or was not. So 9' they said the.whole generic. letter is a backfit. Call it 1 10 that'way. Do a cost benefit, justify it, and let's get on 11 with the business.

12 So when you go through the PDR records, you

13. probably won't find'a piece by piece-by piece description of

14. what part-is, what part isn't. It's.just a blanket cost-15 benefit analysis to justify the backfit.

16 MR. .ROTHBERG: Is there somebody'here .from 17 NUBARG7

'18 VOICE: Yes.

19 MR. ROTHBERG: 'Maybe we're missing something 20 about the business of backfit. I read this letter and the 21 thrust is admit that it's a backfit. Okay, complete the 22 sentence. Then what?

23 VOICE: Until now at the ACRS meetings no one 24 said it.was a backfit. That's in the transcript. So we

-25 didn't know. The backfit analysis was not in the PDR, S

Heritage Reporting Corporation (202) 628-4888 i

i 157 .l 1 therefore, we didn't know. No one wanted to go on record to 2 say that. That's why we asked the questions.

3 MR. ROTHBERG: I referenced the placement of the, 4 I gave them the wrong reference here, but I referenced the 5 placement of the reg analysis and backfit analysis at one of

.6 the ACRS meetings.

7 VOICE: Right, you did, but it wasn't in the PDR 8 so we couldn't read it.

9 MR. ROTHBERG: And you asked for that accession 10 number and they didn't have it?

11 VOICE: Yes.

12 MR. MARSH: We'll make sure you get it. That's 13 the important part.

14 VOICE: I guess what we're getting at partially 15 is on a plant specific basis we'll have to analyze this on 16 plant specific backfit parameters, so we just want to know 17 the different categories, whether you view those individual 18 items as backfits, and we'll have to address that 19 separately, plant by plant.

20 The second emphasis is, if a generic letter is 21 only a recommendation, why do a backfit analysis which is an 22 imposition? We sat that as a dichotomy. We wanted some 23 clarification in that regard.

24 MR. ROTHBERG: The reason we did the analysis 25 that we did is our management demanded that we cover all the Heritage Reporting Corporation (202) 628-4888

t 158

1. bases, that-we approach this thing as if the recommendations 2 that we were making were backfits. Now I don't think that 3 demanding that licensees know that their motor-operated 4 valves function in the plant is a backfit. That's my 5 personal opinion.

6 MR. MARSH: Tom's questioning the vehicle by 7 which.we' transmitted the recommendations,~if you will.

1 8 MR. ROTHBERG: But our management demands that'we 9 do these things, and we did them.

L 10' MR. MARSH: Rather than saying it that way,-let's 11 say it like this. We are changing the basis on which we 12 judge the adequacy of valves. You can see that in the whole 13 thrust of the letter. Regardless of the thrust of the 14 letter, that's sort of evident, that we are questioning the l~ 15 basis under which these pieces of equipment were judged to 16 be. adequate and are judged to be adequate. So there is a L

17 change in regulatory philosophy. There is a big change.

18 The 50.54 F letter,-as you point out, only says 19- you're required to respond. Under the provisions of the 20 regulation you're required to respond to these particular 21 elements. It's trying to say this is how we're going to

- a 22 judge the adequacy of your motor-operated valves and you're 23 required to respond to that.

24 So that's sort of the transmittal of the 25 position. This is how we're going to do it. This is what Heritage Reporting Corporation (202) 628-4888 i

)

i 1

m ,7 -

q l

159 1 'we think the-good program is.- You've got to tell us whether 4 2 'you, under the' 50.54 : language whetherL your language should. ,

l in 3- lxa revoked, modified, or suspended because-of this'new 4 position.

'S I know this is a legal issue and it really talks 6 -about how you transmit generic requirements.. You've got 7- other positions in here that question the' process by.which 8 this was done. All I can say, Tom, is this is the road that we're on.

~

9 There probably could have been other roads. I'm

-10 not sure they could have been as quick as this, and we did 11 want to make sure that we got on with this business of 12- . making sure valves were operable. If we went the regulation 13 route, as you've heard is another route that's still being 14 pursued, we're still looking at regulation changes for 15_ . motor-operated valve operability, I'm not sure where we'd be 16 in_ terms of~ plant safety. I'm really not. I'm not'sure how 17 long we'd take in order to get that part of the process in 18 place.

19 .This was a vehicle, a legal vehicle, and that's 20 the reason _we took it the way we did.

21 VOICE: I think you may understand our concern.

22 Depending on the vehicle that you use, it gives licensees 23 different options in responding. A rulemaking gives you a 24 notice and comment period, which this arguably, depending on i

25 who you talk to, could be a form of generic imposition on Heritage Reporting Corporation (202) 628-4888

_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - - _ .i

n.

160 E 1- all licensees.

2 MR. MARSH: But we did give a chance'for public p 3 comment,.we did go through tho-ACRS process as well. So if:

H '

4 public commenti.s the concern, that concern has already been

'5 addressed.

6 VOICE: The other issues,-such as after 1;

L 7 commitments are~made, the enforcement, I think the question 8 was asked earlier, how you enforce. That changes depending 9 on how the requirement / recommendation is implemented. So

'10 'thats~why-we need clarification, so that we know'what 11' widgets we're playing with here.

12 MR. MARSH: Tom, would you like us to go through 13 the rest of these? They're the same thrust.

14 VOICE: (Inaudible) 15 MR. SCARBROUGH: Number'2A, the generic letter 16 .st'ates in the reporting requirements that "Each licensee 17 shall advise the NRC in writing within six months of the 18 date of the letter that the above criteria and 19 recommendations will be met." In addition, licensees are 20 required to justify why any schedule or date cannot be met.

21 This statement, in addition to.other issues previously 22 raised, transforms all of the generic letter recommendations 23 into requirements and accordingly a backfit.

24 NUBARG believes that if this generic letter 25 contains only recommendations, why are licensees required to Heritage Reporting Corporation (202) 628-4888

161 1 respond with an acceptance of the generic letter provisions 2 or propose equivalent alternatives? If in fact this generic 3 letter were just a 10 CFR 50.54F request for information, 4 the imposition of recommendations would not be necessary.

5 MR. MARSH: No, if you put out a generic letter 6 under 50.54F and you say you are required to address this 7 new acceptance criteria that the staff is promulgating, 8 that's a backfit because you are stating this is our new 9 position. So in that it's a new position, it's a backfit.

10 What licensees are required to do under 50.54F is 11 to address those items. By regulation they're required to 12 under 50.54F. They have to respond. That is the extent of 13 their legal requirement.

14 VOICE: The underlying basis for that question is 15 50.54F does not typically allow the staff to implement new 16 positions or new requirements, and that's what we saw 17 happening here.

18 MR. MARSH: Right.

19 VOICE: You do that via another mechanism.

20 MR. MARSH: Right.

21 VOICE: We think that may have been improper in 22 this case.

23 MR. MARSH: Okay. We hear you.

24 MR. SCARBROUGH: That was it for the NUBARG 25 questions. The next set is Grand Gulf. Is Grand Gulf here?

Heritage Reporting Corporation (202) 628-4888

o i

1 i

~,

V ..

.162

, , .li VOICE: Address 1four and five, please.

1 22 MR. SCARBROUGH: Number four, "Some plants with

'3 later designs 1have an extensive scope for safety-related I

~

l4 valve MOV's up to 300 valves.- This can be a significant

.5 burden on the plant outage schedules 'in order to implement 6' the generic letter actions within the next three outages or 7- five years. Even though the industry. recognizes the need to

8. correct MOV switch. settings and verify operation under 9' design conditions, the proposed generic letter' schedule may

. 10 not be reasonable for plants having numerous MOV's. Is the 11 NRC willing to entertain MOV test schedules where certain 12 MOV's are prioritized for initial'five year testing-and 13 extend the test period beyond five years for MOV's having a 14 lesser safety function and importance?"

15 I think we mentioned this before. But where a 16 licensee has an aggressive program to test as many of the 17 MOV's as it.can and to develop the data for the other MOV's, 18 the NRC is going to give full consideration to a request for 19 extension of a schedule, but that will be made on a case by 20 case basis.

21 The next question was, "Considering --

22 MR. KIESSEL: Tom, let me just throw in that your 23 reference to the fact Of trying to prioritize the valves 24 would have a significant impact on that decision also. So 25 you've thrown two goodies at us and they both work together.

Heritage Reporting Corporation (202) 628-4888 l

- _ _ _ = _ _ _ _ - _ _ _ _ _ _ _ _ _

163 1 :Give'me an~ aggressive: program and a program that addresses 2- .the' valves-which you consider are theLgreatest safety 3 significance to you'first, and it.gets more favorable 4 ' attention.

5 MR. SCARBROUGH That's a good point.

6 .The next question was, "Considering that more

.7 detailed information and direction usually result from the 8 conduct of NRCLworkshops, for example Generic Letter 88-20 9 .and Generic Letter 89-04, is NRC willing to consider the l

10 response-date to begin on the date that Generic Letter'89-10 11 workshop clarifications are issued?"

12 The short answer is no. The transcripts will be 13 available.soon, in a couple of weeks. There's a' lot of

14. things you can already be doing, things that are not 15 controversial. The design basis review, trying to determine 16 which MOV's you can test and which ones you can't test in 17 the plant. There's a lot of work that can already be done.

18 Once individual plants come in and talk about schedules and 19 things of that nature, we'll address it then. But I think 20 we've thought a lot about the five year program, originally 21: it was three years. We extended it to five because of some e

22 of these concerns we talked about today, and we think the 23 schedule is doable and'so we're sticking with what we have.

24 That was it for Grand Gulf. Should we expand on 25 any of that? -

Heritage Reporting Corporation (202) 628-4888 L

l l

l I

- - - - -_- a

164 ,

I l 1 (No response) L 2 MR. SCARBROUGH: The next set came from Florida 3 Power Corporation. Do we have people from Florida' Power ,

here?

4 l

5 The first one Jeff was going to take.

6 MR. JACOBSON: Do you want us to answer all the 7 questions?

8 VOICE: The one about the commitment to 85-03 and 9 -- MR. JACOBSON: That's the only one you want 10 answered? We don't see that one. Do you want to stand up 11 and ask that one specifically?

12 VOICE: Glenn Vaughn, Florida Power Corporation.

13 We've made some commitments to 85-03 Bulletin as far as our 14 trending, and we wanted to know if that commitment still 15 stood as we originally stated. I know 89-10 says it 16 superceded 17 85-03 commitments.

l 18 MR. MARSH: We've talked some about-that.

l 19 VOICE: I believe we have. I just wanted to know 20 if you had anything else to elaborate on it.

21 MR. MARSH: Give me the specific commitment you 22 feel like --

23 VOICE: We were trending our valves that we 24 tested for the 85-03 commitment, we have tested all those l 25 and we were going to trend those over the next three l

Heritage Reporting Corporation (202) 628-4888 l

l

'165 1 refueling outages.- Since.we are testing those. valves.anyway

2' 'onethe:five year' frequency that_the 89-10' letter states, we 3 ,were wanting to know if we still-have to continue that jlL , trending.

5 MR. KIESSEL: By trending you'mean doing a full 6 [ diagnostic test on them?

7> VOICE: Yes.

8 MR. KIESSEL: No, I'd view that the five year, 9 three refueling cycle would in essence re31 eve.you of that.

10 VOICE: Thank you.

11 MR. ROSS: On 85-03 commitments, let me_ elaborate

12. a little bit. In.the spirit of 89-10, any changes to any' 13 schedule or commitments that you've made to 89-10 you're 14 supposed to notify us of those schedule and commitment 15' _ changes.-

'16 : .The same should apply to 85-03. If we have an 17 existing. formal commitment to some aspect of 85-03 that's 18 going to change in-the context of 89-10, send us a letter

.19 - and notify us of that change. That's allowed within the 20 context of 89-10.

21 MR. SCARBROUGH: That was it for Florida Power 22 Corporation.

23 This is the first set we received from Florida 24 Power Corporation. _It looks like this. Are you familiar 25 with this set that came in? There was no name on it.

Heritage Reporting Corporation (202) 628-4888

g ,

y '

166-m

< 'l VOICE: (Inaudible) 2 -- MR.- SCARBROUGH: I think we've addressed those

'3 basically, anyway~ .

4 Why don't we go on-and we_can come back.

5 The next set is Philadelphia Electric. They were 6- kind enough to mark off the ones that we've already touched 7 upon-today,-and I'll jump right to the ones that we haven't 8' -yet.

9 Number four, Dick was going to take that one.

10: MR. KIESSEL:: What is. meant by "This is 11 particularly important if the torque or torque bypass: switch 12 setting has been significantly raised above that required" 13 .concerning monitoring MOV performance to ensure correct 14 switch settings in Section D? Is this inferring that high 15 switch settings are not correct? Why does the importance 16 increase if the switch settings are high?

17 We're reviewing that. If you had to make a 18 radical change to your switch settings during your post-l L '19 maintenance'or lifetime test cycling on these valves, it 20 probably implied there had been a gross change in the 21 characteristics of that valve which would be indicative of 22 either having to do maintenance or that maintenance had been L 23 done improperly, and therefore questioning the availability 24 of margin to still be able to perform a safety function when 25 it is highly loaded during the design basis event.

Heritage Reporting Corporation (202) 628-4888

167-1 Did that answer your question?

2 MR. JACOBSON: A high switch setting alone is not 3 a specific reason for concern. We're talking.about settings 4 that have been raised.

5 MR. KIESSEL: We're talking about the manner of 6 change. Suddenly doubling the switch setting in my mind 7 would imply that you've suddenly found a need for doubling 8 your allowable thrust.

9 MR. ROTHBERG: The next one reads, "In Section l

10 D," of the generic letter, I assume, "it states, 'The ASME 11 code Section 11 stroke timing test required by 10 CFR Part 12 50 is not oriented toward verification of switch settings.'

13 What is the basis for this statement? The Section 11 test 14 parameter (valve stroke time) is extremely sensitive and l 15 affected by switch. settings."

16 That's probably true. The Section 11 stroke 17 timing test is a no load test. It does not give you a 18 satisfactory assurance of operability at design basis 19 conditions. We talked about this INE information in 89-61 20 and I think that's a good example of the problem.

21 MR. KIESSEL: Next question, "Please identify 22 ' pertinent design and installation criteria' that the NRC 23 considers significant hnd should be included in the design 24 basis review reference to Section E of the letter."

25 The items we were considering there was the Heritage Reporting Corporation (202) 623-4888 l

' ~~ ~ -

~

Qf @ + -

[YP ~168 r

,1 location of the-valve.- Is itiin a high temperature area, 2 high rad: area?. Orientation.of the valve. Is it horizontal, 3 vertical,. upside down?- Modes of operation. Is the valve'-

just stroked once and then forgotten about?. Or is the' valve L

'4 5; going lto'be used for maintaining vessel level soJit's going-g 6 to: be cycled several times and repeated. These all would 7; have an impact.on the' design basis and also the L 8 transferability'of prototypical testing over to the valve'in-9' the plant.

~

10. MR. JACOBSON: Question seven, "What is, meant by.

11' ' including the. capability of the MOV's power supply and 12- cables to provide the high initial current needed for 13 operation of the MOV in Section E of the generic _ letter?

14 Operation'of.an MOV normally, in at least one direction, L - 15 closing has a low initial current. Why is the concern 161 raised about initial current?"

17 I'think we're really not concerned about initial 18- current. We're concerned about any high currents that are L 19 required, whether that's in the initial part of the stroke 20 or the end of the stroke. We want to make sure that for 21 instance on your voltage calculations that you use the 22 locked rotor currents for figuring out the voltage drop on 23 the-cables to the MOV and so forth and so on. Whether 24 that's in the initial part of the stroke or the end of the 1

25 stroke really isn't that important.

Heritage Reporting Corporation (202) 628-4888 i

___-_m___---_-____-__________._____._-_____ - - . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ _ _ _ _ _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ - _ _

169 1 MR. KIESSEL: "What is meant by 'NPRDS 2 appropriately modified' in Section H? What type of 3 modifications are considered appropriate or are suggested?"

4 Let me put this into context. This was, in the 5- section where we were suggesting that industry get together 6 to collect the data both for the demonstration of initial 7 operability and also for the maintenance and trending type 8 of information that you'd need for doing your preventive 9 maintenance programs.

10 To the best of our knowledge, the only available 11 data. base on MOV's that was not tied up by a corporation was 12 the NPRDS data system. So therefore, that was the only one 13 we felt free to identify specifically as a potential place 14 where this type of information might be included. It was 15 not our intention to say that IMPO should modify the NPRDS 16 to include this information, but rather to provide a 17 guidance as to perhaps one place where it might be included 18 and in reality, we don't care where the data is. In fact 19 you don't have to have it. But we feel that's really the 20 only way you're going to get this program accomplished in 21 the time frame that's set up and be able to provide a 22 meaningful preventive maintenance data base in the future.

23 So no, I'm not going to tell you exactly how to 24 modify it because I don't know what kinds of information 25 you're going to need.

l Heritage Reporting Corporation l (202) 628-4888 1

L-- - -- - . . . _ . _ _ _ . _ _ _ _ _ _ _ _ . _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ .

s-170 1 Brian, I think you probably have a better feel 2 for it and probably could tell NPRDS how to do it, but we 3 can't.

4 Next. "In the background section of Generic 5 Letter 89-10 it states, 'The NRC staff's evaluation of the 6 data indicates that unless additional measures are taken, 7 failure of safety-related MOV's and position changeable 8 MOV's as defined under recommended actions of this generic 9 letter to operate under design basis conditions will occur 10 much'more often than have previously been estimated.'

11 Please provide the basis for this evaluation. Failure rates i 12 from 85-03 testing, or failure rates from valve worst case 13 operating conditions. Previous estimates of valve failure 14 rates were on the per demand basis including operation at 15 less than worst case conditions. What corrective factors or 16 considerations were taken into account in comparing 85-03 17 data to earlier data?

18 "In the second paragraph of the generic letter it 19 states, "ASME Section 11 stroke time testing provides some 20 measure of on demand reliability. If Section 11 testing 21 provides a measure of on demand reliability and on demand 22 reliability was the basis for previous estimates, why is the 23 85-03 data testing under worst case conditions being used as 24 a standard of comparison?"

25 The primary thrust of the generic letter is to y Heritage Reporting Corporation (202) 628-4888 i

I j

y -a h ,

171 1 , verify that the. valves willifunction under.the worst case 0 L2- -conditions. .The previous numbers,that have been used, L

3 numbers- in the order of magnitude of 10 to.the -3 for PRAs-4 :have not been born out by-the testing results-from 85-03.

p: 5 Now thoseenumbers, when they're used in the PRA, are not an

-6~ average, on demand reliability. .Those are worst case 71 conditions under-which those valves are assumed to function 8 in the PRAs. So'for that reason I think the 8.5 percent 9 failure rate that came out of the bulletin responses is more 10- appropriate than the original 10 to the -3 valve for 11 reliability.

12 Did I cover you?

'13' VOICE: Yes. Can you --

14 MR. KIESSEL: Paragraph 5 of the background 15 section states, "INEL has concluded that diagnostic systems 16 'that measure both static thrust and motor torque are best 17 suited for predicting valve motor performance." How does 18 valve stem thrust indicate or' affect motor performance?- The 19 motor provides torque for operation of the mechanical 20 equipment. It does not directly apply thrust, nor is it in 21 the valve thrust load path. Is this a typographical error 22 where valve motor should have been valve motor operator?

23 Yeah, I think we're talking semantical 24 differences. What INEL was getting at was that simply 25 measuring thrust may not provide you with the full picture, Heritage Reporting Corporation (202) 628-4888

V .

L 173 1 just as simply measuring torque may not provide you the 2 whole picture. Their recommendation was that a system based 3 both on measuring thrust and torque may be a superior 1

4 diagnostic piece of equipment.

l 5 MR. MARSH: I think as you pointed out in the 6 second part of that question, it probably was structured 7 towards the motor operator as opposed to the motor 8 performance itself.

9 MR. SCARBROUGH: The next one is number 18. I l 10 "Could valves equipped with 'special features' such as one, i 11 key lock control switches; two, valve out of position i

l 12 alarms / indicators; three, valve travel stop switches; or i l *

. 13 four, mousetrap control switch covers be excluded from 14 consideration as position changeable?"

15 Possibly. If a combination of these prevents all 16 inadvertent operation, then yes. There are various ways of I

17 accomplishing the provision of blocking inadvertent l l

18 operation and your up to develop any version that you see  !

19 fit.

20 Number 19, Owen was going to do that one.

21 MR. MARSH: I just want to give you the benefit 22 of past discussions, and that is if you remember this 23 morning we talked about an example of an acceptable 24 combination where a position changeable valve could be 25 excluded-from the program. It was locking out power from the Heritage Reporting Corporation (202) 628-4888 hm.__-___ . ___._

173'

1 valve' combined with a locking. device on-the valve hand' wheel' bf' I  ?

so that' local operation couldn't be possible'without taking_

2 k 3 an" aluminum tab or something off, a piece of wire, coupled-

4. with the valve being defeated electrically. So it is a 5 .. combination of some local preventive measures combined with; 6 electrical preventive measures.

7 VOICE: .It would seem then, that if-you had a 8 .- case where you had the valve out of position alarm, you-19 would also not need.a lock on the hand wheel.or on the stem 4' 10- or anything else because'you would go and have an alarm in 11 that case that'the' valve is out of position. And it would 12 seem then,Lthat simply blocking the valve' electrically might 13 be. acceptable.

14 MR.-JACOBSON: If that alarm was a safety-related 15- . alarm and you were sure that you always had it, that'may be 16 an acceptable alternative to the mechanical blocking that we 17 talked about originally.

18 VOICE: The'other. question that's buried in 19 -there, and the phrasing may not have been as clear as 20 possible, is we have a feature in our control circuits or 21 our valves where we're able to operate or interrupt 22 operation of the valve such that if the operators say the 23 valve was supposed to be closed or opened and he started 24 going closed, the alarm came on, where he realized that he 25 could stop it in mid-stroke and reverse it so the valve Heritage Reporting Corporation (202) 628-4888

174 i i

n 1 would not go full closed and would not have to recover with 2 full DP. Is that an acceptable means of dealing with this l 3 problem? l i

l 4 MR. JACOBSON: You're'trying to say that you're i L

i' 5 going to come up with some time frame that the operator is l

6 going to respond and then do your analysis based on that I

l' 7 time frame?

8 VOICE: No, I'm saying the operator made an 9' error. He goes and realizes he's made an error and can 10 correct it immediately. Rather than waiting for the valve 11 to go full closed. Using Davis-Besse as the example. Once 12 those valves started going closed, their operator couldn't 13 do anything. We have, in our control circuitry, the ability 14 to stop it there in mid-stroke and return it to its full 15 open position.

16 MR. KIESSEL: How would the operator know that he 17 had punched the wrong button?

18 VOICE: For one thing, he'd have his hand on it.

19 The other one is --

20 MR. KIESSEL: But he's already got the mindset 21 that he should have done that, so that's not going to tell 22 him.

23 VOICE: We also have our 0-listed alarm that 24 indicates the valve is out of position.

25 MR. KIESSEL: But aren't there going to be Heritage Reporting Corporation (202) 628-4888 l

175 1 additional alarms going off in the control room at the same 2 time?

3 VOICE: It's possible.

4 MR. KIESSEL: In response to the accident 5 scenario?

6 VOICE: I'm not going to dispute that. It's 7 possible.

8 MR. JACOBSON: You have alarms that tell in all 9 circumstances where valves are supposed to be?

10 VOICE: No, we have alarms that indicate that the 11 ~ valve is out of position.

12 MR. JACOBSON: What says what the right position 13 is? I mean different scenarios you need a valve in a 14 different position, right?

15 VOICE: I'd have to go back and look at the 16 logic. I know we do have it in some cases.

17 MR. JACOBSON: You're talking about very specific 18 cases and we're going to have to look at those on a case by 19 case basis, I think. In general we're trying to give a 20 flavor for how we expect these things to be approached.

21 There's always going to be some weird exception out there, 22 and we'll have to look at that on a case by case basis.

23 VOICE: Thahk you.

24 MR. KIESSEL: But I think conceptually you're 25 moving in a favorable direction.

Heritage Reporting Corporation (202) 628-4888

176 1 MR. MARSH: Let us get back to you on whether 2 that's a favorable way to go or not. Because you're relying 3 on the operator acknowledging that it's going in the wrong 4 direction and reversing his actions. Whether or not the 5 valve motor is sized properly depends on when the operator 6 realizes he's made a mistake and when he turns it'around and 7 goes the other direction. You wouldn't want to rely on l

8 operator, system functionality for the operator recognizing 1

9 a problem and within the right time frame turning that valve 10 around. Because if he didn't turn it around, you may lose the system. You don't want that. You don't want to rely on 11 12 recognition and action within a real narrow time frame.

13 Otherwise, the system loses its function. That's my 14 concern.

15 MR. ROTHBERG: Question'19 was Bechtel KWU in 16 presentations to the MOV Users Group and the NRC staff has 17 stated that design basic-differential pressure testing of 18 valves can damage them and reduce their reliability.

19 Industry experience with valves in severe service supports 20 this. The generic letter appears to be recommending 21 practices which may increase valve failure rates, increase 22 valve maintenance costs, decrease valve reliability, and decrease plant availability. Please provide a rationale for 23 24 this recommendation. f 25 The short answer that I wrote down was if you Heritage Reporting Corporation (202) 628-4888

177 1 think that a particular test is going to materially dcyrade 2 or damage your valve, then that's not the test you should 3 perform. I also think that it's kind of interesting and may 4 be very significant that when you have an MOV that might be 5 damaged by stroking it at design _ basis conditions, what the 6 service life of that valve might be. That's something that 7 you have to look at.

8 Are you still on the hook to prove that the MOV 9 is operable? You bet. That's about all I can say about 10 tLat '.

11 VOICE: John Hirsch. This is going back to the 12 lock out question. I just thought I'd bring up this one 13 specific case. On a key lock switch in a control panel, 14 would that constitute a locking of the operator? Would that 15 be sufficient locking?

16 MR. JACOBSON: If the key was not kept in the 17 lock.

18 VOICE: The key would be with the shift 19 supervisor.

20 MR. JACOBSON: That would be sufficient for the 21 electrical aspect of it.

22 VOICE: Now you're talking about spurious 23 signals? We'd have to protect against spurious signals as 24 well?

25 MR. JACOBSON: No, we're talking that we have to Heritage Reporting Corporation (202) 628-4888 i

4 178 1 . protect for one, against an operator mechanically putting it 2 in the wrong position.

3 VOICE: Okay, I'm talking about in the control 4 room operators, in other words to operate the valve you 5 would have to go to the shift supervisor, ask for the key, 6 get the key, go over, put it in the switch.

7 MR. JACOBSON: There's a hand wheel on it too, 8 right? It could be operated locally?

9 VOICE: So in addition the hand wheel would have 10 to be locked up as well. Okay.

11 MR. KIESSEL: And what about the MCC controls?

12 Are those locked out?

13 VOICE: So we'd have to protect against not only 14 inadvertent operator action, I call an actuator is a thing 15 on the valve. An operator is the guy in the room. But we'd 16 have to protect against inadvertent operator work as well as 17 I would say spurious signals or somebody just turning the i

)

18 hand wheel by mistake. All bases would have to be covered 19 before we'd call that valve locked out? )

i 20 MR. SCARBROUGH: Yes, that's why we said all j l

21 forms of inadvertent operation. You can do it by various 22 means.

23 VOICE: In the letter you read it and it kind of 24 looks like examples, and it sounds almost like any one of 25 them would be sufficient by themselves.

Heritage Reporting Corporation (202) 628-4888

, = . . .

j,

.k c Ms 179-I 1= 'MR. SCARBROUGH: But that-doesn't make a lot of. j 2 ' sense.when.you think.about the Davis-Besse event.-

E If'you're 3 only1 going to put a little tag'on the hand wheel, that 4' ;doesn't.do you much for.the Davis-Besse event.

5 VOICE: No, but likewise, the Davis-Besse event 6- was'somebody hitting the wrong switch by mistake. That's 7 something that usually happens over.a fairly short period of '

.8 . time. If.he.were to go and consult with another operator, 9 .you. would' hope that between the two of them they would.know 10- that'wasn't the right valve to hit. Or.that he grabbed the 11 wrong handle.by mistake. That's the kind of accident we 12 'should be protecting against, not some kind of, not a wild 13 maintenance man running through the plant and turning hand 14 wheels indiscriminate 1y.

15 MR. MARSH: One of the concerns we have about

- 16 electrical / mechanical is we understand the generic letter is

- 17 somewhat unclear in this area, where the or applies and 18- whether it's an either statement. But one of the concerns 19 we'had is if power is removed from the valve, you go to the 20 motor control center and rack out the breaker or the valve.

21 Now you've lost-position indication for the valve.

22 . Generally speaking, once that's done there's no power going 23 to its limit switches and you can't tell from the control 24 room what its position may be. The valve may be 25 accidentally locally put in the wrong position and you may Heritage Reporting Corporation f

(202) 628-4888 u-__-2__ ___m-__ .-.m_ - _ _ _ _ ,4.uu.. ___.

a 180 1 not know-it.

2: Infsituations like that,-we were saying put on

'3' some, device which tells local operators don't change the

4 ' position of this valve, like a wire with an aluminum tab on 5 it. Tha weren't talking about stem locking devices, we

,6~ weren't talking about'any larger measures. We're talking-3- , ' 7. 'about thatLkind of combination.

8 We're not looking at big programs to prevent g -9l operation. If-you.can construct the right kind of a story, 10- if the valve is behind a door that's locked and access is 11 prevented by that means, plus you've got a limit switch or 12 -you've,got a key lock switch in the control room, that's

^

13 probablyEall right. I'm not going to say yes, but it's I

14 probably all right.

15 If there are means so-that you can.take those-16 . measures to prevent that valve from being inadvertently 17 operated, then that's generally all right.

18 Keep in mind, though, that if the valve has to be 19- repositioned in the course of an accident or following an 20 accident, that's a different matter.

'21 VOICE: I would assume that would automatically 22' exempt it from the category completely because you would 23 need it at some point.

I 24 MR. MARSH: That's right. If you've got to 1

25 change the position of that valve, and now you're into a 1 l

Heritage Reporting Corporation (202) 628-4888 i

l

r ,

181 L11' different ' framework, you. don't'want-to prevent it from being 2 operated in those cases.

3 VOICE: .There could be'some valves out there with 4 keyflock switches on them that are needed for long term-5 ' recovery, that obviously we shouldn't takh-out of the 6' program.- But we're talking about the kind of valves'like 7 you-were discussing earlier like the maintenance valves and 8 that kind of thing.

9 'MR. MARSH: Exactly.

10 VOICE: 'Thank you.

11 MR. SCARBROUGH: The next one, number 20. "In 12: Paragraph 4 of the background section the NRC clearly 13 l implies.that motor-operated gate valves'in the high energy 14 line break, HELB,' service are likely to be under-sized.

15 This discussion is immediately followed-by a statement that 16 licensees are responsible for the design, testing, and

.17 maintenance of all valves and assuring of their operability.

18 Utilities cannot duplicate this testing in situ.

19 Insufficient data'has been released from or published about 20 this test program for utilities to make an accurate 21 engineering assessment of this test program's results. In 22 addition, the insufficient data was gathered and the program R23 results released to determine the cause and quantify the 24 anomaly. Is this paragraph and statement intended to force 25 utilities to install new and larger MOV's in HELB Heritage Reporting Corporation (202) 628-4888

= - _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ - _ _ _ _ - _ _ _ - _ _ . _ _ _ . ____ __. _ . _ _ _ _ _ _ _ - - _-

V .

182' 1 applications?"

2 No, it wasn't intended to force licensees to 3 install new and larger MOV's. It was intended to ensure 4 that licensees can demonstrate that MOV's can operate under 5 design basis conditions. If through General Letter 89-10 6 program we find out'that the MOV cannot perform that 1 i

7 function, then you're going to have to do something about-  !

i' 8 it. But we weren't trying to force every utility to replace 9 all of its MOV's with larger ones.

10 MR. MARSH: We do think also that we've been  !

11 given a lot of information about these tests, too. This 12 implies that we haven't released information, and I don't 13 think that's the case. I think we have put this type of j 1

I 14 information out.

15 MR. KIESSEL: You're going to have to bear with  ;

16 me again. Brian wrote another long question.

17 "In the first paragraph of the background section 18 of the Generic Letter 89-10 it states, 'NRC staff 19 assessments of the reliability of all safety-related MOV's 20 based on extrapolation of the currently available results of 21 valve surveillance performed in response to Bulletin 85-03 22 indicate that the program to verify switch setting should be 23 expanded in order to ensure operability of all safety-24 related fluid systems.'

25 "Most of these surveillance were performed under Heritage Reporting Corporation (202) 628-4888

_ - _ _ _ = .

183.

y.

1 an. indirect thrust indicating technique. The indirect

.2 technique is' dependent on constant stem factor. Idaho 3_ National Engineering Lab report such and such prepared'for 4 the NRC.has an interim report for contract work to resolve .

5 Generic? Issue 87 and. casts serious doubt on the assumption-6 .that: stem factor remains constant.

7 "Specifically it statesi ' Contrary.to common 8 belief,.the. ratio of operator torque to stem = thrust is not a 9- constant, but change with valve' loading.' Since NRC 10 research casts doubt on the technique used to perform most 11' of the Bulletin'85-03 surveillance, please explain the 12 . justification for using the Bulletin'85-03'resultsias a-13 basis for the generic letter."

14- ~(Laughter) 15 MR. KIESSEL: Brian, every time I read this I 16 seem to answer the question. You say that maybe what we did 17 under 85-03 wasn't good enough and why should we use that as 18 .a basis for going forward with the generic letter to correct 19 that problem?

20 I really don't know how to answer you other than

'21 to say that this concern and the results that we did see out 22 of the bulletin stress the need for additional work on 23 verifying that valves will operate properly when subjected.

24 to the high loads that they'll see under the design basis 25 events. Much beyond that, I'm afraid I'm at a loss unless Heritage Reporting Corporation (2L2) 628-4888

1- f o m

+

184-1' 'you want tojdevelop a' dialogue.

H '

'2 VOICE: I guess the big thing that we see is, 4 1

3J again, is~this thing that we're seeing a very high failure H

4. rate. Were they really failure, or are we getting caught'by 5 this-variation in stem factor? I don't know.

6 MR. KIESSEL: I don't know either. But.the Vi' 7 impression'that, well because I don't.know which way,the 8 error:goes all the time.

3-

9. VOICE: :We may be embarking on a large program 10 based on data which'may be erroneous. That is my big 11- concern. If we make a major commitment of funds on.a small

-12 data: base which may be wrong, that doesn't give me a warm-

13. fuzzy. feeling, and.for my management, they don't like it at 14 all.

15 MR.'KIESSEL: I understand. But remember, 85-03 16 was not the sole basis for going forward with"the generic 17 letter. It merely substantiated a concern. Also there was 18 concern, just the recent LER's that we've been alluding to, 19 the INEL testing results that you. reference don't give me a 20 warm fuzzy feeling about the valves being able to function 21 properly.  !

22 MR. JACOBSON: A lot of what's in the generic -l 23 letter is not new requirements. A lot of that stuff should 24 be at every facility already. You should know the design 25' basis of all your safety-related equipment. You should know q i

Heritage Reporting Corporation (202) 628-4888 1

l l

- _ _ - _ - _ _ _ . _ _ - _ . _ a

~,V 185 i f 1 ;whatt your. torque switches are set at. You should.have F .

g .2 validation for allithese settings.  ;

j' O q. 3 VOICE: I guess one of my biggest concerns is

-4 starting out on a large program for design basis testing-5 valves in the plant which, without a doubt, is very 6 ' difficult. Because based on our experience of diagnostic

'/ _ testing-valves, we have had a phenomenally low' failure rate.

8 I was at a plant prior to this where we'didn't do 9 diagnostics and I do remember seeing operators and valves-10 torn apart regularly. I'm not sure the requirements that we 11 ' design basis test if we're doing a research project, and the

~12 other one is this fixed interval'for retesting. I've not 13 seen the need for'it based on our retest results from our 14 original diagnostic. testing.

15 MR'. KIESSEL: Brian, the generic letter 16- particularly on the retesting, specifically invites you to 17 propose alternatives based on what your testing has shown to 18 date.

19 MR. JACOBSON: We're talking about testing ten 20 years in the future now, on this retest.

21 THE WITNESS: I've got valves that I'm retesting 22 five years from when I did them the first time around.

23 They're coming up with'the same numbers.

24 MR. KIESSEL: Then it sounds like you're starting 25 to provide the basis for extending that time.

Heritage Reporting Corporation (202) 628-4888

= _ - _ _ _ -_

l , .

l 186-1: MR. SCARBROUGH: The last one from Philadelphia o

.{

s

.' 2 Electric was Section'D of the generic' letter recommends j J

3 " Prepare or review procedures to ensure that correct switch 4- settings are determined and retained throughout the life of 5 the' plant." Section L states,."Each licensee shall also e

6 ' submit in writing any future changes to scheduled p

7 commitments." These revised schedules or alternative 8; actions may be implemented without NRC' approval. Section 11-E '9 ' could,be interpreted as requiring'that any changes to MOV-

^

11 0 . design or maintenance procedures be reported to the NRC. Is 11 this NRC's intent?

12 No. I read Section L As just talking about:

13 schedule changes reported to NRC. I don't know of.any 14 intention to have changes to MOV design or maintenance 15 procedures reported to NRC through the mechanism of the 16 generic letter.

17 That's it for Philadelphia Electric. ~ Brian,.is 18 there anything.you want to talk about specifically about 19 those?

20 (No response) 21 We have three more sets of questions to go and 22 these are shorter. So if we can hang in there.

23' The first one is Hope Creek.

24. Page_three of the generic letter, last paragraph 25 on the page. "If a valve is operable from the control room Heritage Reporting Corporation (202) 628-4888

_h__________mm_.___._.-.____ _ _ _ _ _ _ _ _ . _ _ _ . _ _ _ _ _ -

187 1 and it has a local hand wheel operator, is it undesirable to 2 block its operation from the control room or the motor 3 control center? To be considered a non-position changeable 4 MOV must it therefore have a local locking device on the 5 valve?"

6 I think we talked about this before. You do not 7 have to lock the valve from operation. What we're looking 8 at is preventing inadvertent operation. Any possible type 9 of inadvertent operation.

10 There are various ways to do that. We've thrown 11 out various ways. The key lock switch with some sort of 12 mechanical tag. But we're not talking about preventing 13 operation by locking the valve. There are other ways to do 14 it.

15 The second question was " Paragraph L, on pag' 16 five of the generic letter addresses testing under design 17 basis conditions. Is it the intent of this item to test all 18 MOV's or will the testing of one valve for each type of 19 operating condition be sufficient?"

l 20 To start off, you must test where it's j l

21 practicable. That's spelled out in the generic letter and we 22 talked about that. And you must justify the results of ,

)

23 whatever prototype you'used to that MOV, and there are 24 various ways of doing that. I won't go on. I think we've ]

]

25 talked about that quite a bit. {

Heritage Reporting Corporation (202) 628-4888 l

1 1

~108 a 1 IJ VOICE: I'm glad you mentioned practicable. .I-E

-2 assume.that.means whenever possible? Can you clarify your 1 3 definition'of that word?

4 MR. SCARBROUGH: The generic letter says where 5  : practicable. People are very. concerned about us reading the p 16 generic letter word for word. Practicable has, we've had 7 'examplesfof_other trains, testing one train. But the intent t

8 of the generic letter is to test as much aus possible. If 9 you can justify the. application of data from another MOV or i.

10 set of MOV or data to that MOV, I think that goes a long way 11 in. responding to the generic letter.

12 ~ VOICE: I'm more concerned with the kind of

- 13 valves'we've run into where we couldn't get any DP testing, 14 on them at all and-we'd have to go out. I guess that's 15 right, what you're saying is you'd'have to go to the

16. prott'.ypical' testing or the data base type testing. I guess 17 we go as far as we can in the DP test valve just short of 18 putting the plant into some kind of unsafe condition, or 19 modifying our piping to allow for this special kind of 20- testing. I guess that's my interpretation. Is that 21 correct?

22 MR. SCARBROUGH: We would expect some ingenuity.

23 If it's just a little difficult to do it, in trying to set 24 up the system alignment, we would expect a little bit of 25 thought in how to go about doing it. But if you're going to Heritage Reporting Corporation (202) 628-4888

___m___-.__.____ _ _ _ _ _ . _ . _ _ _ _ . _ _ _

'4:

189 1 put the plant in an unsafe condition or you're worried about 2 damaging the MOV, that's a reasonable, practical reason not 3 to do the testing.

4 VOICE: But basically if we can do the test with 5 what we've got, with the physical plant as it is right now, 6 we should try to do it.

7 MR. MARSH: In all its various modes of operation 8 too. You have to consider one through six. You don't just 9 look at one and two and say I can't do it now, therefore 10 it's impracticable. You've got to consider alternatives, 11 when it can be done. Maybe five and six you are able to 12 figure the system so you can do it. These are obvious 13 things. You understand that.

14 VOICE: But that sometimes requires LCOs and 15 things like that. Is that something we should consider?

16 MR. MARSH: You have to be aware of those LCO 17 restrictions. I think you know of our policy on LCOs.

18 VOICE: Thank you.

19 MR. SCARBROUGH: We're going to caucus for two 20 seconds.

21 (Pause) 22 MR. SCARBROUGH: Okay, we're back.

23 The next set came from Duke Power Company. Is 24 Duke here? Great.

25 The first one concerned --

Heritage Reporting Corporation 1 (202) 628-4888

r 190 1 VOICE: Just answer question four.

2 MR. SCARBROUGH: Question four. "Are additional 3 quality assurance requirements being suggested on the

4. ' surveillance adjustment maintenance and repair of safety-5 related MOV's?'"

6 No, we're not establishing new QA requirements.

7 Surveillance adjustment maintenance repair of safety-related 8 MOV's is covered by Appendix B. Your current program should 9 address that. Under diagnostics, you want to make sure your 10 diagnostics are reliable when you perform diagnostics, but 11 Appendix B requirements cover quality assurance. So that 12 would be my response.

13 Is there anything you want to add to that or 14 clarify?

15 VOICE: Thank you.

16 MR. SCARBROUGH: That's it for Duke Power. Thank 17 you.

18 The last set comes from Yankee Atomic. We had a 19 couple of questions earlier.

20 VOICE: Just question three.

21 MR. JACOBSON: When.should verification of switch 22 setting be performed? Immediately after adjustment 23 including packing adjustment? Or the next available time 24 when MOV cycling is permitted by plant technical 25 specifications?

Heritage Reporting Corporation (202) 628-4888 M _ _ - . - - . _ _ - . - _ . _ _ . _ _ . _ _ _ _

l, 191 1 I assume you're referring to performing 2 maintenance on an MOV, when do you need to do switch setting 3 verification? Is that what your question is referring to?

4 VOICE: It's more on the idea of adjusting 5 packing --

6 MR. JACOBSON: Okay, you've already got the plant 7 running and you go down and you have to adjust packing.

8 Your question is do I need to verify switch settings at that 9 time or can I wait until the next available time when MOV 10 cyc1'ing is permitted by plant tech specs.

11 Our answer to that is if you're going to adjust 12 packing you've made a change to the thrust requirement for 13 that motor-operated valve and you need to analyze that right 14 then as to whether or not that valve is still operable.

15 Now if you can verify that this packing 16 adjustment has had no effect on the thrust requirement or 17 that the effect that it has is bounded by the margin or 18 whatever that you've already got the switch settings at, 19 then you wouldn't have to do anything further at that time.

20 But you can't just go and adjust packing without analyzing 21 what effect it's going to have on your particular MOV at 22 that time.

23 MR. MARSH: Hold on a second, I want to ask the 24 group something.

25 (Pause)

Heritage Reporting Corporation (202) 628-4888

192

'l MR. JACOBSON: Have we got another question?

2 VOICE: What if you set your gland for a torque 3 wrench? Do you have specific values for your gland?

4 Setting it back to the original torque that you refitted the 5 valve at and did the original testing, would you have to go 6 back and do another --

7 MR. JACOBSON: If you can verify that whatever 8 you do has not decreased the margin available, or if you can 9' verify that you still have enough margin available to 10 operate that valve under design basis conditions, you're 11 okay.

12 VOICE: Is that a yes or a no?

13 MR. JACOBSON: You have to look at the plant 14 specific basis. I don't know what your torque switch 15 setting is --

16 (Multiple conversations) 17 MR. JACOBSON: --

for the packing is based upon.

18 I can't answer that without knowing the history.

19 If you can verify that that setting is equivalent 20 to a certain thrust requirement, for instance, so forth and 21 so on, that might be acceptable.

22 MR. MARSH: In other words, if you've got data 23 that shows that at the' manufacturer's recommended torque 24 value on the gland the thrust requirements are X --

25 VOICE: Most programs today are based on -- which i

Heritage Reporting Corporation (202) 628-4888

_ - _ _ _ _ _ = _ _ _ _ - _ _ _ _ . _

193 1 is beyond what -- into the gland. But for example, --

i 2 you're given a_certain value. -- range for that, but 3 basically -- you're going to get basically the same gland 4 load. So there really should be no reason to go back and -- l 5

6 MR. JACOBSON: The same load as what? You may be 7 operating based on a test that you did, a full DP test, for I 8 instance, that shows that you've got so much margin in your 9 actuator that can operate the valve under design basis 10 cond'itions. Under where the packing was previously set.

11 Now you go in and you change the packing where you tighten 12 down on the packing. I don't know how much of that margin 13 you've taken away.

14 VOICE: Maybe it's presumptuous on my part, but I 15 would think anybody who would go to the trouble of taking 16 base line signatures would make sure that the valve was set 17 up properly and the operator set up properly before going to 18 the trouble of taking signatures.

19 MR. JACOBSON: Yeah.

20 VOICE: If you're in the middle of a two year 21 cycle, for instance, and two-thirds of the way through your 22 cycle you find you have a packing leak and you go back down 23 and torque that to its' original value --

24 MR. JACOBSON: Original from when? That's what I 25 don't understand.

Heritage Reporting Corporation (202) 628-4888

194 i

• , 11 VOICE: From'when you took your-base line.

2 MR.-JACOBSON: You mean the packing has. loosened-

'3 up and now you torque it back to'theLoriginal value?

4- VOICE: That's right.

5 MR. JACOBSON: I'would say that if you can show 6 that the values-would be the same as long as.the torque is 7 the same, then you'd be okay.

8 MR. SCARBROUGH: Florida Power Corp., do we need 9- -to go over any more of your questions that you had.a chance 10' to't'ke a a look at?-

'll VOICE: One and five.

12 MR. JACOBSON: Number one, "When design basis of 13 an MOV includes a high-energy line break, will 100 percent 114. . design margin for thrust, for example using a valve factor 15 of .6, be acceptable in lieu of differential pressure l

16 testing?" i 1

17 I think we're getting back into what would be an 18 acceptable margin to use in lieu of the full DP testing. I- 'l 19 have-thrown out an example of ten to one, and now you've got

-20 an-example of two to one.

21 I think at two to one we're starting to get into 22 the area where we would be of some concern. A .6, I've seen 23 valve factors that are higher than .6 for certain valves.

24 Maybe not for the type of valves you're looking at. Maybe  ;

1 25 that would be acceptable, maybe it wouldn't be. We'd have  !

l Heritage Reporting Corporation i L (202) 628-4888  ;

l 9

d 195 1 to look at that on a case by case basis.

2 Number five, ."The generic letter states that 3 testing by INEL concludes that MOV diagnostic systems that 4 measure both stem thrust and motor torque are best suited 5 for predicting valve motor performance at design basis 6 conditions. Will systems that measure both thrust and 7 torque be required to meet the intent of the generic letter?

8 Or will one of the parameters be sufficient?"

9 We're not requiring any specific diagnostic 10 system nor any specific parameters be measured. We want you 11 to be comfortable with the diagnostic system that you 12 choose. We're requiring that you assure yourselves that 13 these MOV's will work under the design basis conditions.

14 Now whether you can do that with just thrust or with just 15 torque, you may have to have added margins on ones that 16 measure just one or the other. Maybe if you have both the

, 17 margins don't need to be as great. There's a lot of l

18 flexibility in what can be done. We're not going to say 19 what specific system or parameters need to be measured.

20 I think the more you measure the better you are 21 and the more accurate you are, and the better you'll be 22 because you may not have to oversize valves unnecessarily if 23 you can get a system that's a lot more accurate.

24 MR. SCARBROUGH: We have some cards here. Let us 25 take a minute and decide how we're going to divide these up Heritage Reporting Corporation (202) 628-4888

f& +

s 196-

/

'l .here.-

2 MR. MARSH: I'll'just start'with.one by Brian 3- Curry, Philadelphia. Electric. Brian, I think we've answered.

l 4L your question. It's "How long.after the submittal of our c SL exceptions to the generic letter recommendations will we get 6 the safety evaluation report back?"

7 As I explained, it's not the intent nor the plan 8 to give safety evaluation reports on exceptions. -We look to L 9 you to implement.the program that you ' judge to be' adequate L 10: based on the generic letter. But as.we said earlier, if you 11 need clarification, contact us. If'you need a safety 12 evaluation because it's in license amendment land, you'll 13- get that. Just communicate with us through your project 14- manager as to what your needs are.

15 We've got one from John Hirsch, I think we've-u 16 answered this, too.

17 Question one, "Is prototype testing then not 18 valid any longer based on EG&G tests?" We haven't said 19 that. We don't believe that's true.

20 Question'two, "Can an MOV be left in a

21. ' conservative setting' indefinitely?" I think we discussed 22 that in some detail, too.

23 MR. SCARBROUGH: We had one here from Barry 24' Scorbis. Concerning prevention of -- operation of position 25 . changeable MOV's, is use of a key lock switch by itself i

Heritage Reporting Corporation (202) 628-4888

. 4 a

_ __.m.___._.. _ _ _ . . _ _ _

197 1 acceptable? I think we've said no, it's not. You'll have 2 to have some other means of ensuring that there's not 3 someone down there at the motor control center or the valve 4 itself causing some inadvertent operation.

5 MR. JACOBSON: This one is from Ken Anderson of 6 Pennsylvania Power & Light. "Since the great focus seems 7 top be on the inadequacy of the design determination of 8 thrust required, it would be of interest to know how many 9 valve / actuator manufacturers have provided Part 21

~

10 notification of this inadequacy within their program. If 11 not, why not?"

12 I think I share the same sentiments there, in 13 that I think maybe that is warranted on some of these 14 manufacturers. I know personally I've discussed this issue 15 with a couple of them. There seems to be some answers that 16 they can give that may be legally get out of why they aren't 17 being reported.

18 One of the reasons is that we already know about 19 the problem. Another reason is that some of the people that 20 have found this problem have done specific analysis of the 21 ones that they've shown and have made recommendations and so 22 forth.

23 Certainly this problem would fall under Part 21 24 and I know the vendor branch, for instance, just did an 25 inspection last week at one of the valve manufacturers, and Heritage Reporting Corporation (202) 628-4888 1

1

- - ~ ~ _ _ - - - _ - _ _ - _ - - _ - ,

198 1 this question was brought up there. So I think we are  !

2: somewhat pursuing this issue, and certainly we are providing 3 the information when we become aware of it, of any of these 4 valve factors or so forth that are not what they were 5 originally assumed. We would expect that in your audits of 6 your vendors and so forth that you would ask the same 7 question.of these vendors, as why are they not reporting 8 this information if it is required it be reported.

9' MR. KIESSEL: The next question in from Suzanne 10 Montgomery. I'm afraid I'm going to have to ask you to' 11 expand it a little bit. Let me read it and then if you 12 would expand.

13 VOICE: That question was answered later no when 14 Brian got up. It has to do with the wealth of information 15 that was available in the industry as far as the valve type 16 prototypes that have been done, and the gentleman that was 17 from EPRI here had also talked about it. So it's been 18 answered.

19 MR. MARSH: That's the end of the questions that 20 we've received either in card form or in other forms. We'd 21 be glad to answer any other questions you may have.

22 VOICE: I have one question. It appeared that 23 our questions didn't get it to you. We taxed them, but 24 maybe the fax was down.

25 MR. MARSH: Sorry, no more questions.

Heritage Reporting Corporation (202) 628-4888 h

_______ __ ____-__-- l

F 199 L (Laughter) 1 2 VOICE: I have a statement to make then.

'3 (Laughter) 4 VOICE: There's one thing that we didn't even 5 breathe about and that was there's one word in Paragraph B 6 of the generic letter, and there's that one word that says 7 " overload." The question is, is it the intent, I didn't 8 write this question, by the way, is it the intent of the NRC 9 that utilities consider overload settings if the overloads 10 are currently bypassed except in a test mode? If so, what 11 criteria do you suggest to be used in setting the overloads?

12 MR. JACOBSON: I'll take that one. First of all, 13 If the overloads are bypassed, you certainly wouldn't worry 14 about them. If they're bypassed in all times when you'll 15 need them. I know some plants have arrangements where on 16 safety ejectior, signals or so forth they are bypassed.

17 The other part of the question is, if they're not 18 bypassed, certainly in part of your design review you have 19 to ensure that they are appropriately set and will not 20 prematurely or spuriously trip out. That's inferred. Not 21 necessarily specifically brought out in the generic letter, 22 but that is a piece of safety-related equipment in your 23 plant and you have to ensure that it's set properly.

24 MR. MARSH: We had a lot of discussion about 25 overloads and the type of things that may be involved in Heritage Reporting Corporation (202) 628-4888

200 1 overload settings in the Chicago meeting. There was a 1 2 gentleman there who wanted to talk about what do you 3 protect? Do you protect the motor or do you protect, what 4 was the other aspect?

5 MR. JACOBSON: The question was brought up 6 whether you protect the motor or... In most cases you can 7 set the overloads properly to provide both motor protection 8 and prevent spuriously tripping. In almost all cases they 9 can be set properly.

10 The question in Chicago was he had a case where 11 he felt he could not do that, and he had to either choose 12 between whether or not the overload would provide protection 13 to the motor, or whether that it should be sized such that 14 it wouldn't spuriously trip out. Our position was that in 15 the extreme cases where for some reason you can't size them 16 properly, they should be sized such that the MOV should 17 complete its stroke. But that shouldn't occur.

18 MR. MARSH: More questions?

19 VOICE: I also had a couple of questions that got 20 faxed but apparently didn't get received. The first one I 21 have is also about overloads.

~

22 MR. MARSH: Make sure we get those questions, 23 please, so that when we form the meeting summary even 24 though, we'll get them involved, we'll associate them with 25 the other answers that we've given so your questions get Heritage Reporting Corporation (202) 628-4888

6 201 1 recognized in the summary.

2 (Discussion about getting questions) 3 VOICE: One of the questions I had pertained to 4 overloads. I wanted to know were they also considered to be 5 part of the initial and periodic testing requirements other l 6 than if we size an overload, are we also going to have to 7 test it initially and periodically?

8 MR. JACOBSON: I think our reg guide on 9 overloads, if you're committed to that, requires overload 10 testing. You're saying what if you're not committed to it?

11 Are you required to do overload testing?

12 VOICE: Correct.

13 MR. JACOBSON: I would say that the generic 14 letter does not address that topic.

15 MR. MARSH: We need to be real careful about what 16 your commitments are, what this letter is intended to do. I 17 want to be real hesitant on answering that kind of question.

18 MR. JACOBSON: Certainly it's a good idea to do 19 it.

20 MR. MARSH: But the intent of the generic letter 21 was not to require that type of testing, but I want to go no 22 further than that.

23 VOICE: My other question, we've discussed quite 24 a bit about the operability and you stated that Section 11 25 testing alone is not sufficient to provide assurance of MOV Heritage Reporting Corporation (202) 628-4888

___ A

%.o' 7' ' - -

]

4 ,

l, 202-

,p '1. operability'under design basis' conditions. If a. utility is n

2: using success'ful Section 11~ testing to demonstrate.

3 operability, then is the~ utility required to' change the i

7.

4 working definitionfof f operability even if: they have a e

'5 satisfactory MOV program?

6 - MR. MARSH: .The-generic letter says this is a way 7 of determining acceptable design and operation of the motor-8 operated valve switch setting and the motor itself. Do you 9 have to change your method by which you determine 10 operability? Yes. Do you have to change the technical 11 ~ specifications? No, you don't because they still exist in 12 first their IST form and also the broad, general terms of 13 operability of equipment and associated switches and 14- whatever else.

15 You may have to change your own. internal 16 procedures-in terms of operability of motor-operated valves 17 to make sure that if you have such a procedure that says any

. 18 time any piece of equipment is determined to be outside ofi 19 its-design basis then it's not operable. If that's the type 20 of procedure that you're talking about, yes. You should l 21 reference, it seems to me, a valve program where you l

22 determine operability of motors and of switches. )l 1

- 23 Did I answer your question?

24 VOICE: I think you did, yes.

! )

ll 25 VOICE: Mike Edson, Alabama Power Company.

[ Heritage Reporting Corporation (202) 628-4888

)

1 r

---

.-__m_A- _ _ _ _ _ _ _ _ - ._..__..a._..

r 203

[ 1 I'd like toJgo back to this group of MOV's that 2 ore included in the generic letter but they're not included 3 in tech specs and Section 11 testing programs. Your Item J,

4. I believe it is, refers to this periodic testing of MOV's 5 afterwards. Does periodic testing now include a broader 6 scope? In other words, are we including all these MOV's

-7 that.aren't already in tech specs and aren't already 8 approved by Section 11 and now we have to periodically test 9 all safety-related MOV's and position changeable?

10 MR. SCARBROUGH: Yeah, the ones that aren't 11 locked out by some means of inadvertent operation. If you 12 can block them out so they aren't in the program anymore, 13 then you don't have to worry about them. But if they're in 14 the program, they're included in Item J just like any other 15 MOV that's in the program.

16 VOICE: We've already gone through a lot of I 17 wailing and gnashing of teeth to determine the scope of 18 MOV's covered by Section 11 testing, and then you say now 19 Section 11's no good and not only your definition of 20 operability is defined by Section 11's no good, but now the 21 scope of Section 11 is no good. Isn't that what you're 22 saying now?

23 MR. ROTHBERG: My question to you is do you 24 really, actually believe that Section 11 testing 25 demonstrates safety-related MOV operability at the design i

Heritage Reporting Corporation (202) 628-4888

jI

~

204

basis! conditions?

21 VOICE: ' Personally?

'3 MR. ROTHBERG: -Yes, q 4 VOICE: Of course not. But I'm trying to assess 5 the need to do some amount of periodic testing. I'm trying l

6 to' weigh.the pros and cons'and the safety. benefits. a 7 MR. MARSH: Your basic thrust of the question was 8 -do~we really.have to consider these non-tech spec, maybe not-9' even non-safety-related valves under this. paragraph J,' item.

10 VOICE: Exactly.

11 MR. MARSH: The thrust of the answer was yes you 12 do. _Why'do you? Decause the mispositioning of that valve 13 can defeat a safety function. The entire safety function of

. 14 that system. If that valve is operated throughout'the five

~

year cycle then it can degrade and you may not be able to 16 recover the valve from this position. So that's the reason 17: why-the. answer is yes, it does-need to be in the program.

18 If you feel that gee, we're really gnashing our 19 teeth over the scope of Section 11 and scope of'the generic 20 _ letter, I apologize, but that's sort of where we are. We do 21 believe that the scope that we've laid out by this generic 22 letter is necessary to meet the regulations. We do believe 23 that_this generic letter is necessary in terms of 24 operability of valves, that the in-service testing 25 requirements for motor-operated valves is not sufficient, Heritage Reporting Corporation (202) 628-4888

' _ m_ __________-___.--__.__mm___ _ _ _

205 1 and that this is a program that is necessary.

2 VOICE: So it's also an expansion of the Section 3 11.

4 MR. MARSH: You're absolutely right. It is.

5 There's no doubt about it. If you look at the scope 6 requirements under Section 11, it is narrow. We're taking 7 care of that problem as best we can through our regulatory 8 process.

9 VOICE: But we're not expanding the scope of 10 Appendix A and Appendix B to Part 50. That talks about 11 importance of safety and safety-related. Those have testing 12 requirements in them. So wo're not expanding that scope.

13 but we're still within the boundary of the Part 50. But now.

we're talking more about the other valves that weren't I 14 15 within the Section 11 scope.

16 MR. MARSH: The Generic Letter 89-04 that we just 17 promulgated on in-service testing, if you're familiar with 18 it, talks about this generic letter. It talks about the 19 difficulty we have with scope on Section 11. If you went to 20 the workshops we had on in-service testing you'll remember 21 that we told everyone that we have a lot of difficulties 22 with the current scope of the ASME code with respect to 23 valves, with respect tb pumps themselves. We're pushing 24 that process as best we can to bring that up to where we l 25 want it. And yes, this does extend the scope of Section 11  ;

l Heritage Reporting Corporation (202) 628-4888 l __ _ _ - _ _ _ _ _ _ -

206 l' testing'..'Section'11. 'That's-true.

2.  : Let me clarify that. This does not extend the 3 ' scope of Section 11 testing. You understand that. This

.4 does:not say you have to do Section 11 testing on these 5 'other valves.'

6 VOICE: I understand.that.

7 MR. MARSH: This only extends testing beyond the" 8 Section-11 scope.

9 VOICE: Right. I had some periodic testings that

. 10 wereldefined by my tech specs and I had a scope of MOV's

.11 that'were included in.that program, and now I've gotxa new 12 scope of MOV's that I'm going to have to test at some 13 'different--

-14 MR. JACOBSON: Are there that many MOV's that 15 we're talking about.here? Maybe I'm - .

16 VOICE: -There's a lot. There's a considerable 17 number. You're looking at.200 MOV's maybe that are' safety--

18 related'per unit. Out of those 200 MOV's you may have a 19 total of maybe 75 that have some type of in-service testing 20 requirements.

21 MR. JACOBSON: How many that aren't safety-22 related that we're classifying as position changeable? I 23 don't know that there's very many.

24 VOICE: I'm not talking position changeable. I'm 25 talking safety-related.

I '

Heritage Reporting Corporation (202) 628-4888 e

-_----u - _ _ . - _ - - _ - - - -

A i

207 l l

1 MR. MARSH: Let's probe that. Let me understand i i

2 .the difference between scope in Section 11 and scope of this i l

3 generic letter, putting aside the matter of position 4 changeable because that is clearly beyond Section 11 scope.

5 VOICE: It's clearly beyond.

6 MR. MARSH: So you're saying take that matter l

7 aside. What is the difference in numbers between Section 11 4 8 scope for IWV for example, and this generic letter scope?

9 VOICE: I would say at least a factor of 200 10 percent.

11 MR. MARSH: Can you give me an example of valves 12 which are not included in in-service testing but are 13 included here? I'm going to be real careful about this 14 because the scope on IWV with putting the matter aside of 15 position changeable, requires valves that are class one, 16 two, and three that have a specified function in shutting 17 down the reactor or needed to mitigate the consequences of 18 an accident. You're almost the same.

19 VOICE: The accumulator MOV isolation valves on a 20 Westinghouse PWR.

21 MR. MARSH: They are generally locked out in 22 Westinghouse PWR. They're prevented from operation.

23 Generally speaking. Power is removed.

24 VOICE: By one method. Your criteria requires --

25 Heritage Reporting Corporation (202) 628-4888

'208

1. -MR. MARSH 2- That's locked inside of containment.

2 'We're.not going to_make you --

13. VOICE: Okay. Let's,go with a different one.

L4 :How about, no,.those;are included.

5 'We're running across several'because we made a 6; lot of efforts'to go back and' redetermine all our design-7 . basis stroke times and now we're starting to run into a lot 8 of valves that we didn't have to do that'on.

-9 MR. MARSH: We can discuss it later. I am 10 inte'ested r in knowing the wide extensive difference. If 11 it's containment isolation valves, for example --

12 VOICE: No sir, all the containment isolation 1

-13 valves --

14 MR. MARSH: That's right. I1think the big 15 difference'is position changeable. I think that is, 16 generally speaking, beyond where we are.

17 VOICC: I can get you an exact listing. We can 18 go back and_get that for our facility. But there is a 19 significant' increase in the number of MOV's.

20 MR. MARSH: Okay.

21 VOICE: That's all..

22 VOICE: We've heard a lot today on if we have 23 enough margin and we feel comfortable with it, or we feel 24 comfortable with how we've set up the valve that's okay and 25 that's acceptable and things like this. When our resident Heritage Reporting Corporation (202) 628-4888

__._______._____-__m_ _ _ _ _ . _ _ _ _ _ _ . __ __

o

• 209 .

11 .comes in and'he doesn't' feel' comfortable about it, who wins?

'2? MR.~JACOBSON: I don't know that we.said'if you-

)

j; 3 . feel comfortable'it's acceptable. We said a certain margin 1 4 .would'be acceptable. I. guess we're going to be the deciders.

S - ultimately on whether that margin is acceptable to us or 6 not. I' doubt it's going to be the resident. If you have a

'7 disagreement with-the resident on what is acceptable and 8 what is not acceptable, I'm sure that would be escalated:and 9 the proper. people would make that decision based on similar 10 decisions we've made at other facilities.

11 If you have a problem with'something your 12' resident says, there are ways to handle that. I would hope

'13 that our residents are not going to be making these type of

-14. ' decisions. I wouldn't expect it would be left to them to do.

15L that..

16 MR. ROSS: I wanted to elaborate a little bit on 17 the question we had earlier about when the submittals come 18 in with alternative recommendations or schedules that are

~

19 different than those identified in the generic letter. I 20 ' don't want to leave you with the feeling that you're going 21 to'be left hanging out there. That basically the focus of 22 our verification or confirmation processes are going to be 23 with the TI that's beeh referenced earlier, and all that is 24 going to be done on-site with the records that are

'25 identified in the generic letter, they will be retained on-1 Heritage Reporting Corporation (202) 628-4888 L '

L _ _ _ _ _ __ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ . . _ _ _ _ _ _ _ _ _ _ _ _

210 1 site. But we will and are evolving some internal 2 methodology on how we're going to screen the responses that 3 come in that identify schedules different than the generic 4 letter, that identify alternative measures other than the 5 recommendations that are provided in the generic letter.

6 The level of that screening and how exactly 7 that's going to be conducted is still evolving and we'll 8 need to provide'that type of guidance to all the project 9 managers in a timely enough fashion that we will be able to 10 prov'de i some type of feedback, either generically or on a 11 plant specific basis via the project managers that gives you 12 some type of confidence that what you proposed meets or from 13 our perspective is consistent with the way we understand the 14 generic letter should be applied.

15 The proof of the pudding will be on-site, but 16 don't feel that you're going to be left hanging out that 17 gosh, we proposed a whole lot of different alternatives than 18 the generic letter and we're not going to find out until we 19 get a team inspection at our site that says okay, let's see 20 what you're doing and you're doing it all wrong. And you 21 say gee, we wrote you a letter two years ago that said we're 22 doing it this way. That's not at all the intent.

23 MR. MARSH: Other questions?

24 (No response) 25 MR. MARSH: Thank you very much for attending.

Heritage Reporting Corporation (202) 628-4888

211 1 We appreciate it. We hope you feel like this has been 2 useful to you. It is a frustration on our part, and I'm 3 sure it is on your part, not to have a lot of these answers, 4 not'to have the detailed guidelines, not to have the data, 5 not to know exactly what forms the family, what forms the 6 boundaries, and would it be a perfect world we would have 7 that information. But we do look to the industry to form 8 this data as necessary. This is, in my opinion, the right 9 type of endeavor that calls for a good industry, cooperative 10 effort.

11 So look for the meeting minutes in about two 12 months and the transcripts will be available probably in 13 about a week or a week and a half. Beyond that, any 14 questions or answers on the generic letter contact us 15 through your project manager.

16 Thank you a lot.

17 (Whereupon, at 4:45 p.m. the meeting was 18 adjourned.)

19 20 21 22 23 24 25 Heritage Reporting Corporation 4 (202) 628-4888 1

.g _- - ... , - - . - - . . - - - - . - - - - - - -- - - - - - - . - - - - - - -

L W

~C r ._1n f,y .

l .- REPORTER'S CERTIFICATE

@ ' - g; . .

it i ~ ' ,

'3 - .' DOCKET NO.:

.u 1

~4~ ' CASE' TITLE: Generic Letter 89-10 Workshop w

HEARING'DATE: September 26, 1989 Lf' -5 Rockville,. Maryland-

[ 6 ' LOCATION:

'7

.8, I hereby certify that'the proceedings and evidence are contained fully and accurately on the tapes and notes 9

10 repo'rted by me at the hearing in the'above case before the United States Nuclear Regulatory Commission.

12-Dat : 10-03 13 n.

r ,

15

'VV

/

@ /l/

" /,/ ~ j

-16 Official Reporte _ [

/

17 Heritage Reportibg Corporation 18 1220 L Street, N.W.

w.,

19 Washington, DC 20005 20  ;

21 22 1

23

j. 24 Heritage Reporting Corporation (202) 628-4888 25 m

__ A a' d '.i.. b k [ h # d b5