ML20138K593
| ML20138K593 | |
| Person / Time | |
|---|---|
| Site: | Comanche Peak  |
| Issue date: | 08/23/1984 |
| From: | NRC |
| To: | |
| References | |
| NUDOCS 8512190001 | |
| Download: ML20138K593 (79) | |
Text
-a. ~
l f
ic fl I;
PoR<
)
- f. *
-'.4 .
3 4
5 COMANCHE PEAK MEETING BETWEEN .
. -6 7 NUCLEAR REGULATORY COMMISSION STAFF 8
'9 AND TEXAS UTILITIES 10 11 MOTIONS FOR
SUMMARY
DISPOSITION
~
-12 13 * * *
- (
14 15 23 August, 1984 16 17 * *
- 18 19 20 t
21 22 23 p\
24 p 25 REPORTED BY: Pam Al fo rd , C . 5 . R.
n,emam%- ORIGML ,
w
.. a :
- i. ... . 1 2
1 APPEARANCES 2
3 Spottswood B. Burwell NRC
-4 Thomas A. Ippolito NRC 5 Richard Vollmer NRC ..
.6 Robert Bosnak- NRC
-7. David-Terao NRC 8 William Horir Bishop, Liberman, Cook, Purcell & Reynolds 9
David Wade TUGCO ,
10 Lon Fikar TUGCO 11 -
Robert Iotti Consultant to TUGCO 12 Bob Masterson Consultant to TUGCO
,. 13 .
Vic Farrarini Consultant to TUGCO 14
-Barbara Boltz CASE 15 Dr. David Boltz CASE i 16 John Holt REA 17 Bill May REA 18 -
19 20
~21 22 23 24 25
- i
- w. -
- a; 3 "r ,
3
~1. P,R_ O C E_ E,D_ I N_ G S_
- s "
I " '
,2 MR.'BURNELL: Okay. My name is Spott Burwell.
3- I mn?the licensing project manager for the Nuclear
~4 .- Regulatory. Commission. We are gathered here_today 5 for a meeting.between.the NRC Staff and Texas .
..- 6: Utilities to discuss, I believe it's two or tnree of 7 the Motions for Summary Disposition which Texas 8 Utilities has filed with the board, and I believe
. 9; these were filed in June or July. The meeting today
-10 ~is open to the public in accordance with our proce-
- 11. dure, and there has:been a meeting notice issued. I
~
212 doubt seriously if the parties down here have re.-
.l :
13' ceived'it since when I-left, why, we-were one sec-11<4 retary short and stil.'. trying to get the thing in 15- the mail.- The meeting. notice said that the. meeting 16 was open to the public.- .The= meeting notice also
'17: . stated that there was some possibility that it might 18 be necess'ary for the members'~of the staff to go on
.19 site with the applicant to look at certain of chese
- 20 pipes and that any members of the public who needed
, 21: or' desired to attend that should make arrangements 22 ~ with the applicant. Now, I am not sure that will be 123- necessary today. In any event, I think we are ready
-24 .to basically proceed with the first Motion'for Sum-25 - mary' Disposition, which was a motion filed on June
=1
a e 4
1 the 17th. This relates to the applicant's motion 2 for summary dispos'ition regarding stability on pipe 3 supports.
4 This discussion on this item is a continuation 5 of a discussion held on August the 8th and August 6 the 9th in Bethesda, meeting again between Texas 7 Utilities and the staff. With that, I turn the 3 opening of the discussion over to Mr. Ippolito.
9 MR. IPPOLITO: As you know, I set up this 10 meeting, and the purpose fo'r me setting up this 11- meeting is that I' read the transcripts of the pre-12 .vious meeting, and even thoughI'm not expert in
[ 13 --
pipe support design, it was readily apparent to me 14 that we don't seem to be able to converge on the 15 resolutions of two issues. One that Spott just 16 mentioned that I will call stability, and the second 17 issue is generic stiffness, I believe. I felt that
- 18 both the staff and the appli6 ants have been talking 19 about these two issues for about two or three meet-20 ,
ings, and I felt .that was enough and that something
, ~21 had to be done. We have present here, I think, the 22 people who could make these things happen. I think 23 that this meeting will conclude with a resolution.
24 And as I perceive the resolution, it's going to be 25 as follows:
.e- e ;.-
5
-1 One, let me speak -- I'm going to delay any 2 ' discussion on' generic stiffness until Mr. Fair 3- arrives,.he is down at the plant right now. I am 4 focusing now only on the issue of stability. As I 5 see the issue, we will resolve it in one of two or
- 6- possibly-three ways. One is let's take the pipe 7 supports in question.and fix them in the tradition-8 al sense. Two, this is an "or," or provide an anal-
~9 ysis -- the applicants to provide an analysis from 10 the responsible design organization. I guess what-11 I mean here is that if Grinnell was the responsible 12 organization, we want it from Grinnell. If Gibbs-i 13 Hill was the principal design organization, then we
.14 want it from -- okay? And then as you would guess, 15 these analyses must address all of the uncertainties ,
16 all of the appropriate uncertainties. -Juul the third
-17 . option is,.and'it's an option to you, you could do 18 both. '
-19 Now, I believe we're prepared to identify the 20 ' types of supports that*we're talking about, and that 21 will help us focus on what we are talking about.
22 And I guess we leave it up to the applicant in dis-23- cussion as to how to proceed from that point.
24 MR. VOLLMER: One element in that would be 25 ' preceding those options, w Would want you to
6 1 establish-a program and tell us what that program is ,
2 too, for you to go out and identify those supports 3 -in the plant which would fall under the criteria
- 4 _that we're giving. So it would be your responsi-5 .bility to go out and identify where they are and 6 what needs to be done either the fixing or the
-7 analysis.- Okay. Dave, why don't you --
8 MR. FIKAR: Let me just make some remarks.
- 9. . Tom, you made my opening remark awhile ago.
10 Obviously, we have been at this a long time.
11 We started filing these summary motions three months 12 ago, and we finally got them all. We haven't got
/ 13 -
any of your responses. We're going to be ready to 14L load fuel here in the early part of October. The 15 hatch wiil be closed this weekend, and we're ready 16 and willing if you will specify just exactly what 17 you want us to do. We've got the resources here to 18- do it, so I'm right on with 9our
/ approach, Tom, and 19 hopefully, while we're here today, tomorrow, or how-20 ever long that takes, we'll have resolved this thing .
'21 I suggest you be specific and stay with the motions 22 that were filed and not generate any new ones. That 23 always helps speed up things. Narrow the issues 24 rather than, as you pointed out, we need to work 25 towards some conclusion rather than a divergent set
~
L -
- g. .
na- s en
~ -
g ,.
7 1, of J guestions.or so forth. With that, we are here, 2 have'at us.
'3 IGl.-TERAO: I think one'of the major problems 4 .w e-had from the August 8th and 9th meeting as.be-5 tween.the staff and the applicant is that we never
. .- 6: completely agreed on the definition of unstable.
. -.7 . pipeJsupports. I believe we reached some under-
, 8- standing in concurrence on instability of piping 9; systems, but when-it came to-unstable individual 110 . pipe supports, we never seemed to agree on a-defini-11' ' tion .' We proposed a definition, and apparently it
~ 12 - . was rejected by the applicant. So, in order to 13 - achieve a viable means'of resolution of this issue, we.believe it's necessary.to first clearly-establish
.14 '
15: a mutually ' agreeable definition of instability- and 7
.16; then~ proceed with an analytical or physical _ fixes.
17; for;these, quote, unstable pipe supports. Now,'the-
- 18 definition that we proposed-6uring 'the last meeting
~19 was that instability would be the ability of a pipe 20 : support to shift or move.to an unqualified position.
'21' 'In this-respect, we aren't saying that it isi 22 ..necessarily a- safety concern- to the extent that we'
~
23 aren't implying..that it necessarily results in an 1 24 unstable piping system or other collapsed mode 25 failure, but we're 'saying that-if the-support does p
-j . s
$p-. p m;;u w =. , . .= -
s .
, 4 g
~ ~l f^ 1 not-have a positive controlled. clamping mechanism
- l 2 . .to assure that the' support cannot slide or rotate 13; along the pipes, then we have difficulty accepting 4; .the analysis for that. support because of the uncer-
'. 5 tainties-involved in that design. . ,
. 6; MR. IOTTI: I' don't think'we fundamentally y 17j disagreetwith the definition. We tend to disagree
', - _ 8- .in the word; unqualified position, because it 9 depends- on how you define unqualified as to where
- 10 we may have.some --
m (11 - MR. BOSNAK: -
To me unqualified means unanalyzed --
.12i- MR.:IOTTI: -Unanalyzable. 'If it hasn't been -
[13 g --
analyzed and all analyses -- any analysis has cer-n
- 14s ,
tain qualities. ~A support has to be in a.certain 15: ' location the analysis puts it.
It doesn't effect 16 - the result of .the - analysis if the ! support :is off a i
17 certain amount. - So if you could defi'ne unqualified
!18- . as beinglwithin those tolerafices. And then I don ' t 19 think you yourself would-have a problem'with the
- 20 ,
~ definition of the. word, as long as the control 21 ~ exists that prevents that support from moving ti22 . farther out than this region, within which the 23- -analysis-would hold, then you would consider that
. 24 -~ stable.-
/
._ 25 - MR. BOSNAK: I could accept that.
Ay-- *u-e--g---T me a--w-., 4,,-re-'a-- e si- w m'-e -etm- m -- - ---a---- ' --
- . . s 9
ll 1 MR. IOTTI: In that sense, I think we would 2 agree with yo.u. Also, in~the second sense, is in 3 the word positive control. There's two things 4 that can be. defined as positive control. One is 5~ where you design a positive control intentionally, 16'
~
in other words, you make sure -- absolutely sure 7- by design that that positive control exists. And
~8 another way for positive control to exist where 9- the geometry of the system gives you positive 10 control and you don't necessarily have to design
~
ill for it.
12 .MR. BOSNAK: That's a case-by-case situatio.n.
-13 The other is generic.
14' MR. IOTTI: We.can agree -- with those two 15 caveats 'we can agree with that definition. That's 16 been the board's disagreement in the last meeting.
17 We can't agree on the definition of positive control .
- 18 We can't agree on the definiElon of unqualified 19 with -- another way.of putting it is if there is 20 no control, and this thing can move in some manner
'21 where the system becomes unanalyzable because the 22 -uncertainties become so many that you can't possibly 23 address them.
y 24- MR. BOSNAK: There are uncertainties which 25 you might be able to identify, but you might not be
10 L
-1 able to model.
1 2- MR. IOTTI: To me, it's the same thing as 3' unanalyzable. When I defined unanalyzable, I 4
just wanted to' state for the applicant where we 5 disagree with'the definition of instability that 6' was put forth by the Commission, by the staff, 7 last time. We don't disagree with the definition 8 itself, we disagree with some of the terms within 9 the definition.
10- MR. BOSNAK: Probably w'e didn't communicate.
11 -
In other words, we didn't get into the details 12 enough to understand how you could have solved
~~
13 the problem, and, therefore, we couldn't -- we 14 could not agree.
- 15. MR. IPPOLITO: I think we have reached an 16' agreement. Am I wrong? I think Dave has defined 17 what we believe to be instability. You have said 18' you agree, provided that we' agree in the two areas
'19 mentioned, and I think staff would agree to that 20 as stated by Mr. Iotti.
'21 MR. BOSNAK: One is a generic sense and the 22 other is a case by case.
' 23 MR. IOTTI: I mentioned yesterday the easy way 24 to proceed right now is to address those two areas, 15 positive control, and whether the problem is
, v.
11 1 analyzable or not. We have to talk specific, if 2- I recall that.from our last meeting. We are not 3 talking about a large number of supports falling 4 into the category. You can' correct me if I'm 5 . wrong. You have a concern over a particular one,
.- 6 there is only one of those. ,
i 7 MR.-IPPOLITO: If we can agree to a definition, 8 I.think tha't would be a milestone. Once we agree 9 with the definition, I want that for the record.
10' Do we agree with that definition? Then you may
~
11 want to proceed with the area in which you are 12 going.
- 13. MR. VOLLMER: One point needs clarification.
14 I think what Iotti had indicated in.his concept 15 of definition, there was an implication that design t
16 included'certain margins under which the design 17 w ould still be applicable if there was motions 18 of the pipe support. Is thiE marging quantified 19 in design or is this something that is an inherent
[-
r 20 process? In other words --
21: MR. IOTTI: Can be inherent in the geometry 22- of the design. That's where you have to discuss 23 each specific case.
24 MR. VOLLMER: You're saying that in the design 25 process that there'are numbers called out that the
____=__________________
7,. .g: # .- . - -
12-1 pipe support is normally plus or minus so much 2 that the design is applicable.
3 MR. BOSNAK: Those would be positive tole-4 rances.
5 MR. IOTTI: Yes. .;.
6 MR. BOSNAK: -Quantifiable.
7 MR. IOTTI: Positive and negative tolerances.
4 8 MR. BOSNAK: But quantifiable.
9- MR. IOTTI: Yes, because you can analyze the
'10 Problem. So I guess we agree. If the staff.can 11- agree to those caveats on the definition, we 'll 12 agree with the definition.
MR. BOSNAK:
13 When you go out and identify-14 each of the problem supports, you would have to 15 then identify how you are going to solve it, whether
-16 it be by the generic sense or --
17 MR. IOTTI: That's correct. That's consistent-18 ly what Tom has said. We wo61d tell you what our 19 plan of attack would be, if it's just changing it 20 ,
because it's a lot faster. We might not even
'21 agree with you that it's potentially unstable, 22 but it's a hell of a lot easier to change. Or we 23 may decide to analyze. If I am correct, we're 24 down to four supports. It's not worth spending 25 a month arguing over four supports.
E e j 4 ~
13 1- MR. IPPOLITO: I think we have reached a I :2 milestone. I think we've agreed on a definition.
3 MR. IOTTI: I don't know that I can speak for 4 .. David, I recognize that that is my own opinion.
5 MR. IPPOLITO: What was before, was before.
. 6- We have a~ definition. I think now we can proceed
- 7. . with the definition. I think, David, what we.were
. 8 going to do based on this definition, you are going 9 to identify certain groups or categories that you 10 feel fall within this definition.
11 MR. VOLLMER: I think his definition was 12 generic, and whatever it applied, I guess we could s
( 13 discuss examples of that.
14 MR. IPPOLITO: .Okay. Fine.
15 MR. VOLLMER:- One point I think that needs to
-16 be made, and it's perhaps inherent of what you said, 17 but I think it's an important one. We would not
- 18 expect, I say not expect, TU$CO to take a look at 19 the design drawings with the tolerances and say, 20 " Gee, that support fits within that," because any 21 lateral movement could be constrained to this 22 amount. We would expect that, I think as Tom 23 indicated, that the original design organization 24 would likewise find these acceptable, because that 23 would be needed to' complete the design control
y ., ,. ::- a . : - -- . . - . . . ... . - - , - .. - a . .-
14 l 1 process. -
2 MR. IOTTI: In essence, what you really are
'a 3 telling me is that the initial-design organization 4- must have gone through that thought process when
. 5- they designed or else -- p.
6- MR. VOLIMER: Well, that's --
C7 MR.-IOTTI: We agree with that.
8 MR. BOSNAK: In other words, if they specified 9l a clamp,- and it got changed somewhere . along the 10- line, then whatever you ended up with has to go
- 11 back-and.be acceptable.
12 MR.'IOTTI: That's correct. We don't disagree-4 13 with.that.
.14 MR. TERAO: I think what I would also-like to 115 : add to your caveat is that in'the specific supports 16 to have for each' specific support some statement 17 from the effective --' effected' design groups 18 regarding the acceptability If.that support from 19 either I.T.T. Grinnell Home Office or M.P.S.I.
20 Home' Office, whether they believe that support will
- 21 perform-its intended function and not lose its t22 functionality because of any significant uncer-23 tainties, and from the Gibbs-Hill offices an
'24 assurance that the particular, quote, unstable sup-25- port design will not effect the validity of their
, ,=
==%+ -
_g., .'-ww..v.- .
- y. 4 .'---s.
v 1* * +~ee e- '- - - * * -
15 1 . piping analysis.
g i :-
2 'MR. IOTTI: Well,'I'm not so sure I can inter-3- pret that, because let's take a hypothetical that 4~ that support is done by I.T.T. Grinnell, and I.T.T.
.5 Grinnell, the home office, whoever designed it,.
. -6 tells you in their opinion that support is stable,
-7 it'will behave stably.
L8 MR. TERAO: I didn't say in their opinion.
9 .MR. IOTTI: Whatever. They say it's stable, 10: why should Gibbs-Hill do anything?
11 'MR. TERAO:- Because ideally the supports, the 12 specific evaluations evaluated by the effected --
13 MR. IOTTI: The piping analysis group never 14 conducts that evaluation.
15- MR. BOSNAK: Who has the responsibilityiof the 1<6 whole piping. system. You've got the supports and
. 17 you've got the piping, and there's always one or-18 ganization that's responsible for the-system as a 19 whole, and its performance. Piping and supports, 20 obviously one has an effect on the other.
21 MR. IOTTI: That's right.
22 kR. BOSNAK: So there's one group that does the
- 23 initial design, lays out the design specifications.
124 MR. IOTTI: That group is Gibbs-Hill. I 25 think what David is' driving at, he would like
^
[ ., ; i . - ._. ;.s. . _ . . - - - . . - - - . . - - - _ , . . ._ --
-)
16 1 Gibbs-Hill to concur to that particular support 2- which is being stated as beingstable. Gibbs-Hill 13 concurs with that. Gibbs-Hill is not responsible.
4 MR. TERAO: I would like the design group, 5 :- the effected design group to evaluate the speci.fic I
6 -unstable support in that design area from a func-7 tionality standpoint of the pipe support, it would 8 be the I.T.T. Grinnell Home Office or M.P.S.I. Home 9 office, but the effect of the support on the 10 validity of the piping anal'ysis that would have to 11 be Gibbs-Hill.
'12 MR. IOTTI: I understand. What I'm saying,
,- 13 either I.T.T. comes back and says that support 14' we. don't consider it stable, in which case the 15 - unly way Gibbs-Hill would have would be either
..t 161 to remove it or simply confirm that that support 17 -
is assumed as being stable. Therefore, there is 18 no impact on the analysis. -Ne can do At a lot 19 easier for us to have Gibbs-Hill review. I know 20 ,
what the answer is already. The reason I keep 21 asking questions, I'm not so sure I understand 22 what you want.
23 MR. FIKAR: I am confused, also.
24' MR. TERAO: There's quite a bit of difference 25 when Gibbs-Hill reviews an as-built drawing for i \
R.: i 4
ag g,g; - m% ; - _.. , .- .. - - -
g ; ..
, n,, .; .
i
'17-
- ~
- 1' the general acceptability of that support verses
-a
, 12: frevealingiajpotential concern, a specific' pot'ential '
+3 concern, of:a' specific typed support' configuration 14 .ltocassess then'that uncertainty on the validity of ij -
351 .the piping--analysis.: .
> +: . . 6 .MR. IOTTI:: What you would like us to do if
- 7) -there is a particular support'which is alleged to
.~
- . 8
- be potentially unstable, you would like to see-9 what;the consequences of that^unstability if it a ,
10 were in' fact unstable would be on the system..
..U 111.. MR. TERAO: .If there's anything-that'.the 12I piping stress analysis engineering group is aware ff 13 lof'that could either cause it to be more unstable
. . .. i
- 14.- or whether or not.they can see that.
15'. MR. .IOTTI: LOkay. I had misunderstood what ,
16 you were --
- 17 ,
MR. BOSNAK: .If you takeLa generic approach,-
18 what you're.saying is correcE, you know, the other 19 groups will-establish the stability, but if you're.
20 taking.theicase by case approach, you hdse set up
, ~
C .
- 21- tolerances, and then those have to be --
22 MR. IOTTI: I understand. I misunderstood 23 what Dave was telling me. Now, I understand.
[I 24 There's no problem for us to do that. But to a 25 larger extent, it will depend on the choice that
'" l h ^l I' '
U ,
.',..v-
~. - ..,_ _. .., _ . -. -- ~ _
18 1- we make fo,r each particular-support, because if we c '
', 2 opt not to go through this route, we'll just fix
~3 it to-conform to what everybody will agree to, if~
?4 :it'sistable, that may be the simplest way. If we
~5- choose-the other way, then I agree we'll do so.,
6 MR.>TERAO: I think.we have reached an'under-17 2 . standing.
I 10R. IOTTI: Okay.
9 MR. WADE: Do we need to talk about specific
=10 types or are we set from the previous meeting as 111'
- to what these particular types are that you have a 12 concern with? Are they exactly the-same ones we
'(
~
.13 . discussed before?
14 MR. IOTTI: 'I think it's-worth going back'and-15 . summarizing which one we have a concern so that 16 we --
.17 MR. WADE: We don't need any misunderstandings 18 of what -the scope of this is'2 19 MR. IOTTI: I'm going to refer back to the 20 affidavit, David..
' 21 ' MR. WADE: That would be the easiest way, the 22 affidavit classifies them into different so called 23 types, so we'll refer to those type numbers.
t' 24- MR. TERAO: For the type one supports, these 25' are the box frames with single structure snubbers.
.., .;. 9- - . . . .- ~. - . - . 2, _ _
19 1 MR. BURNELL: I believe you're discussing --
2 14R. IOTTI: Page nine of'the affidavit.
3 MR. BURWELL: Figure on page nine.
4 MR. IOTTI:. Figure three, page nine.
5 MR. TERAO: Figure four,,page thirteen. .
. 6 These are the modifications made to the single 7 struts.
8 MR. IOTTI: Referring to the original design, 9 which is a box frame on a single strut. We no
- 10. longer have any of those, and the proposed modi-fications or existing modifications to those are 12 listed on page thirteen, figure four, type one,-
13 - two, and three.
{
14 MR. TERAO; I.'believe from our August 8th 15 and August 9th meeting, we commented about the 16 uncertainties involved in the modifications using 17 .the index lugs and the additional strut. But the 18 modification used in U bolts'we concurred that 19 from a stability standpoint, we found that the
- 20. U bolts when added to the box frame will preclude 21 either sliding or rotation of the box frames, the 22 pipe, so our concerns are only related to the first 23 two modifications.
24 f fR. IOTTI: Okay. With regard to the index 25 _ lugs, at least I am prepared to tell you there is
p- =:
. . k . g >$4_ g '. -- ~ ~ -
c,'
I' ,
[
^
20
- 1. only one. We haven't determined what we intend to i '
2 do. .And I think one of the things.we may want to t
3 do is go look at it in the field. This is maybe 4: one-of those situations where geometry is such that 5 it can't go anywhere in any case, and that which 6 way we're going to proceed --
'7 MR. BOSNAK: Where is that location?
8 MR. IOTTI: I don't recall, but we'll take
, 9 you there.
10 MR. TERAO: Let's move on to type two.
11 MR. IOTTI: Could you again refresh my memory 12 on your concerns for the type two, the ones for 13 the additional struts.
4' There I don't think it's 14 -so much lateral or axial movement.
15 MR. TERAO: Both of those were either the 16 . uncertainty involved in accounting for the friction.
17 MR. IOTTI: Unfortunately I don' t have the
- 18. total number of these, but t6ey are few. Okay.
19 So the second type is the one that had --
, 20 ,
MR. TERAO: .The type two or the U bolts-l L 21 single struts with thermal gap.
22 MR. IOTTI: Figure five, page fifteen of tla p
23 affidavit.
24 MR. TERAO: The modifications that ysu have 25 proposed were to either sinch U bolts or to add l-L
- ..ri ...4, m ._ m .,_ __
,j , . . , . , .
21 1 supplenentary steel.
2 Right. -
MR. IOTTI:
3 MR. TERAO: I believe our comments from the
- 4 stability standpoint on sinched U bolts, we-felt 5 that your testing and analysis performed under,the
. 6 different dispositions adequately precluded the
'7 ' stability, the instability concern with this parti-7 8 ,cular support, and with. respect to the bumpers, 9 the supplementary steel, although we did not agree 10 with that fix,. resolved the stability standpoint, 11 the stability concern, your analysis, which showed
~12 that even without the supports on the main stream J. 13 line, the piping stresses were still acceptable to 14 us, makes this a mu.te point in the sense that I
-15 think-it's not worthwhile to argue whether or not 16 .this is an acceptable size of restraint if the 17 . analysis, even with this restraint excluded, shows 18 the stresses to be acceptabfh, and I don' t think 19 we should argue the point any more.
20 MR. IOTTI: I agree.
21 MR. FIKAR: Good. We can make some copies of 22 this, of the affidavit.
23 MR. IPPOLITO: We've got one.~
f 24 MR. TERAO: I believe the one with the 25 bumpers, there was'only one of those in the entire L -
C' 22 1 plant. .
2- MR. IOTTI: That's correct.
3 MR. TERAO: Type three, unstable supports, the 4 ' double-strutted frames. And as we pointed out at 5 the last meeting, I believe these are either double 6 strutted or triple strutted, but the concern being 7 addressed in the double-strutted frame dispositions 8 related to really the single plane -- hard single-9 plane frames that are supported by 'trutss or snub-10 bers, probably struts, I doubt if they are snubbers.
11 -
MR. IOTTI: To the best of my knowledge, I 12 haven't seen any with the snubbers.
13 MR. TERAO: These we didn't know at the time 14 how many large single-plane frames there were.
15 MR. IOTTI: There'.s a total of three, and they I 16' are all triple struts -- yeah, they are all triple
-17 struts. And again, I would propose that before we 18 do anything else, we go look'at them. There are 19 some in the odds building and some in the electrical 20 building or_near.the electrical building. I have
-21 the drawings. We can go to elevation below eight 22 seventy-three and look at all of them.
23 MR. TERAO: We still have a concern on these 24 particular frames.
25 MR. IOTTI: I guess what we had discussed, was k.
w 9,;, q. . in n. . ~ . - ~-: -L--~ - - - ' ~ ~ .
-~. us. --
b h 23 t-g 'I an option of.doing a bounding analysis to show that
- v. -.=
p' "- 2 even if the frame were to move,-you wouldn't have 3' reverse consequences. After we look at'them, we-4 lmay stay with'the same option, we may change ~the 5 - option. In any case, I think it would be worthwhile
.. 6' for all~of you to see the supports, these gang 7- supports.
- 8. MR..VOLLMER: We saw two of them yesterday, 9
we didn't see.the other one.
L 10 ' 'MR. IOTTI: --Those are it.
11 iMR. TERAO: But our concern with the home i~
11 2 . offices of I.T.T.Grinnell and M.P.S.I. and Gibbs-
-13 Hill to' comment on.these frame designs not only 14' through a bounding analysis --
15 MR.'IOTTI: Well, I guess it would apply if 16' our intent is to do an analysis. If our intent is 17 .to change them, of course, I'm not so sure whether 18 it would be worthwhile to go'back.
19 MR. VOLLMER: Fix them where they are?
20 MR. IOTTI: I guess I have to confirm how F
21 many of these double-strutted box frames there are.
'22 MR. BURWELL: Off the record.
23 - [A short time was had off the record. There-24 after, the following proceedings were had.]
25- MR. BURWELL: 'On the record, please.
e;- :
+ -g. .
.- ..n.+. + - ~ . .. . . - - . - ~ . ~ . .
24 1 MR. TERAO: The type four restraints are the 2 single-strut snubb'ers with snub U bolts. From the
.3 last meeting I believe our comments were that from 4 a stability standpoint and from the testing and 5 - analysis that were performed on the U bolt sinching 15 summary. disposition, we found that these restraints
' 7. were stable. At this point, what I would like to 8 do is to go back to our agreed upon definition of 9 instability and.have the applicants then identify
- 10. if there's any other supports that fall under that 11- .
definition that are not included under these four
- 12. categories.
+
13 -
MR. IOTTI: Dave, I don't have the answer to 14 that question. We are going to be looking. We 15 don' t think there are any, but we'll have to go back 16 and provide you with that answer.
17 MR. VOLLMER: Part of what we .said we would 18 ask you for was a program us'ing this definition to 19 go out and make sure that everything ha'd been 20 looked out and identified.
'21 MR. TERAO: So then there's the fifth cate-22 gory that we don' t know what's in there at this 23 time?
24 MR. IOTTI: Whether there's Any?
25 MR. TERAO: Whether there's any, if it per-
b-- -
m t -
l, 25 El forms under our agreed definition even it should be 2 addressed in this fifth category.
3 MR. HORIR: The point I would like to make
- 4. is that with respect to this fif th category that. is 5 .an area where we are getting beyond the scope of
. 6 the actual-motion itself. My question to you, I 7: -would like to get your opinion, we have our opinion, 8 .is that when we go into areas that are beyond the 9 questions that we are raising in the motion itself, 10 perhaps.it would be best to address those on a
- _ 11 separate track as opposed to leaving the motion -
12 itself awaiting a resolution until we get to those?
'13 :MR. TERAO: I believe that's a legal questien 14 that our lawyer should be answering. Unfortunately 15 we don' t have our lawyer with us today.
16 MR. IPPOLITO: Let me pursue that. As far as 17- ;I am concerned I think at this meeting we finall'y 18 defined what stability is. /The motion concerns 19 the issue of stability. Now that we understand 20 each other, it's -- to me it's everything that fits 21 within that definition. I mean to me it properly 22 belongs to your motion.
23 MR. HORIR: The motion addresses the types 24 of supports which were identified by the inter-25 vener as being potentially unstable, and the motion
e .
26 1 - was filed as part of a plan to resolve the inter-2 vener's allegation'.
jL MR. BOSNAK: We're interested in making sure b
4- it meets all of the requirements.
5 MR.-HORIR: We don't disagree with that at all.
6 We're just questioning whether the proper attack is 7 to include that within-the motion itself.
8 MR. IOTTI: We keep talking as if there were-9 some that fall within category five. Why don't we
.10 ' find out if there are any.' I don't.think the.re are
~
11 . any. The other thing I would like to suggest, we
-. 12 define functional ability by the definition, be-13 -- - cause someone along the line is going to challenge
. 14 our definition of stability.
15 MR. BOSNAK: We would prefer that.
'16 MR. TERAO: The way I desbribed these supporto ,
17 I said that the word unstable is ac.tually a mis-
. 18 nomer when applied to pipe s3pports individually, 19 that really the concern is that the precariousness 20 of the support that there is uncertainty, and it 21 does not necessarily fall under the classical de-22 finition of stability -- instability.
23 MR. HORIR: Tom, before we leave the point, 24 could we get an agreement that will at least dis-25 cuss with your legal staff, the proper approach for
27 1 this stiff category.
2 MR. IPPOLITO: I'll assure you that's what 3 we'll do.
4 MR. TERAO: Leaving our established definition 5 of instability on individual pipe supports, ,
B h? J .1 6 funtional ability, precarious supports.
i; 7 MR. IOTTI: It doesn' t matter- what we call it 8 here.
9 MR. TERAO: From our last meeting, we did ask 10 for some further information on the acceptability 11 of the supports with the extended mass that we are 12 still looking for, but I do want to point out that
.( 13 it doesn't fall under this specific category of 14 " unstable pipe supports".
15 ~ MR. IOTTI: .I'm sorry. I am missing a point.
16 Maybe'your name -- the names you have given ex-17 , tended mass.
18 MR. TERAO: There was is extended mass with a 19 trunnion and U bolts holding that extended mass on 20 to the pipe.
21 MR. IOTTI: Yes. I understand. We are 22 scheduled to have the information to you at the 23 end of next week. Okay. We haven't forgotten. I 24 just needed to refresh my memory as to what it was t
25 that~we are talkin'g about.
.__g__,_ _ . . . ___ _ _ _ _ _.
28 l' MR. IPPOLITO: Does this mean this is another 2 technical issue wh'ich.is or is not within the scope 3 of the summary disposition?
4- MR. TERAO: No. I'm saying this is within the 5 scope of stability, but it's not under the scope of 6 our established definition of instability. This 7 actually falls under the classicat sense of the 8 definition of instability that indeed these sup-9 - ports are unstable from a classical sense, and we 10' just wanted some verification of what was done to 11- . justify that design.
12 MR. IPPOLITO: It is within or out. Do I
. 13 --
have to talk to the lawyers as to whether or not
~14 this is to be discussed as our responsibility to 15 the summary disposition or do' we put that in an-16 other box?
17 MR. HORIR: That's another one.we think should 18 be in another box. It wasn.'t addressed within the 19 scope of --
20 ,
MR. BOLTZ: .May I ask a question? On the 21 responsos within the scope of the contention that 22 the utility is going to make the NRC, does case 23 receive a copy of this or is case going to receive 24 a copy?
25 MR. IPPOLITO: Yes.
r-
ry
. , - . + . ,-
,~, . . _ . . _ _ , _ . . ~ ,
29 1 ~MR. BOLTZ: And_ matters outside the scope of 2' the contention, if there are any, depending JLf there
'3 are any supports that fall into the category,.and
-4 depending on what the two lawyers do, would case get 5 a copy of those? ,
a 6 MR. IPPOLITO: Yes.
7 -Let me propose the following: We have not had m
8: an opportunity to caucus with John. What I propose 9 is like a twenty minute break caucus with John and 10 we would proceed with the generic stiffness as we
~
11 have hopefully proceeded with the stability ques-12 -
tion. Okay.
- 13 MR. BURWELL: Before we go off, may I raise 14 the question, I believe Dr. Iotti recommended that
'15 we go look at the gang hangers. In any event, do
~16 .w e desire to do that or not? 'I'm not comfortable 17 .that a decision was reached on that.
18 MR. BOSNAK: We would 1The to go later after 19 this is over this afternoon.
20 MR. IOTTI: Not just the gang hanger, but
'21 also what we call the keyed box frame if you would, 22 index lugs.
23' MR. VOLLMER: I would like to see the shield 24 as a separate issue, that is a separate issue. I 25 have time problems'.
t
r: 1
.._ ,- _. , _ r_ __ . . - . - _
30 1 MR.fIPPOLITO: Okay.
2- MR. VOLLMER: I also would like to _ set it up, 3 I have a five o' clock plane.
4' MR. FIKAR: That has nothing to do with this 5 hearing. s.c 6- MR. BURWELL: What have you just told me? Are 7 you saying you want to do it now or do you want to 8 wait or try to do Mr. Fair next?
9 MR. VOLLMER: I would like to do Mr. Fair 10 next, and if it could be handled as expeditiously 11 -
as that was.
12 (A short time was had off the record. Thereafter, the fol-13 lowing proceedings were had.]
14 MR. BURWELL: I believe we are ready at this 15 -point and time to discuss the motion for summary 16 disposition entitled Applicant's Motion for Summary 17 Disposition regarding use of Generic Stiffnesses 18 instead of Actual stiffnessef in Piping Analysis, 19 my accent on that, that 's A-c-t-u-a-1, actual 20 , stiffnesses in piping analysis. This motion was
'21 filed on May 21, 1984. I believe we received some 22 . addit.'.onal information on this item by letter 23 ' dated July 16, 1984.
24 MR. IPPOLITO: Okay. What I'd like to do is 25 to make sure we are all at the right -- all have e
3 3 -. ..
. 31 1 the ' correct understanding of where we are at this i' 2 point in_ time, because we hav'e had again a number p 3 of meetings. I would like to have John Fair 4 summarize for the staff where we believe we are 51 and see if we all agree, and then we will proceed w 6 'from there. Okay. John.
7 MR. FAIR: Okay._ The applicants have used
, ,.8 generic stiffness values as a modeling assumption-9 Lin piping analysis for class two and class three
~10 piping systems. In designing these supports, how-11 e v e r', they used a deflection criteria with the 12 applied load from.the piping analysis. Therefore,
.y n 13 there was no direct correlation between the modelinc' -
14 assumption and piping analysis and the supports 15- design. The iten raised an issue with this assump-16 tion used for the support design whether it ade-17 ,quately represented the actual support stiffness'es 18- as built in tne field. In dider to answer this, 19 the applicants analyzed a few sample piping analy-20 sis problems in their summary disposition. In one 21 of the-piping analysis problems, it turned out that 22 a support was overloaded by approximately fifty- ,
23 seven percent. This staff did not agree that this 24 set of analyses demonstrated adequacy of the gene-25 ric stiffness used'. The applicant did another l
l 1
n._ ., _ . . , . . _ _ _._ _-
.5-
, 32
'l sample of four piping analysis, and again identified 2_ one case where the supports would be loaded above 3- the allowances, At the last meeting, we were still 14 - not in agreement that this demonstrated the adequacy 5 of the design practice on stiffness. Therefore, the 6 applicants were going to.go back and propose an 7; additional screening criteria to try to determine 8 which supports were of concern and how to resolve 9 the issue.
10 MR. IPPOLITO: Is that where we are?
11 -
MR. IOTTI: That's where we are. This is
- 12 . precisely correc:. Yes.
13 MR. IPPOLITO: I hope the applicant is able to 14 respond.
15 MR. IOTTI: The applicant is partly able to 16 respond, because our schedule to respond fully to 17 you was to respond by the 30th. Unfortunately, it's 18 not the 30th. I have some mfterial with me that I 19 can -- if you wish to discuss it, but I uould like.
20 ,
to go back a little bit to the beginning. Applicants
'21 have submitted all of these analysis, not because 22 they believe there was anything wrong with using 23 the generic stiffness, that s being used by other 24 applicants in other plants, but there was a con-25 cern expressed both by the intervener and by the
W _m W
mz '
- w:- ,
- g. ;.
.= - - - .-
_ i '
33 r
a .
~
.[
"M-
~
1 -
staff. We still don' t'believe there's anything V 16 .
( 2' : wrong with-the approach ~of.using' generic stiffnesses .
At 7' i3 -You have to understand that:some of the results that
. ,- 4: 1 have been achieved 'by the, most recent-analysis, . stil1~
'5- have very significant conservatisms within them,
..- f6' ' which we have chosen not to remove because'.we think
~
.7 : 'they are still applicable. We could still have the '
8' opportunity-to go-back and' remove the conservatisms,
- 9 andithose results will probably show no overloading
.10 of supports. I say probably, because we haven't-11 done'so. Just my own opinion. - - We have to distin-guish, I think, between what -- I don' t want to, cal]
$ 12-$ .
- .p -
13 it a standard industry practice, because. unfor-I l<4 t ~tunately there doesn't seem to exist a real-stand-
~
15 ard industry practice:with regard to using generic 16' stiffnesses.with-class two and three piping. Every-
- 17- . body' does it in a different manner. In this regard 18 we are no different than mady. A lot of people:
19 just-use the assumption that these supports are-20: tctally rigid. It's the very same people that em .
21 ploy snubbers, the same snubbers wa are employing
'22 here. One of the conclusions that we have arrived 23' at from our analysis that the snubber turns out to 24 be the weak link in determining the actual stiff-25 ness'of the system', not always, but in a large num-o
. , :, . - . . . .- - .a ..
34 1 ber of cases. So again, let's distinguish those 2- two facts. Our analysis was done in the same 3- manner as being done in many many plants, the ma-14 jority of plants in this country including the 5 plants of the vintage of Comanche Peak. What Me 6 then' have done is attempt to demonstrate even if 7 you were to use the actual stiffness, the same 8 approach employed there would be no safety concern.
9 It is true we have identified for some of the 10' analysis that some of the s'upports if you were to
- 11. . do the analysis 'with the " Actual Stiffness", and 12 you have to bear in mind when we're referring to
~13 --
actual stiffness, we're referring to calculated
- 14. stiffness, which may or may not be the real stiff-15 nesses. They are out there. Okay. These are 16 calculated stiffnesses. Then you might calculate 17 some of the supports to be overloaded. Those that 18 - we have found so far, we believe even though over-19 loaded in the strictest sense, would not fail. So 20 again we don't think there is a safety impact. But 21 we also found as part of our work, that invariably 22 the supports that turned out to be not just over-23 loaded, but overloaded by the rated load, actual 24 load verses rated load, turned out to be snubbers 25 and also snubbers which were initially very lightly
~ _
35 1 ' loaded. The combination of the light loading is 2' what' prompted the stiffness of this particular 3 support to be~much lower than the generic ~ stiffness.
'4 Let.me take you through-a numerical exercise if you
'5. will. Let's take a particular support and le.t'.s J6 suppose that'this' support is comprised of a frame 7 and snubber attached to that frame. Because-the
.. ~
8 applicant uses the deflection criterion, they 9 would take the load as determined from the piping 10c analysis,-and the load in the piping analysis you 11 have to understand has implicit in it that the 1
- 12 seismic deflection of the piping is very very small.
'13 --
For instance, an average seismic deflection of 14 piping at the support point is generally left 15 in six mil, that gives you a load, but that load is 16 applied, as John Fair so well put, to a frame that 17: has to satisfy a. deflection criterion of one-six-18 .teenth. Let's suppose that'Uhat i load turns out to 19 be a low load.
If you take a low load and you use 20 a one-sixteenth criterion of the frame, you can 21 have. a frame that's very very low stiffness. Let's
~
22' say'the frame deflects .06 inches, close to one-23 sixteenth, to get stiffness, you just divide the
.24 load by the deflection. So if the load let's say
-25 is fifty pounds, you divide fifty pounds by .06.
~ } I. . 5Q -'-
~
..y
,.;. .y_ e-
, u. e *
-36
- 1) You end up with a stiffness of nine hund ced peunds
. 2: per inch. ' Let's say that this is _a four-inch line.
3 ~. The generic' stiffness -- I think it's two times ten
.x '
4 to the fifth. Two times ten to.the fifth. So right l
5= there:you.have more than two orders of magnitud,e 6/ change in stiffness. Of course, when you re-analyze -
s,.
'7- . with.the actual stiffness,'you may see the load of !
l8 that particular support increase that drastically..
9 on the'.other hand, if you don't-load it to fifty-
.10 pounds, so you don't get. increase of a. factor of i
11 ten,Jso you don't load five hundred pounds.. If it 12 shows'a snubber, let's say.a PFA one-half snubber, t i
( <
13, ' you. increase a factor of ten. This is all hypo- .
[
14 thetical. We haven't seen this CASE, I am just 15 taking you through a numerical exercise. That snub-16' ber would be calculated as oeing a load. So what 17l that does,..what I'm telling you, is that that snub-u; _
a 18 ber wasn't needed to begin with. The load was so 19 low you didn't need support to begin with. If you
-20 .take that support out,.nothing happens, stresses in
[21, the pipe are still okay. The other supports acco-
'22 modate the-remainder of the load. What all of our m 23 -sensitivity studies have shown that type of support 24' tends to be predicted as being overloaded. Our y' 25 supports, which ard initially low loads and tend to I
f
I l
37 1 be standard components, which have rated low, and 2- the designer rather than using a conservative ap-3 proach will take the standard components which is 4- next to it, so you normally have a little margin be-5 cause of tire particular set of circumstances., .;And 6 this is what prompted our suggestion that we could 7 arrive at a set of criteria that would enable us to 8 determine without doing an analysis which particular 9 supports per pipe size might be candidates were you 10 to take an analysis with these actual stiffnesses 11 -
for potentially overloading. I am prepared to dis-12 cuss some of the initial findings, for instance, we
, 13 don't need to look at any lines that are four inches 14 or less. We just -- just no way that the normal 15 design process would lead you to under design sup-16 ports now, and then in lines that are larger than 17 four inches but le'ss than twelve inches, we would 18- be concerned with, if the ap$11 cant has used snub-19 _ bers like a PSA one-half or PSA three-fourths, be-20: ,
cause the loads were very low. Then those snubbers
~21 may not have significant capacity to accomodate 22 other load increases that might accrue out of actua]
23 analysis. I am just describing for you the set of
'24 criteria you will come up with. We have not fin-25 'ished-the work. Even if the work is completed at
q
_ m _.,... -..y,_. _".'
p
_,' 38 1 I least what I have done, somebody has to check-it, 2- verify.it. We need to try it out and see how well 3: -it' works. That hasn't been done yet. As a matter 4' of fact,-I was going to complete that and call John
~
5 !and sit down and explain it to him'what we'have
" .: 6 Edone and see if he would agree with it. So'that's i . 7 .where we stand at that point. I would just like to r Si reiterate that.so far none of the results-that we 9 ,
have'seen, would tell you that we have a safety
- 10 problem. In other words, even if we were to remove
[11 the snubbers that turned out to be overloaded, the
.12 stress analysis would say your piping system is,
.' 13- okay, the remainder of the supports.are capable of' 14 '~ accepting the loads. And to us that's really what
~
15 it boils down.to.. There's no question in our minds 16, 'that .the generic stiffnesses approximation--- it's 17- ..a. convenient approximation that's been made by many 18 others and ourselves. WhethEr it's a good assump-tion, whether it's good' engineering approach to be.
~
'19 20 taken for the' design of the piping system, depends.
k '
21 .on whether if you use it you have a piping system
' 22 ~ Lwhich is safe as opposed to having a piping system 23 tha't's specified to the inth degree. You will 24 :never get there. If we were to use the actual 25; stiffnesses, those'are not likely to be the actual 1
0 - . _ _ _ _ . - _ . _ . . . _ . , - - , . . . _ . ., . . _ , , _
v 4 ,y -- - . . . .
I
~
39 1 stiffnesses that you will see on sight, because 2' th'ere is a small gap in any support or there is a 3 small place. In reality, the true actual stiffnessea 4 bear only some resemblance to calculated actual 5 stiffnesses, so we .should not lose track of th,at 6 also, and that's why we thinx at that point, we've 7 gone as far as we can go other than completing the 8 lastest efforts that we've committed to do, which 9 we'll complete by the end of next week. I guess I've 10 said my piece at this point.
11 MR. WADE: It boils down, we think that we're 12 well withih what has been done in the past in the 13 --
' industry and that because of inherent conservatism-14 in the piping analysis that our approach is per-t 15 ~ fectly acceptable. . We have no reason to believe
- 16. o therwis e.
17- MR. IOTTI: There's one other item that I 181 .would like to ask.for completeness. Where the gene-19 " ric verses actul stiffness has the most pronounced 20 effect on the results of the piping analysis and
~21 the possible safety implications of piping analysis 22- in the' seismic analysis. You have to bear in mind
~
23 that this, because we also submitted affidavits on 24 safety factors. The reason we did so is so that
, 25 at least you will be aware that the seismic analy-e
. ,, ,,7..
40 1 sis has inherent conservatisms which are present 2 ' in the analysis that are being performed. So what 3 I'm saying is there is conservatism in the analysis.
4 There is different types of conservatisms, some is 5 in'the method of analysis itself. But in addition
. 6 to that to pinpoint to the analysis the input mo-7f tions are already conservative. It's true you.can 8 even play this game. The problem with generic 9 verses actual stiffness is that you can play this
'10 game forever. In other words, we can go back and 11 analyze every piping system in the plant, and two
~12 months after we do, we're analyzing them again be-
/
. s.-
L l3 cause some information says, "Well, the stiffness i
.14 there is really not.what you should assume, you 15- should assume something else."
.i g 16 MR. WADE: Two different analyst can come up 17- ;with different results' depending on how he models
.18 -the piping system itself. TNese things all come in 19 to play. Simplifying the assumption in generic 20 stiffness is certainly' acceptable with the conser-
'21 vatisms that are built in to account for those un-L.
22 certainties.
23- MR. FAIR: I said before that I agreed that 24 generic stiffness was a standard practice used by
.t-
'25 . the industry. I have no problem with the generic
41 1 stiffness using those assumptions in piping analysis 2 I disagree with your statement that you do it ex-3 actly like the rest of the industry. There are 4 variations amongst the industry. Some of the in-5 dustry actually calculated the supports stiffne,ss by other stiffness criteria.
7 MR. WADE: A lot do not.
8 MR. FAIR: I disagree, that having to evaluate 9 the stiffnesses.is a non-ending loop. We have 10- established that you can va'ry quite a wide range 11 . with those assumed generic stiffnesses with no im-12 pact on piping analysis. And as we have already 13 --
identified, it's just those cases that come out
(^
14 very lightly loaded of supports that have very high 15 flexibility that can have an impact on the piping 16 analysis.
17' MR. FIKAR: It's a non-ending loop, Tom. We 18 filed this thing in May. We're going to load fuel 19 .in six weeks. We ain' t there. We would like to get 2'O there if there's,some -- if we could narrow that
'21 down, let's close that loop, whatever it is. You 22 know, it's really not -- it's, I guess, the way 23 some sampling was done or some analysis.
34 im. IOTTI: I think it's categorized correctly, 4
25 I didn't mean an unending loop you and I could
q
~
g-. #. - ..'
, . -.. . . ..e 42 1 Jnever agree as to when we would finish. Rather I it am saying there will always-be a third person come 3: in.and say, "I disagree of how you've done_the cal-14 Lculation of stiffness, and you'll have to do it over 5- again." .And in that sense, it's not something .that
- .. 6 can'be precisely addressed.
7- MR. FAIR: We're not talking about precision
, 8_ here.- We have established that within your -- with-9 in an order magnitude of your generic value, you 10, didn' t have any -significant impact on piping analy-
. 11 ' sis based on.your samples. I agree with you that 12! .
you cannot' precisely calculate the stiffness of l/;; . 13 supports in'the field to three decimal places. But
-14 in order of magnitude, is a reasonable number.to 15- shoot for.
16 'M]R. . IOTTI: Well now, I would agree with you,
, 17' but-you can certainly calculate it'within a factor
- 18 of two. I .think our disagreements are getting less w
19- 'and-less. ~ The question is what. do we do next.
~20 MR. WADE: We need a conclusive practice to put 21' this issue to bed so' that when we finish -this effort p /. .
i22' we know that results~are' going to be the'end, and L
'23' there'is not going to be another loop that we have
- 24 to go'through.
(25 - MR. IOTTI: Ou'r intention is to complete this 1
, .,7 . _. _ . _ ._ u.. _ _ _ __
43
' 1' designing criterion definition which will tell you 2, how you would identify per pipe size the supports 3 which could be susceptible to overloading were you 4 to do an analysis. The designing criteria will ex-5 clude certain pipe sizes, because there doesn't.seem 6 to be a way in which you will ever get into an area 7 of specific. loads at which -- there is no reason to 8 believe that there may be some supports as being 9 overloaded.- The load itself is not the criteria, 10 the load-in combination witIh the type of standard 11 -
components is really the criterion. For instance, 12 there will be variability loads. Let me give you
( 13 an example. We may say for a sixteen inch line, 14 any support which -- if there are any supports which 15 are loaded less than three hundred pounds, we ought 16 to look at that particular stress problem, because 17 there would be a support which will.have both a very 18_ low stiffness, something of-Ihe order of two orders 19 of magnitude less than generic stiffness. That 20 particular support happens to be a standard component
~21- instead of PSA3, which you would expect to see on 22 size line maybe a PSAl or one-half then that could 23 be overloaded. That's the type of criterion we'll 24 have for*you. I want to confirm that's what we're 25 gearing to.
~
p .
n ,. i g:. . - .. a.. -. .
's r ** ,
44 1: MR. WADE: .What is your acceptance criteria
~
yp:, 4 2 for all that? What does it take to satisfy you.
jc 3 MR. FAIR: We would-like to see what you are 4 proposing. I am in agreement with Dr. Iotti that'
. :5f the~ correct-criteria is the lighly loaded support-
.. z-
.; 6 -and it should be based on pipe size.
7L .MR. IOTTI:
To go further, what then we pro-
'8-pose tio do is, one, we id.entify that particular
,9 ' stress problem that is a candidate to have some of
~
10 these supports. We have two options to go, one of
~
.11 them to take that support out and rerun the analy-
~
12 sis without calculating the stiffness,.take out,to
- [.
13 see.if anything happens. Okay. If you-don't need s.
14 it to begin with, who cares. The other option is 15 actually calculate the actual stiffness and re-
,16 rou'te'that particular stress: problem and then de-17 stermine what the consequences would be,'and they
' 18 might chow us an' overload or"not. And what Dave is 19 .trying.to" add that wa're shooting an elusive target, 20 we don't exactly know how far you want us to'go in
'21 this or what would be acceptable to you. In other.
_ a 22' words, where do we draw the-line? We will review
.23: every stress problem from this standpoint looking 24 at the results, identifying the stress from this 25- criteria, define wiiich stress problem may have to
g g ....m -.- . . , . ~ . . .
e 45
- 1. be looked at again.
That far we can go.
C 2 MR. WADE: We don't want to proceed with the 3 re-analysis until we know what the accepted criteria 4 is, once we have that, we can proceed beyond that 5 '.-
point. ..
- 6. MR. FAIR: We certainly have acceptance crite-7 ria for the design supports and pipe at the facility .
8- I don't see there is any question of what the o
L!
9 acceptance criteria is.
10 MR. IOTTI: I think he's looking at something 11~ -
else. Let's say there are two thousand stress pro-12 blems in the plant. We've seen seventeen thousand if ' 13
~' '
supports, Okay. And we can look at all two thou-14 sand stress problems. That's one way to do it. You 15 - can only-look at a percentage of those and draw a 16 conclusion that you -have a certain confidence. Say 17 if you look at two hundred and fifty of those, and 18 you find nothing within two-Eundred and fifty with
. 19 regard to safety, you can draw a conclusion, you car.
20 say, "Well, if I were to look at the other .one thou-t
'21 sand, whatever, one thousand and six hundred or so, s
22 you would find no more than maybe a certain number 23 of supports which would fall in that category." You 24 can say that with a certain confidence level. What
-25 he's saying, what is the criterion? One hundred
r =
g ., .. ._.
_. _ _, _ ~ __. __. .- . --
[ .
46 cii -
1- -percent confident there is none, ninety-five' percent
-2 confident there is less than one percent.
~3 MR. FAIR: You just said:a couple of thousand L4; , . stress' problems. How many. exactly greater than four f5 _ inch piping analysis problems. .
. :6L MR. IOTTI: I don't know, that's part of the
-7 stuff I.have got to do next week.
-18 MR. FAIR: I thought it was in the order of a 4
9 '- couple of-hundred.
~ '
3 101 -MR. FIKAR: We don't_want to do everything and 11' satisfy _there is no problem. That's a waist of J12' time. Maybe I am addressing Vollmer and Bosnak.
..3e 1
f( 13 - Go out there and look at th'at plant.- You have
~14' never seen hangers as massive in any plant.you have 15 been'into as you have seen here. The seismic. zone 16' is.the. lowest-in the country. Conservatism, con-17 .servatism, conservatism.- We don't want to waist-L18 our resources doing a bunch-$f studies which we are 19 absolutely. confident are going to come out all 20 - right.. That's what we're pleading for. We didn't
~
21 put_ snubbers in, because we didn't think they were l 22 needed based on our criteria. Again if you want to I
_r
- 23. re-analyze'and somebody says you can't do that, )
I
'24 'you'll probably analyze that one way. I don't want !
l 1
125; to give the impression we put something in there we ;
i
^
g . _ , , - _. ,_ _ .__ ._ __ _.
47 1 didn't need. And it's generally standard industry 2 practice. That's what's concerning us.
3 MR. FAIR: I haven' t agreed in total that your 4 total design practice.being standard industry 5 practice all the'way through. I believe there.is v.
6 part of it that do comform with the standard in-7 dustry practice. The other side is as far as 8 acceptance criteria once we determine a proper 9 screening criteria to look at the supports where -
o10- ever we come across the sup' ports, we ought to agree 11 on --
12 MR. IOTTI: Okay, I mentioned two options,
< 13 ~~
one of them is re-analysis, the other one is re-14 moval and re-analysis with generic. Where the 15 most uncertainty comes.in, in my opinion, is not 16 re-analysis with the generic stiffness, because 17 that's standard method. But if you then get down
'18 to the calculations of all the actul stiffnesses-19 for that'particular systems -- I mean what's the 20- nerit of doing that if your approach is to just
'21 take that support out.
22 !)R. FAIR: Well, I'm not in agreement with 23' that statement. If we had determined from your 24- affidavit and studies that there is a range of 25 stiffness which has no significant impact on the
48 1
1 piping analysis, then why do you have to calculate 2 the stiffness of every support that passed your
-3 screening criteria.
4 MR. IOTTI: Then I don't have any disagreement 5- with you. Then we'll just either take the supp, ort
. 6 out or -- in other words, it's our option at that 7 point as to what we're going to do. So I guess what 8 we have agreed to do, is that we'll wait until we 9 formalize the screening criteria. I can go over 10 with you my scratch notes. I haven't formalized 11- them'yet. You will know all of what it is you're 12 going to get at the end of next week, it'will come 13 as no surprise. Particularly if you don't agree
~ 14 with it, I would like to know now, because it will 15 save me an exercise-in futility. I have tested
- 16 over a-few piping systems, it seems to work. In
-17 ,other words, I have been able with this criterion,
.18 to come to the same conclusi6ns as I did with all 19 of the analysis that has been submitted to you. We 20 would predict a few more supports than what we have 21 actually found to be overloads. It would identify y.
22- the stress problems that we analyzed but found E
j_ 23 nothing wrong with. So at least it meets that test.
I 24 But I want to do it a little bit more. What I 25 don't have is how many stress problems we'll end up p
l' t
Y
- g- -
49 C
- 1. with when we apply this screening criteria. I don't 2 know that yet, because that's the part that we'll 3 be doing.next week. He'll go back and look at 4 every stress problem and identify the supports, and
~5 then either include or exclude that particulap.,
'6 . stress problem.
7- MR. FAIR: Let me ask you a question. If you 8 were to select the option where you would just as-9 sume the support is not in the analysis, what would 10 you do in the field with that support?
11 MR. IOTTI: Remove.
-12 MR. FAIR: Okay.
13 MR. IOTTI: I don't want it there., If we 14- can' t remove it, then that option is not viable to 15 us. We would re-analyze the actul stiffness.
16 MR. FIKAR: Well, I-am concerned we are going 17- to get involved with a whole bunch of re-analysis 18 and overly cautious concern-i~ and it's going to cost
, 19 us time, money, and effort. That's my problem.
20 You all after having hearing this, maybe we can get
~21 together.
22 MR. IPPOLITO: I'm going to say what I think 4
23 I-heard, and-I will demonstrate my ignorance of the 24- subject. I think I can summarize what I'm hearing.
25 I summarize what I'm hearing is that you propose to
.. ; -. . ,.A = , . :- . . - .-. . . - . - -
50 1 use a method of analysis to consider generic stiff-
.2 ness.
3 MR. IOTTI: Well, that method of analysis. It
<4 was reviewed. It was approved'by the F.E.R.
-5 MR. IPPOLITO: Fine, but in going back and,
.- 6 looking, we find that generic stiffness, does that 7 envelope all of-the system or all of'the supports?
8 The question is now that we know that, what do we 9 do~about it.now? I change that statement. Now that 10- we know that, what are you going to do about it?
11 And what upsets me a little bit about this conver--
121 sation, I'll tell you frankly, you seem to be tell-
/J K13 ing us, the N.R.C., tell me how much is safety. I 14 think the applicant has to sit down and say, "Given 15- this problem, how.do I resalve this problem." Okay.
.16 And then we'll review that - probleN, I mean your -
~
17 recommended solution to that problem.
18 MR. FIKAR: I don't eved know if we have a 19 ' problem.
~20 MR. IPPOLITO: That's fine.
21 MR. IOTTI: I'll make a stronger statement 22 than that. We have done enough in our opinion to.
23 have satisfied ouruelves that there was no safety 24 implications, so we didn't -- if we had taken the 25 attitude that you a're trying to characterize us
g , - - ~- ~ - - ~ . _n-- _ _ . . .. _ , _ _ _,
51 1 with, what we would have done, we could have stone-2 walled this. We have done what other industry has, s .
3- why should you do anything with actul stiffness, we 4 .didn't do. We put in the actul stiffnesses. We did 5 sampling, and on that sampling basis, we determined
, 6 -- applicants did, and I concur with that. I am 7 'not the--applicant per say. There is no safety con-8 cern. That's my opinion at that point. How much 9 further, well, I can go further. I can do more
-10 samples, and again I expect' the result to be the 11 . same.
12 -MR.~ VOLLMER: You drew that conclusion based I / 13 -
on the fact that even though you had supports over-14 loaded, it was your presumption that enough con-15 servatisms existed so that th'ere would have been no 16 failure.
17 MR. FIKAR: No safety issue.
18 MR. VOLLMER: Hear.my words. I said you drew 19 a conclusion based on an assumption that otherwise 20 existed. I would like a demonstration of that.
- 21. MR. IOTTI: Well, it was based on two things.
22 One, that even though the supports were overloaded, 23 it wouldn't fail. That was one thing, because the 24 type of overloading we saw was not so large as to l 1
.25' cause the particular standard component in our L I
7 ,. __ _
52
-1 opinion to really fail, clearly' exceeded the rated 2 load. And the second portion of our position is 3 - that there_ was inherent in the input of analysis and
. the. method of analysis conservatisms that far out 5_ weighed the nonconservatisms introduced by these
-- 6 gneric stiffnesses. I want to make a statement that 7- I don't think.it's fair to characterize the ap-8 plicant that they don't have the concern. They did
-9 what they thought was sufficient to satisfy the 10 commission; there was no safety problem. What we're 11 disagreeing is the degree to which we need to 12 satisfy the staff. That's why we have_to ask the 13 -question, because obviously we were satisfiediour -
14 selves.
15 - ER. FIKAR: We want to satisfy you within 16 some reasonable bounds, that's all I'm saying.
'17: , MR. IPPOLITO: Let me ask John a question.
18 Were you presented with the-$ asis upon which these 19 conservatisms were considered to be bound, the 20 problem?
'21' MR. FAIR: Well, they had two separate basis.
- ' 22 - Dr. Iotti had the test data on a particular snubber 23- which showed that it would not fail under the load 24 they had-calculated.
25 MR. IOTTI: That's the same snubber that was
s
,c--
53 1 calculated to be overloaded, PSA1.
2 -
MR. FAIR: .And the second basis was another 3 summary. disposition on safety factors of the seismic 4 design, which is somewhat of a -- we characterize
. 5 them as medium margins of safety in various ana-6- lytical processes in the' seismic design. There was 7 enough. uncertainty associated with them that we 8 wouldn't use those numbers straight from that affi-19 ' davit. -
10 MR. IOTTI: That, of course, was your char-11 acterization. The medium margin of safety we're
-12 talking about is a factor of safety of forty-four
- ( -
13' -~
zero. Maybe we still consider-that medium. We
-14 never made a statement that that was the true mar-15- gin of safety there, but it's of that order. We 16 are not talking factual twenty percent, forty per-
- 17. cent, fifty percent, we are talking large factors 18 in input. Primarily, that dIesn't come from'the 19 method of analysis, it comes from the actual seis-20 , mic input. ,
'21 MR. FIKAR: That's my point, again, that the 22 seismic criteria, we get such a large safety factor, 23 which is really what this is all about, looking at
- > 24 the plant and knowing what we've got here, there's 25 - got to be some reason to all of this.
Am --
-~ . . .
54
'l MR. BOSNAK: When we were here two years ago, 2 this was the problem at that time, and it sounds 3 like it still is. We haven't made much progress.
4' And the point was . that it was not that you couldn't
- 5. use generic stiffness, but it was the deflection
-- 6 criteria on the supports particularly where you had 7 a light loaded support, and we recognize that, you 8 know, seismic load. The definition that goes into 9 the load is very conservative, but in a wa'y, that's 10 an apple and an orange we're looking at.
11 MR. IOTTI: I agree with that.
' ~
12 MR. BOSNAK: It sound like you're making so.me
( 13 progress on how do you identify the supports that 14 we need to worry about, and I don't think we want 15 to introduce the -- you know, the effect of seismic 16- analysis and the conservatism there. We are trying 17- to look at a particular support load path and which 18 ones are we going to be concIrned about.
19 MR. IOTTI: The reason I made that statement, 20 was because Tom was a little offended at what he 21 thought was the applicants' attitude. That's-not 22 'the applicants' attitude. We have been concerned, 23 what'do we do now that actul stiffnesses are dif-24 ferent. We concluded in our own mind that it wasn't 25 a safety problem, and we stopped.
I
--- - w
C' ;i-
'd '
' ~ ' ~ ~ ' ~ - - ' -- ' - -
I:
ar 55
- 1 MR. BOSNAK: I think you're going down the p , ,
2: right path. The thing is when you have finished what'you have started, then we need to take a look
~
3
'4: at what you have.
. 5: MR. IOTTI: I agree with that. The reason
'6 we're asking the question is that,.for instance, 7 when I complete the designing criteria and I apply 4 -!
- 8. it, I would consider -- I will go far enough to 9 satisfy myself. I am assuming I will find no sur-10 prises. I am assuming that' it has convinced me.
11 . So far that again has been worn out. I will be 12 very suprised if I don't. Given that, I will say i 13 I have done enough, I'm happy, there's no safety 14 problem. That'might not satisfy.you.
15 MR.-BOSNAK: We may or may not agree depending 16 on what you - have looked at, T' guess.
17- MR. IOTTI: I guess to the extent that you have
-18 agreed that that's the propen way of proceeding, 19 I'm. happy. At least I know what. I'm -shooting for.
20 ,
I am not' shooting at an open-ended . type of approach,
~21 MR.-IPPOLITO: Are we sending that message?
22 .Is that the message you want to give them?
23 MR.-BOSNAK: Yes.
.24 MR. IPPOLITO: Okay. I just wanted to make it il5 clear.
r
,...- , . ~ -- . . . . . - . ..
56 1- MR. IOTTI: Well, maybe the 30th of this month, 2 this whole thing will be put to bed. I hope so.
3 MR. IPPOLITO: Okay. Good.
4 MR. FIKAR: I do too, I hope so.
5 MR. IPPOLITO: Anything else on this subje,ct?
.. 6 Okay. We're ready to go on to another of your 7 summary dispositions. John, correct me if I char-8 acterize it wrong. It's'U bolts or U bolts used as 9 two-way restraints.
10 MR. FAIR: That's cotJact.
11 }Ut . IPPOLITO: John was in the process of com-12 pleting his evaluation, and what he wanted to do is
( 13 to apparently go look at what were identified as 14 -maybe eight or so of the most significant of these 15 U bolts and just confirm that, yes, in fact the 16 plant is built as you have indicated in your de-17 . sign and in your drawing. He went this morning with 18 your people, and the-thermal'~agra , whatever was 19 over.it, was removed in front with John present.
-20 John made some measurements. And I would like to 21 have John tell you what he discovered while he was 22 there. Did you see all eight?
23 MR. FAIR: Yes.
24 MR. IOTTI: John, are these the same eight 25 that we have identified as the --
.w
.. o 57 11 MR.' FAIR: Yes, I was going to lead this with 2 .a little recap of the motion These are the cases 3 where you use U bolts for one-way restraints with J4 the presumption that they provided no lateral re-
~
~5 straints to the piping system. There were a limited
^
6' number which you identified'in the affidavit. The 7- . argument in the affidavit,.the steps of the argue-8 ment,.went this way: there was a sixteenth of an
~9 inch gap designed'in these 0 bolts. You identified
-10 eight'of them where the total movement exceeded 11 -
that one sixteenth of an inch total seismic plus
, 12 thermal. You did some sample re-analysis, seismic s
13 --
re-analysis, assuming no gap present for two of the 14 U bolts that had . the largest total displacement,
.15 .found no problem with the piping analysis. Tested 16- a couple of sizes of U. bolts, determine that they 17- could take substantial lateral deflections, and 18 then concluded that there wa'd no safety concern 19 with.these seventy U bolts. I have no problems with
'20 your technical arguments that you put forth. The
'll 'only thing I . wanted to do was to go out to the 22 field and take these eight U bolts and confirm the
- 23. existence and nonexistence of these gaps. As it 24 turns out, when I took the measurements this morning ,
~25 the sixteenth of an inch gap existed in the direc-o
a . . . . ,_., __ -.. -
58
'l tion of restraint and not in the lateral direction.
2 As a matter of fact, six out of the eight supports 3 I checked had lateral gaps less than one sixteenth 4 of an inch, and only two of them had a sixteenth of an inch gap or better in the lateral direction.
. 6 MR. IOTTI: I'm not sure I follow. What do 7 you: consider-directional restraints? Directional 8 restraint is up?
9 MR. FAIR: That's correct.
-10 MR. IOTTI: Can you resummarize it again? You 11 said'six out of eight had no gap laterally?
.12 MR. FAIR: Had lateral gaps less than one sixteenth of an inch, this is total lateral gap on 13:
14 both sides.
15 MR. IOTTI: Right. That's the way it's 16 stated here.
17 MR. FAIR: There was one that had practical'ly 18 zero gap in the lateral dird$ tion.
19 MR. IOTTI: That's total?
20 MR. FAIR: Total, both sides. Therefore, I 21 was unable to confirm that the sixteenth of an 22 inch gap exists on these U bolts.
23 - MR. IOTTI: You said something else about the 24 restraint direction having a sixteenth of an inch--
25 12. FAIR: I also looked at the drawings as m._
,: 1 =-i
, 59 1 I was reviewing these supports, and the sixteenth
- 2. of an inch gap that is specified on the drawing is 3 in the direction of the restraint and not the 4 lateral direction.
5 MR. IOTTI: 'Which systems were they on? ,Do 6 you recall?
7 MR. FAIR: Majority of them were component 8 tooling.
9 MR. IOTTI: Okay. So that's been tested, 10 hasn't it?
11 .
MR. FIKAR: Uh-huh.
12 MR. IOTTI: All right. Well, I don't have an
(' 13 --
answer to this question for you. I would like to 14- discuss what' the impact of this would be on any of 15 our technical results, -because I don' t believe it 16 would have any in terms of the lateral capability 17~ - even if it acts as a restraint, which it would, it-18 would still not impact, becaEse if anything, the 19 gaps are small, and we assume the loads on lateral 20 would be even smaller. There would be some impact 21 on thermal, which had been discussed. So far as
- 22. the conclusion on our affidavit, that would not 23 change. We may have a different thing that we need 24- to address as to why the sixteenth of an inch of a 25 gap is not there. Did you follow what I said?
f L
T' ';
,,.~- g : . ,.- - .-. . . - . . . . . . . }
50 1 MR. IPPOLITO: I think that's what John said
- c _
2 'was he had no problems with your analysis, but the 3 plant apparently does not conform to your analysis.
- 4 MR.' FAIR: Well, 'I'm not sure that that's the i5 . case. -
1 -
. -6 MR. IOTTI: What I'm trying to say if' we were '
7 ~~
- to analyze, but that I would come out even better.
8= I've made worse assumptions. To me the larger the 9
gap'-- when I analyzed it, I analyzed it without'any
~10 gap. The larger a gap, the more a deflection. What 11 I' assume that lateral load is higher, because I'm 12: letting lLt deflect to a larger portion, so with the J
7.;. - 13' gap, there's even..less -- I would-have even less of V 14 a ' load.- The results don't change. His problem is.
.=
15 -where's the sixteenth inch of a gap.
116 MR.. FAIR:- Let me say I have.two problems. One 17 problem is that the .. summary disposition' motion with 18 it states that there is a.si'xteenth of an inch gap,
- 19 that leads- me to the conclusion that the sixteenth 20 Lof an inch. gap was in a' lateral direction, and there--
21-fore, we were only concerned with.eight total sup-
-22, ports. We.tooki a sample of the worst two and showed 23 no problem, and I thought it was a pretty good
-24
. sample. -The second is in the basic assumption of 25 -ignoring the lateral restraints. Now, if you do t
^ -
,,, g . - .: . -.. =
61 E' l have:a gap in there and it did exceed the movement,
'2 . I -have no problem with making that assumption. '
3 'dowever, 'if the gap does not exist there, I consider 4 that a bad . assumption even ~if it does not turn out L5 to_be a safety problem. I think that was a poor
~
6 engineering practice.
7 MR. IOTTI: Well, yeah, maybe it was. I'm not.
7 8 quarreling with that, because I don't see any rea-9 son to go' over spilled' milk. I mean that has al-
- 10 ' ready been made abundantly clear before. But'is 11 -
there an adverse consequence from a bad assumption?
12 The answer-is.no in any. case. I don't necessarily.
-13 agree with your characterization either when you 14' say the gap is less than sixteenth of an -inch. But 15 how much less. Clearly in the case where there is 16 no gap, I may agree:with you, il7 MR.' FAIR: Let me state I only looked at those
-18 supports, those eight, which the movements were 19- greater than one-sixteenth.
20
, MR. .IOTTI: Total thermal plus seismic. I-
'21 guess implicitly we agree with you in that instance, 7 ..
- 22 it would act.as a lateral restraint. Okay. But 23.: there is no safety implication for adverse -- you l
24 know, implication reversed to safety from that I 25 U bolt acting as a restraint. I would like-to keep l
_ _. _m _ _. _ _ .. - _ -
w+
,. . ,i m-62 the two issues separate whether the gap is really
- 2 there or not from the. conclusion of the. affidavit.
3 ' if.I can. .If I may from the work that I did, this 4 is my own personal work. .From what you tell me, 5 that doesn't invalidate any of this other than,the
., .6 dimensions may not be correct in this particular 7 support. The conclusion would still be the same
'8 that there is ample ability of the bolt to with-9 stand the lateral load that would be imposed on it,
- 10 in fact the lateral load might even be left less.
11 Secondly, that the stresses on the piping system
-12 would not be adversely affected, because we as' sum led
(_ 13s in our analysis that there was no gap.
14 MR. FAIR: You analyzed just the seismic, and!
't 115 at the meeting I asked you why you didn't analyze
- 1<6 the thermal, and you said the gap existed.'
-17 ,
MR. IOTTI: Yes, that's true. The thermal had
= 18- been analyzed, they had analized the thermal.
19' . MR. FAIR:- They had no sample,
'20 MR. VOLLMER: No gap.
l 21 MR. IOTTI: I have to go back. For some reasor j l-l22 this.whole section has disappeared from my affi-T 23 davit. Do you have the affidavit-with you? I think 24 we need to take some time to go-find my attachments.
25 ,
.They were attached,' the table was one, and all of
- ._ - .-. . ~ - . . . - .-
63
-1 the attachments later, I don't-remember the number 2 _
now, disappeared from my entire package here, so I 3 need to refresh my memory.
_4 MR. BOSNAK: I.think normal-practice would be 5 to do it without a gap.- I don't know what they did.
6- ' s- MR. IOTTI: Rathar than make a fool out of 7 .myself and.me say one way or the other, I would say 8 no.
. .y 9: MR. FAIR: I believe from the last meeting we 10 haddeterminedthatthosedarticularanalysesmay 11 .
be superceded by some field work.
12 MR. .IOTTI: Unfortunately, I don't have the I' C 13 stuff with me. It would seem to me it was probably 14 -
analyzed without the gap, but.I can't assure you of.
15' that.
16 MR. FAIR: I'would assume that that would have 17 been correct, but as I was saying, I be$ieve that
- 18. those analyses may be superceded because of some 19 field modifications.
7y 20 MR. IOTTI: That's correct. The prior
- '21- analysis where the thermal analysis indicated that s
22~ . the restraints -- that the U-bolts would act as c
, s
-23 two-way restraint where the thermal expansion was 24- larger, so through that sixteenth of an inch gap, c25 they modified the support.
1..
. 64
- 1' [A short time was had off the record. Thereaftnr, the fol-2' lowing proceedings were had.]
3-4 MR. IOTTI: When we left off, there was some
'5 ~ question as to what Gibbs-Hill might have done,with
.- , :6_ regard to : some of the analyses with the two-way 7 restraining action of the U-bolt had been con-
, 3 sidered in determining stresses, independent of 9 that later on, whatever supports had been found 10 constrainJng the motion was modified. There was 11 also~-- I was also to confirm what we had done
. 12 with regard to re-analysis in the affidavit. And r 13 .let me add the second portion first in re-
\
14: assessing typing stresses increased in supports 15 other than those that would act as two-way 16 restraints and also increased in support loads as j 17 added to it as would be happening to the supports
- 11g that act as two-way restraints. The applicants 19 only re-analyzed.the seismic portion.
And the
'20 reason we did so is because it was believed that
, . 21 the-sixteenth inch gap or whatever lateral would 22- not effect' the thermal analysis at all. So we 23 were interested as to how the seismic load would
. 24 be effected, seismic stresses, seismic loads that 25 would remain tube support, that's one item.
L'- ..] ]
E~ -
.,. e 65
- 1- However, in assessing.the capability of the U-bolt 2 to take whatever loads might be imparted to it and 3 what margin it would have on that load, I want to
'4 again clarify that applicants used the combination J
5l of thermal' plus seismic. We assumed the total,.
6' displacement as calculated on the basis .there would 7 be no lateral combination of thermal and seismic 8 would be applied, so insofar as a U-bolt on accepted 9 loads, I don't think there is any question in my 10 mind that it is capable of doing so with regard to 11 -
possible effect of pipe stresses since no analysis 12 has been redone with regard to the effect on thermal 13 if the gap that is supposed to be there is not
(
- 24 there. All I could do is go back to some of the 15 initial analyses of the same problem, for instance, 16 problem AB-162E was re-analyzed by applicants for 17 .the. purpose of the affidavit. It was analyzed by 18 Gibbs-Hill. When Gibbs dill'ran the analysis with 19 the gap as being non-existent to find out what the 20 ,
effect of the restraining action would be insofar
'21 as thermal results, the stresses of the equasion 22 ten went up to about twenty-five percent, not 23 everywhere, but the highest stresses went up to 24 about twenty-five percent. If you were to take
(
25 that as being the characteristic of the responses, r
I d.h .' _
7, ,.
- 66
- 1 you would get that you might see that is added on 2 to the stresses , if you were to calculate the
-3 normal stresses, just the normal. If you were to 41 . do just. thermal, it'would go up twenty-five percent.
5- Thermal is not a significant portion of the analy-
. 6 . sis, but I haven't done an analysis, so you have
- 7 to take . this with a grain of salt. I did not. do 8 the prior analysis, Gibbs-Hill did, so I can only 9 look at their results 'and attempt to draw conclu-10 sions from those results.
11 'MR. FAIR: I think that we're going to have 12 to get this particular motion straightened out
-( 13 with what actually exists where the applicants 14 designed their sixteenth of an inch gap in the 15 supports and a conclusion on the safety conse-16 quences.
17 MR. IOTTI: Well, I don't anticipate any
'18 problem with'that, and it's'E short turnaround.
19 All we'll have to do is add the thermal analysis, 20 the same problem again., that is with regard to the 21 safety issues, with regard to why the gap that's 22 shown on the drawing is not there. I consider
' 23 that a separate issue.
24 MR. FAIR: Let's get that straightened out.
25 I happen to.have ail the drawings with me. The gap t
I
g- ,' ..
-w
- p 67
- 1. that's'shown on the drawing in_the direction of the 2 restraint. I have gone and essentially confirmed m
3 that that gap does not exist in the direction of 4 ' the' restraint. The concern here, of course, is 5 not the restraint, but the unrestrained directipn.
6 And I have measured various gaps from practically 7- zero to slightly above a sixteenth of an inch 3 -- total-on each side. You may want to reconfirm 9- that by your own measurements, but it wasn't very 10 easy to do on some of those supports.
11' _ .
.MR. IOTTI: I think the most straightforward
- 12 way is for.us to assume there is no gap and tell
- 13 you what the consequences are.
14- MR. FAIR: I also have to conclude that it 15 would have been appropr.iate whether or not this
!!6 turns out to be a safety concern, it was an
, _ 17 inappropriate assumption not to assume that these 18 supports provided lateral restraint. There was 19 no basis for making the assumption.
(;
_ 20 MR.'IOTTI: I guess I would have to agree 121 with you for that fact that they were also -intended 22- not to have any gap. The conclusion is fairly 23 inescapable.
-24 MR. WADE: According to John Byner that I 25 just talked to, he said that's intended for there
}T i.
m-
- +(r.. := s. .2. . - - : .. . . . .
~ JE . 68 l-1 :to be a restraint rack on the drawings.
ea:
2, .MR.~IOTTI: He's talking'about unrestrained.
?3 These confirmed.that that is not the problem. The
.4 problem is.the other.. Okay. I guess we'll have to
'5 modify the affidavit to reflect that new analysis.
_a . 6 That's-a short turnaround. We can have-that done 7- by- the end of next week. That's not the problem.
,'8 The T other part, I' guess I have ~ to talk' to some z ~ 9. more people here why it's not there and why the 110: drawings:show it'as'such.-
,11; MR. FAIR: I want to. reiterate the gap as.
12 exists 'as shown in the drawings. .
' 13 - MR. IOTTI: Your question is why do we say
'.14 ' there was no restraining action where it wasn't P
15- supposed to be? ' Yeah, I think it's a point we 16 have'to address. clearly. It was obviously'our 17: understanding that it was, we would have never 18 written it otherwise. - '~
c 19 MR. IPPOLITO: I would appreciate you resiewing
.20 those drawings.
t 21 .MR. IOTTI: Well, we'll do.so, you can rest
'22 assured.
-(=
23 - MR . IPPOLITO:' As a double check, because if p
- 24 we are wrong, I would like to know we are wrong, 115 but, you know, we' will look at those drawings , and u
r:.
69 1 if they age the correct drawing and all that shows -
that its dimension's not given.
3- MR._IOTTI: We understand your point, and we 4; willLaddress it. Does the: approach we have out-5 lined satisfy you? Same problems, give you the 6 results' drawing the same conclusions as we did in 7- the affidavit plus the question of the drawings lg and the appropriateness of making that assumption.
.9- MR. FAIR: Yes.
10 MR. IPPOLITO: Is ther'e another agenda item?
11 .
MR. BURWELL: I believe'you wanted to get a 12
- status report on the original insert information, "ft
.s
. 13' is-that true?
- 14. MR. IOTTI: Well, again, I don't have all the 15 answers, but I've got enough, I think, that I can 16 give you status report on it.
17- . ER . BURWELL: I just wanted to clar:fy that.
13 ~ The next item that we are dis' cussing is the 19 applicants' motion for summary disposition regard -
20 ing the design of. Richmond insert and the applica-
'21 tion to support design. This motion was filed with 22 us on June the 2nd, 1984.
23 MR. IOTTI: As we lef t the meeting of- the 8th 24' and 9th, we had agreed upon a set of action items.
25 Applicants were supposed to provide additional G: a
( _.
- cc ,
70 1- information to.the staff. One of the items we
.2~ don' t have today,t ' his is the' details of the 8th 3 Richmond insert pattern, and I am not in a position, 4 even if I wanted to, to give you the plan on it.
5 That is something that is going to have to wait 6 until John gets back, again our schedule of com-p.
7 pleting it and getting it by the end of next week.
8 You will have all this. One additional item was --
9 we did go back and the particular elements -- the 10 stresses are computed as are outlined elements.
. 11 There is a finite layer 'on the outside. They are
' 12 -not an extreme fiber, on the other hand, it's a 13 fin element. There were two aspects to the 14 question. Whether one should even attempt to 15 derive the bending movement to the finite element.
16 The other one is, of course, where the stress is
' 17 .being calculated. To attempt to answer the 18 question as to why we felt i't was necessary to go 19 to finite element analysis. We believed there was 20 sheer contribution to this particular stress 21 pattern in the bolts, it wasn't just pure bending.
22 As I . pointed out to you, we had two test cases I 23 that were run, one under tortion and the other one h-
. 24 under sheer in the field with the tube steel off-
- 25 set -- the hole in 'the tube steel of fset to the ,
e
y j _ ._ .-
71 1= Point in one particular instance you could only
- 2. see: pure bending. 'If you are ever' going to get 3 pure bending in the bolt, that's going to be the 14 case. We then looked at the results of the test 5- and attempted to superimpose on those results Ahe-6 loads. The one we calculated assuming pure 7 bending, MC over I, and we did so by simulating g the bolt as a simple cantilever and also a guided 9 -cantilever, because its behavior as a guided 10 cantilever was a difference of the two between
'll. them.- Again, these results are fresh off the 12 press, if you will, and they haven't been reviewed,
' j- 13 so 'I'm somewhat hesitant to show them to you. If V
~14 you look at that particular one, it shows in 15 'either' case the cantilever, the guided cantilever 16- are not quite full. The loads depicted by the 17 test is higher than what.you would predict by the 13 analytical approach using MC'over I. That means 19' there's some resistance there. It isn't just 20 bending resistance, that's why we elected to go
'21 to finite elements to find out how much of that 22 resistance was due to bending and how much was v.
f" 23 due to sheer. We don't know of any other way to 24 do so other than to do a finite analysis. You are
. 23 - -welcome to glance at these results. Again, bear u=
.= : 2: ~ .. ... - ~
. 72 1 ~in mind, they haven't even been reviewed, they 2 have just become available.
3 MR. FAIR: I'm not sure what you just told me.
4 You are relying 'on the load reflection on.:the test 5 to verify -- .
.? 6 MR. IOTTI: To verify that there is more to L e 7 the bolt -- that simulating the bolt as if it --
8
[A short time is had off the
~9 record. Thereafter, the fol-lowing proceedings were had.]
10 111 ~ MR. IOTTI: What I was saying is that as you' 12 P l ot load reflection, you can equally plot move-i 13- ment rotation if in fact what you see in the bolt 14 is essentially pure. bending, then that bolt 15' essentially behaves as a cantilever or guided 16 cantilever. You should be able to match the 17 ,results of the test fairly well to the results
.18 you have there. If that bo1T doesn't behave, a
'19 flectual member -- it's a short stout flectual
-20 member, as well as both sheer, you have to deter-21 mine which part is due to sheer, which part is due 22 to flectual resistance. The only way we know how 23 to do that is to find finite element analysis, 24 that's a way of rationalizing where we sent to 25 finite analysis of 'the bolt. We suspected that it u
=
s 73 i
II' was something more than just M over I effects, 2- and the 'results thatL we got.ou of finita clements 3 did confirm that. .cc :: 'ht.x this addresses the 4- particular action as to why we can reduce the
- 5 stresses that you would compute from bending if 6' you just use the MC over I approach. That's what 7 'I thought was one of the action items that we 8 agreed on, and this will:be provided to you as a 9 rationalization.' If you don't agree, your next
~10 question will be how many more of those supports 11 -
of table A or B will be added to the number that
- 12 will have to investigate further. I don't recall j' 13 the exact number,but it seems like twelve more-14~ supports will have to be added, if I recall correct-15 ly, so we are not talking about a large number, in 16 any-case. The third action item is confirmed with 17 the fatigue analysis that we had done included.in 18 the affidavit was performed'$ sing S836 or had in 19 fact accounted for the high strength bolt. We did 20 , go back, and it was done strictly on S836. Now,
~ 21 we have gone back and redone a fatigue analysis for 22 those few high strength bolts. We were also to 23 - go back and verify whether in some tube steels are 24' there still instances with the MZ Movement -- the
- 25 MZ Movement rotating is not being released. And L
r
,p ,r p .=-
y7
- x .
~ 'cw g -
y-1 :- - ~~- - ~'- -
fn +
en
, 74
>i_
., y g
n,
- 1! :the answer-to-that isEthere'aren't any MZ Move-WG ' -2 'ments:also released..-
af 4 y-. ,
P
_' ;3 MR. - . FAIR : This'is current design practice, i
~ ~
. 4 :. .
MR. f IOTTI: Well, no. This.is'something that 5 I1 don' t want_ to get _too _ muc.a into it, but we may
^
_ b.} . i, 6 thave:to modify our affidavit. There,was 'some mis--
W
^
- .75 ,
und'erstanding of: what we ' believe- the Mz design
~^
. [8 people -- the analysis was misinterpreted. -They w -
'91 'in fact gave us the Mz Movement. The Mz Movement
-10 -apparently was always released. So that's the-2;.
sil! , information I have as of today, but I haven't had -
_ ( 12 .
- a chance to go back and. talk to the; people who .
. . e --
' ':% : 113: actually did the analysis, so take _ that with a
.y
- 14 grain of salt again. The last time'it was;-- on
'15
~
providing additional'information, the prying t
- 16i action for larger size tube steel, that would be 17 forty.-eight' inches as opposed to twenty-four inci.
Again,;I'llgiveyouanunreNiewedresult,'butLit
- - J18-
.,4
'" ~
appears. not to be a problem from the standpoint of -
~
_ l19L.
. 20' '
_ prying force, not necessarily the prying action,
, . , , . 21' the prying ratio _. Okay. Because ultimately it's.
m -' . _.
74- ,22
\7 the prying force. The six-by-six tube steel with w
-23 Lone-inch inserts. We will be providing you with
+
- U24-the table that shows the various tube sizes-from x,
25! four-by.-four, three'and three eighths, all the way.
\ s 3
c n ,
i + <
,._,.~~_--+~,.-,AE~<
m- _
75 1; to eight-by-eight by a half, and it will show the 2 reaction ~with or without the end restra"ints and 3 give you prime factor for each of them. So I think 4 that addresses all of the action items that we 5 agreed upon at our last m uting. As I said , ,I.
6 don't have any information at all on this eighth 7 insert pattern. If you want, we'll try and define 8 it this afternoon and go look at it, but I have 9 got nothing at all with regard to that.
10 MR. FAIR: I think tha't may be worthwhile to 11 .
walk out and take a look at it.
12 MR. IOTTI: First, we've got. to identify where
( 13 --
it is. I haven't the fcggiest idea. It will take 14 us some time. Once we identify it, we'll take 15 -you there. .
16 MR. FAIR: Let me recap on'this MZ Movement.
17 If you confirm that your information is correct, 18 it was never modeled as fixed.
19 MR. IOTTI: That's correct, which means that 20 our affidavit mak.es an unwarranted assumption that 21 the information that we have to be correct or to
' 2:2 our best belief may in retrospect turn out to be 23 incorrect, because the analysts were confusing in 24 the MS Movement.
25 MR. FAIR: The other side , there are some h
e . - - -
e
, ,. S a_-/ -l- h .a a r
- mt~ LwA '=*?
76 d
I longer spans-than your. typical twenty to twenty-4
-2 ~ four inches,1and the-longest is forty-eight~ inch.
3: MR. IOTTI: The longest-is forty-eight.- We
[4 _ looked at all'different type tube steels to-address, t , 5 .
your question on prying action, and we have thAt t .
6' tabulated both 1in terms of pryint factors as well
, 7 as the-prying force.
i,
( 8 MR. FAIR: Juui that was done in a generic x9 fashion versus looking at specific supports.
y 10 MR. IOTTI: Well, it was a combination. The
- 11 - study was done. in ~ generic f ashion, but what was 12 doneLspecifically is to confirm there was nothing
_. '13 - longer than-forty-eight. There are some six-by-s 14 six-by-three eighths with a one-inch insert.
.15 MR. FAIR: I.think that's it.
16: MR. IPPOLITO: Just let me ask, this is going 17' to be submitted when?
[ ,18 , MR. IOTTI: The whole pIckage, everything that-19 we have agreed upon the meeting of the 8th will be.
L 20 - turned over to.you hopefully by the 30th.
21~ MR. IPPOLITO: And this includes the revision r-22 of the affidavit, too?
23 MR. IOTTI: 'Yes, if in fact we can confirm L24 that.- You can understand now, I am thoroughly c.
- 25. ' confused , because I' investigated that quite L
_ _ , _ _ - . - - . ~ . _ _ . _ .
+: a;~ . .a .- - -
-;:,;; g;; a. ~ -
77 l' ' extensively and 'actually sat down with the 2- analysts'as we were doing_'t. :I mean,1really,11f-
- y .
3
- 3~ Lin fact'we have to. revise th affidavit, the 30th-14 submittalLwill also includegtha e end of nexti g., ,
- 5. - week. <
g 6 :. 'MR. BURWELL: IIbelieve'the remaining part-of lH 7" Jit would~be a' visit to the plant to get a-look at
~
8[ some of'the hangers. I believe those three gang L9 hangers _were specifically one of the areas and the 10' vertical. hangers that was represented by' figure
~
11 .
one on page -- or type one on page nine of your 12- affidavit on stability.
v 13- MR. TADE: Before we leave, we have another 14 ' action planning going on axial restraint rotational 15 resistance question that we discussed a couple of
- 16 weeks ago. - I. am assuming since we ?are not discus-
.17 sing that here, that plan'inyour mind 'will solve a-18: plan that will bring that is$ue to an end-orcare 19 there other: issues there that we ought to discuss 20 ,
- right now., .
'21 MR. IPPOLITO: Let me make this cle .r , I
- 22 called a meeting primarily to resolve what I~
23 thought were the two significant areas where the
'24' . staff and the applicants could not converge. When 25~ I was' briefed last week, when I was back in L
k
m -
p" '" .
78 1 Washington, these were identified as areas that, 2 you know, we are never going to get there unless 3 .something is done, so we came h'ere prepared to do
~
.4 .that while we're here. John said that, " Hey, I
- 5. can clean up one of the areas by going down and,
.- . 6 looking at this gap." So that's the reason why, 7 you know, we included on the agenda, and I don't 8 know'whether we're prepared to talk about anything 9 else cn: not. Staff, tell me we're not, and we'll
,10 cut it off right here.
11- ~MR. FAIR: We're not.
12 LMR. BURWELL: We're not on the other, so in
(
- 13. answer te your-question ---
14 MR. IPPOLITO: .In answer to your question, I
-15 don't know if we've got a problem, but it wasn't 16 presen'ted to me as a problem that I had to, you 17 know, focus on.
18 (A short time is had off the 19 record. Thereafter, the fol-lowing proceedings are had.]
20 1, 21 MR. IPPOLITO: We have agreed that we will 22 tour the site at two o' clock, and that will close
'23 the meeting up.
.24 * **
25 END O. F PR'OCEEDINGS
o r 79 1 THE STATE OF TEXAS 1 s
2 COUNTY OF TARRANT 1 3-4 I, PAM ALFORD, Certified Shorthand Reporter and a 5 Notary Public in and for the State of Texas, do hereby ..v, 6 certify that the above and foregoing 78 typewritten pages, 7 contain a full, true and accurate transcription of my 8 stenographic notes taken upon the occasion set forth, and 9 reduced to typewriting by Ine or under my direction.
10 WITNESS MY HAND, this 5th day of September, A.D.
11 1984.
12 13 -
{ ,/
14 st. - $Y, PAM ALFORD., C.S .
15 Notary Public and for the State of xas.
16 17 18 -
19 20
-21 22 23 24 25 A